From a29ccf6f823a84d89e1c7aaaf221cf7282022024 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Tue, 3 Jun 2008 14:59:40 +0100 Subject: [PATCH 001/857] Make console charset translation optional By turning off the new CONSOLE_TRANSLATIONS option and dropping the associated code and tables from the kernel, we can save about 7KiB. Taken from linux-tiny project by Tim Bird and mangled further by dwmw2. Signed-off-by: Tim Bird Signed-off-by: David Woodhouse --- drivers/char/Kconfig | 8 ++++++++ drivers/char/Makefile | 4 ++-- drivers/char/vt.c | 2 +- include/linux/consolemap.h | 14 ++++++++++++++ include/linux/vt_kern.h | 19 +++++++++++++++++++ 5 files changed, 44 insertions(+), 3 deletions(-) diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig index 595a925c62a9..b7f7371dee73 100644 --- a/drivers/char/Kconfig +++ b/drivers/char/Kconfig @@ -36,6 +36,14 @@ config VT If unsure, say Y, or else you won't be able to do much with your new shiny Linux system :-) +config CONSOLE_TRANSLATIONS + depends on VT + default y + bool "Enable character translations in console" if EMBEDDED + ---help--- + This enables support for font mapping and Unicode translation + on virtual consoles. + config VT_CONSOLE bool "Support for console on virtual terminal" if EMBEDDED depends on VT diff --git a/drivers/char/Makefile b/drivers/char/Makefile index 4c1c584e9eb6..6ef173cab144 100644 --- a/drivers/char/Makefile +++ b/drivers/char/Makefile @@ -12,8 +12,8 @@ obj-y += mem.o random.o tty_io.o n_tty.o tty_ioctl.o obj-$(CONFIG_LEGACY_PTYS) += pty.o obj-$(CONFIG_UNIX98_PTYS) += pty.o obj-y += misc.o -obj-$(CONFIG_VT) += vt_ioctl.o vc_screen.o consolemap.o \ - consolemap_deftbl.o selection.o keyboard.o +obj-$(CONFIG_VT) += vt_ioctl.o vc_screen.o selection.o keyboard.o +obj-$(CONFIG_CONSOLE_TRANSLATIONS) += consolemap.o consolemap_deftbl.o obj-$(CONFIG_HW_CONSOLE) += vt.o defkeymap.o obj-$(CONFIG_AUDIT) += tty_audit.o obj-$(CONFIG_MAGIC_SYSRQ) += sysrq.o diff --git a/drivers/char/vt.c b/drivers/char/vt.c index fa1ffbf2c621..18b7fb06dace 100644 --- a/drivers/char/vt.c +++ b/drivers/char/vt.c @@ -2208,7 +2208,7 @@ static int do_con_write(struct tty_struct *tty, const unsigned char *buf, int co c = 0xfffd; tc = c; } else { /* no utf or alternate charset mode */ - tc = vc->vc_translate[vc->vc_toggle_meta ? (c | 0x80) : c]; + tc = vc_translate(vc, c); } param.c = tc; diff --git a/include/linux/consolemap.h b/include/linux/consolemap.h index e2bf7e5db39a..c4811da1338b 100644 --- a/include/linux/consolemap.h +++ b/include/linux/consolemap.h @@ -3,6 +3,9 @@ * * Interface between console.c, selection.c and consolemap.c */ +#ifndef __LINUX_CONSOLEMAP_H__ +#define __LINUX_CONSOLEMAP_H__ + #define LAT1_MAP 0 #define GRAF_MAP 1 #define IBMPC_MAP 2 @@ -10,6 +13,7 @@ #include +#ifdef CONFIG_CONSOLE_TRANSLATIONS struct vc_data; extern u16 inverse_translate(struct vc_data *conp, int glyph, int use_unicode); @@ -18,3 +22,13 @@ extern int conv_uni_to_pc(struct vc_data *conp, long ucs); extern u32 conv_8bit_to_uni(unsigned char c); extern int conv_uni_to_8bit(u32 uni); void console_map_init(void); +#else +#define inverse_translate(conp, glyph, uni) ((uint16_t)glyph) +#define set_translate(m, vc) ((unsigned short *)NULL) +#define conv_uni_to_pc(conp, ucs) ((int) (ucs > 0xff ? -1: ucs)) +#define conv_8bit_to_uni(c) ((uint32_t)(c)) +#define conv_uni_to_8bit(c) ((int) ((c) & 0xff)) +#define console_map_init(c) do { ; } while (0) +#endif /* CONFIG_CONSOLE_TRANSLATIONS */ + +#endif /* __LINUX_CONSOLEMAP_H__ */ diff --git a/include/linux/vt_kern.h b/include/linux/vt_kern.h index 9448ffbdcbf6..14c0e91be9b5 100644 --- a/include/linux/vt_kern.h +++ b/include/linux/vt_kern.h @@ -12,6 +12,7 @@ #include #include #include +#include /* * Presently, a lot of graphics programs do not restore the contents of @@ -54,6 +55,7 @@ void redraw_screen(struct vc_data *vc, int is_switch); struct tty_struct; int tioclinux(struct tty_struct *tty, unsigned long arg); +#ifdef CONFIG_CONSOLE_TRANSLATIONS /* consolemap.c */ struct unimapinit; @@ -71,6 +73,23 @@ void con_free_unimap(struct vc_data *vc); void con_protect_unimap(struct vc_data *vc, int rdonly); int con_copy_unimap(struct vc_data *dst_vc, struct vc_data *src_vc); +#define vc_translate(vc, c) ((vc)->vc_translate[(c) | \ + (vc)->vc_toggle_meta ? 0x80 : 0]) +#else +#define con_set_trans_old(arg) (0) +#define con_get_trans_old(arg) (-EINVAL) +#define con_set_trans_new(arg) (0) +#define con_get_trans_new(arg) (-EINVAL) +#define con_clear_unimap(vc, ui) (0) +#define con_set_unimap(vc, ct, list) (0) +#define con_set_default_unimap(vc) (0) +#define con_copy_unimap(d, s) (0) +#define con_get_unimap(vc, ct, uct, list) (-EINVAL) +#define con_free_unimap(vc) do { ; } while (0) + +#define vc_translate(vc, c) (c) +#endif + /* vt.c */ int vt_waitactive(int vt); void change_console(struct vc_data *new_vc); From 4a5e3638b11978262ab76bbb2062e57fefaaedba Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Tue, 15 Jul 2008 09:42:57 -0600 Subject: [PATCH 002/857] ACPI: stop complaints about interrupt link End Tags and blank IRQ descriptors Silently ignore _PRS End Tags. We already ignore Start Dependent Functions in _PRS, and we already ignore End Tags in _CRS, so we might as well ignore End Tags in _PRS as well. Silently ignore _PRS IRQ descriptors that mention no interrupts. The spec allows this (section 6.4.2.1 in ACPI 3.0b spec), and it probably means the interrupt link can't be configured at all. This patch doesn't change any functional behavior; it just removes confusing complaints like these: ACPI: Blank IRQ resource ACPI: Resource is not an IRQ entry when parsing _PRS data "23 00 00 18 79 00" from an IBM xSeries 335 dual Pentium IV Xeon 2.40 GHz machine. For more details, see http://bugzilla.kernel.org/show_bug.cgi?id=11049 The "23 00 00 18" part is a three-byte-long small IRQ resource with no bits set in the IRQ mask ("00 00"), and level-triggered, active low, shareable ("18"). The "79 00" is an End Tag (type 0x7). It is superfluous since there is no Start Dependent Function tag and there are no resources after it, but it is harmless. Thanks to Gabriele Trombetti (aka Kurk) for reporting this and testing the patch. Signed-off-by: Bjorn Helgaas Signed-off-by: Andi Kleen --- drivers/acpi/pci_link.c | 31 ++++++++++++++++++++----------- 1 file changed, 20 insertions(+), 11 deletions(-) diff --git a/drivers/acpi/pci_link.c b/drivers/acpi/pci_link.c index 233c40c51684..89f3b2abfdc7 100644 --- a/drivers/acpi/pci_link.c +++ b/drivers/acpi/pci_link.c @@ -113,20 +113,23 @@ acpi_pci_link_check_possible(struct acpi_resource *resource, void *context) switch (resource->type) { case ACPI_RESOURCE_TYPE_START_DEPENDENT: + case ACPI_RESOURCE_TYPE_END_TAG: return AE_OK; case ACPI_RESOURCE_TYPE_IRQ: { struct acpi_resource_irq *p = &resource->data.irq; if (!p || !p->interrupt_count) { - printk(KERN_WARNING PREFIX "Blank IRQ resource\n"); + ACPI_DEBUG_PRINT((ACPI_DB_INFO, + "Blank _PRS IRQ resource\n")); return AE_OK; } for (i = 0; (i < p->interrupt_count && i < ACPI_PCI_LINK_MAX_POSSIBLE); i++) { if (!p->interrupts[i]) { - printk(KERN_WARNING PREFIX "Invalid IRQ %d\n", - p->interrupts[i]); + printk(KERN_WARNING PREFIX + "Invalid _PRS IRQ %d\n", + p->interrupts[i]); continue; } link->irq.possible[i] = p->interrupts[i]; @@ -143,15 +146,16 @@ acpi_pci_link_check_possible(struct acpi_resource *resource, void *context) &resource->data.extended_irq; if (!p || !p->interrupt_count) { printk(KERN_WARNING PREFIX - "Blank EXT IRQ resource\n"); + "Blank _PRS EXT IRQ resource\n"); return AE_OK; } for (i = 0; (i < p->interrupt_count && i < ACPI_PCI_LINK_MAX_POSSIBLE); i++) { if (!p->interrupts[i]) { - printk(KERN_WARNING PREFIX "Invalid IRQ %d\n", - p->interrupts[i]); + printk(KERN_WARNING PREFIX + "Invalid _PRS IRQ %d\n", + p->interrupts[i]); continue; } link->irq.possible[i] = p->interrupts[i]; @@ -163,7 +167,8 @@ acpi_pci_link_check_possible(struct acpi_resource *resource, void *context) break; } default: - printk(KERN_ERR PREFIX "Resource is not an IRQ entry\n"); + printk(KERN_ERR PREFIX "_PRS resource type 0x%x isn't an IRQ\n", + resource->type); return AE_OK; } @@ -199,6 +204,9 @@ acpi_pci_link_check_current(struct acpi_resource *resource, void *context) switch (resource->type) { + case ACPI_RESOURCE_TYPE_START_DEPENDENT: + case ACPI_RESOURCE_TYPE_END_TAG: + return AE_OK; case ACPI_RESOURCE_TYPE_IRQ: { struct acpi_resource_irq *p = &resource->data.irq; @@ -208,7 +216,7 @@ acpi_pci_link_check_current(struct acpi_resource *resource, void *context) * particularly those those w/ _STA disabled */ ACPI_DEBUG_PRINT((ACPI_DB_INFO, - "Blank IRQ resource\n")); + "Blank _CRS IRQ resource\n")); return AE_OK; } *irq = p->interrupts[0]; @@ -224,7 +232,7 @@ acpi_pci_link_check_current(struct acpi_resource *resource, void *context) * return at least 1 IRQ */ printk(KERN_WARNING PREFIX - "Blank EXT IRQ resource\n"); + "Blank _CRS EXT IRQ resource\n"); return AE_OK; } *irq = p->interrupts[0]; @@ -232,10 +240,11 @@ acpi_pci_link_check_current(struct acpi_resource *resource, void *context) } break; default: - printk(KERN_ERR PREFIX "Resource %d isn't an IRQ\n", resource->type); - case ACPI_RESOURCE_TYPE_END_TAG: + printk(KERN_ERR PREFIX "_CRS resource type 0x%x isn't an IRQ\n", + resource->type); return AE_OK; } + return AE_CTRL_TERMINATE; } From c2c789057f075022658b38b498755c29c1ba8055 Mon Sep 17 00:00:00 2001 From: Zhao Yakui Date: Thu, 17 Jul 2008 10:46:05 +0800 Subject: [PATCH 003/857] ACPI: Ignore _BQC object when registering backlight device According to acpi spec , the objectes of _BCL and _BCM are required if integrated LCD is present and supports brightness level and the _BQC is the optional object. So the _BQC object will be ignored when the backlight device is registered. At the same time when there is no _BQC object, the current brightness will be set to the maximum. http://bugzilla.kernel.org/show_bug.cgi?id=10206 Signed-off-by: Zhao Yakui Signed-off-by: Zhang Rui Signed-off-by: Andi Kleen --- drivers/acpi/video.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) diff --git a/drivers/acpi/video.c b/drivers/acpi/video.c index 64c889331f3b..e32b6c14d928 100644 --- a/drivers/acpi/video.c +++ b/drivers/acpi/video.c @@ -741,7 +741,7 @@ static void acpi_video_device_find_cap(struct acpi_video_device *device) max_level = acpi_video_init_brightness(device); - if (device->cap._BCL && device->cap._BCM && device->cap._BQC && max_level > 0){ + if (device->cap._BCL && device->cap._BCM && max_level > 0) { int result; static int count = 0; char *name; @@ -753,7 +753,17 @@ static void acpi_video_device_find_cap(struct acpi_video_device *device) device->backlight = backlight_device_register(name, NULL, device, &acpi_backlight_ops); device->backlight->props.max_brightness = device->brightness->count-3; - device->backlight->props.brightness = acpi_video_get_brightness(device->backlight); + /* + * If there exists the _BQC object, the _BQC object will be + * called to get the current backlight brightness. Otherwise + * the brightness will be set to the maximum. + */ + if (device->cap._BQC) + device->backlight->props.brightness = + acpi_video_get_brightness(device->backlight); + else + device->backlight->props.brightness = + device->backlight->props.max_brightness; backlight_update_status(device->backlight); kfree(name); From ea51011a27db48ea0a80a5e20de3969b292d5d4d Mon Sep 17 00:00:00 2001 From: Zhao Yakui Date: Mon, 14 Jul 2008 15:14:03 +0800 Subject: [PATCH 004/857] ACPI : Set FAN device to correct state in boot phase Subject:ACPI: Set FAN device to correct state in boot phase From: Zhao Yakui On some laptops when ACPI FAN driver is loaded, maybe the FAN device will be turned on. But if the temperature is below the threshold, the corresponding FAN device should be turned off in the course of loading thermal driver. So it is necessary to set the FAN device to the correct state in course of loading the thermal driver. http://bugzilla.kernel.org/show_bug.cgi?id=8049 Signed-off-by: Zhao Yakui Signed-off-by: Zhang Rui Signed-off-by: Andi Kleen --- drivers/acpi/thermal.c | 43 ++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 43 insertions(+) diff --git a/drivers/acpi/thermal.c b/drivers/acpi/thermal.c index 84c795fb9b1e..9adfd180df6a 100644 --- a/drivers/acpi/thermal.c +++ b/drivers/acpi/thermal.c @@ -769,6 +769,47 @@ static void acpi_thermal_run(unsigned long data) acpi_os_execute(OSL_GPE_HANDLER, acpi_thermal_check, (void *)data); } +static void acpi_thermal_active_off(void *data) +{ + int result = 0; + struct acpi_thermal *tz = data; + int i = 0; + int j = 0; + struct acpi_thermal_active *active = NULL; + + if (!tz) { + printk(KERN_ERR PREFIX "Invalid (NULL) context\n"); + return; + } + + result = acpi_thermal_get_temperature(tz); + if (result) + return; + + for (i = 0; i < ACPI_THERMAL_MAX_ACTIVE; i++) { + active = &(tz->trips.active[i]); + if (!active || !active->flags.valid) + break; + if (tz->temperature >= active->temperature) { + /* + * If the thermal temperature is greater than the + * active threshod, unnecessary to turn off the + * the active cooling device. + */ + continue; + } + /* + * Below Threshold? + * ---------------- + * Turn OFF all cooling devices associated with this + * threshold. + */ + for (j = 0; j < active->devices.count; j++) + result = acpi_bus_set_power(active->devices.handles[j], + ACPI_STATE_D3); + } +} + static void acpi_thermal_check(void *data) { int result = 0; @@ -1624,6 +1665,8 @@ static int acpi_thermal_add(struct acpi_device *device) init_timer(&tz->timer); + acpi_thermal_active_off(tz); + acpi_thermal_check(tz); status = acpi_install_notify_handler(device->handle, From b1d77fae0c429d1be84ca0c9e627d9ab0e2a6d0b Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Fri, 18 Jul 2008 01:42:20 +0200 Subject: [PATCH 005/857] Revert "Fix FADT parsing" This reverts commit 01a5bba576b9364b33f61f0cd9fa70c2cf5535e2. There seem to be some FADTs around with bogus information in the v2 fields. Revert this patch for now until this can be properly resolved. Signed-off-by: Andi Kleen --- drivers/acpi/tables/tbfadt.c | 17 ++++++----------- 1 file changed, 6 insertions(+), 11 deletions(-) diff --git a/drivers/acpi/tables/tbfadt.c b/drivers/acpi/tables/tbfadt.c index ccb5b64bbef3..a4a41ba2484b 100644 --- a/drivers/acpi/tables/tbfadt.c +++ b/drivers/acpi/tables/tbfadt.c @@ -124,7 +124,7 @@ static struct acpi_fadt_info fadt_info_table[] = { static void inline acpi_tb_init_generic_address(struct acpi_generic_address *generic_address, - u8 byte_width, u64 address) + u8 bit_width, u64 address) { /* @@ -136,7 +136,7 @@ acpi_tb_init_generic_address(struct acpi_generic_address *generic_address, /* All other fields are byte-wide */ generic_address->space_id = ACPI_ADR_SPACE_SYSTEM_IO; - generic_address->bit_width = byte_width << 3; + generic_address->bit_width = bit_width; generic_address->bit_offset = 0; generic_address->access_width = 0; } @@ -343,11 +343,9 @@ static void acpi_tb_convert_fadt(void) * * The PM event blocks are split into two register blocks, first is the * PM Status Register block, followed immediately by the PM Enable Register - * block. Each is of length (xpm1x_event_block.bit_width/2) + * block. Each is of length (pm1_event_length/2) */ - WARN_ON(ACPI_MOD_16(acpi_gbl_FADT.xpm1a_event_block.bit_width)); - pm1_register_length = (u8) ACPI_DIV_16(acpi_gbl_FADT - .xpm1a_event_block.bit_width); + pm1_register_length = (u8) ACPI_DIV_2(acpi_gbl_FADT.pm1_event_length); /* The PM1A register block is required */ @@ -362,17 +360,14 @@ static void acpi_tb_convert_fadt(void) /* The PM1B register block is optional, ignore if not present */ if (acpi_gbl_FADT.xpm1b_event_block.address) { - WARN_ON(ACPI_MOD_16(acpi_gbl_FADT.xpm1b_event_block.bit_width)); - pm1_register_length = (u8) ACPI_DIV_16(acpi_gbl_FADT - .xpm1b_event_block - .bit_width); acpi_tb_init_generic_address(&acpi_gbl_xpm1b_enable, pm1_register_length, (acpi_gbl_FADT.xpm1b_event_block. address + pm1_register_length)); /* Don't forget to copy space_id of the GAS */ acpi_gbl_xpm1b_enable.space_id = - acpi_gbl_FADT.xpm1b_event_block.space_id; + acpi_gbl_FADT.xpm1a_event_block.space_id; + } } From e1469c34eb623cd1945ef09bfd7de7bc2f9ff6b3 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Fri, 18 Jul 2008 01:43:08 +0200 Subject: [PATCH 006/857] Revert "dock: bay: Don't call acpi_walk_namespace() when ACPI is disabled." Revert double commit by mistake. Noticed by Thomas Gleixner. This reverts commit cc7e51666d82aedfd6b9a033ca1a10d71c21f1ca. Signed-off-by: Andi Kleen --- drivers/acpi/bay.c | 3 --- drivers/acpi/dock.c | 3 --- 2 files changed, 6 deletions(-) diff --git a/drivers/acpi/bay.c b/drivers/acpi/bay.c index e6caf5d42e0e..61b6c5beb2d3 100644 --- a/drivers/acpi/bay.c +++ b/drivers/acpi/bay.c @@ -377,9 +377,6 @@ static int __init bay_init(void) INIT_LIST_HEAD(&drive_bays); - if (acpi_disabled) - return -ENODEV; - if (acpi_disabled) return -ENODEV; diff --git a/drivers/acpi/dock.c b/drivers/acpi/dock.c index 1e872e79db33..bb7c51f712bd 100644 --- a/drivers/acpi/dock.c +++ b/drivers/acpi/dock.c @@ -917,9 +917,6 @@ static int __init dock_init(void) dock_station = NULL; - if (acpi_disabled) - return 0; - if (acpi_disabled) return 0; From 725c3a2d70f958adee807c178178819a50f68a56 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Fri, 18 Jul 2008 09:12:49 +0200 Subject: [PATCH 007/857] Revert "ACPI: don't walk tables if ACPI was disabled" This reverts commit d1857056904d5f313f11184fcfa624652ff9620a. Double commit, noticed by Thomas Gleixner. Signed-off-by: Andi Kleen --- drivers/acpi/glue.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/acpi/glue.c b/drivers/acpi/glue.c index 0f2dd81736bd..2f173e83f8a7 100644 --- a/drivers/acpi/glue.c +++ b/drivers/acpi/glue.c @@ -335,9 +335,6 @@ static int __init acpi_rtc_init(void) { struct device *dev = get_rtc_dev(); - if (acpi_disabled) - return 0; - if (acpi_disabled) return 0; From 8b95d9172be7146c87e7a998310ce2919c851adc Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Mon, 14 Jul 2008 23:32:32 +0200 Subject: [PATCH 008/857] fix core/stacktrace changes on avr32, mips, sh Fixes this type of problem: CC arch/s390/kernel/stacktrace.o arch/s390/kernel/stacktrace.c:84: warning: data definition has no type or storage class arch/s390/kernel/stacktrace.c:84: warning: type defaults to 'int' in declaration of 'EXPORT_SYMBOL_GPL' arch/s390/kernel/stacktrace.c:84: warning: parameter names (without types) in function declaration arch/s390/kernel/stacktrace.c:97: warning: data definition has no type or storage class arch/s390/kernel/stacktrace.c:97: warning: type defaults to 'int' in declaration of 'EXPORT_SYMBOL_GPL' arch/s390/kernel/stacktrace.c:97: warning: parameter names (without types) in function declaration caused by "stacktrace: export save_stack_trace[_tsk]" Signed-off-by: Heiko Carstens Cc: Stephen Rothwell Cc: Linus Torvalds Cc: Andrew Morton Signed-off-by: Ingo Molnar --- arch/avr32/kernel/stacktrace.c | 1 + arch/mips/kernel/stacktrace.c | 1 + arch/sh/kernel/stacktrace.c | 1 + 3 files changed, 3 insertions(+) diff --git a/arch/avr32/kernel/stacktrace.c b/arch/avr32/kernel/stacktrace.c index f4bdb448049c..c09f0d8dd679 100644 --- a/arch/avr32/kernel/stacktrace.c +++ b/arch/avr32/kernel/stacktrace.c @@ -10,6 +10,7 @@ #include #include #include +#include register unsigned long current_frame_pointer asm("r7"); diff --git a/arch/mips/kernel/stacktrace.c b/arch/mips/kernel/stacktrace.c index 5eb4681a73d2..702e2e92a1ca 100644 --- a/arch/mips/kernel/stacktrace.c +++ b/arch/mips/kernel/stacktrace.c @@ -7,6 +7,7 @@ */ #include #include +#include /* diff --git a/arch/sh/kernel/stacktrace.c b/arch/sh/kernel/stacktrace.c index 1b2ae35c4a76..54d1f61aa007 100644 --- a/arch/sh/kernel/stacktrace.c +++ b/arch/sh/kernel/stacktrace.c @@ -12,6 +12,7 @@ #include #include #include +#include #include /* From b8f8c3cf0a4ac0632ec3f0e15e9dc0c29de917af Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 18 Jul 2008 17:27:28 +0200 Subject: [PATCH 009/857] nohz: prevent tick stop outside of the idle loop Jack Ren and Eric Miao tracked down the following long standing problem in the NOHZ code: scheduler switch to idle task enable interrupts Window starts here ----> interrupt happens (does not set NEED_RESCHED) irq_exit() stops the tick ----> interrupt happens (does set NEED_RESCHED) return from schedule() cpu_idle(): preempt_disable(); Window ends here The interrupts can happen at any point inside the race window. The first interrupt stops the tick, the second one causes the scheduler to rerun and switch away from idle again and we end up with the tick disabled. The fact that it needs two interrupts where the first one does not set NEED_RESCHED and the second one does made the bug obscure and extremly hard to reproduce and analyse. Kudos to Jack and Eric. Solution: Limit the NOHZ functionality to the idle loop to make sure that we can not run into such a situation ever again. cpu_idle() { preempt_disable(); while(1) { tick_nohz_stop_sched_tick(1); <- tell NOHZ code that we are in the idle loop while (!need_resched()) halt(); tick_nohz_restart_sched_tick(); <- disables NOHZ mode preempt_enable_no_resched(); schedule(); preempt_disable(); } } In hindsight we should have done this forever, but ... /me grabs a large brown paperbag. Debugged-by: Jack Ren , Debugged-by: eric miao Signed-off-by: Thomas Gleixner --- arch/arm/kernel/process.c | 2 +- arch/avr32/kernel/process.c | 2 +- arch/blackfin/kernel/process.c | 2 +- arch/mips/kernel/process.c | 2 +- arch/powerpc/kernel/idle.c | 2 +- arch/powerpc/platforms/iseries/setup.c | 4 ++-- arch/sh/kernel/process_32.c | 2 +- arch/sparc64/kernel/process.c | 2 +- arch/um/kernel/process.c | 2 +- arch/x86/kernel/process_32.c | 2 +- arch/x86/kernel/process_64.c | 2 +- include/linux/tick.h | 5 +++-- kernel/softirq.c | 2 +- kernel/time/tick-sched.c | 12 ++++++++++-- 14 files changed, 26 insertions(+), 17 deletions(-) diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c index 46bf2ede6128..84f5a4c778fb 100644 --- a/arch/arm/kernel/process.c +++ b/arch/arm/kernel/process.c @@ -164,7 +164,7 @@ void cpu_idle(void) if (!idle) idle = default_idle; leds_event(led_idle_start); - tick_nohz_stop_sched_tick(); + tick_nohz_stop_sched_tick(1); while (!need_resched()) idle(); leds_event(led_idle_end); diff --git a/arch/avr32/kernel/process.c b/arch/avr32/kernel/process.c index 6cf9df176274..ff820a9e743a 100644 --- a/arch/avr32/kernel/process.c +++ b/arch/avr32/kernel/process.c @@ -31,7 +31,7 @@ void cpu_idle(void) { /* endless idle loop with no priority at all */ while (1) { - tick_nohz_stop_sched_tick(); + tick_nohz_stop_sched_tick(1); while (!need_resched()) cpu_idle_sleep(); tick_nohz_restart_sched_tick(); diff --git a/arch/blackfin/kernel/process.c b/arch/blackfin/kernel/process.c index 53c2cd255441..77800dd83e57 100644 --- a/arch/blackfin/kernel/process.c +++ b/arch/blackfin/kernel/process.c @@ -105,7 +105,7 @@ void cpu_idle(void) #endif if (!idle) idle = default_idle; - tick_nohz_stop_sched_tick(); + tick_nohz_stop_sched_tick(1); while (!need_resched()) idle(); tick_nohz_restart_sched_tick(); diff --git a/arch/mips/kernel/process.c b/arch/mips/kernel/process.c index 2c09a442e5e5..bdead3aad253 100644 --- a/arch/mips/kernel/process.c +++ b/arch/mips/kernel/process.c @@ -53,7 +53,7 @@ void __noreturn cpu_idle(void) { /* endless idle loop with no priority at all */ while (1) { - tick_nohz_stop_sched_tick(); + tick_nohz_stop_sched_tick(1); while (!need_resched()) { #ifdef CONFIG_SMTC_IDLE_HOOK_DEBUG extern void smtc_idle_loop_hook(void); diff --git a/arch/powerpc/kernel/idle.c b/arch/powerpc/kernel/idle.c index c3cf0e8f3ac1..d308a9f70f1b 100644 --- a/arch/powerpc/kernel/idle.c +++ b/arch/powerpc/kernel/idle.c @@ -60,7 +60,7 @@ void cpu_idle(void) set_thread_flag(TIF_POLLING_NRFLAG); while (1) { - tick_nohz_stop_sched_tick(); + tick_nohz_stop_sched_tick(1); while (!need_resched() && !cpu_should_die()) { ppc64_runlatch_off(); diff --git a/arch/powerpc/platforms/iseries/setup.c b/arch/powerpc/platforms/iseries/setup.c index b72120751bbe..70b688c1aefb 100644 --- a/arch/powerpc/platforms/iseries/setup.c +++ b/arch/powerpc/platforms/iseries/setup.c @@ -561,7 +561,7 @@ static void yield_shared_processor(void) static void iseries_shared_idle(void) { while (1) { - tick_nohz_stop_sched_tick(); + tick_nohz_stop_sched_tick(1); while (!need_resched() && !hvlpevent_is_pending()) { local_irq_disable(); ppc64_runlatch_off(); @@ -591,7 +591,7 @@ static void iseries_dedicated_idle(void) set_thread_flag(TIF_POLLING_NRFLAG); while (1) { - tick_nohz_stop_sched_tick(); + tick_nohz_stop_sched_tick(1); if (!need_resched()) { while (!need_resched()) { ppc64_runlatch_off(); diff --git a/arch/sh/kernel/process_32.c b/arch/sh/kernel/process_32.c index b98e37a1f54c..921892c351da 100644 --- a/arch/sh/kernel/process_32.c +++ b/arch/sh/kernel/process_32.c @@ -86,7 +86,7 @@ void cpu_idle(void) if (!idle) idle = default_idle; - tick_nohz_stop_sched_tick(); + tick_nohz_stop_sched_tick(1); while (!need_resched()) idle(); tick_nohz_restart_sched_tick(); diff --git a/arch/sparc64/kernel/process.c b/arch/sparc64/kernel/process.c index 2084f81a76e1..0798928ba361 100644 --- a/arch/sparc64/kernel/process.c +++ b/arch/sparc64/kernel/process.c @@ -97,7 +97,7 @@ void cpu_idle(void) set_thread_flag(TIF_POLLING_NRFLAG); while(1) { - tick_nohz_stop_sched_tick(); + tick_nohz_stop_sched_tick(1); while (!need_resched() && !cpu_is_offline(cpu)) sparc64_yield(cpu); diff --git a/arch/um/kernel/process.c b/arch/um/kernel/process.c index 83603cfbde81..a1c6d07cac3e 100644 --- a/arch/um/kernel/process.c +++ b/arch/um/kernel/process.c @@ -243,7 +243,7 @@ void default_idle(void) if (need_resched()) schedule(); - tick_nohz_stop_sched_tick(); + tick_nohz_stop_sched_tick(1); nsecs = disable_timer(); idle_sleep(nsecs); tick_nohz_restart_sched_tick(); diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index f8476dfbb60d..1f5fa1cf16dd 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -166,7 +166,7 @@ void cpu_idle(void) /* endless idle loop with no priority at all */ while (1) { - tick_nohz_stop_sched_tick(); + tick_nohz_stop_sched_tick(1); while (!need_resched()) { void (*idle)(void); diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index e2319f39988b..c0a5c2a687e6 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -148,7 +148,7 @@ void cpu_idle(void) current_thread_info()->status |= TS_POLLING; /* endless idle loop with no priority at all */ while (1) { - tick_nohz_stop_sched_tick(); + tick_nohz_stop_sched_tick(1); while (!need_resched()) { void (*idle)(void); diff --git a/include/linux/tick.h b/include/linux/tick.h index a881c652f7e9..d3c02695dc5d 100644 --- a/include/linux/tick.h +++ b/include/linux/tick.h @@ -49,6 +49,7 @@ struct tick_sched { unsigned long check_clocks; enum tick_nohz_mode nohz_mode; ktime_t idle_tick; + int inidle; int tick_stopped; unsigned long idle_jiffies; unsigned long idle_calls; @@ -105,14 +106,14 @@ static inline int tick_check_oneshot_change(int allow_nohz) { return 0; } #endif /* !CONFIG_GENERIC_CLOCKEVENTS */ # ifdef CONFIG_NO_HZ -extern void tick_nohz_stop_sched_tick(void); +extern void tick_nohz_stop_sched_tick(int inidle); extern void tick_nohz_restart_sched_tick(void); extern void tick_nohz_update_jiffies(void); extern ktime_t tick_nohz_get_sleep_length(void); extern void tick_nohz_stop_idle(int cpu); extern u64 get_cpu_idle_time_us(int cpu, u64 *last_update_time); # else -static inline void tick_nohz_stop_sched_tick(void) { } +static inline void tick_nohz_stop_sched_tick(int inidle) { } static inline void tick_nohz_restart_sched_tick(void) { } static inline void tick_nohz_update_jiffies(void) { } static inline ktime_t tick_nohz_get_sleep_length(void) diff --git a/kernel/softirq.c b/kernel/softirq.c index 36e061740047..05f248039d77 100644 --- a/kernel/softirq.c +++ b/kernel/softirq.c @@ -312,7 +312,7 @@ void irq_exit(void) #ifdef CONFIG_NO_HZ /* Make sure that timer wheel updates are propagated */ if (!in_interrupt() && idle_cpu(smp_processor_id()) && !need_resched()) - tick_nohz_stop_sched_tick(); + tick_nohz_stop_sched_tick(0); rcu_irq_exit(); #endif preempt_enable_no_resched(); diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index 86baa4f0dfe4..ee962d11107b 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -195,7 +195,7 @@ u64 get_cpu_idle_time_us(int cpu, u64 *last_update_time) * Called either from the idle loop or from irq_exit() when an idle period was * just interrupted by an interrupt which did not cause a reschedule. */ -void tick_nohz_stop_sched_tick(void) +void tick_nohz_stop_sched_tick(int inidle) { unsigned long seq, last_jiffies, next_jiffies, delta_jiffies, flags; struct tick_sched *ts; @@ -224,6 +224,11 @@ void tick_nohz_stop_sched_tick(void) if (unlikely(ts->nohz_mode == NOHZ_MODE_INACTIVE)) goto end; + if (!inidle && !ts->inidle) + goto end; + + ts->inidle = 1; + if (need_resched()) goto end; @@ -372,11 +377,14 @@ void tick_nohz_restart_sched_tick(void) local_irq_disable(); tick_nohz_stop_idle(cpu); - if (!ts->tick_stopped) { + if (!ts->inidle || !ts->tick_stopped) { + ts->inidle = 0; local_irq_enable(); return; } + ts->inidle = 0; + rcu_exit_nohz(); /* Update jiffies first */ From 8df185a95c9b84fc0c3c02224e64fdc5b83bae34 Mon Sep 17 00:00:00 2001 From: Mike Travis Date: Tue, 8 Jul 2008 15:55:48 -0700 Subject: [PATCH 010/857] kthread: reduce stack pressure in create_kthread and kthreadd * Replace: set_cpus_allowed(..., CPU_MASK_ALL) with: set_cpus_allowed_ptr(..., CPU_MASK_ALL_PTR) to remove excessive stack requirements when NR_CPUS=4096. Signed-off-by: Mike Travis Cc: Andrew Morton Signed-off-by: Ingo Molnar --- kernel/kthread.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/kernel/kthread.c b/kernel/kthread.c index ac3fb7326641..6111c27491b1 100644 --- a/kernel/kthread.c +++ b/kernel/kthread.c @@ -106,7 +106,7 @@ static void create_kthread(struct kthread_create_info *create) */ sched_setscheduler(create->result, SCHED_NORMAL, ¶m); set_user_nice(create->result, KTHREAD_NICE_LEVEL); - set_cpus_allowed(create->result, CPU_MASK_ALL); + set_cpus_allowed_ptr(create->result, CPU_MASK_ALL_PTR); } complete(&create->done); } @@ -233,7 +233,7 @@ int kthreadd(void *unused) set_task_comm(tsk, "kthreadd"); ignore_signals(tsk); set_user_nice(tsk, KTHREAD_NICE_LEVEL); - set_cpus_allowed(tsk, CPU_MASK_ALL); + set_cpus_allowed_ptr(tsk, CPU_MASK_ALL_PTR); current->flags |= PF_NOFREEZE | PF_FREEZER_NOSIG; From e338125b8a886923ba8367207c144764dc352584 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sat, 19 Jul 2008 09:33:21 +0200 Subject: [PATCH 011/857] nohz: adjust tick_nohz_stop_sched_tick() call of s390 as well Signed-off-by: Thomas Gleixner --- arch/s390/kernel/process.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c index 85defd01d293..9839767d0842 100644 --- a/arch/s390/kernel/process.c +++ b/arch/s390/kernel/process.c @@ -142,7 +142,7 @@ static void default_idle(void) void cpu_idle(void) { for (;;) { - tick_nohz_stop_sched_tick(); + tick_nohz_stop_sched_tick(1); while (!need_resched()) default_idle(); tick_nohz_restart_sched_tick(); From 3a87208028ef59215a88a143c723ac0b83c11df0 Mon Sep 17 00:00:00 2001 From: Henrique de Moraes Holschuh Date: Mon, 21 Jul 2008 09:15:49 -0300 Subject: [PATCH 012/857] ACPI: thinkpad-acpi: minor refactor on radio switch init Change the code of hotkey_init, wan_init and bluetooth_init a bit to make it much easier to add some Kconfig-selected debugging code later. Signed-off-by: Henrique de Moraes Holschuh --- drivers/misc/thinkpad_acpi.c | 49 +++++++++++++++++++----------------- 1 file changed, 26 insertions(+), 23 deletions(-) diff --git a/drivers/misc/thinkpad_acpi.c b/drivers/misc/thinkpad_acpi.c index b5969298f3d3..c800855be27f 100644 --- a/drivers/misc/thinkpad_acpi.c +++ b/drivers/misc/thinkpad_acpi.c @@ -2167,9 +2167,10 @@ static int __init hotkey_init(struct ibm_init_struct *iibm) printk(TPACPI_INFO "radio switch found; radios are %s\n", enabled(status, 0)); + } + if (tp_features.hotkey_wlsw) res = add_to_attr_set(hotkey_dev_attributes, &dev_attr_hotkey_radio_sw.attr); - } /* For X41t, X60t, X61t Tablets... */ if (!res && acpi_evalf(hkey_handle, &status, "MHKG", "qd")) { @@ -2646,18 +2647,19 @@ static int __init bluetooth_init(struct ibm_init_struct *iibm) str_supported(tp_features.bluetooth), status); + if (tp_features.bluetooth && + !(status & TP_ACPI_BLUETOOTH_HWPRESENT)) { + /* no bluetooth hardware present in system */ + tp_features.bluetooth = 0; + dbg_printk(TPACPI_DBG_INIT, + "bluetooth hardware not installed\n"); + } + if (tp_features.bluetooth) { - if (!(status & TP_ACPI_BLUETOOTH_HWPRESENT)) { - /* no bluetooth hardware present in system */ - tp_features.bluetooth = 0; - dbg_printk(TPACPI_DBG_INIT, - "bluetooth hardware not installed\n"); - } else { - res = sysfs_create_group(&tpacpi_pdev->dev.kobj, - &bluetooth_attr_group); - if (res) - return res; - } + res = sysfs_create_group(&tpacpi_pdev->dev.kobj, + &bluetooth_attr_group); + if (res) + return res; } return (tp_features.bluetooth)? 0 : 1; @@ -2818,18 +2820,19 @@ static int __init wan_init(struct ibm_init_struct *iibm) str_supported(tp_features.wan), status); + if (tp_features.wan && + !(status & TP_ACPI_WANCARD_HWPRESENT)) { + /* no wan hardware present in system */ + tp_features.wan = 0; + dbg_printk(TPACPI_DBG_INIT, + "wan hardware not installed\n"); + } + if (tp_features.wan) { - if (!(status & TP_ACPI_WANCARD_HWPRESENT)) { - /* no wan hardware present in system */ - tp_features.wan = 0; - dbg_printk(TPACPI_DBG_INIT, - "wan hardware not installed\n"); - } else { - res = sysfs_create_group(&tpacpi_pdev->dev.kobj, - &wan_attr_group); - if (res) - return res; - } + res = sysfs_create_group(&tpacpi_pdev->dev.kobj, + &wan_attr_group); + if (res) + return res; } return (tp_features.wan)? 0 : 1; From 733e27c1cc86afae2d9481838693661b3d839950 Mon Sep 17 00:00:00 2001 From: Henrique de Moraes Holschuh Date: Mon, 21 Jul 2008 09:15:49 -0300 Subject: [PATCH 013/857] ACPI: thinkpad-acpi: consolidate wlsw notification function Rename tpacpi_input_send_radiosw() to tpacpi_send_radiosw_update(), and make it a central point to issue "radio switch changed state" notifications by consolidating also the poll() notification in the same function. Signed-off-by: Henrique de Moraes Holschuh --- drivers/misc/thinkpad_acpi.c | 39 ++++++++++++++++++------------------ 1 file changed, 19 insertions(+), 20 deletions(-) diff --git a/drivers/misc/thinkpad_acpi.c b/drivers/misc/thinkpad_acpi.c index c800855be27f..9179f2367d42 100644 --- a/drivers/misc/thinkpad_acpi.c +++ b/drivers/misc/thinkpad_acpi.c @@ -1285,21 +1285,6 @@ static int hotkey_status_set(int status) return 0; } -static void tpacpi_input_send_radiosw(void) -{ - int wlsw; - - if (tp_features.hotkey_wlsw && !hotkey_get_wlsw(&wlsw)) { - mutex_lock(&tpacpi_inputdev_send_mutex); - - input_report_switch(tpacpi_inputdev, - SW_RFKILL_ALL, !!wlsw); - input_sync(tpacpi_inputdev); - - mutex_unlock(&tpacpi_inputdev_send_mutex); - } -} - static void tpacpi_input_send_tabletsw(void) { int state; @@ -1921,6 +1906,22 @@ static struct attribute *hotkey_mask_attributes[] __initdata = { &dev_attr_hotkey_wakeup_hotunplug_complete.attr, }; +static void tpacpi_send_radiosw_update(void) +{ + int wlsw; + + if (tp_features.hotkey_wlsw && !hotkey_get_wlsw(&wlsw)) { + mutex_lock(&tpacpi_inputdev_send_mutex); + + input_report_switch(tpacpi_inputdev, + SW_RFKILL_ALL, !!wlsw); + input_sync(tpacpi_inputdev); + + mutex_unlock(&tpacpi_inputdev_send_mutex); + } + hotkey_radio_sw_notify_change(); +} + static void hotkey_exit(void) { #ifdef CONFIG_THINKPAD_ACPI_HOTKEY_POLL @@ -2288,7 +2289,7 @@ static int __init hotkey_init(struct ibm_init_struct *iibm) tpacpi_inputdev->close = &hotkey_inputdev_close; hotkey_poll_setup_safe(1); - tpacpi_input_send_radiosw(); + tpacpi_send_radiosw_update(); tpacpi_input_send_tabletsw(); return 0; @@ -2420,8 +2421,7 @@ static void hotkey_notify(struct ibm_struct *ibm, u32 event) case 7: /* 0x7000-0x7FFF: misc */ if (tp_features.hotkey_wlsw && hkey == 0x7000) { - tpacpi_input_send_radiosw(); - hotkey_radio_sw_notify_change(); + tpacpi_send_radiosw_update(); send_acpi_ev = 0; break; } @@ -2464,8 +2464,7 @@ static void hotkey_resume(void) printk(TPACPI_ERR "error while trying to read hot key mask " "from firmware\n"); - tpacpi_input_send_radiosw(); - hotkey_radio_sw_notify_change(); + tpacpi_send_radiosw_update(); hotkey_tablet_mode_notify_change(); hotkey_wakeup_reason_notify_change(); hotkey_wakeup_hotunplug_complete_notify_change(); From 07431ec82bf9dc74b470a1d820b41c92c4d86e6f Mon Sep 17 00:00:00 2001 From: Henrique de Moraes Holschuh Date: Mon, 21 Jul 2008 09:15:50 -0300 Subject: [PATCH 014/857] ACPI: thinkpad-acpi: prepare for bluetooth and wwan rfkill support Get rid of some forward definitions by moving code around, this will make the rfkill conversion of wwan and bluetooth a bit cleaner. Signed-off-by: Henrique de Moraes Holschuh --- drivers/misc/thinkpad_acpi.c | 154 +++++++++++++++++------------------ 1 file changed, 74 insertions(+), 80 deletions(-) diff --git a/drivers/misc/thinkpad_acpi.c b/drivers/misc/thinkpad_acpi.c index 9179f2367d42..743a4d6098e8 100644 --- a/drivers/misc/thinkpad_acpi.c +++ b/drivers/misc/thinkpad_acpi.c @@ -2581,8 +2581,37 @@ enum { TP_ACPI_BLUETOOTH_UNK = 0x04, /* unknown function */ }; -static int bluetooth_get_radiosw(void); -static int bluetooth_set_radiosw(int radio_on); +static int bluetooth_get_radiosw(void) +{ + int status; + + if (!tp_features.bluetooth) + return -ENODEV; + + if (!acpi_evalf(hkey_handle, &status, "GBDC", "d")) + return -EIO; + + return (status & TP_ACPI_BLUETOOTH_RADIOSSW) != 0; +} + +static int bluetooth_set_radiosw(int radio_on) +{ + int status; + + if (!tp_features.bluetooth) + return -ENODEV; + + if (!acpi_evalf(hkey_handle, &status, "GBDC", "d")) + return -EIO; + if (radio_on) + status |= TP_ACPI_BLUETOOTH_RADIOSSW; + else + status &= ~TP_ACPI_BLUETOOTH_RADIOSSW; + if (!acpi_evalf(hkey_handle, NULL, "SBDC", "vd", status)) + return -EIO; + + return 0; +} /* sysfs bluetooth enable ---------------------------------------------- */ static ssize_t bluetooth_enable_show(struct device *dev, @@ -2628,6 +2657,12 @@ static const struct attribute_group bluetooth_attr_group = { .attrs = bluetooth_attributes, }; +static void bluetooth_exit(void) +{ + sysfs_remove_group(&tpacpi_pdev->dev.kobj, + &bluetooth_attr_group); +} + static int __init bluetooth_init(struct ibm_init_struct *iibm) { int res; @@ -2664,44 +2699,6 @@ static int __init bluetooth_init(struct ibm_init_struct *iibm) return (tp_features.bluetooth)? 0 : 1; } -static void bluetooth_exit(void) -{ - sysfs_remove_group(&tpacpi_pdev->dev.kobj, - &bluetooth_attr_group); -} - -static int bluetooth_get_radiosw(void) -{ - int status; - - if (!tp_features.bluetooth) - return -ENODEV; - - if (!acpi_evalf(hkey_handle, &status, "GBDC", "d")) - return -EIO; - - return ((status & TP_ACPI_BLUETOOTH_RADIOSSW) != 0); -} - -static int bluetooth_set_radiosw(int radio_on) -{ - int status; - - if (!tp_features.bluetooth) - return -ENODEV; - - if (!acpi_evalf(hkey_handle, &status, "GBDC", "d")) - return -EIO; - if (radio_on) - status |= TP_ACPI_BLUETOOTH_RADIOSSW; - else - status &= ~TP_ACPI_BLUETOOTH_RADIOSSW; - if (!acpi_evalf(hkey_handle, NULL, "SBDC", "vd", status)) - return -EIO; - - return 0; -} - /* procfs -------------------------------------------------------------- */ static int bluetooth_read(char *p) { @@ -2756,8 +2753,37 @@ enum { TP_ACPI_WANCARD_UNK = 0x04, /* unknown function */ }; -static int wan_get_radiosw(void); -static int wan_set_radiosw(int radio_on); +static int wan_get_radiosw(void) +{ + int status; + + if (!tp_features.wan) + return -ENODEV; + + if (!acpi_evalf(hkey_handle, &status, "GWAN", "d")) + return -EIO; + + return (status & TP_ACPI_WANCARD_RADIOSSW) != 0; +} + +static int wan_set_radiosw(int radio_on) +{ + int status; + + if (!tp_features.wan) + return -ENODEV; + + if (!acpi_evalf(hkey_handle, &status, "GWAN", "d")) + return -EIO; + if (radio_on) + status |= TP_ACPI_WANCARD_RADIOSSW; + else + status &= ~TP_ACPI_WANCARD_RADIOSSW; + if (!acpi_evalf(hkey_handle, NULL, "SWAN", "vd", status)) + return -EIO; + + return 0; +} /* sysfs wan enable ---------------------------------------------------- */ static ssize_t wan_enable_show(struct device *dev, @@ -2803,6 +2829,12 @@ static const struct attribute_group wan_attr_group = { .attrs = wan_attributes, }; +static void wan_exit(void) +{ + sysfs_remove_group(&tpacpi_pdev->dev.kobj, + &wan_attr_group); +} + static int __init wan_init(struct ibm_init_struct *iibm) { int res; @@ -2837,44 +2869,6 @@ static int __init wan_init(struct ibm_init_struct *iibm) return (tp_features.wan)? 0 : 1; } -static void wan_exit(void) -{ - sysfs_remove_group(&tpacpi_pdev->dev.kobj, - &wan_attr_group); -} - -static int wan_get_radiosw(void) -{ - int status; - - if (!tp_features.wan) - return -ENODEV; - - if (!acpi_evalf(hkey_handle, &status, "GWAN", "d")) - return -EIO; - - return ((status & TP_ACPI_WANCARD_RADIOSSW) != 0); -} - -static int wan_set_radiosw(int radio_on) -{ - int status; - - if (!tp_features.wan) - return -ENODEV; - - if (!acpi_evalf(hkey_handle, &status, "GWAN", "d")) - return -EIO; - if (radio_on) - status |= TP_ACPI_WANCARD_RADIOSSW; - else - status &= ~TP_ACPI_WANCARD_RADIOSSW; - if (!acpi_evalf(hkey_handle, NULL, "SWAN", "vd", status)) - return -EIO; - - return 0; -} - /* procfs -------------------------------------------------------------- */ static int wan_read(char *p) { From 133ec3bd3ae409895eacdce326cdc8d73c249e8a Mon Sep 17 00:00:00 2001 From: Henrique de Moraes Holschuh Date: Mon, 21 Jul 2008 09:15:50 -0300 Subject: [PATCH 015/857] ACPI: thinkpad-acpi: WLSW overrides other rfkill switches On ThinkPads where the WLSW switch exists, the firmware or the hardware ANDs the WLSW state with the device-specific switches (WWAN, Bluetooth). It is downright impossible to enable WWAN or Bluetooth when WLSW is blocking the radios. This reality does not necessarily carry over to the WWAN and Bluetooth firmware interfaces, though... so the state thinkpad-acpi was reporting could be incorrect. Tie the three switches in the driver so that we keep their state sane. When WLSL is off, force the other switches to off as well. Signed-off-by: Henrique de Moraes Holschuh --- drivers/misc/thinkpad_acpi.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) diff --git a/drivers/misc/thinkpad_acpi.c b/drivers/misc/thinkpad_acpi.c index 743a4d6098e8..202d63e1b391 100644 --- a/drivers/misc/thinkpad_acpi.c +++ b/drivers/misc/thinkpad_acpi.c @@ -2588,6 +2588,10 @@ static int bluetooth_get_radiosw(void) if (!tp_features.bluetooth) return -ENODEV; + /* WLSW overrides bluetooth in firmware/hardware, reflect that */ + if (tp_features.hotkey_wlsw && !hotkey_get_wlsw(&status) && !status) + return 0; + if (!acpi_evalf(hkey_handle, &status, "GBDC", "d")) return -EIO; @@ -2601,6 +2605,12 @@ static int bluetooth_set_radiosw(int radio_on) if (!tp_features.bluetooth) return -ENODEV; + /* WLSW overrides bluetooth in firmware/hardware, but there is no + * reason to risk weird behaviour. */ + if (tp_features.hotkey_wlsw && !hotkey_get_wlsw(&status) && !status + && radio_on) + return -EPERM; + if (!acpi_evalf(hkey_handle, &status, "GBDC", "d")) return -EIO; if (radio_on) @@ -2760,6 +2770,10 @@ static int wan_get_radiosw(void) if (!tp_features.wan) return -ENODEV; + /* WLSW overrides WWAN in firmware/hardware, reflect that */ + if (tp_features.hotkey_wlsw && !hotkey_get_wlsw(&status) && !status) + return 0; + if (!acpi_evalf(hkey_handle, &status, "GWAN", "d")) return -EIO; @@ -2773,6 +2787,12 @@ static int wan_set_radiosw(int radio_on) if (!tp_features.wan) return -ENODEV; + /* WLSW overrides bluetooth in firmware/hardware, but there is no + * reason to risk weird behaviour. */ + if (tp_features.hotkey_wlsw && !hotkey_get_wlsw(&status) && !status + && radio_on) + return -EPERM; + if (!acpi_evalf(hkey_handle, &status, "GWAN", "d")) return -EIO; if (radio_on) From 0e74dc2646db04b644faa8ea10ff4f408d55cf90 Mon Sep 17 00:00:00 2001 From: Henrique de Moraes Holschuh Date: Mon, 21 Jul 2008 09:15:51 -0300 Subject: [PATCH 016/857] ACPI: thinkpad-acpi: add bluetooth and WWAN rfkill support Add a read/write rfkill interface to the bluetooth radio switch on the bluetooth submodule, and one for the wireless wan radio switch to the wan submodule. Since rfkill does care for when a switch changes state, use WLSW notifications to also check if the WWAN or Bluetooth switches did not change state (due to them being slaves of WLSW in firmware/hardware, but that reality not being always properly exported by the thinkpad firmware). Signed-off-by: Henrique de Moraes Holschuh Cc: Ivo van Doorn Cc: John W. Linville --- Documentation/laptops/thinkpad-acpi.txt | 22 ++- drivers/misc/Kconfig | 2 + drivers/misc/thinkpad_acpi.c | 208 +++++++++++++++++++++--- 3 files changed, 200 insertions(+), 32 deletions(-) diff --git a/Documentation/laptops/thinkpad-acpi.txt b/Documentation/laptops/thinkpad-acpi.txt index 64b3f146e4b0..1c1c0217ebd1 100644 --- a/Documentation/laptops/thinkpad-acpi.txt +++ b/Documentation/laptops/thinkpad-acpi.txt @@ -621,7 +621,8 @@ Bluetooth --------- procfs: /proc/acpi/ibm/bluetooth -sysfs device attribute: bluetooth_enable +sysfs device attribute: bluetooth_enable (deprecated) +sysfs rfkill class: switch "tpacpi_bluetooth_sw" This feature shows the presence and current state of a ThinkPad Bluetooth device in the internal ThinkPad CDC slot. @@ -643,8 +644,12 @@ Sysfs notes: 0: disables Bluetooth / Bluetooth is disabled 1: enables Bluetooth / Bluetooth is enabled. - Note: this interface will be probably be superseded by the - generic rfkill class, so it is NOT to be considered stable yet. + Note: this interface has been superseded by the generic rfkill + class. It has been deprecated, and it will be removed in year + 2010. + + rfkill controller switch "tpacpi_bluetooth_sw": refer to + Documentation/rfkill.txt for details. Video output control -- /proc/acpi/ibm/video -------------------------------------------- @@ -1374,7 +1379,8 @@ EXPERIMENTAL: WAN ----------------- procfs: /proc/acpi/ibm/wan -sysfs device attribute: wwan_enable +sysfs device attribute: wwan_enable (deprecated) +sysfs rfkill class: switch "tpacpi_wwan_sw" This feature is marked EXPERIMENTAL because the implementation directly accesses hardware registers and may not work as expected. USE @@ -1404,8 +1410,12 @@ Sysfs notes: 0: disables WWAN card / WWAN card is disabled 1: enables WWAN card / WWAN card is enabled. - Note: this interface will be probably be superseded by the - generic rfkill class, so it is NOT to be considered stable yet. + Note: this interface has been superseded by the generic rfkill + class. It has been deprecated, and it will be removed in year + 2010. + + rfkill controller switch "tpacpi_wwan_sw": refer to + Documentation/rfkill.txt for details. Multiple Commands, Module Parameters ------------------------------------ diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig index 1921b8dbb242..b27ca91fd15e 100644 --- a/drivers/misc/Kconfig +++ b/drivers/misc/Kconfig @@ -279,6 +279,8 @@ config THINKPAD_ACPI select INPUT select NEW_LEDS select LEDS_CLASS + select NET + select RFKILL ---help--- This is a driver for the IBM and Lenovo ThinkPad laptops. It adds support for Fn-Fx key combinations, Bluetooth control, video diff --git a/drivers/misc/thinkpad_acpi.c b/drivers/misc/thinkpad_acpi.c index 202d63e1b391..dc8d00a45701 100644 --- a/drivers/misc/thinkpad_acpi.c +++ b/drivers/misc/thinkpad_acpi.c @@ -68,6 +68,7 @@ #include #include #include +#include #include #include @@ -144,6 +145,12 @@ enum { #define TPACPI_MAX_ACPI_ARGS 3 +/* rfkill switches */ +enum { + TPACPI_RFK_BLUETOOTH_SW_ID = 0, + TPACPI_RFK_WWAN_SW_ID, +}; + /* Debugging */ #define TPACPI_LOG TPACPI_FILE ": " #define TPACPI_ERR KERN_ERR TPACPI_LOG @@ -905,6 +912,43 @@ static int __init tpacpi_check_std_acpi_brightness_support(void) return 0; } +static int __init tpacpi_new_rfkill(const unsigned int id, + struct rfkill **rfk, + const enum rfkill_type rfktype, + const char *name, + int (*toggle_radio)(void *, enum rfkill_state), + int (*get_state)(void *, enum rfkill_state *)) +{ + int res; + enum rfkill_state initial_state; + + *rfk = rfkill_allocate(&tpacpi_pdev->dev, rfktype); + if (!*rfk) { + printk(TPACPI_ERR + "failed to allocate memory for rfkill class\n"); + return -ENOMEM; + } + + (*rfk)->name = name; + (*rfk)->get_state = get_state; + (*rfk)->toggle_radio = toggle_radio; + + if (!get_state(NULL, &initial_state)) + (*rfk)->state = initial_state; + + res = rfkill_register(*rfk); + if (res < 0) { + printk(TPACPI_ERR + "failed to register %s rfkill switch: %d\n", + name, res); + rfkill_free(*rfk); + *rfk = NULL; + return res; + } + + return 0; +} + /************************************************************************* * thinkpad-acpi driver attributes */ @@ -1906,10 +1950,18 @@ static struct attribute *hotkey_mask_attributes[] __initdata = { &dev_attr_hotkey_wakeup_hotunplug_complete.attr, }; +static void bluetooth_update_rfk(void); +static void wan_update_rfk(void); static void tpacpi_send_radiosw_update(void) { int wlsw; + /* Sync these BEFORE sending any rfkill events */ + if (tp_features.bluetooth) + bluetooth_update_rfk(); + if (tp_features.wan) + wan_update_rfk(); + if (tp_features.hotkey_wlsw && !hotkey_get_wlsw(&wlsw)) { mutex_lock(&tpacpi_inputdev_send_mutex); @@ -2581,6 +2633,8 @@ enum { TP_ACPI_BLUETOOTH_UNK = 0x04, /* unknown function */ }; +static struct rfkill *tpacpi_bluetooth_rfkill; + static int bluetooth_get_radiosw(void) { int status; @@ -2590,15 +2644,29 @@ static int bluetooth_get_radiosw(void) /* WLSW overrides bluetooth in firmware/hardware, reflect that */ if (tp_features.hotkey_wlsw && !hotkey_get_wlsw(&status) && !status) - return 0; + return RFKILL_STATE_HARD_BLOCKED; if (!acpi_evalf(hkey_handle, &status, "GBDC", "d")) return -EIO; - return (status & TP_ACPI_BLUETOOTH_RADIOSSW) != 0; + return ((status & TP_ACPI_BLUETOOTH_RADIOSSW) != 0) ? + RFKILL_STATE_UNBLOCKED : RFKILL_STATE_SOFT_BLOCKED; } -static int bluetooth_set_radiosw(int radio_on) +static void bluetooth_update_rfk(void) +{ + int status; + + if (!tpacpi_bluetooth_rfkill) + return; + + status = bluetooth_get_radiosw(); + if (status < 0) + return; + rfkill_force_state(tpacpi_bluetooth_rfkill, status); +} + +static int bluetooth_set_radiosw(int radio_on, int update_rfk) { int status; @@ -2620,6 +2688,9 @@ static int bluetooth_set_radiosw(int radio_on) if (!acpi_evalf(hkey_handle, NULL, "SBDC", "vd", status)) return -EIO; + if (update_rfk) + bluetooth_update_rfk(); + return 0; } @@ -2634,7 +2705,8 @@ static ssize_t bluetooth_enable_show(struct device *dev, if (status < 0) return status; - return snprintf(buf, PAGE_SIZE, "%d\n", status ? 1 : 0); + return snprintf(buf, PAGE_SIZE, "%d\n", + (status == RFKILL_STATE_UNBLOCKED) ? 1 : 0); } static ssize_t bluetooth_enable_store(struct device *dev, @@ -2647,7 +2719,7 @@ static ssize_t bluetooth_enable_store(struct device *dev, if (parse_strtoul(buf, 1, &t)) return -EINVAL; - res = bluetooth_set_radiosw(t); + res = bluetooth_set_radiosw(t, 1); return (res) ? res : count; } @@ -2667,8 +2739,27 @@ static const struct attribute_group bluetooth_attr_group = { .attrs = bluetooth_attributes, }; +static int tpacpi_bluetooth_rfk_get(void *data, enum rfkill_state *state) +{ + int bts = bluetooth_get_radiosw(); + + if (bts < 0) + return bts; + + *state = bts; + return 0; +} + +static int tpacpi_bluetooth_rfk_set(void *data, enum rfkill_state state) +{ + return bluetooth_set_radiosw((state == RFKILL_STATE_UNBLOCKED), 0); +} + static void bluetooth_exit(void) { + if (tpacpi_bluetooth_rfkill) + rfkill_unregister(tpacpi_bluetooth_rfkill); + sysfs_remove_group(&tpacpi_pdev->dev.kobj, &bluetooth_attr_group); } @@ -2699,14 +2790,26 @@ static int __init bluetooth_init(struct ibm_init_struct *iibm) "bluetooth hardware not installed\n"); } - if (tp_features.bluetooth) { - res = sysfs_create_group(&tpacpi_pdev->dev.kobj, + if (!tp_features.bluetooth) + return 1; + + res = sysfs_create_group(&tpacpi_pdev->dev.kobj, &bluetooth_attr_group); - if (res) - return res; + if (res) + return res; + + res = tpacpi_new_rfkill(TPACPI_RFK_BLUETOOTH_SW_ID, + &tpacpi_bluetooth_rfkill, + RFKILL_TYPE_BLUETOOTH, + "tpacpi_bluetooth_sw", + tpacpi_bluetooth_rfk_set, + tpacpi_bluetooth_rfk_get); + if (res) { + bluetooth_exit(); + return res; } - return (tp_features.bluetooth)? 0 : 1; + return 0; } /* procfs -------------------------------------------------------------- */ @@ -2719,7 +2822,8 @@ static int bluetooth_read(char *p) len += sprintf(p + len, "status:\t\tnot supported\n"); else { len += sprintf(p + len, "status:\t\t%s\n", - (status)? "enabled" : "disabled"); + (status == RFKILL_STATE_UNBLOCKED) ? + "enabled" : "disabled"); len += sprintf(p + len, "commands:\tenable, disable\n"); } @@ -2735,9 +2839,9 @@ static int bluetooth_write(char *buf) while ((cmd = next_cmd(&buf))) { if (strlencmp(cmd, "enable") == 0) { - bluetooth_set_radiosw(1); + bluetooth_set_radiosw(1, 1); } else if (strlencmp(cmd, "disable") == 0) { - bluetooth_set_radiosw(0); + bluetooth_set_radiosw(0, 1); } else return -EINVAL; } @@ -2763,6 +2867,8 @@ enum { TP_ACPI_WANCARD_UNK = 0x04, /* unknown function */ }; +static struct rfkill *tpacpi_wan_rfkill; + static int wan_get_radiosw(void) { int status; @@ -2772,15 +2878,29 @@ static int wan_get_radiosw(void) /* WLSW overrides WWAN in firmware/hardware, reflect that */ if (tp_features.hotkey_wlsw && !hotkey_get_wlsw(&status) && !status) - return 0; + return RFKILL_STATE_HARD_BLOCKED; if (!acpi_evalf(hkey_handle, &status, "GWAN", "d")) return -EIO; - return (status & TP_ACPI_WANCARD_RADIOSSW) != 0; + return ((status & TP_ACPI_WANCARD_RADIOSSW) != 0) ? + RFKILL_STATE_UNBLOCKED : RFKILL_STATE_SOFT_BLOCKED; } -static int wan_set_radiosw(int radio_on) +static void wan_update_rfk(void) +{ + int status; + + if (!tpacpi_wan_rfkill) + return; + + status = wan_get_radiosw(); + if (status < 0) + return; + rfkill_force_state(tpacpi_wan_rfkill, status); +} + +static int wan_set_radiosw(int radio_on, int update_rfk) { int status; @@ -2802,6 +2922,9 @@ static int wan_set_radiosw(int radio_on) if (!acpi_evalf(hkey_handle, NULL, "SWAN", "vd", status)) return -EIO; + if (update_rfk) + wan_update_rfk(); + return 0; } @@ -2816,7 +2939,8 @@ static ssize_t wan_enable_show(struct device *dev, if (status < 0) return status; - return snprintf(buf, PAGE_SIZE, "%d\n", status ? 1 : 0); + return snprintf(buf, PAGE_SIZE, "%d\n", + (status == RFKILL_STATE_UNBLOCKED) ? 1 : 0); } static ssize_t wan_enable_store(struct device *dev, @@ -2829,7 +2953,7 @@ static ssize_t wan_enable_store(struct device *dev, if (parse_strtoul(buf, 1, &t)) return -EINVAL; - res = wan_set_radiosw(t); + res = wan_set_radiosw(t, 1); return (res) ? res : count; } @@ -2849,8 +2973,27 @@ static const struct attribute_group wan_attr_group = { .attrs = wan_attributes, }; +static int tpacpi_wan_rfk_get(void *data, enum rfkill_state *state) +{ + int wans = wan_get_radiosw(); + + if (wans < 0) + return wans; + + *state = wans; + return 0; +} + +static int tpacpi_wan_rfk_set(void *data, enum rfkill_state state) +{ + return wan_set_radiosw((state == RFKILL_STATE_UNBLOCKED), 0); +} + static void wan_exit(void) { + if (tpacpi_wan_rfkill) + rfkill_unregister(tpacpi_wan_rfkill); + sysfs_remove_group(&tpacpi_pdev->dev.kobj, &wan_attr_group); } @@ -2879,14 +3022,26 @@ static int __init wan_init(struct ibm_init_struct *iibm) "wan hardware not installed\n"); } - if (tp_features.wan) { - res = sysfs_create_group(&tpacpi_pdev->dev.kobj, + if (!tp_features.wan) + return 1; + + res = sysfs_create_group(&tpacpi_pdev->dev.kobj, &wan_attr_group); - if (res) - return res; + if (res) + return res; + + res = tpacpi_new_rfkill(TPACPI_RFK_WWAN_SW_ID, + &tpacpi_wan_rfkill, + RFKILL_TYPE_WWAN, + "tpacpi_wwan_sw", + tpacpi_wan_rfk_set, + tpacpi_wan_rfk_get); + if (res) { + wan_exit(); + return res; } - return (tp_features.wan)? 0 : 1; + return 0; } /* procfs -------------------------------------------------------------- */ @@ -2899,7 +3054,8 @@ static int wan_read(char *p) len += sprintf(p + len, "status:\t\tnot supported\n"); else { len += sprintf(p + len, "status:\t\t%s\n", - (status)? "enabled" : "disabled"); + (status == RFKILL_STATE_UNBLOCKED) ? + "enabled" : "disabled"); len += sprintf(p + len, "commands:\tenable, disable\n"); } @@ -2915,9 +3071,9 @@ static int wan_write(char *buf) while ((cmd = next_cmd(&buf))) { if (strlencmp(cmd, "enable") == 0) { - wan_set_radiosw(1); + wan_set_radiosw(1, 1); } else if (strlencmp(cmd, "disable") == 0) { - wan_set_radiosw(0); + wan_set_radiosw(0, 1); } else return -EINVAL; } From 490673dc98adfc7de1703cc88508902bd10f446b Mon Sep 17 00:00:00 2001 From: Henrique de Moraes Holschuh Date: Mon, 21 Jul 2008 09:15:51 -0300 Subject: [PATCH 017/857] ACPI: thinkpad-acpi: bump up version to 0.21 rfkill support deserves a new version checkpoint... Signed-off-by: Henrique de Moraes Holschuh --- Documentation/laptops/thinkpad-acpi.txt | 4 ++-- drivers/misc/thinkpad_acpi.c | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/Documentation/laptops/thinkpad-acpi.txt b/Documentation/laptops/thinkpad-acpi.txt index 1c1c0217ebd1..02dc748b76c4 100644 --- a/Documentation/laptops/thinkpad-acpi.txt +++ b/Documentation/laptops/thinkpad-acpi.txt @@ -1,7 +1,7 @@ ThinkPad ACPI Extras Driver - Version 0.20 - April 09th, 2008 + Version 0.21 + May 29th, 2008 Borislav Deianov Henrique de Moraes Holschuh diff --git a/drivers/misc/thinkpad_acpi.c b/drivers/misc/thinkpad_acpi.c index dc8d00a45701..3eb01afe4306 100644 --- a/drivers/misc/thinkpad_acpi.c +++ b/drivers/misc/thinkpad_acpi.c @@ -21,7 +21,7 @@ * 02110-1301, USA. */ -#define TPACPI_VERSION "0.20" +#define TPACPI_VERSION "0.21" #define TPACPI_SYSFS_VERSION 0x020200 /* From bf20e740a4bcc686de02e2fd1c1810a58872f46e Mon Sep 17 00:00:00 2001 From: Henrique de Moraes Holschuh Date: Mon, 21 Jul 2008 09:15:51 -0300 Subject: [PATCH 018/857] ACPI: thinkpad-acpi: don't misdetect in get_thinkpad_model_data() on -ENOMEM Explicitly check for memory allocation failures, and return status to indicate that we could not collect data due to errors. This lets the driver have a far more predictable failure mode on ENOMEM in that codepath: it will refuse to load. This is far better than trying to proceed with missing data which is used to detect quirks, etc. Signed-off-by: Henrique de Moraes Holschuh --- drivers/misc/thinkpad_acpi.c | 47 +++++++++++++++++++++++++----------- 1 file changed, 33 insertions(+), 14 deletions(-) diff --git a/drivers/misc/thinkpad_acpi.c b/drivers/misc/thinkpad_acpi.c index 3eb01afe4306..d3eb7903c346 100644 --- a/drivers/misc/thinkpad_acpi.c +++ b/drivers/misc/thinkpad_acpi.c @@ -6340,13 +6340,18 @@ static int __init ibm_init(struct ibm_init_struct *iibm) /* Probing */ -static void __init get_thinkpad_model_data(struct thinkpad_id_data *tp) +/* returns 0 - probe ok, or < 0 - probe error. + * Probe ok doesn't mean thinkpad found. + * On error, kfree() cleanup on tp->* is not performed, caller must do it */ +static int __must_check __init get_thinkpad_model_data( + struct thinkpad_id_data *tp) { const struct dmi_device *dev = NULL; char ec_fw_string[18]; + char const *s; if (!tp) - return; + return -EINVAL; memset(tp, 0, sizeof(*tp)); @@ -6355,12 +6360,14 @@ static void __init get_thinkpad_model_data(struct thinkpad_id_data *tp) else if (dmi_name_in_vendors("LENOVO")) tp->vendor = PCI_VENDOR_ID_LENOVO; else - return; + return 0; - tp->bios_version_str = kstrdup(dmi_get_system_info(DMI_BIOS_VERSION), - GFP_KERNEL); + s = dmi_get_system_info(DMI_BIOS_VERSION); + tp->bios_version_str = kstrdup(s, GFP_KERNEL); + if (s && !tp->bios_version_str) + return -ENOMEM; if (!tp->bios_version_str) - return; + return 0; tp->bios_model = tp->bios_version_str[0] | (tp->bios_version_str[1] << 8); @@ -6379,21 +6386,27 @@ static void __init get_thinkpad_model_data(struct thinkpad_id_data *tp) ec_fw_string[strcspn(ec_fw_string, " ]")] = 0; tp->ec_version_str = kstrdup(ec_fw_string, GFP_KERNEL); + if (!tp->ec_version_str) + return -ENOMEM; tp->ec_model = ec_fw_string[0] | (ec_fw_string[1] << 8); break; } } - tp->model_str = kstrdup(dmi_get_system_info(DMI_PRODUCT_VERSION), - GFP_KERNEL); - if (tp->model_str && strnicmp(tp->model_str, "ThinkPad", 8) != 0) { - kfree(tp->model_str); - tp->model_str = NULL; + s = dmi_get_system_info(DMI_PRODUCT_VERSION); + if (s && !strnicmp(s, "ThinkPad", 8)) { + tp->model_str = kstrdup(s, GFP_KERNEL); + if (!tp->model_str) + return -ENOMEM; } - tp->nummodel_str = kstrdup(dmi_get_system_info(DMI_PRODUCT_NAME), - GFP_KERNEL); + s = dmi_get_system_info(DMI_PRODUCT_NAME); + tp->nummodel_str = kstrdup(s, GFP_KERNEL); + if (s && !tp->nummodel_str) + return -ENOMEM; + + return 0; } static int __init probe_for_thinkpad(void) @@ -6656,7 +6669,13 @@ static int __init thinkpad_acpi_module_init(void) /* Driver-level probe */ - get_thinkpad_model_data(&thinkpad_id); + ret = get_thinkpad_model_data(&thinkpad_id); + if (ret) { + printk(TPACPI_ERR + "unable to get DMI data: %d\n", ret); + thinkpad_acpi_module_exit(); + return ret; + } ret = probe_for_thinkpad(); if (ret) { thinkpad_acpi_module_exit(); From f88133d76ea38761b7379d6233b752ed82250a4a Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Mon, 21 Jul 2008 15:57:45 +0200 Subject: [PATCH 019/857] acpi: fix crash in core ACPI code, triggered by CONFIG_ACPI_PCI_SLOT=y -tip testing found the following boot crash on 32-bit x86 (Core2Duo laptop) yesterday: [ 5.606664] scsi4 : ata_piix [ 5.606664] scsi5 : ata_piix [ 5.606664] ACPI Error (psargs-0358): [\_SB_.PCI0.LPC_.EC__.BSTA] Namespace lookup failure, AE_NOT_FOUND [ 5.606664] ACPI Error (psparse-0530): ACPI Error (nsnames-0186): Invalid NS Node (f7c0e960) while traversing path [20080609] [ 5.606664] BUG: unable to handle kernel NULL pointer dereference at 0000000f [ 5.606664] IP: [<80339e2f>] acpi_ns_build_external_path+0x1f/0x80 [ 5.609997] *pdpt = 0000000000a03001 *pde = 0000000000000000 [ 5.609997] Oops: 0002 [#1] SMP [ 5.609997] [ 5.609997] Pid: 1, comm: swapper Not tainted (2.6.26-tip-03965-gbbfb62e-dirty #3153) [ 5.609997] EIP: 0060:[<80339e2f>] EFLAGS: 00010286 CPU: 0 [ 5.609997] EIP is at acpi_ns_build_external_path+0x1f/0x80 [ 5.609997] EAX: f7c18c18 EBX: ffffffff ECX: 00000010 EDX: 00000000 [ 5.609997] ESI: f7c18c18 EDI: 00000010 EBP: f7c4dc28 ESP: f7c4dc18 [ 5.609997] DS: 007b ES: 007b FS: 00d8 GS: 0000 SS: 0068 [ 5.609997] Process swapper (pid: 1, ti=f7c4c000 task=f7c50000 task.ti=f7c4c000) [ 5.609997] Stack: 00000000 00000000 f7c18c18 f7c4dc48 f7c4dc40 80339ed0 00000000 f7c18c18 [ 5.609997] 8084c1b6 8084c1b6 f7c4dc58 8033a60a 00000000 00000010 00000000 f7c18c18 [ 5.609997] f7c4dc70 8033a68f f7c18c18 00000000 f6de7600 00000005 f7c4dc98 8033c34d [ 5.609997] Call Trace: [ 5.609997] [<80339ed0>] ? acpi_ns_handle_to_pathname+0x40/0x72 [ 5.609997] [<8033a60a>] ? acpi_ns_print_node_pathname+0x2c/0x61 [ 5.609997] [<8033a68f>] ? acpi_ns_report_method_error+0x50/0x6d [ 5.609997] [<8033c34d>] ? acpi_ps_parse_aml+0x149/0x2f9 [ 5.609997] [<8033d6dd>] ? acpi_ps_execute_method+0x132/0x201 [ 5.609997] [<80339d19>] ? acpi_ns_evaluate+0x1ad/0x258 [ 5.609997] [<803406c4>] ? acpi_ut_evaluate_object+0x55/0x18f [ 5.609997] [<803408b7>] ? acpi_ut_execute_STA+0x22/0x7a [ 5.609997] [<8033a907>] ? acpi_get_object_info+0x131/0x1be [ 5.609997] [<80344bb2>] ? do_acpi_find_child+0x22/0x4b [ 5.609997] [<8033b855>] ? acpi_ns_walk_namespace+0xa5/0x124 [ 5.609997] [<803394f3>] ? acpi_walk_namespace+0x54/0x74 [ 5.609997] [<80344b90>] ? do_acpi_find_child+0x0/0x4b [ 5.609997] [<80344b85>] ? acpi_get_child+0x38/0x43 [ 5.609997] [<80344b90>] ? do_acpi_find_child+0x0/0x4b [ 5.609997] [<804d0148>] ? ata_acpi_associate+0xb5/0x1b5 [ 5.609997] [<804c6ecb>] ? ata_scsi_add_hosts+0x8e/0xdc [ 5.609997] [<804c40c8>] ? ata_host_register+0x9f/0x1d6 [ 5.609997] [<804cbc7f>] ? ata_pci_sff_activate_host+0x179/0x19f [ 5.609997] [<804cdd45>] ? ata_sff_interrupt+0x0/0x1c7 [ 5.609997] [<8069b033>] ? piix_init_one+0x569/0x5b0 [ 5.609997] [<801bd400>] ? sysfs_ilookup_test+0x0/0x11 [ 5.609997] [<801987d7>] ? ilookup5_nowait+0x29/0x30 [ 5.609997] [<802efc7e>] ? pci_match_device+0x99/0xa3 [ 5.609997] [<802efd3c>] ? pci_device_probe+0x39/0x59 [ 5.609997] [<803bc4af>] ? driver_probe_device+0xa0/0x11b [ 5.609997] [<803bc564>] ? __driver_attach+0x3a/0x59 [ 5.609997] [<803bbde3>] ? bus_for_each_dev+0x36/0x58 [ 5.609997] [<803bc354>] ? driver_attach+0x14/0x16 [ 5.609997] [<803bc52a>] ? __driver_attach+0x0/0x59 [ 5.609997] [<803bc161>] ? bus_add_driver+0x93/0x196 [ 5.609997] [<803bc773>] ? driver_register+0x71/0xcd [ 5.609997] [<802eff05>] ? __pci_register_driver+0x3f/0x6e [ 5.609997] [<809af7ff>] ? piix_init+0x14/0x24 [ 5.609997] [<80984568>] ? kernel_init+0x128/0x269 [ 5.609997] [<809af7eb>] ? piix_init+0x0/0x24 [ 5.609997] [<802e2758>] ? trace_hardirqs_on_thunk+0xc/0x10 [ 5.609997] [<80116aef>] ? restore_nocheck_notrace+0x0/0xe [ 5.609997] [<80984440>] ? kernel_init+0x0/0x269 [ 5.609997] [<80984440>] ? kernel_init+0x0/0x269 [ 5.609997] [<80117d87>] ? kernel_thread_helper+0x7/0x10 [ 5.609997] ======================= [ 5.609997] Code: 75 02 b3 01 8d 43 01 8b 5d fc c9 c3 55 89 e5 57 89 cf 56 53 89 d3 4b 83 ec 04 83 fb 03 89 55 f0 77 09 c6 01 5c c6 41 01 00 eb 59 04 19 00 8b 55 f0 8d 34 11 89 c2 eb 19 8b 42 08 83 eb 05 89 [ 5.609997] EIP: [<80339e2f>] acpi_ns_build_external_path+0x1f/0x80 SS:ESP 0068:f7c4dc18 [ 5.613331] Kernel panic - not syncing: Fatal exception [ 5.613331] Rebooting in 1 seconds..[ 4.646664] ata1: SATA link up 1.5 Gbps (SStatus 113 SControl 300) I have bisected it down to: # bad: [5b664cbe] Merge branch 'upstream-linus' of git://git.kernel. # good: [bce7f795] Linux 2.6.26 # good: [e18425ab] Merge branch 'tracing/for-linus' of git://git.kern # good: [cadc7236] Merge branch 'bkl-removal' into next # good: [4515889a] Merge branch 'merge' of git://git.kernel.org/pub/s # good: [42fdd14e] Merge git://git.kernel.org/pub/scm/linux/kernel/gi # good: [8a0ca91f] Merge branch 'for-linus' of git://git.kernel.org/p # bad: [0af4b8cb] ACPI: Introduce new device wakeup flag 'prepared' # good: [fe997407] PCI: construct one fakephp slot per PCI slot # bad: [531f254a] PCIE: aer: use dev_printk when possible # bad: [15650a20] x86/PCI: fixup early quirk probing # good: [0e6859d9] ACPI PM: Remove obsolete Toshiba workaround # bad: [8344b566] PCI: ACPI PCI slot detection driver # good: [f46753c9] PCI: introduce pci_slot | 8344b568f5bdc7ee1bba909de3294c6348c36056 is first bad commit | commit 8344b568f5bdc7ee1bba909de3294c6348c36056 | Author: Alex Chiang | Date: Tue Jun 10 15:30:42 2008 -0600 | | PCI: ACPI PCI slot detection driver | | Detect all physical PCI slots as described by ACPI, and create entries in | /sys/bus/pci/slots/. I.e. the new CONFIG_ACPI_PCI_SLOT=y option was causing this crash. But the bug is not mainly in this new PCI code - that code was just hitting the ACPI code in a new way which made ACPI break. The crash signature shows that we are crashing on this instruction: movb $0x0, (%ecx, %ebx, 1) ECX and EBX are 0x10 and -1. It's this line in drivers/acpi/namespace/nsnames.c's acpi_ns_build_external_path(): name_buffer[index] = 0; I.e. name_buffer is 0x10 and index is -1. index -1 corresponds to size 0, and name_buffer 0x10 is slab's ZERO_SIZE_PTR special-case for zero-sized allocations. I.e. when we called acpi_ns_handle_to_pathname(), we got required_size of 0 due to an error condition, but this is passed to the ACPI allocator unconditionally: required_size = acpi_ns_get_pathname_length(node); /* Validate/Allocate/Clear caller buffer */ status = acpi_ut_initialize_buffer(buffer, required_size); if (ACPI_FAILURE(status)) { return_ACPI_STATUS(status); } Where acpi_ut_initialize_buffer(), through many (unnecessary) layers, ends up calling kzalloc(0). Which returns 0x10 and that then causes the crash later on. So fix both callers of acpi_ns_get_pathname_length(), which can return 0 in case of an invalid node. Also add a WARN_ON() against zero sized allocations in acpi_ut_initialize_buffer() to make it easier to find similar instances of this bug. I have tested this patch for the past 24 hours and the crash has not reappeared. Signed-off-by: Ingo Molnar Signed-off-by: Andi Kleen --- drivers/acpi/namespace/nsnames.c | 8 ++++++++ drivers/acpi/utilities/utalloc.c | 4 ++++ 2 files changed, 12 insertions(+) diff --git a/drivers/acpi/namespace/nsnames.c b/drivers/acpi/namespace/nsnames.c index cffef1bcbdbc..549db42f16cf 100644 --- a/drivers/acpi/namespace/nsnames.c +++ b/drivers/acpi/namespace/nsnames.c @@ -137,6 +137,10 @@ char *acpi_ns_get_external_pathname(struct acpi_namespace_node *node) /* Calculate required buffer size based on depth below root */ size = acpi_ns_get_pathname_length(node); + if (!size) { + ACPI_ERROR((AE_INFO, "Invalid node failure")); + return_PTR(NULL); + } /* Allocate a buffer to be returned to caller */ @@ -229,6 +233,10 @@ acpi_ns_handle_to_pathname(acpi_handle target_handle, /* Determine size required for the caller buffer */ required_size = acpi_ns_get_pathname_length(node); + if (!required_size) { + ACPI_ERROR((AE_INFO, "Invalid node failure")); + return_ACPI_STATUS(AE_ERROR); + } /* Validate/Allocate/Clear caller buffer */ diff --git a/drivers/acpi/utilities/utalloc.c b/drivers/acpi/utilities/utalloc.c index 3dfb8a442b26..e7bf34a7b1d2 100644 --- a/drivers/acpi/utilities/utalloc.c +++ b/drivers/acpi/utilities/utalloc.c @@ -242,6 +242,10 @@ acpi_ut_initialize_buffer(struct acpi_buffer * buffer, { acpi_status status = AE_OK; + if (!required_length) { + WARN_ON(1); + return AE_ERROR; + } switch (buffer->length) { case ACPI_NO_BUFFER: From 11d579ee0a19052a5a90ebfe0c39e7ed8ce8a9dc Mon Sep 17 00:00:00 2001 From: Wolfram Sang Date: Sat, 28 Jun 2008 20:31:52 +0200 Subject: [PATCH 020/857] powerpc/mpc5200: Fix wrong 'no interrupt' handling in of_i2c If an I2C device node does not specify an interrupt, the .irq member of the board_info struct was set to -1. This caused crashes on following irq_dispose_mappings. Leave it NO_IRQ as returned from irq_of_parse_and_map. (Suggesting -1 as 'i2c-no-irq' used to be a bug in linux/i2c.h.) Signed-off-by: Wolfram Sang Acked-by: Sean MacLennan Signed-off-by: Grant Likely --- drivers/of/of_i2c.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/of/of_i2c.c b/drivers/of/of_i2c.c index 5c015d310d4a..344e1b03dd8b 100644 --- a/drivers/of/of_i2c.c +++ b/drivers/of/of_i2c.c @@ -91,8 +91,6 @@ void of_register_i2c_devices(struct i2c_adapter *adap, } info.irq = irq_of_parse_and_map(node, 0); - if (info.irq == NO_IRQ) - info.irq = -1; if (of_find_i2c_driver(node, &info) < 0) { irq_dispose_mapping(info.irq); From 6a4a636fad018500c5db7a2b56a00caeb21cbb2c Mon Sep 17 00:00:00 2001 From: Jon Smirl Date: Sun, 20 Jul 2008 11:27:22 -0400 Subject: [PATCH 021/857] powerpc/mpc5200: Add AC97 register definitions for the MPC52xx PSC Needed by the PSC AC97 sound driver Signed-off-by: Jon Smirl Signed-off-by: Grant Likely --- include/asm-powerpc/mpc52xx_psc.h | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/include/asm-powerpc/mpc52xx_psc.h b/include/asm-powerpc/mpc52xx_psc.h index 710c5d36efaa..5467c2c0faa7 100644 --- a/include/asm-powerpc/mpc52xx_psc.h +++ b/include/asm-powerpc/mpc52xx_psc.h @@ -132,8 +132,12 @@ struct mpc52xx_psc { u8 reserved5[3]; u8 ctlr; /* PSC + 0x1c */ u8 reserved6[3]; - u16 ccr; /* PSC + 0x20 */ - u8 reserved7[14]; + /* BitClkDiv field of CCR is byte swapped in + * the hardware for mpc5200/b compatibility */ + u32 ccr; /* PSC + 0x20 */ + u32 ac97_slots; /* PSC + 0x24 */ + u32 ac97_cmd; /* PSC + 0x28 */ + u32 ac97_data; /* PSC + 0x2c */ u8 ivr; /* PSC + 0x30 */ u8 reserved8[3]; u8 ip; /* PSC + 0x34 */ From 78f56bd3d2dbe173bf1a946b353bf72ab9c0b94e Mon Sep 17 00:00:00 2001 From: Jon Smirl Date: Sun, 20 Jul 2008 11:30:08 -0400 Subject: [PATCH 022/857] powerpc/mpc5200: Remove fsl-soc.c from mpc5200 build, it is not needed. Signed-off-by: Jon Smirl Signed-off-by: Grant Likely --- arch/powerpc/platforms/52xx/Kconfig | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/powerpc/platforms/52xx/Kconfig b/arch/powerpc/platforms/52xx/Kconfig index acd2fc8cf492..981b84b7599a 100644 --- a/arch/powerpc/platforms/52xx/Kconfig +++ b/arch/powerpc/platforms/52xx/Kconfig @@ -1,7 +1,6 @@ config PPC_MPC52xx bool "52xx-based boards" depends on PPC_MULTIPLATFORM && PPC32 - select FSL_SOC select PPC_CLOCK config PPC_MPC5200_SIMPLE From 6d5509babce654fd9ce0ff6689dbdf6ce56c43ae Mon Sep 17 00:00:00 2001 From: Grant Likely Date: Thu, 15 May 2008 17:04:53 -0600 Subject: [PATCH 023/857] powerpc/mpc5200: Make mpc5200 GPIO driver select the GENERIC_GPIO config CONFIG_GENERIC_GPIO is needed for the gpio driver to work. Signed-off-by: Grant Likely --- arch/powerpc/platforms/52xx/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/powerpc/platforms/52xx/Kconfig b/arch/powerpc/platforms/52xx/Kconfig index 981b84b7599a..c1ca4f3ee5da 100644 --- a/arch/powerpc/platforms/52xx/Kconfig +++ b/arch/powerpc/platforms/52xx/Kconfig @@ -46,6 +46,7 @@ config PPC_MPC5200_BUGFIX config PPC_MPC5200_GPIO bool "MPC5200 GPIO support" depends on PPC_MPC52xx + select GENERIC_GPIO select HAVE_GPIO_LIB help Enable gpiolib support for mpc5200 based boards From a19dd1bd7df839c52a668abcf288c2239442c3c9 Mon Sep 17 00:00:00 2001 From: Grant Likely Date: Tue, 22 Jul 2008 01:13:54 -0600 Subject: [PATCH 024/857] powerpc/mpc5200: add PSC SICR bit definitions Required by the PSC I2S audio driver. Signed-off-by: Grant Likely --- include/asm-powerpc/mpc52xx_psc.h | 32 ++++++++++++++++++++++++++++++- 1 file changed, 31 insertions(+), 1 deletion(-) diff --git a/include/asm-powerpc/mpc52xx_psc.h b/include/asm-powerpc/mpc52xx_psc.h index 5467c2c0faa7..8917ed630565 100644 --- a/include/asm-powerpc/mpc52xx_psc.h +++ b/include/asm-powerpc/mpc52xx_psc.h @@ -60,10 +60,12 @@ #define MPC52xx_PSC_RXTX_FIFO_ALARM 0x0002 #define MPC52xx_PSC_RXTX_FIFO_EMPTY 0x0001 -/* PSC interrupt mask bits */ +/* PSC interrupt status/mask bits */ #define MPC52xx_PSC_IMR_TXRDY 0x0100 #define MPC52xx_PSC_IMR_RXRDY 0x0200 #define MPC52xx_PSC_IMR_DB 0x0400 +#define MPC52xx_PSC_IMR_TXEMP 0x0800 +#define MPC52xx_PSC_IMR_ORERR 0x1000 #define MPC52xx_PSC_IMR_IPC 0x8000 /* PSC input port change bit */ @@ -92,6 +94,34 @@ #define MPC52xx_PSC_RFNUM_MASK 0x01ff +#define MPC52xx_PSC_SICR_DTS1 (1 << 29) +#define MPC52xx_PSC_SICR_SHDR (1 << 28) +#define MPC52xx_PSC_SICR_SIM_MASK (0xf << 24) +#define MPC52xx_PSC_SICR_SIM_UART (0x0 << 24) +#define MPC52xx_PSC_SICR_SIM_UART_DCD (0x8 << 24) +#define MPC52xx_PSC_SICR_SIM_CODEC_8 (0x1 << 24) +#define MPC52xx_PSC_SICR_SIM_CODEC_16 (0x2 << 24) +#define MPC52xx_PSC_SICR_SIM_AC97 (0x3 << 24) +#define MPC52xx_PSC_SICR_SIM_SIR (0x8 << 24) +#define MPC52xx_PSC_SICR_SIM_SIR_DCD (0xc << 24) +#define MPC52xx_PSC_SICR_SIM_MIR (0x5 << 24) +#define MPC52xx_PSC_SICR_SIM_FIR (0x6 << 24) +#define MPC52xx_PSC_SICR_SIM_CODEC_24 (0x7 << 24) +#define MPC52xx_PSC_SICR_SIM_CODEC_32 (0xf << 24) +#define MPC52xx_PSC_SICR_GENCLK (1 << 23) +#define MPC52xx_PSC_SICR_I2S (1 << 22) +#define MPC52xx_PSC_SICR_CLKPOL (1 << 21) +#define MPC52xx_PSC_SICR_SYNCPOL (1 << 20) +#define MPC52xx_PSC_SICR_CELLSLAVE (1 << 19) +#define MPC52xx_PSC_SICR_CELL2XCLK (1 << 18) +#define MPC52xx_PSC_SICR_ESAI (1 << 17) +#define MPC52xx_PSC_SICR_ENAC97 (1 << 16) +#define MPC52xx_PSC_SICR_SPI (1 << 15) +#define MPC52xx_PSC_SICR_MSTR (1 << 14) +#define MPC52xx_PSC_SICR_CPOL (1 << 13) +#define MPC52xx_PSC_SICR_CPHA (1 << 12) +#define MPC52xx_PSC_SICR_USEEOF (1 << 11) +#define MPC52xx_PSC_SICR_DISABLEEOF (1 << 10) /* Structure of the hardware registers */ struct mpc52xx_psc { From 79c28acb2b7d66ca48d23e1c8b5e9e043aa634f8 Mon Sep 17 00:00:00 2001 From: Grant Likely Date: Fri, 11 Jul 2008 16:17:57 -0600 Subject: [PATCH 025/857] of-bindings: Add binding documentation for SPI busses and devices Add documentation about how to describe SPI busses in the device tree. Signed-off-by: Grant Likely Acked-by: Segher Boessenkool --- Documentation/powerpc/booting-without-of.txt | 57 ++++++++++++++++++++ 1 file changed, 57 insertions(+) diff --git a/Documentation/powerpc/booting-without-of.txt b/Documentation/powerpc/booting-without-of.txt index aee243a846a2..ee92fedada1a 100644 --- a/Documentation/powerpc/booting-without-of.txt +++ b/Documentation/powerpc/booting-without-of.txt @@ -59,6 +59,7 @@ Table of Contents p) Freescale Synchronous Serial Interface q) USB EHCI controllers r) MDIO on GPIOs + s) SPI busses VII - Marvell Discovery mv64[345]6x System Controller chips 1) The /system-controller node @@ -1881,6 +1882,62 @@ platforms are moved over to use the flattened-device-tree model. &qe_pio_c 6>; }; + s) SPI (Serial Peripheral Interface) busses + + SPI busses can be described with a node for the SPI master device + and a set of child nodes for each SPI slave on the bus. For this + discussion, it is assumed that the system's SPI controller is in + SPI master mode. This binding does not describe SPI controllers + in slave mode. + + The SPI master node requires the following properties: + - #address-cells - number of cells required to define a chip select + address on the SPI bus. + - #size-cells - should be zero. + - compatible - name of SPI bus controller following generic names + recommended practice. + No other properties are required in the SPI bus node. It is assumed + that a driver for an SPI bus device will understand that it is an SPI bus. + However, the binding does not attempt to define the specific method for + assigning chip select numbers. Since SPI chip select configuration is + flexible and non-standardized, it is left out of this binding with the + assumption that board specific platform code will be used to manage + chip selects. Individual drivers can define additional properties to + support describing the chip select layout. + + SPI slave nodes must be children of the SPI master node and can + contain the following properties. + - reg - (required) chip select address of device. + - compatible - (required) name of SPI device following generic names + recommended practice + - spi-max-frequency - (required) Maximum SPI clocking speed of device in Hz + - spi-cpol - (optional) Empty property indicating device requires + inverse clock polarity (CPOL) mode + - spi-cpha - (optional) Empty property indicating device requires + shifted clock phase (CPHA) mode + + SPI example for an MPC5200 SPI bus: + spi@f00 { + #address-cells = <1>; + #size-cells = <0>; + compatible = "fsl,mpc5200b-spi","fsl,mpc5200-spi"; + reg = <0xf00 0x20>; + interrupts = <2 13 0 2 14 0>; + interrupt-parent = <&mpc5200_pic>; + + ethernet-switch@0 { + compatible = "micrel,ks8995m"; + spi-max-frequency = <1000000>; + reg = <0>; + }; + + codec@1 { + compatible = "ti,tlv320aic26"; + spi-max-frequency = <100000>; + reg = <1>; + }; + }; + VII - Marvell Discovery mv64[345]6x System Controller chips =========================================================== From e4268aad42e9f37d01925022830b16bab3d0d5af Mon Sep 17 00:00:00 2001 From: Alex Chiang Date: Thu, 17 Jul 2008 11:13:32 -0600 Subject: [PATCH 026/857] PCI hotplug: fix error path in pci_slot's register_slot Juha Leppnen noticed that an error path in register_slot() wasn't returning appropriately, leading to a condition where we might access a kfree'ed pointer, so let's fix that. Additionally, fix up the copyright information in the file while we're in there. Signed-off-by: Alex Chiang Signed-off-by: Jesse Barnes --- drivers/acpi/pci_slot.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/acpi/pci_slot.c b/drivers/acpi/pci_slot.c index b9ab030a52d5..dd376f7ad090 100644 --- a/drivers/acpi/pci_slot.c +++ b/drivers/acpi/pci_slot.c @@ -6,8 +6,8 @@ * Thanks to Kenji Kaneshige for code * review and fixes. * - * Copyright (C) 2007 Alex Chiang - * Copyright (C) 2007 Hewlett-Packard Development Company, L.P. + * Copyright (C) 2007-2008 Hewlett-Packard Development Company, L.P. + * Alex Chiang * * This program is free software; you can redistribute it and/or modify it * under the terms and conditions of the GNU General Public License, @@ -158,6 +158,7 @@ register_slot(acpi_handle handle, u32 lvl, void *context, void **rv) if (IS_ERR(pci_slot)) { err("pci_create_slot returned %ld\n", PTR_ERR(pci_slot)); kfree(slot); + return AE_OK; } slot->root_handle = parent_context->root_handle; From f42e86d95fa53d3a62b2795515da18b4f41b0480 Mon Sep 17 00:00:00 2001 From: Yong Wang Date: Tue, 22 Jul 2008 14:14:18 -0700 Subject: [PATCH 027/857] PCI/DMAR: don't assume presence of RMRRs RMRRs do not necessarily have to be present on all VT-d capable platforms. The printk is just informational and does not need to be followed by an error return. Signed-off-by: Yong Y Wang Cc: Fenghua Yu Cc: mark gross Cc: Keshavamurthy, Anil S Signed-off-by: Andrew Morton Signed-off-by: Jesse Barnes --- drivers/pci/dmar.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c index f941f609dbf3..8bf86ae2333f 100644 --- a/drivers/pci/dmar.c +++ b/drivers/pci/dmar.c @@ -317,10 +317,8 @@ int __init dmar_table_init(void) return -ENODEV; } - if (list_empty(&dmar_rmrr_units)) { + if (list_empty(&dmar_rmrr_units)) printk(KERN_INFO PREFIX "No RMRR found\n"); - return -ENODEV; - } return 0; } From dd5bdff83b19d9174126e0398b47117c3a80e22d Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Tue, 22 Jul 2008 14:14:22 -0700 Subject: [PATCH 028/857] RDMA/cma: Add RDMA_CM_EVENT_ADDR_CHANGE event Add an RDMA_CM_EVENT_ADDR_CHANGE event can be used by rdma-cm consumers that wish to have their RDMA sessions always use the same links (eg ) as the IP stack does. In the current code, this does not happen when bonding is used and fail-over happened but the IB link used by an already existing session is operating fine. Use the netevent notification for sensing that a change has happened in the IP stack, then scan the rdma-cm ID list to see if there is an ID that is "misaligned" with respect to the IP stack, and deliver RDMA_CM_EVENT_ADDR_CHANGE for this ID. The consumer can act on the event or just ignore it. Signed-off-by: Or Gerlitz Signed-off-by: Roland Dreier --- drivers/infiniband/core/cma.c | 92 +++++++++++++++++++++++++++++++++++ include/rdma/rdma_cm.h | 3 +- 2 files changed, 94 insertions(+), 1 deletion(-) diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c index ae11d5cc74d0..79792c92e6fb 100644 --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -168,6 +168,12 @@ struct cma_work { struct rdma_cm_event event; }; +struct cma_ndev_work { + struct work_struct work; + struct rdma_id_private *id; + struct rdma_cm_event event; +}; + union cma_ip_addr { struct in6_addr ip6; struct { @@ -1598,6 +1604,30 @@ static void cma_work_handler(struct work_struct *_work) kfree(work); } +static void cma_ndev_work_handler(struct work_struct *_work) +{ + struct cma_ndev_work *work = container_of(_work, struct cma_ndev_work, work); + struct rdma_id_private *id_priv = work->id; + int destroy = 0; + + mutex_lock(&id_priv->handler_mutex); + if (id_priv->state == CMA_DESTROYING || + id_priv->state == CMA_DEVICE_REMOVAL) + goto out; + + if (id_priv->id.event_handler(&id_priv->id, &work->event)) { + cma_exch(id_priv, CMA_DESTROYING); + destroy = 1; + } + +out: + mutex_unlock(&id_priv->handler_mutex); + cma_deref_id(id_priv); + if (destroy) + rdma_destroy_id(&id_priv->id); + kfree(work); +} + static int cma_resolve_ib_route(struct rdma_id_private *id_priv, int timeout_ms) { struct rdma_route *route = &id_priv->id.route; @@ -2723,6 +2753,65 @@ void rdma_leave_multicast(struct rdma_cm_id *id, struct sockaddr *addr) } EXPORT_SYMBOL(rdma_leave_multicast); +static int cma_netdev_change(struct net_device *ndev, struct rdma_id_private *id_priv) +{ + struct rdma_dev_addr *dev_addr; + struct cma_ndev_work *work; + + dev_addr = &id_priv->id.route.addr.dev_addr; + + if ((dev_addr->src_dev == ndev) && + memcmp(dev_addr->src_dev_addr, ndev->dev_addr, ndev->addr_len)) { + printk(KERN_INFO "RDMA CM addr change for ndev %s used by id %p\n", + ndev->name, &id_priv->id); + work = kzalloc(sizeof *work, GFP_KERNEL); + if (!work) + return -ENOMEM; + + INIT_WORK(&work->work, cma_ndev_work_handler); + work->id = id_priv; + work->event.event = RDMA_CM_EVENT_ADDR_CHANGE; + atomic_inc(&id_priv->refcount); + queue_work(cma_wq, &work->work); + } + + return 0; +} + +static int cma_netdev_callback(struct notifier_block *self, unsigned long event, + void *ctx) +{ + struct net_device *ndev = (struct net_device *)ctx; + struct cma_device *cma_dev; + struct rdma_id_private *id_priv; + int ret = NOTIFY_DONE; + + if (dev_net(ndev) != &init_net) + return NOTIFY_DONE; + + if (event != NETDEV_BONDING_FAILOVER) + return NOTIFY_DONE; + + if (!(ndev->flags & IFF_MASTER) || !(ndev->priv_flags & IFF_BONDING)) + return NOTIFY_DONE; + + mutex_lock(&lock); + list_for_each_entry(cma_dev, &dev_list, list) + list_for_each_entry(id_priv, &cma_dev->id_list, list) { + ret = cma_netdev_change(ndev, id_priv); + if (ret) + goto out; + } + +out: + mutex_unlock(&lock); + return ret; +} + +static struct notifier_block cma_nb = { + .notifier_call = cma_netdev_callback +}; + static void cma_add_one(struct ib_device *device) { struct cma_device *cma_dev; @@ -2831,6 +2920,7 @@ static int cma_init(void) ib_sa_register_client(&sa_client); rdma_addr_register_client(&addr_client); + register_netdevice_notifier(&cma_nb); ret = ib_register_client(&cma_client); if (ret) @@ -2838,6 +2928,7 @@ static int cma_init(void) return 0; err: + unregister_netdevice_notifier(&cma_nb); rdma_addr_unregister_client(&addr_client); ib_sa_unregister_client(&sa_client); destroy_workqueue(cma_wq); @@ -2847,6 +2938,7 @@ static int cma_init(void) static void cma_cleanup(void) { ib_unregister_client(&cma_client); + unregister_netdevice_notifier(&cma_nb); rdma_addr_unregister_client(&addr_client); ib_sa_unregister_client(&sa_client); destroy_workqueue(cma_wq); diff --git a/include/rdma/rdma_cm.h b/include/rdma/rdma_cm.h index 22bb2e7bab1a..001d606517ff 100644 --- a/include/rdma/rdma_cm.h +++ b/include/rdma/rdma_cm.h @@ -57,7 +57,8 @@ enum rdma_cm_event_type { RDMA_CM_EVENT_DISCONNECTED, RDMA_CM_EVENT_DEVICE_REMOVAL, RDMA_CM_EVENT_MULTICAST_JOIN, - RDMA_CM_EVENT_MULTICAST_ERROR + RDMA_CM_EVENT_MULTICAST_ERROR, + RDMA_CM_EVENT_ADDR_CHANGE }; enum rdma_port_space { From 38ca83a588662f0af684ba2567dd910a564268ab Mon Sep 17 00:00:00 2001 From: Amir Vadai Date: Tue, 22 Jul 2008 14:14:23 -0700 Subject: [PATCH 029/857] RDMA/cma: Add RDMA_CM_EVENT_TIMEWAIT_EXIT event Consumers that want to re-use their QPs in new connections need to know when the QP has exited the timewait state. Report the timewait event through the rdma_cm. Signed-off-by: Amir Vadai Acked-by: Sean Hefty Signed-off-by: Roland Dreier --- drivers/infiniband/core/cma.c | 7 ++++++- include/rdma/rdma_cm.h | 3 ++- 2 files changed, 8 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c index 79792c92e6fb..e980ff3335db 100644 --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -920,7 +920,10 @@ static int cma_ib_handler(struct ib_cm_id *cm_id, struct ib_cm_event *ib_event) struct rdma_cm_event event; int ret = 0; - if (cma_disable_callback(id_priv, CMA_CONNECT)) + if ((ib_event->event != IB_CM_TIMEWAIT_EXIT && + cma_disable_callback(id_priv, CMA_CONNECT)) || + (ib_event->event == IB_CM_TIMEWAIT_EXIT && + cma_disable_callback(id_priv, CMA_DISCONNECT))) return 0; memset(&event, 0, sizeof event); @@ -956,6 +959,8 @@ static int cma_ib_handler(struct ib_cm_id *cm_id, struct ib_cm_event *ib_event) event.event = RDMA_CM_EVENT_DISCONNECTED; break; case IB_CM_TIMEWAIT_EXIT: + event.event = RDMA_CM_EVENT_TIMEWAIT_EXIT; + break; case IB_CM_MRA_RECEIVED: /* ignore event */ goto out; diff --git a/include/rdma/rdma_cm.h b/include/rdma/rdma_cm.h index 001d606517ff..df7faf09d66f 100644 --- a/include/rdma/rdma_cm.h +++ b/include/rdma/rdma_cm.h @@ -58,7 +58,8 @@ enum rdma_cm_event_type { RDMA_CM_EVENT_DEVICE_REMOVAL, RDMA_CM_EVENT_MULTICAST_JOIN, RDMA_CM_EVENT_MULTICAST_ERROR, - RDMA_CM_EVENT_ADDR_CHANGE + RDMA_CM_EVENT_ADDR_CHANGE, + RDMA_CM_EVENT_TIMEWAIT_EXIT }; enum rdma_port_space { From 2f5de1512884da8c74bec2c76e8f114b972ab4be Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Tue, 22 Jul 2008 14:16:21 -0700 Subject: [PATCH 030/857] IB/iser: Add support for RDMA_CM_EVENT_ADDR_CHANGE event Enhance iser to act upon notification on network stack changes that make its RDMA connection unaligned with the link used by the stack for the IPs used to establish the connection. When RDMA_CM_EVENT_ADDR_CHANGE arrives, just disconnect the connection, assuming that the user space iscsid daemon will reconnect, and the new connection will be aligned with the IP stack. Signed-off-by: Or Gerlitz Signed-off-by: Roland Dreier --- drivers/infiniband/ulp/iser/iser_verbs.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/infiniband/ulp/iser/iser_verbs.c b/drivers/infiniband/ulp/iser/iser_verbs.c index 3a917c1f796f..63462ecca147 100644 --- a/drivers/infiniband/ulp/iser/iser_verbs.c +++ b/drivers/infiniband/ulp/iser/iser_verbs.c @@ -483,6 +483,7 @@ static int iser_cma_handler(struct rdma_cm_id *cma_id, struct rdma_cm_event *eve break; case RDMA_CM_EVENT_DISCONNECTED: case RDMA_CM_EVENT_DEVICE_REMOVAL: + case RDMA_CM_EVENT_ADDR_CHANGE: iser_disconnected_handler(cma_id); break; default: From 5b673b71c8ca0fbdb99dc1b1434cfb554212d6ff Mon Sep 17 00:00:00 2001 From: Joachim Fenkes Date: Tue, 22 Jul 2008 14:18:07 -0700 Subject: [PATCH 031/857] IB/ehca: Filter PATH_MIG events if QP was never armed Certain firmware versions sometimes cause spurious PATH_MIG events to occur during QP creation. Filter these events by making sure PATH_MIG events are only handed down when they actually make sense (i.e. when the QP has been armed at least once). Signed-off-by: Joachim Fenkes Signed-off-by: Roland Dreier --- drivers/infiniband/hw/ehca/ehca_classes.h | 1 + drivers/infiniband/hw/ehca/ehca_irq.c | 4 ++++ drivers/infiniband/hw/ehca/ehca_qp.c | 2 ++ 3 files changed, 7 insertions(+) diff --git a/drivers/infiniband/hw/ehca/ehca_classes.h b/drivers/infiniband/hw/ehca/ehca_classes.h index 1e9e99a13933..0b0618edd645 100644 --- a/drivers/infiniband/hw/ehca/ehca_classes.h +++ b/drivers/infiniband/hw/ehca/ehca_classes.h @@ -194,6 +194,7 @@ struct ehca_qp { u32 packet_count; atomic_t nr_events; /* events seen */ wait_queue_head_t wait_completion; + int mig_armed; }; #define IS_SRQ(qp) (qp->ext_type == EQPT_SRQ) diff --git a/drivers/infiniband/hw/ehca/ehca_irq.c b/drivers/infiniband/hw/ehca/ehca_irq.c index 0792d930c481..99642a6e17c4 100644 --- a/drivers/infiniband/hw/ehca/ehca_irq.c +++ b/drivers/infiniband/hw/ehca/ehca_irq.c @@ -178,6 +178,10 @@ static void dispatch_qp_event(struct ehca_shca *shca, struct ehca_qp *qp, { struct ib_event event; + /* PATH_MIG without the QP ever having been armed is false alarm */ + if (event_type == IB_EVENT_PATH_MIG && !qp->mig_armed) + return; + event.device = &shca->ib_device; event.event = event_type; diff --git a/drivers/infiniband/hw/ehca/ehca_qp.c b/drivers/infiniband/hw/ehca/ehca_qp.c index 3f59587338ea..ea13efddf175 100644 --- a/drivers/infiniband/hw/ehca/ehca_qp.c +++ b/drivers/infiniband/hw/ehca/ehca_qp.c @@ -1460,6 +1460,8 @@ static int internal_modify_qp(struct ib_qp *ibqp, goto modify_qp_exit2; } mqpcb->path_migration_state = attr->path_mig_state + 1; + if (attr->path_mig_state == IB_MIG_REARM) + my_qp->mig_armed = 1; update_mask |= EHCA_BMASK_SET(MQPCB_MASK_PATH_MIGRATION_STATE, 1); } From 593e4d4a05c8263a6dbd5452c21d47c5bdadd40c Mon Sep 17 00:00:00 2001 From: Joachim Fenkes Date: Tue, 22 Jul 2008 14:18:08 -0700 Subject: [PATCH 032/857] IB/ehca: Use default value for Local CA ACK Delay if FW returns 0 Some firmware versions report a Local CA ACK Delay of 0. In that case, return a more sensible default value of 12 (-> 16 msec) instead. Signed-off-by: Joachim Fenkes Signed-off-by: Roland Dreier --- drivers/infiniband/hw/ehca/ehca_hca.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/ehca/ehca_hca.c b/drivers/infiniband/hw/ehca/ehca_hca.c index bc3b37d2070f..46288220cfbb 100644 --- a/drivers/infiniband/hw/ehca/ehca_hca.c +++ b/drivers/infiniband/hw/ehca/ehca_hca.c @@ -114,7 +114,9 @@ int ehca_query_device(struct ib_device *ibdev, struct ib_device_attr *props) } props->max_pkeys = 16; - props->local_ca_ack_delay = min_t(u8, rblock->local_ca_ack_delay, 255); + /* Some FW versions say 0 here; insert sensible value in that case */ + props->local_ca_ack_delay = rblock->local_ca_ack_delay ? + min_t(u8, rblock->local_ca_ack_delay, 255) : 12; props->max_raw_ipv6_qp = limit_uint(rblock->max_raw_ipv6_qp); props->max_raw_ethy_qp = limit_uint(rblock->max_raw_ethy_qp); props->max_mcast_grp = limit_uint(rblock->max_mcast_grp); From 1a867c33bb65f2921351a9bdd98548bb96f0ff8c Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Tue, 22 Jul 2008 14:18:10 -0700 Subject: [PATCH 033/857] IB/ehca: Release mutex in error path of alloc_small_queue_page() The pd->lock mutex is released on a successful return, so it should be released on an error return as well. The semantic patch that makes this change is as follows: (http://www.emn.fr/x-info/coccinelle/) // @@ expression l; @@ mutex_lock(l); ... when != mutex_unlock(l) when any when strict ( if (...) { ... when != mutex_unlock(l) + mutex_unlock(l); return ...; } | mutex_unlock(l); ) // Signed-off-by: Julia Lawall Signed-off-by: Roland Dreier --- drivers/infiniband/hw/ehca/ipz_pt_fn.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/infiniband/hw/ehca/ipz_pt_fn.c b/drivers/infiniband/hw/ehca/ipz_pt_fn.c index 661f8db62706..c3a328465431 100644 --- a/drivers/infiniband/hw/ehca/ipz_pt_fn.c +++ b/drivers/infiniband/hw/ehca/ipz_pt_fn.c @@ -163,6 +163,7 @@ static int alloc_small_queue_page(struct ipz_queue *queue, struct ehca_pd *pd) out: ehca_err(pd->ib_pd.device, "failed to allocate small queue page"); + mutex_unlock(&pd->lock); return 0; } From 64b784b583061ebfe1d484dd1fdc5a26c6d4293f Mon Sep 17 00:00:00 2001 From: Ralph Campbell Date: Tue, 22 Jul 2008 14:18:33 -0700 Subject: [PATCH 034/857] IB/sa_query: Check if sm_ah is NULL in ib_sa_remove_one() If update_sm_ah() fails, it leaves the port's sm_ah as NULL. Then if the device or module is removed, ib_sa_remove_one() will dereference a NULL pointer when it calls kref_put(). Fix this by testing if sm_ah is NULL before dropping the reference. Signed-off-by: Ralph Campbell Signed-off-by: Roland Dreier --- drivers/infiniband/core/sa_query.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/infiniband/core/sa_query.c b/drivers/infiniband/core/sa_query.c index 1341de793e51..7863a50d56f2 100644 --- a/drivers/infiniband/core/sa_query.c +++ b/drivers/infiniband/core/sa_query.c @@ -1064,7 +1064,8 @@ static void ib_sa_remove_one(struct ib_device *device) for (i = 0; i <= sa_dev->end_port - sa_dev->start_port; ++i) { ib_unregister_mad_agent(sa_dev->port[i].agent); - kref_put(&sa_dev->port[i].sm_ah->ref, free_sm_ah); + if (sa_dev->port[i].sm_ah) + kref_put(&sa_dev->port[i].sm_ah->ref, free_sm_ah); } kfree(sa_dev); From 01b3fc8b15432f7931e40fe099839e1559fb0e09 Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Tue, 22 Jul 2008 14:18:34 -0700 Subject: [PATCH 035/857] IPoIB: Include err code in trace message for ib_sa_path_rec_get() failures Print the return code of ib_sa_path_rec_get() if it fails to help debug errors. Signed-off-by: Or Gerlitz Signed-off-by: Roland Dreier --- drivers/infiniband/ulp/ipoib/ipoib_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c index 8be9ea0436e6..f51201b17bfd 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_main.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c @@ -548,7 +548,7 @@ static int path_rec_start(struct net_device *dev, path_rec_completion, path, &path->query); if (path->query_id < 0) { - ipoib_warn(priv, "ib_sa_path_rec_get failed\n"); + ipoib_warn(priv, "ib_sa_path_rec_get failed: %d\n", path->query_id); path->query = NULL; return path->query_id; } From 1ca8d15619f725e223c19137350b0336b9196193 Mon Sep 17 00:00:00 2001 From: Dotan Barak Date: Tue, 22 Jul 2008 14:18:34 -0700 Subject: [PATCH 036/857] RDMA/iwcm: Remove IB_ACCESS_LOCAL_WRITE from remote QP attributes Remove IB_ACCESS_LOCAL_WRITE from qp.qp_access_flags because this attribute is only used to set remote permissions. Signed-off-by: Dotan Barak Signed-off-by: Roland Dreier --- drivers/infiniband/core/iwcm.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/infiniband/core/iwcm.c b/drivers/infiniband/core/iwcm.c index 81c9195b512a..8f9509e1ebf7 100644 --- a/drivers/infiniband/core/iwcm.c +++ b/drivers/infiniband/core/iwcm.c @@ -942,8 +942,7 @@ static int iwcm_init_qp_init_attr(struct iwcm_id_private *cm_id_priv, case IW_CM_STATE_CONN_RECV: case IW_CM_STATE_ESTABLISHED: *qp_attr_mask = IB_QP_STATE | IB_QP_ACCESS_FLAGS; - qp_attr->qp_access_flags = IB_ACCESS_LOCAL_WRITE | - IB_ACCESS_REMOTE_WRITE| + qp_attr->qp_access_flags = IB_ACCESS_REMOTE_WRITE| IB_ACCESS_REMOTE_READ; ret = 0; break; From 51f5f0ee22b98980f7816d42647467cd5f4b3b45 Mon Sep 17 00:00:00 2001 From: Jack Morgenstein Date: Tue, 22 Jul 2008 14:19:37 -0700 Subject: [PATCH 037/857] mlx4_core: Add module parameter to enable QoS support Add a module parameter "enable_qos" to mlx4_core. If this param is set, enable support for QoS in the INIT_HCA command. By default, the parameter is set to 0 (disabled). Signed-off-by: Jack Morgenstein Signed-off-by: Roland Dreier --- drivers/net/mlx4/fw.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/net/mlx4/fw.c b/drivers/net/mlx4/fw.c index 2b5006b9be67..0851ebdddfd4 100644 --- a/drivers/net/mlx4/fw.c +++ b/drivers/net/mlx4/fw.c @@ -46,6 +46,10 @@ enum { extern void __buggy_use_of_MLX4_GET(void); extern void __buggy_use_of_MLX4_PUT(void); +static int enable_qos; +module_param(enable_qos, bool, 0444); +MODULE_PARM_DESC(enable_qos, "Enable Quality of Service support in the HCA (default: off)"); + #define MLX4_GET(dest, source, offset) \ do { \ void *__p = (char *) (source) + (offset); \ @@ -737,6 +741,10 @@ int mlx4_INIT_HCA(struct mlx4_dev *dev, struct mlx4_init_hca_param *param) if (dev->caps.flags & MLX4_DEV_CAP_FLAG_IPOIB_CSUM) *(inbox + INIT_HCA_FLAGS_OFFSET / 4) |= cpu_to_be32(1 << 3); + /* Enable QoS support if module parameter set */ + if (enable_qos) + *(inbox + INIT_HCA_FLAGS_OFFSET / 4) |= cpu_to_be32(1 << 2); + /* QPC/EEC/CQC/EQC/RDMARC attributes */ MLX4_PUT(inbox, param->qpc_base, INIT_HCA_QPC_BASE_OFFSET); From 47b374752aed1c029f995473c7c463ee3ae5fbaa Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Tue, 22 Jul 2008 14:19:39 -0700 Subject: [PATCH 038/857] IB/mlx4: Rename struct mlx4_lso_seg to mlx4_wqe_lso_seg Make the struct name consistent with other WQE segment struct types defined in . Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mlx4/qp.c | 2 +- include/linux/mlx4/qp.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c index 89eb6cbe592e..bda0859a5ac5 100644 --- a/drivers/infiniband/hw/mlx4/qp.c +++ b/drivers/infiniband/hw/mlx4/qp.c @@ -1395,7 +1395,7 @@ static void __set_data_seg(struct mlx4_wqe_data_seg *dseg, struct ib_sge *sg) dseg->addr = cpu_to_be64(sg->addr); } -static int build_lso_seg(struct mlx4_lso_seg *wqe, struct ib_send_wr *wr, +static int build_lso_seg(struct mlx4_wqe_lso_seg *wqe, struct ib_send_wr *wr, struct mlx4_ib_qp *qp, unsigned *lso_seg_len) { unsigned halign = ALIGN(sizeof *wqe + wr->wr.ud.hlen, 16); diff --git a/include/linux/mlx4/qp.h b/include/linux/mlx4/qp.h index 7f128b266faa..f02e9ed36cfa 100644 --- a/include/linux/mlx4/qp.h +++ b/include/linux/mlx4/qp.h @@ -219,7 +219,7 @@ struct mlx4_wqe_datagram_seg { __be32 reservd[2]; }; -struct mlx4_lso_seg { +struct mlx4_wqe_lso_seg { __be32 mss_hdr_size; __be32 header[0]; }; From 899698dad72340b562478b8b770317f2f0fe0c09 Mon Sep 17 00:00:00 2001 From: Jack Morgenstein Date: Tue, 22 Jul 2008 14:19:39 -0700 Subject: [PATCH 039/857] mlx4_code: Add missing FW status return code Add ICM_ERROR firmware status code. In mapping to errnos, -ENFILE seems closest. This is in preparation for providing more detailed log info using mlx4_err() in low-level driver when a non-zero status is returned. Signed-off-by: Jack Morgenstein Signed-off-by: Roland Dreier --- drivers/net/mlx4/cmd.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/mlx4/cmd.c b/drivers/net/mlx4/cmd.c index 70dff94a8bc6..04d5bc69a6f8 100644 --- a/drivers/net/mlx4/cmd.c +++ b/drivers/net/mlx4/cmd.c @@ -67,6 +67,8 @@ enum { CMD_STAT_BAD_INDEX = 0x0a, /* FW image corrupted: */ CMD_STAT_BAD_NVMEM = 0x0b, + /* Error in ICM mapping (e.g. not enough auxiliary ICM pages to execute command): */ + CMD_STAT_ICM_ERROR = 0x0c, /* Attempt to modify a QP/EE which is not in the presumed state: */ CMD_STAT_BAD_QP_STATE = 0x10, /* Bad segment parameters (Address/Size): */ @@ -119,6 +121,7 @@ static int mlx4_status_to_errno(u8 status) [CMD_STAT_BAD_RES_STATE] = -EBADF, [CMD_STAT_BAD_INDEX] = -EBADF, [CMD_STAT_BAD_NVMEM] = -EFAULT, + [CMD_STAT_ICM_ERROR] = -ENFILE, [CMD_STAT_BAD_QP_STATE] = -EINVAL, [CMD_STAT_BAD_SEG_PARAM] = -EFAULT, [CMD_STAT_REG_BOUND] = -EBUSY, From e4044cfc493338cd09870bd45dc646336bb66e9f Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Tue, 22 Jul 2008 14:19:40 -0700 Subject: [PATCH 040/857] mlx4_core: Keep free count for MTT buddy allocator MTT entries are allocated with a buddy allocator, which just keeps bitmaps for each level of the buddy table. However, all free space starts out at the highest order, and small allocations start scanning from the lowest order. When the lowest order tables have no free space, this can lead to scanning potentially millions of bits before finding a free entry at a higher order. We can avoid this by just keeping a count of how many free entries each order has, and skipping the bitmap scan when an order is completely empty. This provides a nice performance boost for a negligible increase in memory usage. Signed-off-by: Roland Dreier --- drivers/net/mlx4/mlx4.h | 1 + drivers/net/mlx4/mr.c | 28 +++++++++++++++++++--------- 2 files changed, 20 insertions(+), 9 deletions(-) diff --git a/drivers/net/mlx4/mlx4.h b/drivers/net/mlx4/mlx4.h index a4023c2dd050..78038499cff5 100644 --- a/drivers/net/mlx4/mlx4.h +++ b/drivers/net/mlx4/mlx4.h @@ -118,6 +118,7 @@ struct mlx4_bitmap { struct mlx4_buddy { unsigned long **bits; + unsigned int *num_free; int max_order; spinlock_t lock; }; diff --git a/drivers/net/mlx4/mr.c b/drivers/net/mlx4/mr.c index 03a9abcce524..b3ea93b98689 100644 --- a/drivers/net/mlx4/mr.c +++ b/drivers/net/mlx4/mr.c @@ -79,23 +79,26 @@ static u32 mlx4_buddy_alloc(struct mlx4_buddy *buddy, int order) spin_lock(&buddy->lock); - for (o = order; o <= buddy->max_order; ++o) { - m = 1 << (buddy->max_order - o); - seg = find_first_bit(buddy->bits[o], m); - if (seg < m) - goto found; - } + for (o = order; o <= buddy->max_order; ++o) + if (buddy->num_free[o]) { + m = 1 << (buddy->max_order - o); + seg = find_first_bit(buddy->bits[o], m); + if (seg < m) + goto found; + } spin_unlock(&buddy->lock); return -1; found: clear_bit(seg, buddy->bits[o]); + --buddy->num_free[o]; while (o > order) { --o; seg <<= 1; set_bit(seg ^ 1, buddy->bits[o]); + ++buddy->num_free[o]; } spin_unlock(&buddy->lock); @@ -113,11 +116,13 @@ static void mlx4_buddy_free(struct mlx4_buddy *buddy, u32 seg, int order) while (test_bit(seg ^ 1, buddy->bits[order])) { clear_bit(seg ^ 1, buddy->bits[order]); + --buddy->num_free[order]; seg >>= 1; ++order; } set_bit(seg, buddy->bits[order]); + ++buddy->num_free[order]; spin_unlock(&buddy->lock); } @@ -131,7 +136,9 @@ static int mlx4_buddy_init(struct mlx4_buddy *buddy, int max_order) buddy->bits = kzalloc((buddy->max_order + 1) * sizeof (long *), GFP_KERNEL); - if (!buddy->bits) + buddy->num_free = kzalloc((buddy->max_order + 1) * sizeof (int *), + GFP_KERNEL); + if (!buddy->bits || !buddy->num_free) goto err_out; for (i = 0; i <= buddy->max_order; ++i) { @@ -143,6 +150,7 @@ static int mlx4_buddy_init(struct mlx4_buddy *buddy, int max_order) } set_bit(0, buddy->bits[buddy->max_order]); + buddy->num_free[buddy->max_order] = 1; return 0; @@ -150,9 +158,10 @@ static int mlx4_buddy_init(struct mlx4_buddy *buddy, int max_order) for (i = 0; i <= buddy->max_order; ++i) kfree(buddy->bits[i]); - kfree(buddy->bits); - err_out: + kfree(buddy->bits); + kfree(buddy->num_free); + return -ENOMEM; } @@ -164,6 +173,7 @@ static void mlx4_buddy_cleanup(struct mlx4_buddy *buddy) kfree(buddy->bits[i]); kfree(buddy->bits); + kfree(buddy->num_free); } static u32 mlx4_alloc_mtt_range(struct mlx4_dev *dev, int order) From e8bb4beb2b1f90d499134f2849727ed04c3bedc4 Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Tue, 22 Jul 2008 14:20:05 -0700 Subject: [PATCH 041/857] IB/mthca: Keep free count for MTT buddy allocator MTT entries are allocated with a buddy allocator, which just keeps bitmaps for each level of the buddy table. However, all free space starts out at the highest order, and small allocations start scanning from the lowest order. When the lowest order tables have no free space, this can lead to scanning potentially millions of bits before finding a free entry at a higher order. We can avoid this by just keeping a count of how many free entries each order has, and skipping the bitmap scan when an order is completely empty. This provides a nice performance boost for a negligible increase in memory usage. Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mthca/mthca_dev.h | 1 + drivers/infiniband/hw/mthca/mthca_mr.c | 28 +++++++++++++++++-------- 2 files changed, 20 insertions(+), 9 deletions(-) diff --git a/drivers/infiniband/hw/mthca/mthca_dev.h b/drivers/infiniband/hw/mthca/mthca_dev.h index ee4d073c889f..252590116df5 100644 --- a/drivers/infiniband/hw/mthca/mthca_dev.h +++ b/drivers/infiniband/hw/mthca/mthca_dev.h @@ -202,6 +202,7 @@ struct mthca_pd_table { struct mthca_buddy { unsigned long **bits; + int *num_free; int max_order; spinlock_t lock; }; diff --git a/drivers/infiniband/hw/mthca/mthca_mr.c b/drivers/infiniband/hw/mthca/mthca_mr.c index 8489b1e81c0f..882e6b735915 100644 --- a/drivers/infiniband/hw/mthca/mthca_mr.c +++ b/drivers/infiniband/hw/mthca/mthca_mr.c @@ -89,23 +89,26 @@ static u32 mthca_buddy_alloc(struct mthca_buddy *buddy, int order) spin_lock(&buddy->lock); - for (o = order; o <= buddy->max_order; ++o) { - m = 1 << (buddy->max_order - o); - seg = find_first_bit(buddy->bits[o], m); - if (seg < m) - goto found; - } + for (o = order; o <= buddy->max_order; ++o) + if (buddy->num_free[o]) { + m = 1 << (buddy->max_order - o); + seg = find_first_bit(buddy->bits[o], m); + if (seg < m) + goto found; + } spin_unlock(&buddy->lock); return -1; found: clear_bit(seg, buddy->bits[o]); + --buddy->num_free[o]; while (o > order) { --o; seg <<= 1; set_bit(seg ^ 1, buddy->bits[o]); + ++buddy->num_free[o]; } spin_unlock(&buddy->lock); @@ -123,11 +126,13 @@ static void mthca_buddy_free(struct mthca_buddy *buddy, u32 seg, int order) while (test_bit(seg ^ 1, buddy->bits[order])) { clear_bit(seg ^ 1, buddy->bits[order]); + --buddy->num_free[order]; seg >>= 1; ++order; } set_bit(seg, buddy->bits[order]); + ++buddy->num_free[order]; spin_unlock(&buddy->lock); } @@ -141,7 +146,9 @@ static int mthca_buddy_init(struct mthca_buddy *buddy, int max_order) buddy->bits = kzalloc((buddy->max_order + 1) * sizeof (long *), GFP_KERNEL); - if (!buddy->bits) + buddy->num_free = kzalloc((buddy->max_order + 1) * sizeof (int *), + GFP_KERNEL); + if (!buddy->bits || !buddy->num_free) goto err_out; for (i = 0; i <= buddy->max_order; ++i) { @@ -154,6 +161,7 @@ static int mthca_buddy_init(struct mthca_buddy *buddy, int max_order) } set_bit(0, buddy->bits[buddy->max_order]); + buddy->num_free[buddy->max_order] = 1; return 0; @@ -161,9 +169,10 @@ static int mthca_buddy_init(struct mthca_buddy *buddy, int max_order) for (i = 0; i <= buddy->max_order; ++i) kfree(buddy->bits[i]); - kfree(buddy->bits); - err_out: + kfree(buddy->bits); + kfree(buddy->num_free); + return -ENOMEM; } @@ -175,6 +184,7 @@ static void mthca_buddy_cleanup(struct mthca_buddy *buddy) kfree(buddy->bits[i]); kfree(buddy->bits); + kfree(buddy->num_free); } static u32 mthca_alloc_mtt_range(struct mthca_dev *dev, int order, From e5899e1b7d73e67de758a32174a859cc2586c0b9 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 19 Jul 2008 14:39:24 +0200 Subject: [PATCH 042/857] PCI PM: make more PCI PM core functionality available to drivers Make more PCI PM core functionality available to drivers * Export pci_pme_capable() so that it can be called directly by drivers (for example, tg3 needs that). * Move the state choosing part of pci_prepare_to_sleep() to a separate function, pci_target_state(), that can be called directly by drivers (for example, tg3 needs that). Signed-off-by: Rafael J. Wysocki Signed-off-by: Jesse Barnes --- drivers/pci/pci.c | 34 ++++++++++++++++++++++++---------- include/linux/pci.h | 2 ++ 2 files changed, 26 insertions(+), 10 deletions(-) diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index d00f0e0d8453..e9c356236d27 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -1040,7 +1040,7 @@ int pci_set_pcie_reset_state(struct pci_dev *dev, enum pcie_reset_state state) * @dev: PCI device to handle. * @state: PCI state from which device will issue PME#. */ -static bool pci_pme_capable(struct pci_dev *dev, pci_power_t state) +bool pci_pme_capable(struct pci_dev *dev, pci_power_t state) { if (!dev->pm_cap) return false; @@ -1123,17 +1123,10 @@ int pci_enable_wake(struct pci_dev *dev, pci_power_t state, int enable) } /** - * pci_prepare_to_sleep - prepare PCI device for system-wide transition into a sleep state - * @dev: Device to handle. - * - * Choose the power state appropriate for the device depending on whether - * it can wake up the system and/or is power manageable by the platform - * (PCI_D3hot is the default) and put the device into that state. */ -int pci_prepare_to_sleep(struct pci_dev *dev) +pci_power_t pci_target_state(struct pci_dev *dev) { pci_power_t target_state = PCI_D3hot; - int error; if (platform_pci_power_manageable(dev)) { /* @@ -1160,7 +1153,7 @@ int pci_prepare_to_sleep(struct pci_dev *dev) * to generate PME#. */ if (!dev->pm_cap) - return -EIO; + return PCI_POWER_ERROR; if (dev->pme_support) { while (target_state @@ -1169,6 +1162,25 @@ int pci_prepare_to_sleep(struct pci_dev *dev) } } + return target_state; +} + +/** + * pci_prepare_to_sleep - prepare PCI device for system-wide transition into a sleep state + * @dev: Device to handle. + * + * Choose the power state appropriate for the device depending on whether + * it can wake up the system and/or is power manageable by the platform + * (PCI_D3hot is the default) and put the device into that state. + */ +int pci_prepare_to_sleep(struct pci_dev *dev) +{ + pci_power_t target_state = pci_target_state(dev); + int error; + + if (target_state == PCI_POWER_ERROR) + return -EIO; + pci_enable_wake(dev, target_state, true); error = pci_set_power_state(dev, target_state); @@ -1918,7 +1930,9 @@ EXPORT_SYMBOL(pci_select_bars); EXPORT_SYMBOL(pci_set_power_state); EXPORT_SYMBOL(pci_save_state); EXPORT_SYMBOL(pci_restore_state); +EXPORT_SYMBOL(pci_pme_capable); EXPORT_SYMBOL(pci_enable_wake); +EXPORT_SYMBOL(pci_target_state); EXPORT_SYMBOL(pci_prepare_to_sleep); EXPORT_SYMBOL(pci_back_from_sleep); EXPORT_SYMBOL_GPL(pci_set_pcie_reset_state); diff --git a/include/linux/pci.h b/include/linux/pci.h index a6a088e1a804..1d296d31abe0 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -638,7 +638,9 @@ int pci_save_state(struct pci_dev *dev); int pci_restore_state(struct pci_dev *dev); int pci_set_power_state(struct pci_dev *dev, pci_power_t state); pci_power_t pci_choose_state(struct pci_dev *dev, pm_message_t state); +bool pci_pme_capable(struct pci_dev *dev, pci_power_t state); int pci_enable_wake(struct pci_dev *dev, pci_power_t state, int enable); +pci_power_t pci_target_state(struct pci_dev *dev); int pci_prepare_to_sleep(struct pci_dev *dev); int pci_back_from_sleep(struct pci_dev *dev); From f17a077e61b627e58db5926bc474cf308318dad9 Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Tue, 22 Jul 2008 14:40:47 -0700 Subject: [PATCH 043/857] PCI: fixup sparse endianness warnings in proc.c drivers/pci/proc.c:91:3: warning: cast from restricted __le16 drivers/pci/proc.c:100:3: warning: cast from restricted __le32 drivers/pci/proc.c:109:3: warning: cast from restricted __le16 drivers/pci/proc.c:161:40: warning: cast to restricted __le16 drivers/pci/proc.c:170:41: warning: cast to restricted __le32 drivers/pci/proc.c:179:40: warning: cast to restricted __le16 Signed-off-by: Harvey Harrison Signed-off-by: Jesse Barnes --- drivers/pci/proc.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) diff --git a/drivers/pci/proc.c b/drivers/pci/proc.c index 4400dffbd93a..e1098c302c45 100644 --- a/drivers/pci/proc.c +++ b/drivers/pci/proc.c @@ -88,7 +88,7 @@ proc_bus_pci_read(struct file *file, char __user *buf, size_t nbytes, loff_t *pp if ((pos & 3) && cnt > 2) { unsigned short val; pci_user_read_config_word(dev, pos, &val); - __put_user(cpu_to_le16(val), (unsigned short __user *) buf); + __put_user(cpu_to_le16(val), (__le16 __user *) buf); buf += 2; pos += 2; cnt -= 2; @@ -97,7 +97,7 @@ proc_bus_pci_read(struct file *file, char __user *buf, size_t nbytes, loff_t *pp while (cnt >= 4) { unsigned int val; pci_user_read_config_dword(dev, pos, &val); - __put_user(cpu_to_le32(val), (unsigned int __user *) buf); + __put_user(cpu_to_le32(val), (__le32 __user *) buf); buf += 4; pos += 4; cnt -= 4; @@ -106,7 +106,7 @@ proc_bus_pci_read(struct file *file, char __user *buf, size_t nbytes, loff_t *pp if (cnt >= 2) { unsigned short val; pci_user_read_config_word(dev, pos, &val); - __put_user(cpu_to_le16(val), (unsigned short __user *) buf); + __put_user(cpu_to_le16(val), (__le16 __user *) buf); buf += 2; pos += 2; cnt -= 2; @@ -156,8 +156,8 @@ proc_bus_pci_write(struct file *file, const char __user *buf, size_t nbytes, lof } if ((pos & 3) && cnt > 2) { - unsigned short val; - __get_user(val, (unsigned short __user *) buf); + __le16 val; + __get_user(val, (__le16 __user *) buf); pci_user_write_config_word(dev, pos, le16_to_cpu(val)); buf += 2; pos += 2; @@ -165,8 +165,8 @@ proc_bus_pci_write(struct file *file, const char __user *buf, size_t nbytes, lof } while (cnt >= 4) { - unsigned int val; - __get_user(val, (unsigned int __user *) buf); + __le32 val; + __get_user(val, (__le32 __user *) buf); pci_user_write_config_dword(dev, pos, le32_to_cpu(val)); buf += 4; pos += 4; @@ -174,8 +174,8 @@ proc_bus_pci_write(struct file *file, const char __user *buf, size_t nbytes, lof } if (cnt >= 2) { - unsigned short val; - __get_user(val, (unsigned short __user *) buf); + __le16 val; + __get_user(val, (__le16 __user *) buf); pci_user_write_config_word(dev, pos, le16_to_cpu(val)); buf += 2; pos += 2; From 9bcab8405c98c34849c5795c717b7e6a3e2d3875 Mon Sep 17 00:00:00 2001 From: Milton Miller Date: Fri, 11 Jul 2008 09:03:09 +1000 Subject: [PATCH 044/857] powerpc/spufs: correct kcalloc usage kcalloc is supposed to be called with the count as its first argument and the element size as the second. Signed-off-by: Milton Miller Signed-off-by: Jeremy Kerr --- arch/powerpc/platforms/cell/spufs/sputrace.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/arch/powerpc/platforms/cell/spufs/sputrace.c b/arch/powerpc/platforms/cell/spufs/sputrace.c index 8c0e95766a62..92d20e993ede 100644 --- a/arch/powerpc/platforms/cell/spufs/sputrace.c +++ b/arch/powerpc/platforms/cell/spufs/sputrace.c @@ -196,8 +196,7 @@ static int __init sputrace_init(void) struct proc_dir_entry *entry; int i, error = -ENOMEM; - sputrace_log = kcalloc(sizeof(struct sputrace), - bufsize, GFP_KERNEL); + sputrace_log = kcalloc(bufsize, sizeof(struct sputrace), GFP_KERNEL); if (!sputrace_log) goto out; From 8a6d2ea0cd121e3bfff4dbce5bc111874cf9e9d2 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 22 Jul 2008 21:53:40 -0700 Subject: [PATCH 045/857] sky2: don't stop queue on shutdown It is unnecessary, to stop queue and turn off carrier in shutdown routine. With new netdev_queue this causes warnings. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- drivers/net/sky2.c | 5 ----- 1 file changed, 5 deletions(-) diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index 711e4a8948e0..5257cf464f1a 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -1829,9 +1829,6 @@ static int sky2_down(struct net_device *dev) if (netif_msg_ifdown(sky2)) printk(KERN_INFO PFX "%s: disabling interface\n", dev->name); - /* Stop more packets from being queued */ - netif_stop_queue(dev); - /* Disable port IRQ */ imask = sky2_read32(hw, B0_IMSK); imask &= ~portirq_msk[port]; @@ -1887,8 +1884,6 @@ static int sky2_down(struct net_device *dev) sky2_phy_power_down(hw, port); - netif_carrier_off(dev); - /* turn off LED's */ sky2_write16(hw, B0_Y2LED, LED_STAT_OFF); From deca05c3e81df4fcc38aa891eb8d8add14bce68b Mon Sep 17 00:00:00 2001 From: Greg Ungerer Date: Wed, 4 Jun 2008 21:20:32 +1000 Subject: [PATCH 046/857] m68knommu: change to a configs directory for board configurations Remove the old example m68knommu defconfig. Create a configs directory for specific board configurations. Make the m5208evb the default. Signed-off-by: Greg Ungerer --- arch/m68knommu/configs/m5208evb_defconfig | 610 ++++++++++++++++++++++ 1 file changed, 610 insertions(+) create mode 100644 arch/m68knommu/configs/m5208evb_defconfig diff --git a/arch/m68knommu/configs/m5208evb_defconfig b/arch/m68knommu/configs/m5208evb_defconfig new file mode 100644 index 000000000000..6fae33a05e2a --- /dev/null +++ b/arch/m68knommu/configs/m5208evb_defconfig @@ -0,0 +1,610 @@ +# +# Automatically generated make config: don't edit +# Linux kernel version: 2.6.26-rc1 +# +CONFIG_M68K=y +# CONFIG_MMU is not set +# CONFIG_FPU is not set +CONFIG_ZONE_DMA=y +CONFIG_RWSEM_GENERIC_SPINLOCK=y +# CONFIG_RWSEM_XCHGADD_ALGORITHM is not set +# CONFIG_ARCH_HAS_ILOG2_U32 is not set +# CONFIG_ARCH_HAS_ILOG2_U64 is not set +CONFIG_GENERIC_FIND_NEXT_BIT=y +CONFIG_GENERIC_HWEIGHT=y +CONFIG_GENERIC_HARDIRQS=y +CONFIG_GENERIC_CALIBRATE_DELAY=y +CONFIG_GENERIC_TIME=y +CONFIG_TIME_LOW_RES=y +CONFIG_NO_IOPORT=y +CONFIG_ARCH_SUPPORTS_AOUT=y +CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config" + +# +# General setup +# +CONFIG_EXPERIMENTAL=y +CONFIG_BROKEN_ON_SMP=y +CONFIG_INIT_ENV_ARG_LIMIT=32 +CONFIG_LOCALVERSION="" +CONFIG_LOCALVERSION_AUTO=y +# CONFIG_SYSVIPC is not set +# CONFIG_POSIX_MQUEUE is not set +# CONFIG_BSD_PROCESS_ACCT is not set +# CONFIG_TASKSTATS is not set +# CONFIG_AUDIT is not set +# CONFIG_IKCONFIG is not set +CONFIG_LOG_BUF_SHIFT=14 +# CONFIG_CGROUPS is not set +# CONFIG_GROUP_SCHED is not set +# CONFIG_RELAY is not set +# CONFIG_NAMESPACES is not set +# CONFIG_BLK_DEV_INITRD is not set +# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set +CONFIG_SYSCTL=y +CONFIG_EMBEDDED=y +# CONFIG_UID16 is not set +# CONFIG_SYSCTL_SYSCALL is not set +# CONFIG_KALLSYMS is not set +# CONFIG_HOTPLUG is not set +CONFIG_PRINTK=y +CONFIG_BUG=y +CONFIG_ELF_CORE=y +# CONFIG_COMPAT_BRK is not set +CONFIG_BASE_FULL=y +# CONFIG_FUTEX is not set +# CONFIG_EPOLL is not set +# CONFIG_SIGNALFD is not set +# CONFIG_TIMERFD is not set +# CONFIG_EVENTFD is not set +# CONFIG_VM_EVENT_COUNTERS is not set +CONFIG_SLAB=y +# CONFIG_SLUB is not set +# CONFIG_SLOB is not set +# CONFIG_PROFILING is not set +# CONFIG_MARKERS is not set +# CONFIG_HAVE_OPROFILE is not set +# CONFIG_HAVE_KPROBES is not set +# CONFIG_HAVE_KRETPROBES is not set +# CONFIG_HAVE_DMA_ATTRS is not set +CONFIG_SLABINFO=y +CONFIG_TINY_SHMEM=y +CONFIG_BASE_SMALL=0 +CONFIG_MODULES=y +CONFIG_MODULE_UNLOAD=y +# CONFIG_MODULE_FORCE_UNLOAD is not set +# CONFIG_MODVERSIONS is not set +# CONFIG_MODULE_SRCVERSION_ALL is not set +# CONFIG_KMOD is not set +CONFIG_BLOCK=y +# CONFIG_LBD is not set +# CONFIG_LSF is not set +# CONFIG_BLK_DEV_BSG is not set + +# +# IO Schedulers +# +CONFIG_IOSCHED_NOOP=y +# CONFIG_IOSCHED_AS is not set +# CONFIG_IOSCHED_DEADLINE is not set +# CONFIG_IOSCHED_CFQ is not set +# CONFIG_DEFAULT_AS is not set +# CONFIG_DEFAULT_DEADLINE is not set +# CONFIG_DEFAULT_CFQ is not set +CONFIG_DEFAULT_NOOP=y +CONFIG_DEFAULT_IOSCHED="noop" +CONFIG_CLASSIC_RCU=y + +# +# Processor type and features +# +# CONFIG_M68328 is not set +# CONFIG_M68EZ328 is not set +# CONFIG_M68VZ328 is not set +# CONFIG_M68360 is not set +# CONFIG_M5206 is not set +# CONFIG_M5206e is not set +CONFIG_M520x=y +# CONFIG_M523x is not set +# CONFIG_M5249 is not set +# CONFIG_M5271 is not set +# CONFIG_M5272 is not set +# CONFIG_M5275 is not set +# CONFIG_M528x is not set +# CONFIG_M5307 is not set +# CONFIG_M532x is not set +# CONFIG_M5407 is not set +CONFIG_COLDFIRE=y +CONFIG_CLOCK_SET=y +CONFIG_CLOCK_FREQ=166666666 +CONFIG_CLOCK_DIV=2 + +# +# Platform +# +CONFIG_M5208EVB=y +CONFIG_FREESCALE=y +# CONFIG_4KSTACKS is not set +CONFIG_HZ=100 + +# +# RAM configuration +# +CONFIG_RAMBASE=0x40000000 +CONFIG_RAMSIZE=0x2000000 +CONFIG_VECTORBASE=0x40000000 +CONFIG_KERNELBASE=0x40020000 +# CONFIG_RAMAUTOBIT is not set +# CONFIG_RAM8BIT is not set +CONFIG_RAM16BIT=y +# CONFIG_RAM32BIT is not set + +# +# ROM configuration +# +# CONFIG_ROM is not set +CONFIG_RAMKERNEL=y +# CONFIG_ROMKERNEL is not set +CONFIG_SELECT_MEMORY_MODEL=y +CONFIG_FLATMEM_MANUAL=y +# CONFIG_DISCONTIGMEM_MANUAL is not set +# CONFIG_SPARSEMEM_MANUAL is not set +CONFIG_FLATMEM=y +CONFIG_FLAT_NODE_MEM_MAP=y +# CONFIG_SPARSEMEM_STATIC is not set +# CONFIG_SPARSEMEM_VMEMMAP_ENABLE is not set +CONFIG_PAGEFLAGS_EXTENDED=y +CONFIG_SPLIT_PTLOCK_CPUS=4 +# CONFIG_RESOURCES_64BIT is not set +CONFIG_ZONE_DMA_FLAG=1 +CONFIG_VIRT_TO_BUS=y +CONFIG_ISA_DMA_API=y + +# +# Bus options (PCI, PCMCIA, EISA, MCA, ISA) +# +# CONFIG_PCI is not set +# CONFIG_ARCH_SUPPORTS_MSI is not set + +# +# Executable file formats +# +CONFIG_BINFMT_FLAT=y +# CONFIG_BINFMT_ZFLAT is not set +# CONFIG_BINFMT_SHARED_FLAT is not set +# CONFIG_BINFMT_AOUT is not set +# CONFIG_BINFMT_MISC is not set + +# +# Power management options +# +# CONFIG_PM is not set + +# +# Networking +# +CONFIG_NET=y + +# +# Networking options +# +CONFIG_PACKET=y +# CONFIG_PACKET_MMAP is not set +CONFIG_UNIX=y +# CONFIG_NET_KEY is not set +CONFIG_INET=y +# CONFIG_IP_MULTICAST is not set +# CONFIG_IP_ADVANCED_ROUTER is not set +CONFIG_IP_FIB_HASH=y +# CONFIG_IP_PNP is not set +# CONFIG_NET_IPIP is not set +# CONFIG_NET_IPGRE is not set +# CONFIG_ARPD is not set +# CONFIG_SYN_COOKIES is not set +# CONFIG_INET_AH is not set +# CONFIG_INET_ESP is not set +# CONFIG_INET_IPCOMP is not set +# CONFIG_INET_XFRM_TUNNEL is not set +# CONFIG_INET_TUNNEL is not set +# CONFIG_INET_XFRM_MODE_TRANSPORT is not set +# CONFIG_INET_XFRM_MODE_TUNNEL is not set +# CONFIG_INET_XFRM_MODE_BEET is not set +# CONFIG_INET_LRO is not set +# CONFIG_INET_DIAG is not set +# CONFIG_TCP_CONG_ADVANCED is not set +CONFIG_TCP_CONG_CUBIC=y +CONFIG_DEFAULT_TCP_CONG="cubic" +# CONFIG_TCP_MD5SIG is not set +# CONFIG_IPV6 is not set +# CONFIG_NETWORK_SECMARK is not set +# CONFIG_NETFILTER is not set +# CONFIG_IP_DCCP is not set +# CONFIG_IP_SCTP is not set +# CONFIG_TIPC is not set +# CONFIG_ATM is not set +# CONFIG_BRIDGE is not set +# CONFIG_VLAN_8021Q is not set +# CONFIG_DECNET is not set +# CONFIG_LLC2 is not set +# CONFIG_IPX is not set +# CONFIG_ATALK is not set +# CONFIG_X25 is not set +# CONFIG_LAPB is not set +# CONFIG_ECONET is not set +# CONFIG_WAN_ROUTER is not set +# CONFIG_NET_SCHED is not set + +# +# Network testing +# +# CONFIG_NET_PKTGEN is not set +# CONFIG_HAMRADIO is not set +# CONFIG_CAN is not set +# CONFIG_IRDA is not set +# CONFIG_BT is not set +# CONFIG_AF_RXRPC is not set + +# +# Wireless +# +# CONFIG_CFG80211 is not set +# CONFIG_WIRELESS_EXT is not set +# CONFIG_MAC80211 is not set +# CONFIG_IEEE80211 is not set +# CONFIG_RFKILL is not set +# CONFIG_NET_9P is not set + +# +# Device Drivers +# + +# +# Generic Driver Options +# +CONFIG_STANDALONE=y +CONFIG_PREVENT_FIRMWARE_BUILD=y +# CONFIG_SYS_HYPERVISOR is not set +# CONFIG_CONNECTOR is not set +CONFIG_MTD=y +# CONFIG_MTD_DEBUG is not set +# CONFIG_MTD_CONCAT is not set +CONFIG_MTD_PARTITIONS=y +# CONFIG_MTD_REDBOOT_PARTS is not set +# CONFIG_MTD_CMDLINE_PARTS is not set +# CONFIG_MTD_AR7_PARTS is not set + +# +# User Modules And Translation Layers +# +CONFIG_MTD_CHAR=y +CONFIG_MTD_BLKDEVS=y +CONFIG_MTD_BLOCK=y +# CONFIG_FTL is not set +# CONFIG_NFTL is not set +# CONFIG_INFTL is not set +# CONFIG_RFD_FTL is not set +# CONFIG_SSFDC is not set +# CONFIG_MTD_OOPS is not set + +# +# RAM/ROM/Flash chip drivers +# +CONFIG_MTD_CFI=y +# CONFIG_MTD_JEDECPROBE is not set +CONFIG_MTD_GEN_PROBE=y +# CONFIG_MTD_CFI_ADV_OPTIONS is not set +CONFIG_MTD_MAP_BANK_WIDTH_1=y +CONFIG_MTD_MAP_BANK_WIDTH_2=y +CONFIG_MTD_MAP_BANK_WIDTH_4=y +# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set +# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set +# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set +CONFIG_MTD_CFI_I1=y +CONFIG_MTD_CFI_I2=y +# CONFIG_MTD_CFI_I4 is not set +# CONFIG_MTD_CFI_I8 is not set +# CONFIG_MTD_CFI_INTELEXT is not set +CONFIG_MTD_CFI_AMDSTD=y +# CONFIG_MTD_CFI_STAA is not set +CONFIG_MTD_CFI_UTIL=y +CONFIG_MTD_RAM=y +# CONFIG_MTD_ROM is not set +# CONFIG_MTD_ABSENT is not set + +# +# Mapping drivers for chip access +# +# CONFIG_MTD_COMPLEX_MAPPINGS is not set +# CONFIG_MTD_PHYSMAP is not set +CONFIG_MTD_UCLINUX=y +# CONFIG_MTD_PLATRAM is not set + +# +# Self-contained MTD device drivers +# +# CONFIG_MTD_SLRAM is not set +# CONFIG_MTD_PHRAM is not set +# CONFIG_MTD_MTDRAM is not set +# CONFIG_MTD_BLOCK2MTD is not set + +# +# Disk-On-Chip Device Drivers +# +# CONFIG_MTD_DOC2000 is not set +# CONFIG_MTD_DOC2001 is not set +# CONFIG_MTD_DOC2001PLUS is not set +# CONFIG_MTD_NAND is not set +# CONFIG_MTD_ONENAND is not set + +# +# UBI - Unsorted block images +# +# CONFIG_MTD_UBI is not set +# CONFIG_PARPORT is not set +CONFIG_BLK_DEV=y +# CONFIG_BLK_DEV_COW_COMMON is not set +# CONFIG_BLK_DEV_LOOP is not set +# CONFIG_BLK_DEV_NBD is not set +CONFIG_BLK_DEV_RAM=y +CONFIG_BLK_DEV_RAM_COUNT=16 +CONFIG_BLK_DEV_RAM_SIZE=4096 +# CONFIG_BLK_DEV_XIP is not set +# CONFIG_CDROM_PKTCDVD is not set +# CONFIG_ATA_OVER_ETH is not set +# CONFIG_MISC_DEVICES is not set +CONFIG_HAVE_IDE=y +# CONFIG_IDE is not set + +# +# SCSI device support +# +# CONFIG_RAID_ATTRS is not set +# CONFIG_SCSI is not set +# CONFIG_SCSI_DMA is not set +# CONFIG_SCSI_NETLINK is not set +# CONFIG_MD is not set +CONFIG_NETDEVICES=y +# CONFIG_NETDEVICES_MULTIQUEUE is not set +# CONFIG_DUMMY is not set +# CONFIG_BONDING is not set +# CONFIG_MACVLAN is not set +# CONFIG_EQUALIZER is not set +# CONFIG_TUN is not set +# CONFIG_VETH is not set +# CONFIG_PHYLIB is not set +CONFIG_NET_ETHERNET=y +# CONFIG_MII is not set +# CONFIG_IBM_NEW_EMAC_ZMII is not set +# CONFIG_IBM_NEW_EMAC_RGMII is not set +# CONFIG_IBM_NEW_EMAC_TAH is not set +# CONFIG_IBM_NEW_EMAC_EMAC4 is not set +# CONFIG_B44 is not set +CONFIG_FEC=y +# CONFIG_FEC2 is not set +# CONFIG_NETDEV_1000 is not set +# CONFIG_NETDEV_10000 is not set + +# +# Wireless LAN +# +# CONFIG_WLAN_PRE80211 is not set +# CONFIG_WLAN_80211 is not set +# CONFIG_IWLWIFI is not set +# CONFIG_IWLWIFI_LEDS is not set +# CONFIG_WAN is not set +# CONFIG_PPP is not set +# CONFIG_SLIP is not set +# CONFIG_NETCONSOLE is not set +# CONFIG_NETPOLL is not set +# CONFIG_NET_POLL_CONTROLLER is not set +# CONFIG_ISDN is not set +# CONFIG_PHONE is not set + +# +# Input device support +# +# CONFIG_INPUT is not set + +# +# Hardware I/O ports +# +# CONFIG_SERIO is not set +# CONFIG_GAMEPORT is not set + +# +# Character devices +# +# CONFIG_VT is not set +# CONFIG_DEVKMEM is not set +# CONFIG_SERIAL_NONSTANDARD is not set + +# +# Serial drivers +# +# CONFIG_SERIAL_8250 is not set + +# +# Non-8250 serial port support +# +CONFIG_SERIAL_CORE=y +CONFIG_SERIAL_CORE_CONSOLE=y +# CONFIG_SERIAL_COLDFIRE is not set +CONFIG_SERIAL_MCF=y +CONFIG_SERIAL_MCF_BAUDRATE=115200 +CONFIG_SERIAL_MCF_CONSOLE=y +# CONFIG_UNIX98_PTYS is not set +CONFIG_LEGACY_PTYS=y +CONFIG_LEGACY_PTY_COUNT=256 +# CONFIG_IPMI_HANDLER is not set +# CONFIG_HW_RANDOM is not set +# CONFIG_GEN_RTC is not set +# CONFIG_R3964 is not set +# CONFIG_RAW_DRIVER is not set +# CONFIG_TCG_TPM is not set +# CONFIG_I2C is not set +# CONFIG_SPI is not set +# CONFIG_W1 is not set +# CONFIG_POWER_SUPPLY is not set +# CONFIG_HWMON is not set +# CONFIG_THERMAL is not set +# CONFIG_WATCHDOG is not set + +# +# Sonics Silicon Backplane +# +CONFIG_SSB_POSSIBLE=y +# CONFIG_SSB is not set + +# +# Multifunction device drivers +# +# CONFIG_MFD_SM501 is not set +# CONFIG_HTC_PASIC3 is not set + +# +# Multimedia devices +# + +# +# Multimedia core support +# +# CONFIG_VIDEO_DEV is not set +# CONFIG_DVB_CORE is not set + +# +# Multimedia drivers +# +# CONFIG_DAB is not set + +# +# Graphics support +# +# CONFIG_VGASTATE is not set +# CONFIG_VIDEO_OUTPUT_CONTROL is not set +# CONFIG_FB is not set +# CONFIG_BACKLIGHT_LCD_SUPPORT is not set + +# +# Display device support +# +# CONFIG_DISPLAY_SUPPORT is not set + +# +# Sound +# +# CONFIG_SOUND is not set +# CONFIG_USB_SUPPORT is not set +# CONFIG_MMC is not set +# CONFIG_MEMSTICK is not set +# CONFIG_NEW_LEDS is not set +# CONFIG_ACCESSIBILITY is not set +# CONFIG_RTC_CLASS is not set +# CONFIG_UIO is not set + +# +# File systems +# +CONFIG_EXT2_FS=y +# CONFIG_EXT2_FS_XATTR is not set +# CONFIG_EXT3_FS is not set +# CONFIG_EXT4DEV_FS is not set +# CONFIG_REISERFS_FS is not set +# CONFIG_JFS_FS is not set +# CONFIG_FS_POSIX_ACL is not set +# CONFIG_XFS_FS is not set +# CONFIG_DNOTIFY is not set +# CONFIG_INOTIFY is not set +# CONFIG_QUOTA is not set +# CONFIG_AUTOFS_FS is not set +# CONFIG_AUTOFS4_FS is not set +# CONFIG_FUSE_FS is not set + +# +# CD-ROM/DVD Filesystems +# +# CONFIG_ISO9660_FS is not set +# CONFIG_UDF_FS is not set + +# +# DOS/FAT/NT Filesystems +# +# CONFIG_MSDOS_FS is not set +# CONFIG_VFAT_FS is not set +# CONFIG_NTFS_FS is not set + +# +# Pseudo filesystems +# +CONFIG_PROC_FS=y +CONFIG_PROC_SYSCTL=y +# CONFIG_SYSFS is not set +# CONFIG_TMPFS is not set +# CONFIG_HUGETLB_PAGE is not set + +# +# Miscellaneous filesystems +# +# CONFIG_ADFS_FS is not set +# CONFIG_AFFS_FS is not set +# CONFIG_HFS_FS is not set +# CONFIG_HFSPLUS_FS is not set +# CONFIG_BEFS_FS is not set +# CONFIG_BFS_FS is not set +# CONFIG_EFS_FS is not set +# CONFIG_JFFS2_FS is not set +# CONFIG_CRAMFS is not set +# CONFIG_VXFS_FS is not set +# CONFIG_MINIX_FS is not set +# CONFIG_HPFS_FS is not set +# CONFIG_QNX4FS_FS is not set +CONFIG_ROMFS_FS=y +# CONFIG_SYSV_FS is not set +# CONFIG_UFS_FS is not set +# CONFIG_NETWORK_FILESYSTEMS is not set + +# +# Partition Types +# +# CONFIG_PARTITION_ADVANCED is not set +CONFIG_MSDOS_PARTITION=y +# CONFIG_NLS is not set + +# +# Kernel hacking +# +# CONFIG_PRINTK_TIME is not set +CONFIG_ENABLE_WARN_DEPRECATED=y +CONFIG_ENABLE_MUST_CHECK=y +CONFIG_FRAME_WARN=1024 +# CONFIG_MAGIC_SYSRQ is not set +# CONFIG_UNUSED_SYMBOLS is not set +# CONFIG_HEADERS_CHECK is not set +# CONFIG_DEBUG_KERNEL is not set +# CONFIG_DEBUG_BUGVERBOSE is not set +# CONFIG_SAMPLES is not set +CONFIG_FULLDEBUG=y +# CONFIG_HIGHPROFILE is not set +# CONFIG_BOOTPARAM is not set +# CONFIG_NO_KERNEL_MSG is not set +# CONFIG_BDM_DISABLE is not set + +# +# Security options +# +# CONFIG_KEYS is not set +# CONFIG_SECURITY_FILE_CAPABILITIES is not set +# CONFIG_CRYPTO is not set + +# +# Library routines +# +CONFIG_BITREVERSE=y +# CONFIG_GENERIC_FIND_FIRST_BIT is not set +# CONFIG_CRC_CCITT is not set +# CONFIG_CRC16 is not set +# CONFIG_CRC_ITU_T is not set +CONFIG_CRC32=y +# CONFIG_CRC7 is not set +# CONFIG_LIBCRC32C is not set +CONFIG_HAS_IOMEM=y +CONFIG_HAS_DMA=y From 8c81b0574fd8877b5214f2d33816b199c62e3335 Mon Sep 17 00:00:00 2001 From: Greg Ungerer Date: Wed, 4 Jun 2008 21:22:18 +1000 Subject: [PATCH 047/857] m68knommu: defconfig for M5249EVB board Add a defconfig for the Freescale M5249EVB board. Signed-off-by: Greg Ungerer --- arch/m68knommu/configs/m5249evb_defconfig | 497 ++++++++++++++++++++++ 1 file changed, 497 insertions(+) create mode 100644 arch/m68knommu/configs/m5249evb_defconfig diff --git a/arch/m68knommu/configs/m5249evb_defconfig b/arch/m68knommu/configs/m5249evb_defconfig new file mode 100644 index 000000000000..cc6458333d67 --- /dev/null +++ b/arch/m68knommu/configs/m5249evb_defconfig @@ -0,0 +1,497 @@ +# +# Automatically generated make config: don't edit +# Linux kernel version: 2.6.26-rc1 +# +CONFIG_M68K=y +# CONFIG_MMU is not set +# CONFIG_FPU is not set +CONFIG_ZONE_DMA=y +CONFIG_RWSEM_GENERIC_SPINLOCK=y +# CONFIG_RWSEM_XCHGADD_ALGORITHM is not set +# CONFIG_ARCH_HAS_ILOG2_U32 is not set +# CONFIG_ARCH_HAS_ILOG2_U64 is not set +CONFIG_GENERIC_FIND_NEXT_BIT=y +CONFIG_GENERIC_HWEIGHT=y +CONFIG_GENERIC_HARDIRQS=y +CONFIG_GENERIC_CALIBRATE_DELAY=y +CONFIG_GENERIC_TIME=y +CONFIG_TIME_LOW_RES=y +CONFIG_NO_IOPORT=y +CONFIG_ARCH_SUPPORTS_AOUT=y +CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config" + +# +# General setup +# +CONFIG_EXPERIMENTAL=y +CONFIG_BROKEN_ON_SMP=y +CONFIG_INIT_ENV_ARG_LIMIT=32 +CONFIG_LOCALVERSION="" +CONFIG_LOCALVERSION_AUTO=y +# CONFIG_SYSVIPC is not set +# CONFIG_BSD_PROCESS_ACCT is not set +# CONFIG_IKCONFIG is not set +CONFIG_LOG_BUF_SHIFT=14 +# CONFIG_CGROUPS is not set +# CONFIG_GROUP_SCHED is not set +# CONFIG_SYSFS_DEPRECATED_V2 is not set +# CONFIG_RELAY is not set +# CONFIG_NAMESPACES is not set +# CONFIG_BLK_DEV_INITRD is not set +# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set +CONFIG_SYSCTL=y +CONFIG_EMBEDDED=y +# CONFIG_UID16 is not set +# CONFIG_SYSCTL_SYSCALL is not set +# CONFIG_KALLSYMS is not set +# CONFIG_HOTPLUG is not set +CONFIG_PRINTK=y +CONFIG_BUG=y +CONFIG_ELF_CORE=y +# CONFIG_COMPAT_BRK is not set +CONFIG_BASE_FULL=y +# CONFIG_FUTEX is not set +# CONFIG_EPOLL is not set +# CONFIG_SIGNALFD is not set +# CONFIG_TIMERFD is not set +# CONFIG_EVENTFD is not set +# CONFIG_VM_EVENT_COUNTERS is not set +CONFIG_SLAB=y +# CONFIG_SLUB is not set +# CONFIG_SLOB is not set +# CONFIG_PROFILING is not set +# CONFIG_MARKERS is not set +# CONFIG_HAVE_OPROFILE is not set +# CONFIG_HAVE_KPROBES is not set +# CONFIG_HAVE_KRETPROBES is not set +# CONFIG_HAVE_DMA_ATTRS is not set +CONFIG_SLABINFO=y +CONFIG_TINY_SHMEM=y +CONFIG_BASE_SMALL=0 +CONFIG_MODULES=y +CONFIG_MODULE_UNLOAD=y +# CONFIG_MODULE_FORCE_UNLOAD is not set +# CONFIG_MODVERSIONS is not set +# CONFIG_MODULE_SRCVERSION_ALL is not set +# CONFIG_KMOD is not set +CONFIG_BLOCK=y +# CONFIG_LBD is not set +# CONFIG_BLK_DEV_IO_TRACE is not set +# CONFIG_LSF is not set +# CONFIG_BLK_DEV_BSG is not set + +# +# IO Schedulers +# +CONFIG_IOSCHED_NOOP=y +# CONFIG_IOSCHED_AS is not set +# CONFIG_IOSCHED_DEADLINE is not set +# CONFIG_IOSCHED_CFQ is not set +# CONFIG_DEFAULT_AS is not set +# CONFIG_DEFAULT_DEADLINE is not set +# CONFIG_DEFAULT_CFQ is not set +CONFIG_DEFAULT_NOOP=y +CONFIG_DEFAULT_IOSCHED="noop" +CONFIG_CLASSIC_RCU=y + +# +# Processor type and features +# +# CONFIG_M68328 is not set +# CONFIG_M68EZ328 is not set +# CONFIG_M68VZ328 is not set +# CONFIG_M68360 is not set +# CONFIG_M5206 is not set +# CONFIG_M5206e is not set +# CONFIG_M520x is not set +# CONFIG_M523x is not set +CONFIG_M5249=y +# CONFIG_M5271 is not set +# CONFIG_M5272 is not set +# CONFIG_M5275 is not set +# CONFIG_M528x is not set +# CONFIG_M5307 is not set +# CONFIG_M532x is not set +# CONFIG_M5407 is not set +CONFIG_COLDFIRE=y +CONFIG_CLOCK_SET=y +CONFIG_CLOCK_FREQ=140000000 +CONFIG_CLOCK_DIV=2 + +# +# Platform +# +CONFIG_M5249C3=y +CONFIG_FREESCALE=y +CONFIG_4KSTACKS=y +CONFIG_HZ=100 + +# +# RAM configuration +# +CONFIG_RAMBASE=0x00000000 +CONFIG_RAMSIZE=0x00800000 +CONFIG_VECTORBASE=0x00000000 +CONFIG_KERNELBASE=0x00020000 +CONFIG_RAMAUTOBIT=y +# CONFIG_RAM8BIT is not set +# CONFIG_RAM16BIT is not set +# CONFIG_RAM32BIT is not set + +# +# ROM configuration +# +# CONFIG_ROM is not set +CONFIG_RAMKERNEL=y +# CONFIG_ROMKERNEL is not set +CONFIG_SELECT_MEMORY_MODEL=y +CONFIG_FLATMEM_MANUAL=y +# CONFIG_DISCONTIGMEM_MANUAL is not set +# CONFIG_SPARSEMEM_MANUAL is not set +CONFIG_FLATMEM=y +CONFIG_FLAT_NODE_MEM_MAP=y +# CONFIG_SPARSEMEM_STATIC is not set +# CONFIG_SPARSEMEM_VMEMMAP_ENABLE is not set +CONFIG_PAGEFLAGS_EXTENDED=y +CONFIG_SPLIT_PTLOCK_CPUS=4 +# CONFIG_RESOURCES_64BIT is not set +CONFIG_ZONE_DMA_FLAG=1 +CONFIG_VIRT_TO_BUS=y +CONFIG_ISA_DMA_API=y + +# +# Bus options (PCI, PCMCIA, EISA, MCA, ISA) +# +# CONFIG_PCI is not set +# CONFIG_ARCH_SUPPORTS_MSI is not set + +# +# Executable file formats +# +CONFIG_BINFMT_FLAT=y +# CONFIG_BINFMT_ZFLAT is not set +# CONFIG_BINFMT_SHARED_FLAT is not set +# CONFIG_BINFMT_AOUT is not set +# CONFIG_BINFMT_MISC is not set + +# +# Power management options +# +# CONFIG_PM is not set + +# +# Networking +# +# CONFIG_NET is not set + +# +# Device Drivers +# + +# +# Generic Driver Options +# +CONFIG_STANDALONE=y +CONFIG_PREVENT_FIRMWARE_BUILD=y +# CONFIG_SYS_HYPERVISOR is not set +CONFIG_MTD=y +# CONFIG_MTD_DEBUG is not set +# CONFIG_MTD_CONCAT is not set +CONFIG_MTD_PARTITIONS=y +# CONFIG_MTD_REDBOOT_PARTS is not set +# CONFIG_MTD_CMDLINE_PARTS is not set +# CONFIG_MTD_AR7_PARTS is not set + +# +# User Modules And Translation Layers +# +CONFIG_MTD_CHAR=y +CONFIG_MTD_BLKDEVS=y +CONFIG_MTD_BLOCK=y +# CONFIG_FTL is not set +# CONFIG_NFTL is not set +# CONFIG_INFTL is not set +# CONFIG_RFD_FTL is not set +# CONFIG_SSFDC is not set +# CONFIG_MTD_OOPS is not set + +# +# RAM/ROM/Flash chip drivers +# +# CONFIG_MTD_CFI is not set +# CONFIG_MTD_JEDECPROBE is not set +CONFIG_MTD_MAP_BANK_WIDTH_1=y +CONFIG_MTD_MAP_BANK_WIDTH_2=y +CONFIG_MTD_MAP_BANK_WIDTH_4=y +# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set +# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set +# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set +CONFIG_MTD_CFI_I1=y +CONFIG_MTD_CFI_I2=y +# CONFIG_MTD_CFI_I4 is not set +# CONFIG_MTD_CFI_I8 is not set +CONFIG_MTD_RAM=y +# CONFIG_MTD_ROM is not set +# CONFIG_MTD_ABSENT is not set + +# +# Mapping drivers for chip access +# +# CONFIG_MTD_COMPLEX_MAPPINGS is not set +CONFIG_MTD_UCLINUX=y +# CONFIG_MTD_PLATRAM is not set + +# +# Self-contained MTD device drivers +# +# CONFIG_MTD_SLRAM is not set +# CONFIG_MTD_PHRAM is not set +# CONFIG_MTD_MTDRAM is not set +# CONFIG_MTD_BLOCK2MTD is not set + +# +# Disk-On-Chip Device Drivers +# +# CONFIG_MTD_DOC2000 is not set +# CONFIG_MTD_DOC2001 is not set +# CONFIG_MTD_DOC2001PLUS is not set +# CONFIG_MTD_NAND is not set +# CONFIG_MTD_ONENAND is not set + +# +# UBI - Unsorted block images +# +# CONFIG_MTD_UBI is not set +# CONFIG_PARPORT is not set +CONFIG_BLK_DEV=y +# CONFIG_BLK_DEV_COW_COMMON is not set +# CONFIG_BLK_DEV_LOOP is not set +CONFIG_BLK_DEV_RAM=y +CONFIG_BLK_DEV_RAM_COUNT=16 +CONFIG_BLK_DEV_RAM_SIZE=4096 +# CONFIG_BLK_DEV_XIP is not set +# CONFIG_CDROM_PKTCDVD is not set +CONFIG_MISC_DEVICES=y +# CONFIG_EEPROM_93CX6 is not set +# CONFIG_ENCLOSURE_SERVICES is not set +CONFIG_HAVE_IDE=y +# CONFIG_IDE is not set + +# +# SCSI device support +# +# CONFIG_RAID_ATTRS is not set +# CONFIG_SCSI is not set +# CONFIG_SCSI_DMA is not set +# CONFIG_SCSI_NETLINK is not set +# CONFIG_MD is not set +# CONFIG_PHONE is not set + +# +# Input device support +# +# CONFIG_INPUT is not set + +# +# Hardware I/O ports +# +# CONFIG_SERIO is not set +# CONFIG_GAMEPORT is not set + +# +# Character devices +# +# CONFIG_VT is not set +# CONFIG_DEVKMEM is not set +# CONFIG_SERIAL_NONSTANDARD is not set + +# +# Serial drivers +# +# CONFIG_SERIAL_8250 is not set + +# +# Non-8250 serial port support +# +CONFIG_SERIAL_CORE=y +CONFIG_SERIAL_CORE_CONSOLE=y +# CONFIG_SERIAL_COLDFIRE is not set +CONFIG_SERIAL_MCF=y +CONFIG_SERIAL_MCF_BAUDRATE=19200 +CONFIG_SERIAL_MCF_CONSOLE=y +# CONFIG_UNIX98_PTYS is not set +CONFIG_LEGACY_PTYS=y +CONFIG_LEGACY_PTY_COUNT=256 +# CONFIG_IPMI_HANDLER is not set +# CONFIG_HW_RANDOM is not set +# CONFIG_GEN_RTC is not set +# CONFIG_R3964 is not set +# CONFIG_RAW_DRIVER is not set +# CONFIG_TCG_TPM is not set +# CONFIG_I2C is not set +# CONFIG_SPI is not set +# CONFIG_W1 is not set +# CONFIG_POWER_SUPPLY is not set +# CONFIG_HWMON is not set +# CONFIG_THERMAL is not set +# CONFIG_WATCHDOG is not set + +# +# Sonics Silicon Backplane +# +CONFIG_SSB_POSSIBLE=y +# CONFIG_SSB is not set + +# +# Multifunction device drivers +# +# CONFIG_MFD_SM501 is not set +# CONFIG_HTC_PASIC3 is not set + +# +# Multimedia devices +# + +# +# Multimedia core support +# +# CONFIG_VIDEO_DEV is not set + +# +# Multimedia drivers +# +# CONFIG_DAB is not set + +# +# Graphics support +# +# CONFIG_VGASTATE is not set +# CONFIG_VIDEO_OUTPUT_CONTROL is not set +# CONFIG_FB is not set +# CONFIG_BACKLIGHT_LCD_SUPPORT is not set + +# +# Display device support +# +# CONFIG_DISPLAY_SUPPORT is not set + +# +# Sound +# +# CONFIG_SOUND is not set +# CONFIG_USB_SUPPORT is not set +# CONFIG_MMC is not set +# CONFIG_MEMSTICK is not set +# CONFIG_NEW_LEDS is not set +# CONFIG_ACCESSIBILITY is not set +# CONFIG_RTC_CLASS is not set +# CONFIG_UIO is not set + +# +# File systems +# +CONFIG_EXT2_FS=y +# CONFIG_EXT2_FS_XATTR is not set +# CONFIG_EXT3_FS is not set +# CONFIG_EXT4DEV_FS is not set +# CONFIG_REISERFS_FS is not set +# CONFIG_JFS_FS is not set +# CONFIG_FS_POSIX_ACL is not set +# CONFIG_XFS_FS is not set +# CONFIG_DNOTIFY is not set +# CONFIG_INOTIFY is not set +# CONFIG_QUOTA is not set +# CONFIG_AUTOFS_FS is not set +# CONFIG_AUTOFS4_FS is not set +# CONFIG_FUSE_FS is not set + +# +# CD-ROM/DVD Filesystems +# +# CONFIG_ISO9660_FS is not set +# CONFIG_UDF_FS is not set + +# +# DOS/FAT/NT Filesystems +# +# CONFIG_MSDOS_FS is not set +# CONFIG_VFAT_FS is not set +# CONFIG_NTFS_FS is not set + +# +# Pseudo filesystems +# +CONFIG_PROC_FS=y +CONFIG_PROC_SYSCTL=y +CONFIG_SYSFS=y +# CONFIG_TMPFS is not set +# CONFIG_HUGETLB_PAGE is not set +# CONFIG_CONFIGFS_FS is not set + +# +# Miscellaneous filesystems +# +# CONFIG_ADFS_FS is not set +# CONFIG_AFFS_FS is not set +# CONFIG_HFS_FS is not set +# CONFIG_HFSPLUS_FS is not set +# CONFIG_BEFS_FS is not set +# CONFIG_BFS_FS is not set +# CONFIG_EFS_FS is not set +# CONFIG_JFFS2_FS is not set +# CONFIG_CRAMFS is not set +# CONFIG_VXFS_FS is not set +# CONFIG_MINIX_FS is not set +# CONFIG_HPFS_FS is not set +# CONFIG_QNX4FS_FS is not set +CONFIG_ROMFS_FS=y +# CONFIG_SYSV_FS is not set +# CONFIG_UFS_FS is not set + +# +# Partition Types +# +# CONFIG_PARTITION_ADVANCED is not set +CONFIG_MSDOS_PARTITION=y +# CONFIG_NLS is not set + +# +# Kernel hacking +# +# CONFIG_PRINTK_TIME is not set +CONFIG_ENABLE_WARN_DEPRECATED=y +CONFIG_ENABLE_MUST_CHECK=y +CONFIG_FRAME_WARN=1024 +# CONFIG_MAGIC_SYSRQ is not set +# CONFIG_UNUSED_SYMBOLS is not set +# CONFIG_DEBUG_FS is not set +# CONFIG_HEADERS_CHECK is not set +# CONFIG_DEBUG_KERNEL is not set +# CONFIG_DEBUG_BUGVERBOSE is not set +# CONFIG_SAMPLES is not set +# CONFIG_FULLDEBUG is not set +# CONFIG_HIGHPROFILE is not set +# CONFIG_BOOTPARAM is not set +# CONFIG_NO_KERNEL_MSG is not set +# CONFIG_BDM_DISABLE is not set + +# +# Security options +# +# CONFIG_KEYS is not set +# CONFIG_SECURITY is not set +# CONFIG_SECURITY_FILE_CAPABILITIES is not set +# CONFIG_CRYPTO is not set + +# +# Library routines +# +# CONFIG_GENERIC_FIND_FIRST_BIT is not set +# CONFIG_CRC_CCITT is not set +# CONFIG_CRC16 is not set +# CONFIG_CRC_ITU_T is not set +# CONFIG_CRC32 is not set +# CONFIG_CRC7 is not set +# CONFIG_LIBCRC32C is not set +CONFIG_HAS_IOMEM=y +CONFIG_HAS_DMA=y From c26def07f97d662ee96362795400cbf6f44d11ce Mon Sep 17 00:00:00 2001 From: Greg Ungerer Date: Wed, 4 Jun 2008 21:23:43 +1000 Subject: [PATCH 048/857] m68knommu: defconfig for M5275EVB board Add a defconfig for the Freescale M5275EVB board. Signed-off-by: Greg Ungerer --- arch/m68knommu/configs/m5275evb_defconfig | 627 ++++++++++++++++++++++ 1 file changed, 627 insertions(+) create mode 100644 arch/m68knommu/configs/m5275evb_defconfig diff --git a/arch/m68knommu/configs/m5275evb_defconfig b/arch/m68knommu/configs/m5275evb_defconfig new file mode 100644 index 000000000000..0d1256f5addb --- /dev/null +++ b/arch/m68knommu/configs/m5275evb_defconfig @@ -0,0 +1,627 @@ +# +# Automatically generated make config: don't edit +# Linux kernel version: 2.6.26-rc1 +# +CONFIG_M68K=y +# CONFIG_MMU is not set +# CONFIG_FPU is not set +CONFIG_ZONE_DMA=y +CONFIG_RWSEM_GENERIC_SPINLOCK=y +# CONFIG_RWSEM_XCHGADD_ALGORITHM is not set +# CONFIG_ARCH_HAS_ILOG2_U32 is not set +# CONFIG_ARCH_HAS_ILOG2_U64 is not set +CONFIG_GENERIC_FIND_NEXT_BIT=y +CONFIG_GENERIC_HWEIGHT=y +CONFIG_GENERIC_HARDIRQS=y +CONFIG_GENERIC_CALIBRATE_DELAY=y +CONFIG_GENERIC_TIME=y +CONFIG_TIME_LOW_RES=y +CONFIG_NO_IOPORT=y +CONFIG_ARCH_SUPPORTS_AOUT=y +CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config" + +# +# General setup +# +CONFIG_EXPERIMENTAL=y +CONFIG_BROKEN_ON_SMP=y +CONFIG_INIT_ENV_ARG_LIMIT=32 +CONFIG_LOCALVERSION="" +CONFIG_LOCALVERSION_AUTO=y +# CONFIG_SYSVIPC is not set +# CONFIG_POSIX_MQUEUE is not set +# CONFIG_BSD_PROCESS_ACCT is not set +# CONFIG_TASKSTATS is not set +# CONFIG_AUDIT is not set +# CONFIG_IKCONFIG is not set +CONFIG_LOG_BUF_SHIFT=14 +# CONFIG_CGROUPS is not set +# CONFIG_GROUP_SCHED is not set +# CONFIG_SYSFS_DEPRECATED_V2 is not set +# CONFIG_RELAY is not set +# CONFIG_NAMESPACES is not set +# CONFIG_BLK_DEV_INITRD is not set +# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set +CONFIG_SYSCTL=y +CONFIG_EMBEDDED=y +# CONFIG_UID16 is not set +# CONFIG_SYSCTL_SYSCALL is not set +# CONFIG_KALLSYMS is not set +# CONFIG_HOTPLUG is not set +CONFIG_PRINTK=y +CONFIG_BUG=y +CONFIG_ELF_CORE=y +# CONFIG_COMPAT_BRK is not set +CONFIG_BASE_FULL=y +# CONFIG_FUTEX is not set +# CONFIG_EPOLL is not set +# CONFIG_SIGNALFD is not set +# CONFIG_TIMERFD is not set +# CONFIG_EVENTFD is not set +# CONFIG_VM_EVENT_COUNTERS is not set +CONFIG_SLAB=y +# CONFIG_SLUB is not set +# CONFIG_SLOB is not set +# CONFIG_PROFILING is not set +# CONFIG_MARKERS is not set +# CONFIG_HAVE_OPROFILE is not set +# CONFIG_HAVE_KPROBES is not set +# CONFIG_HAVE_KRETPROBES is not set +# CONFIG_HAVE_DMA_ATTRS is not set +CONFIG_SLABINFO=y +CONFIG_TINY_SHMEM=y +CONFIG_BASE_SMALL=0 +CONFIG_MODULES=y +CONFIG_MODULE_UNLOAD=y +# CONFIG_MODULE_FORCE_UNLOAD is not set +# CONFIG_MODVERSIONS is not set +# CONFIG_MODULE_SRCVERSION_ALL is not set +# CONFIG_KMOD is not set +CONFIG_BLOCK=y +# CONFIG_LBD is not set +# CONFIG_BLK_DEV_IO_TRACE is not set +# CONFIG_LSF is not set +# CONFIG_BLK_DEV_BSG is not set + +# +# IO Schedulers +# +CONFIG_IOSCHED_NOOP=y +# CONFIG_IOSCHED_AS is not set +# CONFIG_IOSCHED_DEADLINE is not set +# CONFIG_IOSCHED_CFQ is not set +# CONFIG_DEFAULT_AS is not set +# CONFIG_DEFAULT_DEADLINE is not set +# CONFIG_DEFAULT_CFQ is not set +CONFIG_DEFAULT_NOOP=y +CONFIG_DEFAULT_IOSCHED="noop" +CONFIG_CLASSIC_RCU=y + +# +# Processor type and features +# +# CONFIG_M68328 is not set +# CONFIG_M68EZ328 is not set +# CONFIG_M68VZ328 is not set +# CONFIG_M68360 is not set +# CONFIG_M5206 is not set +# CONFIG_M5206e is not set +# CONFIG_M520x is not set +# CONFIG_M523x is not set +# CONFIG_M5249 is not set +# CONFIG_M5271 is not set +# CONFIG_M5272 is not set +CONFIG_M5275=y +# CONFIG_M528x is not set +# CONFIG_M5307 is not set +# CONFIG_M532x is not set +# CONFIG_M5407 is not set +CONFIG_M527x=y +CONFIG_COLDFIRE=y +CONFIG_CLOCK_SET=y +CONFIG_CLOCK_FREQ=150000000 +CONFIG_CLOCK_DIV=2 + +# +# Platform +# +CONFIG_M5275EVB=y +CONFIG_FREESCALE=y +# CONFIG_4KSTACKS is not set +CONFIG_HZ=100 + +# +# RAM configuration +# +CONFIG_RAMBASE=0x00000000 +CONFIG_RAMSIZE=0x00000000 +CONFIG_VECTORBASE=0x00000000 +CONFIG_KERNELBASE=0x00020000 +CONFIG_RAMAUTOBIT=y +# CONFIG_RAM8BIT is not set +# CONFIG_RAM16BIT is not set +# CONFIG_RAM32BIT is not set + +# +# ROM configuration +# +# CONFIG_ROM is not set +CONFIG_RAMKERNEL=y +# CONFIG_ROMKERNEL is not set +CONFIG_SELECT_MEMORY_MODEL=y +CONFIG_FLATMEM_MANUAL=y +# CONFIG_DISCONTIGMEM_MANUAL is not set +# CONFIG_SPARSEMEM_MANUAL is not set +CONFIG_FLATMEM=y +CONFIG_FLAT_NODE_MEM_MAP=y +# CONFIG_SPARSEMEM_STATIC is not set +# CONFIG_SPARSEMEM_VMEMMAP_ENABLE is not set +CONFIG_PAGEFLAGS_EXTENDED=y +CONFIG_SPLIT_PTLOCK_CPUS=4 +# CONFIG_RESOURCES_64BIT is not set +CONFIG_ZONE_DMA_FLAG=1 +CONFIG_VIRT_TO_BUS=y +CONFIG_ISA_DMA_API=y + +# +# Bus options (PCI, PCMCIA, EISA, MCA, ISA) +# +# CONFIG_PCI is not set +# CONFIG_ARCH_SUPPORTS_MSI is not set + +# +# Executable file formats +# +CONFIG_BINFMT_FLAT=y +# CONFIG_BINFMT_ZFLAT is not set +# CONFIG_BINFMT_SHARED_FLAT is not set +# CONFIG_BINFMT_AOUT is not set +# CONFIG_BINFMT_MISC is not set + +# +# Power management options +# +# CONFIG_PM is not set + +# +# Networking +# +CONFIG_NET=y + +# +# Networking options +# +CONFIG_PACKET=y +# CONFIG_PACKET_MMAP is not set +CONFIG_UNIX=y +# CONFIG_NET_KEY is not set +CONFIG_INET=y +# CONFIG_IP_MULTICAST is not set +# CONFIG_IP_ADVANCED_ROUTER is not set +CONFIG_IP_FIB_HASH=y +# CONFIG_IP_PNP is not set +# CONFIG_NET_IPIP is not set +# CONFIG_NET_IPGRE is not set +# CONFIG_ARPD is not set +# CONFIG_SYN_COOKIES is not set +# CONFIG_INET_AH is not set +# CONFIG_INET_ESP is not set +# CONFIG_INET_IPCOMP is not set +# CONFIG_INET_XFRM_TUNNEL is not set +# CONFIG_INET_TUNNEL is not set +# CONFIG_INET_XFRM_MODE_TRANSPORT is not set +# CONFIG_INET_XFRM_MODE_TUNNEL is not set +# CONFIG_INET_XFRM_MODE_BEET is not set +# CONFIG_INET_LRO is not set +# CONFIG_INET_DIAG is not set +# CONFIG_TCP_CONG_ADVANCED is not set +CONFIG_TCP_CONG_CUBIC=y +CONFIG_DEFAULT_TCP_CONG="cubic" +# CONFIG_TCP_MD5SIG is not set +# CONFIG_IPV6 is not set +# CONFIG_NETWORK_SECMARK is not set +# CONFIG_NETFILTER is not set +# CONFIG_IP_DCCP is not set +# CONFIG_IP_SCTP is not set +# CONFIG_TIPC is not set +# CONFIG_ATM is not set +# CONFIG_BRIDGE is not set +# CONFIG_VLAN_8021Q is not set +# CONFIG_DECNET is not set +# CONFIG_LLC2 is not set +# CONFIG_IPX is not set +# CONFIG_ATALK is not set +# CONFIG_X25 is not set +# CONFIG_LAPB is not set +# CONFIG_ECONET is not set +# CONFIG_WAN_ROUTER is not set +# CONFIG_NET_SCHED is not set + +# +# Network testing +# +# CONFIG_NET_PKTGEN is not set +# CONFIG_HAMRADIO is not set +# CONFIG_CAN is not set +# CONFIG_IRDA is not set +# CONFIG_BT is not set +# CONFIG_AF_RXRPC is not set + +# +# Wireless +# +# CONFIG_CFG80211 is not set +# CONFIG_WIRELESS_EXT is not set +# CONFIG_MAC80211 is not set +# CONFIG_IEEE80211 is not set +# CONFIG_RFKILL is not set +# CONFIG_NET_9P is not set + +# +# Device Drivers +# + +# +# Generic Driver Options +# +CONFIG_STANDALONE=y +CONFIG_PREVENT_FIRMWARE_BUILD=y +# CONFIG_SYS_HYPERVISOR is not set +# CONFIG_CONNECTOR is not set +CONFIG_MTD=y +# CONFIG_MTD_DEBUG is not set +# CONFIG_MTD_CONCAT is not set +CONFIG_MTD_PARTITIONS=y +# CONFIG_MTD_REDBOOT_PARTS is not set +# CONFIG_MTD_CMDLINE_PARTS is not set +# CONFIG_MTD_AR7_PARTS is not set + +# +# User Modules And Translation Layers +# +CONFIG_MTD_CHAR=y +CONFIG_MTD_BLKDEVS=y +CONFIG_MTD_BLOCK=y +# CONFIG_FTL is not set +# CONFIG_NFTL is not set +# CONFIG_INFTL is not set +# CONFIG_RFD_FTL is not set +# CONFIG_SSFDC is not set +# CONFIG_MTD_OOPS is not set + +# +# RAM/ROM/Flash chip drivers +# +# CONFIG_MTD_CFI is not set +# CONFIG_MTD_JEDECPROBE is not set +CONFIG_MTD_MAP_BANK_WIDTH_1=y +CONFIG_MTD_MAP_BANK_WIDTH_2=y +CONFIG_MTD_MAP_BANK_WIDTH_4=y +# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set +# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set +# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set +CONFIG_MTD_CFI_I1=y +CONFIG_MTD_CFI_I2=y +# CONFIG_MTD_CFI_I4 is not set +# CONFIG_MTD_CFI_I8 is not set +CONFIG_MTD_RAM=y +# CONFIG_MTD_ROM is not set +# CONFIG_MTD_ABSENT is not set + +# +# Mapping drivers for chip access +# +# CONFIG_MTD_COMPLEX_MAPPINGS is not set +CONFIG_MTD_UCLINUX=y +# CONFIG_MTD_PLATRAM is not set + +# +# Self-contained MTD device drivers +# +# CONFIG_MTD_SLRAM is not set +# CONFIG_MTD_PHRAM is not set +# CONFIG_MTD_MTDRAM is not set +# CONFIG_MTD_BLOCK2MTD is not set + +# +# Disk-On-Chip Device Drivers +# +# CONFIG_MTD_DOC2000 is not set +# CONFIG_MTD_DOC2001 is not set +# CONFIG_MTD_DOC2001PLUS is not set +# CONFIG_MTD_NAND is not set +# CONFIG_MTD_ONENAND is not set + +# +# UBI - Unsorted block images +# +# CONFIG_MTD_UBI is not set +# CONFIG_PARPORT is not set +CONFIG_BLK_DEV=y +# CONFIG_BLK_DEV_COW_COMMON is not set +# CONFIG_BLK_DEV_LOOP is not set +# CONFIG_BLK_DEV_NBD is not set +CONFIG_BLK_DEV_RAM=y +CONFIG_BLK_DEV_RAM_COUNT=16 +CONFIG_BLK_DEV_RAM_SIZE=4096 +# CONFIG_BLK_DEV_XIP is not set +# CONFIG_CDROM_PKTCDVD is not set +# CONFIG_ATA_OVER_ETH is not set +# CONFIG_MISC_DEVICES is not set +CONFIG_HAVE_IDE=y +# CONFIG_IDE is not set + +# +# SCSI device support +# +# CONFIG_RAID_ATTRS is not set +# CONFIG_SCSI is not set +# CONFIG_SCSI_DMA is not set +# CONFIG_SCSI_NETLINK is not set +# CONFIG_MD is not set +CONFIG_NETDEVICES=y +# CONFIG_NETDEVICES_MULTIQUEUE is not set +# CONFIG_DUMMY is not set +# CONFIG_BONDING is not set +# CONFIG_MACVLAN is not set +# CONFIG_EQUALIZER is not set +# CONFIG_TUN is not set +# CONFIG_VETH is not set +# CONFIG_PHYLIB is not set +CONFIG_NET_ETHERNET=y +# CONFIG_MII is not set +# CONFIG_IBM_NEW_EMAC_ZMII is not set +# CONFIG_IBM_NEW_EMAC_RGMII is not set +# CONFIG_IBM_NEW_EMAC_TAH is not set +# CONFIG_IBM_NEW_EMAC_EMAC4 is not set +# CONFIG_B44 is not set +CONFIG_FEC=y +CONFIG_FEC2=y +# CONFIG_NETDEV_1000 is not set +# CONFIG_NETDEV_10000 is not set + +# +# Wireless LAN +# +# CONFIG_WLAN_PRE80211 is not set +# CONFIG_WLAN_80211 is not set +# CONFIG_IWLWIFI is not set +# CONFIG_IWLWIFI_LEDS is not set +# CONFIG_WAN is not set +CONFIG_PPP=y +# CONFIG_PPP_MULTILINK is not set +# CONFIG_PPP_FILTER is not set +# CONFIG_PPP_ASYNC is not set +# CONFIG_PPP_SYNC_TTY is not set +# CONFIG_PPP_DEFLATE is not set +# CONFIG_PPP_BSDCOMP is not set +# CONFIG_PPP_MPPE is not set +# CONFIG_PPPOE is not set +# CONFIG_PPPOL2TP is not set +# CONFIG_SLIP is not set +CONFIG_SLHC=y +# CONFIG_NETCONSOLE is not set +# CONFIG_NETPOLL is not set +# CONFIG_NET_POLL_CONTROLLER is not set +# CONFIG_ISDN is not set +# CONFIG_PHONE is not set + +# +# Input device support +# +# CONFIG_INPUT is not set + +# +# Hardware I/O ports +# +# CONFIG_SERIO is not set +# CONFIG_GAMEPORT is not set + +# +# Character devices +# +# CONFIG_VT is not set +# CONFIG_DEVKMEM is not set +# CONFIG_SERIAL_NONSTANDARD is not set + +# +# Serial drivers +# +# CONFIG_SERIAL_8250 is not set + +# +# Non-8250 serial port support +# +CONFIG_SERIAL_CORE=y +CONFIG_SERIAL_CORE_CONSOLE=y +# CONFIG_SERIAL_COLDFIRE is not set +CONFIG_SERIAL_MCF=y +CONFIG_SERIAL_MCF_BAUDRATE=19200 +CONFIG_SERIAL_MCF_CONSOLE=y +# CONFIG_UNIX98_PTYS is not set +CONFIG_LEGACY_PTYS=y +CONFIG_LEGACY_PTY_COUNT=256 +# CONFIG_IPMI_HANDLER is not set +# CONFIG_HW_RANDOM is not set +# CONFIG_GEN_RTC is not set +# CONFIG_R3964 is not set +# CONFIG_RAW_DRIVER is not set +# CONFIG_TCG_TPM is not set +# CONFIG_I2C is not set +# CONFIG_SPI is not set +# CONFIG_W1 is not set +# CONFIG_POWER_SUPPLY is not set +# CONFIG_HWMON is not set +# CONFIG_THERMAL is not set +# CONFIG_WATCHDOG is not set + +# +# Sonics Silicon Backplane +# +CONFIG_SSB_POSSIBLE=y +# CONFIG_SSB is not set + +# +# Multifunction device drivers +# +# CONFIG_MFD_SM501 is not set +# CONFIG_HTC_PASIC3 is not set + +# +# Multimedia devices +# + +# +# Multimedia core support +# +# CONFIG_VIDEO_DEV is not set +# CONFIG_DVB_CORE is not set + +# +# Multimedia drivers +# +CONFIG_DAB=y + +# +# Graphics support +# +# CONFIG_VGASTATE is not set +# CONFIG_VIDEO_OUTPUT_CONTROL is not set +# CONFIG_FB is not set +# CONFIG_BACKLIGHT_LCD_SUPPORT is not set + +# +# Display device support +# +# CONFIG_DISPLAY_SUPPORT is not set + +# +# Sound +# +# CONFIG_SOUND is not set +# CONFIG_USB_SUPPORT is not set +# CONFIG_MMC is not set +# CONFIG_MEMSTICK is not set +# CONFIG_NEW_LEDS is not set +# CONFIG_ACCESSIBILITY is not set +# CONFIG_RTC_CLASS is not set +# CONFIG_UIO is not set + +# +# File systems +# +CONFIG_EXT2_FS=y +# CONFIG_EXT2_FS_XATTR is not set +# CONFIG_EXT3_FS is not set +# CONFIG_EXT4DEV_FS is not set +# CONFIG_REISERFS_FS is not set +# CONFIG_JFS_FS is not set +# CONFIG_FS_POSIX_ACL is not set +# CONFIG_XFS_FS is not set +# CONFIG_OCFS2_FS is not set +# CONFIG_DNOTIFY is not set +# CONFIG_INOTIFY is not set +# CONFIG_QUOTA is not set +# CONFIG_AUTOFS_FS is not set +# CONFIG_AUTOFS4_FS is not set +# CONFIG_FUSE_FS is not set + +# +# CD-ROM/DVD Filesystems +# +# CONFIG_ISO9660_FS is not set +# CONFIG_UDF_FS is not set + +# +# DOS/FAT/NT Filesystems +# +# CONFIG_MSDOS_FS is not set +# CONFIG_VFAT_FS is not set +# CONFIG_NTFS_FS is not set + +# +# Pseudo filesystems +# +CONFIG_PROC_FS=y +CONFIG_PROC_SYSCTL=y +CONFIG_SYSFS=y +# CONFIG_TMPFS is not set +# CONFIG_HUGETLB_PAGE is not set +# CONFIG_CONFIGFS_FS is not set + +# +# Miscellaneous filesystems +# +# CONFIG_ADFS_FS is not set +# CONFIG_AFFS_FS is not set +# CONFIG_HFS_FS is not set +# CONFIG_HFSPLUS_FS is not set +# CONFIG_BEFS_FS is not set +# CONFIG_BFS_FS is not set +# CONFIG_EFS_FS is not set +# CONFIG_JFFS2_FS is not set +# CONFIG_CRAMFS is not set +# CONFIG_VXFS_FS is not set +# CONFIG_MINIX_FS is not set +# CONFIG_HPFS_FS is not set +# CONFIG_QNX4FS_FS is not set +CONFIG_ROMFS_FS=y +# CONFIG_SYSV_FS is not set +# CONFIG_UFS_FS is not set +CONFIG_NETWORK_FILESYSTEMS=y +# CONFIG_NFS_FS is not set +# CONFIG_NFSD is not set +# CONFIG_SMB_FS is not set +# CONFIG_CIFS is not set +# CONFIG_NCP_FS is not set +# CONFIG_CODA_FS is not set +# CONFIG_AFS_FS is not set + +# +# Partition Types +# +# CONFIG_PARTITION_ADVANCED is not set +CONFIG_MSDOS_PARTITION=y +# CONFIG_NLS is not set +# CONFIG_DLM is not set + +# +# Kernel hacking +# +# CONFIG_PRINTK_TIME is not set +CONFIG_ENABLE_WARN_DEPRECATED=y +CONFIG_ENABLE_MUST_CHECK=y +CONFIG_FRAME_WARN=1024 +# CONFIG_MAGIC_SYSRQ is not set +# CONFIG_UNUSED_SYMBOLS is not set +# CONFIG_DEBUG_FS is not set +# CONFIG_HEADERS_CHECK is not set +# CONFIG_DEBUG_KERNEL is not set +# CONFIG_DEBUG_BUGVERBOSE is not set +# CONFIG_SAMPLES is not set +# CONFIG_FULLDEBUG is not set +# CONFIG_HIGHPROFILE is not set +# CONFIG_BOOTPARAM is not set +# CONFIG_NO_KERNEL_MSG is not set +# CONFIG_BDM_DISABLE is not set + +# +# Security options +# +# CONFIG_KEYS is not set +# CONFIG_SECURITY is not set +# CONFIG_SECURITY_FILE_CAPABILITIES is not set +# CONFIG_CRYPTO is not set + +# +# Library routines +# +# CONFIG_GENERIC_FIND_FIRST_BIT is not set +# CONFIG_CRC_CCITT is not set +# CONFIG_CRC16 is not set +# CONFIG_CRC_ITU_T is not set +# CONFIG_CRC32 is not set +# CONFIG_CRC7 is not set +# CONFIG_LIBCRC32C is not set +CONFIG_HAS_IOMEM=y +CONFIG_HAS_DMA=y From f3a64eaf9ea87d70487b482a6d733cfd52ae4499 Mon Sep 17 00:00:00 2001 From: Greg Ungerer Date: Wed, 4 Jun 2008 21:24:54 +1000 Subject: [PATCH 049/857] m68knommu: defconfig for M5307C3 board Add a defconfig for the Freescale M5307C3 board. Signed-off-by: Greg Ungerer --- arch/m68knommu/configs/m5307c3_defconfig | 580 +++++++++++++++++++++++ 1 file changed, 580 insertions(+) create mode 100644 arch/m68knommu/configs/m5307c3_defconfig diff --git a/arch/m68knommu/configs/m5307c3_defconfig b/arch/m68knommu/configs/m5307c3_defconfig new file mode 100644 index 000000000000..fe2acdfa4d76 --- /dev/null +++ b/arch/m68knommu/configs/m5307c3_defconfig @@ -0,0 +1,580 @@ +# +# Automatically generated make config: don't edit +# Linux kernel version: 2.6.26-rc1 +# +CONFIG_M68K=y +# CONFIG_MMU is not set +# CONFIG_FPU is not set +CONFIG_ZONE_DMA=y +CONFIG_RWSEM_GENERIC_SPINLOCK=y +# CONFIG_RWSEM_XCHGADD_ALGORITHM is not set +# CONFIG_ARCH_HAS_ILOG2_U32 is not set +# CONFIG_ARCH_HAS_ILOG2_U64 is not set +CONFIG_GENERIC_FIND_NEXT_BIT=y +CONFIG_GENERIC_HWEIGHT=y +CONFIG_GENERIC_HARDIRQS=y +CONFIG_GENERIC_CALIBRATE_DELAY=y +CONFIG_GENERIC_TIME=y +CONFIG_TIME_LOW_RES=y +CONFIG_NO_IOPORT=y +CONFIG_ARCH_SUPPORTS_AOUT=y +CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config" + +# +# General setup +# +CONFIG_EXPERIMENTAL=y +CONFIG_BROKEN_ON_SMP=y +CONFIG_INIT_ENV_ARG_LIMIT=32 +CONFIG_LOCALVERSION="" +CONFIG_LOCALVERSION_AUTO=y +# CONFIG_SYSVIPC is not set +# CONFIG_POSIX_MQUEUE is not set +# CONFIG_BSD_PROCESS_ACCT is not set +# CONFIG_TASKSTATS is not set +# CONFIG_AUDIT is not set +# CONFIG_IKCONFIG is not set +CONFIG_LOG_BUF_SHIFT=14 +# CONFIG_CGROUPS is not set +# CONFIG_GROUP_SCHED is not set +# CONFIG_SYSFS_DEPRECATED_V2 is not set +# CONFIG_RELAY is not set +# CONFIG_NAMESPACES is not set +# CONFIG_BLK_DEV_INITRD is not set +# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set +CONFIG_SYSCTL=y +CONFIG_EMBEDDED=y +# CONFIG_UID16 is not set +# CONFIG_SYSCTL_SYSCALL is not set +# CONFIG_KALLSYMS is not set +# CONFIG_HOTPLUG is not set +CONFIG_PRINTK=y +CONFIG_BUG=y +CONFIG_ELF_CORE=y +# CONFIG_COMPAT_BRK is not set +CONFIG_BASE_FULL=y +# CONFIG_FUTEX is not set +# CONFIG_EPOLL is not set +# CONFIG_SIGNALFD is not set +# CONFIG_TIMERFD is not set +# CONFIG_EVENTFD is not set +# CONFIG_VM_EVENT_COUNTERS is not set +CONFIG_SLAB=y +# CONFIG_SLUB is not set +# CONFIG_SLOB is not set +# CONFIG_PROFILING is not set +# CONFIG_MARKERS is not set +# CONFIG_HAVE_OPROFILE is not set +# CONFIG_HAVE_KPROBES is not set +# CONFIG_HAVE_KRETPROBES is not set +# CONFIG_HAVE_DMA_ATTRS is not set +CONFIG_SLABINFO=y +CONFIG_TINY_SHMEM=y +CONFIG_BASE_SMALL=0 +CONFIG_MODULES=y +CONFIG_MODULE_UNLOAD=y +# CONFIG_MODULE_FORCE_UNLOAD is not set +# CONFIG_MODVERSIONS is not set +# CONFIG_MODULE_SRCVERSION_ALL is not set +# CONFIG_KMOD is not set +CONFIG_BLOCK=y +# CONFIG_LBD is not set +# CONFIG_BLK_DEV_IO_TRACE is not set +# CONFIG_LSF is not set +# CONFIG_BLK_DEV_BSG is not set + +# +# IO Schedulers +# +CONFIG_IOSCHED_NOOP=y +# CONFIG_IOSCHED_AS is not set +# CONFIG_IOSCHED_DEADLINE is not set +# CONFIG_IOSCHED_CFQ is not set +# CONFIG_DEFAULT_AS is not set +# CONFIG_DEFAULT_DEADLINE is not set +# CONFIG_DEFAULT_CFQ is not set +CONFIG_DEFAULT_NOOP=y +CONFIG_DEFAULT_IOSCHED="noop" +CONFIG_CLASSIC_RCU=y + +# +# Processor type and features +# +# CONFIG_M68328 is not set +# CONFIG_M68EZ328 is not set +# CONFIG_M68VZ328 is not set +# CONFIG_M68360 is not set +# CONFIG_M5206 is not set +# CONFIG_M5206e is not set +# CONFIG_M520x is not set +# CONFIG_M523x is not set +# CONFIG_M5249 is not set +# CONFIG_M5271 is not set +# CONFIG_M5272 is not set +# CONFIG_M5275 is not set +# CONFIG_M528x is not set +CONFIG_M5307=y +# CONFIG_M532x is not set +# CONFIG_M5407 is not set +CONFIG_COLDFIRE=y +CONFIG_CLOCK_SET=y +CONFIG_CLOCK_FREQ=90000000 +CONFIG_CLOCK_DIV=2 +# CONFIG_OLDMASK is not set + +# +# Platform +# +# CONFIG_ARN5307 is not set +CONFIG_M5307C3=y +# CONFIG_eLIA is not set +# CONFIG_SECUREEDGEMP3 is not set +# CONFIG_CLEOPATRA is not set +# CONFIG_NETtel is not set +CONFIG_FREESCALE=y +# CONFIG_4KSTACKS is not set +CONFIG_HZ=100 + +# +# RAM configuration +# +CONFIG_RAMBASE=0x00000000 +CONFIG_RAMSIZE=0x00800000 +CONFIG_VECTORBASE=0x00000000 +CONFIG_KERNELBASE=0x00020000 +CONFIG_RAMAUTOBIT=y +# CONFIG_RAM8BIT is not set +# CONFIG_RAM16BIT is not set +# CONFIG_RAM32BIT is not set + +# +# ROM configuration +# +# CONFIG_ROM is not set +CONFIG_RAMKERNEL=y +# CONFIG_ROMKERNEL is not set +CONFIG_SELECT_MEMORY_MODEL=y +CONFIG_FLATMEM_MANUAL=y +# CONFIG_DISCONTIGMEM_MANUAL is not set +# CONFIG_SPARSEMEM_MANUAL is not set +CONFIG_FLATMEM=y +CONFIG_FLAT_NODE_MEM_MAP=y +# CONFIG_SPARSEMEM_STATIC is not set +# CONFIG_SPARSEMEM_VMEMMAP_ENABLE is not set +CONFIG_PAGEFLAGS_EXTENDED=y +CONFIG_SPLIT_PTLOCK_CPUS=4 +# CONFIG_RESOURCES_64BIT is not set +CONFIG_ZONE_DMA_FLAG=1 +CONFIG_VIRT_TO_BUS=y +CONFIG_ISA_DMA_API=y + +# +# Bus options (PCI, PCMCIA, EISA, MCA, ISA) +# +# CONFIG_PCI is not set +# CONFIG_COMEMPCI is not set +# CONFIG_ARCH_SUPPORTS_MSI is not set + +# +# Executable file formats +# +CONFIG_BINFMT_FLAT=y +# CONFIG_BINFMT_ZFLAT is not set +# CONFIG_BINFMT_SHARED_FLAT is not set +# CONFIG_BINFMT_AOUT is not set +# CONFIG_BINFMT_MISC is not set + +# +# Power management options +# +# CONFIG_PM is not set + +# +# Networking +# +CONFIG_NET=y + +# +# Networking options +# +CONFIG_PACKET=y +# CONFIG_PACKET_MMAP is not set +CONFIG_UNIX=y +# CONFIG_NET_KEY is not set +CONFIG_INET=y +# CONFIG_IP_MULTICAST is not set +# CONFIG_IP_ADVANCED_ROUTER is not set +CONFIG_IP_FIB_HASH=y +# CONFIG_IP_PNP is not set +# CONFIG_NET_IPIP is not set +# CONFIG_NET_IPGRE is not set +# CONFIG_ARPD is not set +# CONFIG_SYN_COOKIES is not set +# CONFIG_INET_AH is not set +# CONFIG_INET_ESP is not set +# CONFIG_INET_IPCOMP is not set +# CONFIG_INET_XFRM_TUNNEL is not set +# CONFIG_INET_TUNNEL is not set +# CONFIG_INET_XFRM_MODE_TRANSPORT is not set +# CONFIG_INET_XFRM_MODE_TUNNEL is not set +# CONFIG_INET_XFRM_MODE_BEET is not set +# CONFIG_INET_LRO is not set +# CONFIG_INET_DIAG is not set +# CONFIG_TCP_CONG_ADVANCED is not set +CONFIG_TCP_CONG_CUBIC=y +CONFIG_DEFAULT_TCP_CONG="cubic" +# CONFIG_TCP_MD5SIG is not set +# CONFIG_IPV6 is not set +# CONFIG_NETWORK_SECMARK is not set +# CONFIG_NETFILTER is not set +# CONFIG_IP_DCCP is not set +# CONFIG_IP_SCTP is not set +# CONFIG_TIPC is not set +# CONFIG_ATM is not set +# CONFIG_BRIDGE is not set +# CONFIG_VLAN_8021Q is not set +# CONFIG_DECNET is not set +# CONFIG_LLC2 is not set +# CONFIG_IPX is not set +# CONFIG_ATALK is not set +# CONFIG_X25 is not set +# CONFIG_LAPB is not set +# CONFIG_ECONET is not set +# CONFIG_WAN_ROUTER is not set +# CONFIG_NET_SCHED is not set + +# +# Network testing +# +# CONFIG_NET_PKTGEN is not set +# CONFIG_HAMRADIO is not set +# CONFIG_CAN is not set +# CONFIG_IRDA is not set +# CONFIG_BT is not set +# CONFIG_AF_RXRPC is not set + +# +# Wireless +# +# CONFIG_CFG80211 is not set +# CONFIG_WIRELESS_EXT is not set +# CONFIG_MAC80211 is not set +# CONFIG_IEEE80211 is not set +# CONFIG_RFKILL is not set +# CONFIG_NET_9P is not set + +# +# Device Drivers +# + +# +# Generic Driver Options +# +CONFIG_STANDALONE=y +CONFIG_PREVENT_FIRMWARE_BUILD=y +# CONFIG_SYS_HYPERVISOR is not set +# CONFIG_CONNECTOR is not set +# CONFIG_MTD is not set +# CONFIG_PARPORT is not set +CONFIG_BLK_DEV=y +# CONFIG_BLK_DEV_COW_COMMON is not set +# CONFIG_BLK_DEV_LOOP is not set +# CONFIG_BLK_DEV_NBD is not set +CONFIG_BLK_DEV_RAM=y +CONFIG_BLK_DEV_RAM_COUNT=16 +CONFIG_BLK_DEV_RAM_SIZE=4096 +# CONFIG_BLK_DEV_XIP is not set +# CONFIG_CDROM_PKTCDVD is not set +# CONFIG_ATA_OVER_ETH is not set +# CONFIG_MISC_DEVICES is not set +CONFIG_HAVE_IDE=y +# CONFIG_IDE is not set + +# +# SCSI device support +# +# CONFIG_RAID_ATTRS is not set +# CONFIG_SCSI is not set +# CONFIG_SCSI_DMA is not set +# CONFIG_SCSI_NETLINK is not set +# CONFIG_MD is not set +CONFIG_NETDEVICES=y +# CONFIG_NETDEVICES_MULTIQUEUE is not set +# CONFIG_DUMMY is not set +# CONFIG_BONDING is not set +# CONFIG_MACVLAN is not set +# CONFIG_EQUALIZER is not set +# CONFIG_TUN is not set +# CONFIG_VETH is not set +# CONFIG_PHYLIB is not set +CONFIG_NET_ETHERNET=y +# CONFIG_MII is not set +# CONFIG_IBM_NEW_EMAC_ZMII is not set +# CONFIG_IBM_NEW_EMAC_RGMII is not set +# CONFIG_IBM_NEW_EMAC_TAH is not set +# CONFIG_IBM_NEW_EMAC_EMAC4 is not set +# CONFIG_B44 is not set +# CONFIG_NETDEV_1000 is not set +# CONFIG_NETDEV_10000 is not set + +# +# Wireless LAN +# +# CONFIG_WLAN_PRE80211 is not set +# CONFIG_WLAN_80211 is not set +# CONFIG_IWLWIFI is not set +# CONFIG_IWLWIFI_LEDS is not set +# CONFIG_WAN is not set +CONFIG_PPP=y +# CONFIG_PPP_MULTILINK is not set +# CONFIG_PPP_FILTER is not set +# CONFIG_PPP_ASYNC is not set +# CONFIG_PPP_SYNC_TTY is not set +# CONFIG_PPP_DEFLATE is not set +# CONFIG_PPP_BSDCOMP is not set +# CONFIG_PPP_MPPE is not set +# CONFIG_PPPOE is not set +# CONFIG_PPPOL2TP is not set +CONFIG_SLIP=y +CONFIG_SLIP_COMPRESSED=y +CONFIG_SLHC=y +# CONFIG_SLIP_SMART is not set +# CONFIG_SLIP_MODE_SLIP6 is not set +# CONFIG_NETCONSOLE is not set +# CONFIG_NETPOLL is not set +# CONFIG_NET_POLL_CONTROLLER is not set +# CONFIG_ISDN is not set +# CONFIG_PHONE is not set + +# +# Input device support +# +CONFIG_INPUT=y +# CONFIG_INPUT_FF_MEMLESS is not set +# CONFIG_INPUT_POLLDEV is not set + +# +# Userland interfaces +# +# CONFIG_INPUT_MOUSEDEV is not set +# CONFIG_INPUT_JOYDEV is not set +# CONFIG_INPUT_EVDEV is not set +# CONFIG_INPUT_EVBUG is not set + +# +# Input Device Drivers +# +# CONFIG_INPUT_KEYBOARD is not set +# CONFIG_INPUT_MOUSE is not set +# CONFIG_INPUT_JOYSTICK is not set +# CONFIG_INPUT_TABLET is not set +# CONFIG_INPUT_TOUCHSCREEN is not set +# CONFIG_INPUT_MISC is not set + +# +# Hardware I/O ports +# +# CONFIG_SERIO is not set +# CONFIG_GAMEPORT is not set + +# +# Character devices +# +# CONFIG_VT is not set +# CONFIG_DEVKMEM is not set +# CONFIG_SERIAL_NONSTANDARD is not set + +# +# Serial drivers +# +# CONFIG_SERIAL_8250 is not set + +# +# Non-8250 serial port support +# +CONFIG_SERIAL_CORE=y +CONFIG_SERIAL_CORE_CONSOLE=y +# CONFIG_SERIAL_COLDFIRE is not set +CONFIG_SERIAL_MCF=y +CONFIG_SERIAL_MCF_BAUDRATE=19200 +CONFIG_SERIAL_MCF_CONSOLE=y +CONFIG_UNIX98_PTYS=y +CONFIG_LEGACY_PTYS=y +CONFIG_LEGACY_PTY_COUNT=256 +# CONFIG_IPMI_HANDLER is not set +# CONFIG_HW_RANDOM is not set +# CONFIG_GEN_RTC is not set +# CONFIG_R3964 is not set +# CONFIG_RAW_DRIVER is not set +# CONFIG_TCG_TPM is not set +# CONFIG_I2C is not set +# CONFIG_SPI is not set +# CONFIG_W1 is not set +# CONFIG_POWER_SUPPLY is not set +# CONFIG_HWMON is not set +# CONFIG_THERMAL is not set +# CONFIG_WATCHDOG is not set + +# +# Sonics Silicon Backplane +# +CONFIG_SSB_POSSIBLE=y +# CONFIG_SSB is not set + +# +# Multifunction device drivers +# +# CONFIG_MFD_SM501 is not set +# CONFIG_HTC_PASIC3 is not set + +# +# Multimedia devices +# + +# +# Multimedia core support +# +# CONFIG_VIDEO_DEV is not set +# CONFIG_DVB_CORE is not set + +# +# Multimedia drivers +# +CONFIG_DAB=y + +# +# Graphics support +# +# CONFIG_VGASTATE is not set +# CONFIG_VIDEO_OUTPUT_CONTROL is not set +# CONFIG_FB is not set +# CONFIG_BACKLIGHT_LCD_SUPPORT is not set + +# +# Display device support +# +# CONFIG_DISPLAY_SUPPORT is not set + +# +# Sound +# +# CONFIG_SOUND is not set +# CONFIG_HID_SUPPORT is not set +# CONFIG_USB_SUPPORT is not set +# CONFIG_MMC is not set +# CONFIG_MEMSTICK is not set +# CONFIG_NEW_LEDS is not set +# CONFIG_ACCESSIBILITY is not set +# CONFIG_RTC_CLASS is not set +# CONFIG_UIO is not set + +# +# File systems +# +CONFIG_EXT2_FS=y +# CONFIG_EXT2_FS_XATTR is not set +# CONFIG_EXT3_FS is not set +# CONFIG_EXT4DEV_FS is not set +# CONFIG_REISERFS_FS is not set +# CONFIG_JFS_FS is not set +# CONFIG_FS_POSIX_ACL is not set +# CONFIG_XFS_FS is not set +# CONFIG_OCFS2_FS is not set +# CONFIG_DNOTIFY is not set +# CONFIG_INOTIFY is not set +# CONFIG_QUOTA is not set +# CONFIG_AUTOFS_FS is not set +# CONFIG_AUTOFS4_FS is not set +# CONFIG_FUSE_FS is not set + +# +# CD-ROM/DVD Filesystems +# +# CONFIG_ISO9660_FS is not set +# CONFIG_UDF_FS is not set + +# +# DOS/FAT/NT Filesystems +# +# CONFIG_MSDOS_FS is not set +# CONFIG_VFAT_FS is not set +# CONFIG_NTFS_FS is not set + +# +# Pseudo filesystems +# +CONFIG_PROC_FS=y +CONFIG_PROC_SYSCTL=y +CONFIG_SYSFS=y +# CONFIG_TMPFS is not set +# CONFIG_HUGETLB_PAGE is not set +# CONFIG_CONFIGFS_FS is not set + +# +# Miscellaneous filesystems +# +# CONFIG_ADFS_FS is not set +# CONFIG_AFFS_FS is not set +# CONFIG_HFS_FS is not set +# CONFIG_HFSPLUS_FS is not set +# CONFIG_BEFS_FS is not set +# CONFIG_BFS_FS is not set +# CONFIG_EFS_FS is not set +# CONFIG_CRAMFS is not set +# CONFIG_VXFS_FS is not set +# CONFIG_MINIX_FS is not set +# CONFIG_HPFS_FS is not set +# CONFIG_QNX4FS_FS is not set +CONFIG_ROMFS_FS=y +# CONFIG_SYSV_FS is not set +# CONFIG_UFS_FS is not set +# CONFIG_NETWORK_FILESYSTEMS is not set + +# +# Partition Types +# +# CONFIG_PARTITION_ADVANCED is not set +CONFIG_MSDOS_PARTITION=y +# CONFIG_NLS is not set +# CONFIG_DLM is not set + +# +# Kernel hacking +# +# CONFIG_PRINTK_TIME is not set +CONFIG_ENABLE_WARN_DEPRECATED=y +CONFIG_ENABLE_MUST_CHECK=y +CONFIG_FRAME_WARN=1024 +# CONFIG_MAGIC_SYSRQ is not set +# CONFIG_UNUSED_SYMBOLS is not set +# CONFIG_DEBUG_FS is not set +# CONFIG_HEADERS_CHECK is not set +# CONFIG_DEBUG_KERNEL is not set +# CONFIG_DEBUG_BUGVERBOSE is not set +# CONFIG_SAMPLES is not set +CONFIG_FULLDEBUG=y +# CONFIG_HIGHPROFILE is not set +# CONFIG_BOOTPARAM is not set +# CONFIG_NO_KERNEL_MSG is not set +# CONFIG_BDM_DISABLE is not set + +# +# Security options +# +# CONFIG_KEYS is not set +# CONFIG_SECURITY is not set +# CONFIG_SECURITY_FILE_CAPABILITIES is not set +# CONFIG_CRYPTO is not set + +# +# Library routines +# +# CONFIG_GENERIC_FIND_FIRST_BIT is not set +# CONFIG_CRC_CCITT is not set +# CONFIG_CRC16 is not set +# CONFIG_CRC_ITU_T is not set +# CONFIG_CRC32 is not set +# CONFIG_CRC7 is not set +# CONFIG_LIBCRC32C is not set +CONFIG_HAS_IOMEM=y +CONFIG_HAS_DMA=y From ab88e474c8ffa300660f03a8e6b08ea660956bef Mon Sep 17 00:00:00 2001 From: Greg Ungerer Date: Wed, 4 Jun 2008 21:26:38 +1000 Subject: [PATCH 050/857] m68knommu: defconfig for M5407C3 board Add a defconfig for the Freescale M5407C3 board. Signed-off-by: Greg Ungerer --- arch/m68knommu/configs/m5407c3_defconfig | 641 +++++++++++++++++++++++ 1 file changed, 641 insertions(+) create mode 100644 arch/m68knommu/configs/m5407c3_defconfig diff --git a/arch/m68knommu/configs/m5407c3_defconfig b/arch/m68knommu/configs/m5407c3_defconfig new file mode 100644 index 000000000000..1118936d20e3 --- /dev/null +++ b/arch/m68knommu/configs/m5407c3_defconfig @@ -0,0 +1,641 @@ +# +# Automatically generated make config: don't edit +# Linux kernel version: 2.6.26-rc1 +# Wed May 7 10:25:16 2008 +# +CONFIG_M68K=y +# CONFIG_MMU is not set +# CONFIG_FPU is not set +CONFIG_ZONE_DMA=y +CONFIG_RWSEM_GENERIC_SPINLOCK=y +# CONFIG_RWSEM_XCHGADD_ALGORITHM is not set +# CONFIG_ARCH_HAS_ILOG2_U32 is not set +# CONFIG_ARCH_HAS_ILOG2_U64 is not set +CONFIG_GENERIC_FIND_NEXT_BIT=y +CONFIG_GENERIC_HWEIGHT=y +CONFIG_GENERIC_HARDIRQS=y +CONFIG_GENERIC_CALIBRATE_DELAY=y +CONFIG_GENERIC_TIME=y +CONFIG_TIME_LOW_RES=y +CONFIG_NO_IOPORT=y +CONFIG_ARCH_SUPPORTS_AOUT=y +CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config" + +# +# General setup +# +CONFIG_EXPERIMENTAL=y +CONFIG_BROKEN_ON_SMP=y +CONFIG_INIT_ENV_ARG_LIMIT=32 +CONFIG_LOCALVERSION="" +CONFIG_LOCALVERSION_AUTO=y +# CONFIG_SYSVIPC is not set +# CONFIG_POSIX_MQUEUE is not set +# CONFIG_BSD_PROCESS_ACCT is not set +# CONFIG_TASKSTATS is not set +# CONFIG_AUDIT is not set +# CONFIG_IKCONFIG is not set +CONFIG_LOG_BUF_SHIFT=14 +# CONFIG_CGROUPS is not set +# CONFIG_GROUP_SCHED is not set +# CONFIG_SYSFS_DEPRECATED_V2 is not set +# CONFIG_RELAY is not set +# CONFIG_NAMESPACES is not set +# CONFIG_BLK_DEV_INITRD is not set +# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set +CONFIG_SYSCTL=y +CONFIG_EMBEDDED=y +# CONFIG_UID16 is not set +# CONFIG_SYSCTL_SYSCALL is not set +# CONFIG_KALLSYMS is not set +# CONFIG_HOTPLUG is not set +CONFIG_PRINTK=y +CONFIG_BUG=y +CONFIG_ELF_CORE=y +# CONFIG_COMPAT_BRK is not set +CONFIG_BASE_FULL=y +# CONFIG_FUTEX is not set +# CONFIG_EPOLL is not set +# CONFIG_SIGNALFD is not set +# CONFIG_TIMERFD is not set +# CONFIG_EVENTFD is not set +# CONFIG_VM_EVENT_COUNTERS is not set +CONFIG_SLAB=y +# CONFIG_SLUB is not set +# CONFIG_SLOB is not set +# CONFIG_PROFILING is not set +# CONFIG_MARKERS is not set +# CONFIG_HAVE_OPROFILE is not set +# CONFIG_HAVE_KPROBES is not set +# CONFIG_HAVE_KRETPROBES is not set +# CONFIG_HAVE_DMA_ATTRS is not set +CONFIG_SLABINFO=y +CONFIG_TINY_SHMEM=y +CONFIG_BASE_SMALL=0 +CONFIG_MODULES=y +CONFIG_MODULE_UNLOAD=y +# CONFIG_MODULE_FORCE_UNLOAD is not set +# CONFIG_MODVERSIONS is not set +# CONFIG_MODULE_SRCVERSION_ALL is not set +# CONFIG_KMOD is not set +CONFIG_BLOCK=y +# CONFIG_LBD is not set +# CONFIG_BLK_DEV_IO_TRACE is not set +# CONFIG_LSF is not set +# CONFIG_BLK_DEV_BSG is not set + +# +# IO Schedulers +# +CONFIG_IOSCHED_NOOP=y +# CONFIG_IOSCHED_AS is not set +# CONFIG_IOSCHED_DEADLINE is not set +# CONFIG_IOSCHED_CFQ is not set +# CONFIG_DEFAULT_AS is not set +# CONFIG_DEFAULT_DEADLINE is not set +# CONFIG_DEFAULT_CFQ is not set +CONFIG_DEFAULT_NOOP=y +CONFIG_DEFAULT_IOSCHED="noop" +CONFIG_CLASSIC_RCU=y + +# +# Processor type and features +# +# CONFIG_M68328 is not set +# CONFIG_M68EZ328 is not set +# CONFIG_M68VZ328 is not set +# CONFIG_M68360 is not set +# CONFIG_M5206 is not set +# CONFIG_M5206e is not set +# CONFIG_M520x is not set +# CONFIG_M523x is not set +# CONFIG_M5249 is not set +# CONFIG_M5271 is not set +# CONFIG_M5272 is not set +# CONFIG_M5275 is not set +# CONFIG_M528x is not set +# CONFIG_M5307 is not set +# CONFIG_M532x is not set +CONFIG_M5407=y +CONFIG_COLDFIRE=y +CONFIG_CLOCK_SET=y +CONFIG_CLOCK_FREQ=50000000 +CONFIG_CLOCK_DIV=1 + +# +# Platform +# +CONFIG_M5407C3=y +# CONFIG_CLEOPATRA is not set +CONFIG_FREESCALE=y +CONFIG_4KSTACKS=y +CONFIG_HZ=100 + +# +# RAM configuration +# +CONFIG_RAMBASE=0x00000000 +CONFIG_RAMSIZE=0x00000000 +CONFIG_VECTORBASE=0x00000000 +CONFIG_KERNELBASE=0x00020000 +CONFIG_RAMAUTOBIT=y +# CONFIG_RAM8BIT is not set +# CONFIG_RAM16BIT is not set +# CONFIG_RAM32BIT is not set + +# +# ROM configuration +# +# CONFIG_ROM is not set +CONFIG_RAMKERNEL=y +# CONFIG_ROMKERNEL is not set +CONFIG_SELECT_MEMORY_MODEL=y +CONFIG_FLATMEM_MANUAL=y +# CONFIG_DISCONTIGMEM_MANUAL is not set +# CONFIG_SPARSEMEM_MANUAL is not set +CONFIG_FLATMEM=y +CONFIG_FLAT_NODE_MEM_MAP=y +# CONFIG_SPARSEMEM_STATIC is not set +# CONFIG_SPARSEMEM_VMEMMAP_ENABLE is not set +CONFIG_PAGEFLAGS_EXTENDED=y +CONFIG_SPLIT_PTLOCK_CPUS=4 +# CONFIG_RESOURCES_64BIT is not set +CONFIG_ZONE_DMA_FLAG=1 +CONFIG_VIRT_TO_BUS=y +CONFIG_ISA_DMA_API=y + +# +# Bus options (PCI, PCMCIA, EISA, MCA, ISA) +# +# CONFIG_PCI is not set +# CONFIG_COMEMPCI is not set +# CONFIG_ARCH_SUPPORTS_MSI is not set + +# +# Executable file formats +# +CONFIG_BINFMT_FLAT=y +# CONFIG_BINFMT_ZFLAT is not set +# CONFIG_BINFMT_SHARED_FLAT is not set +# CONFIG_BINFMT_AOUT is not set +# CONFIG_BINFMT_MISC is not set + +# +# Power management options +# +# CONFIG_PM is not set + +# +# Networking +# +CONFIG_NET=y + +# +# Networking options +# +CONFIG_PACKET=y +# CONFIG_PACKET_MMAP is not set +CONFIG_UNIX=y +# CONFIG_NET_KEY is not set +CONFIG_INET=y +# CONFIG_IP_MULTICAST is not set +# CONFIG_IP_ADVANCED_ROUTER is not set +CONFIG_IP_FIB_HASH=y +# CONFIG_IP_PNP is not set +# CONFIG_NET_IPIP is not set +# CONFIG_NET_IPGRE is not set +# CONFIG_ARPD is not set +# CONFIG_SYN_COOKIES is not set +# CONFIG_INET_AH is not set +# CONFIG_INET_ESP is not set +# CONFIG_INET_IPCOMP is not set +# CONFIG_INET_XFRM_TUNNEL is not set +# CONFIG_INET_TUNNEL is not set +# CONFIG_INET_XFRM_MODE_TRANSPORT is not set +# CONFIG_INET_XFRM_MODE_TUNNEL is not set +# CONFIG_INET_XFRM_MODE_BEET is not set +# CONFIG_INET_LRO is not set +# CONFIG_INET_DIAG is not set +# CONFIG_TCP_CONG_ADVANCED is not set +CONFIG_TCP_CONG_CUBIC=y +CONFIG_DEFAULT_TCP_CONG="cubic" +# CONFIG_TCP_MD5SIG is not set +# CONFIG_IPV6 is not set +# CONFIG_NETWORK_SECMARK is not set +# CONFIG_NETFILTER is not set +# CONFIG_IP_DCCP is not set +# CONFIG_IP_SCTP is not set +# CONFIG_TIPC is not set +# CONFIG_ATM is not set +# CONFIG_BRIDGE is not set +# CONFIG_VLAN_8021Q is not set +# CONFIG_DECNET is not set +# CONFIG_LLC2 is not set +# CONFIG_IPX is not set +# CONFIG_ATALK is not set +# CONFIG_X25 is not set +# CONFIG_LAPB is not set +# CONFIG_ECONET is not set +# CONFIG_WAN_ROUTER is not set +# CONFIG_NET_SCHED is not set + +# +# Network testing +# +# CONFIG_NET_PKTGEN is not set +# CONFIG_HAMRADIO is not set +# CONFIG_CAN is not set +# CONFIG_IRDA is not set +# CONFIG_BT is not set +# CONFIG_AF_RXRPC is not set + +# +# Wireless +# +# CONFIG_CFG80211 is not set +# CONFIG_WIRELESS_EXT is not set +# CONFIG_MAC80211 is not set +# CONFIG_IEEE80211 is not set +# CONFIG_RFKILL is not set +# CONFIG_NET_9P is not set + +# +# Device Drivers +# + +# +# Generic Driver Options +# +CONFIG_STANDALONE=y +CONFIG_PREVENT_FIRMWARE_BUILD=y +# CONFIG_SYS_HYPERVISOR is not set +# CONFIG_CONNECTOR is not set +CONFIG_MTD=y +# CONFIG_MTD_DEBUG is not set +# CONFIG_MTD_CONCAT is not set +CONFIG_MTD_PARTITIONS=y +# CONFIG_MTD_REDBOOT_PARTS is not set +# CONFIG_MTD_CMDLINE_PARTS is not set +# CONFIG_MTD_AR7_PARTS is not set + +# +# User Modules And Translation Layers +# +CONFIG_MTD_CHAR=y +CONFIG_MTD_BLKDEVS=y +CONFIG_MTD_BLOCK=y +# CONFIG_FTL is not set +# CONFIG_NFTL is not set +# CONFIG_INFTL is not set +# CONFIG_RFD_FTL is not set +# CONFIG_SSFDC is not set +# CONFIG_MTD_OOPS is not set + +# +# RAM/ROM/Flash chip drivers +# +# CONFIG_MTD_CFI is not set +# CONFIG_MTD_JEDECPROBE is not set +CONFIG_MTD_MAP_BANK_WIDTH_1=y +CONFIG_MTD_MAP_BANK_WIDTH_2=y +CONFIG_MTD_MAP_BANK_WIDTH_4=y +# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set +# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set +# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set +CONFIG_MTD_CFI_I1=y +CONFIG_MTD_CFI_I2=y +# CONFIG_MTD_CFI_I4 is not set +# CONFIG_MTD_CFI_I8 is not set +CONFIG_MTD_RAM=y +# CONFIG_MTD_ROM is not set +# CONFIG_MTD_ABSENT is not set + +# +# Mapping drivers for chip access +# +# CONFIG_MTD_COMPLEX_MAPPINGS is not set +CONFIG_MTD_UCLINUX=y +# CONFIG_MTD_PLATRAM is not set + +# +# Self-contained MTD device drivers +# +# CONFIG_MTD_SLRAM is not set +# CONFIG_MTD_PHRAM is not set +# CONFIG_MTD_MTDRAM is not set +# CONFIG_MTD_BLOCK2MTD is not set + +# +# Disk-On-Chip Device Drivers +# +# CONFIG_MTD_DOC2000 is not set +# CONFIG_MTD_DOC2001 is not set +# CONFIG_MTD_DOC2001PLUS is not set +# CONFIG_MTD_NAND is not set +# CONFIG_MTD_ONENAND is not set + +# +# UBI - Unsorted block images +# +# CONFIG_MTD_UBI is not set +# CONFIG_PARPORT is not set +CONFIG_BLK_DEV=y +# CONFIG_BLK_DEV_COW_COMMON is not set +# CONFIG_BLK_DEV_LOOP is not set +# CONFIG_BLK_DEV_NBD is not set +CONFIG_BLK_DEV_RAM=y +CONFIG_BLK_DEV_RAM_COUNT=16 +CONFIG_BLK_DEV_RAM_SIZE=4096 +# CONFIG_BLK_DEV_XIP is not set +# CONFIG_CDROM_PKTCDVD is not set +# CONFIG_ATA_OVER_ETH is not set +# CONFIG_MISC_DEVICES is not set +CONFIG_HAVE_IDE=y +# CONFIG_IDE is not set + +# +# SCSI device support +# +# CONFIG_RAID_ATTRS is not set +# CONFIG_SCSI is not set +# CONFIG_SCSI_DMA is not set +# CONFIG_SCSI_NETLINK is not set +# CONFIG_MD is not set +CONFIG_NETDEVICES=y +# CONFIG_NETDEVICES_MULTIQUEUE is not set +# CONFIG_DUMMY is not set +# CONFIG_BONDING is not set +# CONFIG_MACVLAN is not set +# CONFIG_EQUALIZER is not set +# CONFIG_TUN is not set +# CONFIG_VETH is not set +# CONFIG_PHYLIB is not set +CONFIG_NET_ETHERNET=y +# CONFIG_MII is not set +# CONFIG_IBM_NEW_EMAC_ZMII is not set +# CONFIG_IBM_NEW_EMAC_RGMII is not set +# CONFIG_IBM_NEW_EMAC_TAH is not set +# CONFIG_IBM_NEW_EMAC_EMAC4 is not set +# CONFIG_B44 is not set +# CONFIG_NETDEV_1000 is not set +# CONFIG_NETDEV_10000 is not set + +# +# Wireless LAN +# +# CONFIG_WLAN_PRE80211 is not set +# CONFIG_WLAN_80211 is not set +# CONFIG_IWLWIFI is not set +# CONFIG_IWLWIFI_LEDS is not set +# CONFIG_WAN is not set +CONFIG_PPP=y +# CONFIG_PPP_MULTILINK is not set +# CONFIG_PPP_FILTER is not set +# CONFIG_PPP_ASYNC is not set +# CONFIG_PPP_SYNC_TTY is not set +# CONFIG_PPP_DEFLATE is not set +# CONFIG_PPP_BSDCOMP is not set +# CONFIG_PPP_MPPE is not set +# CONFIG_PPPOE is not set +# CONFIG_PPPOL2TP is not set +# CONFIG_SLIP is not set +CONFIG_SLHC=y +# CONFIG_NETCONSOLE is not set +# CONFIG_NETPOLL is not set +# CONFIG_NET_POLL_CONTROLLER is not set +# CONFIG_ISDN is not set +# CONFIG_PHONE is not set + +# +# Input device support +# +CONFIG_INPUT=y +# CONFIG_INPUT_FF_MEMLESS is not set +# CONFIG_INPUT_POLLDEV is not set + +# +# Userland interfaces +# +# CONFIG_INPUT_MOUSEDEV is not set +# CONFIG_INPUT_JOYDEV is not set +# CONFIG_INPUT_EVDEV is not set +# CONFIG_INPUT_EVBUG is not set + +# +# Input Device Drivers +# +# CONFIG_INPUT_KEYBOARD is not set +# CONFIG_INPUT_MOUSE is not set +# CONFIG_INPUT_JOYSTICK is not set +# CONFIG_INPUT_TABLET is not set +# CONFIG_INPUT_TOUCHSCREEN is not set +# CONFIG_INPUT_MISC is not set + +# +# Hardware I/O ports +# +# CONFIG_SERIO is not set +# CONFIG_GAMEPORT is not set + +# +# Character devices +# +# CONFIG_VT is not set +# CONFIG_DEVKMEM is not set +# CONFIG_SERIAL_NONSTANDARD is not set + +# +# Serial drivers +# +# CONFIG_SERIAL_8250 is not set + +# +# Non-8250 serial port support +# +CONFIG_SERIAL_CORE=y +CONFIG_SERIAL_CORE_CONSOLE=y +# CONFIG_SERIAL_COLDFIRE is not set +CONFIG_SERIAL_MCF=y +CONFIG_SERIAL_MCF_BAUDRATE=19200 +CONFIG_SERIAL_MCF_CONSOLE=y +# CONFIG_UNIX98_PTYS is not set +CONFIG_LEGACY_PTYS=y +CONFIG_LEGACY_PTY_COUNT=256 +# CONFIG_IPMI_HANDLER is not set +# CONFIG_HW_RANDOM is not set +# CONFIG_GEN_RTC is not set +# CONFIG_R3964 is not set +# CONFIG_RAW_DRIVER is not set +# CONFIG_TCG_TPM is not set +# CONFIG_I2C is not set +# CONFIG_SPI is not set +# CONFIG_W1 is not set +# CONFIG_POWER_SUPPLY is not set +# CONFIG_HWMON is not set +# CONFIG_THERMAL is not set +# CONFIG_WATCHDOG is not set + +# +# Sonics Silicon Backplane +# +CONFIG_SSB_POSSIBLE=y +# CONFIG_SSB is not set + +# +# Multifunction device drivers +# +# CONFIG_MFD_SM501 is not set +# CONFIG_HTC_PASIC3 is not set + +# +# Multimedia devices +# + +# +# Multimedia core support +# +# CONFIG_VIDEO_DEV is not set +# CONFIG_DVB_CORE is not set + +# +# Multimedia drivers +# +CONFIG_DAB=y + +# +# Graphics support +# +# CONFIG_VGASTATE is not set +# CONFIG_VIDEO_OUTPUT_CONTROL is not set +# CONFIG_FB is not set +# CONFIG_BACKLIGHT_LCD_SUPPORT is not set + +# +# Display device support +# +# CONFIG_DISPLAY_SUPPORT is not set + +# +# Sound +# +# CONFIG_SOUND is not set +# CONFIG_HID_SUPPORT is not set +# CONFIG_USB_SUPPORT is not set +# CONFIG_MMC is not set +# CONFIG_MEMSTICK is not set +# CONFIG_NEW_LEDS is not set +# CONFIG_ACCESSIBILITY is not set +# CONFIG_RTC_CLASS is not set +# CONFIG_UIO is not set + +# +# File systems +# +CONFIG_EXT2_FS=y +# CONFIG_EXT2_FS_XATTR is not set +# CONFIG_EXT3_FS is not set +# CONFIG_EXT4DEV_FS is not set +# CONFIG_REISERFS_FS is not set +# CONFIG_JFS_FS is not set +# CONFIG_FS_POSIX_ACL is not set +# CONFIG_XFS_FS is not set +# CONFIG_OCFS2_FS is not set +# CONFIG_DNOTIFY is not set +# CONFIG_INOTIFY is not set +# CONFIG_QUOTA is not set +# CONFIG_AUTOFS_FS is not set +# CONFIG_AUTOFS4_FS is not set +# CONFIG_FUSE_FS is not set + +# +# CD-ROM/DVD Filesystems +# +# CONFIG_ISO9660_FS is not set +# CONFIG_UDF_FS is not set + +# +# DOS/FAT/NT Filesystems +# +# CONFIG_MSDOS_FS is not set +# CONFIG_VFAT_FS is not set +# CONFIG_NTFS_FS is not set + +# +# Pseudo filesystems +# +CONFIG_PROC_FS=y +CONFIG_PROC_SYSCTL=y +CONFIG_SYSFS=y +# CONFIG_TMPFS is not set +# CONFIG_HUGETLB_PAGE is not set +# CONFIG_CONFIGFS_FS is not set + +# +# Miscellaneous filesystems +# +# CONFIG_ADFS_FS is not set +# CONFIG_AFFS_FS is not set +# CONFIG_HFS_FS is not set +# CONFIG_HFSPLUS_FS is not set +# CONFIG_BEFS_FS is not set +# CONFIG_BFS_FS is not set +# CONFIG_EFS_FS is not set +# CONFIG_JFFS2_FS is not set +# CONFIG_CRAMFS is not set +# CONFIG_VXFS_FS is not set +# CONFIG_MINIX_FS is not set +# CONFIG_HPFS_FS is not set +# CONFIG_QNX4FS_FS is not set +CONFIG_ROMFS_FS=y +# CONFIG_SYSV_FS is not set +# CONFIG_UFS_FS is not set +# CONFIG_NETWORK_FILESYSTEMS is not set + +# +# Partition Types +# +# CONFIG_PARTITION_ADVANCED is not set +CONFIG_MSDOS_PARTITION=y +# CONFIG_NLS is not set +# CONFIG_DLM is not set + +# +# Kernel hacking +# +# CONFIG_PRINTK_TIME is not set +CONFIG_ENABLE_WARN_DEPRECATED=y +CONFIG_ENABLE_MUST_CHECK=y +CONFIG_FRAME_WARN=1024 +# CONFIG_MAGIC_SYSRQ is not set +# CONFIG_UNUSED_SYMBOLS is not set +# CONFIG_DEBUG_FS is not set +# CONFIG_HEADERS_CHECK is not set +# CONFIG_DEBUG_KERNEL is not set +# CONFIG_DEBUG_BUGVERBOSE is not set +# CONFIG_SAMPLES is not set +# CONFIG_FULLDEBUG is not set +# CONFIG_HIGHPROFILE is not set +# CONFIG_BOOTPARAM is not set +# CONFIG_NO_KERNEL_MSG is not set +# CONFIG_BDM_DISABLE is not set + +# +# Security options +# +# CONFIG_KEYS is not set +# CONFIG_SECURITY is not set +# CONFIG_SECURITY_FILE_CAPABILITIES is not set +# CONFIG_CRYPTO is not set + +# +# Library routines +# +# CONFIG_GENERIC_FIND_FIRST_BIT is not set +# CONFIG_CRC_CCITT is not set +# CONFIG_CRC16 is not set +# CONFIG_CRC_ITU_T is not set +# CONFIG_CRC32 is not set +# CONFIG_CRC7 is not set +# CONFIG_LIBCRC32C is not set +CONFIG_HAS_IOMEM=y +CONFIG_HAS_DMA=y From 5732b38ddb770b98110ea218232fc072e5626b87 Mon Sep 17 00:00:00 2001 From: Sebastian Siewior Date: Fri, 9 May 2008 16:18:33 +0200 Subject: [PATCH 051/857] m68knommu: Add Coldfire DMA Timer support This one could be used as a hrtimer. Signed-off-by: Benedikt Spranger Signed-off-by: Sebastian Siewior Signed-off-by: Greg Ungerer --- arch/m68knommu/platform/coldfire/Makefile | 2 +- arch/m68knommu/platform/coldfire/dma_timer.c | 68 ++++++++++++++++++++ 2 files changed, 69 insertions(+), 1 deletion(-) create mode 100644 arch/m68knommu/platform/coldfire/dma_timer.c diff --git a/arch/m68knommu/platform/coldfire/Makefile b/arch/m68knommu/platform/coldfire/Makefile index 40cf20be1b90..4f416a91a829 100644 --- a/arch/m68knommu/platform/coldfire/Makefile +++ b/arch/m68knommu/platform/coldfire/Makefile @@ -18,7 +18,7 @@ obj-$(CONFIG_COLDFIRE) += dma.o entry.o vectors.o obj-$(CONFIG_M5206) += timers.o obj-$(CONFIG_M5206e) += timers.o obj-$(CONFIG_M520x) += pit.o -obj-$(CONFIG_M523x) += pit.o +obj-$(CONFIG_M523x) += pit.o dma_timer.o obj-$(CONFIG_M5249) += timers.o obj-$(CONFIG_M527x) += pit.o obj-$(CONFIG_M5272) += timers.o diff --git a/arch/m68knommu/platform/coldfire/dma_timer.c b/arch/m68knommu/platform/coldfire/dma_timer.c new file mode 100644 index 000000000000..b623c993219e --- /dev/null +++ b/arch/m68knommu/platform/coldfire/dma_timer.c @@ -0,0 +1,68 @@ +/* + * dma_timer.c -- Freescale ColdFire DMA Timer. + * + * Copyright (C) 2007, Benedikt Spranger + * Copyright (C) 2008. Sebastian Siewior, Linutronix + * + */ + +#include +#include + +#include +#include +#include +#include + +#define DMA_TIMER_0 (0x00) +#define DMA_TIMER_1 (0x40) +#define DMA_TIMER_2 (0x80) +#define DMA_TIMER_3 (0xc0) + +#define DTMR0 (MCF_IPSBAR + DMA_TIMER_0 + 0x400) +#define DTXMR0 (MCF_IPSBAR + DMA_TIMER_0 + 0x402) +#define DTER0 (MCF_IPSBAR + DMA_TIMER_0 + 0x403) +#define DTRR0 (MCF_IPSBAR + DMA_TIMER_0 + 0x404) +#define DTCR0 (MCF_IPSBAR + DMA_TIMER_0 + 0x408) +#define DTCN0 (MCF_IPSBAR + DMA_TIMER_0 + 0x40c) + +#define DMA_FREQ ((MCF_CLK / 2) / 16) + +/* DTMR */ +#define DMA_DTMR_RESTART (1 << 3) +#define DMA_DTMR_CLK_DIV_1 (1 << 1) +#define DMA_DTMR_CLK_DIV_16 (2 << 1) +#define DMA_DTMR_ENABLE (1 << 0) + +static cycle_t cf_dt_get_cycles(void) +{ + return __raw_readl(DTCN0); +} + +static struct clocksource clocksource_cf_dt = { + .name = "coldfire_dma_timer", + .rating = 200, + .read = cf_dt_get_cycles, + .mask = CLOCKSOURCE_MASK(32), + .shift = 20, + .flags = CLOCK_SOURCE_IS_CONTINUOUS, +}; + +static int __init init_cf_dt_clocksource(void) +{ + /* + * We setup DMA timer 0 in free run mode. This incrementing counter is + * used as a highly precious clock source. With MCF_CLOCK = 150 MHz we + * get a ~213 ns resolution and the 32bit register will overflow almost + * every 15 minutes. + */ + __raw_writeb(0x00, DTXMR0); + __raw_writeb(0x00, DTER0); + __raw_writel(0x00000000, DTRR0); + __raw_writew(DMA_DTMR_CLK_DIV_16 | DMA_DTMR_ENABLE, DTMR0); + clocksource_cf_dt.mult = clocksource_hz2mult(DMA_FREQ, + clocksource_cf_dt.shift); + return clocksource_register(&clocksource_cf_dt); +} + +arch_initcall(init_cf_dt_clocksource); From 1fda83d83c664ad74bfec8ce093a86d4d962f093 Mon Sep 17 00:00:00 2001 From: Sebastian Siewior Date: Fri, 9 May 2008 16:13:36 +0200 Subject: [PATCH 052/857] m68knommu: m68knommu: add old stack trace method The old method is used when frame pointers are not available. Also fix formating with CONFIG_KALLSYMS=n which eliminates \n. Signed-off-by: Sebastian Siewior Signed-off-by: Greg Ungerer --- arch/m68knommu/kernel/traps.c | 38 ++++++++++++++++++++++++++++++++--- 1 file changed, 35 insertions(+), 3 deletions(-) diff --git a/arch/m68knommu/kernel/traps.c b/arch/m68knommu/kernel/traps.c index ec9aea652e79..46f8f9d0c408 100644 --- a/arch/m68knommu/kernel/traps.c +++ b/arch/m68knommu/kernel/traps.c @@ -103,12 +103,28 @@ asmlinkage void buserr_c(struct frame *fp) force_sig(SIGSEGV, current); } +static void print_this_address(unsigned long addr, int i) +{ +#ifdef CONFIG_KALLSYMS + printk(KERN_EMERG " [%08lx] ", addr); + print_symbol(KERN_CONT "%s\n", addr); +#else + if (i % 5) + printk(KERN_CONT " [%08lx] ", addr); + else + printk(KERN_CONT "\n" KERN_EMERG " [%08lx] ", addr); + i++; +#endif +} + int kstack_depth_to_print = 48; static void __show_stack(struct task_struct *task, unsigned long *stack) { unsigned long *endstack, addr; +#ifdef CONFIG_FRAME_POINTER unsigned long *last_stack; +#endif int i; if (!stack) @@ -126,6 +142,7 @@ static void __show_stack(struct task_struct *task, unsigned long *stack) printk(" %08lx", *(stack + i)); } printk("\n"); + i = 0; #ifdef CONFIG_FRAME_POINTER printk(KERN_EMERG "Call Trace:\n"); @@ -134,15 +151,30 @@ static void __show_stack(struct task_struct *task, unsigned long *stack) while (stack <= endstack && stack > last_stack) { addr = *(stack + 1); - printk(KERN_EMERG " [%08lx] ", addr); - print_symbol(KERN_CONT "%s\n", addr); + print_this_address(addr, i); + i++; last_stack = stack; stack = (unsigned long *)*stack; } printk("\n"); #else - printk(KERN_EMERG "CONFIG_FRAME_POINTER disabled, no symbolic call trace\n"); + printk(KERN_EMERG "Call Trace with CONFIG_FRAME_POINTER disabled:\n"); + while (stack <= endstack) { + addr = *stack++; + /* + * If the address is either in the text segment of the kernel, + * or in a region which is occupied by a module then it *may* + * be the address of a calling routine; if so, print it so that + * someone tracing down the cause of the crash will be able to + * figure out the call path that was taken. + */ + if (__kernel_text_address(addr)) { + print_this_address(addr, i); + i++; + } + } + printk(KERN_CONT "\n"); #endif } From 0df185f5a1430ab8b437be402d286ee0728ef9f8 Mon Sep 17 00:00:00 2001 From: Sebastian Siewior Date: Mon, 28 Apr 2008 11:43:00 +0200 Subject: [PATCH 053/857] m68knommu: move code within time.c This patch creates two functions do_set_rtc() and read_rtc_mmss() based on allready available code. Signed-off-by: Sebastian Siewior Signed-off-by: Greg Ungerer --- arch/m68knommu/kernel/time.c | 40 +++++++++++++++++++++++------------- 1 file changed, 26 insertions(+), 14 deletions(-) diff --git a/arch/m68knommu/kernel/time.c b/arch/m68knommu/kernel/time.c index 0ccfb2ad6380..d33ed9a84cc7 100644 --- a/arch/m68knommu/kernel/time.c +++ b/arch/m68knommu/kernel/time.c @@ -33,22 +33,11 @@ static inline int set_rtc_mmss(unsigned long nowtime) return -1; } -/* - * timer_interrupt() needs to keep up the real-time clock, - * as well as call the "do_timer()" routine every clocktick - */ -irqreturn_t arch_timer_interrupt(int irq, void *dummy) +static inline void do_set_rtc(void) { /* last time the cmos clock got updated */ static long last_rtc_update=0; - if (current->pid) - profile_tick(CPU_PROFILING); - - write_seqlock(&xtime_lock); - - do_timer(1); - /* * If we have an externally synchronized Linux clock, then update * CMOS clock accordingly every ~11 minutes. Set_rtc_mmss() has to be @@ -63,6 +52,23 @@ irqreturn_t arch_timer_interrupt(int irq, void *dummy) else last_rtc_update = xtime.tv_sec - 600; /* do it again in 60 s */ } +} + +/* + * timer_interrupt() needs to keep up the real-time clock, + * as well as call the "do_timer()" routine every clocktick + */ +irqreturn_t arch_timer_interrupt(int irq, void *dummy) +{ + + if (current->pid) + profile_tick(CPU_PROFILING); + + write_seqlock(&xtime_lock); + + do_timer(1); + + do_set_rtc(); write_sequnlock(&xtime_lock); @@ -72,7 +78,7 @@ irqreturn_t arch_timer_interrupt(int irq, void *dummy) return(IRQ_HANDLED); } -void time_init(void) +static unsigned long read_rtc_mmss(void) { unsigned int year, mon, day, hour, min, sec; @@ -83,7 +89,13 @@ void time_init(void) if ((year += 1900) < 1970) year += 100; - xtime.tv_sec = mktime(year, mon, day, hour, min, sec); + + return mktime(year, mon, day, hour, min, sec);; +} + +void time_init(void) +{ + xtime.tv_sec = read_rtc_mmss(); xtime.tv_nsec = 0; wall_to_monotonic.tv_sec = -xtime.tv_sec; From 95469bd64a7a9ab405b566deb8c81d4aaf67ed9e Mon Sep 17 00:00:00 2001 From: Sebastian Siewior Date: Mon, 28 Apr 2008 11:43:01 +0200 Subject: [PATCH 054/857] m68knommu: complete generic time do_set_rtc() isn't required because the work that is handled is allready served if read_persistent_clock() & update_persistent_clock() are implemented and CONFIG_GENERIC_CMOS_UPDATE is. sync_cmos_clock() looks very familiar :) Signed-off-by: Sebastian Siewior Signed-off-by: Greg Ungerer --- arch/m68knommu/Kconfig | 4 ++++ arch/m68knommu/kernel/time.c | 40 ++++++++++-------------------------- 2 files changed, 15 insertions(+), 29 deletions(-) diff --git a/arch/m68knommu/Kconfig b/arch/m68knommu/Kconfig index 8e8441587c22..bfd35304d586 100644 --- a/arch/m68knommu/Kconfig +++ b/arch/m68knommu/Kconfig @@ -58,6 +58,10 @@ config GENERIC_TIME bool default y +config GENERIC_CMOS_UPDATE + bool + default y + config TIME_LOW_RES bool default y diff --git a/arch/m68knommu/kernel/time.c b/arch/m68knommu/kernel/time.c index d33ed9a84cc7..67944aa27280 100644 --- a/arch/m68knommu/kernel/time.c +++ b/arch/m68knommu/kernel/time.c @@ -33,27 +33,6 @@ static inline int set_rtc_mmss(unsigned long nowtime) return -1; } -static inline void do_set_rtc(void) -{ - /* last time the cmos clock got updated */ - static long last_rtc_update=0; - - /* - * If we have an externally synchronized Linux clock, then update - * CMOS clock accordingly every ~11 minutes. Set_rtc_mmss() has to be - * called as close as possible to 500 ms before the new second starts. - */ - if (ntp_synced() && - xtime.tv_sec > last_rtc_update + 660 && - (xtime.tv_nsec / 1000) >= 500000 - ((unsigned) TICK_SIZE) / 2 && - (xtime.tv_nsec / 1000) <= 500000 + ((unsigned) TICK_SIZE) / 2) { - if (set_rtc_mmss(xtime.tv_sec) == 0) - last_rtc_update = xtime.tv_sec; - else - last_rtc_update = xtime.tv_sec - 600; /* do it again in 60 s */ - } -} - /* * timer_interrupt() needs to keep up the real-time clock, * as well as call the "do_timer()" routine every clocktick @@ -68,8 +47,6 @@ irqreturn_t arch_timer_interrupt(int irq, void *dummy) do_timer(1); - do_set_rtc(); - write_sequnlock(&xtime_lock); #ifndef CONFIG_SMP @@ -93,12 +70,17 @@ static unsigned long read_rtc_mmss(void) return mktime(year, mon, day, hour, min, sec);; } -void time_init(void) +unsigned long read_persistent_clock(void) { - xtime.tv_sec = read_rtc_mmss(); - xtime.tv_nsec = 0; - wall_to_monotonic.tv_sec = -xtime.tv_sec; - - hw_timer_init(); + return read_rtc_mmss(); } +int update_persistent_clock(struct timespec now) +{ + return set_rtc_mmss(now.tv_sec); +} + +void time_init(void) +{ + hw_timer_init(); +} From 5bed10a5ee272fbf18ce0ce764245bbb8f28e2e6 Mon Sep 17 00:00:00 2001 From: Sebastian Siewior Date: Mon, 28 Apr 2008 11:43:03 +0200 Subject: [PATCH 055/857] m68knommu: add sched_clock() for the DMA timer with this printk() and other sched_clock() user use the more precise timestamps. The highly optimized math is from arch/x86/kernel/tsc_32.c. Signed-off-by: Sebastian Siewior Signed-off-by: Greg Ungerer --- arch/m68knommu/platform/coldfire/dma_timer.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/arch/m68knommu/platform/coldfire/dma_timer.c b/arch/m68knommu/platform/coldfire/dma_timer.c index b623c993219e..772578b1084f 100644 --- a/arch/m68knommu/platform/coldfire/dma_timer.c +++ b/arch/m68knommu/platform/coldfire/dma_timer.c @@ -66,3 +66,19 @@ static int __init init_cf_dt_clocksource(void) } arch_initcall(init_cf_dt_clocksource); + +#define CYC2NS_SCALE_FACTOR 10 /* 2^10, carefully chosen */ +#define CYC2NS_SCALE ((1000000 << CYC2NS_SCALE_FACTOR) / (DMA_FREQ / 1000)) + +static unsigned long long cycles2ns(unsigned long cycl) +{ + return (unsigned long long) ((unsigned long long)cycl * + CYC2NS_SCALE) >> CYC2NS_SCALE_FACTOR; +} + +unsigned long long sched_clock(void) +{ + unsigned long cycl = __raw_readl(DTCN0); + + return cycles2ns(cycl); +} From a6260ef84103fa8a51a67b6a58e5e16c676e08ad Mon Sep 17 00:00:00 2001 From: Sebastian Siewior Date: Fri, 9 May 2008 16:10:37 +0200 Subject: [PATCH 056/857] m68knommu: add ffs and __ffs plattform which support ISA A+ or ISA C the ff1 and bitrev opcode appears in ISA C and ISA A+ what isn't supported by all plattforms. The assembly optimization is automaticly enabled if the compiler understand the required cpu keyword. My m5235 seems to boot and run fine so far. Signed-off-by: Sebastian Siewior Signed-off-by: Greg Ungerer --- arch/m68knommu/Makefile | 11 +++++++---- include/asm-m68knommu/bitops.h | 30 ++++++++++++++++++++++++++++++ 2 files changed, 37 insertions(+), 4 deletions(-) diff --git a/arch/m68knommu/Makefile b/arch/m68knommu/Makefile index e0b5f62e395c..b63bbcf874ff 100644 --- a/arch/m68knommu/Makefile +++ b/arch/m68knommu/Makefile @@ -8,6 +8,8 @@ # (C) Copyright 2002, Greg Ungerer # +KBUILD_DEFCONFIG := m5208evb_defconfig + platform-$(CONFIG_M68328) := 68328 platform-$(CONFIG_M68EZ328) := 68EZ328 platform-$(CONFIG_M68VZ328) := 68VZ328 @@ -90,13 +92,14 @@ export PLATFORM BOARD MODEL CPUCLASS cflags-$(CONFIG_M5206) := -m5200 cflags-$(CONFIG_M5206e) := -m5200 cflags-$(CONFIG_M520x) := -m5307 -cflags-$(CONFIG_M523x) := -m5307 +cflags-$(CONFIG_M523x) := $(call cc-option,-mcpu=523x,-m5307) cflags-$(CONFIG_M5249) := -m5200 -cflags-$(CONFIG_M527x) := -m5307 +cflags-$(CONFIG_M5271) := $(call cc-option,-mcpu=5271,-m5307) cflags-$(CONFIG_M5272) := -m5307 -cflags-$(CONFIG_M528x) := -m5307 +cflags-$(CONFIG_M5275) := $(call cc-option,-mcpu=5275,-m5307) +cflags-$(CONFIG_M528x) := $(call cc-option,-m528x,-m5307) cflags-$(CONFIG_M5307) := -m5307 -cflags-$(CONFIG_M532x) := -m5307 +cflags-$(CONFIG_M532x) := $(call cc-option,-mcpu=532x,-m5307) cflags-$(CONFIG_M5407) := -m5200 cflags-$(CONFIG_M68328) := -m68000 cflags-$(CONFIG_M68EZ328) := -m68000 diff --git a/include/asm-m68knommu/bitops.h b/include/asm-m68knommu/bitops.h index c142fbf2f376..6f3685eab44c 100644 --- a/include/asm-m68knommu/bitops.h +++ b/include/asm-m68knommu/bitops.h @@ -14,8 +14,38 @@ #error only can be included directly #endif +#if defined (__mcfisaaplus__) || defined (__mcfisac__) +static inline int ffs(unsigned int val) +{ + if (!val) + return 0; + + asm volatile( + "bitrev %0\n\t" + "ff1 %0\n\t" + : "=d" (val) + : "0" (val) + ); + val++; + return val; +} + +static inline int __ffs(unsigned int val) +{ + asm volatile( + "bitrev %0\n\t" + "ff1 %0\n\t" + : "=d" (val) + : "0" (val) + ); + return val; +} + +#else #include #include +#endif + #include #include From e872504b311cec52f7a316a0037fb959080dbea0 Mon Sep 17 00:00:00 2001 From: Sebastian Siewior Date: Sat, 17 May 2008 21:51:15 +0200 Subject: [PATCH 057/857] m68knommu: add byteswap assembly opcode for ISA A+ Signed-off-by: Sebastian Siewior Signed-off-by: Greg Ungerer --- include/asm-m68knommu/byteorder.h | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-) diff --git a/include/asm-m68knommu/byteorder.h b/include/asm-m68knommu/byteorder.h index 8fcde907b0f9..20bb4426b610 100644 --- a/include/asm-m68knommu/byteorder.h +++ b/include/asm-m68knommu/byteorder.h @@ -1,13 +1,27 @@ #ifndef _M68KNOMMU_BYTEORDER_H #define _M68KNOMMU_BYTEORDER_H -#include +#include #if defined(__GNUC__) && !defined(__STRICT_ANSI__) || defined(__KERNEL__) # define __BYTEORDER_HAS_U64__ # define __SWAB_64_THRU_32__ #endif +#if defined (__mcfisaaplus__) || defined (__mcfisac__) +static inline __attribute_const__ __u32 ___arch__swab32(__u32 val) +{ + asm( + "byterev %0" + : "=d" (val) + : "0" (val) + ); + return val; +} + +#define __arch__swab32(x) ___arch__swab32(x) +#endif + #include #endif /* _M68KNOMMU_BYTEORDER_H */ From 6dbeb456baaba05d60e7ca8213da26142062408a Mon Sep 17 00:00:00 2001 From: Sebastian Siewior Date: Tue, 13 May 2008 18:52:44 +0200 Subject: [PATCH 058/857] m68knommu: add read_barrier_depends() and irqs_disabled_flags() /home/bigeasy/git/linux-2.6-ftrace/kernel/trace/trace.c: In function 'tracing_generic_entry_update': /home/bigeasy/git/linux-2.6-ftrace/kernel/trace/trace.c:802: error: implicit declaration of function 'irqs_disabled_flags' make[3]: *** [kernel/trace/trace.o] Error 1 /home/bigeasy/git/linux-2.6-ftrace/kernel/trace/ftrace.c: In function 'ftrace_list_func': /home/bigeasy/git/linux-2.6-ftrace/kernel/trace/ftrace.c:61: error: implicit declaration of function 'read_barrier_depends' Signed-off-by: Sebastian Siewior Signed-off-by: Greg Ungerer --- include/asm-m68knommu/system.h | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/include/asm-m68knommu/system.h b/include/asm-m68knommu/system.h index 64c64432bbb8..40f49de69821 100644 --- a/include/asm-m68knommu/system.h +++ b/include/asm-m68knommu/system.h @@ -118,6 +118,8 @@ asmlinkage void resume(void); #define smp_read_barrier_depends() do { } while(0) #endif +#define read_barrier_depends() ((void)0) + #define xchg(ptr,x) ((__typeof__(*(ptr)))__xchg((unsigned long)(x),(ptr),sizeof(*(ptr)))) struct __xchg_dummy { unsigned long a[100]; }; @@ -310,4 +312,13 @@ static inline unsigned long __xchg(unsigned long x, volatile void * ptr, int siz #endif #define arch_align_stack(x) (x) + +static inline int irqs_disabled_flags(unsigned long flags) +{ + if (flags & 0x0700) + return 0; + else + return 1; +} + #endif /* _M68KNOMMU_SYSTEM_H */ From 2b9a69861c39ae4c232385def833816acc07a0a4 Mon Sep 17 00:00:00 2001 From: Sebastian Siewior Date: Mon, 28 Apr 2008 11:43:04 +0200 Subject: [PATCH 059/857] m68knommu: MCF5307 PIT GENERIC_CLOCKEVENTS support The PIT code has been changed in order to suppport GENERIC_CLOCKEVENTS. The priority of the PIT clocksource has been decreased in favor of the DMA timer. pit_cycles_per_jiffy become a constant (PIT_CYCLES_PER_JIFFY) because it is known at compile time and does not change afterwards. Signed-off-by: Benedikt Spranger Signed-off-by: Sebastian Siewior Signed-off-by: Greg Ungerer --- arch/m68knommu/Kconfig | 11 ++++ arch/m68knommu/kernel/time.c | 2 + arch/m68knommu/platform/coldfire/pit.c | 91 ++++++++++++++++++++++---- 3 files changed, 91 insertions(+), 13 deletions(-) diff --git a/arch/m68knommu/Kconfig b/arch/m68knommu/Kconfig index bfd35304d586..2e7515e8db98 100644 --- a/arch/m68knommu/Kconfig +++ b/arch/m68knommu/Kconfig @@ -66,6 +66,10 @@ config TIME_LOW_RES bool default y +config GENERIC_CLOCKEVENTS + bool + default n + config NO_IOPORT def_bool y @@ -112,11 +116,13 @@ config M5206e config M520x bool "MCF520x" + select GENERIC_CLOCKEVENTS help Freescale Coldfire 5207/5208 processor support. config M523x bool "MCF523x" + select GENERIC_CLOCKEVENTS help Freescale Coldfire 5230/1/2/4/5 processor support @@ -142,6 +148,7 @@ config M5275 config M528x bool "MCF528x" + select GENERIC_CLOCKEVENTS help Motorola ColdFire 5280/5282 processor support. @@ -165,6 +172,7 @@ endchoice config M527x bool depends on (M5271 || M5275) + select GENERIC_CLOCKEVENTS default y config COLDFIRE @@ -678,6 +686,9 @@ endchoice if COLDFIRE source "kernel/Kconfig.preempt" endif + +source "kernel/time/Kconfig" + source "mm/Kconfig" endmenu diff --git a/arch/m68knommu/kernel/time.c b/arch/m68knommu/kernel/time.c index 67944aa27280..d182b2f72211 100644 --- a/arch/m68knommu/kernel/time.c +++ b/arch/m68knommu/kernel/time.c @@ -33,6 +33,7 @@ static inline int set_rtc_mmss(unsigned long nowtime) return -1; } +#ifndef CONFIG_GENERIC_CLOCKEVENTS /* * timer_interrupt() needs to keep up the real-time clock, * as well as call the "do_timer()" routine every clocktick @@ -54,6 +55,7 @@ irqreturn_t arch_timer_interrupt(int irq, void *dummy) #endif return(IRQ_HANDLED); } +#endif static unsigned long read_rtc_mmss(void) { diff --git a/arch/m68knommu/platform/coldfire/pit.c b/arch/m68knommu/platform/coldfire/pit.c index 4290638012e0..c5b916700b22 100644 --- a/arch/m68knommu/platform/coldfire/pit.c +++ b/arch/m68knommu/platform/coldfire/pit.c @@ -18,7 +18,7 @@ #include #include #include -#include +#include #include #include #include @@ -33,22 +33,86 @@ #define FREQ ((MCF_CLK / 2) / 64) #define TA(a) (MCF_IPSBAR + MCFPIT_BASE1 + (a)) #define INTC0 (MCF_IPSBAR + MCFICM_INTC0) +#define PIT_CYCLES_PER_JIFFY (FREQ / HZ) -static u32 pit_cycles_per_jiffy; static u32 pit_cnt; +/* + * Initialize the PIT timer. + * + * This is also called after resume to bring the PIT into operation again. + */ + +static void init_cf_pit_timer(enum clock_event_mode mode, + struct clock_event_device *evt) +{ + switch (mode) { + case CLOCK_EVT_MODE_PERIODIC: + + __raw_writew(MCFPIT_PCSR_DISABLE, TA(MCFPIT_PCSR)); + __raw_writew(PIT_CYCLES_PER_JIFFY, TA(MCFPIT_PMR)); + __raw_writew(MCFPIT_PCSR_EN | MCFPIT_PCSR_PIE | \ + MCFPIT_PCSR_OVW | MCFPIT_PCSR_RLD | \ + MCFPIT_PCSR_CLK64, TA(MCFPIT_PCSR)); + break; + + case CLOCK_EVT_MODE_SHUTDOWN: + case CLOCK_EVT_MODE_UNUSED: + + __raw_writew(MCFPIT_PCSR_DISABLE, TA(MCFPIT_PCSR)); + break; + + case CLOCK_EVT_MODE_ONESHOT: + + __raw_writew(MCFPIT_PCSR_DISABLE, TA(MCFPIT_PCSR)); + __raw_writew(MCFPIT_PCSR_EN | MCFPIT_PCSR_PIE | \ + MCFPIT_PCSR_OVW | MCFPIT_PCSR_CLK64, \ + TA(MCFPIT_PCSR)); + break; + + case CLOCK_EVT_MODE_RESUME: + /* Nothing to do here */ + break; + } +} + +/* + * Program the next event in oneshot mode + * + * Delta is given in PIT ticks + */ +static int cf_pit_next_event(unsigned long delta, + struct clock_event_device *evt) +{ + __raw_writew(delta, TA(MCFPIT_PMR)); + return 0; +} + +struct clock_event_device cf_pit_clockevent = { + .name = "pit", + .features = CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT, + .set_mode = init_cf_pit_timer, + .set_next_event = cf_pit_next_event, + .shift = 32, + .irq = MCFINT_VECBASE + MCFINT_PIT1, +}; + + + /***************************************************************************/ static irqreturn_t pit_tick(int irq, void *dummy) { + struct clock_event_device *evt = &cf_pit_clockevent; u16 pcsr; /* Reset the ColdFire timer */ pcsr = __raw_readw(TA(MCFPIT_PCSR)); __raw_writew(pcsr | MCFPIT_PCSR_PIF, TA(MCFPIT_PCSR)); - pit_cnt += pit_cycles_per_jiffy; - return arch_timer_interrupt(irq, dummy); + pit_cnt += PIT_CYCLES_PER_JIFFY; + evt->event_handler(evt); + return IRQ_HANDLED; } /***************************************************************************/ @@ -72,14 +136,14 @@ static cycle_t pit_read_clk(void) cycles = pit_cnt; local_irq_restore(flags); - return cycles + pit_cycles_per_jiffy - pcntr; + return cycles + PIT_CYCLES_PER_JIFFY - pcntr; } /***************************************************************************/ static struct clocksource pit_clk = { .name = "pit", - .rating = 250, + .rating = 100, .read = pit_read_clk, .shift = 20, .mask = CLOCKSOURCE_MASK(32), @@ -92,6 +156,14 @@ void hw_timer_init(void) { u32 imr; + cf_pit_clockevent.cpumask = cpumask_of_cpu(smp_processor_id()); + cf_pit_clockevent.mult = div_sc(FREQ, NSEC_PER_SEC, 32); + cf_pit_clockevent.max_delta_ns = + clockevent_delta2ns(0xFFFF, &cf_pit_clockevent); + cf_pit_clockevent.min_delta_ns = + clockevent_delta2ns(0x3f, &cf_pit_clockevent); + clockevents_register_device(&cf_pit_clockevent); + setup_irq(MCFINT_VECBASE + MCFINT_PIT1, &pit_irq); __raw_writeb(ICR_INTRCONF, INTC0 + MCFINTC_ICR0 + MCFINT_PIT1); @@ -99,13 +171,6 @@ void hw_timer_init(void) imr &= ~MCFPIT_IMR_IBIT; __raw_writel(imr, INTC0 + MCFPIT_IMR); - /* Set up PIT timer 1 as poll clock */ - pit_cycles_per_jiffy = FREQ / HZ; - __raw_writew(MCFPIT_PCSR_DISABLE, TA(MCFPIT_PCSR)); - __raw_writew(pit_cycles_per_jiffy, TA(MCFPIT_PMR)); - __raw_writew(MCFPIT_PCSR_EN | MCFPIT_PCSR_PIE | MCFPIT_PCSR_OVW | - MCFPIT_PCSR_RLD | MCFPIT_PCSR_CLK64, TA(MCFPIT_PCSR)); - pit_clk.mult = clocksource_hz2mult(FREQ, pit_clk.shift); clocksource_register(&pit_clk); } From 0d176af5b7e658490b75427ccd23ff6a158c472b Mon Sep 17 00:00:00 2001 From: Sebastian Siewior Date: Mon, 28 Apr 2008 11:43:20 +0200 Subject: [PATCH 060/857] m68knommu: fec: remove FADS I found config FADS only in ppc/Kconfig. Bye bye relic. Signed-off-by: Sebastian Siewior Signed-off-by: Greg Ungerer --- drivers/net/fec.c | 10 ---------- 1 file changed, 10 deletions(-) diff --git a/drivers/net/fec.c b/drivers/net/fec.c index 32a4f17d35fc..6abbcd5f7c3a 100644 --- a/drivers/net/fec.c +++ b/drivers/net/fec.c @@ -2,12 +2,6 @@ * Fast Ethernet Controller (FEC) driver for Motorola MPC8xx. * Copyright (c) 1997 Dan Malek (dmalek@jlc.net) * - * This version of the driver is specific to the FADS implementation, - * since the board contains control registers external to the processor - * for the control of the LevelOne LXT970 transceiver. The MPC860T manual - * describes connections using the internal parallel port I/O, which - * is basically all of Port D. - * * Right now, I am very wasteful with the buffers. I allocate memory * pages and then divide them into 2K frame buffers. This way I know I * have buffers large enough to hold one frame within one buffer descriptor. @@ -1809,10 +1803,6 @@ static void __inline__ fec_request_intrs(struct net_device *dev) */ *((uint *) RPX_CSR_ADDR) &= ~BCSR2_FETHLEDMODE; #endif -#ifdef CONFIG_FADS - if (request_8xxirq(SIU_IRQ2, mii_link_interrupt, 0, "mii", dev) != 0) - panic("Could not allocate MII IRQ!"); -#endif } static void __inline__ fec_get_mac(struct net_device *dev) From c1863bed8c88324405dc2a922c153fe5d7df716c Mon Sep 17 00:00:00 2001 From: Sebastian Siewior Date: Mon, 28 Apr 2008 11:43:17 +0200 Subject: [PATCH 061/857] m68knommu: remove RPXCLASSIC from the m68k tree This ifdefs are leftovers from the time as the driver was running on a ppc. Signed-off-by: Sebastian Siewior Signed-off-by: Greg Ungerer --- drivers/net/fec.c | 42 -------------------------------- include/asm-m68knommu/commproc.h | 19 --------------- 2 files changed, 61 deletions(-) diff --git a/drivers/net/fec.c b/drivers/net/fec.c index 6abbcd5f7c3a..0ef7226efd5c 100644 --- a/drivers/net/fec.c +++ b/drivers/net/fec.c @@ -43,17 +43,9 @@ #include #include -#if defined(CONFIG_M523x) || defined(CONFIG_M527x) || \ - defined(CONFIG_M5272) || defined(CONFIG_M528x) || \ - defined(CONFIG_M520x) || defined(CONFIG_M532x) #include #include #include "fec.h" -#else -#include -#include -#include "commproc.h" -#endif #if defined(CONFIG_FEC2) #define FEC_MAX_PORTS 2 @@ -1229,14 +1221,9 @@ static phy_info_t const * const phy_info[] = { /* ------------------------------------------------------------------------- */ #ifdef HAVE_mii_link_interrupt -#ifdef CONFIG_RPXCLASSIC -static void -mii_link_interrupt(void *dev_id); -#else static irqreturn_t mii_link_interrupt(int irq, void * dev_id); #endif -#endif #if defined(CONFIG_M5272) /* @@ -1789,20 +1776,6 @@ static void __inline__ fec_request_intrs(struct net_device *dev) if (request_8xxirq(FEC_INTERRUPT, fec_enet_interrupt, 0, "fec", dev) != 0) panic("Could not allocate FEC IRQ!"); - -#ifdef CONFIG_RPXCLASSIC - /* Make Port C, bit 15 an input that causes interrupts. - */ - immap->im_ioport.iop_pcpar &= ~0x0001; - immap->im_ioport.iop_pcdir &= ~0x0001; - immap->im_ioport.iop_pcso &= ~0x0001; - immap->im_ioport.iop_pcint |= 0x0001; - cpm_install_handler(CPMVEC_PIO_PC15, mii_link_interrupt, dev); - - /* Make LEDS reflect Link status. - */ - *((uint *) RPX_CSR_ADDR) &= ~BCSR2_FETHLEDMODE; -#endif } static void __inline__ fec_get_mac(struct net_device *dev) @@ -1811,16 +1784,6 @@ static void __inline__ fec_get_mac(struct net_device *dev) bd = (bd_t *)__res; memcpy(dev->dev_addr, bd->bi_enetaddr, ETH_ALEN); - -#ifdef CONFIG_RPXCLASSIC - /* The Embedded Planet boards have only one MAC address in - * the EEPROM, but can have two Ethernet ports. For the - * FEC port, we create another address by setting one of - * the address bits above something that would have (up to - * now) been allocated. - */ - dev->dev_adrd[3] |= 0x80; -#endif } static void __inline__ fec_set_mii(struct net_device *dev, struct fec_enet_private *fep) @@ -2099,13 +2062,8 @@ mii_discover_phy(uint mii_reg, struct net_device *dev) /* This interrupt occurs when the PHY detects a link change. */ #ifdef HAVE_mii_link_interrupt -#ifdef CONFIG_RPXCLASSIC -static void -mii_link_interrupt(void *dev_id) -#else static irqreturn_t mii_link_interrupt(int irq, void * dev_id) -#endif { struct net_device *dev = dev_id; struct fec_enet_private *fep = netdev_priv(dev); diff --git a/include/asm-m68knommu/commproc.h b/include/asm-m68knommu/commproc.h index 36e870b468ef..edf5eb6c08d2 100644 --- a/include/asm-m68knommu/commproc.h +++ b/include/asm-m68knommu/commproc.h @@ -519,25 +519,6 @@ typedef struct scc_enet { #define SICR_ENET_CLKRT ((uint)0x00002c00) #endif -#ifdef CONFIG_RPXCLASSIC -/* Bits in parallel I/O port registers that have to be set/cleared - * to configure the pins for SCC1 use. - */ -#define PA_ENET_RXD ((ushort)0x0001) -#define PA_ENET_TXD ((ushort)0x0002) -#define PA_ENET_TCLK ((ushort)0x0200) -#define PA_ENET_RCLK ((ushort)0x0800) -#define PB_ENET_TENA ((uint)0x00001000) -#define PC_ENET_CLSN ((ushort)0x0010) -#define PC_ENET_RENA ((ushort)0x0020) - -/* Control bits in the SICR to route TCLK (CLK2) and RCLK (CLK4) to - * SCC1. Also, make sure GR1 (bit 24) and SC1 (bit 25) are zero. - */ -#define SICR_ENET_MASK ((uint)0x000000ff) -#define SICR_ENET_CLKRT ((uint)0x0000003d) -#endif - /* SCC Event register as used by Ethernet. */ #define SCCE_ENET_GRA ((ushort)0x0080) /* Graceful stop complete */ From 87f4abb45bc640638e6986f0f4d412b2d0ea21e1 Mon Sep 17 00:00:00 2001 From: Greg Ungerer Date: Fri, 6 Jun 2008 15:55:36 +1000 Subject: [PATCH 062/857] m68knommu: remove last use of CONFIG_FADS and CONFIG_RPXCLASSIC They have never been used in this port of the driver. It is has only ever been used on the ColdFire SoC ethernet core. Signed-off-by: Greg Ungerer --- drivers/net/fec.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/fec.c b/drivers/net/fec.c index 0ef7226efd5c..ecd5c71a7a8a 100644 --- a/drivers/net/fec.c +++ b/drivers/net/fec.c @@ -53,7 +53,7 @@ #define FEC_MAX_PORTS 1 #endif -#if defined(CONFIG_FADS) || defined(CONFIG_RPXCLASSIC) || defined(CONFIG_M5272) +#if defined(CONFIG_M5272) #define HAVE_mii_link_interrupt #endif From 9b0e74102494971ca37a425c63031fea68bb5b79 Mon Sep 17 00:00:00 2001 From: Greg Ungerer Date: Fri, 11 Jul 2008 15:29:36 +1000 Subject: [PATCH 063/857] m68knommu: put ColdFire head code into .text.head section Switch the ColdFire head start up code to be in the .text.head segment. And make sure that segment is at the start of the final linked text segment. Fixes the linker warnings about section use mis-matches: WARNING: vmlinux.o(.text+0xa8): Section mismatch in reference from the variable _clear_bss to the function .init.text:start_kernel() The function _clear_bss() references the function __init start_kernel(). This is often because _clear_bss lacks a __init annotation or the annotation of start_kernel is wrong. Signed-off-by: Greg Ungerer --- arch/m68knommu/kernel/vmlinux.lds.S | 1 + arch/m68knommu/platform/coldfire/head.S | 3 ++- 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/m68knommu/kernel/vmlinux.lds.S b/arch/m68knommu/kernel/vmlinux.lds.S index 93e69236ed6f..69ba9b10767a 100644 --- a/arch/m68knommu/kernel/vmlinux.lds.S +++ b/arch/m68knommu/kernel/vmlinux.lds.S @@ -62,6 +62,7 @@ SECTIONS { .text : { _text = .; _stext = . ; + HEAD_TEXT TEXT_TEXT SCHED_TEXT LOCK_TEXT diff --git a/arch/m68knommu/platform/coldfire/head.S b/arch/m68knommu/platform/coldfire/head.S index b9aa0ca29bfb..2b0d73c0cc32 100644 --- a/arch/m68knommu/platform/coldfire/head.S +++ b/arch/m68knommu/platform/coldfire/head.S @@ -10,6 +10,7 @@ #include #include +#include #include #include #include @@ -126,7 +127,7 @@ _ramend: /*****************************************************************************/ -.text +__HEAD /* * This is the codes first entry point. This is where it all From b7c2a75725dee9b5643a0aae3a4cb47f52e00a49 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 22 Jul 2008 22:34:29 -0700 Subject: [PATCH 064/857] sparc64: Fix lockdep issues in LDC protocol layer. We're calling request_irq() with a IRQs disabled. No straightforward fix exists because we want to enable these IRQs and setup state atomically before getting into the IRQ handler the first time. What happens now is that we mark the VIRQ to not be automatically enabled by request_irq(). Then we make explicit enable_irq() calls when we grab the LDC channel. This way we don't need to call request_irq() illegally under the LDC channel lock any more. Bump LDC version and release date. Signed-off-by: David S. Miller --- arch/sparc64/kernel/irq.c | 10 +++++++++- arch/sparc64/kernel/ldc.c | 38 +++++++++++++++++++------------------- 2 files changed, 28 insertions(+), 20 deletions(-) diff --git a/arch/sparc64/kernel/irq.c b/arch/sparc64/kernel/irq.c index b441a26b73b0..c481673d249c 100644 --- a/arch/sparc64/kernel/irq.c +++ b/arch/sparc64/kernel/irq.c @@ -621,8 +621,9 @@ unsigned int sun4v_build_irq(u32 devhandle, unsigned int devino) unsigned int sun4v_build_virq(u32 devhandle, unsigned int devino) { struct irq_handler_data *data; - struct ino_bucket *bucket; unsigned long hv_err, cookie; + struct ino_bucket *bucket; + struct irq_desc *desc; unsigned int virt_irq; bucket = kzalloc(sizeof(struct ino_bucket), GFP_ATOMIC); @@ -643,6 +644,13 @@ unsigned int sun4v_build_virq(u32 devhandle, unsigned int devino) if (unlikely(!data)) return 0; + /* In order to make the LDC channel startup sequence easier, + * especially wrt. locking, we do not let request_irq() enable + * the interrupt. + */ + desc = irq_desc + virt_irq; + desc->status |= IRQ_NOAUTOEN; + set_irq_chip_data(virt_irq, data); /* Catch accidental accesses to these things. IMAP/ICLR handling diff --git a/arch/sparc64/kernel/ldc.c b/arch/sparc64/kernel/ldc.c index 63969f610284..d68982330f66 100644 --- a/arch/sparc64/kernel/ldc.c +++ b/arch/sparc64/kernel/ldc.c @@ -1,6 +1,6 @@ /* ldc.c: Logical Domain Channel link-layer protocol driver. * - * Copyright (C) 2007 David S. Miller + * Copyright (C) 2007, 2008 David S. Miller */ #include @@ -23,8 +23,8 @@ #define DRV_MODULE_NAME "ldc" #define PFX DRV_MODULE_NAME ": " -#define DRV_MODULE_VERSION "1.0" -#define DRV_MODULE_RELDATE "June 25, 2007" +#define DRV_MODULE_VERSION "1.1" +#define DRV_MODULE_RELDATE "July 22, 2008" static char version[] __devinitdata = DRV_MODULE_NAME ".c:v" DRV_MODULE_VERSION " (" DRV_MODULE_RELDATE ")\n"; @@ -1235,13 +1235,9 @@ int ldc_bind(struct ldc_channel *lp, const char *name) unsigned long hv_err, flags; int err = -EINVAL; - spin_lock_irqsave(&lp->lock, flags); - - if (!name) - goto out_err; - - if (lp->state != LDC_STATE_INIT) - goto out_err; + if (!name || + (lp->state != LDC_STATE_INIT)) + return -EINVAL; snprintf(lp->rx_irq_name, LDC_IRQ_NAME_MAX, "%s RX", name); snprintf(lp->tx_irq_name, LDC_IRQ_NAME_MAX, "%s TX", name); @@ -1250,25 +1246,32 @@ int ldc_bind(struct ldc_channel *lp, const char *name) IRQF_SAMPLE_RANDOM | IRQF_SHARED, lp->rx_irq_name, lp); if (err) - goto out_err; + return err; err = request_irq(lp->cfg.tx_irq, ldc_tx, IRQF_SAMPLE_RANDOM | IRQF_SHARED, lp->tx_irq_name, lp); - if (err) - goto out_free_rx_irq; + if (err) { + free_irq(lp->cfg.rx_irq, lp); + return err; + } + spin_lock_irqsave(&lp->lock, flags); + + enable_irq(lp->cfg.rx_irq); + enable_irq(lp->cfg.tx_irq); + lp->flags |= LDC_FLAG_REGISTERED_IRQS; err = -ENODEV; hv_err = sun4v_ldc_tx_qconf(lp->id, 0, 0); if (hv_err) - goto out_free_tx_irq; + goto out_free_irqs; hv_err = sun4v_ldc_tx_qconf(lp->id, lp->tx_ra, lp->tx_num_entries); if (hv_err) - goto out_free_tx_irq; + goto out_free_irqs; hv_err = sun4v_ldc_rx_qconf(lp->id, 0, 0); if (hv_err) @@ -1304,14 +1307,11 @@ int ldc_bind(struct ldc_channel *lp, const char *name) out_unmap_tx: sun4v_ldc_tx_qconf(lp->id, 0, 0); -out_free_tx_irq: +out_free_irqs: lp->flags &= ~LDC_FLAG_REGISTERED_IRQS; free_irq(lp->cfg.tx_irq, lp); - -out_free_rx_irq: free_irq(lp->cfg.rx_irq, lp); -out_err: spin_unlock_irqrestore(&lp->lock, flags); return err; From e14fa82439d33cef67eaafc1a48960bbfa610c8e Mon Sep 17 00:00:00 2001 From: Riku Voipio Date: Sat, 31 May 2008 14:43:41 +0100 Subject: [PATCH 065/857] leds: Add pca9532 led driver NXP pca9532 is a LED dimmer/controller attached to i2c bus. It allows attaching upto 16 leds which can either be on, off or dimmed and/or blinked with the two PWM modulators available. This driver is a "new-style" i2c driver that adheres to the driver model and implements the led framework api. Since the leds connected to the driver are platform specific, it is only useful when platform data is passed to the driver to define what leds are connected to which pins. Signed-off-by: Riku Voipio Signed-off-by: Andrew Morton Signed-off-by: Richard Purdie --- drivers/leds/Kconfig | 8 + drivers/leds/Makefile | 1 + drivers/leds/leds-pca9532.c | 337 +++++++++++++++++++++++++++++++++++ include/linux/leds-pca9532.h | 45 +++++ 4 files changed, 391 insertions(+) create mode 100644 drivers/leds/leds-pca9532.c create mode 100644 include/linux/leds-pca9532.h diff --git a/drivers/leds/Kconfig b/drivers/leds/Kconfig index 86a369bc57d6..1c35dfaef721 100644 --- a/drivers/leds/Kconfig +++ b/drivers/leds/Kconfig @@ -103,6 +103,14 @@ config LEDS_HP6XX This option enables led support for the handheld HP Jornada 620/660/680/690. +config LEDS_PCA9532 + tristate "LED driver for PCA9532 dimmer" + depends on LEDS_CLASS && I2C && INPUT && EXPERIMENTAL + help + This option enables support for NXP pca9532 + led controller. It is generally only usefull + as a platform driver + config LEDS_GPIO tristate "LED Support for GPIO connected LEDs" depends on LEDS_CLASS && GENERIC_GPIO diff --git a/drivers/leds/Makefile b/drivers/leds/Makefile index 973d626f5f4a..7156f9970fa9 100644 --- a/drivers/leds/Makefile +++ b/drivers/leds/Makefile @@ -16,6 +16,7 @@ obj-$(CONFIG_LEDS_WRAP) += leds-wrap.o obj-$(CONFIG_LEDS_H1940) += leds-h1940.o obj-$(CONFIG_LEDS_COBALT_QUBE) += leds-cobalt-qube.o obj-$(CONFIG_LEDS_COBALT_RAQ) += leds-cobalt-raq.o +obj-$(CONFIG_LEDS_PCA9532) += leds-pca9532.o obj-$(CONFIG_LEDS_GPIO) += leds-gpio.o obj-$(CONFIG_LEDS_CM_X270) += leds-cm-x270.o obj-$(CONFIG_LEDS_CLEVO_MAIL) += leds-clevo-mail.o diff --git a/drivers/leds/leds-pca9532.c b/drivers/leds/leds-pca9532.c new file mode 100644 index 000000000000..4064d4f6b33b --- /dev/null +++ b/drivers/leds/leds-pca9532.c @@ -0,0 +1,337 @@ +/* + * pca9532.c - 16-bit Led dimmer + * + * Copyright (C) 2008 Riku Voipio + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * Datasheet: http://www.nxp.com/acrobat/datasheets/PCA9532_3.pdf + * + */ + +#include +#include +#include +#include +#include +#include + +static const unsigned short normal_i2c[] = { /*0x60,*/ I2C_CLIENT_END}; +I2C_CLIENT_INSMOD_1(pca9532); + +#define PCA9532_REG_PSC(i) (0x2+(i)*2) +#define PCA9532_REG_PWM(i) (0x3+(i)*2) +#define PCA9532_REG_LS0 0x6 +#define LED_REG(led) ((led>>2)+PCA9532_REG_LS0) +#define LED_NUM(led) (led & 0x3) + +#define ldev_to_led(c) container_of(c, struct pca9532_led, ldev) + +struct pca9532_data { + struct i2c_client *client; + struct pca9532_led leds[16]; + struct mutex update_lock; + struct input_dev *idev; + u8 pwm[2]; + u8 psc[2]; +}; + +static int pca9532_probe(struct i2c_client *client, + const struct i2c_device_id *id); +static int pca9532_remove(struct i2c_client *client); + +static const struct i2c_device_id pca9532_id[] = { + { "pca9532", 0 }, + { } +}; + +MODULE_DEVICE_TABLE(i2c, pca9532_id); + +static struct i2c_driver pca9532_driver = { + .driver = { + .name = "pca9532", + }, + .probe = pca9532_probe, + .remove = pca9532_remove, + .id_table = pca9532_id, +}; + +/* We have two pwm/blinkers, but 16 possible leds to drive. Additionaly, + * the clever Thecus people are using one pwm to drive the beeper. So, + * as a compromise we average one pwm to the values requested by all + * leds that are not ON/OFF. + * */ +static int pca9532_setpwm(struct i2c_client *client, int pwm, int blink, + enum led_brightness value) +{ + int a = 0, b = 0, i = 0; + struct pca9532_data *data = i2c_get_clientdata(client); + for (i = 0; i < 16; i++) { + if (data->leds[i].type == PCA9532_TYPE_LED && + data->leds[i].state == PCA9532_PWM0+pwm) { + a++; + b += data->leds[i].ldev.brightness; + } + } + if (a == 0) { + dev_err(&client->dev, + "fear of division by zero %d/%d, wanted %d\n", + b, a, value); + return -EINVAL; + } + b = b/a; + if (b > 0xFF) + return -EINVAL; + mutex_lock(&data->update_lock); + data->pwm[pwm] = b; + i2c_smbus_write_byte_data(client, PCA9532_REG_PWM(pwm), + data->pwm[pwm]); + data->psc[pwm] = blink; + i2c_smbus_write_byte_data(client, PCA9532_REG_PSC(pwm), + data->psc[pwm]); + mutex_unlock(&data->update_lock); + return 0; +} + +/* Set LED routing */ +static void pca9532_setled(struct pca9532_led *led) +{ + struct i2c_client *client = led->client; + struct pca9532_data *data = i2c_get_clientdata(client); + char reg; + + mutex_lock(&data->update_lock); + reg = i2c_smbus_read_byte_data(client, LED_REG(led->id)); + /* zero led bits */ + reg = reg & ~(0x3<id)*2); + /* set the new value */ + reg = reg | (led->state << LED_NUM(led->id)*2); + i2c_smbus_write_byte_data(client, LED_REG(led->id), reg); + mutex_unlock(&data->update_lock); +} + +static void pca9532_set_brightness(struct led_classdev *led_cdev, + enum led_brightness value) +{ + int err = 0; + struct pca9532_led *led = ldev_to_led(led_cdev); + + if (value == LED_OFF) + led->state = PCA9532_OFF; + else if (value == LED_FULL) + led->state = PCA9532_ON; + else { + led->state = PCA9532_PWM0; /* Thecus: hardcode one pwm */ + err = pca9532_setpwm(led->client, 0, 0, value); + if (err) + return; /* XXX: led api doesn't allow error code? */ + } + pca9532_setled(led); +} + +static int pca9532_set_blink(struct led_classdev *led_cdev, + unsigned long *delay_on, unsigned long *delay_off) +{ + struct pca9532_led *led = ldev_to_led(led_cdev); + struct i2c_client *client = led->client; + int psc; + + if (*delay_on == 0 && *delay_off == 0) { + /* led subsystem ask us for a blink rate */ + *delay_on = 1000; + *delay_off = 1000; + } + if (*delay_on != *delay_off || *delay_on > 1690 || *delay_on < 6) + return -EINVAL; + + /* Thecus specific: only use PSC/PWM 0 */ + psc = (*delay_on * 152-1)/1000; + return pca9532_setpwm(client, 0, psc, led_cdev->brightness); +} + +int pca9532_event(struct input_dev *dev, unsigned int type, unsigned int code, + int value) +{ + struct pca9532_data *data = input_get_drvdata(dev); + + if (type != EV_SND && (code != SND_BELL || code != SND_TONE)) + return -1; + + /* XXX: allow different kind of beeps with psc/pwm modifications */ + if (value > 1 && value < 32767) + data->pwm[1] = 127; + else + data->pwm[1] = 0; + + dev_info(&dev->dev, "setting beep to %d \n", data->pwm[1]); + mutex_lock(&data->update_lock); + i2c_smbus_write_byte_data(data->client, PCA9532_REG_PWM(1), + data->pwm[1]); + mutex_unlock(&data->update_lock); + + return 0; +} + +static int pca9532_configure(struct i2c_client *client, + struct pca9532_data *data, struct pca9532_platform_data *pdata) +{ + int i, err = 0; + + for (i = 0; i < 2; i++) { + data->pwm[i] = pdata->pwm[i]; + data->psc[i] = pdata->psc[i]; + i2c_smbus_write_byte_data(client, PCA9532_REG_PWM(i), + data->pwm[i]); + i2c_smbus_write_byte_data(client, PCA9532_REG_PSC(i), + data->psc[i]); + } + + for (i = 0; i < 16; i++) { + struct pca9532_led *led = &data->leds[i]; + struct pca9532_led *pled = &pdata->leds[i]; + led->client = client; + led->id = i; + led->type = pled->type; + switch (led->type) { + case PCA9532_TYPE_NONE: + break; + case PCA9532_TYPE_LED: + led->state = pled->state; + led->name = pled->name; + led->ldev.name = led->name; + led->ldev.brightness = LED_OFF; + led->ldev.brightness_set = pca9532_set_brightness; + led->ldev.blink_set = pca9532_set_blink; + if (led_classdev_register(&client->dev, + &led->ldev) < 0) { + dev_err(&client->dev, + "couldn't register LED %s\n", + led->name); + goto exit; + } + pca9532_setled(led); + break; + case PCA9532_TYPE_N2100_BEEP: + BUG_ON(data->idev); + led->state = PCA9532_PWM1; + pca9532_setled(led); + data->idev = input_allocate_device(); + if (data->idev == NULL) { + err = -ENOMEM; + goto exit; + } + data->idev->name = pled->name; + data->idev->phys = "i2c/pca9532"; + data->idev->id.bustype = BUS_HOST; + data->idev->id.vendor = 0x001f; + data->idev->id.product = 0x0001; + data->idev->id.version = 0x0100; + data->idev->evbit[0] = BIT_MASK(EV_SND); + data->idev->sndbit[0] = BIT_MASK(SND_BELL) | + BIT_MASK(SND_TONE); + data->idev->event = pca9532_event; + input_set_drvdata(data->idev, data); + err = input_register_device(data->idev); + if (err) { + input_free_device(data->idev); + data->idev = NULL; + goto exit; + } + break; + } + } + return 0; + +exit: + if (i > 0) + for (i = i - 1; i >= 0; i--) + switch (data->leds[i].type) { + case PCA9532_TYPE_NONE: + break; + case PCA9532_TYPE_LED: + led_classdev_unregister(&data->leds[i].ldev); + break; + case PCA9532_TYPE_N2100_BEEP: + if (data->idev != NULL) { + input_unregister_device(data->idev); + input_free_device(data->idev); + data->idev = NULL; + } + break; + } + + return err; + +} + +static int pca9532_probe(struct i2c_client *client, + const struct i2c_device_id *id) +{ + struct pca9532_data *data = i2c_get_clientdata(client); + struct pca9532_platform_data *pca9532_pdata = client->dev.platform_data; + + if (!i2c_check_functionality(client->adapter, + I2C_FUNC_SMBUS_BYTE_DATA)) + return -EIO; + + data = kzalloc(sizeof(struct pca9532_data), GFP_KERNEL); + if (!data) + return -ENOMEM; + + dev_info(&client->dev, "setting platform data\n"); + i2c_set_clientdata(client, data); + data->client = client; + mutex_init(&data->update_lock); + + if (pca9532_pdata == NULL) + return -EIO; + + pca9532_configure(client, data, pca9532_pdata); + return 0; + +} + +static int pca9532_remove(struct i2c_client *client) +{ + struct pca9532_data *data = i2c_get_clientdata(client); + int i; + for (i = 0; i < 16; i++) + switch (data->leds[i].type) { + case PCA9532_TYPE_NONE: + break; + case PCA9532_TYPE_LED: + led_classdev_unregister(&data->leds[i].ldev); + break; + case PCA9532_TYPE_N2100_BEEP: + if (data->idev != NULL) { + input_unregister_device(data->idev); + input_free_device(data->idev); + data->idev = NULL; + } + break; + } + + kfree(data); + i2c_set_clientdata(client, NULL); + return 0; +} + +static int __init pca9532_init(void) +{ + return i2c_add_driver(&pca9532_driver); +} + +static void __exit pca9532_exit(void) +{ + i2c_del_driver(&pca9532_driver); +} + +MODULE_AUTHOR("Riku Voipio "); +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("PCA 9532 LED dimmer"); + +module_init(pca9532_init); +module_exit(pca9532_exit); + diff --git a/include/linux/leds-pca9532.h b/include/linux/leds-pca9532.h new file mode 100644 index 000000000000..81b4207deb95 --- /dev/null +++ b/include/linux/leds-pca9532.h @@ -0,0 +1,45 @@ +/* + * pca9532.h - platform data structure for pca9532 led controller + * + * Copyright (C) 2008 Riku Voipio + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; version 2 of the License. + * + * Datasheet: http://www.nxp.com/acrobat/datasheets/PCA9532_3.pdf + * + */ + +#ifndef __LINUX_PCA9532_H +#define __LINUX_PCA9532_H + +#include + +enum pca9532_state { + PCA9532_OFF = 0x0, + PCA9532_ON = 0x1, + PCA9532_PWM0 = 0x2, + PCA9532_PWM1 = 0x3 +}; + +enum pca9532_type { PCA9532_TYPE_NONE, PCA9532_TYPE_LED, + PCA9532_TYPE_N2100_BEEP }; + +struct pca9532_led { + u8 id; + struct i2c_client *client; + char *name; + struct led_classdev ldev; + enum pca9532_type type; + enum pca9532_state state; +}; + +struct pca9532_platform_data { + struct pca9532_led leds[16]; + u8 pwm[2]; + u8 psc[2]; +}; + +#endif /* __LINUX_PCA9532_H */ + From 30be0486791fb637e758c771956c8f73bef3467c Mon Sep 17 00:00:00 2001 From: Riku Voipio Date: Sat, 31 May 2008 14:45:16 +0100 Subject: [PATCH 066/857] leds: Add pca9532 platform data for Thecus N2100 Thecus N2100 has leds and a buzzer attached to a pca9532 controller. Attach the driver to the i2c bus and define the pca9532 pin coniguration for this platform in n2100_leds. With this patch, support for N2100 should be complete in mainline Linux. Signed-off-by: Riku Voipio Acked-by: Lennert Buytenhek Signed-off-by: Andrew Morton Signed-off-by: Richard Purdie --- arch/arm/mach-iop32x/n2100.c | 52 ++++++++++++++++++++++++++++++++++++ 1 file changed, 52 insertions(+) diff --git a/arch/arm/mach-iop32x/n2100.c b/arch/arm/mach-iop32x/n2100.c index 2741063bf361..28f164ea4726 100644 --- a/arch/arm/mach-iop32x/n2100.c +++ b/arch/arm/mach-iop32x/n2100.c @@ -17,6 +17,7 @@ #include #include #include +#include #include #include #include @@ -206,6 +207,53 @@ static struct f75375s_platform_data n2100_f75375s = { .pwm_enable = { 0, 0 }, }; +static struct pca9532_platform_data n2100_leds = { + .leds = { + { .name = "n2100:red:satafail0", + .state = PCA9532_OFF, + .type = PCA9532_TYPE_LED, + }, + { .name = "n2100:red:satafail1", + .state = PCA9532_OFF, + .type = PCA9532_TYPE_LED, + }, + { .name = "n2100:blue:usb", + .state = PCA9532_OFF, + .type = PCA9532_TYPE_LED, + }, + { .type = PCA9532_TYPE_NONE }, + + { .type = PCA9532_TYPE_NONE }, + { .type = PCA9532_TYPE_NONE }, + { .type = PCA9532_TYPE_NONE }, + { .name = "n2100:red:usb", + .state = PCA9532_OFF, + .type = PCA9532_TYPE_LED, + }, + + { .type = PCA9532_TYPE_NONE }, /* power OFF gpio */ + { .type = PCA9532_TYPE_NONE }, /* reset gpio */ + { .type = PCA9532_TYPE_NONE }, + { .type = PCA9532_TYPE_NONE }, + + { .type = PCA9532_TYPE_NONE }, + { .name = "n2100:orange:system", + .state = PCA9532_OFF, + .type = PCA9532_TYPE_LED, + }, + { .name = "n2100:red:system", + .state = PCA9532_OFF, + .type = PCA9532_TYPE_LED, + }, + { .name = "N2100 beeper" , + .state = PCA9532_OFF, + .type = PCA9532_TYPE_N2100_BEEP, + }, + }, + .psc = { 0, 0 }, + .pwm = { 0, 0 }, +}; + static struct i2c_board_info __initdata n2100_i2c_devices[] = { { I2C_BOARD_INFO("rs5c372b", 0x32), @@ -214,6 +262,10 @@ static struct i2c_board_info __initdata n2100_i2c_devices[] = { I2C_BOARD_INFO("f75375", 0x2e), .platform_data = &n2100_f75375s, }, + { + I2C_BOARD_INFO("pca9532", 0x60), + .platform_data = &n2100_leds, + }, }; /* From e49575f46cdb40014e14789a18e637f8fb917317 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Sat, 31 May 2008 15:18:55 +0100 Subject: [PATCH 067/857] leds: fix unsigned value overflow in atmel pwm driver Fix an unsigned value overflow in the error handling code in the Atmel PWM driver. Signed-off-by: Li Zefan Signed-off-by: Richard Purdie --- drivers/leds/leds-atmel-pwm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/leds/leds-atmel-pwm.c b/drivers/leds/leds-atmel-pwm.c index 28db6c1444ed..52297c3ab246 100644 --- a/drivers/leds/leds-atmel-pwm.c +++ b/drivers/leds/leds-atmel-pwm.c @@ -37,7 +37,7 @@ static int __init pwmled_probe(struct platform_device *pdev) { const struct gpio_led_platform_data *pdata; struct pwmled *leds; - unsigned i; + int i; int status; pdata = pdev->dev.platform_data; From 781a54e7664cc0089287a90d27086e9656ac68a1 Mon Sep 17 00:00:00 2001 From: Anton Vorontsov Date: Sat, 31 May 2008 15:23:19 +0100 Subject: [PATCH 068/857] leds: mark led_classdev.default_trigger as const LED classdev core doesn't modify memory pointed by the default_trigger, so mark it as const and we'll able to pass const char *s without getting compiler warnings. Signed-off-by: Anton Vorontsov Signed-off-by: Richard Purdie --- include/linux/leds.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/leds.h b/include/linux/leds.h index 519df72e939d..e7a5e89932fe 100644 --- a/include/linux/leds.h +++ b/include/linux/leds.h @@ -48,7 +48,7 @@ struct led_classdev { struct device *dev; struct list_head node; /* LED Device list */ - char *default_trigger; /* Trigger to use */ + const char *default_trigger; /* Trigger to use */ #ifdef CONFIG_LEDS_TRIGGERS /* Protects the trigger data below */ From dd1160dc1842ae172495a6da274a77e35c593ed8 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Mon, 9 Jun 2008 22:00:49 +0100 Subject: [PATCH 069/857] leds: Fix sparse warnings in leds-h1940 driver Fixes the following sparse errors: drivers/leds/leds-h1940.c:26:6: warning: symbol 'h1940_greenled_set' was not declared. Should it be static? drivers/leds/leds-h1940.c:55:6: warning: symbol 'h1940_redled_set' was not declared. Should it be static? drivers/leds/leds-h1940.c:85:6: warning: symbol 'h1940_blueled_set' was not declared. Should it be static? Signed-off-by: Ben Dooks Signed-off-by: Richard Purdie --- drivers/leds/leds-h1940.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/drivers/leds/leds-h1940.c b/drivers/leds/leds-h1940.c index bcec42230389..73c705021686 100644 --- a/drivers/leds/leds-h1940.c +++ b/drivers/leds/leds-h1940.c @@ -23,7 +23,8 @@ /* * Green led. */ -void h1940_greenled_set(struct led_classdev *led_dev, enum led_brightness value) +static void h1940_greenled_set(struct led_classdev *led_dev, + enum led_brightness value) { switch (value) { case LED_HALF: @@ -52,7 +53,8 @@ static struct led_classdev h1940_greenled = { /* * Red led. */ -void h1940_redled_set(struct led_classdev *led_dev, enum led_brightness value) +static void h1940_redled_set(struct led_classdev *led_dev, + enum led_brightness value) { switch (value) { case LED_HALF: @@ -82,7 +84,8 @@ static struct led_classdev h1940_redled = { * Blue led. * (it can only be blue flashing led) */ -void h1940_blueled_set(struct led_classdev *led_dev, enum led_brightness value) +static void h1940_blueled_set(struct led_classdev *led_dev, + enum led_brightness value) { if (value) { /* flashing Blue */ From f46e9203d9a100bae216cc06e17f2e77351aa8d8 Mon Sep 17 00:00:00 2001 From: Nate Case Date: Wed, 16 Jul 2008 22:49:55 +0100 Subject: [PATCH 070/857] leds: Add support for Philips PCA955x I2C LED drivers This driver supports the PCA9550, PCA9551, PCA9552, and PCA9553 LED driver chips. Signed-off-by: Nate Case Signed-off-by: Andrew Morton Signed-off-by: Richard Purdie --- drivers/leds/Kconfig | 8 + drivers/leds/Makefile | 1 + drivers/leds/leds-pca955x.c | 384 ++++++++++++++++++++++++++++++++++++ include/linux/leds.h | 14 ++ 4 files changed, 407 insertions(+) create mode 100644 drivers/leds/leds-pca955x.c diff --git a/drivers/leds/Kconfig b/drivers/leds/Kconfig index 1c35dfaef721..9556262dda5a 100644 --- a/drivers/leds/Kconfig +++ b/drivers/leds/Kconfig @@ -155,6 +155,14 @@ config LEDS_CLEVO_MAIL To compile this driver as a module, choose M here: the module will be called leds-clevo-mail. +config LEDS_PCA955X + tristate "LED Support for PCA955x I2C chips" + depends on LEDS_CLASS && I2C + help + This option enables support for LEDs connected to PCA955x + LED driver chips accessed via the I2C bus. Supported + devices include PCA9550, PCA9551, PCA9552, and PCA9553. + comment "LED Triggers" config LEDS_TRIGGERS diff --git a/drivers/leds/Makefile b/drivers/leds/Makefile index 7156f9970fa9..ff7982b44565 100644 --- a/drivers/leds/Makefile +++ b/drivers/leds/Makefile @@ -22,6 +22,7 @@ obj-$(CONFIG_LEDS_CM_X270) += leds-cm-x270.o obj-$(CONFIG_LEDS_CLEVO_MAIL) += leds-clevo-mail.o obj-$(CONFIG_LEDS_HP6XX) += leds-hp6xx.o obj-$(CONFIG_LEDS_FSG) += leds-fsg.o +obj-$(CONFIG_LEDS_PCA955X) += leds-pca955x.o # LED Triggers obj-$(CONFIG_LEDS_TRIGGER_TIMER) += ledtrig-timer.o diff --git a/drivers/leds/leds-pca955x.c b/drivers/leds/leds-pca955x.c new file mode 100644 index 000000000000..146c06972863 --- /dev/null +++ b/drivers/leds/leds-pca955x.c @@ -0,0 +1,384 @@ +/* + * Copyright 2007-2008 Extreme Engineering Solutions, Inc. + * + * Author: Nate Case + * + * This file is subject to the terms and conditions of version 2 of + * the GNU General Public License. See the file COPYING in the main + * directory of this archive for more details. + * + * LED driver for various PCA955x I2C LED drivers + * + * Supported devices: + * + * Device Description 7-bit slave address + * ------ ----------- ------------------- + * PCA9550 2-bit driver 0x60 .. 0x61 + * PCA9551 8-bit driver 0x60 .. 0x67 + * PCA9552 16-bit driver 0x60 .. 0x67 + * PCA9553/01 4-bit driver 0x62 + * PCA9553/02 4-bit driver 0x63 + * + * Philips PCA955x LED driver chips follow a register map as shown below: + * + * Control Register Description + * ---------------- ----------- + * 0x0 Input register 0 + * .. + * NUM_INPUT_REGS - 1 Last Input register X + * + * NUM_INPUT_REGS Frequency prescaler 0 + * NUM_INPUT_REGS + 1 PWM register 0 + * NUM_INPUT_REGS + 2 Frequency prescaler 1 + * NUM_INPUT_REGS + 3 PWM register 1 + * + * NUM_INPUT_REGS + 4 LED selector 0 + * NUM_INPUT_REGS + 4 + * + NUM_LED_REGS - 1 Last LED selector + * + * where NUM_INPUT_REGS and NUM_LED_REGS vary depending on how many + * bits the chip supports. + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +/* LED select registers determine the source that drives LED outputs */ +#define PCA955X_LS_LED_ON 0x0 /* Output LOW */ +#define PCA955X_LS_LED_OFF 0x1 /* Output HI-Z */ +#define PCA955X_LS_BLINK0 0x2 /* Blink at PWM0 rate */ +#define PCA955X_LS_BLINK1 0x3 /* Blink at PWM1 rate */ + +enum pca955x_type { + pca9550, + pca9551, + pca9552, + pca9553, +}; + +struct pca955x_chipdef { + int bits; + u8 slv_addr; /* 7-bit slave address mask */ + int slv_addr_shift; /* Number of bits to ignore */ +}; + +static struct pca955x_chipdef pca955x_chipdefs[] = { + [pca9550] = { + .bits = 2, + .slv_addr = /* 110000x */ 0x60, + .slv_addr_shift = 1, + }, + [pca9551] = { + .bits = 8, + .slv_addr = /* 1100xxx */ 0x60, + .slv_addr_shift = 3, + }, + [pca9552] = { + .bits = 16, + .slv_addr = /* 1100xxx */ 0x60, + .slv_addr_shift = 3, + }, + [pca9553] = { + .bits = 4, + .slv_addr = /* 110001x */ 0x62, + .slv_addr_shift = 1, + }, +}; + +static const struct i2c_device_id pca955x_id[] = { + { "pca9550", pca9550 }, + { "pca9551", pca9551 }, + { "pca9552", pca9552 }, + { "pca9553", pca9553 }, + { } +}; +MODULE_DEVICE_TABLE(i2c, pca955x_id); + +struct pca955x_led { + struct pca955x_chipdef *chipdef; + struct i2c_client *client; + struct work_struct work; + spinlock_t lock; + enum led_brightness brightness; + struct led_classdev led_cdev; + int led_num; /* 0 .. 15 potentially */ + char name[32]; +}; + +/* 8 bits per input register */ +static inline int pca95xx_num_input_regs(int bits) +{ + return (bits + 7) / 8; +} + +/* 4 bits per LED selector register */ +static inline int pca95xx_num_led_regs(int bits) +{ + return (bits + 3) / 4; +} + +/* + * Return an LED selector register value based on an existing one, with + * the appropriate 2-bit state value set for the given LED number (0-3). + */ +static inline u8 pca955x_ledsel(u8 oldval, int led_num, int state) +{ + return (oldval & (~(0x3 << (led_num << 1)))) | + ((state & 0x3) << (led_num << 1)); +} + +/* + * Write to frequency prescaler register, used to program the + * period of the PWM output. period = (PSCx + 1) / 38 + */ +static void pca955x_write_psc(struct i2c_client *client, int n, u8 val) +{ + struct pca955x_led *pca955x = i2c_get_clientdata(client); + + i2c_smbus_write_byte_data(client, + pca95xx_num_input_regs(pca955x->chipdef->bits) + 2*n, + val); +} + +/* + * Write to PWM register, which determines the duty cycle of the + * output. LED is OFF when the count is less than the value of this + * register, and ON when it is greater. If PWMx == 0, LED is always OFF. + * + * Duty cycle is (256 - PWMx) / 256 + */ +static void pca955x_write_pwm(struct i2c_client *client, int n, u8 val) +{ + struct pca955x_led *pca955x = i2c_get_clientdata(client); + + i2c_smbus_write_byte_data(client, + pca95xx_num_input_regs(pca955x->chipdef->bits) + 1 + 2*n, + val); +} + +/* + * Write to LED selector register, which determines the source that + * drives the LED output. + */ +static void pca955x_write_ls(struct i2c_client *client, int n, u8 val) +{ + struct pca955x_led *pca955x = i2c_get_clientdata(client); + + i2c_smbus_write_byte_data(client, + pca95xx_num_input_regs(pca955x->chipdef->bits) + 4 + n, + val); +} + +/* + * Read the LED selector register, which determines the source that + * drives the LED output. + */ +static u8 pca955x_read_ls(struct i2c_client *client, int n) +{ + struct pca955x_led *pca955x = i2c_get_clientdata(client); + + return (u8) i2c_smbus_read_byte_data(client, + pca95xx_num_input_regs(pca955x->chipdef->bits) + 4 + n); +} + +static void pca955x_led_work(struct work_struct *work) +{ + struct pca955x_led *pca955x; + u8 ls; + int chip_ls; /* which LSx to use (0-3 potentially) */ + int ls_led; /* which set of bits within LSx to use (0-3) */ + + pca955x = container_of(work, struct pca955x_led, work); + chip_ls = pca955x->led_num / 4; + ls_led = pca955x->led_num % 4; + + ls = pca955x_read_ls(pca955x->client, chip_ls); + + switch (pca955x->brightness) { + case LED_FULL: + ls = pca955x_ledsel(ls, ls_led, PCA955X_LS_LED_ON); + break; + case LED_OFF: + ls = pca955x_ledsel(ls, ls_led, PCA955X_LS_LED_OFF); + break; + case LED_HALF: + ls = pca955x_ledsel(ls, ls_led, PCA955X_LS_BLINK0); + break; + default: + /* + * Use PWM1 for all other values. This has the unwanted + * side effect of making all LEDs on the chip share the + * same brightness level if set to a value other than + * OFF, HALF, or FULL. But, this is probably better than + * just turning off for all other values. + */ + pca955x_write_pwm(pca955x->client, 1, 255-pca955x->brightness); + ls = pca955x_ledsel(ls, ls_led, PCA955X_LS_BLINK1); + break; + } + + pca955x_write_ls(pca955x->client, chip_ls, ls); +} + +void pca955x_led_set(struct led_classdev *led_cdev, enum led_brightness value) +{ + struct pca955x_led *pca955x; + + pca955x = container_of(led_cdev, struct pca955x_led, led_cdev); + + spin_lock(&pca955x->lock); + pca955x->brightness = value; + + /* + * Must use workqueue for the actual I/O since I2C operations + * can sleep. + */ + schedule_work(&pca955x->work); + + spin_unlock(&pca955x->lock); +} + +static int __devinit pca955x_probe(struct i2c_client *client, + const struct i2c_device_id *id) +{ + struct pca955x_led *pca955x; + int i; + int err = -ENODEV; + struct pca955x_chipdef *chip; + struct i2c_adapter *adapter; + struct led_platform_data *pdata; + + chip = &pca955x_chipdefs[id->driver_data]; + adapter = to_i2c_adapter(client->dev.parent); + pdata = client->dev.platform_data; + + /* Make sure the slave address / chip type combo given is possible */ + if ((client->addr & ~((1 << chip->slv_addr_shift) - 1)) != + chip->slv_addr) { + dev_err(&client->dev, "invalid slave address %02x\n", + client->addr); + return -ENODEV; + } + + printk(KERN_INFO "leds-pca955x: Using %s %d-bit LED driver at " + "slave address 0x%02x\n", + id->name, chip->bits, client->addr); + + if (!i2c_check_functionality(adapter, I2C_FUNC_I2C)) + return -EIO; + + if (pdata) { + if (pdata->num_leds != chip->bits) { + dev_err(&client->dev, "board info claims %d LEDs" + " on a %d-bit chip\n", + pdata->num_leds, chip->bits); + return -ENODEV; + } + } + + for (i = 0; i < chip->bits; i++) { + pca955x = kzalloc(sizeof(struct pca955x_led), GFP_KERNEL); + if (!pca955x) { + err = -ENOMEM; + goto exit; + } + + pca955x->chipdef = chip; + pca955x->client = client; + pca955x->led_num = i; + /* Platform data can specify LED names and default triggers */ + if (pdata) { + if (pdata->leds[i].name) + snprintf(pca955x->name, 32, "pca955x:%s", + pdata->leds[i].name); + if (pdata->leds[i].default_trigger) + pca955x->led_cdev.default_trigger = + pdata->leds[i].default_trigger; + } else { + snprintf(pca955x->name, 32, "pca955x:%d", i); + } + spin_lock_init(&pca955x->lock); + + pca955x->led_cdev.name = pca955x->name; + pca955x->led_cdev.brightness_set = + pca955x_led_set; + + /* + * Client data is a pointer to the _first_ pca955x_led + * struct + */ + if (i == 0) + i2c_set_clientdata(client, pca955x); + + INIT_WORK(&(pca955x->work), pca955x_led_work); + + led_classdev_register(&client->dev, &(pca955x->led_cdev)); + } + + /* Turn off LEDs */ + for (i = 0; i < pca95xx_num_led_regs(chip->bits); i++) + pca955x_write_ls(client, i, 0x55); + + /* PWM0 is used for half brightness or 50% duty cycle */ + pca955x_write_pwm(client, 0, 255-LED_HALF); + + /* PWM1 is used for variable brightness, default to OFF */ + pca955x_write_pwm(client, 1, 0); + + /* Set to fast frequency so we do not see flashing */ + pca955x_write_psc(client, 0, 0); + pca955x_write_psc(client, 1, 0); + + return 0; +exit: + return err; +} + +static int __devexit pca955x_remove(struct i2c_client *client) +{ + struct pca955x_led *pca955x = i2c_get_clientdata(client); + int leds = pca955x->chipdef->bits; + int i; + + for (i = 0; i < leds; i++) { + led_classdev_unregister(&(pca955x->led_cdev)); + cancel_work_sync(&(pca955x->work)); + kfree(pca955x); + pca955x = pca955x + 1; + } + + return 0; +} + +static struct i2c_driver pca955x_driver = { + .driver = { + .name = "leds-pca955x", + .owner = THIS_MODULE, + }, + .probe = pca955x_probe, + .remove = __devexit_p(pca955x_remove), + .id_table = pca955x_id, +}; + +static int __init pca955x_leds_init(void) +{ + return i2c_add_driver(&pca955x_driver); +} + +static void __exit pca955x_leds_exit(void) +{ + i2c_del_driver(&pca955x_driver); +} + +module_init(pca955x_leds_init); +module_exit(pca955x_leds_exit); + +MODULE_AUTHOR("Nate Case "); +MODULE_DESCRIPTION("PCA955x LED driver"); +MODULE_LICENSE("GPL v2"); diff --git a/include/linux/leds.h b/include/linux/leds.h index e7a5e89932fe..d41ccb56146a 100644 --- a/include/linux/leds.h +++ b/include/linux/leds.h @@ -118,6 +118,20 @@ extern void ledtrig_ide_activity(void); #define ledtrig_ide_activity() do {} while(0) #endif +/* + * Generic LED platform data for describing LED names and default triggers. + */ +struct led_info { + const char *name; + char *default_trigger; + int flags; +}; + +struct led_platform_data { + int num_leds; + struct led_info *leds; +}; + /* For the leds-gpio driver */ struct gpio_led { const char *name; From fe3025b55c8ed06929afe94e9c9095fc19d15aa0 Mon Sep 17 00:00:00 2001 From: Dmitry Baryshkov Date: Wed, 16 Jul 2008 22:51:14 +0100 Subject: [PATCH 071/857] leds: Ensure led->trigger is set earlier Make sure led->trigger is valid before calling trigger->activate Signed-off-by: Dmitry Baryshkov Signed-off-by: Andrew Morton Signed-off-by: Richard Purdie --- drivers/leds/led-triggers.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/leds/led-triggers.c b/drivers/leds/led-triggers.c index 0f242b3f09b6..f910eaffe3a6 100644 --- a/drivers/leds/led-triggers.c +++ b/drivers/leds/led-triggers.c @@ -111,16 +111,17 @@ void led_trigger_set(struct led_classdev *led_cdev, struct led_trigger *trigger) flags); if (led_cdev->trigger->deactivate) led_cdev->trigger->deactivate(led_cdev); + led_cdev->trigger = NULL; led_set_brightness(led_cdev, LED_OFF); } if (trigger) { write_lock_irqsave(&trigger->leddev_list_lock, flags); list_add_tail(&led_cdev->trig_list, &trigger->led_cdevs); write_unlock_irqrestore(&trigger->leddev_list_lock, flags); + led_cdev->trigger = trigger; if (trigger->activate) trigger->activate(led_cdev); } - led_cdev->trigger = trigger; } EXPORT_SYMBOL_GPL(led_trigger_set); From 7be35c72e6454059a33ad844153349973d22fcb7 Mon Sep 17 00:00:00 2001 From: Matthew Garrett Date: Mon, 9 Jun 2008 21:56:16 +0100 Subject: [PATCH 072/857] backlight: Add Nvidia-based Apple Macbook Pro backlight driver Nvidia-based Apple Macbook Pros don't appear to handle backlight control through the graphics card registers or ACPI, but instead trigger changes via SMI calls. This driver registers a generic backlight device that lets existing userspace deal with it. Code derived from Julien Blache's Pommed application. Signed-off-by: Julien Blache Signed-off-by: Matthew Garrett Signed-off-by: Richard Purdie --- drivers/video/backlight/Kconfig | 9 ++ drivers/video/backlight/Makefile | 2 + drivers/video/backlight/mbp_nvidia_bl.c | 116 ++++++++++++++++++++++++ 3 files changed, 127 insertions(+) create mode 100644 drivers/video/backlight/mbp_nvidia_bl.c diff --git a/drivers/video/backlight/Kconfig b/drivers/video/backlight/Kconfig index 30bf7f2f1635..62547bd2ea49 100644 --- a/drivers/video/backlight/Kconfig +++ b/drivers/video/backlight/Kconfig @@ -119,3 +119,12 @@ config BACKLIGHT_PWM help If you have a LCD backlight adjustable by PWM, say Y to enable this driver. + +config BACKLIGHT_MBP_NVIDIA + tristate "MacBook Pro Nvidia Backlight Driver" + depends on BACKLIGHT_CLASS_DEVICE && X86 + default n + help + If you have an Apple Macbook Pro with Nvidia graphics hardware say Y + to enable a driver for its backlight + diff --git a/drivers/video/backlight/Makefile b/drivers/video/backlight/Makefile index b51a7cd12500..c7c4d95fdc1c 100644 --- a/drivers/video/backlight/Makefile +++ b/drivers/video/backlight/Makefile @@ -11,3 +11,5 @@ obj-$(CONFIG_BACKLIGHT_OMAP1) += omap1_bl.o obj-$(CONFIG_BACKLIGHT_PROGEAR) += progear_bl.o obj-$(CONFIG_BACKLIGHT_CARILLO_RANCH) += cr_bllcd.o obj-$(CONFIG_BACKLIGHT_PWM) += pwm_bl.o +obj-$(CONFIG_BACKLIGHT_MBP_NVIDIA) += mbp_nvidia_bl.o + diff --git a/drivers/video/backlight/mbp_nvidia_bl.c b/drivers/video/backlight/mbp_nvidia_bl.c new file mode 100644 index 000000000000..385cba40ea87 --- /dev/null +++ b/drivers/video/backlight/mbp_nvidia_bl.c @@ -0,0 +1,116 @@ +/* + * Backlight Driver for Nvidia 8600 in Macbook Pro + * + * Copyright (c) Red Hat + * Based on code from Pommed: + * Copyright (C) 2006 Nicolas Boichat + * Copyright (C) 2006 Felipe Alfaro Solana + * Copyright (C) 2007 Julien BLACHE + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This driver triggers SMIs which cause the firmware to change the + * backlight brightness. This is icky in many ways, but it's impractical to + * get at the firmware code in order to figure out what it's actually doing. + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +static struct backlight_device *mbp_backlight_device; + +static struct dmi_system_id __initdata mbp_device_table[] = { + { + .ident = "3,1", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "MacBookPro3,1"), + }, + }, + { + .ident = "3,2", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "MacBookPro3,2"), + }, + }, + { + .ident = "4,1", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "MacBookPro4,1"), + }, + }, + { } +}; + +static int mbp_send_intensity(struct backlight_device *bd) +{ + int intensity = bd->props.brightness; + + outb(0x04 | (intensity << 4), 0xb3); + outb(0xbf, 0xb2); + + return 0; +} + +static int mbp_get_intensity(struct backlight_device *bd) +{ + outb(0x03, 0xb3); + outb(0xbf, 0xb2); + return inb(0xb3) >> 4; +} + +static struct backlight_ops mbp_ops = { + .get_brightness = mbp_get_intensity, + .update_status = mbp_send_intensity, +}; + +static int __init mbp_init(void) +{ + if (!dmi_check_system(mbp_device_table)) + return -ENODEV; + + if (!request_region(0xb2, 2, "Macbook Pro backlight")) + return -ENXIO; + + mbp_backlight_device = backlight_device_register("mbp_backlight", + NULL, NULL, + &mbp_ops); + if (IS_ERR(mbp_backlight_device)) { + release_region(0xb2, 2); + return PTR_ERR(mbp_backlight_device); + } + + mbp_backlight_device->props.max_brightness = 15; + mbp_backlight_device->props.brightness = + mbp_get_intensity(mbp_backlight_device); + backlight_update_status(mbp_backlight_device); + + return 0; +} + +static void __exit mbp_exit(void) +{ + backlight_device_unregister(mbp_backlight_device); + + release_region(0xb2, 2); +} + +module_init(mbp_init); +module_exit(mbp_exit); + +MODULE_AUTHOR("Matthew Garrett "); +MODULE_DESCRIPTION("Nvidia-based Macbook Pro Backlight Driver"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("svnAppleInc.:pnMacBookPro3,1"); +MODULE_ALIAS("svnAppleInc.:pnMacBookPro3,2"); +MODULE_ALIAS("svnAppleInc.:pnMacBookPro4,1"); From f6ec2d96796d0accda6c325890206f3629130729 Mon Sep 17 00:00:00 2001 From: Sebastian Siewior Date: Wed, 16 Jul 2008 23:05:49 +0100 Subject: [PATCH 073/857] backlight: Fix missing kernel doc entry Signed-off-by: Sebastian Siewior Signed-off-by: Richard Purdie --- drivers/video/backlight/backlight.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/video/backlight/backlight.c b/drivers/video/backlight/backlight.c index 39394757679c..fab0bc874b58 100644 --- a/drivers/video/backlight/backlight.c +++ b/drivers/video/backlight/backlight.c @@ -191,6 +191,7 @@ static struct device_attribute bl_device_attributes[] = { * backlight_device class. * @name: the name of the new object(must be the same as the name of the * respective framebuffer device). + * @parent: a pointer to the parent device * @devdata: an optional pointer to be stored for private driver use. The * methods may retrieve it by using bl_get_data(bd). * @ops: the backlight operations structure. From 422037bafde8083acc3c539ceba3dfc60a04110c Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Wed, 23 Jul 2008 11:16:38 +0200 Subject: [PATCH 074/857] sched: fix hrtick & generic-ipi dependency Andrew Morton reported this s390 allmodconfig build failure: kernel/built-in.o: In function `hrtick_start_fair': sched.c:(.text+0x69c6): undefined reference to `__smp_call_function_single' the reason is that s390 is not a generic-ipi SMP platform yet, while the hrtick code relies on it. Fix the dependency. Signed-off-by: Ingo Molnar --- kernel/Kconfig.hz | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/Kconfig.hz b/kernel/Kconfig.hz index 2a202a846757..382dd5a8b2d7 100644 --- a/kernel/Kconfig.hz +++ b/kernel/Kconfig.hz @@ -55,4 +55,4 @@ config HZ default 1000 if HZ_1000 config SCHED_HRTICK - def_bool HIGH_RES_TIMERS + def_bool HIGH_RES_TIMERS && USE_GENERIC_SMP_HELPERS From 36bd53d07243ae83c1b73bae549086cea2252854 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 23 Jul 2008 00:58:13 -0700 Subject: [PATCH 075/857] arch/mips/kernel/stacktrace.c: Heiko can't type Signed-off-by: Andrew Morton Signed-off-by: Ingo Molnar --- arch/mips/kernel/stacktrace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/mips/kernel/stacktrace.c b/arch/mips/kernel/stacktrace.c index 702e2e92a1ca..0632e2a849c0 100644 --- a/arch/mips/kernel/stacktrace.c +++ b/arch/mips/kernel/stacktrace.c @@ -7,7 +7,7 @@ */ #include #include -#include #include /* From 95d04f0735b4fc837bff9aedcc3f3efb20ddc3d1 Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Wed, 23 Jul 2008 08:12:26 -0700 Subject: [PATCH 076/857] IB/mlx4: Add support for memory management extensions and local DMA L_Key Add support for the following operations to mlx4 when device firmware supports them: - Send with invalidate and local invalidate send queue work requests; - Allocate/free fast register MRs; - Allocate/free fast register MR page lists; - Fast register MR send queue work requests; - Local DMA L_Key. Signed-off-by: Roland Dreier --- drivers/infiniband/hw/mlx4/cq.c | 12 +++++ drivers/infiniband/hw/mlx4/main.c | 11 +++++ drivers/infiniband/hw/mlx4/mlx4_ib.h | 15 ++++++ drivers/infiniband/hw/mlx4/mr.c | 70 +++++++++++++++++++++++++++ drivers/infiniband/hw/mlx4/qp.c | 72 ++++++++++++++++++++++++++-- drivers/net/mlx4/fw.c | 10 ++-- drivers/net/mlx4/fw.h | 2 +- drivers/net/mlx4/main.c | 2 + drivers/net/mlx4/mr.c | 23 +++++++-- include/linux/mlx4/device.h | 10 ++++ include/linux/mlx4/qp.h | 16 +++++-- 11 files changed, 221 insertions(+), 22 deletions(-) diff --git a/drivers/infiniband/hw/mlx4/cq.c b/drivers/infiniband/hw/mlx4/cq.c index 299f20832ab6..0b191a4842ce 100644 --- a/drivers/infiniband/hw/mlx4/cq.c +++ b/drivers/infiniband/hw/mlx4/cq.c @@ -637,6 +637,7 @@ static int mlx4_ib_poll_one(struct mlx4_ib_cq *cq, case MLX4_OPCODE_SEND_IMM: wc->wc_flags |= IB_WC_WITH_IMM; case MLX4_OPCODE_SEND: + case MLX4_OPCODE_SEND_INVAL: wc->opcode = IB_WC_SEND; break; case MLX4_OPCODE_RDMA_READ: @@ -657,6 +658,12 @@ static int mlx4_ib_poll_one(struct mlx4_ib_cq *cq, case MLX4_OPCODE_LSO: wc->opcode = IB_WC_LSO; break; + case MLX4_OPCODE_FMR: + wc->opcode = IB_WC_FAST_REG_MR; + break; + case MLX4_OPCODE_LOCAL_INVAL: + wc->opcode = IB_WC_LOCAL_INV; + break; } } else { wc->byte_len = be32_to_cpu(cqe->byte_cnt); @@ -667,6 +674,11 @@ static int mlx4_ib_poll_one(struct mlx4_ib_cq *cq, wc->wc_flags = IB_WC_WITH_IMM; wc->ex.imm_data = cqe->immed_rss_invalid; break; + case MLX4_RECV_OPCODE_SEND_INVAL: + wc->opcode = IB_WC_RECV; + wc->wc_flags = IB_WC_WITH_INVALIDATE; + wc->ex.invalidate_rkey = be32_to_cpu(cqe->immed_rss_invalid); + break; case MLX4_RECV_OPCODE_SEND: wc->opcode = IB_WC_RECV; wc->wc_flags = 0; diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c index bcf50648fa18..38d6907ab521 100644 --- a/drivers/infiniband/hw/mlx4/main.c +++ b/drivers/infiniband/hw/mlx4/main.c @@ -104,6 +104,12 @@ static int mlx4_ib_query_device(struct ib_device *ibdev, props->device_cap_flags |= IB_DEVICE_UD_IP_CSUM; if (dev->dev->caps.max_gso_sz) props->device_cap_flags |= IB_DEVICE_UD_TSO; + if (dev->dev->caps.bmme_flags & MLX4_BMME_FLAG_RESERVED_LKEY) + props->device_cap_flags |= IB_DEVICE_LOCAL_DMA_LKEY; + if ((dev->dev->caps.bmme_flags & MLX4_BMME_FLAG_LOCAL_INV) && + (dev->dev->caps.bmme_flags & MLX4_BMME_FLAG_REMOTE_INV) && + (dev->dev->caps.bmme_flags & MLX4_BMME_FLAG_FAST_REG_WR)) + props->device_cap_flags |= IB_DEVICE_MEM_MGT_EXTENSIONS; props->vendor_id = be32_to_cpup((__be32 *) (out_mad->data + 36)) & 0xffffff; @@ -127,6 +133,7 @@ static int mlx4_ib_query_device(struct ib_device *ibdev, props->max_srq = dev->dev->caps.num_srqs - dev->dev->caps.reserved_srqs; props->max_srq_wr = dev->dev->caps.max_srq_wqes - 1; props->max_srq_sge = dev->dev->caps.max_srq_sge; + props->max_fast_reg_page_list_len = PAGE_SIZE / sizeof (u64); props->local_ca_ack_delay = dev->dev->caps.local_ca_ack_delay; props->atomic_cap = dev->dev->caps.flags & MLX4_DEV_CAP_FLAG_ATOMIC ? IB_ATOMIC_HCA : IB_ATOMIC_NONE; @@ -565,6 +572,7 @@ static void *mlx4_ib_add(struct mlx4_dev *dev) strlcpy(ibdev->ib_dev.name, "mlx4_%d", IB_DEVICE_NAME_MAX); ibdev->ib_dev.owner = THIS_MODULE; ibdev->ib_dev.node_type = RDMA_NODE_IB_CA; + ibdev->ib_dev.local_dma_lkey = dev->caps.reserved_lkey; ibdev->ib_dev.phys_port_cnt = dev->caps.num_ports; ibdev->ib_dev.num_comp_vectors = 1; ibdev->ib_dev.dma_device = &dev->pdev->dev; @@ -627,6 +635,9 @@ static void *mlx4_ib_add(struct mlx4_dev *dev) ibdev->ib_dev.get_dma_mr = mlx4_ib_get_dma_mr; ibdev->ib_dev.reg_user_mr = mlx4_ib_reg_user_mr; ibdev->ib_dev.dereg_mr = mlx4_ib_dereg_mr; + ibdev->ib_dev.alloc_fast_reg_mr = mlx4_ib_alloc_fast_reg_mr; + ibdev->ib_dev.alloc_fast_reg_page_list = mlx4_ib_alloc_fast_reg_page_list; + ibdev->ib_dev.free_fast_reg_page_list = mlx4_ib_free_fast_reg_page_list; ibdev->ib_dev.attach_mcast = mlx4_ib_mcg_attach; ibdev->ib_dev.detach_mcast = mlx4_ib_mcg_detach; ibdev->ib_dev.process_mad = mlx4_ib_process_mad; diff --git a/drivers/infiniband/hw/mlx4/mlx4_ib.h b/drivers/infiniband/hw/mlx4/mlx4_ib.h index c4cf5b69eefa..d26a91317d4d 100644 --- a/drivers/infiniband/hw/mlx4/mlx4_ib.h +++ b/drivers/infiniband/hw/mlx4/mlx4_ib.h @@ -83,6 +83,11 @@ struct mlx4_ib_mr { struct ib_umem *umem; }; +struct mlx4_ib_fast_reg_page_list { + struct ib_fast_reg_page_list ibfrpl; + dma_addr_t map; +}; + struct mlx4_ib_fmr { struct ib_fmr ibfmr; struct mlx4_fmr mfmr; @@ -199,6 +204,11 @@ static inline struct mlx4_ib_mr *to_mmr(struct ib_mr *ibmr) return container_of(ibmr, struct mlx4_ib_mr, ibmr); } +static inline struct mlx4_ib_fast_reg_page_list *to_mfrpl(struct ib_fast_reg_page_list *ibfrpl) +{ + return container_of(ibfrpl, struct mlx4_ib_fast_reg_page_list, ibfrpl); +} + static inline struct mlx4_ib_fmr *to_mfmr(struct ib_fmr *ibfmr) { return container_of(ibfmr, struct mlx4_ib_fmr, ibfmr); @@ -239,6 +249,11 @@ struct ib_mr *mlx4_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length, u64 virt_addr, int access_flags, struct ib_udata *udata); int mlx4_ib_dereg_mr(struct ib_mr *mr); +struct ib_mr *mlx4_ib_alloc_fast_reg_mr(struct ib_pd *pd, + int max_page_list_len); +struct ib_fast_reg_page_list *mlx4_ib_alloc_fast_reg_page_list(struct ib_device *ibdev, + int page_list_len); +void mlx4_ib_free_fast_reg_page_list(struct ib_fast_reg_page_list *page_list); int mlx4_ib_modify_cq(struct ib_cq *cq, u16 cq_count, u16 cq_period); int mlx4_ib_resize_cq(struct ib_cq *ibcq, int entries, struct ib_udata *udata); diff --git a/drivers/infiniband/hw/mlx4/mr.c b/drivers/infiniband/hw/mlx4/mr.c index 68e92485fc76..db2086faa4ed 100644 --- a/drivers/infiniband/hw/mlx4/mr.c +++ b/drivers/infiniband/hw/mlx4/mr.c @@ -183,6 +183,76 @@ int mlx4_ib_dereg_mr(struct ib_mr *ibmr) return 0; } +struct ib_mr *mlx4_ib_alloc_fast_reg_mr(struct ib_pd *pd, + int max_page_list_len) +{ + struct mlx4_ib_dev *dev = to_mdev(pd->device); + struct mlx4_ib_mr *mr; + int err; + + mr = kmalloc(sizeof *mr, GFP_KERNEL); + if (!mr) + return ERR_PTR(-ENOMEM); + + err = mlx4_mr_alloc(dev->dev, to_mpd(pd)->pdn, 0, 0, 0, + max_page_list_len, 0, &mr->mmr); + if (err) + goto err_free; + + err = mlx4_mr_enable(dev->dev, &mr->mmr); + if (err) + goto err_mr; + + return &mr->ibmr; + +err_mr: + mlx4_mr_free(dev->dev, &mr->mmr); + +err_free: + kfree(mr); + return ERR_PTR(err); +} + +struct ib_fast_reg_page_list *mlx4_ib_alloc_fast_reg_page_list(struct ib_device *ibdev, + int page_list_len) +{ + struct mlx4_ib_dev *dev = to_mdev(ibdev); + struct mlx4_ib_fast_reg_page_list *mfrpl; + int size = page_list_len * sizeof (u64); + + if (size > PAGE_SIZE) + return ERR_PTR(-EINVAL); + + mfrpl = kmalloc(sizeof *mfrpl, GFP_KERNEL); + if (!mfrpl) + return ERR_PTR(-ENOMEM); + + mfrpl->ibfrpl.page_list = dma_alloc_coherent(&dev->dev->pdev->dev, + size, &mfrpl->map, + GFP_KERNEL); + if (!mfrpl->ibfrpl.page_list) + goto err_free; + + WARN_ON(mfrpl->map & 0x3f); + + return &mfrpl->ibfrpl; + +err_free: + kfree(mfrpl); + return ERR_PTR(-ENOMEM); +} + +void mlx4_ib_free_fast_reg_page_list(struct ib_fast_reg_page_list *page_list) +{ + struct mlx4_ib_dev *dev = to_mdev(page_list->device); + struct mlx4_ib_fast_reg_page_list *mfrpl = to_mfrpl(page_list); + int size = page_list->max_page_list_len * sizeof (u64); + + dma_free_coherent(&dev->dev->pdev->dev, size, page_list->page_list, + mfrpl->map); + kfree(mfrpl); +} + struct ib_fmr *mlx4_ib_fmr_alloc(struct ib_pd *pd, int acc, struct ib_fmr_attr *fmr_attr) { diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c index bda0859a5ac5..02a99bc4442e 100644 --- a/drivers/infiniband/hw/mlx4/qp.c +++ b/drivers/infiniband/hw/mlx4/qp.c @@ -78,6 +78,9 @@ static const __be32 mlx4_ib_opcode[] = { [IB_WR_RDMA_READ] = __constant_cpu_to_be32(MLX4_OPCODE_RDMA_READ), [IB_WR_ATOMIC_CMP_AND_SWP] = __constant_cpu_to_be32(MLX4_OPCODE_ATOMIC_CS), [IB_WR_ATOMIC_FETCH_AND_ADD] = __constant_cpu_to_be32(MLX4_OPCODE_ATOMIC_FA), + [IB_WR_SEND_WITH_INV] = __constant_cpu_to_be32(MLX4_OPCODE_SEND_INVAL), + [IB_WR_LOCAL_INV] = __constant_cpu_to_be32(MLX4_OPCODE_LOCAL_INVAL), + [IB_WR_FAST_REG_MR] = __constant_cpu_to_be32(MLX4_OPCODE_FMR), }; static struct mlx4_ib_sqp *to_msqp(struct mlx4_ib_qp *mqp) @@ -976,6 +979,10 @@ static int __mlx4_ib_modify_qp(struct ib_qp *ibqp, context->pd = cpu_to_be32(to_mpd(ibqp->pd)->pdn); context->params1 = cpu_to_be32(MLX4_IB_ACK_REQ_FREQ << 28); + /* Set "fast registration enabled" for all kernel QPs */ + if (!qp->ibqp.uobject) + context->params1 |= cpu_to_be32(1 << 11); + if (attr_mask & IB_QP_RNR_RETRY) { context->params1 |= cpu_to_be32(attr->rnr_retry << 13); optpar |= MLX4_QP_OPTPAR_RNR_RETRY; @@ -1322,6 +1329,38 @@ static int mlx4_wq_overflow(struct mlx4_ib_wq *wq, int nreq, struct ib_cq *ib_cq return cur + nreq >= wq->max_post; } +static __be32 convert_access(int acc) +{ + return (acc & IB_ACCESS_REMOTE_ATOMIC ? cpu_to_be32(MLX4_WQE_FMR_PERM_ATOMIC) : 0) | + (acc & IB_ACCESS_REMOTE_WRITE ? cpu_to_be32(MLX4_WQE_FMR_PERM_REMOTE_WRITE) : 0) | + (acc & IB_ACCESS_REMOTE_READ ? cpu_to_be32(MLX4_WQE_FMR_PERM_REMOTE_READ) : 0) | + (acc & IB_ACCESS_LOCAL_WRITE ? cpu_to_be32(MLX4_WQE_FMR_PERM_LOCAL_WRITE) : 0) | + cpu_to_be32(MLX4_WQE_FMR_PERM_LOCAL_READ); +} + +static void set_fmr_seg(struct mlx4_wqe_fmr_seg *fseg, struct ib_send_wr *wr) +{ + struct mlx4_ib_fast_reg_page_list *mfrpl = to_mfrpl(wr->wr.fast_reg.page_list); + + fseg->flags = convert_access(wr->wr.fast_reg.access_flags); + fseg->mem_key = cpu_to_be32(wr->wr.fast_reg.rkey); + fseg->buf_list = cpu_to_be64(mfrpl->map); + fseg->start_addr = cpu_to_be64(wr->wr.fast_reg.iova_start); + fseg->reg_len = cpu_to_be64(wr->wr.fast_reg.length); + fseg->offset = 0; /* XXX -- is this just for ZBVA? */ + fseg->page_size = cpu_to_be32(wr->wr.fast_reg.page_shift); + fseg->reserved[0] = 0; + fseg->reserved[1] = 0; +} + +static void set_local_inv_seg(struct mlx4_wqe_local_inval_seg *iseg, u32 rkey) +{ + iseg->flags = 0; + iseg->mem_key = cpu_to_be32(rkey); + iseg->guest_id = 0; + iseg->pa = 0; +} + static __always_inline void set_raddr_seg(struct mlx4_wqe_raddr_seg *rseg, u64 remote_addr, u32 rkey) { @@ -1423,6 +1462,21 @@ static int build_lso_seg(struct mlx4_wqe_lso_seg *wqe, struct ib_send_wr *wr, return 0; } +static __be32 send_ieth(struct ib_send_wr *wr) +{ + switch (wr->opcode) { + case IB_WR_SEND_WITH_IMM: + case IB_WR_RDMA_WRITE_WITH_IMM: + return wr->ex.imm_data; + + case IB_WR_SEND_WITH_INV: + return cpu_to_be32(wr->ex.invalidate_rkey); + + default: + return 0; + } +} + int mlx4_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, struct ib_send_wr **bad_wr) { @@ -1469,11 +1523,7 @@ int mlx4_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, MLX4_WQE_CTRL_TCP_UDP_CSUM) : 0) | qp->sq_signal_bits; - if (wr->opcode == IB_WR_SEND_WITH_IMM || - wr->opcode == IB_WR_RDMA_WRITE_WITH_IMM) - ctrl->imm = wr->ex.imm_data; - else - ctrl->imm = 0; + ctrl->imm = send_ieth(wr); wqe += sizeof *ctrl; size = sizeof *ctrl / 16; @@ -1505,6 +1555,18 @@ int mlx4_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, size += sizeof (struct mlx4_wqe_raddr_seg) / 16; break; + case IB_WR_LOCAL_INV: + set_local_inv_seg(wqe, wr->ex.invalidate_rkey); + wqe += sizeof (struct mlx4_wqe_local_inval_seg); + size += sizeof (struct mlx4_wqe_local_inval_seg) / 16; + break; + + case IB_WR_FAST_REG_MR: + set_fmr_seg(wqe, wr); + wqe += sizeof (struct mlx4_wqe_fmr_seg); + size += sizeof (struct mlx4_wqe_fmr_seg) / 16; + break; + default: /* No extra segments required for sends */ break; diff --git a/drivers/net/mlx4/fw.c b/drivers/net/mlx4/fw.c index 0851ebdddfd4..57278224ba1e 100644 --- a/drivers/net/mlx4/fw.c +++ b/drivers/net/mlx4/fw.c @@ -202,7 +202,7 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap) #define QUERY_DEV_CAP_C_MPT_ENTRY_SZ_OFFSET 0x8e #define QUERY_DEV_CAP_MTT_ENTRY_SZ_OFFSET 0x90 #define QUERY_DEV_CAP_D_MPT_ENTRY_SZ_OFFSET 0x92 -#define QUERY_DEV_CAP_BMME_FLAGS_OFFSET 0x97 +#define QUERY_DEV_CAP_BMME_FLAGS_OFFSET 0x94 #define QUERY_DEV_CAP_RSVD_LKEY_OFFSET 0x98 #define QUERY_DEV_CAP_MAX_ICM_SZ_OFFSET 0xa0 @@ -377,12 +377,8 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap) } } - if (dev_cap->bmme_flags & 1) - mlx4_dbg(dev, "Base MM extensions: yes " - "(flags %d, rsvd L_Key %08x)\n", - dev_cap->bmme_flags, dev_cap->reserved_lkey); - else - mlx4_dbg(dev, "Base MM extensions: no\n"); + mlx4_dbg(dev, "Base MM extensions: flags %08x, rsvd L_Key %08x\n", + dev_cap->bmme_flags, dev_cap->reserved_lkey); /* * Each UAR has 4 EQ doorbells; so if a UAR is reserved, then diff --git a/drivers/net/mlx4/fw.h b/drivers/net/mlx4/fw.h index a0e046c149b7..fbf0e22be122 100644 --- a/drivers/net/mlx4/fw.h +++ b/drivers/net/mlx4/fw.h @@ -98,7 +98,7 @@ struct mlx4_dev_cap { int cmpt_entry_sz; int mtt_entry_sz; int resize_srq; - u8 bmme_flags; + u32 bmme_flags; u32 reserved_lkey; u64 max_icm_sz; int max_gso_sz; diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c index d3736013fe9b..8e1d24cda1b0 100644 --- a/drivers/net/mlx4/main.c +++ b/drivers/net/mlx4/main.c @@ -158,6 +158,8 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap) dev->caps.max_msg_sz = dev_cap->max_msg_sz; dev->caps.page_size_cap = ~(u32) (dev_cap->min_page_sz - 1); dev->caps.flags = dev_cap->flags; + dev->caps.bmme_flags = dev_cap->bmme_flags; + dev->caps.reserved_lkey = dev_cap->reserved_lkey; dev->caps.stat_rate_support = dev_cap->stat_rate_support; dev->caps.max_gso_sz = dev_cap->max_gso_sz; diff --git a/drivers/net/mlx4/mr.c b/drivers/net/mlx4/mr.c index b3ea93b98689..a3c04c5f12c2 100644 --- a/drivers/net/mlx4/mr.c +++ b/drivers/net/mlx4/mr.c @@ -47,7 +47,7 @@ struct mlx4_mpt_entry { __be32 flags; __be32 qpn; __be32 key; - __be32 pd; + __be32 pd_flags; __be64 start; __be64 length; __be32 lkey; @@ -61,11 +61,15 @@ struct mlx4_mpt_entry { } __attribute__((packed)); #define MLX4_MPT_FLAG_SW_OWNS (0xfUL << 28) +#define MLX4_MPT_FLAG_FREE (0x3UL << 28) #define MLX4_MPT_FLAG_MIO (1 << 17) #define MLX4_MPT_FLAG_BIND_ENABLE (1 << 15) #define MLX4_MPT_FLAG_PHYSICAL (1 << 9) #define MLX4_MPT_FLAG_REGION (1 << 8) +#define MLX4_MPT_PD_FLAG_FAST_REG (1 << 26) +#define MLX4_MPT_PD_FLAG_EN_INV (3 << 24) + #define MLX4_MTT_FLAG_PRESENT 1 #define MLX4_MPT_STATUS_SW 0xF0 @@ -324,21 +328,30 @@ int mlx4_mr_enable(struct mlx4_dev *dev, struct mlx4_mr *mr) memset(mpt_entry, 0, sizeof *mpt_entry); - mpt_entry->flags = cpu_to_be32(MLX4_MPT_FLAG_SW_OWNS | - MLX4_MPT_FLAG_MIO | + mpt_entry->flags = cpu_to_be32(MLX4_MPT_FLAG_MIO | MLX4_MPT_FLAG_REGION | mr->access); mpt_entry->key = cpu_to_be32(key_to_hw_index(mr->key)); - mpt_entry->pd = cpu_to_be32(mr->pd); + mpt_entry->pd_flags = cpu_to_be32(mr->pd | MLX4_MPT_PD_FLAG_EN_INV); mpt_entry->start = cpu_to_be64(mr->iova); mpt_entry->length = cpu_to_be64(mr->size); mpt_entry->entity_size = cpu_to_be32(mr->mtt.page_shift); + if (mr->mtt.order < 0) { mpt_entry->flags |= cpu_to_be32(MLX4_MPT_FLAG_PHYSICAL); mpt_entry->mtt_seg = 0; - } else + } else { mpt_entry->mtt_seg = cpu_to_be64(mlx4_mtt_addr(dev, &mr->mtt)); + } + + if (mr->mtt.order >= 0 && mr->mtt.page_shift == 0) { + /* fast register MR in free state */ + mpt_entry->flags |= cpu_to_be32(MLX4_MPT_FLAG_FREE); + mpt_entry->pd_flags |= cpu_to_be32(MLX4_MPT_PD_FLAG_FAST_REG); + } else { + mpt_entry->flags |= cpu_to_be32(MLX4_MPT_FLAG_SW_OWNS); + } err = mlx4_SW2HW_MPT(dev, mailbox, key_to_hw_index(mr->key) & (dev->caps.num_mpts - 1)); diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h index 81b3dd5206e0..655ea0d1ee14 100644 --- a/include/linux/mlx4/device.h +++ b/include/linux/mlx4/device.h @@ -68,6 +68,14 @@ enum { MLX4_DEV_CAP_FLAG_UD_MCAST = 1 << 21 }; +enum { + MLX4_BMME_FLAG_LOCAL_INV = 1 << 6, + MLX4_BMME_FLAG_REMOTE_INV = 1 << 7, + MLX4_BMME_FLAG_TYPE_2_WIN = 1 << 9, + MLX4_BMME_FLAG_RESERVED_LKEY = 1 << 10, + MLX4_BMME_FLAG_FAST_REG_WR = 1 << 11, +}; + enum mlx4_event { MLX4_EVENT_TYPE_COMP = 0x00, MLX4_EVENT_TYPE_PATH_MIG = 0x01, @@ -184,6 +192,8 @@ struct mlx4_caps { u32 max_msg_sz; u32 page_size_cap; u32 flags; + u32 bmme_flags; + u32 reserved_lkey; u16 stat_rate_support; u8 port_width_cap[MLX4_MAX_PORTS + 1]; int max_gso_sz; diff --git a/include/linux/mlx4/qp.h b/include/linux/mlx4/qp.h index f02e9ed36cfa..e27082cd650e 100644 --- a/include/linux/mlx4/qp.h +++ b/include/linux/mlx4/qp.h @@ -233,6 +233,14 @@ struct mlx4_wqe_bind_seg { __be64 length; }; +enum { + MLX4_WQE_FMR_PERM_LOCAL_READ = 1 << 27, + MLX4_WQE_FMR_PERM_LOCAL_WRITE = 1 << 28, + MLX4_WQE_FMR_PERM_REMOTE_READ = 1 << 29, + MLX4_WQE_FMR_PERM_REMOTE_WRITE = 1 << 30, + MLX4_WQE_FMR_PERM_ATOMIC = 1 << 31 +}; + struct mlx4_wqe_fmr_seg { __be32 flags; __be32 mem_key; @@ -255,11 +263,11 @@ struct mlx4_wqe_fmr_ext_seg { }; struct mlx4_wqe_local_inval_seg { - u8 flags; - u8 reserved1[3]; + __be32 flags; + u32 reserved1; __be32 mem_key; - u8 reserved2[3]; - u8 guest_id; + u32 reserved2[2]; + __be32 guest_id; __be64 pa; }; From 76442640829163d0cdb67c2bf0cb4b81a0fe537b Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Wed, 23 Jul 2008 08:12:47 -0700 Subject: [PATCH 077/857] mlx4_core: Improve error message when not enough UAR pages are available If an mlx4 device with default FW (which gives a UAR BAR size of 8 MB) is used in a system with 64 KB pages, then there are only 8192/64==128 UAR pages available. However, the first 128 UAR pages are reserved for use with event queue doorbells, so no UAR pages are available to do anything else with, which means that the driver cannot work. The current driver fails with a fairly cryptic "Failed to allocate driver access region, aborting" message in this situation. Fix the driver to detect the problem earlier and print out a clearer description of the problem and a suggestion of how to fix it (use a new firmware image). Signed-off-by: Roland Dreier --- drivers/net/mlx4/pd.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/net/mlx4/pd.c b/drivers/net/mlx4/pd.c index 3a93c5f0f7ab..aa616892d09c 100644 --- a/drivers/net/mlx4/pd.c +++ b/drivers/net/mlx4/pd.c @@ -91,6 +91,13 @@ EXPORT_SYMBOL_GPL(mlx4_uar_free); int mlx4_init_uar_table(struct mlx4_dev *dev) { + if (dev->caps.num_uars <= 128) { + mlx4_err(dev, "Only %d UAR pages (need more than 128)\n", + dev->caps.num_uars); + mlx4_err(dev, "Increase firmware log2_uar_bar_megabytes?\n"); + return -ENODEV; + } + return mlx4_bitmap_init(&mlx4_priv(dev)->uar_table.bitmap, dev->caps.num_uars, dev->caps.num_uars - 1, max(128, dev->caps.reserved_uars)); From 5b3ab1dbd401b36ba2f9bfee2d2dae252fd62cd8 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 23 Jul 2008 14:01:29 -0700 Subject: [PATCH 078/857] netdev: Remove warning from __netif_schedule(). It isn't helping anything and we aren't going to be able to change all the drivers that do queue wakeups in strange situations. Just letting a noop_qdisc get scheduled will work because when qdisc_run() executes via net_tx_work() it will simply find no packets pending when it makes the ->dequeue() call in qdisc_restart. Signed-off-by: David S. Miller --- net/core/dev.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/net/core/dev.c b/net/core/dev.c index 6bf217da9d8f..ccf97f9f37eb 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1341,9 +1341,6 @@ static void dev_queue_xmit_nit(struct sk_buff *skb, struct net_device *dev) void __netif_schedule(struct Qdisc *q) { - if (WARN_ON_ONCE(q == &noop_qdisc)) - return; - if (!test_and_set_bit(__QDISC_STATE_SCHED, &q->state)) { struct softnet_data *sd; unsigned long flags; From b4942af65028c5eb516fdd9053020ccb2ee186ce Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Wed, 23 Jul 2008 14:06:04 -0700 Subject: [PATCH 079/857] net: Update entry in af_family_clock_key_strings In the merge phase of the CAN subsystem the af_family_clock_key_strings[] have been added to sock.c in commit 443aef0eddfa44c158d1b94ebb431a70638fcab4 (lockdep: fixup sk_callback_lock annotation). This trivial patch adds the missing name for address family 29 (AF_CAN). Signed-off-by: Oliver Hartkopp Signed-off-by: David S. Miller --- net/core/sock.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/core/sock.c b/net/core/sock.c index 10a64d57078c..91f8bbc93526 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -180,7 +180,7 @@ static const char *af_family_clock_key_strings[AF_MAX+1] = { "clock-AF_ASH" , "clock-AF_ECONET" , "clock-AF_ATMSVC" , "clock-21" , "clock-AF_SNA" , "clock-AF_IRDA" , "clock-AF_PPPOX" , "clock-AF_WANPIPE" , "clock-AF_LLC" , - "clock-27" , "clock-28" , "clock-29" , + "clock-27" , "clock-28" , "clock-AF_CAN" , "clock-AF_TIPC" , "clock-AF_BLUETOOTH", "clock-AF_IUCV" , "clock-AF_RXRPC" , "clock-AF_MAX" }; From 1fa6d8181b7bb0361512170c30e436dcc95591ee Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Wed, 23 Jul 2008 14:20:12 -0700 Subject: [PATCH 080/857] MAINTAINERS: Remove Glenn Streiff from NetEffect entry Glenn is no longer at NetEffect. Signed-off-by: Roland Dreier --- MAINTAINERS | 2 -- 1 file changed, 2 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 11944b44c2ff..2a73da0cd075 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2915,8 +2915,6 @@ P: Faisal Latif M: flatif@neteffect.com P: Chien Tung M: ctung@neteffect.com -P: Glenn Streiff -M: gstreiff@neteffect.com L: general@lists.openfabrics.org W: http://www.neteffect.com S: Supported From e8ebe3b893792887317bc24cc4608753f81b81d3 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Wed, 23 Jul 2008 15:30:52 -0700 Subject: [PATCH 081/857] e1000e: fix e1000_netpoll(), remove extraneous e1000_clean_tx_irq() call Evgeniy Polyakov noticed that drivers/net/e1000e/netdev.c:e1000_netpoll() was calling e1000_clean_tx_irq() without taking the TX lock. David Miller suggested to remove the call altogether: since in this callpah there's periodic calls to ->poll() anyway which will do e1000_clean_tx_irq() and will garbage-collect any finished TX ring descriptors. This fix solved the e1000e+netconsole crashes i've been seeing: ============================================================================= BUG skbuff_head_cache: Poison overwritten ----------------------------------------------------------------------------- INFO: 0xf658ae9c-0xf658ae9c. First byte 0x6a instead of 0x6b INFO: Allocated in __alloc_skb+0x2c/0x110 age=0 cpu=0 pid=5098 INFO: Freed in __kfree_skb+0x31/0x80 age=0 cpu=1 pid=4440 INFO: Slab 0xc16cc140 objects=16 used=1 fp=0xf658ae00 flags=0x400000c3 INFO: Object 0xf658ae00 @offset=3584 fp=0xf658af00 Signed-off-by: Ingo Molnar Signed-off-by: David S. Miller --- drivers/net/e1000e/netdev.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/net/e1000e/netdev.c b/drivers/net/e1000e/netdev.c index 869544b8c05c..9c0f56b3c518 100644 --- a/drivers/net/e1000e/netdev.c +++ b/drivers/net/e1000e/netdev.c @@ -4067,8 +4067,6 @@ static void e1000_netpoll(struct net_device *netdev) disable_irq(adapter->pdev->irq); e1000_intr(adapter->pdev->irq, netdev); - e1000_clean_tx_irq(adapter); - enable_irq(adapter->pdev->irq); } #endif From 7ae93f51d7fa8b9130d47e0b7d17979a165c5bc3 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 23 Jul 2008 16:21:07 -0700 Subject: [PATCH 082/857] sparc64: Fix cpufreq notifier registry. Based upon a report by Daniel Smolik. We do it too early, which triggers a BUG in cpufreq_register_notifier(). Signed-off-by: David S. Miller --- arch/sparc64/kernel/time.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) diff --git a/arch/sparc64/kernel/time.c b/arch/sparc64/kernel/time.c index bedc4c159b1c..a0c6a97eec6e 100644 --- a/arch/sparc64/kernel/time.c +++ b/arch/sparc64/kernel/time.c @@ -884,6 +884,16 @@ static struct notifier_block sparc64_cpufreq_notifier_block = { .notifier_call = sparc64_cpufreq_notifier }; +static int __init register_sparc64_cpufreq_notifier(void) +{ + + cpufreq_register_notifier(&sparc64_cpufreq_notifier_block, + CPUFREQ_TRANSITION_NOTIFIER); + return 0; +} + +core_initcall(register_sparc64_cpufreq_notifier); + #endif /* CONFIG_CPU_FREQ */ static int sparc64_next_event(unsigned long delta, @@ -1050,11 +1060,6 @@ void __init time_init(void) sparc64_clockevent.mult, sparc64_clockevent.shift); setup_sparc64_timer(); - -#ifdef CONFIG_CPU_FREQ - cpufreq_register_notifier(&sparc64_cpufreq_notifier_block, - CPUFREQ_TRANSITION_NOTIFIER); -#endif } unsigned long long sched_clock(void) From 4b53fb67e385b856a991d402096379dab462170a Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 23 Jul 2008 16:38:45 -0700 Subject: [PATCH 083/857] tcp: Clear probes_out more aggressively in tcp_ack(). This is based upon an excellent bug report from Eric Dumazet. tcp_ack() should clear ->icsk_probes_out even if there are packets outstanding. Otherwise if we get a sequence of ACKs while we do have packets outstanding over and over again, we'll never clear the probes_out value and eventually think the connection is too sick and we'll reset it. This appears to be some "optimization" added to tcp_ack() in the 2.4.x timeframe. In 2.2.x, probes_out is pretty much always cleared by tcp_ack(). Here is Eric's original report: ---------------------------------------- Apparently, we can in some situations reset TCP connections in a couple of seconds when some frames are lost. In order to reproduce the problem, please try the following program on linux-2.6.25.* Setup some iptables rules to allow two frames per second sent on loopback interface to tcp destination port 12000 iptables -N SLOWLO iptables -A SLOWLO -m hashlimit --hashlimit 2 --hashlimit-burst 1 --hashlimit-mode dstip --hashlimit-name slow2 -j ACCEPT iptables -A SLOWLO -j DROP iptables -A OUTPUT -o lo -p tcp --dport 12000 -j SLOWLO Then run the attached program and see the output : # ./loop State Recv-Q Send-Q Local Address:Port Peer Address:Port ESTAB 0 40 127.0.0.1:54455 127.0.0.1:12000 timer:(persist,200ms,1) State Recv-Q Send-Q Local Address:Port Peer Address:Port ESTAB 0 40 127.0.0.1:54455 127.0.0.1:12000 timer:(persist,200ms,3) State Recv-Q Send-Q Local Address:Port Peer Address:Port ESTAB 0 40 127.0.0.1:54455 127.0.0.1:12000 timer:(persist,200ms,5) State Recv-Q Send-Q Local Address:Port Peer Address:Port ESTAB 0 40 127.0.0.1:54455 127.0.0.1:12000 timer:(persist,200ms,7) State Recv-Q Send-Q Local Address:Port Peer Address:Port ESTAB 0 40 127.0.0.1:54455 127.0.0.1:12000 timer:(persist,200ms,9) State Recv-Q Send-Q Local Address:Port Peer Address:Port ESTAB 0 40 127.0.0.1:54455 127.0.0.1:12000 timer:(persist,200ms,11) State Recv-Q Send-Q Local Address:Port Peer Address:Port ESTAB 0 40 127.0.0.1:54455 127.0.0.1:12000 timer:(persist,201ms,13) State Recv-Q Send-Q Local Address:Port Peer Address:Port ESTAB 0 40 127.0.0.1:54455 127.0.0.1:12000 timer:(persist,188ms,15) write(): Connection timed out wrote 890 bytes but was interrupted after 9 seconds ESTAB 0 0 127.0.0.1:12000 127.0.0.1:54455 Exiting read() because no data available (4000 ms timeout). read 860 bytes While this tcp session makes progress (sending frames with 50 bytes of payload, every 500ms), linux tcp stack decides to reset it, when tcp_retries 2 is reached (default value : 15) tcpdump : 15:30:28.856695 IP 127.0.0.1.56554 > 127.0.0.1.12000: S 33788768:33788768(0) win 32792 15:30:28.856711 IP 127.0.0.1.12000 > 127.0.0.1.56554: S 33899253:33899253(0) ack 33788769 win 32792 15:30:29.356947 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 1:61(60) ack 1 win 257 15:30:29.356966 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 61 win 257 15:30:29.866415 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 61:111(50) ack 1 win 257 15:30:29.866427 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 111 win 257 15:30:30.366516 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 111:161(50) ack 1 win 257 15:30:30.366527 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 161 win 257 15:30:30.876196 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 161:211(50) ack 1 win 257 15:30:30.876207 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 211 win 257 15:30:31.376282 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 211:261(50) ack 1 win 257 15:30:31.376290 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 261 win 257 15:30:31.885619 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 261:311(50) ack 1 win 257 15:30:31.885631 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 311 win 257 15:30:32.385705 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 311:361(50) ack 1 win 257 15:30:32.385715 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 361 win 257 15:30:32.895249 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 361:411(50) ack 1 win 257 15:30:32.895266 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 411 win 257 15:30:33.395341 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 411:461(50) ack 1 win 257 15:30:33.395351 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 461 win 257 15:30:33.918085 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 461:511(50) ack 1 win 257 15:30:33.918096 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 511 win 257 15:30:34.418163 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 511:561(50) ack 1 win 257 15:30:34.418172 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 561 win 257 15:30:34.927685 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 561:611(50) ack 1 win 257 15:30:34.927698 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 611 win 257 15:30:35.427757 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 611:661(50) ack 1 win 257 15:30:35.427766 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 661 win 257 15:30:35.937359 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 661:711(50) ack 1 win 257 15:30:35.937376 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 711 win 257 15:30:36.437451 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 711:761(50) ack 1 win 257 15:30:36.437464 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 761 win 257 15:30:36.947022 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 761:811(50) ack 1 win 257 15:30:36.947039 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 811 win 257 15:30:37.447135 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 811:861(50) ack 1 win 257 15:30:37.447203 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 861 win 257 15:30:41.448171 IP 127.0.0.1.12000 > 127.0.0.1.56554: F 1:1(0) ack 861 win 257 15:30:41.448189 IP 127.0.0.1.56554 > 127.0.0.1.12000: R 33789629:33789629(0) win 0 Source of program : /* * small producer/consumer program. * setup a listener on 127.0.0.1:12000 * Forks a child * child connect to 127.0.0.1, and sends 10 bytes on this tcp socket every 100 ms * Father accepts connection, and read all data */ #include #include #include #include #include #include #include int port = 12000; char buffer[4096]; int main(int argc, char *argv[]) { int lfd = socket(AF_INET, SOCK_STREAM, 0); struct sockaddr_in socket_address; time_t t0, t1; int on = 1, sfd, res; unsigned long total = 0; socklen_t alen = sizeof(socket_address); pid_t pid; time(&t0); socket_address.sin_family = AF_INET; socket_address.sin_port = htons(port); socket_address.sin_addr.s_addr = htonl(INADDR_LOOPBACK); if (lfd == -1) { perror("socket()"); return 1; } setsockopt(lfd, SOL_SOCKET, SO_REUSEADDR, &on, sizeof(int)); if (bind(lfd, (struct sockaddr *)&socket_address, sizeof(socket_address)) == -1) { perror("bind"); close(lfd); return 1; } if (listen(lfd, 1) == -1) { perror("listen()"); close(lfd); return 1; } pid = fork(); if (pid == 0) { int i, cfd = socket(AF_INET, SOCK_STREAM, 0); close(lfd); if (connect(cfd, (struct sockaddr *)&socket_address, sizeof(socket_address)) == -1) { perror("connect()"); return 1; } for (i = 0 ; ;) { res = write(cfd, "blablabla\n", 10); if (res > 0) total += res; else if (res == -1) { perror("write()"); break; } else break; usleep(100000); if (++i == 10) { system("ss -on dst 127.0.0.1:12000"); i = 0; } } time(&t1); fprintf(stderr, "wrote %lu bytes but was interrupted after %g seconds\n", total, difftime(t1, t0)); system("ss -on | grep 127.0.0.1:12000"); close(cfd); return 0; } sfd = accept(lfd, (struct sockaddr *)&socket_address, &alen); if (sfd == -1) { perror("accept"); return 1; } close(lfd); while (1) { struct pollfd pfd[1]; pfd[0].fd = sfd; pfd[0].events = POLLIN; if (poll(pfd, 1, 4000) == 0) { fprintf(stderr, "Exiting read() because no data available (4000 ms timeout).\n"); break; } res = read(sfd, buffer, sizeof(buffer)); if (res > 0) total += res; else if (res == 0) break; else perror("read()"); } fprintf(stderr, "read %lu bytes\n", total); close(sfd); return 0; } ---------------------------------------- Signed-off-by: David S. Miller --- net/ipv4/tcp_input.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 1f5e6049883e..75efd244f2af 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -3292,6 +3292,7 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag) * log. Something worked... */ sk->sk_err_soft = 0; + icsk->icsk_probes_out = 0; tp->rcv_tstamp = tcp_time_stamp; prior_packets = tp->packets_out; if (!prior_packets) @@ -3324,8 +3325,6 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag) return 1; no_queue: - icsk->icsk_probes_out = 0; - /* If this ack opens up a zero window, clear backoff. It was * being used to time the probes, and is probably far higher than * it needs to be for normal retransmission. From 70eed75d76635ba7350651b9bd96529a306ec67a Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Wed, 23 Jul 2008 16:42:42 -0700 Subject: [PATCH 084/857] netfilter: make security table depend on NETFILTER_ADVANCED Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/ipv4/netfilter/Kconfig | 2 +- net/ipv6/netfilter/Kconfig | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index f23e60c93ef9..90eb7cb47e77 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -369,7 +369,7 @@ config IP_NF_SECURITY tristate "Security table" depends on IP_NF_IPTABLES depends on SECURITY - default m if NETFILTER_ADVANCED=n + depends on NETFILTER_ADVANCED help This option adds a `security' table to iptables, for use with Mandatory Access Control (MAC) policy. diff --git a/net/ipv6/netfilter/Kconfig b/net/ipv6/netfilter/Kconfig index 689dec899c57..0cfcce7b18d8 100644 --- a/net/ipv6/netfilter/Kconfig +++ b/net/ipv6/netfilter/Kconfig @@ -213,7 +213,7 @@ config IP6_NF_SECURITY tristate "Security table" depends on IP6_NF_IPTABLES depends on SECURITY - default m if NETFILTER_ADVANCED=n + depends on NETFILTER_ADVANCED help This option adds a `security' table to iptables, for use with Mandatory Access Control (MAC) policy. From 0855b543222e79cbbd9d66dd56cb54740e7d524f Mon Sep 17 00:00:00 2001 From: Andre Detsch Date: Thu, 24 Jul 2008 10:57:26 +1000 Subject: [PATCH 085/857] powerpc/spufs: fix aff_mutex and cbe_spu_info[n].list_mutex deadlock Currenlt,, it is possible to lock aff_mutex and cbe_spu_info[n].list_mutex in different orders, allowing a deadlock to occur. With this change, aff_mutex is not taken within a list_mutex critical section anymore. Signed-off-by: Andre Detsch Signed-off-by: Jeremy Kerr --- arch/powerpc/platforms/cell/spufs/sched.c | 18 ++++++------------ 1 file changed, 6 insertions(+), 12 deletions(-) diff --git a/arch/powerpc/platforms/cell/spufs/sched.c b/arch/powerpc/platforms/cell/spufs/sched.c index 34654743363d..f293963cd855 100644 --- a/arch/powerpc/platforms/cell/spufs/sched.c +++ b/arch/powerpc/platforms/cell/spufs/sched.c @@ -389,6 +389,9 @@ static int has_affinity(struct spu_context *ctx) if (list_empty(&ctx->aff_list)) return 0; + if (atomic_read(&ctx->gang->aff_sched_count) == 0) + ctx->gang->aff_ref_spu = NULL; + if (!gang->aff_ref_spu) { if (!(gang->aff_flags & AFF_MERGED)) aff_merge_remaining_ctxs(gang); @@ -416,14 +419,8 @@ static void spu_unbind_context(struct spu *spu, struct spu_context *ctx) if (spu->ctx->flags & SPU_CREATE_NOSCHED) atomic_dec(&cbe_spu_info[spu->node].reserved_spus); - if (ctx->gang){ - mutex_lock(&ctx->gang->aff_mutex); - if (has_affinity(ctx)) { - if (atomic_dec_and_test(&ctx->gang->aff_sched_count)) - ctx->gang->aff_ref_spu = NULL; - } - mutex_unlock(&ctx->gang->aff_mutex); - } + if (ctx->gang) + atomic_dec_if_positive(&ctx->gang->aff_sched_count); spu_switch_notify(spu, NULL); spu_unmap_mappings(ctx); @@ -562,10 +559,7 @@ static struct spu *spu_get_idle(struct spu_context *ctx) goto found; mutex_unlock(&cbe_spu_info[node].list_mutex); - mutex_lock(&ctx->gang->aff_mutex); - if (atomic_dec_and_test(&ctx->gang->aff_sched_count)) - ctx->gang->aff_ref_spu = NULL; - mutex_unlock(&ctx->gang->aff_mutex); + atomic_dec(&ctx->gang->aff_sched_count); goto not_found; } mutex_unlock(&ctx->gang->aff_mutex); From ad1ede127760d6ca4903f44dfe1a8a38b3bfb36c Mon Sep 17 00:00:00 2001 From: Andre Detsch Date: Thu, 24 Jul 2008 11:01:54 +1000 Subject: [PATCH 086/857] powerpc/spufs: better placement of spu affinity reference context This patch adjusts the placement of a reference context from a spu affinity chain. The reference context can now be placed only on nodes that have enough spus not intended to be used by another gang (already running on the node). Signed-off-by: Andre Detsch Signed-off-by: Jeremy Kerr --- arch/powerpc/platforms/cell/spufs/sched.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/arch/powerpc/platforms/cell/spufs/sched.c b/arch/powerpc/platforms/cell/spufs/sched.c index f293963cd855..2deeeba7eccf 100644 --- a/arch/powerpc/platforms/cell/spufs/sched.c +++ b/arch/powerpc/platforms/cell/spufs/sched.c @@ -312,10 +312,27 @@ static struct spu *aff_ref_location(struct spu_context *ctx, int mem_aff, */ node = cpu_to_node(raw_smp_processor_id()); for (n = 0; n < MAX_NUMNODES; n++, node++) { + int available_spus; + node = (node < MAX_NUMNODES) ? node : 0; if (!node_allowed(ctx, node)) continue; + + available_spus = 0; mutex_lock(&cbe_spu_info[node].list_mutex); + list_for_each_entry(spu, &cbe_spu_info[node].spus, cbe_list) { + if (spu->ctx && spu->ctx->gang + && spu->ctx->aff_offset == 0) + available_spus -= + (spu->ctx->gang->contexts - 1); + else + available_spus++; + } + if (available_spus < ctx->gang->contexts) { + mutex_unlock(&cbe_spu_info[node].list_mutex); + continue; + } + list_for_each_entry(spu, &cbe_spu_info[node].spus, cbe_list) { if ((!mem_aff || spu->has_mem_affinity) && sched_spu(spu)) { From 6f75a9b6426e686649ac440c37ec7c249501f9a5 Mon Sep 17 00:00:00 2001 From: Chas Williams Date: Wed, 23 Jul 2008 20:29:21 -0700 Subject: [PATCH 087/857] atm: [fore200e] use MODULE_FIRMWARE() and other suggested cleanups Signed-off-by: Chas Williams Signed-off-by: David S. Miller --- drivers/atm/fore200e.c | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) diff --git a/drivers/atm/fore200e.c b/drivers/atm/fore200e.c index d5c1bbfbe79d..73338d231db9 100644 --- a/drivers/atm/fore200e.c +++ b/drivers/atm/fore200e.c @@ -2562,7 +2562,8 @@ fore200e_load_and_start_fw(struct fore200e* fore200e) const struct firmware *firmware; struct device *device; struct fw_header *fw_header; - u32 *fw_data, fw_size; + const __le32 *fw_data; + u32 fw_size; u32 __iomem *load_addr; char buf[48]; int err = -ENODEV; @@ -2582,7 +2583,7 @@ fore200e_load_and_start_fw(struct fore200e* fore200e) return err; } - fw_data = (u32 *) firmware->data; + fw_data = (__le32 *) firmware->data; fw_size = firmware->size / sizeof(u32); fw_header = (struct fw_header *) firmware->data; load_addr = fore200e->virt_base + le32_to_cpu(fw_header->load_offset); @@ -3199,6 +3200,14 @@ static const struct fore200e_bus fore200e_bus[] = { {} }; -#ifdef MODULE_LICENSE MODULE_LICENSE("GPL"); +#ifdef CONFIG_PCI +#ifdef __LITTLE_ENDIAN__ +MODULE_FIRMWARE("pca200e.bin"); +#else +MODULE_FIRMWARE("pca200e_ecd.bin2"); +#endif +#endif /* CONFIG_PCI */ +#ifdef CONFIG_SBUS +MODULE_FIRMWARE("sba200e_ecd.bin2"); #endif From f867e6af94239a04ec23aeec2fcda5aa58e41db7 Mon Sep 17 00:00:00 2001 From: Jarek Poplawski Date: Wed, 23 Jul 2008 21:34:27 -0700 Subject: [PATCH 088/857] pkt_sched: sch_sfq: dump a real number of flows Dump the "flows" number according to the number of active flows instead of repeating the "limit". Reported-by: Denys Fedoryshchenko Signed-off-by: Jarek Poplawski Signed-off-by: David S. Miller --- net/sched/sch_sfq.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c index 8589da666568..73f53844ce97 100644 --- a/net/sched/sch_sfq.c +++ b/net/sched/sch_sfq.c @@ -536,7 +536,14 @@ static int sfq_dump(struct Qdisc *sch, struct sk_buff *skb) opt.limit = q->limit; opt.divisor = SFQ_HASH_DIVISOR; - opt.flows = q->limit; + opt.flows = 0; + if (q->tail != SFQ_DEPTH) { + unsigned int i; + + for (i = 0; i < SFQ_HASH_DIVISOR; i++) + if (q->ht[i] != SFQ_DEPTH) + opt.flows++; + } NLA_PUT(skb, TCA_OPTIONS, sizeof(opt), &opt); From 9e882c9282512cc622752f29ec7c29ce338fc1eb Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Mon, 21 Jul 2008 16:49:54 -0700 Subject: [PATCH 089/857] x86: call early_cpu_init at the same point Call early_cpu_init() at the same (early) point in setup_arch(). The x86_64 code was calling it relatively late, after when other arch code need to do cpu-related setup which depends on it. Signed-off-by: Jeremy Fitzhardinge Cc: Mark McLoughlin Cc: Eduardo Habkost Signed-off-by: Ingo Molnar --- arch/x86/kernel/setup.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index b4aacb9f52e3..b520dae02bf4 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -597,11 +597,11 @@ void __init setup_arch(char **cmdline_p) memcpy(&boot_cpu_data, &new_cpu_data, sizeof(new_cpu_data)); visws_early_detect(); pre_setup_arch_hook(); - early_cpu_init(); #else printk(KERN_INFO "Command line: %s\n", boot_command_line); #endif + early_cpu_init(); early_ioremap_init(); ROOT_DEV = old_decode_dev(boot_params.hdr.root_dev); @@ -665,9 +665,6 @@ void __init setup_arch(char **cmdline_p) bss_resource.start = virt_to_phys(&__bss_start); bss_resource.end = virt_to_phys(&__bss_stop)-1; -#ifdef CONFIG_X86_64 - early_cpu_init(); -#endif strlcpy(command_line, boot_command_line, COMMAND_LINE_SIZE); *cmdline_p = command_line; From 2dc1697eb355c34f9f7bcbbb83f490de248c360a Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Mon, 21 Jul 2008 16:49:58 -0700 Subject: [PATCH 090/857] xen: don't use sysret for sysexit32 When implementing sysexit32, don't let Xen use sysret to return to userspace. That results in usermode register state being trashed. Signed-off-by: Jeremy Fitzhardinge Cc: Mark McLoughlin Cc: Eduardo Habkost Signed-off-by: Ingo Molnar --- arch/x86/xen/xen-asm_64.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/xen/xen-asm_64.S b/arch/x86/xen/xen-asm_64.S index 4038cbfe3331..7f58304fafb3 100644 --- a/arch/x86/xen/xen-asm_64.S +++ b/arch/x86/xen/xen-asm_64.S @@ -173,7 +173,7 @@ ENTRY(xen_sysexit) pushq $__USER32_CS pushq %rdx - pushq $VGCF_in_syscall + pushq $0 1: jmp hypercall_iret ENDPATCH(xen_sysexit) RELOC(xen_sysexit, 1b+1) From 979c9296bdcfded58ebac41905c3397317df0355 Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Wed, 14 May 2008 16:10:33 +0300 Subject: [PATCH 091/857] UBI: print error code Print error code if checking failed which is very useful to identify problems. Signed-off-by: Artem Bityutskiy --- drivers/mtd/ubi/vtbl.c | 39 +++++++++++++++++++++------------------ 1 file changed, 21 insertions(+), 18 deletions(-) diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c index af36b12be278..3c4d68f2cfd4 100644 --- a/drivers/mtd/ubi/vtbl.c +++ b/drivers/mtd/ubi/vtbl.c @@ -127,7 +127,7 @@ static int vtbl_check(const struct ubi_device *ubi, const struct ubi_vtbl_record *vtbl) { int i, n, reserved_pebs, alignment, data_pad, vol_type, name_len; - int upd_marker; + int upd_marker, err; uint32_t crc; const char *name; @@ -153,7 +153,7 @@ static int vtbl_check(const struct ubi_device *ubi, if (reserved_pebs == 0) { if (memcmp(&vtbl[i], &empty_vtbl_record, UBI_VTBL_RECORD_SIZE)) { - dbg_err("bad empty record"); + err = 2; goto bad; } continue; @@ -161,56 +161,57 @@ static int vtbl_check(const struct ubi_device *ubi, if (reserved_pebs < 0 || alignment < 0 || data_pad < 0 || name_len < 0) { - dbg_err("negative values"); + err = 3; goto bad; } if (alignment > ubi->leb_size || alignment == 0) { - dbg_err("bad alignment"); + err = 4; goto bad; } n = alignment % ubi->min_io_size; if (alignment != 1 && n) { - dbg_err("alignment is not multiple of min I/O unit"); + err = 5; goto bad; } n = ubi->leb_size % alignment; if (data_pad != n) { dbg_err("bad data_pad, has to be %d", n); + err = 6; goto bad; } if (vol_type != UBI_VID_DYNAMIC && vol_type != UBI_VID_STATIC) { - dbg_err("bad vol_type"); + err = 7; goto bad; } if (upd_marker != 0 && upd_marker != 1) { - dbg_err("bad upd_marker"); + err = 8; goto bad; } if (reserved_pebs > ubi->good_peb_count) { dbg_err("too large reserved_pebs, good PEBs %d", ubi->good_peb_count); + err = 9; goto bad; } if (name_len > UBI_VOL_NAME_MAX) { - dbg_err("too long volume name, max %d", - UBI_VOL_NAME_MAX); + err = 10; goto bad; } if (name[0] == '\0') { - dbg_err("NULL volume name"); + err = 11; goto bad; } if (name_len != strnlen(name, name_len + 1)) { - dbg_err("bad name_len"); + err = 12; goto bad; } } @@ -235,7 +236,7 @@ static int vtbl_check(const struct ubi_device *ubi, return 0; bad: - ubi_err("volume table check failed, record %d", i); + ubi_err("volume table check failed: record %d, error %d", i, err); ubi_dbg_dump_vtbl_record(&vtbl[i], i); return -EINVAL; } @@ -620,30 +621,32 @@ static int init_volumes(struct ubi_device *ubi, const struct ubi_scan_info *si, static int check_sv(const struct ubi_volume *vol, const struct ubi_scan_volume *sv) { + int err; + if (sv->highest_lnum >= vol->reserved_pebs) { - dbg_err("bad highest_lnum"); + err = 1; goto bad; } if (sv->leb_count > vol->reserved_pebs) { - dbg_err("bad leb_count"); + err = 2; goto bad; } if (sv->vol_type != vol->vol_type) { - dbg_err("bad vol_type"); + err = 3; goto bad; } if (sv->used_ebs > vol->reserved_pebs) { - dbg_err("bad used_ebs"); + err = 4; goto bad; } if (sv->data_pad != vol->data_pad) { - dbg_err("bad data_pad"); + err = 5; goto bad; } return 0; bad: - ubi_err("bad scanning information"); + ubi_err("bad scanning information, error %d", err); ubi_dbg_dump_sv(sv); ubi_dbg_dump_vol_info(vol); return -EINVAL; From beeea636030622f6de67d15c61f5b311a03d188c Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Tue, 20 May 2008 09:54:02 +0300 Subject: [PATCH 092/857] UBI: add a comment It is not clear why we schedule PEB for scrubbing in case of -EBADMSG. Elaborate. Requested-by: Kyungmin Park Signed-off-by: Artem Bityutskiy --- drivers/mtd/ubi/vtbl.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c index 3c4d68f2cfd4..42a7815086b7 100644 --- a/drivers/mtd/ubi/vtbl.c +++ b/drivers/mtd/ubi/vtbl.c @@ -385,7 +385,16 @@ static struct ubi_vtbl_record *process_lvol(struct ubi_device *ubi, err = ubi_io_read_data(ubi, leb[seb->lnum], seb->pnum, 0, ubi->vtbl_size); if (err == UBI_IO_BITFLIPS || err == -EBADMSG) - /* Scrub the PEB later */ + /* + * Scrub the PEB later. Note, -EBADMSG indicates an + * uncorrectable ECC error, but we have our own CRC and + * the data will be checked later. If the data is OK, + * the PEB will be scrubbed (because we set + * seb->scrub). If the data is not OK, the contents of + * the PEB will be recovered from the second copy, and + * seb->scrub will be cleared in + * 'ubi_scan_add_used()'. + */ seb->scrub = 1; else if (err) goto out_free; From a0fd1efd488092951f310fdb777b8a540cf84dcb Mon Sep 17 00:00:00 2001 From: Kyungmin Park Date: Wed, 21 May 2008 14:34:56 +0300 Subject: [PATCH 093/857] UBI: fix buffer padding Instead of correctly pad the buffer wich we are writing to the eraseblock during update, we used weird construct: memset(buf + len, 0xFF, len - len); Fix this. Signed-off-by: Kyungmin Park Signed-off-by: Artem Bityutskiy --- drivers/mtd/ubi/upd.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/mtd/ubi/upd.c b/drivers/mtd/ubi/upd.c index ddaa1a56cc69..6fa1ab3f2a70 100644 --- a/drivers/mtd/ubi/upd.c +++ b/drivers/mtd/ubi/upd.c @@ -237,10 +237,10 @@ static int write_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum, int err; if (vol->vol_type == UBI_DYNAMIC_VOLUME) { - len = ALIGN(len, ubi->min_io_size); - memset(buf + len, 0xFF, len - len); + int l = ALIGN(len, ubi->min_io_size); - len = ubi_calc_data_len(ubi, buf, len); + memset(buf + len, 0xFF, l - len); + len = ubi_calc_data_len(ubi, buf, l); if (len == 0) { dbg_msg("all %d bytes contain 0xFF - skip", len); return 0; From cadb40ccc16a26a738f1cbc963e35b21edd93e79 Mon Sep 17 00:00:00 2001 From: Kyungmin Park Date: Thu, 22 May 2008 10:32:18 +0900 Subject: [PATCH 094/857] UBI: avoid unnecessary division operations UBI already checks that @min io size is the power of 2 at io_init. It is save to use bit operations then. Signed-off-by: Kyungmin Park Signed-off-by: Artem Bityutskiy --- drivers/mtd/ubi/build.c | 8 ++++++-- drivers/mtd/ubi/cdev.c | 6 +++--- drivers/mtd/ubi/eba.c | 2 +- drivers/mtd/ubi/kapi.c | 6 +++--- drivers/mtd/ubi/misc.c | 2 +- drivers/mtd/ubi/vmt.c | 2 +- drivers/mtd/ubi/vtbl.c | 5 ++--- drivers/mtd/ubi/wl.c | 3 +-- 8 files changed, 18 insertions(+), 16 deletions(-) diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c index 961416ac0616..ff4425de1527 100644 --- a/drivers/mtd/ubi/build.c +++ b/drivers/mtd/ubi/build.c @@ -530,7 +530,11 @@ static int io_init(struct ubi_device *ubi) ubi->min_io_size = ubi->mtd->writesize; ubi->hdrs_min_io_size = ubi->mtd->writesize >> ubi->mtd->subpage_sft; - /* Make sure minimal I/O unit is power of 2 */ + /* + * Make sure minimal I/O unit is power of 2. Note, there is no + * fundamental reason for this assumption. It is just an optimization + * which allows us to avoid costly division operations. + */ if (!is_power_of_2(ubi->min_io_size)) { ubi_err("min. I/O unit (%d) is not power of 2", ubi->min_io_size); @@ -581,7 +585,7 @@ static int io_init(struct ubi_device *ubi) if (ubi->vid_hdr_offset < UBI_EC_HDR_SIZE || ubi->leb_start < ubi->vid_hdr_offset + UBI_VID_HDR_SIZE || ubi->leb_start > ubi->peb_size - UBI_VID_HDR_SIZE || - ubi->leb_start % ubi->min_io_size) { + ubi->leb_start & (ubi->min_io_size - 1)) { ubi_err("bad VID header (%d) or data offsets (%d)", ubi->vid_hdr_offset, ubi->leb_start); return -EINVAL; diff --git a/drivers/mtd/ubi/cdev.c b/drivers/mtd/ubi/cdev.c index 89193ba9451e..0cdaf9fba7b0 100644 --- a/drivers/mtd/ubi/cdev.c +++ b/drivers/mtd/ubi/cdev.c @@ -295,7 +295,7 @@ static ssize_t vol_cdev_direct_write(struct file *file, const char __user *buf, off = do_div(tmp, vol->usable_leb_size); lnum = tmp; - if (off % ubi->min_io_size) { + if (off & (ubi->min_io_size - 1)) { dbg_err("unaligned position"); return -EINVAL; } @@ -304,7 +304,7 @@ static ssize_t vol_cdev_direct_write(struct file *file, const char __user *buf, count_save = count = vol->used_bytes - *offp; /* We can write only in fractions of the minimum I/O unit */ - if (count % ubi->min_io_size) { + if (count & (ubi->min_io_size - 1)) { dbg_err("unaligned write length"); return -EINVAL; } @@ -564,7 +564,7 @@ static int verify_mkvol_req(const struct ubi_device *ubi, if (req->alignment > ubi->leb_size) goto bad; - n = req->alignment % ubi->min_io_size; + n = req->alignment & (ubi->min_io_size - 1); if (req->alignment != 1 && n) goto bad; diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c index 7ce91ca742b1..37d778447943 100644 --- a/drivers/mtd/ubi/eba.c +++ b/drivers/mtd/ubi/eba.c @@ -752,7 +752,7 @@ int ubi_eba_write_leb_st(struct ubi_device *ubi, struct ubi_volume *vol, /* If this is the last LEB @len may be unaligned */ len = ALIGN(data_size, ubi->min_io_size); else - ubi_assert(len % ubi->min_io_size == 0); + ubi_assert(!(len & (ubi->min_io_size - 1))); vid_hdr = ubi_zalloc_vid_hdr(ubi, GFP_NOFS); if (!vid_hdr) diff --git a/drivers/mtd/ubi/kapi.c b/drivers/mtd/ubi/kapi.c index a70d58823f8d..51508832566d 100644 --- a/drivers/mtd/ubi/kapi.c +++ b/drivers/mtd/ubi/kapi.c @@ -397,8 +397,8 @@ int ubi_leb_write(struct ubi_volume_desc *desc, int lnum, const void *buf, return -EROFS; if (lnum < 0 || lnum >= vol->reserved_pebs || offset < 0 || len < 0 || - offset + len > vol->usable_leb_size || offset % ubi->min_io_size || - len % ubi->min_io_size) + offset + len > vol->usable_leb_size || + offset & (ubi->min_io_size - 1) || len & (ubi->min_io_size - 1)) return -EINVAL; if (dtype != UBI_LONGTERM && dtype != UBI_SHORTTERM && @@ -447,7 +447,7 @@ int ubi_leb_change(struct ubi_volume_desc *desc, int lnum, const void *buf, return -EROFS; if (lnum < 0 || lnum >= vol->reserved_pebs || len < 0 || - len > vol->usable_leb_size || len % ubi->min_io_size) + len > vol->usable_leb_size || len & (ubi->min_io_size - 1)) return -EINVAL; if (dtype != UBI_LONGTERM && dtype != UBI_SHORTTERM && diff --git a/drivers/mtd/ubi/misc.c b/drivers/mtd/ubi/misc.c index 93e052812012..22ad31402945 100644 --- a/drivers/mtd/ubi/misc.c +++ b/drivers/mtd/ubi/misc.c @@ -37,7 +37,7 @@ int ubi_calc_data_len(const struct ubi_device *ubi, const void *buf, { int i; - ubi_assert(length % ubi->min_io_size == 0); + ubi_assert(!(length & (ubi->min_io_size - 1))); for (i = length - 1; i >= 0; i--) if (((const uint8_t *)buf)[i] != 0xFF) diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c index 5be58d85c639..7402025ded94 100644 --- a/drivers/mtd/ubi/vmt.c +++ b/drivers/mtd/ubi/vmt.c @@ -727,7 +727,7 @@ static void paranoid_check_volume(struct ubi_device *ubi, int vol_id) goto fail; } - n = vol->alignment % ubi->min_io_size; + n = vol->alignment & (ubi->min_io_size - 1); if (vol->alignment != 1 && n) { ubi_err("alignment is not multiple of min I/O unit"); goto fail; diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c index 42a7815086b7..d9af11a8682b 100644 --- a/drivers/mtd/ubi/vtbl.c +++ b/drivers/mtd/ubi/vtbl.c @@ -170,7 +170,7 @@ static int vtbl_check(const struct ubi_device *ubi, goto bad; } - n = alignment % ubi->min_io_size; + n = alignment & (ubi->min_io_size - 1); if (alignment != 1 && n) { err = 5; goto bad; @@ -684,14 +684,13 @@ static int check_scanning_info(const struct ubi_device *ubi, return -EINVAL; } - if (si->highest_vol_id >= ubi->vtbl_slots + UBI_INT_VOL_COUNT&& + if (si->highest_vol_id >= ubi->vtbl_slots + UBI_INT_VOL_COUNT && si->highest_vol_id < UBI_INTERNAL_VOL_START) { ubi_err("too large volume ID %d found by scanning", si->highest_vol_id); return -EINVAL; } - for (i = 0; i < ubi->vtbl_slots + UBI_INT_VOL_COUNT; i++) { cond_resched(); diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c index a471a491f0ab..cc8fe2934d2b 100644 --- a/drivers/mtd/ubi/wl.c +++ b/drivers/mtd/ubi/wl.c @@ -1368,7 +1368,7 @@ int ubi_thread(void *u) int err; if (kthread_should_stop()) - goto out; + break; if (try_to_freeze()) continue; @@ -1403,7 +1403,6 @@ int ubi_thread(void *u) cond_resched(); } -out: dbg_wl("background thread \"%s\" is killed", ubi->bgt_name); return 0; } From abc5e92262d87f9c5c628492bffc55f81c7dcb80 Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Wed, 4 Jun 2008 16:48:12 +0300 Subject: [PATCH 095/857] UBI: fix memory leak ubi_free_volume() function sets ubi->volumes[] to NULL, so ubi_eba_close() is useless, it does not free what has to be freed. So zap it and free vol->eba_tbl at the volume release function. Pointed-out-by: Adrian Hunter Signed-off-by: Artem Bityutskiy --- drivers/mtd/ubi/build.c | 2 -- drivers/mtd/ubi/eba.c | 17 ----------------- drivers/mtd/ubi/ubi.h | 1 - drivers/mtd/ubi/vmt.c | 18 +++++++++--------- 4 files changed, 9 insertions(+), 29 deletions(-) diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c index ff4425de1527..7b42b4d05b3a 100644 --- a/drivers/mtd/ubi/build.c +++ b/drivers/mtd/ubi/build.c @@ -840,7 +840,6 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, int vid_hdr_offset) out_uif: uif_close(ubi); out_detach: - ubi_eba_close(ubi); ubi_wl_close(ubi); vfree(ubi->vtbl); out_free: @@ -903,7 +902,6 @@ int ubi_detach_mtd_dev(int ubi_num, int anyway) kthread_stop(ubi->bgt_thread); uif_close(ubi); - ubi_eba_close(ubi); ubi_wl_close(ubi); vfree(ubi->vtbl); put_mtd_device(ubi->mtd); diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c index 37d778447943..623d25f4855f 100644 --- a/drivers/mtd/ubi/eba.c +++ b/drivers/mtd/ubi/eba.c @@ -1233,20 +1233,3 @@ int ubi_eba_init_scan(struct ubi_device *ubi, struct ubi_scan_info *si) } return err; } - -/** - * ubi_eba_close - close EBA unit. - * @ubi: UBI device description object - */ -void ubi_eba_close(const struct ubi_device *ubi) -{ - int i, num_volumes = ubi->vtbl_slots + UBI_INT_VOL_COUNT; - - dbg_eba("close EBA unit"); - - for (i = 0; i < num_volumes; i++) { - if (!ubi->volumes[i]) - continue; - kfree(ubi->volumes[i]->eba_tbl); - } -} diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h index 67dcbd11c15c..940f6b7deec3 100644 --- a/drivers/mtd/ubi/ubi.h +++ b/drivers/mtd/ubi/ubi.h @@ -477,7 +477,6 @@ int ubi_eba_atomic_leb_change(struct ubi_device *ubi, struct ubi_volume *vol, int ubi_eba_copy_leb(struct ubi_device *ubi, int from, int to, struct ubi_vid_hdr *vid_hdr); int ubi_eba_init_scan(struct ubi_device *ubi, struct ubi_scan_info *si); -void ubi_eba_close(const struct ubi_device *ubi); /* wl.c */ int ubi_wl_get_peb(struct ubi_device *ubi, int dtype); diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c index 7402025ded94..367b04176e0a 100644 --- a/drivers/mtd/ubi/vmt.c +++ b/drivers/mtd/ubi/vmt.c @@ -127,6 +127,7 @@ static void vol_release(struct device *dev) { struct ubi_volume *vol = container_of(dev, struct ubi_volume, dev); + kfree(vol->eba_tbl); kfree(vol); } @@ -201,7 +202,7 @@ static void volume_sysfs_close(struct ubi_volume *vol) */ int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req) { - int i, err, vol_id = req->vol_id, dont_free = 0; + int i, err, vol_id = req->vol_id, do_free = 1; struct ubi_volume *vol; struct ubi_vtbl_record vtbl_rec; uint64_t bytes; @@ -365,14 +366,14 @@ int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req) out_sysfs: /* - * We have registered our device, we should not free the volume* + * We have registered our device, we should not free the volume * description object in this function in case of an error - it is * freed by the release function. * * Get device reference to prevent the release function from being * called just after sysfs has been closed. */ - dont_free = 1; + do_free = 0; get_device(&vol->dev); volume_sysfs_close(vol); out_gluebi: @@ -382,17 +383,18 @@ int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req) out_cdev: cdev_del(&vol->cdev); out_mapping: - kfree(vol->eba_tbl); + if (do_free) + kfree(vol->eba_tbl); out_acc: spin_lock(&ubi->volumes_lock); ubi->rsvd_pebs -= vol->reserved_pebs; ubi->avail_pebs += vol->reserved_pebs; out_unlock: spin_unlock(&ubi->volumes_lock); - if (dont_free) - put_device(&vol->dev); - else + if (do_free) kfree(vol); + else + put_device(&vol->dev); ubi_err("cannot create volume %d, error %d", vol_id, err); return err; } @@ -445,8 +447,6 @@ int ubi_remove_volume(struct ubi_volume_desc *desc) goto out_err; } - kfree(vol->eba_tbl); - vol->eba_tbl = NULL; cdev_del(&vol->cdev); volume_sysfs_close(vol); From 505d1caa79cd61a70615e9a7eae2eab85e797a83 Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Wed, 4 Jun 2008 17:00:35 +0300 Subject: [PATCH 096/857] UBI: do not forget to free internal volumes UBI forgets to free internal volumes when detaching MTD device. Fix this. Pointed-out-by: Adrian Hunter Signed-off-by: Artem Bityutskiy --- drivers/mtd/ubi/build.c | 28 +++++++++++++++++++++++++--- 1 file changed, 25 insertions(+), 3 deletions(-) diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c index 7b42b4d05b3a..33205e4c1f5b 100644 --- a/drivers/mtd/ubi/build.c +++ b/drivers/mtd/ubi/build.c @@ -422,6 +422,10 @@ static int uif_init(struct ubi_device *ubi) /** * uif_close - close user interfaces for an UBI device. * @ubi: UBI device description object + * + * Note, since this function un-registers UBI volume device objects (@vol->dev), + * the memory allocated voe the volumes is freed as well (in the release + * function). */ static void uif_close(struct ubi_device *ubi) { @@ -431,6 +435,21 @@ static void uif_close(struct ubi_device *ubi) unregister_chrdev_region(ubi->cdev.dev, ubi->vtbl_slots + 1); } +/** + * free_internal_volumes - free internal volumes. + * @ubi: UBI device description object + */ +static void free_internal_volumes(struct ubi_device *ubi) +{ + int i; + + for (i = ubi->vtbl_slots; + i < ubi->vtbl_slots + UBI_INT_VOL_COUNT; i++) { + kfree(ubi->volumes[i]->eba_tbl); + kfree(ubi->volumes[i]); + } +} + /** * attach_by_scanning - attach an MTD device using scanning method. * @ubi: UBI device descriptor @@ -475,6 +494,7 @@ static int attach_by_scanning(struct ubi_device *ubi) out_wl: ubi_wl_close(ubi); out_vtbl: + free_internal_volumes(ubi); vfree(ubi->vtbl); out_si: ubi_scan_destroy_si(si); @@ -650,7 +670,7 @@ static int autoresize(struct ubi_device *ubi, int vol_id) /* * Clear the auto-resize flag in the volume in-memory copy of the - * volume table, and 'ubi_resize_volume()' will propogate this change + * volume table, and 'ubi_resize_volume()' will propagate this change * to the flash. */ ubi->vtbl[vol_id].flags &= ~UBI_VTBL_AUTORESIZE_FLG; @@ -659,7 +679,7 @@ static int autoresize(struct ubi_device *ubi, int vol_id) struct ubi_vtbl_record vtbl_rec; /* - * No avalilable PEBs to re-size the volume, clear the flag on + * No available PEBs to re-size the volume, clear the flag on * flash and exit. */ memcpy(&vtbl_rec, &ubi->vtbl[vol_id], @@ -692,7 +712,7 @@ static int autoresize(struct ubi_device *ubi, int vol_id) * * This function attaches MTD device @mtd_dev to UBI and assign @ubi_num number * to the newly created UBI device, unless @ubi_num is %UBI_DEV_NUM_AUTO, in - * which case this function finds a vacant device nubert and assings it + * which case this function finds a vacant device number and assigns it * automatically. Returns the new UBI device number in case of success and a * negative error code in case of failure. * @@ -841,6 +861,7 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, int vid_hdr_offset) uif_close(ubi); out_detach: ubi_wl_close(ubi); + free_internal_volumes(ubi); vfree(ubi->vtbl); out_free: vfree(ubi->peb_buf1); @@ -903,6 +924,7 @@ int ubi_detach_mtd_dev(int ubi_num, int anyway) uif_close(ubi); ubi_wl_close(ubi); + free_internal_volumes(ubi); vfree(ubi->vtbl); put_mtd_device(ubi->mtd); vfree(ubi->peb_buf1); From 472018f73e7308a7f29b753ee8c742b6f45f103f Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Wed, 4 Jun 2008 17:58:37 +0300 Subject: [PATCH 097/857] UBI: fix memory leak on error path Normally UBI volumes are freed in the release function of the struct device object. However, on error path they may have to be freed before the struct device objects have been initialized. Signed-off-by: Artem Bityutskiy --- drivers/mtd/ubi/build.c | 34 ++++++++++++++++++++++++++++++---- 1 file changed, 30 insertions(+), 4 deletions(-) diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c index 33205e4c1f5b..a5b19944eca8 100644 --- a/drivers/mtd/ubi/build.c +++ b/drivers/mtd/ubi/build.c @@ -354,16 +354,35 @@ static void kill_volumes(struct ubi_device *ubi) ubi_free_volume(ubi, ubi->volumes[i]); } +/** + * free_user_volumes - free all user volumes. + * @ubi: UBI device description object + * + * Normally the volumes are freed at the release function of the volume device + * objects. However, on error paths the volumes have to be freed before the + * device objects have been initialized. + */ +static void free_user_volumes(struct ubi_device *ubi) +{ + int i; + + for (i = 0; i < ubi->vtbl_slots; i++) + if (ubi->volumes[i]) { + kfree(ubi->volumes[i]->eba_tbl); + kfree(ubi->volumes[i]); + } +} + /** * uif_init - initialize user interfaces for an UBI device. * @ubi: UBI device description object * * This function returns zero in case of success and a negative error code in - * case of failure. + * case of failure. Note, this function destroys all volumes if it failes. */ static int uif_init(struct ubi_device *ubi) { - int i, err; + int i, err, do_free = 0; dev_t dev; sprintf(ubi->ubi_name, UBI_NAME_STR "%d", ubi->ubi_num); @@ -410,10 +429,13 @@ static int uif_init(struct ubi_device *ubi) out_volumes: kill_volumes(ubi); + do_free = 0; out_sysfs: ubi_sysfs_close(ubi); cdev_del(&ubi->cdev); out_unreg: + if (do_free) + free_user_volumes(ubi); unregister_chrdev_region(ubi->cdev.dev, ubi->vtbl_slots + 1); ubi_err("cannot initialize UBI %s, error %d", ubi->ubi_name, err); return err; @@ -722,7 +744,7 @@ static int autoresize(struct ubi_device *ubi, int vol_id) int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, int vid_hdr_offset) { struct ubi_device *ubi; - int i, err; + int i, err, do_free = 1; /* * Check if we already have the same MTD device attached. @@ -822,7 +844,7 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, int vid_hdr_offset) err = uif_init(ubi); if (err) - goto out_detach; + goto out_nofree; ubi->bgt_thread = kthread_create(ubi_thread, ubi, ubi->bgt_name); if (IS_ERR(ubi->bgt_thread)) { @@ -859,8 +881,12 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, int vid_hdr_offset) out_uif: uif_close(ubi); +out_nofree: + do_free = 0; out_detach: ubi_wl_close(ubi); + if (do_free) + free_user_volumes(ubi); free_internal_volumes(ubi); vfree(ubi->vtbl); out_free: From 23add7455c42eef63f8719bd268328047d4aed69 Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Mon, 16 Jun 2008 13:35:23 +0300 Subject: [PATCH 098/857] UBI: fix LEB locking leb_read_unlock() may be called simultaniously by several tasks. The would race at the following code: up_read(&le->mutex); if (free) kfree(le); And it is possible that one task frees 'le' before the other tasks do 'up_read()'. Fix this by doing up_read and free inside the 'ubi->ltree' lock. Below it the oops we had because of this: BUG: spinlock bad magic on CPU#0, integck/7504 BUG: unable to handle kernel paging request at 6b6b6c4f IP: [] spin_bug+0x5c/0xdb *pde = 00000000 Oops: 0000 [#1] PREEMPT SMP Modules linked in: ubifs ubi nandsim nand nand_ids nand_ecc video output Pid: 7504, comm: integck Not tainted (2.6.26-rc3ubifs26 #8) EIP: 0060:[] EFLAGS: 00010002 CPU: 0 EIP is at spin_bug+0x5c/0xdb EAX: 00000032 EBX: 6b6b6b6b ECX: 6b6b6b6b EDX: f7f7ce30 ESI: f76491dc EDI: c044f51f EBP: e8a736cc ESP: e8a736a8 DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068 Process integck (pid: 7504, ti=e8a72000 task=f7f7ce30 task.ti=e8a72000) Stack: c044f754 c044f51f 00000000 f7f7d024 00001d50 00000001 f76491dc 00000296 f6df50e0 e8a736d8 c02112f0 f76491dc e8a736e8 c039157a f7d9e830 f76491d8 e8a7370c c020b975 f76491dc 00000296 f76491f8 00000000 f76491d8 00000000 Call Trace: [] ? _raw_spin_unlock+0x50/0x7c [] ? _spin_unlock_irqrestore+0x20/0x58 [] ? rwsem_wake+0x4b/0x122 [] ? call_rwsem_wake+0xa/0xc [] ? up_read+0x28/0x31 [] ? leb_read_unlock+0x73/0x7b [ubi] [] ? ubi_eba_read_leb+0x195/0x2b0 [ubi] [] ? ubi_leb_read+0xaf/0xf8 [ubi] Signed-off-by: Artem Bityutskiy --- drivers/mtd/ubi/eba.c | 27 +++++++-------------------- 1 file changed, 7 insertions(+), 20 deletions(-) diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c index 623d25f4855f..8dc488fc0cdf 100644 --- a/drivers/mtd/ubi/eba.c +++ b/drivers/mtd/ubi/eba.c @@ -223,22 +223,18 @@ static int leb_read_lock(struct ubi_device *ubi, int vol_id, int lnum) */ static void leb_read_unlock(struct ubi_device *ubi, int vol_id, int lnum) { - int free = 0; struct ubi_ltree_entry *le; spin_lock(&ubi->ltree_lock); le = ltree_lookup(ubi, vol_id, lnum); le->users -= 1; ubi_assert(le->users >= 0); + up_read(&le->mutex); if (le->users == 0) { rb_erase(&le->rb, &ubi->ltree); - free = 1; + kfree(le); } spin_unlock(&ubi->ltree_lock); - - up_read(&le->mutex); - if (free) - kfree(le); } /** @@ -274,7 +270,6 @@ static int leb_write_lock(struct ubi_device *ubi, int vol_id, int lnum) */ static int leb_write_trylock(struct ubi_device *ubi, int vol_id, int lnum) { - int free; struct ubi_ltree_entry *le; le = ltree_add_entry(ubi, vol_id, lnum); @@ -289,12 +284,9 @@ static int leb_write_trylock(struct ubi_device *ubi, int vol_id, int lnum) ubi_assert(le->users >= 0); if (le->users == 0) { rb_erase(&le->rb, &ubi->ltree); - free = 1; - } else - free = 0; - spin_unlock(&ubi->ltree_lock); - if (free) kfree(le); + } + spin_unlock(&ubi->ltree_lock); return 1; } @@ -307,23 +299,18 @@ static int leb_write_trylock(struct ubi_device *ubi, int vol_id, int lnum) */ static void leb_write_unlock(struct ubi_device *ubi, int vol_id, int lnum) { - int free; struct ubi_ltree_entry *le; spin_lock(&ubi->ltree_lock); le = ltree_lookup(ubi, vol_id, lnum); le->users -= 1; ubi_assert(le->users >= 0); + up_write(&le->mutex); if (le->users == 0) { rb_erase(&le->rb, &ubi->ltree); - free = 1; - } else - free = 0; - spin_unlock(&ubi->ltree_lock); - - up_write(&le->mutex); - if (free) kfree(le); + } + spin_unlock(&ubi->ltree_lock); } /** From 73789a3d9fd8e500e121c1d4a5a2b16dd748ab5f Mon Sep 17 00:00:00 2001 From: Bruce Leonard Date: Thu, 3 Jul 2008 10:35:49 +0300 Subject: [PATCH 099/857] UBI: fix 64-bit calculations Signed-off-by: Bruce Leonard Signed-off-by: Artem Bityutskiy --- drivers/mtd/ubi/cdev.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/mtd/ubi/cdev.c b/drivers/mtd/ubi/cdev.c index 0cdaf9fba7b0..3e3449ec07f0 100644 --- a/drivers/mtd/ubi/cdev.c +++ b/drivers/mtd/ubi/cdev.c @@ -437,7 +437,8 @@ static int vol_cdev_ioctl(struct inode *inode, struct file *file, break; } - rsvd_bytes = vol->reserved_pebs * (ubi->leb_size-vol->data_pad); + rsvd_bytes = (long long)vol->reserved_pebs * + ubi->leb_size-vol->data_pad; if (bytes < 0 || bytes > rsvd_bytes) { err = -EINVAL; break; From a5bf6190417cbbf80443a9f71c65b653e13e9982 Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Thu, 10 Jul 2008 18:38:33 +0300 Subject: [PATCH 100/857] UBI: add ubi_sync() interface To flush MTD device caches. Signed-off-by: Artem Bityutskiy --- drivers/mtd/ubi/kapi.c | 24 ++++++++++++++++++++++++ include/linux/mtd/ubi.h | 1 + 2 files changed, 25 insertions(+) diff --git a/drivers/mtd/ubi/kapi.c b/drivers/mtd/ubi/kapi.c index 51508832566d..e65c8e0bcd5d 100644 --- a/drivers/mtd/ubi/kapi.c +++ b/drivers/mtd/ubi/kapi.c @@ -632,3 +632,27 @@ int ubi_is_mapped(struct ubi_volume_desc *desc, int lnum) return vol->eba_tbl[lnum] >= 0; } EXPORT_SYMBOL_GPL(ubi_is_mapped); + +/** + * ubi_sync - synchronize UBI device buffers. + * @ubi_num: UBI device to synchronize + * + * The underlying MTD device may cache data in hardware or in software. This + * function ensures the caches are flushed. Returns zero in case of success and + * a negative error code in case of failure. + */ +int ubi_sync(int ubi_num) +{ + struct ubi_device *ubi; + + ubi = ubi_get_device(ubi_num); + if (!ubi) + return -ENODEV; + + if (ubi->mtd->sync) + ubi->mtd->sync(ubi->mtd); + + ubi_put_device(ubi); + return 0; +} +EXPORT_SYMBOL_GPL(ubi_sync); diff --git a/include/linux/mtd/ubi.h b/include/linux/mtd/ubi.h index f71201d0f3e7..83302bbbddb4 100644 --- a/include/linux/mtd/ubi.h +++ b/include/linux/mtd/ubi.h @@ -152,6 +152,7 @@ int ubi_leb_erase(struct ubi_volume_desc *desc, int lnum); int ubi_leb_unmap(struct ubi_volume_desc *desc, int lnum); int ubi_leb_map(struct ubi_volume_desc *desc, int lnum, int dtype); int ubi_is_mapped(struct ubi_volume_desc *desc, int lnum); +int ubi_sync(int ubi_num); /* * This function is the same as the 'ubi_leb_read()' function, but it does not From a6ea440769e11c46828cddd20f91ab57261701d5 Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Sun, 13 Jul 2008 21:46:24 +0300 Subject: [PATCH 101/857] UBI: improve mkvol request validation Check that volume name is not shorter than 'name_len'. No need to copy the trailing zero byte because whole array was zeroed earlier. Signed-off-by: Artem Bityutskiy --- drivers/mtd/ubi/cdev.c | 7 +++++-- drivers/mtd/ubi/vmt.c | 4 ++-- 2 files changed, 7 insertions(+), 4 deletions(-) diff --git a/drivers/mtd/ubi/cdev.c b/drivers/mtd/ubi/cdev.c index 3e3449ec07f0..4fb84e3e6502 100644 --- a/drivers/mtd/ubi/cdev.c +++ b/drivers/mtd/ubi/cdev.c @@ -574,6 +574,10 @@ static int verify_mkvol_req(const struct ubi_device *ubi, goto bad; } + n = strnlen(req->name, req->name_len + 1); + if (n != req->name_len) + goto bad; + return 0; bad: @@ -629,12 +633,11 @@ static int ubi_cdev_ioctl(struct inode *inode, struct file *file, break; } + req.name[req.name_len] = '\0'; err = verify_mkvol_req(ubi, &req); if (err) break; - req.name[req.name_len] = '\0'; - mutex_lock(&ubi->volumes_mutex); err = ubi_create_volume(ubi, &req); mutex_unlock(&ubi->volumes_mutex); diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c index 367b04176e0a..bfa7c5d2e064 100644 --- a/drivers/mtd/ubi/vmt.c +++ b/drivers/mtd/ubi/vmt.c @@ -275,7 +275,7 @@ int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req) vol->data_pad = ubi->leb_size % vol->alignment; vol->vol_type = req->vol_type; vol->name_len = req->name_len; - memcpy(vol->name, req->name, vol->name_len + 1); + memcpy(vol->name, req->name, vol->name_len); vol->ubi = ubi; /* @@ -350,7 +350,7 @@ int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req) vtbl_rec.vol_type = UBI_VID_DYNAMIC; else vtbl_rec.vol_type = UBI_VID_STATIC; - memcpy(vtbl_rec.name, vol->name, vol->name_len + 1); + memcpy(vtbl_rec.name, vol->name, vol->name_len); err = ubi_change_vtbl_record(ubi, vol_id, &vtbl_rec); if (err) From bb84c1a199558962edf4b4aeb4480fb09aa09b91 Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Mon, 14 Jul 2008 12:57:27 +0300 Subject: [PATCH 102/857] UBI: fix error message The ubi_err() macro will add \n. Signed-off-by: Artem Bityutskiy --- drivers/mtd/ubi/gluebi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/mtd/ubi/gluebi.c b/drivers/mtd/ubi/gluebi.c index e909b390069a..ae76ab638b2f 100644 --- a/drivers/mtd/ubi/gluebi.c +++ b/drivers/mtd/ubi/gluebi.c @@ -299,7 +299,7 @@ int ubi_create_gluebi(struct ubi_device *ubi, struct ubi_volume *vol) mtd->size = vol->used_bytes; if (add_mtd_device(mtd)) { - ubi_err("cannot not add MTD device\n"); + ubi_err("cannot not add MTD device"); kfree(mtd->name); return -ENFILE; } From 85c6e6e28259e9b58b8984db536c45bc3161f40c Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Wed, 16 Jul 2008 10:25:56 +0300 Subject: [PATCH 103/857] UBI: amend commentaries Hch asked not to use "unit" for sub-systems, let it be so. Also some other commentaries modifications. Signed-off-by: Artem Bityutskiy --- drivers/mtd/ubi/build.c | 2 +- drivers/mtd/ubi/debug.h | 6 +-- drivers/mtd/ubi/eba.c | 22 ++++----- drivers/mtd/ubi/io.c | 22 ++++----- drivers/mtd/ubi/scan.c | 28 ++++++----- drivers/mtd/ubi/scan.h | 19 ++++---- drivers/mtd/ubi/ubi-media.h | 23 ++++----- drivers/mtd/ubi/ubi.h | 37 +++++++-------- drivers/mtd/ubi/wl.c | 94 ++++++++++++++++++------------------- include/linux/mtd/ubi.h | 4 +- 10 files changed, 129 insertions(+), 128 deletions(-) diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c index a5b19944eca8..27271fe32e02 100644 --- a/drivers/mtd/ubi/build.c +++ b/drivers/mtd/ubi/build.c @@ -524,7 +524,7 @@ static int attach_by_scanning(struct ubi_device *ubi) } /** - * io_init - initialize I/O unit for a given UBI device. + * io_init - initialize I/O sub-system for a given UBI device. * @ubi: UBI device description object * * If @ubi->vid_hdr_offset or @ubi->leb_start is zero, default offsets are diff --git a/drivers/mtd/ubi/debug.h b/drivers/mtd/ubi/debug.h index 8ea99d8c9e1f..7d8d77c31dfe 100644 --- a/drivers/mtd/ubi/debug.h +++ b/drivers/mtd/ubi/debug.h @@ -76,21 +76,21 @@ void ubi_dbg_dump_mkvol_req(const struct ubi_mkvol_req *req); #endif /* CONFIG_MTD_UBI_DEBUG_MSG */ #ifdef CONFIG_MTD_UBI_DEBUG_MSG_EBA -/* Messages from the eraseblock association unit */ +/* Messages from the eraseblock association sub-system */ #define dbg_eba(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__) #else #define dbg_eba(fmt, ...) ({}) #endif #ifdef CONFIG_MTD_UBI_DEBUG_MSG_WL -/* Messages from the wear-leveling unit */ +/* Messages from the wear-leveling sub-system */ #define dbg_wl(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__) #else #define dbg_wl(fmt, ...) ({}) #endif #ifdef CONFIG_MTD_UBI_DEBUG_MSG_IO -/* Messages from the input/output unit */ +/* Messages from the input/output sub-system */ #define dbg_io(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__) #else #define dbg_io(fmt, ...) ({}) diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c index 8dc488fc0cdf..613cd1e51648 100644 --- a/drivers/mtd/ubi/eba.c +++ b/drivers/mtd/ubi/eba.c @@ -19,20 +19,20 @@ */ /* - * The UBI Eraseblock Association (EBA) unit. + * The UBI Eraseblock Association (EBA) sub-system. * - * This unit is responsible for I/O to/from logical eraseblock. + * This sub-system is responsible for I/O to/from logical eraseblock. * * Although in this implementation the EBA table is fully kept and managed in * RAM, which assumes poor scalability, it might be (partially) maintained on * flash in future implementations. * - * The EBA unit implements per-logical eraseblock locking. Before accessing a - * logical eraseblock it is locked for reading or writing. The per-logical - * eraseblock locking is implemented by means of the lock tree. The lock tree - * is an RB-tree which refers all the currently locked logical eraseblocks. The - * lock tree elements are &struct ubi_ltree_entry objects. They are indexed by - * (@vol_id, @lnum) pairs. + * The EBA sub-system implements per-logical eraseblock locking. Before + * accessing a logical eraseblock it is locked for reading or writing. The + * per-logical eraseblock locking is implemented by means of the lock tree. The + * lock tree is an RB-tree which refers all the currently locked logical + * eraseblocks. The lock tree elements are &struct ubi_ltree_entry objects. + * They are indexed by (@vol_id, @lnum) pairs. * * EBA also maintains the global sequence counter which is incremented each * time a logical eraseblock is mapped to a physical eraseblock and it is @@ -1128,7 +1128,7 @@ int ubi_eba_copy_leb(struct ubi_device *ubi, int from, int to, } /** - * ubi_eba_init_scan - initialize the EBA unit using scanning information. + * ubi_eba_init_scan - initialize the EBA sub-system using scanning information. * @ubi: UBI device description object * @si: scanning information * @@ -1143,7 +1143,7 @@ int ubi_eba_init_scan(struct ubi_device *ubi, struct ubi_scan_info *si) struct ubi_scan_leb *seb; struct rb_node *rb; - dbg_eba("initialize EBA unit"); + dbg_eba("initialize EBA sub-system"); spin_lock_init(&ubi->ltree_lock); mutex_init(&ubi->alc_mutex); @@ -1209,7 +1209,7 @@ int ubi_eba_init_scan(struct ubi_device *ubi, struct ubi_scan_info *si) ubi->rsvd_pebs += ubi->beb_rsvd_pebs; } - dbg_eba("EBA unit is initialized"); + dbg_eba("EBA sub-system is initialized"); return 0; out_free: diff --git a/drivers/mtd/ubi/io.c b/drivers/mtd/ubi/io.c index 4ac11df7b048..561e7b2f96cb 100644 --- a/drivers/mtd/ubi/io.c +++ b/drivers/mtd/ubi/io.c @@ -20,15 +20,15 @@ */ /* - * UBI input/output unit. + * UBI input/output sub-system. * - * This unit provides a uniform way to work with all kinds of the underlying - * MTD devices. It also implements handy functions for reading and writing UBI - * headers. + * This sub-system provides a uniform way to work with all kinds of the + * underlying MTD devices. It also implements handy functions for reading and + * writing UBI headers. * * We are trying to have a paranoid mindset and not to trust to what we read - * from the flash media in order to be more secure and robust. So this unit - * validates every single header it reads from the flash media. + * from the flash media in order to be more secure and robust. So this + * sub-system validates every single header it reads from the flash media. * * Some words about how the eraseblock headers are stored. * @@ -79,11 +79,11 @@ * 512-byte chunks, we have to allocate one more buffer and copy our VID header * to offset 448 of this buffer. * - * The I/O unit does the following trick in order to avoid this extra copy. - * It always allocates a @ubi->vid_hdr_alsize bytes buffer for the VID header - * and returns a pointer to offset @ubi->vid_hdr_shift of this buffer. When the - * VID header is being written out, it shifts the VID header pointer back and - * writes the whole sub-page. + * The I/O sub-system does the following trick in order to avoid this extra + * copy. It always allocates a @ubi->vid_hdr_alsize bytes buffer for the VID + * header and returns a pointer to offset @ubi->vid_hdr_shift of this buffer. + * When the VID header is being written out, it shifts the VID header pointer + * back and writes the whole sub-page. */ #include diff --git a/drivers/mtd/ubi/scan.c b/drivers/mtd/ubi/scan.c index 96d410e106ab..892c2ba49777 100644 --- a/drivers/mtd/ubi/scan.c +++ b/drivers/mtd/ubi/scan.c @@ -19,9 +19,9 @@ */ /* - * UBI scanning unit. + * UBI scanning sub-system. * - * This unit is responsible for scanning the flash media, checking UBI + * This sub-system is responsible for scanning the flash media, checking UBI * headers and providing complete information about the UBI flash image. * * The scanning information is represented by a &struct ubi_scan_info' object. @@ -103,7 +103,7 @@ static int add_to_list(struct ubi_scan_info *si, int pnum, int ec, * non-zero if an inconsistency was found and zero if not. * * Note, UBI does sanity check of everything it reads from the flash media. - * Most of the checks are done in the I/O unit. Here we check that the + * Most of the checks are done in the I/O sub-system. Here we check that the * information in the VID header is consistent to the information in other VID * headers of the same volume. */ @@ -256,8 +256,8 @@ static int compare_lebs(struct ubi_device *ubi, const struct ubi_scan_leb *seb, * that versions that are close to %0xFFFFFFFF are less then * versions that are close to %0. * - * The UBI WL unit guarantees that the number of pending tasks - * is not greater then %0x7FFFFFFF. So, if the difference + * The UBI WL sub-system guarantees that the number of pending + * tasks is not greater then %0x7FFFFFFF. So, if the difference * between any two versions is greater or equivalent to * %0x7FFFFFFF, there was an overflow and the logical * eraseblock with lower version is actually newer then the one @@ -645,9 +645,9 @@ void ubi_scan_rm_volume(struct ubi_scan_info *si, struct ubi_scan_volume *sv) * * This function erases physical eraseblock 'pnum', and writes the erase * counter header to it. This function should only be used on UBI device - * initialization stages, when the EBA unit had not been yet initialized. This - * function returns zero in case of success and a negative error code in case - * of failure. + * initialization stages, when the EBA sub-system had not been yet initialized. + * This function returns zero in case of success and a negative error code in + * case of failure. */ int ubi_scan_erase_peb(struct ubi_device *ubi, const struct ubi_scan_info *si, int pnum, int ec) @@ -687,9 +687,10 @@ int ubi_scan_erase_peb(struct ubi_device *ubi, const struct ubi_scan_info *si, * @si: scanning information * * This function returns a free physical eraseblock. It is supposed to be - * called on the UBI initialization stages when the wear-leveling unit is not - * initialized yet. This function picks a physical eraseblocks from one of the - * lists, writes the EC header if it is needed, and removes it from the list. + * called on the UBI initialization stages when the wear-leveling sub-system is + * not initialized yet. This function picks a physical eraseblocks from one of + * the lists, writes the EC header if it is needed, and removes it from the + * list. * * This function returns scanning physical eraseblock information in case of * success and an error code in case of failure. @@ -764,8 +765,9 @@ static int process_eb(struct ubi_device *ubi, struct ubi_scan_info *si, int pnum return err; else if (err) { /* - * FIXME: this is actually duty of the I/O unit to initialize - * this, but MTD does not provide enough information. + * FIXME: this is actually duty of the I/O sub-system to + * initialize this, but MTD does not provide enough + * information. */ si->bad_peb_count += 1; return 0; diff --git a/drivers/mtd/ubi/scan.h b/drivers/mtd/ubi/scan.h index 966b9b682a42..4e2e3cc0becd 100644 --- a/drivers/mtd/ubi/scan.h +++ b/drivers/mtd/ubi/scan.h @@ -59,16 +59,16 @@ struct ubi_scan_leb { * @leb_count: number of logical eraseblocks in this volume * @vol_type: volume type * @used_ebs: number of used logical eraseblocks in this volume (only for - * static volumes) + * static volumes) * @last_data_size: amount of data in the last logical eraseblock of this - * volume (always equivalent to the usable logical eraseblock size in case of - * dynamic volumes) + * volume (always equivalent to the usable logical eraseblock + * size in case of dynamic volumes) * @data_pad: how many bytes at the end of logical eraseblocks of this volume - * are not used (due to volume alignment) + * are not used (due to volume alignment) * @compat: compatibility flags of this volume * @rb: link in the volume RB-tree * @root: root of the RB-tree containing all the eraseblock belonging to this - * volume (&struct ubi_scan_leb objects) + * volume (&struct ubi_scan_leb objects) * * One object of this type is allocated for each volume during scanning. */ @@ -92,8 +92,8 @@ struct ubi_scan_volume { * @free: list of free physical eraseblocks * @erase: list of physical eraseblocks which have to be erased * @alien: list of physical eraseblocks which should not be used by UBI (e.g., + * those belonging to "preserve"-compatible internal volumes) * @bad_peb_count: count of bad physical eraseblocks - * those belonging to "preserve"-compatible internal volumes) * @vols_found: number of volumes found during scanning * @highest_vol_id: highest volume ID * @alien_peb_count: count of physical eraseblocks in the @alien list @@ -106,8 +106,8 @@ struct ubi_scan_volume { * @ec_count: a temporary variable used when calculating @mean_ec * * This data structure contains the result of scanning and may be used by other - * UBI units to build final UBI data structures, further error-recovery and so - * on. + * UBI sub-systems to build final UBI data structures, further error-recovery + * and so on. */ struct ubi_scan_info { struct rb_root volumes; @@ -132,8 +132,7 @@ struct ubi_device; struct ubi_vid_hdr; /* - * ubi_scan_move_to_list - move a physical eraseblock from the volume tree to a - * list. + * ubi_scan_move_to_list - move a PEB from the volume tree to a list. * * @sv: volume scanning information * @seb: scanning eraseblock infprmation diff --git a/drivers/mtd/ubi/ubi-media.h b/drivers/mtd/ubi/ubi-media.h index c3185d9fd048..26bb7af9787a 100644 --- a/drivers/mtd/ubi/ubi-media.h +++ b/drivers/mtd/ubi/ubi-media.h @@ -98,10 +98,11 @@ enum { * Compatibility constants used by internal volumes. * * @UBI_COMPAT_DELETE: delete this internal volume before anything is written - * to the flash + * to the flash * @UBI_COMPAT_RO: attach this device in read-only mode * @UBI_COMPAT_PRESERVE: preserve this internal volume - do not touch its - * physical eraseblocks, don't allow the wear-leveling unit to move them + * physical eraseblocks, don't allow the wear-leveling + * sub-system to move them * @UBI_COMPAT_REJECT: reject this UBI image */ enum { @@ -123,7 +124,7 @@ enum { * struct ubi_ec_hdr - UBI erase counter header. * @magic: erase counter header magic number (%UBI_EC_HDR_MAGIC) * @version: version of UBI implementation which is supposed to accept this - * UBI image + * UBI image * @padding1: reserved for future, zeroes * @ec: the erase counter * @vid_hdr_offset: where the VID header starts @@ -159,20 +160,20 @@ struct ubi_ec_hdr { * struct ubi_vid_hdr - on-flash UBI volume identifier header. * @magic: volume identifier header magic number (%UBI_VID_HDR_MAGIC) * @version: UBI implementation version which is supposed to accept this UBI - * image (%UBI_VERSION) + * image (%UBI_VERSION) * @vol_type: volume type (%UBI_VID_DYNAMIC or %UBI_VID_STATIC) * @copy_flag: if this logical eraseblock was copied from another physical - * eraseblock (for wear-leveling reasons) + * eraseblock (for wear-leveling reasons) * @compat: compatibility of this volume (%0, %UBI_COMPAT_DELETE, - * %UBI_COMPAT_IGNORE, %UBI_COMPAT_PRESERVE, or %UBI_COMPAT_REJECT) + * %UBI_COMPAT_IGNORE, %UBI_COMPAT_PRESERVE, or %UBI_COMPAT_REJECT) * @vol_id: ID of this volume * @lnum: logical eraseblock number * @leb_ver: version of this logical eraseblock (IMPORTANT: obsolete, to be - * removed, kept only for not breaking older UBI users) + * removed, kept only for not breaking older UBI users) * @data_size: how many bytes of data this logical eraseblock contains * @used_ebs: total number of used logical eraseblocks in this volume * @data_pad: how many bytes at the end of this physical eraseblock are not - * used + * used * @data_crc: CRC checksum of the data stored in this logical eraseblock * @padding1: reserved for future, zeroes * @sqnum: sequence number @@ -248,9 +249,9 @@ struct ubi_ec_hdr { * The @data_crc field contains the CRC checksum of the contents of the logical * eraseblock if this is a static volume. In case of dynamic volumes, it does * not contain the CRC checksum as a rule. The only exception is when the - * data of the physical eraseblock was moved by the wear-leveling unit, then - * the wear-leveling unit calculates the data CRC and stores it in the - * @data_crc field. And of course, the @copy_flag is %in this case. + * data of the physical eraseblock was moved by the wear-leveling sub-system, + * then the wear-leveling sub-system calculates the data CRC and stores it in + * the @data_crc field. And of course, the @copy_flag is %in this case. * * The @data_size field is used only for static volumes because UBI has to know * how many bytes of data are stored in this eraseblock. For dynamic volumes, diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h index 940f6b7deec3..1fc32c863b78 100644 --- a/drivers/mtd/ubi/ubi.h +++ b/drivers/mtd/ubi/ubi.h @@ -74,15 +74,15 @@ #define UBI_IO_RETRIES 3 /* - * Error codes returned by the I/O unit. + * Error codes returned by the I/O sub-system. * * UBI_IO_PEB_EMPTY: the physical eraseblock is empty, i.e. it contains only - * 0xFF bytes + * %0xFF bytes * UBI_IO_PEB_FREE: the physical eraseblock is free, i.e. it contains only a - * valid erase counter header, and the rest are %0xFF bytes + * valid erase counter header, and the rest are %0xFF bytes * UBI_IO_BAD_EC_HDR: the erase counter header is corrupted (bad magic or CRC) * UBI_IO_BAD_VID_HDR: the volume identifier header is corrupted (bad magic or - * CRC) + * CRC) * UBI_IO_BITFLIPS: bit-flips were detected and corrected */ enum { @@ -99,9 +99,9 @@ enum { * @ec: erase counter * @pnum: physical eraseblock number * - * This data structure is used in the WL unit. Each physical eraseblock has a - * corresponding &struct wl_entry object which may be kept in different - * RB-trees. See WL unit for details. + * This data structure is used in the WL sub-system. Each physical eraseblock + * has a corresponding &struct wl_entry object which may be kept in different + * RB-trees. See WL sub-system for details. */ struct ubi_wl_entry { struct rb_node rb; @@ -118,10 +118,10 @@ struct ubi_wl_entry { * @mutex: read/write mutex to implement read/write access serialization to * the (@vol_id, @lnum) logical eraseblock * - * This data structure is used in the EBA unit to implement per-LEB locking. - * When a logical eraseblock is being locked - corresponding + * This data structure is used in the EBA sub-system to implement per-LEB + * locking. When a logical eraseblock is being locked - corresponding * &struct ubi_ltree_entry object is inserted to the lock tree (@ubi->ltree). - * See EBA unit for details. + * See EBA sub-system for details. */ struct ubi_ltree_entry { struct rb_node rb; @@ -225,7 +225,7 @@ struct ubi_volume { #ifdef CONFIG_MTD_UBI_GLUEBI /* * Gluebi-related stuff may be compiled out. - * TODO: this should not be built into UBI but should be a separate + * Note: this should not be built into UBI but should be a separate * ubimtd driver which works on top of UBI and emulates MTD devices. */ struct ubi_volume_desc *gluebi_desc; @@ -235,8 +235,7 @@ struct ubi_volume { }; /** - * struct ubi_volume_desc - descriptor of the UBI volume returned when it is - * opened. + * struct ubi_volume_desc - UBI volume descriptor returned when it is opened. * @vol: reference to the corresponding volume description object * @mode: open mode (%UBI_READONLY, %UBI_READWRITE, or %UBI_EXCLUSIVE) */ @@ -316,11 +315,11 @@ struct ubi_wl_entry; * @ro_mode: if the UBI device is in read-only mode * @leb_size: logical eraseblock size * @leb_start: starting offset of logical eraseblocks within physical - * eraseblocks + * eraseblocks * @ec_hdr_alsize: size of the EC header aligned to @hdrs_min_io_size * @vid_hdr_alsize: size of the VID header aligned to @hdrs_min_io_size * @vid_hdr_offset: starting offset of the volume identifier header (might be - * unaligned) + * unaligned) * @vid_hdr_aloffset: starting offset of the VID header aligned to * @hdrs_min_io_size * @vid_hdr_shift: contains @vid_hdr_offset - @vid_hdr_aloffset @@ -356,16 +355,16 @@ struct ubi_device { struct mutex volumes_mutex; int max_ec; - /* TODO: mean_ec is not updated run-time, fix */ + /* Note, mean_ec is not updated run-time - should be fixed */ int mean_ec; - /* EBA unit's stuff */ + /* EBA sub-system's stuff */ unsigned long long global_sqnum; spinlock_t ltree_lock; struct rb_root ltree; struct mutex alc_mutex; - /* Wear-leveling unit's stuff */ + /* Wear-leveling sub-system's stuff */ struct rb_root used; struct rb_root free; struct rb_root scrub; @@ -388,7 +387,7 @@ struct ubi_device { int thread_enabled; char bgt_name[sizeof(UBI_BGT_NAME_PATTERN)+2]; - /* I/O unit's stuff */ + /* I/O sub-system's stuff */ long long flash_size; int peb_count; int peb_size; diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c index cc8fe2934d2b..761952ba125b 100644 --- a/drivers/mtd/ubi/wl.c +++ b/drivers/mtd/ubi/wl.c @@ -19,22 +19,22 @@ */ /* - * UBI wear-leveling unit. + * UBI wear-leveling sub-system. * - * This unit is responsible for wear-leveling. It works in terms of physical - * eraseblocks and erase counters and knows nothing about logical eraseblocks, - * volumes, etc. From this unit's perspective all physical eraseblocks are of - * two types - used and free. Used physical eraseblocks are those that were - * "get" by the 'ubi_wl_get_peb()' function, and free physical eraseblocks are - * those that were put by the 'ubi_wl_put_peb()' function. + * This sub-system is responsible for wear-leveling. It works in terms of + * physical* eraseblocks and erase counters and knows nothing about logical + * eraseblocks, volumes, etc. From this sub-system's perspective all physical + * eraseblocks are of two types - used and free. Used physical eraseblocks are + * those that were "get" by the 'ubi_wl_get_peb()' function, and free physical + * eraseblocks are those that were put by the 'ubi_wl_put_peb()' function. * * Physical eraseblocks returned by 'ubi_wl_get_peb()' have only erase counter - * header. The rest of the physical eraseblock contains only 0xFF bytes. + * header. The rest of the physical eraseblock contains only %0xFF bytes. * - * When physical eraseblocks are returned to the WL unit by means of the + * When physical eraseblocks are returned to the WL sub-system by means of the * 'ubi_wl_put_peb()' function, they are scheduled for erasure. The erasure is * done asynchronously in context of the per-UBI device background thread, - * which is also managed by the WL unit. + * which is also managed by the WL sub-system. * * The wear-leveling is ensured by means of moving the contents of used * physical eraseblocks with low erase counter to free physical eraseblocks @@ -43,34 +43,36 @@ * The 'ubi_wl_get_peb()' function accepts data type hints which help to pick * an "optimal" physical eraseblock. For example, when it is known that the * physical eraseblock will be "put" soon because it contains short-term data, - * the WL unit may pick a free physical eraseblock with low erase counter, and - * so forth. + * the WL sub-system may pick a free physical eraseblock with low erase + * counter, and so forth. * - * If the WL unit fails to erase a physical eraseblock, it marks it as bad. + * If the WL sub-system fails to erase a physical eraseblock, it marks it as + * bad. * - * This unit is also responsible for scrubbing. If a bit-flip is detected in a - * physical eraseblock, it has to be moved. Technically this is the same as - * moving it for wear-leveling reasons. + * This sub-system is also responsible for scrubbing. If a bit-flip is detected + * in a physical eraseblock, it has to be moved. Technically this is the same + * as moving it for wear-leveling reasons. * - * As it was said, for the UBI unit all physical eraseblocks are either "free" - * or "used". Free eraseblock are kept in the @wl->free RB-tree, while used - * eraseblocks are kept in a set of different RB-trees: @wl->used, + * As it was said, for the UBI sub-system all physical eraseblocks are either + * "free" or "used". Free eraseblock are kept in the @wl->free RB-tree, while + * used eraseblocks are kept in a set of different RB-trees: @wl->used, * @wl->prot.pnum, @wl->prot.aec, and @wl->scrub. * * Note, in this implementation, we keep a small in-RAM object for each physical * eraseblock. This is surely not a scalable solution. But it appears to be good * enough for moderately large flashes and it is simple. In future, one may - * re-work this unit and make it more scalable. + * re-work this sub-system and make it more scalable. * - * At the moment this unit does not utilize the sequence number, which was - * introduced relatively recently. But it would be wise to do this because the - * sequence number of a logical eraseblock characterizes how old is it. For + * At the moment this sub-system does not utilize the sequence number, which + * was introduced relatively recently. But it would be wise to do this because + * the sequence number of a logical eraseblock characterizes how old is it. For * example, when we move a PEB with low erase counter, and we need to pick the * target PEB, we pick a PEB with the highest EC if our PEB is "old" and we * pick target PEB with an average EC if our PEB is not very "old". This is a - * room for future re-works of the WL unit. + * room for future re-works of the WL sub-system. * - * FIXME: looks too complex, should be simplified (later). + * Note: the stuff with protection trees looks too complex and is difficult to + * understand. Should be fixed. */ #include @@ -92,20 +94,21 @@ /* * Maximum difference between two erase counters. If this threshold is - * exceeded, the WL unit starts moving data from used physical eraseblocks with - * low erase counter to free physical eraseblocks with high erase counter. + * exceeded, the WL sub-system starts moving data from used physical + * eraseblocks with low erase counter to free physical eraseblocks with high + * erase counter. */ #define UBI_WL_THRESHOLD CONFIG_MTD_UBI_WL_THRESHOLD /* - * When a physical eraseblock is moved, the WL unit has to pick the target + * When a physical eraseblock is moved, the WL sub-system has to pick the target * physical eraseblock to move to. The simplest way would be just to pick the * one with the highest erase counter. But in certain workloads this could lead * to an unlimited wear of one or few physical eraseblock. Indeed, imagine a * situation when the picked physical eraseblock is constantly erased after the * data is written to it. So, we have a constant which limits the highest erase - * counter of the free physical eraseblock to pick. Namely, the WL unit does - * not pick eraseblocks with erase counter greater then the lowest erase + * counter of the free physical eraseblock to pick. Namely, the WL sub-system + * does not pick eraseblocks with erase counter greater then the lowest erase * counter plus %WL_FREE_MAX_DIFF. */ #define WL_FREE_MAX_DIFF (2*UBI_WL_THRESHOLD) @@ -123,11 +126,11 @@ * @abs_ec: the absolute erase counter value when the protection ends * @e: the wear-leveling entry of the physical eraseblock under protection * - * When the WL unit returns a physical eraseblock, the physical eraseblock is - * protected from being moved for some "time". For this reason, the physical - * eraseblock is not directly moved from the @wl->free tree to the @wl->used - * tree. There is one more tree in between where this physical eraseblock is - * temporarily stored (@wl->prot). + * When the WL sub-system returns a physical eraseblock, the physical + * eraseblock is protected from being moved for some "time". For this reason, + * the physical eraseblock is not directly moved from the @wl->free tree to the + * @wl->used tree. There is one more tree in between where this physical + * eraseblock is temporarily stored (@wl->prot). * * All this protection stuff is needed because: * o we don't want to move physical eraseblocks just after we have given them @@ -175,7 +178,6 @@ struct ubi_wl_prot_entry { * @list: a link in the list of pending works * @func: worker function * @priv: private data of the worker function - * * @e: physical eraseblock to erase * @torture: if the physical eraseblock has to be tortured * @@ -1136,7 +1138,7 @@ static int erase_worker(struct ubi_device *ubi, struct ubi_work *wl_wrk, } /** - * ubi_wl_put_peb - return a physical eraseblock to the wear-leveling unit. + * ubi_wl_put_peb - return a PEB to the wear-leveling sub-system. * @ubi: UBI device description object * @pnum: physical eraseblock to return * @torture: if this physical eraseblock has to be tortured @@ -1175,11 +1177,11 @@ int ubi_wl_put_peb(struct ubi_device *ubi, int pnum, int torture) /* * User is putting the physical eraseblock which was selected * as the target the data is moved to. It may happen if the EBA - * unit already re-mapped the LEB in 'ubi_eba_copy_leb()' but - * the WL unit has not put the PEB to the "used" tree yet, but - * it is about to do this. So we just set a flag which will - * tell the WL worker that the PEB is not needed anymore and - * should be scheduled for erasure. + * sub-system already re-mapped the LEB in 'ubi_eba_copy_leb()' + * but the WL sub-system has not put the PEB to the "used" tree + * yet, but it is about to do this. So we just set a flag which + * will tell the WL worker that the PEB is not needed anymore + * and should be scheduled for erasure. */ dbg_wl("PEB %d is the target of data moving", pnum); ubi_assert(!ubi->move_to_put); @@ -1425,8 +1427,7 @@ static void cancel_pending(struct ubi_device *ubi) } /** - * ubi_wl_init_scan - initialize the wear-leveling unit using scanning - * information. + * ubi_wl_init_scan - initialize the WL sub-system using scanning information. * @ubi: UBI device description object * @si: scanning information * @@ -1583,13 +1584,12 @@ static void protection_trees_destroy(struct ubi_device *ubi) } /** - * ubi_wl_close - close the wear-leveling unit. + * ubi_wl_close - close the wear-leveling sub-system. * @ubi: UBI device description object */ void ubi_wl_close(struct ubi_device *ubi) { - dbg_wl("close the UBI wear-leveling unit"); - + dbg_wl("close the WL sub-system"); cancel_pending(ubi); protection_trees_destroy(ubi); tree_destroy(&ubi->used); diff --git a/include/linux/mtd/ubi.h b/include/linux/mtd/ubi.h index 83302bbbddb4..6316fafe5c2a 100644 --- a/include/linux/mtd/ubi.h +++ b/include/linux/mtd/ubi.h @@ -45,13 +45,13 @@ enum { * @size: how many physical eraseblocks are reserved for this volume * @used_bytes: how many bytes of data this volume contains * @used_ebs: how many physical eraseblocks of this volume actually contain any - * data + * data * @vol_type: volume type (%UBI_DYNAMIC_VOLUME or %UBI_STATIC_VOLUME) * @corrupted: non-zero if the volume is corrupted (static volumes only) * @upd_marker: non-zero if the volume has update marker set * @alignment: volume alignment * @usable_leb_size: how many bytes are available in logical eraseblocks of - * this volume + * this volume * @name_len: volume name length * @name: volume name * @cdev: UBI volume character device major and minor numbers From c8566350a3229ca505b84313c65d1403b4d0cbfc Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Wed, 16 Jul 2008 17:40:22 +0300 Subject: [PATCH 104/857] UBI: fix and re-work debugging stuff Signed-off-by: Artem Bityutskiy --- drivers/mtd/ubi/build.c | 2 +- drivers/mtd/ubi/cdev.c | 26 +++---- drivers/mtd/ubi/debug.c | 160 ++++++++++++++++++++------------------- drivers/mtd/ubi/debug.h | 68 +++++++++++------ drivers/mtd/ubi/gluebi.c | 10 +-- drivers/mtd/ubi/io.c | 4 +- drivers/mtd/ubi/kapi.c | 20 ++--- drivers/mtd/ubi/scan.c | 2 +- drivers/mtd/ubi/upd.c | 16 ++-- drivers/mtd/ubi/vmt.c | 73 +++++++++--------- drivers/mtd/ubi/vtbl.c | 2 +- 11 files changed, 206 insertions(+), 177 deletions(-) diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c index 27271fe32e02..7210e1da1fcb 100644 --- a/drivers/mtd/ubi/build.c +++ b/drivers/mtd/ubi/build.c @@ -403,7 +403,7 @@ static int uif_init(struct ubi_device *ubi) ubi_assert(MINOR(dev) == 0); cdev_init(&ubi->cdev, &ubi_cdev_operations); - dbg_msg("%s major is %u", ubi->ubi_name, MAJOR(dev)); + dbg_gen("%s major is %u", ubi->ubi_name, MAJOR(dev)); ubi->cdev.owner = THIS_MODULE; err = cdev_add(&ubi->cdev, dev, 1); diff --git a/drivers/mtd/ubi/cdev.c b/drivers/mtd/ubi/cdev.c index 4fb84e3e6502..7c19918cc914 100644 --- a/drivers/mtd/ubi/cdev.c +++ b/drivers/mtd/ubi/cdev.c @@ -116,7 +116,7 @@ static int vol_cdev_open(struct inode *inode, struct file *file) else mode = UBI_READONLY; - dbg_msg("open volume %d, mode %d", vol_id, mode); + dbg_gen("open volume %d, mode %d", vol_id, mode); desc = ubi_open_volume(ubi_num, vol_id, mode); unlock_kernel(); @@ -132,7 +132,7 @@ static int vol_cdev_release(struct inode *inode, struct file *file) struct ubi_volume_desc *desc = file->private_data; struct ubi_volume *vol = desc->vol; - dbg_msg("release volume %d, mode %d", vol->vol_id, desc->mode); + dbg_gen("release volume %d, mode %d", vol->vol_id, desc->mode); if (vol->updating) { ubi_warn("update of volume %d not finished, volume is damaged", @@ -141,7 +141,7 @@ static int vol_cdev_release(struct inode *inode, struct file *file) vol->updating = 0; vfree(vol->upd_buf); } else if (vol->changing_leb) { - dbg_msg("only %lld of %lld bytes received for atomic LEB change" + dbg_gen("only %lld of %lld bytes received for atomic LEB change" " for volume %d:%d, cancel", vol->upd_received, vol->upd_bytes, vol->ubi->ubi_num, vol->vol_id); vol->changing_leb = 0; @@ -183,7 +183,7 @@ static loff_t vol_cdev_llseek(struct file *file, loff_t offset, int origin) return -EINVAL; } - dbg_msg("seek volume %d, offset %lld, origin %d, new offset %lld", + dbg_gen("seek volume %d, offset %lld, origin %d, new offset %lld", vol->vol_id, offset, origin, new_offset); file->f_pos = new_offset; @@ -201,7 +201,7 @@ static ssize_t vol_cdev_read(struct file *file, __user char *buf, size_t count, void *tbuf; uint64_t tmp; - dbg_msg("read %zd bytes from offset %lld of volume %d", + dbg_gen("read %zd bytes from offset %lld of volume %d", count, *offp, vol->vol_id); if (vol->updating) { @@ -216,7 +216,7 @@ static ssize_t vol_cdev_read(struct file *file, __user char *buf, size_t count, return 0; if (vol->corrupted) - dbg_msg("read from corrupted volume %d", vol->vol_id); + dbg_gen("read from corrupted volume %d", vol->vol_id); if (*offp + count > vol->used_bytes) count_save = count = vol->used_bytes - *offp; @@ -285,7 +285,7 @@ static ssize_t vol_cdev_direct_write(struct file *file, const char __user *buf, char *tbuf; uint64_t tmp; - dbg_msg("requested: write %zd bytes to offset %lld of volume %u", + dbg_gen("requested: write %zd bytes to offset %lld of volume %u", count, *offp, vol->vol_id); if (vol->vol_type == UBI_STATIC_VOLUME) @@ -514,7 +514,7 @@ static int vol_cdev_ioctl(struct inode *inode, struct file *file, break; } - dbg_msg("erase LEB %d:%d", vol->vol_id, lnum); + dbg_gen("erase LEB %d:%d", vol->vol_id, lnum); err = ubi_eba_unmap_leb(ubi, vol, lnum); if (err) break; @@ -626,7 +626,7 @@ static int ubi_cdev_ioctl(struct inode *inode, struct file *file, { struct ubi_mkvol_req req; - dbg_msg("create volume"); + dbg_gen("create volume"); err = copy_from_user(&req, argp, sizeof(struct ubi_mkvol_req)); if (err) { err = -EFAULT; @@ -656,7 +656,7 @@ static int ubi_cdev_ioctl(struct inode *inode, struct file *file, { int vol_id; - dbg_msg("remove volume"); + dbg_gen("remove volume"); err = get_user(vol_id, (__user int32_t *)argp); if (err) { err = -EFAULT; @@ -689,7 +689,7 @@ static int ubi_cdev_ioctl(struct inode *inode, struct file *file, uint64_t tmp; struct ubi_rsvol_req req; - dbg_msg("re-size volume"); + dbg_gen("re-size volume"); err = copy_from_user(&req, argp, sizeof(struct ubi_rsvol_req)); if (err) { err = -EFAULT; @@ -742,7 +742,7 @@ static int ctrl_cdev_ioctl(struct inode *inode, struct file *file, struct ubi_attach_req req; struct mtd_info *mtd; - dbg_msg("attach MTD device"); + dbg_gen("attach MTD device"); err = copy_from_user(&req, argp, sizeof(struct ubi_attach_req)); if (err) { err = -EFAULT; @@ -782,7 +782,7 @@ static int ctrl_cdev_ioctl(struct inode *inode, struct file *file, { int ubi_num; - dbg_msg("dettach MTD device"); + dbg_gen("dettach MTD device"); err = get_user(ubi_num, (__user int32_t *)argp); if (err) { err = -EFAULT; diff --git a/drivers/mtd/ubi/debug.c b/drivers/mtd/ubi/debug.c index 56956ec2845f..21e0d7d76a44 100644 --- a/drivers/mtd/ubi/debug.c +++ b/drivers/mtd/ubi/debug.c @@ -24,7 +24,7 @@ * changes. */ -#ifdef CONFIG_MTD_UBI_DEBUG_MSG +#ifdef CONFIG_MTD_UBI_DEBUG #include "ubi.h" @@ -34,14 +34,19 @@ */ void ubi_dbg_dump_ec_hdr(const struct ubi_ec_hdr *ec_hdr) { - dbg_msg("erase counter header dump:"); - dbg_msg("magic %#08x", be32_to_cpu(ec_hdr->magic)); - dbg_msg("version %d", (int)ec_hdr->version); - dbg_msg("ec %llu", (long long)be64_to_cpu(ec_hdr->ec)); - dbg_msg("vid_hdr_offset %d", be32_to_cpu(ec_hdr->vid_hdr_offset)); - dbg_msg("data_offset %d", be32_to_cpu(ec_hdr->data_offset)); - dbg_msg("hdr_crc %#08x", be32_to_cpu(ec_hdr->hdr_crc)); - dbg_msg("erase counter header hexdump:"); + printk(KERN_DEBUG "Erase counter header dump:\n"); + printk(KERN_DEBUG "\tmagic %#08x\n", + be32_to_cpu(ec_hdr->magic)); + printk(KERN_DEBUG "\tversion %d\n", (int)ec_hdr->version); + printk(KERN_DEBUG "\tec %llu\n", + (long long)be64_to_cpu(ec_hdr->ec)); + printk(KERN_DEBUG "\tvid_hdr_offset %d\n", + be32_to_cpu(ec_hdr->vid_hdr_offset)); + printk(KERN_DEBUG "\tdata_offset %d\n", + be32_to_cpu(ec_hdr->data_offset)); + printk(KERN_DEBUG "\thdr_crc %#08x\n", + be32_to_cpu(ec_hdr->hdr_crc)); + printk(KERN_DEBUG "erase counter header hexdump:\n"); print_hex_dump(KERN_DEBUG, "", DUMP_PREFIX_OFFSET, 32, 1, ec_hdr, UBI_EC_HDR_SIZE, 1); } @@ -52,22 +57,24 @@ void ubi_dbg_dump_ec_hdr(const struct ubi_ec_hdr *ec_hdr) */ void ubi_dbg_dump_vid_hdr(const struct ubi_vid_hdr *vid_hdr) { - dbg_msg("volume identifier header dump:"); - dbg_msg("magic %08x", be32_to_cpu(vid_hdr->magic)); - dbg_msg("version %d", (int)vid_hdr->version); - dbg_msg("vol_type %d", (int)vid_hdr->vol_type); - dbg_msg("copy_flag %d", (int)vid_hdr->copy_flag); - dbg_msg("compat %d", (int)vid_hdr->compat); - dbg_msg("vol_id %d", be32_to_cpu(vid_hdr->vol_id)); - dbg_msg("lnum %d", be32_to_cpu(vid_hdr->lnum)); - dbg_msg("leb_ver %u", be32_to_cpu(vid_hdr->leb_ver)); - dbg_msg("data_size %d", be32_to_cpu(vid_hdr->data_size)); - dbg_msg("used_ebs %d", be32_to_cpu(vid_hdr->used_ebs)); - dbg_msg("data_pad %d", be32_to_cpu(vid_hdr->data_pad)); - dbg_msg("sqnum %llu", + printk(KERN_DEBUG "Volume identifier header dump:\n"); + printk(KERN_DEBUG "\tmagic %08x\n", be32_to_cpu(vid_hdr->magic)); + printk(KERN_DEBUG "\tversion %d\n", (int)vid_hdr->version); + printk(KERN_DEBUG "\tvol_type %d\n", (int)vid_hdr->vol_type); + printk(KERN_DEBUG "\tcopy_flag %d\n", (int)vid_hdr->copy_flag); + printk(KERN_DEBUG "\tcompat %d\n", (int)vid_hdr->compat); + printk(KERN_DEBUG "\tvol_id %d\n", be32_to_cpu(vid_hdr->vol_id)); + printk(KERN_DEBUG "\tlnum %d\n", be32_to_cpu(vid_hdr->lnum)); + printk(KERN_DEBUG "\tleb_ver %u\n", be32_to_cpu(vid_hdr->leb_ver)); + printk(KERN_DEBUG "\tdata_size %d\n", be32_to_cpu(vid_hdr->data_size)); + printk(KERN_DEBUG "\tused_ebs %d\n", be32_to_cpu(vid_hdr->used_ebs)); + printk(KERN_DEBUG "\tdata_pad %d\n", be32_to_cpu(vid_hdr->data_pad)); + printk(KERN_DEBUG "\tsqnum %llu\n", (unsigned long long)be64_to_cpu(vid_hdr->sqnum)); - dbg_msg("hdr_crc %08x", be32_to_cpu(vid_hdr->hdr_crc)); - dbg_msg("volume identifier header hexdump:"); + printk(KERN_DEBUG "\thdr_crc %08x\n", be32_to_cpu(vid_hdr->hdr_crc)); + printk(KERN_DEBUG "Volume identifier header hexdump:\n"); + print_hex_dump(KERN_DEBUG, "", DUMP_PREFIX_OFFSET, 32, 1, + vid_hdr, UBI_VID_HDR_SIZE, 1); } /** @@ -76,27 +83,27 @@ void ubi_dbg_dump_vid_hdr(const struct ubi_vid_hdr *vid_hdr) */ void ubi_dbg_dump_vol_info(const struct ubi_volume *vol) { - dbg_msg("volume information dump:"); - dbg_msg("vol_id %d", vol->vol_id); - dbg_msg("reserved_pebs %d", vol->reserved_pebs); - dbg_msg("alignment %d", vol->alignment); - dbg_msg("data_pad %d", vol->data_pad); - dbg_msg("vol_type %d", vol->vol_type); - dbg_msg("name_len %d", vol->name_len); - dbg_msg("usable_leb_size %d", vol->usable_leb_size); - dbg_msg("used_ebs %d", vol->used_ebs); - dbg_msg("used_bytes %lld", vol->used_bytes); - dbg_msg("last_eb_bytes %d", vol->last_eb_bytes); - dbg_msg("corrupted %d", vol->corrupted); - dbg_msg("upd_marker %d", vol->upd_marker); + printk(KERN_DEBUG "Volume information dump:\n"); + printk(KERN_DEBUG "\tvol_id %d\n", vol->vol_id); + printk(KERN_DEBUG "\treserved_pebs %d\n", vol->reserved_pebs); + printk(KERN_DEBUG "\talignment %d\n", vol->alignment); + printk(KERN_DEBUG "\tdata_pad %d\n", vol->data_pad); + printk(KERN_DEBUG "\tvol_type %d\n", vol->vol_type); + printk(KERN_DEBUG "\tname_len %d\n", vol->name_len); + printk(KERN_DEBUG "\tusable_leb_size %d\n", vol->usable_leb_size); + printk(KERN_DEBUG "\tused_ebs %d\n", vol->used_ebs); + printk(KERN_DEBUG "\tused_bytes %lld\n", vol->used_bytes); + printk(KERN_DEBUG "\tlast_eb_bytes %d\n", vol->last_eb_bytes); + printk(KERN_DEBUG "\tcorrupted %d\n", vol->corrupted); + printk(KERN_DEBUG "\tupd_marker %d\n", vol->upd_marker); if (vol->name_len <= UBI_VOL_NAME_MAX && strnlen(vol->name, vol->name_len + 1) == vol->name_len) { - dbg_msg("name %s", vol->name); + printk(KERN_DEBUG "\tname %s\n", vol->name); } else { - dbg_msg("the 1st 5 characters of the name: %c%c%c%c%c", - vol->name[0], vol->name[1], vol->name[2], - vol->name[3], vol->name[4]); + printk(KERN_DEBUG "\t1st 5 characters of name: %c%c%c%c%c\n", + vol->name[0], vol->name[1], vol->name[2], + vol->name[3], vol->name[4]); } } @@ -109,28 +116,29 @@ void ubi_dbg_dump_vtbl_record(const struct ubi_vtbl_record *r, int idx) { int name_len = be16_to_cpu(r->name_len); - dbg_msg("volume table record %d dump:", idx); - dbg_msg("reserved_pebs %d", be32_to_cpu(r->reserved_pebs)); - dbg_msg("alignment %d", be32_to_cpu(r->alignment)); - dbg_msg("data_pad %d", be32_to_cpu(r->data_pad)); - dbg_msg("vol_type %d", (int)r->vol_type); - dbg_msg("upd_marker %d", (int)r->upd_marker); - dbg_msg("name_len %d", name_len); + printk(KERN_DEBUG "Volume table record %d dump:\n", idx); + printk(KERN_DEBUG "\treserved_pebs %d\n", + be32_to_cpu(r->reserved_pebs)); + printk(KERN_DEBUG "\talignment %d\n", be32_to_cpu(r->alignment)); + printk(KERN_DEBUG "\tdata_pad %d\n", be32_to_cpu(r->data_pad)); + printk(KERN_DEBUG "\tvol_type %d\n", (int)r->vol_type); + printk(KERN_DEBUG "\tupd_marker %d\n", (int)r->upd_marker); + printk(KERN_DEBUG "\tname_len %d\n", name_len); if (r->name[0] == '\0') { - dbg_msg("name NULL"); + printk(KERN_DEBUG "\tname NULL\n"); return; } if (name_len <= UBI_VOL_NAME_MAX && strnlen(&r->name[0], name_len + 1) == name_len) { - dbg_msg("name %s", &r->name[0]); + printk(KERN_DEBUG "\tname %s\n", &r->name[0]); } else { - dbg_msg("1st 5 characters of the name: %c%c%c%c%c", + printk(KERN_DEBUG "\t1st 5 characters of name: %c%c%c%c%c\n", r->name[0], r->name[1], r->name[2], r->name[3], r->name[4]); } - dbg_msg("crc %#08x", be32_to_cpu(r->crc)); + printk(KERN_DEBUG "\tcrc %#08x\n", be32_to_cpu(r->crc)); } /** @@ -139,15 +147,15 @@ void ubi_dbg_dump_vtbl_record(const struct ubi_vtbl_record *r, int idx) */ void ubi_dbg_dump_sv(const struct ubi_scan_volume *sv) { - dbg_msg("volume scanning information dump:"); - dbg_msg("vol_id %d", sv->vol_id); - dbg_msg("highest_lnum %d", sv->highest_lnum); - dbg_msg("leb_count %d", sv->leb_count); - dbg_msg("compat %d", sv->compat); - dbg_msg("vol_type %d", sv->vol_type); - dbg_msg("used_ebs %d", sv->used_ebs); - dbg_msg("last_data_size %d", sv->last_data_size); - dbg_msg("data_pad %d", sv->data_pad); + printk(KERN_DEBUG "Volume scanning information dump:\n"); + printk(KERN_DEBUG "\tvol_id %d\n", sv->vol_id); + printk(KERN_DEBUG "\thighest_lnum %d\n", sv->highest_lnum); + printk(KERN_DEBUG "\tleb_count %d\n", sv->leb_count); + printk(KERN_DEBUG "\tcompat %d\n", sv->compat); + printk(KERN_DEBUG "\tvol_type %d\n", sv->vol_type); + printk(KERN_DEBUG "\tused_ebs %d\n", sv->used_ebs); + printk(KERN_DEBUG "\tlast_data_size %d\n", sv->last_data_size); + printk(KERN_DEBUG "\tdata_pad %d\n", sv->data_pad); } /** @@ -157,14 +165,14 @@ void ubi_dbg_dump_sv(const struct ubi_scan_volume *sv) */ void ubi_dbg_dump_seb(const struct ubi_scan_leb *seb, int type) { - dbg_msg("eraseblock scanning information dump:"); - dbg_msg("ec %d", seb->ec); - dbg_msg("pnum %d", seb->pnum); + printk(KERN_DEBUG "eraseblock scanning information dump:\n"); + printk(KERN_DEBUG "\tec %d\n", seb->ec); + printk(KERN_DEBUG "\tpnum %d\n", seb->pnum); if (type == 0) { - dbg_msg("lnum %d", seb->lnum); - dbg_msg("scrub %d", seb->scrub); - dbg_msg("sqnum %llu", seb->sqnum); - dbg_msg("leb_ver %u", seb->leb_ver); + printk(KERN_DEBUG "\tlnum %d\n", seb->lnum); + printk(KERN_DEBUG "\tscrub %d\n", seb->scrub); + printk(KERN_DEBUG "\tsqnum %llu\n", seb->sqnum); + printk(KERN_DEBUG "\tleb_ver %u\n", seb->leb_ver); } } @@ -176,16 +184,16 @@ void ubi_dbg_dump_mkvol_req(const struct ubi_mkvol_req *req) { char nm[17]; - dbg_msg("volume creation request dump:"); - dbg_msg("vol_id %d", req->vol_id); - dbg_msg("alignment %d", req->alignment); - dbg_msg("bytes %lld", (long long)req->bytes); - dbg_msg("vol_type %d", req->vol_type); - dbg_msg("name_len %d", req->name_len); + printk(KERN_DEBUG "Volume creation request dump:\n"); + printk(KERN_DEBUG "\tvol_id %d\n", req->vol_id); + printk(KERN_DEBUG "\talignment %d\n", req->alignment); + printk(KERN_DEBUG "\tbytes %lld\n", (long long)req->bytes); + printk(KERN_DEBUG "\tvol_type %d\n", req->vol_type); + printk(KERN_DEBUG "\tname_len %d\n", req->name_len); memcpy(nm, req->name, 16); nm[16] = 0; - dbg_msg("the 1st 16 characters of the name: %s", nm); + printk(KERN_DEBUG "\t1st 16 characters of name: %s\n", nm); } -#endif /* CONFIG_MTD_UBI_DEBUG_MSG */ +#endif /* CONFIG_MTD_UBI_DEBUG */ diff --git a/drivers/mtd/ubi/debug.h b/drivers/mtd/ubi/debug.h index 7d8d77c31dfe..78e914d23ece 100644 --- a/drivers/mtd/ubi/debug.h +++ b/drivers/mtd/ubi/debug.h @@ -24,21 +24,16 @@ #ifdef CONFIG_MTD_UBI_DEBUG #include -#define ubi_assert(expr) BUG_ON(!(expr)) #define dbg_err(fmt, ...) ubi_err(fmt, ##__VA_ARGS__) -#else -#define ubi_assert(expr) ({}) -#define dbg_err(fmt, ...) ({}) -#endif -#ifdef CONFIG_MTD_UBI_DEBUG_DISABLE_BGT -#define DBG_DISABLE_BGT 1 -#else -#define DBG_DISABLE_BGT 0 -#endif +#define ubi_assert(expr) do { \ + if (unlikely(!(expr))) { \ + printk(KERN_CRIT "UBI assert failed in %s at %u (pid %d)\n", \ + __func__, __LINE__, current->pid); \ + ubi_dbg_dump_stack(); \ + } \ +} while (0) -#ifdef CONFIG_MTD_UBI_DEBUG_MSG -/* Generic debugging message */ #define dbg_msg(fmt, ...) \ printk(KERN_DEBUG "UBI DBG (pid %d): %s: " fmt "\n", \ current->pid, __func__, ##__VA_ARGS__) @@ -61,19 +56,12 @@ void ubi_dbg_dump_sv(const struct ubi_scan_volume *sv); void ubi_dbg_dump_seb(const struct ubi_scan_leb *seb, int type); void ubi_dbg_dump_mkvol_req(const struct ubi_mkvol_req *req); +#ifdef CONFIG_MTD_UBI_DEBUG_MSG +/* General debugging messages */ +#define dbg_gen(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__) #else - -#define dbg_msg(fmt, ...) ({}) -#define ubi_dbg_dump_stack() ({}) -#define ubi_dbg_dump_ec_hdr(ec_hdr) ({}) -#define ubi_dbg_dump_vid_hdr(vid_hdr) ({}) -#define ubi_dbg_dump_vol_info(vol) ({}) -#define ubi_dbg_dump_vtbl_record(r, idx) ({}) -#define ubi_dbg_dump_sv(sv) ({}) -#define ubi_dbg_dump_seb(seb, type) ({}) -#define ubi_dbg_dump_mkvol_req(req) ({}) - -#endif /* CONFIG_MTD_UBI_DEBUG_MSG */ +#define dbg_gen(fmt, ...) ({}) +#endif #ifdef CONFIG_MTD_UBI_DEBUG_MSG_EBA /* Messages from the eraseblock association sub-system */ @@ -105,6 +93,12 @@ void ubi_dbg_dump_mkvol_req(const struct ubi_mkvol_req *req); #define UBI_IO_DEBUG 0 #endif +#ifdef CONFIG_MTD_UBI_DEBUG_DISABLE_BGT +#define DBG_DISABLE_BGT 1 +#else +#define DBG_DISABLE_BGT 0 +#endif + #ifdef CONFIG_MTD_UBI_DEBUG_EMULATE_BITFLIPS /** * ubi_dbg_is_bitflip - if it is time to emulate a bit-flip. @@ -149,4 +143,30 @@ static inline int ubi_dbg_is_erase_failure(void) #define ubi_dbg_is_erase_failure() 0 #endif +#else + +#define ubi_assert(expr) ({}) +#define dbg_err(fmt, ...) ({}) +#define dbg_msg(fmt, ...) ({}) +#define dbg_gen(fmt, ...) ({}) +#define dbg_eba(fmt, ...) ({}) +#define dbg_wl(fmt, ...) ({}) +#define dbg_io(fmt, ...) ({}) +#define dbg_bld(fmt, ...) ({}) +#define ubi_dbg_dump_stack() ({}) +#define ubi_dbg_dump_ec_hdr(ec_hdr) ({}) +#define ubi_dbg_dump_vid_hdr(vid_hdr) ({}) +#define ubi_dbg_dump_vol_info(vol) ({}) +#define ubi_dbg_dump_vtbl_record(r, idx) ({}) +#define ubi_dbg_dump_sv(sv) ({}) +#define ubi_dbg_dump_seb(seb, type) ({}) +#define ubi_dbg_dump_mkvol_req(req) ({}) + +#define UBI_IO_DEBUG 0 +#define DBG_DISABLE_BGT 0 +#define ubi_dbg_is_bitflip() 0 +#define ubi_dbg_is_write_failure() 0 +#define ubi_dbg_is_erase_failure() 0 + +#endif /* !CONFIG_MTD_UBI_DEBUG */ #endif /* !__UBI_DEBUG_H__ */ diff --git a/drivers/mtd/ubi/gluebi.c b/drivers/mtd/ubi/gluebi.c index ae76ab638b2f..49f52dceea93 100644 --- a/drivers/mtd/ubi/gluebi.c +++ b/drivers/mtd/ubi/gluebi.c @@ -111,7 +111,7 @@ static int gluebi_read(struct mtd_info *mtd, loff_t from, size_t len, struct ubi_device *ubi; uint64_t tmp = from; - dbg_msg("read %zd bytes from offset %lld", len, from); + dbg_gen("read %zd bytes from offset %lld", len, from); if (len < 0 || from < 0 || from + len > mtd->size) return -EINVAL; @@ -162,7 +162,7 @@ static int gluebi_write(struct mtd_info *mtd, loff_t to, size_t len, struct ubi_device *ubi; uint64_t tmp = to; - dbg_msg("write %zd bytes to offset %lld", len, to); + dbg_gen("write %zd bytes to offset %lld", len, to); if (len < 0 || to < 0 || len + to > mtd->size) return -EINVAL; @@ -215,7 +215,7 @@ static int gluebi_erase(struct mtd_info *mtd, struct erase_info *instr) struct ubi_volume *vol; struct ubi_device *ubi; - dbg_msg("erase %u bytes at offset %u", instr->len, instr->addr); + dbg_gen("erase %u bytes at offset %u", instr->len, instr->addr); if (instr->addr < 0 || instr->addr > mtd->size - mtd->erasesize) return -EINVAL; @@ -304,7 +304,7 @@ int ubi_create_gluebi(struct ubi_device *ubi, struct ubi_volume *vol) return -ENFILE; } - dbg_msg("added mtd%d (\"%s\"), size %u, EB size %u", + dbg_gen("added mtd%d (\"%s\"), size %u, EB size %u", mtd->index, mtd->name, mtd->size, mtd->erasesize); return 0; } @@ -322,7 +322,7 @@ int ubi_destroy_gluebi(struct ubi_volume *vol) int err; struct mtd_info *mtd = &vol->gluebi_mtd; - dbg_msg("remove mtd%d", mtd->index); + dbg_gen("remove mtd%d", mtd->index); err = del_mtd_device(mtd); if (err) return err; diff --git a/drivers/mtd/ubi/io.c b/drivers/mtd/ubi/io.c index 561e7b2f96cb..27b9c2c2fc60 100644 --- a/drivers/mtd/ubi/io.c +++ b/drivers/mtd/ubi/io.c @@ -187,7 +187,7 @@ int ubi_io_read(const struct ubi_device *ubi, void *buf, int pnum, int offset, ubi_assert(len == read); if (ubi_dbg_is_bitflip()) { - dbg_msg("bit-flip (emulated)"); + dbg_gen("bit-flip (emulated)"); err = UBI_IO_BITFLIPS; } } @@ -1256,7 +1256,7 @@ static int paranoid_check_all_ff(struct ubi_device *ubi, int pnum, int offset, fail: ubi_err("paranoid check failed for PEB %d", pnum); - dbg_msg("hex dump of the %d-%d region", offset, offset + len); + ubi_msg("hex dump of the %d-%d region", offset, offset + len); print_hex_dump(KERN_DEBUG, "", DUMP_PREFIX_OFFSET, 32, 1, ubi->dbg_peb_buf, len, 1); err = 1; diff --git a/drivers/mtd/ubi/kapi.c b/drivers/mtd/ubi/kapi.c index e65c8e0bcd5d..5d9bcf109c13 100644 --- a/drivers/mtd/ubi/kapi.c +++ b/drivers/mtd/ubi/kapi.c @@ -106,7 +106,7 @@ struct ubi_volume_desc *ubi_open_volume(int ubi_num, int vol_id, int mode) struct ubi_device *ubi; struct ubi_volume *vol; - dbg_msg("open device %d volume %d, mode %d", ubi_num, vol_id, mode); + dbg_gen("open device %d volume %d, mode %d", ubi_num, vol_id, mode); if (ubi_num < 0 || ubi_num >= UBI_MAX_DEVICES) return ERR_PTR(-EINVAL); @@ -215,7 +215,7 @@ struct ubi_volume_desc *ubi_open_volume_nm(int ubi_num, const char *name, struct ubi_device *ubi; struct ubi_volume_desc *ret; - dbg_msg("open volume %s, mode %d", name, mode); + dbg_gen("open volume %s, mode %d", name, mode); if (!name) return ERR_PTR(-EINVAL); @@ -266,7 +266,7 @@ void ubi_close_volume(struct ubi_volume_desc *desc) struct ubi_volume *vol = desc->vol; struct ubi_device *ubi = vol->ubi; - dbg_msg("close volume %d, mode %d", vol->vol_id, desc->mode); + dbg_gen("close volume %d, mode %d", vol->vol_id, desc->mode); spin_lock(&ubi->volumes_lock); switch (desc->mode) { @@ -323,7 +323,7 @@ int ubi_leb_read(struct ubi_volume_desc *desc, int lnum, char *buf, int offset, struct ubi_device *ubi = vol->ubi; int err, vol_id = vol->vol_id; - dbg_msg("read %d bytes from LEB %d:%d:%d", len, vol_id, lnum, offset); + dbg_gen("read %d bytes from LEB %d:%d:%d", len, vol_id, lnum, offset); if (vol_id < 0 || vol_id >= ubi->vtbl_slots || lnum < 0 || lnum >= vol->used_ebs || offset < 0 || len < 0 || @@ -388,7 +388,7 @@ int ubi_leb_write(struct ubi_volume_desc *desc, int lnum, const void *buf, struct ubi_device *ubi = vol->ubi; int vol_id = vol->vol_id; - dbg_msg("write %d bytes to LEB %d:%d:%d", len, vol_id, lnum, offset); + dbg_gen("write %d bytes to LEB %d:%d:%d", len, vol_id, lnum, offset); if (vol_id < 0 || vol_id >= ubi->vtbl_slots) return -EINVAL; @@ -438,7 +438,7 @@ int ubi_leb_change(struct ubi_volume_desc *desc, int lnum, const void *buf, struct ubi_device *ubi = vol->ubi; int vol_id = vol->vol_id; - dbg_msg("atomically write %d bytes to LEB %d:%d", len, vol_id, lnum); + dbg_gen("atomically write %d bytes to LEB %d:%d", len, vol_id, lnum); if (vol_id < 0 || vol_id >= ubi->vtbl_slots) return -EINVAL; @@ -482,7 +482,7 @@ int ubi_leb_erase(struct ubi_volume_desc *desc, int lnum) struct ubi_device *ubi = vol->ubi; int err; - dbg_msg("erase LEB %d:%d", vol->vol_id, lnum); + dbg_gen("erase LEB %d:%d", vol->vol_id, lnum); if (desc->mode == UBI_READONLY || vol->vol_type == UBI_STATIC_VOLUME) return -EROFS; @@ -542,7 +542,7 @@ int ubi_leb_unmap(struct ubi_volume_desc *desc, int lnum) struct ubi_volume *vol = desc->vol; struct ubi_device *ubi = vol->ubi; - dbg_msg("unmap LEB %d:%d", vol->vol_id, lnum); + dbg_gen("unmap LEB %d:%d", vol->vol_id, lnum); if (desc->mode == UBI_READONLY || vol->vol_type == UBI_STATIC_VOLUME) return -EROFS; @@ -579,7 +579,7 @@ int ubi_leb_map(struct ubi_volume_desc *desc, int lnum, int dtype) struct ubi_volume *vol = desc->vol; struct ubi_device *ubi = vol->ubi; - dbg_msg("unmap LEB %d:%d", vol->vol_id, lnum); + dbg_gen("unmap LEB %d:%d", vol->vol_id, lnum); if (desc->mode == UBI_READONLY || vol->vol_type == UBI_STATIC_VOLUME) return -EROFS; @@ -621,7 +621,7 @@ int ubi_is_mapped(struct ubi_volume_desc *desc, int lnum) { struct ubi_volume *vol = desc->vol; - dbg_msg("test LEB %d:%d", vol->vol_id, lnum); + dbg_gen("test LEB %d:%d", vol->vol_id, lnum); if (lnum < 0 || lnum >= vol->reserved_pebs) return -EINVAL; diff --git a/drivers/mtd/ubi/scan.c b/drivers/mtd/ubi/scan.c index 892c2ba49777..40eca9ce5fab 100644 --- a/drivers/mtd/ubi/scan.c +++ b/drivers/mtd/ubi/scan.c @@ -932,7 +932,7 @@ struct ubi_scan_info *ubi_scan(struct ubi_device *ubi) for (pnum = 0; pnum < ubi->peb_count; pnum++) { cond_resched(); - dbg_msg("process PEB %d", pnum); + dbg_gen("process PEB %d", pnum); err = process_eb(ubi, si, pnum); if (err < 0) goto out_vidh; diff --git a/drivers/mtd/ubi/upd.c b/drivers/mtd/ubi/upd.c index 6fa1ab3f2a70..1230a5e1b530 100644 --- a/drivers/mtd/ubi/upd.c +++ b/drivers/mtd/ubi/upd.c @@ -56,11 +56,11 @@ static int set_update_marker(struct ubi_device *ubi, struct ubi_volume *vol) int err; struct ubi_vtbl_record vtbl_rec; - dbg_msg("set update marker for volume %d", vol->vol_id); + dbg_gen("set update marker for volume %d", vol->vol_id); if (vol->upd_marker) { ubi_assert(ubi->vtbl[vol->vol_id].upd_marker); - dbg_msg("already set"); + dbg_gen("already set"); return 0; } @@ -92,7 +92,7 @@ static int clear_update_marker(struct ubi_device *ubi, struct ubi_volume *vol, uint64_t tmp; struct ubi_vtbl_record vtbl_rec; - dbg_msg("clear update marker for volume %d", vol->vol_id); + dbg_gen("clear update marker for volume %d", vol->vol_id); memcpy(&vtbl_rec, &ubi->vtbl[vol->vol_id], sizeof(struct ubi_vtbl_record)); @@ -133,7 +133,7 @@ int ubi_start_update(struct ubi_device *ubi, struct ubi_volume *vol, int i, err; uint64_t tmp; - dbg_msg("start update of volume %d, %llu bytes", vol->vol_id, bytes); + dbg_gen("start update of volume %d, %llu bytes", vol->vol_id, bytes); ubi_assert(!vol->updating && !vol->changing_leb); vol->updating = 1; @@ -183,7 +183,7 @@ int ubi_start_leb_change(struct ubi_device *ubi, struct ubi_volume *vol, { ubi_assert(!vol->updating && !vol->changing_leb); - dbg_msg("start changing LEB %d:%d, %u bytes", + dbg_gen("start changing LEB %d:%d, %u bytes", vol->vol_id, req->lnum, req->bytes); if (req->bytes == 0) return ubi_eba_atomic_leb_change(ubi, vol, req->lnum, NULL, 0, @@ -242,7 +242,7 @@ static int write_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum, memset(buf + len, 0xFF, l - len); len = ubi_calc_data_len(ubi, buf, l); if (len == 0) { - dbg_msg("all %d bytes contain 0xFF - skip", len); + dbg_gen("all %d bytes contain 0xFF - skip", len); return 0; } @@ -283,7 +283,7 @@ int ubi_more_update_data(struct ubi_device *ubi, struct ubi_volume *vol, uint64_t tmp; int lnum, offs, err = 0, len, to_write = count; - dbg_msg("write %d of %lld bytes, %lld already passed", + dbg_gen("write %d of %lld bytes, %lld already passed", count, vol->upd_bytes, vol->upd_received); if (ubi->ro_mode) @@ -400,7 +400,7 @@ int ubi_more_leb_change_data(struct ubi_device *ubi, struct ubi_volume *vol, { int err; - dbg_msg("write %d of %lld bytes, %lld already passed", + dbg_gen("write %d of %lld bytes, %lld already passed", count, vol->upd_bytes, vol->upd_received); if (ubi->ro_mode) diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c index bfa7c5d2e064..2cd886a5adac 100644 --- a/drivers/mtd/ubi/vmt.c +++ b/drivers/mtd/ubi/vmt.c @@ -28,9 +28,9 @@ #include "ubi.h" #ifdef CONFIG_MTD_UBI_DEBUG_PARANOID -static void paranoid_check_volumes(struct ubi_device *ubi); +static int paranoid_check_volumes(struct ubi_device *ubi); #else -#define paranoid_check_volumes(ubi) +#define paranoid_check_volumes(ubi) 0 #endif static ssize_t vol_attribute_show(struct device *dev, @@ -218,7 +218,7 @@ int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req) spin_lock(&ubi->volumes_lock); if (vol_id == UBI_VOL_NUM_AUTO) { /* Find unused volume ID */ - dbg_msg("search for vacant volume ID"); + dbg_gen("search for vacant volume ID"); for (i = 0; i < ubi->vtbl_slots; i++) if (!ubi->volumes[i]) { vol_id = i; @@ -233,7 +233,7 @@ int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req) req->vol_id = vol_id; } - dbg_msg("volume ID %d, %llu bytes, type %d, name %s", + dbg_gen("volume ID %d, %llu bytes, type %d, name %s", vol_id, (unsigned long long)req->bytes, (int)req->vol_type, req->name); @@ -361,8 +361,8 @@ int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req) ubi->vol_count += 1; spin_unlock(&ubi->volumes_lock); - paranoid_check_volumes(ubi); - return 0; + err = paranoid_check_volumes(ubi); + return err; out_sysfs: /* @@ -414,7 +414,7 @@ int ubi_remove_volume(struct ubi_volume_desc *desc) struct ubi_device *ubi = vol->ubi; int i, err, vol_id = vol->vol_id, reserved_pebs = vol->reserved_pebs; - dbg_msg("remove UBI volume %d", vol_id); + dbg_gen("remove UBI volume %d", vol_id); ubi_assert(desc->mode == UBI_EXCLUSIVE); ubi_assert(vol == ubi->volumes[vol_id]); @@ -465,8 +465,8 @@ int ubi_remove_volume(struct ubi_volume_desc *desc) ubi->vol_count -= 1; spin_unlock(&ubi->volumes_lock); - paranoid_check_volumes(ubi); - return 0; + err = paranoid_check_volumes(ubi); + return err; out_err: ubi_err("cannot remove volume %d, error %d", vol_id, err); @@ -497,7 +497,7 @@ int ubi_resize_volume(struct ubi_volume_desc *desc, int reserved_pebs) if (ubi->ro_mode) return -EROFS; - dbg_msg("re-size volume %d to from %d to %d PEBs", + dbg_gen("re-size volume %d to from %d to %d PEBs", vol_id, vol->reserved_pebs, reserved_pebs); if (vol->vol_type == UBI_STATIC_VOLUME && @@ -586,8 +586,8 @@ int ubi_resize_volume(struct ubi_volume_desc *desc, int reserved_pebs) (long long)vol->used_ebs * vol->usable_leb_size; } - paranoid_check_volumes(ubi); - return 0; + err = paranoid_check_volumes(ubi); + return err; out_acc: if (pebs > 0) { @@ -615,8 +615,7 @@ int ubi_add_volume(struct ubi_device *ubi, struct ubi_volume *vol) int err, vol_id = vol->vol_id; dev_t dev; - dbg_msg("add volume %d", vol_id); - ubi_dbg_dump_vol_info(vol); + dbg_gen("add volume %d", vol_id); /* Register character device for the volume */ cdev_init(&vol->cdev, &ubi_vol_cdev_operations); @@ -650,8 +649,8 @@ int ubi_add_volume(struct ubi_device *ubi, struct ubi_volume *vol) return err; } - paranoid_check_volumes(ubi); - return 0; + err = paranoid_check_volumes(ubi); + return err; out_gluebi: err = ubi_destroy_gluebi(vol); @@ -672,7 +671,7 @@ void ubi_free_volume(struct ubi_device *ubi, struct ubi_volume *vol) { int err; - dbg_msg("free volume %d", vol->vol_id); + dbg_gen("free volume %d", vol->vol_id); ubi->volumes[vol->vol_id] = NULL; err = ubi_destroy_gluebi(vol); @@ -686,8 +685,10 @@ void ubi_free_volume(struct ubi_device *ubi, struct ubi_volume *vol) * paranoid_check_volume - check volume information. * @ubi: UBI device description object * @vol_id: volume ID + * + * Returns zero if volume is all right and a a negative error code if not. */ -static void paranoid_check_volume(struct ubi_device *ubi, int vol_id) +static int paranoid_check_volume(struct ubi_device *ubi, int vol_id) { int idx = vol_id2idx(ubi, vol_id); int reserved_pebs, alignment, data_pad, vol_type, name_len, upd_marker; @@ -705,16 +706,7 @@ static void paranoid_check_volume(struct ubi_device *ubi, int vol_id) goto fail; } spin_unlock(&ubi->volumes_lock); - return; - } - - if (vol->exclusive) { - /* - * The volume may be being created at the moment, do not check - * it (e.g., it may be in the middle of ubi_create_volume(). - */ - spin_unlock(&ubi->volumes_lock); - return; + return 0; } if (vol->reserved_pebs < 0 || vol->alignment < 0 || vol->data_pad < 0 || @@ -830,25 +822,34 @@ static void paranoid_check_volume(struct ubi_device *ubi, int vol_id) } spin_unlock(&ubi->volumes_lock); - return; + return 0; fail: ubi_err("paranoid check failed for volume %d", vol_id); - ubi_dbg_dump_vol_info(vol); - ubi_dbg_dump_vtbl_record(&ubi->vtbl[vol_id], vol_id); + if (vol) { + ubi_dbg_dump_vol_info(vol); + ubi_dbg_dump_vtbl_record(&ubi->vtbl[vol_id], vol_id); + } spin_unlock(&ubi->volumes_lock); - BUG(); + return -EINVAL; } /** * paranoid_check_volumes - check information about all volumes. * @ubi: UBI device description object + * + * Returns zero if volumes are all right and a a negative error code if not. */ -static void paranoid_check_volumes(struct ubi_device *ubi) +static int paranoid_check_volumes(struct ubi_device *ubi) { - int i; + int i, err = 0; - for (i = 0; i < ubi->vtbl_slots; i++) - paranoid_check_volume(ubi, i); + for (i = 0; i < ubi->vtbl_slots; i++) { + err = paranoid_check_volume(ubi, i); + if (err) + break; + } + + return err; } #endif diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c index d9af11a8682b..05fb72fd268f 100644 --- a/drivers/mtd/ubi/vtbl.c +++ b/drivers/mtd/ubi/vtbl.c @@ -371,7 +371,7 @@ static struct ubi_vtbl_record *process_lvol(struct ubi_device *ubi, * to LEB 0. */ - dbg_msg("check layout volume"); + dbg_gen("check layout volume"); /* Read both LEB 0 and LEB 1 into memory */ ubi_rb_for_each_entry(rb, seb, &sv->root, u.rb) { From f40ac9cdf6991287f19bdafe9b0752ee40137908 Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Sun, 13 Jul 2008 21:47:47 +0300 Subject: [PATCH 105/857] UBI: implement multiple volumes rename Quite useful ioctl which allows to make atomic system upgrades. The idea belongs to Richard Titmuss Signed-off-by: Artem Bityutskiy --- drivers/mtd/ubi/build.c | 1 + drivers/mtd/ubi/cdev.c | 188 +++++++++++++++++++++++++++++++++++++++- drivers/mtd/ubi/ubi.h | 33 ++++++- drivers/mtd/ubi/vmt.c | 57 ++++++++++-- drivers/mtd/ubi/vtbl.c | 51 +++++++++++ include/mtd/ubi-user.h | 60 ++++++++++++- 6 files changed, 375 insertions(+), 15 deletions(-) diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c index 7210e1da1fcb..4418a2369b56 100644 --- a/drivers/mtd/ubi/build.c +++ b/drivers/mtd/ubi/build.c @@ -806,6 +806,7 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, int vid_hdr_offset) mutex_init(&ubi->buf_mutex); mutex_init(&ubi->ckvol_mutex); + mutex_init(&ubi->mult_mutex); mutex_init(&ubi->volumes_mutex); spin_lock_init(&ubi->volumes_lock); diff --git a/drivers/mtd/ubi/cdev.c b/drivers/mtd/ubi/cdev.c index 7c19918cc914..bc8199c6a9f5 100644 --- a/drivers/mtd/ubi/cdev.c +++ b/drivers/mtd/ubi/cdev.c @@ -605,6 +605,166 @@ static int verify_rsvol_req(const struct ubi_device *ubi, return 0; } +/** + * rename_volumes - rename UBI volumes. + * @ubi: UBI device description object + * @req: volumes re-name request + * + * This is a helper function for the volume re-name IOCTL which validates the + * the request, opens the volume and calls corresponding volumes management + * function. Returns zero in case of success and a negative error code in case + * of failure. + */ +static int rename_volumes(struct ubi_device *ubi, + struct ubi_rnvol_req *req) +{ + int i, n, err; + struct list_head rename_list; + struct ubi_rename_entry *re, *re1; + + if (req->count < 0 || req->count > UBI_MAX_RNVOL) + return -EINVAL; + + if (req->count == 0) + return 0; + + /* Validate volume IDs and names in the request */ + for (i = 0; i < req->count; i++) { + if (req->ents[i].vol_id < 0 || + req->ents[i].vol_id >= ubi->vtbl_slots) + return -EINVAL; + if (req->ents[i].name_len < 0) + return -EINVAL; + if (req->ents[i].name_len > UBI_VOL_NAME_MAX) + return -ENAMETOOLONG; + req->ents[i].name[req->ents[i].name_len] = '\0'; + n = strlen(req->ents[i].name); + if (n != req->ents[i].name_len) + err = -EINVAL; + } + + /* Make sure volume IDs and names are unique */ + for (i = 0; i < req->count - 1; i++) { + for (n = i + 1; n < req->count; n++) { + if (req->ents[i].vol_id == req->ents[n].vol_id) { + dbg_err("duplicated volume id %d", + req->ents[i].vol_id); + return -EINVAL; + } + if (!strcmp(req->ents[i].name, req->ents[n].name)) { + dbg_err("duplicated volume name \"%s\"", + req->ents[i].name); + return -EINVAL; + } + } + } + + /* Create the re-name list */ + INIT_LIST_HEAD(&rename_list); + for (i = 0; i < req->count; i++) { + int vol_id = req->ents[i].vol_id; + int name_len = req->ents[i].name_len; + const char *name = req->ents[i].name; + + re = kzalloc(sizeof(struct ubi_rename_entry), GFP_KERNEL); + if (!re) { + err = -ENOMEM; + goto out_free; + } + + re->desc = ubi_open_volume(ubi->ubi_num, vol_id, UBI_EXCLUSIVE); + if (IS_ERR(re->desc)) { + err = PTR_ERR(re->desc); + dbg_err("cannot open volume %d, error %d", vol_id, err); + kfree(re); + goto out_free; + } + + /* Skip this re-naming if the name does not really change */ + if (re->desc->vol->name_len == name_len && + !memcmp(re->desc->vol->name, name, name_len)) { + ubi_close_volume(re->desc); + kfree(re); + continue; + } + + re->new_name_len = name_len; + memcpy(re->new_name, name, name_len); + list_add_tail(&re->list, &rename_list); + dbg_msg("will rename volume %d from \"%s\" to \"%s\"", + vol_id, re->desc->vol->name, name); + } + + if (list_empty(&rename_list)) + return 0; + + /* Find out the volumes which have to be removed */ + list_for_each_entry(re, &rename_list, list) { + struct ubi_volume_desc *desc; + int no_remove_needed = 0; + + /* + * Volume @re->vol_id is going to be re-named to + * @re->new_name, while its current name is @name. If a volume + * with name @re->new_name currently exists, it has to be + * removed, unless it is also re-named in the request (@req). + */ + list_for_each_entry(re1, &rename_list, list) { + if (re->new_name_len == re1->desc->vol->name_len && + !memcmp(re->new_name, re1->desc->vol->name, + re1->desc->vol->name_len)) { + no_remove_needed = 1; + break; + } + } + + if (no_remove_needed) + continue; + + /* + * It seems we need to remove volume with name @re->new_name, + * if it exists. + */ + desc = ubi_open_volume_nm(ubi->ubi_num, re->new_name, UBI_EXCLUSIVE); + if (IS_ERR(desc)) { + err = PTR_ERR(desc); + if (err == -ENODEV) + /* Re-naming into a non-existing volume name */ + continue; + + /* The volume exists but busy, or an error occurred */ + dbg_err("cannot open volume \"%s\", error %d", + re->new_name, err); + goto out_free; + } + + re = kzalloc(sizeof(struct ubi_rename_entry), GFP_KERNEL); + if (!re) { + err = -ENOMEM; + ubi_close_volume(desc); + goto out_free; + } + + re->remove = 1; + re->desc = desc; + list_add(&re->list, &rename_list); + dbg_msg("will remove volume %d, name \"%s\"", + re->desc->vol->vol_id, re->desc->vol->name); + } + + mutex_lock(&ubi->volumes_mutex); + err = ubi_rename_volumes(ubi, &rename_list); + mutex_unlock(&ubi->volumes_mutex); + +out_free: + list_for_each_entry_safe(re, re1, &rename_list, list) { + ubi_close_volume(re->desc); + list_del(&re->list); + kfree(re); + } + return err; +} + static int ubi_cdev_ioctl(struct inode *inode, struct file *file, unsigned int cmd, unsigned long arg) { @@ -670,7 +830,7 @@ static int ubi_cdev_ioctl(struct inode *inode, struct file *file, } mutex_lock(&ubi->volumes_mutex); - err = ubi_remove_volume(desc); + err = ubi_remove_volume(desc, 0); mutex_unlock(&ubi->volumes_mutex); /* @@ -717,6 +877,32 @@ static int ubi_cdev_ioctl(struct inode *inode, struct file *file, break; } + /* Re-name volumes command */ + case UBI_IOCRNVOL: + { + struct ubi_rnvol_req *req; + + dbg_msg("re-name volumes"); + req = kmalloc(sizeof(struct ubi_rnvol_req), GFP_KERNEL); + if (!req) { + err = -ENOMEM; + break; + }; + + err = copy_from_user(req, argp, sizeof(struct ubi_rnvol_req)); + if (err) { + err = -EFAULT; + kfree(req); + break; + } + + mutex_lock(&ubi->mult_mutex); + err = rename_volumes(ubi, req); + mutex_unlock(&ubi->mult_mutex); + kfree(req); + break; + } + default: err = -ENOTTY; break; diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h index 1fc32c863b78..274c67916b34 100644 --- a/drivers/mtd/ubi/ubi.h +++ b/drivers/mtd/ubi/ubi.h @@ -131,6 +131,27 @@ struct ubi_ltree_entry { struct rw_semaphore mutex; }; +/** + * struct ubi_rename_entry - volume re-name description data structure. + * @new_name_len: new volume name length + * @new_name: new volume name + * @remove: if not zero, this volume should be removed, not re-named + * @desc: descriptor of the volume + * @list: links re-name entries into a list + * + * This data structure is utilized in the multiple volume re-name code. Namely, + * UBI first creates a list of &struct ubi_rename_entry objects from the + * &struct ubi_rnvol_req request object, and then utilizes this list to do all + * the job. + */ +struct ubi_rename_entry { + int new_name_len; + char new_name[UBI_VOL_NAME_MAX + 1]; + int remove; + struct ubi_volume_desc *desc; + struct list_head list; +}; + struct ubi_volume_desc; /** @@ -206,7 +227,7 @@ struct ubi_volume { int alignment; int data_pad; int name_len; - char name[UBI_VOL_NAME_MAX+1]; + char name[UBI_VOL_NAME_MAX + 1]; int upd_ebs; int ch_lnum; @@ -272,7 +293,7 @@ struct ubi_wl_entry; * @vtbl_size: size of the volume table in bytes * @vtbl: in-RAM volume table copy * @volumes_mutex: protects on-flash volume table and serializes volume - * changes, like creation, deletion, update, resize + * changes, like creation, deletion, update, re-size and re-name * * @max_ec: current highest erase counter value * @mean_ec: current mean erase counter value @@ -330,6 +351,8 @@ struct ubi_wl_entry; * @peb_buf1: a buffer of PEB size used for different purposes * @peb_buf2: another buffer of PEB size used for different purposes * @buf_mutex: proptects @peb_buf1 and @peb_buf2 + * @ckvol_mutex: serializes static volume checking when opening + * @mult_mutex: serializes operations on multiple volumes, like re-nameing * @dbg_peb_buf: buffer of PEB size used for debugging * @dbg_buf_mutex: proptects @dbg_peb_buf */ @@ -410,6 +433,7 @@ struct ubi_device { void *peb_buf2; struct mutex buf_mutex; struct mutex ckvol_mutex; + struct mutex mult_mutex; #ifdef CONFIG_MTD_UBI_DEBUG void *dbg_peb_buf; struct mutex dbg_buf_mutex; @@ -426,12 +450,15 @@ extern struct mutex ubi_devices_mutex; /* vtbl.c */ int ubi_change_vtbl_record(struct ubi_device *ubi, int idx, struct ubi_vtbl_record *vtbl_rec); +int ubi_vtbl_rename_volumes(struct ubi_device *ubi, + struct list_head *rename_list); int ubi_read_volume_table(struct ubi_device *ubi, struct ubi_scan_info *si); /* vmt.c */ int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req); -int ubi_remove_volume(struct ubi_volume_desc *desc); +int ubi_remove_volume(struct ubi_volume_desc *desc, int no_vtbl); int ubi_resize_volume(struct ubi_volume_desc *desc, int reserved_pebs); +int ubi_rename_volumes(struct ubi_device *ubi, struct list_head *rename_list); int ubi_add_volume(struct ubi_device *ubi, struct ubi_volume *vol); void ubi_free_volume(struct ubi_device *ubi, struct ubi_volume *vol); diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c index 2cd886a5adac..4be4014c70df 100644 --- a/drivers/mtd/ubi/vmt.c +++ b/drivers/mtd/ubi/vmt.c @@ -402,13 +402,14 @@ int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req) /** * ubi_remove_volume - remove volume. * @desc: volume descriptor + * @no_vtbl: do not change volume table if not zero * * This function removes volume described by @desc. The volume has to be opened * in "exclusive" mode. Returns zero in case of success and a negative error * code in case of failure. The caller has to have the @ubi->volumes_mutex * locked. */ -int ubi_remove_volume(struct ubi_volume_desc *desc) +int ubi_remove_volume(struct ubi_volume_desc *desc, int no_vtbl) { struct ubi_volume *vol = desc->vol; struct ubi_device *ubi = vol->ubi; @@ -437,9 +438,11 @@ int ubi_remove_volume(struct ubi_volume_desc *desc) if (err) goto out_err; - err = ubi_change_vtbl_record(ubi, vol_id, NULL); - if (err) - goto out_err; + if (!no_vtbl) { + err = ubi_change_vtbl_record(ubi, vol_id, NULL); + if (err) + goto out_err; + } for (i = 0; i < vol->reserved_pebs; i++) { err = ubi_eba_unmap_leb(ubi, vol, i); @@ -465,7 +468,8 @@ int ubi_remove_volume(struct ubi_volume_desc *desc) ubi->vol_count -= 1; spin_unlock(&ubi->volumes_lock); - err = paranoid_check_volumes(ubi); + if (!no_vtbl) + err = paranoid_check_volumes(ubi); return err; out_err: @@ -601,6 +605,44 @@ int ubi_resize_volume(struct ubi_volume_desc *desc, int reserved_pebs) return err; } +/** + * ubi_rename_volumes - re-name UBI volumes. + * @ubi: UBI device description object + * @renam_list: list of &struct ubi_rename_entry objects + * + * This function re-names or removes volumes specified in the re-name list. + * Returns zero in case of success and a negative error code in case of + * failure. + */ +int ubi_rename_volumes(struct ubi_device *ubi, struct list_head *rename_list) +{ + int err; + struct ubi_rename_entry *re; + + err = ubi_vtbl_rename_volumes(ubi, rename_list); + if (err) + return err; + + list_for_each_entry(re, rename_list, list) { + if (re->remove) { + err = ubi_remove_volume(re->desc, 1); + if (err) + break; + } else { + struct ubi_volume *vol = re->desc->vol; + + spin_lock(&ubi->volumes_lock); + vol->name_len = re->new_name_len; + memcpy(vol->name, re->new_name, re->new_name_len + 1); + spin_unlock(&ubi->volumes_lock); + } + } + + if (!err) + paranoid_check_volumes(ubi); + return err; +} + /** * ubi_add_volume - add volume. * @ubi: UBI device description object @@ -826,10 +868,9 @@ static int paranoid_check_volume(struct ubi_device *ubi, int vol_id) fail: ubi_err("paranoid check failed for volume %d", vol_id); - if (vol) { + if (vol) ubi_dbg_dump_vol_info(vol); - ubi_dbg_dump_vtbl_record(&ubi->vtbl[vol_id], vol_id); - } + ubi_dbg_dump_vtbl_record(&ubi->vtbl[vol_id], vol_id); spin_unlock(&ubi->volumes_lock); return -EINVAL; } diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c index 05fb72fd268f..23c5376234b2 100644 --- a/drivers/mtd/ubi/vtbl.c +++ b/drivers/mtd/ubi/vtbl.c @@ -114,6 +114,57 @@ int ubi_change_vtbl_record(struct ubi_device *ubi, int idx, return 0; } +/** + * ubi_vtbl_rename_volumes - rename UBI volumes in the volume table. + * @ubi: UBI device description object + * @renam_list: list of &struct ubi_rename_entry objects + * + * This function re-names multiple volumes specified in @req in the volume + * table. Returns zero in case of success and a negative error code in case of + * failure. + */ +int ubi_vtbl_rename_volumes(struct ubi_device *ubi, + struct list_head *rename_list) +{ + int i, err; + struct ubi_rename_entry *re; + struct ubi_volume *layout_vol; + + list_for_each_entry(re, rename_list, list) { + uint32_t crc; + struct ubi_volume *vol = re->desc->vol; + struct ubi_vtbl_record *vtbl_rec = &ubi->vtbl[vol->vol_id]; + + if (re->remove) { + memcpy(vtbl_rec, &empty_vtbl_record, + sizeof(struct ubi_vtbl_record)); + continue; + } + + vtbl_rec->name_len = cpu_to_be16(re->new_name_len); + memcpy(vtbl_rec->name, re->new_name, re->new_name_len); + memset(vtbl_rec->name + re->new_name_len, 0, + UBI_VOL_NAME_MAX + 1 - re->new_name_len); + crc = crc32(UBI_CRC32_INIT, vtbl_rec, + UBI_VTBL_RECORD_SIZE_CRC); + vtbl_rec->crc = cpu_to_be32(crc); + } + + layout_vol = ubi->volumes[vol_id2idx(ubi, UBI_LAYOUT_VOLUME_ID)]; + for (i = 0; i < UBI_LAYOUT_VOLUME_EBS; i++) { + err = ubi_eba_unmap_leb(ubi, layout_vol, i); + if (err) + return err; + + err = ubi_eba_write_leb(ubi, layout_vol, i, ubi->vtbl, 0, + ubi->vtbl_size, UBI_LONGTERM); + if (err) + return err; + } + + return 0; +} + /** * vtbl_check - check if volume table is not corrupted and contains sensible * data. diff --git a/include/mtd/ubi-user.h b/include/mtd/ubi-user.h index a7421f130cc0..e8e57c3dfcdb 100644 --- a/include/mtd/ubi-user.h +++ b/include/mtd/ubi-user.h @@ -58,6 +58,13 @@ * device should be used. A &struct ubi_rsvol_req object has to be properly * filled and a pointer to it has to be passed to the IOCTL. * + * UBI volumes re-name + * ~~~~~~~~~~~~~~~~~~~ + * + * To re-name several volumes atomically at one go, the %UBI_IOCRNVOL command + * of the UBI character device should be used. A &struct ubi_rnvol_req object + * has to be properly filled and a pointer to it has to be passed to the IOCTL. + * * UBI volume update * ~~~~~~~~~~~~~~~~~ * @@ -104,6 +111,8 @@ #define UBI_IOCRMVOL _IOW(UBI_IOC_MAGIC, 1, int32_t) /* Re-size an UBI volume */ #define UBI_IOCRSVOL _IOW(UBI_IOC_MAGIC, 2, struct ubi_rsvol_req) +/* Re-name volumes */ +#define UBI_IOCRNVOL _IOW(UBI_IOC_MAGIC, 3, struct ubi_rnvol_req) /* IOCTL commands of the UBI control character device */ @@ -128,6 +137,9 @@ /* Maximum MTD device name length supported by UBI */ #define MAX_UBI_MTD_NAME_LEN 127 +/* Maximum amount of UBI volumes that can be re-named at one go */ +#define UBI_MAX_RNVOL 32 + /* * UBI data type hint constants. * @@ -189,7 +201,7 @@ struct ubi_attach_req { int32_t ubi_num; int32_t mtd_num; int32_t vid_hdr_offset; - uint8_t padding[12]; + int8_t padding[12]; }; /** @@ -250,6 +262,48 @@ struct ubi_rsvol_req { int32_t vol_id; } __attribute__ ((packed)); +/** + * struct ubi_rnvol_req - volumes re-name request. + * @count: count of volumes to re-name + * @padding1: reserved for future, not used, has to be zeroed + * @vol_id: ID of the volume to re-name + * @name_len: name length + * @padding2: reserved for future, not used, has to be zeroed + * @name: new volume name + * + * UBI allows to re-name up to %32 volumes at one go. The count of volumes to + * re-name is specified in the @count field. The ID of the volumes to re-name + * and the new names are specified in the @vol_id and @name fields. + * + * The UBI volume re-name operation is atomic, which means that should power cut + * happen, the volumes will have either old name or new name. So the possible + * use-cases of this command is atomic upgrade. Indeed, to upgrade, say, volumes + * A and B one may create temporary volumes %A1 and %B1 with the new contents, + * then atomically re-name A1->A and B1->B, in which case old %A and %B will + * be removed. + * + * If it is not desirable to remove old A and B, the re-name request has to + * contain 4 entries: A1->A, A->A1, B1->B, B->B1, in which case old A1 and B1 + * become A and B, and old A and B will become A1 and B1. + * + * It is also OK to request: A1->A, A1->X, B1->B, B->Y, in which case old A1 + * and B1 become A and B, and old A and B become X and Y. + * + * In other words, in case of re-naming into an existing volume name, the + * existing volume is removed, unless it is re-named as well at the same + * re-name request. + */ +struct ubi_rnvol_req { + int32_t count; + int8_t padding1[12]; + struct { + int32_t vol_id; + int16_t name_len; + int8_t padding2[2]; + char name[UBI_MAX_VOLUME_NAME + 1]; + } ents[UBI_MAX_RNVOL]; +} __attribute__ ((packed)); + /** * struct ubi_leb_change_req - a data structure used in atomic logical * eraseblock change requests. @@ -261,8 +315,8 @@ struct ubi_rsvol_req { struct ubi_leb_change_req { int32_t lnum; int32_t bytes; - uint8_t dtype; - uint8_t padding[7]; + int8_t dtype; + int8_t padding[7]; } __attribute__ ((packed)); #endif /* __UBI_USER_H__ */ From 8c1e6ee10bd87d70faada065a8d1f70732c17382 Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Fri, 18 Jul 2008 12:20:23 +0300 Subject: [PATCH 106/857] UBI: rework scrubbing messages If bit-flips happen often, UBI prints to many messages. Lessen the amount by only printing the messages when the PEB has been scrubbed. Also, print torturing messages. Signed-off-by: Artem Bityutskiy --- drivers/mtd/ubi/io.c | 8 +++++++- drivers/mtd/ubi/wl.c | 6 +++++- 2 files changed, 12 insertions(+), 2 deletions(-) diff --git a/drivers/mtd/ubi/io.c b/drivers/mtd/ubi/io.c index 27b9c2c2fc60..2bebb39d19b3 100644 --- a/drivers/mtd/ubi/io.c +++ b/drivers/mtd/ubi/io.c @@ -156,8 +156,12 @@ int ubi_io_read(const struct ubi_device *ubi, void *buf, int pnum, int offset, /* * -EUCLEAN is reported if there was a bit-flip which * was corrected, so this is harmless. + * + * We do not report about it here unless debugging is + * enabled. A corresponding message will be printed + * later, when it is has been scrubbed. */ - ubi_msg("fixable bit-flip detected at PEB %d", pnum); + dbg_msg("fixable bit-flip detected at PEB %d", pnum); ubi_assert(len == read); return UBI_IO_BITFLIPS; } @@ -391,6 +395,7 @@ static int torture_peb(struct ubi_device *ubi, int pnum) { int err, i, patt_count; + ubi_msg("run torture test for PEB %d", pnum); patt_count = ARRAY_SIZE(patterns); ubi_assert(patt_count > 0); @@ -434,6 +439,7 @@ static int torture_peb(struct ubi_device *ubi, int pnum) } err = patt_count; + ubi_msg("PEB %d passed torture test, do not mark it a bad", pnum); out: mutex_unlock(&ubi->buf_mutex); diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c index 761952ba125b..6821952bcdb8 100644 --- a/drivers/mtd/ubi/wl.c +++ b/drivers/mtd/ubi/wl.c @@ -873,6 +873,10 @@ static int wear_leveling_worker(struct ubi_device *ubi, struct ubi_work *wrk, } ubi_free_vid_hdr(ubi, vid_hdr); + if (scrubbing && !protect) + ubi_msg("scrubbed PEB %d, data moved to PEB %d", + e1->pnum, e2->pnum); + spin_lock(&ubi->wl_lock); if (protect) prot_tree_add(ubi, e1, pe, protect); @@ -1231,7 +1235,7 @@ int ubi_wl_scrub_peb(struct ubi_device *ubi, int pnum) { struct ubi_wl_entry *e; - ubi_msg("schedule PEB %d for scrubbing", pnum); + dbg_msg("schedule PEB %d for scrubbing", pnum); retry: spin_lock(&ubi->wl_lock); From 4d88de4beb6f327dfc7c2221eab532dad5b2bb3e Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Fri, 18 Jul 2008 12:42:14 +0300 Subject: [PATCH 107/857] UBI: bugfix - do not torture PEB needlessly This is probably a copy-paste bug - we torture the old PEB in the atomic LEB change function, but we should not do this. Signed-off-by: Artem Bityutskiy --- drivers/mtd/ubi/eba.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c index 613cd1e51648..e14208152c36 100644 --- a/drivers/mtd/ubi/eba.c +++ b/drivers/mtd/ubi/eba.c @@ -906,7 +906,7 @@ int ubi_eba_atomic_leb_change(struct ubi_device *ubi, struct ubi_volume *vol, } if (vol->eba_tbl[lnum] >= 0) { - err = ubi_wl_put_peb(ubi, vol->eba_tbl[lnum], 1); + err = ubi_wl_put_peb(ubi, vol->eba_tbl[lnum], 0); if (err) goto out_leb_unlock; } From 9c9ec147709e63e4e8ac6a037c6bb50688ff8e9c Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Fri, 18 Jul 2008 13:19:52 +0300 Subject: [PATCH 108/857] UBI: fix checkpatch.pl errors and warnings Just out or curiousity ran checkpatch.pl for whole UBI, and discovered there are quite a few of stylistic issues. Fix them. Signed-off-by: Artem Bityutskiy --- drivers/mtd/ubi/build.c | 8 ++-- drivers/mtd/ubi/cdev.c | 4 +- drivers/mtd/ubi/eba.c | 7 +-- drivers/mtd/ubi/gluebi.c | 4 +- drivers/mtd/ubi/io.c | 8 ++-- drivers/mtd/ubi/scan.c | 9 ++-- drivers/mtd/ubi/ubi.h | 3 +- drivers/mtd/ubi/upd.c | 8 ++-- drivers/mtd/ubi/vmt.c | 4 +- drivers/mtd/ubi/vtbl.c | 12 +++--- drivers/mtd/ubi/wl.c | 92 +++++++++++++++++++--------------------- include/mtd/ubi-user.h | 16 +++---- 12 files changed, 86 insertions(+), 89 deletions(-) diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c index 4418a2369b56..535d9a8a6ba8 100644 --- a/drivers/mtd/ubi/build.c +++ b/drivers/mtd/ubi/build.c @@ -51,14 +51,13 @@ * @name: MTD device name or number string * @vid_hdr_offs: VID header offset */ -struct mtd_dev_param -{ +struct mtd_dev_param { char name[MTD_PARAM_LEN_MAX]; int vid_hdr_offs; }; /* Numbers of elements set in the @mtd_dev_param array */ -static int mtd_devs = 0; +static int mtd_devs; /* MTD devices specification parameters */ static struct mtd_dev_param mtd_dev_param[UBI_MAX_DEVICES]; @@ -781,7 +780,8 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, int vid_hdr_offset) if (!ubi_devices[ubi_num]) break; if (ubi_num == UBI_MAX_DEVICES) { - dbg_err("only %d UBI devices may be created", UBI_MAX_DEVICES); + dbg_err("only %d UBI devices may be created", + UBI_MAX_DEVICES); return -ENFILE; } } else { diff --git a/drivers/mtd/ubi/cdev.c b/drivers/mtd/ubi/cdev.c index bc8199c6a9f5..03c759b4eeb5 100644 --- a/drivers/mtd/ubi/cdev.c +++ b/drivers/mtd/ubi/cdev.c @@ -39,9 +39,9 @@ #include #include #include +#include #include #include -#include #include #include "ubi.h" @@ -352,7 +352,7 @@ static ssize_t vol_cdev_direct_write(struct file *file, const char __user *buf, } #else -#define vol_cdev_direct_write(file, buf, count, offp) -EPERM +#define vol_cdev_direct_write(file, buf, count, offp) (-EPERM) #endif /* CONFIG_MTD_UBI_DEBUG_USERSPACE_IO */ static ssize_t vol_cdev_write(struct file *file, const char __user *buf, diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c index e14208152c36..e04bcf1dff87 100644 --- a/drivers/mtd/ubi/eba.c +++ b/drivers/mtd/ubi/eba.c @@ -189,9 +189,7 @@ static struct ubi_ltree_entry *ltree_add_entry(struct ubi_device *ubi, le->users += 1; spin_unlock(&ubi->ltree_lock); - if (le_free) - kfree(le_free); - + kfree(le_free); return le; } @@ -503,9 +501,8 @@ static int recover_peb(struct ubi_device *ubi, int pnum, int vol_id, int lnum, struct ubi_vid_hdr *vid_hdr; vid_hdr = ubi_zalloc_vid_hdr(ubi, GFP_NOFS); - if (!vid_hdr) { + if (!vid_hdr) return -ENOMEM; - } mutex_lock(&ubi->buf_mutex); diff --git a/drivers/mtd/ubi/gluebi.c b/drivers/mtd/ubi/gluebi.c index 49f52dceea93..605812bb0b1a 100644 --- a/drivers/mtd/ubi/gluebi.c +++ b/drivers/mtd/ubi/gluebi.c @@ -249,8 +249,8 @@ static int gluebi_erase(struct mtd_info *mtd, struct erase_info *instr) if (err) goto out_err; - instr->state = MTD_ERASE_DONE; - mtd_erase_callback(instr); + instr->state = MTD_ERASE_DONE; + mtd_erase_callback(instr); return 0; out_err: diff --git a/drivers/mtd/ubi/io.c b/drivers/mtd/ubi/io.c index 2bebb39d19b3..a84f0db0a03a 100644 --- a/drivers/mtd/ubi/io.c +++ b/drivers/mtd/ubi/io.c @@ -167,8 +167,8 @@ int ubi_io_read(const struct ubi_device *ubi, void *buf, int pnum, int offset, } if (read != len && retries++ < UBI_IO_RETRIES) { - dbg_io("error %d while reading %d bytes from PEB %d:%d, " - "read only %zd bytes, retry", + dbg_io("error %d while reading %d bytes from PEB %d:%d," + " read only %zd bytes, retry", err, len, pnum, offset, read); yield(); goto retry; @@ -705,8 +705,8 @@ int ubi_io_read_ec_hdr(struct ubi_device *ubi, int pnum, if (hdr_crc != crc) { if (verbose) { - ubi_warn("bad EC header CRC at PEB %d, calculated %#08x," - " read %#08x", pnum, crc, hdr_crc); + ubi_warn("bad EC header CRC at PEB %d, calculated " + "%#08x, read %#08x", pnum, crc, hdr_crc); ubi_dbg_dump_ec_hdr(ec_hdr); } return UBI_IO_BAD_EC_HDR; diff --git a/drivers/mtd/ubi/scan.c b/drivers/mtd/ubi/scan.c index 40eca9ce5fab..0bb7488862d2 100644 --- a/drivers/mtd/ubi/scan.c +++ b/drivers/mtd/ubi/scan.c @@ -248,7 +248,8 @@ static int compare_lebs(struct ubi_device *ubi, const struct ubi_scan_leb *seb, unsigned long long sqnum2 = be64_to_cpu(vid_hdr->sqnum); if (seb->sqnum == 0 && sqnum2 == 0) { - long long abs, v1 = seb->leb_ver, v2 = be32_to_cpu(vid_hdr->leb_ver); + long long abs; + long long v1 = seb->leb_ver, v2 = be32_to_cpu(vid_hdr->leb_ver); /* * UBI constantly increases the logical eraseblock version @@ -752,7 +753,8 @@ struct ubi_scan_leb *ubi_scan_get_free_peb(struct ubi_device *ubi, * This function returns a zero if the physical eraseblock was successfully * handled and a negative error code in case of failure. */ -static int process_eb(struct ubi_device *ubi, struct ubi_scan_info *si, int pnum) +static int process_eb(struct ubi_device *ubi, struct ubi_scan_info *si, + int pnum) { long long uninitialized_var(ec); int err, bitflips = 0, vol_id, ec_corr = 0; @@ -1301,8 +1303,7 @@ static int paranoid_check_si(struct ubi_device *ubi, struct ubi_scan_info *si) if (err < 0) { kfree(buf); return err; - } - else if (err) + } else if (err) buf[pnum] = 1; } diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h index 274c67916b34..14a5596d2d9e 100644 --- a/drivers/mtd/ubi/ubi.h +++ b/drivers/mtd/ubi/ubi.h @@ -473,7 +473,8 @@ int ubi_more_leb_change_data(struct ubi_device *ubi, struct ubi_volume *vol, const void __user *buf, int count); /* misc.c */ -int ubi_calc_data_len(const struct ubi_device *ubi, const void *buf, int length); +int ubi_calc_data_len(const struct ubi_device *ubi, const void *buf, + int length); int ubi_check_volume(struct ubi_device *ubi, int vol_id); void ubi_calculate_reserved(struct ubi_device *ubi); diff --git a/drivers/mtd/ubi/upd.c b/drivers/mtd/ubi/upd.c index 1230a5e1b530..3b8beb8545c4 100644 --- a/drivers/mtd/ubi/upd.c +++ b/drivers/mtd/ubi/upd.c @@ -39,7 +39,7 @@ */ #include -#include +#include #include #include "ubi.h" @@ -246,7 +246,8 @@ static int write_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum, return 0; } - err = ubi_eba_write_leb(ubi, vol, lnum, buf, 0, len, UBI_UNKNOWN); + err = ubi_eba_write_leb(ubi, vol, lnum, buf, 0, len, + UBI_UNKNOWN); } else { /* * When writing static volume, and this is the last logical @@ -418,7 +419,8 @@ int ubi_more_leb_change_data(struct ubi_device *ubi, struct ubi_volume *vol, if (vol->upd_received == vol->upd_bytes) { int len = ALIGN((int)vol->upd_bytes, ubi->min_io_size); - memset(vol->upd_buf + vol->upd_bytes, 0xFF, len - vol->upd_bytes); + memset(vol->upd_buf + vol->upd_bytes, 0xFF, + len - vol->upd_bytes); len = ubi_calc_data_len(ubi, vol->upd_buf, len); err = ubi_eba_atomic_leb_change(ubi, vol, vol->ch_lnum, vol->upd_buf, len, UBI_UNKNOWN); diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c index 4be4014c70df..852482d8b18f 100644 --- a/drivers/mtd/ubi/vmt.c +++ b/drivers/mtd/ubi/vmt.c @@ -253,7 +253,7 @@ int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req) goto out_unlock; } - /* Calculate how many eraseblocks are requested */ + /* Calculate how many eraseblocks are requested */ vol->usable_leb_size = ubi->leb_size - ubi->leb_size % req->alignment; bytes = req->bytes; if (do_div(bytes, vol->usable_leb_size)) @@ -858,7 +858,7 @@ static int paranoid_check_volume(struct ubi_device *ubi, int vol_id) if (alignment != vol->alignment || data_pad != vol->data_pad || upd_marker != vol->upd_marker || vol_type != vol->vol_type || - name_len!= vol->name_len || strncmp(name, vol->name, name_len)) { + name_len != vol->name_len || strncmp(name, vol->name, name_len)) { ubi_err("volume info is different"); goto fail; } diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c index 23c5376234b2..10c22257f601 100644 --- a/drivers/mtd/ubi/vtbl.c +++ b/drivers/mtd/ubi/vtbl.c @@ -461,7 +461,8 @@ static struct ubi_vtbl_record *process_lvol(struct ubi_device *ubi, if (!leb_corrupted[0]) { /* LEB 0 is OK */ if (leb[1]) - leb_corrupted[1] = memcmp(leb[0], leb[1], ubi->vtbl_size); + leb_corrupted[1] = memcmp(leb[0], leb[1], + ubi->vtbl_size); if (leb_corrupted[1]) { ubi_warn("volume table copy #2 is corrupted"); err = create_vtbl(ubi, si, 1, leb[0]); @@ -859,11 +860,10 @@ int ubi_read_volume_table(struct ubi_device *ubi, struct ubi_scan_info *si) out_free: vfree(ubi->vtbl); - for (i = 0; i < ubi->vtbl_slots + UBI_INT_VOL_COUNT; i++) - if (ubi->volumes[i]) { - kfree(ubi->volumes[i]); - ubi->volumes[i] = NULL; - } + for (i = 0; i < ubi->vtbl_slots + UBI_INT_VOL_COUNT; i++) { + kfree(ubi->volumes[i]); + ubi->volumes[i] = NULL; + } return err; } diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c index 6821952bcdb8..2a5d2a0e14ad 100644 --- a/drivers/mtd/ubi/wl.c +++ b/drivers/mtd/ubi/wl.c @@ -475,52 +475,47 @@ int ubi_wl_get_peb(struct ubi_device *ubi, int dtype) } switch (dtype) { - case UBI_LONGTERM: - /* - * For long term data we pick a physical eraseblock - * with high erase counter. But the highest erase - * counter we can pick is bounded by the the lowest - * erase counter plus %WL_FREE_MAX_DIFF. - */ - e = find_wl_entry(&ubi->free, WL_FREE_MAX_DIFF); - protect = LT_PROTECTION; - break; - case UBI_UNKNOWN: - /* - * For unknown data we pick a physical eraseblock with - * medium erase counter. But we by no means can pick a - * physical eraseblock with erase counter greater or - * equivalent than the lowest erase counter plus - * %WL_FREE_MAX_DIFF. - */ - first = rb_entry(rb_first(&ubi->free), - struct ubi_wl_entry, rb); - last = rb_entry(rb_last(&ubi->free), - struct ubi_wl_entry, rb); + case UBI_LONGTERM: + /* + * For long term data we pick a physical eraseblock with high + * erase counter. But the highest erase counter we can pick is + * bounded by the the lowest erase counter plus + * %WL_FREE_MAX_DIFF. + */ + e = find_wl_entry(&ubi->free, WL_FREE_MAX_DIFF); + protect = LT_PROTECTION; + break; + case UBI_UNKNOWN: + /* + * For unknown data we pick a physical eraseblock with medium + * erase counter. But we by no means can pick a physical + * eraseblock with erase counter greater or equivalent than the + * lowest erase counter plus %WL_FREE_MAX_DIFF. + */ + first = rb_entry(rb_first(&ubi->free), struct ubi_wl_entry, rb); + last = rb_entry(rb_last(&ubi->free), struct ubi_wl_entry, rb); - if (last->ec - first->ec < WL_FREE_MAX_DIFF) - e = rb_entry(ubi->free.rb_node, - struct ubi_wl_entry, rb); - else { - medium_ec = (first->ec + WL_FREE_MAX_DIFF)/2; - e = find_wl_entry(&ubi->free, medium_ec); - } - protect = U_PROTECTION; - break; - case UBI_SHORTTERM: - /* - * For short term data we pick a physical eraseblock - * with the lowest erase counter as we expect it will - * be erased soon. - */ - e = rb_entry(rb_first(&ubi->free), - struct ubi_wl_entry, rb); - protect = ST_PROTECTION; - break; - default: - protect = 0; - e = NULL; - BUG(); + if (last->ec - first->ec < WL_FREE_MAX_DIFF) + e = rb_entry(ubi->free.rb_node, + struct ubi_wl_entry, rb); + else { + medium_ec = (first->ec + WL_FREE_MAX_DIFF)/2; + e = find_wl_entry(&ubi->free, medium_ec); + } + protect = U_PROTECTION; + break; + case UBI_SHORTTERM: + /* + * For short term data we pick a physical eraseblock with the + * lowest erase counter as we expect it will be erased soon. + */ + e = rb_entry(rb_first(&ubi->free), struct ubi_wl_entry, rb); + protect = ST_PROTECTION; + break; + default: + protect = 0; + e = NULL; + BUG(); } /* @@ -584,7 +579,8 @@ static int prot_tree_del(struct ubi_device *ubi, int pnum) * This function returns zero in case of success and a negative error code in * case of failure. */ -static int sync_erase(struct ubi_device *ubi, struct ubi_wl_entry *e, int torture) +static int sync_erase(struct ubi_device *ubi, struct ubi_wl_entry *e, + int torture) { int err; struct ubi_ec_hdr *ec_hdr; @@ -1060,8 +1056,8 @@ static int erase_worker(struct ubi_device *ubi, struct ubi_work *wl_wrk, spin_unlock(&ubi->wl_lock); /* - * One more erase operation has happened, take care about protected - * physical eraseblocks. + * One more erase operation has happened, take care about + * protected physical eraseblocks. */ check_protection_over(ubi); diff --git a/include/mtd/ubi-user.h b/include/mtd/ubi-user.h index e8e57c3dfcdb..ccdc562e444e 100644 --- a/include/mtd/ubi-user.h +++ b/include/mtd/ubi-user.h @@ -188,14 +188,14 @@ enum { * it will be 512 in case of a 2KiB page NAND flash with 4 512-byte sub-pages. * * But in rare cases, if this optimizes things, the VID header may be placed to - * a different offset. For example, the boot-loader might do things faster if the - * VID header sits at the end of the first 2KiB NAND page with 4 sub-pages. As - * the boot-loader would not normally need to read EC headers (unless it needs - * UBI in RW mode), it might be faster to calculate ECC. This is weird example, - * but it real-life example. So, in this example, @vid_hdr_offer would be - * 2KiB-64 bytes = 1984. Note, that this position is not even 512-bytes - * aligned, which is OK, as UBI is clever enough to realize this is 4th sub-page - * of the first page and add needed padding. + * a different offset. For example, the boot-loader might do things faster if + * the VID header sits at the end of the first 2KiB NAND page with 4 sub-pages. + * As the boot-loader would not normally need to read EC headers (unless it + * needs UBI in RW mode), it might be faster to calculate ECC. This is weird + * example, but it real-life example. So, in this example, @vid_hdr_offer would + * be 2KiB-64 bytes = 1984. Note, that this position is not even 512-bytes + * aligned, which is OK, as UBI is clever enough to realize this is 4th + * sub-page of the first page and add needed padding. */ struct ubi_attach_req { int32_t ubi_num; From ebaaf1af3e9ef05c4fb7c61e4530c15e1ad10e3b Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Fri, 18 Jul 2008 13:34:32 +0300 Subject: [PATCH 109/857] UBI: fix kernel-doc errors and warnings No functional changes, just tweak comments to make kernel-doc work fine and stop complaining. Signed-off-by: Artem Bityutskiy --- drivers/mtd/ubi/build.c | 8 +++----- drivers/mtd/ubi/io.c | 6 ++---- drivers/mtd/ubi/scan.c | 18 ++++++------------ drivers/mtd/ubi/ubi.h | 1 + drivers/mtd/ubi/upd.c | 2 ++ drivers/mtd/ubi/vmt.c | 2 +- drivers/mtd/ubi/vtbl.c | 8 +++----- drivers/mtd/ubi/wl.c | 13 +++++-------- 8 files changed, 23 insertions(+), 35 deletions(-) diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c index 535d9a8a6ba8..eba760b3b8c7 100644 --- a/drivers/mtd/ubi/build.c +++ b/drivers/mtd/ubi/build.c @@ -159,8 +159,7 @@ void ubi_put_device(struct ubi_device *ubi) } /** - * ubi_get_by_major - get UBI device description object by character device - * major number. + * ubi_get_by_major - get UBI device by character device major number. * @major: major number * * This function is similar to 'ubi_get_device()', but it searches the device @@ -727,7 +726,7 @@ static int autoresize(struct ubi_device *ubi, int vol_id) /** * ubi_attach_mtd_dev - attach an MTD device. - * @mtd_dev: MTD device description object + * @mtd: MTD device description object * @ubi_num: number to assign to the new UBI device * @vid_hdr_offset: VID header offset * @@ -1095,8 +1094,7 @@ static void __exit ubi_exit(void) module_exit(ubi_exit); /** - * bytes_str_to_int - convert a string representing number of bytes to an - * integer. + * bytes_str_to_int - convert a number of bytes string into an integer. * @str: the string to convert * * This function returns positive resulting integer in case of success and a diff --git a/drivers/mtd/ubi/io.c b/drivers/mtd/ubi/io.c index a84f0db0a03a..2fb64be44f1b 100644 --- a/drivers/mtd/ubi/io.c +++ b/drivers/mtd/ubi/io.c @@ -1101,8 +1101,7 @@ static int paranoid_check_ec_hdr(const struct ubi_device *ubi, int pnum, } /** - * paranoid_check_peb_ec_hdr - check that the erase counter header of a - * physical eraseblock is in-place and is all right. + * paranoid_check_peb_ec_hdr - check erase counter header. * @ubi: UBI device description object * @pnum: the physical eraseblock number to check * @@ -1180,8 +1179,7 @@ static int paranoid_check_vid_hdr(const struct ubi_device *ubi, int pnum, } /** - * paranoid_check_peb_vid_hdr - check that the volume identifier header of a - * physical eraseblock is in-place and is all right. + * paranoid_check_peb_vid_hdr - check volume identifier header. * @ubi: UBI device description object * @pnum: the physical eraseblock number to check * diff --git a/drivers/mtd/ubi/scan.c b/drivers/mtd/ubi/scan.c index 0bb7488862d2..4dfbf27b0656 100644 --- a/drivers/mtd/ubi/scan.c +++ b/drivers/mtd/ubi/scan.c @@ -93,8 +93,7 @@ static int add_to_list(struct ubi_scan_info *si, int pnum, int ec, } /** - * validate_vid_hdr - check that volume identifier header is correct and - * consistent. + * validate_vid_hdr - check volume identifier header. * @vid_hdr: the volume identifier header to check * @sv: information about the volume this logical eraseblock belongs to * @pnum: physical eraseblock number the VID header came from @@ -380,8 +379,7 @@ static int compare_lebs(struct ubi_device *ubi, const struct ubi_scan_leb *seb, } /** - * ubi_scan_add_used - add information about a physical eraseblock to the - * scanning information. + * ubi_scan_add_used - add physical eraseblock to the scanning information. * @ubi: UBI device description object * @si: scanning information * @pnum: the physical eraseblock number @@ -555,8 +553,7 @@ int ubi_scan_add_used(struct ubi_device *ubi, struct ubi_scan_info *si, } /** - * ubi_scan_find_sv - find information about a particular volume in the - * scanning information. + * ubi_scan_find_sv - find volume in the scanning information. * @si: scanning information * @vol_id: the requested volume ID * @@ -585,8 +582,7 @@ struct ubi_scan_volume *ubi_scan_find_sv(const struct ubi_scan_info *si, } /** - * ubi_scan_find_seb - find information about a particular logical - * eraseblock in the volume scanning information. + * ubi_scan_find_seb - find LEB in the volume scanning information. * @sv: a pointer to the volume scanning information * @lnum: the requested logical eraseblock * @@ -744,8 +740,7 @@ struct ubi_scan_leb *ubi_scan_get_free_peb(struct ubi_device *ubi, } /** - * process_eb - read UBI headers, check them and add corresponding data - * to the scanning information. + * process_eb - read, check UBI headers, and add them to scanning information. * @ubi: UBI device description object * @si: scanning information * @pnum: the physical eraseblock number @@ -1083,8 +1078,7 @@ void ubi_scan_destroy_si(struct ubi_scan_info *si) #ifdef CONFIG_MTD_UBI_DEBUG_PARANOID /** - * paranoid_check_si - check if the scanning information is correct and - * consistent. + * paranoid_check_si - check the scanning information. * @ubi: UBI device description object * @si: scanning information * diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h index 14a5596d2d9e..1c3fa18c26a7 100644 --- a/drivers/mtd/ubi/ubi.h +++ b/drivers/mtd/ubi/ubi.h @@ -313,6 +313,7 @@ struct ubi_wl_entry; * @move_to, @move_to_put @erase_pending, @wl_scheduled, and @works * fields * @move_mutex: serializes eraseblock moves + * @work_sem: sycnhronizes the WL worker with use tasks * @wl_scheduled: non-zero if the wear-leveling was scheduled * @lookuptbl: a table to quickly find a &struct ubi_wl_entry object for any * physical eraseblock diff --git a/drivers/mtd/ubi/upd.c b/drivers/mtd/ubi/upd.c index 3b8beb8545c4..8b89cc18ff0b 100644 --- a/drivers/mtd/ubi/upd.c +++ b/drivers/mtd/ubi/upd.c @@ -268,6 +268,7 @@ static int write_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum, /** * ubi_more_update_data - write more update data. + * @ubi: UBI device description object * @vol: volume description object * @buf: write data (user-space memory buffer) * @count: how much bytes to write @@ -385,6 +386,7 @@ int ubi_more_update_data(struct ubi_device *ubi, struct ubi_volume *vol, /** * ubi_more_leb_change_data - accept more data for atomic LEB change. + * @ubi: UBI device description object * @vol: volume description object * @buf: write data (user-space memory buffer) * @count: how much bytes to write diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c index 852482d8b18f..d40066833abb 100644 --- a/drivers/mtd/ubi/vmt.c +++ b/drivers/mtd/ubi/vmt.c @@ -608,7 +608,7 @@ int ubi_resize_volume(struct ubi_volume_desc *desc, int reserved_pebs) /** * ubi_rename_volumes - re-name UBI volumes. * @ubi: UBI device description object - * @renam_list: list of &struct ubi_rename_entry objects + * @rename_list: list of &struct ubi_rename_entry objects * * This function re-names or removes volumes specified in the re-name list. * Returns zero in case of success and a negative error code in case of diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c index 10c22257f601..4e1c489a3bae 100644 --- a/drivers/mtd/ubi/vtbl.c +++ b/drivers/mtd/ubi/vtbl.c @@ -117,7 +117,7 @@ int ubi_change_vtbl_record(struct ubi_device *ubi, int idx, /** * ubi_vtbl_rename_volumes - rename UBI volumes in the volume table. * @ubi: UBI device description object - * @renam_list: list of &struct ubi_rename_entry objects + * @rename_list: list of &struct ubi_rename_entry objects * * This function re-names multiple volumes specified in @req in the volume * table. Returns zero in case of success and a negative error code in case of @@ -166,8 +166,7 @@ int ubi_vtbl_rename_volumes(struct ubi_device *ubi, } /** - * vtbl_check - check if volume table is not corrupted and contains sensible - * data. + * vtbl_check - check if volume table is not corrupted and sensible. * @ubi: UBI device description object * @vtbl: volume table * @@ -780,8 +779,7 @@ static int check_scanning_info(const struct ubi_device *ubi, } /** - * ubi_read_volume_table - read volume table. - * information. + * ubi_read_volume_table - read the volume table. * @ubi: UBI device description object * @si: scanning information * diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c index 2a5d2a0e14ad..05d70937b543 100644 --- a/drivers/mtd/ubi/wl.c +++ b/drivers/mtd/ubi/wl.c @@ -632,8 +632,7 @@ static int sync_erase(struct ubi_device *ubi, struct ubi_wl_entry *e, } /** - * check_protection_over - check if it is time to stop protecting some - * physical eraseblocks. + * check_protection_over - check if it is time to stop protecting some PEBs. * @ubi: UBI device description object * * This function is called after each erase operation, when the absolute erase @@ -1601,8 +1600,7 @@ void ubi_wl_close(struct ubi_device *ubi) #ifdef CONFIG_MTD_UBI_DEBUG_PARANOID /** - * paranoid_check_ec - make sure that the erase counter of a physical eraseblock - * is correct. + * paranoid_check_ec - make sure that the erase counter of a PEB is correct. * @ubi: UBI device description object * @pnum: the physical eraseblock number to check * @ec: the erase counter to check @@ -1643,13 +1641,12 @@ static int paranoid_check_ec(struct ubi_device *ubi, int pnum, int ec) } /** - * paranoid_check_in_wl_tree - make sure that a wear-leveling entry is present - * in a WL RB-tree. + * paranoid_check_in_wl_tree - check that wear-leveling entry is in WL RB-tree. * @e: the wear-leveling entry to check * @root: the root of the tree * - * This function returns zero if @e is in the @root RB-tree and %1 if it - * is not. + * This function returns zero if @e is in the @root RB-tree and %1 if it is + * not. */ static int paranoid_check_in_wl_tree(struct ubi_wl_entry *e, struct rb_root *root) From 9869cd801c107bbae91663c3f4edbb6b5715919f Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Fri, 18 Jul 2008 13:53:39 +0300 Subject: [PATCH 110/857] UBI: remove pre-sqnum images support Before UBI got into mainline, there was a slight flash format change - we did not have sequence number support, then added it. We have carried full support of those ancient images till this moment. Now the support is removed, well, not fully removed. Now UBI will support only _clean_ old images, which were cleanly detached last time (just before kernel upgrade). This is most likely the case. But we will not support unclean ancient images. Surprisingly, this allows us to remove a big chunk of legacy code. And the same should be true for downgrading: clean images should downgrade fine, but unclean ones will not. Signed-off-by: Artem Bityutskiy --- drivers/mtd/ubi/debug.c | 2 - drivers/mtd/ubi/scan.c | 87 ++++++++++--------------------------- drivers/mtd/ubi/scan.h | 2 - drivers/mtd/ubi/ubi-media.h | 17 +++----- drivers/mtd/ubi/vtbl.c | 1 - 5 files changed, 30 insertions(+), 79 deletions(-) diff --git a/drivers/mtd/ubi/debug.c b/drivers/mtd/ubi/debug.c index 21e0d7d76a44..c0ed60e8ade9 100644 --- a/drivers/mtd/ubi/debug.c +++ b/drivers/mtd/ubi/debug.c @@ -65,7 +65,6 @@ void ubi_dbg_dump_vid_hdr(const struct ubi_vid_hdr *vid_hdr) printk(KERN_DEBUG "\tcompat %d\n", (int)vid_hdr->compat); printk(KERN_DEBUG "\tvol_id %d\n", be32_to_cpu(vid_hdr->vol_id)); printk(KERN_DEBUG "\tlnum %d\n", be32_to_cpu(vid_hdr->lnum)); - printk(KERN_DEBUG "\tleb_ver %u\n", be32_to_cpu(vid_hdr->leb_ver)); printk(KERN_DEBUG "\tdata_size %d\n", be32_to_cpu(vid_hdr->data_size)); printk(KERN_DEBUG "\tused_ebs %d\n", be32_to_cpu(vid_hdr->used_ebs)); printk(KERN_DEBUG "\tdata_pad %d\n", be32_to_cpu(vid_hdr->data_pad)); @@ -172,7 +171,6 @@ void ubi_dbg_dump_seb(const struct ubi_scan_leb *seb, int type) printk(KERN_DEBUG "\tlnum %d\n", seb->lnum); printk(KERN_DEBUG "\tscrub %d\n", seb->scrub); printk(KERN_DEBUG "\tsqnum %llu\n", seb->sqnum); - printk(KERN_DEBUG "\tleb_ver %u\n", seb->leb_ver); } } diff --git a/drivers/mtd/ubi/scan.c b/drivers/mtd/ubi/scan.c index 4dfbf27b0656..967bb4406df9 100644 --- a/drivers/mtd/ubi/scan.c +++ b/drivers/mtd/ubi/scan.c @@ -246,46 +246,21 @@ static int compare_lebs(struct ubi_device *ubi, const struct ubi_scan_leb *seb, struct ubi_vid_hdr *vh = NULL; unsigned long long sqnum2 = be64_to_cpu(vid_hdr->sqnum); - if (seb->sqnum == 0 && sqnum2 == 0) { - long long abs; - long long v1 = seb->leb_ver, v2 = be32_to_cpu(vid_hdr->leb_ver); - + if (sqnum2 == seb->sqnum) { /* - * UBI constantly increases the logical eraseblock version - * number and it can overflow. Thus, we have to bear in mind - * that versions that are close to %0xFFFFFFFF are less then - * versions that are close to %0. - * - * The UBI WL sub-system guarantees that the number of pending - * tasks is not greater then %0x7FFFFFFF. So, if the difference - * between any two versions is greater or equivalent to - * %0x7FFFFFFF, there was an overflow and the logical - * eraseblock with lower version is actually newer then the one - * with higher version. - * - * FIXME: but this is anyway obsolete and will be removed at - * some point. + * This must be a really ancient UBI image which has been + * created before sequence numbers support has been added. At + * that times we used 32-bit LEB versions stored in logical + * eraseblocks. That was before UBI got into mainline. We do not + * support these images anymore. Well, those images will work + * still work, but only if no unclean reboots happened. */ - dbg_bld("using old crappy leb_ver stuff"); + ubi_err("unsupported on-flash UBI format\n"); + return -EINVAL; + } - if (v1 == v2) { - ubi_err("PEB %d and PEB %d have the same version %lld", - seb->pnum, pnum, v1); - return -EINVAL; - } - - abs = v1 - v2; - if (abs < 0) - abs = -abs; - - if (abs < 0x7FFFFFFF) - /* Non-overflow situation */ - second_is_newer = (v2 > v1); - else - second_is_newer = (v2 < v1); - } else - /* Obviously the LEB with lower sequence counter is older */ - second_is_newer = sqnum2 > seb->sqnum; + /* Obviously the LEB with lower sequence counter is older */ + second_is_newer = !!(sqnum2 > seb->sqnum); /* * Now we know which copy is newer. If the copy flag of the PEB with @@ -293,7 +268,7 @@ static int compare_lebs(struct ubi_device *ubi, const struct ubi_scan_leb *seb, * check data CRC. For the second PEB we already have the VID header, * for the first one - we'll need to re-read it from flash. * - * FIXME: this may be optimized so that we wouldn't read twice. + * Note: this may be optimized so that we wouldn't read twice. */ if (second_is_newer) { @@ -399,7 +374,6 @@ int ubi_scan_add_used(struct ubi_device *ubi, struct ubi_scan_info *si, int bitflips) { int err, vol_id, lnum; - uint32_t leb_ver; unsigned long long sqnum; struct ubi_scan_volume *sv; struct ubi_scan_leb *seb; @@ -408,10 +382,9 @@ int ubi_scan_add_used(struct ubi_device *ubi, struct ubi_scan_info *si, vol_id = be32_to_cpu(vid_hdr->vol_id); lnum = be32_to_cpu(vid_hdr->lnum); sqnum = be64_to_cpu(vid_hdr->sqnum); - leb_ver = be32_to_cpu(vid_hdr->leb_ver); - dbg_bld("PEB %d, LEB %d:%d, EC %d, sqnum %llu, ver %u, bitflips %d", - pnum, vol_id, lnum, ec, sqnum, leb_ver, bitflips); + dbg_bld("PEB %d, LEB %d:%d, EC %d, sqnum %llu, bitflips %d", + pnum, vol_id, lnum, ec, sqnum, bitflips); sv = add_volume(si, vol_id, pnum, vid_hdr); if (IS_ERR(sv) < 0) @@ -444,25 +417,20 @@ int ubi_scan_add_used(struct ubi_device *ubi, struct ubi_scan_info *si, */ dbg_bld("this LEB already exists: PEB %d, sqnum %llu, " - "LEB ver %u, EC %d", seb->pnum, seb->sqnum, - seb->leb_ver, seb->ec); - - /* - * Make sure that the logical eraseblocks have different - * versions. Otherwise the image is bad. - */ - if (seb->leb_ver == leb_ver && leb_ver != 0) { - ubi_err("two LEBs with same version %u", leb_ver); - ubi_dbg_dump_seb(seb, 0); - ubi_dbg_dump_vid_hdr(vid_hdr); - return -EINVAL; - } + "EC %d", seb->pnum, seb->sqnum, seb->ec); /* * Make sure that the logical eraseblocks have different * sequence numbers. Otherwise the image is bad. * - * FIXME: remove 'sqnum != 0' check when leb_ver is removed. + * However, if the sequence number is zero, we assume it must + * be an ancient UBI image from the era when UBI did not have + * sequence numbers. We still can attach these images, unless + * there is a need to distinguish between old and new + * eraseblocks, in which case we'll refuse the image in + * 'compare_lebs()'. In other words, we attach old clean + * images, but refuse attaching old images with duplicated + * logical eraseblocks because there was an unclean reboot. */ if (seb->sqnum == sqnum && sqnum != 0) { ubi_err("two LEBs with same sequence number %llu", @@ -502,7 +470,6 @@ int ubi_scan_add_used(struct ubi_device *ubi, struct ubi_scan_info *si, seb->pnum = pnum; seb->scrub = ((cmp_res & 2) || bitflips); seb->sqnum = sqnum; - seb->leb_ver = leb_ver; if (sv->highest_lnum == lnum) sv->last_data_size = @@ -539,7 +506,6 @@ int ubi_scan_add_used(struct ubi_device *ubi, struct ubi_scan_info *si, seb->lnum = lnum; seb->sqnum = sqnum; seb->scrub = bitflips; - seb->leb_ver = leb_ver; if (sv->highest_lnum <= lnum) { sv->highest_lnum = lnum; @@ -1263,11 +1229,6 @@ static int paranoid_check_si(struct ubi_device *ubi, struct ubi_scan_info *si) ubi_err("bad data_pad %d", sv->data_pad); goto bad_vid_hdr; } - - if (seb->leb_ver != be32_to_cpu(vidh->leb_ver)) { - ubi_err("bad leb_ver %u", seb->leb_ver); - goto bad_vid_hdr; - } } if (!last_seb) diff --git a/drivers/mtd/ubi/scan.h b/drivers/mtd/ubi/scan.h index 4e2e3cc0becd..61df208e2f20 100644 --- a/drivers/mtd/ubi/scan.h +++ b/drivers/mtd/ubi/scan.h @@ -34,7 +34,6 @@ * @u: unions RB-tree or @list links * @u.rb: link in the per-volume RB-tree of &struct ubi_scan_leb objects * @u.list: link in one of the eraseblock lists - * @leb_ver: logical eraseblock version (obsolete) * * One object of this type is allocated for each physical eraseblock during * scanning. @@ -49,7 +48,6 @@ struct ubi_scan_leb { struct rb_node rb; struct list_head list; } u; - uint32_t leb_ver; }; /** diff --git a/drivers/mtd/ubi/ubi-media.h b/drivers/mtd/ubi/ubi-media.h index 26bb7af9787a..2ad940409053 100644 --- a/drivers/mtd/ubi/ubi-media.h +++ b/drivers/mtd/ubi/ubi-media.h @@ -168,16 +168,15 @@ struct ubi_ec_hdr { * %UBI_COMPAT_IGNORE, %UBI_COMPAT_PRESERVE, or %UBI_COMPAT_REJECT) * @vol_id: ID of this volume * @lnum: logical eraseblock number - * @leb_ver: version of this logical eraseblock (IMPORTANT: obsolete, to be - * removed, kept only for not breaking older UBI users) + * @padding1: reserved for future, zeroes * @data_size: how many bytes of data this logical eraseblock contains * @used_ebs: total number of used logical eraseblocks in this volume * @data_pad: how many bytes at the end of this physical eraseblock are not * used * @data_crc: CRC checksum of the data stored in this logical eraseblock - * @padding1: reserved for future, zeroes - * @sqnum: sequence number * @padding2: reserved for future, zeroes + * @sqnum: sequence number + * @padding3: reserved for future, zeroes * @hdr_crc: volume identifier header CRC checksum * * The @sqnum is the value of the global sequence counter at the time when this @@ -225,10 +224,6 @@ struct ubi_ec_hdr { * checksum is correct, this physical eraseblock is selected (P1). Otherwise * the older one (P) is selected. * - * Note, there is an obsolete @leb_ver field which was used instead of @sqnum - * in the past. But it is not used anymore and we keep it in order to be able - * to deal with old UBI images. It will be removed at some point. - * * There are 2 sorts of volumes in UBI: user volumes and internal volumes. * Internal volumes are not seen from outside and are used for various internal * UBI purposes. In this implementation there is only one internal volume - the @@ -278,14 +273,14 @@ struct ubi_vid_hdr { __u8 compat; __be32 vol_id; __be32 lnum; - __be32 leb_ver; /* obsolete, to be removed, don't use */ + __u8 padding1[4]; __be32 data_size; __be32 used_ebs; __be32 data_pad; __be32 data_crc; - __u8 padding1[4]; + __u8 padding2[4]; __be64 sqnum; - __u8 padding2[12]; + __u8 padding3[12]; __be32 hdr_crc; } __attribute__ ((packed)); diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c index 4e1c489a3bae..217d0e111b2a 100644 --- a/drivers/mtd/ubi/vtbl.c +++ b/drivers/mtd/ubi/vtbl.c @@ -338,7 +338,6 @@ static int create_vtbl(struct ubi_device *ubi, struct ubi_scan_info *si, vid_hdr->data_pad = cpu_to_be32(0); vid_hdr->lnum = cpu_to_be32(copy); vid_hdr->sqnum = cpu_to_be64(++si->max_sqnum); - vid_hdr->leb_ver = cpu_to_be32(old_seb ? old_seb->leb_ver + 1: 0); /* The EC header is already there, write the VID header */ err = ubi_io_write_vid_hdr(ubi, new_seb->pnum, vid_hdr); From eeb16e87b6747c9a4f5769f33467c9d173e9f5ee Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Wed, 23 Jul 2008 15:51:46 +0300 Subject: [PATCH 111/857] UBI: fix gcc warning Fix the following warning: drivers/mtd/ubi/vmt.c: In function 'ubi_rename_volumes': drivers/mtd/ubi/vmt.c:642: warning: statement with no effect Signed-off-by: Artem Bityutskiy --- drivers/mtd/ubi/vmt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c index d40066833abb..3531ca9a1e24 100644 --- a/drivers/mtd/ubi/vmt.c +++ b/drivers/mtd/ubi/vmt.c @@ -639,7 +639,7 @@ int ubi_rename_volumes(struct ubi_device *ubi, struct list_head *rename_list) } if (!err) - paranoid_check_volumes(ubi); + err = paranoid_check_volumes(ubi); return err; } From 9d25d4db81833029d30b7b03cc1000cbbe09e192 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Mon, 21 Jul 2008 18:41:26 +0100 Subject: [PATCH 112/857] x86: BUILD_IRQ say .text to avoid .data.percpu When I edit the x86_64 Makefile to -fno-unit-at-a-time, bootup panics on 0xCCs in IRQ0x3e_interrupt(): IRQ0x20_interrupt etc. have got linked into .data.percpu. Perhaps there are other ways of triggering that: specify ".text" in the BUILD_IRQ() macro for safety. I've been using -fno-unit-at-a-time (to lessen inlining, for easier debugging) for a long time. Signed-off-by: Hugh Dickins Cc: Mike Travis Signed-off-by: Ingo Molnar --- arch/x86/kernel/irqinit_64.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/kernel/irqinit_64.c b/arch/x86/kernel/irqinit_64.c index 0373e88de95a..9414125f19ce 100644 --- a/arch/x86/kernel/irqinit_64.c +++ b/arch/x86/kernel/irqinit_64.c @@ -43,7 +43,7 @@ #define BUILD_IRQ(nr) \ asmlinkage void IRQ_NAME(nr); \ - asm("\n.p2align\n" \ + asm("\n.text\n.p2align\n" \ "IRQ" #nr "_interrupt:\n\t" \ "push $~(" #nr ") ; " \ "jmp common_interrupt"); From 04bbe430f73c6c31bbd067349c029e907e153a8d Mon Sep 17 00:00:00 2001 From: Vegard Nossum Date: Mon, 21 Jul 2008 15:06:35 +0200 Subject: [PATCH 113/857] x86: fix header export, asm-x86/processor-flags.h, CONFIG_* leaks Apparently, commit 6330a30a76c1e62d4b4ec238368957f8febf9113 Author: Vegard Nossum Date: Wed May 28 09:46:19 2008 +0200 x86: break mutual header inclusion introduced some CONFIG names to processor-flags.h, which was exported in commit 6093015db2bd9e70cf20cdd23be1a50733baafdd Author: Ingo Molnar Date: Sun Mar 30 11:45:23 2008 +0200 x86: cleanup replace most vm86 flags with flags from processor-flags.h, fix Fix it by wrapping the CONFIG parts in __KERNEL__. Reported-by: Olaf Hering Signed-off-by: Vegard Nossum Cc: Adrian Bunk Cc: Olaf Hering Signed-off-by: Ingo Molnar --- include/asm-x86/processor-flags.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/include/asm-x86/processor-flags.h b/include/asm-x86/processor-flags.h index 092b39b3a7e6..eff2ecd7fff0 100644 --- a/include/asm-x86/processor-flags.h +++ b/include/asm-x86/processor-flags.h @@ -88,10 +88,12 @@ #define CX86_ARR_BASE 0xc4 #define CX86_RCR_BASE 0xdc +#ifdef __KERNEL__ #ifdef CONFIG_VM86 #define X86_VM_MASK X86_EFLAGS_VM #else #define X86_VM_MASK 0 /* No VM86 support */ #endif +#endif #endif /* __ASM_I386_PROCESSOR_FLAGS_H */ From 58838cf3ca3337d76141c33d6c68376490263468 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 24 Jul 2008 12:43:13 +0200 Subject: [PATCH 114/857] sched: clean up compiler warning Reported-by: Daniel Walker Signed-off-by: Peter Zijlstra Signed-off-by: Ingo Molnar --- kernel/sched_rt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index 147004c651c0..93ac8ee08271 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -253,7 +253,7 @@ static int do_balance_runtime(struct rt_rq *rt_rq) diff = iter->rt_runtime - iter->rt_time; if (diff > 0) { - do_div(diff, weight); + diff = div_u64((u64)diff, weight); if (rt_rq->rt_runtime + diff > rt_period) diff = rt_period - rt_rq->rt_runtime; iter->rt_runtime -= diff; From 78305de2f99e9f43ab860dd95bb430b20e26c695 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Wed, 23 Apr 2008 07:20:41 -0400 Subject: [PATCH 115/857] Remove mention of semaphores from kernel-locking Since the consensus seems to be to eliminate semaphores where possible, we shouldn't be educating people about how to use them as locks. Use mutexes instead. Semaphores should be described in a separate document if we end up keeping them. Signed-off-by: Matthew Wilcox Acked-by: Rusty Russell --- Documentation/DocBook/kernel-locking.tmpl | 57 ++++++++++------------- 1 file changed, 24 insertions(+), 33 deletions(-) diff --git a/Documentation/DocBook/kernel-locking.tmpl b/Documentation/DocBook/kernel-locking.tmpl index 2510763295d0..084f6ad7b7a0 100644 --- a/Documentation/DocBook/kernel-locking.tmpl +++ b/Documentation/DocBook/kernel-locking.tmpl @@ -219,10 +219,10 @@ - Three Main Types of Kernel Locks: Spinlocks, Mutexes and Semaphores + Two Main Types of Kernel Locks: Spinlocks and Mutexes - There are three main types of kernel locks. The fundamental type + There are two main types of kernel locks. The fundamental type is the spinlock (include/asm/spinlock.h), which is a very simple single-holder lock: if you can't get the @@ -239,14 +239,6 @@ can't sleep (see ), and so have to use a spinlock instead. - - The third type is a semaphore - (include/linux/semaphore.h): it - can have more than one holder at any time (the number decided at - initialization time), although it is most commonly used as a - single-holder lock (a mutex). If you can't get a semaphore, your - task will be suspended and later on woken up - just like for mutexes. - Neither type of lock is recursive: see . @@ -278,7 +270,7 @@ - Semaphores still exist, because they are required for + Mutexes still exist, because they are required for synchronization between user contexts, as we will see below. @@ -289,18 +281,17 @@ If you have a data structure which is only ever accessed from - user context, then you can use a simple semaphore - (linux/linux/semaphore.h) to protect it. This - is the most trivial case: you initialize the semaphore to the number - of resources available (usually 1), and call - down_interruptible() to grab the semaphore, and - up() to release it. There is also a - down(), which should be avoided, because it + user context, then you can use a simple mutex + (include/linux/mutex.h) to protect it. This + is the most trivial case: you initialize the mutex. Then you can + call mutex_lock_interruptible() to grab the mutex, + and mutex_unlock() to release it. There is also a + mutex_lock(), which should be avoided, because it will not return if a signal is received. - Example: linux/net/core/netfilter.c allows + Example: net/netfilter/nf_sockopt.c allows registration of new setsockopt() and getsockopt() calls, with nf_register_sockopt(). Registration and @@ -515,7 +506,7 @@ If you are in a process context (any syscall) and want to - lock other process out, use a semaphore. You can take a semaphore + lock other process out, use a mutex. You can take a mutex and sleep (copy_from_user*( or kmalloc(x,GFP_KERNEL)). @@ -662,7 +653,7 @@ SLBH SLBH SLBH -DI +MLI None @@ -692,8 +683,8 @@ spin_lock_bh -DI -down_interruptible +MLI +mutex_lock_interruptible @@ -1310,7 +1301,7 @@ as Alan Cox says, Lock data, not code. There is a coding bug where a piece of code tries to grab a spinlock twice: it will spin forever, waiting for the lock to - be released (spinlocks, rwlocks and semaphores are not + be released (spinlocks, rwlocks and mutexes are not recursive in Linux). This is trivial to diagnose: not a stay-up-five-nights-talk-to-fluffy-code-bunnies kind of problem. @@ -1335,7 +1326,7 @@ as Alan Cox says, Lock data, not code. This complete lockup is easy to diagnose: on SMP boxes the - watchdog timer or compiling with DEBUG_SPINLOCKS set + watchdog timer or compiling with DEBUG_SPINLOCK set (include/linux/spinlock.h) will show this up immediately when it happens. @@ -1558,7 +1549,7 @@ the amount of locking which needs to be done. Read/Write Lock Variants - Both spinlocks and semaphores have read/write variants: + Both spinlocks and mutexes have read/write variants: rwlock_t and struct rw_semaphore. These divide users into two classes: the readers and the writers. If you are only reading the data, you can get a read lock, but to write to @@ -1681,7 +1672,7 @@ the amount of locking which needs to be done. #include <linux/slab.h> #include <linux/string.h> +#include <linux/rcupdate.h> - #include <linux/semaphore.h> + #include <linux/mutex.h> #include <asm/errno.h> struct object @@ -1913,7 +1904,7 @@ machines due to caching. - put_user() + put_user() @@ -1927,13 +1918,13 @@ machines due to caching. - down_interruptible() and - down() + mutex_lock_interruptible() and + mutex_lock() - There is a down_trylock() which can be + There is a mutex_trylock() which can be used inside interrupt context, as it will not sleep. - up() will also never sleep. + mutex_unlock() will also never sleep. @@ -2023,7 +2014,7 @@ machines due to caching. Prior to 2.5, or when CONFIG_PREEMPT is unset, processes in user context inside the kernel would not - preempt each other (ie. you had that CPU until you have it up, + preempt each other (ie. you had that CPU until you gave it up, except for interrupts). With the addition of CONFIG_PREEMPT in 2.5.4, this changed: when in user context, higher priority tasks can "cut in": spinlocks From 0f17e4c796e89d1f69f13b653aba60e6ccfb8ae0 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Thu, 24 Jul 2008 08:30:48 -0400 Subject: [PATCH 116/857] Add missing semaphore.h includes These files use semaphores but don't include semaphore.h Signed-off-by: Matthew Wilcox Acked-by: Geert Uytterhoeven --- drivers/input/keyboard/hil_kbd.c | 1 + drivers/input/misc/hp_sdc_rtc.c | 1 + drivers/input/serio/hp_sdc.c | 1 + 3 files changed, 3 insertions(+) diff --git a/drivers/input/keyboard/hil_kbd.c b/drivers/input/keyboard/hil_kbd.c index adbf29f0169d..71c1971abf80 100644 --- a/drivers/input/keyboard/hil_kbd.c +++ b/drivers/input/keyboard/hil_kbd.c @@ -37,6 +37,7 @@ #include #include #include +#include #include #include diff --git a/drivers/input/misc/hp_sdc_rtc.c b/drivers/input/misc/hp_sdc_rtc.c index 49d8abfe38fe..daa9d4220331 100644 --- a/drivers/input/misc/hp_sdc_rtc.c +++ b/drivers/input/misc/hp_sdc_rtc.c @@ -44,6 +44,7 @@ #include #include #include +#include MODULE_AUTHOR("Brian S. Julin "); MODULE_DESCRIPTION("HP i8042 SDC + MSM-58321 RTC Driver"); diff --git a/drivers/input/serio/hp_sdc.c b/drivers/input/serio/hp_sdc.c index 7b233a492ad5..aad664d5259f 100644 --- a/drivers/input/serio/hp_sdc.c +++ b/drivers/input/serio/hp_sdc.c @@ -67,6 +67,7 @@ #include #include #include +#include #include #include #include From 6310e472717ed736c9bff9840febb71f7bb400ed Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Thu, 24 Jul 2008 08:08:09 -0400 Subject: [PATCH 117/857] Remove use of asm/semaphore.h Change to use linux/semaphore.h Signed-off-by: Matthew Wilcox --- arch/arm/mach-ns9xxx/clock.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/mach-ns9xxx/clock.c b/arch/arm/mach-ns9xxx/clock.c index f8639161068f..44ed20d4a388 100644 --- a/arch/arm/mach-ns9xxx/clock.c +++ b/arch/arm/mach-ns9xxx/clock.c @@ -14,8 +14,8 @@ #include #include #include +#include -#include #include "clock.h" static LIST_HEAD(clocks); From 2351ec533ed0dd56052ab96988d2161d5ecc8ed9 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Thu, 24 Jul 2008 08:09:32 -0400 Subject: [PATCH 118/857] Remove asm/semaphore.h All users have now been converted to linux/semaphore.h and we don't need to keep these files around any longer. Signed-off-by: Matthew Wilcox --- Documentation/feature-removal-schedule.txt | 8 -------- include/asm-alpha/semaphore.h | 1 - include/asm-arm/semaphore.h | 1 - include/asm-avr32/semaphore.h | 1 - include/asm-blackfin/semaphore.h | 1 - include/asm-cris/semaphore.h | 1 - include/asm-frv/semaphore.h | 1 - include/asm-h8300/semaphore.h | 1 - include/asm-ia64/semaphore.h | 1 - include/asm-m32r/semaphore.h | 1 - include/asm-m68k/semaphore.h | 1 - include/asm-m68knommu/semaphore.h | 1 - include/asm-mips/semaphore.h | 1 - include/asm-mn10300/semaphore.h | 1 - include/asm-parisc/semaphore.h | 1 - include/asm-powerpc/semaphore.h | 1 - include/asm-s390/semaphore.h | 1 - include/asm-sh/semaphore.h | 1 - include/asm-sparc/semaphore.h | 1 - include/asm-sparc64/semaphore.h | 1 - include/asm-um/semaphore.h | 1 - include/asm-v850/semaphore.h | 1 - include/asm-x86/semaphore.h | 1 - include/asm-xtensa/semaphore.h | 1 - 24 files changed, 31 deletions(-) delete mode 100644 include/asm-alpha/semaphore.h delete mode 100644 include/asm-arm/semaphore.h delete mode 100644 include/asm-avr32/semaphore.h delete mode 100644 include/asm-blackfin/semaphore.h delete mode 100644 include/asm-cris/semaphore.h delete mode 100644 include/asm-frv/semaphore.h delete mode 100644 include/asm-h8300/semaphore.h delete mode 100644 include/asm-ia64/semaphore.h delete mode 100644 include/asm-m32r/semaphore.h delete mode 100644 include/asm-m68k/semaphore.h delete mode 100644 include/asm-m68knommu/semaphore.h delete mode 100644 include/asm-mips/semaphore.h delete mode 100644 include/asm-mn10300/semaphore.h delete mode 100644 include/asm-parisc/semaphore.h delete mode 100644 include/asm-powerpc/semaphore.h delete mode 100644 include/asm-s390/semaphore.h delete mode 100644 include/asm-sh/semaphore.h delete mode 100644 include/asm-sparc/semaphore.h delete mode 100644 include/asm-sparc64/semaphore.h delete mode 100644 include/asm-um/semaphore.h delete mode 100644 include/asm-v850/semaphore.h delete mode 100644 include/asm-x86/semaphore.h delete mode 100644 include/asm-xtensa/semaphore.h diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index 9f73587219e8..09c4a1efb8e3 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -300,14 +300,6 @@ Who: ocfs2-devel@oss.oracle.com --------------------------- -What: asm/semaphore.h -When: 2.6.26 -Why: Implementation became generic; users should now include - linux/semaphore.h instead. -Who: Matthew Wilcox - ---------------------------- - What: SCTP_GET_PEER_ADDRS_NUM_OLD, SCTP_GET_PEER_ADDRS_OLD, SCTP_GET_LOCAL_ADDRS_NUM_OLD, SCTP_GET_LOCAL_ADDRS_OLD When: June 2009 diff --git a/include/asm-alpha/semaphore.h b/include/asm-alpha/semaphore.h deleted file mode 100644 index d9b2034ed1d2..000000000000 --- a/include/asm-alpha/semaphore.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-arm/semaphore.h b/include/asm-arm/semaphore.h deleted file mode 100644 index d9b2034ed1d2..000000000000 --- a/include/asm-arm/semaphore.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-avr32/semaphore.h b/include/asm-avr32/semaphore.h deleted file mode 100644 index d9b2034ed1d2..000000000000 --- a/include/asm-avr32/semaphore.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-blackfin/semaphore.h b/include/asm-blackfin/semaphore.h deleted file mode 100644 index d9b2034ed1d2..000000000000 --- a/include/asm-blackfin/semaphore.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-cris/semaphore.h b/include/asm-cris/semaphore.h deleted file mode 100644 index d9b2034ed1d2..000000000000 --- a/include/asm-cris/semaphore.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-frv/semaphore.h b/include/asm-frv/semaphore.h deleted file mode 100644 index d9b2034ed1d2..000000000000 --- a/include/asm-frv/semaphore.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-h8300/semaphore.h b/include/asm-h8300/semaphore.h deleted file mode 100644 index d9b2034ed1d2..000000000000 --- a/include/asm-h8300/semaphore.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-ia64/semaphore.h b/include/asm-ia64/semaphore.h deleted file mode 100644 index d9b2034ed1d2..000000000000 --- a/include/asm-ia64/semaphore.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-m32r/semaphore.h b/include/asm-m32r/semaphore.h deleted file mode 100644 index d9b2034ed1d2..000000000000 --- a/include/asm-m32r/semaphore.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-m68k/semaphore.h b/include/asm-m68k/semaphore.h deleted file mode 100644 index d9b2034ed1d2..000000000000 --- a/include/asm-m68k/semaphore.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-m68knommu/semaphore.h b/include/asm-m68knommu/semaphore.h deleted file mode 100644 index d9b2034ed1d2..000000000000 --- a/include/asm-m68knommu/semaphore.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-mips/semaphore.h b/include/asm-mips/semaphore.h deleted file mode 100644 index d9b2034ed1d2..000000000000 --- a/include/asm-mips/semaphore.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-mn10300/semaphore.h b/include/asm-mn10300/semaphore.h deleted file mode 100644 index d9b2034ed1d2..000000000000 --- a/include/asm-mn10300/semaphore.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-parisc/semaphore.h b/include/asm-parisc/semaphore.h deleted file mode 100644 index d9b2034ed1d2..000000000000 --- a/include/asm-parisc/semaphore.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-powerpc/semaphore.h b/include/asm-powerpc/semaphore.h deleted file mode 100644 index d9b2034ed1d2..000000000000 --- a/include/asm-powerpc/semaphore.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-s390/semaphore.h b/include/asm-s390/semaphore.h deleted file mode 100644 index d9b2034ed1d2..000000000000 --- a/include/asm-s390/semaphore.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-sh/semaphore.h b/include/asm-sh/semaphore.h deleted file mode 100644 index d9b2034ed1d2..000000000000 --- a/include/asm-sh/semaphore.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-sparc/semaphore.h b/include/asm-sparc/semaphore.h deleted file mode 100644 index d9b2034ed1d2..000000000000 --- a/include/asm-sparc/semaphore.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-sparc64/semaphore.h b/include/asm-sparc64/semaphore.h deleted file mode 100644 index 39362afde5fe..000000000000 --- a/include/asm-sparc64/semaphore.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-um/semaphore.h b/include/asm-um/semaphore.h deleted file mode 100644 index d9b2034ed1d2..000000000000 --- a/include/asm-um/semaphore.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-v850/semaphore.h b/include/asm-v850/semaphore.h deleted file mode 100644 index d9b2034ed1d2..000000000000 --- a/include/asm-v850/semaphore.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-x86/semaphore.h b/include/asm-x86/semaphore.h deleted file mode 100644 index d9b2034ed1d2..000000000000 --- a/include/asm-x86/semaphore.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-xtensa/semaphore.h b/include/asm-xtensa/semaphore.h deleted file mode 100644 index d9b2034ed1d2..000000000000 --- a/include/asm-xtensa/semaphore.h +++ /dev/null @@ -1 +0,0 @@ -#include From b552068999b0b05087c454e525b30b785c79dc9b Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Wed, 23 Apr 2008 10:07:27 -0400 Subject: [PATCH 119/857] Remove __DECLARE_SEMAPHORE_GENERIC There are no users of __DECLARE_SEMAPHORE_GENERIC in the kernel Signed-off-by: Matthew Wilcox --- include/linux/semaphore.h | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/include/linux/semaphore.h b/include/linux/semaphore.h index 9cae64b00d6b..7415839ac890 100644 --- a/include/linux/semaphore.h +++ b/include/linux/semaphore.h @@ -26,10 +26,8 @@ struct semaphore { .wait_list = LIST_HEAD_INIT((name).wait_list), \ } -#define __DECLARE_SEMAPHORE_GENERIC(name, count) \ - struct semaphore name = __SEMAPHORE_INITIALIZER(name, count) - -#define DECLARE_MUTEX(name) __DECLARE_SEMAPHORE_GENERIC(name, 1) +#define DECLARE_MUTEX(name) \ + struct semaphore name = __SEMAPHORE_INITIALIZER(name, 1) static inline void sema_init(struct semaphore *sem, int val) { From e108526e77aa41c89b3be96f75d97615db2b751c Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Wed, 23 Jul 2008 21:26:44 -0700 Subject: [PATCH 120/857] move memory_read_from_buffer() from fs.h to string.h James Bottomley warns that inclusion of linux/fs.h in a low level driver was always a danger signal. This patch moves memory_read_from_buffer() from fs.h to string.h and fixes includes in existing memory_read_from_buffer() users. Signed-off-by: Akinobu Mita Cc: James Bottomley Cc: Geert Uytterhoeven Cc: Zhang Rui Cc: Bob Moore Cc: Thomas Renninger Cc: Len Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/acpi/system.c | 1 + drivers/zorro/zorro-sysfs.c | 1 - include/linux/fs.h | 2 -- include/linux/string.h | 3 +++ 4 files changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/acpi/system.c b/drivers/acpi/system.c index d8e3f153b295..91dec448b3ed 100644 --- a/drivers/acpi/system.c +++ b/drivers/acpi/system.c @@ -26,6 +26,7 @@ #include #include #include +#include #include #include diff --git a/drivers/zorro/zorro-sysfs.c b/drivers/zorro/zorro-sysfs.c index 3da712cc7708..5290552d2ef7 100644 --- a/drivers/zorro/zorro-sysfs.c +++ b/drivers/zorro/zorro-sysfs.c @@ -15,7 +15,6 @@ #include #include #include -#include #include "zorro.h" diff --git a/include/linux/fs.h b/include/linux/fs.h index 9c2ac5c0ef5c..ff54ae4933f3 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2006,8 +2006,6 @@ extern void simple_release_fs(struct vfsmount **mount, int *count); extern ssize_t simple_read_from_buffer(void __user *to, size_t count, loff_t *ppos, const void *from, size_t available); -extern ssize_t memory_read_from_buffer(void *to, size_t count, loff_t *ppos, - const void *from, size_t available); #ifdef CONFIG_MIGRATION extern int buffer_migrate_page(struct address_space *, diff --git a/include/linux/string.h b/include/linux/string.h index efdc44593b52..810d80df0a1d 100644 --- a/include/linux/string.h +++ b/include/linux/string.h @@ -111,5 +111,8 @@ extern void argv_free(char **argv); extern bool sysfs_streq(const char *s1, const char *s2); +extern ssize_t memory_read_from_buffer(void *to, size_t count, loff_t *ppos, + const void *from, size_t available); + #endif #endif /* _LINUX_STRING_H_ */ From d36e74c4392b5f26a5c4d94d7881a156ddc8e593 Mon Sep 17 00:00:00 2001 From: Clemens Ladisch Date: Wed, 23 Jul 2008 21:26:46 -0700 Subject: [PATCH 121/857] hpet: clarify maintainer entry The existing HPET maintainer entries are somewhat unclear about which one applies to what part of the kernel. Signed-off-by: Clemens Ladisch Cc: Thomas Gleixner Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 5d8971c76a7f..7ffd78c4e277 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1984,7 +1984,7 @@ P: Carlos Corbacho M: carlos@strangeworlds.co.uk S: Odd Fixes -HPET: High Precision Event Timers driver (hpet.c) +HPET: High Precision Event Timers driver (drivers/char/hpet.c) P: Clemens Ladisch M: clemens@ladisch.de S: Maintained From d7ce20b2024d318b9ba88859226af1441270d99f Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 23 Jul 2008 21:26:47 -0700 Subject: [PATCH 122/857] remove is_tty() This patch removes the no longer used is_tty(). Signed-off-by: Adrian Bunk Acked-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/tty_io.c | 13 ------------- 1 file changed, 13 deletions(-) diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index fa48dba5ba5e..6f4d856df987 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -1119,19 +1119,6 @@ int tty_hung_up_p(struct file *filp) EXPORT_SYMBOL(tty_hung_up_p); -/** - * is_tty - checker whether file is a TTY - * @filp: file handle that may be a tty - * - * Check if the file handle is a tty handle. - */ - -int is_tty(struct file *filp) -{ - return filp->f_op->read == tty_read - || filp->f_op->read == hung_up_tty_read; -} - static void session_clear_tty(struct pid *session) { struct task_struct *p; From 9483a578df27fe7603605d565eefe039c1ba5845 Mon Sep 17 00:00:00 2001 From: David Brownell Date: Wed, 23 Jul 2008 21:26:48 -0700 Subject: [PATCH 123/857] add HAVE_CLK to Kconfig, for driver dependencies Flag platforms as HAVE_CLK (or not) in Kconfig, based on whether they support calls, so that otherwise portable drivers which need those calls can list that dependency. Something like this is a prerequisite for merging the musb_hdrc driver, currently used on platforms including Davinci, OMAP2430, OMAP3xx ... and the discrete TUSB6010 chip, which doesn't have a natural platform dependency. (Used with OMAP 2420 in current Nokia N8x0 tablets.) Signed-off-by: David Brownell Cc: Russell King Acked-by: Haavard Skinnemoen Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Paul Mundt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/Kconfig | 7 +++++++ arch/arm/Kconfig | 13 +++++++++++++ arch/avr32/Kconfig | 1 + arch/powerpc/Kconfig | 1 + arch/sh/Kconfig | 1 + 5 files changed, 23 insertions(+) diff --git a/arch/Kconfig b/arch/Kconfig index ad89a33d8c6e..4d5ebbc1e72b 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -42,3 +42,10 @@ config HAVE_DMA_ATTRS config USE_GENERIC_SMP_HELPERS def_bool n + +config HAVE_CLK + def_bool n + help + The calls support software clock gating and + thus are a key power management tool on many systems. + diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index d048f6887d0b..6fb4f03369f2 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -198,12 +198,14 @@ choice config ARCH_AAEC2000 bool "Agilent AAEC-2000 based" select ARM_AMBA + select HAVE_CLK help This enables support for systems based on the Agilent AAEC-2000 config ARCH_INTEGRATOR bool "ARM Ltd. Integrator family" select ARM_AMBA + select HAVE_CLK select ICST525 help Support for ARM's Integrator platform. @@ -211,6 +213,7 @@ config ARCH_INTEGRATOR config ARCH_REALVIEW bool "ARM Ltd. RealView family" select ARM_AMBA + select HAVE_CLK select ICST307 select GENERIC_TIME select GENERIC_CLOCKEVENTS @@ -221,6 +224,7 @@ config ARCH_VERSATILE bool "ARM Ltd. Versatile family" select ARM_AMBA select ARM_VIC + select HAVE_CLK select ICST307 select GENERIC_TIME select GENERIC_CLOCKEVENTS @@ -262,6 +266,8 @@ config ARCH_EP93XX select ARM_AMBA select ARM_VIC select GENERIC_GPIO + select HAVE_CLK + select HAVE_CLK select HAVE_GPIO_LIB help This enables support for the Cirrus EP93xx series of CPUs. @@ -381,6 +387,7 @@ config ARCH_NS9XXX select GENERIC_GPIO select GENERIC_TIME select GENERIC_CLOCKEVENTS + select HAVE_CLK help Say Y here if you intend to run this kernel on a NetSilicon NS9xxx System. @@ -430,6 +437,7 @@ config ARCH_ORION5X config ARCH_PNX4008 bool "Philips Nexperia PNX4008 Mobile" + select HAVE_CLK help This enables support for Philips PNX4008 mobile platform. @@ -438,6 +446,7 @@ config ARCH_PXA depends on MMU select ARCH_MTD_XIP select GENERIC_GPIO + select HAVE_CLK select HAVE_GPIO_LIB select GENERIC_TIME select GENERIC_CLOCKEVENTS @@ -468,6 +477,7 @@ config ARCH_SA1100 select GENERIC_GPIO select GENERIC_TIME select GENERIC_CLOCKEVENTS + select HAVE_CLK select TICK_ONESHOT select HAVE_GPIO_LIB help @@ -476,6 +486,7 @@ config ARCH_SA1100 config ARCH_S3C2410 bool "Samsung S3C2410, S3C2412, S3C2413, S3C2440, S3C2442, S3C2443" select GENERIC_GPIO + select HAVE_CLK help Samsung S3C2410X CPU based systems, such as the Simtec Electronics BAST (), the IPAQ 1940 or @@ -503,12 +514,14 @@ config ARCH_DAVINCI select GENERIC_TIME select GENERIC_CLOCKEVENTS select GENERIC_GPIO + select HAVE_CLK help Support for TI's DaVinci platform. config ARCH_OMAP bool "TI OMAP" select GENERIC_GPIO + select HAVE_CLK select HAVE_GPIO_LIB select GENERIC_TIME select GENERIC_CLOCKEVENTS diff --git a/arch/avr32/Kconfig b/arch/avr32/Kconfig index 45d63c986015..df4adefedb42 100644 --- a/arch/avr32/Kconfig +++ b/arch/avr32/Kconfig @@ -10,6 +10,7 @@ config AVR32 # With EMBEDDED=n, we get lots of stuff automatically selected # that we usually don't need on AVR32. select EMBEDDED + select HAVE_CLK select HAVE_OPROFILE select HAVE_KPROBES help diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 737ebf9d12bb..4d7e2ba10bae 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -843,6 +843,7 @@ source "crypto/Kconfig" config PPC_CLOCK bool default n + select HAVE_CLK config PPC_LIB_RHEAP bool diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig index 3e7384f4619c..8879938f3356 100644 --- a/arch/sh/Kconfig +++ b/arch/sh/Kconfig @@ -8,6 +8,7 @@ mainmenu "Linux/SuperH Kernel Configuration" config SUPERH def_bool y select EMBEDDED + select HAVE_CLK select HAVE_IDE select HAVE_OPROFILE help From 6b74ab97bc12ce74acec900f1d89a4aee2e4d70d Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Wed, 23 Jul 2008 21:26:49 -0700 Subject: [PATCH 124/857] mm: add a basic debugging framework for memory initialisation Boot initialisation is very complex, with significant numbers of architecture-specific routines, hooks and code ordering. While significant amounts of the initialisation is architecture-independent, it trusts the data received from the architecture layer. This is a mistake, and has resulted in a number of difficult-to-diagnose bugs. This patchset adds some validation and tracing to memory initialisation. It also introduces a few basic defensive measures. The validation code can be explicitly disabled for embedded systems. This patch: Add additional debugging and verification code for memory initialisation. Once enabled, the verification checks are always run and when required additional debugging information may be outputted via a mminit_loglevel= command-line parameter. The verification code is placed in a new file mm/mm_init.c. Ideally other mm initialisation code will be moved here over time. Signed-off-by: Mel Gorman Cc: Christoph Lameter Cc: Andy Whitcroft Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/kernel-parameters.txt | 8 ++++++++ lib/Kconfig.debug | 12 ++++++++++++ mm/Makefile | 1 + mm/internal.h | 27 +++++++++++++++++++++++++++ mm/mm_init.c | 18 ++++++++++++++++++ mm/page_alloc.c | 22 +++++++++++++--------- 6 files changed, 79 insertions(+), 9 deletions(-) create mode 100644 mm/mm_init.c diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 47e7d8794fc6..5e20ccb5a736 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1225,6 +1225,14 @@ and is between 256 and 4096 characters. It is defined in the file mga= [HW,DRM] + mminit_loglevel= + [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this + parameter allows control of the logging verbosity for + the additional memory initialisation checks. A value + of 0 disables mminit logging and a level of 4 will + log everything. Information is printed at KERN_DEBUG + so loglevel=8 may also need to be specified. + mousedev.tap_time= [MOUSE] Maximum time between finger touching and leaving touchpad surface for touch to be considered diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 882c51048993..e1d4764435ed 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -505,6 +505,18 @@ config DEBUG_WRITECOUNT If unsure, say N. +config DEBUG_MEMORY_INIT + bool "Debug memory initialisation" if EMBEDDED + default !EMBEDDED + help + Enable this for additional checks during memory initialisation. + The sanity checks verify aspects of the VM such as the memory model + and other information provided by the architecture. Verbose + information will be printed at KERN_DEBUG loglevel depending + on the mminit_loglevel= command-line option. + + If unsure, say Y + config DEBUG_LIST bool "Debug linked list manipulation" depends on DEBUG_KERNEL diff --git a/mm/Makefile b/mm/Makefile index 18c143b3c46c..4bbc8f094ff0 100644 --- a/mm/Makefile +++ b/mm/Makefile @@ -26,6 +26,7 @@ obj-$(CONFIG_TMPFS_POSIX_ACL) += shmem_acl.o obj-$(CONFIG_TINY_SHMEM) += tiny-shmem.o obj-$(CONFIG_SLOB) += slob.o obj-$(CONFIG_SLAB) += slab.o +obj-$(CONFIG_DEBUG_MEMORY_INIT) += mm_init.o obj-$(CONFIG_SLUB) += slub.o obj-$(CONFIG_MEMORY_HOTPLUG) += memory_hotplug.o obj-$(CONFIG_FS_XIP) += filemap_xip.o diff --git a/mm/internal.h b/mm/internal.h index 0034e947e4bc..a7ee05253294 100644 --- a/mm/internal.h +++ b/mm/internal.h @@ -59,4 +59,31 @@ static inline unsigned long page_order(struct page *page) #define __paginginit __init #endif +/* Memory initialisation debug and verification */ +enum mminit_level { + MMINIT_WARNING, + MMINIT_VERIFY, + MMINIT_TRACE +}; + +#ifdef CONFIG_DEBUG_MEMORY_INIT + +extern int mminit_loglevel; + +#define mminit_dprintk(level, prefix, fmt, arg...) \ +do { \ + if (level < mminit_loglevel) { \ + printk(level <= MMINIT_WARNING ? KERN_WARNING : KERN_DEBUG); \ + printk(KERN_CONT "mminit::" prefix " " fmt, ##arg); \ + } \ +} while (0) + +#else + +static inline void mminit_dprintk(enum mminit_level level, + const char *prefix, const char *fmt, ...) +{ +} + +#endif /* CONFIG_DEBUG_MEMORY_INIT */ #endif diff --git a/mm/mm_init.c b/mm/mm_init.c new file mode 100644 index 000000000000..c01d8dfec817 --- /dev/null +++ b/mm/mm_init.c @@ -0,0 +1,18 @@ +/* + * mm_init.c - Memory initialisation verification and debugging + * + * Copyright 2008 IBM Corporation, 2008 + * Author Mel Gorman + * + */ +#include +#include + +int __meminitdata mminit_loglevel; + +static __init int set_mminit_loglevel(char *str) +{ + get_option(&str, &mminit_loglevel); + return 0; +} +early_param("mminit_loglevel", set_mminit_loglevel); diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 79ac4afc908c..0908352ba727 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2975,7 +2975,8 @@ void __init sparse_memory_present_with_active_regions(int nid) void __init push_node_boundaries(unsigned int nid, unsigned long start_pfn, unsigned long end_pfn) { - printk(KERN_DEBUG "Entering push_node_boundaries(%u, %lu, %lu)\n", + mminit_dprintk(MMINIT_TRACE, "zoneboundary", + "Entering push_node_boundaries(%u, %lu, %lu)\n", nid, start_pfn, end_pfn); /* Initialise the boundary for this node if necessary */ @@ -2993,7 +2994,8 @@ void __init push_node_boundaries(unsigned int nid, static void __meminit account_node_boundary(unsigned int nid, unsigned long *start_pfn, unsigned long *end_pfn) { - printk(KERN_DEBUG "Entering account_node_boundary(%u, %lu, %lu)\n", + mminit_dprintk(MMINIT_TRACE, "zoneboundary", + "Entering account_node_boundary(%u, %lu, %lu)\n", nid, *start_pfn, *end_pfn); /* Return if boundary information has not been provided */ @@ -3368,8 +3370,8 @@ static void __paginginit free_area_init_core(struct pglist_data *pgdat, PAGE_ALIGN(size * sizeof(struct page)) >> PAGE_SHIFT; if (realsize >= memmap_pages) { realsize -= memmap_pages; - printk(KERN_DEBUG - " %s zone: %lu pages used for memmap\n", + mminit_dprintk(MMINIT_TRACE, "memmap_init", + "%s zone: %lu pages used for memmap\n", zone_names[j], memmap_pages); } else printk(KERN_WARNING @@ -3379,7 +3381,8 @@ static void __paginginit free_area_init_core(struct pglist_data *pgdat, /* Account for reserved pages */ if (j == 0 && realsize > dma_reserve) { realsize -= dma_reserve; - printk(KERN_DEBUG " %s zone: %lu pages reserved\n", + mminit_dprintk(MMINIT_TRACE, "memmap_init", + "%s zone: %lu pages reserved\n", zone_names[0], dma_reserve); } @@ -3520,10 +3523,11 @@ void __init add_active_range(unsigned int nid, unsigned long start_pfn, { int i; - printk(KERN_DEBUG "Entering add_active_range(%d, %#lx, %#lx) " - "%d entries of %d used\n", - nid, start_pfn, end_pfn, - nr_nodemap_entries, MAX_ACTIVE_REGIONS); + mminit_dprintk(MMINIT_TRACE, "memory_register", + "Entering add_active_range(%d, %#lx, %#lx) " + "%d entries of %d used\n", + nid, start_pfn, end_pfn, + nr_nodemap_entries, MAX_ACTIVE_REGIONS); /* Merge with existing active regions if possible */ for (i = 0; i < nr_nodemap_entries; i++) { From 708614e6180f398cd307ea0048d48ba6fa274610 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Wed, 23 Jul 2008 21:26:51 -0700 Subject: [PATCH 125/857] mm: verify the page links and memory model Print out information on how the page flags are being used if mminit_loglevel is MMINIT_VERIFY or higher and unconditionally performs sanity checks on the flags regardless of loglevel. When the page flags are updated with section, node and zone information, a check are made to ensure the values can be retrieved correctly. Finally we confirm that pfn_to_page and page_to_pfn are the correct inverse functions. [akpm@linux-foundation.org: fix printk warnings] Signed-off-by: Mel Gorman Cc: Christoph Lameter Cc: Andy Whitcroft Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/internal.h | 12 +++++++++ mm/mm_init.c | 71 +++++++++++++++++++++++++++++++++++++++++++++++++ mm/page_alloc.c | 8 ++++++ 3 files changed, 91 insertions(+) diff --git a/mm/internal.h b/mm/internal.h index a7ee05253294..7a4a2885dc8e 100644 --- a/mm/internal.h +++ b/mm/internal.h @@ -78,6 +78,10 @@ do { \ } \ } while (0) +extern void mminit_verify_pageflags_layout(void); +extern void mminit_verify_page_links(struct page *page, + enum zone_type zone, unsigned long nid, unsigned long pfn); + #else static inline void mminit_dprintk(enum mminit_level level, @@ -85,5 +89,13 @@ static inline void mminit_dprintk(enum mminit_level level, { } +static inline void mminit_verify_pageflags_layout(void) +{ +} + +static inline void mminit_verify_page_links(struct page *page, + enum zone_type zone, unsigned long nid, unsigned long pfn) +{ +} #endif /* CONFIG_DEBUG_MEMORY_INIT */ #endif diff --git a/mm/mm_init.c b/mm/mm_init.c index c01d8dfec817..e16990d629e6 100644 --- a/mm/mm_init.c +++ b/mm/mm_init.c @@ -7,9 +7,80 @@ */ #include #include +#include "internal.h" int __meminitdata mminit_loglevel; +void __init mminit_verify_pageflags_layout(void) +{ + int shift, width; + unsigned long or_mask, add_mask; + + shift = 8 * sizeof(unsigned long); + width = shift - SECTIONS_WIDTH - NODES_WIDTH - ZONES_WIDTH; + mminit_dprintk(MMINIT_TRACE, "pageflags_layout_widths", + "Section %d Node %d Zone %d Flags %d\n", + SECTIONS_WIDTH, + NODES_WIDTH, + ZONES_WIDTH, + NR_PAGEFLAGS); + mminit_dprintk(MMINIT_TRACE, "pageflags_layout_shifts", + "Section %d Node %d Zone %d\n", +#ifdef SECTIONS_SHIFT + SECTIONS_SHIFT, +#else + 0, +#endif + NODES_SHIFT, + ZONES_SHIFT); + mminit_dprintk(MMINIT_TRACE, "pageflags_layout_offsets", + "Section %lu Node %lu Zone %lu\n", + (unsigned long)SECTIONS_PGSHIFT, + (unsigned long)NODES_PGSHIFT, + (unsigned long)ZONES_PGSHIFT); + mminit_dprintk(MMINIT_TRACE, "pageflags_layout_zoneid", + "Zone ID: %lu -> %lu\n", + (unsigned long)ZONEID_PGOFF, + (unsigned long)(ZONEID_PGOFF + ZONEID_SHIFT)); + mminit_dprintk(MMINIT_TRACE, "pageflags_layout_usage", + "location: %d -> %d unused %d -> %d flags %d -> %d\n", + shift, width, width, NR_PAGEFLAGS, NR_PAGEFLAGS, 0); +#ifdef NODE_NOT_IN_PAGE_FLAGS + mminit_dprintk(MMINIT_TRACE, "pageflags_layout_nodeflags", + "Node not in page flags"); +#endif + + if (SECTIONS_WIDTH) { + shift -= SECTIONS_WIDTH; + BUG_ON(shift != SECTIONS_PGSHIFT); + } + if (NODES_WIDTH) { + shift -= NODES_WIDTH; + BUG_ON(shift != NODES_PGSHIFT); + } + if (ZONES_WIDTH) { + shift -= ZONES_WIDTH; + BUG_ON(shift != ZONES_PGSHIFT); + } + + /* Check for bitmask overlaps */ + or_mask = (ZONES_MASK << ZONES_PGSHIFT) | + (NODES_MASK << NODES_PGSHIFT) | + (SECTIONS_MASK << SECTIONS_PGSHIFT); + add_mask = (ZONES_MASK << ZONES_PGSHIFT) + + (NODES_MASK << NODES_PGSHIFT) + + (SECTIONS_MASK << SECTIONS_PGSHIFT); + BUG_ON(or_mask != add_mask); +} + +void __meminit mminit_verify_page_links(struct page *page, enum zone_type zone, + unsigned long nid, unsigned long pfn) +{ + BUG_ON(page_to_nid(page) != nid); + BUG_ON(page_zonenum(page) != zone); + BUG_ON(page_to_pfn(page) != pfn); +} + static __init int set_mminit_loglevel(char *str) { get_option(&str, &mminit_loglevel); diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 0908352ba727..acab6ad326df 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2534,6 +2534,7 @@ void __meminit memmap_init_zone(unsigned long size, int nid, unsigned long zone, } page = pfn_to_page(pfn); set_page_links(page, zone, nid, pfn); + mminit_verify_page_links(page, zone, nid, pfn); init_page_count(page); reset_page_mapcount(page); SetPageReserved(page); @@ -2836,6 +2837,12 @@ __meminit int init_currently_empty_zone(struct zone *zone, zone->zone_start_pfn = zone_start_pfn; + mminit_dprintk(MMINIT_TRACE, "memmap_init", + "Initialising map node %d zone %lu pfns %lu -> %lu\n", + pgdat->node_id, + (unsigned long)zone_idx(zone), + zone_start_pfn, (zone_start_pfn + size)); + zone_init_free_lists(zone); return 0; @@ -3961,6 +3968,7 @@ void __init free_area_init_nodes(unsigned long *max_zone_pfn) early_node_map[i].end_pfn); /* Initialise every node */ + mminit_verify_pageflags_layout(); setup_nr_node_ids(); for_each_online_node(nid) { pg_data_t *pgdat = NODE_DATA(nid); From 2dbb51c49f4fecb8330e43247a0edfbc4b2b8974 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Wed, 23 Jul 2008 21:26:52 -0700 Subject: [PATCH 126/857] mm: make defensive checks around PFN values registered for memory usage There are a number of different views to how much memory is currently active. There is the arch-independent zone-sizing view, the bootmem allocator and memory models view. Architectures register this information at different times and is not necessarily in sync particularly with respect to some SPARSEMEM limitations. This patch introduces mminit_validate_memmodel_limits() which is able to validate and correct PFN ranges with respect to the memory model. It is only SPARSEMEM that currently validates itself. Signed-off-by: Mel Gorman Cc: Christoph Lameter Cc: Andy Whitcroft Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/bootmem.c | 1 + mm/internal.h | 12 ++++++++++++ mm/page_alloc.c | 2 ++ mm/sparse.c | 37 +++++++++++++++++++++++++++++-------- 4 files changed, 44 insertions(+), 8 deletions(-) diff --git a/mm/bootmem.c b/mm/bootmem.c index 8d9f60e06f62..9f4bbc5da73f 100644 --- a/mm/bootmem.c +++ b/mm/bootmem.c @@ -91,6 +91,7 @@ static unsigned long __init init_bootmem_core(pg_data_t *pgdat, bootmem_data_t *bdata = pgdat->bdata; unsigned long mapsize; + mminit_validate_memmodel_limits(&start, &end); bdata->node_bootmem_map = phys_to_virt(PFN_PHYS(mapstart)); bdata->node_boot_start = PFN_PHYS(start); bdata->node_low_pfn = end; diff --git a/mm/internal.h b/mm/internal.h index 7a4a2885dc8e..5d17f3efac41 100644 --- a/mm/internal.h +++ b/mm/internal.h @@ -98,4 +98,16 @@ static inline void mminit_verify_page_links(struct page *page, { } #endif /* CONFIG_DEBUG_MEMORY_INIT */ + +/* mminit_validate_memmodel_limits is independent of CONFIG_DEBUG_MEMORY_INIT */ +#if defined(CONFIG_SPARSEMEM) +extern void mminit_validate_memmodel_limits(unsigned long *start_pfn, + unsigned long *end_pfn); +#else +static inline void mminit_validate_memmodel_limits(unsigned long *start_pfn, + unsigned long *end_pfn) +{ +} +#endif /* CONFIG_SPARSEMEM */ + #endif diff --git a/mm/page_alloc.c b/mm/page_alloc.c index acab6ad326df..0adb66e711e6 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -3536,6 +3536,8 @@ void __init add_active_range(unsigned int nid, unsigned long start_pfn, nid, start_pfn, end_pfn, nr_nodemap_entries, MAX_ACTIVE_REGIONS); + mminit_validate_memmodel_limits(&start_pfn, &end_pfn); + /* Merge with existing active regions if possible */ for (i = 0; i < nr_nodemap_entries; i++) { if (early_node_map[i].nid != nid) diff --git a/mm/sparse.c b/mm/sparse.c index 36511c7b5e2c..7a3650923d9a 100644 --- a/mm/sparse.c +++ b/mm/sparse.c @@ -12,6 +12,7 @@ #include #include #include +#include "internal.h" /* * Permanent SPARSEMEM data: @@ -147,22 +148,41 @@ static inline int sparse_early_nid(struct mem_section *section) return (section->section_mem_map >> SECTION_NID_SHIFT); } -/* Record a memory area against a node. */ -void __init memory_present(int nid, unsigned long start, unsigned long end) +/* Validate the physical addressing limitations of the model */ +void __meminit mminit_validate_memmodel_limits(unsigned long *start_pfn, + unsigned long *end_pfn) { - unsigned long max_arch_pfn = 1UL << (MAX_PHYSMEM_BITS-PAGE_SHIFT); - unsigned long pfn; + unsigned long max_sparsemem_pfn = 1UL << (MAX_PHYSMEM_BITS-PAGE_SHIFT); /* * Sanity checks - do not allow an architecture to pass * in larger pfns than the maximum scope of sparsemem: */ - if (start >= max_arch_pfn) - return; - if (end >= max_arch_pfn) - end = max_arch_pfn; + if (*start_pfn > max_sparsemem_pfn) { + mminit_dprintk(MMINIT_WARNING, "pfnvalidation", + "Start of range %lu -> %lu exceeds SPARSEMEM max %lu\n", + *start_pfn, *end_pfn, max_sparsemem_pfn); + WARN_ON_ONCE(1); + *start_pfn = max_sparsemem_pfn; + *end_pfn = max_sparsemem_pfn; + } + + if (*end_pfn > max_sparsemem_pfn) { + mminit_dprintk(MMINIT_WARNING, "pfnvalidation", + "End of range %lu -> %lu exceeds SPARSEMEM max %lu\n", + *start_pfn, *end_pfn, max_sparsemem_pfn); + WARN_ON_ONCE(1); + *end_pfn = max_sparsemem_pfn; + } +} + +/* Record a memory area against a node. */ +void __init memory_present(int nid, unsigned long start, unsigned long end) +{ + unsigned long pfn; start &= PAGE_SECTION_MASK; + mminit_validate_memmodel_limits(&start, &end); for (pfn = start; pfn < end; pfn += PAGES_PER_SECTION) { unsigned long section = pfn_to_section_nr(pfn); struct mem_section *ms; @@ -187,6 +207,7 @@ unsigned long __init node_memmap_size_bytes(int nid, unsigned long start_pfn, unsigned long pfn; unsigned long nr_pages = 0; + mminit_validate_memmodel_limits(&start_pfn, &end_pfn); for (pfn = start_pfn; pfn < end_pfn; pfn += PAGES_PER_SECTION) { if (nid != early_pfn_to_nid(pfn)) continue; From 68ad8df42e12037c3894c9706ab428bf5cd6426b Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Wed, 23 Jul 2008 21:26:52 -0700 Subject: [PATCH 127/857] mm: print out the zonelists on request for manual verification This patch prints out the zonelists during boot for manual verification by the user if the mminit_loglevel is MMINIT_VERIFY or higher. Signed-off-by: Mel Gorman Cc: Christoph Lameter Cc: Andy Whitcroft Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/internal.h | 5 +++++ mm/mm_init.c | 45 +++++++++++++++++++++++++++++++++++++++++++++ mm/page_alloc.c | 1 + 3 files changed, 51 insertions(+) diff --git a/mm/internal.h b/mm/internal.h index 5d17f3efac41..50807e12490e 100644 --- a/mm/internal.h +++ b/mm/internal.h @@ -81,6 +81,7 @@ do { \ extern void mminit_verify_pageflags_layout(void); extern void mminit_verify_page_links(struct page *page, enum zone_type zone, unsigned long nid, unsigned long pfn); +extern void mminit_verify_zonelist(void); #else @@ -97,6 +98,10 @@ static inline void mminit_verify_page_links(struct page *page, enum zone_type zone, unsigned long nid, unsigned long pfn) { } + +static inline void mminit_verify_zonelist(void) +{ +} #endif /* CONFIG_DEBUG_MEMORY_INIT */ /* mminit_validate_memmodel_limits is independent of CONFIG_DEBUG_MEMORY_INIT */ diff --git a/mm/mm_init.c b/mm/mm_init.c index e16990d629e6..ce445ca097e7 100644 --- a/mm/mm_init.c +++ b/mm/mm_init.c @@ -11,6 +11,51 @@ int __meminitdata mminit_loglevel; +/* The zonelists are simply reported, validation is manual. */ +void mminit_verify_zonelist(void) +{ + int nid; + + if (mminit_loglevel < MMINIT_VERIFY) + return; + + for_each_online_node(nid) { + pg_data_t *pgdat = NODE_DATA(nid); + struct zone *zone; + struct zoneref *z; + struct zonelist *zonelist; + int i, listid, zoneid; + + BUG_ON(MAX_ZONELISTS > 2); + for (i = 0; i < MAX_ZONELISTS * MAX_NR_ZONES; i++) { + + /* Identify the zone and nodelist */ + zoneid = i % MAX_NR_ZONES; + listid = i / MAX_NR_ZONES; + zonelist = &pgdat->node_zonelists[listid]; + zone = &pgdat->node_zones[zoneid]; + if (!populated_zone(zone)) + continue; + + /* Print information about the zonelist */ + printk(KERN_DEBUG "mminit::zonelist %s %d:%s = ", + listid > 0 ? "thisnode" : "general", nid, + zone->name); + + /* Iterate the zonelist */ + for_each_zone_zonelist(zone, z, zonelist, zoneid) { +#ifdef CONFIG_NUMA + printk(KERN_CONT "%d:%s ", + zone->node, zone->name); +#else + printk(KERN_CONT "0:%s ", zone->name); +#endif /* CONFIG_NUMA */ + } + printk(KERN_CONT "\n"); + } + } +} + void __init mminit_verify_pageflags_layout(void) { int shift, width; diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 0adb66e711e6..9ece07ce65b0 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2352,6 +2352,7 @@ void build_all_zonelists(void) if (system_state == SYSTEM_BOOTING) { __build_all_zonelists(NULL); + mminit_verify_zonelist(); cpuset_init_current_mems_allowed(); } else { /* we have to stop all cpus to guarantee there is no user From 8b05c7e6e159d2f33c9275281b8b909a89eb7c5d Mon Sep 17 00:00:00 2001 From: FUJITA Tomonori Date: Wed, 23 Jul 2008 21:26:53 -0700 Subject: [PATCH 128/857] add a helper function to test if an object is on the stack lib/debugobjects.c has a function to test if an object is on the stack. The block layer and ide needs it (they need to avoid DMA from/to stack buffers). This patch moves the function to include/linux/sched.h so that everyone can use it. lib/debugobjects.c uses current->stack but this patch uses a task_stack_page() accessor, which is a preferable way to access the stack. Signed-off-by: FUJITA Tomonori Cc: Christoph Lameter Cc: Andy Whitcroft Cc: Ingo Molnar Cc: Thomas Gleixner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 7 +++++++ lib/debugobjects.c | 4 +--- 2 files changed, 8 insertions(+), 3 deletions(-) diff --git a/include/linux/sched.h b/include/linux/sched.h index dc7e592c473a..6aca4a16e377 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1983,6 +1983,13 @@ static inline unsigned long *end_of_stack(struct task_struct *p) #endif +static inline int object_is_on_stack(void *obj) +{ + void *stack = task_stack_page(current); + + return (obj >= stack) && (obj < (stack + THREAD_SIZE)); +} + extern void thread_info_cache_init(void); /* set thread flags in other task's structures diff --git a/lib/debugobjects.c b/lib/debugobjects.c index 85b18d79be89..f86196390cfd 100644 --- a/lib/debugobjects.c +++ b/lib/debugobjects.c @@ -226,15 +226,13 @@ debug_object_fixup(int (*fixup)(void *addr, enum debug_obj_state state), static void debug_object_is_on_stack(void *addr, int onstack) { - void *stack = current->stack; int is_on_stack; static int limit; if (limit > 4) return; - is_on_stack = (addr >= stack && addr < (stack + THREAD_SIZE)); - + is_on_stack = object_is_on_stack(addr); if (is_on_stack == onstack) return; From b61bfa3c462671c48a51fb5c31af337c5a996a04 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 23 Jul 2008 21:26:55 -0700 Subject: [PATCH 129/857] mm: move bootmem descriptors definition to a single place There are a lot of places that define either a single bootmem descriptor or an array of them. Use only one central array with MAX_NUMNODES items instead. Signed-off-by: Johannes Weiner Acked-by: Ralf Baechle Cc: Ingo Molnar Cc: Richard Henderson Cc: Russell King Cc: Tony Luck Cc: Hirokazu Takata Cc: Geert Uytterhoeven Cc: Kyle McMartin Cc: Paul Mackerras Cc: Paul Mundt Cc: David S. Miller Cc: Yinghai Lu Cc: Christoph Lameter Cc: Mel Gorman Cc: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/alpha/mm/numa.c | 8 ++++---- arch/arm/mm/discontig.c | 34 +++++++++++++++----------------- arch/ia64/mm/discontig.c | 11 +++++------ arch/m32r/mm/discontig.c | 4 +--- arch/m68k/mm/init.c | 4 +--- arch/mips/sgi-ip27/ip27-memory.c | 4 +--- arch/parisc/mm/init.c | 3 +-- arch/powerpc/mm/numa.c | 3 +-- arch/sh/mm/numa.c | 5 ++--- arch/sparc64/mm/init.c | 3 +-- arch/x86/mm/discontig_32.c | 3 +-- arch/x86/mm/numa_64.c | 4 +--- include/linux/bootmem.h | 2 ++ mm/bootmem.c | 2 ++ mm/page_alloc.c | 4 +--- 15 files changed, 40 insertions(+), 54 deletions(-) diff --git a/arch/alpha/mm/numa.c b/arch/alpha/mm/numa.c index 10ab7833e83c..a53fda0481ca 100644 --- a/arch/alpha/mm/numa.c +++ b/arch/alpha/mm/numa.c @@ -19,7 +19,6 @@ #include pg_data_t node_data[MAX_NUMNODES]; -bootmem_data_t node_bdata[MAX_NUMNODES]; EXPORT_SYMBOL(node_data); #undef DEBUG_DISCONTIG @@ -141,7 +140,7 @@ setup_memory_node(int nid, void *kernel_end) printk(" not enough mem to reserve NODE_DATA"); return; } - NODE_DATA(nid)->bdata = &node_bdata[nid]; + NODE_DATA(nid)->bdata = &bootmem_node_data[nid]; printk(" Detected node memory: start %8lu, end %8lu\n", node_min_pfn, node_max_pfn); @@ -304,8 +303,9 @@ void __init paging_init(void) dma_local_pfn = virt_to_phys((char *)MAX_DMA_ADDRESS) >> PAGE_SHIFT; for_each_online_node(nid) { - unsigned long start_pfn = node_bdata[nid].node_boot_start >> PAGE_SHIFT; - unsigned long end_pfn = node_bdata[nid].node_low_pfn; + bootmem_data_t *bdata = &bootmem_node_data[nid]; + unsigned long start_pfn = bdata->node_boot_start >> PAGE_SHIFT; + unsigned long end_pfn = bdata->node_low_pfn; if (dma_local_pfn >= end_pfn - start_pfn) zones_size[ZONE_DMA] = end_pfn - start_pfn; diff --git a/arch/arm/mm/discontig.c b/arch/arm/mm/discontig.c index 1e5602189507..c8c0c4b0f0a3 100644 --- a/arch/arm/mm/discontig.c +++ b/arch/arm/mm/discontig.c @@ -21,26 +21,24 @@ * Our node_data structure for discontiguous memory. */ -static bootmem_data_t node_bootmem_data[MAX_NUMNODES]; - pg_data_t discontig_node_data[MAX_NUMNODES] = { - { .bdata = &node_bootmem_data[0] }, - { .bdata = &node_bootmem_data[1] }, - { .bdata = &node_bootmem_data[2] }, - { .bdata = &node_bootmem_data[3] }, + { .bdata = &bootmem_node_data[0] }, + { .bdata = &bootmem_node_data[1] }, + { .bdata = &bootmem_node_data[2] }, + { .bdata = &bootmem_node_data[3] }, #if MAX_NUMNODES == 16 - { .bdata = &node_bootmem_data[4] }, - { .bdata = &node_bootmem_data[5] }, - { .bdata = &node_bootmem_data[6] }, - { .bdata = &node_bootmem_data[7] }, - { .bdata = &node_bootmem_data[8] }, - { .bdata = &node_bootmem_data[9] }, - { .bdata = &node_bootmem_data[10] }, - { .bdata = &node_bootmem_data[11] }, - { .bdata = &node_bootmem_data[12] }, - { .bdata = &node_bootmem_data[13] }, - { .bdata = &node_bootmem_data[14] }, - { .bdata = &node_bootmem_data[15] }, + { .bdata = &bootmem_node_data[4] }, + { .bdata = &bootmem_node_data[5] }, + { .bdata = &bootmem_node_data[6] }, + { .bdata = &bootmem_node_data[7] }, + { .bdata = &bootmem_node_data[8] }, + { .bdata = &bootmem_node_data[9] }, + { .bdata = &bootmem_node_data[10] }, + { .bdata = &bootmem_node_data[11] }, + { .bdata = &bootmem_node_data[12] }, + { .bdata = &bootmem_node_data[13] }, + { .bdata = &bootmem_node_data[14] }, + { .bdata = &bootmem_node_data[15] }, #endif }; diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c index 544dc420c65e..2fcf8464331e 100644 --- a/arch/ia64/mm/discontig.c +++ b/arch/ia64/mm/discontig.c @@ -36,7 +36,6 @@ struct early_node_data { struct ia64_node_data *node_data; unsigned long pernode_addr; unsigned long pernode_size; - struct bootmem_data bootmem_data; unsigned long num_physpages; #ifdef CONFIG_ZONE_DMA unsigned long num_dma_physpages; @@ -76,7 +75,7 @@ static int __init build_node_maps(unsigned long start, unsigned long len, int node) { unsigned long cstart, epfn, end = start + len; - struct bootmem_data *bdp = &mem_data[node].bootmem_data; + struct bootmem_data *bdp = &bootmem_node_data[node]; epfn = GRANULEROUNDUP(end) >> PAGE_SHIFT; cstart = GRANULEROUNDDOWN(start); @@ -167,7 +166,7 @@ static void __init fill_pernode(int node, unsigned long pernode, { void *cpu_data; int cpus = early_nr_cpus_node(node); - struct bootmem_data *bdp = &mem_data[node].bootmem_data; + struct bootmem_data *bdp = &bootmem_node_data[node]; mem_data[node].pernode_addr = pernode; mem_data[node].pernode_size = pernodesize; @@ -224,7 +223,7 @@ static int __init find_pernode_space(unsigned long start, unsigned long len, { unsigned long epfn; unsigned long pernodesize = 0, pernode, pages, mapsize; - struct bootmem_data *bdp = &mem_data[node].bootmem_data; + struct bootmem_data *bdp = &bootmem_node_data[node]; epfn = (start + len) >> PAGE_SHIFT; @@ -440,7 +439,7 @@ void __init find_memory(void) efi_memmap_walk(find_max_min_low_pfn, NULL); for_each_online_node(node) - if (mem_data[node].bootmem_data.node_low_pfn) { + if (bootmem_node_data[node].node_low_pfn) { node_clear(node, memory_less_mask); mem_data[node].min_pfn = ~0UL; } @@ -460,7 +459,7 @@ void __init find_memory(void) else if (node_isset(node, memory_less_mask)) continue; - bdp = &mem_data[node].bootmem_data; + bdp = &bootmem_node_data[node]; pernode = mem_data[node].pernode_addr; pernodesize = mem_data[node].pernode_size; map = pernode + pernodesize; diff --git a/arch/m32r/mm/discontig.c b/arch/m32r/mm/discontig.c index 07c1af7dc0e2..aa9145ef6cca 100644 --- a/arch/m32r/mm/discontig.c +++ b/arch/m32r/mm/discontig.c @@ -20,7 +20,6 @@ extern char _end[]; struct pglist_data *node_data[MAX_NUMNODES]; EXPORT_SYMBOL(node_data); -static bootmem_data_t node_bdata[MAX_NUMNODES] __initdata; pg_data_t m32r_node_data[MAX_NUMNODES]; @@ -81,7 +80,7 @@ unsigned long __init setup_memory(void) for_each_online_node(nid) { mp = &mem_prof[nid]; NODE_DATA(nid)=(pg_data_t *)&m32r_node_data[nid]; - NODE_DATA(nid)->bdata = &node_bdata[nid]; + NODE_DATA(nid)->bdata = &bootmem_node_data[nid]; min_pfn = mp->start_pfn; max_pfn = mp->start_pfn + mp->pages; bootmap_size = init_bootmem_node(NODE_DATA(nid), mp->free_pfn, @@ -163,4 +162,3 @@ unsigned long __init zone_sizes_init(void) return holes; } - diff --git a/arch/m68k/mm/init.c b/arch/m68k/mm/init.c index d8fb9c5303cc..79f5f94d4800 100644 --- a/arch/m68k/mm/init.c +++ b/arch/m68k/mm/init.c @@ -32,8 +32,6 @@ DEFINE_PER_CPU(struct mmu_gather, mmu_gathers); -static bootmem_data_t __initdata bootmem_data[MAX_NUMNODES]; - pg_data_t pg_data_map[MAX_NUMNODES]; EXPORT_SYMBOL(pg_data_map); @@ -58,7 +56,7 @@ void __init m68k_setup_node(int node) pg_data_table[i] = pg_data_map + node; } #endif - pg_data_map[node].bdata = bootmem_data + node; + pg_data_map[node].bdata = bootmem_node_data + node; node_set_online(node); } diff --git a/arch/mips/sgi-ip27/ip27-memory.c b/arch/mips/sgi-ip27/ip27-memory.c index 42cd10956306..060d853d7b35 100644 --- a/arch/mips/sgi-ip27/ip27-memory.c +++ b/arch/mips/sgi-ip27/ip27-memory.c @@ -33,8 +33,6 @@ #define SLOT_PFNSHIFT (SLOT_SHIFT - PAGE_SHIFT) #define PFN_NASIDSHFT (NASID_SHFT - PAGE_SHIFT) -static struct bootmem_data __initdata plat_node_bdata[MAX_COMPACT_NODES]; - struct node_data *__node_data[MAX_COMPACT_NODES]; EXPORT_SYMBOL(__node_data); @@ -403,7 +401,7 @@ static void __init node_mem_init(cnodeid_t node) */ __node_data[node] = __va(slot_freepfn << PAGE_SHIFT); - NODE_DATA(node)->bdata = &plat_node_bdata[node]; + NODE_DATA(node)->bdata = &bootmem_node_data[node]; NODE_DATA(node)->node_start_pfn = start_pfn; NODE_DATA(node)->node_spanned_pages = end_pfn - start_pfn; diff --git a/arch/parisc/mm/init.c b/arch/parisc/mm/init.c index b4d6c8777ed0..0ddf4904640a 100644 --- a/arch/parisc/mm/init.c +++ b/arch/parisc/mm/init.c @@ -36,7 +36,6 @@ extern int data_start; #ifdef CONFIG_DISCONTIGMEM struct node_map_data node_data[MAX_NUMNODES] __read_mostly; -bootmem_data_t bmem_data[MAX_NUMNODES] __read_mostly; unsigned char pfnnid_map[PFNNID_MAP_MAX] __read_mostly; #endif @@ -262,7 +261,7 @@ static void __init setup_bootmem(void) #ifdef CONFIG_DISCONTIGMEM for (i = 0; i < MAX_PHYSMEM_RANGES; i++) { memset(NODE_DATA(i), 0, sizeof(pg_data_t)); - NODE_DATA(i)->bdata = &bmem_data[i]; + NODE_DATA(i)->bdata = &bootmem_node_data[i]; } memset(pfnnid_map, 0xff, sizeof(pfnnid_map)); diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c index cf4bffba6f7c..d9a181351332 100644 --- a/arch/powerpc/mm/numa.c +++ b/arch/powerpc/mm/numa.c @@ -39,7 +39,6 @@ EXPORT_SYMBOL(numa_cpu_lookup_table); EXPORT_SYMBOL(numa_cpumask_lookup_table); EXPORT_SYMBOL(node_data); -static bootmem_data_t __initdata plat_node_bdata[MAX_NUMNODES]; static int min_common_depth; static int n_mem_addr_cells, n_mem_size_cells; @@ -816,7 +815,7 @@ void __init do_init_bootmem(void) dbg("node %d\n", nid); dbg("NODE_DATA() = %p\n", NODE_DATA(nid)); - NODE_DATA(nid)->bdata = &plat_node_bdata[nid]; + NODE_DATA(nid)->bdata = &bootmem_node_data[nid]; NODE_DATA(nid)->node_start_pfn = start_pfn; NODE_DATA(nid)->node_spanned_pages = end_pfn - start_pfn; diff --git a/arch/sh/mm/numa.c b/arch/sh/mm/numa.c index 1663199ce888..095d93bec7cd 100644 --- a/arch/sh/mm/numa.c +++ b/arch/sh/mm/numa.c @@ -14,7 +14,6 @@ #include #include -static bootmem_data_t plat_node_bdata[MAX_NUMNODES]; struct pglist_data *node_data[MAX_NUMNODES] __read_mostly; EXPORT_SYMBOL_GPL(node_data); @@ -35,7 +34,7 @@ void __init setup_memory(void) NODE_DATA(0) = pfn_to_kaddr(free_pfn); memset(NODE_DATA(0), 0, sizeof(struct pglist_data)); free_pfn += PFN_UP(sizeof(struct pglist_data)); - NODE_DATA(0)->bdata = &plat_node_bdata[0]; + NODE_DATA(0)->bdata = &bootmem_node_data[0]; /* Set up node 0 */ setup_bootmem_allocator(free_pfn); @@ -66,7 +65,7 @@ void __init setup_bootmem_node(int nid, unsigned long start, unsigned long end) free_pfn += PFN_UP(sizeof(struct pglist_data)); memset(NODE_DATA(nid), 0, sizeof(struct pglist_data)); - NODE_DATA(nid)->bdata = &plat_node_bdata[nid]; + NODE_DATA(nid)->bdata = &bootmem_node_data[nid]; NODE_DATA(nid)->node_start_pfn = start_pfn; NODE_DATA(nid)->node_spanned_pages = end_pfn - start_pfn; diff --git a/arch/sparc64/mm/init.c b/arch/sparc64/mm/init.c index 84898c44dd4d..713297473951 100644 --- a/arch/sparc64/mm/init.c +++ b/arch/sparc64/mm/init.c @@ -788,7 +788,6 @@ int numa_cpu_lookup_table[NR_CPUS]; cpumask_t numa_cpumask_lookup_table[MAX_NUMNODES]; #ifdef CONFIG_NEED_MULTIPLE_NODES -static bootmem_data_t plat_node_bdata[MAX_NUMNODES]; struct mdesc_mblock { u64 base; @@ -871,7 +870,7 @@ static void __init allocate_node_data(int nid) NODE_DATA(nid) = __va(paddr); memset(NODE_DATA(nid), 0, sizeof(struct pglist_data)); - NODE_DATA(nid)->bdata = &plat_node_bdata[nid]; + NODE_DATA(nid)->bdata = &bootmem_node_data[nid]; #endif p = NODE_DATA(nid); diff --git a/arch/x86/mm/discontig_32.c b/arch/x86/mm/discontig_32.c index 5dfef9fa061a..62fa440678d8 100644 --- a/arch/x86/mm/discontig_32.c +++ b/arch/x86/mm/discontig_32.c @@ -42,7 +42,6 @@ struct pglist_data *node_data[MAX_NUMNODES] __read_mostly; EXPORT_SYMBOL(node_data); -static bootmem_data_t node0_bdata; /* * numa interface - we expect the numa architecture specific code to have @@ -385,7 +384,7 @@ void __init initmem_init(unsigned long start_pfn, for_each_online_node(nid) memset(NODE_DATA(nid), 0, sizeof(struct pglist_data)); - NODE_DATA(0)->bdata = &node0_bdata; + NODE_DATA(0)->bdata = &bootmem_node_data[0]; setup_bootmem_allocator(); } diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c index 9782f42dd319..a4dd793d6003 100644 --- a/arch/x86/mm/numa_64.c +++ b/arch/x86/mm/numa_64.c @@ -23,8 +23,6 @@ struct pglist_data *node_data[MAX_NUMNODES] __read_mostly; EXPORT_SYMBOL(node_data); -static bootmem_data_t plat_node_bdata[MAX_NUMNODES]; - struct memnode memnode; s16 apicid_to_node[MAX_LOCAL_APIC] __cpuinitdata = { @@ -198,7 +196,7 @@ void __init setup_node_bootmem(int nodeid, unsigned long start, nodedata_phys + pgdat_size - 1); memset(NODE_DATA(nodeid), 0, sizeof(pg_data_t)); - NODE_DATA(nodeid)->bdata = &plat_node_bdata[nodeid]; + NODE_DATA(nodeid)->bdata = &bootmem_node_data[nodeid]; NODE_DATA(nodeid)->node_start_pfn = start_pfn; NODE_DATA(nodeid)->node_spanned_pages = last_pfn - start_pfn; diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h index a1d9b79078ea..2599c741405e 100644 --- a/include/linux/bootmem.h +++ b/include/linux/bootmem.h @@ -38,6 +38,8 @@ typedef struct bootmem_data { struct list_head list; } bootmem_data_t; +extern bootmem_data_t bootmem_node_data[]; + extern unsigned long bootmem_bootmap_pages(unsigned long); extern unsigned long init_bootmem(unsigned long addr, unsigned long memend); extern void free_bootmem(unsigned long addr, unsigned long size); diff --git a/mm/bootmem.c b/mm/bootmem.c index 9f4bbc5da73f..35b3cb667036 100644 --- a/mm/bootmem.c +++ b/mm/bootmem.c @@ -36,6 +36,8 @@ static LIST_HEAD(bdata_list); unsigned long saved_max_pfn; #endif +bootmem_data_t bootmem_node_data[MAX_NUMNODES] __initdata; + /* return the number of _pages_ that will be allocated for the boot bitmap */ unsigned long __init bootmem_bootmap_pages(unsigned long pages) { diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 9ece07ce65b0..e089b92cdfff 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -4040,9 +4040,7 @@ void __init set_dma_reserve(unsigned long new_dma_reserve) } #ifndef CONFIG_NEED_MULTIPLE_NODES -static bootmem_data_t contig_bootmem_data; -struct pglist_data contig_page_data = { .bdata = &contig_bootmem_data }; - +struct pglist_data contig_page_data = { .bdata = &bootmem_node_data[0] }; EXPORT_SYMBOL(contig_page_data); #endif From 6b312c0e6e2f44b020e12953d1dd37eed60e3609 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 23 Jul 2008 21:26:58 -0700 Subject: [PATCH 130/857] mm: fix free_all_bootmem_core alignment check The check for node_boot_start is bogus because we start freeing at the corresponding pfn. So check if the pfn is properly aligned instead in a more readable way and adjust the documentation. Also remove an unneeded accounting variable. Signed-off-by: Johannes Weiner Cc: Ingo Molnar Cc: Yinghai Lu Cc: Christoph Lameter Cc: Mel Gorman Cc: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/bootmem.c | 21 ++++++++++----------- 1 file changed, 10 insertions(+), 11 deletions(-) diff --git a/mm/bootmem.c b/mm/bootmem.c index 35b3cb667036..319a79bce7cd 100644 --- a/mm/bootmem.c +++ b/mm/bootmem.c @@ -377,7 +377,7 @@ static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat) struct page *page; unsigned long pfn; bootmem_data_t *bdata = pgdat->bdata; - unsigned long i, count, total = 0; + unsigned long i, count; unsigned long idx; unsigned long *map; int gofast = 0; @@ -389,10 +389,13 @@ static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat) pfn = PFN_DOWN(bdata->node_boot_start); idx = bdata->node_low_pfn - pfn; map = bdata->node_bootmem_map; - /* Check physaddr is O(LOG2(BITS_PER_LONG)) page aligned */ - if (bdata->node_boot_start == 0 || - ffs(bdata->node_boot_start) - PAGE_SHIFT > ffs(BITS_PER_LONG)) + /* + * Check if we are aligned to BITS_PER_LONG pages. If so, we might + * be able to free page orders of that size at once. + */ + if (!(pfn & (BITS_PER_LONG-1))) gofast = 1; + for (i = 0; i < idx; ) { unsigned long v = ~map[i / BITS_PER_LONG]; @@ -420,23 +423,19 @@ static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat) } pfn += BITS_PER_LONG; } - total += count; /* * Now free the allocator bitmap itself, it's not * needed anymore: */ page = virt_to_page(bdata->node_bootmem_map); - count = 0; idx = (get_mapsize(bdata) + PAGE_SIZE-1) >> PAGE_SHIFT; - for (i = 0; i < idx; i++, page++) { + for (i = 0; i < idx; i++, page++) __free_pages_bootmem(page, 0); - count++; - } - total += count; + count += i; bdata->node_bootmem_map = NULL; - return total; + return count; } unsigned long __init init_bootmem_node(pg_data_t *pgdat, unsigned long freepfn, From 8ae04463077324ed9f6b04ab3a5b17ae1ee4dd35 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 23 Jul 2008 21:26:59 -0700 Subject: [PATCH 131/857] mm: normalize internal argument passing of bootmem data All _core functions only need the bootmem data, not the whole node descriptor. Adjust the two functions that take the node descriptor unneededly. Signed-off-by: Johannes Weiner Cc: Ingo Molnar Cc: Yinghai Lu Cc: Christoph Lameter Cc: Mel Gorman Cc: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/bootmem.c | 14 ++++++-------- 1 file changed, 6 insertions(+), 8 deletions(-) diff --git a/mm/bootmem.c b/mm/bootmem.c index 319a79bce7cd..251c66c5d96a 100644 --- a/mm/bootmem.c +++ b/mm/bootmem.c @@ -87,10 +87,9 @@ static unsigned long __init get_mapsize(bootmem_data_t *bdata) /* * Called once to set up the allocator itself. */ -static unsigned long __init init_bootmem_core(pg_data_t *pgdat, +static unsigned long __init init_bootmem_core(bootmem_data_t *bdata, unsigned long mapstart, unsigned long start, unsigned long end) { - bootmem_data_t *bdata = pgdat->bdata; unsigned long mapsize; mminit_validate_memmodel_limits(&start, &end); @@ -372,11 +371,10 @@ __alloc_bootmem_core(struct bootmem_data *bdata, unsigned long size, return ret; } -static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat) +static unsigned long __init free_all_bootmem_core(bootmem_data_t *bdata) { struct page *page; unsigned long pfn; - bootmem_data_t *bdata = pgdat->bdata; unsigned long i, count; unsigned long idx; unsigned long *map; @@ -441,7 +439,7 @@ static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat) unsigned long __init init_bootmem_node(pg_data_t *pgdat, unsigned long freepfn, unsigned long startpfn, unsigned long endpfn) { - return init_bootmem_core(pgdat, freepfn, startpfn, endpfn); + return init_bootmem_core(pgdat->bdata, freepfn, startpfn, endpfn); } int __init reserve_bootmem_node(pg_data_t *pgdat, unsigned long physaddr, @@ -466,14 +464,14 @@ void __init free_bootmem_node(pg_data_t *pgdat, unsigned long physaddr, unsigned long __init free_all_bootmem_node(pg_data_t *pgdat) { register_page_bootmem_info_node(pgdat); - return free_all_bootmem_core(pgdat); + return free_all_bootmem_core(pgdat->bdata); } unsigned long __init init_bootmem(unsigned long start, unsigned long pages) { max_low_pfn = pages; min_low_pfn = start; - return init_bootmem_core(NODE_DATA(0), start, 0, pages); + return init_bootmem_core(NODE_DATA(0)->bdata, start, 0, pages); } #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE @@ -504,7 +502,7 @@ void __init free_bootmem(unsigned long addr, unsigned long size) unsigned long __init free_all_bootmem(void) { - return free_all_bootmem_core(NODE_DATA(0)); + return free_all_bootmem_core(NODE_DATA(0)->bdata); } void * __init __alloc_bootmem_nopanic(unsigned long size, unsigned long align, From ffc6421f0720f433b5b35b89ff56e998eabff93b Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 23 Jul 2008 21:26:59 -0700 Subject: [PATCH 132/857] mm: unexport __alloc_bootmem_core() This function has no external callers, so unexport it. Also fix its naming inconsistency. Signed-off-by: Johannes Weiner Cc: Ingo Molnar Cc: Yinghai Lu Cc: Christoph Lameter Cc: Mel Gorman Cc: Andy Whitcroft Cc: Mel Gorman Cc: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/bootmem.h | 5 ----- mm/bootmem.c | 24 ++++++++++++------------ 2 files changed, 12 insertions(+), 17 deletions(-) diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h index 2599c741405e..dd8fee6c46d9 100644 --- a/include/linux/bootmem.h +++ b/include/linux/bootmem.h @@ -56,11 +56,6 @@ extern void *__alloc_bootmem_low_node(pg_data_t *pgdat, unsigned long size, unsigned long align, unsigned long goal); -extern void *__alloc_bootmem_core(struct bootmem_data *bdata, - unsigned long size, - unsigned long align, - unsigned long goal, - unsigned long limit); /* * flags for reserve_bootmem (also if CONFIG_HAVE_ARCH_BOOTMEM_NODE, diff --git a/mm/bootmem.c b/mm/bootmem.c index 251c66c5d96a..4bc6ae2fbaa3 100644 --- a/mm/bootmem.c +++ b/mm/bootmem.c @@ -234,9 +234,9 @@ static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr, * * NOTE: This function is _not_ reentrant. */ -void * __init -__alloc_bootmem_core(struct bootmem_data *bdata, unsigned long size, - unsigned long align, unsigned long goal, unsigned long limit) +static void * __init +alloc_bootmem_core(struct bootmem_data *bdata, unsigned long size, + unsigned long align, unsigned long goal, unsigned long limit) { unsigned long areasize, preferred; unsigned long i, start = 0, incr, eidx, end_pfn; @@ -245,7 +245,7 @@ __alloc_bootmem_core(struct bootmem_data *bdata, unsigned long size, void *node_bootmem_map; if (!size) { - printk("__alloc_bootmem_core(): zero-sized request\n"); + printk("alloc_bootmem_core(): zero-sized request\n"); BUG(); } BUG_ON(align & (align-1)); @@ -512,7 +512,7 @@ void * __init __alloc_bootmem_nopanic(unsigned long size, unsigned long align, void *ptr; list_for_each_entry(bdata, &bdata_list, list) { - ptr = __alloc_bootmem_core(bdata, size, align, goal, 0); + ptr = alloc_bootmem_core(bdata, size, align, goal, 0); if (ptr) return ptr; } @@ -540,7 +540,7 @@ void * __init __alloc_bootmem_node(pg_data_t *pgdat, unsigned long size, { void *ptr; - ptr = __alloc_bootmem_core(pgdat->bdata, size, align, goal, 0); + ptr = alloc_bootmem_core(pgdat->bdata, size, align, goal, 0); if (ptr) return ptr; @@ -559,8 +559,8 @@ void * __init alloc_bootmem_section(unsigned long size, goal = PFN_PHYS(pfn); limit = PFN_PHYS(section_nr_to_pfn(section_nr + 1)) - 1; pgdat = NODE_DATA(early_pfn_to_nid(pfn)); - ptr = __alloc_bootmem_core(pgdat->bdata, size, SMP_CACHE_BYTES, goal, - limit); + ptr = alloc_bootmem_core(pgdat->bdata, size, SMP_CACHE_BYTES, goal, + limit); if (!ptr) return NULL; @@ -589,8 +589,8 @@ void * __init __alloc_bootmem_low(unsigned long size, unsigned long align, void *ptr; list_for_each_entry(bdata, &bdata_list, list) { - ptr = __alloc_bootmem_core(bdata, size, align, goal, - ARCH_LOW_ADDRESS_LIMIT); + ptr = alloc_bootmem_core(bdata, size, align, goal, + ARCH_LOW_ADDRESS_LIMIT); if (ptr) return ptr; } @@ -606,6 +606,6 @@ void * __init __alloc_bootmem_low(unsigned long size, unsigned long align, void * __init __alloc_bootmem_low_node(pg_data_t *pgdat, unsigned long size, unsigned long align, unsigned long goal) { - return __alloc_bootmem_core(pgdat->bdata, size, align, goal, - ARCH_LOW_ADDRESS_LIMIT); + return alloc_bootmem_core(pgdat->bdata, size, align, goal, + ARCH_LOW_ADDRESS_LIMIT); } From efe9e77997f6e0306fedc6efa98df491dcf5ecb0 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Wed, 23 Jul 2008 21:27:00 -0700 Subject: [PATCH 133/857] mspec: convert nopfn to fault [akpm@linux-foundation.org: remove unused variable] Signed-off-by: Nick Piggin Acked-by: Jes Sorensen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/mspec.c | 23 ++++++++++++++--------- 1 file changed, 14 insertions(+), 9 deletions(-) diff --git a/drivers/char/mspec.c b/drivers/char/mspec.c index fe2a95b5d3c0..30f095a8c2d4 100644 --- a/drivers/char/mspec.c +++ b/drivers/char/mspec.c @@ -193,25 +193,23 @@ mspec_close(struct vm_area_struct *vma) } /* - * mspec_nopfn + * mspec_fault * * Creates a mspec page and maps it to user space. */ -static unsigned long -mspec_nopfn(struct vm_area_struct *vma, unsigned long address) +static int +mspec_fault(struct vm_area_struct *vma, struct vm_fault *vmf) { unsigned long paddr, maddr; unsigned long pfn; - int index; + pgoff_t index = vmf->pgoff; struct vma_data *vdata = vma->vm_private_data; - BUG_ON(address < vdata->vm_start || address >= vdata->vm_end); - index = (address - vdata->vm_start) >> PAGE_SHIFT; maddr = (volatile unsigned long) vdata->maddr[index]; if (maddr == 0) { maddr = uncached_alloc_page(numa_node_id(), 1); if (maddr == 0) - return NOPFN_OOM; + return VM_FAULT_OOM; spin_lock(&vdata->lock); if (vdata->maddr[index] == 0) { @@ -231,13 +229,20 @@ mspec_nopfn(struct vm_area_struct *vma, unsigned long address) pfn = paddr >> PAGE_SHIFT; - return pfn; + /* + * vm_insert_pfn can fail with -EBUSY, but in that case it will + * be because another thread has installed the pte first, so it + * is no problem. + */ + vm_insert_pfn(vma, (unsigned long)vmf->virtual_address, pfn); + + return VM_FAULT_NOPAGE; } static struct vm_operations_struct mspec_vm_ops = { .open = mspec_open, .close = mspec_close, - .nopfn = mspec_nopfn + .fault = mspec_fault, }; /* From e4048e5dc4aecec670f48ed007a28779f09cebd6 Mon Sep 17 00:00:00 2001 From: KOSAKI Motohiro Date: Wed, 23 Jul 2008 21:27:01 -0700 Subject: [PATCH 134/857] page allocator: inline some __alloc_pages() wrappers Two zonelist patch series rewrote __page_alloc() largely. Now, it is just a wrapper function. Inlining them will save a function call. [akpm@linux-foundation.org: export __alloc_pages_internal] Cc: Lee Schermerhorn Cc: Mel Gorman Signed-off-by: KOSAKI Motohiro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/gfp.h | 21 +++++++++++++++++---- mm/page_alloc.c | 19 ++----------------- 2 files changed, 19 insertions(+), 21 deletions(-) diff --git a/include/linux/gfp.h b/include/linux/gfp.h index b414be387180..f640ed241422 100644 --- a/include/linux/gfp.h +++ b/include/linux/gfp.h @@ -173,11 +173,24 @@ static inline void arch_free_page(struct page *page, int order) { } static inline void arch_alloc_page(struct page *page, int order) { } #endif -extern struct page *__alloc_pages(gfp_t, unsigned int, struct zonelist *); +struct page * +__alloc_pages_internal(gfp_t gfp_mask, unsigned int order, + struct zonelist *zonelist, nodemask_t *nodemask); + +static inline struct page * +__alloc_pages(gfp_t gfp_mask, unsigned int order, + struct zonelist *zonelist) +{ + return __alloc_pages_internal(gfp_mask, order, zonelist, NULL); +} + +static inline struct page * +__alloc_pages_nodemask(gfp_t gfp_mask, unsigned int order, + struct zonelist *zonelist, nodemask_t *nodemask) +{ + return __alloc_pages_internal(gfp_mask, order, zonelist, nodemask); +} -extern struct page * -__alloc_pages_nodemask(gfp_t, unsigned int, - struct zonelist *, nodemask_t *nodemask); static inline struct page *alloc_pages_node(int nid, gfp_t gfp_mask, unsigned int order) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index e089b92cdfff..35b1347d81bb 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1429,7 +1429,7 @@ get_page_from_freelist(gfp_t gfp_mask, nodemask_t *nodemask, unsigned int order, /* * This is the 'heart' of the zoned buddy allocator. */ -static struct page * +struct page * __alloc_pages_internal(gfp_t gfp_mask, unsigned int order, struct zonelist *zonelist, nodemask_t *nodemask) { @@ -1632,22 +1632,7 @@ __alloc_pages_internal(gfp_t gfp_mask, unsigned int order, got_pg: return page; } - -struct page * -__alloc_pages(gfp_t gfp_mask, unsigned int order, - struct zonelist *zonelist) -{ - return __alloc_pages_internal(gfp_mask, order, zonelist, NULL); -} - -struct page * -__alloc_pages_nodemask(gfp_t gfp_mask, unsigned int order, - struct zonelist *zonelist, nodemask_t *nodemask) -{ - return __alloc_pages_internal(gfp_mask, order, zonelist, nodemask); -} - -EXPORT_SYMBOL(__alloc_pages); +EXPORT_SYMBOL(__alloc_pages_internal); /* * Common helper functions. From 4f5ca265788973e3f5a1129a96ee4a9cbf587f2b Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 23 Jul 2008 21:27:02 -0700 Subject: [PATCH 135/857] mm/migrate.c should #include Every file should include the headers containing the externs for its global functions (in this case for sys_move_pages()). Signed-off-by: Adrian Bunk Acked-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/migrate.c | 1 + 1 file changed, 1 insertion(+) diff --git a/mm/migrate.c b/mm/migrate.c index 55bd355d170d..e7d13a708da0 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -30,6 +30,7 @@ #include #include #include +#include #include "internal.h" From c748e1340e0de3fa7fed86f8bdf499be9242afff Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 23 Jul 2008 21:27:03 -0700 Subject: [PATCH 136/857] mm/vmstat.c: proper externs This patch adds proper extern declarations for five variables in include/linux/vmstat.h Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/proc_misc.c | 4 ---- include/linux/vmstat.h | 6 ++++++ kernel/sysctl.c | 2 +- mm/vmstat.c | 1 + 4 files changed, 8 insertions(+), 5 deletions(-) diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c index c652d469dc08..b14f43d25e9e 100644 --- a/fs/proc/proc_misc.c +++ b/fs/proc/proc_misc.c @@ -232,7 +232,6 @@ static int meminfo_read_proc(char *page, char **start, off_t off, #undef K } -extern const struct seq_operations fragmentation_op; static int fragmentation_open(struct inode *inode, struct file *file) { (void)inode; @@ -246,7 +245,6 @@ static const struct file_operations fragmentation_file_operations = { .release = seq_release, }; -extern const struct seq_operations pagetypeinfo_op; static int pagetypeinfo_open(struct inode *inode, struct file *file) { return seq_open(file, &pagetypeinfo_op); @@ -259,7 +257,6 @@ static const struct file_operations pagetypeinfo_file_ops = { .release = seq_release, }; -extern const struct seq_operations zoneinfo_op; static int zoneinfo_open(struct inode *inode, struct file *file) { return seq_open(file, &zoneinfo_op); @@ -356,7 +353,6 @@ static const struct file_operations proc_devinfo_operations = { .release = seq_release, }; -extern const struct seq_operations vmstat_op; static int vmstat_open(struct inode *inode, struct file *file) { return seq_open(file, &vmstat_op); diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h index e83b69346d23..58334d439516 100644 --- a/include/linux/vmstat.h +++ b/include/linux/vmstat.h @@ -44,6 +44,12 @@ enum vm_event_item { PGPGIN, PGPGOUT, PSWPIN, PSWPOUT, NR_VM_EVENT_ITEMS }; +extern const struct seq_operations fragmentation_op; +extern const struct seq_operations pagetypeinfo_op; +extern const struct seq_operations zoneinfo_op; +extern const struct seq_operations vmstat_op; +extern int sysctl_stat_interval; + #ifdef CONFIG_VM_EVENT_COUNTERS /* * Light weight per cpu counter implementation. diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 2a7b9d88706b..1f7b3b76a166 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -43,6 +43,7 @@ #include #include #include +#include #include #include #include @@ -80,7 +81,6 @@ extern int sysctl_drop_caches; extern int percpu_pagelist_fraction; extern int compat_log; extern int maps_protect; -extern int sysctl_stat_interval; extern int latencytop_enabled; extern int sysctl_nr_open_min, sysctl_nr_open_max; #ifdef CONFIG_RCU_TORTURE_TEST diff --git a/mm/vmstat.c b/mm/vmstat.c index c3d4a781802f..b0d08e667ece 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #ifdef CONFIG_VM_EVENT_COUNTERS From 75353bed36cfbbfb55bbde0896bbf5a02d9ba355 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 23 Jul 2008 21:27:03 -0700 Subject: [PATCH 137/857] mm/hugetlb.c: fix duplicate variable It's confusing that set_max_huge_pages() contained two different variables named "ret", and although the code works correctly this should be fixed. The inner of the two variables can simply be removed. Spotted by sparse. Signed-off-by: Adrian Bunk Cc: "KOSAKI Motohiro" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/hugetlb.c | 1 - 1 file changed, 1 deletion(-) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index ab171274ef21..2c5c9ee4220d 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -603,7 +603,6 @@ static unsigned long set_max_huge_pages(unsigned long count) } while (count > persistent_huge_pages) { - int ret; /* * If this allocation races such that we no longer need the * page, free_huge_page will handle it by freeing the page From a969e903a944f69309ee5cc9e7c7b08310d1151e Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 23 Jul 2008 21:27:04 -0700 Subject: [PATCH 138/857] kill generic_file_direct_IO() generic_file_direct_IO is a common helper around the invocation of ->direct_IO. But there's almost nothing shared between the read and write side, so we're better off without this helper. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/filemap.c | 117 ++++++++++++++++++++++----------------------------- 1 file changed, 51 insertions(+), 66 deletions(-) diff --git a/mm/filemap.c b/mm/filemap.c index 65d9d9e2b755..6343f3c841b7 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -42,9 +42,6 @@ #include -static ssize_t -generic_file_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, - loff_t offset, unsigned long nr_segs); /* * Shared mappings implemented 30.11.1994. It's not fully working yet, @@ -1205,8 +1202,11 @@ generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov, goto out; /* skip atime */ size = i_size_read(inode); if (pos < size) { - retval = generic_file_direct_IO(READ, iocb, - iov, pos, nr_segs); + retval = filemap_write_and_wait(mapping); + if (!retval) { + retval = mapping->a_ops->direct_IO(READ, iocb, + iov, pos, nr_segs); + } if (retval > 0) *ppos = pos + retval; } @@ -2004,11 +2004,55 @@ generic_file_direct_write(struct kiocb *iocb, const struct iovec *iov, struct address_space *mapping = file->f_mapping; struct inode *inode = mapping->host; ssize_t written; + size_t write_len; + pgoff_t end; if (count != ocount) *nr_segs = iov_shorten((struct iovec *)iov, *nr_segs, count); - written = generic_file_direct_IO(WRITE, iocb, iov, pos, *nr_segs); + /* + * Unmap all mmappings of the file up-front. + * + * This will cause any pte dirty bits to be propagated into the + * pageframes for the subsequent filemap_write_and_wait(). + */ + write_len = iov_length(iov, *nr_segs); + end = (pos + write_len - 1) >> PAGE_CACHE_SHIFT; + if (mapping_mapped(mapping)) + unmap_mapping_range(mapping, pos, write_len, 0); + + written = filemap_write_and_wait(mapping); + if (written) + goto out; + + /* + * After a write we want buffered reads to be sure to go to disk to get + * the new data. We invalidate clean cached page from the region we're + * about to write. We do this *before* the write so that we can return + * -EIO without clobbering -EIOCBQUEUED from ->direct_IO(). + */ + if (mapping->nrpages) { + written = invalidate_inode_pages2_range(mapping, + pos >> PAGE_CACHE_SHIFT, end); + if (written) + goto out; + } + + written = mapping->a_ops->direct_IO(WRITE, iocb, iov, pos, *nr_segs); + + /* + * Finally, try again to invalidate clean pages which might have been + * cached by non-direct readahead, or faulted in by get_user_pages() + * if the source of the write was an mmap'ed region of the file + * we're writing. Either one is a pretty crazy thing to do, + * so we don't support it 100%. If this invalidation + * fails, tough, the write still worked... + */ + if (mapping->nrpages) { + invalidate_inode_pages2_range(mapping, + pos >> PAGE_CACHE_SHIFT, end); + } + if (written > 0) { loff_t end = pos + written; if (end > i_size_read(inode) && !S_ISBLK(inode->i_mode)) { @@ -2024,6 +2068,7 @@ generic_file_direct_write(struct kiocb *iocb, const struct iovec *iov, * i_mutex is held, which protects generic_osync_inode() from * livelocking. AIO O_DIRECT ops attempt to sync metadata here. */ +out: if ((written >= 0 || written == -EIOCBQUEUED) && ((file->f_flags & O_SYNC) || IS_SYNC(inode))) { int err = generic_osync_inode(inode, mapping, OSYNC_METADATA); @@ -2511,66 +2556,6 @@ ssize_t generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov, } EXPORT_SYMBOL(generic_file_aio_write); -/* - * Called under i_mutex for writes to S_ISREG files. Returns -EIO if something - * went wrong during pagecache shootdown. - */ -static ssize_t -generic_file_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, - loff_t offset, unsigned long nr_segs) -{ - struct file *file = iocb->ki_filp; - struct address_space *mapping = file->f_mapping; - ssize_t retval; - size_t write_len; - pgoff_t end = 0; /* silence gcc */ - - /* - * If it's a write, unmap all mmappings of the file up-front. This - * will cause any pte dirty bits to be propagated into the pageframes - * for the subsequent filemap_write_and_wait(). - */ - if (rw == WRITE) { - write_len = iov_length(iov, nr_segs); - end = (offset + write_len - 1) >> PAGE_CACHE_SHIFT; - if (mapping_mapped(mapping)) - unmap_mapping_range(mapping, offset, write_len, 0); - } - - retval = filemap_write_and_wait(mapping); - if (retval) - goto out; - - /* - * After a write we want buffered reads to be sure to go to disk to get - * the new data. We invalidate clean cached page from the region we're - * about to write. We do this *before* the write so that we can return - * -EIO without clobbering -EIOCBQUEUED from ->direct_IO(). - */ - if (rw == WRITE && mapping->nrpages) { - retval = invalidate_inode_pages2_range(mapping, - offset >> PAGE_CACHE_SHIFT, end); - if (retval) - goto out; - } - - retval = mapping->a_ops->direct_IO(rw, iocb, iov, offset, nr_segs); - - /* - * Finally, try again to invalidate clean pages which might have been - * cached by non-direct readahead, or faulted in by get_user_pages() - * if the source of the write was an mmap'ed region of the file - * we're writing. Either one is a pretty crazy thing to do, - * so we don't support it 100%. If this invalidation - * fails, tough, the write still worked... - */ - if (rw == WRITE && mapping->nrpages) { - invalidate_inode_pages2_range(mapping, offset >> PAGE_CACHE_SHIFT, end); - } -out: - return retval; -} - /** * try_to_release_page() - release old fs-specific metadata on a page * From 0d71d10a4252a3938e6b70189bc776171c02e076 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Wed, 23 Jul 2008 21:27:05 -0700 Subject: [PATCH 139/857] mm: remove nopfn There are no users of nopfn in the tree. Remove it. [hugh@veritas.com: fix build error] Signed-off-by: Nick Piggin Signed-off-by: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 9 ------- mm/memory.c | 67 +++++----------------------------------------- 2 files changed, 7 insertions(+), 69 deletions(-) diff --git a/include/linux/mm.h b/include/linux/mm.h index 2128ef7780c6..eb815cfc1b35 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -166,8 +166,6 @@ struct vm_operations_struct { void (*open)(struct vm_area_struct * area); void (*close)(struct vm_area_struct * area); int (*fault)(struct vm_area_struct *vma, struct vm_fault *vmf); - unsigned long (*nopfn)(struct vm_area_struct *area, - unsigned long address); /* notification that a previously read-only page is about to become * writable, if an error is returned it will cause a SIGBUS */ @@ -674,13 +672,6 @@ static inline int page_mapped(struct page *page) return atomic_read(&(page)->_mapcount) >= 0; } -/* - * Error return values for the *_nopfn functions - */ -#define NOPFN_SIGBUS ((unsigned long) -1) -#define NOPFN_OOM ((unsigned long) -2) -#define NOPFN_REFAULT ((unsigned long) -3) - /* * Different kinds of faults, as returned by handle_mm_fault(). * Used to decide whether a process gets delivered SIGBUS or diff --git a/mm/memory.c b/mm/memory.c index 2302d228fe04..46dbed4b7446 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -1058,11 +1058,9 @@ static inline int use_zero_page(struct vm_area_struct *vma) if (vma->vm_flags & (VM_LOCKED | VM_SHARED)) return 0; /* - * And if we have a fault or a nopfn routine, it's not an - * anonymous region. + * And if we have a fault routine, it's not an anonymous region. */ - return !vma->vm_ops || - (!vma->vm_ops->fault && !vma->vm_ops->nopfn); + return !vma->vm_ops || !vma->vm_ops->fault; } int get_user_pages(struct task_struct *tsk, struct mm_struct *mm, @@ -1338,6 +1336,11 @@ static int insert_pfn(struct vm_area_struct *vma, unsigned long addr, * * This function should only be called from a vm_ops->fault handler, and * in that case the handler should return NULL. + * + * vma cannot be a COW mapping. + * + * As this is called only for pages that do not currently exist, we + * do not need to flush old virtual caches or the TLB. */ int vm_insert_pfn(struct vm_area_struct *vma, unsigned long addr, unsigned long pfn) @@ -2501,59 +2504,6 @@ static int do_linear_fault(struct mm_struct *mm, struct vm_area_struct *vma, return __do_fault(mm, vma, address, pmd, pgoff, flags, orig_pte); } - -/* - * do_no_pfn() tries to create a new page mapping for a page without - * a struct_page backing it - * - * As this is called only for pages that do not currently exist, we - * do not need to flush old virtual caches or the TLB. - * - * We enter with non-exclusive mmap_sem (to exclude vma changes, - * but allow concurrent faults), and pte mapped but not yet locked. - * We return with mmap_sem still held, but pte unmapped and unlocked. - * - * It is expected that the ->nopfn handler always returns the same pfn - * for a given virtual mapping. - * - * Mark this `noinline' to prevent it from bloating the main pagefault code. - */ -static noinline int do_no_pfn(struct mm_struct *mm, struct vm_area_struct *vma, - unsigned long address, pte_t *page_table, pmd_t *pmd, - int write_access) -{ - spinlock_t *ptl; - pte_t entry; - unsigned long pfn; - - pte_unmap(page_table); - BUG_ON(!(vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP))); - BUG_ON((vma->vm_flags & VM_PFNMAP) && is_cow_mapping(vma->vm_flags)); - - pfn = vma->vm_ops->nopfn(vma, address & PAGE_MASK); - - BUG_ON((vma->vm_flags & VM_MIXEDMAP) && pfn_valid(pfn)); - - if (unlikely(pfn == NOPFN_OOM)) - return VM_FAULT_OOM; - else if (unlikely(pfn == NOPFN_SIGBUS)) - return VM_FAULT_SIGBUS; - else if (unlikely(pfn == NOPFN_REFAULT)) - return 0; - - page_table = pte_offset_map_lock(mm, pmd, address, &ptl); - - /* Only go through if we didn't race with anybody else... */ - if (pte_none(*page_table)) { - entry = pfn_pte(pfn, vma->vm_page_prot); - if (write_access) - entry = maybe_mkwrite(pte_mkdirty(entry), vma); - set_pte_at(mm, address, page_table, entry); - } - pte_unmap_unlock(page_table, ptl); - return 0; -} - /* * Fault of a previously existing named mapping. Repopulate the pte * from the encoded file_pte if possible. This enables swappable @@ -2614,9 +2564,6 @@ static inline int handle_pte_fault(struct mm_struct *mm, if (likely(vma->vm_ops->fault)) return do_linear_fault(mm, vma, address, pte, pmd, write_access, entry); - if (unlikely(vma->vm_ops->nopfn)) - return do_no_pfn(mm, vma, address, pte, - pmd, write_access); } return do_anonymous_page(mm, vma, address, pte, pmd, write_access); From 28b2ee20c7cba812b6f2ccf6d722cf86d00a84dc Mon Sep 17 00:00:00 2001 From: Rik van Riel Date: Wed, 23 Jul 2008 21:27:05 -0700 Subject: [PATCH 140/857] access_process_vm device memory infrastructure In order to be able to debug things like the X server and programs using the PPC Cell SPUs, the debugger needs to be able to access device memory through ptrace and /proc/pid/mem. This patch: Add the generic_access_phys access function and put the hooks in place to allow access_process_vm to access device or PPC Cell SPU memory. [riel@redhat.com: Add documentation for the vm_ops->access function] Signed-off-by: Rik van Riel Signed-off-by: Benjamin Herrensmidt Cc: Dave Airlie Cc: Hugh Dickins Cc: Paul Mackerras Cc: Arnd Bergmann Acked-by: Peter Zijlstra Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/filesystems/Locking | 7 ++ arch/Kconfig | 3 + arch/x86/Kconfig | 1 + arch/x86/mm/ioremap.c | 8 ++ include/asm-x86/io_32.h | 2 + include/asm-x86/io_64.h | 2 + include/linux/mm.h | 8 ++ mm/memory.c | 131 ++++++++++++++++++++++++++---- 8 files changed, 144 insertions(+), 18 deletions(-) diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking index 8b22d7d8b991..680fb566b928 100644 --- a/Documentation/filesystems/Locking +++ b/Documentation/filesystems/Locking @@ -510,6 +510,7 @@ prototypes: void (*close)(struct vm_area_struct*); int (*fault)(struct vm_area_struct*, struct vm_fault *); int (*page_mkwrite)(struct vm_area_struct *, struct page *); + int (*access)(struct vm_area_struct *, unsigned long, void*, int, int); locking rules: BKL mmap_sem PageLocked(page) @@ -517,6 +518,7 @@ open: no yes close: no yes fault: no yes page_mkwrite: no yes no +access: no yes ->page_mkwrite() is called when a previously read-only page is about to become writeable. The file system is responsible for @@ -525,6 +527,11 @@ taking to lock out truncate, the page range should be verified to be within i_size. The page mapping should also be checked that it is not NULL. + ->access() is called when get_user_pages() fails in +acces_process_vm(), typically used to debug a process through +/proc/pid/mem or ptrace. This function is needed only for +VM_IO | VM_PFNMAP VMAs. + ================================================================================ Dubious stuff diff --git a/arch/Kconfig b/arch/Kconfig index 4d5ebbc1e72b..6093c0be58b0 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -31,6 +31,9 @@ config KRETPROBES def_bool y depends on KPROBES && HAVE_KRETPROBES +config HAVE_IOREMAP_PROT + def_bool n + config HAVE_KPROBES def_bool n diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 03980cb04291..b2ddfcf01728 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -21,6 +21,7 @@ config X86 select HAVE_UNSTABLE_SCHED_CLOCK select HAVE_IDE select HAVE_OPROFILE + select HAVE_IOREMAP_PROT select HAVE_KPROBES select HAVE_KRETPROBES select HAVE_DYNAMIC_FTRACE diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c index 24c1d3c30186..016f335bbeea 100644 --- a/arch/x86/mm/ioremap.c +++ b/arch/x86/mm/ioremap.c @@ -330,6 +330,14 @@ static void __iomem *ioremap_default(resource_size_t phys_addr, return (void __iomem *)ret; } +void __iomem *ioremap_prot(resource_size_t phys_addr, unsigned long size, + unsigned long prot_val) +{ + return __ioremap_caller(phys_addr, size, (prot_val & _PAGE_CACHE_MASK), + __builtin_return_address(0)); +} +EXPORT_SYMBOL(ioremap_prot); + /** * iounmap - Free a IO remapping * @addr: virtual address from ioremap_* diff --git a/include/asm-x86/io_32.h b/include/asm-x86/io_32.h index 4df44ed54077..e876d89ac156 100644 --- a/include/asm-x86/io_32.h +++ b/include/asm-x86/io_32.h @@ -110,6 +110,8 @@ static inline void *phys_to_virt(unsigned long address) */ extern void __iomem *ioremap_nocache(resource_size_t offset, unsigned long size); extern void __iomem *ioremap_cache(resource_size_t offset, unsigned long size); +extern void __iomem *ioremap_prot(resource_size_t offset, unsigned long size, + unsigned long prot_val); /* * The default ioremap() behavior is non-cached: diff --git a/include/asm-x86/io_64.h b/include/asm-x86/io_64.h index ddd8058a5026..22995c5c5adc 100644 --- a/include/asm-x86/io_64.h +++ b/include/asm-x86/io_64.h @@ -175,6 +175,8 @@ extern void early_iounmap(void *addr, unsigned long size); */ extern void __iomem *ioremap_nocache(resource_size_t offset, unsigned long size); extern void __iomem *ioremap_cache(resource_size_t offset, unsigned long size); +extern void __iomem *ioremap_prot(resource_size_t offset, unsigned long size, + unsigned long prot_val); /* * The default ioremap() behavior is non-cached: diff --git a/include/linux/mm.h b/include/linux/mm.h index eb815cfc1b35..5c7f8f64f70e 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -170,6 +170,12 @@ struct vm_operations_struct { /* notification that a previously read-only page is about to become * writable, if an error is returned it will cause a SIGBUS */ int (*page_mkwrite)(struct vm_area_struct *vma, struct page *page); + + /* called by access_process_vm when get_user_pages() fails, typically + * for use by special VMAs that can switch between memory and hardware + */ + int (*access)(struct vm_area_struct *vma, unsigned long addr, + void *buf, int len, int write); #ifdef CONFIG_NUMA /* * set_policy() op must add a reference to any non-NULL @new mempolicy @@ -771,6 +777,8 @@ int copy_page_range(struct mm_struct *dst, struct mm_struct *src, struct vm_area_struct *vma); void unmap_mapping_range(struct address_space *mapping, loff_t const holebegin, loff_t const holelen, int even_cows); +int generic_access_phys(struct vm_area_struct *vma, unsigned long addr, + void *buf, int len, int write); static inline void unmap_shared_mapping_range(struct address_space *mapping, loff_t const holebegin, loff_t const holelen) diff --git a/mm/memory.c b/mm/memory.c index 46dbed4b7446..87350321e66f 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -2751,6 +2751,86 @@ int in_gate_area_no_task(unsigned long addr) #endif /* __HAVE_ARCH_GATE_AREA */ +#ifdef CONFIG_HAVE_IOREMAP_PROT +static resource_size_t follow_phys(struct vm_area_struct *vma, + unsigned long address, unsigned int flags, + unsigned long *prot) +{ + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *ptep, pte; + spinlock_t *ptl; + resource_size_t phys_addr = 0; + struct mm_struct *mm = vma->vm_mm; + + VM_BUG_ON(!(vma->vm_flags & (VM_IO | VM_PFNMAP))); + + pgd = pgd_offset(mm, address); + if (pgd_none(*pgd) || unlikely(pgd_bad(*pgd))) + goto no_page_table; + + pud = pud_offset(pgd, address); + if (pud_none(*pud) || unlikely(pud_bad(*pud))) + goto no_page_table; + + pmd = pmd_offset(pud, address); + if (pmd_none(*pmd) || unlikely(pmd_bad(*pmd))) + goto no_page_table; + + /* We cannot handle huge page PFN maps. Luckily they don't exist. */ + if (pmd_huge(*pmd)) + goto no_page_table; + + ptep = pte_offset_map_lock(mm, pmd, address, &ptl); + if (!ptep) + goto out; + + pte = *ptep; + if (!pte_present(pte)) + goto unlock; + if ((flags & FOLL_WRITE) && !pte_write(pte)) + goto unlock; + phys_addr = pte_pfn(pte); + phys_addr <<= PAGE_SHIFT; /* Shift here to avoid overflow on PAE */ + + *prot = pgprot_val(pte_pgprot(pte)); + +unlock: + pte_unmap_unlock(ptep, ptl); +out: + return phys_addr; +no_page_table: + return 0; +} + +int generic_access_phys(struct vm_area_struct *vma, unsigned long addr, + void *buf, int len, int write) +{ + resource_size_t phys_addr; + unsigned long prot = 0; + void *maddr; + int offset = addr & (PAGE_SIZE-1); + + if (!(vma->vm_flags & (VM_IO | VM_PFNMAP))) + return -EINVAL; + + phys_addr = follow_phys(vma, addr, write, &prot); + + if (!phys_addr) + return -EINVAL; + + maddr = ioremap_prot(phys_addr, PAGE_SIZE, prot); + if (write) + memcpy_toio(maddr + offset, buf, len); + else + memcpy_fromio(buf, maddr + offset, len); + iounmap(maddr); + + return len; +} +#endif + /* * Access another process' address space. * Source/target buffer must be kernel space, @@ -2760,7 +2840,6 @@ int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, in { struct mm_struct *mm; struct vm_area_struct *vma; - struct page *page; void *old_buf = buf; mm = get_task_mm(tsk); @@ -2772,28 +2851,44 @@ int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, in while (len) { int bytes, ret, offset; void *maddr; + struct page *page = NULL; ret = get_user_pages(tsk, mm, addr, 1, write, 1, &page, &vma); - if (ret <= 0) - break; - - bytes = len; - offset = addr & (PAGE_SIZE-1); - if (bytes > PAGE_SIZE-offset) - bytes = PAGE_SIZE-offset; - - maddr = kmap(page); - if (write) { - copy_to_user_page(vma, page, addr, - maddr + offset, buf, bytes); - set_page_dirty_lock(page); + if (ret <= 0) { + /* + * Check if this is a VM_IO | VM_PFNMAP VMA, which + * we can access using slightly different code. + */ +#ifdef CONFIG_HAVE_IOREMAP_PROT + vma = find_vma(mm, addr); + if (!vma) + break; + if (vma->vm_ops && vma->vm_ops->access) + ret = vma->vm_ops->access(vma, addr, buf, + len, write); + if (ret <= 0) +#endif + break; + bytes = ret; } else { - copy_from_user_page(vma, page, addr, - buf, maddr + offset, bytes); + bytes = len; + offset = addr & (PAGE_SIZE-1); + if (bytes > PAGE_SIZE-offset) + bytes = PAGE_SIZE-offset; + + maddr = kmap(page); + if (write) { + copy_to_user_page(vma, page, addr, + maddr + offset, buf, bytes); + set_page_dirty_lock(page); + } else { + copy_from_user_page(vma, page, addr, + buf, maddr + offset, bytes); + } + kunmap(page); + page_cache_release(page); } - kunmap(page); - page_cache_release(page); len -= bytes; buf += bytes; addr += bytes; From 7ae8ed5053a39082d224a3f48409e016baca9c16 Mon Sep 17 00:00:00 2001 From: Rik van Riel Date: Wed, 23 Jul 2008 21:27:07 -0700 Subject: [PATCH 141/857] use generic_access_phys for /dev/mem mappings Use generic_access_phys as the access_process_vm access function for /dev/mem mappings. This makes it possible to debug the X server. [akpm@linux-foundation.org: repair all the architectures which broke] Signed-off-by: Rik van Riel Cc: Benjamin Herrensmidt Cc: Dave Airlie Cc: Hugh Dickins Cc: Paul Mackerras Cc: Arnd Bergmann Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/pci/i386.c | 1 + drivers/char/mem.c | 5 ++++- 2 files changed, 5 insertions(+), 1 deletion(-) diff --git a/arch/x86/pci/i386.c b/arch/x86/pci/i386.c index 2aafb67dc5f1..a09505806b82 100644 --- a/arch/x86/pci/i386.c +++ b/arch/x86/pci/i386.c @@ -280,6 +280,7 @@ static void pci_track_mmap_page_range(struct vm_area_struct *vma) static struct vm_operations_struct pci_mmap_ops = { .open = pci_track_mmap_page_range, .close = pci_unmap_page_range, + .access = generic_access_phys, }; int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma, diff --git a/drivers/char/mem.c b/drivers/char/mem.c index c2dba82eb5f7..672b08e694d0 100644 --- a/drivers/char/mem.c +++ b/drivers/char/mem.c @@ -327,7 +327,10 @@ static void mmap_mem_close(struct vm_area_struct *vma) static struct vm_operations_struct mmap_mem_ops = { .open = mmap_mem_open, - .close = mmap_mem_close + .close = mmap_mem_close, +#ifdef CONFIG_HAVE_IOREMAP_PROT + .access = generic_access_phys +#endif }; static int mmap_mem(struct file * file, struct vm_area_struct * vma) From a1f242ff460e4b50a045fa237c3c56cce9eabf83 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Wed, 23 Jul 2008 21:27:08 -0700 Subject: [PATCH 142/857] powerpc ioremap_prot This adds ioremap_prot and pte_pgprot() so that one can extract protection bits from a PTE and use them to ioremap_prot() (in order to support ptrace of VM_IO | VM_PFNMAP as per Rik's patch). This moves a couple of flag checks around in the ioremap implementations of arch/powerpc. There's a side effect of allowing non-cacheable and non-guarded mappings on ppc32 which before would always have _PAGE_GUARDED set whenever _PAGE_NO_CACHE is. (standard ioremap will still set _PAGE_GUARDED, but ioremap_prot will be capable of setting such a non guarded mapping). Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Rik van Riel Cc: Dave Airlie Cc: Hugh Dickins Cc: Paul Mackerras Cc: Arnd Bergmann Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/powerpc/Kconfig | 1 + arch/powerpc/mm/pgtable_32.c | 22 ++++++++++++++++------ arch/powerpc/mm/pgtable_64.c | 16 ++++++++++++++++ include/asm-powerpc/io.h | 5 ++++- include/asm-powerpc/pgtable-4k.h | 3 +++ include/asm-powerpc/pgtable-ppc32.h | 16 ++++++++++++++++ include/asm-powerpc/pgtable-ppc64.h | 8 ++++++++ 7 files changed, 64 insertions(+), 7 deletions(-) diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 4d7e2ba10bae..a487671c282f 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -111,6 +111,7 @@ config PPC select HAVE_DYNAMIC_FTRACE select HAVE_FTRACE select HAVE_IDE + select HAVE_IOREMAP_PROT select HAVE_KPROBES select HAVE_ARCH_KGDB select HAVE_KRETPROBES diff --git a/arch/powerpc/mm/pgtable_32.c b/arch/powerpc/mm/pgtable_32.c index c7584072dfcc..2001abdb1912 100644 --- a/arch/powerpc/mm/pgtable_32.c +++ b/arch/powerpc/mm/pgtable_32.c @@ -145,13 +145,20 @@ void pte_free(struct mm_struct *mm, pgtable_t ptepage) void __iomem * ioremap(phys_addr_t addr, unsigned long size) { - return __ioremap(addr, size, _PAGE_NO_CACHE); + return __ioremap(addr, size, _PAGE_NO_CACHE | _PAGE_GUARDED); } EXPORT_SYMBOL(ioremap); void __iomem * ioremap_flags(phys_addr_t addr, unsigned long size, unsigned long flags) { + /* writeable implies dirty for kernel addresses */ + if (flags & _PAGE_RW) + flags |= _PAGE_DIRTY | _PAGE_HWWRITE; + + /* we don't want to let _PAGE_USER and _PAGE_EXEC leak out */ + flags &= ~(_PAGE_USER | _PAGE_EXEC | _PAGE_HWEXEC); + return __ioremap(addr, size, flags); } EXPORT_SYMBOL(ioremap_flags); @@ -163,6 +170,14 @@ __ioremap(phys_addr_t addr, unsigned long size, unsigned long flags) phys_addr_t p; int err; + /* Make sure we have the base flags */ + if ((flags & _PAGE_PRESENT) == 0) + flags |= _PAGE_KERNEL; + + /* Non-cacheable page cannot be coherent */ + if (flags & _PAGE_NO_CACHE) + flags &= ~_PAGE_COHERENT; + /* * Choose an address to map it to. * Once the vmalloc system is running, we use it. @@ -219,11 +234,6 @@ __ioremap(phys_addr_t addr, unsigned long size, unsigned long flags) v = (ioremap_bot -= size); } - if ((flags & _PAGE_PRESENT) == 0) - flags |= _PAGE_KERNEL; - if (flags & _PAGE_NO_CACHE) - flags |= _PAGE_GUARDED; - /* * Should check if it is a candidate for a BAT mapping */ diff --git a/arch/powerpc/mm/pgtable_64.c b/arch/powerpc/mm/pgtable_64.c index 3ef0ad2f9ca0..365e61ae5dbc 100644 --- a/arch/powerpc/mm/pgtable_64.c +++ b/arch/powerpc/mm/pgtable_64.c @@ -107,9 +107,18 @@ void __iomem * __ioremap_at(phys_addr_t pa, void *ea, unsigned long size, { unsigned long i; + /* Make sure we have the base flags */ if ((flags & _PAGE_PRESENT) == 0) flags |= pgprot_val(PAGE_KERNEL); + /* Non-cacheable page cannot be coherent */ + if (flags & _PAGE_NO_CACHE) + flags &= ~_PAGE_COHERENT; + + /* We don't support the 4K PFN hack with ioremap */ + if (flags & _PAGE_4K_PFN) + return NULL; + WARN_ON(pa & ~PAGE_MASK); WARN_ON(((unsigned long)ea) & ~PAGE_MASK); WARN_ON(size & ~PAGE_MASK); @@ -190,6 +199,13 @@ void __iomem * ioremap(phys_addr_t addr, unsigned long size) void __iomem * ioremap_flags(phys_addr_t addr, unsigned long size, unsigned long flags) { + /* writeable implies dirty for kernel addresses */ + if (flags & _PAGE_RW) + flags |= _PAGE_DIRTY; + + /* we don't want to let _PAGE_USER and _PAGE_EXEC leak out */ + flags &= ~(_PAGE_USER | _PAGE_EXEC); + if (ppc_md.ioremap) return ppc_md.ioremap(addr, size, flags); return __ioremap(addr, size, flags); diff --git a/include/asm-powerpc/io.h b/include/asm-powerpc/io.h index 8b627823f5f9..77c7fa025e65 100644 --- a/include/asm-powerpc/io.h +++ b/include/asm-powerpc/io.h @@ -617,7 +617,8 @@ static inline void iosync(void) * and can be hooked by the platform via ppc_md * * * ioremap_flags allows to specify the page flags as an argument and can - * also be hooked by the platform via ppc_md + * also be hooked by the platform via ppc_md. ioremap_prot is the exact + * same thing as ioremap_flags. * * * ioremap_nocache is identical to ioremap * @@ -639,6 +640,8 @@ extern void __iomem *ioremap(phys_addr_t address, unsigned long size); extern void __iomem *ioremap_flags(phys_addr_t address, unsigned long size, unsigned long flags); #define ioremap_nocache(addr, size) ioremap((addr), (size)) +#define ioremap_prot(addr, size, prot) ioremap_flags((addr), (size), (prot)) + extern void iounmap(volatile void __iomem *addr); extern void __iomem *__ioremap(phys_addr_t, unsigned long size, diff --git a/include/asm-powerpc/pgtable-4k.h b/include/asm-powerpc/pgtable-4k.h index fd2090dc1dce..c9601dfb4a1e 100644 --- a/include/asm-powerpc/pgtable-4k.h +++ b/include/asm-powerpc/pgtable-4k.h @@ -51,6 +51,9 @@ #define _PAGE_HPTEFLAGS (_PAGE_BUSY | _PAGE_HASHPTE | \ _PAGE_SECONDARY | _PAGE_GROUP_IX) +/* There is no 4K PFN hack on 4K pages */ +#define _PAGE_4K_PFN 0 + /* PAGE_MASK gives the right answer below, but only by accident */ /* It should be preserving the high 48 bits and then specifically */ /* preserving _PAGE_SECONDARY | _PAGE_GROUP_IX */ diff --git a/include/asm-powerpc/pgtable-ppc32.h b/include/asm-powerpc/pgtable-ppc32.h index 3a96d001cb75..bdbab72f3ebc 100644 --- a/include/asm-powerpc/pgtable-ppc32.h +++ b/include/asm-powerpc/pgtable-ppc32.h @@ -395,6 +395,12 @@ extern int icache_44x_need_flush; #ifndef _PAGE_EXEC #define _PAGE_EXEC 0 #endif +#ifndef _PAGE_ENDIAN +#define _PAGE_ENDIAN 0 +#endif +#ifndef _PAGE_COHERENT +#define _PAGE_COHERENT 0 +#endif #ifndef _PMD_PRESENT_MASK #define _PMD_PRESENT_MASK _PMD_PRESENT #endif @@ -405,6 +411,12 @@ extern int icache_44x_need_flush; #define _PAGE_CHG_MASK (PAGE_MASK | _PAGE_ACCESSED | _PAGE_DIRTY) + +#define PAGE_PROT_BITS __pgprot(_PAGE_GUARDED | _PAGE_COHERENT | _PAGE_NO_CACHE | \ + _PAGE_WRITETHRU | _PAGE_ENDIAN | \ + _PAGE_USER | _PAGE_ACCESSED | \ + _PAGE_RW | _PAGE_HWWRITE | _PAGE_DIRTY | \ + _PAGE_EXEC | _PAGE_HWEXEC) /* * Note: the _PAGE_COHERENT bit automatically gets set in the hardware * PTE if CONFIG_SMP is defined (hash_page does this); there is no need @@ -538,6 +550,10 @@ static inline pte_t pte_mkyoung(pte_t pte) { pte_val(pte) |= _PAGE_ACCESSED; return pte; } static inline pte_t pte_mkspecial(pte_t pte) { return pte; } +static inline unsigned long pte_pgprot(pte_t pte) +{ + return __pgprot(pte_val(pte)) & PAGE_PROT_BITS; +} static inline pte_t pte_modify(pte_t pte, pgprot_t newprot) { diff --git a/include/asm-powerpc/pgtable-ppc64.h b/include/asm-powerpc/pgtable-ppc64.h index ab98a9c80b28..ba8000352b9a 100644 --- a/include/asm-powerpc/pgtable-ppc64.h +++ b/include/asm-powerpc/pgtable-ppc64.h @@ -117,6 +117,10 @@ #define PAGE_AGP __pgprot(_PAGE_BASE | _PAGE_WRENABLE | _PAGE_NO_CACHE) #define HAVE_PAGE_AGP +#define PAGE_PROT_BITS __pgprot(_PAGE_GUARDED | _PAGE_COHERENT | \ + _PAGE_NO_CACHE | _PAGE_WRITETHRU | \ + _PAGE_4K_PFN | _PAGE_RW | _PAGE_USER | \ + _PAGE_ACCESSED | _PAGE_DIRTY | _PAGE_EXEC) /* PTEIDX nibble */ #define _PTEIDX_SECONDARY 0x8 #define _PTEIDX_GROUP_IX 0x7 @@ -262,6 +266,10 @@ static inline pte_t pte_mkhuge(pte_t pte) { return pte; } static inline pte_t pte_mkspecial(pte_t pte) { return pte; } +static inline unsigned long pte_pgprot(pte_t pte) +{ + return __pgprot(pte_val(pte)) & PAGE_PROT_BITS; +} /* Atomic PTE updates */ static inline unsigned long pte_update(struct mm_struct *mm, From a352894d07059649398c4769dc8b645e1a1dad88 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Wed, 23 Jul 2008 21:27:09 -0700 Subject: [PATCH 143/857] spufs: use new vm_ops->access to allow local state access from gdb This uses the new vm_ops->access to allow gdb to access the SPU local store. We currently prevent access to problem state registers, this can be done later if really needed but it's safer not to. [akpm@linux-foundation.org: fix typo] Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Rik van Riel Cc: Dave Airlie Cc: Hugh Dickins Cc: Paul Mackerras Cc: Arnd Bergmann Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/powerpc/platforms/cell/spufs/file.c | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) diff --git a/arch/powerpc/platforms/cell/spufs/file.c b/arch/powerpc/platforms/cell/spufs/file.c index 99c73066b82f..010a51f59796 100644 --- a/arch/powerpc/platforms/cell/spufs/file.c +++ b/arch/powerpc/platforms/cell/spufs/file.c @@ -288,9 +288,32 @@ spufs_mem_mmap_fault(struct vm_area_struct *vma, struct vm_fault *vmf) return VM_FAULT_NOPAGE; } +static int spufs_mem_mmap_access(struct vm_area_struct *vma, + unsigned long address, + void *buf, int len, int write) +{ + struct spu_context *ctx = vma->vm_file->private_data; + unsigned long offset = address - vma->vm_start; + char *local_store; + + if (write && !(vma->vm_flags & VM_WRITE)) + return -EACCES; + if (spu_acquire(ctx)) + return -EINTR; + if ((offset + len) > vma->vm_end) + len = vma->vm_end - offset; + local_store = ctx->ops->get_ls(ctx); + if (write) + memcpy_toio(local_store + offset, buf, len); + else + memcpy_fromio(buf, local_store + offset, len); + spu_release(ctx); + return len; +} static struct vm_operations_struct spufs_mem_mmap_vmops = { .fault = spufs_mem_mmap_fault, + .access = spufs_mem_mmap_access, }; static int spufs_mem_mmap(struct file *file, struct vm_area_struct *vma) From 42b7772812d15b86543a23b82bd6070eef9a08b1 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Wed, 23 Jul 2008 21:27:10 -0700 Subject: [PATCH 144/857] mm: remove double indirection on tlb parameter to free_pgd_range() & Co The double indirection here is not needed anywhere and hence (at least) confusing. Signed-off-by: Jan Beulich Cc: Hugh Dickins Cc: Nick Piggin Cc: Christoph Lameter Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: "Luck, Tony" Cc: Paul Mundt Cc: "David S. Miller" Acked-by: Jeremy Fitzhardinge Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/mm/hugetlbpage.c | 2 +- arch/powerpc/mm/hugetlbpage.c | 8 ++++---- fs/exec.c | 4 ++-- include/asm-ia64/hugetlb.h | 2 +- include/asm-powerpc/hugetlb.h | 2 +- include/asm-sh/hugetlb.h | 2 +- include/asm-sparc/hugetlb.h | 2 +- include/asm-x86/hugetlb.h | 2 +- include/linux/mm.h | 4 +--- mm/internal.h | 3 +++ mm/memory.c | 10 ++++++---- mm/mmap.c | 6 ++++-- 12 files changed, 26 insertions(+), 21 deletions(-) diff --git a/arch/ia64/mm/hugetlbpage.c b/arch/ia64/mm/hugetlbpage.c index d3ce8f3bcaa6..cd49e2860eef 100644 --- a/arch/ia64/mm/hugetlbpage.c +++ b/arch/ia64/mm/hugetlbpage.c @@ -112,7 +112,7 @@ follow_huge_pmd(struct mm_struct *mm, unsigned long address, pmd_t *pmd, int wri return NULL; } -void hugetlb_free_pgd_range(struct mmu_gather **tlb, +void hugetlb_free_pgd_range(struct mmu_gather *tlb, unsigned long addr, unsigned long end, unsigned long floor, unsigned long ceiling) { diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c index 0d12fba31bc5..1a96cc891cf5 100644 --- a/arch/powerpc/mm/hugetlbpage.c +++ b/arch/powerpc/mm/hugetlbpage.c @@ -255,7 +255,7 @@ static void hugetlb_free_pud_range(struct mmu_gather *tlb, pgd_t *pgd, * * Must be called with pagetable lock held. */ -void hugetlb_free_pgd_range(struct mmu_gather **tlb, +void hugetlb_free_pgd_range(struct mmu_gather *tlb, unsigned long addr, unsigned long end, unsigned long floor, unsigned long ceiling) { @@ -315,13 +315,13 @@ void hugetlb_free_pgd_range(struct mmu_gather **tlb, return; start = addr; - pgd = pgd_offset((*tlb)->mm, addr); + pgd = pgd_offset(tlb->mm, addr); do { - BUG_ON(get_slice_psize((*tlb)->mm, addr) != mmu_huge_psize); + BUG_ON(get_slice_psize(tlb->mm, addr) != mmu_huge_psize); next = pgd_addr_end(addr, end); if (pgd_none_or_clear_bad(pgd)) continue; - hugetlb_free_pud_range(*tlb, pgd, addr, next, floor, ceiling); + hugetlb_free_pud_range(tlb, pgd, addr, next, floor, ceiling); } while (pgd++, addr = next, addr != end); } diff --git a/fs/exec.c b/fs/exec.c index fd9234379e8d..190ed1f92774 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -541,7 +541,7 @@ static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift) /* * when the old and new regions overlap clear from new_end. */ - free_pgd_range(&tlb, new_end, old_end, new_end, + free_pgd_range(tlb, new_end, old_end, new_end, vma->vm_next ? vma->vm_next->vm_start : 0); } else { /* @@ -550,7 +550,7 @@ static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift) * have constraints on va-space that make this illegal (IA64) - * for the others its just a little faster. */ - free_pgd_range(&tlb, old_start, old_end, new_end, + free_pgd_range(tlb, old_start, old_end, new_end, vma->vm_next ? vma->vm_next->vm_start : 0); } tlb_finish_mmu(tlb, new_end, old_end); diff --git a/include/asm-ia64/hugetlb.h b/include/asm-ia64/hugetlb.h index f28a9701f1cf..e9d1e5e2382d 100644 --- a/include/asm-ia64/hugetlb.h +++ b/include/asm-ia64/hugetlb.h @@ -4,7 +4,7 @@ #include -void hugetlb_free_pgd_range(struct mmu_gather **tlb, unsigned long addr, +void hugetlb_free_pgd_range(struct mmu_gather *tlb, unsigned long addr, unsigned long end, unsigned long floor, unsigned long ceiling); diff --git a/include/asm-powerpc/hugetlb.h b/include/asm-powerpc/hugetlb.h index be32ff02f4a0..0a37aa5ecaa5 100644 --- a/include/asm-powerpc/hugetlb.h +++ b/include/asm-powerpc/hugetlb.h @@ -7,7 +7,7 @@ int is_hugepage_only_range(struct mm_struct *mm, unsigned long addr, unsigned long len); -void hugetlb_free_pgd_range(struct mmu_gather **tlb, unsigned long addr, +void hugetlb_free_pgd_range(struct mmu_gather *tlb, unsigned long addr, unsigned long end, unsigned long floor, unsigned long ceiling); diff --git a/include/asm-sh/hugetlb.h b/include/asm-sh/hugetlb.h index 02402303d89b..fb30018938c7 100644 --- a/include/asm-sh/hugetlb.h +++ b/include/asm-sh/hugetlb.h @@ -26,7 +26,7 @@ static inline int prepare_hugepage_range(unsigned long addr, unsigned long len) static inline void hugetlb_prefault_arch_hook(struct mm_struct *mm) { } -static inline void hugetlb_free_pgd_range(struct mmu_gather **tlb, +static inline void hugetlb_free_pgd_range(struct mmu_gather *tlb, unsigned long addr, unsigned long end, unsigned long floor, unsigned long ceiling) diff --git a/include/asm-sparc/hugetlb.h b/include/asm-sparc/hugetlb.h index 412af58926a0..aeb92374ca3d 100644 --- a/include/asm-sparc/hugetlb.h +++ b/include/asm-sparc/hugetlb.h @@ -31,7 +31,7 @@ static inline int prepare_hugepage_range(unsigned long addr, unsigned long len) return 0; } -static inline void hugetlb_free_pgd_range(struct mmu_gather **tlb, +static inline void hugetlb_free_pgd_range(struct mmu_gather *tlb, unsigned long addr, unsigned long end, unsigned long floor, unsigned long ceiling) diff --git a/include/asm-x86/hugetlb.h b/include/asm-x86/hugetlb.h index 14171a4924f6..7eed6e0883bf 100644 --- a/include/asm-x86/hugetlb.h +++ b/include/asm-x86/hugetlb.h @@ -26,7 +26,7 @@ static inline int prepare_hugepage_range(unsigned long addr, unsigned long len) static inline void hugetlb_prefault_arch_hook(struct mm_struct *mm) { } -static inline void hugetlb_free_pgd_range(struct mmu_gather **tlb, +static inline void hugetlb_free_pgd_range(struct mmu_gather *tlb, unsigned long addr, unsigned long end, unsigned long floor, unsigned long ceiling) diff --git a/include/linux/mm.h b/include/linux/mm.h index 5c7f8f64f70e..f8071097302a 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -769,10 +769,8 @@ struct mm_walk { int walk_page_range(unsigned long addr, unsigned long end, struct mm_walk *walk); -void free_pgd_range(struct mmu_gather **tlb, unsigned long addr, +void free_pgd_range(struct mmu_gather *tlb, unsigned long addr, unsigned long end, unsigned long floor, unsigned long ceiling); -void free_pgtables(struct mmu_gather **tlb, struct vm_area_struct *start_vma, - unsigned long floor, unsigned long ceiling); int copy_page_range(struct mm_struct *dst, struct mm_struct *src, struct vm_area_struct *vma); void unmap_mapping_range(struct address_space *mapping, diff --git a/mm/internal.h b/mm/internal.h index 50807e12490e..858ad01864dc 100644 --- a/mm/internal.h +++ b/mm/internal.h @@ -13,6 +13,9 @@ #include +void free_pgtables(struct mmu_gather *tlb, struct vm_area_struct *start_vma, + unsigned long floor, unsigned long ceiling); + static inline void set_page_count(struct page *page, int v) { atomic_set(&page->_count, v); diff --git a/mm/memory.c b/mm/memory.c index 87350321e66f..82f3f1c5cf17 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -61,6 +61,8 @@ #include #include +#include "internal.h" + #ifndef CONFIG_NEED_MULTIPLE_NODES /* use the per-pgdat data instead for discontigmem - mbligh */ unsigned long max_mapnr; @@ -211,7 +213,7 @@ static inline void free_pud_range(struct mmu_gather *tlb, pgd_t *pgd, * * Must be called with pagetable lock held. */ -void free_pgd_range(struct mmu_gather **tlb, +void free_pgd_range(struct mmu_gather *tlb, unsigned long addr, unsigned long end, unsigned long floor, unsigned long ceiling) { @@ -262,16 +264,16 @@ void free_pgd_range(struct mmu_gather **tlb, return; start = addr; - pgd = pgd_offset((*tlb)->mm, addr); + pgd = pgd_offset(tlb->mm, addr); do { next = pgd_addr_end(addr, end); if (pgd_none_or_clear_bad(pgd)) continue; - free_pud_range(*tlb, pgd, addr, next, floor, ceiling); + free_pud_range(tlb, pgd, addr, next, floor, ceiling); } while (pgd++, addr = next, addr != end); } -void free_pgtables(struct mmu_gather **tlb, struct vm_area_struct *vma, +void free_pgtables(struct mmu_gather *tlb, struct vm_area_struct *vma, unsigned long floor, unsigned long ceiling) { while (vma) { diff --git a/mm/mmap.c b/mm/mmap.c index 1d102b956fd8..75e0d0673d78 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -32,6 +32,8 @@ #include #include +#include "internal.h" + #ifndef arch_mmap_check #define arch_mmap_check(addr, len, flags) (0) #endif @@ -1763,7 +1765,7 @@ static void unmap_region(struct mm_struct *mm, update_hiwater_rss(mm); unmap_vmas(&tlb, vma, start, end, &nr_accounted, NULL); vm_unacct_memory(nr_accounted); - free_pgtables(&tlb, vma, prev? prev->vm_end: FIRST_USER_ADDRESS, + free_pgtables(tlb, vma, prev? prev->vm_end: FIRST_USER_ADDRESS, next? next->vm_start: 0); tlb_finish_mmu(tlb, start, end); } @@ -2063,7 +2065,7 @@ void exit_mmap(struct mm_struct *mm) /* Use -1 here to ensure all VMAs in the mm are unmapped */ end = unmap_vmas(&tlb, vma, 0, -1, &nr_accounted, NULL); vm_unacct_memory(nr_accounted); - free_pgtables(&tlb, vma, FIRST_USER_ADDRESS, 0); + free_pgtables(tlb, vma, FIRST_USER_ADDRESS, 0); tlb_finish_mmu(tlb, 0, end); /* From 3c82d0ce2c4f642b2f24ef98707a030543b06b90 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:27:11 -0700 Subject: [PATCH 145/857] buddy: clarify comments describing buddy merge In __free_one_page(), the comment "Move the buddy up one level" appears attached to the break and by implication when the break is taken we are moving it up one level: if (!page_is_buddy(page, buddy, order)) break; /* Move the buddy up one level. */ In reality the inverse is true, we break out when we can no longer merge this page with its buddy. Looking back into pre-history (into the full git history) it appears that these two lines accidentally got joined as part of another change. Move the comment down where it belongs below the if and clarify its language. Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 35b1347d81bb..24aa3d1b9d96 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -432,8 +432,9 @@ static inline void __free_one_page(struct page *page, buddy = __page_find_buddy(page, page_idx, order); if (!page_is_buddy(page, buddy, order)) - break; /* Move the buddy up one level. */ + break; + /* Our buddy is free, merge with it and move up one order. */ list_del(&buddy->lru); zone->free_area[order].nr_free--; rmv_page_order(buddy); From da3bbdd4632c0171406b2677e31494afa5bde2f8 Mon Sep 17 00:00:00 2001 From: Kentaro Makita Date: Wed, 23 Jul 2008 21:27:13 -0700 Subject: [PATCH 146/857] fix soft lock up at NFS mount via per-SB LRU-list of unused dentries [Summary] Split LRU-list of unused dentries to one per superblock to avoid soft lock up during NFS mounts and remounting of any filesystem. Previously I posted here: http://lkml.org/lkml/2008/3/5/590 [Descriptions] - background dentry_unused is a list of dentries which are not referenced. dentry_unused grows up when references on directories or files are released. This list can be very long if there is huge free memory. - the problem When shrink_dcache_sb() is called, it scans all dentry_unused linearly under spin_lock(), and if dentry->d_sb is differnt from given superblock, scan next dentry. This scan costs very much if there are many entries, and very ineffective if there are many superblocks. IOW, When we need to shrink unused dentries on one dentry, but scans unused dentries on all superblocks in the system. For example, we scan 500 dentries to unmount a filesystem, but scans 1,000,000 or more unused dentries on other superblocks. In our case , At mounting NFS*, shrink_dcache_sb() is called to shrink unused dentries on NFS, but scans 100,000,000 unused dentries on superblocks in the system such as local ext3 filesystems. I hear NFS mounting took 1 min on some system in use. * : NFS uses virtual filesystem in rpc layer, so NFS is affected by this problem. 100,000,000 is possible number on large systems. Per-superblock LRU of unused dentried can reduce the cost in reasonable manner. - How to fix I found this problem is solved by David Chinner's "Per-superblock unused dentry LRU lists V3"(1), so I rebase it and add some fix to reclaim with fairness, which is in Andrew Morton's comments(2). 1) http://lkml.org/lkml/2006/5/25/318 2) http://lkml.org/lkml/2006/5/25/320 Split LRU-list of unused dentries to each superblocks. Then, NFS mounting will check dentries under a superblock instead of all. But this spliting will break LRU of dentry-unused. So, I've attempted to make reclaim unused dentrins with fairness by calculate number of dentries to scan on this sb based on following way number of dentries to scan on this sb = count * (number of dentries on this sb / number of dentries in the machine) - ToDo - I have to measuring performance number and do stress tests. - When unmount occurs during prune_dcache(), scanning on same superblock, It is unable to reach next superblock because it is gone away. We restart scannig superblock from first one, it causes unfairness of reclaim unused dentries on first superblock. But I think this happens very rarely. - Test Results Result on 6GB boxes with excessive unused dentries. Without patch: $ cat /proc/sys/fs/dentry-state 10181835 10180203 45 0 0 0 # mount -t nfs 10.124.60.70:/work/kernel-src nfs real 0m1.830s user 0m0.001s sys 0m1.653s With this patch: $ cat /proc/sys/fs/dentry-state 10236610 10234751 45 0 0 0 # mount -t nfs 10.124.60.70:/work/kernel-src nfs real 0m0.106s user 0m0.002s sys 0m0.032s [akpm@linux-foundation.org: fix comments] Signed-off-by: Kentaro Makita Cc: Neil Brown Cc: Trond Myklebust Cc: David Chinner Cc: "J. Bruce Fields" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/dcache.c | 343 ++++++++++++++++++++++++--------------------- fs/super.c | 1 + include/linux/fs.h | 4 + 3 files changed, 189 insertions(+), 159 deletions(-) diff --git a/fs/dcache.c b/fs/dcache.c index 6068c25b393c..3818d6ab76ca 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -61,7 +61,6 @@ static struct kmem_cache *dentry_cache __read_mostly; static unsigned int d_hash_mask __read_mostly; static unsigned int d_hash_shift __read_mostly; static struct hlist_head *dentry_hashtable __read_mostly; -static LIST_HEAD(dentry_unused); /* Statistics gathering. */ struct dentry_stat_t dentry_stat = { @@ -96,14 +95,6 @@ static void d_free(struct dentry *dentry) call_rcu(&dentry->d_u.d_rcu, d_callback); } -static void dentry_lru_remove(struct dentry *dentry) -{ - if (!list_empty(&dentry->d_lru)) { - list_del_init(&dentry->d_lru); - dentry_stat.nr_unused--; - } -} - /* * Release the dentry's inode, using the filesystem * d_iput() operation if defined. @@ -130,6 +121,41 @@ static void dentry_iput(struct dentry * dentry) } } +/* + * dentry_lru_(add|add_tail|del|del_init) must be called with dcache_lock held. + */ +static void dentry_lru_add(struct dentry *dentry) +{ + list_add(&dentry->d_lru, &dentry->d_sb->s_dentry_lru); + dentry->d_sb->s_nr_dentry_unused++; + dentry_stat.nr_unused++; +} + +static void dentry_lru_add_tail(struct dentry *dentry) +{ + list_add_tail(&dentry->d_lru, &dentry->d_sb->s_dentry_lru); + dentry->d_sb->s_nr_dentry_unused++; + dentry_stat.nr_unused++; +} + +static void dentry_lru_del(struct dentry *dentry) +{ + if (!list_empty(&dentry->d_lru)) { + list_del(&dentry->d_lru); + dentry->d_sb->s_nr_dentry_unused--; + dentry_stat.nr_unused--; + } +} + +static void dentry_lru_del_init(struct dentry *dentry) +{ + if (likely(!list_empty(&dentry->d_lru))) { + list_del_init(&dentry->d_lru); + dentry->d_sb->s_nr_dentry_unused--; + dentry_stat.nr_unused--; + } +} + /** * d_kill - kill dentry and return parent * @dentry: dentry to kill @@ -212,8 +238,7 @@ void dput(struct dentry *dentry) goto kill_it; if (list_empty(&dentry->d_lru)) { dentry->d_flags |= DCACHE_REFERENCED; - list_add(&dentry->d_lru, &dentry_unused); - dentry_stat.nr_unused++; + dentry_lru_add(dentry); } spin_unlock(&dentry->d_lock); spin_unlock(&dcache_lock); @@ -222,7 +247,8 @@ void dput(struct dentry *dentry) unhash_it: __d_drop(dentry); kill_it: - dentry_lru_remove(dentry); + /* if dentry was on the d_lru list delete it from there */ + dentry_lru_del(dentry); dentry = d_kill(dentry); if (dentry) goto repeat; @@ -290,7 +316,7 @@ int d_invalidate(struct dentry * dentry) static inline struct dentry * __dget_locked(struct dentry *dentry) { atomic_inc(&dentry->d_count); - dentry_lru_remove(dentry); + dentry_lru_del_init(dentry); return dentry; } @@ -406,132 +432,166 @@ static void prune_one_dentry(struct dentry * dentry) if (dentry->d_op && dentry->d_op->d_delete) dentry->d_op->d_delete(dentry); - dentry_lru_remove(dentry); + dentry_lru_del_init(dentry); __d_drop(dentry); dentry = d_kill(dentry); spin_lock(&dcache_lock); } } -/** - * prune_dcache - shrink the dcache - * @count: number of entries to try and free - * @sb: if given, ignore dentries for other superblocks - * which are being unmounted. - * - * Shrink the dcache. This is done when we need - * more memory, or simply when we need to unmount - * something (at which point we need to unuse - * all dentries). - * - * This function may fail to free any resources if - * all the dentries are in use. +/* + * Shrink the dentry LRU on a given superblock. + * @sb : superblock to shrink dentry LRU. + * @count: If count is NULL, we prune all dentries on superblock. + * @flags: If flags is non-zero, we need to do special processing based on + * which flags are set. This means we don't need to maintain multiple + * similar copies of this loop. */ - -static void prune_dcache(int count, struct super_block *sb) +static void __shrink_dcache_sb(struct super_block *sb, int *count, int flags) { + LIST_HEAD(referenced); + LIST_HEAD(tmp); + struct dentry *dentry; + int cnt = 0; + + BUG_ON(!sb); + BUG_ON((flags & DCACHE_REFERENCED) && count == NULL); spin_lock(&dcache_lock); - for (; count ; count--) { - struct dentry *dentry; - struct list_head *tmp; - struct rw_semaphore *s_umount; + if (count != NULL) + /* called from prune_dcache() and shrink_dcache_parent() */ + cnt = *count; +restart: + if (count == NULL) + list_splice_init(&sb->s_dentry_lru, &tmp); + else { + while (!list_empty(&sb->s_dentry_lru)) { + dentry = list_entry(sb->s_dentry_lru.prev, + struct dentry, d_lru); + BUG_ON(dentry->d_sb != sb); - cond_resched_lock(&dcache_lock); - - tmp = dentry_unused.prev; - if (sb) { - /* Try to find a dentry for this sb, but don't try - * too hard, if they aren't near the tail they will - * be moved down again soon + spin_lock(&dentry->d_lock); + /* + * If we are honouring the DCACHE_REFERENCED flag and + * the dentry has this flag set, don't free it. Clear + * the flag and put it back on the LRU. */ - int skip = count; - while (skip && tmp != &dentry_unused && - list_entry(tmp, struct dentry, d_lru)->d_sb != sb) { - skip--; - tmp = tmp->prev; + if ((flags & DCACHE_REFERENCED) + && (dentry->d_flags & DCACHE_REFERENCED)) { + dentry->d_flags &= ~DCACHE_REFERENCED; + list_move_tail(&dentry->d_lru, &referenced); + spin_unlock(&dentry->d_lock); + } else { + list_move_tail(&dentry->d_lru, &tmp); + spin_unlock(&dentry->d_lock); + cnt--; + if (!cnt) + break; } } - if (tmp == &dentry_unused) - break; - list_del_init(tmp); - prefetch(dentry_unused.prev); - dentry_stat.nr_unused--; - dentry = list_entry(tmp, struct dentry, d_lru); - - spin_lock(&dentry->d_lock); + } + while (!list_empty(&tmp)) { + dentry = list_entry(tmp.prev, struct dentry, d_lru); + dentry_lru_del_init(dentry); + spin_lock(&dentry->d_lock); /* * We found an inuse dentry which was not removed from - * dentry_unused because of laziness during lookup. Do not free - * it - just keep it off the dentry_unused list. + * the LRU because of laziness during lookup. Do not free + * it - just keep it off the LRU list. */ - if (atomic_read(&dentry->d_count)) { - spin_unlock(&dentry->d_lock); + if (atomic_read(&dentry->d_count)) { + spin_unlock(&dentry->d_lock); continue; } - /* If the dentry was recently referenced, don't free it. */ - if (dentry->d_flags & DCACHE_REFERENCED) { - dentry->d_flags &= ~DCACHE_REFERENCED; - list_add(&dentry->d_lru, &dentry_unused); - dentry_stat.nr_unused++; - spin_unlock(&dentry->d_lock); - continue; - } - /* - * If the dentry is not DCACHED_REFERENCED, it is time - * to remove it from the dcache, provided the super block is - * NULL (which means we are trying to reclaim memory) - * or this dentry belongs to the same super block that - * we want to shrink. - */ - /* - * If this dentry is for "my" filesystem, then I can prune it - * without taking the s_umount lock (I already hold it). - */ - if (sb && dentry->d_sb == sb) { - prune_one_dentry(dentry); - continue; - } - /* - * ...otherwise we need to be sure this filesystem isn't being - * unmounted, otherwise we could race with - * generic_shutdown_super(), and end up holding a reference to - * an inode while the filesystem is unmounted. - * So we try to get s_umount, and make sure s_root isn't NULL. - * (Take a local copy of s_umount to avoid a use-after-free of - * `dentry'). - */ - s_umount = &dentry->d_sb->s_umount; - if (down_read_trylock(s_umount)) { - if (dentry->d_sb->s_root != NULL) { - prune_one_dentry(dentry); - up_read(s_umount); - continue; - } - up_read(s_umount); - } - spin_unlock(&dentry->d_lock); - /* - * Insert dentry at the head of the list as inserting at the - * tail leads to a cycle. - */ - list_add(&dentry->d_lru, &dentry_unused); - dentry_stat.nr_unused++; + prune_one_dentry(dentry); + /* dentry->d_lock was dropped in prune_one_dentry() */ + cond_resched_lock(&dcache_lock); } + if (count == NULL && !list_empty(&sb->s_dentry_lru)) + goto restart; + if (count != NULL) + *count = cnt; + if (!list_empty(&referenced)) + list_splice(&referenced, &sb->s_dentry_lru); spin_unlock(&dcache_lock); } -/* - * Shrink the dcache for the specified super block. - * This allows us to unmount a device without disturbing - * the dcache for the other devices. +/** + * prune_dcache - shrink the dcache + * @count: number of entries to try to free * - * This implementation makes just two traversals of the - * unused list. On the first pass we move the selected - * dentries to the most recent end, and on the second - * pass we free them. The second pass must restart after - * each dput(), but since the target dentries are all at - * the end, it's really just a single traversal. + * Shrink the dcache. This is done when we need more memory, or simply when we + * need to unmount something (at which point we need to unuse all dentries). + * + * This function may fail to free any resources if all the dentries are in use. */ +static void prune_dcache(int count) +{ + struct super_block *sb; + int w_count; + int unused = dentry_stat.nr_unused; + int prune_ratio; + int pruned; + + if (unused == 0 || count == 0) + return; + spin_lock(&dcache_lock); +restart: + if (count >= unused) + prune_ratio = 1; + else + prune_ratio = unused / count; + spin_lock(&sb_lock); + list_for_each_entry(sb, &super_blocks, s_list) { + if (sb->s_nr_dentry_unused == 0) + continue; + sb->s_count++; + /* Now, we reclaim unused dentrins with fairness. + * We reclaim them same percentage from each superblock. + * We calculate number of dentries to scan on this sb + * as follows, but the implementation is arranged to avoid + * overflows: + * number of dentries to scan on this sb = + * count * (number of dentries on this sb / + * number of dentries in the machine) + */ + spin_unlock(&sb_lock); + if (prune_ratio != 1) + w_count = (sb->s_nr_dentry_unused / prune_ratio) + 1; + else + w_count = sb->s_nr_dentry_unused; + pruned = w_count; + /* + * We need to be sure this filesystem isn't being unmounted, + * otherwise we could race with generic_shutdown_super(), and + * end up holding a reference to an inode while the filesystem + * is unmounted. So we try to get s_umount, and make sure + * s_root isn't NULL. + */ + if (down_read_trylock(&sb->s_umount)) { + if ((sb->s_root != NULL) && + (!list_empty(&sb->s_dentry_lru))) { + spin_unlock(&dcache_lock); + __shrink_dcache_sb(sb, &w_count, + DCACHE_REFERENCED); + pruned -= w_count; + spin_lock(&dcache_lock); + } + up_read(&sb->s_umount); + } + spin_lock(&sb_lock); + count -= pruned; + /* + * restart only when sb is no longer on the list and + * we have more work to do. + */ + if (__put_super_and_need_restart(sb) && count > 0) { + spin_unlock(&sb_lock); + goto restart; + } + } + spin_unlock(&sb_lock); + spin_unlock(&dcache_lock); +} /** * shrink_dcache_sb - shrink dcache for a superblock @@ -541,44 +601,9 @@ static void prune_dcache(int count, struct super_block *sb) * is used to free the dcache before unmounting a file * system */ - void shrink_dcache_sb(struct super_block * sb) { - struct list_head *tmp, *next; - struct dentry *dentry; - - /* - * Pass one ... move the dentries for the specified - * superblock to the most recent end of the unused list. - */ - spin_lock(&dcache_lock); - list_for_each_prev_safe(tmp, next, &dentry_unused) { - dentry = list_entry(tmp, struct dentry, d_lru); - if (dentry->d_sb != sb) - continue; - list_move_tail(tmp, &dentry_unused); - } - - /* - * Pass two ... free the dentries for this superblock. - */ -repeat: - list_for_each_prev_safe(tmp, next, &dentry_unused) { - dentry = list_entry(tmp, struct dentry, d_lru); - if (dentry->d_sb != sb) - continue; - dentry_stat.nr_unused--; - list_del_init(tmp); - spin_lock(&dentry->d_lock); - if (atomic_read(&dentry->d_count)) { - spin_unlock(&dentry->d_lock); - continue; - } - prune_one_dentry(dentry); - cond_resched_lock(&dcache_lock); - goto repeat; - } - spin_unlock(&dcache_lock); + __shrink_dcache_sb(sb, NULL, 0); } /* @@ -595,7 +620,7 @@ static void shrink_dcache_for_umount_subtree(struct dentry *dentry) /* detach this root from the system */ spin_lock(&dcache_lock); - dentry_lru_remove(dentry); + dentry_lru_del_init(dentry); __d_drop(dentry); spin_unlock(&dcache_lock); @@ -609,7 +634,7 @@ static void shrink_dcache_for_umount_subtree(struct dentry *dentry) spin_lock(&dcache_lock); list_for_each_entry(loop, &dentry->d_subdirs, d_u.d_child) { - dentry_lru_remove(loop); + dentry_lru_del_init(loop); __d_drop(loop); cond_resched_lock(&dcache_lock); } @@ -791,14 +816,13 @@ static int select_parent(struct dentry * parent) struct dentry *dentry = list_entry(tmp, struct dentry, d_u.d_child); next = tmp->next; - dentry_lru_remove(dentry); + dentry_lru_del_init(dentry); /* * move only zero ref count dentries to the end * of the unused list for prune_dcache */ if (!atomic_read(&dentry->d_count)) { - list_add_tail(&dentry->d_lru, &dentry_unused); - dentry_stat.nr_unused++; + dentry_lru_add_tail(dentry); found++; } @@ -840,10 +864,11 @@ static int select_parent(struct dentry * parent) void shrink_dcache_parent(struct dentry * parent) { + struct super_block *sb = parent->d_sb; int found; while ((found = select_parent(parent)) != 0) - prune_dcache(found, parent->d_sb); + __shrink_dcache_sb(sb, &found, 0); } /* @@ -863,7 +888,7 @@ static int shrink_dcache_memory(int nr, gfp_t gfp_mask) if (nr) { if (!(gfp_mask & __GFP_FS)) return -1; - prune_dcache(nr, NULL); + prune_dcache(nr); } return (dentry_stat.nr_unused / 100) * sysctl_vfs_cache_pressure; } @@ -1215,7 +1240,7 @@ struct dentry *d_splice_alias(struct inode *inode, struct dentry *dentry) * rcu_read_lock() and rcu_read_unlock() are used to disable preemption while * lookup is going on. * - * dentry_unused list is not updated even if lookup finds the required dentry + * The dentry unused LRU is not updated even if lookup finds the required dentry * in there. It is updated in places such as prune_dcache, shrink_dcache_sb, * select_parent and __dget_locked. This laziness saves lookup from dcache_lock * acquisition. diff --git a/fs/super.c b/fs/super.c index 453877c5697b..e931ae9511fe 100644 --- a/fs/super.c +++ b/fs/super.c @@ -70,6 +70,7 @@ static struct super_block *alloc_super(struct file_system_type *type) INIT_LIST_HEAD(&s->s_instances); INIT_HLIST_HEAD(&s->s_anon); INIT_LIST_HEAD(&s->s_inodes); + INIT_LIST_HEAD(&s->s_dentry_lru); init_rwsem(&s->s_umount); mutex_init(&s->s_lock); lockdep_set_class(&s->s_umount, &type->s_umount_key); diff --git a/include/linux/fs.h b/include/linux/fs.h index ff54ae4933f3..e5e6a244096c 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1025,6 +1025,7 @@ extern int send_sigurg(struct fown_struct *fown); extern struct list_head super_blocks; extern spinlock_t sb_lock; +#define sb_entry(list) list_entry((list), struct super_block, s_list) #define S_BIAS (1<<30) struct super_block { struct list_head s_list; /* Keep this first */ @@ -1058,6 +1059,9 @@ struct super_block { struct list_head s_more_io; /* parked for more writeback */ struct hlist_head s_anon; /* anonymous dentries for (nfs) exporting */ struct list_head s_files; + /* s_dentry_lru and s_nr_dentry_unused are protected by dcache_lock */ + struct list_head s_dentry_lru; /* unused dentry lru */ + int s_nr_dentry_unused; /* # of dentry on lru */ struct block_device *s_bdev; struct mtd_info *s_mtd; From 0cad47cf13bc2e9142d3a11d9f50523797d0d4ea Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:27:16 -0700 Subject: [PATCH 147/857] page-flags: record page flag overlays explicitly With the recent page flag reorganisation we have a single enum which defines the valid page flags and their values, nice and clear. However there are a number of bits which are overloaded by different subsystems. Firstly there is PG_owner_priv_1 which is used by filesystems and by XEN. Secondly both SLOB and SLUB use a couple of extra page bits to manage internal state for pages they own; both overlay other bits. All of these "aliases" are scattered about the source making it very hard for a reader to know if the bits are safe to rely on in all contexts; confusion here is bad. As we now have a single place where the bits are clearly assigned it makes sense to clarify the reuse of bits by making the aliases explicit and visible with the original bit assignments. This patch creates explicit aliases within the enum itself for the overloaded bits, creates standard bit accessors PageFoo etc. and uses those throughout. This version pulls the bit manipulation out to standard named page bit accessors as suggested by Christoph, it retains the explicit mapping to the overlayed bits. A fusion of both ideas. This has been SLUB and SLOB have been compile tested on x86_64 only, and SLUB boot tested. If people feel this is worth doing then I can run a fuller set of testing. This patch: Some page flags are used for more than one purpose, for example PG_owner_priv_1. Currently there are individual accessors for each user, each built using the common flag name far away from the bit definitions. This makes it hard to see all possible uses of these bits. Now that we have a single enum to generate the bit orders it makes sense to express overlays in the same place. So create per use aliases for this bit in the main page-flags enum and use those in the accessors. [akpm@linux-foundation.org: fix xen] Signed-off-by: Andy Whitcroft Cc: Pekka Enberg Cc: Christoph Lameter Cc: Matt Mackall Cc: Nick Piggin Cc: KAMEZAWA Hiroyuki Reviewed-by: KOSAKI Motohiro Cc: Rik van Riel Cc: Jeremy Fitzhardinge Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/page-flags.h | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h index 0d2a4e7012aa..7d8db1233e44 100644 --- a/include/linux/page-flags.h +++ b/include/linux/page-flags.h @@ -96,7 +96,14 @@ enum pageflags { #ifdef CONFIG_IA64_UNCACHED_ALLOCATOR PG_uncached, /* Page has been mapped as uncached */ #endif - __NR_PAGEFLAGS + __NR_PAGEFLAGS, + + /* Filesystems */ + PG_checked = PG_owner_priv_1, + + /* XEN */ + PG_pinned = PG_owner_priv_1, + PG_savepinned = PG_dirty, }; #ifndef __GENERATING_BOUNDS_H @@ -155,9 +162,9 @@ PAGEFLAG(Dirty, dirty) TESTSCFLAG(Dirty, dirty) __CLEARPAGEFLAG(Dirty, dirty) PAGEFLAG(LRU, lru) __CLEARPAGEFLAG(LRU, lru) PAGEFLAG(Active, active) __CLEARPAGEFLAG(Active, active) __PAGEFLAG(Slab, slab) -PAGEFLAG(Checked, owner_priv_1) /* Used by some filesystems */ -PAGEFLAG(Pinned, owner_priv_1) TESTSCFLAG(Pinned, owner_priv_1) /* Xen */ -PAGEFLAG(SavePinned, dirty); /* Xen */ +PAGEFLAG(Checked, checked) /* Used by some filesystems */ +PAGEFLAG(Pinned, pinned) TESTSCFLAG(Pinned, pinned) /* Xen */ +PAGEFLAG(SavePinned, savepinned); /* Xen */ PAGEFLAG(Reserved, reserved) __CLEARPAGEFLAG(Reserved, reserved) PAGEFLAG(Private, private) __CLEARPAGEFLAG(Private, private) __SETPAGEFLAG(Private, private) From 8a38082d21cbc5ec961da7dda195e98a9a064dcf Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:27:18 -0700 Subject: [PATCH 148/857] slub: record page flag overlays explicitly SLUB reuses two page bits for internal purposes, it overlays PG_active and PG_error. This is hidden away in slub.c. Document these overlays explicitly in the main page-flags enum along with all the others. Signed-off-by: Andy Whitcroft Cc: Pekka Enberg Cc: Christoph Lameter Cc: Matt Mackall Cc: Nick Piggin Tested-by: KOSAKI Motohiro Cc: KOSAKI Motohiro Cc: Rik van Riel Cc: Jeremy Fitzhardinge Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/page-flags.h | 7 ++++ mm/slub.c | 65 ++++++++++---------------------------- 2 files changed, 24 insertions(+), 48 deletions(-) diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h index 7d8db1233e44..3fc586b7b90b 100644 --- a/include/linux/page-flags.h +++ b/include/linux/page-flags.h @@ -104,6 +104,10 @@ enum pageflags { /* XEN */ PG_pinned = PG_owner_priv_1, PG_savepinned = PG_dirty, + + /* SLUB */ + PG_slub_frozen = PG_active, + PG_slub_debug = PG_error, }; #ifndef __GENERATING_BOUNDS_H @@ -169,6 +173,9 @@ PAGEFLAG(Reserved, reserved) __CLEARPAGEFLAG(Reserved, reserved) PAGEFLAG(Private, private) __CLEARPAGEFLAG(Private, private) __SETPAGEFLAG(Private, private) +__PAGEFLAG(SlubFrozen, slub_frozen) +__PAGEFLAG(SlubDebug, slub_debug) + /* * Only test-and-set exist for PG_writeback. The unconditional operators are * risky: they bypass page accounting. diff --git a/mm/slub.c b/mm/slub.c index 6d4a49c1ff2f..77c21cf53ff9 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -102,44 +102,12 @@ * the fast path and disables lockless freelists. */ -#define FROZEN (1 << PG_active) - #ifdef CONFIG_SLUB_DEBUG -#define SLABDEBUG (1 << PG_error) +#define SLABDEBUG 1 #else #define SLABDEBUG 0 #endif -static inline int SlabFrozen(struct page *page) -{ - return page->flags & FROZEN; -} - -static inline void SetSlabFrozen(struct page *page) -{ - page->flags |= FROZEN; -} - -static inline void ClearSlabFrozen(struct page *page) -{ - page->flags &= ~FROZEN; -} - -static inline int SlabDebug(struct page *page) -{ - return page->flags & SLABDEBUG; -} - -static inline void SetSlabDebug(struct page *page) -{ - page->flags |= SLABDEBUG; -} - -static inline void ClearSlabDebug(struct page *page) -{ - page->flags &= ~SLABDEBUG; -} - /* * Issues still to be resolved: * @@ -971,7 +939,7 @@ static int free_debug_processing(struct kmem_cache *s, struct page *page, } /* Special debug activities for freeing objects */ - if (!SlabFrozen(page) && !page->freelist) + if (!PageSlubFrozen(page) && !page->freelist) remove_full(s, page); if (s->flags & SLAB_STORE_USER) set_track(s, object, TRACK_FREE, addr); @@ -1157,7 +1125,7 @@ static struct page *new_slab(struct kmem_cache *s, gfp_t flags, int node) page->flags |= 1 << PG_slab; if (s->flags & (SLAB_DEBUG_FREE | SLAB_RED_ZONE | SLAB_POISON | SLAB_STORE_USER | SLAB_TRACE)) - SetSlabDebug(page); + __SetPageSlubDebug(page); start = page_address(page); @@ -1184,14 +1152,14 @@ static void __free_slab(struct kmem_cache *s, struct page *page) int order = compound_order(page); int pages = 1 << order; - if (unlikely(SlabDebug(page))) { + if (unlikely(SLABDEBUG && PageSlubDebug(page))) { void *p; slab_pad_check(s, page); for_each_object(p, s, page_address(page), page->objects) check_object(s, page, p, 0); - ClearSlabDebug(page); + __ClearPageSlubDebug(page); } mod_zone_page_state(page_zone(page), @@ -1288,7 +1256,7 @@ static inline int lock_and_freeze_slab(struct kmem_cache_node *n, if (slab_trylock(page)) { list_del(&page->lru); n->nr_partial--; - SetSlabFrozen(page); + __SetPageSlubFrozen(page); return 1; } return 0; @@ -1398,7 +1366,7 @@ static void unfreeze_slab(struct kmem_cache *s, struct page *page, int tail) struct kmem_cache_node *n = get_node(s, page_to_nid(page)); struct kmem_cache_cpu *c = get_cpu_slab(s, smp_processor_id()); - ClearSlabFrozen(page); + __ClearPageSlubFrozen(page); if (page->inuse) { if (page->freelist) { @@ -1406,7 +1374,8 @@ static void unfreeze_slab(struct kmem_cache *s, struct page *page, int tail) stat(c, tail ? DEACTIVATE_TO_TAIL : DEACTIVATE_TO_HEAD); } else { stat(c, DEACTIVATE_FULL); - if (SlabDebug(page) && (s->flags & SLAB_STORE_USER)) + if (SLABDEBUG && PageSlubDebug(page) && + (s->flags & SLAB_STORE_USER)) add_full(n, page); } slab_unlock(page); @@ -1551,7 +1520,7 @@ static void *__slab_alloc(struct kmem_cache *s, object = c->page->freelist; if (unlikely(!object)) goto another_slab; - if (unlikely(SlabDebug(c->page))) + if (unlikely(SLABDEBUG && PageSlubDebug(c->page))) goto debug; c->freelist = object[c->offset]; @@ -1588,7 +1557,7 @@ static void *__slab_alloc(struct kmem_cache *s, if (c->page) flush_slab(s, c); slab_lock(new); - SetSlabFrozen(new); + __SetPageSlubFrozen(new); c->page = new; goto load_freelist; } @@ -1674,7 +1643,7 @@ static void __slab_free(struct kmem_cache *s, struct page *page, stat(c, FREE_SLOWPATH); slab_lock(page); - if (unlikely(SlabDebug(page))) + if (unlikely(SLABDEBUG && PageSlubDebug(page))) goto debug; checks_ok: @@ -1682,7 +1651,7 @@ static void __slab_free(struct kmem_cache *s, struct page *page, page->freelist = object; page->inuse--; - if (unlikely(SlabFrozen(page))) { + if (unlikely(PageSlubFrozen(page))) { stat(c, FREE_FROZEN); goto out_unlock; } @@ -3317,12 +3286,12 @@ static void validate_slab_slab(struct kmem_cache *s, struct page *page, s->name, page); if (s->flags & DEBUG_DEFAULT_FLAGS) { - if (!SlabDebug(page)) - printk(KERN_ERR "SLUB %s: SlabDebug not set " + if (!PageSlubDebug(page)) + printk(KERN_ERR "SLUB %s: SlubDebug not set " "on slab 0x%p\n", s->name, page); } else { - if (SlabDebug(page)) - printk(KERN_ERR "SLUB %s: SlabDebug set on " + if (PageSlubDebug(page)) + printk(KERN_ERR "SLUB %s: SlubDebug set on " "slab 0x%p\n", s->name, page); } } From 9023cb7e8564d95a1893f8cb6895a293be9a71fe Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:27:19 -0700 Subject: [PATCH 149/857] slob: record page flag overlays explicitly SLOB reuses two page bits for internal purposes, it overlays PG_active and PG_private. This is hidden away in slob.c. Document these overlays explicitly in the main page-flags enum along with all the others. Signed-off-by: Andy Whitcroft Cc: Pekka Enberg Cc: Christoph Lameter Cc: Matt Mackall Cc: Nick Piggin Reviewed-by: KOSAKI Motohiro Cc: KOSAKI Motohiro Cc: Rik van Riel Cc: Jeremy Fitzhardinge Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/page-flags.h | 7 +++++++ mm/slob.c | 12 ++++++------ 2 files changed, 13 insertions(+), 6 deletions(-) diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h index 3fc586b7b90b..54590a9a103e 100644 --- a/include/linux/page-flags.h +++ b/include/linux/page-flags.h @@ -105,6 +105,10 @@ enum pageflags { PG_pinned = PG_owner_priv_1, PG_savepinned = PG_dirty, + /* SLOB */ + PG_slob_page = PG_active, + PG_slob_free = PG_private, + /* SLUB */ PG_slub_frozen = PG_active, PG_slub_debug = PG_error, @@ -173,6 +177,9 @@ PAGEFLAG(Reserved, reserved) __CLEARPAGEFLAG(Reserved, reserved) PAGEFLAG(Private, private) __CLEARPAGEFLAG(Private, private) __SETPAGEFLAG(Private, private) +__PAGEFLAG(SlobPage, slob_page) +__PAGEFLAG(SlobFree, slob_free) + __PAGEFLAG(SlubFrozen, slub_frozen) __PAGEFLAG(SlubDebug, slub_debug) diff --git a/mm/slob.c b/mm/slob.c index a3ad6671adf1..de268eb7ac70 100644 --- a/mm/slob.c +++ b/mm/slob.c @@ -130,17 +130,17 @@ static LIST_HEAD(free_slob_large); */ static inline int slob_page(struct slob_page *sp) { - return test_bit(PG_active, &sp->flags); + return PageSlobPage((struct page *)sp); } static inline void set_slob_page(struct slob_page *sp) { - __set_bit(PG_active, &sp->flags); + __SetPageSlobPage((struct page *)sp); } static inline void clear_slob_page(struct slob_page *sp) { - __clear_bit(PG_active, &sp->flags); + __ClearPageSlobPage((struct page *)sp); } /* @@ -148,19 +148,19 @@ static inline void clear_slob_page(struct slob_page *sp) */ static inline int slob_page_free(struct slob_page *sp) { - return test_bit(PG_private, &sp->flags); + return PageSlobFree((struct page *)sp); } static void set_slob_page_free(struct slob_page *sp, struct list_head *list) { list_add(&sp->list, list); - __set_bit(PG_private, &sp->flags); + __SetPageSlobFree((struct page *)sp); } static inline void clear_slob_page_free(struct slob_page *sp) { list_del(&sp->list); - __clear_bit(PG_private, &sp->flags); + __ClearPageSlobFree((struct page *)sp); } #define SLOB_UNIT sizeof(slob_t) From 2185e69f680ae8c8496b6fc15e20c889d5b39b67 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 23 Jul 2008 21:27:19 -0700 Subject: [PATCH 150/857] mapping_set_error: add unlikely() This is called on a per-page basis and in the vast majority of cases `error' is zero. Cc: Guillaume Chazarain Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/pagemap.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h index d2fca802f809..ee1ec2c7723c 100644 --- a/include/linux/pagemap.h +++ b/include/linux/pagemap.h @@ -22,7 +22,7 @@ static inline void mapping_set_error(struct address_space *mapping, int error) { - if (error) { + if (unlikely(error)) { if (error == -ENOSPC) set_bit(AS_ENOSPC, &mapping->flags); else From 9109fb7b3520de187ebc3646c209d66a233f7169 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 23 Jul 2008 21:27:20 -0700 Subject: [PATCH 151/857] mm: drop unneeded pgdat argument from free_area_init_node() free_area_init_node() gets passed in the node id as well as the node descriptor. This is redundant as the function can trivially get the node descriptor itself by means of NODE_DATA() and the node's id. I checked all the users and NODE_DATA() seems to be usable everywhere from where this function is called. Signed-off-by: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/alpha/mm/numa.c | 2 +- arch/arm/mm/init.c | 2 +- arch/avr32/mm/init.c | 2 +- arch/cris/arch-v10/mm/init.c | 2 +- arch/cris/arch-v32/mm/init.c | 2 +- arch/m32r/mm/discontig.c | 3 +-- arch/m32r/mm/init.c | 2 +- arch/m68k/mm/motorola.c | 2 +- arch/m68k/mm/sun3mmu.c | 2 +- arch/parisc/mm/init.c | 2 +- arch/sparc/mm/srmmu.c | 3 +-- arch/sparc/mm/sun4c.c | 3 +-- arch/v850/kernel/setup.c | 3 +-- include/linux/mm.h | 5 ++--- mm/memory_hotplug.c | 2 +- mm/page_alloc.c | 11 ++++++----- 16 files changed, 22 insertions(+), 26 deletions(-) diff --git a/arch/alpha/mm/numa.c b/arch/alpha/mm/numa.c index a53fda0481ca..def0c74a78a8 100644 --- a/arch/alpha/mm/numa.c +++ b/arch/alpha/mm/numa.c @@ -313,7 +313,7 @@ void __init paging_init(void) zones_size[ZONE_DMA] = dma_local_pfn; zones_size[ZONE_NORMAL] = (end_pfn - start_pfn) - dma_local_pfn; } - free_area_init_node(nid, NODE_DATA(nid), zones_size, start_pfn, NULL); + free_area_init_node(nid, zones_size, start_pfn, NULL); } /* Initialize the kernel's ZERO_PGE. */ diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c index b657f1719af0..e6352946dde0 100644 --- a/arch/arm/mm/init.c +++ b/arch/arm/mm/init.c @@ -284,7 +284,7 @@ bootmem_init_node(int node, int initrd_node, struct meminfo *mi) */ arch_adjust_zones(node, zone_size, zhole_size); - free_area_init_node(node, pgdat, zone_size, start_pfn, zhole_size); + free_area_init_node(node, zone_size, start_pfn, zhole_size); return end_pfn; } diff --git a/arch/avr32/mm/init.c b/arch/avr32/mm/init.c index 3f90a87527bb..786de88a82a7 100644 --- a/arch/avr32/mm/init.c +++ b/arch/avr32/mm/init.c @@ -129,7 +129,7 @@ void __init paging_init(void) printk("Node %u: start_pfn = 0x%lx, low = 0x%lx\n", nid, start_pfn, low); - free_area_init_node(nid, pgdat, zones_size, start_pfn, NULL); + free_area_init_node(nid, zones_size, start_pfn, NULL); printk("Node %u: mem_map starts at %p\n", pgdat->node_id, pgdat->node_mem_map); diff --git a/arch/cris/arch-v10/mm/init.c b/arch/cris/arch-v10/mm/init.c index e0fcd1a9bfd5..742fd1974c2e 100644 --- a/arch/cris/arch-v10/mm/init.c +++ b/arch/cris/arch-v10/mm/init.c @@ -182,7 +182,7 @@ paging_init(void) * mem_map page array. */ - free_area_init_node(0, &contig_page_data, zones_size, PAGE_OFFSET >> PAGE_SHIFT, 0); + free_area_init_node(0, zones_size, PAGE_OFFSET >> PAGE_SHIFT, 0); } /* Initialize remaps of some I/O-ports. It is important that this diff --git a/arch/cris/arch-v32/mm/init.c b/arch/cris/arch-v32/mm/init.c index 5a9ac5834647..8a34b8b74293 100644 --- a/arch/cris/arch-v32/mm/init.c +++ b/arch/cris/arch-v32/mm/init.c @@ -162,7 +162,7 @@ paging_init(void) * substantially higher than 0, like us (we start at PAGE_OFFSET). This * saves space in the mem_map page array. */ - free_area_init_node(0, &contig_page_data, zones_size, PAGE_OFFSET >> PAGE_SHIFT, 0); + free_area_init_node(0, zones_size, PAGE_OFFSET >> PAGE_SHIFT, 0); mem_map = contig_page_data.node_mem_map; } diff --git a/arch/m32r/mm/discontig.c b/arch/m32r/mm/discontig.c index aa9145ef6cca..cc23934bc41e 100644 --- a/arch/m32r/mm/discontig.c +++ b/arch/m32r/mm/discontig.c @@ -147,8 +147,7 @@ unsigned long __init zone_sizes_init(void) zholes_size[ZONE_DMA] = mp->holes; holes += zholes_size[ZONE_DMA]; - free_area_init_node(nid, NODE_DATA(nid), zones_size, - start_pfn, zholes_size); + free_area_init_node(nid, zones_size, start_pfn, zholes_size); } /* diff --git a/arch/m32r/mm/init.c b/arch/m32r/mm/init.c index bbd97c85bc5d..28799af15e95 100644 --- a/arch/m32r/mm/init.c +++ b/arch/m32r/mm/init.c @@ -123,7 +123,7 @@ unsigned long __init zone_sizes_init(void) start_pfn = __MEMORY_START >> PAGE_SHIFT; #endif /* CONFIG_MMU */ - free_area_init_node(0, NODE_DATA(0), zones_size, start_pfn, 0); + free_area_init_node(0, zones_size, start_pfn, 0); return 0; } diff --git a/arch/m68k/mm/motorola.c b/arch/m68k/mm/motorola.c index 226795bdf355..c5dbb9bdb322 100644 --- a/arch/m68k/mm/motorola.c +++ b/arch/m68k/mm/motorola.c @@ -296,7 +296,7 @@ void __init paging_init(void) #endif for (i = 0; i < m68k_num_memory; i++) { zones_size[ZONE_DMA] = m68k_memory[i].size >> PAGE_SHIFT; - free_area_init_node(i, pg_data_map + i, zones_size, + free_area_init_node(i, zones_size, m68k_memory[i].addr >> PAGE_SHIFT, NULL); } } diff --git a/arch/m68k/mm/sun3mmu.c b/arch/m68k/mm/sun3mmu.c index edceefc18870..1b902dbd4376 100644 --- a/arch/m68k/mm/sun3mmu.c +++ b/arch/m68k/mm/sun3mmu.c @@ -94,7 +94,7 @@ void __init paging_init(void) /* I really wish I knew why the following change made things better... -- Sam */ /* free_area_init(zones_size); */ - free_area_init_node(0, NODE_DATA(0), zones_size, + free_area_init_node(0, zones_size, (__pa(PAGE_OFFSET) >> PAGE_SHIFT) + 1, NULL); diff --git a/arch/parisc/mm/init.c b/arch/parisc/mm/init.c index 0ddf4904640a..7c155c254e72 100644 --- a/arch/parisc/mm/init.c +++ b/arch/parisc/mm/init.c @@ -887,7 +887,7 @@ void __init paging_init(void) } #endif - free_area_init_node(i, NODE_DATA(i), zones_size, + free_area_init_node(i, zones_size, pmem_ranges[i].start_pfn, NULL); } } diff --git a/arch/sparc/mm/srmmu.c b/arch/sparc/mm/srmmu.c index c624e04ff03e..ee30462598fc 100644 --- a/arch/sparc/mm/srmmu.c +++ b/arch/sparc/mm/srmmu.c @@ -1352,8 +1352,7 @@ void __init srmmu_paging_init(void) zones_size[ZONE_HIGHMEM] = npages; zholes_size[ZONE_HIGHMEM] = npages - calc_highpages(); - free_area_init_node(0, &contig_page_data, zones_size, - pfn_base, zholes_size); + free_area_init_node(0, zones_size, pfn_base, zholes_size); } } diff --git a/arch/sparc/mm/sun4c.c b/arch/sparc/mm/sun4c.c index 2375fe9dc312..d1782f6368be 100644 --- a/arch/sparc/mm/sun4c.c +++ b/arch/sparc/mm/sun4c.c @@ -2123,8 +2123,7 @@ void __init sun4c_paging_init(void) zones_size[ZONE_HIGHMEM] = npages; zholes_size[ZONE_HIGHMEM] = npages - calc_highpages(); - free_area_init_node(0, &contig_page_data, zones_size, - pfn_base, zholes_size); + free_area_init_node(0, zones_size, pfn_base, zholes_size); } cnt = 0; diff --git a/arch/v850/kernel/setup.c b/arch/v850/kernel/setup.c index a0a8456a8430..10335cecf7bd 100644 --- a/arch/v850/kernel/setup.c +++ b/arch/v850/kernel/setup.c @@ -295,8 +295,7 @@ init_mem_alloc (unsigned long ram_start, unsigned long ram_len) #error MAX_ORDER is too large for given PAGE_OFFSET (use CONFIG_FORCE_MAX_ZONEORDER to change it) #endif NODE_DATA(0)->node_mem_map = NULL; - free_area_init_node (0, NODE_DATA(0), zones_size, - ADDR_TO_PAGE (PAGE_OFFSET), 0); + free_area_init_node(0, zones_size, ADDR_TO_PAGE (PAGE_OFFSET), 0); } diff --git a/include/linux/mm.h b/include/linux/mm.h index f8071097302a..196924b657bc 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -962,9 +962,8 @@ static inline void pgtable_page_dtor(struct page *page) NULL: pte_offset_kernel(pmd, address)) extern void free_area_init(unsigned long * zones_size); -extern void free_area_init_node(int nid, pg_data_t *pgdat, - unsigned long * zones_size, unsigned long zone_start_pfn, - unsigned long *zholes_size); +extern void free_area_init_node(int nid, unsigned long * zones_size, + unsigned long zone_start_pfn, unsigned long *zholes_size); #ifdef CONFIG_ARCH_POPULATES_NODE_MAP /* * With CONFIG_ARCH_POPULATES_NODE_MAP set, an architecture may initialise its diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index 833f854eabe5..6e26adc08f14 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -455,7 +455,7 @@ static pg_data_t *hotadd_new_pgdat(int nid, u64 start) /* we can use NODE_DATA(nid) from here */ /* init node's zones as empty zones, we don't have any present pages.*/ - free_area_init_node(nid, pgdat, zones_size, start_pfn, zholes_size); + free_area_init_node(nid, zones_size, start_pfn, zholes_size); return pgdat; } diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 24aa3d1b9d96..e43aae135b38 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -3461,10 +3461,11 @@ static void __init_refok alloc_node_mem_map(struct pglist_data *pgdat) #endif /* CONFIG_FLAT_NODE_MEM_MAP */ } -void __paginginit free_area_init_node(int nid, struct pglist_data *pgdat, - unsigned long *zones_size, unsigned long node_start_pfn, - unsigned long *zholes_size) +void __paginginit free_area_init_node(int nid, unsigned long *zones_size, + unsigned long node_start_pfn, unsigned long *zholes_size) { + pg_data_t *pgdat = NODE_DATA(nid); + pgdat->node_id = nid; pgdat->node_start_pfn = node_start_pfn; calculate_node_totalpages(pgdat, zones_size, zholes_size); @@ -3961,7 +3962,7 @@ void __init free_area_init_nodes(unsigned long *max_zone_pfn) setup_nr_node_ids(); for_each_online_node(nid) { pg_data_t *pgdat = NODE_DATA(nid); - free_area_init_node(nid, pgdat, NULL, + free_area_init_node(nid, NULL, find_min_pfn_for_node(nid), NULL); /* Any memory on that node */ @@ -4032,7 +4033,7 @@ EXPORT_SYMBOL(contig_page_data); void __init free_area_init(unsigned long *zones_size) { - free_area_init_node(0, NODE_DATA(0), zones_size, + free_area_init_node(0, zones_size, __pa(PAGE_OFFSET) >> PAGE_SHIFT, NULL); } From fc1b8a73dd71226902a11928dd5500326e101df9 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Wed, 23 Jul 2008 21:27:22 -0700 Subject: [PATCH 152/857] hugetlb: move hugetlb_acct_memory() This is a patchset to give reliable behaviour to a process that successfully calls mmap(MAP_PRIVATE) on a hugetlbfs file. Currently, it is possible for the process to be killed due to a small hugepage pool size even if it calls mlock(). MAP_SHARED mappings on hugetlbfs reserve huge pages at mmap() time. This guarantees all future faults against the mapping will succeed. This allows local allocations at first use improving NUMA locality whilst retaining reliability. MAP_PRIVATE mappings do not reserve pages. This can result in an application being SIGKILLed later if a huge page is not available at fault time. This makes huge pages usage very ill-advised in some cases as the unexpected application failure cannot be detected and handled as it is immediately fatal. Although an application may force instantiation of the pages using mlock(), this may lead to poor memory placement and the process may still be killed when performing COW. This patchset introduces a reliability guarantee for the process which creates a private mapping, i.e. the process that calls mmap() on a hugetlbfs file successfully. The first patch of the set is purely mechanical code move to make later diffs easier to read. The second patch will guarantee faults up until the process calls fork(). After patch two, as long as the child keeps the mappings, the parent is no longer guaranteed to be reliable. Patch 3 guarantees that the parent will always successfully COW by unmapping the pages from the child in the event there are insufficient pages in the hugepage pool in allocate a new page, be it via a static or dynamic pool. Existing hugepage-aware applications are unlikely to be affected by this change. For much of hugetlbfs's history, pages were pre-faulted at mmap() time or mmap() failed which acts in a reserve-like manner. If the pool is sized correctly already so that parent and child can fault reliably, the application will not even notice the reserves. It's only when the pool is too small for the application to function perfectly reliably that the reserves come into play. Credit goes to Andy Whitcroft for cleaning up a number of mistakes during review before the patches were released. This patch: A later patch in this set needs to call hugetlb_acct_memory() before it is defined. This patch moves the function without modification. This makes later diffs easier to read. Signed-off-by: Mel Gorman Acked-by: Adam Litke Cc: Andy Whitcroft Cc: William Lee Irwin III Cc: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/hugetlb.c | 82 ++++++++++++++++++++++++++-------------------------- 1 file changed, 41 insertions(+), 41 deletions(-) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 2c5c9ee4220d..a4dbba8965f3 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -716,6 +716,47 @@ unsigned long hugetlb_total_pages(void) return nr_huge_pages * (HPAGE_SIZE / PAGE_SIZE); } +static int hugetlb_acct_memory(long delta) +{ + int ret = -ENOMEM; + + spin_lock(&hugetlb_lock); + /* + * When cpuset is configured, it breaks the strict hugetlb page + * reservation as the accounting is done on a global variable. Such + * reservation is completely rubbish in the presence of cpuset because + * the reservation is not checked against page availability for the + * current cpuset. Application can still potentially OOM'ed by kernel + * with lack of free htlb page in cpuset that the task is in. + * Attempt to enforce strict accounting with cpuset is almost + * impossible (or too ugly) because cpuset is too fluid that + * task or memory node can be dynamically moved between cpusets. + * + * The change of semantics for shared hugetlb mapping with cpuset is + * undesirable. However, in order to preserve some of the semantics, + * we fall back to check against current free page availability as + * a best attempt and hopefully to minimize the impact of changing + * semantics that cpuset has. + */ + if (delta > 0) { + if (gather_surplus_pages(delta) < 0) + goto out; + + if (delta > cpuset_mems_nr(free_huge_pages_node)) { + return_unused_surplus_pages(delta); + goto out; + } + } + + ret = 0; + if (delta < 0) + return_unused_surplus_pages((unsigned long) -delta); + +out: + spin_unlock(&hugetlb_lock); + return ret; +} + /* * We cannot handle pagefaults against hugetlb pages at all. They cause * handle_mm_fault() to try to instantiate regular-sized pages in the @@ -1248,47 +1289,6 @@ static long region_truncate(struct list_head *head, long end) return chg; } -static int hugetlb_acct_memory(long delta) -{ - int ret = -ENOMEM; - - spin_lock(&hugetlb_lock); - /* - * When cpuset is configured, it breaks the strict hugetlb page - * reservation as the accounting is done on a global variable. Such - * reservation is completely rubbish in the presence of cpuset because - * the reservation is not checked against page availability for the - * current cpuset. Application can still potentially OOM'ed by kernel - * with lack of free htlb page in cpuset that the task is in. - * Attempt to enforce strict accounting with cpuset is almost - * impossible (or too ugly) because cpuset is too fluid that - * task or memory node can be dynamically moved between cpusets. - * - * The change of semantics for shared hugetlb mapping with cpuset is - * undesirable. However, in order to preserve some of the semantics, - * we fall back to check against current free page availability as - * a best attempt and hopefully to minimize the impact of changing - * semantics that cpuset has. - */ - if (delta > 0) { - if (gather_surplus_pages(delta) < 0) - goto out; - - if (delta > cpuset_mems_nr(free_huge_pages_node)) { - return_unused_surplus_pages(delta); - goto out; - } - } - - ret = 0; - if (delta < 0) - return_unused_surplus_pages((unsigned long) -delta); - -out: - spin_unlock(&hugetlb_lock); - return ret; -} - int hugetlb_reserve_pages(struct inode *inode, long from, long to) { long ret, chg; From a1e78772d72b2616ed20e54896e68e0e7044854e Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Wed, 23 Jul 2008 21:27:23 -0700 Subject: [PATCH 153/857] hugetlb: reserve huge pages for reliable MAP_PRIVATE hugetlbfs mappings until fork() This patch reserves huge pages at mmap() time for MAP_PRIVATE mappings in a similar manner to the reservations taken for MAP_SHARED mappings. The reserve count is accounted both globally and on a per-VMA basis for private mappings. This guarantees that a process that successfully calls mmap() will successfully fault all pages in the future unless fork() is called. The characteristics of private mappings of hugetlbfs files behaviour after this patch are; 1. The process calling mmap() is guaranteed to succeed all future faults until it forks(). 2. On fork(), the parent may die due to SIGKILL on writes to the private mapping if enough pages are not available for the COW. For reasonably reliable behaviour in the face of a small huge page pool, children of hugepage-aware processes should not reference the mappings; such as might occur when fork()ing to exec(). 3. On fork(), the child VMAs inherit no reserves. Reads on pages already faulted by the parent will succeed. Successful writes will depend on enough huge pages being free in the pool. 4. Quotas of the hugetlbfs mount are checked at reserve time for the mapper and at fault time otherwise. Before this patch, all reads or writes in the child potentially needs page allocations that can later lead to the death of the parent. This applies to reads and writes of uninstantiated pages as well as COW. After the patch it is only a write to an instantiated page that causes problems. Signed-off-by: Mel Gorman Acked-by: Adam Litke Cc: Andy Whitcroft Cc: William Lee Irwin III Cc: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/hugetlbfs/inode.c | 8 +- include/linux/hugetlb.h | 9 ++- kernel/fork.c | 9 +++ mm/hugetlb.c | 174 +++++++++++++++++++++++++++++----------- 4 files changed, 148 insertions(+), 52 deletions(-) diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index aeabf80f81a5..1576bbecd084 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -103,9 +103,9 @@ static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma) ret = -ENOMEM; len = vma_len + ((loff_t)vma->vm_pgoff << PAGE_SHIFT); - if (vma->vm_flags & VM_MAYSHARE && - hugetlb_reserve_pages(inode, vma->vm_pgoff >> (HPAGE_SHIFT-PAGE_SHIFT), - len >> HPAGE_SHIFT)) + if (hugetlb_reserve_pages(inode, + vma->vm_pgoff >> (HPAGE_SHIFT-PAGE_SHIFT), + len >> HPAGE_SHIFT, vma)) goto out; ret = 0; @@ -942,7 +942,7 @@ struct file *hugetlb_file_setup(const char *name, size_t size) goto out_dentry; error = -ENOMEM; - if (hugetlb_reserve_pages(inode, 0, size >> HPAGE_SHIFT)) + if (hugetlb_reserve_pages(inode, 0, size >> HPAGE_SHIFT, NULL)) goto out_inode; d_instantiate(dentry, inode); diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index a79e80b689d8..185b14c9f021 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -17,6 +17,7 @@ static inline int is_vm_hugetlb_page(struct vm_area_struct *vma) return vma->vm_flags & VM_HUGETLB; } +void reset_vma_resv_huge_pages(struct vm_area_struct *vma); int hugetlb_sysctl_handler(struct ctl_table *, int, struct file *, void __user *, size_t *, loff_t *); int hugetlb_overcommit_handler(struct ctl_table *, int, struct file *, void __user *, size_t *, loff_t *); int hugetlb_treat_movable_handler(struct ctl_table *, int, struct file *, void __user *, size_t *, loff_t *); @@ -30,7 +31,8 @@ int hugetlb_report_node_meminfo(int, char *); unsigned long hugetlb_total_pages(void); int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long address, int write_access); -int hugetlb_reserve_pages(struct inode *inode, long from, long to); +int hugetlb_reserve_pages(struct inode *inode, long from, long to, + struct vm_area_struct *vma); void hugetlb_unreserve_pages(struct inode *inode, long offset, long freed); extern unsigned long max_huge_pages; @@ -58,6 +60,11 @@ static inline int is_vm_hugetlb_page(struct vm_area_struct *vma) { return 0; } + +static inline void reset_vma_resv_huge_pages(struct vm_area_struct *vma) +{ +} + static inline unsigned long hugetlb_total_pages(void) { return 0; diff --git a/kernel/fork.c b/kernel/fork.c index adefc1131f27..552c8d8e77ad 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -33,6 +33,7 @@ #include #include #include +#include #include #include #include @@ -306,6 +307,14 @@ static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm) spin_unlock(&file->f_mapping->i_mmap_lock); } + /* + * Clear hugetlb-related page reserves for children. This only + * affects MAP_PRIVATE mappings. Faults generated by the child + * are not guaranteed to succeed, even if read-only + */ + if (is_vm_hugetlb_page(tmp)) + reset_vma_resv_huge_pages(tmp); + /* * Link in the new vma and copy the page table entries. */ diff --git a/mm/hugetlb.c b/mm/hugetlb.c index a4dbba8965f3..0af500db3632 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -40,6 +40,69 @@ static int hugetlb_next_nid; */ static DEFINE_SPINLOCK(hugetlb_lock); +/* + * These helpers are used to track how many pages are reserved for + * faults in a MAP_PRIVATE mapping. Only the process that called mmap() + * is guaranteed to have their future faults succeed. + * + * With the exception of reset_vma_resv_huge_pages() which is called at fork(), + * the reserve counters are updated with the hugetlb_lock held. It is safe + * to reset the VMA at fork() time as it is not in use yet and there is no + * chance of the global counters getting corrupted as a result of the values. + */ +static unsigned long vma_resv_huge_pages(struct vm_area_struct *vma) +{ + VM_BUG_ON(!is_vm_hugetlb_page(vma)); + if (!(vma->vm_flags & VM_SHARED)) + return (unsigned long)vma->vm_private_data; + return 0; +} + +static void set_vma_resv_huge_pages(struct vm_area_struct *vma, + unsigned long reserve) +{ + VM_BUG_ON(!is_vm_hugetlb_page(vma)); + VM_BUG_ON(vma->vm_flags & VM_SHARED); + + vma->vm_private_data = (void *)reserve; +} + +/* Decrement the reserved pages in the hugepage pool by one */ +static void decrement_hugepage_resv_vma(struct vm_area_struct *vma) +{ + if (vma->vm_flags & VM_SHARED) { + /* Shared mappings always use reserves */ + resv_huge_pages--; + } else { + /* + * Only the process that called mmap() has reserves for + * private mappings. + */ + if (vma_resv_huge_pages(vma)) { + resv_huge_pages--; + reserve = (unsigned long)vma->vm_private_data - 1; + vma->vm_private_data = (void *)reserve; + } + } +} + +void reset_vma_resv_huge_pages(struct vm_area_struct *vma) +{ + VM_BUG_ON(!is_vm_hugetlb_page(vma)); + if (!(vma->vm_flags & VM_SHARED)) + vma->vm_private_data = (void *)0; +} + +/* Returns true if the VMA has associated reserve pages */ +static int vma_has_private_reserves(struct vm_area_struct *vma) +{ + if (vma->vm_flags & VM_SHARED) + return 0; + if (!vma_resv_huge_pages(vma)) + return 0; + return 1; +} + static void clear_huge_page(struct page *page, unsigned long addr) { int i; @@ -101,6 +164,15 @@ static struct page *dequeue_huge_page_vma(struct vm_area_struct *vma, struct zone *zone; struct zoneref *z; + /* + * A child process with MAP_PRIVATE mappings created by their parent + * have no page reserves. This check ensures that reservations are + * not "stolen". The child may still get SIGKILLed + */ + if (!vma_has_private_reserves(vma) && + free_huge_pages - resv_huge_pages == 0) + return NULL; + for_each_zone_zonelist_nodemask(zone, z, zonelist, MAX_NR_ZONES - 1, nodemask) { nid = zone_to_nid(zone); @@ -111,8 +183,8 @@ static struct page *dequeue_huge_page_vma(struct vm_area_struct *vma, list_del(&page->lru); free_huge_pages--; free_huge_pages_node[nid]--; - if (vma && vma->vm_flags & VM_MAYSHARE) - resv_huge_pages--; + decrement_hugepage_resv_vma(vma); + break; } } @@ -461,55 +533,40 @@ static void return_unused_surplus_pages(unsigned long unused_resv_pages) } } - -static struct page *alloc_huge_page_shared(struct vm_area_struct *vma, - unsigned long addr) -{ - struct page *page; - - spin_lock(&hugetlb_lock); - page = dequeue_huge_page_vma(vma, addr); - spin_unlock(&hugetlb_lock); - return page ? page : ERR_PTR(-VM_FAULT_OOM); -} - -static struct page *alloc_huge_page_private(struct vm_area_struct *vma, - unsigned long addr) -{ - struct page *page = NULL; - - if (hugetlb_get_quota(vma->vm_file->f_mapping, 1)) - return ERR_PTR(-VM_FAULT_SIGBUS); - - spin_lock(&hugetlb_lock); - if (free_huge_pages > resv_huge_pages) - page = dequeue_huge_page_vma(vma, addr); - spin_unlock(&hugetlb_lock); - if (!page) { - page = alloc_buddy_huge_page(vma, addr); - if (!page) { - hugetlb_put_quota(vma->vm_file->f_mapping, 1); - return ERR_PTR(-VM_FAULT_OOM); - } - } - return page; -} - static struct page *alloc_huge_page(struct vm_area_struct *vma, unsigned long addr) { struct page *page; struct address_space *mapping = vma->vm_file->f_mapping; + struct inode *inode = mapping->host; + unsigned int chg = 0; - if (vma->vm_flags & VM_MAYSHARE) - page = alloc_huge_page_shared(vma, addr); - else - page = alloc_huge_page_private(vma, addr); - - if (!IS_ERR(page)) { - set_page_refcounted(page); - set_page_private(page, (unsigned long) mapping); + /* + * Processes that did not create the mapping will have no reserves and + * will not have accounted against quota. Check that the quota can be + * made before satisfying the allocation + */ + if (!vma_has_private_reserves(vma)) { + chg = 1; + if (hugetlb_get_quota(inode->i_mapping, chg)) + return ERR_PTR(-ENOSPC); } + + spin_lock(&hugetlb_lock); + page = dequeue_huge_page_vma(vma, addr); + spin_unlock(&hugetlb_lock); + + if (!page) { + page = alloc_buddy_huge_page(vma, addr); + if (!page) { + hugetlb_put_quota(inode->i_mapping, chg); + return ERR_PTR(-VM_FAULT_OOM); + } + } + + set_page_refcounted(page); + set_page_private(page, (unsigned long) mapping); + return page; } @@ -757,6 +814,13 @@ static int hugetlb_acct_memory(long delta) return ret; } +static void hugetlb_vm_op_close(struct vm_area_struct *vma) +{ + unsigned long reserve = vma_resv_huge_pages(vma); + if (reserve) + hugetlb_acct_memory(-reserve); +} + /* * We cannot handle pagefaults against hugetlb pages at all. They cause * handle_mm_fault() to try to instantiate regular-sized pages in the @@ -771,6 +835,7 @@ static int hugetlb_vm_op_fault(struct vm_area_struct *vma, struct vm_fault *vmf) struct vm_operations_struct hugetlb_vm_ops = { .fault = hugetlb_vm_op_fault, + .close = hugetlb_vm_op_close, }; static pte_t make_huge_pte(struct vm_area_struct *vma, struct page *page, @@ -1289,11 +1354,25 @@ static long region_truncate(struct list_head *head, long end) return chg; } -int hugetlb_reserve_pages(struct inode *inode, long from, long to) +int hugetlb_reserve_pages(struct inode *inode, + long from, long to, + struct vm_area_struct *vma) { long ret, chg; - chg = region_chg(&inode->i_mapping->private_list, from, to); + /* + * Shared mappings base their reservation on the number of pages that + * are already allocated on behalf of the file. Private mappings need + * to reserve the full area even if read-only as mprotect() may be + * called to make the mapping read-write. Assume !vma is a shm mapping + */ + if (!vma || vma->vm_flags & VM_SHARED) + chg = region_chg(&inode->i_mapping->private_list, from, to); + else { + chg = to - from; + set_vma_resv_huge_pages(vma, chg); + } + if (chg < 0) return chg; @@ -1304,7 +1383,8 @@ int hugetlb_reserve_pages(struct inode *inode, long from, long to) hugetlb_put_quota(inode->i_mapping, chg); return ret; } - region_add(&inode->i_mapping->private_list, from, to); + if (!vma || vma->vm_flags & VM_SHARED) + region_add(&inode->i_mapping->private_list, from, to); return 0; } From 04f2cbe35699d22dbf428373682ead85ca1240f5 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Wed, 23 Jul 2008 21:27:25 -0700 Subject: [PATCH 154/857] hugetlb: guarantee that COW faults for a process that called mmap(MAP_PRIVATE) on hugetlbfs will succeed After patch 2 in this series, a process that successfully calls mmap() for a MAP_PRIVATE mapping will be guaranteed to successfully fault until a process calls fork(). At that point, the next write fault from the parent could fail due to COW if the child still has a reference. We only reserve pages for the parent but a copy must be made to avoid leaking data from the parent to the child after fork(). Reserves could be taken for both parent and child at fork time to guarantee faults but if the mapping is large it is highly likely we will not have sufficient pages for the reservation, and it is common to fork only to exec() immediatly after. A failure here would be very undesirable. Note that the current behaviour of mainline with MAP_PRIVATE pages is pretty bad. The following situation is allowed to occur today. 1. Process calls mmap(MAP_PRIVATE) 2. Process calls mlock() to fault all pages and makes sure it succeeds 3. Process forks() 4. Process writes to MAP_PRIVATE mapping while child still exists 5. If the COW fails at this point, the process gets SIGKILLed even though it had taken care to ensure the pages existed This patch improves the situation by guaranteeing the reliability of the process that successfully calls mmap(). When the parent performs COW, it will try to satisfy the allocation without using reserves. If that fails the parent will steal the page leaving any children without a page. Faults from the child after that point will result in failure. If the child COW happens first, an attempt will be made to allocate the page without reserves and the child will get SIGKILLed on failure. To summarise the new behaviour: 1. If the original mapper performs COW on a private mapping with multiple references, it will attempt to allocate a hugepage from the pool or the buddy allocator without using the existing reserves. On fail, VMAs mapping the same area are traversed and the page being COW'd is unmapped where found. It will then steal the original page as the last mapper in the normal way. 2. The VMAs the pages were unmapped from are flagged to note that pages with data no longer exist. Future no-page faults on those VMAs will terminate the process as otherwise it would appear that data was corrupted. A warning is printed to the console that this situation occured. 2. If the child performs COW first, it will attempt to satisfy the COW from the pool if there are enough pages or via the buddy allocator if overcommit is allowed and the buddy allocator can satisfy the request. If it fails, the child will be killed. If the pool is large enough, existing applications will not notice that the reserves were a factor. Existing applications depending on the no-reserves been set are unlikely to exist as for much of the history of hugetlbfs, pages were prefaulted at mmap(), allocating the pages at that point or failing the mmap(). [npiggin@suse.de: fix CONFIG_HUGETLB=n build] Signed-off-by: Mel Gorman Acked-by: Adam Litke Cc: Andy Whitcroft Cc: William Lee Irwin III Cc: Hugh Dickins Cc: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/hugetlbfs/inode.c | 2 +- include/linux/hugetlb.h | 8 +- mm/hugetlb.c | 201 ++++++++++++++++++++++++++++++++++++---- mm/memory.c | 2 +- 4 files changed, 190 insertions(+), 23 deletions(-) diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 1576bbecd084..428eff5b73f3 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -441,7 +441,7 @@ hugetlb_vmtruncate_list(struct prio_tree_root *root, pgoff_t pgoff) v_offset = 0; __unmap_hugepage_range(vma, - vma->vm_start + v_offset, vma->vm_end); + vma->vm_start + v_offset, vma->vm_end, NULL); } } diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index 185b14c9f021..abbc187193a1 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -23,8 +23,10 @@ int hugetlb_overcommit_handler(struct ctl_table *, int, struct file *, void __us int hugetlb_treat_movable_handler(struct ctl_table *, int, struct file *, void __user *, size_t *, loff_t *); int copy_hugetlb_page_range(struct mm_struct *, struct mm_struct *, struct vm_area_struct *); int follow_hugetlb_page(struct mm_struct *, struct vm_area_struct *, struct page **, struct vm_area_struct **, unsigned long *, int *, int, int); -void unmap_hugepage_range(struct vm_area_struct *, unsigned long, unsigned long); -void __unmap_hugepage_range(struct vm_area_struct *, unsigned long, unsigned long); +void unmap_hugepage_range(struct vm_area_struct *, + unsigned long, unsigned long, struct page *); +void __unmap_hugepage_range(struct vm_area_struct *, + unsigned long, unsigned long, struct page *); int hugetlb_prefault(struct address_space *, struct vm_area_struct *); int hugetlb_report_meminfo(char *); int hugetlb_report_node_meminfo(int, char *); @@ -74,7 +76,7 @@ static inline unsigned long hugetlb_total_pages(void) #define follow_huge_addr(mm, addr, write) ERR_PTR(-EINVAL) #define copy_hugetlb_page_range(src, dst, vma) ({ BUG(); 0; }) #define hugetlb_prefault(mapping, vma) ({ BUG(); 0; }) -#define unmap_hugepage_range(vma, start, end) BUG() +#define unmap_hugepage_range(vma, start, end, page) BUG() #define hugetlb_report_meminfo(buf) 0 #define hugetlb_report_node_meminfo(n, buf) 0 #define follow_huge_pmd(mm, addr, pmd, write) NULL diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 0af500db3632..a2d29b84501f 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -40,6 +40,9 @@ static int hugetlb_next_nid; */ static DEFINE_SPINLOCK(hugetlb_lock); +#define HPAGE_RESV_OWNER (1UL << (BITS_PER_LONG - 1)) +#define HPAGE_RESV_UNMAPPED (1UL << (BITS_PER_LONG - 2)) +#define HPAGE_RESV_MASK (HPAGE_RESV_OWNER | HPAGE_RESV_UNMAPPED) /* * These helpers are used to track how many pages are reserved for * faults in a MAP_PRIVATE mapping. Only the process that called mmap() @@ -54,17 +57,32 @@ static unsigned long vma_resv_huge_pages(struct vm_area_struct *vma) { VM_BUG_ON(!is_vm_hugetlb_page(vma)); if (!(vma->vm_flags & VM_SHARED)) - return (unsigned long)vma->vm_private_data; + return (unsigned long)vma->vm_private_data & ~HPAGE_RESV_MASK; return 0; } static void set_vma_resv_huge_pages(struct vm_area_struct *vma, unsigned long reserve) { + unsigned long flags; VM_BUG_ON(!is_vm_hugetlb_page(vma)); VM_BUG_ON(vma->vm_flags & VM_SHARED); - vma->vm_private_data = (void *)reserve; + flags = (unsigned long)vma->vm_private_data & HPAGE_RESV_MASK; + vma->vm_private_data = (void *)(reserve | flags); +} + +static void set_vma_resv_flags(struct vm_area_struct *vma, unsigned long flags) +{ + unsigned long reserveflags = (unsigned long)vma->vm_private_data; + VM_BUG_ON(!is_vm_hugetlb_page(vma)); + vma->vm_private_data = (void *)(reserveflags | flags); +} + +static int is_vma_resv_set(struct vm_area_struct *vma, unsigned long flag) +{ + VM_BUG_ON(!is_vm_hugetlb_page(vma)); + return ((unsigned long)vma->vm_private_data & flag) != 0; } /* Decrement the reserved pages in the hugepage pool by one */ @@ -78,14 +96,18 @@ static void decrement_hugepage_resv_vma(struct vm_area_struct *vma) * Only the process that called mmap() has reserves for * private mappings. */ - if (vma_resv_huge_pages(vma)) { + if (is_vma_resv_set(vma, HPAGE_RESV_OWNER)) { + unsigned long flags, reserve; resv_huge_pages--; + flags = (unsigned long)vma->vm_private_data & + HPAGE_RESV_MASK; reserve = (unsigned long)vma->vm_private_data - 1; - vma->vm_private_data = (void *)reserve; + vma->vm_private_data = (void *)(reserve | flags); } } } +/* Reset counters to 0 and clear all HPAGE_RESV_* flags */ void reset_vma_resv_huge_pages(struct vm_area_struct *vma) { VM_BUG_ON(!is_vm_hugetlb_page(vma)); @@ -153,7 +175,7 @@ static struct page *dequeue_huge_page(void) } static struct page *dequeue_huge_page_vma(struct vm_area_struct *vma, - unsigned long address) + unsigned long address, int avoid_reserve) { int nid; struct page *page = NULL; @@ -173,6 +195,10 @@ static struct page *dequeue_huge_page_vma(struct vm_area_struct *vma, free_huge_pages - resv_huge_pages == 0) return NULL; + /* If reserves cannot be used, ensure enough pages are in the pool */ + if (avoid_reserve && free_huge_pages - resv_huge_pages == 0) + return NULL; + for_each_zone_zonelist_nodemask(zone, z, zonelist, MAX_NR_ZONES - 1, nodemask) { nid = zone_to_nid(zone); @@ -183,7 +209,9 @@ static struct page *dequeue_huge_page_vma(struct vm_area_struct *vma, list_del(&page->lru); free_huge_pages--; free_huge_pages_node[nid]--; - decrement_hugepage_resv_vma(vma); + + if (!avoid_reserve) + decrement_hugepage_resv_vma(vma); break; } @@ -534,7 +562,7 @@ static void return_unused_surplus_pages(unsigned long unused_resv_pages) } static struct page *alloc_huge_page(struct vm_area_struct *vma, - unsigned long addr) + unsigned long addr, int avoid_reserve) { struct page *page; struct address_space *mapping = vma->vm_file->f_mapping; @@ -546,14 +574,15 @@ static struct page *alloc_huge_page(struct vm_area_struct *vma, * will not have accounted against quota. Check that the quota can be * made before satisfying the allocation */ - if (!vma_has_private_reserves(vma)) { + if (!(vma->vm_flags & VM_SHARED) && + !is_vma_resv_set(vma, HPAGE_RESV_OWNER)) { chg = 1; if (hugetlb_get_quota(inode->i_mapping, chg)) return ERR_PTR(-ENOSPC); } spin_lock(&hugetlb_lock); - page = dequeue_huge_page_vma(vma, addr); + page = dequeue_huge_page_vma(vma, addr, avoid_reserve); spin_unlock(&hugetlb_lock); if (!page) { @@ -909,7 +938,7 @@ int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src, } void __unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start, - unsigned long end) + unsigned long end, struct page *ref_page) { struct mm_struct *mm = vma->vm_mm; unsigned long address; @@ -937,6 +966,27 @@ void __unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start, if (huge_pmd_unshare(mm, &address, ptep)) continue; + /* + * If a reference page is supplied, it is because a specific + * page is being unmapped, not a range. Ensure the page we + * are about to unmap is the actual page of interest. + */ + if (ref_page) { + pte = huge_ptep_get(ptep); + if (huge_pte_none(pte)) + continue; + page = pte_page(pte); + if (page != ref_page) + continue; + + /* + * Mark the VMA as having unmapped its page so that + * future faults in this VMA will fail rather than + * looking like data was lost + */ + set_vma_resv_flags(vma, HPAGE_RESV_UNMAPPED); + } + pte = huge_ptep_get_and_clear(mm, address, ptep); if (huge_pte_none(pte)) continue; @@ -955,7 +1005,7 @@ void __unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start, } void unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start, - unsigned long end) + unsigned long end, struct page *ref_page) { /* * It is undesirable to test vma->vm_file as it should be non-null @@ -967,19 +1017,68 @@ void unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start, */ if (vma->vm_file) { spin_lock(&vma->vm_file->f_mapping->i_mmap_lock); - __unmap_hugepage_range(vma, start, end); + __unmap_hugepage_range(vma, start, end, ref_page); spin_unlock(&vma->vm_file->f_mapping->i_mmap_lock); } } +/* + * This is called when the original mapper is failing to COW a MAP_PRIVATE + * mappping it owns the reserve page for. The intention is to unmap the page + * from other VMAs and let the children be SIGKILLed if they are faulting the + * same region. + */ +int unmap_ref_private(struct mm_struct *mm, + struct vm_area_struct *vma, + struct page *page, + unsigned long address) +{ + struct vm_area_struct *iter_vma; + struct address_space *mapping; + struct prio_tree_iter iter; + pgoff_t pgoff; + + /* + * vm_pgoff is in PAGE_SIZE units, hence the different calculation + * from page cache lookup which is in HPAGE_SIZE units. + */ + address = address & huge_page_mask(hstate_vma(vma)); + pgoff = ((address - vma->vm_start) >> PAGE_SHIFT) + + (vma->vm_pgoff >> PAGE_SHIFT); + mapping = (struct address_space *)page_private(page); + + vma_prio_tree_foreach(iter_vma, &iter, &mapping->i_mmap, pgoff, pgoff) { + /* Do not unmap the current VMA */ + if (iter_vma == vma) + continue; + + /* + * Unmap the page from other VMAs without their own reserves. + * They get marked to be SIGKILLed if they fault in these + * areas. This is because a future no-page fault on this VMA + * could insert a zeroed page instead of the data existing + * from the time of fork. This would look like data corruption + */ + if (!is_vma_resv_set(iter_vma, HPAGE_RESV_OWNER)) + unmap_hugepage_range(iter_vma, + address, address + HPAGE_SIZE, + page); + } + + return 1; +} + static int hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma, - unsigned long address, pte_t *ptep, pte_t pte) + unsigned long address, pte_t *ptep, pte_t pte, + struct page *pagecache_page) { struct page *old_page, *new_page; int avoidcopy; + int outside_reserve = 0; old_page = pte_page(pte); +retry_avoidcopy: /* If no-one else is actually using this page, avoid the copy * and just make the page writable */ avoidcopy = (page_count(old_page) == 1); @@ -988,11 +1087,43 @@ static int hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma, return 0; } + /* + * If the process that created a MAP_PRIVATE mapping is about to + * perform a COW due to a shared page count, attempt to satisfy + * the allocation without using the existing reserves. The pagecache + * page is used to determine if the reserve at this address was + * consumed or not. If reserves were used, a partial faulted mapping + * at the time of fork() could consume its reserves on COW instead + * of the full address range. + */ + if (!(vma->vm_flags & VM_SHARED) && + is_vma_resv_set(vma, HPAGE_RESV_OWNER) && + old_page != pagecache_page) + outside_reserve = 1; + page_cache_get(old_page); - new_page = alloc_huge_page(vma, address); + new_page = alloc_huge_page(vma, address, outside_reserve); if (IS_ERR(new_page)) { page_cache_release(old_page); + + /* + * If a process owning a MAP_PRIVATE mapping fails to COW, + * it is due to references held by a child and an insufficient + * huge page pool. To guarantee the original mappers + * reliability, unmap the page from child processes. The child + * may get SIGKILLed if it later faults. + */ + if (outside_reserve) { + BUG_ON(huge_pte_none(pte)); + if (unmap_ref_private(mm, vma, old_page, address)) { + BUG_ON(page_count(old_page) != 1); + BUG_ON(huge_pte_none(pte)); + goto retry_avoidcopy; + } + WARN_ON_ONCE(1); + } + return -PTR_ERR(new_page); } @@ -1015,6 +1146,20 @@ static int hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma, return 0; } +/* Return the pagecache page at a given address within a VMA */ +static struct page *hugetlbfs_pagecache_page(struct vm_area_struct *vma, + unsigned long address) +{ + struct address_space *mapping; + unsigned long idx; + + mapping = vma->vm_file->f_mapping; + idx = ((address - vma->vm_start) >> HPAGE_SHIFT) + + (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT)); + + return find_lock_page(mapping, idx); +} + static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long address, pte_t *ptep, int write_access) { @@ -1025,6 +1170,18 @@ static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma, struct address_space *mapping; pte_t new_pte; + /* + * Currently, we are forced to kill the process in the event the + * original mapper has unmapped pages from the child due to a failed + * COW. Warn that such a situation has occured as it may not be obvious + */ + if (is_vma_resv_set(vma, HPAGE_RESV_UNMAPPED)) { + printk(KERN_WARNING + "PID %d killed due to inadequate hugepage pool\n", + current->pid); + return ret; + } + mapping = vma->vm_file->f_mapping; idx = ((address - vma->vm_start) >> HPAGE_SHIFT) + (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT)); @@ -1039,7 +1196,7 @@ static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma, size = i_size_read(mapping->host) >> HPAGE_SHIFT; if (idx >= size) goto out; - page = alloc_huge_page(vma, address); + page = alloc_huge_page(vma, address, 0); if (IS_ERR(page)) { ret = -PTR_ERR(page); goto out; @@ -1081,7 +1238,7 @@ static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma, if (write_access && !(vma->vm_flags & VM_SHARED)) { /* Optimization, do the COW without a second fault */ - ret = hugetlb_cow(mm, vma, address, ptep, new_pte); + ret = hugetlb_cow(mm, vma, address, ptep, new_pte, page); } spin_unlock(&mm->page_table_lock); @@ -1126,8 +1283,15 @@ int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma, spin_lock(&mm->page_table_lock); /* Check for a racing update before calling hugetlb_cow */ if (likely(pte_same(entry, huge_ptep_get(ptep)))) - if (write_access && !pte_write(entry)) - ret = hugetlb_cow(mm, vma, address, ptep, entry); + if (write_access && !pte_write(entry)) { + struct page *page; + page = hugetlbfs_pagecache_page(vma, address); + ret = hugetlb_cow(mm, vma, address, ptep, entry, page); + if (page) { + unlock_page(page); + put_page(page); + } + } spin_unlock(&mm->page_table_lock); mutex_unlock(&hugetlb_instantiation_mutex); @@ -1371,6 +1535,7 @@ int hugetlb_reserve_pages(struct inode *inode, else { chg = to - from; set_vma_resv_huge_pages(vma, chg); + set_vma_resv_flags(vma, HPAGE_RESV_OWNER); } if (chg < 0) diff --git a/mm/memory.c b/mm/memory.c index 82f3f1c5cf17..72932489a082 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -901,7 +901,7 @@ unsigned long unmap_vmas(struct mmu_gather **tlbp, } if (unlikely(is_vm_hugetlb_page(vma))) { - unmap_hugepage_range(vma, start, end); + unmap_hugepage_range(vma, start, end, NULL); zap_work -= (end - start) / (HPAGE_SIZE / PAGE_SIZE); start = end; From e7c4b0bfd025f71cf7624b7c1be174f63caade33 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:27:26 -0700 Subject: [PATCH 155/857] huge page private reservation review cleanups Create some new accessors for vma private data to cut down on and contain the casts. Encapsulates the huge and small page offset calculations. Also adds a couple of VM_BUG_ONs for consistency. [akpm@linux-foundation.org: Make things static] Signed-off-by: Andy Whitcroft Acked-by: Mel Gorman Cc: Adam Litke Cc: Johannes Weiner Cc: Andy Whitcroft Cc: William Lee Irwin III Cc: Hugh Dickins Cc: Michael Kerrisk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/hugetlb.c | 58 ++++++++++++++++++++++++++++++++++++++++------------ 1 file changed, 45 insertions(+), 13 deletions(-) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index a2d29b84501f..3e873f0101fb 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -40,6 +40,28 @@ static int hugetlb_next_nid; */ static DEFINE_SPINLOCK(hugetlb_lock); +/* + * Convert the address within this vma to the page offset within + * the mapping, in base page units. + */ +static pgoff_t vma_page_offset(struct vm_area_struct *vma, + unsigned long address) +{ + return ((address - vma->vm_start) >> PAGE_SHIFT) + + (vma->vm_pgoff >> PAGE_SHIFT); +} + +/* + * Convert the address within this vma to the page offset within + * the mapping, in pagecache page units; huge pages here. + */ +static pgoff_t vma_pagecache_offset(struct vm_area_struct *vma, + unsigned long address) +{ + return ((address - vma->vm_start) >> HPAGE_SHIFT) + + (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT)); +} + #define HPAGE_RESV_OWNER (1UL << (BITS_PER_LONG - 1)) #define HPAGE_RESV_UNMAPPED (1UL << (BITS_PER_LONG - 2)) #define HPAGE_RESV_MASK (HPAGE_RESV_OWNER | HPAGE_RESV_UNMAPPED) @@ -53,36 +75,48 @@ static DEFINE_SPINLOCK(hugetlb_lock); * to reset the VMA at fork() time as it is not in use yet and there is no * chance of the global counters getting corrupted as a result of the values. */ +static unsigned long get_vma_private_data(struct vm_area_struct *vma) +{ + return (unsigned long)vma->vm_private_data; +} + +static void set_vma_private_data(struct vm_area_struct *vma, + unsigned long value) +{ + vma->vm_private_data = (void *)value; +} + static unsigned long vma_resv_huge_pages(struct vm_area_struct *vma) { VM_BUG_ON(!is_vm_hugetlb_page(vma)); if (!(vma->vm_flags & VM_SHARED)) - return (unsigned long)vma->vm_private_data & ~HPAGE_RESV_MASK; + return get_vma_private_data(vma) & ~HPAGE_RESV_MASK; return 0; } static void set_vma_resv_huge_pages(struct vm_area_struct *vma, unsigned long reserve) { - unsigned long flags; VM_BUG_ON(!is_vm_hugetlb_page(vma)); VM_BUG_ON(vma->vm_flags & VM_SHARED); - flags = (unsigned long)vma->vm_private_data & HPAGE_RESV_MASK; - vma->vm_private_data = (void *)(reserve | flags); + set_vma_private_data(vma, + (get_vma_private_data(vma) & HPAGE_RESV_MASK) | reserve); } static void set_vma_resv_flags(struct vm_area_struct *vma, unsigned long flags) { - unsigned long reserveflags = (unsigned long)vma->vm_private_data; VM_BUG_ON(!is_vm_hugetlb_page(vma)); - vma->vm_private_data = (void *)(reserveflags | flags); + VM_BUG_ON(vma->vm_flags & VM_SHARED); + + set_vma_private_data(vma, get_vma_private_data(vma) | flags); } static int is_vma_resv_set(struct vm_area_struct *vma, unsigned long flag) { VM_BUG_ON(!is_vm_hugetlb_page(vma)); - return ((unsigned long)vma->vm_private_data & flag) != 0; + + return (get_vma_private_data(vma) & flag) != 0; } /* Decrement the reserved pages in the hugepage pool by one */ @@ -1151,11 +1185,10 @@ static struct page *hugetlbfs_pagecache_page(struct vm_area_struct *vma, unsigned long address) { struct address_space *mapping; - unsigned long idx; + pgoff_t idx; mapping = vma->vm_file->f_mapping; - idx = ((address - vma->vm_start) >> HPAGE_SHIFT) - + (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT)); + idx = vma_pagecache_offset(vma, address); return find_lock_page(mapping, idx); } @@ -1164,7 +1197,7 @@ static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long address, pte_t *ptep, int write_access) { int ret = VM_FAULT_SIGBUS; - unsigned long idx; + pgoff_t idx; unsigned long size; struct page *page; struct address_space *mapping; @@ -1183,8 +1216,7 @@ static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma, } mapping = vma->vm_file->f_mapping; - idx = ((address - vma->vm_start) >> HPAGE_SHIFT) - + (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT)); + idx = vma_pagecache_offset(vma, address); /* * Use page lock to guard against racing truncation From cdfd4325c0d878679bd6a3ba8285b71d9980e3c0 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:27:28 -0700 Subject: [PATCH 156/857] mm: record MAP_NORESERVE status on vmas and fix small page mprotect reservations With Mel's hugetlb private reservation support patches applied, strict overcommit semantics are applied to both shared and private huge page mappings. This can be a problem if an application relied on unlimited overcommit semantics for private mappings. An example of this would be an application which maps a huge area with the intention of using it very sparsely. These application would benefit from being able to opt-out of the strict overcommit. It should be noted that prior to hugetlb supporting demand faulting all mappings were fully populated and so applications of this type should be rare. This patch stack implements the MAP_NORESERVE mmap() flag for huge page mappings. This flag has the same meaning as for small page mappings, suppressing reservations for that mapping. Thanks to Mel Gorman for reviewing a number of early versions of these patches. This patch: When a small page mapping is created with mmap() reservations are created by default for any memory pages required. When the region is read/write the reservation is increased for every page, no reservation is needed for read-only regions (as they implicitly share the zero page). Reservations are tracked via the VM_ACCOUNT vma flag which is present when the region has reservation backing it. When we convert a region from read-only to read-write new reservations are aquired and VM_ACCOUNT is set. However, when a read-only map is created with MAP_NORESERVE it is indistinguishable from a normal mapping. When we then convert that to read/write we are forced to incorrectly create reservations for it as we have no record of the original MAP_NORESERVE. This patch introduces a new vma flag VM_NORESERVE which records the presence of the original MAP_NORESERVE flag. This allows us to distinguish these two circumstances and correctly account the reserve. As well as fixing this FIXME in the code, this makes it much easier to introduce MAP_NORESERVE support for huge pages as this flag is available consistantly for the life of the mapping. VM_ACCOUNT on the other hand is heavily used at the generic level in association with small pages. Signed-off-by: Andy Whitcroft Cc: Mel Gorman Cc: Adam Litke Cc: Johannes Weiner Cc: Andy Whitcroft Cc: William Lee Irwin III Cc: Hugh Dickins Cc: Michael Kerrisk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 1 + mm/mmap.c | 3 +++ mm/mprotect.c | 6 ++---- 3 files changed, 6 insertions(+), 4 deletions(-) diff --git a/include/linux/mm.h b/include/linux/mm.h index 196924b657bc..df322fb4df31 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -100,6 +100,7 @@ extern unsigned int kobjsize(const void *objp); #define VM_DONTEXPAND 0x00040000 /* Cannot expand with mremap() */ #define VM_RESERVED 0x00080000 /* Count as reserved_vm like IO */ #define VM_ACCOUNT 0x00100000 /* Is a VM accounted object */ +#define VM_NORESERVE 0x00200000 /* should the VM suppress accounting */ #define VM_HUGETLB 0x00400000 /* Huge TLB Page VM */ #define VM_NONLINEAR 0x00800000 /* Is non-linear (remap_file_pages) */ #define VM_MAPPED_COPY 0x01000000 /* T if mapped copy of data (nommu mmap) */ diff --git a/mm/mmap.c b/mm/mmap.c index 75e0d0673d78..57d3b6097deb 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1110,6 +1110,9 @@ unsigned long mmap_region(struct file *file, unsigned long addr, if (!may_expand_vm(mm, len >> PAGE_SHIFT)) return -ENOMEM; + if (flags & MAP_NORESERVE) + vm_flags |= VM_NORESERVE; + if (accountable && (!(flags & MAP_NORESERVE) || sysctl_overcommit_memory == OVERCOMMIT_NEVER)) { if (vm_flags & VM_SHARED) { diff --git a/mm/mprotect.c b/mm/mprotect.c index 360d9cc8b38c..abd645a3b0a0 100644 --- a/mm/mprotect.c +++ b/mm/mprotect.c @@ -153,12 +153,10 @@ mprotect_fixup(struct vm_area_struct *vma, struct vm_area_struct **pprev, * If we make a private mapping writable we increase our commit; * but (without finer accounting) cannot reduce our commit if we * make it unwritable again. - * - * FIXME? We haven't defined a VM_NORESERVE flag, so mprotecting - * a MAP_NORESERVE private mapping to writable will now reserve. */ if (newflags & VM_WRITE) { - if (!(oldflags & (VM_ACCOUNT|VM_WRITE|VM_SHARED))) { + if (!(oldflags & (VM_ACCOUNT|VM_WRITE| + VM_SHARED|VM_NORESERVE))) { charged = nrpages; if (security_vm_enough_memory(charged)) return -ENOMEM; From 9682290484370ce68ba23cd2ec2838e301934199 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:27:29 -0700 Subject: [PATCH 157/857] hugetlb: move reservation region support earlier The following patch will require use of the reservation regions support. Move this earlier in the file. No changes have been made to this code. Signed-off-by: Andy Whitcroft Cc: Mel Gorman Acked-by: Adam Litke Cc: Johannes Weiner Cc: Andy Whitcroft Cc: William Lee Irwin III Cc: Hugh Dickins Cc: Michael Kerrisk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/hugetlb.c | 246 ++++++++++++++++++++++++++------------------------- 1 file changed, 125 insertions(+), 121 deletions(-) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 3e873f0101fb..05bc9af4fca9 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -40,6 +40,131 @@ static int hugetlb_next_nid; */ static DEFINE_SPINLOCK(hugetlb_lock); +/* + * Region tracking -- allows tracking of reservations and instantiated pages + * across the pages in a mapping. + */ +struct file_region { + struct list_head link; + long from; + long to; +}; + +static long region_add(struct list_head *head, long f, long t) +{ + struct file_region *rg, *nrg, *trg; + + /* Locate the region we are either in or before. */ + list_for_each_entry(rg, head, link) + if (f <= rg->to) + break; + + /* Round our left edge to the current segment if it encloses us. */ + if (f > rg->from) + f = rg->from; + + /* Check for and consume any regions we now overlap with. */ + nrg = rg; + list_for_each_entry_safe(rg, trg, rg->link.prev, link) { + if (&rg->link == head) + break; + if (rg->from > t) + break; + + /* If this area reaches higher then extend our area to + * include it completely. If this is not the first area + * which we intend to reuse, free it. */ + if (rg->to > t) + t = rg->to; + if (rg != nrg) { + list_del(&rg->link); + kfree(rg); + } + } + nrg->from = f; + nrg->to = t; + return 0; +} + +static long region_chg(struct list_head *head, long f, long t) +{ + struct file_region *rg, *nrg; + long chg = 0; + + /* Locate the region we are before or in. */ + list_for_each_entry(rg, head, link) + if (f <= rg->to) + break; + + /* If we are below the current region then a new region is required. + * Subtle, allocate a new region at the position but make it zero + * size such that we can guarantee to record the reservation. */ + if (&rg->link == head || t < rg->from) { + nrg = kmalloc(sizeof(*nrg), GFP_KERNEL); + if (!nrg) + return -ENOMEM; + nrg->from = f; + nrg->to = f; + INIT_LIST_HEAD(&nrg->link); + list_add(&nrg->link, rg->link.prev); + + return t - f; + } + + /* Round our left edge to the current segment if it encloses us. */ + if (f > rg->from) + f = rg->from; + chg = t - f; + + /* Check for and consume any regions we now overlap with. */ + list_for_each_entry(rg, rg->link.prev, link) { + if (&rg->link == head) + break; + if (rg->from > t) + return chg; + + /* We overlap with this area, if it extends futher than + * us then we must extend ourselves. Account for its + * existing reservation. */ + if (rg->to > t) { + chg += rg->to - t; + t = rg->to; + } + chg -= rg->to - rg->from; + } + return chg; +} + +static long region_truncate(struct list_head *head, long end) +{ + struct file_region *rg, *trg; + long chg = 0; + + /* Locate the region we are either in or before. */ + list_for_each_entry(rg, head, link) + if (end <= rg->to) + break; + if (&rg->link == head) + return 0; + + /* If we are in the middle of a region then adjust it. */ + if (end > rg->from) { + chg = rg->to - end; + rg->to = end; + rg = list_entry(rg->link.next, typeof(*rg), link); + } + + /* Drop any remaining regions. */ + list_for_each_entry_safe(rg, trg, rg->link.prev, link) { + if (&rg->link == head) + break; + chg += rg->to - rg->from; + list_del(&rg->link); + kfree(rg); + } + return chg; +} + /* * Convert the address within this vma to the page offset within * the mapping, in base page units. @@ -1429,127 +1554,6 @@ void hugetlb_change_protection(struct vm_area_struct *vma, flush_tlb_range(vma, start, end); } -struct file_region { - struct list_head link; - long from; - long to; -}; - -static long region_add(struct list_head *head, long f, long t) -{ - struct file_region *rg, *nrg, *trg; - - /* Locate the region we are either in or before. */ - list_for_each_entry(rg, head, link) - if (f <= rg->to) - break; - - /* Round our left edge to the current segment if it encloses us. */ - if (f > rg->from) - f = rg->from; - - /* Check for and consume any regions we now overlap with. */ - nrg = rg; - list_for_each_entry_safe(rg, trg, rg->link.prev, link) { - if (&rg->link == head) - break; - if (rg->from > t) - break; - - /* If this area reaches higher then extend our area to - * include it completely. If this is not the first area - * which we intend to reuse, free it. */ - if (rg->to > t) - t = rg->to; - if (rg != nrg) { - list_del(&rg->link); - kfree(rg); - } - } - nrg->from = f; - nrg->to = t; - return 0; -} - -static long region_chg(struct list_head *head, long f, long t) -{ - struct file_region *rg, *nrg; - long chg = 0; - - /* Locate the region we are before or in. */ - list_for_each_entry(rg, head, link) - if (f <= rg->to) - break; - - /* If we are below the current region then a new region is required. - * Subtle, allocate a new region at the position but make it zero - * size such that we can guarantee to record the reservation. */ - if (&rg->link == head || t < rg->from) { - nrg = kmalloc(sizeof(*nrg), GFP_KERNEL); - if (!nrg) - return -ENOMEM; - nrg->from = f; - nrg->to = f; - INIT_LIST_HEAD(&nrg->link); - list_add(&nrg->link, rg->link.prev); - - return t - f; - } - - /* Round our left edge to the current segment if it encloses us. */ - if (f > rg->from) - f = rg->from; - chg = t - f; - - /* Check for and consume any regions we now overlap with. */ - list_for_each_entry(rg, rg->link.prev, link) { - if (&rg->link == head) - break; - if (rg->from > t) - return chg; - - /* We overlap with this area, if it extends futher than - * us then we must extend ourselves. Account for its - * existing reservation. */ - if (rg->to > t) { - chg += rg->to - t; - t = rg->to; - } - chg -= rg->to - rg->from; - } - return chg; -} - -static long region_truncate(struct list_head *head, long end) -{ - struct file_region *rg, *trg; - long chg = 0; - - /* Locate the region we are either in or before. */ - list_for_each_entry(rg, head, link) - if (end <= rg->to) - break; - if (&rg->link == head) - return 0; - - /* If we are in the middle of a region then adjust it. */ - if (end > rg->from) { - chg = rg->to - end; - rg->to = end; - rg = list_entry(rg->link.next, typeof(*rg), link); - } - - /* Drop any remaining regions. */ - list_for_each_entry_safe(rg, trg, rg->link.prev, link) { - if (&rg->link == head) - break; - chg += rg->to - rg->from; - list_del(&rg->link); - kfree(rg); - } - return chg; -} - int hugetlb_reserve_pages(struct inode *inode, long from, long to, struct vm_area_struct *vma) From c37f9fb11c976ffc08200d631dada6dcbfd07ea4 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:27:30 -0700 Subject: [PATCH 158/857] hugetlb: allow huge page mappings to be created without reservations By default all shared mappings and most private mappings now have reservations associated with them. This improves semantics by providing allocation guarentees to the mapper. However a small number of applications may attempt to make very large sparse mappings, with these strict reservations the system will never be able to honour the mapping. This patch set brings MAP_NORESERVE support to hugetlb files. This allows new mappings to be made to hugetlbfs files without an associated reservation, for both shared and private mappings. This allows applications which want to create very sparse mappings to opt-out of the reservation system. Obviously as there is no reservation they are liable to fault at runtime if the huge page pool becomes exhausted; buyer beware. Signed-off-by: Andy Whitcroft Cc: Mel Gorman Cc: Adam Litke Cc: Johannes Weiner Cc: Andy Whitcroft Cc: William Lee Irwin III Cc: Hugh Dickins Cc: Michael Kerrisk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/hugetlb.c | 58 +++++++++++++++++++++++++++++++++++++++++++++++----- 1 file changed, 53 insertions(+), 5 deletions(-) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 05bc9af4fca9..72acbb29d2cc 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -247,6 +247,9 @@ static int is_vma_resv_set(struct vm_area_struct *vma, unsigned long flag) /* Decrement the reserved pages in the hugepage pool by one */ static void decrement_hugepage_resv_vma(struct vm_area_struct *vma) { + if (vma->vm_flags & VM_NORESERVE) + return; + if (vma->vm_flags & VM_SHARED) { /* Shared mappings always use reserves */ resv_huge_pages--; @@ -720,25 +723,65 @@ static void return_unused_surplus_pages(unsigned long unused_resv_pages) } } +/* + * Determine if the huge page at addr within the vma has an associated + * reservation. Where it does not we will need to logically increase + * reservation and actually increase quota before an allocation can occur. + * Where any new reservation would be required the reservation change is + * prepared, but not committed. Once the page has been quota'd allocated + * an instantiated the change should be committed via vma_commit_reservation. + * No action is required on failure. + */ +static int vma_needs_reservation(struct vm_area_struct *vma, unsigned long addr) +{ + struct address_space *mapping = vma->vm_file->f_mapping; + struct inode *inode = mapping->host; + + if (vma->vm_flags & VM_SHARED) { + pgoff_t idx = vma_pagecache_offset(vma, addr); + return region_chg(&inode->i_mapping->private_list, + idx, idx + 1); + + } else { + if (!is_vma_resv_set(vma, HPAGE_RESV_OWNER)) + return 1; + } + + return 0; +} +static void vma_commit_reservation(struct vm_area_struct *vma, + unsigned long addr) +{ + struct address_space *mapping = vma->vm_file->f_mapping; + struct inode *inode = mapping->host; + + if (vma->vm_flags & VM_SHARED) { + pgoff_t idx = vma_pagecache_offset(vma, addr); + region_add(&inode->i_mapping->private_list, idx, idx + 1); + } +} + static struct page *alloc_huge_page(struct vm_area_struct *vma, unsigned long addr, int avoid_reserve) { struct page *page; struct address_space *mapping = vma->vm_file->f_mapping; struct inode *inode = mapping->host; - unsigned int chg = 0; + unsigned int chg; /* * Processes that did not create the mapping will have no reserves and * will not have accounted against quota. Check that the quota can be * made before satisfying the allocation + * MAP_NORESERVE mappings may also need pages and quota allocated + * if no reserve mapping overlaps. */ - if (!(vma->vm_flags & VM_SHARED) && - !is_vma_resv_set(vma, HPAGE_RESV_OWNER)) { - chg = 1; + chg = vma_needs_reservation(vma, addr); + if (chg < 0) + return ERR_PTR(chg); + if (chg) if (hugetlb_get_quota(inode->i_mapping, chg)) return ERR_PTR(-ENOSPC); - } spin_lock(&hugetlb_lock); page = dequeue_huge_page_vma(vma, addr, avoid_reserve); @@ -755,6 +798,8 @@ static struct page *alloc_huge_page(struct vm_area_struct *vma, set_page_refcounted(page); set_page_private(page, (unsigned long) mapping); + vma_commit_reservation(vma, addr); + return page; } @@ -1560,6 +1605,9 @@ int hugetlb_reserve_pages(struct inode *inode, { long ret, chg; + if (vma && vma->vm_flags & VM_NORESERVE) + return 0; + /* * Shared mappings base their reservation on the number of pages that * are already allocated on behalf of the file. Private mappings need From 84afd99b8398c9d73af8238aa3cd835858e3097a Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:27:32 -0700 Subject: [PATCH 159/857] hugetlb reservations: fix hugetlb MAP_PRIVATE reservations across vma splits When a hugetlb mapping with a reservation is split, a new VMA is cloned from the original. This new VMA is a direct copy of the original including the reservation count. When this pair of VMAs are unmapped we will incorrect double account the unused reservation and the overall reservation count will be incorrect, in extreme cases it will wrap. The problem occurs when we split an existing VMA say to unmap a page in the middle. split_vma() will create a new VMA copying all fields from the original. As we are storing our reservation count in vm_private_data this is also copies, endowing the new VMA with a duplicate of the original VMA's reservation. Neither of the new VMAs can exhaust these reservations as they are too small, but when we unmap and close these VMAs we will incorrect credit the remainder twice and resv_huge_pages will become out of sync. This can lead to allocation failures on mappings with reservations and even to resv_huge_pages wrapping which prevents all subsequent hugepage allocations. The simple fix would be to correctly apportion the remaining reservation count when the split is made. However the only hook we have vm_ops->open only has the new VMA we do not know the identity of the preceeding VMA. Also even if we did have that VMA to hand we do not know how much of the reservation was consumed each side of the split. This patch therefore takes a different tack. We know that the whole of any private mapping (which has a reservation) has a reservation over its whole size. Any present pages represent consumed reservation. Therefore if we track the instantiated pages we can calculate the remaining reservation. This patch reuses the existing regions code to track the regions for which we have consumed reservation (ie. the instantiated pages), as each page is faulted in we record the consumption of reservation for the new page. When we need to return unused reservations at unmap time we simply count the consumed reservation region subtracting that from the whole of the map. During a VMA split the newly opened VMA will point to the same region map, as this map is offset oriented it remains valid for both of the split VMAs. This map is referenced counted so that it is removed when all VMAs which are part of the mmap are gone. Thanks to Adam Litke and Mel Gorman for their review feedback. Signed-off-by: Andy Whitcroft Acked-by: Mel Gorman Cc: Adam Litke Cc: Johannes Weiner Cc: Andy Whitcroft Cc: William Lee Irwin III Cc: Hugh Dickins Cc: Michael Kerrisk Cc: Jon Tollefson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/hugetlb.c | 172 +++++++++++++++++++++++++++++++++++++++++++-------- 1 file changed, 145 insertions(+), 27 deletions(-) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 72acbb29d2cc..65616941a383 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -43,6 +43,16 @@ static DEFINE_SPINLOCK(hugetlb_lock); /* * Region tracking -- allows tracking of reservations and instantiated pages * across the pages in a mapping. + * + * The region data structures are protected by a combination of the mmap_sem + * and the hugetlb_instantion_mutex. To access or modify a region the caller + * must either hold the mmap_sem for write, or the mmap_sem for read and + * the hugetlb_instantiation mutex: + * + * down_write(&mm->mmap_sem); + * or + * down_read(&mm->mmap_sem); + * mutex_lock(&hugetlb_instantiation_mutex); */ struct file_region { struct list_head link; @@ -165,6 +175,30 @@ static long region_truncate(struct list_head *head, long end) return chg; } +static long region_count(struct list_head *head, long f, long t) +{ + struct file_region *rg; + long chg = 0; + + /* Locate each segment we overlap with, and count that overlap. */ + list_for_each_entry(rg, head, link) { + int seg_from; + int seg_to; + + if (rg->to <= f) + continue; + if (rg->from >= t) + break; + + seg_from = max(rg->from, f); + seg_to = min(rg->to, t); + + chg += seg_to - seg_from; + } + + return chg; +} + /* * Convert the address within this vma to the page offset within * the mapping, in base page units. @@ -187,9 +221,15 @@ static pgoff_t vma_pagecache_offset(struct vm_area_struct *vma, (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT)); } -#define HPAGE_RESV_OWNER (1UL << (BITS_PER_LONG - 1)) -#define HPAGE_RESV_UNMAPPED (1UL << (BITS_PER_LONG - 2)) +/* + * Flags for MAP_PRIVATE reservations. These are stored in the bottom + * bits of the reservation map pointer, which are always clear due to + * alignment. + */ +#define HPAGE_RESV_OWNER (1UL << 0) +#define HPAGE_RESV_UNMAPPED (1UL << 1) #define HPAGE_RESV_MASK (HPAGE_RESV_OWNER | HPAGE_RESV_UNMAPPED) + /* * These helpers are used to track how many pages are reserved for * faults in a MAP_PRIVATE mapping. Only the process that called mmap() @@ -199,6 +239,15 @@ static pgoff_t vma_pagecache_offset(struct vm_area_struct *vma, * the reserve counters are updated with the hugetlb_lock held. It is safe * to reset the VMA at fork() time as it is not in use yet and there is no * chance of the global counters getting corrupted as a result of the values. + * + * The private mapping reservation is represented in a subtly different + * manner to a shared mapping. A shared mapping has a region map associated + * with the underlying file, this region map represents the backing file + * pages which have ever had a reservation assigned which this persists even + * after the page is instantiated. A private mapping has a region map + * associated with the original mmap which is attached to all VMAs which + * reference it, this region map represents those offsets which have consumed + * reservation ie. where pages have been instantiated. */ static unsigned long get_vma_private_data(struct vm_area_struct *vma) { @@ -211,22 +260,48 @@ static void set_vma_private_data(struct vm_area_struct *vma, vma->vm_private_data = (void *)value; } -static unsigned long vma_resv_huge_pages(struct vm_area_struct *vma) +struct resv_map { + struct kref refs; + struct list_head regions; +}; + +struct resv_map *resv_map_alloc(void) +{ + struct resv_map *resv_map = kmalloc(sizeof(*resv_map), GFP_KERNEL); + if (!resv_map) + return NULL; + + kref_init(&resv_map->refs); + INIT_LIST_HEAD(&resv_map->regions); + + return resv_map; +} + +void resv_map_release(struct kref *ref) +{ + struct resv_map *resv_map = container_of(ref, struct resv_map, refs); + + /* Clear out any active regions before we release the map. */ + region_truncate(&resv_map->regions, 0); + kfree(resv_map); +} + +static struct resv_map *vma_resv_map(struct vm_area_struct *vma) { VM_BUG_ON(!is_vm_hugetlb_page(vma)); if (!(vma->vm_flags & VM_SHARED)) - return get_vma_private_data(vma) & ~HPAGE_RESV_MASK; + return (struct resv_map *)(get_vma_private_data(vma) & + ~HPAGE_RESV_MASK); return 0; } -static void set_vma_resv_huge_pages(struct vm_area_struct *vma, - unsigned long reserve) +static void set_vma_resv_map(struct vm_area_struct *vma, struct resv_map *map) { VM_BUG_ON(!is_vm_hugetlb_page(vma)); VM_BUG_ON(vma->vm_flags & VM_SHARED); - set_vma_private_data(vma, - (get_vma_private_data(vma) & HPAGE_RESV_MASK) | reserve); + set_vma_private_data(vma, (get_vma_private_data(vma) & + HPAGE_RESV_MASK) | (unsigned long)map); } static void set_vma_resv_flags(struct vm_area_struct *vma, unsigned long flags) @@ -253,19 +328,12 @@ static void decrement_hugepage_resv_vma(struct vm_area_struct *vma) if (vma->vm_flags & VM_SHARED) { /* Shared mappings always use reserves */ resv_huge_pages--; - } else { + } else if (is_vma_resv_set(vma, HPAGE_RESV_OWNER)) { /* * Only the process that called mmap() has reserves for * private mappings. */ - if (is_vma_resv_set(vma, HPAGE_RESV_OWNER)) { - unsigned long flags, reserve; - resv_huge_pages--; - flags = (unsigned long)vma->vm_private_data & - HPAGE_RESV_MASK; - reserve = (unsigned long)vma->vm_private_data - 1; - vma->vm_private_data = (void *)(reserve | flags); - } + resv_huge_pages--; } } @@ -282,7 +350,7 @@ static int vma_has_private_reserves(struct vm_area_struct *vma) { if (vma->vm_flags & VM_SHARED) return 0; - if (!vma_resv_huge_pages(vma)) + if (!is_vma_resv_set(vma, HPAGE_RESV_OWNER)) return 0; return 1; } @@ -742,12 +810,19 @@ static int vma_needs_reservation(struct vm_area_struct *vma, unsigned long addr) return region_chg(&inode->i_mapping->private_list, idx, idx + 1); - } else { - if (!is_vma_resv_set(vma, HPAGE_RESV_OWNER)) - return 1; - } + } else if (!is_vma_resv_set(vma, HPAGE_RESV_OWNER)) { + return 1; - return 0; + } else { + int err; + pgoff_t idx = vma_pagecache_offset(vma, addr); + struct resv_map *reservations = vma_resv_map(vma); + + err = region_chg(&reservations->regions, idx, idx + 1); + if (err < 0) + return err; + return 0; + } } static void vma_commit_reservation(struct vm_area_struct *vma, unsigned long addr) @@ -758,6 +833,13 @@ static void vma_commit_reservation(struct vm_area_struct *vma, if (vma->vm_flags & VM_SHARED) { pgoff_t idx = vma_pagecache_offset(vma, addr); region_add(&inode->i_mapping->private_list, idx, idx + 1); + + } else if (is_vma_resv_set(vma, HPAGE_RESV_OWNER)) { + pgoff_t idx = vma_pagecache_offset(vma, addr); + struct resv_map *reservations = vma_resv_map(vma); + + /* Mark this page used in the map. */ + region_add(&reservations->regions, idx, idx + 1); } } @@ -1047,11 +1129,41 @@ static int hugetlb_acct_memory(long delta) return ret; } +static void hugetlb_vm_op_open(struct vm_area_struct *vma) +{ + struct resv_map *reservations = vma_resv_map(vma); + + /* + * This new VMA should share its siblings reservation map if present. + * The VMA will only ever have a valid reservation map pointer where + * it is being copied for another still existing VMA. As that VMA + * has a reference to the reservation map it cannot dissappear until + * after this open call completes. It is therefore safe to take a + * new reference here without additional locking. + */ + if (reservations) + kref_get(&reservations->refs); +} + static void hugetlb_vm_op_close(struct vm_area_struct *vma) { - unsigned long reserve = vma_resv_huge_pages(vma); - if (reserve) - hugetlb_acct_memory(-reserve); + struct resv_map *reservations = vma_resv_map(vma); + unsigned long reserve; + unsigned long start; + unsigned long end; + + if (reservations) { + start = vma_pagecache_offset(vma, vma->vm_start); + end = vma_pagecache_offset(vma, vma->vm_end); + + reserve = (end - start) - + region_count(&reservations->regions, start, end); + + kref_put(&reservations->refs, resv_map_release); + + if (reserve) + hugetlb_acct_memory(-reserve); + } } /* @@ -1068,6 +1180,7 @@ static int hugetlb_vm_op_fault(struct vm_area_struct *vma, struct vm_fault *vmf) struct vm_operations_struct hugetlb_vm_ops = { .fault = hugetlb_vm_op_fault, + .open = hugetlb_vm_op_open, .close = hugetlb_vm_op_close, }; @@ -1617,8 +1730,13 @@ int hugetlb_reserve_pages(struct inode *inode, if (!vma || vma->vm_flags & VM_SHARED) chg = region_chg(&inode->i_mapping->private_list, from, to); else { + struct resv_map *resv_map = resv_map_alloc(); + if (!resv_map) + return -ENOMEM; + chg = to - from; - set_vma_resv_huge_pages(vma, chg); + + set_vma_resv_map(vma, resv_map); set_vma_resv_flags(vma, HPAGE_RESV_OWNER); } From a858f7b2e9bb4eb665176dde5cf32eeaaf90f153 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 23 Jul 2008 21:27:33 -0700 Subject: [PATCH 160/857] vma_page_offset() has no callees: drop it Hugh adds: vma_pagecache_offset() has a dangerously misleading name, since it's using hugepage units: rename it to vma_hugecache_offset(). [apw@shadowen.org: restack onto fixed MAP_PRIVATE reservations] [akpm@linux-foundation.org: vma_split conversion] Signed-off-by: Johannes Weiner Signed-off-by: Hugh Dickins Cc: Adam Litke Cc: Nishanth Aravamudan Cc: Andi Kleen Cc: Nick Piggin Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/hugetlb.c | 29 +++++++++-------------------- 1 file changed, 9 insertions(+), 20 deletions(-) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 65616941a383..eda9642254a0 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -199,22 +199,11 @@ static long region_count(struct list_head *head, long f, long t) return chg; } -/* - * Convert the address within this vma to the page offset within - * the mapping, in base page units. - */ -static pgoff_t vma_page_offset(struct vm_area_struct *vma, - unsigned long address) -{ - return ((address - vma->vm_start) >> PAGE_SHIFT) + - (vma->vm_pgoff >> PAGE_SHIFT); -} - /* * Convert the address within this vma to the page offset within * the mapping, in pagecache page units; huge pages here. */ -static pgoff_t vma_pagecache_offset(struct vm_area_struct *vma, +static pgoff_t vma_hugecache_offset(struct vm_area_struct *vma, unsigned long address) { return ((address - vma->vm_start) >> HPAGE_SHIFT) + @@ -806,7 +795,7 @@ static int vma_needs_reservation(struct vm_area_struct *vma, unsigned long addr) struct inode *inode = mapping->host; if (vma->vm_flags & VM_SHARED) { - pgoff_t idx = vma_pagecache_offset(vma, addr); + pgoff_t idx = vma_hugecache_offset(vma, addr); return region_chg(&inode->i_mapping->private_list, idx, idx + 1); @@ -815,7 +804,7 @@ static int vma_needs_reservation(struct vm_area_struct *vma, unsigned long addr) } else { int err; - pgoff_t idx = vma_pagecache_offset(vma, addr); + pgoff_t idx = vma_hugecache_offset(vma, addr); struct resv_map *reservations = vma_resv_map(vma); err = region_chg(&reservations->regions, idx, idx + 1); @@ -831,11 +820,11 @@ static void vma_commit_reservation(struct vm_area_struct *vma, struct inode *inode = mapping->host; if (vma->vm_flags & VM_SHARED) { - pgoff_t idx = vma_pagecache_offset(vma, addr); + pgoff_t idx = vma_hugecache_offset(vma, addr); region_add(&inode->i_mapping->private_list, idx, idx + 1); } else if (is_vma_resv_set(vma, HPAGE_RESV_OWNER)) { - pgoff_t idx = vma_pagecache_offset(vma, addr); + pgoff_t idx = vma_hugecache_offset(vma, addr); struct resv_map *reservations = vma_resv_map(vma); /* Mark this page used in the map. */ @@ -1153,8 +1142,8 @@ static void hugetlb_vm_op_close(struct vm_area_struct *vma) unsigned long end; if (reservations) { - start = vma_pagecache_offset(vma, vma->vm_start); - end = vma_pagecache_offset(vma, vma->vm_end); + start = vma_hugecache_offset(vma, vma->vm_start); + end = vma_hugecache_offset(vma, vma->vm_end); reserve = (end - start) - region_count(&reservations->regions, start, end); @@ -1471,7 +1460,7 @@ static struct page *hugetlbfs_pagecache_page(struct vm_area_struct *vma, pgoff_t idx; mapping = vma->vm_file->f_mapping; - idx = vma_pagecache_offset(vma, address); + idx = vma_hugecache_offset(vma, address); return find_lock_page(mapping, idx); } @@ -1499,7 +1488,7 @@ static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma, } mapping = vma->vm_file->f_mapping; - idx = vma_pagecache_offset(vma, address); + idx = vma_hugecache_offset(vma, address); /* * Use page lock to guard against racing truncation From 11fa977ecde652ab324dd79c179deb52e82a8df1 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 23 Jul 2008 21:27:34 -0700 Subject: [PATCH 161/857] generic_file_aio_read() cleanups As akpm points out, there's really no need for generic_file_aio_read to make a special case of count 0: just loop through nr_segs doing nothing. And as Harvey Harrison points out, there's no need to reset retval to 0 where it's already 0. Setting count (or ocount) to 0 before calling generic_segment_checks is unnecessary too; but reluctantly I'll leave that removal to someone with a wider range of gcc versions to hand - 4.1.2 and 4.2.1 don't warn about it, but perhaps others do - I forget which are the warniest versions. Signed-off-by: Hugh Dickins Tested-by: Lawrence Greenfield Cc: Christoph Rohland Cc: Badari Pulavarty Cc: Zach Brown Cc: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/filemap.c | 42 +++++++++++++++++++----------------------- 1 file changed, 19 insertions(+), 23 deletions(-) diff --git a/mm/filemap.c b/mm/filemap.c index 6343f3c841b7..7675b91f4f63 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -1197,7 +1197,6 @@ generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov, mapping = filp->f_mapping; inode = mapping->host; - retval = 0; if (!count) goto out; /* skip atime */ size = i_size_read(inode); @@ -1209,33 +1208,30 @@ generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov, } if (retval > 0) *ppos = pos + retval; - } - if (likely(retval != 0)) { - file_accessed(filp); - goto out; + if (retval) { + file_accessed(filp); + goto out; + } } } - retval = 0; - if (count) { - for (seg = 0; seg < nr_segs; seg++) { - read_descriptor_t desc; + for (seg = 0; seg < nr_segs; seg++) { + read_descriptor_t desc; - desc.written = 0; - desc.arg.buf = iov[seg].iov_base; - desc.count = iov[seg].iov_len; - if (desc.count == 0) - continue; - desc.error = 0; - do_generic_file_read(filp,ppos,&desc,file_read_actor); - retval += desc.written; - if (desc.error) { - retval = retval ?: desc.error; - break; - } - if (desc.count > 0) - break; + desc.written = 0; + desc.arg.buf = iov[seg].iov_base; + desc.count = iov[seg].iov_len; + if (desc.count == 0) + continue; + desc.error = 0; + do_generic_file_read(filp, ppos, &desc, file_read_actor); + retval += desc.written; + if (desc.error) { + retval = retval ?: desc.error; + break; } + if (desc.count > 0) + break; } out: return retval; From bcd78e49613c41b5bed96fa288e983876f286a59 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 23 Jul 2008 21:27:35 -0700 Subject: [PATCH 162/857] tmpfs: support aio We have a request for tmpfs to support the AIO interface: easily done, no more than replacing the old shmem_file_read by shmem_file_aio_read, cribbed from generic_file_aio_read. (In 2.6.25 its write side was already changed to use generic_file_aio_write.) Incorporate cleanups from Andrew Morton and Harvey Harrison. Tests out fine with LTP's ltp-aiodio.sh, given hacks (not included) to support O_DIRECT. tmpfs cannot honestly support O_DIRECT: its cache-avoiding-IO nature is at odds with direct IO-avoiding-cache. Signed-off-by: Hugh Dickins Tested-by: Lawrence Greenfield Cc: Christoph Rohland Cc: Badari Pulavarty Cc: Zach Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/shmem.c | 47 ++++++++++++++++++++++++++++++----------------- 1 file changed, 30 insertions(+), 17 deletions(-) diff --git a/mm/shmem.c b/mm/shmem.c index e2a6ae1a44e9..9ffbea9b79e1 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -1690,26 +1690,38 @@ static void do_shmem_file_read(struct file *filp, loff_t *ppos, read_descriptor_ file_accessed(filp); } -static ssize_t shmem_file_read(struct file *filp, char __user *buf, size_t count, loff_t *ppos) +static ssize_t shmem_file_aio_read(struct kiocb *iocb, + const struct iovec *iov, unsigned long nr_segs, loff_t pos) { - read_descriptor_t desc; + struct file *filp = iocb->ki_filp; + ssize_t retval; + unsigned long seg; + size_t count; + loff_t *ppos = &iocb->ki_pos; - if ((ssize_t) count < 0) - return -EINVAL; - if (!access_ok(VERIFY_WRITE, buf, count)) - return -EFAULT; - if (!count) - return 0; + retval = generic_segment_checks(iov, &nr_segs, &count, VERIFY_WRITE); + if (retval) + return retval; - desc.written = 0; - desc.count = count; - desc.arg.buf = buf; - desc.error = 0; + for (seg = 0; seg < nr_segs; seg++) { + read_descriptor_t desc; - do_shmem_file_read(filp, ppos, &desc, file_read_actor); - if (desc.written) - return desc.written; - return desc.error; + desc.written = 0; + desc.arg.buf = iov[seg].iov_base; + desc.count = iov[seg].iov_len; + if (desc.count == 0) + continue; + desc.error = 0; + do_shmem_file_read(filp, ppos, &desc, file_read_actor); + retval += desc.written; + if (desc.error) { + retval = retval ?: desc.error; + break; + } + if (desc.count > 0) + break; + } + return retval; } static int shmem_statfs(struct dentry *dentry, struct kstatfs *buf) @@ -2369,8 +2381,9 @@ static const struct file_operations shmem_file_operations = { .mmap = shmem_mmap, #ifdef CONFIG_TMPFS .llseek = generic_file_llseek, - .read = shmem_file_read, + .read = do_sync_read, .write = do_sync_write, + .aio_read = shmem_file_aio_read, .aio_write = generic_file_aio_write, .fsync = simple_sync_file, .splice_read = generic_file_splice_read, From cce770815869e9209171d819dfde89bcc738ab62 Mon Sep 17 00:00:00 2001 From: Pavel Machek Date: Wed, 23 Jul 2008 21:27:36 -0700 Subject: [PATCH 163/857] SYNC_FILE_RANGE_WRITE may and will block. Document that. [akpm@linux-foundation.org: fix comment text] Signed-off-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/sync.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/sync.c b/fs/sync.c index 228e17b5e9ee..2967562d416f 100644 --- a/fs/sync.c +++ b/fs/sync.c @@ -139,7 +139,8 @@ asmlinkage long sys_fdatasync(unsigned int fd) * before performing the write. * * SYNC_FILE_RANGE_WRITE: initiate writeout of all those dirty pages in the - * range which are not presently under writeback. + * range which are not presently under writeback. Note that this may block for + * significant periods due to exhaustion of disk request structures. * * SYNC_FILE_RANGE_WAIT_AFTER: wait upon writeout of all pages in the range * after performing the write. From a47a126ad5ea072aca3e611ed8f8dc6adad24bab Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 23 Jul 2008 21:27:38 -0700 Subject: [PATCH 164/857] vmallocinfo: add NUMA information Christoph recently added /proc/vmallocinfo file to get information about vmalloc allocations. This patch adds NUMA specific information, giving number of pages allocated on each memory node. This should help to check that vmalloc() is able to respect NUMA policies. Example of output on a four nodes machine (one cpu per node) 1) network hash tables are evenly spreaded on four nodes (OK) (Same point for inodes and dentries hash tables) 2) iptables tables (x_tables) are correctly allocated on each cpu node (OK). 3) sys_swapon() allocates its memory from one node only. 4) each loaded module is using memory on one node. Sysadmins could tune their setup to change points 3) and 4) if necessary. grep "pages=" /proc/vmallocinfo 0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204/0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128 0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204/0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64 0xffffc2000031a000-0xffffc2000031d000 12288 alloc_large_system_hash+0x204/0x2c0 pages=2 vmalloc N1=1 N2=1 0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e/0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3 0xffffc2000033e000-0xffffc20000341000 12288 sys_swapon+0x640/0xac0 pages=2 vmalloc N0=2 0xffffc20000341000-0xffffc20000344000 12288 xt_alloc_table_info+0xfe/0x130 [x_tables] pages=2 vmalloc N0=2 0xffffc20000344000-0xffffc20000347000 12288 xt_alloc_table_info+0xfe/0x130 [x_tables] pages=2 vmalloc N1=2 0xffffc20000347000-0xffffc2000034a000 12288 xt_alloc_table_info+0xfe/0x130 [x_tables] pages=2 vmalloc N2=2 0xffffc2000034a000-0xffffc2000034d000 12288 xt_alloc_table_info+0xfe/0x130 [x_tables] pages=2 vmalloc N3=2 0xffffc20004381000-0xffffc20004402000 528384 alloc_large_system_hash+0x204/0x2c0 pages=128 vmalloc N0=32 N1=32 N2=32 N3=32 0xffffc20004402000-0xffffc20004803000 4198400 alloc_large_system_hash+0x204/0x2c0 pages=1024 vmalloc vpages N0=256 N1=256 N2=256 N3=256 0xffffc20004803000-0xffffc20004904000 1052672 alloc_large_system_hash+0x204/0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64 0xffffc20004904000-0xffffc20004bec000 3047424 sys_swapon+0x640/0xac0 pages=743 vmalloc vpages N0=743 0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 pages=14 vmalloc N1=14 0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 pages=4 vmalloc N0=4 0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 pages=2 vmalloc N0=2 0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 pages=10 vmalloc N1=10 0xffffffffa0022000-0xffffffffa0028000 24576 sys_init_module+0xc27/0x1d00 pages=5 vmalloc N3=5 0xffffffffa0028000-0xffffffffa0050000 163840 sys_init_module+0xc27/0x1d00 pages=39 vmalloc N1=39 0xffffffffa0050000-0xffffffffa0052000 8192 sys_init_module+0xc27/0x1d00 pages=1 vmalloc N1=1 0xffffffffa0052000-0xffffffffa0056000 16384 sys_init_module+0xc27/0x1d00 pages=3 vmalloc N1=3 0xffffffffa0056000-0xffffffffa0081000 176128 sys_init_module+0xc27/0x1d00 pages=42 vmalloc N3=42 0xffffffffa0081000-0xffffffffa00ae000 184320 sys_init_module+0xc27/0x1d00 pages=44 vmalloc N3=44 0xffffffffa00ae000-0xffffffffa00b1000 12288 sys_init_module+0xc27/0x1d00 pages=2 vmalloc N3=2 0xffffffffa00b1000-0xffffffffa00b9000 32768 sys_init_module+0xc27/0x1d00 pages=7 vmalloc N0=7 0xffffffffa00b9000-0xffffffffa00c4000 45056 sys_init_module+0xc27/0x1d00 pages=10 vmalloc N3=10 0xffffffffa00c6000-0xffffffffa00e0000 106496 sys_init_module+0xc27/0x1d00 pages=25 vmalloc N2=25 0xffffffffa00e0000-0xffffffffa00f1000 69632 sys_init_module+0xc27/0x1d00 pages=16 vmalloc N2=16 0xffffffffa00f1000-0xffffffffa00f4000 12288 sys_init_module+0xc27/0x1d00 pages=2 vmalloc N3=2 0xffffffffa00f4000-0xffffffffa00f7000 12288 sys_init_module+0xc27/0x1d00 pages=2 vmalloc N3=2 [akpm@linux-foundation.org: fix comment] Signed-off-by: Eric Dumazet Cc: Christoph Lameter Cc: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/filesystems/proc.txt | 44 ++++++++++++++++++++++++++++++ fs/proc/proc_misc.c | 15 ++++++++-- mm/vmalloc.c | 20 ++++++++++++++ 3 files changed, 77 insertions(+), 2 deletions(-) diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt index 7f268f327d75..8c6384bdfed4 100644 --- a/Documentation/filesystems/proc.txt +++ b/Documentation/filesystems/proc.txt @@ -296,6 +296,7 @@ Table 1-4: Kernel info in /proc uptime System uptime version Kernel version video bttv info of video resources (2.4) + vmallocinfo Show vmalloced areas .............................................................................. You can, for example, check which interrupts are currently in use and what @@ -557,6 +558,49 @@ VmallocTotal: total size of vmalloc memory area VmallocUsed: amount of vmalloc area which is used VmallocChunk: largest contigious block of vmalloc area which is free +.............................................................................. + +vmallocinfo: + +Provides information about vmalloced/vmaped areas. One line per area, +containing the virtual address range of the area, size in bytes, +caller information of the creator, and optional information depending +on the kind of area : + + pages=nr number of pages + phys=addr if a physical address was specified + ioremap I/O mapping (ioremap() and friends) + vmalloc vmalloc() area + vmap vmap()ed pages + user VM_USERMAP area + vpages buffer for pages pointers was vmalloced (huge area) + N=nr (Only on NUMA kernels) + Number of pages allocated on memory node + +> cat /proc/vmallocinfo +0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ... + /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128 +0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ... + /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64 +0xffffc20000302000-0xffffc20000304000 8192 acpi_tb_verify_table+0x21/0x4f... + phys=7fee8000 ioremap +0xffffc20000304000-0xffffc20000307000 12288 acpi_tb_verify_table+0x21/0x4f... + phys=7fee7000 ioremap +0xffffc2000031d000-0xffffc2000031f000 8192 init_vdso_vars+0x112/0x210 +0xffffc2000031f000-0xffffc2000032b000 49152 cramfs_uncompress_init+0x2e ... + /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3 +0xffffc2000033a000-0xffffc2000033d000 12288 sys_swapon+0x640/0xac0 ... + pages=2 vmalloc N1=2 +0xffffc20000347000-0xffffc2000034c000 20480 xt_alloc_table_info+0xfe ... + /0x130 [x_tables] pages=4 vmalloc N0=4 +0xffffffffa0000000-0xffffffffa000f000 61440 sys_init_module+0xc27/0x1d00 ... + pages=14 vmalloc N2=14 +0xffffffffa000f000-0xffffffffa0014000 20480 sys_init_module+0xc27/0x1d00 ... + pages=4 vmalloc N1=4 +0xffffffffa0014000-0xffffffffa0017000 12288 sys_init_module+0xc27/0x1d00 ... + pages=2 vmalloc N1=2 +0xffffffffa0017000-0xffffffffa0022000 45056 sys_init_module+0xc27/0x1d00 ... + pages=10 vmalloc N0=10 1.3 IDE devices in /proc/ide ---------------------------- diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c index b14f43d25e9e..ded969862960 100644 --- a/fs/proc/proc_misc.c +++ b/fs/proc/proc_misc.c @@ -464,14 +464,25 @@ static const struct file_operations proc_slabstats_operations = { #ifdef CONFIG_MMU static int vmalloc_open(struct inode *inode, struct file *file) { - return seq_open(file, &vmalloc_op); + unsigned int *ptr = NULL; + int ret; + + if (NUMA_BUILD) + ptr = kmalloc(nr_node_ids * sizeof(unsigned int), GFP_KERNEL); + ret = seq_open(file, &vmalloc_op); + if (!ret) { + struct seq_file *m = file->private_data; + m->private = ptr; + } else + kfree(ptr); + return ret; } static const struct file_operations proc_vmalloc_operations = { .open = vmalloc_open, .read = seq_read, .llseek = seq_lseek, - .release = seq_release, + .release = seq_release_private, }; #endif diff --git a/mm/vmalloc.c b/mm/vmalloc.c index 6e45b0f3d125..35f293816294 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -931,6 +931,25 @@ static void s_stop(struct seq_file *m, void *p) read_unlock(&vmlist_lock); } +static void show_numa_info(struct seq_file *m, struct vm_struct *v) +{ + if (NUMA_BUILD) { + unsigned int nr, *counters = m->private; + + if (!counters) + return; + + memset(counters, 0, nr_node_ids * sizeof(unsigned int)); + + for (nr = 0; nr < v->nr_pages; nr++) + counters[page_to_nid(v->pages[nr])]++; + + for_each_node_state(nr, N_HIGH_MEMORY) + if (counters[nr]) + seq_printf(m, " N%u=%u", nr, counters[nr]); + } +} + static int s_show(struct seq_file *m, void *p) { struct vm_struct *v = p; @@ -967,6 +986,7 @@ static int s_show(struct seq_file *m, void *p) if (v->flags & VM_VPAGES) seq_printf(m, " vpages"); + show_numa_info(m, v); seq_putc(m, '\n'); return 0; } From 5e9426abe209cf134adbbd62c5e73ef534eb73e9 Mon Sep 17 00:00:00 2001 From: Nishanth Aravamudan Date: Wed, 23 Jul 2008 21:27:39 -0700 Subject: [PATCH 165/857] mm: remove mm_init compilation dependency on CONFIG_DEBUG_MEMORY_INIT Towards the end of putting all core mm initialization in mm_init.c, I plan on putting the creation of a mm kobject in a function in that file. However, the file is currently only compiled if CONFIG_DEBUG_MEMORY_INIT is set. Remove this dependency, but put the code under an #ifdef on the same config option. This should result in no functional changes. Signed-off-by: Nishanth Aravamudan Cc: Nick Piggin Cc: Mel Gorman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/Makefile | 3 +-- mm/mm_init.c | 2 ++ 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/mm/Makefile b/mm/Makefile index 4bbc8f094ff0..06ca2381fef1 100644 --- a/mm/Makefile +++ b/mm/Makefile @@ -11,7 +11,7 @@ obj-y := bootmem.o filemap.o mempool.o oom_kill.o fadvise.o \ maccess.o page_alloc.o page-writeback.o pdflush.o \ readahead.o swap.o truncate.o vmscan.o \ prio_tree.o util.o mmzone.o vmstat.o backing-dev.o \ - page_isolation.o $(mmu-y) + page_isolation.o mm_init.o $(mmu-y) obj-$(CONFIG_PROC_PAGE_MONITOR) += pagewalk.o obj-$(CONFIG_BOUNCE) += bounce.o @@ -26,7 +26,6 @@ obj-$(CONFIG_TMPFS_POSIX_ACL) += shmem_acl.o obj-$(CONFIG_TINY_SHMEM) += tiny-shmem.o obj-$(CONFIG_SLOB) += slob.o obj-$(CONFIG_SLAB) += slab.o -obj-$(CONFIG_DEBUG_MEMORY_INIT) += mm_init.o obj-$(CONFIG_SLUB) += slub.o obj-$(CONFIG_MEMORY_HOTPLUG) += memory_hotplug.o obj-$(CONFIG_FS_XIP) += filemap_xip.o diff --git a/mm/mm_init.c b/mm/mm_init.c index ce445ca097e7..eaf0d3b47099 100644 --- a/mm/mm_init.c +++ b/mm/mm_init.c @@ -9,6 +9,7 @@ #include #include "internal.h" +#ifdef CONFIG_DEBUG_MEMORY_INIT int __meminitdata mminit_loglevel; /* The zonelists are simply reported, validation is manual. */ @@ -132,3 +133,4 @@ static __init int set_mminit_loglevel(char *str) return 0; } early_param("mminit_loglevel", set_mminit_loglevel); +#endif /* CONFIG_DEBUG_MEMORY_INIT */ From ff7ea79cf7c3a481851bd4b2185fdeb6ce4afa29 Mon Sep 17 00:00:00 2001 From: Nishanth Aravamudan Date: Wed, 23 Jul 2008 21:27:39 -0700 Subject: [PATCH 166/857] mm: create /sys/kernel/mm Add a kobject to create /sys/kernel/mm when sysfs is mounted. The kobject will exist regardless. This will allow for the hugepage related sysfs directories to exist under the mm "subsystem" directory. Add an ABI file appropriately. [kosaki.motohiro@jp.fujitsu.com: fix build] Signed-off-by: Nishanth Aravamudan Cc: Nick Piggin Cc: Mel Gorman Signed-off-by: KOSAKI Motohiro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/ABI/testing/sysfs-kernel-mm | 6 ++++++ include/linux/kobject.h | 2 ++ mm/mm_init.c | 16 ++++++++++++++++ 3 files changed, 24 insertions(+) create mode 100644 Documentation/ABI/testing/sysfs-kernel-mm diff --git a/Documentation/ABI/testing/sysfs-kernel-mm b/Documentation/ABI/testing/sysfs-kernel-mm new file mode 100644 index 000000000000..190d523ac159 --- /dev/null +++ b/Documentation/ABI/testing/sysfs-kernel-mm @@ -0,0 +1,6 @@ +What: /sys/kernel/mm +Date: July 2008 +Contact: Nishanth Aravamudan , VM maintainers +Description: + /sys/kernel/mm/ should contain any and all VM + related information in /sys/kernel/. diff --git a/include/linux/kobject.h b/include/linux/kobject.h index 60f0d418ae32..5437ac0276e2 100644 --- a/include/linux/kobject.h +++ b/include/linux/kobject.h @@ -186,6 +186,8 @@ extern struct kobject *kset_find_obj(struct kset *, const char *); /* The global /sys/kernel/ kobject for people to chain off of */ extern struct kobject *kernel_kobj; +/* The global /sys/kernel/mm/ kobject for people to chain off of */ +extern struct kobject *mm_kobj; /* The global /sys/hypervisor/ kobject for people to chain off of */ extern struct kobject *hypervisor_kobj; /* The global /sys/power/ kobject for people to chain off of */ diff --git a/mm/mm_init.c b/mm/mm_init.c index eaf0d3b47099..c6af41ea9994 100644 --- a/mm/mm_init.c +++ b/mm/mm_init.c @@ -7,6 +7,8 @@ */ #include #include +#include +#include #include "internal.h" #ifdef CONFIG_DEBUG_MEMORY_INIT @@ -134,3 +136,17 @@ static __init int set_mminit_loglevel(char *str) } early_param("mminit_loglevel", set_mminit_loglevel); #endif /* CONFIG_DEBUG_MEMORY_INIT */ + +struct kobject *mm_kobj; +EXPORT_SYMBOL_GPL(mm_kobj); + +static int __init mm_sysfs_init(void) +{ + mm_kobj = kobject_create_and_add("mm", kernel_kobj); + if (!mm_kobj) + return -ENOMEM; + + return 0; +} + +__initcall(mm_sysfs_init); From b7ba30c679ed1eb7ed3ed8f281f6493282042bd4 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 23 Jul 2008 21:27:40 -0700 Subject: [PATCH 167/857] hugetlb: factor out prep_new_huge_page Needed to avoid code duplication in follow up patches. Acked-by: Adam Litke Acked-by: Nishanth Aravamudan Signed-off-by: Andi Kleen Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/hugetlb.c | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index eda9642254a0..32dff4290c66 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -513,6 +513,16 @@ static int adjust_pool_surplus(int delta) return ret; } +static void prep_new_huge_page(struct page *page, int nid) +{ + set_compound_page_dtor(page, free_huge_page); + spin_lock(&hugetlb_lock); + nr_huge_pages++; + nr_huge_pages_node[nid]++; + spin_unlock(&hugetlb_lock); + put_page(page); /* free it into the hugepage allocator */ +} + static struct page *alloc_fresh_huge_page_node(int nid) { struct page *page; @@ -526,12 +536,7 @@ static struct page *alloc_fresh_huge_page_node(int nid) __free_pages(page, HUGETLB_PAGE_ORDER); return NULL; } - set_compound_page_dtor(page, free_huge_page); - spin_lock(&hugetlb_lock); - nr_huge_pages++; - nr_huge_pages_node[nid]++; - spin_unlock(&hugetlb_lock); - put_page(page); /* free it into the hugepage allocator */ + prep_new_huge_page(page, nid); } return page; From a5516438959d90b071ff0a484ce4f3f523dc3152 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 23 Jul 2008 21:27:41 -0700 Subject: [PATCH 168/857] hugetlb: modular state for hugetlb page size The goal of this patchset is to support multiple hugetlb page sizes. This is achieved by introducing a new struct hstate structure, which encapsulates the important hugetlb state and constants (eg. huge page size, number of huge pages currently allocated, etc). The hstate structure is then passed around the code which requires these fields, they will do the right thing regardless of the exact hstate they are operating on. This patch adds the hstate structure, with a single global instance of it (default_hstate), and does the basic work of converting hugetlb to use the hstate. Future patches will add more hstate structures to allow for different hugetlbfs mounts to have different page sizes. [akpm@linux-foundation.org: coding-style fixes] Acked-by: Adam Litke Acked-by: Nishanth Aravamudan Signed-off-by: Andi Kleen Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/mm/hugetlbpage.c | 7 +- arch/powerpc/mm/hugetlbpage.c | 3 +- arch/s390/mm/hugetlbpage.c | 3 +- arch/sh/mm/hugetlbpage.c | 3 +- arch/sparc64/mm/hugetlbpage.c | 5 +- arch/x86/mm/hugetlbpage.c | 5 +- fs/hugetlbfs/inode.c | 52 +++-- include/asm-ia64/hugetlb.h | 3 +- include/asm-powerpc/hugetlb.h | 3 +- include/asm-s390/hugetlb.h | 3 +- include/asm-sh/hugetlb.h | 3 +- include/asm-sparc/hugetlb.h | 3 +- include/asm-x86/hugetlb.h | 8 +- include/linux/hugetlb.h | 88 +++++++- ipc/shm.c | 3 +- mm/hugetlb.c | 370 +++++++++++++++++++--------------- mm/memory.c | 2 +- mm/mempolicy.c | 9 +- mm/mmap.c | 3 +- 19 files changed, 357 insertions(+), 219 deletions(-) diff --git a/arch/ia64/mm/hugetlbpage.c b/arch/ia64/mm/hugetlbpage.c index cd49e2860eef..6170f097d255 100644 --- a/arch/ia64/mm/hugetlbpage.c +++ b/arch/ia64/mm/hugetlbpage.c @@ -24,7 +24,7 @@ unsigned int hpage_shift=HPAGE_SHIFT_DEFAULT; pte_t * -huge_pte_alloc (struct mm_struct *mm, unsigned long addr) +huge_pte_alloc(struct mm_struct *mm, unsigned long addr, unsigned long sz) { unsigned long taddr = htlbpage_to_page(addr); pgd_t *pgd; @@ -75,7 +75,8 @@ int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) * Don't actually need to do any preparation, but need to make sure * the address is in the right region. */ -int prepare_hugepage_range(unsigned long addr, unsigned long len) +int prepare_hugepage_range(struct file *file, + unsigned long addr, unsigned long len) { if (len & ~HPAGE_MASK) return -EINVAL; @@ -149,7 +150,7 @@ unsigned long hugetlb_get_unmapped_area(struct file *file, unsigned long addr, u /* Handle MAP_FIXED */ if (flags & MAP_FIXED) { - if (prepare_hugepage_range(addr, len)) + if (prepare_hugepage_range(file, addr, len)) return -EINVAL; return addr; } diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c index 1a96cc891cf5..c94dc71af989 100644 --- a/arch/powerpc/mm/hugetlbpage.c +++ b/arch/powerpc/mm/hugetlbpage.c @@ -128,7 +128,8 @@ pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) return NULL; } -pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) +pte_t *huge_pte_alloc(struct mm_struct *mm, + unsigned long addr, unsigned long sz) { pgd_t *pg; pud_t *pu; diff --git a/arch/s390/mm/hugetlbpage.c b/arch/s390/mm/hugetlbpage.c index f4b6124fdb75..9162dc84f77f 100644 --- a/arch/s390/mm/hugetlbpage.c +++ b/arch/s390/mm/hugetlbpage.c @@ -72,7 +72,8 @@ void arch_release_hugepage(struct page *page) page[1].index = 0; } -pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) +pte_t *huge_pte_alloc(struct mm_struct *mm, + unsigned long addr, unsigned long sz) { pgd_t *pgdp; pud_t *pudp; diff --git a/arch/sh/mm/hugetlbpage.c b/arch/sh/mm/hugetlbpage.c index ae8c321d6e2a..2f9dbe0ef4ac 100644 --- a/arch/sh/mm/hugetlbpage.c +++ b/arch/sh/mm/hugetlbpage.c @@ -22,7 +22,8 @@ #include #include -pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) +pte_t *huge_pte_alloc(struct mm_struct *mm, + unsigned long addr, unsigned long sz) { pgd_t *pgd; pud_t *pud; diff --git a/arch/sparc64/mm/hugetlbpage.c b/arch/sparc64/mm/hugetlbpage.c index ebefd2a14375..1307b23f6a76 100644 --- a/arch/sparc64/mm/hugetlbpage.c +++ b/arch/sparc64/mm/hugetlbpage.c @@ -175,7 +175,7 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr, return -ENOMEM; if (flags & MAP_FIXED) { - if (prepare_hugepage_range(addr, len)) + if (prepare_hugepage_range(file, addr, len)) return -EINVAL; return addr; } @@ -195,7 +195,8 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr, pgoff, flags); } -pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) +pte_t *huge_pte_alloc(struct mm_struct *mm, + unsigned long addr, unsigned long sz) { pgd_t *pgd; pud_t *pud; diff --git a/arch/x86/mm/hugetlbpage.c b/arch/x86/mm/hugetlbpage.c index 0b3d567e686d..52476fde8996 100644 --- a/arch/x86/mm/hugetlbpage.c +++ b/arch/x86/mm/hugetlbpage.c @@ -124,7 +124,8 @@ int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) return 1; } -pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr) +pte_t *huge_pte_alloc(struct mm_struct *mm, + unsigned long addr, unsigned long sz) { pgd_t *pgd; pud_t *pud; @@ -368,7 +369,7 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr, return -ENOMEM; if (flags & MAP_FIXED) { - if (prepare_hugepage_range(addr, len)) + if (prepare_hugepage_range(file, addr, len)) return -EINVAL; return addr; } diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 428eff5b73f3..516c581b5371 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -80,6 +80,7 @@ static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma) struct inode *inode = file->f_path.dentry->d_inode; loff_t len, vma_len; int ret; + struct hstate *h = hstate_file(file); /* * vma address alignment (but not the pgoff alignment) has @@ -92,7 +93,7 @@ static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma) vma->vm_flags |= VM_HUGETLB | VM_RESERVED; vma->vm_ops = &hugetlb_vm_ops; - if (vma->vm_pgoff & ~(HPAGE_MASK >> PAGE_SHIFT)) + if (vma->vm_pgoff & ~(huge_page_mask(h) >> PAGE_SHIFT)) return -EINVAL; vma_len = (loff_t)(vma->vm_end - vma->vm_start); @@ -104,8 +105,8 @@ static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma) len = vma_len + ((loff_t)vma->vm_pgoff << PAGE_SHIFT); if (hugetlb_reserve_pages(inode, - vma->vm_pgoff >> (HPAGE_SHIFT-PAGE_SHIFT), - len >> HPAGE_SHIFT, vma)) + vma->vm_pgoff >> huge_page_order(h), + len >> huge_page_shift(h), vma)) goto out; ret = 0; @@ -130,20 +131,21 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr, struct mm_struct *mm = current->mm; struct vm_area_struct *vma; unsigned long start_addr; + struct hstate *h = hstate_file(file); - if (len & ~HPAGE_MASK) + if (len & ~huge_page_mask(h)) return -EINVAL; if (len > TASK_SIZE) return -ENOMEM; if (flags & MAP_FIXED) { - if (prepare_hugepage_range(addr, len)) + if (prepare_hugepage_range(file, addr, len)) return -EINVAL; return addr; } if (addr) { - addr = ALIGN(addr, HPAGE_SIZE); + addr = ALIGN(addr, huge_page_size(h)); vma = find_vma(mm, addr); if (TASK_SIZE - len >= addr && (!vma || addr + len <= vma->vm_start)) @@ -156,7 +158,7 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr, start_addr = TASK_UNMAPPED_BASE; full_search: - addr = ALIGN(start_addr, HPAGE_SIZE); + addr = ALIGN(start_addr, huge_page_size(h)); for (vma = find_vma(mm, addr); ; vma = vma->vm_next) { /* At this point: (!vma || addr < vma->vm_end). */ @@ -174,7 +176,7 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr, if (!vma || addr + len <= vma->vm_start) return addr; - addr = ALIGN(vma->vm_end, HPAGE_SIZE); + addr = ALIGN(vma->vm_end, huge_page_size(h)); } } #endif @@ -225,10 +227,11 @@ hugetlbfs_read_actor(struct page *page, unsigned long offset, static ssize_t hugetlbfs_read(struct file *filp, char __user *buf, size_t len, loff_t *ppos) { + struct hstate *h = hstate_file(filp); struct address_space *mapping = filp->f_mapping; struct inode *inode = mapping->host; - unsigned long index = *ppos >> HPAGE_SHIFT; - unsigned long offset = *ppos & ~HPAGE_MASK; + unsigned long index = *ppos >> huge_page_shift(h); + unsigned long offset = *ppos & ~huge_page_mask(h); unsigned long end_index; loff_t isize; ssize_t retval = 0; @@ -243,17 +246,17 @@ static ssize_t hugetlbfs_read(struct file *filp, char __user *buf, if (!isize) goto out; - end_index = (isize - 1) >> HPAGE_SHIFT; + end_index = (isize - 1) >> huge_page_shift(h); for (;;) { struct page *page; - int nr, ret; + unsigned long nr, ret; /* nr is the maximum number of bytes to copy from this page */ - nr = HPAGE_SIZE; + nr = huge_page_size(h); if (index >= end_index) { if (index > end_index) goto out; - nr = ((isize - 1) & ~HPAGE_MASK) + 1; + nr = ((isize - 1) & ~huge_page_mask(h)) + 1; if (nr <= offset) { goto out; } @@ -287,8 +290,8 @@ static ssize_t hugetlbfs_read(struct file *filp, char __user *buf, offset += ret; retval += ret; len -= ret; - index += offset >> HPAGE_SHIFT; - offset &= ~HPAGE_MASK; + index += offset >> huge_page_shift(h); + offset &= ~huge_page_mask(h); if (page) page_cache_release(page); @@ -298,7 +301,7 @@ static ssize_t hugetlbfs_read(struct file *filp, char __user *buf, break; } out: - *ppos = ((loff_t)index << HPAGE_SHIFT) + offset; + *ppos = ((loff_t)index << huge_page_shift(h)) + offset; mutex_unlock(&inode->i_mutex); return retval; } @@ -339,8 +342,9 @@ static void truncate_huge_page(struct page *page) static void truncate_hugepages(struct inode *inode, loff_t lstart) { + struct hstate *h = hstate_inode(inode); struct address_space *mapping = &inode->i_data; - const pgoff_t start = lstart >> HPAGE_SHIFT; + const pgoff_t start = lstart >> huge_page_shift(h); struct pagevec pvec; pgoff_t next; int i, freed = 0; @@ -449,8 +453,9 @@ static int hugetlb_vmtruncate(struct inode *inode, loff_t offset) { pgoff_t pgoff; struct address_space *mapping = inode->i_mapping; + struct hstate *h = hstate_inode(inode); - BUG_ON(offset & ~HPAGE_MASK); + BUG_ON(offset & ~huge_page_mask(h)); pgoff = offset >> PAGE_SHIFT; i_size_write(inode, offset); @@ -465,6 +470,7 @@ static int hugetlb_vmtruncate(struct inode *inode, loff_t offset) static int hugetlbfs_setattr(struct dentry *dentry, struct iattr *attr) { struct inode *inode = dentry->d_inode; + struct hstate *h = hstate_inode(inode); int error; unsigned int ia_valid = attr->ia_valid; @@ -476,7 +482,7 @@ static int hugetlbfs_setattr(struct dentry *dentry, struct iattr *attr) if (ia_valid & ATTR_SIZE) { error = -EINVAL; - if (!(attr->ia_size & ~HPAGE_MASK)) + if (!(attr->ia_size & ~huge_page_mask(h))) error = hugetlb_vmtruncate(inode, attr->ia_size); if (error) goto out; @@ -610,9 +616,10 @@ static int hugetlbfs_set_page_dirty(struct page *page) static int hugetlbfs_statfs(struct dentry *dentry, struct kstatfs *buf) { struct hugetlbfs_sb_info *sbinfo = HUGETLBFS_SB(dentry->d_sb); + struct hstate *h = hstate_inode(dentry->d_inode); buf->f_type = HUGETLBFS_MAGIC; - buf->f_bsize = HPAGE_SIZE; + buf->f_bsize = huge_page_size(h); if (sbinfo) { spin_lock(&sbinfo->stat_lock); /* If no limits set, just report 0 for max/free/used @@ -942,7 +949,8 @@ struct file *hugetlb_file_setup(const char *name, size_t size) goto out_dentry; error = -ENOMEM; - if (hugetlb_reserve_pages(inode, 0, size >> HPAGE_SHIFT, NULL)) + if (hugetlb_reserve_pages(inode, 0, + size >> huge_page_shift(hstate_inode(inode)), NULL)) goto out_inode; d_instantiate(dentry, inode); diff --git a/include/asm-ia64/hugetlb.h b/include/asm-ia64/hugetlb.h index e9d1e5e2382d..da55c63728e0 100644 --- a/include/asm-ia64/hugetlb.h +++ b/include/asm-ia64/hugetlb.h @@ -8,7 +8,8 @@ void hugetlb_free_pgd_range(struct mmu_gather *tlb, unsigned long addr, unsigned long end, unsigned long floor, unsigned long ceiling); -int prepare_hugepage_range(unsigned long addr, unsigned long len); +int prepare_hugepage_range(struct file *file, + unsigned long addr, unsigned long len); static inline int is_hugepage_only_range(struct mm_struct *mm, unsigned long addr, diff --git a/include/asm-powerpc/hugetlb.h b/include/asm-powerpc/hugetlb.h index 0a37aa5ecaa5..ca37c4af27b1 100644 --- a/include/asm-powerpc/hugetlb.h +++ b/include/asm-powerpc/hugetlb.h @@ -21,7 +21,8 @@ pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, * If the arch doesn't supply something else, assume that hugepage * size aligned regions are ok without further preparation. */ -static inline int prepare_hugepage_range(unsigned long addr, unsigned long len) +static inline int prepare_hugepage_range(struct file *file, + unsigned long addr, unsigned long len) { if (len & ~HPAGE_MASK) return -EINVAL; diff --git a/include/asm-s390/hugetlb.h b/include/asm-s390/hugetlb.h index 600a776f8f75..670a1d1745d2 100644 --- a/include/asm-s390/hugetlb.h +++ b/include/asm-s390/hugetlb.h @@ -22,7 +22,8 @@ void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, * If the arch doesn't supply something else, assume that hugepage * size aligned regions are ok without further preparation. */ -static inline int prepare_hugepage_range(unsigned long addr, unsigned long len) +static inline int prepare_hugepage_range(struct file *file, + unsigned long addr, unsigned long len) { if (len & ~HPAGE_MASK) return -EINVAL; diff --git a/include/asm-sh/hugetlb.h b/include/asm-sh/hugetlb.h index fb30018938c7..967068fb79ac 100644 --- a/include/asm-sh/hugetlb.h +++ b/include/asm-sh/hugetlb.h @@ -14,7 +14,8 @@ static inline int is_hugepage_only_range(struct mm_struct *mm, * If the arch doesn't supply something else, assume that hugepage * size aligned regions are ok without further preparation. */ -static inline int prepare_hugepage_range(unsigned long addr, unsigned long len) +static inline int prepare_hugepage_range(struct file *file, + unsigned long addr, unsigned long len) { if (len & ~HPAGE_MASK) return -EINVAL; diff --git a/include/asm-sparc/hugetlb.h b/include/asm-sparc/hugetlb.h index aeb92374ca3d..177061064ee6 100644 --- a/include/asm-sparc/hugetlb.h +++ b/include/asm-sparc/hugetlb.h @@ -22,7 +22,8 @@ static inline int is_hugepage_only_range(struct mm_struct *mm, * If the arch doesn't supply something else, assume that hugepage * size aligned regions are ok without further preparation. */ -static inline int prepare_hugepage_range(unsigned long addr, unsigned long len) +static inline int prepare_hugepage_range(struct file *file, + unsigned long addr, unsigned long len) { if (len & ~HPAGE_MASK) return -EINVAL; diff --git a/include/asm-x86/hugetlb.h b/include/asm-x86/hugetlb.h index 7eed6e0883bf..439a9acc132d 100644 --- a/include/asm-x86/hugetlb.h +++ b/include/asm-x86/hugetlb.h @@ -14,11 +14,13 @@ static inline int is_hugepage_only_range(struct mm_struct *mm, * If the arch doesn't supply something else, assume that hugepage * size aligned regions are ok without further preparation. */ -static inline int prepare_hugepage_range(unsigned long addr, unsigned long len) +static inline int prepare_hugepage_range(struct file *file, + unsigned long addr, unsigned long len) { - if (len & ~HPAGE_MASK) + struct hstate *h = hstate_file(file); + if (len & ~huge_page_mask(h)) return -EINVAL; - if (addr & ~HPAGE_MASK) + if (addr & ~huge_page_mask(h)) return -EINVAL; return 0; } diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index abbc187193a1..ad2271e11f9b 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -8,7 +8,6 @@ #include #include #include -#include struct ctl_table; @@ -45,7 +44,8 @@ extern int sysctl_hugetlb_shm_group; /* arch callbacks */ -pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr); +pte_t *huge_pte_alloc(struct mm_struct *mm, + unsigned long addr, unsigned long sz); pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr); int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep); struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address, @@ -80,7 +80,7 @@ static inline unsigned long hugetlb_total_pages(void) #define hugetlb_report_meminfo(buf) 0 #define hugetlb_report_node_meminfo(n, buf) 0 #define follow_huge_pmd(mm, addr, pmd, write) NULL -#define prepare_hugepage_range(addr,len) (-EINVAL) +#define prepare_hugepage_range(file, addr, len) (-EINVAL) #define pmd_huge(x) 0 #define is_hugepage_only_range(mm, addr, len) 0 #define hugetlb_free_pgd_range(tlb, addr, end, floor, ceiling) ({BUG(); 0; }) @@ -134,8 +134,6 @@ struct file *hugetlb_file_setup(const char *name, size_t); int hugetlb_get_quota(struct address_space *mapping, long delta); void hugetlb_put_quota(struct address_space *mapping, long delta); -#define BLOCKS_PER_HUGEPAGE (HPAGE_SIZE / 512) - static inline int is_file_hugepages(struct file *file) { if (file->f_op == &hugetlbfs_file_operations) @@ -164,4 +162,84 @@ unsigned long hugetlb_get_unmapped_area(struct file *file, unsigned long addr, unsigned long flags); #endif /* HAVE_ARCH_HUGETLB_UNMAPPED_AREA */ +#ifdef CONFIG_HUGETLB_PAGE + +/* Defines one hugetlb page size */ +struct hstate { + int hugetlb_next_nid; + unsigned int order; + unsigned long mask; + unsigned long max_huge_pages; + unsigned long nr_huge_pages; + unsigned long free_huge_pages; + unsigned long resv_huge_pages; + unsigned long surplus_huge_pages; + unsigned long nr_overcommit_huge_pages; + struct list_head hugepage_freelists[MAX_NUMNODES]; + unsigned int nr_huge_pages_node[MAX_NUMNODES]; + unsigned int free_huge_pages_node[MAX_NUMNODES]; + unsigned int surplus_huge_pages_node[MAX_NUMNODES]; +}; + +extern struct hstate default_hstate; + +static inline struct hstate *hstate_vma(struct vm_area_struct *vma) +{ + return &default_hstate; +} + +static inline struct hstate *hstate_file(struct file *f) +{ + return &default_hstate; +} + +static inline struct hstate *hstate_inode(struct inode *i) +{ + return &default_hstate; +} + +static inline unsigned long huge_page_size(struct hstate *h) +{ + return (unsigned long)PAGE_SIZE << h->order; +} + +static inline unsigned long huge_page_mask(struct hstate *h) +{ + return h->mask; +} + +static inline unsigned int huge_page_order(struct hstate *h) +{ + return h->order; +} + +static inline unsigned huge_page_shift(struct hstate *h) +{ + return h->order + PAGE_SHIFT; +} + +static inline unsigned int pages_per_huge_page(struct hstate *h) +{ + return 1 << h->order; +} + +static inline unsigned int blocks_per_huge_page(struct hstate *h) +{ + return huge_page_size(h) / 512; +} + +#include + +#else +struct hstate {}; +#define hstate_file(f) NULL +#define hstate_vma(v) NULL +#define hstate_inode(i) NULL +#define huge_page_size(h) PAGE_SIZE +#define huge_page_mask(h) PAGE_MASK +#define huge_page_order(h) 0 +#define huge_page_shift(h) PAGE_SHIFT +#define pages_per_huge_page(h) 1 +#endif + #endif /* _LINUX_HUGETLB_H */ diff --git a/ipc/shm.c b/ipc/shm.c index 790240cd067f..a726aebce7d7 100644 --- a/ipc/shm.c +++ b/ipc/shm.c @@ -577,7 +577,8 @@ static void shm_get_stat(struct ipc_namespace *ns, unsigned long *rss, if (is_file_hugepages(shp->shm_file)) { struct address_space *mapping = inode->i_mapping; - *rss += (HPAGE_SIZE/PAGE_SIZE)*mapping->nrpages; + struct hstate *h = hstate_file(shp->shm_file); + *rss += pages_per_huge_page(h) * mapping->nrpages; } else { struct shmem_inode_info *info = SHMEM_I(inode); spin_lock(&info->lock); diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 32dff4290c66..0d8153e25f09 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -22,18 +22,12 @@ #include "internal.h" const unsigned long hugetlb_zero = 0, hugetlb_infinity = ~0UL; -static unsigned long nr_huge_pages, free_huge_pages, resv_huge_pages; -static unsigned long surplus_huge_pages; -static unsigned long nr_overcommit_huge_pages; unsigned long max_huge_pages; unsigned long sysctl_overcommit_huge_pages; -static struct list_head hugepage_freelists[MAX_NUMNODES]; -static unsigned int nr_huge_pages_node[MAX_NUMNODES]; -static unsigned int free_huge_pages_node[MAX_NUMNODES]; -static unsigned int surplus_huge_pages_node[MAX_NUMNODES]; static gfp_t htlb_alloc_mask = GFP_HIGHUSER; unsigned long hugepages_treat_as_movable; -static int hugetlb_next_nid; + +struct hstate default_hstate; /* * Protects updates to hugepage_freelists, nr_huge_pages, and free_huge_pages @@ -203,11 +197,11 @@ static long region_count(struct list_head *head, long f, long t) * Convert the address within this vma to the page offset within * the mapping, in pagecache page units; huge pages here. */ -static pgoff_t vma_hugecache_offset(struct vm_area_struct *vma, - unsigned long address) +static pgoff_t vma_hugecache_offset(struct hstate *h, + struct vm_area_struct *vma, unsigned long address) { - return ((address - vma->vm_start) >> HPAGE_SHIFT) + - (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT)); + return ((address - vma->vm_start) >> huge_page_shift(h)) + + (vma->vm_pgoff >> huge_page_order(h)); } /* @@ -309,20 +303,21 @@ static int is_vma_resv_set(struct vm_area_struct *vma, unsigned long flag) } /* Decrement the reserved pages in the hugepage pool by one */ -static void decrement_hugepage_resv_vma(struct vm_area_struct *vma) +static void decrement_hugepage_resv_vma(struct hstate *h, + struct vm_area_struct *vma) { if (vma->vm_flags & VM_NORESERVE) return; if (vma->vm_flags & VM_SHARED) { /* Shared mappings always use reserves */ - resv_huge_pages--; + h->resv_huge_pages--; } else if (is_vma_resv_set(vma, HPAGE_RESV_OWNER)) { /* * Only the process that called mmap() has reserves for * private mappings. */ - resv_huge_pages--; + h->resv_huge_pages--; } } @@ -344,12 +339,13 @@ static int vma_has_private_reserves(struct vm_area_struct *vma) return 1; } -static void clear_huge_page(struct page *page, unsigned long addr) +static void clear_huge_page(struct page *page, + unsigned long addr, unsigned long sz) { int i; might_sleep(); - for (i = 0; i < (HPAGE_SIZE/PAGE_SIZE); i++) { + for (i = 0; i < sz/PAGE_SIZE; i++) { cond_resched(); clear_user_highpage(page + i, addr + i * PAGE_SIZE); } @@ -359,41 +355,43 @@ static void copy_huge_page(struct page *dst, struct page *src, unsigned long addr, struct vm_area_struct *vma) { int i; + struct hstate *h = hstate_vma(vma); might_sleep(); - for (i = 0; i < HPAGE_SIZE/PAGE_SIZE; i++) { + for (i = 0; i < pages_per_huge_page(h); i++) { cond_resched(); copy_user_highpage(dst + i, src + i, addr + i*PAGE_SIZE, vma); } } -static void enqueue_huge_page(struct page *page) +static void enqueue_huge_page(struct hstate *h, struct page *page) { int nid = page_to_nid(page); - list_add(&page->lru, &hugepage_freelists[nid]); - free_huge_pages++; - free_huge_pages_node[nid]++; + list_add(&page->lru, &h->hugepage_freelists[nid]); + h->free_huge_pages++; + h->free_huge_pages_node[nid]++; } -static struct page *dequeue_huge_page(void) +static struct page *dequeue_huge_page(struct hstate *h) { int nid; struct page *page = NULL; for (nid = 0; nid < MAX_NUMNODES; ++nid) { - if (!list_empty(&hugepage_freelists[nid])) { - page = list_entry(hugepage_freelists[nid].next, + if (!list_empty(&h->hugepage_freelists[nid])) { + page = list_entry(h->hugepage_freelists[nid].next, struct page, lru); list_del(&page->lru); - free_huge_pages--; - free_huge_pages_node[nid]--; + h->free_huge_pages--; + h->free_huge_pages_node[nid]--; break; } } return page; } -static struct page *dequeue_huge_page_vma(struct vm_area_struct *vma, +static struct page *dequeue_huge_page_vma(struct hstate *h, + struct vm_area_struct *vma, unsigned long address, int avoid_reserve) { int nid; @@ -411,26 +409,26 @@ static struct page *dequeue_huge_page_vma(struct vm_area_struct *vma, * not "stolen". The child may still get SIGKILLed */ if (!vma_has_private_reserves(vma) && - free_huge_pages - resv_huge_pages == 0) + h->free_huge_pages - h->resv_huge_pages == 0) return NULL; /* If reserves cannot be used, ensure enough pages are in the pool */ - if (avoid_reserve && free_huge_pages - resv_huge_pages == 0) + if (avoid_reserve && h->free_huge_pages - h->resv_huge_pages == 0) return NULL; for_each_zone_zonelist_nodemask(zone, z, zonelist, MAX_NR_ZONES - 1, nodemask) { nid = zone_to_nid(zone); if (cpuset_zone_allowed_softwall(zone, htlb_alloc_mask) && - !list_empty(&hugepage_freelists[nid])) { - page = list_entry(hugepage_freelists[nid].next, + !list_empty(&h->hugepage_freelists[nid])) { + page = list_entry(h->hugepage_freelists[nid].next, struct page, lru); list_del(&page->lru); - free_huge_pages--; - free_huge_pages_node[nid]--; + h->free_huge_pages--; + h->free_huge_pages_node[nid]--; if (!avoid_reserve) - decrement_hugepage_resv_vma(vma); + decrement_hugepage_resv_vma(h, vma); break; } @@ -439,12 +437,13 @@ static struct page *dequeue_huge_page_vma(struct vm_area_struct *vma, return page; } -static void update_and_free_page(struct page *page) +static void update_and_free_page(struct hstate *h, struct page *page) { int i; - nr_huge_pages--; - nr_huge_pages_node[page_to_nid(page)]--; - for (i = 0; i < (HPAGE_SIZE / PAGE_SIZE); i++) { + + h->nr_huge_pages--; + h->nr_huge_pages_node[page_to_nid(page)]--; + for (i = 0; i < pages_per_huge_page(h); i++) { page[i].flags &= ~(1 << PG_locked | 1 << PG_error | 1 << PG_referenced | 1 << PG_dirty | 1 << PG_active | 1 << PG_reserved | 1 << PG_private | 1<< PG_writeback); @@ -452,11 +451,16 @@ static void update_and_free_page(struct page *page) set_compound_page_dtor(page, NULL); set_page_refcounted(page); arch_release_hugepage(page); - __free_pages(page, HUGETLB_PAGE_ORDER); + __free_pages(page, huge_page_order(h)); } static void free_huge_page(struct page *page) { + /* + * Can't pass hstate in here because it is called from the + * compound page destructor. + */ + struct hstate *h = &default_hstate; int nid = page_to_nid(page); struct address_space *mapping; @@ -466,12 +470,12 @@ static void free_huge_page(struct page *page) INIT_LIST_HEAD(&page->lru); spin_lock(&hugetlb_lock); - if (surplus_huge_pages_node[nid]) { - update_and_free_page(page); - surplus_huge_pages--; - surplus_huge_pages_node[nid]--; + if (h->surplus_huge_pages_node[nid]) { + update_and_free_page(h, page); + h->surplus_huge_pages--; + h->surplus_huge_pages_node[nid]--; } else { - enqueue_huge_page(page); + enqueue_huge_page(h, page); } spin_unlock(&hugetlb_lock); if (mapping) @@ -483,7 +487,7 @@ static void free_huge_page(struct page *page) * balanced by operating on them in a round-robin fashion. * Returns 1 if an adjustment was made. */ -static int adjust_pool_surplus(int delta) +static int adjust_pool_surplus(struct hstate *h, int delta) { static int prev_nid; int nid = prev_nid; @@ -496,15 +500,15 @@ static int adjust_pool_surplus(int delta) nid = first_node(node_online_map); /* To shrink on this node, there must be a surplus page */ - if (delta < 0 && !surplus_huge_pages_node[nid]) + if (delta < 0 && !h->surplus_huge_pages_node[nid]) continue; /* Surplus cannot exceed the total number of pages */ - if (delta > 0 && surplus_huge_pages_node[nid] >= - nr_huge_pages_node[nid]) + if (delta > 0 && h->surplus_huge_pages_node[nid] >= + h->nr_huge_pages_node[nid]) continue; - surplus_huge_pages += delta; - surplus_huge_pages_node[nid] += delta; + h->surplus_huge_pages += delta; + h->surplus_huge_pages_node[nid] += delta; ret = 1; break; } while (nid != prev_nid); @@ -513,46 +517,46 @@ static int adjust_pool_surplus(int delta) return ret; } -static void prep_new_huge_page(struct page *page, int nid) +static void prep_new_huge_page(struct hstate *h, struct page *page, int nid) { set_compound_page_dtor(page, free_huge_page); spin_lock(&hugetlb_lock); - nr_huge_pages++; - nr_huge_pages_node[nid]++; + h->nr_huge_pages++; + h->nr_huge_pages_node[nid]++; spin_unlock(&hugetlb_lock); put_page(page); /* free it into the hugepage allocator */ } -static struct page *alloc_fresh_huge_page_node(int nid) +static struct page *alloc_fresh_huge_page_node(struct hstate *h, int nid) { struct page *page; page = alloc_pages_node(nid, htlb_alloc_mask|__GFP_COMP|__GFP_THISNODE| __GFP_REPEAT|__GFP_NOWARN, - HUGETLB_PAGE_ORDER); + huge_page_order(h)); if (page) { if (arch_prepare_hugepage(page)) { __free_pages(page, HUGETLB_PAGE_ORDER); return NULL; } - prep_new_huge_page(page, nid); + prep_new_huge_page(h, page, nid); } return page; } -static int alloc_fresh_huge_page(void) +static int alloc_fresh_huge_page(struct hstate *h) { struct page *page; int start_nid; int next_nid; int ret = 0; - start_nid = hugetlb_next_nid; + start_nid = h->hugetlb_next_nid; do { - page = alloc_fresh_huge_page_node(hugetlb_next_nid); + page = alloc_fresh_huge_page_node(h, h->hugetlb_next_nid); if (page) ret = 1; /* @@ -566,11 +570,11 @@ static int alloc_fresh_huge_page(void) * if we just successfully allocated a hugepage so that * the next caller gets hugepages on the next node. */ - next_nid = next_node(hugetlb_next_nid, node_online_map); + next_nid = next_node(h->hugetlb_next_nid, node_online_map); if (next_nid == MAX_NUMNODES) next_nid = first_node(node_online_map); - hugetlb_next_nid = next_nid; - } while (!page && hugetlb_next_nid != start_nid); + h->hugetlb_next_nid = next_nid; + } while (!page && h->hugetlb_next_nid != start_nid); if (ret) count_vm_event(HTLB_BUDDY_PGALLOC); @@ -580,8 +584,8 @@ static int alloc_fresh_huge_page(void) return ret; } -static struct page *alloc_buddy_huge_page(struct vm_area_struct *vma, - unsigned long address) +static struct page *alloc_buddy_huge_page(struct hstate *h, + struct vm_area_struct *vma, unsigned long address) { struct page *page; unsigned int nid; @@ -610,18 +614,18 @@ static struct page *alloc_buddy_huge_page(struct vm_area_struct *vma, * per-node value is checked there. */ spin_lock(&hugetlb_lock); - if (surplus_huge_pages >= nr_overcommit_huge_pages) { + if (h->surplus_huge_pages >= h->nr_overcommit_huge_pages) { spin_unlock(&hugetlb_lock); return NULL; } else { - nr_huge_pages++; - surplus_huge_pages++; + h->nr_huge_pages++; + h->surplus_huge_pages++; } spin_unlock(&hugetlb_lock); page = alloc_pages(htlb_alloc_mask|__GFP_COMP| __GFP_REPEAT|__GFP_NOWARN, - HUGETLB_PAGE_ORDER); + huge_page_order(h)); spin_lock(&hugetlb_lock); if (page) { @@ -636,12 +640,12 @@ static struct page *alloc_buddy_huge_page(struct vm_area_struct *vma, /* * We incremented the global counters already */ - nr_huge_pages_node[nid]++; - surplus_huge_pages_node[nid]++; + h->nr_huge_pages_node[nid]++; + h->surplus_huge_pages_node[nid]++; __count_vm_event(HTLB_BUDDY_PGALLOC); } else { - nr_huge_pages--; - surplus_huge_pages--; + h->nr_huge_pages--; + h->surplus_huge_pages--; __count_vm_event(HTLB_BUDDY_PGALLOC_FAIL); } spin_unlock(&hugetlb_lock); @@ -653,16 +657,16 @@ static struct page *alloc_buddy_huge_page(struct vm_area_struct *vma, * Increase the hugetlb pool such that it can accomodate a reservation * of size 'delta'. */ -static int gather_surplus_pages(int delta) +static int gather_surplus_pages(struct hstate *h, int delta) { struct list_head surplus_list; struct page *page, *tmp; int ret, i; int needed, allocated; - needed = (resv_huge_pages + delta) - free_huge_pages; + needed = (h->resv_huge_pages + delta) - h->free_huge_pages; if (needed <= 0) { - resv_huge_pages += delta; + h->resv_huge_pages += delta; return 0; } @@ -673,7 +677,7 @@ static int gather_surplus_pages(int delta) retry: spin_unlock(&hugetlb_lock); for (i = 0; i < needed; i++) { - page = alloc_buddy_huge_page(NULL, 0); + page = alloc_buddy_huge_page(h, NULL, 0); if (!page) { /* * We were not able to allocate enough pages to @@ -694,7 +698,8 @@ static int gather_surplus_pages(int delta) * because either resv_huge_pages or free_huge_pages may have changed. */ spin_lock(&hugetlb_lock); - needed = (resv_huge_pages + delta) - (free_huge_pages + allocated); + needed = (h->resv_huge_pages + delta) - + (h->free_huge_pages + allocated); if (needed > 0) goto retry; @@ -707,7 +712,7 @@ static int gather_surplus_pages(int delta) * before they are reserved. */ needed += allocated; - resv_huge_pages += delta; + h->resv_huge_pages += delta; ret = 0; free: /* Free the needed pages to the hugetlb pool */ @@ -715,7 +720,7 @@ static int gather_surplus_pages(int delta) if ((--needed) < 0) break; list_del(&page->lru); - enqueue_huge_page(page); + enqueue_huge_page(h, page); } /* Free unnecessary surplus pages to the buddy allocator */ @@ -743,7 +748,8 @@ static int gather_surplus_pages(int delta) * allocated to satisfy the reservation must be explicitly freed if they were * never used. */ -static void return_unused_surplus_pages(unsigned long unused_resv_pages) +static void return_unused_surplus_pages(struct hstate *h, + unsigned long unused_resv_pages) { static int nid = -1; struct page *page; @@ -758,27 +764,27 @@ static void return_unused_surplus_pages(unsigned long unused_resv_pages) unsigned long remaining_iterations = num_online_nodes(); /* Uncommit the reservation */ - resv_huge_pages -= unused_resv_pages; + h->resv_huge_pages -= unused_resv_pages; - nr_pages = min(unused_resv_pages, surplus_huge_pages); + nr_pages = min(unused_resv_pages, h->surplus_huge_pages); while (remaining_iterations-- && nr_pages) { nid = next_node(nid, node_online_map); if (nid == MAX_NUMNODES) nid = first_node(node_online_map); - if (!surplus_huge_pages_node[nid]) + if (!h->surplus_huge_pages_node[nid]) continue; - if (!list_empty(&hugepage_freelists[nid])) { - page = list_entry(hugepage_freelists[nid].next, + if (!list_empty(&h->hugepage_freelists[nid])) { + page = list_entry(h->hugepage_freelists[nid].next, struct page, lru); list_del(&page->lru); - update_and_free_page(page); - free_huge_pages--; - free_huge_pages_node[nid]--; - surplus_huge_pages--; - surplus_huge_pages_node[nid]--; + update_and_free_page(h, page); + h->free_huge_pages--; + h->free_huge_pages_node[nid]--; + h->surplus_huge_pages--; + h->surplus_huge_pages_node[nid]--; nr_pages--; remaining_iterations = num_online_nodes(); } @@ -794,13 +800,14 @@ static void return_unused_surplus_pages(unsigned long unused_resv_pages) * an instantiated the change should be committed via vma_commit_reservation. * No action is required on failure. */ -static int vma_needs_reservation(struct vm_area_struct *vma, unsigned long addr) +static int vma_needs_reservation(struct hstate *h, + struct vm_area_struct *vma, unsigned long addr) { struct address_space *mapping = vma->vm_file->f_mapping; struct inode *inode = mapping->host; if (vma->vm_flags & VM_SHARED) { - pgoff_t idx = vma_hugecache_offset(vma, addr); + pgoff_t idx = vma_hugecache_offset(h, vma, addr); return region_chg(&inode->i_mapping->private_list, idx, idx + 1); @@ -809,7 +816,7 @@ static int vma_needs_reservation(struct vm_area_struct *vma, unsigned long addr) } else { int err; - pgoff_t idx = vma_hugecache_offset(vma, addr); + pgoff_t idx = vma_hugecache_offset(h, vma, addr); struct resv_map *reservations = vma_resv_map(vma); err = region_chg(&reservations->regions, idx, idx + 1); @@ -818,18 +825,18 @@ static int vma_needs_reservation(struct vm_area_struct *vma, unsigned long addr) return 0; } } -static void vma_commit_reservation(struct vm_area_struct *vma, - unsigned long addr) +static void vma_commit_reservation(struct hstate *h, + struct vm_area_struct *vma, unsigned long addr) { struct address_space *mapping = vma->vm_file->f_mapping; struct inode *inode = mapping->host; if (vma->vm_flags & VM_SHARED) { - pgoff_t idx = vma_hugecache_offset(vma, addr); + pgoff_t idx = vma_hugecache_offset(h, vma, addr); region_add(&inode->i_mapping->private_list, idx, idx + 1); } else if (is_vma_resv_set(vma, HPAGE_RESV_OWNER)) { - pgoff_t idx = vma_hugecache_offset(vma, addr); + pgoff_t idx = vma_hugecache_offset(h, vma, addr); struct resv_map *reservations = vma_resv_map(vma); /* Mark this page used in the map. */ @@ -840,6 +847,7 @@ static void vma_commit_reservation(struct vm_area_struct *vma, static struct page *alloc_huge_page(struct vm_area_struct *vma, unsigned long addr, int avoid_reserve) { + struct hstate *h = hstate_vma(vma); struct page *page; struct address_space *mapping = vma->vm_file->f_mapping; struct inode *inode = mapping->host; @@ -852,7 +860,7 @@ static struct page *alloc_huge_page(struct vm_area_struct *vma, * MAP_NORESERVE mappings may also need pages and quota allocated * if no reserve mapping overlaps. */ - chg = vma_needs_reservation(vma, addr); + chg = vma_needs_reservation(h, vma, addr); if (chg < 0) return ERR_PTR(chg); if (chg) @@ -860,11 +868,11 @@ static struct page *alloc_huge_page(struct vm_area_struct *vma, return ERR_PTR(-ENOSPC); spin_lock(&hugetlb_lock); - page = dequeue_huge_page_vma(vma, addr, avoid_reserve); + page = dequeue_huge_page_vma(h, vma, addr, avoid_reserve); spin_unlock(&hugetlb_lock); if (!page) { - page = alloc_buddy_huge_page(vma, addr); + page = alloc_buddy_huge_page(h, vma, addr); if (!page) { hugetlb_put_quota(inode->i_mapping, chg); return ERR_PTR(-VM_FAULT_OOM); @@ -874,7 +882,7 @@ static struct page *alloc_huge_page(struct vm_area_struct *vma, set_page_refcounted(page); set_page_private(page, (unsigned long) mapping); - vma_commit_reservation(vma, addr); + vma_commit_reservation(h, vma, addr); return page; } @@ -882,21 +890,28 @@ static struct page *alloc_huge_page(struct vm_area_struct *vma, static int __init hugetlb_init(void) { unsigned long i; + struct hstate *h = &default_hstate; if (HPAGE_SHIFT == 0) return 0; - for (i = 0; i < MAX_NUMNODES; ++i) - INIT_LIST_HEAD(&hugepage_freelists[i]); + if (!h->order) { + h->order = HPAGE_SHIFT - PAGE_SHIFT; + h->mask = HPAGE_MASK; + } - hugetlb_next_nid = first_node(node_online_map); + for (i = 0; i < MAX_NUMNODES; ++i) + INIT_LIST_HEAD(&h->hugepage_freelists[i]); + + h->hugetlb_next_nid = first_node(node_online_map); for (i = 0; i < max_huge_pages; ++i) { - if (!alloc_fresh_huge_page()) + if (!alloc_fresh_huge_page(h)) break; } - max_huge_pages = free_huge_pages = nr_huge_pages = i; - printk("Total HugeTLB memory allocated, %ld\n", free_huge_pages); + max_huge_pages = h->free_huge_pages = h->nr_huge_pages = i; + printk(KERN_INFO "Total HugeTLB memory allocated, %ld\n", + h->free_huge_pages); return 0; } module_init(hugetlb_init); @@ -922,34 +937,36 @@ static unsigned int cpuset_mems_nr(unsigned int *array) #ifdef CONFIG_SYSCTL #ifdef CONFIG_HIGHMEM -static void try_to_free_low(unsigned long count) +static void try_to_free_low(struct hstate *h, unsigned long count) { int i; for (i = 0; i < MAX_NUMNODES; ++i) { struct page *page, *next; - list_for_each_entry_safe(page, next, &hugepage_freelists[i], lru) { - if (count >= nr_huge_pages) + struct list_head *freel = &h->hugepage_freelists[i]; + list_for_each_entry_safe(page, next, freel, lru) { + if (count >= h->nr_huge_pages) return; if (PageHighMem(page)) continue; list_del(&page->lru); update_and_free_page(page); - free_huge_pages--; - free_huge_pages_node[page_to_nid(page)]--; + h->free_huge_pages--; + h->free_huge_pages_node[page_to_nid(page)]--; } } } #else -static inline void try_to_free_low(unsigned long count) +static inline void try_to_free_low(struct hstate *h, unsigned long count) { } #endif -#define persistent_huge_pages (nr_huge_pages - surplus_huge_pages) +#define persistent_huge_pages(h) (h->nr_huge_pages - h->surplus_huge_pages) static unsigned long set_max_huge_pages(unsigned long count) { unsigned long min_count, ret; + struct hstate *h = &default_hstate; /* * Increase the pool size @@ -963,19 +980,19 @@ static unsigned long set_max_huge_pages(unsigned long count) * within all the constraints specified by the sysctls. */ spin_lock(&hugetlb_lock); - while (surplus_huge_pages && count > persistent_huge_pages) { - if (!adjust_pool_surplus(-1)) + while (h->surplus_huge_pages && count > persistent_huge_pages(h)) { + if (!adjust_pool_surplus(h, -1)) break; } - while (count > persistent_huge_pages) { + while (count > persistent_huge_pages(h)) { /* * If this allocation races such that we no longer need the * page, free_huge_page will handle it by freeing the page * and reducing the surplus. */ spin_unlock(&hugetlb_lock); - ret = alloc_fresh_huge_page(); + ret = alloc_fresh_huge_page(h); spin_lock(&hugetlb_lock); if (!ret) goto out; @@ -997,21 +1014,21 @@ static unsigned long set_max_huge_pages(unsigned long count) * and won't grow the pool anywhere else. Not until one of the * sysctls are changed, or the surplus pages go out of use. */ - min_count = resv_huge_pages + nr_huge_pages - free_huge_pages; + min_count = h->resv_huge_pages + h->nr_huge_pages - h->free_huge_pages; min_count = max(count, min_count); - try_to_free_low(min_count); - while (min_count < persistent_huge_pages) { - struct page *page = dequeue_huge_page(); + try_to_free_low(h, min_count); + while (min_count < persistent_huge_pages(h)) { + struct page *page = dequeue_huge_page(h); if (!page) break; - update_and_free_page(page); + update_and_free_page(h, page); } - while (count < persistent_huge_pages) { - if (!adjust_pool_surplus(1)) + while (count < persistent_huge_pages(h)) { + if (!adjust_pool_surplus(h, 1)) break; } out: - ret = persistent_huge_pages; + ret = persistent_huge_pages(h); spin_unlock(&hugetlb_lock); return ret; } @@ -1041,9 +1058,10 @@ int hugetlb_overcommit_handler(struct ctl_table *table, int write, struct file *file, void __user *buffer, size_t *length, loff_t *ppos) { + struct hstate *h = &default_hstate; proc_doulongvec_minmax(table, write, file, buffer, length, ppos); spin_lock(&hugetlb_lock); - nr_overcommit_huge_pages = sysctl_overcommit_huge_pages; + h->nr_overcommit_huge_pages = sysctl_overcommit_huge_pages; spin_unlock(&hugetlb_lock); return 0; } @@ -1052,37 +1070,40 @@ int hugetlb_overcommit_handler(struct ctl_table *table, int write, int hugetlb_report_meminfo(char *buf) { + struct hstate *h = &default_hstate; return sprintf(buf, "HugePages_Total: %5lu\n" "HugePages_Free: %5lu\n" "HugePages_Rsvd: %5lu\n" "HugePages_Surp: %5lu\n" "Hugepagesize: %5lu kB\n", - nr_huge_pages, - free_huge_pages, - resv_huge_pages, - surplus_huge_pages, - HPAGE_SIZE/1024); + h->nr_huge_pages, + h->free_huge_pages, + h->resv_huge_pages, + h->surplus_huge_pages, + 1UL << (huge_page_order(h) + PAGE_SHIFT - 10)); } int hugetlb_report_node_meminfo(int nid, char *buf) { + struct hstate *h = &default_hstate; return sprintf(buf, "Node %d HugePages_Total: %5u\n" "Node %d HugePages_Free: %5u\n" "Node %d HugePages_Surp: %5u\n", - nid, nr_huge_pages_node[nid], - nid, free_huge_pages_node[nid], - nid, surplus_huge_pages_node[nid]); + nid, h->nr_huge_pages_node[nid], + nid, h->free_huge_pages_node[nid], + nid, h->surplus_huge_pages_node[nid]); } /* Return the number pages of memory we physically have, in PAGE_SIZE units. */ unsigned long hugetlb_total_pages(void) { - return nr_huge_pages * (HPAGE_SIZE / PAGE_SIZE); + struct hstate *h = &default_hstate; + return h->nr_huge_pages * pages_per_huge_page(h); } -static int hugetlb_acct_memory(long delta) +static int hugetlb_acct_memory(struct hstate *h, long delta) { int ret = -ENOMEM; @@ -1105,18 +1126,18 @@ static int hugetlb_acct_memory(long delta) * semantics that cpuset has. */ if (delta > 0) { - if (gather_surplus_pages(delta) < 0) + if (gather_surplus_pages(h, delta) < 0) goto out; - if (delta > cpuset_mems_nr(free_huge_pages_node)) { - return_unused_surplus_pages(delta); + if (delta > cpuset_mems_nr(h->free_huge_pages_node)) { + return_unused_surplus_pages(h, delta); goto out; } } ret = 0; if (delta < 0) - return_unused_surplus_pages((unsigned long) -delta); + return_unused_surplus_pages(h, (unsigned long) -delta); out: spin_unlock(&hugetlb_lock); @@ -1141,14 +1162,15 @@ static void hugetlb_vm_op_open(struct vm_area_struct *vma) static void hugetlb_vm_op_close(struct vm_area_struct *vma) { + struct hstate *h = hstate_vma(vma); struct resv_map *reservations = vma_resv_map(vma); unsigned long reserve; unsigned long start; unsigned long end; if (reservations) { - start = vma_hugecache_offset(vma, vma->vm_start); - end = vma_hugecache_offset(vma, vma->vm_end); + start = vma_hugecache_offset(h, vma, vma->vm_start); + end = vma_hugecache_offset(h, vma, vma->vm_end); reserve = (end - start) - region_count(&reservations->regions, start, end); @@ -1156,7 +1178,7 @@ static void hugetlb_vm_op_close(struct vm_area_struct *vma) kref_put(&reservations->refs, resv_map_release); if (reserve) - hugetlb_acct_memory(-reserve); + hugetlb_acct_memory(h, -reserve); } } @@ -1214,14 +1236,16 @@ int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src, struct page *ptepage; unsigned long addr; int cow; + struct hstate *h = hstate_vma(vma); + unsigned long sz = huge_page_size(h); cow = (vma->vm_flags & (VM_SHARED | VM_MAYWRITE)) == VM_MAYWRITE; - for (addr = vma->vm_start; addr < vma->vm_end; addr += HPAGE_SIZE) { + for (addr = vma->vm_start; addr < vma->vm_end; addr += sz) { src_pte = huge_pte_offset(src, addr); if (!src_pte) continue; - dst_pte = huge_pte_alloc(dst, addr); + dst_pte = huge_pte_alloc(dst, addr, sz); if (!dst_pte) goto nomem; @@ -1257,6 +1281,9 @@ void __unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start, pte_t pte; struct page *page; struct page *tmp; + struct hstate *h = hstate_vma(vma); + unsigned long sz = huge_page_size(h); + /* * A page gathering list, protected by per file i_mmap_lock. The * lock is used to avoid list corruption from multiple unmapping @@ -1265,11 +1292,11 @@ void __unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start, LIST_HEAD(page_list); WARN_ON(!is_vm_hugetlb_page(vma)); - BUG_ON(start & ~HPAGE_MASK); - BUG_ON(end & ~HPAGE_MASK); + BUG_ON(start & ~huge_page_mask(h)); + BUG_ON(end & ~huge_page_mask(h)); spin_lock(&mm->page_table_lock); - for (address = start; address < end; address += HPAGE_SIZE) { + for (address = start; address < end; address += sz) { ptep = huge_pte_offset(mm, address); if (!ptep) continue; @@ -1383,6 +1410,7 @@ static int hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long address, pte_t *ptep, pte_t pte, struct page *pagecache_page) { + struct hstate *h = hstate_vma(vma); struct page *old_page, *new_page; int avoidcopy; int outside_reserve = 0; @@ -1443,7 +1471,7 @@ static int hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma, __SetPageUptodate(new_page); spin_lock(&mm->page_table_lock); - ptep = huge_pte_offset(mm, address & HPAGE_MASK); + ptep = huge_pte_offset(mm, address & huge_page_mask(h)); if (likely(pte_same(huge_ptep_get(ptep), pte))) { /* Break COW */ huge_ptep_clear_flush(vma, address, ptep); @@ -1458,14 +1486,14 @@ static int hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma, } /* Return the pagecache page at a given address within a VMA */ -static struct page *hugetlbfs_pagecache_page(struct vm_area_struct *vma, - unsigned long address) +static struct page *hugetlbfs_pagecache_page(struct hstate *h, + struct vm_area_struct *vma, unsigned long address) { struct address_space *mapping; pgoff_t idx; mapping = vma->vm_file->f_mapping; - idx = vma_hugecache_offset(vma, address); + idx = vma_hugecache_offset(h, vma, address); return find_lock_page(mapping, idx); } @@ -1473,6 +1501,7 @@ static struct page *hugetlbfs_pagecache_page(struct vm_area_struct *vma, static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long address, pte_t *ptep, int write_access) { + struct hstate *h = hstate_vma(vma); int ret = VM_FAULT_SIGBUS; pgoff_t idx; unsigned long size; @@ -1493,7 +1522,7 @@ static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma, } mapping = vma->vm_file->f_mapping; - idx = vma_hugecache_offset(vma, address); + idx = vma_hugecache_offset(h, vma, address); /* * Use page lock to guard against racing truncation @@ -1502,7 +1531,7 @@ static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma, retry: page = find_lock_page(mapping, idx); if (!page) { - size = i_size_read(mapping->host) >> HPAGE_SHIFT; + size = i_size_read(mapping->host) >> huge_page_shift(h); if (idx >= size) goto out; page = alloc_huge_page(vma, address, 0); @@ -1510,7 +1539,7 @@ static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma, ret = -PTR_ERR(page); goto out; } - clear_huge_page(page, address); + clear_huge_page(page, address, huge_page_size(h)); __SetPageUptodate(page); if (vma->vm_flags & VM_SHARED) { @@ -1526,14 +1555,14 @@ static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma, } spin_lock(&inode->i_lock); - inode->i_blocks += BLOCKS_PER_HUGEPAGE; + inode->i_blocks += blocks_per_huge_page(h); spin_unlock(&inode->i_lock); } else lock_page(page); } spin_lock(&mm->page_table_lock); - size = i_size_read(mapping->host) >> HPAGE_SHIFT; + size = i_size_read(mapping->host) >> huge_page_shift(h); if (idx >= size) goto backout; @@ -1569,8 +1598,9 @@ int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma, pte_t entry; int ret; static DEFINE_MUTEX(hugetlb_instantiation_mutex); + struct hstate *h = hstate_vma(vma); - ptep = huge_pte_alloc(mm, address); + ptep = huge_pte_alloc(mm, address, huge_page_size(h)); if (!ptep) return VM_FAULT_OOM; @@ -1594,7 +1624,7 @@ int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma, if (likely(pte_same(entry, huge_ptep_get(ptep)))) if (write_access && !pte_write(entry)) { struct page *page; - page = hugetlbfs_pagecache_page(vma, address); + page = hugetlbfs_pagecache_page(h, vma, address); ret = hugetlb_cow(mm, vma, address, ptep, entry, page); if (page) { unlock_page(page); @@ -1615,6 +1645,7 @@ int follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long pfn_offset; unsigned long vaddr = *position; int remainder = *length; + struct hstate *h = hstate_vma(vma); spin_lock(&mm->page_table_lock); while (vaddr < vma->vm_end && remainder) { @@ -1626,7 +1657,7 @@ int follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma, * each hugepage. We have to make * sure we get the * first, for the page indexing below to work. */ - pte = huge_pte_offset(mm, vaddr & HPAGE_MASK); + pte = huge_pte_offset(mm, vaddr & huge_page_mask(h)); if (!pte || huge_pte_none(huge_ptep_get(pte)) || (write && !pte_write(huge_ptep_get(pte)))) { @@ -1644,7 +1675,7 @@ int follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma, break; } - pfn_offset = (vaddr & ~HPAGE_MASK) >> PAGE_SHIFT; + pfn_offset = (vaddr & ~huge_page_mask(h)) >> PAGE_SHIFT; page = pte_page(huge_ptep_get(pte)); same_page: if (pages) { @@ -1660,7 +1691,7 @@ int follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma, --remainder; ++i; if (vaddr < vma->vm_end && remainder && - pfn_offset < HPAGE_SIZE/PAGE_SIZE) { + pfn_offset < pages_per_huge_page(h)) { /* * We use pfn_offset to avoid touching the pageframes * of this compound page. @@ -1682,13 +1713,14 @@ void hugetlb_change_protection(struct vm_area_struct *vma, unsigned long start = address; pte_t *ptep; pte_t pte; + struct hstate *h = hstate_vma(vma); BUG_ON(address >= end); flush_cache_range(vma, address, end); spin_lock(&vma->vm_file->f_mapping->i_mmap_lock); spin_lock(&mm->page_table_lock); - for (; address < end; address += HPAGE_SIZE) { + for (; address < end; address += huge_page_size(h)) { ptep = huge_pte_offset(mm, address); if (!ptep) continue; @@ -1711,6 +1743,7 @@ int hugetlb_reserve_pages(struct inode *inode, struct vm_area_struct *vma) { long ret, chg; + struct hstate *h = hstate_inode(inode); if (vma && vma->vm_flags & VM_NORESERVE) return 0; @@ -1739,7 +1772,7 @@ int hugetlb_reserve_pages(struct inode *inode, if (hugetlb_get_quota(inode->i_mapping, chg)) return -ENOSPC; - ret = hugetlb_acct_memory(chg); + ret = hugetlb_acct_memory(h, chg); if (ret < 0) { hugetlb_put_quota(inode->i_mapping, chg); return ret; @@ -1751,12 +1784,13 @@ int hugetlb_reserve_pages(struct inode *inode, void hugetlb_unreserve_pages(struct inode *inode, long offset, long freed) { + struct hstate *h = hstate_inode(inode); long chg = region_truncate(&inode->i_mapping->private_list, offset); spin_lock(&inode->i_lock); - inode->i_blocks -= BLOCKS_PER_HUGEPAGE * freed; + inode->i_blocks -= blocks_per_huge_page(h); spin_unlock(&inode->i_lock); hugetlb_put_quota(inode->i_mapping, (chg - freed)); - hugetlb_acct_memory(-(chg - freed)); + hugetlb_acct_memory(h, -(chg - freed)); } diff --git a/mm/memory.c b/mm/memory.c index 72932489a082..c1c1d6d8c22b 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -903,7 +903,7 @@ unsigned long unmap_vmas(struct mmu_gather **tlbp, if (unlikely(is_vm_hugetlb_page(vma))) { unmap_hugepage_range(vma, start, end, NULL); zap_work -= (end - start) / - (HPAGE_SIZE / PAGE_SIZE); + pages_per_huge_page(hstate_vma(vma)); start = end; } else start = unmap_page_range(*tlbp, vma, diff --git a/mm/mempolicy.c b/mm/mempolicy.c index c94e58b192c3..e550bec20582 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -1481,7 +1481,7 @@ struct zonelist *huge_zonelist(struct vm_area_struct *vma, unsigned long addr, if (unlikely((*mpol)->mode == MPOL_INTERLEAVE)) { zl = node_zonelist(interleave_nid(*mpol, vma, addr, - HPAGE_SHIFT), gfp_flags); + huge_page_shift(hstate_vma(vma))), gfp_flags); } else { zl = policy_zonelist(gfp_flags, *mpol); if ((*mpol)->mode == MPOL_BIND) @@ -2220,9 +2220,12 @@ static void check_huge_range(struct vm_area_struct *vma, { unsigned long addr; struct page *page; + struct hstate *h = hstate_vma(vma); + unsigned long sz = huge_page_size(h); - for (addr = start; addr < end; addr += HPAGE_SIZE) { - pte_t *ptep = huge_pte_offset(vma->vm_mm, addr & HPAGE_MASK); + for (addr = start; addr < end; addr += sz) { + pte_t *ptep = huge_pte_offset(vma->vm_mm, + addr & huge_page_mask(h)); pte_t pte; if (!ptep) diff --git a/mm/mmap.c b/mm/mmap.c index 57d3b6097deb..5e0cc99e9cd5 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1812,7 +1812,8 @@ int split_vma(struct mm_struct * mm, struct vm_area_struct * vma, struct mempolicy *pol; struct vm_area_struct *new; - if (is_vm_hugetlb_page(vma) && (addr & ~HPAGE_MASK)) + if (is_vm_hugetlb_page(vma) && (addr & + ~(huge_page_mask(hstate_vma(vma))))) return -EINVAL; if (mm->map_count >= sysctl_max_map_count) From e5ff215941d59f8ae6bf58f6428dc5c26745a612 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 23 Jul 2008 21:27:42 -0700 Subject: [PATCH 169/857] hugetlb: multiple hstates for multiple page sizes Add basic support for more than one hstate in hugetlbfs. This is the key to supporting multiple hugetlbfs page sizes at once. - Rather than a single hstate, we now have an array, with an iterator - default_hstate continues to be the struct hstate which we use by default - Add functions for architectures to register new hstates [akpm@linux-foundation.org: coding-style fixes] Acked-by: Adam Litke Acked-by: Nishanth Aravamudan Signed-off-by: Andi Kleen Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/hugetlb.h | 19 +++++- kernel/sysctl.c | 8 ++- mm/hugetlb.c | 148 ++++++++++++++++++++++++++++++++-------- 3 files changed, 142 insertions(+), 33 deletions(-) diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index ad2271e11f9b..b75bdb4deba3 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -36,8 +36,6 @@ int hugetlb_reserve_pages(struct inode *inode, long from, long to, struct vm_area_struct *vma); void hugetlb_unreserve_pages(struct inode *inode, long offset, long freed); -extern unsigned long max_huge_pages; -extern unsigned long sysctl_overcommit_huge_pages; extern unsigned long hugepages_treat_as_movable; extern const unsigned long hugetlb_zero, hugetlb_infinity; extern int sysctl_hugetlb_shm_group; @@ -181,7 +179,17 @@ struct hstate { unsigned int surplus_huge_pages_node[MAX_NUMNODES]; }; -extern struct hstate default_hstate; +void __init hugetlb_add_hstate(unsigned order); +struct hstate *size_to_hstate(unsigned long size); + +#ifndef HUGE_MAX_HSTATE +#define HUGE_MAX_HSTATE 1 +#endif + +extern struct hstate hstates[HUGE_MAX_HSTATE]; +extern unsigned int default_hstate_idx; + +#define default_hstate (hstates[default_hstate_idx]) static inline struct hstate *hstate_vma(struct vm_area_struct *vma) { @@ -230,6 +238,11 @@ static inline unsigned int blocks_per_huge_page(struct hstate *h) #include +static inline struct hstate *page_hstate(struct page *page) +{ + return size_to_hstate(PAGE_SIZE << compound_order(page)); +} + #else struct hstate {}; #define hstate_file(f) NULL diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 1f7b3b76a166..1a8299d1fe59 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -959,7 +959,7 @@ static struct ctl_table vm_table[] = { #ifdef CONFIG_HUGETLB_PAGE { .procname = "nr_hugepages", - .data = &max_huge_pages, + .data = NULL, .maxlen = sizeof(unsigned long), .mode = 0644, .proc_handler = &hugetlb_sysctl_handler, @@ -985,10 +985,12 @@ static struct ctl_table vm_table[] = { { .ctl_name = CTL_UNNUMBERED, .procname = "nr_overcommit_hugepages", - .data = &sysctl_overcommit_huge_pages, - .maxlen = sizeof(sysctl_overcommit_huge_pages), + .data = NULL, + .maxlen = sizeof(unsigned long), .mode = 0644, .proc_handler = &hugetlb_overcommit_handler, + .extra1 = (void *)&hugetlb_zero, + .extra2 = (void *)&hugetlb_infinity, }, #endif { diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 0d8153e25f09..82378d44a0c5 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -22,12 +22,19 @@ #include "internal.h" const unsigned long hugetlb_zero = 0, hugetlb_infinity = ~0UL; -unsigned long max_huge_pages; -unsigned long sysctl_overcommit_huge_pages; static gfp_t htlb_alloc_mask = GFP_HIGHUSER; unsigned long hugepages_treat_as_movable; -struct hstate default_hstate; +static int max_hstate; +unsigned int default_hstate_idx; +struct hstate hstates[HUGE_MAX_HSTATE]; + +/* for command line parsing */ +static struct hstate * __initdata parsed_hstate; +static unsigned long __initdata default_hstate_max_huge_pages; + +#define for_each_hstate(h) \ + for ((h) = hstates; (h) < &hstates[max_hstate]; (h)++) /* * Protects updates to hugepage_freelists, nr_huge_pages, and free_huge_pages @@ -454,13 +461,24 @@ static void update_and_free_page(struct hstate *h, struct page *page) __free_pages(page, huge_page_order(h)); } +struct hstate *size_to_hstate(unsigned long size) +{ + struct hstate *h; + + for_each_hstate(h) { + if (huge_page_size(h) == size) + return h; + } + return NULL; +} + static void free_huge_page(struct page *page) { /* * Can't pass hstate in here because it is called from the * compound page destructor. */ - struct hstate *h = &default_hstate; + struct hstate *h = page_hstate(page); int nid = page_to_nid(page); struct address_space *mapping; @@ -887,39 +905,94 @@ static struct page *alloc_huge_page(struct vm_area_struct *vma, return page; } -static int __init hugetlb_init(void) +static void __init hugetlb_init_one_hstate(struct hstate *h) { unsigned long i; - struct hstate *h = &default_hstate; - - if (HPAGE_SHIFT == 0) - return 0; - - if (!h->order) { - h->order = HPAGE_SHIFT - PAGE_SHIFT; - h->mask = HPAGE_MASK; - } for (i = 0; i < MAX_NUMNODES; ++i) INIT_LIST_HEAD(&h->hugepage_freelists[i]); h->hugetlb_next_nid = first_node(node_online_map); - for (i = 0; i < max_huge_pages; ++i) { + for (i = 0; i < h->max_huge_pages; ++i) { if (!alloc_fresh_huge_page(h)) break; } - max_huge_pages = h->free_huge_pages = h->nr_huge_pages = i; - printk(KERN_INFO "Total HugeTLB memory allocated, %ld\n", - h->free_huge_pages); + h->max_huge_pages = h->free_huge_pages = h->nr_huge_pages = i; +} + +static void __init hugetlb_init_hstates(void) +{ + struct hstate *h; + + for_each_hstate(h) { + hugetlb_init_one_hstate(h); + } +} + +static void __init report_hugepages(void) +{ + struct hstate *h; + + for_each_hstate(h) { + printk(KERN_INFO "Total HugeTLB memory allocated, " + "%ld %dMB pages\n", + h->free_huge_pages, + 1 << (h->order + PAGE_SHIFT - 20)); + } +} + +static int __init hugetlb_init(void) +{ + BUILD_BUG_ON(HPAGE_SHIFT == 0); + + if (!size_to_hstate(HPAGE_SIZE)) { + hugetlb_add_hstate(HUGETLB_PAGE_ORDER); + parsed_hstate->max_huge_pages = default_hstate_max_huge_pages; + } + default_hstate_idx = size_to_hstate(HPAGE_SIZE) - hstates; + + hugetlb_init_hstates(); + + report_hugepages(); + return 0; } module_init(hugetlb_init); +/* Should be called on processing a hugepagesz=... option */ +void __init hugetlb_add_hstate(unsigned order) +{ + struct hstate *h; + if (size_to_hstate(PAGE_SIZE << order)) { + printk(KERN_WARNING "hugepagesz= specified twice, ignoring\n"); + return; + } + BUG_ON(max_hstate >= HUGE_MAX_HSTATE); + BUG_ON(order == 0); + h = &hstates[max_hstate++]; + h->order = order; + h->mask = ~((1ULL << (order + PAGE_SHIFT)) - 1); + hugetlb_init_one_hstate(h); + parsed_hstate = h; +} + static int __init hugetlb_setup(char *s) { - if (sscanf(s, "%lu", &max_huge_pages) <= 0) - max_huge_pages = 0; + unsigned long *mhp; + + /* + * !max_hstate means we haven't parsed a hugepagesz= parameter yet, + * so this hugepages= parameter goes to the "default hstate". + */ + if (!max_hstate) + mhp = &default_hstate_max_huge_pages; + else + mhp = &parsed_hstate->max_huge_pages; + + if (sscanf(s, "%lu", mhp) <= 0) + *mhp = 0; + return 1; } __setup("hugepages=", hugetlb_setup); @@ -950,7 +1023,7 @@ static void try_to_free_low(struct hstate *h, unsigned long count) if (PageHighMem(page)) continue; list_del(&page->lru); - update_and_free_page(page); + update_and_free_page(h, page); h->free_huge_pages--; h->free_huge_pages_node[page_to_nid(page)]--; } @@ -963,10 +1036,9 @@ static inline void try_to_free_low(struct hstate *h, unsigned long count) #endif #define persistent_huge_pages(h) (h->nr_huge_pages - h->surplus_huge_pages) -static unsigned long set_max_huge_pages(unsigned long count) +static unsigned long set_max_huge_pages(struct hstate *h, unsigned long count) { unsigned long min_count, ret; - struct hstate *h = &default_hstate; /* * Increase the pool size @@ -1037,8 +1109,19 @@ int hugetlb_sysctl_handler(struct ctl_table *table, int write, struct file *file, void __user *buffer, size_t *length, loff_t *ppos) { + struct hstate *h = &default_hstate; + unsigned long tmp; + + if (!write) + tmp = h->max_huge_pages; + + table->data = &tmp; + table->maxlen = sizeof(unsigned long); proc_doulongvec_minmax(table, write, file, buffer, length, ppos); - max_huge_pages = set_max_huge_pages(max_huge_pages); + + if (write) + h->max_huge_pages = set_max_huge_pages(h, tmp); + return 0; } @@ -1059,10 +1142,21 @@ int hugetlb_overcommit_handler(struct ctl_table *table, int write, size_t *length, loff_t *ppos) { struct hstate *h = &default_hstate; + unsigned long tmp; + + if (!write) + tmp = h->nr_overcommit_huge_pages; + + table->data = &tmp; + table->maxlen = sizeof(unsigned long); proc_doulongvec_minmax(table, write, file, buffer, length, ppos); - spin_lock(&hugetlb_lock); - h->nr_overcommit_huge_pages = sysctl_overcommit_huge_pages; - spin_unlock(&hugetlb_lock); + + if (write) { + spin_lock(&hugetlb_lock); + h->nr_overcommit_huge_pages = tmp; + spin_unlock(&hugetlb_lock); + } + return 0; } From a137e1cc6d6e7d315fef03962a2a5a113348b13b Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 23 Jul 2008 21:27:43 -0700 Subject: [PATCH 170/857] hugetlbfs: per mount huge page sizes Add the ability to configure the hugetlb hstate used on a per mount basis. - Add a new pagesize= option to the hugetlbfs mount that allows setting the page size - This option causes the mount code to find the hstate corresponding to the specified size, and sets up a pointer to the hstate in the mount's superblock. - Change the hstate accessors to use this information rather than the global_hstate they were using (requires a slight change in mm/memory.c so we don't NULL deref in the error-unmap path -- see comments). [np: take hstate out of hugetlbfs inode and vma->vm_private_data] Acked-by: Adam Litke Acked-by: Nishanth Aravamudan Signed-off-by: Andi Kleen Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/hugetlbfs/inode.c | 45 ++++++++++++++++++++++++++++++++--------- include/linux/hugetlb.h | 14 ++++++++----- mm/hugetlb.c | 16 +++------------ mm/memory.c | 18 +++++++++++++++-- 4 files changed, 64 insertions(+), 29 deletions(-) diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 516c581b5371..dbd01d262ca4 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -53,6 +53,7 @@ int sysctl_hugetlb_shm_group; enum { Opt_size, Opt_nr_inodes, Opt_mode, Opt_uid, Opt_gid, + Opt_pagesize, Opt_err, }; @@ -62,6 +63,7 @@ static match_table_t tokens = { {Opt_mode, "mode=%o"}, {Opt_uid, "uid=%u"}, {Opt_gid, "gid=%u"}, + {Opt_pagesize, "pagesize=%s"}, {Opt_err, NULL}, }; @@ -750,6 +752,8 @@ hugetlbfs_parse_options(char *options, struct hugetlbfs_config *pconfig) char *p, *rest; substring_t args[MAX_OPT_ARGS]; int option; + unsigned long long size = 0; + enum { NO_SIZE, SIZE_STD, SIZE_PERCENT } setsize = NO_SIZE; if (!options) return 0; @@ -780,17 +784,13 @@ hugetlbfs_parse_options(char *options, struct hugetlbfs_config *pconfig) break; case Opt_size: { - unsigned long long size; /* memparse() will accept a K/M/G without a digit */ if (!isdigit(*args[0].from)) goto bad_val; size = memparse(args[0].from, &rest); - if (*rest == '%') { - size <<= HPAGE_SHIFT; - size *= max_huge_pages; - do_div(size, 100); - } - pconfig->nr_blocks = (size >> HPAGE_SHIFT); + setsize = SIZE_STD; + if (*rest == '%') + setsize = SIZE_PERCENT; break; } @@ -801,6 +801,19 @@ hugetlbfs_parse_options(char *options, struct hugetlbfs_config *pconfig) pconfig->nr_inodes = memparse(args[0].from, &rest); break; + case Opt_pagesize: { + unsigned long ps; + ps = memparse(args[0].from, &rest); + pconfig->hstate = size_to_hstate(ps); + if (!pconfig->hstate) { + printk(KERN_ERR + "hugetlbfs: Unsupported page size %lu MB\n", + ps >> 20); + return -EINVAL; + } + break; + } + default: printk(KERN_ERR "hugetlbfs: Bad mount option: \"%s\"\n", p); @@ -808,6 +821,18 @@ hugetlbfs_parse_options(char *options, struct hugetlbfs_config *pconfig) break; } } + + /* Do size after hstate is set up */ + if (setsize > NO_SIZE) { + struct hstate *h = pconfig->hstate; + if (setsize == SIZE_PERCENT) { + size <<= huge_page_shift(h); + size *= h->max_huge_pages; + do_div(size, 100); + } + pconfig->nr_blocks = (size >> huge_page_shift(h)); + } + return 0; bad_val: @@ -832,6 +857,7 @@ hugetlbfs_fill_super(struct super_block *sb, void *data, int silent) config.uid = current->fsuid; config.gid = current->fsgid; config.mode = 0755; + config.hstate = &default_hstate; ret = hugetlbfs_parse_options(data, &config); if (ret) return ret; @@ -840,14 +866,15 @@ hugetlbfs_fill_super(struct super_block *sb, void *data, int silent) if (!sbinfo) return -ENOMEM; sb->s_fs_info = sbinfo; + sbinfo->hstate = config.hstate; spin_lock_init(&sbinfo->stat_lock); sbinfo->max_blocks = config.nr_blocks; sbinfo->free_blocks = config.nr_blocks; sbinfo->max_inodes = config.nr_inodes; sbinfo->free_inodes = config.nr_inodes; sb->s_maxbytes = MAX_LFS_FILESIZE; - sb->s_blocksize = HPAGE_SIZE; - sb->s_blocksize_bits = HPAGE_SHIFT; + sb->s_blocksize = huge_page_size(config.hstate); + sb->s_blocksize_bits = huge_page_shift(config.hstate); sb->s_magic = HUGETLBFS_MAGIC; sb->s_op = &hugetlbfs_ops; sb->s_time_gran = 1; diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index b75bdb4deba3..ba9263e631b9 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -100,6 +100,7 @@ struct hugetlbfs_config { umode_t mode; long nr_blocks; long nr_inodes; + struct hstate *hstate; }; struct hugetlbfs_sb_info { @@ -108,6 +109,7 @@ struct hugetlbfs_sb_info { long max_inodes; /* inodes allowed */ long free_inodes; /* inodes free */ spinlock_t stat_lock; + struct hstate *hstate; }; @@ -191,19 +193,21 @@ extern unsigned int default_hstate_idx; #define default_hstate (hstates[default_hstate_idx]) -static inline struct hstate *hstate_vma(struct vm_area_struct *vma) +static inline struct hstate *hstate_inode(struct inode *i) { - return &default_hstate; + struct hugetlbfs_sb_info *hsb; + hsb = HUGETLBFS_SB(i->i_sb); + return hsb->hstate; } static inline struct hstate *hstate_file(struct file *f) { - return &default_hstate; + return hstate_inode(f->f_dentry->d_inode); } -static inline struct hstate *hstate_inode(struct inode *i) +static inline struct hstate *hstate_vma(struct vm_area_struct *vma) { - return &default_hstate; + return hstate_file(vma->vm_file); } static inline unsigned long huge_page_size(struct hstate *h) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 82378d44a0c5..4cf7a90e9140 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -1439,19 +1439,9 @@ void __unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start, void unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start, unsigned long end, struct page *ref_page) { - /* - * It is undesirable to test vma->vm_file as it should be non-null - * for valid hugetlb area. However, vm_file will be NULL in the error - * cleanup path of do_mmap_pgoff. When hugetlbfs ->mmap method fails, - * do_mmap_pgoff() nullifies vma->vm_file before calling this function - * to clean up. Since no pte has actually been setup, it is safe to - * do nothing in this case. - */ - if (vma->vm_file) { - spin_lock(&vma->vm_file->f_mapping->i_mmap_lock); - __unmap_hugepage_range(vma, start, end, ref_page); - spin_unlock(&vma->vm_file->f_mapping->i_mmap_lock); - } + spin_lock(&vma->vm_file->f_mapping->i_mmap_lock); + __unmap_hugepage_range(vma, start, end, ref_page); + spin_unlock(&vma->vm_file->f_mapping->i_mmap_lock); } /* diff --git a/mm/memory.c b/mm/memory.c index c1c1d6d8c22b..02fc6b1047b0 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -901,9 +901,23 @@ unsigned long unmap_vmas(struct mmu_gather **tlbp, } if (unlikely(is_vm_hugetlb_page(vma))) { - unmap_hugepage_range(vma, start, end, NULL); - zap_work -= (end - start) / + /* + * It is undesirable to test vma->vm_file as it + * should be non-null for valid hugetlb area. + * However, vm_file will be NULL in the error + * cleanup path of do_mmap_pgoff. When + * hugetlbfs ->mmap method fails, + * do_mmap_pgoff() nullifies vma->vm_file + * before calling this function to clean up. + * Since no pte has actually been setup, it is + * safe to do nothing in this case. + */ + if (vma->vm_file) { + unmap_hugepage_range(vma, start, end, NULL); + zap_work -= (end - start) / pages_per_huge_page(hstate_vma(vma)); + } + start = end; } else start = unmap_page_range(*tlbp, vma, From a3437870160cf2caaac6bdd76c7377a5a4145a8c Mon Sep 17 00:00:00 2001 From: Nishanth Aravamudan Date: Wed, 23 Jul 2008 21:27:44 -0700 Subject: [PATCH 171/857] hugetlb: new sysfs interface Provide new hugepages user APIs that are more suited to multiple hstates in sysfs. There is a new directory, /sys/kernel/hugepages. Underneath that directory there will be a directory per-supported hugepage size, e.g.: /sys/kernel/hugepages/hugepages-64kB /sys/kernel/hugepages/hugepages-16384kB /sys/kernel/hugepages/hugepages-16777216kB corresponding to 64k, 16m and 16g respectively. Within each hugepages-size directory there are a number of files, corresponding to the tracked counters in the hstate, e.g.: /sys/kernel/hugepages/hugepages-64/nr_hugepages /sys/kernel/hugepages/hugepages-64/nr_overcommit_hugepages /sys/kernel/hugepages/hugepages-64/free_hugepages /sys/kernel/hugepages/hugepages-64/resv_hugepages /sys/kernel/hugepages/hugepages-64/surplus_hugepages Of these files, the first two are read-write and the latter three are read-only. The size of the hugepage being manipulated is trivially deducible from the enclosing directory and is always expressed in kB (to match meminfo). [dave@linux.vnet.ibm.com: fix build] [nacc@us.ibm.com: hugetlb: hang off of /sys/kernel/mm rather than /sys/kernel] [nacc@us.ibm.com: hugetlb: remove CONFIG_SYSFS dependency] Acked-by: Greg Kroah-Hartman Signed-off-by: Nishanth Aravamudan Signed-off-by: Nick Piggin Cc: Dave Hansen Signed-off-by: Nishanth Aravamudan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- .../ABI/testing/sysfs-kernel-mm-hugepages | 15 + Documentation/vm/hugetlbpage.txt | 23 ++ include/linux/hugetlb.h | 2 + mm/hugetlb.c | 288 ++++++++++++++---- 4 files changed, 262 insertions(+), 66 deletions(-) create mode 100644 Documentation/ABI/testing/sysfs-kernel-mm-hugepages diff --git a/Documentation/ABI/testing/sysfs-kernel-mm-hugepages b/Documentation/ABI/testing/sysfs-kernel-mm-hugepages new file mode 100644 index 000000000000..e21c00571cf4 --- /dev/null +++ b/Documentation/ABI/testing/sysfs-kernel-mm-hugepages @@ -0,0 +1,15 @@ +What: /sys/kernel/mm/hugepages/ +Date: June 2008 +Contact: Nishanth Aravamudan , hugetlb maintainers +Description: + /sys/kernel/mm/hugepages/ contains a number of subdirectories + of the form hugepages-kB, where is the page size + of the hugepages supported by the kernel/CPU combination. + + Under these directories are a number of files: + nr_hugepages + nr_overcommit_hugepages + free_hugepages + surplus_hugepages + resv_hugepages + See Documentation/vm/hugetlbpage.txt for details. diff --git a/Documentation/vm/hugetlbpage.txt b/Documentation/vm/hugetlbpage.txt index 3102b81bef88..8a5b5763f0fe 100644 --- a/Documentation/vm/hugetlbpage.txt +++ b/Documentation/vm/hugetlbpage.txt @@ -95,6 +95,29 @@ this condition holds, however, no more surplus huge pages will be allowed on the system until one of the two sysctls are increased sufficiently, or the surplus huge pages go out of use and are freed. +With support for multiple hugepage pools at run-time available, much of +the hugepage userspace interface has been duplicated in sysfs. The above +information applies to the default hugepage size (which will be +controlled by the proc interfaces for backwards compatibility). The root +hugepage control directory is + + /sys/kernel/mm/hugepages + +For each hugepage size supported by the running kernel, a subdirectory +will exist, of the form + + hugepages-${size}kB + +Inside each of these directories, the same set of files will exist: + + nr_hugepages + nr_overcommit_hugepages + free_hugepages + resv_hugepages + surplus_hugepages + +which function as described above for the default hugepage-sized case. + If the user applications are going to request hugepages using mmap system call, then it is required that system administrator mount a file system of type hugetlbfs: diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index ba9263e631b9..58c0de32e7f0 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -164,6 +164,7 @@ unsigned long hugetlb_get_unmapped_area(struct file *file, unsigned long addr, #ifdef CONFIG_HUGETLB_PAGE +#define HSTATE_NAME_LEN 32 /* Defines one hugetlb page size */ struct hstate { int hugetlb_next_nid; @@ -179,6 +180,7 @@ struct hstate { unsigned int nr_huge_pages_node[MAX_NUMNODES]; unsigned int free_huge_pages_node[MAX_NUMNODES]; unsigned int surplus_huge_pages_node[MAX_NUMNODES]; + char name[HSTATE_NAME_LEN]; }; void __init hugetlb_add_hstate(unsigned order); diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 4cf7a90e9140..bb49ce5d0067 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -14,6 +14,7 @@ #include #include #include +#include #include #include @@ -942,72 +943,6 @@ static void __init report_hugepages(void) } } -static int __init hugetlb_init(void) -{ - BUILD_BUG_ON(HPAGE_SHIFT == 0); - - if (!size_to_hstate(HPAGE_SIZE)) { - hugetlb_add_hstate(HUGETLB_PAGE_ORDER); - parsed_hstate->max_huge_pages = default_hstate_max_huge_pages; - } - default_hstate_idx = size_to_hstate(HPAGE_SIZE) - hstates; - - hugetlb_init_hstates(); - - report_hugepages(); - - return 0; -} -module_init(hugetlb_init); - -/* Should be called on processing a hugepagesz=... option */ -void __init hugetlb_add_hstate(unsigned order) -{ - struct hstate *h; - if (size_to_hstate(PAGE_SIZE << order)) { - printk(KERN_WARNING "hugepagesz= specified twice, ignoring\n"); - return; - } - BUG_ON(max_hstate >= HUGE_MAX_HSTATE); - BUG_ON(order == 0); - h = &hstates[max_hstate++]; - h->order = order; - h->mask = ~((1ULL << (order + PAGE_SHIFT)) - 1); - hugetlb_init_one_hstate(h); - parsed_hstate = h; -} - -static int __init hugetlb_setup(char *s) -{ - unsigned long *mhp; - - /* - * !max_hstate means we haven't parsed a hugepagesz= parameter yet, - * so this hugepages= parameter goes to the "default hstate". - */ - if (!max_hstate) - mhp = &default_hstate_max_huge_pages; - else - mhp = &parsed_hstate->max_huge_pages; - - if (sscanf(s, "%lu", mhp) <= 0) - *mhp = 0; - - return 1; -} -__setup("hugepages=", hugetlb_setup); - -static unsigned int cpuset_mems_nr(unsigned int *array) -{ - int node; - unsigned int nr = 0; - - for_each_node_mask(node, cpuset_current_mems_allowed) - nr += array[node]; - - return nr; -} - #ifdef CONFIG_SYSCTL #ifdef CONFIG_HIGHMEM static void try_to_free_low(struct hstate *h, unsigned long count) @@ -1105,6 +1040,227 @@ static unsigned long set_max_huge_pages(struct hstate *h, unsigned long count) return ret; } +#define HSTATE_ATTR_RO(_name) \ + static struct kobj_attribute _name##_attr = __ATTR_RO(_name) + +#define HSTATE_ATTR(_name) \ + static struct kobj_attribute _name##_attr = \ + __ATTR(_name, 0644, _name##_show, _name##_store) + +static struct kobject *hugepages_kobj; +static struct kobject *hstate_kobjs[HUGE_MAX_HSTATE]; + +static struct hstate *kobj_to_hstate(struct kobject *kobj) +{ + int i; + for (i = 0; i < HUGE_MAX_HSTATE; i++) + if (hstate_kobjs[i] == kobj) + return &hstates[i]; + BUG(); + return NULL; +} + +static ssize_t nr_hugepages_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + struct hstate *h = kobj_to_hstate(kobj); + return sprintf(buf, "%lu\n", h->nr_huge_pages); +} +static ssize_t nr_hugepages_store(struct kobject *kobj, + struct kobj_attribute *attr, const char *buf, size_t count) +{ + int err; + unsigned long input; + struct hstate *h = kobj_to_hstate(kobj); + + err = strict_strtoul(buf, 10, &input); + if (err) + return 0; + + h->max_huge_pages = set_max_huge_pages(h, input); + + return count; +} +HSTATE_ATTR(nr_hugepages); + +static ssize_t nr_overcommit_hugepages_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + struct hstate *h = kobj_to_hstate(kobj); + return sprintf(buf, "%lu\n", h->nr_overcommit_huge_pages); +} +static ssize_t nr_overcommit_hugepages_store(struct kobject *kobj, + struct kobj_attribute *attr, const char *buf, size_t count) +{ + int err; + unsigned long input; + struct hstate *h = kobj_to_hstate(kobj); + + err = strict_strtoul(buf, 10, &input); + if (err) + return 0; + + spin_lock(&hugetlb_lock); + h->nr_overcommit_huge_pages = input; + spin_unlock(&hugetlb_lock); + + return count; +} +HSTATE_ATTR(nr_overcommit_hugepages); + +static ssize_t free_hugepages_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + struct hstate *h = kobj_to_hstate(kobj); + return sprintf(buf, "%lu\n", h->free_huge_pages); +} +HSTATE_ATTR_RO(free_hugepages); + +static ssize_t resv_hugepages_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + struct hstate *h = kobj_to_hstate(kobj); + return sprintf(buf, "%lu\n", h->resv_huge_pages); +} +HSTATE_ATTR_RO(resv_hugepages); + +static ssize_t surplus_hugepages_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + struct hstate *h = kobj_to_hstate(kobj); + return sprintf(buf, "%lu\n", h->surplus_huge_pages); +} +HSTATE_ATTR_RO(surplus_hugepages); + +static struct attribute *hstate_attrs[] = { + &nr_hugepages_attr.attr, + &nr_overcommit_hugepages_attr.attr, + &free_hugepages_attr.attr, + &resv_hugepages_attr.attr, + &surplus_hugepages_attr.attr, + NULL, +}; + +static struct attribute_group hstate_attr_group = { + .attrs = hstate_attrs, +}; + +static int __init hugetlb_sysfs_add_hstate(struct hstate *h) +{ + int retval; + + hstate_kobjs[h - hstates] = kobject_create_and_add(h->name, + hugepages_kobj); + if (!hstate_kobjs[h - hstates]) + return -ENOMEM; + + retval = sysfs_create_group(hstate_kobjs[h - hstates], + &hstate_attr_group); + if (retval) + kobject_put(hstate_kobjs[h - hstates]); + + return retval; +} + +static void __init hugetlb_sysfs_init(void) +{ + struct hstate *h; + int err; + + hugepages_kobj = kobject_create_and_add("hugepages", mm_kobj); + if (!hugepages_kobj) + return; + + for_each_hstate(h) { + err = hugetlb_sysfs_add_hstate(h); + if (err) + printk(KERN_ERR "Hugetlb: Unable to add hstate %s", + h->name); + } +} + +static void __exit hugetlb_exit(void) +{ + struct hstate *h; + + for_each_hstate(h) { + kobject_put(hstate_kobjs[h - hstates]); + } + + kobject_put(hugepages_kobj); +} +module_exit(hugetlb_exit); + +static int __init hugetlb_init(void) +{ + BUILD_BUG_ON(HPAGE_SHIFT == 0); + + if (!size_to_hstate(HPAGE_SIZE)) { + hugetlb_add_hstate(HUGETLB_PAGE_ORDER); + parsed_hstate->max_huge_pages = default_hstate_max_huge_pages; + } + default_hstate_idx = size_to_hstate(HPAGE_SIZE) - hstates; + + hugetlb_init_hstates(); + + report_hugepages(); + + hugetlb_sysfs_init(); + + return 0; +} +module_init(hugetlb_init); + +/* Should be called on processing a hugepagesz=... option */ +void __init hugetlb_add_hstate(unsigned order) +{ + struct hstate *h; + if (size_to_hstate(PAGE_SIZE << order)) { + printk(KERN_WARNING "hugepagesz= specified twice, ignoring\n"); + return; + } + BUG_ON(max_hstate >= HUGE_MAX_HSTATE); + BUG_ON(order == 0); + h = &hstates[max_hstate++]; + h->order = order; + h->mask = ~((1ULL << (order + PAGE_SHIFT)) - 1); + snprintf(h->name, HSTATE_NAME_LEN, "hugepages-%lukB", + huge_page_size(h)/1024); + hugetlb_init_one_hstate(h); + parsed_hstate = h; +} + +static int __init hugetlb_setup(char *s) +{ + unsigned long *mhp; + + /* + * !max_hstate means we haven't parsed a hugepagesz= parameter yet, + * so this hugepages= parameter goes to the "default hstate". + */ + if (!max_hstate) + mhp = &default_hstate_max_huge_pages; + else + mhp = &parsed_hstate->max_huge_pages; + + if (sscanf(s, "%lu", mhp) <= 0) + *mhp = 0; + + return 1; +} +__setup("hugepages=", hugetlb_setup); + +static unsigned int cpuset_mems_nr(unsigned int *array) +{ + int node; + unsigned int nr = 0; + + for_each_node_mask(node, cpuset_current_mems_allowed) + nr += array[node]; + + return nr; +} + int hugetlb_sysctl_handler(struct ctl_table *table, int write, struct file *file, void __user *buffer, size_t *length, loff_t *ppos) From 5ced66c901f1cf0b684feb15c2cd8b126e263d07 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 23 Jul 2008 21:27:45 -0700 Subject: [PATCH 172/857] hugetlb: abstract numa round robin selection Need this as a separate function for a future patch. No behaviour change. Acked-by: Adam Litke Acked-by: Nishanth Aravamudan Signed-off-by: Andi Kleen Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/hugetlb.c | 37 ++++++++++++++++++++++--------------- 1 file changed, 22 insertions(+), 15 deletions(-) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index bb49ce5d0067..5e620e25cf08 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -565,6 +565,27 @@ static struct page *alloc_fresh_huge_page_node(struct hstate *h, int nid) return page; } +/* + * Use a helper variable to find the next node and then + * copy it back to hugetlb_next_nid afterwards: + * otherwise there's a window in which a racer might + * pass invalid nid MAX_NUMNODES to alloc_pages_node. + * But we don't need to use a spin_lock here: it really + * doesn't matter if occasionally a racer chooses the + * same nid as we do. Move nid forward in the mask even + * if we just successfully allocated a hugepage so that + * the next caller gets hugepages on the next node. + */ +static int hstate_next_node(struct hstate *h) +{ + int next_nid; + next_nid = next_node(h->hugetlb_next_nid, node_online_map); + if (next_nid == MAX_NUMNODES) + next_nid = first_node(node_online_map); + h->hugetlb_next_nid = next_nid; + return next_nid; +} + static int alloc_fresh_huge_page(struct hstate *h) { struct page *page; @@ -578,21 +599,7 @@ static int alloc_fresh_huge_page(struct hstate *h) page = alloc_fresh_huge_page_node(h, h->hugetlb_next_nid); if (page) ret = 1; - /* - * Use a helper variable to find the next node and then - * copy it back to hugetlb_next_nid afterwards: - * otherwise there's a window in which a racer might - * pass invalid nid MAX_NUMNODES to alloc_pages_node. - * But we don't need to use a spin_lock here: it really - * doesn't matter if occasionally a racer chooses the - * same nid as we do. Move nid forward in the mask even - * if we just successfully allocated a hugepage so that - * the next caller gets hugepages on the next node. - */ - next_nid = next_node(h->hugetlb_next_nid, node_online_map); - if (next_nid == MAX_NUMNODES) - next_nid = first_node(node_online_map); - h->hugetlb_next_nid = next_nid; + next_nid = hstate_next_node(h); } while (!page && h->hugetlb_next_nid != start_nid); if (ret) From b54bbf7b81170f03597c17dd0b559e3006bc9868 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 23 Jul 2008 21:27:45 -0700 Subject: [PATCH 173/857] mm: introduce non panic alloc_bootmem Straight forward variant of the existing __alloc_bootmem_node, only subsequent patch when allocating giant hugepages at boot -- don't want to panic if we can't allocate as many as the user asked for. Signed-off-by: Andi Kleen Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/bootmem.h | 4 ++++ mm/bootmem.c | 12 ++++++++++++ 2 files changed, 16 insertions(+) diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h index dd8fee6c46d9..f352c5f125b4 100644 --- a/include/linux/bootmem.h +++ b/include/linux/bootmem.h @@ -89,6 +89,10 @@ extern void *__alloc_bootmem_node(pg_data_t *pgdat, unsigned long size, unsigned long align, unsigned long goal); +extern void *__alloc_bootmem_node_nopanic(pg_data_t *pgdat, + unsigned long size, + unsigned long align, + unsigned long goal); extern unsigned long init_bootmem_node(pg_data_t *pgdat, unsigned long freepfn, unsigned long startpfn, diff --git a/mm/bootmem.c b/mm/bootmem.c index 4bc6ae2fbaa3..9ac972535fff 100644 --- a/mm/bootmem.c +++ b/mm/bootmem.c @@ -578,6 +578,18 @@ void * __init alloc_bootmem_section(unsigned long size, } #endif +void * __init __alloc_bootmem_node_nopanic(pg_data_t *pgdat, unsigned long size, + unsigned long align, unsigned long goal) +{ + void *ptr; + + ptr = alloc_bootmem_core(pgdat->bdata, size, align, goal, 0); + if (ptr) + return ptr; + + return __alloc_bootmem_nopanic(size, align, goal); +} + #ifndef ARCH_LOW_ADDRESS_LIMIT #define ARCH_LOW_ADDRESS_LIMIT 0xffffffffUL #endif From 01ad1c0827db5b3695c53e296dbb2c1da16a0911 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 23 Jul 2008 21:27:46 -0700 Subject: [PATCH 174/857] mm: export prep_compound_page to mm hugetlb will need to get compound pages from bootmem to handle the case of them being greater than or equal to MAX_ORDER. Export the constructor function needed for this. Acked-by: Adam Litke Signed-off-by: Andi Kleen Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/internal.h | 2 ++ mm/page_alloc.c | 2 +- 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/mm/internal.h b/mm/internal.h index 858ad01864dc..1f43f7416972 100644 --- a/mm/internal.h +++ b/mm/internal.h @@ -16,6 +16,8 @@ void free_pgtables(struct mmu_gather *tlb, struct vm_area_struct *start_vma, unsigned long floor, unsigned long ceiling); +extern void prep_compound_page(struct page *page, unsigned long order); + static inline void set_page_count(struct page *page, int v) { atomic_set(&page->_count, v); diff --git a/mm/page_alloc.c b/mm/page_alloc.c index e43aae135b38..eaa86671ebbd 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -264,7 +264,7 @@ static void free_compound_page(struct page *page) __free_pages_ok(page, compound_order(page)); } -static void prep_compound_page(struct page *page, unsigned long order) +void prep_compound_page(struct page *page, unsigned long order) { int i; int nr_pages = 1 << order; From aa888a74977a8f2120ae9332376e179c39a6b07d Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 23 Jul 2008 21:27:47 -0700 Subject: [PATCH 175/857] hugetlb: support larger than MAX_ORDER This is needed on x86-64 to handle GB pages in hugetlbfs, because it is not practical to enlarge MAX_ORDER to 1GB. Instead the 1GB pages are only allocated at boot using the bootmem allocator using the hugepages=... option. These 1G bootmem pages are never freed. In theory it would be possible to implement that with some complications, but since it would be a one-way street (>= MAX_ORDER pages cannot be allocated later) I decided not to currently. The >= MAX_ORDER code is not ifdef'ed per architecture. It is not very big and the ifdef uglyness seemed not be worth it. Known problems: /proc/meminfo and "free" do not display the memory allocated for gb pages in "Total". This is a little confusing for the user. Acked-by: Andrew Hastings Signed-off-by: Andi Kleen Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/hugetlb.c | 83 ++++++++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 81 insertions(+), 2 deletions(-) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 5e620e25cf08..1a6fe87555b2 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -14,6 +14,7 @@ #include #include #include +#include #include #include @@ -489,7 +490,7 @@ static void free_huge_page(struct page *page) INIT_LIST_HEAD(&page->lru); spin_lock(&hugetlb_lock); - if (h->surplus_huge_pages_node[nid]) { + if (h->surplus_huge_pages_node[nid] && huge_page_order(h) < MAX_ORDER) { update_and_free_page(h, page); h->surplus_huge_pages--; h->surplus_huge_pages_node[nid]--; @@ -550,6 +551,9 @@ static struct page *alloc_fresh_huge_page_node(struct hstate *h, int nid) { struct page *page; + if (h->order >= MAX_ORDER) + return NULL; + page = alloc_pages_node(nid, htlb_alloc_mask|__GFP_COMP|__GFP_THISNODE| __GFP_REPEAT|__GFP_NOWARN, @@ -616,6 +620,9 @@ static struct page *alloc_buddy_huge_page(struct hstate *h, struct page *page; unsigned int nid; + if (h->order >= MAX_ORDER) + return NULL; + /* * Assume we will successfully allocate the surplus page to * prevent racing processes from causing the surplus to exceed @@ -792,6 +799,10 @@ static void return_unused_surplus_pages(struct hstate *h, /* Uncommit the reservation */ h->resv_huge_pages -= unused_resv_pages; + /* Cannot return gigantic pages currently */ + if (h->order >= MAX_ORDER) + return; + nr_pages = min(unused_resv_pages, h->surplus_huge_pages); while (remaining_iterations-- && nr_pages) { @@ -913,6 +924,63 @@ static struct page *alloc_huge_page(struct vm_area_struct *vma, return page; } +static __initdata LIST_HEAD(huge_boot_pages); + +struct huge_bootmem_page { + struct list_head list; + struct hstate *hstate; +}; + +static int __init alloc_bootmem_huge_page(struct hstate *h) +{ + struct huge_bootmem_page *m; + int nr_nodes = nodes_weight(node_online_map); + + while (nr_nodes) { + void *addr; + + addr = __alloc_bootmem_node_nopanic( + NODE_DATA(h->hugetlb_next_nid), + huge_page_size(h), huge_page_size(h), 0); + + if (addr) { + /* + * Use the beginning of the huge page to store the + * huge_bootmem_page struct (until gather_bootmem + * puts them into the mem_map). + */ + m = addr; + if (m) + goto found; + } + hstate_next_node(h); + nr_nodes--; + } + return 0; + +found: + BUG_ON((unsigned long)virt_to_phys(m) & (huge_page_size(h) - 1)); + /* Put them into a private list first because mem_map is not up yet */ + list_add(&m->list, &huge_boot_pages); + m->hstate = h; + return 1; +} + +/* Put bootmem huge pages into the standard lists after mem_map is up */ +static void __init gather_bootmem_prealloc(void) +{ + struct huge_bootmem_page *m; + + list_for_each_entry(m, &huge_boot_pages, list) { + struct page *page = virt_to_page(m); + struct hstate *h = m->hstate; + __ClearPageReserved(page); + WARN_ON(page_count(page) != 1); + prep_compound_page(page, h->order); + prep_new_huge_page(h, page, page_to_nid(page)); + } +} + static void __init hugetlb_init_one_hstate(struct hstate *h) { unsigned long i; @@ -923,7 +991,10 @@ static void __init hugetlb_init_one_hstate(struct hstate *h) h->hugetlb_next_nid = first_node(node_online_map); for (i = 0; i < h->max_huge_pages; ++i) { - if (!alloc_fresh_huge_page(h)) + if (h->order >= MAX_ORDER) { + if (!alloc_bootmem_huge_page(h)) + break; + } else if (!alloc_fresh_huge_page(h)) break; } h->max_huge_pages = h->free_huge_pages = h->nr_huge_pages = i; @@ -956,6 +1027,9 @@ static void try_to_free_low(struct hstate *h, unsigned long count) { int i; + if (h->order >= MAX_ORDER) + return; + for (i = 0; i < MAX_NUMNODES; ++i) { struct page *page, *next; struct list_head *freel = &h->hugepage_freelists[i]; @@ -982,6 +1056,9 @@ static unsigned long set_max_huge_pages(struct hstate *h, unsigned long count) { unsigned long min_count, ret; + if (h->order >= MAX_ORDER) + return h->max_huge_pages; + /* * Increase the pool size * First take pages out of surplus state. Then make up the @@ -1210,6 +1287,8 @@ static int __init hugetlb_init(void) hugetlb_init_hstates(); + gather_bootmem_prealloc(); + report_hugepages(); hugetlb_sysfs_init(); From 8faa8b077b2cdc4e4646842fe50b07840955a013 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 23 Jul 2008 21:27:48 -0700 Subject: [PATCH 176/857] hugetlb: support boot allocate different sizes Make some infrastructure changes to allow boot-time allocation of different hugepage page sizes. - move all basic hstate initialisation into hugetlb_add_hstate - create a new function hugetlb_hstate_alloc_pages() to do the actual initial page allocations. Call this function early in order to allocate giant pages from bootmem. - Check for multiple hugepages= parameters Acked-by: Adam Litke Acked-by: Nishanth Aravamudan Acked-by: Andrew Hastings Signed-off-by: Andi Kleen Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/hugetlb.c | 39 ++++++++++++++++++++++++++++++--------- 1 file changed, 30 insertions(+), 9 deletions(-) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 1a6fe87555b2..243a8684d180 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -981,15 +981,10 @@ static void __init gather_bootmem_prealloc(void) } } -static void __init hugetlb_init_one_hstate(struct hstate *h) +static void __init hugetlb_hstate_alloc_pages(struct hstate *h) { unsigned long i; - for (i = 0; i < MAX_NUMNODES; ++i) - INIT_LIST_HEAD(&h->hugepage_freelists[i]); - - h->hugetlb_next_nid = first_node(node_online_map); - for (i = 0; i < h->max_huge_pages; ++i) { if (h->order >= MAX_ORDER) { if (!alloc_bootmem_huge_page(h)) @@ -997,7 +992,7 @@ static void __init hugetlb_init_one_hstate(struct hstate *h) } else if (!alloc_fresh_huge_page(h)) break; } - h->max_huge_pages = h->free_huge_pages = h->nr_huge_pages = i; + h->max_huge_pages = i; } static void __init hugetlb_init_hstates(void) @@ -1005,7 +1000,9 @@ static void __init hugetlb_init_hstates(void) struct hstate *h; for_each_hstate(h) { - hugetlb_init_one_hstate(h); + /* oversize hugepages were init'ed in early boot */ + if (h->order < MAX_ORDER) + hugetlb_hstate_alloc_pages(h); } } @@ -1301,6 +1298,8 @@ module_init(hugetlb_init); void __init hugetlb_add_hstate(unsigned order) { struct hstate *h; + unsigned long i; + if (size_to_hstate(PAGE_SIZE << order)) { printk(KERN_WARNING "hugepagesz= specified twice, ignoring\n"); return; @@ -1310,15 +1309,21 @@ void __init hugetlb_add_hstate(unsigned order) h = &hstates[max_hstate++]; h->order = order; h->mask = ~((1ULL << (order + PAGE_SHIFT)) - 1); + h->nr_huge_pages = 0; + h->free_huge_pages = 0; + for (i = 0; i < MAX_NUMNODES; ++i) + INIT_LIST_HEAD(&h->hugepage_freelists[i]); + h->hugetlb_next_nid = first_node(node_online_map); snprintf(h->name, HSTATE_NAME_LEN, "hugepages-%lukB", huge_page_size(h)/1024); - hugetlb_init_one_hstate(h); + parsed_hstate = h; } static int __init hugetlb_setup(char *s) { unsigned long *mhp; + static unsigned long *last_mhp; /* * !max_hstate means we haven't parsed a hugepagesz= parameter yet, @@ -1329,9 +1334,25 @@ static int __init hugetlb_setup(char *s) else mhp = &parsed_hstate->max_huge_pages; + if (mhp == last_mhp) { + printk(KERN_WARNING "hugepages= specified twice without " + "interleaving hugepagesz=, ignoring\n"); + return 1; + } + if (sscanf(s, "%lu", mhp) <= 0) *mhp = 0; + /* + * Global state is always initialized later in hugetlb_init. + * But we need to allocate >= MAX_ORDER hstates here early to still + * use the bootmem allocator. + */ + if (max_hstate && parsed_hstate->order >= MAX_ORDER) + hugetlb_hstate_alloc_pages(parsed_hstate); + + last_mhp = mhp; + return 1; } __setup("hugepages=", hugetlb_setup); From 4abd32dbab201c3ced0b0af12accea77cd9eeffc Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 23 Jul 2008 21:27:49 -0700 Subject: [PATCH 177/857] hugetlb: printk cleanup - Reword sentence to clarify meaning with multiple options - Add support for using GB prefixes for the page size - Add extra printk to delayed > MAX_ORDER allocation code Acked-by: Adam Litke Acked-by: Nishanth Aravamudan Signed-off-by: Andi Kleen Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/hugetlb.c | 20 ++++++++++++++++---- 1 file changed, 16 insertions(+), 4 deletions(-) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 243a8684d180..0c74c14dd2f7 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -1006,15 +1006,27 @@ static void __init hugetlb_init_hstates(void) } } +static char * __init memfmt(char *buf, unsigned long n) +{ + if (n >= (1UL << 30)) + sprintf(buf, "%lu GB", n >> 30); + else if (n >= (1UL << 20)) + sprintf(buf, "%lu MB", n >> 20); + else + sprintf(buf, "%lu KB", n >> 10); + return buf; +} + static void __init report_hugepages(void) { struct hstate *h; for_each_hstate(h) { - printk(KERN_INFO "Total HugeTLB memory allocated, " - "%ld %dMB pages\n", - h->free_huge_pages, - 1 << (h->order + PAGE_SHIFT - 20)); + char buf[32]; + printk(KERN_INFO "HugeTLB registered %s page size, " + "pre-allocated %ld pages\n", + memfmt(buf, huge_page_size(h)), + h->free_huge_pages); } } From ceb868796181dc95ea01a110e123afd391639873 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 23 Jul 2008 21:27:50 -0700 Subject: [PATCH 178/857] hugetlb: introduce pud_huge Straight forward extensions for huge pages located in the PUD instead of PMDs. Signed-off-by: Andi Kleen Signed-off-by: Nick Piggin Cc: Martin Schwidefsky Cc: Heiko Carstens Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/mm/hugetlbpage.c | 6 ++++++ arch/powerpc/mm/hugetlbpage.c | 5 +++++ arch/s390/mm/hugetlbpage.c | 5 +++++ arch/sh/mm/hugetlbpage.c | 5 +++++ arch/sparc64/mm/hugetlbpage.c | 5 +++++ arch/x86/mm/hugetlbpage.c | 25 ++++++++++++++++++++++++- include/linux/hugetlb.h | 5 +++++ mm/hugetlb.c | 9 +++++++++ mm/memory.c | 15 +++++++++++---- 9 files changed, 75 insertions(+), 5 deletions(-) diff --git a/arch/ia64/mm/hugetlbpage.c b/arch/ia64/mm/hugetlbpage.c index 6170f097d255..c45fc7f5a979 100644 --- a/arch/ia64/mm/hugetlbpage.c +++ b/arch/ia64/mm/hugetlbpage.c @@ -107,6 +107,12 @@ int pmd_huge(pmd_t pmd) { return 0; } + +int pud_huge(pud_t pud) +{ + return 0; +} + struct page * follow_huge_pmd(struct mm_struct *mm, unsigned long address, pmd_t *pmd, int write) { diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c index c94dc71af989..63db7adce717 100644 --- a/arch/powerpc/mm/hugetlbpage.c +++ b/arch/powerpc/mm/hugetlbpage.c @@ -369,6 +369,11 @@ int pmd_huge(pmd_t pmd) return 0; } +int pud_huge(pud_t pud) +{ + return 0; +} + struct page * follow_huge_pmd(struct mm_struct *mm, unsigned long address, pmd_t *pmd, int write) diff --git a/arch/s390/mm/hugetlbpage.c b/arch/s390/mm/hugetlbpage.c index 9162dc84f77f..f28c43d2f61d 100644 --- a/arch/s390/mm/hugetlbpage.c +++ b/arch/s390/mm/hugetlbpage.c @@ -120,6 +120,11 @@ int pmd_huge(pmd_t pmd) return !!(pmd_val(pmd) & _SEGMENT_ENTRY_LARGE); } +int pud_huge(pud_t pud) +{ + return 0; +} + struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address, pmd_t *pmdp, int write) { diff --git a/arch/sh/mm/hugetlbpage.c b/arch/sh/mm/hugetlbpage.c index 2f9dbe0ef4ac..9304117039c4 100644 --- a/arch/sh/mm/hugetlbpage.c +++ b/arch/sh/mm/hugetlbpage.c @@ -79,6 +79,11 @@ int pmd_huge(pmd_t pmd) return 0; } +int pud_huge(pud_t pud) +{ + return 0; +} + struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address, pmd_t *pmd, int write) { diff --git a/arch/sparc64/mm/hugetlbpage.c b/arch/sparc64/mm/hugetlbpage.c index 1307b23f6a76..f27d10369e0c 100644 --- a/arch/sparc64/mm/hugetlbpage.c +++ b/arch/sparc64/mm/hugetlbpage.c @@ -295,6 +295,11 @@ int pmd_huge(pmd_t pmd) return 0; } +int pud_huge(pud_t pud) +{ + return 0; +} + struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address, pmd_t *pmd, int write) { diff --git a/arch/x86/mm/hugetlbpage.c b/arch/x86/mm/hugetlbpage.c index 52476fde8996..a4789e87a315 100644 --- a/arch/x86/mm/hugetlbpage.c +++ b/arch/x86/mm/hugetlbpage.c @@ -189,6 +189,11 @@ int pmd_huge(pmd_t pmd) return 0; } +int pud_huge(pud_t pud) +{ + return 0; +} + struct page * follow_huge_pmd(struct mm_struct *mm, unsigned long address, pmd_t *pmd, int write) @@ -209,6 +214,11 @@ int pmd_huge(pmd_t pmd) return !!(pmd_val(pmd) & _PAGE_PSE); } +int pud_huge(pud_t pud) +{ + return 0; +} + struct page * follow_huge_pmd(struct mm_struct *mm, unsigned long address, pmd_t *pmd, int write) @@ -217,9 +227,22 @@ follow_huge_pmd(struct mm_struct *mm, unsigned long address, page = pte_page(*(pte_t *)pmd); if (page) - page += ((address & ~HPAGE_MASK) >> PAGE_SHIFT); + page += ((address & ~PMD_MASK) >> PAGE_SHIFT); return page; } + +struct page * +follow_huge_pud(struct mm_struct *mm, unsigned long address, + pud_t *pud, int write) +{ + struct page *page; + + page = pte_page(*(pte_t *)pud); + if (page) + page += ((address & ~PUD_MASK) >> PAGE_SHIFT); + return page; +} + #endif /* x86_64 also uses this file */ diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index 58c0de32e7f0..b2c17f62cacb 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -50,7 +50,10 @@ struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address, int write); struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address, pmd_t *pmd, int write); +struct page *follow_huge_pud(struct mm_struct *mm, unsigned long address, + pud_t *pud, int write); int pmd_huge(pmd_t pmd); +int pud_huge(pud_t pmd); void hugetlb_change_protection(struct vm_area_struct *vma, unsigned long address, unsigned long end, pgprot_t newprot); @@ -78,8 +81,10 @@ static inline unsigned long hugetlb_total_pages(void) #define hugetlb_report_meminfo(buf) 0 #define hugetlb_report_node_meminfo(n, buf) 0 #define follow_huge_pmd(mm, addr, pmd, write) NULL +#define follow_huge_pud(mm, addr, pud, write) NULL #define prepare_hugepage_range(file, addr, len) (-EINVAL) #define pmd_huge(x) 0 +#define pud_huge(x) 0 #define is_hugepage_only_range(mm, addr, len) 0 #define hugetlb_free_pgd_range(tlb, addr, end, floor, ceiling) ({BUG(); 0; }) #define hugetlb_fault(mm, vma, addr, write) ({ BUG(); 0; }) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 0c74c14dd2f7..107c1ce223cb 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -1996,6 +1996,15 @@ int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma, return ret; } +/* Can be overriden by architectures */ +__attribute__((weak)) struct page * +follow_huge_pud(struct mm_struct *mm, unsigned long address, + pud_t *pud, int write) +{ + BUG(); + return NULL; +} + int follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma, struct page **pages, struct vm_area_struct **vmas, unsigned long *position, int *length, int i, diff --git a/mm/memory.c b/mm/memory.c index 02fc6b1047b0..262e3eb6601a 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -998,19 +998,24 @@ struct page *follow_page(struct vm_area_struct *vma, unsigned long address, goto no_page_table; pud = pud_offset(pgd, address); - if (pud_none(*pud) || unlikely(pud_bad(*pud))) + if (pud_none(*pud)) goto no_page_table; - + if (pud_huge(*pud)) { + BUG_ON(flags & FOLL_GET); + page = follow_huge_pud(mm, address, pud, flags & FOLL_WRITE); + goto out; + } + if (unlikely(pud_bad(*pud))) + goto no_page_table; + pmd = pmd_offset(pud, address); if (pmd_none(*pmd)) goto no_page_table; - if (pmd_huge(*pmd)) { BUG_ON(flags & FOLL_GET); page = follow_huge_pmd(mm, address, pmd, flags & FOLL_WRITE); goto out; } - if (unlikely(pmd_bad(*pmd))) goto no_page_table; @@ -1567,6 +1572,8 @@ static int apply_to_pmd_range(struct mm_struct *mm, pud_t *pud, unsigned long next; int err; + BUG_ON(pud_huge(*pud)); + pmd = pmd_alloc(mm, pud, addr); if (!pmd) return -ENOMEM; From 39c11e6c05b7fedbf7ed4df3908b25f622d56204 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 23 Jul 2008 21:27:50 -0700 Subject: [PATCH 179/857] x86: support GB hugepages on 64-bit Acked-by: Adam Litke Signed-off-by: Andi Kleen Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/mm/hugetlbpage.c | 33 ++++++++++++++++++++++----------- 1 file changed, 22 insertions(+), 11 deletions(-) diff --git a/arch/x86/mm/hugetlbpage.c b/arch/x86/mm/hugetlbpage.c index a4789e87a315..b7a65a07af03 100644 --- a/arch/x86/mm/hugetlbpage.c +++ b/arch/x86/mm/hugetlbpage.c @@ -134,9 +134,14 @@ pte_t *huge_pte_alloc(struct mm_struct *mm, pgd = pgd_offset(mm, addr); pud = pud_alloc(mm, pgd, addr); if (pud) { - if (pud_none(*pud)) - huge_pmd_share(mm, addr, pud); - pte = (pte_t *) pmd_alloc(mm, pud, addr); + if (sz == PUD_SIZE) { + pte = (pte_t *)pud; + } else { + BUG_ON(sz != PMD_SIZE); + if (pud_none(*pud)) + huge_pmd_share(mm, addr, pud); + pte = (pte_t *) pmd_alloc(mm, pud, addr); + } } BUG_ON(pte && !pte_none(*pte) && !pte_huge(*pte)); @@ -152,8 +157,11 @@ pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) pgd = pgd_offset(mm, addr); if (pgd_present(*pgd)) { pud = pud_offset(pgd, addr); - if (pud_present(*pud)) + if (pud_present(*pud)) { + if (pud_large(*pud)) + return (pte_t *)pud; pmd = pmd_offset(pud, addr); + } } return (pte_t *) pmd; } @@ -216,7 +224,7 @@ int pmd_huge(pmd_t pmd) int pud_huge(pud_t pud) { - return 0; + return !!(pud_val(pud) & _PAGE_PSE); } struct page * @@ -252,6 +260,7 @@ static unsigned long hugetlb_get_unmapped_area_bottomup(struct file *file, unsigned long addr, unsigned long len, unsigned long pgoff, unsigned long flags) { + struct hstate *h = hstate_file(file); struct mm_struct *mm = current->mm; struct vm_area_struct *vma; unsigned long start_addr; @@ -264,7 +273,7 @@ static unsigned long hugetlb_get_unmapped_area_bottomup(struct file *file, } full_search: - addr = ALIGN(start_addr, HPAGE_SIZE); + addr = ALIGN(start_addr, huge_page_size(h)); for (vma = find_vma(mm, addr); ; vma = vma->vm_next) { /* At this point: (!vma || addr < vma->vm_end). */ @@ -286,7 +295,7 @@ static unsigned long hugetlb_get_unmapped_area_bottomup(struct file *file, } if (addr + mm->cached_hole_size < vma->vm_start) mm->cached_hole_size = vma->vm_start - addr; - addr = ALIGN(vma->vm_end, HPAGE_SIZE); + addr = ALIGN(vma->vm_end, huge_page_size(h)); } } @@ -294,6 +303,7 @@ static unsigned long hugetlb_get_unmapped_area_topdown(struct file *file, unsigned long addr0, unsigned long len, unsigned long pgoff, unsigned long flags) { + struct hstate *h = hstate_file(file); struct mm_struct *mm = current->mm; struct vm_area_struct *vma, *prev_vma; unsigned long base = mm->mmap_base, addr = addr0; @@ -314,7 +324,7 @@ static unsigned long hugetlb_get_unmapped_area_topdown(struct file *file, goto fail; /* either no address requested or cant fit in requested address hole */ - addr = (mm->free_area_cache - len) & HPAGE_MASK; + addr = (mm->free_area_cache - len) & huge_page_mask(h); do { /* * Lookup failure means no vma is above this address, @@ -345,7 +355,7 @@ static unsigned long hugetlb_get_unmapped_area_topdown(struct file *file, largest_hole = vma->vm_start - addr; /* try just below the current vma->vm_start */ - addr = (vma->vm_start - len) & HPAGE_MASK; + addr = (vma->vm_start - len) & huge_page_mask(h); } while (len <= vma->vm_start); fail: @@ -383,10 +393,11 @@ unsigned long hugetlb_get_unmapped_area(struct file *file, unsigned long addr, unsigned long len, unsigned long pgoff, unsigned long flags) { + struct hstate *h = hstate_file(file); struct mm_struct *mm = current->mm; struct vm_area_struct *vma; - if (len & ~HPAGE_MASK) + if (len & ~huge_page_mask(h)) return -EINVAL; if (len > TASK_SIZE) return -ENOMEM; @@ -398,7 +409,7 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr, } if (addr) { - addr = ALIGN(addr, HPAGE_SIZE); + addr = ALIGN(addr, huge_page_size(h)); vma = find_vma(mm, addr); if (TASK_SIZE - len >= addr && (!vma || addr + len <= vma->vm_start)) From b4718e628dbf68a2dee23b5709e2aa3190409c56 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 23 Jul 2008 21:27:51 -0700 Subject: [PATCH 180/857] x86: add hugepagesz option on 64-bit Add an hugepagesz=... option similar to IA64, PPC etc. to x86-64. This finally allows to select GB pages for hugetlbfs in x86 now that all the infrastructure is in place. Signed-off-by: Andi Kleen Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/kernel-parameters.txt | 11 +++++++++-- arch/x86/mm/hugetlbpage.c | 17 +++++++++++++++++ include/asm-x86/page.h | 2 ++ 3 files changed, 28 insertions(+), 2 deletions(-) diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 5e20ccb5a736..d55fd88fd0a9 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -774,8 +774,15 @@ and is between 256 and 4096 characters. It is defined in the file hisax= [HW,ISDN] See Documentation/isdn/README.HiSax. - hugepages= [HW,X86-32,IA-64] Maximal number of HugeTLB pages. - hugepagesz= [HW,IA-64,PPC] The size of the HugeTLB pages. + hugepages= [HW,X86-32,IA-64] HugeTLB pages to allocate at boot. + hugepagesz= [HW,IA-64,PPC,X86-64] The size of the HugeTLB pages. + On x86 this option can be specified multiple times + interleaved with hugepages= to reserve huge pages + of different sizes. Valid pages sizes on x86-64 + are 2M (when the CPU supports "pse") and 1G (when the + CPU supports the "pdpe1gb" cpuinfo flag) + Note that 1GB pages can only be allocated at boot time + using hugepages= and not freed afterwards. i8042.direct [HW] Put keyboard port into non-translated mode i8042.dumbkbd [HW] Pretend that controller can only read data from diff --git a/arch/x86/mm/hugetlbpage.c b/arch/x86/mm/hugetlbpage.c index b7a65a07af03..8f307d914c2e 100644 --- a/arch/x86/mm/hugetlbpage.c +++ b/arch/x86/mm/hugetlbpage.c @@ -425,3 +425,20 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr, #endif /*HAVE_ARCH_HUGETLB_UNMAPPED_AREA*/ +#ifdef CONFIG_X86_64 +static __init int setup_hugepagesz(char *opt) +{ + unsigned long ps = memparse(opt, &opt); + if (ps == PMD_SIZE) { + hugetlb_add_hstate(PMD_SHIFT - PAGE_SHIFT); + } else if (ps == PUD_SIZE && cpu_has_gbpages) { + hugetlb_add_hstate(PUD_SHIFT - PAGE_SHIFT); + } else { + printk(KERN_ERR "hugepagesz: Unsupported page size %lu M\n", + ps >> 20); + return 0; + } + return 1; +} +__setup("hugepagesz=", setup_hugepagesz); +#endif diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h index 6c846228948d..6e02098b1605 100644 --- a/include/asm-x86/page.h +++ b/include/asm-x86/page.h @@ -32,6 +32,8 @@ #define HPAGE_MASK (~(HPAGE_SIZE - 1)) #define HUGETLB_PAGE_ORDER (HPAGE_SHIFT - PAGE_SHIFT) +#define HUGE_MAX_HSTATE 2 + /* to align the pointer to the (next) page boundary */ #define PAGE_ALIGN(addr) (((addr)+PAGE_SIZE-1)&PAGE_MASK) From e11bfbfcb08ef4223b863799897c19cdf7c5bc00 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Wed, 23 Jul 2008 21:27:52 -0700 Subject: [PATCH 181/857] hugetlb: override default huge page size Allow configurations with the default huge page size which is different to the traditional HPAGE_SIZE size. The default huge page size is the one represented in the legacy /proc ABIs, SHM, and which is defaulted to when mounting hugetlbfs filesystems. This is implemented with a new kernel option default_hugepagesz=, which defaults to HPAGE_SIZE if not specified. Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/kernel-parameters.txt | 7 +++++++ mm/hugetlb.c | 23 +++++++++++++++++------ 2 files changed, 24 insertions(+), 6 deletions(-) diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index d55fd88fd0a9..30278e9e5211 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -783,6 +783,13 @@ and is between 256 and 4096 characters. It is defined in the file CPU supports the "pdpe1gb" cpuinfo flag) Note that 1GB pages can only be allocated at boot time using hugepages= and not freed afterwards. + default_hugepagesz= + [same as hugepagesz=] The size of the default + HugeTLB page size. This is the size represented by + the legacy /proc/ hugepages APIs, used for SHM, and + default size when mounting hugetlbfs filesystems. + Defaults to the default architecture's huge page size + if not specified. i8042.direct [HW] Put keyboard port into non-translated mode i8042.dumbkbd [HW] Pretend that controller can only read data from diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 107c1ce223cb..2a2f6e869401 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -34,6 +34,7 @@ struct hstate hstates[HUGE_MAX_HSTATE]; /* for command line parsing */ static struct hstate * __initdata parsed_hstate; static unsigned long __initdata default_hstate_max_huge_pages; +static unsigned long __initdata default_hstate_size; #define for_each_hstate(h) \ for ((h) = hstates; (h) < &hstates[max_hstate]; (h)++) @@ -1288,11 +1289,14 @@ static int __init hugetlb_init(void) { BUILD_BUG_ON(HPAGE_SHIFT == 0); - if (!size_to_hstate(HPAGE_SIZE)) { - hugetlb_add_hstate(HUGETLB_PAGE_ORDER); - parsed_hstate->max_huge_pages = default_hstate_max_huge_pages; + if (!size_to_hstate(default_hstate_size)) { + default_hstate_size = HPAGE_SIZE; + if (!size_to_hstate(default_hstate_size)) + hugetlb_add_hstate(HUGETLB_PAGE_ORDER); } - default_hstate_idx = size_to_hstate(HPAGE_SIZE) - hstates; + default_hstate_idx = size_to_hstate(default_hstate_size) - hstates; + if (default_hstate_max_huge_pages) + default_hstate.max_huge_pages = default_hstate_max_huge_pages; hugetlb_init_hstates(); @@ -1332,7 +1336,7 @@ void __init hugetlb_add_hstate(unsigned order) parsed_hstate = h; } -static int __init hugetlb_setup(char *s) +static int __init hugetlb_nrpages_setup(char *s) { unsigned long *mhp; static unsigned long *last_mhp; @@ -1367,7 +1371,14 @@ static int __init hugetlb_setup(char *s) return 1; } -__setup("hugepages=", hugetlb_setup); +__setup("hugepages=", hugetlb_nrpages_setup); + +static int __init hugetlb_default_setup(char *s) +{ + default_hstate_size = memparse(s, &s); + return 1; +} +__setup("default_hugepagesz=", hugetlb_default_setup); static unsigned int cpuset_mems_nr(unsigned int *array) { From 53ba51d21d6e048424ab8aadfebdb1f25ae07b60 Mon Sep 17 00:00:00 2001 From: Jon Tollefson Date: Wed, 23 Jul 2008 21:27:52 -0700 Subject: [PATCH 182/857] hugetlb: allow arch overridden hugepage allocation Allow alloc_bootmem_huge_page() to be overridden by architectures that can't always use bootmem. This requires huge_boot_pages to be available for use by this function. This is required for powerpc 16G pages, which have to be reserved prior to boot-time. The location of these pages are indicated in the device tree. Acked-by: Adam Litke Signed-off-by: Jon Tollefson Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/hugetlb.h | 10 ++++++++++ mm/hugetlb.c | 11 +++-------- 2 files changed, 13 insertions(+), 8 deletions(-) diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index b2c17f62cacb..9a71d4cc88c8 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -39,6 +39,7 @@ void hugetlb_unreserve_pages(struct inode *inode, long offset, long freed); extern unsigned long hugepages_treat_as_movable; extern const unsigned long hugetlb_zero, hugetlb_infinity; extern int sysctl_hugetlb_shm_group; +extern struct list_head huge_boot_pages; /* arch callbacks */ @@ -188,6 +189,14 @@ struct hstate { char name[HSTATE_NAME_LEN]; }; +struct huge_bootmem_page { + struct list_head list; + struct hstate *hstate; +}; + +/* arch callback */ +int __init alloc_bootmem_huge_page(struct hstate *h); + void __init hugetlb_add_hstate(unsigned order); struct hstate *size_to_hstate(unsigned long size); @@ -256,6 +265,7 @@ static inline struct hstate *page_hstate(struct page *page) #else struct hstate {}; +#define alloc_bootmem_huge_page(h) NULL #define hstate_file(f) NULL #define hstate_vma(v) NULL #define hstate_inode(i) NULL diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 2a2f6e869401..3e1506b808a3 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -31,6 +31,8 @@ static int max_hstate; unsigned int default_hstate_idx; struct hstate hstates[HUGE_MAX_HSTATE]; +__initdata LIST_HEAD(huge_boot_pages); + /* for command line parsing */ static struct hstate * __initdata parsed_hstate; static unsigned long __initdata default_hstate_max_huge_pages; @@ -925,14 +927,7 @@ static struct page *alloc_huge_page(struct vm_area_struct *vma, return page; } -static __initdata LIST_HEAD(huge_boot_pages); - -struct huge_bootmem_page { - struct list_head list; - struct hstate *hstate; -}; - -static int __init alloc_bootmem_huge_page(struct hstate *h) +__attribute__((weak)) int alloc_bootmem_huge_page(struct hstate *h) { struct huge_bootmem_page *m; int nr_nodes = nodes_weight(node_online_map); From ec4b2c0c8312d1118c2acd00c89988ecf955d5cc Mon Sep 17 00:00:00 2001 From: Jon Tollefson Date: Wed, 23 Jul 2008 21:27:53 -0700 Subject: [PATCH 183/857] powerpc: function to allocate gigantic hugepages The 16G page locations have been saved during early boot in an array. The alloc_bootmem_huge_page() function adds a page from here to the huge_boot_pages list. Acked-by: Adam Litke Signed-off-by: Jon Tollefson Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/powerpc/mm/hugetlbpage.c | 21 +++++++++++++++++++++ 1 file changed, 21 insertions(+) diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c index 63db7adce717..5df82186fc93 100644 --- a/arch/powerpc/mm/hugetlbpage.c +++ b/arch/powerpc/mm/hugetlbpage.c @@ -29,6 +29,12 @@ #define NUM_LOW_AREAS (0x100000000UL >> SID_SHIFT) #define NUM_HIGH_AREAS (PGTABLE_RANGE >> HTLB_AREA_SHIFT) +#define MAX_NUMBER_GPAGES 1024 + +/* Tracks the 16G pages after the device tree is scanned and before the + * huge_boot_pages list is ready. */ +static unsigned long gpage_freearray[MAX_NUMBER_GPAGES]; +static unsigned nr_gpages; unsigned int hugepte_shift; #define PTRS_PER_HUGEPTE (1 << hugepte_shift) @@ -104,6 +110,21 @@ pmd_t *hpmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long addr) } #endif +/* Moves the gigantic page addresses from the temporary list to the + * huge_boot_pages list. */ +int alloc_bootmem_huge_page(struct hstate *h) +{ + struct huge_bootmem_page *m; + if (nr_gpages == 0) + return 0; + m = phys_to_virt(gpage_freearray[--nr_gpages]); + gpage_freearray[nr_gpages] = 0; + list_add(&m->list, &huge_boot_pages); + m->hstate = h; + return 1; +} + + /* Modelled after find_linux_pte() */ pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) { From 658013e93eb70494f7300bc90457b09a807232a4 Mon Sep 17 00:00:00 2001 From: Jon Tollefson Date: Wed, 23 Jul 2008 21:27:54 -0700 Subject: [PATCH 184/857] powerpc: scan device tree for gigantic pages The 16G huge pages have to be reserved in the HMC prior to boot. The location of the pages are placed in the device tree. This patch adds code to scan the device tree during very early boot and save these page locations until hugetlbfs is ready for them. Acked-by: Adam Litke Signed-off-by: Jon Tollefson Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/powerpc/mm/hash_utils_64.c | 44 +++++++++++++++++++++++++++++++- arch/powerpc/mm/hugetlbpage.c | 16 ++++++++++++ include/asm-powerpc/mmu-hash64.h | 2 ++ 3 files changed, 61 insertions(+), 1 deletion(-) diff --git a/arch/powerpc/mm/hash_utils_64.c b/arch/powerpc/mm/hash_utils_64.c index 8d3b58ebd38e..ae4c717243a5 100644 --- a/arch/powerpc/mm/hash_utils_64.c +++ b/arch/powerpc/mm/hash_utils_64.c @@ -68,6 +68,7 @@ #define KB (1024) #define MB (1024*KB) +#define GB (1024L*MB) /* * Note: pte --> Linux PTE @@ -329,6 +330,44 @@ static int __init htab_dt_scan_page_sizes(unsigned long node, return 0; } +/* Scan for 16G memory blocks that have been set aside for huge pages + * and reserve those blocks for 16G huge pages. + */ +static int __init htab_dt_scan_hugepage_blocks(unsigned long node, + const char *uname, int depth, + void *data) { + char *type = of_get_flat_dt_prop(node, "device_type", NULL); + unsigned long *addr_prop; + u32 *page_count_prop; + unsigned int expected_pages; + long unsigned int phys_addr; + long unsigned int block_size; + + /* We are scanning "memory" nodes only */ + if (type == NULL || strcmp(type, "memory") != 0) + return 0; + + /* This property is the log base 2 of the number of virtual pages that + * will represent this memory block. */ + page_count_prop = of_get_flat_dt_prop(node, "ibm,expected#pages", NULL); + if (page_count_prop == NULL) + return 0; + expected_pages = (1 << page_count_prop[0]); + addr_prop = of_get_flat_dt_prop(node, "reg", NULL); + if (addr_prop == NULL) + return 0; + phys_addr = addr_prop[0]; + block_size = addr_prop[1]; + if (block_size != (16 * GB)) + return 0; + printk(KERN_INFO "Huge page(16GB) memory: " + "addr = 0x%lX size = 0x%lX pages = %d\n", + phys_addr, block_size, expected_pages); + lmb_reserve(phys_addr, block_size * expected_pages); + add_gpage(phys_addr, block_size, expected_pages); + return 0; +} + static void __init htab_init_page_sizes(void) { int rc; @@ -418,7 +457,10 @@ static void __init htab_init_page_sizes(void) ); #ifdef CONFIG_HUGETLB_PAGE - /* Init large page size. Currently, we pick 16M or 1M depending + /* Reserve 16G huge page memory sections for huge pages */ + of_scan_flat_dt(htab_dt_scan_hugepage_blocks, NULL); + +/* Init large page size. Currently, we pick 16M or 1M depending * on what is available */ if (mmu_psize_defs[MMU_PAGE_16M].shift) diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c index 5df82186fc93..e2a650a9e533 100644 --- a/arch/powerpc/mm/hugetlbpage.c +++ b/arch/powerpc/mm/hugetlbpage.c @@ -110,6 +110,22 @@ pmd_t *hpmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long addr) } #endif +/* Build list of addresses of gigantic pages. This function is used in early + * boot before the buddy or bootmem allocator is setup. + */ +void add_gpage(unsigned long addr, unsigned long page_size, + unsigned long number_of_pages) +{ + if (!addr) + return; + while (number_of_pages > 0) { + gpage_freearray[nr_gpages] = addr; + nr_gpages++; + number_of_pages--; + addr += page_size; + } +} + /* Moves the gigantic page addresses from the temporary list to the * huge_boot_pages list. */ int alloc_bootmem_huge_page(struct hstate *h) diff --git a/include/asm-powerpc/mmu-hash64.h b/include/asm-powerpc/mmu-hash64.h index d1dc16afb118..b61181aa7746 100644 --- a/include/asm-powerpc/mmu-hash64.h +++ b/include/asm-powerpc/mmu-hash64.h @@ -281,6 +281,8 @@ extern int htab_bolt_mapping(unsigned long vstart, unsigned long vend, unsigned long pstart, unsigned long mode, int psize, int ssize); extern void set_huge_psize(int psize); +extern void add_gpage(unsigned long addr, unsigned long page_size, + unsigned long number_of_pages); extern void demote_segment_4k(struct mm_struct *mm, unsigned long addr); extern void htab_initialize(void); From 91224346aa8c1cdaa660300a98e0b074a3a95030 Mon Sep 17 00:00:00 2001 From: Jon Tollefson Date: Wed, 23 Jul 2008 21:27:55 -0700 Subject: [PATCH 185/857] powerpc: define support for 16G hugepages The huge page size is defined for 16G pages. If a hugepagesz of 16G is specified at boot-time then it becomes the huge page size instead of the default 16M. The change in pgtable-64K.h is to the macro pte_iterate_hashed_subpages to make the increment to va (the 1 being shifted) be a long so that it is not shifted to 0. Otherwise it would create an infinite loop when the shift value is for a 16G page (when base page size is 64K). Signed-off-by: Jon Tollefson Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/powerpc/mm/hugetlbpage.c | 62 ++++++++++++++++++++++--------- include/asm-powerpc/pgtable-64k.h | 2 +- 2 files changed, 45 insertions(+), 19 deletions(-) diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c index e2a650a9e533..19b1a9cec6d5 100644 --- a/arch/powerpc/mm/hugetlbpage.c +++ b/arch/powerpc/mm/hugetlbpage.c @@ -24,8 +24,9 @@ #include #include -#define HPAGE_SHIFT_64K 16 -#define HPAGE_SHIFT_16M 24 +#define PAGE_SHIFT_64K 16 +#define PAGE_SHIFT_16M 24 +#define PAGE_SHIFT_16G 34 #define NUM_LOW_AREAS (0x100000000UL >> SID_SHIFT) #define NUM_HIGH_AREAS (PGTABLE_RANGE >> HTLB_AREA_SHIFT) @@ -95,7 +96,7 @@ static int __hugepte_alloc(struct mm_struct *mm, hugepd_t *hpdp, static inline pmd_t *hpmd_offset(pud_t *pud, unsigned long addr) { - if (HPAGE_SHIFT == HPAGE_SHIFT_64K) + if (HPAGE_SHIFT == PAGE_SHIFT_64K) return pmd_offset(pud, addr); else return (pmd_t *) pud; @@ -103,7 +104,7 @@ pmd_t *hpmd_offset(pud_t *pud, unsigned long addr) static inline pmd_t *hpmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long addr) { - if (HPAGE_SHIFT == HPAGE_SHIFT_64K) + if (HPAGE_SHIFT == PAGE_SHIFT_64K) return pmd_alloc(mm, pud, addr); else return (pmd_t *) pud; @@ -260,7 +261,7 @@ static void hugetlb_free_pud_range(struct mmu_gather *tlb, pgd_t *pgd, continue; hugetlb_free_pmd_range(tlb, pud, addr, next, floor, ceiling); #else - if (HPAGE_SHIFT == HPAGE_SHIFT_64K) { + if (HPAGE_SHIFT == PAGE_SHIFT_64K) { if (pud_none_or_clear_bad(pud)) continue; hugetlb_free_pmd_range(tlb, pud, addr, next, floor, ceiling); @@ -592,20 +593,40 @@ void set_huge_psize(int psize) { /* Check that it is a page size supported by the hardware and * that it fits within pagetable limits. */ - if (mmu_psize_defs[psize].shift && mmu_psize_defs[psize].shift < SID_SHIFT && + if (mmu_psize_defs[psize].shift && + mmu_psize_defs[psize].shift < SID_SHIFT_1T && (mmu_psize_defs[psize].shift > MIN_HUGEPTE_SHIFT || - mmu_psize_defs[psize].shift == HPAGE_SHIFT_64K)) { + mmu_psize_defs[psize].shift == PAGE_SHIFT_64K || + mmu_psize_defs[psize].shift == PAGE_SHIFT_16G)) { + /* Return if huge page size is the same as the + * base page size. */ + if (mmu_psize_defs[psize].shift == PAGE_SHIFT) + return; + HPAGE_SHIFT = mmu_psize_defs[psize].shift; mmu_huge_psize = psize; -#ifdef CONFIG_PPC_64K_PAGES - hugepte_shift = (PMD_SHIFT-HPAGE_SHIFT); -#else - if (HPAGE_SHIFT == HPAGE_SHIFT_64K) - hugepte_shift = (PMD_SHIFT-HPAGE_SHIFT); - else - hugepte_shift = (PUD_SHIFT-HPAGE_SHIFT); -#endif + switch (HPAGE_SHIFT) { + case PAGE_SHIFT_64K: + /* We only allow 64k hpages with 4k base page, + * which was checked above, and always put them + * at the PMD */ + hugepte_shift = PMD_SHIFT; + break; + case PAGE_SHIFT_16M: + /* 16M pages can be at two different levels + * of pagestables based on base page size */ + if (PAGE_SHIFT == PAGE_SHIFT_64K) + hugepte_shift = PMD_SHIFT; + else /* 4k base page */ + hugepte_shift = PUD_SHIFT; + break; + case PAGE_SHIFT_16G: + /* 16G pages are always at PGD level */ + hugepte_shift = PGDIR_SHIFT; + break; + } + hugepte_shift -= HPAGE_SHIFT; } else HPAGE_SHIFT = 0; } @@ -621,17 +642,22 @@ static int __init hugepage_setup_sz(char *str) shift = __ffs(size); switch (shift) { #ifndef CONFIG_PPC_64K_PAGES - case HPAGE_SHIFT_64K: + case PAGE_SHIFT_64K: mmu_psize = MMU_PAGE_64K; break; #endif - case HPAGE_SHIFT_16M: + case PAGE_SHIFT_16M: mmu_psize = MMU_PAGE_16M; break; + case PAGE_SHIFT_16G: + mmu_psize = MMU_PAGE_16G; + break; } - if (mmu_psize >=0 && mmu_psize_defs[mmu_psize].shift) + if (mmu_psize >= 0 && mmu_psize_defs[mmu_psize].shift) { set_huge_psize(mmu_psize); + hugetlb_add_hstate(shift - PAGE_SHIFT); + } else printk(KERN_WARNING "Invalid huge page size specified(%llu)\n", size); diff --git a/include/asm-powerpc/pgtable-64k.h b/include/asm-powerpc/pgtable-64k.h index c5007712473f..7e54adb35596 100644 --- a/include/asm-powerpc/pgtable-64k.h +++ b/include/asm-powerpc/pgtable-64k.h @@ -138,7 +138,7 @@ static inline struct subpage_prot_table *pgd_subpage_prot(pgd_t *pgd) unsigned __split = (psize == MMU_PAGE_4K || \ psize == MMU_PAGE_64K_AP); \ shift = mmu_psize_defs[psize].shift; \ - for (index = 0; va < __end; index++, va += (1 << shift)) { \ + for (index = 0; va < __end; index++, va += (1L << shift)) { \ if (!__split || __rpte_sub_valid(rpte, index)) do { \ #define pte_iterate_hashed_end() } while(0); } } while(0) From f4a67cceee4a6f5ed38011a698c9e34747270ae5 Mon Sep 17 00:00:00 2001 From: Jon Tollefson Date: Wed, 23 Jul 2008 21:27:55 -0700 Subject: [PATCH 186/857] fs: check for statfs overflow Adds a check for an overflow in the filesystem size so if someone is checking with statfs() on a 16G blocksize hugetlbfs in a 32bit binary that it will report back EOVERFLOW instead of a size of 0. Acked-by: Nishanth Aravamudan Signed-off-by: Jon Tollefson Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/compat.c | 8 ++++---- fs/open.c | 3 ++- 2 files changed, 6 insertions(+), 5 deletions(-) diff --git a/fs/compat.c b/fs/compat.c index ed43e17a5dc6..b46604281766 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -197,8 +197,8 @@ static int put_compat_statfs(struct compat_statfs __user *ubuf, struct kstatfs * { if (sizeof ubuf->f_blocks == 4) { - if ((kbuf->f_blocks | kbuf->f_bfree | kbuf->f_bavail) & - 0xffffffff00000000ULL) + if ((kbuf->f_blocks | kbuf->f_bfree | kbuf->f_bavail | + kbuf->f_bsize | kbuf->f_frsize) & 0xffffffff00000000ULL) return -EOVERFLOW; /* f_files and f_ffree may be -1; it's okay * to stuff that into 32 bits */ @@ -271,8 +271,8 @@ asmlinkage long compat_sys_fstatfs(unsigned int fd, struct compat_statfs __user static int put_compat_statfs64(struct compat_statfs64 __user *ubuf, struct kstatfs *kbuf) { if (sizeof ubuf->f_blocks == 4) { - if ((kbuf->f_blocks | kbuf->f_bfree | kbuf->f_bavail) & - 0xffffffff00000000ULL) + if ((kbuf->f_blocks | kbuf->f_bfree | kbuf->f_bavail | + kbuf->f_bsize | kbuf->f_frsize) & 0xffffffff00000000ULL) return -EOVERFLOW; /* f_files and f_ffree may be -1; it's okay * to stuff that into 32 bits */ diff --git a/fs/open.c b/fs/open.c index a99ad09c3197..bb98d2fe809f 100644 --- a/fs/open.c +++ b/fs/open.c @@ -64,7 +64,8 @@ static int vfs_statfs_native(struct dentry *dentry, struct statfs *buf) memcpy(buf, &st, sizeof(st)); else { if (sizeof buf->f_blocks == 4) { - if ((st.f_blocks | st.f_bfree | st.f_bavail) & + if ((st.f_blocks | st.f_bfree | st.f_bavail | + st.f_bsize | st.f_frsize) & 0xffffffff00000000ULL) return -EOVERFLOW; /* From 0d9ea75443dc7e37843e656b8ebc947a6d16d618 Mon Sep 17 00:00:00 2001 From: Jon Tollefson Date: Wed, 23 Jul 2008 21:27:56 -0700 Subject: [PATCH 187/857] powerpc: support multiple hugepage sizes Instead of using the variable mmu_huge_psize to keep track of the huge page size we use an array of MMU_PAGE_* values. For each supported huge page size we need to know the hugepte_shift value and have a pgtable_cache. The hstate or an mmu_huge_psizes index is passed to functions so that they know which huge page size they should use. The hugepage sizes 16M and 64K are setup(if available on the hardware) so that they don't have to be set on the boot cmd line in order to use them. The number of 16G pages have to be specified at boot-time though (e.g. hugepagesz=16G hugepages=5). Signed-off-by: Jon Tollefson Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/kernel-parameters.txt | 10 +- arch/powerpc/mm/hash_utils_64.c | 9 +- arch/powerpc/mm/hugetlbpage.c | 272 ++++++++++++++++++---------- arch/powerpc/mm/init_64.c | 8 +- arch/powerpc/mm/tlb_64.c | 2 +- include/asm-powerpc/hugetlb.h | 5 +- include/asm-powerpc/mmu-hash64.h | 4 +- include/asm-powerpc/page_64.h | 1 + include/asm-powerpc/pgalloc-64.h | 4 +- 9 files changed, 198 insertions(+), 117 deletions(-) diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 30278e9e5211..01a2992b5754 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -776,11 +776,11 @@ and is between 256 and 4096 characters. It is defined in the file hugepages= [HW,X86-32,IA-64] HugeTLB pages to allocate at boot. hugepagesz= [HW,IA-64,PPC,X86-64] The size of the HugeTLB pages. - On x86 this option can be specified multiple times - interleaved with hugepages= to reserve huge pages - of different sizes. Valid pages sizes on x86-64 - are 2M (when the CPU supports "pse") and 1G (when the - CPU supports the "pdpe1gb" cpuinfo flag) + On x86-64 and powerpc, this option can be specified + multiple times interleaved with hugepages= to reserve + huge pages of different sizes. Valid pages sizes on + x86-64 are 2M (when the CPU supports "pse") and 1G + (when the CPU supports the "pdpe1gb" cpuinfo flag) Note that 1GB pages can only be allocated at boot time using hugepages= and not freed afterwards. default_hugepagesz= diff --git a/arch/powerpc/mm/hash_utils_64.c b/arch/powerpc/mm/hash_utils_64.c index ae4c717243a5..5ce5a4dcd008 100644 --- a/arch/powerpc/mm/hash_utils_64.c +++ b/arch/powerpc/mm/hash_utils_64.c @@ -103,7 +103,6 @@ int mmu_kernel_ssize = MMU_SEGSIZE_256M; int mmu_highuser_ssize = MMU_SEGSIZE_256M; u16 mmu_slb_size = 64; #ifdef CONFIG_HUGETLB_PAGE -int mmu_huge_psize = MMU_PAGE_16M; unsigned int HPAGE_SHIFT; #endif #ifdef CONFIG_PPC_64K_PAGES @@ -460,15 +459,15 @@ static void __init htab_init_page_sizes(void) /* Reserve 16G huge page memory sections for huge pages */ of_scan_flat_dt(htab_dt_scan_hugepage_blocks, NULL); -/* Init large page size. Currently, we pick 16M or 1M depending +/* Set default large page size. Currently, we pick 16M or 1M depending * on what is available */ if (mmu_psize_defs[MMU_PAGE_16M].shift) - set_huge_psize(MMU_PAGE_16M); + HPAGE_SHIFT = mmu_psize_defs[MMU_PAGE_16M].shift; /* With 4k/4level pagetables, we can't (for now) cope with a * huge page size < PMD_SIZE */ else if (mmu_psize_defs[MMU_PAGE_1M].shift) - set_huge_psize(MMU_PAGE_1M); + HPAGE_SHIFT = mmu_psize_defs[MMU_PAGE_1M].shift; #endif /* CONFIG_HUGETLB_PAGE */ } @@ -889,7 +888,7 @@ int hash_page(unsigned long ea, unsigned long access, unsigned long trap) #ifdef CONFIG_HUGETLB_PAGE /* Handle hugepage regions */ - if (HPAGE_SHIFT && psize == mmu_huge_psize) { + if (HPAGE_SHIFT && mmu_huge_psizes[psize]) { DBG_LOW(" -> huge page !\n"); return hash_huge_page(mm, access, ea, vsid, local, trap); } diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c index 19b1a9cec6d5..fb42c4dd3217 100644 --- a/arch/powerpc/mm/hugetlbpage.c +++ b/arch/powerpc/mm/hugetlbpage.c @@ -37,15 +37,30 @@ static unsigned long gpage_freearray[MAX_NUMBER_GPAGES]; static unsigned nr_gpages; -unsigned int hugepte_shift; -#define PTRS_PER_HUGEPTE (1 << hugepte_shift) -#define HUGEPTE_TABLE_SIZE (sizeof(pte_t) << hugepte_shift) +/* Array of valid huge page sizes - non-zero value(hugepte_shift) is + * stored for the huge page sizes that are valid. + */ +unsigned int mmu_huge_psizes[MMU_PAGE_COUNT] = { }; /* initialize all to 0 */ -#define HUGEPD_SHIFT (HPAGE_SHIFT + hugepte_shift) -#define HUGEPD_SIZE (1UL << HUGEPD_SHIFT) -#define HUGEPD_MASK (~(HUGEPD_SIZE-1)) +#define hugepte_shift mmu_huge_psizes +#define PTRS_PER_HUGEPTE(psize) (1 << hugepte_shift[psize]) +#define HUGEPTE_TABLE_SIZE(psize) (sizeof(pte_t) << hugepte_shift[psize]) -#define huge_pgtable_cache (pgtable_cache[HUGEPTE_CACHE_NUM]) +#define HUGEPD_SHIFT(psize) (mmu_psize_to_shift(psize) \ + + hugepte_shift[psize]) +#define HUGEPD_SIZE(psize) (1UL << HUGEPD_SHIFT(psize)) +#define HUGEPD_MASK(psize) (~(HUGEPD_SIZE(psize)-1)) + +/* Subtract one from array size because we don't need a cache for 4K since + * is not a huge page size */ +#define huge_pgtable_cache(psize) (pgtable_cache[HUGEPTE_CACHE_NUM \ + + psize-1]) +#define HUGEPTE_CACHE_NAME(psize) (huge_pgtable_cache_name[psize]) + +static const char *huge_pgtable_cache_name[MMU_PAGE_COUNT] = { + "unused_4K", "hugepte_cache_64K", "unused_64K_AP", + "hugepte_cache_1M", "hugepte_cache_16M", "hugepte_cache_16G" +}; /* Flag to mark huge PD pointers. This means pmd_bad() and pud_bad() * will choke on pointers to hugepte tables, which is handy for @@ -56,24 +71,49 @@ typedef struct { unsigned long pd; } hugepd_t; #define hugepd_none(hpd) ((hpd).pd == 0) +static inline int shift_to_mmu_psize(unsigned int shift) +{ + switch (shift) { +#ifndef CONFIG_PPC_64K_PAGES + case PAGE_SHIFT_64K: + return MMU_PAGE_64K; +#endif + case PAGE_SHIFT_16M: + return MMU_PAGE_16M; + case PAGE_SHIFT_16G: + return MMU_PAGE_16G; + } + return -1; +} + +static inline unsigned int mmu_psize_to_shift(unsigned int mmu_psize) +{ + if (mmu_psize_defs[mmu_psize].shift) + return mmu_psize_defs[mmu_psize].shift; + BUG(); +} + static inline pte_t *hugepd_page(hugepd_t hpd) { BUG_ON(!(hpd.pd & HUGEPD_OK)); return (pte_t *)(hpd.pd & ~HUGEPD_OK); } -static inline pte_t *hugepte_offset(hugepd_t *hpdp, unsigned long addr) +static inline pte_t *hugepte_offset(hugepd_t *hpdp, unsigned long addr, + struct hstate *hstate) { - unsigned long idx = ((addr >> HPAGE_SHIFT) & (PTRS_PER_HUGEPTE-1)); + unsigned int shift = huge_page_shift(hstate); + int psize = shift_to_mmu_psize(shift); + unsigned long idx = ((addr >> shift) & (PTRS_PER_HUGEPTE(psize)-1)); pte_t *dir = hugepd_page(*hpdp); return dir + idx; } static int __hugepte_alloc(struct mm_struct *mm, hugepd_t *hpdp, - unsigned long address) + unsigned long address, unsigned int psize) { - pte_t *new = kmem_cache_alloc(huge_pgtable_cache, + pte_t *new = kmem_cache_alloc(huge_pgtable_cache(psize), GFP_KERNEL|__GFP_REPEAT); if (! new) @@ -81,7 +121,7 @@ static int __hugepte_alloc(struct mm_struct *mm, hugepd_t *hpdp, spin_lock(&mm->page_table_lock); if (!hugepd_none(*hpdp)) - kmem_cache_free(huge_pgtable_cache, new); + kmem_cache_free(huge_pgtable_cache(psize), new); else hpdp->pd = (unsigned long)new | HUGEPD_OK; spin_unlock(&mm->page_table_lock); @@ -90,21 +130,22 @@ static int __hugepte_alloc(struct mm_struct *mm, hugepd_t *hpdp, /* Base page size affects how we walk hugetlb page tables */ #ifdef CONFIG_PPC_64K_PAGES -#define hpmd_offset(pud, addr) pmd_offset(pud, addr) -#define hpmd_alloc(mm, pud, addr) pmd_alloc(mm, pud, addr) +#define hpmd_offset(pud, addr, h) pmd_offset(pud, addr) +#define hpmd_alloc(mm, pud, addr, h) pmd_alloc(mm, pud, addr) #else static inline -pmd_t *hpmd_offset(pud_t *pud, unsigned long addr) +pmd_t *hpmd_offset(pud_t *pud, unsigned long addr, struct hstate *hstate) { - if (HPAGE_SHIFT == PAGE_SHIFT_64K) + if (huge_page_shift(hstate) == PAGE_SHIFT_64K) return pmd_offset(pud, addr); else return (pmd_t *) pud; } static inline -pmd_t *hpmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long addr) +pmd_t *hpmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long addr, + struct hstate *hstate) { - if (HPAGE_SHIFT == PAGE_SHIFT_64K) + if (huge_page_shift(hstate) == PAGE_SHIFT_64K) return pmd_alloc(mm, pud, addr); else return (pmd_t *) pud; @@ -128,8 +169,9 @@ void add_gpage(unsigned long addr, unsigned long page_size, } /* Moves the gigantic page addresses from the temporary list to the - * huge_boot_pages list. */ -int alloc_bootmem_huge_page(struct hstate *h) + * huge_boot_pages list. + */ +int alloc_bootmem_huge_page(struct hstate *hstate) { struct huge_bootmem_page *m; if (nr_gpages == 0) @@ -137,7 +179,7 @@ int alloc_bootmem_huge_page(struct hstate *h) m = phys_to_virt(gpage_freearray[--nr_gpages]); gpage_freearray[nr_gpages] = 0; list_add(&m->list, &huge_boot_pages); - m->hstate = h; + m->hstate = hstate; return 1; } @@ -149,17 +191,25 @@ pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) pud_t *pu; pmd_t *pm; - BUG_ON(get_slice_psize(mm, addr) != mmu_huge_psize); + unsigned int psize; + unsigned int shift; + unsigned long sz; + struct hstate *hstate; + psize = get_slice_psize(mm, addr); + shift = mmu_psize_to_shift(psize); + sz = ((1UL) << shift); + hstate = size_to_hstate(sz); - addr &= HPAGE_MASK; + addr &= hstate->mask; pg = pgd_offset(mm, addr); if (!pgd_none(*pg)) { pu = pud_offset(pg, addr); if (!pud_none(*pu)) { - pm = hpmd_offset(pu, addr); + pm = hpmd_offset(pu, addr, hstate); if (!pmd_none(*pm)) - return hugepte_offset((hugepd_t *)pm, addr); + return hugepte_offset((hugepd_t *)pm, addr, + hstate); } } @@ -173,16 +223,20 @@ pte_t *huge_pte_alloc(struct mm_struct *mm, pud_t *pu; pmd_t *pm; hugepd_t *hpdp = NULL; + struct hstate *hstate; + unsigned int psize; + hstate = size_to_hstate(sz); - BUG_ON(get_slice_psize(mm, addr) != mmu_huge_psize); + psize = get_slice_psize(mm, addr); + BUG_ON(!mmu_huge_psizes[psize]); - addr &= HPAGE_MASK; + addr &= hstate->mask; pg = pgd_offset(mm, addr); pu = pud_alloc(mm, pg, addr); if (pu) { - pm = hpmd_alloc(mm, pu, addr); + pm = hpmd_alloc(mm, pu, addr, hstate); if (pm) hpdp = (hugepd_t *)pm; } @@ -190,10 +244,10 @@ pte_t *huge_pte_alloc(struct mm_struct *mm, if (! hpdp) return NULL; - if (hugepd_none(*hpdp) && __hugepte_alloc(mm, hpdp, addr)) + if (hugepd_none(*hpdp) && __hugepte_alloc(mm, hpdp, addr, psize)) return NULL; - return hugepte_offset(hpdp, addr); + return hugepte_offset(hpdp, addr, hstate); } int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) @@ -201,19 +255,22 @@ int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep) return 0; } -static void free_hugepte_range(struct mmu_gather *tlb, hugepd_t *hpdp) +static void free_hugepte_range(struct mmu_gather *tlb, hugepd_t *hpdp, + unsigned int psize) { pte_t *hugepte = hugepd_page(*hpdp); hpdp->pd = 0; tlb->need_flush = 1; - pgtable_free_tlb(tlb, pgtable_free_cache(hugepte, HUGEPTE_CACHE_NUM, + pgtable_free_tlb(tlb, pgtable_free_cache(hugepte, + HUGEPTE_CACHE_NUM+psize-1, PGF_CACHENUM_MASK)); } static void hugetlb_free_pmd_range(struct mmu_gather *tlb, pud_t *pud, unsigned long addr, unsigned long end, - unsigned long floor, unsigned long ceiling) + unsigned long floor, unsigned long ceiling, + unsigned int psize) { pmd_t *pmd; unsigned long next; @@ -225,7 +282,7 @@ static void hugetlb_free_pmd_range(struct mmu_gather *tlb, pud_t *pud, next = pmd_addr_end(addr, end); if (pmd_none(*pmd)) continue; - free_hugepte_range(tlb, (hugepd_t *)pmd); + free_hugepte_range(tlb, (hugepd_t *)pmd, psize); } while (pmd++, addr = next, addr != end); start &= PUD_MASK; @@ -251,6 +308,9 @@ static void hugetlb_free_pud_range(struct mmu_gather *tlb, pgd_t *pgd, pud_t *pud; unsigned long next; unsigned long start; + unsigned int shift; + unsigned int psize = get_slice_psize(tlb->mm, addr); + shift = mmu_psize_to_shift(psize); start = addr; pud = pud_offset(pgd, addr); @@ -259,16 +319,18 @@ static void hugetlb_free_pud_range(struct mmu_gather *tlb, pgd_t *pgd, #ifdef CONFIG_PPC_64K_PAGES if (pud_none_or_clear_bad(pud)) continue; - hugetlb_free_pmd_range(tlb, pud, addr, next, floor, ceiling); + hugetlb_free_pmd_range(tlb, pud, addr, next, floor, ceiling, + psize); #else - if (HPAGE_SHIFT == PAGE_SHIFT_64K) { + if (shift == PAGE_SHIFT_64K) { if (pud_none_or_clear_bad(pud)) continue; - hugetlb_free_pmd_range(tlb, pud, addr, next, floor, ceiling); + hugetlb_free_pmd_range(tlb, pud, addr, next, floor, + ceiling, psize); } else { if (pud_none(*pud)) continue; - free_hugepte_range(tlb, (hugepd_t *)pud); + free_hugepte_range(tlb, (hugepd_t *)pud, psize); } #endif } while (pud++, addr = next, addr != end); @@ -336,27 +398,29 @@ void hugetlb_free_pgd_range(struct mmu_gather *tlb, * now has no other vmas using it, so can be freed, we don't * bother to round floor or end up - the tests don't need that. */ + unsigned int psize = get_slice_psize(tlb->mm, addr); - addr &= HUGEPD_MASK; + addr &= HUGEPD_MASK(psize); if (addr < floor) { - addr += HUGEPD_SIZE; + addr += HUGEPD_SIZE(psize); if (!addr) return; } if (ceiling) { - ceiling &= HUGEPD_MASK; + ceiling &= HUGEPD_MASK(psize); if (!ceiling) return; } if (end - 1 > ceiling - 1) - end -= HUGEPD_SIZE; + end -= HUGEPD_SIZE(psize); if (addr > end - 1) return; start = addr; pgd = pgd_offset(tlb->mm, addr); do { - BUG_ON(get_slice_psize(tlb->mm, addr) != mmu_huge_psize); + psize = get_slice_psize(tlb->mm, addr); + BUG_ON(!mmu_huge_psizes[psize]); next = pgd_addr_end(addr, end); if (pgd_none_or_clear_bad(pgd)) continue; @@ -373,7 +437,11 @@ void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, * necessary anymore if we make hpte_need_flush() get the * page size from the slices */ - pte_update(mm, addr & HPAGE_MASK, ptep, ~0UL, 1); + unsigned int psize = get_slice_psize(mm, addr); + unsigned int shift = mmu_psize_to_shift(psize); + unsigned long sz = ((1UL) << shift); + struct hstate *hstate = size_to_hstate(sz); + pte_update(mm, addr & hstate->mask, ptep, ~0UL, 1); } *ptep = __pte(pte_val(pte) & ~_PAGE_HPTEFLAGS); } @@ -390,14 +458,19 @@ follow_huge_addr(struct mm_struct *mm, unsigned long address, int write) { pte_t *ptep; struct page *page; + unsigned int mmu_psize = get_slice_psize(mm, address); - if (get_slice_psize(mm, address) != mmu_huge_psize) + /* Verify it is a huge page else bail. */ + if (!mmu_huge_psizes[mmu_psize]) return ERR_PTR(-EINVAL); ptep = huge_pte_offset(mm, address); page = pte_page(*ptep); - if (page) - page += (address % HPAGE_SIZE) / PAGE_SIZE; + if (page) { + unsigned int shift = mmu_psize_to_shift(mmu_psize); + unsigned long sz = ((1UL) << shift); + page += (address % sz) / PAGE_SIZE; + } return page; } @@ -425,15 +498,16 @@ unsigned long hugetlb_get_unmapped_area(struct file *file, unsigned long addr, unsigned long len, unsigned long pgoff, unsigned long flags) { - return slice_get_unmapped_area(addr, len, flags, - mmu_huge_psize, 1, 0); + struct hstate *hstate = hstate_file(file); + int mmu_psize = shift_to_mmu_psize(huge_page_shift(hstate)); + return slice_get_unmapped_area(addr, len, flags, mmu_psize, 1, 0); } /* * Called by asm hashtable.S for doing lazy icache flush */ static unsigned int hash_huge_page_do_lazy_icache(unsigned long rflags, - pte_t pte, int trap) + pte_t pte, int trap, unsigned long sz) { struct page *page; int i; @@ -446,7 +520,7 @@ static unsigned int hash_huge_page_do_lazy_icache(unsigned long rflags, /* page is dirty */ if (!test_bit(PG_arch_1, &page->flags) && !PageReserved(page)) { if (trap == 0x400) { - for (i = 0; i < (HPAGE_SIZE / PAGE_SIZE); i++) + for (i = 0; i < (sz / PAGE_SIZE); i++) __flush_dcache_icache(page_address(page+i)); set_bit(PG_arch_1, &page->flags); } else { @@ -462,11 +536,16 @@ int hash_huge_page(struct mm_struct *mm, unsigned long access, { pte_t *ptep; unsigned long old_pte, new_pte; - unsigned long va, rflags, pa; + unsigned long va, rflags, pa, sz; long slot; int err = 1; int ssize = user_segment_size(ea); + unsigned int mmu_psize; + int shift; + mmu_psize = get_slice_psize(mm, ea); + if (!mmu_huge_psizes[mmu_psize]) + goto out; ptep = huge_pte_offset(mm, ea); /* Search the Linux page table for a match with va */ @@ -509,30 +588,32 @@ int hash_huge_page(struct mm_struct *mm, unsigned long access, rflags = 0x2 | (!(new_pte & _PAGE_RW)); /* _PAGE_EXEC -> HW_NO_EXEC since it's inverted */ rflags |= ((new_pte & _PAGE_EXEC) ? 0 : HPTE_R_N); + shift = mmu_psize_to_shift(mmu_psize); + sz = ((1UL) << shift); if (!cpu_has_feature(CPU_FTR_COHERENT_ICACHE)) /* No CPU has hugepages but lacks no execute, so we * don't need to worry about that case */ rflags = hash_huge_page_do_lazy_icache(rflags, __pte(old_pte), - trap); + trap, sz); /* Check if pte already has an hpte (case 2) */ if (unlikely(old_pte & _PAGE_HASHPTE)) { /* There MIGHT be an HPTE for this pte */ unsigned long hash, slot; - hash = hpt_hash(va, HPAGE_SHIFT, ssize); + hash = hpt_hash(va, shift, ssize); if (old_pte & _PAGE_F_SECOND) hash = ~hash; slot = (hash & htab_hash_mask) * HPTES_PER_GROUP; slot += (old_pte & _PAGE_F_GIX) >> 12; - if (ppc_md.hpte_updatepp(slot, rflags, va, mmu_huge_psize, + if (ppc_md.hpte_updatepp(slot, rflags, va, mmu_psize, ssize, local) == -1) old_pte &= ~_PAGE_HPTEFLAGS; } if (likely(!(old_pte & _PAGE_HASHPTE))) { - unsigned long hash = hpt_hash(va, HPAGE_SHIFT, ssize); + unsigned long hash = hpt_hash(va, shift, ssize); unsigned long hpte_group; pa = pte_pfn(__pte(old_pte)) << PAGE_SHIFT; @@ -553,7 +634,7 @@ int hash_huge_page(struct mm_struct *mm, unsigned long access, /* Insert into the hash table, primary slot */ slot = ppc_md.hpte_insert(hpte_group, va, pa, rflags, 0, - mmu_huge_psize, ssize); + mmu_psize, ssize); /* Primary is full, try the secondary */ if (unlikely(slot == -1)) { @@ -561,7 +642,7 @@ int hash_huge_page(struct mm_struct *mm, unsigned long access, HPTES_PER_GROUP) & ~0x7UL; slot = ppc_md.hpte_insert(hpte_group, va, pa, rflags, HPTE_V_SECONDARY, - mmu_huge_psize, ssize); + mmu_psize, ssize); if (slot == -1) { if (mftb() & 0x1) hpte_group = ((hash & htab_hash_mask) * @@ -598,66 +679,50 @@ void set_huge_psize(int psize) (mmu_psize_defs[psize].shift > MIN_HUGEPTE_SHIFT || mmu_psize_defs[psize].shift == PAGE_SHIFT_64K || mmu_psize_defs[psize].shift == PAGE_SHIFT_16G)) { - /* Return if huge page size is the same as the - * base page size. */ - if (mmu_psize_defs[psize].shift == PAGE_SHIFT) + /* Return if huge page size has already been setup or is the + * same as the base page size. */ + if (mmu_huge_psizes[psize] || + mmu_psize_defs[psize].shift == PAGE_SHIFT) return; + hugetlb_add_hstate(mmu_psize_defs[psize].shift - PAGE_SHIFT); - HPAGE_SHIFT = mmu_psize_defs[psize].shift; - mmu_huge_psize = psize; - - switch (HPAGE_SHIFT) { + switch (mmu_psize_defs[psize].shift) { case PAGE_SHIFT_64K: /* We only allow 64k hpages with 4k base page, * which was checked above, and always put them * at the PMD */ - hugepte_shift = PMD_SHIFT; + hugepte_shift[psize] = PMD_SHIFT; break; case PAGE_SHIFT_16M: /* 16M pages can be at two different levels * of pagestables based on base page size */ if (PAGE_SHIFT == PAGE_SHIFT_64K) - hugepte_shift = PMD_SHIFT; + hugepte_shift[psize] = PMD_SHIFT; else /* 4k base page */ - hugepte_shift = PUD_SHIFT; + hugepte_shift[psize] = PUD_SHIFT; break; case PAGE_SHIFT_16G: /* 16G pages are always at PGD level */ - hugepte_shift = PGDIR_SHIFT; + hugepte_shift[psize] = PGDIR_SHIFT; break; } - hugepte_shift -= HPAGE_SHIFT; + hugepte_shift[psize] -= mmu_psize_defs[psize].shift; } else - HPAGE_SHIFT = 0; + hugepte_shift[psize] = 0; } static int __init hugepage_setup_sz(char *str) { unsigned long long size; - int mmu_psize = -1; + int mmu_psize; int shift; size = memparse(str, &str); shift = __ffs(size); - switch (shift) { -#ifndef CONFIG_PPC_64K_PAGES - case PAGE_SHIFT_64K: - mmu_psize = MMU_PAGE_64K; - break; -#endif - case PAGE_SHIFT_16M: - mmu_psize = MMU_PAGE_16M; - break; - case PAGE_SHIFT_16G: - mmu_psize = MMU_PAGE_16G; - break; - } - - if (mmu_psize >= 0 && mmu_psize_defs[mmu_psize].shift) { + mmu_psize = shift_to_mmu_psize(shift); + if (mmu_psize >= 0 && mmu_psize_defs[mmu_psize].shift) set_huge_psize(mmu_psize); - hugetlb_add_hstate(shift - PAGE_SHIFT); - } else printk(KERN_WARNING "Invalid huge page size specified(%llu)\n", size); @@ -672,16 +737,31 @@ static void zero_ctor(struct kmem_cache *cache, void *addr) static int __init hugetlbpage_init(void) { + unsigned int psize; + if (!cpu_has_feature(CPU_FTR_16M_PAGE)) return -ENODEV; + /* Add supported huge page sizes. Need to change HUGE_MAX_HSTATE + * and adjust PTE_NONCACHE_NUM if the number of supported huge page + * sizes changes. + */ + set_huge_psize(MMU_PAGE_16M); + set_huge_psize(MMU_PAGE_64K); + set_huge_psize(MMU_PAGE_16G); - huge_pgtable_cache = kmem_cache_create("hugepte_cache", - HUGEPTE_TABLE_SIZE, - HUGEPTE_TABLE_SIZE, - 0, - zero_ctor); - if (! huge_pgtable_cache) - panic("hugetlbpage_init(): could not create hugepte cache\n"); + for (psize = 0; psize < MMU_PAGE_COUNT; ++psize) { + if (mmu_huge_psizes[psize]) { + huge_pgtable_cache(psize) = kmem_cache_create( + HUGEPTE_CACHE_NAME(psize), + HUGEPTE_TABLE_SIZE(psize), + HUGEPTE_TABLE_SIZE(psize), + 0, + zero_ctor); + if (!huge_pgtable_cache(psize)) + panic("hugetlbpage_init(): could not create %s"\ + "\n", HUGEPTE_CACHE_NAME(psize)); + } + } return 0; } diff --git a/arch/powerpc/mm/init_64.c b/arch/powerpc/mm/init_64.c index 6ef63caca682..a41bc5aa2043 100644 --- a/arch/powerpc/mm/init_64.c +++ b/arch/powerpc/mm/init_64.c @@ -153,10 +153,10 @@ static const char *pgtable_cache_name[ARRAY_SIZE(pgtable_cache_size)] = { }; #ifdef CONFIG_HUGETLB_PAGE -/* Hugepages need one extra cache, initialized in hugetlbpage.c. We - * can't put into the tables above, because HPAGE_SHIFT is not compile - * time constant. */ -struct kmem_cache *pgtable_cache[ARRAY_SIZE(pgtable_cache_size)+1]; +/* Hugepages need an extra cache per hugepagesize, initialized in + * hugetlbpage.c. We can't put into the tables above, because HPAGE_SHIFT + * is not compile time constant. */ +struct kmem_cache *pgtable_cache[ARRAY_SIZE(pgtable_cache_size)+MMU_PAGE_COUNT]; #else struct kmem_cache *pgtable_cache[ARRAY_SIZE(pgtable_cache_size)]; #endif diff --git a/arch/powerpc/mm/tlb_64.c b/arch/powerpc/mm/tlb_64.c index a01b5c608ff9..409fcc7b63ce 100644 --- a/arch/powerpc/mm/tlb_64.c +++ b/arch/powerpc/mm/tlb_64.c @@ -147,7 +147,7 @@ void hpte_need_flush(struct mm_struct *mm, unsigned long addr, */ if (huge) { #ifdef CONFIG_HUGETLB_PAGE - psize = mmu_huge_psize; + psize = get_slice_psize(mm, addr);; #else BUG(); psize = pte_pagesize_index(mm, addr, pte); /* shutup gcc */ diff --git a/include/asm-powerpc/hugetlb.h b/include/asm-powerpc/hugetlb.h index ca37c4af27b1..26f0d0ab27a5 100644 --- a/include/asm-powerpc/hugetlb.h +++ b/include/asm-powerpc/hugetlb.h @@ -24,9 +24,10 @@ pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, static inline int prepare_hugepage_range(struct file *file, unsigned long addr, unsigned long len) { - if (len & ~HPAGE_MASK) + struct hstate *h = hstate_file(file); + if (len & ~huge_page_mask(h)) return -EINVAL; - if (addr & ~HPAGE_MASK) + if (addr & ~huge_page_mask(h)) return -EINVAL; return 0; } diff --git a/include/asm-powerpc/mmu-hash64.h b/include/asm-powerpc/mmu-hash64.h index b61181aa7746..19c7a9403490 100644 --- a/include/asm-powerpc/mmu-hash64.h +++ b/include/asm-powerpc/mmu-hash64.h @@ -194,9 +194,9 @@ extern int mmu_ci_restrictions; #ifdef CONFIG_HUGETLB_PAGE /* - * The page size index of the huge pages for use by hugetlbfs + * The page size indexes of the huge pages for use by hugetlbfs */ -extern int mmu_huge_psize; +extern unsigned int mmu_huge_psizes[MMU_PAGE_COUNT]; #endif /* CONFIG_HUGETLB_PAGE */ diff --git a/include/asm-powerpc/page_64.h b/include/asm-powerpc/page_64.h index 02fd80710e9d..043bfdfe4f73 100644 --- a/include/asm-powerpc/page_64.h +++ b/include/asm-powerpc/page_64.h @@ -90,6 +90,7 @@ extern unsigned int HPAGE_SHIFT; #define HPAGE_SIZE ((1UL) << HPAGE_SHIFT) #define HPAGE_MASK (~(HPAGE_SIZE - 1)) #define HUGETLB_PAGE_ORDER (HPAGE_SHIFT - PAGE_SHIFT) +#define HUGE_MAX_HSTATE 3 #endif /* __ASSEMBLY__ */ diff --git a/include/asm-powerpc/pgalloc-64.h b/include/asm-powerpc/pgalloc-64.h index 68980990f62a..812a1d8f35cb 100644 --- a/include/asm-powerpc/pgalloc-64.h +++ b/include/asm-powerpc/pgalloc-64.h @@ -22,7 +22,7 @@ extern struct kmem_cache *pgtable_cache[]; #define PUD_CACHE_NUM 1 #define PMD_CACHE_NUM 1 #define HUGEPTE_CACHE_NUM 2 -#define PTE_NONCACHE_NUM 3 /* from GFP rather than kmem_cache */ +#define PTE_NONCACHE_NUM 7 /* from GFP rather than kmem_cache */ static inline pgd_t *pgd_alloc(struct mm_struct *mm) { @@ -119,7 +119,7 @@ static inline void pte_free(struct mm_struct *mm, pgtable_t ptepage) __free_page(ptepage); } -#define PGF_CACHENUM_MASK 0x3 +#define PGF_CACHENUM_MASK 0x7 typedef struct pgtable_free { unsigned long val; From 7f09ca51e925ba62e9ebfd4979f093e97e38adeb Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Wed, 23 Jul 2008 21:27:58 -0700 Subject: [PATCH 188/857] hugetlb: fix a hugepage reservation check for MAP_SHARED When removing a huge page from the hugepage pool for a fault the system checks to see if the mapping requires additional pages to be reserved, and if it does whether there are any unreserved pages remaining. If not, the allocation fails without even attempting to get a page. In order to determine whether to apply this check we call vma_has_private_reserves() which tells us if this vma is MAP_PRIVATE and is the owner. This incorrectly triggers the remaining reservation test for MAP_SHARED mappings which prevents allocation of the final page in the pool even though it is reserved for this mapping. In reality we only want to check this for MAP_PRIVATE mappings where the process is not the original mapper. Replace vma_has_private_reserves() with vma_has_reserves() which indicates whether further reserves are required, and update the caller. Signed-off-by: Mel Gorman Acked-by: Adam Litke Acked-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/hugetlb.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 3e1506b808a3..8c20aed62b9c 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -342,13 +342,13 @@ void reset_vma_resv_huge_pages(struct vm_area_struct *vma) } /* Returns true if the VMA has associated reserve pages */ -static int vma_has_private_reserves(struct vm_area_struct *vma) +static int vma_has_reserves(struct vm_area_struct *vma) { if (vma->vm_flags & VM_SHARED) - return 0; - if (!is_vma_resv_set(vma, HPAGE_RESV_OWNER)) - return 0; - return 1; + return 1; + if (is_vma_resv_set(vma, HPAGE_RESV_OWNER)) + return 1; + return 0; } static void clear_huge_page(struct page *page, @@ -420,7 +420,7 @@ static struct page *dequeue_huge_page_vma(struct hstate *h, * have no page reserves. This check ensures that reservations are * not "stolen". The child may still get SIGKILLed */ - if (!vma_has_private_reserves(vma) && + if (!vma_has_reserves(vma) && h->free_huge_pages - h->resv_huge_pages == 0) return NULL; From 7251ff78b94c2a68d267623d09b32672b20662c1 Mon Sep 17 00:00:00 2001 From: Adam Litke Date: Wed, 23 Jul 2008 21:27:59 -0700 Subject: [PATCH 189/857] hugetlb: quota is not freed for unused reserved private huge pages With shared reservations (and now also with private reservations), we reserve huge pages at mmap time. We also account for the mapping against fs quota to prevent a reservation from being preempted by quota exhaustion. When testing with the libhugetlbfs test suite, I found a problem with quota accounting. FS quota for allocated pages is handled correctly but we are not releasing quota for private pages that were reserved but never allocated. Do this in hugetlb_vm_op_close() at the same time as unused page reservations are released. Signed-off-by: Adam Litke Cc: Mel Gorman Cc: Johannes Weiner Cc: William Lee Irwin III Cc: Hugh Dickins Acked-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/hugetlb.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 8c20aed62b9c..41341c414194 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -1552,8 +1552,10 @@ static void hugetlb_vm_op_close(struct vm_area_struct *vma) kref_put(&reservations->refs, resv_map_release); - if (reserve) + if (reserve) { hugetlb_acct_memory(h, -reserve); + hugetlb_put_quota(vma->vm_file->f_mapping, reserve); + } } } From 223e8dc9249c9e15f6c8b638d73fcad78ccb0a88 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 23 Jul 2008 21:28:00 -0700 Subject: [PATCH 190/857] bootmem: reorder code to match new bootmem structure This only reorders functions so that further patches will be easier to read. No code changed. Signed-off-by: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/bootmem.h | 86 +++++----- mm/bootmem.c | 362 ++++++++++++++++++++-------------------- 2 files changed, 225 insertions(+), 223 deletions(-) diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h index f352c5f125b4..5000fd70b04f 100644 --- a/include/linux/bootmem.h +++ b/include/linux/bootmem.h @@ -41,36 +41,62 @@ typedef struct bootmem_data { extern bootmem_data_t bootmem_node_data[]; extern unsigned long bootmem_bootmap_pages(unsigned long); + +extern unsigned long init_bootmem_node(pg_data_t *pgdat, + unsigned long freepfn, + unsigned long startpfn, + unsigned long endpfn); extern unsigned long init_bootmem(unsigned long addr, unsigned long memend); + +extern unsigned long free_all_bootmem_node(pg_data_t *pgdat); +extern unsigned long free_all_bootmem(void); + +extern void free_bootmem_node(pg_data_t *pgdat, + unsigned long addr, + unsigned long size); extern void free_bootmem(unsigned long addr, unsigned long size); -extern void *__alloc_bootmem(unsigned long size, + +/* + * Flags for reserve_bootmem (also if CONFIG_HAVE_ARCH_BOOTMEM_NODE, + * the architecture-specific code should honor this). + * + * If flags is 0, then the return value is always 0 (success). If + * flags contains BOOTMEM_EXCLUSIVE, then -EBUSY is returned if the + * memory already was reserved. + */ +#define BOOTMEM_DEFAULT 0 +#define BOOTMEM_EXCLUSIVE (1<<0) + +extern int reserve_bootmem_node(pg_data_t *pgdat, + unsigned long physaddr, + unsigned long size, + int flags); +#ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE +extern int reserve_bootmem(unsigned long addr, unsigned long size, int flags); +#endif + +extern void *__alloc_bootmem_nopanic(unsigned long size, unsigned long align, unsigned long goal); -extern void *__alloc_bootmem_nopanic(unsigned long size, +extern void *__alloc_bootmem(unsigned long size, unsigned long align, unsigned long goal); extern void *__alloc_bootmem_low(unsigned long size, unsigned long align, unsigned long goal); +extern void *__alloc_bootmem_node(pg_data_t *pgdat, + unsigned long size, + unsigned long align, + unsigned long goal); +extern void *__alloc_bootmem_node_nopanic(pg_data_t *pgdat, + unsigned long size, + unsigned long align, + unsigned long goal); extern void *__alloc_bootmem_low_node(pg_data_t *pgdat, unsigned long size, unsigned long align, unsigned long goal); - -/* - * flags for reserve_bootmem (also if CONFIG_HAVE_ARCH_BOOTMEM_NODE, - * the architecture-specific code should honor this) - */ -#define BOOTMEM_DEFAULT 0 -#define BOOTMEM_EXCLUSIVE (1<<0) - #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE -/* - * If flags is 0, then the return value is always 0 (success). If - * flags contains BOOTMEM_EXCLUSIVE, then -EBUSY is returned if the - * memory already was reserved. - */ -extern int reserve_bootmem(unsigned long addr, unsigned long size, int flags); #define alloc_bootmem(x) \ __alloc_bootmem(x, SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS)) #define alloc_bootmem_low(x) \ @@ -83,38 +109,16 @@ extern int reserve_bootmem(unsigned long addr, unsigned long size, int flags); extern int reserve_bootmem_generic(unsigned long addr, unsigned long size, int flags); -extern unsigned long free_all_bootmem(void); -extern unsigned long free_all_bootmem_node(pg_data_t *pgdat); -extern void *__alloc_bootmem_node(pg_data_t *pgdat, - unsigned long size, - unsigned long align, - unsigned long goal); -extern void *__alloc_bootmem_node_nopanic(pg_data_t *pgdat, - unsigned long size, - unsigned long align, - unsigned long goal); -extern unsigned long init_bootmem_node(pg_data_t *pgdat, - unsigned long freepfn, - unsigned long startpfn, - unsigned long endpfn); -extern int reserve_bootmem_node(pg_data_t *pgdat, - unsigned long physaddr, - unsigned long size, - int flags); -extern void free_bootmem_node(pg_data_t *pgdat, - unsigned long addr, - unsigned long size); -extern void *alloc_bootmem_section(unsigned long size, - unsigned long section_nr); -#ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE #define alloc_bootmem_node(pgdat, x) \ __alloc_bootmem_node(pgdat, x, SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS)) #define alloc_bootmem_pages_node(pgdat, x) \ __alloc_bootmem_node(pgdat, x, PAGE_SIZE, __pa(MAX_DMA_ADDRESS)) #define alloc_bootmem_low_pages_node(pgdat, x) \ __alloc_bootmem_low_node(pgdat, x, PAGE_SIZE, 0) -#endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */ + +extern void *alloc_bootmem_section(unsigned long size, + unsigned long section_nr); #ifdef CONFIG_HAVE_ARCH_ALLOC_REMAP extern void *alloc_remap(int nid, unsigned long size); diff --git a/mm/bootmem.c b/mm/bootmem.c index 9ac972535fff..24eacf52c50e 100644 --- a/mm/bootmem.c +++ b/mm/bootmem.c @@ -38,6 +38,19 @@ unsigned long saved_max_pfn; bootmem_data_t bootmem_node_data[MAX_NUMNODES] __initdata; +/* + * Given an initialised bdata, it returns the size of the boot bitmap + */ +static unsigned long __init get_mapsize(bootmem_data_t *bdata) +{ + unsigned long mapsize; + unsigned long start = PFN_DOWN(bdata->node_boot_start); + unsigned long end = bdata->node_low_pfn; + + mapsize = ((end - start) + 7) / 8; + return ALIGN(mapsize, sizeof(long)); +} + /* return the number of _pages_ that will be allocated for the boot bitmap */ unsigned long __init bootmem_bootmap_pages(unsigned long pages) { @@ -71,19 +84,6 @@ static void __init link_bootmem(bootmem_data_t *bdata) list_add_tail(&bdata->list, &bdata_list); } -/* - * Given an initialised bdata, it returns the size of the boot bitmap - */ -static unsigned long __init get_mapsize(bootmem_data_t *bdata) -{ - unsigned long mapsize; - unsigned long start = PFN_DOWN(bdata->node_boot_start); - unsigned long end = bdata->node_low_pfn; - - mapsize = ((end - start) + 7) / 8; - return ALIGN(mapsize, sizeof(long)); -} - /* * Called once to set up the allocator itself. */ @@ -108,6 +108,146 @@ static unsigned long __init init_bootmem_core(bootmem_data_t *bdata, return mapsize; } +unsigned long __init init_bootmem_node(pg_data_t *pgdat, unsigned long freepfn, + unsigned long startpfn, unsigned long endpfn) +{ + return init_bootmem_core(pgdat->bdata, freepfn, startpfn, endpfn); +} + +unsigned long __init init_bootmem(unsigned long start, unsigned long pages) +{ + max_low_pfn = pages; + min_low_pfn = start; + return init_bootmem_core(NODE_DATA(0)->bdata, start, 0, pages); +} + +static unsigned long __init free_all_bootmem_core(bootmem_data_t *bdata) +{ + struct page *page; + unsigned long pfn; + unsigned long i, count; + unsigned long idx; + unsigned long *map; + int gofast = 0; + + BUG_ON(!bdata->node_bootmem_map); + + count = 0; + /* first extant page of the node */ + pfn = PFN_DOWN(bdata->node_boot_start); + idx = bdata->node_low_pfn - pfn; + map = bdata->node_bootmem_map; + /* + * Check if we are aligned to BITS_PER_LONG pages. If so, we might + * be able to free page orders of that size at once. + */ + if (!(pfn & (BITS_PER_LONG-1))) + gofast = 1; + + for (i = 0; i < idx; ) { + unsigned long v = ~map[i / BITS_PER_LONG]; + + if (gofast && v == ~0UL) { + int order; + + page = pfn_to_page(pfn); + count += BITS_PER_LONG; + order = ffs(BITS_PER_LONG) - 1; + __free_pages_bootmem(page, order); + i += BITS_PER_LONG; + page += BITS_PER_LONG; + } else if (v) { + unsigned long m; + + page = pfn_to_page(pfn); + for (m = 1; m && i < idx; m<<=1, page++, i++) { + if (v & m) { + count++; + __free_pages_bootmem(page, 0); + } + } + } else { + i += BITS_PER_LONG; + } + pfn += BITS_PER_LONG; + } + + /* + * Now free the allocator bitmap itself, it's not + * needed anymore: + */ + page = virt_to_page(bdata->node_bootmem_map); + idx = (get_mapsize(bdata) + PAGE_SIZE-1) >> PAGE_SHIFT; + for (i = 0; i < idx; i++, page++) + __free_pages_bootmem(page, 0); + count += i; + bdata->node_bootmem_map = NULL; + + return count; +} + +unsigned long __init free_all_bootmem_node(pg_data_t *pgdat) +{ + register_page_bootmem_info_node(pgdat); + return free_all_bootmem_core(pgdat->bdata); +} + +unsigned long __init free_all_bootmem(void) +{ + return free_all_bootmem_core(NODE_DATA(0)->bdata); +} + +static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr, + unsigned long size) +{ + unsigned long sidx, eidx; + unsigned long i; + + BUG_ON(!size); + + /* out range */ + if (addr + size < bdata->node_boot_start || + PFN_DOWN(addr) > bdata->node_low_pfn) + return; + /* + * round down end of usable mem, partially free pages are + * considered reserved. + */ + + if (addr >= bdata->node_boot_start && addr < bdata->last_success) + bdata->last_success = addr; + + /* + * Round up to index to the range. + */ + if (PFN_UP(addr) > PFN_DOWN(bdata->node_boot_start)) + sidx = PFN_UP(addr) - PFN_DOWN(bdata->node_boot_start); + else + sidx = 0; + + eidx = PFN_DOWN(addr + size - bdata->node_boot_start); + if (eidx > bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start)) + eidx = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start); + + for (i = sidx; i < eidx; i++) { + if (unlikely(!test_and_clear_bit(i, bdata->node_bootmem_map))) + BUG(); + } +} + +void __init free_bootmem_node(pg_data_t *pgdat, unsigned long physaddr, + unsigned long size) +{ + free_bootmem_core(pgdat->bdata, physaddr, size); +} + +void __init free_bootmem(unsigned long addr, unsigned long size) +{ + bootmem_data_t *bdata; + list_for_each_entry(bdata, &bdata_list, list) + free_bootmem_core(bdata, addr, size); +} + /* * Marks a particular physical memory range as unallocatable. Usable RAM * might be used for boot-time allocations - or it might get added @@ -183,44 +323,37 @@ static void __init reserve_bootmem_core(bootmem_data_t *bdata, } } -static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr, - unsigned long size) +int __init reserve_bootmem_node(pg_data_t *pgdat, unsigned long physaddr, + unsigned long size, int flags) { - unsigned long sidx, eidx; - unsigned long i; + int ret; - BUG_ON(!size); - - /* out range */ - if (addr + size < bdata->node_boot_start || - PFN_DOWN(addr) > bdata->node_low_pfn) - return; - /* - * round down end of usable mem, partially free pages are - * considered reserved. - */ - - if (addr >= bdata->node_boot_start && addr < bdata->last_success) - bdata->last_success = addr; - - /* - * Round up to index to the range. - */ - if (PFN_UP(addr) > PFN_DOWN(bdata->node_boot_start)) - sidx = PFN_UP(addr) - PFN_DOWN(bdata->node_boot_start); - else - sidx = 0; - - eidx = PFN_DOWN(addr + size - bdata->node_boot_start); - if (eidx > bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start)) - eidx = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start); - - for (i = sidx; i < eidx; i++) { - if (unlikely(!test_and_clear_bit(i, bdata->node_bootmem_map))) - BUG(); - } + ret = can_reserve_bootmem_core(pgdat->bdata, physaddr, size, flags); + if (ret < 0) + return -ENOMEM; + reserve_bootmem_core(pgdat->bdata, physaddr, size, flags); + return 0; } +#ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE +int __init reserve_bootmem(unsigned long addr, unsigned long size, + int flags) +{ + bootmem_data_t *bdata; + int ret; + + list_for_each_entry(bdata, &bdata_list, list) { + ret = can_reserve_bootmem_core(bdata, addr, size, flags); + if (ret < 0) + return ret; + } + list_for_each_entry(bdata, &bdata_list, list) + reserve_bootmem_core(bdata, addr, size, flags); + + return 0; +} +#endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */ + /* * We 'merge' subsequent allocations to save space. We might 'lose' * some fraction of a page if allocations cannot be satisfied due to @@ -371,140 +504,6 @@ alloc_bootmem_core(struct bootmem_data *bdata, unsigned long size, return ret; } -static unsigned long __init free_all_bootmem_core(bootmem_data_t *bdata) -{ - struct page *page; - unsigned long pfn; - unsigned long i, count; - unsigned long idx; - unsigned long *map; - int gofast = 0; - - BUG_ON(!bdata->node_bootmem_map); - - count = 0; - /* first extant page of the node */ - pfn = PFN_DOWN(bdata->node_boot_start); - idx = bdata->node_low_pfn - pfn; - map = bdata->node_bootmem_map; - /* - * Check if we are aligned to BITS_PER_LONG pages. If so, we might - * be able to free page orders of that size at once. - */ - if (!(pfn & (BITS_PER_LONG-1))) - gofast = 1; - - for (i = 0; i < idx; ) { - unsigned long v = ~map[i / BITS_PER_LONG]; - - if (gofast && v == ~0UL) { - int order; - - page = pfn_to_page(pfn); - count += BITS_PER_LONG; - order = ffs(BITS_PER_LONG) - 1; - __free_pages_bootmem(page, order); - i += BITS_PER_LONG; - page += BITS_PER_LONG; - } else if (v) { - unsigned long m; - - page = pfn_to_page(pfn); - for (m = 1; m && i < idx; m<<=1, page++, i++) { - if (v & m) { - count++; - __free_pages_bootmem(page, 0); - } - } - } else { - i += BITS_PER_LONG; - } - pfn += BITS_PER_LONG; - } - - /* - * Now free the allocator bitmap itself, it's not - * needed anymore: - */ - page = virt_to_page(bdata->node_bootmem_map); - idx = (get_mapsize(bdata) + PAGE_SIZE-1) >> PAGE_SHIFT; - for (i = 0; i < idx; i++, page++) - __free_pages_bootmem(page, 0); - count += i; - bdata->node_bootmem_map = NULL; - - return count; -} - -unsigned long __init init_bootmem_node(pg_data_t *pgdat, unsigned long freepfn, - unsigned long startpfn, unsigned long endpfn) -{ - return init_bootmem_core(pgdat->bdata, freepfn, startpfn, endpfn); -} - -int __init reserve_bootmem_node(pg_data_t *pgdat, unsigned long physaddr, - unsigned long size, int flags) -{ - int ret; - - ret = can_reserve_bootmem_core(pgdat->bdata, physaddr, size, flags); - if (ret < 0) - return -ENOMEM; - reserve_bootmem_core(pgdat->bdata, physaddr, size, flags); - - return 0; -} - -void __init free_bootmem_node(pg_data_t *pgdat, unsigned long physaddr, - unsigned long size) -{ - free_bootmem_core(pgdat->bdata, physaddr, size); -} - -unsigned long __init free_all_bootmem_node(pg_data_t *pgdat) -{ - register_page_bootmem_info_node(pgdat); - return free_all_bootmem_core(pgdat->bdata); -} - -unsigned long __init init_bootmem(unsigned long start, unsigned long pages) -{ - max_low_pfn = pages; - min_low_pfn = start; - return init_bootmem_core(NODE_DATA(0)->bdata, start, 0, pages); -} - -#ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE -int __init reserve_bootmem(unsigned long addr, unsigned long size, - int flags) -{ - bootmem_data_t *bdata; - int ret; - - list_for_each_entry(bdata, &bdata_list, list) { - ret = can_reserve_bootmem_core(bdata, addr, size, flags); - if (ret < 0) - return ret; - } - list_for_each_entry(bdata, &bdata_list, list) - reserve_bootmem_core(bdata, addr, size, flags); - - return 0; -} -#endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */ - -void __init free_bootmem(unsigned long addr, unsigned long size) -{ - bootmem_data_t *bdata; - list_for_each_entry(bdata, &bdata_list, list) - free_bootmem_core(bdata, addr, size); -} - -unsigned long __init free_all_bootmem(void) -{ - return free_all_bootmem_core(NODE_DATA(0)->bdata); -} - void * __init __alloc_bootmem_nopanic(unsigned long size, unsigned long align, unsigned long goal) { @@ -534,7 +533,6 @@ void * __init __alloc_bootmem(unsigned long size, unsigned long align, return NULL; } - void * __init __alloc_bootmem_node(pg_data_t *pgdat, unsigned long size, unsigned long align, unsigned long goal) { From 57cfc29efac6670355ee0e107c8dbae8237d406b Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 23 Jul 2008 21:28:00 -0700 Subject: [PATCH 191/857] bootmem: clean up bootmem.c file header Change the description, move a misplaced comment about the allocator itself and add me to the list of copyright holders. Signed-off-by: Johannes Weiner Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/bootmem.c | 14 +++++--------- 1 file changed, 5 insertions(+), 9 deletions(-) diff --git a/mm/bootmem.c b/mm/bootmem.c index 24eacf52c50e..286e12c536ae 100644 --- a/mm/bootmem.c +++ b/mm/bootmem.c @@ -1,12 +1,12 @@ /* - * linux/mm/bootmem.c + * bootmem - A boot-time physical memory allocator and configurator * * Copyright (C) 1999 Ingo Molnar - * Discontiguous memory support, Kanoj Sarcar, SGI, Nov 1999 + * 1999 Kanoj Sarcar, SGI + * 2008 Johannes Weiner * - * simple boot-time physical memory area allocator and - * free memory collector. It's used to deal with reserved - * system memory and memory holes as well. + * Access to this subsystem has to be serialized externally (which is true + * for the boot process anyway). */ #include #include @@ -19,10 +19,6 @@ #include "internal.h" -/* - * Access to this subsystem has to be serialized externally. (this is - * true for the boot process anyway) - */ unsigned long max_low_pfn; unsigned long min_low_pfn; unsigned long max_pfn; From a66fd7daec1f40c1f0eac466f0da9206b615fe2a Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 23 Jul 2008 21:28:01 -0700 Subject: [PATCH 192/857] bootmem: add documentation to API functions Signed-off-by: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/bootmem.c | 150 ++++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 149 insertions(+), 1 deletion(-) diff --git a/mm/bootmem.c b/mm/bootmem.c index 286e12c536ae..105ad4cff2e1 100644 --- a/mm/bootmem.c +++ b/mm/bootmem.c @@ -47,7 +47,10 @@ static unsigned long __init get_mapsize(bootmem_data_t *bdata) return ALIGN(mapsize, sizeof(long)); } -/* return the number of _pages_ that will be allocated for the boot bitmap */ +/** + * bootmem_bootmap_pages - calculate bitmap size in pages + * @pages: number of pages the bitmap has to represent + */ unsigned long __init bootmem_bootmap_pages(unsigned long pages) { unsigned long mapsize; @@ -104,12 +107,28 @@ static unsigned long __init init_bootmem_core(bootmem_data_t *bdata, return mapsize; } +/** + * init_bootmem_node - register a node as boot memory + * @pgdat: node to register + * @freepfn: pfn where the bitmap for this node is to be placed + * @startpfn: first pfn on the node + * @endpfn: first pfn after the node + * + * Returns the number of bytes needed to hold the bitmap for this node. + */ unsigned long __init init_bootmem_node(pg_data_t *pgdat, unsigned long freepfn, unsigned long startpfn, unsigned long endpfn) { return init_bootmem_core(pgdat->bdata, freepfn, startpfn, endpfn); } +/** + * init_bootmem - register boot memory + * @start: pfn where the bitmap is to be placed + * @pages: number of available physical pages + * + * Returns the number of bytes needed to hold the bitmap. + */ unsigned long __init init_bootmem(unsigned long start, unsigned long pages) { max_low_pfn = pages; @@ -182,12 +201,23 @@ static unsigned long __init free_all_bootmem_core(bootmem_data_t *bdata) return count; } +/** + * free_all_bootmem_node - release a node's free pages to the buddy allocator + * @pgdat: node to be released + * + * Returns the number of pages actually released. + */ unsigned long __init free_all_bootmem_node(pg_data_t *pgdat) { register_page_bootmem_info_node(pgdat); return free_all_bootmem_core(pgdat->bdata); } +/** + * free_all_bootmem - release free pages to the buddy allocator + * + * Returns the number of pages actually released. + */ unsigned long __init free_all_bootmem(void) { return free_all_bootmem_core(NODE_DATA(0)->bdata); @@ -231,12 +261,32 @@ static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr, } } +/** + * free_bootmem_node - mark a page range as usable + * @pgdat: node the range resides on + * @physaddr: starting address of the range + * @size: size of the range in bytes + * + * Partial pages will be considered reserved and left as they are. + * + * Only physical pages that actually reside on @pgdat are marked. + */ void __init free_bootmem_node(pg_data_t *pgdat, unsigned long physaddr, unsigned long size) { free_bootmem_core(pgdat->bdata, physaddr, size); } +/** + * free_bootmem - mark a page range as usable + * @addr: starting address of the range + * @size: size of the range in bytes + * + * Partial pages will be considered reserved and left as they are. + * + * All physical pages within the range are marked, no matter what + * node they reside on. + */ void __init free_bootmem(unsigned long addr, unsigned long size) { bootmem_data_t *bdata; @@ -319,6 +369,17 @@ static void __init reserve_bootmem_core(bootmem_data_t *bdata, } } +/** + * reserve_bootmem_node - mark a page range as reserved + * @pgdat: node the range resides on + * @physaddr: starting address of the range + * @size: size of the range in bytes + * @flags: reservation flags (see linux/bootmem.h) + * + * Partial pages will be reserved. + * + * Only physical pages that actually reside on @pgdat are marked. + */ int __init reserve_bootmem_node(pg_data_t *pgdat, unsigned long physaddr, unsigned long size, int flags) { @@ -332,6 +393,17 @@ int __init reserve_bootmem_node(pg_data_t *pgdat, unsigned long physaddr, } #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE +/** + * reserve_bootmem - mark a page range as usable + * @addr: starting address of the range + * @size: size of the range in bytes + * @flags: reservation flags (see linux/bootmem.h) + * + * Partial pages will be reserved. + * + * All physical pages within the range are marked, no matter what + * node they reside on. + */ int __init reserve_bootmem(unsigned long addr, unsigned long size, int flags) { @@ -500,6 +572,19 @@ alloc_bootmem_core(struct bootmem_data *bdata, unsigned long size, return ret; } +/** + * __alloc_bootmem_nopanic - allocate boot memory without panicking + * @size: size of the request in bytes + * @align: alignment of the region + * @goal: preferred starting address of the region + * + * The goal is dropped if it can not be satisfied and the allocation will + * fall back to memory below @goal. + * + * Allocation may happen on any node in the system. + * + * Returns NULL on failure. + */ void * __init __alloc_bootmem_nopanic(unsigned long size, unsigned long align, unsigned long goal) { @@ -514,6 +599,19 @@ void * __init __alloc_bootmem_nopanic(unsigned long size, unsigned long align, return NULL; } +/** + * __alloc_bootmem - allocate boot memory + * @size: size of the request in bytes + * @align: alignment of the region + * @goal: preferred starting address of the region + * + * The goal is dropped if it can not be satisfied and the allocation will + * fall back to memory below @goal. + * + * Allocation may happen on any node in the system. + * + * The function panics if the request can not be satisfied. + */ void * __init __alloc_bootmem(unsigned long size, unsigned long align, unsigned long goal) { @@ -529,6 +627,21 @@ void * __init __alloc_bootmem(unsigned long size, unsigned long align, return NULL; } +/** + * __alloc_bootmem_node - allocate boot memory from a specific node + * @pgdat: node to allocate from + * @size: size of the request in bytes + * @align: alignment of the region + * @goal: preferred starting address of the region + * + * The goal is dropped if it can not be satisfied and the allocation will + * fall back to memory below @goal. + * + * Allocation may fall back to any node in the system if the specified node + * can not hold the requested memory. + * + * The function panics if the request can not be satisfied. + */ void * __init __alloc_bootmem_node(pg_data_t *pgdat, unsigned long size, unsigned long align, unsigned long goal) { @@ -542,6 +655,13 @@ void * __init __alloc_bootmem_node(pg_data_t *pgdat, unsigned long size, } #ifdef CONFIG_SPARSEMEM +/** + * alloc_bootmem_section - allocate boot memory from a specific section + * @size: size of the request in bytes + * @section_nr: sparse map section to allocate from + * + * Return NULL on failure. + */ void * __init alloc_bootmem_section(unsigned long size, unsigned long section_nr) { @@ -588,6 +708,19 @@ void * __init __alloc_bootmem_node_nopanic(pg_data_t *pgdat, unsigned long size, #define ARCH_LOW_ADDRESS_LIMIT 0xffffffffUL #endif +/** + * __alloc_bootmem_low - allocate low boot memory + * @size: size of the request in bytes + * @align: alignment of the region + * @goal: preferred starting address of the region + * + * The goal is dropped if it can not be satisfied and the allocation will + * fall back to memory below @goal. + * + * Allocation may happen on any node in the system. + * + * The function panics if the request can not be satisfied. + */ void * __init __alloc_bootmem_low(unsigned long size, unsigned long align, unsigned long goal) { @@ -609,6 +742,21 @@ void * __init __alloc_bootmem_low(unsigned long size, unsigned long align, return NULL; } +/** + * __alloc_bootmem_low_node - allocate low boot memory from a specific node + * @pgdat: node to allocate from + * @size: size of the request in bytes + * @align: alignment of the region + * @goal: preferred starting address of the region + * + * The goal is dropped if it can not be satisfied and the allocation will + * fall back to memory below @goal. + * + * Allocation may fall back to any node in the system if the specified node + * can not hold the requested memory. + * + * The function panics if the request can not be satisfied. + */ void * __init __alloc_bootmem_low_node(pg_data_t *pgdat, unsigned long size, unsigned long align, unsigned long goal) { From 2e5237daf0cc3c8d87762f53f704dc54fa91dcf6 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 23 Jul 2008 21:28:02 -0700 Subject: [PATCH 193/857] bootmem: add debugging framework Introduce the bootmem_debug kernel parameter that enables very verbose diagnostics regarding all range operations of bootmem as well as the initialization and release of nodes. [akpm@linux-foundation.org: fix printk warnings] Signed-off-by: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/bootmem.c | 51 ++++++++++++++++++++++++++++++++++++++++++++------- 1 file changed, 44 insertions(+), 7 deletions(-) diff --git a/mm/bootmem.c b/mm/bootmem.c index 105ad4cff2e1..4e085ee1d98e 100644 --- a/mm/bootmem.c +++ b/mm/bootmem.c @@ -34,6 +34,22 @@ unsigned long saved_max_pfn; bootmem_data_t bootmem_node_data[MAX_NUMNODES] __initdata; +static int bootmem_debug; + +static int __init bootmem_debug_setup(char *buf) +{ + bootmem_debug = 1; + return 0; +} +early_param("bootmem_debug", bootmem_debug_setup); + +#define bdebug(fmt, args...) ({ \ + if (unlikely(bootmem_debug)) \ + printk(KERN_INFO \ + "bootmem::%s " fmt, \ + __FUNCTION__, ## args); \ +}) + /* * Given an initialised bdata, it returns the size of the boot bitmap */ @@ -104,6 +120,9 @@ static unsigned long __init init_bootmem_core(bootmem_data_t *bdata, mapsize = get_mapsize(bdata); memset(bdata->node_bootmem_map, 0xff, mapsize); + bdebug("nid=%td start=%lx map=%lx end=%lx mapsize=%lx\n", + bdata - bootmem_node_data, start, mapstart, end, mapsize); + return mapsize; } @@ -198,6 +217,8 @@ static unsigned long __init free_all_bootmem_core(bootmem_data_t *bdata) count += i; bdata->node_bootmem_map = NULL; + bdebug("nid=%td released=%lx\n", bdata - bootmem_node_data, count); + return count; } @@ -255,6 +276,10 @@ static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr, if (eidx > bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start)) eidx = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start); + bdebug("nid=%td start=%lx end=%lx\n", bdata - bootmem_node_data, + sidx + PFN_DOWN(bdata->node_boot_start), + eidx + PFN_DOWN(bdata->node_boot_start)); + for (i = sidx; i < eidx; i++) { if (unlikely(!test_and_clear_bit(i, bdata->node_bootmem_map))) BUG(); @@ -360,13 +385,16 @@ static void __init reserve_bootmem_core(bootmem_data_t *bdata, if (eidx > bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start)) eidx = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start); - for (i = sidx; i < eidx; i++) { - if (test_and_set_bit(i, bdata->node_bootmem_map)) { -#ifdef CONFIG_DEBUG_BOOTMEM - printk("hm, page %08lx reserved twice.\n", i*PAGE_SIZE); -#endif - } - } + bdebug("nid=%td start=%lx end=%lx flags=%x\n", + bdata - bootmem_node_data, + sidx + PFN_DOWN(bdata->node_boot_start), + eidx + PFN_DOWN(bdata->node_boot_start), + flags); + + for (i = sidx; i < eidx; i++) + if (test_and_set_bit(i, bdata->node_bootmem_map)) + bdebug("hm, page %lx reserved twice.\n", + PFN_DOWN(bdata->node_boot_start) + i); } /** @@ -455,6 +483,10 @@ alloc_bootmem_core(struct bootmem_data *bdata, unsigned long size, if (!bdata->node_bootmem_map) return NULL; + bdebug("nid=%td size=%lx [%lu pages] align=%lx goal=%lx limit=%lx\n", + bdata - bootmem_node_data, size, PAGE_ALIGN(size) >> PAGE_SHIFT, + align, goal, limit); + /* bdata->node_boot_start is supposed to be (12+6)bits alignment on x86_64 ? */ node_boot_start = bdata->node_boot_start; node_bootmem_map = bdata->node_bootmem_map; @@ -562,6 +594,11 @@ alloc_bootmem_core(struct bootmem_data *bdata, unsigned long size, ret = phys_to_virt(start * PAGE_SIZE + node_boot_start); } + bdebug("nid=%td start=%lx end=%lx\n", + bdata - bootmem_node_data, + start + PFN_DOWN(bdata->node_boot_start), + start + areasize + PFN_DOWN(bdata->node_boot_start)); + /* * Reserve the area now: */ From df049a5f41a3b2eee2131221959e3b558ba7c705 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 23 Jul 2008 21:28:02 -0700 Subject: [PATCH 194/857] bootmem: revisit bitmap size calculations Reincarnate get_mapsize as bootmap_bytes and implement bootmem_bootmap_pages on top of it. Adjust users of these helpers and make free_all_bootmem_core use bootmem_bootmap_pages instead of open-coding it. Signed-off-by: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/bootmem.c | 27 +++++++++------------------ 1 file changed, 9 insertions(+), 18 deletions(-) diff --git a/mm/bootmem.c b/mm/bootmem.c index 4e085ee1d98e..484849bfc8c4 100644 --- a/mm/bootmem.c +++ b/mm/bootmem.c @@ -50,17 +50,11 @@ early_param("bootmem_debug", bootmem_debug_setup); __FUNCTION__, ## args); \ }) -/* - * Given an initialised bdata, it returns the size of the boot bitmap - */ -static unsigned long __init get_mapsize(bootmem_data_t *bdata) +static unsigned long __init bootmap_bytes(unsigned long pages) { - unsigned long mapsize; - unsigned long start = PFN_DOWN(bdata->node_boot_start); - unsigned long end = bdata->node_low_pfn; + unsigned long bytes = (pages + 7) / 8; - mapsize = ((end - start) + 7) / 8; - return ALIGN(mapsize, sizeof(long)); + return ALIGN(bytes, sizeof(long)); } /** @@ -69,13 +63,9 @@ static unsigned long __init get_mapsize(bootmem_data_t *bdata) */ unsigned long __init bootmem_bootmap_pages(unsigned long pages) { - unsigned long mapsize; + unsigned long bytes = bootmap_bytes(pages); - mapsize = (pages+7)/8; - mapsize = (mapsize + ~PAGE_MASK) & PAGE_MASK; - mapsize >>= PAGE_SHIFT; - - return mapsize; + return PAGE_ALIGN(bytes) >> PAGE_SHIFT; } /* @@ -117,7 +107,7 @@ static unsigned long __init init_bootmem_core(bootmem_data_t *bdata, * Initially all pages are reserved - setup_arch() has to * register free RAM areas explicitly. */ - mapsize = get_mapsize(bdata); + mapsize = bootmap_bytes(end - start); memset(bdata->node_bootmem_map, 0xff, mapsize); bdebug("nid=%td start=%lx map=%lx end=%lx mapsize=%lx\n", @@ -160,7 +150,7 @@ static unsigned long __init free_all_bootmem_core(bootmem_data_t *bdata) struct page *page; unsigned long pfn; unsigned long i, count; - unsigned long idx; + unsigned long idx, pages; unsigned long *map; int gofast = 0; @@ -211,7 +201,8 @@ static unsigned long __init free_all_bootmem_core(bootmem_data_t *bdata) * needed anymore: */ page = virt_to_page(bdata->node_bootmem_map); - idx = (get_mapsize(bdata) + PAGE_SIZE-1) >> PAGE_SHIFT; + pages = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start); + idx = bootmem_bootmap_pages(pages); for (i = 0; i < idx; i++, page++) __free_pages_bootmem(page, 0); count += i; From 636cc40cb79f511d9caa27ef098a83e4fa4971fb Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 23 Jul 2008 21:28:03 -0700 Subject: [PATCH 195/857] bootmem: revisit bootmem descriptor list handling link_bootmem handles an insertion of a new descriptor into the sorted list in more or less three explicit branches; empty list, insert in between and append. These cases can be expressed implicite. Also mark the sorted list as initdata as it can be thrown away after boot as well. Signed-off-by: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/bootmem.c | 23 ++++++++++------------- 1 file changed, 10 insertions(+), 13 deletions(-) diff --git a/mm/bootmem.c b/mm/bootmem.c index 484849bfc8c4..9da7d4097810 100644 --- a/mm/bootmem.c +++ b/mm/bootmem.c @@ -23,7 +23,6 @@ unsigned long max_low_pfn; unsigned long min_low_pfn; unsigned long max_pfn; -static LIST_HEAD(bdata_list); #ifdef CONFIG_CRASH_DUMP /* * If we have booted due to a crash, max_pfn will be a very low value. We need @@ -34,6 +33,8 @@ unsigned long saved_max_pfn; bootmem_data_t bootmem_node_data[MAX_NUMNODES] __initdata; +static struct list_head bdata_list __initdata = LIST_HEAD_INIT(bdata_list); + static int bootmem_debug; static int __init bootmem_debug_setup(char *buf) @@ -73,20 +74,16 @@ unsigned long __init bootmem_bootmap_pages(unsigned long pages) */ static void __init link_bootmem(bootmem_data_t *bdata) { - bootmem_data_t *ent; + struct list_head *iter; - if (list_empty(&bdata_list)) { - list_add(&bdata->list, &bdata_list); - return; + list_for_each(iter, &bdata_list) { + bootmem_data_t *ent; + + ent = list_entry(iter, bootmem_data_t, list); + if (bdata->node_boot_start < ent->node_boot_start) + break; } - /* insert in order */ - list_for_each_entry(ent, &bdata_list, list) { - if (bdata->node_boot_start < ent->node_boot_start) { - list_add_tail(&bdata->list, &ent->list); - return; - } - } - list_add_tail(&bdata->list, &bdata_list); + list_add_tail(&bdata->list, iter); } /* From 41546c17418fba08ece978bad72a33072715b8f3 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 23 Jul 2008 21:28:03 -0700 Subject: [PATCH 196/857] bootmem: clean up free_all_bootmem_core Rewrite the code in a more concise way using less variables. [akpm@linux-foundation.org: fix printk warnings] Signed-off-by: Johannes Weiner Cc: Ingo Molnar Cc: Yinghai Lu Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/bootmem.c | 83 ++++++++++++++++++++++++---------------------------- 1 file changed, 38 insertions(+), 45 deletions(-) diff --git a/mm/bootmem.c b/mm/bootmem.c index 9da7d4097810..300d126ec533 100644 --- a/mm/bootmem.c +++ b/mm/bootmem.c @@ -144,66 +144,59 @@ unsigned long __init init_bootmem(unsigned long start, unsigned long pages) static unsigned long __init free_all_bootmem_core(bootmem_data_t *bdata) { + int aligned; struct page *page; - unsigned long pfn; - unsigned long i, count; - unsigned long idx, pages; - unsigned long *map; - int gofast = 0; + unsigned long start, end, pages, count = 0; - BUG_ON(!bdata->node_bootmem_map); + if (!bdata->node_bootmem_map) + return 0; + + start = PFN_DOWN(bdata->node_boot_start); + end = bdata->node_low_pfn; - count = 0; - /* first extant page of the node */ - pfn = PFN_DOWN(bdata->node_boot_start); - idx = bdata->node_low_pfn - pfn; - map = bdata->node_bootmem_map; /* - * Check if we are aligned to BITS_PER_LONG pages. If so, we might - * be able to free page orders of that size at once. + * If the start is aligned to the machines wordsize, we might + * be able to free pages in bulks of that order. */ - if (!(pfn & (BITS_PER_LONG-1))) - gofast = 1; + aligned = !(start & (BITS_PER_LONG - 1)); - for (i = 0; i < idx; ) { - unsigned long v = ~map[i / BITS_PER_LONG]; + bdebug("nid=%td start=%lx end=%lx aligned=%d\n", + bdata - bootmem_node_data, start, end, aligned); - if (gofast && v == ~0UL) { - int order; + while (start < end) { + unsigned long *map, idx, vec; - page = pfn_to_page(pfn); + map = bdata->node_bootmem_map; + idx = start - PFN_DOWN(bdata->node_boot_start); + vec = ~map[idx / BITS_PER_LONG]; + + if (aligned && vec == ~0UL && start + BITS_PER_LONG < end) { + int order = ilog2(BITS_PER_LONG); + + __free_pages_bootmem(pfn_to_page(start), order); count += BITS_PER_LONG; - order = ffs(BITS_PER_LONG) - 1; - __free_pages_bootmem(page, order); - i += BITS_PER_LONG; - page += BITS_PER_LONG; - } else if (v) { - unsigned long m; - - page = pfn_to_page(pfn); - for (m = 1; m && i < idx; m<<=1, page++, i++) { - if (v & m) { - count++; - __free_pages_bootmem(page, 0); - } - } } else { - i += BITS_PER_LONG; + unsigned long off = 0; + + while (vec && off < BITS_PER_LONG) { + if (vec & 1) { + page = pfn_to_page(start + off); + __free_pages_bootmem(page, 0); + count++; + } + vec >>= 1; + off++; + } } - pfn += BITS_PER_LONG; + start += BITS_PER_LONG; } - /* - * Now free the allocator bitmap itself, it's not - * needed anymore: - */ page = virt_to_page(bdata->node_bootmem_map); pages = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start); - idx = bootmem_bootmap_pages(pages); - for (i = 0; i < idx; i++, page++) - __free_pages_bootmem(page, 0); - count += i; - bdata->node_bootmem_map = NULL; + pages = bootmem_bootmap_pages(pages); + count += pages; + while (pages--) + __free_pages_bootmem(page++, 0); bdebug("nid=%td released=%lx\n", bdata - bootmem_node_data, count); From 5f2809e69c7128f86316048221cf45146f69a4a0 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 23 Jul 2008 21:28:05 -0700 Subject: [PATCH 197/857] bootmem: clean up alloc_bootmem_core alloc_bootmem_core has become quite nasty to read over time. This is a clean rewrite that keeps the semantics. bdata->last_pos has been dropped. bdata->last_success has been renamed to hint_idx and it is now an index relative to the node's range. Since further block searching might start at this index, it is now set to the end of a succeeded allocation rather than its beginning. bdata->last_offset has been renamed to last_end_off to be more clear that it represents the ending address of the last allocation relative to the node. [y-goto@jp.fujitsu.com: fix new alloc_bootmem_core()] Signed-off-by: Johannes Weiner Signed-off-by: Yasunori Goto Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/bootmem.h | 6 +- mm/bootmem.c | 210 ++++++++++++++-------------------------- 2 files changed, 77 insertions(+), 139 deletions(-) diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h index 5000fd70b04f..90921d10ffa2 100644 --- a/include/linux/bootmem.h +++ b/include/linux/bootmem.h @@ -31,10 +31,8 @@ typedef struct bootmem_data { unsigned long node_boot_start; unsigned long node_low_pfn; void *node_bootmem_map; - unsigned long last_offset; - unsigned long last_pos; - unsigned long last_success; /* Previous allocation point. To speed - * up searching */ + unsigned long last_end_off; + unsigned long hint_idx; struct list_head list; } bootmem_data_t; diff --git a/mm/bootmem.c b/mm/bootmem.c index 300d126ec533..94ea612deccf 100644 --- a/mm/bootmem.c +++ b/mm/bootmem.c @@ -242,8 +242,9 @@ static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr, * considered reserved. */ - if (addr >= bdata->node_boot_start && addr < bdata->last_success) - bdata->last_success = addr; + if (addr >= bdata->node_boot_start && + PFN_DOWN(addr - bdata->node_boot_start) < bdata->hint_idx) + bdata->hint_idx = PFN_DOWN(addr - bdata->node_boot_start); /* * Round up to index to the range. @@ -431,36 +432,16 @@ int __init reserve_bootmem(unsigned long addr, unsigned long size, } #endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */ -/* - * We 'merge' subsequent allocations to save space. We might 'lose' - * some fraction of a page if allocations cannot be satisfied due to - * size constraints on boxes where there is physical RAM space - * fragmentation - in these cases (mostly large memory boxes) this - * is not a problem. - * - * On low memory boxes we get it right in 100% of the cases. - * - * alignment has to be a power of 2 value. - * - * NOTE: This function is _not_ reentrant. - */ -static void * __init -alloc_bootmem_core(struct bootmem_data *bdata, unsigned long size, - unsigned long align, unsigned long goal, unsigned long limit) +static void * __init alloc_bootmem_core(struct bootmem_data *bdata, + unsigned long size, unsigned long align, + unsigned long goal, unsigned long limit) { - unsigned long areasize, preferred; - unsigned long i, start = 0, incr, eidx, end_pfn; - void *ret; - unsigned long node_boot_start; - void *node_bootmem_map; + unsigned long min, max, start, sidx, midx, step; - if (!size) { - printk("alloc_bootmem_core(): zero-sized request\n"); - BUG(); - } - BUG_ON(align & (align-1)); + BUG_ON(!size); + BUG_ON(align & (align - 1)); + BUG_ON(limit && goal + size > limit); - /* on nodes without memory - bootmem_map is NULL */ if (!bdata->node_bootmem_map) return NULL; @@ -468,126 +449,85 @@ alloc_bootmem_core(struct bootmem_data *bdata, unsigned long size, bdata - bootmem_node_data, size, PAGE_ALIGN(size) >> PAGE_SHIFT, align, goal, limit); - /* bdata->node_boot_start is supposed to be (12+6)bits alignment on x86_64 ? */ - node_boot_start = bdata->node_boot_start; - node_bootmem_map = bdata->node_bootmem_map; - if (align) { - node_boot_start = ALIGN(bdata->node_boot_start, align); - if (node_boot_start > bdata->node_boot_start) - node_bootmem_map = (unsigned long *)bdata->node_bootmem_map + - PFN_DOWN(node_boot_start - bdata->node_boot_start)/BITS_PER_LONG; - } + min = PFN_DOWN(bdata->node_boot_start); + max = bdata->node_low_pfn; - if (limit && node_boot_start >= limit) + goal >>= PAGE_SHIFT; + limit >>= PAGE_SHIFT; + + if (limit && max > limit) + max = limit; + if (max <= min) return NULL; - end_pfn = bdata->node_low_pfn; - limit = PFN_DOWN(limit); - if (limit && end_pfn > limit) - end_pfn = limit; + step = max(align >> PAGE_SHIFT, 1UL); - eidx = end_pfn - PFN_DOWN(node_boot_start); + if (goal && min < goal && goal < max) + start = ALIGN(goal, step); + else + start = ALIGN(min, step); - /* - * We try to allocate bootmem pages above 'goal' - * first, then we try to allocate lower pages. - */ - preferred = 0; - if (goal && PFN_DOWN(goal) < end_pfn) { - if (goal > node_boot_start) - preferred = goal - node_boot_start; + sidx = start - PFN_DOWN(bdata->node_boot_start); + midx = max - PFN_DOWN(bdata->node_boot_start); - if (bdata->last_success > node_boot_start && - bdata->last_success - node_boot_start >= preferred) - if (!limit || (limit && limit > bdata->last_success)) - preferred = bdata->last_success - node_boot_start; + if (bdata->hint_idx > sidx) { + /* Make sure we retry on failure */ + goal = 1; + sidx = ALIGN(bdata->hint_idx, step); } - preferred = PFN_DOWN(ALIGN(preferred, align)); - areasize = (size + PAGE_SIZE-1) / PAGE_SIZE; - incr = align >> PAGE_SHIFT ? : 1; + while (1) { + int merge; + void *region; + unsigned long eidx, i, start_off, end_off; +find_block: + sidx = find_next_zero_bit(bdata->node_bootmem_map, midx, sidx); + sidx = ALIGN(sidx, step); + eidx = sidx + PFN_UP(size); -restart_scan: - for (i = preferred; i < eidx;) { - unsigned long j; - - i = find_next_zero_bit(node_bootmem_map, eidx, i); - i = ALIGN(i, incr); - if (i >= eidx) + if (sidx >= midx || eidx > midx) break; - if (test_bit(i, node_bootmem_map)) { - i += incr; - continue; - } - for (j = i + 1; j < i + areasize; ++j) { - if (j >= eidx) - goto fail_block; - if (test_bit(j, node_bootmem_map)) - goto fail_block; - } - start = i; - goto found; - fail_block: - i = ALIGN(j, incr); - if (i == j) - i += incr; + + for (i = sidx; i < eidx; i++) + if (test_bit(i, bdata->node_bootmem_map)) { + sidx = ALIGN(i, step); + if (sidx == i) + sidx += step; + goto find_block; + } + + if (bdata->last_end_off && + PFN_DOWN(bdata->last_end_off) + 1 == sidx) + start_off = ALIGN(bdata->last_end_off, align); + else + start_off = PFN_PHYS(sidx); + + merge = PFN_DOWN(start_off) < sidx; + end_off = start_off + size; + + bdata->last_end_off = end_off; + bdata->hint_idx = PFN_UP(end_off); + + /* + * Reserve the area now: + */ + for (i = PFN_DOWN(start_off) + merge; + i < PFN_UP(end_off); i++) + if (test_and_set_bit(i, bdata->node_bootmem_map)) + BUG(); + + region = phys_to_virt(bdata->node_boot_start + start_off); + memset(region, 0, size); + return region; } - if (preferred > 0) { - preferred = 0; - goto restart_scan; + if (goal) { + goal = 0; + sidx = 0; + goto find_block; } + return NULL; - -found: - bdata->last_success = PFN_PHYS(start) + node_boot_start; - BUG_ON(start >= eidx); - - /* - * Is the next page of the previous allocation-end the start - * of this allocation's buffer? If yes then we can 'merge' - * the previous partial page with this allocation. - */ - if (align < PAGE_SIZE && - bdata->last_offset && bdata->last_pos+1 == start) { - unsigned long offset, remaining_size; - offset = ALIGN(bdata->last_offset, align); - BUG_ON(offset > PAGE_SIZE); - remaining_size = PAGE_SIZE - offset; - if (size < remaining_size) { - areasize = 0; - /* last_pos unchanged */ - bdata->last_offset = offset + size; - ret = phys_to_virt(bdata->last_pos * PAGE_SIZE + - offset + node_boot_start); - } else { - remaining_size = size - remaining_size; - areasize = (remaining_size + PAGE_SIZE-1) / PAGE_SIZE; - ret = phys_to_virt(bdata->last_pos * PAGE_SIZE + - offset + node_boot_start); - bdata->last_pos = start + areasize - 1; - bdata->last_offset = remaining_size; - } - bdata->last_offset &= ~PAGE_MASK; - } else { - bdata->last_pos = start + areasize - 1; - bdata->last_offset = size & ~PAGE_MASK; - ret = phys_to_virt(start * PAGE_SIZE + node_boot_start); - } - - bdebug("nid=%td start=%lx end=%lx\n", - bdata - bootmem_node_data, - start + PFN_DOWN(bdata->node_boot_start), - start + areasize + PFN_DOWN(bdata->node_boot_start)); - - /* - * Reserve the area now: - */ - for (i = start; i < start + areasize; i++) - if (unlikely(test_and_set_bit(i, node_bootmem_map))) - BUG(); - memset(ret, 0, size); - return ret; } /** From d747fa4bcebcf3696607b86a6b0dafa644be0676 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 23 Jul 2008 21:28:05 -0700 Subject: [PATCH 198/857] bootmem: free/reserve helpers Factor out the common operation of marking a range on the bitmap. [akpm@linux-foundation.org: fix various warnings] Signed-off-by: Johannes Weiner Cc: Ingo Molnar Cc: Yinghai Lu Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/bootmem.c | 65 ++++++++++++++++++++++++++++++++++------------------ 1 file changed, 43 insertions(+), 22 deletions(-) diff --git a/mm/bootmem.c b/mm/bootmem.c index 94ea612deccf..9d03ff651359 100644 --- a/mm/bootmem.c +++ b/mm/bootmem.c @@ -225,6 +225,44 @@ unsigned long __init free_all_bootmem(void) return free_all_bootmem_core(NODE_DATA(0)->bdata); } +static void __init __free(bootmem_data_t *bdata, + unsigned long sidx, unsigned long eidx) +{ + unsigned long idx; + + bdebug("nid=%td start=%lx end=%lx\n", bdata - bootmem_node_data, + sidx + PFN_DOWN(bdata->node_boot_start), + eidx + PFN_DOWN(bdata->node_boot_start)); + + for (idx = sidx; idx < eidx; idx++) + if (!test_and_clear_bit(idx, bdata->node_bootmem_map)) + BUG(); +} + +static int __init __reserve(bootmem_data_t *bdata, unsigned long sidx, + unsigned long eidx, int flags) +{ + unsigned long idx; + int exclusive = flags & BOOTMEM_EXCLUSIVE; + + bdebug("nid=%td start=%lx end=%lx flags=%x\n", + bdata - bootmem_node_data, + sidx + PFN_DOWN(bdata->node_boot_start), + eidx + PFN_DOWN(bdata->node_boot_start), + flags); + + for (idx = sidx; idx < eidx; idx++) + if (test_and_set_bit(idx, bdata->node_bootmem_map)) { + if (exclusive) { + __free(bdata, sidx, idx); + return -EBUSY; + } + bdebug("silent double reserve of PFN %lx\n", + idx + PFN_DOWN(bdata->node_boot_start)); + } + return 0; +} + static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr, unsigned long size) { @@ -258,14 +296,7 @@ static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr, if (eidx > bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start)) eidx = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start); - bdebug("nid=%td start=%lx end=%lx\n", bdata - bootmem_node_data, - sidx + PFN_DOWN(bdata->node_boot_start), - eidx + PFN_DOWN(bdata->node_boot_start)); - - for (i = sidx; i < eidx; i++) { - if (unlikely(!test_and_clear_bit(i, bdata->node_bootmem_map))) - BUG(); - } + __free(bdata, sidx, eidx); } /** @@ -367,16 +398,7 @@ static void __init reserve_bootmem_core(bootmem_data_t *bdata, if (eidx > bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start)) eidx = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start); - bdebug("nid=%td start=%lx end=%lx flags=%x\n", - bdata - bootmem_node_data, - sidx + PFN_DOWN(bdata->node_boot_start), - eidx + PFN_DOWN(bdata->node_boot_start), - flags); - - for (i = sidx; i < eidx; i++) - if (test_and_set_bit(i, bdata->node_bootmem_map)) - bdebug("hm, page %lx reserved twice.\n", - PFN_DOWN(bdata->node_boot_start) + i); + return __reserve(bdata, sidx, eidx, flags); } /** @@ -511,10 +533,9 @@ static void * __init alloc_bootmem_core(struct bootmem_data *bdata, /* * Reserve the area now: */ - for (i = PFN_DOWN(start_off) + merge; - i < PFN_UP(end_off); i++) - if (test_and_set_bit(i, bdata->node_bootmem_map)) - BUG(); + if (__reserve(bdata, PFN_DOWN(start_off) + merge, + PFN_UP(end_off), BOOTMEM_EXCLUSIVE)) + BUG(); region = phys_to_virt(bdata->node_boot_start + start_off); memset(region, 0, size); From e2bf3cae515090fefe28329e71230dfe7ab873b1 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 23 Jul 2008 21:28:06 -0700 Subject: [PATCH 199/857] bootmem: factor out the marking of a PFN range Introduce new helpers that mark a range that resides completely on a node or node-agnostic ranges that might also span node boundaries. The free/reserve API functions will then directly use these helpers. Note that the free/reserve semantics become more strict: while the prior code took basically arbitrary range arguments and marked the PFNs that happen to fall into that range, the new code requires node-specific ranges to be completely on the node. The node-agnostic requests might span node boundaries as long as the nodes are contiguous. Passing ranges that do not satisfy these criteria is a bug. [akpm@linux-foundation.org: fix printk warnings] Signed-off-by: Johannes Weiner Cc: Ingo Molnar Cc: Yinghai Lu Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/bootmem.c | 188 +++++++++++++++++++-------------------------------- 1 file changed, 69 insertions(+), 119 deletions(-) diff --git a/mm/bootmem.c b/mm/bootmem.c index 9d03ff651359..e5415a5414a5 100644 --- a/mm/bootmem.c +++ b/mm/bootmem.c @@ -234,6 +234,9 @@ static void __init __free(bootmem_data_t *bdata, sidx + PFN_DOWN(bdata->node_boot_start), eidx + PFN_DOWN(bdata->node_boot_start)); + if (bdata->hint_idx > sidx) + bdata->hint_idx = sidx; + for (idx = sidx; idx < eidx; idx++) if (!test_and_clear_bit(idx, bdata->node_bootmem_map)) BUG(); @@ -263,40 +266,57 @@ static int __init __reserve(bootmem_data_t *bdata, unsigned long sidx, return 0; } -static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr, - unsigned long size) +static int __init mark_bootmem_node(bootmem_data_t *bdata, + unsigned long start, unsigned long end, + int reserve, int flags) { unsigned long sidx, eidx; - unsigned long i; - BUG_ON(!size); + bdebug("nid=%td start=%lx end=%lx reserve=%d flags=%x\n", + bdata - bootmem_node_data, start, end, reserve, flags); - /* out range */ - if (addr + size < bdata->node_boot_start || - PFN_DOWN(addr) > bdata->node_low_pfn) - return; - /* - * round down end of usable mem, partially free pages are - * considered reserved. - */ + BUG_ON(start < PFN_DOWN(bdata->node_boot_start)); + BUG_ON(end > bdata->node_low_pfn); - if (addr >= bdata->node_boot_start && - PFN_DOWN(addr - bdata->node_boot_start) < bdata->hint_idx) - bdata->hint_idx = PFN_DOWN(addr - bdata->node_boot_start); + sidx = start - PFN_DOWN(bdata->node_boot_start); + eidx = end - PFN_DOWN(bdata->node_boot_start); - /* - * Round up to index to the range. - */ - if (PFN_UP(addr) > PFN_DOWN(bdata->node_boot_start)) - sidx = PFN_UP(addr) - PFN_DOWN(bdata->node_boot_start); + if (reserve) + return __reserve(bdata, sidx, eidx, flags); else - sidx = 0; + __free(bdata, sidx, eidx); + return 0; +} - eidx = PFN_DOWN(addr + size - bdata->node_boot_start); - if (eidx > bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start)) - eidx = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start); +static int __init mark_bootmem(unsigned long start, unsigned long end, + int reserve, int flags) +{ + unsigned long pos; + bootmem_data_t *bdata; - __free(bdata, sidx, eidx); + pos = start; + list_for_each_entry(bdata, &bdata_list, list) { + int err; + unsigned long max; + + if (pos < PFN_DOWN(bdata->node_boot_start)) { + BUG_ON(pos != start); + continue; + } + + max = min(bdata->node_low_pfn, end); + + err = mark_bootmem_node(bdata, pos, max, reserve, flags); + if (reserve && err) { + mark_bootmem(start, pos, 0, 0); + return err; + } + + if (max == end) + return 0; + pos = bdata->node_low_pfn; + } + BUG(); } /** @@ -307,12 +327,17 @@ static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr, * * Partial pages will be considered reserved and left as they are. * - * Only physical pages that actually reside on @pgdat are marked. + * The range must reside completely on the specified node. */ void __init free_bootmem_node(pg_data_t *pgdat, unsigned long physaddr, unsigned long size) { - free_bootmem_core(pgdat->bdata, physaddr, size); + unsigned long start, end; + + start = PFN_UP(physaddr); + end = PFN_DOWN(physaddr + size); + + mark_bootmem_node(pgdat->bdata, start, end, 0, 0); } /** @@ -322,83 +347,16 @@ void __init free_bootmem_node(pg_data_t *pgdat, unsigned long physaddr, * * Partial pages will be considered reserved and left as they are. * - * All physical pages within the range are marked, no matter what - * node they reside on. + * The range must be contiguous but may span node boundaries. */ void __init free_bootmem(unsigned long addr, unsigned long size) { - bootmem_data_t *bdata; - list_for_each_entry(bdata, &bdata_list, list) - free_bootmem_core(bdata, addr, size); -} + unsigned long start, end; -/* - * Marks a particular physical memory range as unallocatable. Usable RAM - * might be used for boot-time allocations - or it might get added - * to the free page pool later on. - */ -static int __init can_reserve_bootmem_core(bootmem_data_t *bdata, - unsigned long addr, unsigned long size, int flags) -{ - unsigned long sidx, eidx; - unsigned long i; + start = PFN_UP(addr); + end = PFN_DOWN(addr + size); - BUG_ON(!size); - - /* out of range, don't hold other */ - if (addr + size < bdata->node_boot_start || - PFN_DOWN(addr) > bdata->node_low_pfn) - return 0; - - /* - * Round up to index to the range. - */ - if (addr > bdata->node_boot_start) - sidx= PFN_DOWN(addr - bdata->node_boot_start); - else - sidx = 0; - - eidx = PFN_UP(addr + size - bdata->node_boot_start); - if (eidx > bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start)) - eidx = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start); - - for (i = sidx; i < eidx; i++) { - if (test_bit(i, bdata->node_bootmem_map)) { - if (flags & BOOTMEM_EXCLUSIVE) - return -EBUSY; - } - } - - return 0; - -} - -static void __init reserve_bootmem_core(bootmem_data_t *bdata, - unsigned long addr, unsigned long size, int flags) -{ - unsigned long sidx, eidx; - unsigned long i; - - BUG_ON(!size); - - /* out of range */ - if (addr + size < bdata->node_boot_start || - PFN_DOWN(addr) > bdata->node_low_pfn) - return; - - /* - * Round up to index to the range. - */ - if (addr > bdata->node_boot_start) - sidx= PFN_DOWN(addr - bdata->node_boot_start); - else - sidx = 0; - - eidx = PFN_UP(addr + size - bdata->node_boot_start); - if (eidx > bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start)) - eidx = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start); - - return __reserve(bdata, sidx, eidx, flags); + mark_bootmem(start, end, 0, 0); } /** @@ -410,18 +368,17 @@ static void __init reserve_bootmem_core(bootmem_data_t *bdata, * * Partial pages will be reserved. * - * Only physical pages that actually reside on @pgdat are marked. + * The range must reside completely on the specified node. */ int __init reserve_bootmem_node(pg_data_t *pgdat, unsigned long physaddr, unsigned long size, int flags) { - int ret; + unsigned long start, end; - ret = can_reserve_bootmem_core(pgdat->bdata, physaddr, size, flags); - if (ret < 0) - return -ENOMEM; - reserve_bootmem_core(pgdat->bdata, physaddr, size, flags); - return 0; + start = PFN_DOWN(physaddr); + end = PFN_UP(physaddr + size); + + return mark_bootmem_node(pgdat->bdata, start, end, 1, flags); } #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE @@ -433,24 +390,17 @@ int __init reserve_bootmem_node(pg_data_t *pgdat, unsigned long physaddr, * * Partial pages will be reserved. * - * All physical pages within the range are marked, no matter what - * node they reside on. + * The range must be contiguous but may span node boundaries. */ int __init reserve_bootmem(unsigned long addr, unsigned long size, int flags) { - bootmem_data_t *bdata; - int ret; + unsigned long start, end; - list_for_each_entry(bdata, &bdata_list, list) { - ret = can_reserve_bootmem_core(bdata, addr, size, flags); - if (ret < 0) - return ret; - } - list_for_each_entry(bdata, &bdata_list, list) - reserve_bootmem_core(bdata, addr, size, flags); + start = PFN_DOWN(addr); + end = PFN_UP(addr + size); - return 0; + return mark_bootmem(start, end, 1, flags); } #endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */ @@ -663,7 +613,7 @@ void * __init alloc_bootmem_section(unsigned long size, if (start_nr != section_nr || end_nr != section_nr) { printk(KERN_WARNING "alloc_bootmem failed on section %ld.\n", section_nr); - free_bootmem_core(pgdat->bdata, __pa(ptr), size); + free_bootmem_node(pgdat, __pa(ptr), size); ptr = NULL; } From 0f3caba211babef6e3fbde1ba76ddc79321bc92f Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 23 Jul 2008 21:28:07 -0700 Subject: [PATCH 200/857] bootmem: respect goal more likely The old node-agnostic code tried allocating on all nodes starting from the one with the lowest range. alloc_bootmem_core retried without the goal if it could not satisfy it and so the goal was only respected at all when it happened to be on the first (lowest page numbers) node (or theoretically if allocations failed on all nodes before to the one holding the goal). Introduce a non-panicking helper that starts allocating from the node holding the goal and falls back only after all thes tries failed, thus moving the goal fallback code out of alloc_bootmem_core. Make all other allocation functions benefit from this new helper. Signed-off-by: Johannes Weiner Cc: Ingo Molnar Cc: Yinghai Lu Cc: Andi Kleen Cc: Yasunori Goto Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/bootmem.c | 92 ++++++++++++++++++++++++++++++---------------------- 1 file changed, 54 insertions(+), 38 deletions(-) diff --git a/mm/bootmem.c b/mm/bootmem.c index e5415a5414a5..89646f77b427 100644 --- a/mm/bootmem.c +++ b/mm/bootmem.c @@ -408,6 +408,7 @@ static void * __init alloc_bootmem_core(struct bootmem_data *bdata, unsigned long size, unsigned long align, unsigned long goal, unsigned long limit) { + unsigned long fallback = 0; unsigned long min, max, start, sidx, midx, step; BUG_ON(!size); @@ -443,8 +444,11 @@ static void * __init alloc_bootmem_core(struct bootmem_data *bdata, midx = max - PFN_DOWN(bdata->node_boot_start); if (bdata->hint_idx > sidx) { - /* Make sure we retry on failure */ - goal = 1; + /* + * Handle the valid case of sidx being zero and still + * catch the fallback below. + */ + fallback = sidx + 1; sidx = ALIGN(bdata->hint_idx, step); } @@ -492,10 +496,39 @@ static void * __init alloc_bootmem_core(struct bootmem_data *bdata, return region; } + if (fallback) { + sidx = ALIGN(fallback - 1, step); + fallback = 0; + goto find_block; + } + + return NULL; +} + +static void * __init ___alloc_bootmem_nopanic(unsigned long size, + unsigned long align, + unsigned long goal, + unsigned long limit) +{ + bootmem_data_t *bdata; + +restart: + list_for_each_entry(bdata, &bdata_list, list) { + void *region; + + if (goal && bdata->node_low_pfn <= PFN_DOWN(goal)) + continue; + if (limit && bdata->node_boot_start >= limit) + break; + + region = alloc_bootmem_core(bdata, size, align, goal, limit); + if (region) + return region; + } + if (goal) { goal = 0; - sidx = 0; - goto find_block; + goto restart; } return NULL; @@ -515,16 +548,23 @@ static void * __init alloc_bootmem_core(struct bootmem_data *bdata, * Returns NULL on failure. */ void * __init __alloc_bootmem_nopanic(unsigned long size, unsigned long align, - unsigned long goal) + unsigned long goal) { - bootmem_data_t *bdata; - void *ptr; + return ___alloc_bootmem_nopanic(size, align, goal, 0); +} - list_for_each_entry(bdata, &bdata_list, list) { - ptr = alloc_bootmem_core(bdata, size, align, goal, 0); - if (ptr) - return ptr; - } +static void * __init ___alloc_bootmem(unsigned long size, unsigned long align, + unsigned long goal, unsigned long limit) +{ + void *mem = ___alloc_bootmem_nopanic(size, align, goal, limit); + + if (mem) + return mem; + /* + * Whoops, we cannot satisfy the allocation request. + */ + printk(KERN_ALERT "bootmem alloc of %lu bytes failed!\n", size); + panic("Out of memory"); return NULL; } @@ -544,16 +584,7 @@ void * __init __alloc_bootmem_nopanic(unsigned long size, unsigned long align, void * __init __alloc_bootmem(unsigned long size, unsigned long align, unsigned long goal) { - void *mem = __alloc_bootmem_nopanic(size,align,goal); - - if (mem) - return mem; - /* - * Whoops, we cannot satisfy the allocation request. - */ - printk(KERN_ALERT "bootmem alloc of %lu bytes failed!\n", size); - panic("Out of memory"); - return NULL; + return ___alloc_bootmem(size, align, goal, 0); } /** @@ -653,22 +684,7 @@ void * __init __alloc_bootmem_node_nopanic(pg_data_t *pgdat, unsigned long size, void * __init __alloc_bootmem_low(unsigned long size, unsigned long align, unsigned long goal) { - bootmem_data_t *bdata; - void *ptr; - - list_for_each_entry(bdata, &bdata_list, list) { - ptr = alloc_bootmem_core(bdata, size, align, goal, - ARCH_LOW_ADDRESS_LIMIT); - if (ptr) - return ptr; - } - - /* - * Whoops, we cannot satisfy the allocation request. - */ - printk(KERN_ALERT "low bootmem alloc of %lu bytes failed!\n", size); - panic("Out of low memory"); - return NULL; + return ___alloc_bootmem(size, align, goal, ARCH_LOW_ADDRESS_LIMIT); } /** From 4cc278b721d5bf3569dfc5f1100253042e097bc3 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 23 Jul 2008 21:28:08 -0700 Subject: [PATCH 201/857] bootmem: Make __alloc_bootmem_low_node fall back to other nodes __alloc_bootmem_node already does this, make the interface consistent. Signed-off-by: Johannes Weiner Cc: Ingo Molnar Cc: Yinghai Lu Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/bootmem.c | 25 ++++++++++++++++--------- 1 file changed, 16 insertions(+), 9 deletions(-) diff --git a/mm/bootmem.c b/mm/bootmem.c index 89646f77b427..459da4710b8f 100644 --- a/mm/bootmem.c +++ b/mm/bootmem.c @@ -587,6 +587,19 @@ void * __init __alloc_bootmem(unsigned long size, unsigned long align, return ___alloc_bootmem(size, align, goal, 0); } +static void * __init ___alloc_bootmem_node(bootmem_data_t *bdata, + unsigned long size, unsigned long align, + unsigned long goal, unsigned long limit) +{ + void *ptr; + + ptr = alloc_bootmem_core(bdata, size, align, goal, limit); + if (ptr) + return ptr; + + return ___alloc_bootmem(size, align, goal, limit); +} + /** * __alloc_bootmem_node - allocate boot memory from a specific node * @pgdat: node to allocate from @@ -605,13 +618,7 @@ void * __init __alloc_bootmem(unsigned long size, unsigned long align, void * __init __alloc_bootmem_node(pg_data_t *pgdat, unsigned long size, unsigned long align, unsigned long goal) { - void *ptr; - - ptr = alloc_bootmem_core(pgdat->bdata, size, align, goal, 0); - if (ptr) - return ptr; - - return __alloc_bootmem(size, align, goal); + return ___alloc_bootmem_node(pgdat->bdata, size, align, goal, 0); } #ifdef CONFIG_SPARSEMEM @@ -705,6 +712,6 @@ void * __init __alloc_bootmem_low(unsigned long size, unsigned long align, void * __init __alloc_bootmem_low_node(pg_data_t *pgdat, unsigned long size, unsigned long align, unsigned long goal) { - return alloc_bootmem_core(pgdat->bdata, size, align, goal, - ARCH_LOW_ADDRESS_LIMIT); + return ___alloc_bootmem_node(pgdat->bdata, size, align, + goal, ARCH_LOW_ADDRESS_LIMIT); } From 75a56cfe9fdb064d1db1cfbc564315fddb756fb1 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 23 Jul 2008 21:28:09 -0700 Subject: [PATCH 202/857] bootmem: revisit alloc_bootmem_section Since alloc_bootmem_core does no goal-fallback anymore and just returns NULL if the allocation fails, we might now use it in alloc_bootmem_section without all the fixup code for a misplaced allocation. Also, the limit can be the first PFN of the next section as the semantics is that the limit is _above_ the allocated region, not within. Signed-off-by: Johannes Weiner Cc: Yasunori Goto Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/bootmem.c | 27 ++++++--------------------- 1 file changed, 6 insertions(+), 21 deletions(-) diff --git a/mm/bootmem.c b/mm/bootmem.c index 459da4710b8f..282b786c2b15 100644 --- a/mm/bootmem.c +++ b/mm/bootmem.c @@ -632,30 +632,15 @@ void * __init __alloc_bootmem_node(pg_data_t *pgdat, unsigned long size, void * __init alloc_bootmem_section(unsigned long size, unsigned long section_nr) { - void *ptr; - unsigned long limit, goal, start_nr, end_nr, pfn; - struct pglist_data *pgdat; + bootmem_data_t *bdata; + unsigned long pfn, goal, limit; pfn = section_nr_to_pfn(section_nr); - goal = PFN_PHYS(pfn); - limit = PFN_PHYS(section_nr_to_pfn(section_nr + 1)) - 1; - pgdat = NODE_DATA(early_pfn_to_nid(pfn)); - ptr = alloc_bootmem_core(pgdat->bdata, size, SMP_CACHE_BYTES, goal, - limit); + goal = pfn << PAGE_SHIFT; + limit = section_nr_to_pfn(section_nr + 1) << PAGE_SHIFT; + bdata = &bootmem_node_data[early_pfn_to_nid(pfn)]; - if (!ptr) - return NULL; - - start_nr = pfn_to_section_nr(PFN_DOWN(__pa(ptr))); - end_nr = pfn_to_section_nr(PFN_DOWN(__pa(ptr) + size)); - if (start_nr != section_nr || end_nr != section_nr) { - printk(KERN_WARNING "alloc_bootmem failed on section %ld.\n", - section_nr); - free_bootmem_node(pgdat, __pa(ptr), size); - ptr = NULL; - } - - return ptr; + return alloc_bootmem_core(bdata, size, SMP_CACHE_BYTES, goal, limit); } #endif From 3560e249abda6bee41a07a7bf0383a6e193e2839 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Wed, 23 Jul 2008 21:28:09 -0700 Subject: [PATCH 203/857] bootmem: replace node_boot_start in struct bootmem_data Almost all users of this field need a PFN instead of a physical address, so replace node_boot_start with node_min_pfn. [Lee.Schermerhorn@hp.com: fix spurious BUG_ON() in mark_bootmem()] Signed-off-by: Johannes Weiner Cc: Signed-off-by: Lee Schermerhorn Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/alpha/mm/numa.c | 2 +- arch/arm/plat-omap/fb.c | 4 +--- arch/avr32/mm/init.c | 3 +-- arch/ia64/mm/discontig.c | 19 ++++++++++--------- arch/m32r/mm/discontig.c | 3 +-- arch/m32r/mm/init.c | 4 +--- arch/mn10300/mm/init.c | 6 +++--- arch/sh/mm/init.c | 2 +- include/linux/bootmem.h | 2 +- mm/bootmem.c | 40 +++++++++++++++++++++------------------- 10 files changed, 41 insertions(+), 44 deletions(-) diff --git a/arch/alpha/mm/numa.c b/arch/alpha/mm/numa.c index def0c74a78a8..d8c4ceaf00b9 100644 --- a/arch/alpha/mm/numa.c +++ b/arch/alpha/mm/numa.c @@ -304,7 +304,7 @@ void __init paging_init(void) for_each_online_node(nid) { bootmem_data_t *bdata = &bootmem_node_data[nid]; - unsigned long start_pfn = bdata->node_boot_start >> PAGE_SHIFT; + unsigned long start_pfn = bdata->node_min_pfn; unsigned long end_pfn = bdata->node_low_pfn; if (dma_local_pfn >= end_pfn - start_pfn) diff --git a/arch/arm/plat-omap/fb.c b/arch/arm/plat-omap/fb.c index 7854f19b77cf..96d6f0619733 100644 --- a/arch/arm/plat-omap/fb.c +++ b/arch/arm/plat-omap/fb.c @@ -182,7 +182,7 @@ void __init omapfb_reserve_sdram(void) return; bdata = NODE_DATA(0)->bdata; - sdram_start = bdata->node_boot_start; + sdram_start = bdata->node_min_pfn << PAGE_SHIFT; sdram_size = (bdata->node_low_pfn << PAGE_SHIFT) - sdram_start; reserved = 0; for (i = 0; ; i++) { @@ -340,5 +340,3 @@ unsigned long omapfb_reserve_sram(unsigned long sram_pstart, #endif - - diff --git a/arch/avr32/mm/init.c b/arch/avr32/mm/init.c index 786de88a82a7..3c85fdaa9487 100644 --- a/arch/avr32/mm/init.c +++ b/arch/avr32/mm/init.c @@ -119,8 +119,7 @@ void __init paging_init(void) unsigned long zones_size[MAX_NR_ZONES]; unsigned long low, start_pfn; - start_pfn = pgdat->bdata->node_boot_start; - start_pfn >>= PAGE_SHIFT; + start_pfn = pgdat->bdata->node_min_pfn; low = pgdat->bdata->node_low_pfn; memset(zones_size, 0, sizeof(zones_size)); diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c index 2fcf8464331e..d83125e1ed27 100644 --- a/arch/ia64/mm/discontig.c +++ b/arch/ia64/mm/discontig.c @@ -74,17 +74,17 @@ pg_data_t *pgdat_list[MAX_NUMNODES]; static int __init build_node_maps(unsigned long start, unsigned long len, int node) { - unsigned long cstart, epfn, end = start + len; + unsigned long spfn, epfn, end = start + len; struct bootmem_data *bdp = &bootmem_node_data[node]; epfn = GRANULEROUNDUP(end) >> PAGE_SHIFT; - cstart = GRANULEROUNDDOWN(start); + spfn = GRANULEROUNDDOWN(start) >> PAGE_SHIFT; if (!bdp->node_low_pfn) { - bdp->node_boot_start = cstart; + bdp->node_min_pfn = spfn; bdp->node_low_pfn = epfn; } else { - bdp->node_boot_start = min(cstart, bdp->node_boot_start); + bdp->node_min_pfn = min(spfn, bdp->node_min_pfn); bdp->node_low_pfn = max(epfn, bdp->node_low_pfn); } @@ -221,20 +221,21 @@ static void __init fill_pernode(int node, unsigned long pernode, static int __init find_pernode_space(unsigned long start, unsigned long len, int node) { - unsigned long epfn; + unsigned long spfn, epfn; unsigned long pernodesize = 0, pernode, pages, mapsize; struct bootmem_data *bdp = &bootmem_node_data[node]; + spfn = start >> PAGE_SHIFT; epfn = (start + len) >> PAGE_SHIFT; - pages = bdp->node_low_pfn - (bdp->node_boot_start >> PAGE_SHIFT); + pages = bdp->node_low_pfn - bdp->node_min_pfn; mapsize = bootmem_bootmap_pages(pages) << PAGE_SHIFT; /* * Make sure this memory falls within this node's usable memory * since we may have thrown some away in build_maps(). */ - if (start < bdp->node_boot_start || epfn > bdp->node_low_pfn) + if (spfn < bdp->node_min_pfn || epfn > bdp->node_low_pfn) return 0; /* Don't setup this node's local space twice... */ @@ -296,7 +297,7 @@ static void __init reserve_pernode_space(void) bdp = pdp->bdata; /* First the bootmem_map itself */ - pages = bdp->node_low_pfn - (bdp->node_boot_start>>PAGE_SHIFT); + pages = bdp->node_low_pfn - bdp->node_min_pfn; size = bootmem_bootmap_pages(pages) << PAGE_SHIFT; base = __pa(bdp->node_bootmem_map); reserve_bootmem_node(pdp, base, size, BOOTMEM_DEFAULT); @@ -466,7 +467,7 @@ void __init find_memory(void) init_bootmem_node(pgdat_list[node], map>>PAGE_SHIFT, - bdp->node_boot_start>>PAGE_SHIFT, + bdp->node_min_pfn, bdp->node_low_pfn); } diff --git a/arch/m32r/mm/discontig.c b/arch/m32r/mm/discontig.c index cc23934bc41e..cbc3c4c54566 100644 --- a/arch/m32r/mm/discontig.c +++ b/arch/m32r/mm/discontig.c @@ -123,8 +123,7 @@ unsigned long __init setup_memory(void) return max_low_pfn; } -#define START_PFN(nid) \ - (NODE_DATA(nid)->bdata->node_boot_start >> PAGE_SHIFT) +#define START_PFN(nid) (NODE_DATA(nid)->bdata->node_min_pfn) #define MAX_LOW_PFN(nid) (NODE_DATA(nid)->bdata->node_low_pfn) unsigned long __init zone_sizes_init(void) diff --git a/arch/m32r/mm/init.c b/arch/m32r/mm/init.c index 28799af15e95..2554eb59cfef 100644 --- a/arch/m32r/mm/init.c +++ b/arch/m32r/mm/init.c @@ -93,8 +93,7 @@ void free_initrd_mem(unsigned long, unsigned long); #endif /* It'd be good if these lines were in the standard header file. */ -#define START_PFN(nid) \ - (NODE_DATA(nid)->bdata->node_boot_start >> PAGE_SHIFT) +#define START_PFN(nid) (NODE_DATA(nid)->bdata->node_min_pfn) #define MAX_LOW_PFN(nid) (NODE_DATA(nid)->bdata->node_low_pfn) #ifndef CONFIG_DISCONTIGMEM @@ -252,4 +251,3 @@ void free_initrd_mem(unsigned long start, unsigned long end) printk (KERN_INFO "Freeing initrd memory: %ldk freed\n", (end - start) >> 10); } #endif - diff --git a/arch/mn10300/mm/init.c b/arch/mn10300/mm/init.c index 8c5d88c7b90a..8cee387a24fd 100644 --- a/arch/mn10300/mm/init.c +++ b/arch/mn10300/mm/init.c @@ -67,8 +67,8 @@ void __init paging_init(void) /* declare the sizes of the RAM zones (only use the normal zone) */ zones_size[ZONE_NORMAL] = - (contig_page_data.bdata->node_low_pfn) - - (contig_page_data.bdata->node_boot_start >> PAGE_SHIFT); + contig_page_data.bdata->node_low_pfn - + contig_page_data.bdata->node_min_pfn; /* pass the memory from the bootmem allocator to the main allocator */ free_area_init(zones_size); @@ -87,7 +87,7 @@ void __init mem_init(void) if (!mem_map) BUG(); -#define START_PFN (contig_page_data.bdata->node_boot_start >> PAGE_SHIFT) +#define START_PFN (contig_page_data.bdata->node_min_pfn) #define MAX_LOW_PFN (contig_page_data.bdata->node_low_pfn) max_mapnr = num_physpages = MAX_LOW_PFN - START_PFN; diff --git a/arch/sh/mm/init.c b/arch/sh/mm/init.c index d7df26bd1e54..d652d375eb1e 100644 --- a/arch/sh/mm/init.c +++ b/arch/sh/mm/init.c @@ -191,7 +191,7 @@ void __init paging_init(void) pg_data_t *pgdat = NODE_DATA(nid); unsigned long low, start_pfn; - start_pfn = pgdat->bdata->node_boot_start >> PAGE_SHIFT; + start_pfn = pgdat->bdata->node_min_pfn; low = pgdat->bdata->node_low_pfn; if (max_zone_pfns[ZONE_NORMAL] < low) diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h index 90921d10ffa2..4ddf2922fc8d 100644 --- a/include/linux/bootmem.h +++ b/include/linux/bootmem.h @@ -28,7 +28,7 @@ extern unsigned long saved_max_pfn; * memory pages (including holes) on the node. */ typedef struct bootmem_data { - unsigned long node_boot_start; + unsigned long node_min_pfn; unsigned long node_low_pfn; void *node_bootmem_map; unsigned long last_end_off; diff --git a/mm/bootmem.c b/mm/bootmem.c index 282b786c2b15..4af15d0340ad 100644 --- a/mm/bootmem.c +++ b/mm/bootmem.c @@ -80,7 +80,7 @@ static void __init link_bootmem(bootmem_data_t *bdata) bootmem_data_t *ent; ent = list_entry(iter, bootmem_data_t, list); - if (bdata->node_boot_start < ent->node_boot_start) + if (bdata->node_min_pfn < ent->node_min_pfn) break; } list_add_tail(&bdata->list, iter); @@ -96,7 +96,7 @@ static unsigned long __init init_bootmem_core(bootmem_data_t *bdata, mminit_validate_memmodel_limits(&start, &end); bdata->node_bootmem_map = phys_to_virt(PFN_PHYS(mapstart)); - bdata->node_boot_start = PFN_PHYS(start); + bdata->node_min_pfn = start; bdata->node_low_pfn = end; link_bootmem(bdata); @@ -151,7 +151,7 @@ static unsigned long __init free_all_bootmem_core(bootmem_data_t *bdata) if (!bdata->node_bootmem_map) return 0; - start = PFN_DOWN(bdata->node_boot_start); + start = bdata->node_min_pfn; end = bdata->node_low_pfn; /* @@ -167,7 +167,7 @@ static unsigned long __init free_all_bootmem_core(bootmem_data_t *bdata) unsigned long *map, idx, vec; map = bdata->node_bootmem_map; - idx = start - PFN_DOWN(bdata->node_boot_start); + idx = start - bdata->node_min_pfn; vec = ~map[idx / BITS_PER_LONG]; if (aligned && vec == ~0UL && start + BITS_PER_LONG < end) { @@ -192,7 +192,7 @@ static unsigned long __init free_all_bootmem_core(bootmem_data_t *bdata) } page = virt_to_page(bdata->node_bootmem_map); - pages = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start); + pages = bdata->node_low_pfn - bdata->node_min_pfn; pages = bootmem_bootmap_pages(pages); count += pages; while (pages--) @@ -231,8 +231,8 @@ static void __init __free(bootmem_data_t *bdata, unsigned long idx; bdebug("nid=%td start=%lx end=%lx\n", bdata - bootmem_node_data, - sidx + PFN_DOWN(bdata->node_boot_start), - eidx + PFN_DOWN(bdata->node_boot_start)); + sidx + bdata->node_min_pfn, + eidx + bdata->node_min_pfn); if (bdata->hint_idx > sidx) bdata->hint_idx = sidx; @@ -250,8 +250,8 @@ static int __init __reserve(bootmem_data_t *bdata, unsigned long sidx, bdebug("nid=%td start=%lx end=%lx flags=%x\n", bdata - bootmem_node_data, - sidx + PFN_DOWN(bdata->node_boot_start), - eidx + PFN_DOWN(bdata->node_boot_start), + sidx + bdata->node_min_pfn, + eidx + bdata->node_min_pfn, flags); for (idx = sidx; idx < eidx; idx++) @@ -261,7 +261,7 @@ static int __init __reserve(bootmem_data_t *bdata, unsigned long sidx, return -EBUSY; } bdebug("silent double reserve of PFN %lx\n", - idx + PFN_DOWN(bdata->node_boot_start)); + idx + bdata->node_min_pfn); } return 0; } @@ -275,11 +275,11 @@ static int __init mark_bootmem_node(bootmem_data_t *bdata, bdebug("nid=%td start=%lx end=%lx reserve=%d flags=%x\n", bdata - bootmem_node_data, start, end, reserve, flags); - BUG_ON(start < PFN_DOWN(bdata->node_boot_start)); + BUG_ON(start < bdata->node_min_pfn); BUG_ON(end > bdata->node_low_pfn); - sidx = start - PFN_DOWN(bdata->node_boot_start); - eidx = end - PFN_DOWN(bdata->node_boot_start); + sidx = start - bdata->node_min_pfn; + eidx = end - bdata->node_min_pfn; if (reserve) return __reserve(bdata, sidx, eidx, flags); @@ -299,7 +299,8 @@ static int __init mark_bootmem(unsigned long start, unsigned long end, int err; unsigned long max; - if (pos < PFN_DOWN(bdata->node_boot_start)) { + if (pos < bdata->node_min_pfn || + pos >= bdata->node_low_pfn) { BUG_ON(pos != start); continue; } @@ -422,7 +423,7 @@ static void * __init alloc_bootmem_core(struct bootmem_data *bdata, bdata - bootmem_node_data, size, PAGE_ALIGN(size) >> PAGE_SHIFT, align, goal, limit); - min = PFN_DOWN(bdata->node_boot_start); + min = bdata->node_min_pfn; max = bdata->node_low_pfn; goal >>= PAGE_SHIFT; @@ -440,8 +441,8 @@ static void * __init alloc_bootmem_core(struct bootmem_data *bdata, else start = ALIGN(min, step); - sidx = start - PFN_DOWN(bdata->node_boot_start); - midx = max - PFN_DOWN(bdata->node_boot_start); + sidx = start - bdata->node_min_pfn;; + midx = max - bdata->node_min_pfn; if (bdata->hint_idx > sidx) { /* @@ -491,7 +492,8 @@ static void * __init alloc_bootmem_core(struct bootmem_data *bdata, PFN_UP(end_off), BOOTMEM_EXCLUSIVE)) BUG(); - region = phys_to_virt(bdata->node_boot_start + start_off); + region = phys_to_virt(PFN_PHYS(bdata->node_min_pfn) + + start_off); memset(region, 0, size); return region; } @@ -518,7 +520,7 @@ static void * __init ___alloc_bootmem_nopanic(unsigned long size, if (goal && bdata->node_low_pfn <= PFN_DOWN(goal)) continue; - if (limit && bdata->node_boot_start >= limit) + if (limit && bdata->node_min_pfn >= PFN_DOWN(limit)) break; region = alloc_bootmem_core(bdata, size, align, goal, limit); From 2be0ffe2b29bd31d3debd0877797892ff2d91f4c Mon Sep 17 00:00:00 2001 From: Timur Tabi Date: Wed, 23 Jul 2008 21:28:11 -0700 Subject: [PATCH 204/857] mm: add alloc_pages_exact() and free_pages_exact() alloc_pages_exact() is similar to alloc_pages(), except that it allocates the minimum number of pages to fulfill the request. This is useful if you want to allocate a very large buffer that is slightly larger than an even power-of-two number of pages. In that case, alloc_pages() will waste a lot of memory. I have a video driver that wants to allocate a 5MB buffer. alloc_pages() wiill waste 3MB of physically-contiguous memory. Signed-off-by: Timur Tabi Cc: Andi Kleen Acked-by: Mel Gorman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/gfp.h | 3 +++ mm/page_alloc.c | 53 +++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 56 insertions(+) diff --git a/include/linux/gfp.h b/include/linux/gfp.h index f640ed241422..e8003afeffba 100644 --- a/include/linux/gfp.h +++ b/include/linux/gfp.h @@ -228,6 +228,9 @@ extern struct page *alloc_page_vma(gfp_t gfp_mask, extern unsigned long __get_free_pages(gfp_t gfp_mask, unsigned int order); extern unsigned long get_zeroed_page(gfp_t gfp_mask); +void *alloc_pages_exact(size_t size, gfp_t gfp_mask); +void free_pages_exact(void *virt, size_t size); + #define __get_free_page(gfp_mask) \ __get_free_pages((gfp_mask),0) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index eaa86671ebbd..8d528d57b403 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1697,6 +1697,59 @@ void free_pages(unsigned long addr, unsigned int order) EXPORT_SYMBOL(free_pages); +/** + * alloc_pages_exact - allocate an exact number physically-contiguous pages. + * @size: the number of bytes to allocate + * @gfp_mask: GFP flags for the allocation + * + * This function is similar to alloc_pages(), except that it allocates the + * minimum number of pages to satisfy the request. alloc_pages() can only + * allocate memory in power-of-two pages. + * + * This function is also limited by MAX_ORDER. + * + * Memory allocated by this function must be released by free_pages_exact(). + */ +void *alloc_pages_exact(size_t size, gfp_t gfp_mask) +{ + unsigned int order = get_order(size); + unsigned long addr; + + addr = __get_free_pages(gfp_mask, order); + if (addr) { + unsigned long alloc_end = addr + (PAGE_SIZE << order); + unsigned long used = addr + PAGE_ALIGN(size); + + split_page(virt_to_page(addr), order); + while (used < alloc_end) { + free_page(used); + used += PAGE_SIZE; + } + } + + return (void *)addr; +} +EXPORT_SYMBOL(alloc_pages_exact); + +/** + * free_pages_exact - release memory allocated via alloc_pages_exact() + * @virt: the value returned by alloc_pages_exact. + * @size: size of allocation, same value as passed to alloc_pages_exact(). + * + * Release the memory allocated by a previous call to alloc_pages_exact. + */ +void free_pages_exact(void *virt, size_t size) +{ + unsigned long addr = (unsigned long)virt; + unsigned long end = addr + PAGE_ALIGN(size); + + while (addr < end) { + free_page(addr); + addr += PAGE_SIZE; + } +} +EXPORT_SYMBOL(free_pages_exact); + static unsigned int nr_free_zone_pages(int offset) { struct zoneref *z; From b69a7288ea7bf171328f313f0edae629f50e3bdb Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 23 Jul 2008 21:28:12 -0700 Subject: [PATCH 205/857] mm/page_alloc.c: cleanups This patch contains the following cleanups: - make the following needlessly global variables static: - required_kernelcore - zone_movable_pfn[] - make the following needlessly global functions static: - move_freepages() - move_freepages_block() - setup_pageset() - find_usable_zone_for_movable() - adjust_zone_range_for_zone_movable() - __absent_pages_in_range() - find_min_pfn_for_node() - find_zone_movable_pfns_for_nodes() Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 25 +++++++++++++------------ 1 file changed, 13 insertions(+), 12 deletions(-) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 8d528d57b403..cd4c41432ef6 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -153,9 +153,9 @@ static unsigned long __meminitdata dma_reserve; static unsigned long __meminitdata node_boundary_start_pfn[MAX_NUMNODES]; static unsigned long __meminitdata node_boundary_end_pfn[MAX_NUMNODES]; #endif /* CONFIG_MEMORY_HOTPLUG_RESERVE */ - unsigned long __initdata required_kernelcore; + static unsigned long __initdata required_kernelcore; static unsigned long __initdata required_movablecore; - unsigned long __meminitdata zone_movable_pfn[MAX_NUMNODES]; + static unsigned long __meminitdata zone_movable_pfn[MAX_NUMNODES]; /* movable_zone is the "real" zone pages in ZONE_MOVABLE are taken from */ int movable_zone; @@ -674,9 +674,9 @@ static int fallbacks[MIGRATE_TYPES][MIGRATE_TYPES-1] = { * Note that start_page and end_pages are not aligned on a pageblock * boundary. If alignment is required, use move_freepages_block() */ -int move_freepages(struct zone *zone, - struct page *start_page, struct page *end_page, - int migratetype) +static int move_freepages(struct zone *zone, + struct page *start_page, struct page *end_page, + int migratetype) { struct page *page; unsigned long order; @@ -715,7 +715,8 @@ int move_freepages(struct zone *zone, return pages_moved; } -int move_freepages_block(struct zone *zone, struct page *page, int migratetype) +static int move_freepages_block(struct zone *zone, struct page *page, + int migratetype) { unsigned long start_pfn, end_pfn; struct page *start_page, *end_page; @@ -2652,7 +2653,7 @@ static int zone_batchsize(struct zone *zone) return batch; } -inline void setup_pageset(struct per_cpu_pageset *p, unsigned long batch) +static void setup_pageset(struct per_cpu_pageset *p, unsigned long batch) { struct per_cpu_pages *pcp; @@ -3099,7 +3100,7 @@ void __meminit get_pfn_range_for_nid(unsigned int nid, * assumption is made that zones within a node are ordered in monotonic * increasing memory addresses so that the "highest" populated zone is used */ -void __init find_usable_zone_for_movable(void) +static void __init find_usable_zone_for_movable(void) { int zone_index; for (zone_index = MAX_NR_ZONES - 1; zone_index >= 0; zone_index--) { @@ -3125,7 +3126,7 @@ void __init find_usable_zone_for_movable(void) * highest usable zone for ZONE_MOVABLE. This preserves the assumption that * zones within a node are in order of monotonic increases memory addresses */ -void __meminit adjust_zone_range_for_zone_movable(int nid, +static void __meminit adjust_zone_range_for_zone_movable(int nid, unsigned long zone_type, unsigned long node_start_pfn, unsigned long node_end_pfn, @@ -3186,7 +3187,7 @@ static unsigned long __meminit zone_spanned_pages_in_node(int nid, * Return the number of holes in a range on a node. If nid is MAX_NUMNODES, * then all holes in the requested range will be accounted for. */ -unsigned long __meminit __absent_pages_in_range(int nid, +static unsigned long __meminit __absent_pages_in_range(int nid, unsigned long range_start_pfn, unsigned long range_end_pfn) { @@ -3723,7 +3724,7 @@ static void __init sort_node_map(void) } /* Find the lowest pfn for a node */ -unsigned long __init find_min_pfn_for_node(int nid) +static unsigned long __init find_min_pfn_for_node(int nid) { int i; unsigned long min_pfn = ULONG_MAX; @@ -3795,7 +3796,7 @@ static unsigned long __init early_calculate_totalpages(void) * memory. When they don't, some nodes will have more kernelcore than * others */ -void __init find_zone_movable_pfns_for_nodes(unsigned long *movable_pfn) +static void __init find_zone_movable_pfns_for_nodes(unsigned long *movable_pfn) { int i, nid; unsigned long usable_startpfn; From d92bc318547507a944a22e7ef936793dc0fe167f Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 23 Jul 2008 21:28:12 -0700 Subject: [PATCH 206/857] mm: make register_page_bootmem_info_section() static Make the needlessly global register_page_bootmem_info_section() static. Signed-off-by: Adrian Bunk Acked-by: Yasunori Goto Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memory_hotplug.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index 6e26adc08f14..ec85c37dcfb9 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -86,7 +86,7 @@ void put_page_bootmem(struct page *page) } -void register_page_bootmem_info_section(unsigned long start_pfn) +static void register_page_bootmem_info_section(unsigned long start_pfn) { unsigned long *usemap, mapsize, section_nr, i; struct mem_section *ms; From 27ac792ca0b0a1e7e65f20342260650516c95864 Mon Sep 17 00:00:00 2001 From: Andrea Righi Date: Wed, 23 Jul 2008 21:28:13 -0700 Subject: [PATCH 207/857] PAGE_ALIGN(): correctly handle 64-bit values on 32-bit architectures On 32-bit architectures PAGE_ALIGN() truncates 64-bit values to the 32-bit boundary. For example: u64 val = PAGE_ALIGN(size); always returns a value < 4GB even if size is greater than 4GB. The problem resides in PAGE_MASK definition (from include/asm-x86/page.h for example): #define PAGE_SHIFT 12 #define PAGE_SIZE (_AC(1,UL) << PAGE_SHIFT) #define PAGE_MASK (~(PAGE_SIZE-1)) ... #define PAGE_ALIGN(addr) (((addr)+PAGE_SIZE-1)&PAGE_MASK) The "~" is performed on a 32-bit value, so everything in "and" with PAGE_MASK greater than 4GB will be truncated to the 32-bit boundary. Using the ALIGN() macro seems to be the right way, because it uses typeof(addr) for the mask. Also move the PAGE_ALIGN() definitions out of include/asm-*/page.h in include/linux/mm.h. See also lkml discussion: http://lkml.org/lkml/2008/6/11/237 [akpm@linux-foundation.org: fix drivers/media/video/uvc/uvc_queue.c] [akpm@linux-foundation.org: fix v850] [akpm@linux-foundation.org: fix powerpc] [akpm@linux-foundation.org: fix arm] [akpm@linux-foundation.org: fix mips] [akpm@linux-foundation.org: fix drivers/media/video/pvrusb2/pvrusb2-dvb.c] [akpm@linux-foundation.org: fix drivers/mtd/maps/uclinux.c] [akpm@linux-foundation.org: fix powerpc] Signed-off-by: Andrea Righi Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/kernel/module.c | 1 + arch/arm/plat-omap/fb.c | 1 + arch/avr32/mm/ioremap.c | 1 + arch/h8300/kernel/setup.c | 1 + arch/m68k/amiga/chipram.c | 1 + arch/m68knommu/kernel/setup.c | 1 + arch/mips/kernel/module.c | 1 + arch/mips/sgi-ip27/ip27-klnuma.c | 1 + arch/powerpc/kernel/suspend.c | 1 + arch/powerpc/lib/code-patching.c | 1 + arch/sparc64/kernel/iommu_common.h | 2 +- arch/x86/kernel/module_64.c | 1 + arch/xtensa/kernel/setup.c | 1 + drivers/char/random.c | 1 + drivers/ieee1394/iso.c | 1 + drivers/media/video/pvrusb2/pvrusb2-dvb.c | 1 + drivers/media/video/pvrusb2/pvrusb2-ioread.c | 1 + drivers/media/video/uvc/uvc_queue.c | 1 + drivers/media/video/videobuf-core.c | 1 + drivers/mtd/maps/uclinux.c | 1 + drivers/net/mlx4/eq.c | 1 + drivers/pcmcia/electra_cf.c | 1 + drivers/scsi/sun_esp.c | 1 + drivers/video/acornfb.c | 1 + drivers/video/imxfb.c | 1 + drivers/video/omap/dispc.c | 1 + drivers/video/omap/omapfb_main.c | 1 + drivers/video/pxafb.c | 1 + drivers/video/sa1100fb.c | 1 + include/asm-alpha/page.h | 3 --- include/asm-arm/page-nommu.h | 4 +--- include/asm-arm/page.h | 3 --- include/asm-avr32/page.h | 3 --- include/asm-blackfin/page.h | 3 --- include/asm-cris/page.h | 3 --- include/asm-frv/page.h | 3 --- include/asm-h8300/page.h | 3 --- include/asm-ia64/page.h | 1 - include/asm-m32r/page.h | 3 --- include/asm-m68k/dvma.h | 2 +- include/asm-m68k/page.h | 3 --- include/asm-m68knommu/page.h | 3 --- include/asm-mips/page.h | 3 --- include/asm-mips/processor.h | 2 +- include/asm-mn10300/page.h | 3 --- include/asm-parisc/page.h | 4 ---- include/asm-powerpc/page.h | 3 --- include/asm-s390/page.h | 3 --- include/asm-sh/page.h | 3 --- include/asm-sparc/page_32.h | 3 --- include/asm-sparc/page_64.h | 3 --- include/asm-um/page.h | 3 --- include/asm-v850/page.h | 4 ---- include/asm-x86/page.h | 3 --- include/asm-xtensa/page.h | 2 -- include/linux/mm.h | 3 +++ sound/core/info.c | 1 + 57 files changed, 36 insertions(+), 74 deletions(-) diff --git a/arch/arm/kernel/module.c b/arch/arm/kernel/module.c index 79b7e5cf5416..a68259a0cccd 100644 --- a/arch/arm/kernel/module.c +++ b/arch/arm/kernel/module.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include #include diff --git a/arch/arm/plat-omap/fb.c b/arch/arm/plat-omap/fb.c index 96d6f0619733..5d107520e6b9 100644 --- a/arch/arm/plat-omap/fb.c +++ b/arch/arm/plat-omap/fb.c @@ -23,6 +23,7 @@ #include #include +#include #include #include #include diff --git a/arch/avr32/mm/ioremap.c b/arch/avr32/mm/ioremap.c index 3437c82434ac..f03b79f0e0ab 100644 --- a/arch/avr32/mm/ioremap.c +++ b/arch/avr32/mm/ioremap.c @@ -6,6 +6,7 @@ * published by the Free Software Foundation. */ #include +#include #include #include diff --git a/arch/h8300/kernel/setup.c b/arch/h8300/kernel/setup.c index b1f25c20a5db..7fda657110eb 100644 --- a/arch/h8300/kernel/setup.c +++ b/arch/h8300/kernel/setup.c @@ -20,6 +20,7 @@ #include #include #include +#include #include #include #include diff --git a/arch/m68k/amiga/chipram.c b/arch/m68k/amiga/chipram.c index cbe36538af47..61df1d33c050 100644 --- a/arch/m68k/amiga/chipram.c +++ b/arch/m68k/amiga/chipram.c @@ -9,6 +9,7 @@ #include #include +#include #include #include #include diff --git a/arch/m68knommu/kernel/setup.c b/arch/m68knommu/kernel/setup.c index 03f4fe6a2fc0..5985f1989021 100644 --- a/arch/m68knommu/kernel/setup.c +++ b/arch/m68knommu/kernel/setup.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include #include diff --git a/arch/mips/kernel/module.c b/arch/mips/kernel/module.c index e7ed0ac48537..1f60e27523d9 100644 --- a/arch/mips/kernel/module.c +++ b/arch/mips/kernel/module.c @@ -22,6 +22,7 @@ #include #include +#include #include #include #include diff --git a/arch/mips/sgi-ip27/ip27-klnuma.c b/arch/mips/sgi-ip27/ip27-klnuma.c index 48932ce1d730..d9c79d8be81d 100644 --- a/arch/mips/sgi-ip27/ip27-klnuma.c +++ b/arch/mips/sgi-ip27/ip27-klnuma.c @@ -4,6 +4,7 @@ * Copyright 2000 - 2001 Kanoj Sarcar (kanoj@sgi.com) */ #include +#include #include #include #include diff --git a/arch/powerpc/kernel/suspend.c b/arch/powerpc/kernel/suspend.c index 8cee57107541..6fc6328dc626 100644 --- a/arch/powerpc/kernel/suspend.c +++ b/arch/powerpc/kernel/suspend.c @@ -7,6 +7,7 @@ * Copyright (c) 2001 Patrick Mochel */ +#include #include /* References to section boundaries */ diff --git a/arch/powerpc/lib/code-patching.c b/arch/powerpc/lib/code-patching.c index 0559fe086eb4..7c975d43e3f3 100644 --- a/arch/powerpc/lib/code-patching.c +++ b/arch/powerpc/lib/code-patching.c @@ -10,6 +10,7 @@ #include #include #include +#include #include #include diff --git a/arch/sparc64/kernel/iommu_common.h b/arch/sparc64/kernel/iommu_common.h index f3575a614fa2..53b19c8231a9 100644 --- a/arch/sparc64/kernel/iommu_common.h +++ b/arch/sparc64/kernel/iommu_common.h @@ -23,7 +23,7 @@ #define IO_PAGE_SHIFT 13 #define IO_PAGE_SIZE (1UL << IO_PAGE_SHIFT) #define IO_PAGE_MASK (~(IO_PAGE_SIZE-1)) -#define IO_PAGE_ALIGN(addr) (((addr)+IO_PAGE_SIZE-1)&IO_PAGE_MASK) +#define IO_PAGE_ALIGN(addr) ALIGN(addr, IO_PAGE_SIZE) #define IO_TSB_ENTRIES (128*1024) #define IO_TSB_SIZE (IO_TSB_ENTRIES * 8) diff --git a/arch/x86/kernel/module_64.c b/arch/x86/kernel/module_64.c index 0e867676b5a5..6ba87830d4b1 100644 --- a/arch/x86/kernel/module_64.c +++ b/arch/x86/kernel/module_64.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include diff --git a/arch/xtensa/kernel/setup.c b/arch/xtensa/kernel/setup.c index 5e6d75c9f92b..a00359e8f7a8 100644 --- a/arch/xtensa/kernel/setup.c +++ b/arch/xtensa/kernel/setup.c @@ -16,6 +16,7 @@ #include #include +#include #include #include #include diff --git a/drivers/char/random.c b/drivers/char/random.c index 0cf98bd4f2d2..e0d0e371909c 100644 --- a/drivers/char/random.c +++ b/drivers/char/random.c @@ -236,6 +236,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/ieee1394/iso.c b/drivers/ieee1394/iso.c index 07ca35c98f96..1cf6487b65ba 100644 --- a/drivers/ieee1394/iso.c +++ b/drivers/ieee1394/iso.c @@ -11,6 +11,7 @@ #include #include +#include #include #include "hosts.h" diff --git a/drivers/media/video/pvrusb2/pvrusb2-dvb.c b/drivers/media/video/pvrusb2/pvrusb2-dvb.c index 6ec4bf81fc7f..77b3c3385066 100644 --- a/drivers/media/video/pvrusb2/pvrusb2-dvb.c +++ b/drivers/media/video/pvrusb2/pvrusb2-dvb.c @@ -20,6 +20,7 @@ #include #include +#include #include "dvbdev.h" #include "pvrusb2-debug.h" #include "pvrusb2-hdw-internal.h" diff --git a/drivers/media/video/pvrusb2/pvrusb2-ioread.c b/drivers/media/video/pvrusb2/pvrusb2-ioread.c index 05a1376405e7..b4824782d858 100644 --- a/drivers/media/video/pvrusb2/pvrusb2-ioread.c +++ b/drivers/media/video/pvrusb2/pvrusb2-ioread.c @@ -22,6 +22,7 @@ #include "pvrusb2-debug.h" #include #include +#include #include #include #include diff --git a/drivers/media/video/uvc/uvc_queue.c b/drivers/media/video/uvc/uvc_queue.c index 7388d0cee3d4..5646a6a32939 100644 --- a/drivers/media/video/uvc/uvc_queue.c +++ b/drivers/media/video/uvc/uvc_queue.c @@ -13,6 +13,7 @@ #include #include +#include #include #include #include diff --git a/drivers/media/video/videobuf-core.c b/drivers/media/video/videobuf-core.c index 0a88c44ace00..b7b05842cf28 100644 --- a/drivers/media/video/videobuf-core.c +++ b/drivers/media/video/videobuf-core.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include diff --git a/drivers/mtd/maps/uclinux.c b/drivers/mtd/maps/uclinux.c index c42f4b83f686..3fcf92130aa4 100644 --- a/drivers/mtd/maps/uclinux.c +++ b/drivers/mtd/maps/uclinux.c @@ -15,6 +15,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/net/mlx4/eq.c b/drivers/net/mlx4/eq.c index e141a1513f07..ea3a09aaa844 100644 --- a/drivers/net/mlx4/eq.c +++ b/drivers/net/mlx4/eq.c @@ -33,6 +33,7 @@ #include #include +#include #include #include diff --git a/drivers/pcmcia/electra_cf.c b/drivers/pcmcia/electra_cf.c index c21f9a9c3e3f..a34284b1482a 100644 --- a/drivers/pcmcia/electra_cf.c +++ b/drivers/pcmcia/electra_cf.c @@ -28,6 +28,7 @@ #include #include #include +#include #include #include diff --git a/drivers/scsi/sun_esp.c b/drivers/scsi/sun_esp.c index 2c87db98cdfb..f9cf70151366 100644 --- a/drivers/scsi/sun_esp.c +++ b/drivers/scsi/sun_esp.c @@ -7,6 +7,7 @@ #include #include #include +#include #include #include diff --git a/drivers/video/acornfb.c b/drivers/video/acornfb.c index eedb8285e32f..017233d0c481 100644 --- a/drivers/video/acornfb.c +++ b/drivers/video/acornfb.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/video/imxfb.c b/drivers/video/imxfb.c index 94e4d3ac1a05..0c5a475c1cae 100644 --- a/drivers/video/imxfb.c +++ b/drivers/video/imxfb.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/video/omap/dispc.c b/drivers/video/omap/dispc.c index ab32ceb06178..ab77c51fe9d6 100644 --- a/drivers/video/omap/dispc.c +++ b/drivers/video/omap/dispc.c @@ -20,6 +20,7 @@ */ #include #include +#include #include #include #include diff --git a/drivers/video/omap/omapfb_main.c b/drivers/video/omap/omapfb_main.c index 14d0f7a11145..f85af5c4fa68 100644 --- a/drivers/video/omap/omapfb_main.c +++ b/drivers/video/omap/omapfb_main.c @@ -25,6 +25,7 @@ * 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */ #include +#include #include #include diff --git a/drivers/video/pxafb.c b/drivers/video/pxafb.c index bb2514369507..5e8a140399fc 100644 --- a/drivers/video/pxafb.c +++ b/drivers/video/pxafb.c @@ -30,6 +30,7 @@ #include #include #include +#include #include #include #include diff --git a/drivers/video/sa1100fb.c b/drivers/video/sa1100fb.c index ab2b2110478b..4a9f7e121807 100644 --- a/drivers/video/sa1100fb.c +++ b/drivers/video/sa1100fb.c @@ -167,6 +167,7 @@ #include #include #include +#include #include #include #include diff --git a/include/asm-alpha/page.h b/include/asm-alpha/page.h index 22ff9762d17b..0995f9d13417 100644 --- a/include/asm-alpha/page.h +++ b/include/asm-alpha/page.h @@ -80,9 +80,6 @@ typedef struct page *pgtable_t; #endif /* !__ASSEMBLY__ */ -/* to align the pointer to the (next) page boundary */ -#define PAGE_ALIGN(addr) (((addr)+PAGE_SIZE-1)&PAGE_MASK) - #define __pa(x) ((unsigned long) (x) - PAGE_OFFSET) #define __va(x) ((void *)((unsigned long) (x) + PAGE_OFFSET)) #ifndef CONFIG_DISCONTIGMEM diff --git a/include/asm-arm/page-nommu.h b/include/asm-arm/page-nommu.h index a1bcad060480..ea1cde84f500 100644 --- a/include/asm-arm/page-nommu.h +++ b/include/asm-arm/page-nommu.h @@ -7,6 +7,7 @@ * it under the terms of the GNU General Public License version 2 as * published by the Free Software Foundation. */ + #ifndef _ASMARM_PAGE_NOMMU_H #define _ASMARM_PAGE_NOMMU_H @@ -42,9 +43,6 @@ typedef unsigned long pgprot_t; #define __pmd(x) (x) #define __pgprot(x) (x) -/* to align the pointer to the (next) page boundary */ -#define PAGE_ALIGN(addr) (((addr)+PAGE_SIZE-1)&PAGE_MASK) - extern unsigned long memory_start; extern unsigned long memory_end; diff --git a/include/asm-arm/page.h b/include/asm-arm/page.h index 8e05bdb5f12f..7c5fc5582e5d 100644 --- a/include/asm-arm/page.h +++ b/include/asm-arm/page.h @@ -15,9 +15,6 @@ #define PAGE_SIZE (1UL << PAGE_SHIFT) #define PAGE_MASK (~(PAGE_SIZE-1)) -/* to align the pointer to the (next) page boundary */ -#define PAGE_ALIGN(addr) (((addr)+PAGE_SIZE-1)&PAGE_MASK) - #ifndef __ASSEMBLY__ #ifndef CONFIG_MMU diff --git a/include/asm-avr32/page.h b/include/asm-avr32/page.h index cbbc5ca9728b..f805d1cb11bc 100644 --- a/include/asm-avr32/page.h +++ b/include/asm-avr32/page.h @@ -57,9 +57,6 @@ static inline int get_order(unsigned long size) #endif /* !__ASSEMBLY__ */ -/* Align the pointer to the (next) page boundary */ -#define PAGE_ALIGN(addr) (((addr) + PAGE_SIZE - 1) & PAGE_MASK) - /* * The hardware maps the virtual addresses 0x80000000 -> 0x9fffffff * permanently to the physical addresses 0x00000000 -> 0x1fffffff when diff --git a/include/asm-blackfin/page.h b/include/asm-blackfin/page.h index c7db0220fbd6..344f6a8c1f22 100644 --- a/include/asm-blackfin/page.h +++ b/include/asm-blackfin/page.h @@ -51,9 +51,6 @@ typedef struct page *pgtable_t; #define __pgd(x) ((pgd_t) { (x) } ) #define __pgprot(x) ((pgprot_t) { (x) } ) -/* to align the pointer to the (next) page boundary */ -#define PAGE_ALIGN(addr) (((addr)+PAGE_SIZE-1)&PAGE_MASK) - extern unsigned long memory_start; extern unsigned long memory_end; diff --git a/include/asm-cris/page.h b/include/asm-cris/page.h index c45bb1ef397c..d19272ba6b69 100644 --- a/include/asm-cris/page.h +++ b/include/asm-cris/page.h @@ -60,9 +60,6 @@ typedef struct page *pgtable_t; #define page_to_phys(page) __pa((((page) - mem_map) << PAGE_SHIFT) + PAGE_OFFSET) -/* to align the pointer to the (next) page boundary */ -#define PAGE_ALIGN(addr) (((addr)+PAGE_SIZE-1)&PAGE_MASK) - #ifndef __ASSEMBLY__ #endif /* __ASSEMBLY__ */ diff --git a/include/asm-frv/page.h b/include/asm-frv/page.h index c2c1e89e747d..bd9c220094c7 100644 --- a/include/asm-frv/page.h +++ b/include/asm-frv/page.h @@ -40,9 +40,6 @@ typedef struct page *pgtable_t; #define __pgprot(x) ((pgprot_t) { (x) } ) #define PTE_MASK PAGE_MASK -/* to align the pointer to the (next) page boundary */ -#define PAGE_ALIGN(addr) (((addr) + PAGE_SIZE - 1) & PAGE_MASK) - #define devmem_is_allowed(pfn) 1 #define __pa(vaddr) virt_to_phys((void *) (unsigned long) (vaddr)) diff --git a/include/asm-h8300/page.h b/include/asm-h8300/page.h index d6a3eaf3b27e..0b6acf0b03aa 100644 --- a/include/asm-h8300/page.h +++ b/include/asm-h8300/page.h @@ -43,9 +43,6 @@ typedef struct page *pgtable_t; #define __pgd(x) ((pgd_t) { (x) } ) #define __pgprot(x) ((pgprot_t) { (x) } ) -/* to align the pointer to the (next) page boundary */ -#define PAGE_ALIGN(addr) (((addr)+PAGE_SIZE-1)&PAGE_MASK) - extern unsigned long memory_start; extern unsigned long memory_end; diff --git a/include/asm-ia64/page.h b/include/asm-ia64/page.h index 36f39321b768..5f271bc712ee 100644 --- a/include/asm-ia64/page.h +++ b/include/asm-ia64/page.h @@ -40,7 +40,6 @@ #define PAGE_SIZE (__IA64_UL_CONST(1) << PAGE_SHIFT) #define PAGE_MASK (~(PAGE_SIZE - 1)) -#define PAGE_ALIGN(addr) (((addr) + PAGE_SIZE - 1) & PAGE_MASK) #define PERCPU_PAGE_SHIFT 16 /* log2() of max. size of per-CPU area */ #define PERCPU_PAGE_SIZE (__IA64_UL_CONST(1) << PERCPU_PAGE_SHIFT) diff --git a/include/asm-m32r/page.h b/include/asm-m32r/page.h index 8a677f3fca68..c9333089fe11 100644 --- a/include/asm-m32r/page.h +++ b/include/asm-m32r/page.h @@ -41,9 +41,6 @@ typedef struct page *pgtable_t; #endif /* !__ASSEMBLY__ */ -/* to align the pointer to the (next) page boundary */ -#define PAGE_ALIGN(addr) (((addr) + PAGE_SIZE - 1) & PAGE_MASK) - /* * This handles the memory map.. We could make this a config * option, but too many people screw it up, and too few need diff --git a/include/asm-m68k/dvma.h b/include/asm-m68k/dvma.h index 4fff408d0150..890bbf7e7758 100644 --- a/include/asm-m68k/dvma.h +++ b/include/asm-m68k/dvma.h @@ -13,7 +13,7 @@ #define DVMA_PAGE_SHIFT 13 #define DVMA_PAGE_SIZE (1UL << DVMA_PAGE_SHIFT) #define DVMA_PAGE_MASK (~(DVMA_PAGE_SIZE-1)) -#define DVMA_PAGE_ALIGN(addr) (((addr)+DVMA_PAGE_SIZE-1)&DVMA_PAGE_MASK) +#define DVMA_PAGE_ALIGN(addr) ALIGN(addr, DVMA_PAGE_SIZE) extern void dvma_init(void); extern int dvma_map_iommu(unsigned long kaddr, unsigned long baddr, diff --git a/include/asm-m68k/page.h b/include/asm-m68k/page.h index 880c2cbff8a6..a34b8bad7847 100644 --- a/include/asm-m68k/page.h +++ b/include/asm-m68k/page.h @@ -103,9 +103,6 @@ typedef struct page *pgtable_t; #define __pgd(x) ((pgd_t) { (x) } ) #define __pgprot(x) ((pgprot_t) { (x) } ) -/* to align the pointer to the (next) page boundary */ -#define PAGE_ALIGN(addr) (((addr)+PAGE_SIZE-1)&PAGE_MASK) - #endif /* !__ASSEMBLY__ */ #include diff --git a/include/asm-m68knommu/page.h b/include/asm-m68knommu/page.h index 1e82ebb7d644..3a1ede4544cb 100644 --- a/include/asm-m68knommu/page.h +++ b/include/asm-m68knommu/page.h @@ -43,9 +43,6 @@ typedef struct page *pgtable_t; #define __pgd(x) ((pgd_t) { (x) } ) #define __pgprot(x) ((pgprot_t) { (x) } ) -/* to align the pointer to the (next) page boundary */ -#define PAGE_ALIGN(addr) (((addr)+PAGE_SIZE-1)&PAGE_MASK) - extern unsigned long memory_start; extern unsigned long memory_end; diff --git a/include/asm-mips/page.h b/include/asm-mips/page.h index 494f00ba9541..fe7a88ea066e 100644 --- a/include/asm-mips/page.h +++ b/include/asm-mips/page.h @@ -137,9 +137,6 @@ typedef struct { unsigned long pgprot; } pgprot_t; #endif /* !__ASSEMBLY__ */ -/* to align the pointer to the (next) page boundary */ -#define PAGE_ALIGN(addr) (((addr) + PAGE_SIZE - 1) & PAGE_MASK) - /* * __pa()/__va() should be used only during mem init. */ diff --git a/include/asm-mips/processor.h b/include/asm-mips/processor.h index 58cbac5a64e4..a1e4453469f9 100644 --- a/include/asm-mips/processor.h +++ b/include/asm-mips/processor.h @@ -45,7 +45,7 @@ extern unsigned int vced_count, vcei_count; * This decides where the kernel will search for a free chunk of vm * space during mmap's. */ -#define TASK_UNMAPPED_BASE (PAGE_ALIGN(TASK_SIZE / 3)) +#define TASK_UNMAPPED_BASE ((TASK_SIZE / 3) & ~(PAGE_SIZE)) #endif #ifdef CONFIG_64BIT diff --git a/include/asm-mn10300/page.h b/include/asm-mn10300/page.h index 124971b9fb9b..8288e124165b 100644 --- a/include/asm-mn10300/page.h +++ b/include/asm-mn10300/page.h @@ -61,9 +61,6 @@ typedef struct page *pgtable_t; #endif /* !__ASSEMBLY__ */ -/* to align the pointer to the (next) page boundary */ -#define PAGE_ALIGN(addr) (((addr) + PAGE_SIZE - 1) & PAGE_MASK) - /* * This handles the memory map.. We could make this a config * option, but too many people screw it up, and too few need diff --git a/include/asm-parisc/page.h b/include/asm-parisc/page.h index 27d50b859541..c3941f09a878 100644 --- a/include/asm-parisc/page.h +++ b/include/asm-parisc/page.h @@ -119,10 +119,6 @@ extern int npmem_ranges; #define PMD_ENTRY_SIZE (1UL << BITS_PER_PMD_ENTRY) #define PTE_ENTRY_SIZE (1UL << BITS_PER_PTE_ENTRY) -/* to align the pointer to the (next) page boundary */ -#define PAGE_ALIGN(addr) (((addr)+PAGE_SIZE-1)&PAGE_MASK) - - #define LINUX_GATEWAY_SPACE 0 /* This governs the relationship between virtual and physical addresses. diff --git a/include/asm-powerpc/page.h b/include/asm-powerpc/page.h index cffdf0eb0df6..e088545cb3f5 100644 --- a/include/asm-powerpc/page.h +++ b/include/asm-powerpc/page.h @@ -119,9 +119,6 @@ extern phys_addr_t kernstart_addr; /* align addr on a size boundary - adjust address up if needed */ #define _ALIGN(addr,size) _ALIGN_UP(addr,size) -/* to align the pointer to the (next) page boundary */ -#define PAGE_ALIGN(addr) _ALIGN(addr, PAGE_SIZE) - /* * Don't compare things with KERNELBASE or PAGE_OFFSET to test for * "kernelness", use is_kernel_addr() - it should do what you want. diff --git a/include/asm-s390/page.h b/include/asm-s390/page.h index 12fd9c4f0f15..991ba939408c 100644 --- a/include/asm-s390/page.h +++ b/include/asm-s390/page.h @@ -138,9 +138,6 @@ void arch_alloc_page(struct page *page, int order); #endif /* !__ASSEMBLY__ */ -/* to align the pointer to the (next) page boundary */ -#define PAGE_ALIGN(addr) (((addr)+PAGE_SIZE-1)&PAGE_MASK) - #define __PAGE_OFFSET 0x0UL #define PAGE_OFFSET 0x0UL #define __pa(x) (unsigned long)(x) diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h index 304c30b5d947..5dc01d2fcc4c 100644 --- a/include/asm-sh/page.h +++ b/include/asm-sh/page.h @@ -22,9 +22,6 @@ #define PAGE_MASK (~(PAGE_SIZE-1)) #define PTE_MASK PAGE_MASK -/* to align the pointer to the (next) page boundary */ -#define PAGE_ALIGN(addr) (((addr)+PAGE_SIZE-1)&PAGE_MASK) - #if defined(CONFIG_HUGETLB_PAGE_SIZE_64K) #define HPAGE_SHIFT 16 #elif defined(CONFIG_HUGETLB_PAGE_SIZE_256K) diff --git a/include/asm-sparc/page_32.h b/include/asm-sparc/page_32.h index 14de518cc38f..cf5fb70ca1c1 100644 --- a/include/asm-sparc/page_32.h +++ b/include/asm-sparc/page_32.h @@ -134,9 +134,6 @@ BTFIXUPDEF_SETHI(sparc_unmapped_base) #endif /* !(__ASSEMBLY__) */ -/* to align the pointer to the (next) page boundary */ -#define PAGE_ALIGN(addr) (((addr)+PAGE_SIZE-1)&PAGE_MASK) - #define PAGE_OFFSET 0xf0000000 #ifndef __ASSEMBLY__ extern unsigned long phys_base; diff --git a/include/asm-sparc/page_64.h b/include/asm-sparc/page_64.h index a8a2bba032c1..b579b910ef51 100644 --- a/include/asm-sparc/page_64.h +++ b/include/asm-sparc/page_64.h @@ -106,9 +106,6 @@ typedef struct page *pgtable_t; #endif /* !(__ASSEMBLY__) */ -/* to align the pointer to the (next) page boundary */ -#define PAGE_ALIGN(addr) (((addr)+PAGE_SIZE-1)&PAGE_MASK) - /* We used to stick this into a hard-coded global register (%g4) * but that does not make sense anymore. */ diff --git a/include/asm-um/page.h b/include/asm-um/page.h index 916e1a61999f..335c57383c02 100644 --- a/include/asm-um/page.h +++ b/include/asm-um/page.h @@ -92,9 +92,6 @@ typedef struct page *pgtable_t; #define __pgd(x) ((pgd_t) { (x) } ) #define __pgprot(x) ((pgprot_t) { (x) } ) -/* to align the pointer to the (next) page boundary */ -#define PAGE_ALIGN(addr) (((addr)+PAGE_SIZE-1)&PAGE_MASK) - extern unsigned long uml_physmem; #define PAGE_OFFSET (uml_physmem) diff --git a/include/asm-v850/page.h b/include/asm-v850/page.h index 74a539a9bd59..f9de35d873fa 100644 --- a/include/asm-v850/page.h +++ b/include/asm-v850/page.h @@ -94,10 +94,6 @@ typedef unsigned long pgprot_t; #endif /* !__ASSEMBLY__ */ -/* to align the pointer to the (next) page boundary */ -#define PAGE_ALIGN(addr) (((addr) + PAGE_SIZE - 1) & PAGE_MASK) - - /* No current v850 processor has virtual memory. */ #define __virt_to_phys(addr) (addr) #define __phys_to_virt(addr) (addr) diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h index 6e02098b1605..49982110e4d9 100644 --- a/include/asm-x86/page.h +++ b/include/asm-x86/page.h @@ -34,9 +34,6 @@ #define HUGE_MAX_HSTATE 2 -/* to align the pointer to the (next) page boundary */ -#define PAGE_ALIGN(addr) (((addr)+PAGE_SIZE-1)&PAGE_MASK) - #ifndef __ASSEMBLY__ #include #endif diff --git a/include/asm-xtensa/page.h b/include/asm-xtensa/page.h index 80a6ae0dd259..11f7dc2dbec7 100644 --- a/include/asm-xtensa/page.h +++ b/include/asm-xtensa/page.h @@ -26,13 +26,11 @@ /* * PAGE_SHIFT determines the page size - * PAGE_ALIGN(x) aligns the pointer to the (next) page boundary */ #define PAGE_SHIFT 12 #define PAGE_SIZE (__XTENSA_UL_CONST(1) << PAGE_SHIFT) #define PAGE_MASK (~(PAGE_SIZE-1)) -#define PAGE_ALIGN(addr) (((addr)+PAGE_SIZE - 1) & PAGE_MASK) #define PAGE_OFFSET XCHAL_KSEG_CACHED_VADDR #define MAX_MEM_PFN XCHAL_KSEG_SIZE diff --git a/include/linux/mm.h b/include/linux/mm.h index df322fb4df31..d87a5a5fe87d 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -41,6 +41,9 @@ extern unsigned long mmap_min_addr; #define nth_page(page,n) pfn_to_page(page_to_pfn((page)) + (n)) +/* to align the pointer to the (next) page boundary */ +#define PAGE_ALIGN(addr) ALIGN(addr, PAGE_SIZE) + /* * Linux kernel virtual memory manager primitives. * The idea being to have a "virtual" mm in the same way diff --git a/sound/core/info.c b/sound/core/info.c index cb5ead3e202d..c67773ad9298 100644 --- a/sound/core/info.c +++ b/sound/core/info.c @@ -21,6 +21,7 @@ #include #include +#include #include #include #include From f84f9504bddeec33a72d64ebe95143d3aaeb3f9b Mon Sep 17 00:00:00 2001 From: Vegard Nossum Date: Wed, 23 Jul 2008 21:28:14 -0700 Subject: [PATCH 208/857] mm: remove initialization of static per-cpu variables This was required by some old, no-longer-used gcc on sparc. Signed-off-by: Vegard Nossum Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/swap.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/mm/swap.c b/mm/swap.c index 45c9f25a8a3b..dd89234ee51f 100644 --- a/mm/swap.c +++ b/mm/swap.c @@ -34,9 +34,9 @@ /* How many pages do we try to swap or page in/out together? */ int page_cluster; -static DEFINE_PER_CPU(struct pagevec, lru_add_pvecs) = { 0, }; -static DEFINE_PER_CPU(struct pagevec, lru_add_active_pvecs) = { 0, }; -static DEFINE_PER_CPU(struct pagevec, lru_rotate_pvecs) = { 0, }; +static DEFINE_PER_CPU(struct pagevec, lru_add_pvecs); +static DEFINE_PER_CPU(struct pagevec, lru_add_active_pvecs); +static DEFINE_PER_CPU(struct pagevec, lru_rotate_pvecs); /* * This path almost never happens for VM activity - pages are normally @@ -493,7 +493,7 @@ EXPORT_SYMBOL(pagevec_lookup_tag); */ #define ACCT_THRESHOLD max(16, NR_CPUS * 2) -static DEFINE_PER_CPU(long, committed_space) = 0; +static DEFINE_PER_CPU(long, committed_space); void vm_acct_memory(long pages) { From 48c906823f3927b981db9f0b03c2e2499977ee93 Mon Sep 17 00:00:00 2001 From: Yasunori Goto Date: Wed, 23 Jul 2008 21:28:15 -0700 Subject: [PATCH 209/857] memory hotplug: allocate usemap on the section with pgdat Usemaps are allocated on the section which has pgdat by this. Because usemap size is very small, many other sections usemaps are allocated on only one page. If a section has usemap, it can't be removed until removing other sections. This dependency is not desirable for memory removing. Pgdat has similar feature. When a section has pgdat area, it must be the last section for removing on the node. So, if section A has pgdat and section B has usemap for section A, Both sections can't be removed due to dependency each other. To solve this issue, this patch collects usemap on same section with pgdat as much as possible. If other sections doesn't have any dependency, this section will be able to be removed finally. Signed-off-by: Yasunori Goto Cc: Mel Gorman Cc: Andy Whitcroft Cc: David Miller Cc: Badari Pulavarty Cc: Heiko Carstens Cc: Hiroyuki KAMEZAWA Cc: Tony Breeds Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/sparse.c | 78 ++++++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 77 insertions(+), 1 deletion(-) diff --git a/mm/sparse.c b/mm/sparse.c index 7a3650923d9a..8ffc08990008 100644 --- a/mm/sparse.c +++ b/mm/sparse.c @@ -269,16 +269,92 @@ static unsigned long *__kmalloc_section_usemap(void) } #endif /* CONFIG_MEMORY_HOTPLUG */ +#ifdef CONFIG_MEMORY_HOTREMOVE +static unsigned long * __init +sparse_early_usemap_alloc_pgdat_section(struct pglist_data *pgdat) +{ + unsigned long section_nr; + + /* + * A page may contain usemaps for other sections preventing the + * page being freed and making a section unremovable while + * other sections referencing the usemap retmain active. Similarly, + * a pgdat can prevent a section being removed. If section A + * contains a pgdat and section B contains the usemap, both + * sections become inter-dependent. This allocates usemaps + * from the same section as the pgdat where possible to avoid + * this problem. + */ + section_nr = pfn_to_section_nr(__pa(pgdat) >> PAGE_SHIFT); + return alloc_bootmem_section(usemap_size(), section_nr); +} + +static void __init check_usemap_section_nr(int nid, unsigned long *usemap) +{ + unsigned long usemap_snr, pgdat_snr; + static unsigned long old_usemap_snr = NR_MEM_SECTIONS; + static unsigned long old_pgdat_snr = NR_MEM_SECTIONS; + struct pglist_data *pgdat = NODE_DATA(nid); + int usemap_nid; + + usemap_snr = pfn_to_section_nr(__pa(usemap) >> PAGE_SHIFT); + pgdat_snr = pfn_to_section_nr(__pa(pgdat) >> PAGE_SHIFT); + if (usemap_snr == pgdat_snr) + return; + + if (old_usemap_snr == usemap_snr && old_pgdat_snr == pgdat_snr) + /* skip redundant message */ + return; + + old_usemap_snr = usemap_snr; + old_pgdat_snr = pgdat_snr; + + usemap_nid = sparse_early_nid(__nr_to_section(usemap_snr)); + if (usemap_nid != nid) { + printk(KERN_INFO + "node %d must be removed before remove section %ld\n", + nid, usemap_snr); + return; + } + /* + * There is a circular dependency. + * Some platforms allow un-removable section because they will just + * gather other removable sections for dynamic partitioning. + * Just notify un-removable section's number here. + */ + printk(KERN_INFO "Section %ld and %ld (node %d)", usemap_snr, + pgdat_snr, nid); + printk(KERN_CONT + " have a circular dependency on usemap and pgdat allocations\n"); +} +#else +static unsigned long * __init +sparse_early_usemap_alloc_pgdat_section(struct pglist_data *pgdat) +{ + return NULL; +} + +static void __init check_usemap_section_nr(int nid, unsigned long *usemap) +{ +} +#endif /* CONFIG_MEMORY_HOTREMOVE */ + static unsigned long *__init sparse_early_usemap_alloc(unsigned long pnum) { unsigned long *usemap; struct mem_section *ms = __nr_to_section(pnum); int nid = sparse_early_nid(ms); - usemap = alloc_bootmem_node(NODE_DATA(nid), usemap_size()); + usemap = sparse_early_usemap_alloc_pgdat_section(NODE_DATA(nid)); if (usemap) return usemap; + usemap = alloc_bootmem_node(NODE_DATA(nid), usemap_size()); + if (usemap) { + check_usemap_section_nr(nid, usemap); + return usemap; + } + /* Stupid: suppress gcc warning for SPARSEMEM && !NUMA */ nid = 0; From af370fb8cb3031f20438f246798d5f0d98089f29 Mon Sep 17 00:00:00 2001 From: Yasunori Goto Date: Wed, 23 Jul 2008 21:28:17 -0700 Subject: [PATCH 210/857] memory hotplug: small fixes to bootmem freeing for memory hotremove - Change some naming * Magic -> types * MIX_INFO -> MIX_SECTION_INFO * Change definition of bootmem type from direct hex value - __free_pages_bootmem() becomes __meminit. Signed-off-by: Yasunori Goto Cc: Andy Whitcroft Cc: Badari Pulavarty Cc: Yinghai Lu Cc: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/memory_hotplug.h | 8 ++++---- mm/memory_hotplug.c | 12 ++++++------ mm/page_alloc.c | 2 +- 3 files changed, 11 insertions(+), 11 deletions(-) diff --git a/include/linux/memory_hotplug.h b/include/linux/memory_hotplug.h index ea9f5ad9ec8e..3628e5088f64 100644 --- a/include/linux/memory_hotplug.h +++ b/include/linux/memory_hotplug.h @@ -13,12 +13,12 @@ struct mem_section; #ifdef CONFIG_MEMORY_HOTPLUG /* - * Magic number for free bootmem. + * Types for free bootmem. * The normal smallest mapcount is -1. Here is smaller value than it. */ -#define SECTION_INFO 0xfffffffe -#define MIX_INFO 0xfffffffd -#define NODE_INFO 0xfffffffc +#define SECTION_INFO (-1 - 1) +#define MIX_SECTION_INFO (-1 - 2) +#define NODE_INFO (-1 - 3) /* * pgdat resizing functions diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index ec85c37dcfb9..0fb05b258f0c 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -62,9 +62,9 @@ static void release_memory_resource(struct resource *res) #ifdef CONFIG_MEMORY_HOTPLUG_SPARSE #ifndef CONFIG_SPARSEMEM_VMEMMAP -static void get_page_bootmem(unsigned long info, struct page *page, int magic) +static void get_page_bootmem(unsigned long info, struct page *page, int type) { - atomic_set(&page->_mapcount, magic); + atomic_set(&page->_mapcount, type); SetPagePrivate(page); set_page_private(page, info); atomic_inc(&page->_count); @@ -72,10 +72,10 @@ static void get_page_bootmem(unsigned long info, struct page *page, int magic) void put_page_bootmem(struct page *page) { - int magic; + int type; - magic = atomic_read(&page->_mapcount); - BUG_ON(magic >= -1); + type = atomic_read(&page->_mapcount); + BUG_ON(type >= -1); if (atomic_dec_return(&page->_count) == 1) { ClearPagePrivate(page); @@ -119,7 +119,7 @@ static void register_page_bootmem_info_section(unsigned long start_pfn) mapsize = PAGE_ALIGN(usemap_size()) >> PAGE_SHIFT; for (i = 0; i < mapsize; i++, page++) - get_page_bootmem(section_nr, page, MIX_INFO); + get_page_bootmem(section_nr, page, MIX_SECTION_INFO); } diff --git a/mm/page_alloc.c b/mm/page_alloc.c index cd4c41432ef6..6da667274df5 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -533,7 +533,7 @@ static void __free_pages_ok(struct page *page, unsigned int order) /* * permit the bootmem allocator to evade page validation on high-order frees */ -void __free_pages_bootmem(struct page *page, unsigned int order) +void __meminit __free_pages_bootmem(struct page *page, unsigned int order) { if (order == 0) { __ClearPageReserved(page); From 2f7f24eca31c4fc2fdb134b2ef743ccd67cfb9a9 Mon Sep 17 00:00:00 2001 From: Kent Liu Date: Wed, 23 Jul 2008 21:28:18 -0700 Subject: [PATCH 211/857] memory-hotplug: don't calculate vm_total_pages twice when rebuilding zonelists in online_pages() If zonelist is required to be rebuilt in online_pages(), there is no need to recalculate vm_total_pages in that function, as it has been updated in the call build_all_zonelists(). Signed-off-by: Kent Liu Acked-by: KAMEZAWA Hiroyuki Cc: Yasunori Goto Cc: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memory_hotplug.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index 0fb05b258f0c..93aba78dc8b6 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -429,7 +429,9 @@ int online_pages(unsigned long pfn, unsigned long nr_pages) if (need_zonelists_rebuild) build_all_zonelists(); - vm_total_pages = nr_free_pagecache_pages(); + else + vm_total_pages = nr_free_pagecache_pages(); + writeback_set_ratelimit(); if (onlined_pages) From 5c755e9fd813810680abd56ec09a5f90143e815b Mon Sep 17 00:00:00 2001 From: Badari Pulavarty Date: Wed, 23 Jul 2008 21:28:19 -0700 Subject: [PATCH 212/857] memory-hotplug: add sysfs removable attribute for hotplug memory remove Memory may be hot-removed on a per-memory-block basis, particularly on POWER where the SPARSEMEM section size often matches the memory-block size. A user-level agent must be able to identify which sections of memory are likely to be removable before attempting the potentially expensive operation. This patch adds a file called "removable" to the memory directory in sysfs to help such an agent. In this patch, a memory block is considered removable if; o It contains only MOVABLE pageblocks o It contains only pageblocks with free pages regardless of pageblock type On the other hand, a memory block starting with a PageReserved() page will never be considered removable. Without this patch, the user-agent is forced to choose a memory block to remove randomly. Sample output of the sysfs files: ./memory/memory0/removable: 0 ./memory/memory1/removable: 0 ./memory/memory2/removable: 0 ./memory/memory3/removable: 0 ./memory/memory4/removable: 0 ./memory/memory5/removable: 0 ./memory/memory6/removable: 0 ./memory/memory7/removable: 1 ./memory/memory8/removable: 0 ./memory/memory9/removable: 0 ./memory/memory10/removable: 0 ./memory/memory11/removable: 0 ./memory/memory12/removable: 0 ./memory/memory13/removable: 0 ./memory/memory14/removable: 0 ./memory/memory15/removable: 0 ./memory/memory16/removable: 0 ./memory/memory17/removable: 1 ./memory/memory18/removable: 1 ./memory/memory19/removable: 1 ./memory/memory20/removable: 1 ./memory/memory21/removable: 1 ./memory/memory22/removable: 1 Signed-off-by: Badari Pulavarty Signed-off-by: Mel Gorman Acked-by: KAMEZAWA Hiroyuki Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- .../ABI/testing/sysfs-devices-memory | 24 ++++++++ drivers/base/memory.c | 19 ++++++ include/linux/memory_hotplug.h | 12 ++++ mm/memory_hotplug.c | 60 +++++++++++++++++++ 4 files changed, 115 insertions(+) create mode 100644 Documentation/ABI/testing/sysfs-devices-memory diff --git a/Documentation/ABI/testing/sysfs-devices-memory b/Documentation/ABI/testing/sysfs-devices-memory new file mode 100644 index 000000000000..7a16fe1e2270 --- /dev/null +++ b/Documentation/ABI/testing/sysfs-devices-memory @@ -0,0 +1,24 @@ +What: /sys/devices/system/memory +Date: June 2008 +Contact: Badari Pulavarty +Description: + The /sys/devices/system/memory contains a snapshot of the + internal state of the kernel memory blocks. Files could be + added or removed dynamically to represent hot-add/remove + operations. + +Users: hotplug memory add/remove tools + https://w3.opensource.ibm.com/projects/powerpc-utils/ + +What: /sys/devices/system/memory/memoryX/removable +Date: June 2008 +Contact: Badari Pulavarty +Description: + The file /sys/devices/system/memory/memoryX/removable + indicates whether this memory block is removable or not. + This is useful for a user-level agent to determine + identify removable sections of the memory before attempting + potentially expensive hot-remove memory operation + +Users: hotplug memory remove tools + https://w3.opensource.ibm.com/projects/powerpc-utils/ diff --git a/drivers/base/memory.c b/drivers/base/memory.c index 4d4e0e7b6e92..855ed1a9f97b 100644 --- a/drivers/base/memory.c +++ b/drivers/base/memory.c @@ -100,6 +100,21 @@ static ssize_t show_mem_phys_index(struct sys_device *dev, return sprintf(buf, "%08lx\n", mem->phys_index); } +/* + * Show whether the section of memory is likely to be hot-removable + */ +static ssize_t show_mem_removable(struct sys_device *dev, char *buf) +{ + unsigned long start_pfn; + int ret; + struct memory_block *mem = + container_of(dev, struct memory_block, sysdev); + + start_pfn = section_nr_to_pfn(mem->phys_index); + ret = is_mem_section_removable(start_pfn, PAGES_PER_SECTION); + return sprintf(buf, "%d\n", ret); +} + /* * online, offline, going offline, etc. */ @@ -262,6 +277,7 @@ static ssize_t show_phys_device(struct sys_device *dev, static SYSDEV_ATTR(phys_index, 0444, show_mem_phys_index, NULL); static SYSDEV_ATTR(state, 0644, show_mem_state, store_mem_state); static SYSDEV_ATTR(phys_device, 0444, show_phys_device, NULL); +static SYSDEV_ATTR(removable, 0444, show_mem_removable, NULL); #define mem_create_simple_file(mem, attr_name) \ sysdev_create_file(&mem->sysdev, &attr_##attr_name) @@ -350,6 +366,8 @@ static int add_memory_block(unsigned long node_id, struct mem_section *section, ret = mem_create_simple_file(mem, state); if (!ret) ret = mem_create_simple_file(mem, phys_device); + if (!ret) + ret = mem_create_simple_file(mem, removable); return ret; } @@ -394,6 +412,7 @@ int remove_memory_block(unsigned long node_id, struct mem_section *section, mem_remove_simple_file(mem, phys_index); mem_remove_simple_file(mem, state); mem_remove_simple_file(mem, phys_device); + mem_remove_simple_file(mem, removable); unregister_memory(mem, section); return 0; diff --git a/include/linux/memory_hotplug.h b/include/linux/memory_hotplug.h index 3628e5088f64..763ba81fc0f0 100644 --- a/include/linux/memory_hotplug.h +++ b/include/linux/memory_hotplug.h @@ -199,6 +199,18 @@ extern int walk_memory_resource(unsigned long start_pfn, unsigned long nr_pages, void *arg, int (*func)(unsigned long, unsigned long, void *)); +#ifdef CONFIG_MEMORY_HOTREMOVE + +extern int is_mem_section_removable(unsigned long pfn, unsigned long nr_pages); + +#else +static inline int is_mem_section_removable(unsigned long pfn, + unsigned long nr_pages) +{ + return 0; +} +#endif /* CONFIG_MEMORY_HOTREMOVE */ + extern int add_memory(int nid, u64 start, u64 size); extern int arch_add_memory(int nid, u64 start, u64 size); extern int remove_memory(u64 start, u64 size); diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index 93aba78dc8b6..89fee2dcb039 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -522,6 +522,66 @@ int add_memory(int nid, u64 start, u64 size) EXPORT_SYMBOL_GPL(add_memory); #ifdef CONFIG_MEMORY_HOTREMOVE +/* + * A free page on the buddy free lists (not the per-cpu lists) has PageBuddy + * set and the size of the free page is given by page_order(). Using this, + * the function determines if the pageblock contains only free pages. + * Due to buddy contraints, a free page at least the size of a pageblock will + * be located at the start of the pageblock + */ +static inline int pageblock_free(struct page *page) +{ + return PageBuddy(page) && page_order(page) >= pageblock_order; +} + +/* Return the start of the next active pageblock after a given page */ +static struct page *next_active_pageblock(struct page *page) +{ + int pageblocks_stride; + + /* Ensure the starting page is pageblock-aligned */ + BUG_ON(page_to_pfn(page) & (pageblock_nr_pages - 1)); + + /* Move forward by at least 1 * pageblock_nr_pages */ + pageblocks_stride = 1; + + /* If the entire pageblock is free, move to the end of free page */ + if (pageblock_free(page)) + pageblocks_stride += page_order(page) - pageblock_order; + + return page + (pageblocks_stride * pageblock_nr_pages); +} + +/* Checks if this range of memory is likely to be hot-removable. */ +int is_mem_section_removable(unsigned long start_pfn, unsigned long nr_pages) +{ + int type; + struct page *page = pfn_to_page(start_pfn); + struct page *end_page = page + nr_pages; + + /* Check the starting page of each pageblock within the range */ + for (; page < end_page; page = next_active_pageblock(page)) { + type = get_pageblock_migratetype(page); + + /* + * A pageblock containing MOVABLE or free pages is considered + * removable + */ + if (type != MIGRATE_MOVABLE && !pageblock_free(page)) + return 0; + + /* + * A pageblock starting with a PageReserved page is not + * considered removable. + */ + if (PageReserved(page)) + return 0; + } + + /* All pageblocks in the memory block are likely to be hot-removable */ + return 1; +} + /* * Confirm all pages in a range [start, end) is belongs to the same zone. */ From 9ca908f47bc784c90e17a553ce33e756c73feac4 Mon Sep 17 00:00:00 2001 From: Milton Miller Date: Wed, 23 Jul 2008 21:28:20 -0700 Subject: [PATCH 213/857] kcalloc: remove runtime division While in all cases in the kernel we know the size of the elements to be created, we don't always know the count of elements. By commuting the size and count in the overflow check, the compiler can reduce the runtime division of size_t with a compare to a (unique) constant in these cases. Signed-off-by: Milton Miller Cc: Takashi Iwai Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/slab.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/slab.h b/include/linux/slab.h index 9aa90a6f20e0..41103910f8a2 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -180,7 +180,7 @@ size_t ksize(const void *); */ static inline void *kcalloc(size_t n, size_t size, gfp_t flags) { - if (n != 0 && size > ULONG_MAX / n) + if (size != 0 && n > ULONG_MAX / size) return NULL; return __kmalloc(n * size, flags | __GFP_ZERO); } From 83d1674a946141c3c59d430e96c224f7937e6158 Mon Sep 17 00:00:00 2001 From: Gerald Schaefer Date: Wed, 23 Jul 2008 21:28:22 -0700 Subject: [PATCH 214/857] mm: make CONFIG_MIGRATION available w/o CONFIG_NUMA We'd like to support CONFIG_MEMORY_HOTREMOVE on s390, which depends on CONFIG_MIGRATION. So far, CONFIG_MIGRATION is only available with NUMA support. This patch makes CONFIG_MIGRATION selectable for architectures that define ARCH_ENABLE_MEMORY_HOTREMOVE. When MIGRATION is enabled w/o NUMA, the kernel won't compile because migrate_vmas() does not know about vm_ops->migrate() and vma_migratable() does not know about policy_zone. To fix this, those two functions can be restricted to '#ifdef CONFIG_NUMA' because they are not being used w/o NUMA. vma_migratable() is moved over from migrate.h to mempolicy.h. [kosaki.motohiro@jp.fujitsu.com: build fix] Acked-by: Christoph Lameter Signed-off-by: Gerald Schaefer Cc: Martin Schwidefsky Cc: Heiko Carstens Signed-off-by: KOSAKI Motorhiro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mempolicy.h | 19 +++++++++++++++++++ include/linux/migrate.h | 21 --------------------- mm/Kconfig | 2 +- mm/migrate.c | 2 +- 4 files changed, 21 insertions(+), 23 deletions(-) diff --git a/include/linux/mempolicy.h b/include/linux/mempolicy.h index 3a39570b81b8..085c903fe0f1 100644 --- a/include/linux/mempolicy.h +++ b/include/linux/mempolicy.h @@ -59,6 +59,7 @@ enum { #include #include #include +#include struct mm_struct; @@ -220,6 +221,24 @@ extern int mpol_parse_str(char *str, struct mempolicy **mpol, int no_context); extern int mpol_to_str(char *buffer, int maxlen, struct mempolicy *pol, int no_context); #endif + +/* Check if a vma is migratable */ +static inline int vma_migratable(struct vm_area_struct *vma) +{ + if (vma->vm_flags & (VM_IO|VM_HUGETLB|VM_PFNMAP|VM_RESERVED)) + return 0; + /* + * Migration allocates pages in the highest zone. If we cannot + * do so then migration (at least from node to node) is not + * possible. + */ + if (vma->vm_file && + gfp_zone(mapping_gfp_mask(vma->vm_file->f_mapping)) + < policy_zone) + return 0; + return 1; +} + #else struct mempolicy {}; diff --git a/include/linux/migrate.h b/include/linux/migrate.h index e10a90a93b5d..03aea612d284 100644 --- a/include/linux/migrate.h +++ b/include/linux/migrate.h @@ -3,28 +3,10 @@ #include #include -#include typedef struct page *new_page_t(struct page *, unsigned long private, int **); #ifdef CONFIG_MIGRATION -/* Check if a vma is migratable */ -static inline int vma_migratable(struct vm_area_struct *vma) -{ - if (vma->vm_flags & (VM_IO|VM_HUGETLB|VM_PFNMAP|VM_RESERVED)) - return 0; - /* - * Migration allocates pages in the highest zone. If we cannot - * do so then migration (at least from node to node) is not - * possible. - */ - if (vma->vm_file && - gfp_zone(mapping_gfp_mask(vma->vm_file->f_mapping)) - < policy_zone) - return 0; - return 1; -} - extern int isolate_lru_page(struct page *p, struct list_head *pagelist); extern int putback_lru_pages(struct list_head *l); extern int migrate_page(struct address_space *, @@ -39,9 +21,6 @@ extern int migrate_vmas(struct mm_struct *mm, const nodemask_t *from, const nodemask_t *to, unsigned long flags); #else -static inline int vma_migratable(struct vm_area_struct *vma) - { return 0; } - static inline int isolate_lru_page(struct page *p, struct list_head *list) { return -ENOSYS; } static inline int putback_lru_pages(struct list_head *l) { return 0; } diff --git a/mm/Kconfig b/mm/Kconfig index c4de85285bb4..aa799007a11b 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -174,7 +174,7 @@ config SPLIT_PTLOCK_CPUS config MIGRATION bool "Page migration" def_bool y - depends on NUMA + depends on NUMA || ARCH_ENABLE_MEMORY_HOTREMOVE help Allows the migration of the physical location of pages of processes while the virtual addresses are not changed. This is useful for diff --git a/mm/migrate.c b/mm/migrate.c index e7d13a708da0..376cceba82f9 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -1071,7 +1071,6 @@ asmlinkage long sys_move_pages(pid_t pid, unsigned long nr_pages, mmput(mm); return err; } -#endif /* * Call migration functions in the vma_ops that may prepare @@ -1093,3 +1092,4 @@ int migrate_vmas(struct mm_struct *mm, const nodemask_t *to, } return err; } +#endif From 78ecba081224a2db5876b6b81cfed0b78f58adc7 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 23 Jul 2008 21:28:23 -0700 Subject: [PATCH 215/857] mm: fix ever-decreasing swap priority Vegard Nossum has noticed the ever-decreasing negative priority in a swapon /swapoff loop, which eventually would misprioritize when int wraps positive. Not worth spending much code on, but probably better fixed. It's easy to handle the swapping on and off of just one area, but there's not much point if a pair or more still misbehave. To handle the general case, swapoff should compact negative priorities, keeping them always from -1 to -MAX_SWAPFILES. That's a change, but should cause no regression, since these negative (unspecified) priorities are disjoint from the the positive specified priorities 0 to 32767. One small functional difference, which seems appropriate: when swapoff fails to free all swap from a negative priority area, that area is now reinserted at lowest priority, rather than at its original priority. In moving down swapon's setting of priority, I notice that an area is visible to /proc/swaps when it has swap_map set, yet that was being set before all the visible fields were properly filled in: corrected. Signed-off-by: Hugh Dickins Reviewed-by: KOSAKI Motohiro Reported-by: Vegard Nossum Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/swapfile.c | 49 +++++++++++++++++++++++++------------------------ 1 file changed, 25 insertions(+), 24 deletions(-) diff --git a/mm/swapfile.c b/mm/swapfile.c index bd1bb5920306..2f33edb8bee9 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -37,6 +37,7 @@ DEFINE_SPINLOCK(swap_lock); unsigned int nr_swapfiles; long total_swap_pages; static int swap_overflow; +static int least_priority; static const char Bad_file[] = "Bad swap file entry "; static const char Unused_file[] = "Unused swap file entry "; @@ -1260,6 +1261,11 @@ asmlinkage long sys_swapoff(const char __user * specialfile) /* just pick something that's safe... */ swap_list.next = swap_list.head; } + if (p->prio < 0) { + for (i = p->next; i >= 0; i = swap_info[i].next) + swap_info[i].prio = p->prio--; + least_priority++; + } nr_swap_pages -= p->pages; total_swap_pages -= p->pages; p->flags &= ~SWP_WRITEOK; @@ -1272,9 +1278,14 @@ asmlinkage long sys_swapoff(const char __user * specialfile) if (err) { /* re-insert swap space back into swap_list */ spin_lock(&swap_lock); - for (prev = -1, i = swap_list.head; i >= 0; prev = i, i = swap_info[i].next) + if (p->prio < 0) + p->prio = --least_priority; + prev = -1; + for (i = swap_list.head; i >= 0; i = swap_info[i].next) { if (p->prio >= swap_info[i].prio) break; + prev = i; + } p->next = i; if (prev < 0) swap_list.head = swap_list.next = p - swap_info; @@ -1447,7 +1458,6 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags) unsigned int type; int i, prev; int error; - static int least_priority; union swap_header *swap_header = NULL; int swap_header_version; unsigned int nr_good_pages = 0; @@ -1455,7 +1465,7 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags) sector_t span; unsigned long maxpages = 1; int swapfilesize; - unsigned short *swap_map; + unsigned short *swap_map = NULL; struct page *page = NULL; struct inode *inode = NULL; int did_down = 0; @@ -1474,22 +1484,10 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags) } if (type >= nr_swapfiles) nr_swapfiles = type+1; + memset(p, 0, sizeof(*p)); INIT_LIST_HEAD(&p->extent_list); p->flags = SWP_USED; - p->swap_file = NULL; - p->old_block_size = 0; - p->swap_map = NULL; - p->lowest_bit = 0; - p->highest_bit = 0; - p->cluster_nr = 0; - p->inuse_pages = 0; p->next = -1; - if (swap_flags & SWAP_FLAG_PREFER) { - p->prio = - (swap_flags & SWAP_FLAG_PRIO_MASK)>>SWAP_FLAG_PRIO_SHIFT; - } else { - p->prio = --least_priority; - } spin_unlock(&swap_lock); name = getname(specialfile); error = PTR_ERR(name); @@ -1632,19 +1630,20 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags) goto bad_swap; /* OK, set up the swap map and apply the bad block list */ - if (!(p->swap_map = vmalloc(maxpages * sizeof(short)))) { + swap_map = vmalloc(maxpages * sizeof(short)); + if (!swap_map) { error = -ENOMEM; goto bad_swap; } error = 0; - memset(p->swap_map, 0, maxpages * sizeof(short)); + memset(swap_map, 0, maxpages * sizeof(short)); for (i = 0; i < swap_header->info.nr_badpages; i++) { int page_nr = swap_header->info.badpages[i]; if (page_nr <= 0 || page_nr >= swap_header->info.last_page) error = -EINVAL; else - p->swap_map[page_nr] = SWAP_MAP_BAD; + swap_map[page_nr] = SWAP_MAP_BAD; } nr_good_pages = swap_header->info.last_page - swap_header->info.nr_badpages - @@ -1654,7 +1653,7 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags) } if (nr_good_pages) { - p->swap_map[0] = SWAP_MAP_BAD; + swap_map[0] = SWAP_MAP_BAD; p->max = maxpages; p->pages = nr_good_pages; nr_extents = setup_swap_extents(p, &span); @@ -1672,6 +1671,12 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags) mutex_lock(&swapon_mutex); spin_lock(&swap_lock); + if (swap_flags & SWAP_FLAG_PREFER) + p->prio = + (swap_flags & SWAP_FLAG_PRIO_MASK) >> SWAP_FLAG_PRIO_SHIFT; + else + p->prio = --least_priority; + p->swap_map = swap_map; p->flags = SWP_ACTIVE; nr_swap_pages += nr_good_pages; total_swap_pages += nr_good_pages; @@ -1707,12 +1712,8 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags) destroy_swap_extents(p); bad_swap_2: spin_lock(&swap_lock); - swap_map = p->swap_map; p->swap_file = NULL; - p->swap_map = NULL; p->flags = 0; - if (!(swap_flags & SWAP_FLAG_PREFER)) - ++least_priority; spin_unlock(&swap_lock); vfree(swap_map); if (swap_file) From 5459c164f0591ee75ed0203bb8f3817f25948e2f Mon Sep 17 00:00:00 2001 From: "Andrew G. Morgan" Date: Wed, 23 Jul 2008 21:28:24 -0700 Subject: [PATCH 216/857] security: protect legacy applications from executing with insufficient privilege When cap_bset suppresses some of the forced (fP) capabilities of a file, it is generally only safe to execute the program if it understands how to recognize it doesn't have enough privilege to work correctly. For legacy applications (fE!=0), which have no non-destructive way to determine that they are missing privilege, we fail to execute (EPERM) any executable that requires fP capabilities, but would otherwise get pP' < fP. This is a fail-safe permission check. For some discussion of why it is problematic for (legacy) privileged applications to run with less than the set of capabilities requested for them, see: http://userweb.kernel.org/~morgan/sendmail-capabilities-war-story.html With this iteration of this support, we do not include setuid-0 based privilege protection from the bounding set. That is, the admin can still (ab)use the bounding set to suppress the privileges of a setuid-0 program. [akpm@linux-foundation.org: coding-style fixes] [akpm@linux-foundation.org: cleanup] Signed-off-by: Andrew G. Morgan Acked-by: Serge Hallyn Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/binfmts.h | 2 +- security/commoncap.c | 108 ++++++++++++++++++++++------------------ 2 files changed, 60 insertions(+), 50 deletions(-) diff --git a/include/linux/binfmts.h b/include/linux/binfmts.h index ee0ed48e8348..826f62350805 100644 --- a/include/linux/binfmts.h +++ b/include/linux/binfmts.h @@ -38,7 +38,7 @@ struct linux_binprm{ misc_bang:1; struct file * file; int e_uid, e_gid; - kernel_cap_t cap_inheritable, cap_permitted; + kernel_cap_t cap_post_exec_permitted; bool cap_effective; void *security; int argc, envc; diff --git a/security/commoncap.c b/security/commoncap.c index 0b6537a3672d..4afbece37a08 100644 --- a/security/commoncap.c +++ b/security/commoncap.c @@ -162,8 +162,7 @@ void cap_capset_set (struct task_struct *target, kernel_cap_t *effective, static inline void bprm_clear_caps(struct linux_binprm *bprm) { - cap_clear(bprm->cap_inheritable); - cap_clear(bprm->cap_permitted); + cap_clear(bprm->cap_post_exec_permitted); bprm->cap_effective = false; } @@ -198,6 +197,7 @@ static inline int cap_from_disk(struct vfs_cap_data *caps, { __u32 magic_etc; unsigned tocopy, i; + int ret; if (size < sizeof(magic_etc)) return -EINVAL; @@ -225,19 +225,40 @@ static inline int cap_from_disk(struct vfs_cap_data *caps, bprm->cap_effective = false; } - for (i = 0; i < tocopy; ++i) { - bprm->cap_permitted.cap[i] = - le32_to_cpu(caps->data[i].permitted); - bprm->cap_inheritable.cap[i] = - le32_to_cpu(caps->data[i].inheritable); - } - while (i < VFS_CAP_U32) { - bprm->cap_permitted.cap[i] = 0; - bprm->cap_inheritable.cap[i] = 0; - i++; + ret = 0; + + CAP_FOR_EACH_U32(i) { + __u32 value_cpu; + + if (i >= tocopy) { + /* + * Legacy capability sets have no upper bits + */ + bprm->cap_post_exec_permitted.cap[i] = 0; + continue; + } + /* + * pP' = (X & fP) | (pI & fI) + */ + value_cpu = le32_to_cpu(caps->data[i].permitted); + bprm->cap_post_exec_permitted.cap[i] = + (current->cap_bset.cap[i] & value_cpu) | + (current->cap_inheritable.cap[i] & + le32_to_cpu(caps->data[i].inheritable)); + if (value_cpu & ~bprm->cap_post_exec_permitted.cap[i]) { + /* + * insufficient to execute correctly + */ + ret = -EPERM; + } } - return 0; + /* + * For legacy apps, with no internal support for recognizing they + * do not have enough capabilities, we return an error if they are + * missing some "forced" (aka file-permitted) capabilities. + */ + return bprm->cap_effective ? ret : 0; } /* Locate any VFS capabilities: */ @@ -269,9 +290,9 @@ static int get_file_caps(struct linux_binprm *bprm) goto out; rc = cap_from_disk(&vcaps, bprm, rc); - if (rc) + if (rc == -EINVAL) printk(KERN_NOTICE "%s: cap_from_disk returned %d for %s\n", - __func__, rc, bprm->filename); + __func__, rc, bprm->filename); out: dput(dentry); @@ -304,25 +325,24 @@ int cap_bprm_set_security (struct linux_binprm *bprm) int ret; ret = get_file_caps(bprm); - if (ret) - printk(KERN_NOTICE "%s: get_file_caps returned %d for %s\n", - __func__, ret, bprm->filename); - /* To support inheritance of root-permissions and suid-root - * executables under compatibility mode, we raise all three - * capability sets for the file. - * - * If only the real uid is 0, we only raise the inheritable - * and permitted sets of the executable file. - */ - - if (!issecure (SECURE_NOROOT)) { + if (!issecure(SECURE_NOROOT)) { + /* + * To support inheritance of root-permissions and suid-root + * executables under compatibility mode, we override the + * capability sets for the file. + * + * If only the real uid is 0, we do not set the effective + * bit. + */ if (bprm->e_uid == 0 || current->uid == 0) { - cap_set_full (bprm->cap_inheritable); - cap_set_full (bprm->cap_permitted); + /* pP' = (cap_bset & ~0) | (pI & ~0) */ + bprm->cap_post_exec_permitted = cap_combine( + current->cap_bset, current->cap_inheritable + ); + bprm->cap_effective = (bprm->e_uid == 0); + ret = 0; } - if (bprm->e_uid == 0) - bprm->cap_effective = true; } return ret; @@ -330,17 +350,9 @@ int cap_bprm_set_security (struct linux_binprm *bprm) void cap_bprm_apply_creds (struct linux_binprm *bprm, int unsafe) { - /* Derived from fs/exec.c:compute_creds. */ - kernel_cap_t new_permitted, working; - - new_permitted = cap_intersect(bprm->cap_permitted, - current->cap_bset); - working = cap_intersect(bprm->cap_inheritable, - current->cap_inheritable); - new_permitted = cap_combine(new_permitted, working); - if (bprm->e_uid != current->uid || bprm->e_gid != current->gid || - !cap_issubset (new_permitted, current->cap_permitted)) { + !cap_issubset(bprm->cap_post_exec_permitted, + current->cap_permitted)) { set_dumpable(current->mm, suid_dumpable); current->pdeath_signal = 0; @@ -350,9 +362,9 @@ void cap_bprm_apply_creds (struct linux_binprm *bprm, int unsafe) bprm->e_gid = current->gid; } if (cap_limit_ptraced_target()) { - new_permitted = - cap_intersect(new_permitted, - current->cap_permitted); + bprm->cap_post_exec_permitted = cap_intersect( + bprm->cap_post_exec_permitted, + current->cap_permitted); } } } @@ -364,9 +376,9 @@ void cap_bprm_apply_creds (struct linux_binprm *bprm, int unsafe) * in the init_task struct. Thus we skip the usual * capability rules */ if (!is_global_init(current)) { - current->cap_permitted = new_permitted; + current->cap_permitted = bprm->cap_post_exec_permitted; if (bprm->cap_effective) - current->cap_effective = new_permitted; + current->cap_effective = bprm->cap_post_exec_permitted; else cap_clear(current->cap_effective); } @@ -381,9 +393,7 @@ int cap_bprm_secureexec (struct linux_binprm *bprm) if (current->uid != 0) { if (bprm->cap_effective) return 1; - if (!cap_isclear(bprm->cap_permitted)) - return 1; - if (!cap_isclear(bprm->cap_inheritable)) + if (!cap_isclear(bprm->cap_post_exec_permitted)) return 1; } From ab763c7112ce0e2559c73f921617c81dc7287ca6 Mon Sep 17 00:00:00 2001 From: "Andrew G. Morgan" Date: Wed, 23 Jul 2008 21:28:25 -0700 Subject: [PATCH 217/857] security: filesystem capabilities refactor kernel code To date, we've tried hard to confine filesystem support for capabilities to the security modules. This has left a lot of the code in kernel/capability.c in a state where it looks like it supports something that filesystem support for capabilities actually suppresses when the LSM security/commmoncap.c code runs. What is left is a lot of code that uses sub-optimal locking in the main kernel With this change we refactor the main kernel code and make it explicit which locks are needed and that the only remaining kernel races in this area are associated with non-filesystem capability code. Signed-off-by: Andrew G. Morgan Acked-by: Serge Hallyn Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/capability.c | 338 +++++++++++++++++++++++++++++--------------- 1 file changed, 221 insertions(+), 117 deletions(-) diff --git a/kernel/capability.c b/kernel/capability.c index 901e0fdc3fff..0101e847603e 100644 --- a/kernel/capability.c +++ b/kernel/capability.c @@ -115,11 +115,208 @@ static int cap_validate_magic(cap_user_header_t header, unsigned *tocopy) return 0; } +#ifndef CONFIG_SECURITY_FILE_CAPABILITIES + /* - * For sys_getproccap() and sys_setproccap(), any of the three - * capability set pointers may be NULL -- indicating that that set is - * uninteresting and/or not to be changed. + * Without filesystem capability support, we nominally support one process + * setting the capabilities of another */ +static inline int cap_get_target_pid(pid_t pid, kernel_cap_t *pEp, + kernel_cap_t *pIp, kernel_cap_t *pPp) +{ + struct task_struct *target; + int ret; + + spin_lock(&task_capability_lock); + read_lock(&tasklist_lock); + + if (pid && pid != task_pid_vnr(current)) { + target = find_task_by_vpid(pid); + if (!target) { + ret = -ESRCH; + goto out; + } + } else + target = current; + + ret = security_capget(target, pEp, pIp, pPp); + +out: + read_unlock(&tasklist_lock); + spin_unlock(&task_capability_lock); + + return ret; +} + +/* + * cap_set_pg - set capabilities for all processes in a given process + * group. We call this holding task_capability_lock and tasklist_lock. + */ +static inline int cap_set_pg(int pgrp_nr, kernel_cap_t *effective, + kernel_cap_t *inheritable, + kernel_cap_t *permitted) +{ + struct task_struct *g, *target; + int ret = -EPERM; + int found = 0; + struct pid *pgrp; + + spin_lock(&task_capability_lock); + read_lock(&tasklist_lock); + + pgrp = find_vpid(pgrp_nr); + do_each_pid_task(pgrp, PIDTYPE_PGID, g) { + target = g; + while_each_thread(g, target) { + if (!security_capset_check(target, effective, + inheritable, permitted)) { + security_capset_set(target, effective, + inheritable, permitted); + ret = 0; + } + found = 1; + } + } while_each_pid_task(pgrp, PIDTYPE_PGID, g); + + read_unlock(&tasklist_lock); + spin_unlock(&task_capability_lock); + + if (!found) + ret = 0; + return ret; +} + +/* + * cap_set_all - set capabilities for all processes other than init + * and self. We call this holding task_capability_lock and tasklist_lock. + */ +static inline int cap_set_all(kernel_cap_t *effective, + kernel_cap_t *inheritable, + kernel_cap_t *permitted) +{ + struct task_struct *g, *target; + int ret = -EPERM; + int found = 0; + + spin_lock(&task_capability_lock); + read_lock(&tasklist_lock); + + do_each_thread(g, target) { + if (target == current + || is_container_init(target->group_leader)) + continue; + found = 1; + if (security_capset_check(target, effective, inheritable, + permitted)) + continue; + ret = 0; + security_capset_set(target, effective, inheritable, permitted); + } while_each_thread(g, target); + + read_unlock(&tasklist_lock); + spin_unlock(&task_capability_lock); + + if (!found) + ret = 0; + + return ret; +} + +/* + * Given the target pid does not refer to the current process we + * need more elaborate support... (This support is not present when + * filesystem capabilities are configured.) + */ +static inline int do_sys_capset_other_tasks(pid_t pid, kernel_cap_t *effective, + kernel_cap_t *inheritable, + kernel_cap_t *permitted) +{ + struct task_struct *target; + int ret; + + if (!capable(CAP_SETPCAP)) + return -EPERM; + + if (pid == -1) /* all procs other than current and init */ + return cap_set_all(effective, inheritable, permitted); + + else if (pid < 0) /* all procs in process group */ + return cap_set_pg(-pid, effective, inheritable, permitted); + + /* target != current */ + spin_lock(&task_capability_lock); + read_lock(&tasklist_lock); + + target = find_task_by_vpid(pid); + if (!target) + ret = -ESRCH; + else { + ret = security_capset_check(target, effective, inheritable, + permitted); + + /* having verified that the proposed changes are legal, + we now put them into effect. */ + if (!ret) + security_capset_set(target, effective, inheritable, + permitted); + } + + read_unlock(&tasklist_lock); + spin_unlock(&task_capability_lock); + + return ret; +} + +#else /* ie., def CONFIG_SECURITY_FILE_CAPABILITIES */ + +/* + * If we have configured with filesystem capability support, then the + * only thing that can change the capabilities of the current process + * is the current process. As such, we can't be in this code at the + * same time as we are in the process of setting capabilities in this + * process. The net result is that we can limit our use of locks to + * when we are reading the caps of another process. + */ +static inline int cap_get_target_pid(pid_t pid, kernel_cap_t *pEp, + kernel_cap_t *pIp, kernel_cap_t *pPp) +{ + int ret; + + if (pid && (pid != task_pid_vnr(current))) { + struct task_struct *target; + + spin_lock(&task_capability_lock); + read_lock(&tasklist_lock); + + target = find_task_by_vpid(pid); + if (!target) + ret = -ESRCH; + else + ret = security_capget(target, pEp, pIp, pPp); + + read_unlock(&tasklist_lock); + spin_unlock(&task_capability_lock); + } else + ret = security_capget(current, pEp, pIp, pPp); + + return ret; +} + +/* + * With filesystem capability support configured, the kernel does not + * permit the changing of capabilities in one process by another + * process. (CAP_SETPCAP has much less broad semantics when configured + * this way.) + */ +static inline int do_sys_capset_other_tasks(pid_t pid, + kernel_cap_t *effective, + kernel_cap_t *inheritable, + kernel_cap_t *permitted) +{ + return -EPERM; +} + +#endif /* ie., ndef CONFIG_SECURITY_FILE_CAPABILITIES */ /* * Atomically modify the effective capabilities returning the original @@ -155,7 +352,6 @@ asmlinkage long sys_capget(cap_user_header_t header, cap_user_data_t dataptr) { int ret = 0; pid_t pid; - struct task_struct *target; unsigned tocopy; kernel_cap_t pE, pI, pP; @@ -169,23 +365,7 @@ asmlinkage long sys_capget(cap_user_header_t header, cap_user_data_t dataptr) if (pid < 0) return -EINVAL; - spin_lock(&task_capability_lock); - read_lock(&tasklist_lock); - - if (pid && pid != task_pid_vnr(current)) { - target = find_task_by_vpid(pid); - if (!target) { - ret = -ESRCH; - goto out; - } - } else - target = current; - - ret = security_capget(target, &pE, &pI, &pP); - -out: - read_unlock(&tasklist_lock); - spin_unlock(&task_capability_lock); + ret = cap_get_target_pid(pid, &pE, &pI, &pP); if (!ret) { struct __user_cap_data_struct kdata[_KERNEL_CAPABILITY_U32S]; @@ -216,7 +396,6 @@ asmlinkage long sys_capget(cap_user_header_t header, cap_user_data_t dataptr) * before modification is attempted and the application * fails. */ - if (copy_to_user(dataptr, kdata, tocopy * sizeof(struct __user_cap_data_struct))) { return -EFAULT; @@ -226,70 +405,8 @@ asmlinkage long sys_capget(cap_user_header_t header, cap_user_data_t dataptr) return ret; } -/* - * cap_set_pg - set capabilities for all processes in a given process - * group. We call this holding task_capability_lock and tasklist_lock. - */ -static inline int cap_set_pg(int pgrp_nr, kernel_cap_t *effective, - kernel_cap_t *inheritable, - kernel_cap_t *permitted) -{ - struct task_struct *g, *target; - int ret = -EPERM; - int found = 0; - struct pid *pgrp; - - pgrp = find_vpid(pgrp_nr); - do_each_pid_task(pgrp, PIDTYPE_PGID, g) { - target = g; - while_each_thread(g, target) { - if (!security_capset_check(target, effective, - inheritable, - permitted)) { - security_capset_set(target, effective, - inheritable, - permitted); - ret = 0; - } - found = 1; - } - } while_each_pid_task(pgrp, PIDTYPE_PGID, g); - - if (!found) - ret = 0; - return ret; -} - -/* - * cap_set_all - set capabilities for all processes other than init - * and self. We call this holding task_capability_lock and tasklist_lock. - */ -static inline int cap_set_all(kernel_cap_t *effective, - kernel_cap_t *inheritable, - kernel_cap_t *permitted) -{ - struct task_struct *g, *target; - int ret = -EPERM; - int found = 0; - - do_each_thread(g, target) { - if (target == current || is_container_init(target->group_leader)) - continue; - found = 1; - if (security_capset_check(target, effective, inheritable, - permitted)) - continue; - ret = 0; - security_capset_set(target, effective, inheritable, permitted); - } while_each_thread(g, target); - - if (!found) - ret = 0; - return ret; -} - /** - * sys_capset - set capabilities for a process or a group of processes + * sys_capset - set capabilities for a process or (*) a group of processes * @header: pointer to struct that contains capability version and * target pid data * @data: pointer to struct that contains the effective, permitted, @@ -313,7 +430,6 @@ asmlinkage long sys_capset(cap_user_header_t header, const cap_user_data_t data) struct __user_cap_data_struct kdata[_KERNEL_CAPABILITY_U32S]; unsigned i, tocopy; kernel_cap_t inheritable, permitted, effective; - struct task_struct *target; int ret; pid_t pid; @@ -324,9 +440,6 @@ asmlinkage long sys_capset(cap_user_header_t header, const cap_user_data_t data) if (get_user(pid, &header->pid)) return -EFAULT; - if (pid && pid != task_pid_vnr(current) && !capable(CAP_SETPCAP)) - return -EPERM; - if (copy_from_user(&kdata, data, tocopy * sizeof(struct __user_cap_data_struct))) { return -EFAULT; @@ -344,40 +457,31 @@ asmlinkage long sys_capset(cap_user_header_t header, const cap_user_data_t data) i++; } - spin_lock(&task_capability_lock); - read_lock(&tasklist_lock); + if (pid && (pid != task_pid_vnr(current))) + ret = do_sys_capset_other_tasks(pid, &effective, &inheritable, + &permitted); + else { + /* + * This lock is required even when filesystem + * capability support is configured - it protects the + * sys_capget() call from returning incorrect data in + * the case that the targeted process is not the + * current one. + */ + spin_lock(&task_capability_lock); - if (pid > 0 && pid != task_pid_vnr(current)) { - target = find_task_by_vpid(pid); - if (!target) { - ret = -ESRCH; - goto out; - } - } else - target = current; - - ret = 0; - - /* having verified that the proposed changes are legal, - we now put them into effect. */ - if (pid < 0) { - if (pid == -1) /* all procs other than current and init */ - ret = cap_set_all(&effective, &inheritable, &permitted); - - else /* all procs in process group */ - ret = cap_set_pg(-pid, &effective, &inheritable, - &permitted); - } else { - ret = security_capset_check(target, &effective, &inheritable, + ret = security_capset_check(current, &effective, &inheritable, &permitted); + /* + * Having verified that the proposed changes are + * legal, we now put them into effect. + */ if (!ret) - security_capset_set(target, &effective, &inheritable, + security_capset_set(current, &effective, &inheritable, &permitted); + spin_unlock(&task_capability_lock); } -out: - read_unlock(&tasklist_lock); - spin_unlock(&task_capability_lock); return ret; } From 84aaa7ab4c40b66d6dd9aa393901551ad50ec640 Mon Sep 17 00:00:00 2001 From: "Andrew G. Morgan" Date: Wed, 23 Jul 2008 21:28:25 -0700 Subject: [PATCH 218/857] security: filesystem capabilities no longer experimental Filesystem capabilities have come of age. Remove the experimental tag for configuring filesystem capabilities. Signed-off-by: Andrew G. Morgan Acked-by: Serge Hallyn Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- security/Kconfig | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/security/Kconfig b/security/Kconfig index 62ed4717d334..559293922a47 100644 --- a/security/Kconfig +++ b/security/Kconfig @@ -74,8 +74,7 @@ config SECURITY_NETWORK_XFRM If you are unsure how to answer this question, answer N. config SECURITY_FILE_CAPABILITIES - bool "File POSIX Capabilities (EXPERIMENTAL)" - depends on EXPERIMENTAL + bool "File POSIX Capabilities" default n help This enables filesystem capabilities, allowing you to give From 9b3e43a747c74029b0acf6acf4666601f132f471 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 23 Jul 2008 21:28:26 -0700 Subject: [PATCH 219/857] security: remove unused forwards Why would linux/security.h need forward declarations for nfsctl_arg and swap_info_struct? It's hard to imagine: remove them. Signed-off-by: Hugh Dickins Acked-by: James Morris Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/security.h | 2 -- 1 file changed, 2 deletions(-) diff --git a/include/linux/security.h b/include/linux/security.h index 31c8851ec5d0..f0e9adb22ac2 100644 --- a/include/linux/security.h +++ b/include/linux/security.h @@ -102,9 +102,7 @@ extern unsigned long mmap_min_addr; #define LSM_SETID_FS 8 /* forward declares to avoid warnings */ -struct nfsctl_arg; struct sched_param; -struct swap_info_struct; struct request_sock; /* bprm_apply_creds unsafe reasons */ From 5002779d37b261271da9883e06c14b097d4781c4 Mon Sep 17 00:00:00 2001 From: Tilman Schmidt Date: Wed, 23 Jul 2008 21:28:27 -0700 Subject: [PATCH 220/857] gigaset: use dev_ macros for messages The info() / warn() / err() macros from usb.h for generating kernel messages are considered inferior to dev_info() / dev_warn() / dev_err() from device.h. Replace them where possible. Also correct the severity level and improve the text of one message. Signed-off-by: Tilman Schmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/isdn/gigaset/asyncdata.c | 3 +- drivers/isdn/gigaset/common.c | 2 +- drivers/isdn/gigaset/gigaset.h | 3 -- drivers/isdn/gigaset/i4l.c | 56 ++++++++++++++++++------------ drivers/isdn/gigaset/interface.c | 25 ++++++------- drivers/isdn/gigaset/usb-gigaset.c | 7 ++-- 6 files changed, 53 insertions(+), 43 deletions(-) diff --git a/drivers/isdn/gigaset/asyncdata.c b/drivers/isdn/gigaset/asyncdata.c index 091deb9d1c47..c2bd97d29273 100644 --- a/drivers/isdn/gigaset/asyncdata.c +++ b/drivers/isdn/gigaset/asyncdata.c @@ -575,7 +575,8 @@ int gigaset_m10x_send_skb(struct bc_state *bcs, struct sk_buff *skb) else skb = iraw_encode(skb, HW_HDR_LEN, 0); if (!skb) { - err("unable to allocate memory for encoding!\n"); + dev_err(bcs->cs->dev, + "unable to allocate memory for encoding!\n"); return -ENOMEM; } diff --git a/drivers/isdn/gigaset/common.c b/drivers/isdn/gigaset/common.c index 827c32c16795..9d3ce7718e58 100644 --- a/drivers/isdn/gigaset/common.c +++ b/drivers/isdn/gigaset/common.c @@ -287,7 +287,7 @@ struct event_t *gigaset_add_event(struct cardstate *cs, tail = cs->ev_tail; next = (tail + 1) % MAX_EVENTS; if (unlikely(next == cs->ev_head)) - err("event queue full"); + dev_err(cs->dev, "event queue full\n"); else { event = cs->events + tail; event->type = type; diff --git a/drivers/isdn/gigaset/gigaset.h b/drivers/isdn/gigaset/gigaset.h index f365993161fc..003752954993 100644 --- a/drivers/isdn/gigaset/gigaset.h +++ b/drivers/isdn/gigaset/gigaset.h @@ -106,7 +106,6 @@ enum debuglevel { #undef err #undef info #undef warn -#undef notice #define err(format, arg...) printk(KERN_ERR KBUILD_MODNAME ": " \ format "\n" , ## arg) @@ -114,8 +113,6 @@ enum debuglevel { format "\n" , ## arg) #define warn(format, arg...) printk(KERN_WARNING KBUILD_MODNAME ": " \ format "\n" , ## arg) -#define notice(format, arg...) printk(KERN_NOTICE KBUILD_MODNAME ": " \ - format "\n" , ## arg) #ifdef CONFIG_GIGASET_DEBUG diff --git a/drivers/isdn/gigaset/i4l.c b/drivers/isdn/gigaset/i4l.c index 9e089f06a942..3c127a8cbaf2 100644 --- a/drivers/isdn/gigaset/i4l.c +++ b/drivers/isdn/gigaset/i4l.c @@ -46,7 +46,8 @@ static int writebuf_from_LL(int driverID, int channel, int ack, return -ENODEV; } if (channel < 0 || channel >= cs->channels) { - err("%s: invalid channel ID (%d)", __func__, channel); + dev_err(cs->dev, "%s: invalid channel ID (%d)\n", + __func__, channel); return -ENODEV; } bcs = &cs->bcs[channel]; @@ -58,11 +59,13 @@ static int writebuf_from_LL(int driverID, int channel, int ack, if (!len) { if (ack) - notice("%s: not ACKing empty packet", __func__); + dev_notice(cs->dev, "%s: not ACKing empty packet\n", + __func__); return 0; } if (len > MAX_BUF_SIZE) { - err("%s: packet too large (%d bytes)", __func__, len); + dev_err(cs->dev, "%s: packet too large (%d bytes)\n", + __func__, len); return -EINVAL; } @@ -116,8 +119,7 @@ static int command_from_LL(isdn_ctrl *cntrl) gigaset_debugdrivers(); if (!cs) { - warn("LL tried to access unknown device with nr. %d", - cntrl->driver); + err("%s: invalid driver ID (%d)", __func__, cntrl->driver); return -ENODEV; } @@ -126,7 +128,7 @@ static int command_from_LL(isdn_ctrl *cntrl) gig_dbg(DEBUG_ANY, "ISDN_CMD_IOCTL (driver: %d, arg: %ld)", cntrl->driver, cntrl->arg); - warn("ISDN_CMD_IOCTL is not supported."); + dev_warn(cs->dev, "ISDN_CMD_IOCTL not supported\n"); return -EINVAL; case ISDN_CMD_DIAL: @@ -138,22 +140,23 @@ static int command_from_LL(isdn_ctrl *cntrl) cntrl->parm.setup.si1, cntrl->parm.setup.si2); if (cntrl->arg >= cs->channels) { - err("ISDN_CMD_DIAL: invalid channel (%d)", - (int) cntrl->arg); + dev_err(cs->dev, + "ISDN_CMD_DIAL: invalid channel (%d)\n", + (int) cntrl->arg); return -EINVAL; } bcs = cs->bcs + cntrl->arg; if (!gigaset_get_channel(bcs)) { - err("ISDN_CMD_DIAL: channel not free"); + dev_err(cs->dev, "ISDN_CMD_DIAL: channel not free\n"); return -EBUSY; } sp = kmalloc(sizeof *sp, GFP_ATOMIC); if (!sp) { gigaset_free_channel(bcs); - err("ISDN_CMD_DIAL: out of memory"); + dev_err(cs->dev, "ISDN_CMD_DIAL: out of memory\n"); return -ENOMEM; } *sp = cntrl->parm.setup; @@ -173,8 +176,9 @@ static int command_from_LL(isdn_ctrl *cntrl) gig_dbg(DEBUG_ANY, "ISDN_CMD_ACCEPTD"); if (cntrl->arg >= cs->channels) { - err("ISDN_CMD_ACCEPTD: invalid channel (%d)", - (int) cntrl->arg); + dev_err(cs->dev, + "ISDN_CMD_ACCEPTD: invalid channel (%d)\n", + (int) cntrl->arg); return -EINVAL; } @@ -196,8 +200,9 @@ static int command_from_LL(isdn_ctrl *cntrl) (int) cntrl->arg); if (cntrl->arg >= cs->channels) { - err("ISDN_CMD_HANGUP: invalid channel (%u)", - (unsigned) cntrl->arg); + dev_err(cs->dev, + "ISDN_CMD_HANGUP: invalid channel (%d)\n", + (int) cntrl->arg); return -EINVAL; } @@ -224,8 +229,9 @@ static int command_from_LL(isdn_ctrl *cntrl) cntrl->arg & 0xff, (cntrl->arg >> 8)); if ((cntrl->arg & 0xff) >= cs->channels) { - err("ISDN_CMD_SETL2: invalid channel (%u)", - (unsigned) cntrl->arg & 0xff); + dev_err(cs->dev, + "ISDN_CMD_SETL2: invalid channel (%d)\n", + (int) cntrl->arg & 0xff); return -EINVAL; } @@ -244,14 +250,16 @@ static int command_from_LL(isdn_ctrl *cntrl) cntrl->arg & 0xff, (cntrl->arg >> 8)); if ((cntrl->arg & 0xff) >= cs->channels) { - err("ISDN_CMD_SETL3: invalid channel (%u)", - (unsigned) cntrl->arg & 0xff); + dev_err(cs->dev, + "ISDN_CMD_SETL3: invalid channel (%d)\n", + (int) cntrl->arg & 0xff); return -EINVAL; } if (cntrl->arg >> 8 != ISDN_PROTO_L3_TRANS) { - err("ISDN_CMD_SETL3: invalid protocol %lu", - cntrl->arg >> 8); + dev_err(cs->dev, + "ISDN_CMD_SETL3: invalid protocol %lu\n", + cntrl->arg >> 8); return -EINVAL; } @@ -262,8 +270,9 @@ static int command_from_LL(isdn_ctrl *cntrl) case ISDN_CMD_ALERT: gig_dbg(DEBUG_ANY, "ISDN_CMD_ALERT"); //FIXME if (cntrl->arg >= cs->channels) { - err("ISDN_CMD_ALERT: invalid channel (%d)", - (int) cntrl->arg); + dev_err(cs->dev, + "ISDN_CMD_ALERT: invalid channel (%d)\n", + (int) cntrl->arg); return -EINVAL; } //bcs = cs->bcs + cntrl->arg; @@ -295,7 +304,8 @@ static int command_from_LL(isdn_ctrl *cntrl) gig_dbg(DEBUG_ANY, "ISDN_CMD_GETSIL"); break; default: - err("unknown command %d from LL", cntrl->command); + dev_err(cs->dev, "unknown command %d from LL\n", + cntrl->command); return -EINVAL; } diff --git a/drivers/isdn/gigaset/interface.c b/drivers/isdn/gigaset/interface.c index af195b07c191..521951a898ec 100644 --- a/drivers/isdn/gigaset/interface.c +++ b/drivers/isdn/gigaset/interface.c @@ -197,7 +197,7 @@ static void if_close(struct tty_struct *tty, struct file *filp) mutex_lock(&cs->mutex); if (!cs->open_count) - warn("%s: device not opened", __func__); + dev_warn(cs->dev, "%s: device not opened\n", __func__); else { if (!--cs->open_count) { spin_lock_irqsave(&cs->lock, flags); @@ -232,7 +232,7 @@ static int if_ioctl(struct tty_struct *tty, struct file *file, return -ERESTARTSYS; // FIXME -EINTR? if (!cs->open_count) - warn("%s: device not opened", __func__); + dev_warn(cs->dev, "%s: device not opened\n", __func__); else { retval = 0; switch (cmd) { @@ -364,9 +364,9 @@ static int if_write(struct tty_struct *tty, const unsigned char *buf, int count) return -ERESTARTSYS; // FIXME -EINTR? if (!cs->open_count) - warn("%s: device not opened", __func__); + dev_warn(cs->dev, "%s: device not opened\n", __func__); else if (cs->mstate != MS_LOCKED) { - warn("can't write to unlocked device"); + dev_warn(cs->dev, "can't write to unlocked device\n"); retval = -EBUSY; } else if (!cs->connected) { gig_dbg(DEBUG_ANY, "can't write to unplugged device"); @@ -398,9 +398,9 @@ static int if_write_room(struct tty_struct *tty) return -ERESTARTSYS; // FIXME -EINTR? if (!cs->open_count) - warn("%s: device not opened", __func__); + dev_warn(cs->dev, "%s: device not opened\n", __func__); else if (cs->mstate != MS_LOCKED) { - warn("can't write to unlocked device"); + dev_warn(cs->dev, "can't write to unlocked device\n"); retval = -EBUSY; } else if (!cs->connected) { gig_dbg(DEBUG_ANY, "can't write to unplugged device"); @@ -430,9 +430,9 @@ static int if_chars_in_buffer(struct tty_struct *tty) return -ERESTARTSYS; // FIXME -EINTR? if (!cs->open_count) - warn("%s: device not opened", __func__); + dev_warn(cs->dev, "%s: device not opened\n", __func__); else if (cs->mstate != MS_LOCKED) { - warn("can't write to unlocked device"); + dev_warn(cs->dev, "can't write to unlocked device\n"); retval = -EBUSY; } else if (!cs->connected) { gig_dbg(DEBUG_ANY, "can't write to unplugged device"); @@ -460,7 +460,7 @@ static void if_throttle(struct tty_struct *tty) mutex_lock(&cs->mutex); if (!cs->open_count) - warn("%s: device not opened", __func__); + dev_warn(cs->dev, "%s: device not opened\n", __func__); else { //FIXME } @@ -483,7 +483,7 @@ static void if_unthrottle(struct tty_struct *tty) mutex_lock(&cs->mutex); if (!cs->open_count) - warn("%s: device not opened", __func__); + dev_warn(cs->dev, "%s: device not opened\n", __func__); else { //FIXME } @@ -510,7 +510,7 @@ static void if_set_termios(struct tty_struct *tty, struct ktermios *old) mutex_lock(&cs->mutex); if (!cs->open_count) { - warn("%s: device not opened", __func__); + dev_warn(cs->dev, "%s: device not opened\n", __func__); goto out; } @@ -623,7 +623,8 @@ void gigaset_if_init(struct cardstate *cs) if (!IS_ERR(cs->tty_dev)) dev_set_drvdata(cs->tty_dev, cs); else { - warn("could not register device to the tty subsystem"); + dev_warn(cs->dev, + "could not register device to the tty subsystem\n"); cs->tty_dev = NULL; } mutex_unlock(&cs->mutex); diff --git a/drivers/isdn/gigaset/usb-gigaset.c b/drivers/isdn/gigaset/usb-gigaset.c index 77d20ab0cd4d..4661830a49db 100644 --- a/drivers/isdn/gigaset/usb-gigaset.c +++ b/drivers/isdn/gigaset/usb-gigaset.c @@ -498,8 +498,9 @@ static int send_cb(struct cardstate *cs, struct cmdbuf_t *cb) if (status) { ucs->busy = 0; - err("could not submit urb (error %d)\n", - -status); + dev_err(cs->dev, + "could not submit urb (error %d)\n", + -status); cb->len = 0; /* skip urb => remove cb+wakeup in next loop cycle */ } @@ -670,7 +671,7 @@ static int write_modem(struct cardstate *cs) spin_unlock_irqrestore(&cs->lock, flags); if (ret) { - err("could not submit urb (error %d)\n", -ret); + dev_err(cs->dev, "could not submit urb (error %d)\n", -ret); ucs->busy = 0; } From 5f09c4c797d00bef5700e1ca085b4efcedaf34b8 Mon Sep 17 00:00:00 2001 From: Tilman Schmidt Date: Wed, 23 Jul 2008 21:28:27 -0700 Subject: [PATCH 221/857] gigaset: gigaset_isowbuf_getbytes() may return signed unnoticed ifd->offset is unsigned. gigaset_isowbuf_getbytes() may return signed unnoticed. Revised version of patch originally submitted by Roel Kluin <12o3l@tiscali.nl>. Signed-off-by: Tilman Schmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/isdn/gigaset/bas-gigaset.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/isdn/gigaset/bas-gigaset.c b/drivers/isdn/gigaset/bas-gigaset.c index 5255b5e20e13..3f11910c7ccd 100644 --- a/drivers/isdn/gigaset/bas-gigaset.c +++ b/drivers/isdn/gigaset/bas-gigaset.c @@ -1050,10 +1050,9 @@ static int submit_iso_write_urb(struct isow_urbctx_t *ucx) } /* retrieve block of data to send */ - ifd->offset = gigaset_isowbuf_getbytes(ubc->isooutbuf, - ifd->length); - if (ifd->offset < 0) { - if (ifd->offset == -EBUSY) { + rc = gigaset_isowbuf_getbytes(ubc->isooutbuf, ifd->length); + if (rc < 0) { + if (rc == -EBUSY) { gig_dbg(DEBUG_ISO, "%s: buffer busy at frame %d", __func__, nframe); @@ -1062,11 +1061,12 @@ static int submit_iso_write_urb(struct isow_urbctx_t *ucx) } else { dev_err(ucx->bcs->cs->dev, "%s: buffer error %d at frame %d\n", - __func__, ifd->offset, nframe); - return ifd->offset; + __func__, rc, nframe); + return rc; } break; } + ifd->offset = rc; ucx->limit = ubc->isooutbuf->nextread; ifd->status = 0; ifd->actual_length = 0; From e53f12cc6c43b69f54937f15c5706d83f67c2fdd Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 23 Jul 2008 21:28:28 -0700 Subject: [PATCH 222/857] remove include/asm-h8300/keyboard.h This patch removes the unused include/asm-h8300/keyboard.h Signed-off-by: Adrian Bunk Acked-by: Yoshinori Sato Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-h8300/keyboard.h | 24 ------------------------ 1 file changed, 24 deletions(-) delete mode 100644 include/asm-h8300/keyboard.h diff --git a/include/asm-h8300/keyboard.h b/include/asm-h8300/keyboard.h deleted file mode 100644 index 90efbd655390..000000000000 --- a/include/asm-h8300/keyboard.h +++ /dev/null @@ -1,24 +0,0 @@ -/* - * linux/include/asm-h8300/keyboard.h - * Created 04 Dec 2001 by Khaled Hassounah - * This file contains the Dragonball architecture specific keyboard definitions - */ - -#ifndef _H8300_KEYBOARD_H -#define _H8300_KEYBOARD_H - - -/* dummy i.e. no real keyboard */ -#define kbd_setkeycode(x...) (-ENOSYS) -#define kbd_getkeycode(x...) (-ENOSYS) -#define kbd_translate(x...) (0) -#define kbd_unexpected_up(x...) (1) -#define kbd_leds(x...) do {;} while (0) -#define kbd_init_hw(x...) do {;} while (0) -#define kbd_enable_irq(x...) do {;} while (0) -#define kbd_disable_irq(x...) do {;} while (0) - -#endif /* _H8300_KEYBOARD_H */ - - - From fb9ba4e95921f71d874beee2d0964fc2322b47a2 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 23 Jul 2008 21:28:29 -0700 Subject: [PATCH 223/857] alpha: remove the unused ALPHA_CORE_AGP option The real option is named AGP_ALPHA_CORE. Reviewed-by: Robert P. J. Day Signed-off-by: Adrian Bunk Cc: Richard Henderson Cc: Ivan Kokshaysky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/alpha/Kconfig | 5 ----- 1 file changed, 5 deletions(-) diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig index dbe8c280fea9..1bec55d63ef6 100644 --- a/arch/alpha/Kconfig +++ b/arch/alpha/Kconfig @@ -333,11 +333,6 @@ config PCI_SYSCALL config IOMMU_HELPER def_bool PCI -config ALPHA_CORE_AGP - bool - depends on ALPHA_GENERIC || ALPHA_TITAN || ALPHA_MARVEL - default y - config ALPHA_NONAME bool depends on ALPHA_BOOK1 || ALPHA_NONAME_CH From 0d63081d418c73cc187c893069e0f24c4c6eecd3 Mon Sep 17 00:00:00 2001 From: Pavel Machek Date: Wed, 23 Jul 2008 21:28:32 -0700 Subject: [PATCH 224/857] swsusp: provide users with a hint about the no_console_suspend option Tell the user about the no_console_suspend option, so that we don't have to tell each bug reporter personally. [akpm@linux-foundation.org: clarify the text a little] Signed-off-by: Pavel Machek Cc: "Rafael J. Wysocki" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/printk.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/printk.c b/kernel/printk.c index 07ad9e7f7a66..3f7a2a94583b 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -933,7 +933,7 @@ void suspend_console(void) { if (!console_suspend_enabled) return; - printk("Suspending console(s)\n"); + printk("Suspending console(s) (use no_console_suspend to debug)\n"); acquire_console_sem(); console_suspended = 1; } From 77437fd4e61f87cc94d9314baa5cbf50e3ccdf54 Mon Sep 17 00:00:00 2001 From: David Brownell Date: Wed, 23 Jul 2008 21:28:33 -0700 Subject: [PATCH 225/857] pm: boot time suspend selftest Boot-time test for system suspend states (STR or standby). The generic RTC framework triggers wakeup alarms, which are used to exit those states. - Measures some aspects of suspend time ... this uses "jiffies" until someone converts it to use a timebase that works properly even while timer IRQs are disabled. - Triggered by a command line parameter. By default nothing even vaguely troublesome will happen, but "test_suspend=mem" will give you a brief STR test during system boot. (Or you may need to use "test_suspend=standby" instead, if your hardware needs that.) This isn't without problems. It fires early enough during boot that for example both PCMCIA and MMC stacks have misbehaved. The workaround in those cases was to boot without such media cards inserted. [matthltc@us.ibm.com: fix compile failure in boot time suspend selftest] Signed-off-by: David Brownell Cc: Ingo Molnar Cc: Pavel Machek Cc: "Rafael J. Wysocki" Signed-off-by: Matt Helsley Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/kernel-parameters.txt | 9 +- kernel/power/Kconfig | 11 ++ kernel/power/main.c | 194 +++++++++++++++++++++++++++- 3 files changed, 212 insertions(+), 2 deletions(-) diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 01a2992b5754..4d705713cabc 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -87,7 +87,8 @@ parameter is applicable: SH SuperH architecture is enabled. SMP The kernel is an SMP kernel. SPARC Sparc architecture is enabled. - SWSUSP Software suspend is enabled. + SWSUSP Software suspend (hibernation) is enabled. + SUSPEND System suspend states are enabled. TS Appropriate touchscreen support is enabled. USB USB support is enabled. USBHID USB Human Interface Device support is enabled. @@ -2123,6 +2124,12 @@ and is between 256 and 4096 characters. It is defined in the file tdfx= [HW,DRM] + test_suspend= [SUSPEND] + Specify "mem" (for Suspend-to-RAM) or "standby" (for + standby suspend) as the system sleep state to briefly + enter during system startup. The system is woken from + this state using a wakeup-capable RTC alarm. + thash_entries= [KNL,NET] Set number of hash buckets for TCP connection diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig index 59dfdf1e1d20..dcd165f92a88 100644 --- a/kernel/power/Kconfig +++ b/kernel/power/Kconfig @@ -94,6 +94,17 @@ config SUSPEND powered and thus its contents are preserved, such as the suspend-to-RAM state (e.g. the ACPI S3 state). +config PM_TEST_SUSPEND + bool "Test suspend/resume and wakealarm during bootup" + depends on SUSPEND && PM_DEBUG && RTC_LIB=y + ---help--- + This option will let you suspend your machine during bootup, and + make it wake up a few seconds later using an RTC wakeup alarm. + Enable this with a kernel parameter like "test_suspend=mem". + + You probably want to have your system's RTC driver statically + linked, ensuring that it's available when this test runs. + config SUSPEND_FREEZER bool "Enable freezer for suspend to RAM/standby" \ if ARCH_WANTS_FREEZER_CONTROL || BROKEN diff --git a/kernel/power/main.c b/kernel/power/main.c index 3398f4651aa1..95bff23ecdaa 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -132,6 +132,61 @@ static inline int suspend_test(int level) { return 0; } #ifdef CONFIG_SUSPEND +#ifdef CONFIG_PM_TEST_SUSPEND + +/* + * We test the system suspend code by setting an RTC wakealarm a short + * time in the future, then suspending. Suspending the devices won't + * normally take long ... some systems only need a few milliseconds. + * + * The time it takes is system-specific though, so when we test this + * during system bootup we allow a LOT of time. + */ +#define TEST_SUSPEND_SECONDS 5 + +static unsigned long suspend_test_start_time; + +static void suspend_test_start(void) +{ + /* FIXME Use better timebase than "jiffies", ideally a clocksource. + * What we want is a hardware counter that will work correctly even + * during the irqs-are-off stages of the suspend/resume cycle... + */ + suspend_test_start_time = jiffies; +} + +static void suspend_test_finish(const char *label) +{ + long nj = jiffies - suspend_test_start_time; + unsigned msec; + + msec = jiffies_to_msecs(abs(nj)); + pr_info("PM: %s took %d.%03d seconds\n", label, + msec / 1000, msec % 1000); + + /* Warning on suspend means the RTC alarm period needs to be + * larger -- the system was sooo slooowwww to suspend that the + * alarm (should have) fired before the system went to sleep! + * + * Warning on either suspend or resume also means the system + * has some performance issues. The stack dump of a WARN_ON + * is more likely to get the right attention than a printk... + */ + WARN_ON(msec > (TEST_SUSPEND_SECONDS * 1000)); +} + +#else + +static void suspend_test_start(void) +{ +} + +static void suspend_test_finish(const char *label) +{ +} + +#endif + /* This is just an arbitrary number */ #define FREE_PAGE_NUMBER (100) @@ -266,12 +321,13 @@ int suspend_devices_and_enter(suspend_state_t state) goto Close; } suspend_console(); + suspend_test_start(); error = device_suspend(PMSG_SUSPEND); if (error) { printk(KERN_ERR "PM: Some devices failed to suspend\n"); goto Recover_platform; } - + suspend_test_finish("suspend devices"); if (suspend_test(TEST_DEVICES)) goto Recover_platform; @@ -293,7 +349,9 @@ int suspend_devices_and_enter(suspend_state_t state) if (suspend_ops->finish) suspend_ops->finish(); Resume_devices: + suspend_test_start(); device_resume(PMSG_RESUME); + suspend_test_finish("resume devices"); resume_console(); Close: if (suspend_ops->end) @@ -521,3 +579,137 @@ static int __init pm_init(void) } core_initcall(pm_init); + + +#ifdef CONFIG_PM_TEST_SUSPEND + +#include + +/* + * To test system suspend, we need a hands-off mechanism to resume the + * system. RTCs wake alarms are a common self-contained mechanism. + */ + +static void __init test_wakealarm(struct rtc_device *rtc, suspend_state_t state) +{ + static char err_readtime[] __initdata = + KERN_ERR "PM: can't read %s time, err %d\n"; + static char err_wakealarm [] __initdata = + KERN_ERR "PM: can't set %s wakealarm, err %d\n"; + static char err_suspend[] __initdata = + KERN_ERR "PM: suspend test failed, error %d\n"; + static char info_test[] __initdata = + KERN_INFO "PM: test RTC wakeup from '%s' suspend\n"; + + unsigned long now; + struct rtc_wkalrm alm; + int status; + + /* this may fail if the RTC hasn't been initialized */ + status = rtc_read_time(rtc, &alm.time); + if (status < 0) { + printk(err_readtime, rtc->dev.bus_id, status); + return; + } + rtc_tm_to_time(&alm.time, &now); + + memset(&alm, 0, sizeof alm); + rtc_time_to_tm(now + TEST_SUSPEND_SECONDS, &alm.time); + alm.enabled = true; + + status = rtc_set_alarm(rtc, &alm); + if (status < 0) { + printk(err_wakealarm, rtc->dev.bus_id, status); + return; + } + + if (state == PM_SUSPEND_MEM) { + printk(info_test, pm_states[state]); + status = pm_suspend(state); + if (status == -ENODEV) + state = PM_SUSPEND_STANDBY; + } + if (state == PM_SUSPEND_STANDBY) { + printk(info_test, pm_states[state]); + status = pm_suspend(state); + } + if (status < 0) + printk(err_suspend, status); +} + +static int __init has_wakealarm(struct device *dev, void *name_ptr) +{ + struct rtc_device *candidate = to_rtc_device(dev); + + if (!candidate->ops->set_alarm) + return 0; + if (!device_may_wakeup(candidate->dev.parent)) + return 0; + + *(char **)name_ptr = dev->bus_id; + return 1; +} + +/* + * Kernel options like "test_suspend=mem" force suspend/resume sanity tests + * at startup time. They're normally disabled, for faster boot and because + * we can't know which states really work on this particular system. + */ +static suspend_state_t test_state __initdata = PM_SUSPEND_ON; + +static char warn_bad_state[] __initdata = + KERN_WARNING "PM: can't test '%s' suspend state\n"; + +static int __init setup_test_suspend(char *value) +{ + unsigned i; + + /* "=mem" ==> "mem" */ + value++; + for (i = 0; i < PM_SUSPEND_MAX; i++) { + if (!pm_states[i]) + continue; + if (strcmp(pm_states[i], value) != 0) + continue; + test_state = (__force suspend_state_t) i; + return 0; + } + printk(warn_bad_state, value); + return 0; +} +__setup("test_suspend", setup_test_suspend); + +static int __init test_suspend(void) +{ + static char warn_no_rtc[] __initdata = + KERN_WARNING "PM: no wakealarm-capable RTC driver is ready\n"; + + char *pony = NULL; + struct rtc_device *rtc = NULL; + + /* PM is initialized by now; is that state testable? */ + if (test_state == PM_SUSPEND_ON) + goto done; + if (!valid_state(test_state)) { + printk(warn_bad_state, pm_states[test_state]); + goto done; + } + + /* RTCs have initialized by now too ... can we use one? */ + class_find_device(rtc_class, NULL, &pony, has_wakealarm); + if (pony) + rtc = rtc_class_open(pony); + if (!rtc) { + printk(warn_no_rtc); + goto done; + } + + /* go for it */ + test_wakealarm(rtc, test_state); + rtc_class_close(rtc); +done: + return 0; +} +late_initcall(test_suspend); + +#endif /* CONFIG_PM_TEST_SUSPEND */ From d75f65fd247fe85d90a3880d143b1bb22fe13a48 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 23 Jul 2008 21:28:34 -0700 Subject: [PATCH 226/857] remove include/linux/pm_legacy.h Remove the obsolete and no longer used include/linux/pm_legacy.h Reviewed-by: Robert P. J. Day Signed-off-by: Adrian Bunk Cc: Pavel Machek Acked-by: "Rafael J. Wysocki" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/frv/kernel/pm.c | 1 - arch/mips/au1000/common/power.c | 1 - arch/x86/kernel/apm_32.c | 1 - include/linux/pm_legacy.h | 35 --------------------------------- 4 files changed, 38 deletions(-) delete mode 100644 include/linux/pm_legacy.h diff --git a/arch/frv/kernel/pm.c b/arch/frv/kernel/pm.c index 73f3aeefd203..d1113c5031f5 100644 --- a/arch/frv/kernel/pm.c +++ b/arch/frv/kernel/pm.c @@ -14,7 +14,6 @@ #include #include #include -#include #include #include #include diff --git a/arch/mips/au1000/common/power.c b/arch/mips/au1000/common/power.c index 2166b9e1e80c..bd854a6d1d89 100644 --- a/arch/mips/au1000/common/power.c +++ b/arch/mips/au1000/common/power.c @@ -31,7 +31,6 @@ #include #include -#include #include #include diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c index bf9b441331e9..9ee24e6bc4b0 100644 --- a/arch/x86/kernel/apm_32.c +++ b/arch/x86/kernel/apm_32.c @@ -219,7 +219,6 @@ #include #include #include -#include #include #include #include diff --git a/include/linux/pm_legacy.h b/include/linux/pm_legacy.h deleted file mode 100644 index 446f4f42b952..000000000000 --- a/include/linux/pm_legacy.h +++ /dev/null @@ -1,35 +0,0 @@ -#ifndef __LINUX_PM_LEGACY_H__ -#define __LINUX_PM_LEGACY_H__ - - -#ifdef CONFIG_PM_LEGACY - -/* - * Register a device with power management - */ -struct pm_dev __deprecated * -pm_register(pm_dev_t type, unsigned long id, pm_callback callback); - -/* - * Send a request to all devices - */ -int __deprecated pm_send_all(pm_request_t rqst, void *data); - -#else /* CONFIG_PM_LEGACY */ - -static inline struct pm_dev *pm_register(pm_dev_t type, - unsigned long id, - pm_callback callback) -{ - return NULL; -} - -static inline int pm_send_all(pm_request_t rqst, void *data) -{ - return 0; -} - -#endif /* CONFIG_PM_LEGACY */ - -#endif /* __LINUX_PM_LEGACY_H__ */ - From 558481f038e587b22d02167af58914c814ce9de5 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Wed, 23 Jul 2008 21:28:35 -0700 Subject: [PATCH 227/857] pm: remove definition of struct pm_dev Remove the definition of 'struct pm_dev', which is not used any more, along with some related stuff from include/linux/pm.h . Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/pm.h | 24 ------------------------ 1 file changed, 24 deletions(-) diff --git a/include/linux/pm.h b/include/linux/pm.h index 4ad9de94449a..5bf1ce89cfbb 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -68,30 +68,6 @@ enum */ #define PM_PCI_ID(dev) ((dev)->bus->number << 16 | (dev)->devfn) -/* - * Request handler callback - */ -struct pm_dev; - -typedef int (*pm_callback)(struct pm_dev *dev, pm_request_t rqst, void *data); - -/* - * Dynamic device information - */ -struct pm_dev -{ - pm_dev_t type; - unsigned long id; - pm_callback callback; - void *data; - - unsigned long flags; - unsigned long state; - unsigned long prev_state; - - struct list_head entry; -}; - /* Functions above this comment are list-based old-style power * management. Please avoid using them. */ From e7ecb331e11d1f7aa66aeef9170fc20781c9bb55 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Wed, 23 Jul 2008 21:28:35 -0700 Subject: [PATCH 228/857] pm: remove remaining obsolete definitions from pm.h Remove the remaining obsolete definitions from include/linux/pm.h and move the definitions of PM_SUSPEND and PM_RESUME to the header of h3600 which is the only user of them. Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-arm/arch-sa1100/h3600.h | 5 ++++ include/linux/pm.h | 46 ----------------------------- 2 files changed, 5 insertions(+), 46 deletions(-) diff --git a/include/asm-arm/arch-sa1100/h3600.h b/include/asm-arm/arch-sa1100/h3600.h index 1b6355971574..3ca0ecf095e6 100644 --- a/include/asm-arm/arch-sa1100/h3600.h +++ b/include/asm-arm/arch-sa1100/h3600.h @@ -23,6 +23,11 @@ #ifndef _INCLUDE_H3600_H_ #define _INCLUDE_H3600_H_ +typedef int __bitwise pm_request_t; + +#define PM_SUSPEND ((__force pm_request_t) 1) /* enter D1-D3 */ +#define PM_RESUME ((__force pm_request_t) 2) /* enter D0 */ + /* generalized support for H3xxx series Compaq Pocket PC's */ #define machine_is_h3xxx() (machine_is_h3100() || machine_is_h3600() || machine_is_h3800()) diff --git a/include/linux/pm.h b/include/linux/pm.h index 5bf1ce89cfbb..390dd95a375e 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -25,52 +25,6 @@ #include #include -/* - * Power management requests... these are passed to pm_send_all() and friends. - * - * these functions are old and deprecated, see below. - */ -typedef int __bitwise pm_request_t; - -#define PM_SUSPEND ((__force pm_request_t) 1) /* enter D1-D3 */ -#define PM_RESUME ((__force pm_request_t) 2) /* enter D0 */ - - -/* - * Device types... these are passed to pm_register - */ -typedef int __bitwise pm_dev_t; - -#define PM_UNKNOWN_DEV ((__force pm_dev_t) 0) /* generic */ -#define PM_SYS_DEV ((__force pm_dev_t) 1) /* system device (fan, KB controller, ...) */ -#define PM_PCI_DEV ((__force pm_dev_t) 2) /* PCI device */ -#define PM_USB_DEV ((__force pm_dev_t) 3) /* USB device */ -#define PM_SCSI_DEV ((__force pm_dev_t) 4) /* SCSI device */ -#define PM_ISA_DEV ((__force pm_dev_t) 5) /* ISA device */ -#define PM_MTD_DEV ((__force pm_dev_t) 6) /* Memory Technology Device */ - -/* - * System device hardware ID (PnP) values - */ -enum -{ - PM_SYS_UNKNOWN = 0x00000000, /* generic */ - PM_SYS_KBC = 0x41d00303, /* keyboard controller */ - PM_SYS_COM = 0x41d00500, /* serial port */ - PM_SYS_IRDA = 0x41d00510, /* IRDA controller */ - PM_SYS_FDC = 0x41d00700, /* floppy controller */ - PM_SYS_VGA = 0x41d00900, /* VGA controller */ - PM_SYS_PCMCIA = 0x41d00e00, /* PCMCIA controller */ -}; - -/* - * Device identifier - */ -#define PM_PCI_ID(dev) ((dev)->bus->number << 16 | (dev)->devfn) - -/* Functions above this comment are list-based old-style power - * management. Please avoid using them. */ - /* * Callbacks for platform drivers to implement. */ From 40b4ac33b4d1bdd5cbeb2241be2399c550fa3696 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Wed, 23 Jul 2008 21:28:36 -0700 Subject: [PATCH 229/857] pm: remove obsolete piece of PM documentation Remove some obsolete PM documentation. The majority of contents of Documentation/power/pm.txt are outdated. Remove the outdated parts of this file and move the rest to Documentation/power/apm-acpi.txt . Update the index in Documentation/power/ as appropriate. Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Acked-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/power/00-INDEX | 4 +- Documentation/power/apm-acpi.txt | 32 ++++ Documentation/power/pm.txt | 257 ------------------------------- 3 files changed, 34 insertions(+), 259 deletions(-) create mode 100644 Documentation/power/apm-acpi.txt delete mode 100644 Documentation/power/pm.txt diff --git a/Documentation/power/00-INDEX b/Documentation/power/00-INDEX index a55d7f1c836d..fb742c213c9e 100644 --- a/Documentation/power/00-INDEX +++ b/Documentation/power/00-INDEX @@ -1,5 +1,7 @@ 00-INDEX - This file +apm-acpi.txt + - basic info about the APM and ACPI support. basic-pm-debugging.txt - Debugging suspend and resume devices.txt @@ -14,8 +16,6 @@ notifiers.txt - Registering suspend notifiers in device drivers pci.txt - How the PCI Subsystem Does Power Management -pm.txt - - info on Linux power management support. pm_qos_interface.txt - info on Linux PM Quality of Service interface power_supply_class.txt diff --git a/Documentation/power/apm-acpi.txt b/Documentation/power/apm-acpi.txt new file mode 100644 index 000000000000..1bd799dc17e8 --- /dev/null +++ b/Documentation/power/apm-acpi.txt @@ -0,0 +1,32 @@ +APM or ACPI? +------------ +If you have a relatively recent x86 mobile, desktop, or server system, +odds are it supports either Advanced Power Management (APM) or +Advanced Configuration and Power Interface (ACPI). ACPI is the newer +of the two technologies and puts power management in the hands of the +operating system, allowing for more intelligent power management than +is possible with BIOS controlled APM. + +The best way to determine which, if either, your system supports is to +build a kernel with both ACPI and APM enabled (as of 2.3.x ACPI is +enabled by default). If a working ACPI implementation is found, the +ACPI driver will override and disable APM, otherwise the APM driver +will be used. + +No, sorry, you cannot have both ACPI and APM enabled and running at +once. Some people with broken ACPI or broken APM implementations +would like to use both to get a full set of working features, but you +simply cannot mix and match the two. Only one power management +interface can be in control of the machine at once. Think about it.. + +User-space Daemons +------------------ +Both APM and ACPI rely on user-space daemons, apmd and acpid +respectively, to be completely functional. Obtain both of these +daemons from your Linux distribution or from the Internet (see below) +and be sure that they are started sometime in the system boot process. +Go ahead and start both. If ACPI or APM is not available on your +system the associated daemon will exit gracefully. + + apmd: http://worldvisions.ca/~apenwarr/apmd/ + acpid: http://acpid.sf.net/ diff --git a/Documentation/power/pm.txt b/Documentation/power/pm.txt deleted file mode 100644 index be841507e43f..000000000000 --- a/Documentation/power/pm.txt +++ /dev/null @@ -1,257 +0,0 @@ - Linux Power Management Support - -This document briefly describes how to use power management with your -Linux system and how to add power management support to Linux drivers. - -APM or ACPI? ------------- -If you have a relatively recent x86 mobile, desktop, or server system, -odds are it supports either Advanced Power Management (APM) or -Advanced Configuration and Power Interface (ACPI). ACPI is the newer -of the two technologies and puts power management in the hands of the -operating system, allowing for more intelligent power management than -is possible with BIOS controlled APM. - -The best way to determine which, if either, your system supports is to -build a kernel with both ACPI and APM enabled (as of 2.3.x ACPI is -enabled by default). If a working ACPI implementation is found, the -ACPI driver will override and disable APM, otherwise the APM driver -will be used. - -No, sorry, you cannot have both ACPI and APM enabled and running at -once. Some people with broken ACPI or broken APM implementations -would like to use both to get a full set of working features, but you -simply cannot mix and match the two. Only one power management -interface can be in control of the machine at once. Think about it.. - -User-space Daemons ------------------- -Both APM and ACPI rely on user-space daemons, apmd and acpid -respectively, to be completely functional. Obtain both of these -daemons from your Linux distribution or from the Internet (see below) -and be sure that they are started sometime in the system boot process. -Go ahead and start both. If ACPI or APM is not available on your -system the associated daemon will exit gracefully. - - apmd: http://worldvisions.ca/~apenwarr/apmd/ - acpid: http://acpid.sf.net/ - -Driver Interface -- OBSOLETE, DO NOT USE! -----------------************************* - -Note: pm_register(), pm_access(), pm_dev_idle() and friends are -obsolete. Please do not use them. Instead you should properly hook -your driver into the driver model, and use its suspend()/resume() -callbacks to do this kind of stuff. - -If you are writing a new driver or maintaining an old driver, it -should include power management support. Without power management -support, a single driver may prevent a system with power management -capabilities from ever being able to suspend (safely). - -Overview: -1) Register each instance of a device with "pm_register" -2) Call "pm_access" before accessing the hardware. - (this will ensure that the hardware is awake and ready) -3) Your "pm_callback" is called before going into a - suspend state (ACPI D1-D3) or after resuming (ACPI D0) - from a suspend. -4) Call "pm_dev_idle" when the device is not being used - (optional but will improve device idle detection) -5) When unloaded, unregister the device with "pm_unregister" - -/* - * Description: Register a device with the power-management subsystem - * - * Parameters: - * type - device type (PCI device, system device, ...) - * id - instance number or unique identifier - * cback - request handler callback (suspend, resume, ...) - * - * Returns: Registered PM device or NULL on error - * - * Examples: - * dev = pm_register(PM_SYS_DEV, PM_SYS_VGA, vga_callback); - * - * struct pci_dev *pci_dev = pci_find_dev(...); - * dev = pm_register(PM_PCI_DEV, PM_PCI_ID(pci_dev), callback); - */ -struct pm_dev *pm_register(pm_dev_t type, unsigned long id, pm_callback cback); - -/* - * Description: Unregister a device with the power management subsystem - * - * Parameters: - * dev - PM device previously returned from pm_register - */ -void pm_unregister(struct pm_dev *dev); - -/* - * Description: Unregister all devices with a matching callback function - * - * Parameters: - * cback - previously registered request callback - * - * Notes: Provided for easier porting from old APM interface - */ -void pm_unregister_all(pm_callback cback); - -/* - * Power management request callback - * - * Parameters: - * dev - PM device previously returned from pm_register - * rqst - request type - * data - data, if any, associated with the request - * - * Returns: 0 if the request is successful - * EINVAL if the request is not supported - * EBUSY if the device is now busy and cannot handle the request - * ENOMEM if the device was unable to handle the request due to memory - * - * Details: The device request callback will be called before the - * device/system enters a suspend state (ACPI D1-D3) or - * or after the device/system resumes from suspend (ACPI D0). - * For PM_SUSPEND, the ACPI D-state being entered is passed - * as the "data" argument to the callback. The device - * driver should save (PM_SUSPEND) or restore (PM_RESUME) - * device context when the request callback is called. - * - * Once a driver returns 0 (success) from a suspend - * request, it should not process any further requests or - * access the device hardware until a call to "pm_access" is made. - */ -typedef int (*pm_callback)(struct pm_dev *dev, pm_request_t rqst, void *data); - -Driver Details --------------- -This is just a quick Q&A as a stopgap until a real driver writers' -power management guide is available. - -Q: When is a device suspended? - -Devices can be suspended based on direct user request (eg. laptop lid -closes), system power policy (eg. sleep after 30 minutes of console -inactivity), or device power policy (eg. power down device after 5 -minutes of inactivity) - -Q: Must a driver honor a suspend request? - -No, a driver can return -EBUSY from a suspend request and this -will stop the system from suspending. When a suspend request -fails, all suspended devices are resumed and the system continues -to run. Suspend can be retried at a later time. - -Q: Can the driver block suspend/resume requests? - -Yes, a driver can delay its return from a suspend or resume -request until the device is ready to handle requests. It -is advantageous to return as quickly as possible from a -request as suspend/resume are done serially. - -Q: What context is a suspend/resume initiated from? - -A suspend or resume is initiated from a kernel thread context. -It is safe to block, allocate memory, initiate requests -or anything else you can do within the kernel. - -Q: Will requests continue to arrive after a suspend? - -Possibly. It is the driver's responsibility to queue(*), -fail, or drop any requests that arrive after returning -success to a suspend request. It is important that the -driver not access its device until after it receives -a resume request as the device's bus may no longer -be active. - -(*) If a driver queues requests for processing after - resume be aware that the device, network, etc. - might be in a different state than at suspend time. - It's probably better to drop requests unless - the driver is a storage device. - -Q: Do I have to manage bus-specific power management registers - -No. It is the responsibility of the bus driver to manage -PCI, USB, etc. power management registers. The bus driver -or the power management subsystem will also enable any -wake-on functionality that the device has. - -Q: So, really, what do I need to do to support suspend/resume? - -You need to save any device context that would -be lost if the device was powered off and then restore -it at resume time. When ACPI is active, there are -three levels of device suspend states; D1, D2, and D3. -(The suspend state is passed as the "data" argument -to the device callback.) With D3, the device is powered -off and loses all context, D1 and D2 are shallower power -states and require less device context to be saved. To -play it safe, just save everything at suspend and restore -everything at resume. - -Q: Where do I store device context for suspend? - -Anywhere in memory, kmalloc a buffer or store it -in the device descriptor. You are guaranteed that the -contents of memory will be restored and accessible -before resume, even when the system suspends to disk. - -Q: What do I need to do for ACPI vs. APM vs. etc? - -Drivers need not be aware of the specific power management -technology that is active. They just need to be aware -of when the overlying power management system requests -that they suspend or resume. - -Q: What about device dependencies? - -When a driver registers a device, the power management -subsystem uses the information provided to build a -tree of device dependencies (eg. USB device X is on -USB controller Y which is on PCI bus Z) When power -management wants to suspend a device, it first sends -a suspend request to its driver, then the bus driver, -and so on up to the system bus. Device resumes -proceed in the opposite direction. - -Q: Who do I contact for additional information about - enabling power management for my specific driver/device? - -ACPI Development mailing list: linux-acpi@vger.kernel.org - -System Interface -- OBSOLETE, DO NOT USE! -----------------************************* -If you are providing new power management support to Linux (ie. -adding support for something like APM or ACPI), you should -communicate with drivers through the existing generic power -management interface. - -/* - * Send a request to all devices - * - * Parameters: - * rqst - request type - * data - data, if any, associated with the request - * - * Returns: 0 if the request is successful - * See "pm_callback" return for errors - * - * Details: Walk list of registered devices and call pm_send - * for each until complete or an error is encountered. - * If an error is encountered for a suspend request, - * return all devices to the state they were in before - * the suspend request. - */ -int pm_send_all(pm_request_t rqst, void *data); - -/* - * Find a matching device - * - * Parameters: - * type - device type (PCI device, system device, or 0 to match all devices) - * from - previous match or NULL to start from the beginning - * - * Returns: Matching device or NULL if none found - */ -struct pm_dev *pm_find(pm_dev_t type, struct pm_dev *from); From 8c363265d57d755e62053e9f69a1f2164e83f7ea Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Wed, 23 Jul 2008 21:28:37 -0700 Subject: [PATCH 230/857] pm: drop unnecessary includes from pm.h Drop unnecessary includes from include/linux/pm.h . Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/pm.h | 2 -- 1 file changed, 2 deletions(-) diff --git a/include/linux/pm.h b/include/linux/pm.h index 390dd95a375e..ed98d967f9fb 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -22,8 +22,6 @@ #define _LINUX_PM_H #include -#include -#include /* * Callbacks for platform drivers to implement. From 8111d1b552349921aae1acf73e4e8cea98e80970 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Wed, 23 Jul 2008 21:28:37 -0700 Subject: [PATCH 231/857] pm: add new PM_EVENT codes for runtime power transitions This patch (as1112) adds some new PM_EVENT_* codes for use by kernel subsystems. They describe runtime power-state transitions of the sort already implemented by the USB subsystem. Signed-off-by: Alan Stern Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/pm.h | 37 +++++++++++++++++++++++++++++++++++-- 1 file changed, 35 insertions(+), 2 deletions(-) diff --git a/include/linux/pm.h b/include/linux/pm.h index ed98d967f9fb..4dcce54b6d76 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -245,6 +245,21 @@ struct pm_ext_ops { * RECOVER Creation of a hibernation image or restoration of the main * memory contents from a hibernation image has failed, call * ->thaw() and ->complete() for all devices. + * + * The following PM_EVENT_ messages are defined for internal use by + * kernel subsystems. They are never issued by the PM core. + * + * USER_SUSPEND Manual selective suspend was issued by userspace. + * + * USER_RESUME Manual selective resume was issued by userspace. + * + * REMOTE_WAKEUP Remote-wakeup request was received from the device. + * + * AUTO_SUSPEND Automatic (device idle) runtime suspend was + * initiated by the subsystem. + * + * AUTO_RESUME Automatic (device needed) runtime resume was + * requested by a driver. */ #define PM_EVENT_ON 0x0000 @@ -256,9 +271,18 @@ struct pm_ext_ops { #define PM_EVENT_THAW 0x0020 #define PM_EVENT_RESTORE 0x0040 #define PM_EVENT_RECOVER 0x0080 +#define PM_EVENT_USER 0x0100 +#define PM_EVENT_REMOTE 0x0200 +#define PM_EVENT_AUTO 0x0400 -#define PM_EVENT_SLEEP (PM_EVENT_SUSPEND | PM_EVENT_HIBERNATE) +#define PM_EVENT_SLEEP (PM_EVENT_SUSPEND | PM_EVENT_HIBERNATE) +#define PM_EVENT_USER_SUSPEND (PM_EVENT_USER | PM_EVENT_SUSPEND) +#define PM_EVENT_USER_RESUME (PM_EVENT_USER | PM_EVENT_RESUME) +#define PM_EVENT_REMOTE_WAKEUP (PM_EVENT_REMOTE | PM_EVENT_RESUME) +#define PM_EVENT_AUTO_SUSPEND (PM_EVENT_AUTO | PM_EVENT_SUSPEND) +#define PM_EVENT_AUTO_RESUME (PM_EVENT_AUTO | PM_EVENT_RESUME) +#define PMSG_ON ((struct pm_message){ .event = PM_EVENT_ON, }) #define PMSG_FREEZE ((struct pm_message){ .event = PM_EVENT_FREEZE, }) #define PMSG_QUIESCE ((struct pm_message){ .event = PM_EVENT_QUIESCE, }) #define PMSG_SUSPEND ((struct pm_message){ .event = PM_EVENT_SUSPEND, }) @@ -267,7 +291,16 @@ struct pm_ext_ops { #define PMSG_THAW ((struct pm_message){ .event = PM_EVENT_THAW, }) #define PMSG_RESTORE ((struct pm_message){ .event = PM_EVENT_RESTORE, }) #define PMSG_RECOVER ((struct pm_message){ .event = PM_EVENT_RECOVER, }) -#define PMSG_ON ((struct pm_message){ .event = PM_EVENT_ON, }) +#define PMSG_USER_SUSPEND ((struct pm_messge) \ + { .event = PM_EVENT_USER_SUSPEND, }) +#define PMSG_USER_RESUME ((struct pm_messge) \ + { .event = PM_EVENT_USER_RESUME, }) +#define PMSG_REMOTE_RESUME ((struct pm_messge) \ + { .event = PM_EVENT_REMOTE_RESUME, }) +#define PMSG_AUTO_SUSPEND ((struct pm_messge) \ + { .event = PM_EVENT_AUTO_SUSPEND, }) +#define PMSG_AUTO_RESUME ((struct pm_messge) \ + { .event = PM_EVENT_AUTO_RESUME, }) /** * Device power management states From 0d83304c7e7bd3b05be90281b3a47841bc8f057a Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Wed, 23 Jul 2008 21:28:38 -0700 Subject: [PATCH 232/857] pm: hibernation: simplify memory bitmap This patch simplifies the memory bitmap manipulations. - remove the member size in struct bm_block It is not necessary for struct bm_block to have the number of bit chunks that can be calculated by using end_pfn and start_pfn. - use find_next_bit() for memory_bm_next_pfn No need to invent the bitmap library only for the memory bitmap. Signed-off-by: Akinobu Mita Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/power/snapshot.c | 86 ++++++++++------------------------------- 1 file changed, 20 insertions(+), 66 deletions(-) diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index 5f91a07c4eac..5d2ab836e998 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -205,8 +205,7 @@ static void chain_free(struct chain_allocator *ca, int clear_page_nosave) * objects. The main list's elements are of type struct zone_bitmap * and each of them corresonds to one zone. For each zone bitmap * object there is a list of objects of type struct bm_block that - * represent each blocks of bit chunks in which information is - * stored. + * represent each blocks of bitmap in which information is stored. * * struct memory_bitmap contains a pointer to the main list of zone * bitmap objects, a struct bm_position used for browsing the bitmap, @@ -224,26 +223,27 @@ static void chain_free(struct chain_allocator *ca, int clear_page_nosave) * pfns that correspond to the start and end of the represented zone. * * struct bm_block contains a pointer to the memory page in which - * information is stored (in the form of a block of bit chunks - * of type unsigned long each). It also contains the pfns that - * correspond to the start and end of the represented memory area and - * the number of bit chunks in the block. + * information is stored (in the form of a block of bitmap) + * It also contains the pfns that correspond to the start and end of + * the represented memory area. */ #define BM_END_OF_MAP (~0UL) -#define BM_CHUNKS_PER_BLOCK (PAGE_SIZE / sizeof(long)) -#define BM_BITS_PER_CHUNK (sizeof(long) << 3) #define BM_BITS_PER_BLOCK (PAGE_SIZE << 3) struct bm_block { struct bm_block *next; /* next element of the list */ unsigned long start_pfn; /* pfn represented by the first bit */ unsigned long end_pfn; /* pfn represented by the last bit plus 1 */ - unsigned int size; /* number of bit chunks */ - unsigned long *data; /* chunks of bits representing pages */ + unsigned long *data; /* bitmap representing pages */ }; +static inline unsigned long bm_block_bits(struct bm_block *bb) +{ + return bb->end_pfn - bb->start_pfn; +} + struct zone_bitmap { struct zone_bitmap *next; /* next element of the list */ unsigned long start_pfn; /* minimal pfn in this zone */ @@ -257,7 +257,6 @@ struct zone_bitmap { struct bm_position { struct zone_bitmap *zone_bm; struct bm_block *block; - int chunk; int bit; }; @@ -272,12 +271,6 @@ struct memory_bitmap { /* Functions that operate on memory bitmaps */ -static inline void memory_bm_reset_chunk(struct memory_bitmap *bm) -{ - bm->cur.chunk = 0; - bm->cur.bit = -1; -} - static void memory_bm_position_reset(struct memory_bitmap *bm) { struct zone_bitmap *zone_bm; @@ -285,7 +278,7 @@ static void memory_bm_position_reset(struct memory_bitmap *bm) zone_bm = bm->zone_bm_list; bm->cur.zone_bm = zone_bm; bm->cur.block = zone_bm->bm_blocks; - memory_bm_reset_chunk(bm); + bm->cur.bit = 0; } static void memory_bm_free(struct memory_bitmap *bm, int clear_nosave_free); @@ -394,12 +387,10 @@ memory_bm_create(struct memory_bitmap *bm, gfp_t gfp_mask, int safe_needed) bb->start_pfn = pfn; if (nr >= BM_BITS_PER_BLOCK) { pfn += BM_BITS_PER_BLOCK; - bb->size = BM_CHUNKS_PER_BLOCK; nr -= BM_BITS_PER_BLOCK; } else { /* This is executed only once in the loop */ pfn += nr; - bb->size = DIV_ROUND_UP(nr, BM_BITS_PER_CHUNK); } bb->end_pfn = pfn; bb = bb->next; @@ -478,8 +469,8 @@ static int memory_bm_find_bit(struct memory_bitmap *bm, unsigned long pfn, } zone_bm->cur_block = bb; pfn -= bb->start_pfn; - *bit_nr = pfn % BM_BITS_PER_CHUNK; - *addr = bb->data + pfn / BM_BITS_PER_CHUNK; + *bit_nr = pfn; + *addr = bb->data; return 0; } @@ -528,36 +519,6 @@ static int memory_bm_test_bit(struct memory_bitmap *bm, unsigned long pfn) return test_bit(bit, addr); } -/* Two auxiliary functions for memory_bm_next_pfn */ - -/* Find the first set bit in the given chunk, if there is one */ - -static inline int next_bit_in_chunk(int bit, unsigned long *chunk_p) -{ - bit++; - while (bit < BM_BITS_PER_CHUNK) { - if (test_bit(bit, chunk_p)) - return bit; - - bit++; - } - return -1; -} - -/* Find a chunk containing some bits set in given block of bits */ - -static inline int next_chunk_in_block(int n, struct bm_block *bb) -{ - n++; - while (n < bb->size) { - if (bb->data[n]) - return n; - - n++; - } - return -1; -} - /** * memory_bm_next_pfn - find the pfn that corresponds to the next set bit * in the bitmap @bm. If the pfn cannot be found, BM_END_OF_MAP is @@ -571,40 +532,33 @@ static unsigned long memory_bm_next_pfn(struct memory_bitmap *bm) { struct zone_bitmap *zone_bm; struct bm_block *bb; - int chunk; int bit; do { bb = bm->cur.block; do { - chunk = bm->cur.chunk; bit = bm->cur.bit; - do { - bit = next_bit_in_chunk(bit, bb->data + chunk); - if (bit >= 0) - goto Return_pfn; + bit = find_next_bit(bb->data, bm_block_bits(bb), bit); + if (bit < bm_block_bits(bb)) + goto Return_pfn; - chunk = next_chunk_in_block(chunk, bb); - bit = -1; - } while (chunk >= 0); bb = bb->next; bm->cur.block = bb; - memory_bm_reset_chunk(bm); + bm->cur.bit = 0; } while (bb); zone_bm = bm->cur.zone_bm->next; if (zone_bm) { bm->cur.zone_bm = zone_bm; bm->cur.block = zone_bm->bm_blocks; - memory_bm_reset_chunk(bm); + bm->cur.bit = 0; } } while (zone_bm); memory_bm_position_reset(bm); return BM_END_OF_MAP; Return_pfn: - bm->cur.chunk = chunk; - bm->cur.bit = bit; - return bb->start_pfn + chunk * BM_BITS_PER_CHUNK + bit; + bm->cur.bit = bit + 1; + return bb->start_pfn + bit; } /** From c1a220e7acf8ad2c03504891f4a70cd9c32c904b Mon Sep 17 00:00:00 2001 From: Zhang Rui Date: Wed, 23 Jul 2008 21:28:39 -0700 Subject: [PATCH 233/857] pm: introduce new interfaces schedule_work_on() and queue_work_on() This interface allows adding a job on a specific cpu. Although a work struct on a cpu will be scheduled to other cpu if the cpu dies, there is a recursion if a work task tries to offline the cpu it's running on. we need to schedule the task to a specific cpu in this case. http://bugzilla.kernel.org/show_bug.cgi?id=10897 [oleg@tv-sign.ru: cleanups] Signed-off-by: Zhang Rui Tested-by: Rus Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Signed-off-by: Oleg Nesterov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/workqueue.h | 3 +++ kernel/workqueue.c | 39 ++++++++++++++++++++++++++++++++++++++- 2 files changed, 41 insertions(+), 1 deletion(-) diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h index 542526c6e8ef..14d47120682b 100644 --- a/include/linux/workqueue.h +++ b/include/linux/workqueue.h @@ -179,6 +179,8 @@ __create_workqueue_key(const char *name, int singlethread, extern void destroy_workqueue(struct workqueue_struct *wq); extern int queue_work(struct workqueue_struct *wq, struct work_struct *work); +extern int queue_work_on(int cpu, struct workqueue_struct *wq, + struct work_struct *work); extern int queue_delayed_work(struct workqueue_struct *wq, struct delayed_work *work, unsigned long delay); extern int queue_delayed_work_on(int cpu, struct workqueue_struct *wq, @@ -188,6 +190,7 @@ extern void flush_workqueue(struct workqueue_struct *wq); extern void flush_scheduled_work(void); extern int schedule_work(struct work_struct *work); +extern int schedule_work_on(int cpu, struct work_struct *work); extern int schedule_delayed_work(struct delayed_work *work, unsigned long delay); extern int schedule_delayed_work_on(int cpu, struct delayed_work *work, unsigned long delay); diff --git a/kernel/workqueue.c b/kernel/workqueue.c index a6d36346d10a..6fd158b21026 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -140,7 +140,6 @@ static void insert_work(struct cpu_workqueue_struct *cwq, wake_up(&cwq->more_work); } -/* Preempt must be disabled. */ static void __queue_work(struct cpu_workqueue_struct *cwq, struct work_struct *work) { @@ -175,6 +174,31 @@ int queue_work(struct workqueue_struct *wq, struct work_struct *work) } EXPORT_SYMBOL_GPL(queue_work); +/** + * queue_work_on - queue work on specific cpu + * @cpu: CPU number to execute work on + * @wq: workqueue to use + * @work: work to queue + * + * Returns 0 if @work was already on a queue, non-zero otherwise. + * + * We queue the work to a specific CPU, the caller must ensure it + * can't go away. + */ +int +queue_work_on(int cpu, struct workqueue_struct *wq, struct work_struct *work) +{ + int ret = 0; + + if (!test_and_set_bit(WORK_STRUCT_PENDING, work_data_bits(work))) { + BUG_ON(!list_empty(&work->entry)); + __queue_work(wq_per_cpu(wq, cpu), work); + ret = 1; + } + return ret; +} +EXPORT_SYMBOL_GPL(queue_work_on); + static void delayed_work_timer_fn(unsigned long __data) { struct delayed_work *dwork = (struct delayed_work *)__data; @@ -553,6 +577,19 @@ int schedule_work(struct work_struct *work) } EXPORT_SYMBOL(schedule_work); +/* + * schedule_work_on - put work task on a specific cpu + * @cpu: cpu to put the work task on + * @work: job to be done + * + * This puts a job on a specific cpu + */ +int schedule_work_on(int cpu, struct work_struct *work) +{ + return queue_work_on(cpu, keventd_wq, work); +} +EXPORT_SYMBOL(schedule_work_on); + /** * schedule_delayed_work - put work task in global workqueue after delay * @dwork: job to be done From 2f15fc4bdf91eb399da3f47a09c55831d9f22826 Mon Sep 17 00:00:00 2001 From: Zhang Rui Date: Wed, 23 Jul 2008 21:28:40 -0700 Subject: [PATCH 234/857] pm: schedule sysrq poweroff on boot cpu schedule sysrq poweroff on boot cpu. sysrq poweroff needs to disable nonboot cpus, and we need to run this on boot cpu to avoid any recursion. http://bugzilla.kernel.org/show_bug.cgi?id=10897 [kosaki.motohiro@jp.fujitsu.com: build fix] Signed-off-by: Zhang Rui Tested-by: Rus Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Signed-off-by: KOSAKI Motohiro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/power/poweroff.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/kernel/power/poweroff.c b/kernel/power/poweroff.c index 678ec736076b..72016f051477 100644 --- a/kernel/power/poweroff.c +++ b/kernel/power/poweroff.c @@ -10,6 +10,7 @@ #include #include #include +#include /* * When the user hits Sys-Rq o to power down the machine this is the @@ -25,7 +26,8 @@ static DECLARE_WORK(poweroff_work, do_poweroff); static void handle_poweroff(int key, struct tty_struct *tty) { - schedule_work(&poweroff_work); + /* run sysrq poweroff on boot cpu */ + schedule_work_on(first_cpu(cpu_online_map), &poweroff_work); } static struct sysrq_key_op sysrq_poweroff_op = { From bdfe6b7c681669148dae4db27eb24ee5408ba371 Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Wed, 23 Jul 2008 21:28:41 -0700 Subject: [PATCH 235/857] pm: acpi hibernation: utilize hardware signature ACPI defines a hardware signature. BIOS calculates the signature according to hardware configure and if hardware changes while hibernated, the signature will change. In that case, S4 resume should fail. Still, there may be systems on which this mechanism does not work correctly, so it is better to provide a workaround for them. For this reason, add a new switch to the acpi_sleep= command line argument allowing one to disable hardware signature checking. [shaohua.li@intel.com: build fix] Signed-off-by: Shaohua Li Signed-off-by: Rafael J. Wysocki Cc: Andi Kleen Cc: Len Brown Acked-by: Pavel Machek Cc: Cc: Shaohua Li Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/kernel-parameters.txt | 4 +++- arch/x86/kernel/acpi/sleep.c | 4 ++++ drivers/acpi/sleep/main.c | 22 ++++++++++++++++++++++ include/linux/acpi.h | 1 + 4 files changed, 30 insertions(+), 1 deletion(-) diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 4d705713cabc..497a98dafdaa 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -148,10 +148,12 @@ and is between 256 and 4096 characters. It is defined in the file default: 0 acpi_sleep= [HW,ACPI] Sleep options - Format: { s3_bios, s3_mode, s3_beep, old_ordering } + Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig, old_ordering } See Documentation/power/video.txt for s3_bios and s3_mode. s3_beep is for debugging; it makes the PC's speaker beep as soon as the kernel's real-mode entry point is called. + s4_nohwsig prevents ACPI hardware signature from being + used during resume from hibernation. old_ordering causes the ACPI 1.0 ordering of the _PTS control method, wrt putting devices into low power states, to be enforced (the ACPI 2.0 ordering of _PTS is diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c index a3ddad18aaa3..fa2161d5003b 100644 --- a/arch/x86/kernel/acpi/sleep.c +++ b/arch/x86/kernel/acpi/sleep.c @@ -150,6 +150,10 @@ static int __init acpi_sleep_setup(char *str) acpi_realmode_flags |= 2; if (strncmp(str, "s3_beep", 7) == 0) acpi_realmode_flags |= 4; +#ifdef CONFIG_HIBERNATION + if (strncmp(str, "s4_nohwsig", 10) == 0) + acpi_no_s4_hw_signature(); +#endif if (strncmp(str, "old_ordering", 12) == 0) acpi_old_suspend_ordering(); str = strchr(str, ','); diff --git a/drivers/acpi/sleep/main.c b/drivers/acpi/sleep/main.c index 0489a7d1d42c..313507accf18 100644 --- a/drivers/acpi/sleep/main.c +++ b/drivers/acpi/sleep/main.c @@ -283,6 +283,15 @@ static struct platform_suspend_ops acpi_suspend_ops_old = { #endif /* CONFIG_SUSPEND */ #ifdef CONFIG_HIBERNATION +static unsigned long s4_hardware_signature; +static struct acpi_table_facs *facs; +static bool nosigcheck; + +void __init acpi_no_s4_hw_signature(void) +{ + nosigcheck = true; +} + static int acpi_hibernation_begin(void) { acpi_target_sleep_state = ACPI_STATE_S4; @@ -316,6 +325,12 @@ static void acpi_hibernation_leave(void) acpi_enable(); /* Reprogram control registers and execute _BFS */ acpi_leave_sleep_state_prep(ACPI_STATE_S4); + /* Check the hardware signature */ + if (facs && s4_hardware_signature != facs->hardware_signature) { + printk(KERN_EMERG "ACPI: Hardware changed while hibernated, " + "cannot resume!\n"); + panic("ACPI S4 hardware signature mismatch"); + } } static void acpi_pm_enable_gpes(void) @@ -544,6 +559,13 @@ int __init acpi_sleep_init(void) &acpi_hibernation_ops_old : &acpi_hibernation_ops); sleep_states[ACPI_STATE_S4] = 1; printk(" S4"); + if (!nosigcheck) { + acpi_get_table_by_index(ACPI_TABLE_INDEX_FACS, + (struct acpi_table_header **)&facs); + if (facs) + s4_hardware_signature = + facs->hardware_signature; + } } #endif status = acpi_get_sleep_type_data(ACPI_STATE_S5, &type_a, &type_b); diff --git a/include/linux/acpi.h b/include/linux/acpi.h index a17177639376..702f79dad16a 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -236,6 +236,7 @@ int acpi_check_mem_region(resource_size_t start, resource_size_t n, const char *name); #ifdef CONFIG_PM_SLEEP +void __init acpi_no_s4_hw_signature(void); void __init acpi_old_suspend_ordering(void); #endif /* CONFIG_PM_SLEEP */ #else /* CONFIG_ACPI */ From e41fb7c58e3ca18ec5c9c9bb7bb68e8e653c9e8e Mon Sep 17 00:00:00 2001 From: Carlos Corbacho Date: Wed, 23 Jul 2008 21:28:43 -0700 Subject: [PATCH 236/857] pm: acpi pm: add DMI quirk list for ACPI 1.0 suspend ordering There are a few BIOSes that we know of already that need to use the ACPI 1.0 suspend order. This appears to be only be a small minority of mostly nVidia based systems. Based on observation of Windows behaviour, it's clear that Windows is also doing maintaining its own list of broken hardware that needs this workaround. Signed-off-by: Carlos Corbacho Signed-off-by: Rafael J. Wysocki Cc: Andi Kleen Cc: Len Brown Acked-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/acpi/sleep/main.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) diff --git a/drivers/acpi/sleep/main.c b/drivers/acpi/sleep/main.c index 313507accf18..d13194a031bf 100644 --- a/drivers/acpi/sleep/main.c +++ b/drivers/acpi/sleep/main.c @@ -280,6 +280,24 @@ static struct platform_suspend_ops acpi_suspend_ops_old = { .end = acpi_pm_end, .recover = acpi_pm_finish, }; + +static int __init init_old_suspend_ordering(const struct dmi_system_id *d) +{ + old_suspend_ordering = true; + return 0; +} + +static struct dmi_system_id __initdata acpisleep_dmi_table[] = { + { + .callback = init_old_suspend_ordering, + .ident = "Abit KN9 (nForce4 variant)", + .matches = { + DMI_MATCH(DMI_BOARD_VENDOR, "http://www.abit.com.tw/"), + DMI_MATCH(DMI_BOARD_NAME, "KN9 Series(NF-CK804)"), + }, + }, + {}, +}; #endif /* CONFIG_SUSPEND */ #ifdef CONFIG_HIBERNATION @@ -531,6 +549,8 @@ int __init acpi_sleep_init(void) u8 type_a, type_b; #ifdef CONFIG_SUSPEND int i = 0; + + dmi_check_system(acpisleep_dmi_table); #endif if (acpi_disabled) From f0af566da6e9a4a2f5a83c5a70f3d0a772050e21 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 23 Jul 2008 21:28:44 -0700 Subject: [PATCH 237/857] pm: fix try_to_freeze_tasks()'s use of do_div() Fix try_to_freeze_tasks()'s use of do_div() on an s64 by making elapsed_csecs64 a u64 instead and dividing that. Possibly this should be guarded lest the interval calculation turn up negative, but the possible negativity of the result of the division is cast away anyway. This was introduced by patch 438e2ce68dfd4af4cfcec2f873564fb921db4bb5. Signed-off-by: David Howells Acked-by: "Rafael J. Wysocki" Acked-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/power/process.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/power/process.c b/kernel/power/process.c index 5fb87652f214..278946aecaf0 100644 --- a/kernel/power/process.c +++ b/kernel/power/process.c @@ -149,7 +149,7 @@ static int try_to_freeze_tasks(bool sig_only) unsigned long end_time; unsigned int todo; struct timeval start, end; - s64 elapsed_csecs64; + u64 elapsed_csecs64; unsigned int elapsed_csecs; do_gettimeofday(&start); From 912019572180f287e85b5534fbb1c1e3ca6df6c9 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 23 Jul 2008 21:28:45 -0700 Subject: [PATCH 238/857] mn10300: move sg_dma_{address,len}() to asm/scatterlist.h mn10300 was the only architecture where sg_dma_{address,len}() were not in asm/scatterlist.h, and it's not a big surprise that this caused a compile error somewhere: /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/media/video/videobuf-dma-sg.c: In function `videobuf_dma_map': /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/media/video/videobuf-dma-sg.c:238: error: implicit declaration of function 'sg_dma_address' Acked-by: David Howells Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-mn10300/pci.h | 9 --------- include/asm-mn10300/scatterlist.h | 9 +++++++++ 2 files changed, 9 insertions(+), 9 deletions(-) diff --git a/include/asm-mn10300/pci.h b/include/asm-mn10300/pci.h index 205192c52bb5..cd9cc5c89cea 100644 --- a/include/asm-mn10300/pci.h +++ b/include/asm-mn10300/pci.h @@ -74,15 +74,6 @@ struct pci_dev; /* This is always fine. */ #define pci_dac_dma_supported(pci_dev, mask) (0) -/* - * These macros should be used after a pci_map_sg call has been done - * to get bus addresses of each of the SG entries and their lengths. - * You should only work with the number of sg entries pci_map_sg - * returns. - */ -#define sg_dma_address(sg) ((sg)->dma_address) -#define sg_dma_len(sg) ((sg)->length) - /* Return the index of the PCI controller for device. */ static inline int pci_controller_num(struct pci_dev *dev) { diff --git a/include/asm-mn10300/scatterlist.h b/include/asm-mn10300/scatterlist.h index e29d91dbcf2b..67535901b9ff 100644 --- a/include/asm-mn10300/scatterlist.h +++ b/include/asm-mn10300/scatterlist.h @@ -43,4 +43,13 @@ struct scatterlist { #define ISA_DMA_THRESHOLD (0x00ffffff) +/* + * These macros should be used after a pci_map_sg call has been done + * to get bus addresses of each of the SG entries and their lengths. + * You should only work with the number of sg entries pci_map_sg + * returns. + */ +#define sg_dma_address(sg) ((sg)->dma_address) +#define sg_dma_len(sg) ((sg)->length) + #endif /* _ASM_SCATTERLIST_H */ From d50004b0867a59f8a81116f000edb352595343d9 Mon Sep 17 00:00:00 2001 From: Fernando Luis Vazquez Cao Date: Wed, 23 Jul 2008 21:28:45 -0700 Subject: [PATCH 239/857] cris: remove unused global_flush_tlb global_flush_tlb is declared but never used. Signed-off-by: Fernando Luis Vazquez Cao Cc: Mikael Starvik Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-cris/cacheflush.h | 1 - 1 file changed, 1 deletion(-) diff --git a/include/asm-cris/cacheflush.h b/include/asm-cris/cacheflush.h index 01af2de27c5b..cf60e3f69f8d 100644 --- a/include/asm-cris/cacheflush.h +++ b/include/asm-cris/cacheflush.h @@ -26,7 +26,6 @@ #define copy_from_user_page(vma, page, vaddr, dst, src, len) \ memcpy(dst, src, len) -void global_flush_tlb(void); int change_page_attr(struct page *page, int numpages, pgprot_t prot); #endif /* _CRIS_CACHEFLUSH_H */ From ed62f77bb631bc4a2d8acb0521b720cb55e58183 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Wed, 23 Jul 2008 21:28:46 -0700 Subject: [PATCH 240/857] cris: use simple_read_from_buffer() Signed-off-by: Akinobu Mita Cc: Mikael Starvik Cc: Jesper Nilsson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/cris/kernel/profile.c | 17 +++++++---------- 1 file changed, 7 insertions(+), 10 deletions(-) diff --git a/arch/cris/kernel/profile.c b/arch/cris/kernel/profile.c index 44f7b4f79476..9aa571169bcc 100644 --- a/arch/cris/kernel/profile.c +++ b/arch/cris/kernel/profile.c @@ -35,19 +35,16 @@ read_cris_profile(struct file *file, char __user *buf, size_t count, loff_t *ppos) { unsigned long p = *ppos; + ssize_t ret; - if (p > SAMPLE_BUFFER_SIZE) - return 0; + ret = simple_read_from_buffer(buf, count, ppos, sample_buffer, + SAMPLE_BUFFER_SIZE); + if (ret < 0) + return ret; - if (p + count > SAMPLE_BUFFER_SIZE) - count = SAMPLE_BUFFER_SIZE - p; - if (copy_to_user(buf, sample_buffer + p,count)) - return -EFAULT; + memset(sample_buffer + p, 0, ret); - memset(sample_buffer + p, 0, count); - *ppos += count; - - return count; + return ret; } static ssize_t From 4c182ae7810f3fe444e666f3f78c209a7c116fdf Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Wed, 23 Jul 2008 21:28:47 -0700 Subject: [PATCH 241/857] arch/um/kernel/irq.c: clean up some functions Make activate_fd() and free_irq_by_irq_and_dev() static. Remove init_aio_irq() since it has no users. Cc: Jeff Dike Signed-off-by: WANG Cong Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/include/irq_kern.h | 2 -- arch/um/include/irq_user.h | 2 -- arch/um/kernel/irq.c | 35 ++--------------------------------- 3 files changed, 2 insertions(+), 37 deletions(-) diff --git a/arch/um/include/irq_kern.h b/arch/um/include/irq_kern.h index 4f775597fd5f..fba3895274f9 100644 --- a/arch/um/include/irq_kern.h +++ b/arch/um/include/irq_kern.h @@ -13,8 +13,6 @@ extern int um_request_irq(unsigned int irq, int fd, int type, irq_handler_t handler, unsigned long irqflags, const char * devname, void *dev_id); -extern int init_aio_irq(int irq, char *name, - irq_handler_t handler); #endif diff --git a/arch/um/include/irq_user.h b/arch/um/include/irq_user.h index e60b31873de1..c6c784df2673 100644 --- a/arch/um/include/irq_user.h +++ b/arch/um/include/irq_user.h @@ -21,8 +21,6 @@ struct irq_fd { enum { IRQ_READ, IRQ_WRITE }; extern void sigio_handler(int sig, struct uml_pt_regs *regs); -extern int activate_fd(int irq, int fd, int type, void *dev_id); -extern void free_irq_by_irq_and_dev(unsigned int irq, void *dev_id); extern void free_irq_by_fd(int fd); extern void reactivate_fd(int fd, int irqnum); extern void deactivate_fd(int fd, int irqnum); diff --git a/arch/um/kernel/irq.c b/arch/um/kernel/irq.c index 91587f8db340..3d7aad09b171 100644 --- a/arch/um/kernel/irq.c +++ b/arch/um/kernel/irq.c @@ -102,7 +102,7 @@ void sigio_handler(int sig, struct uml_pt_regs *regs) static DEFINE_SPINLOCK(irq_lock); -int activate_fd(int irq, int fd, int type, void *dev_id) +static int activate_fd(int irq, int fd, int type, void *dev_id) { struct pollfd *tmp_pfd; struct irq_fd *new_fd, *irq_fd; @@ -216,7 +216,7 @@ static int same_irq_and_dev(struct irq_fd *irq, void *d) return ((irq->irq == data->irq) && (irq->id == data->dev)); } -void free_irq_by_irq_and_dev(unsigned int irq, void *dev) +static void free_irq_by_irq_and_dev(unsigned int irq, void *dev) { struct irq_and_dev data = ((struct irq_and_dev) { .irq = irq, .dev = dev }); @@ -403,37 +403,6 @@ void __init init_IRQ(void) } } -int init_aio_irq(int irq, char *name, irq_handler_t handler) -{ - int fds[2], err; - - err = os_pipe(fds, 1, 1); - if (err) { - printk(KERN_ERR "init_aio_irq - os_pipe failed, err = %d\n", - -err); - goto out; - } - - err = um_request_irq(irq, fds[0], IRQ_READ, handler, - IRQF_DISABLED | IRQF_SAMPLE_RANDOM, name, - (void *) (long) fds[0]); - if (err) { - printk(KERN_ERR "init_aio_irq - : um_request_irq failed, " - "err = %d\n", - err); - goto out_close; - } - - err = fds[1]; - goto out; - - out_close: - os_close_file(fds[0]); - os_close_file(fds[1]); - out: - return err; -} - /* * IRQ stack entry and exit: * From 4a5675820436e4ad738dd442c1cc8a165101509b Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Wed, 23 Jul 2008 21:28:49 -0700 Subject: [PATCH 242/857] arch/um/kernel/mem.c: remove arch_validate() - Remove arch_validate(), because no one uses it. - Remove useless macro HAVE_ARCH_VALIDATE. - Make the variable 'empty_bad_page' static. Cc: Jeff Dike Signed-off-by: WANG Cong Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/kernel/ksyms.c | 1 - arch/um/kernel/mem.c | 33 +-------------------------------- include/asm-um/page.h | 3 --- 3 files changed, 1 insertion(+), 36 deletions(-) diff --git a/arch/um/kernel/ksyms.c b/arch/um/kernel/ksyms.c index ccc02a616c22..836fc9b94707 100644 --- a/arch/um/kernel/ksyms.c +++ b/arch/um/kernel/ksyms.c @@ -18,7 +18,6 @@ EXPORT_SYMBOL(get_signals); EXPORT_SYMBOL(kernel_thread); EXPORT_SYMBOL(sys_waitpid); EXPORT_SYMBOL(flush_tlb_range); -EXPORT_SYMBOL(arch_validate); EXPORT_SYMBOL(high_physmem); EXPORT_SYMBOL(empty_zero_page); diff --git a/arch/um/kernel/mem.c b/arch/um/kernel/mem.c index b0ee64622ff7..e2274ef3155d 100644 --- a/arch/um/kernel/mem.c +++ b/arch/um/kernel/mem.c @@ -21,7 +21,7 @@ /* allocated in paging_init, zeroed in mem_init, and unchanged thereafter */ unsigned long *empty_zero_page = NULL; /* allocated in paging_init and unchanged thereafter */ -unsigned long *empty_bad_page = NULL; +static unsigned long *empty_bad_page = NULL; /* * Initialized during boot, and readonly for initializing page tables @@ -240,37 +240,6 @@ void __init paging_init(void) #endif } -struct page *arch_validate(struct page *page, gfp_t mask, int order) -{ - unsigned long addr, zero = 0; - int i; - - again: - if (page == NULL) - return page; - if (PageHighMem(page)) - return page; - - addr = (unsigned long) page_address(page); - for (i = 0; i < (1 << order); i++) { - current->thread.fault_addr = (void *) addr; - if (__do_copy_to_user((void __user *) addr, &zero, - sizeof(zero), - ¤t->thread.fault_addr, - ¤t->thread.fault_catcher)) { - if (!(mask & __GFP_WAIT)) - return NULL; - else break; - } - addr += PAGE_SIZE; - } - - if (i == (1 << order)) - return page; - page = alloc_pages(mask, order); - goto again; -} - /* * This can't do anything because nothing in the kernel image can be freed * since it's not in kernel physical memory. diff --git a/include/asm-um/page.h b/include/asm-um/page.h index 335c57383c02..a6df1f13d732 100644 --- a/include/asm-um/page.h +++ b/include/asm-um/page.h @@ -115,9 +115,6 @@ extern unsigned long uml_physmem; #define pfn_valid(pfn) ((pfn) < max_mapnr) #define virt_addr_valid(v) pfn_valid(phys_to_pfn(__pa(v))) -extern struct page *arch_validate(struct page *page, gfp_t mask, int order); -#define HAVE_ARCH_VALIDATE - #include #include From 99764fa4ceeecba8b9e0a8a5565b418a2e94f83b Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Wed, 23 Jul 2008 21:28:49 -0700 Subject: [PATCH 243/857] UML: make several more things static - Make some variables and functions static, since they don't need to be global. - Remove an unused function - arch/um/kernel/time.c::sched_clock(). - Clean the style a bit as complained by checkpatch.pl. Cc: Jeff Dike Signed-off-by: WANG Cong Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/include/skas/skas.h | 1 - arch/um/include/um_uaccess.h | 1 - arch/um/kernel/physmem.c | 2 +- arch/um/kernel/ptrace.c | 2 +- arch/um/kernel/time.c | 8 -------- arch/um/kernel/uaccess.c | 2 +- arch/um/os-Linux/sigio.c | 2 +- arch/um/os-Linux/signal.c | 2 +- arch/um/os-Linux/skas/process.c | 2 +- arch/um/os-Linux/umid.c | 2 +- arch/um/sys-i386/bugs.c | 2 +- arch/um/sys-i386/checksum.S | 5 ++--- arch/um/sys-i386/ldt.c | 4 ++-- include/asm-um/ptrace-generic.h | 3 --- 14 files changed, 12 insertions(+), 26 deletions(-) diff --git a/arch/um/include/skas/skas.h b/arch/um/include/skas/skas.h index b073f8a86bd3..64d2c7443306 100644 --- a/arch/um/include/skas/skas.h +++ b/arch/um/include/skas/skas.h @@ -16,7 +16,6 @@ extern int user_thread(unsigned long stack, int flags); extern void new_thread_handler(void); extern void handle_syscall(struct uml_pt_regs *regs); extern int new_mm(unsigned long stack); -extern void get_skas_faultinfo(int pid, struct faultinfo * fi); extern long execute_syscall_skas(void *r); extern unsigned long current_stub_stack(void); diff --git a/arch/um/include/um_uaccess.h b/arch/um/include/um_uaccess.h index 2b6fc8e0f071..45c04999d670 100644 --- a/arch/um/include/um_uaccess.h +++ b/arch/um/include/um_uaccess.h @@ -34,7 +34,6 @@ extern int copy_to_user(void __user *to, const void *from, int n); extern int __do_copy_to_user(void *to, const void *from, int n, void **fault_addr, jmp_buf **fault_catcher); -extern void __do_copy(void *to, const void *from, int n); /* * strncpy_from_user: - Copy a NUL terminated string from userspace. diff --git a/arch/um/kernel/physmem.c b/arch/um/kernel/physmem.c index 9757085a0220..a1a9090254c2 100644 --- a/arch/um/kernel/physmem.c +++ b/arch/um/kernel/physmem.c @@ -185,7 +185,7 @@ unsigned long find_iomem(char *driver, unsigned long *len_out) return 0; } -int setup_iomem(void) +static int setup_iomem(void) { struct iomem_region *region = iomem_regions; unsigned long iomem_start = high_physmem + PAGE_SIZE; diff --git a/arch/um/kernel/ptrace.c b/arch/um/kernel/ptrace.c index 47b57b497d55..15e8b7c4de13 100644 --- a/arch/um/kernel/ptrace.c +++ b/arch/um/kernel/ptrace.c @@ -225,7 +225,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) return ret; } -void send_sigtrap(struct task_struct *tsk, struct uml_pt_regs *regs, +static void send_sigtrap(struct task_struct *tsk, struct uml_pt_regs *regs, int error_code) { struct siginfo info; diff --git a/arch/um/kernel/time.c b/arch/um/kernel/time.c index c3e2f369c33c..47f04f4a3464 100644 --- a/arch/um/kernel/time.c +++ b/arch/um/kernel/time.c @@ -13,14 +13,6 @@ #include "kern_util.h" #include "os.h" -/* - * Scheduler clock - returns current time in nanosec units. - */ -unsigned long long sched_clock(void) -{ - return (unsigned long long)jiffies_64 * (NSEC_PER_SEC / HZ); -} - void timer_handler(int sig, struct uml_pt_regs *regs) { unsigned long flags; diff --git a/arch/um/kernel/uaccess.c b/arch/um/kernel/uaccess.c index f0f4b040d7c5..dd33f040c526 100644 --- a/arch/um/kernel/uaccess.c +++ b/arch/um/kernel/uaccess.c @@ -12,7 +12,7 @@ #include #include "os.h" -void __do_copy(void *to, const void *from, int n) +static void __do_copy(void *to, const void *from, int n) { memcpy(to, from, n); } diff --git a/arch/um/os-Linux/sigio.c b/arch/um/os-Linux/sigio.c index eb8f2e4be192..63d299df152b 100644 --- a/arch/um/os-Linux/sigio.c +++ b/arch/um/os-Linux/sigio.c @@ -530,7 +530,7 @@ static void tty_close(int master, int slave) printk(UM_KERN_CONT "No, enabling workaround\n"); } -void __init check_sigio(void) +static void __init check_sigio(void) { if ((access("/dev/ptmx", R_OK) < 0) && (access("/dev/ptyp0", R_OK) < 0)) { diff --git a/arch/um/os-Linux/signal.c b/arch/um/os-Linux/signal.c index 5aade6027e40..6ae180703a63 100644 --- a/arch/um/os-Linux/signal.c +++ b/arch/um/os-Linux/signal.c @@ -126,7 +126,7 @@ void set_sigstack(void *sig_stack, int size) panic("enabling signal stack failed, errno = %d\n", errno); } -void (*handlers[_NSIG])(int sig, struct sigcontext *sc); +static void (*handlers[_NSIG])(int sig, struct sigcontext *sc); void handle_signal(int sig, struct sigcontext *sc) { diff --git a/arch/um/os-Linux/skas/process.c b/arch/um/os-Linux/skas/process.c index 172ad8f72e12..d6e0a2234b86 100644 --- a/arch/um/os-Linux/skas/process.c +++ b/arch/um/os-Linux/skas/process.c @@ -96,7 +96,7 @@ void wait_stub_done(int pid) extern unsigned long current_stub_stack(void); -void get_skas_faultinfo(int pid, struct faultinfo * fi) +static void get_skas_faultinfo(int pid, struct faultinfo *fi) { int err; diff --git a/arch/um/os-Linux/umid.c b/arch/um/os-Linux/umid.c index 106fa8641553..a27defb81884 100644 --- a/arch/um/os-Linux/umid.c +++ b/arch/um/os-Linux/umid.c @@ -245,7 +245,7 @@ int __init set_umid(char *name) /* Changed in make_umid, which is called during early boot */ static int umid_setup = 0; -int __init make_umid(void) +static int __init make_umid(void) { int fd, err; char tmp[256]; diff --git a/arch/um/sys-i386/bugs.c b/arch/um/sys-i386/bugs.c index a74442d13762..2c6d0d731c12 100644 --- a/arch/um/sys-i386/bugs.c +++ b/arch/um/sys-i386/bugs.c @@ -12,7 +12,7 @@ #include "sysdep/ptrace.h" /* Set during early boot */ -int host_has_cmov = 1; +static int host_has_cmov = 1; static jmp_buf cmov_test_return; static void cmov_sigill_test_handler(int sig) diff --git a/arch/um/sys-i386/checksum.S b/arch/um/sys-i386/checksum.S index 62c7e564f22e..f058d2f82e18 100644 --- a/arch/um/sys-i386/checksum.S +++ b/arch/um/sys-i386/checksum.S @@ -243,13 +243,12 @@ unsigned int csum_partial_copy_generic (const char *src, char *dst, .previous .align 4 -.globl csum_partial_copy_generic_i386 - + #ifndef CONFIG_X86_USE_PPRO_CHECKSUM #define ARGBASE 16 #define FP 12 - + csum_partial_copy_generic_i386: subl $4,%esp pushl %edi diff --git a/arch/um/sys-i386/ldt.c b/arch/um/sys-i386/ldt.c index a34263e6b08d..a4846a84a7be 100644 --- a/arch/um/sys-i386/ldt.c +++ b/arch/um/sys-i386/ldt.c @@ -14,8 +14,8 @@ extern int modify_ldt(int func, void *ptr, unsigned long bytecount); -long write_ldt_entry(struct mm_id * mm_idp, int func, struct user_desc * desc, - void **addr, int done) +static long write_ldt_entry(struct mm_id *mm_idp, int func, + struct user_desc *desc, void **addr, int done) { long res; diff --git a/include/asm-um/ptrace-generic.h b/include/asm-um/ptrace-generic.h index 6aefcd32fc61..315749705ea1 100644 --- a/include/asm-um/ptrace-generic.h +++ b/include/asm-um/ptrace-generic.h @@ -47,9 +47,6 @@ extern int set_fpregs(struct user_i387_struct __user *buf, extern void show_regs(struct pt_regs *regs); -extern void send_sigtrap(struct task_struct *tsk, struct uml_pt_regs *regs, - int error_code); - extern int arch_copy_tls(struct task_struct *new); extern void clear_flushed_tls(struct task_struct *task); From f606ddf42fd4edc558eeb48bfee66d2c591571d2 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 23 Jul 2008 21:28:50 -0700 Subject: [PATCH 244/857] remove the v850 port Trying to compile the v850 port brings many compile errors, one of them exists since at least kernel 2.6.19. There also seems to be noone willing to bring this port back into a usable state. This patch therefore removes the v850 port. If anyone ever decides to revive the v850 port the code will still be available from older kernels, and it wouldn't be impossible for the port to reenter the kernel if it would become actively maintained again. Signed-off-by: Adrian Bunk Acked-by: Greg Ungerer Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- MAINTAINERS | 3 - arch/v850/Kconfig | 353 -------- arch/v850/Kconfig.debug | 10 - arch/v850/Makefile | 54 -- arch/v850/README | 44 - arch/v850/configs/rte-ma1-cb_defconfig | 617 ------------- arch/v850/configs/rte-me2-cb_defconfig | 462 ---------- arch/v850/configs/sim_defconfig | 451 ---------- arch/v850/kernel/Makefile | 40 - arch/v850/kernel/anna-rom.ld | 16 - arch/v850/kernel/anna.c | 202 ----- arch/v850/kernel/anna.ld | 20 - arch/v850/kernel/as85ep1-rom.ld | 21 - arch/v850/kernel/as85ep1.c | 234 ----- arch/v850/kernel/as85ep1.ld | 49 -- arch/v850/kernel/asm-offsets.c | 58 -- arch/v850/kernel/bug.c | 142 --- arch/v850/kernel/entry.S | 1121 ------------------------ arch/v850/kernel/fpga85e2c.c | 167 ---- arch/v850/kernel/fpga85e2c.ld | 62 -- arch/v850/kernel/gbus_int.c | 271 ------ arch/v850/kernel/head.S | 128 --- arch/v850/kernel/highres_timer.c | 132 --- arch/v850/kernel/init_task.c | 48 - arch/v850/kernel/intv.S | 87 -- arch/v850/kernel/irq.c | 123 --- arch/v850/kernel/ma.c | 69 -- arch/v850/kernel/mach.c | 17 - arch/v850/kernel/mach.h | 56 -- arch/v850/kernel/me2.c | 73 -- arch/v850/kernel/memcons.c | 135 --- arch/v850/kernel/module.c | 237 ----- arch/v850/kernel/process.c | 217 ----- arch/v850/kernel/procfs.c | 67 -- arch/v850/kernel/ptrace.c | 235 ----- arch/v850/kernel/rte_cb.c | 193 ---- arch/v850/kernel/rte_cb_leds.c | 137 --- arch/v850/kernel/rte_cb_multi.c | 121 --- arch/v850/kernel/rte_ma1_cb-rom.ld | 14 - arch/v850/kernel/rte_ma1_cb.c | 107 --- arch/v850/kernel/rte_ma1_cb.ld | 57 -- arch/v850/kernel/rte_mb_a_pci.c | 819 ----------------- arch/v850/kernel/rte_me2_cb.c | 298 ------- arch/v850/kernel/rte_me2_cb.ld | 30 - arch/v850/kernel/rte_nb85e_cb-multi.ld | 57 -- arch/v850/kernel/rte_nb85e_cb.c | 81 -- arch/v850/kernel/rte_nb85e_cb.ld | 22 - arch/v850/kernel/setup.c | 329 ------- arch/v850/kernel/signal.c | 523 ----------- arch/v850/kernel/sim.c | 172 ---- arch/v850/kernel/sim.ld | 13 - arch/v850/kernel/sim85e2.c | 195 ----- arch/v850/kernel/sim85e2.ld | 36 - arch/v850/kernel/simcons.c | 161 ---- arch/v850/kernel/syscalls.c | 196 ----- arch/v850/kernel/teg.c | 62 -- arch/v850/kernel/time.c | 106 --- arch/v850/kernel/v850_ksyms.c | 51 -- arch/v850/kernel/v850e2_cache.c | 127 --- arch/v850/kernel/v850e_cache.c | 174 ---- arch/v850/kernel/v850e_intc.c | 104 --- arch/v850/kernel/v850e_timer_d.c | 54 -- arch/v850/kernel/v850e_utils.c | 62 -- arch/v850/kernel/vmlinux.lds.S | 306 ------- arch/v850/lib/Makefile | 6 - arch/v850/lib/ashldi3.c | 62 -- arch/v850/lib/ashrdi3.c | 63 -- arch/v850/lib/checksum.c | 155 ---- arch/v850/lib/lshrdi3.c | 62 -- arch/v850/lib/memcpy.c | 92 -- arch/v850/lib/memset.c | 68 -- arch/v850/lib/muldi3.c | 61 -- arch/v850/lib/negdi2.c | 25 - drivers/serial/Kconfig | 16 - drivers/watchdog/Kconfig | 2 - drivers/watchdog/Makefile | 2 - include/asm-v850/Kbuild | 1 - include/asm-v850/a.out.h | 21 - include/asm-v850/anna.h | 137 --- include/asm-v850/as85ep1.h | 152 ---- include/asm-v850/asm.h | 32 - include/asm-v850/atomic.h | 131 --- include/asm-v850/auxvec.h | 4 - include/asm-v850/bitops.h | 161 ---- include/asm-v850/bug.h | 25 - include/asm-v850/bugs.h | 16 - include/asm-v850/byteorder.h | 48 - include/asm-v850/cache.h | 26 - include/asm-v850/cacheflush.h | 70 -- include/asm-v850/checksum.h | 112 --- include/asm-v850/clinkage.h | 26 - include/asm-v850/cputime.h | 6 - include/asm-v850/current.h | 47 - include/asm-v850/delay.h | 47 - include/asm-v850/device.h | 7 - include/asm-v850/div64.h | 1 - include/asm-v850/dma-mapping.h | 11 - include/asm-v850/dma.h | 18 - include/asm-v850/elf.h | 99 --- include/asm-v850/emergency-restart.h | 6 - include/asm-v850/entry.h | 113 --- include/asm-v850/errno.h | 6 - include/asm-v850/fb.h | 12 - include/asm-v850/fcntl.h | 11 - include/asm-v850/flat.h | 133 --- include/asm-v850/fpga85e2c.h | 82 -- include/asm-v850/futex.h | 6 - include/asm-v850/gbus_int.h | 97 -- include/asm-v850/hardirq.h | 28 - include/asm-v850/highres_timer.h | 44 - include/asm-v850/hw_irq.h | 4 - include/asm-v850/io.h | 142 --- include/asm-v850/ioctl.h | 1 - include/asm-v850/ioctls.h | 84 -- include/asm-v850/ipcbuf.h | 29 - include/asm-v850/irq.h | 55 -- include/asm-v850/irq_regs.h | 1 - include/asm-v850/kdebug.h | 1 - include/asm-v850/kmap_types.h | 19 - include/asm-v850/kvm.h | 6 - include/asm-v850/linkage.h | 8 - include/asm-v850/local.h | 6 - include/asm-v850/ma.h | 101 --- include/asm-v850/ma1.h | 50 -- include/asm-v850/machdep.h | 60 -- include/asm-v850/macrology.h | 17 - include/asm-v850/me2.h | 182 ---- include/asm-v850/mman.h | 15 - include/asm-v850/mmu.h | 11 - include/asm-v850/mmu_context.h | 13 - include/asm-v850/module.h | 62 -- include/asm-v850/msgbuf.h | 31 - include/asm-v850/mutex.h | 9 - include/asm-v850/page.h | 124 --- include/asm-v850/param.h | 33 - include/asm-v850/pci.h | 119 --- include/asm-v850/percpu.h | 14 - include/asm-v850/pgalloc.h | 22 - include/asm-v850/pgtable.h | 59 -- include/asm-v850/poll.h | 9 - include/asm-v850/posix_types.h | 72 -- include/asm-v850/processor.h | 120 --- include/asm-v850/ptrace.h | 121 --- include/asm-v850/resource.h | 6 - include/asm-v850/rte_cb.h | 78 -- include/asm-v850/rte_ma1_cb.h | 128 --- include/asm-v850/rte_mb_a_pci.h | 56 -- include/asm-v850/rte_me2_cb.h | 202 ----- include/asm-v850/rte_nb85e_cb.h | 111 --- include/asm-v850/scatterlist.h | 31 - include/asm-v850/sections.h | 6 - include/asm-v850/segment.h | 36 - include/asm-v850/semaphore.h | 1 - include/asm-v850/sembuf.h | 25 - include/asm-v850/serial.h | 56 -- include/asm-v850/setup.h | 6 - include/asm-v850/shmbuf.h | 42 - include/asm-v850/shmparam.h | 6 - include/asm-v850/sigcontext.h | 25 - include/asm-v850/siginfo.h | 6 - include/asm-v850/signal.h | 168 ---- include/asm-v850/sim.h | 47 - include/asm-v850/sim85e2.h | 69 -- include/asm-v850/sim85e2c.h | 26 - include/asm-v850/sim85e2s.h | 28 - include/asm-v850/simsyscall.h | 99 --- include/asm-v850/socket.h | 57 -- include/asm-v850/sockios.h | 13 - include/asm-v850/stat.h | 73 -- include/asm-v850/statfs.h | 6 - include/asm-v850/string.h | 25 - include/asm-v850/system.h | 123 --- include/asm-v850/teg.h | 101 --- include/asm-v850/termbits.h | 200 ----- include/asm-v850/termios.h | 90 -- include/asm-v850/thread_info.h | 129 --- include/asm-v850/timex.h | 18 - include/asm-v850/tlb.h | 21 - include/asm-v850/tlbflush.h | 64 -- include/asm-v850/topology.h | 6 - include/asm-v850/types.h | 36 - include/asm-v850/uaccess.h | 159 ---- include/asm-v850/ucontext.h | 14 - include/asm-v850/unaligned.h | 22 - include/asm-v850/unistd.h | 244 ------ include/asm-v850/user.h | 52 -- include/asm-v850/v850e.h | 21 - include/asm-v850/v850e2.h | 69 -- include/asm-v850/v850e2_cache.h | 75 -- include/asm-v850/v850e_cache.h | 48 - include/asm-v850/v850e_intc.h | 133 --- include/asm-v850/v850e_timer_c.h | 48 - include/asm-v850/v850e_timer_d.h | 62 -- include/asm-v850/v850e_uart.h | 76 -- include/asm-v850/v850e_uarta.h | 278 ------ include/asm-v850/v850e_uartb.h | 262 ------ include/asm-v850/v850e_utils.h | 35 - include/linux/audit.h | 1 - include/linux/module.h | 2 +- include/linux/serial_core.h | 3 - include/linux/syscalls.h | 2 +- scripts/genksyms/genksyms.c | 3 +- scripts/mod/file2alias.c | 2 +- scripts/mod/mk_elfconfig.c | 2 +- 204 files changed, 5 insertions(+), 18406 deletions(-) delete mode 100644 arch/v850/Kconfig delete mode 100644 arch/v850/Kconfig.debug delete mode 100644 arch/v850/Makefile delete mode 100644 arch/v850/README delete mode 100644 arch/v850/configs/rte-ma1-cb_defconfig delete mode 100644 arch/v850/configs/rte-me2-cb_defconfig delete mode 100644 arch/v850/configs/sim_defconfig delete mode 100644 arch/v850/kernel/Makefile delete mode 100644 arch/v850/kernel/anna-rom.ld delete mode 100644 arch/v850/kernel/anna.c delete mode 100644 arch/v850/kernel/anna.ld delete mode 100644 arch/v850/kernel/as85ep1-rom.ld delete mode 100644 arch/v850/kernel/as85ep1.c delete mode 100644 arch/v850/kernel/as85ep1.ld delete mode 100644 arch/v850/kernel/asm-offsets.c delete mode 100644 arch/v850/kernel/bug.c delete mode 100644 arch/v850/kernel/entry.S delete mode 100644 arch/v850/kernel/fpga85e2c.c delete mode 100644 arch/v850/kernel/fpga85e2c.ld delete mode 100644 arch/v850/kernel/gbus_int.c delete mode 100644 arch/v850/kernel/head.S delete mode 100644 arch/v850/kernel/highres_timer.c delete mode 100644 arch/v850/kernel/init_task.c delete mode 100644 arch/v850/kernel/intv.S delete mode 100644 arch/v850/kernel/irq.c delete mode 100644 arch/v850/kernel/ma.c delete mode 100644 arch/v850/kernel/mach.c delete mode 100644 arch/v850/kernel/mach.h delete mode 100644 arch/v850/kernel/me2.c delete mode 100644 arch/v850/kernel/memcons.c delete mode 100644 arch/v850/kernel/module.c delete mode 100644 arch/v850/kernel/process.c delete mode 100644 arch/v850/kernel/procfs.c delete mode 100644 arch/v850/kernel/ptrace.c delete mode 100644 arch/v850/kernel/rte_cb.c delete mode 100644 arch/v850/kernel/rte_cb_leds.c delete mode 100644 arch/v850/kernel/rte_cb_multi.c delete mode 100644 arch/v850/kernel/rte_ma1_cb-rom.ld delete mode 100644 arch/v850/kernel/rte_ma1_cb.c delete mode 100644 arch/v850/kernel/rte_ma1_cb.ld delete mode 100644 arch/v850/kernel/rte_mb_a_pci.c delete mode 100644 arch/v850/kernel/rte_me2_cb.c delete mode 100644 arch/v850/kernel/rte_me2_cb.ld delete mode 100644 arch/v850/kernel/rte_nb85e_cb-multi.ld delete mode 100644 arch/v850/kernel/rte_nb85e_cb.c delete mode 100644 arch/v850/kernel/rte_nb85e_cb.ld delete mode 100644 arch/v850/kernel/setup.c delete mode 100644 arch/v850/kernel/signal.c delete mode 100644 arch/v850/kernel/sim.c delete mode 100644 arch/v850/kernel/sim.ld delete mode 100644 arch/v850/kernel/sim85e2.c delete mode 100644 arch/v850/kernel/sim85e2.ld delete mode 100644 arch/v850/kernel/simcons.c delete mode 100644 arch/v850/kernel/syscalls.c delete mode 100644 arch/v850/kernel/teg.c delete mode 100644 arch/v850/kernel/time.c delete mode 100644 arch/v850/kernel/v850_ksyms.c delete mode 100644 arch/v850/kernel/v850e2_cache.c delete mode 100644 arch/v850/kernel/v850e_cache.c delete mode 100644 arch/v850/kernel/v850e_intc.c delete mode 100644 arch/v850/kernel/v850e_timer_d.c delete mode 100644 arch/v850/kernel/v850e_utils.c delete mode 100644 arch/v850/kernel/vmlinux.lds.S delete mode 100644 arch/v850/lib/Makefile delete mode 100644 arch/v850/lib/ashldi3.c delete mode 100644 arch/v850/lib/ashrdi3.c delete mode 100644 arch/v850/lib/checksum.c delete mode 100644 arch/v850/lib/lshrdi3.c delete mode 100644 arch/v850/lib/memcpy.c delete mode 100644 arch/v850/lib/memset.c delete mode 100644 arch/v850/lib/muldi3.c delete mode 100644 arch/v850/lib/negdi2.c delete mode 100644 include/asm-v850/Kbuild delete mode 100644 include/asm-v850/a.out.h delete mode 100644 include/asm-v850/anna.h delete mode 100644 include/asm-v850/as85ep1.h delete mode 100644 include/asm-v850/asm.h delete mode 100644 include/asm-v850/atomic.h delete mode 100644 include/asm-v850/auxvec.h delete mode 100644 include/asm-v850/bitops.h delete mode 100644 include/asm-v850/bug.h delete mode 100644 include/asm-v850/bugs.h delete mode 100644 include/asm-v850/byteorder.h delete mode 100644 include/asm-v850/cache.h delete mode 100644 include/asm-v850/cacheflush.h delete mode 100644 include/asm-v850/checksum.h delete mode 100644 include/asm-v850/clinkage.h delete mode 100644 include/asm-v850/cputime.h delete mode 100644 include/asm-v850/current.h delete mode 100644 include/asm-v850/delay.h delete mode 100644 include/asm-v850/device.h delete mode 100644 include/asm-v850/div64.h delete mode 100644 include/asm-v850/dma-mapping.h delete mode 100644 include/asm-v850/dma.h delete mode 100644 include/asm-v850/elf.h delete mode 100644 include/asm-v850/emergency-restart.h delete mode 100644 include/asm-v850/entry.h delete mode 100644 include/asm-v850/errno.h delete mode 100644 include/asm-v850/fb.h delete mode 100644 include/asm-v850/fcntl.h delete mode 100644 include/asm-v850/flat.h delete mode 100644 include/asm-v850/fpga85e2c.h delete mode 100644 include/asm-v850/futex.h delete mode 100644 include/asm-v850/gbus_int.h delete mode 100644 include/asm-v850/hardirq.h delete mode 100644 include/asm-v850/highres_timer.h delete mode 100644 include/asm-v850/hw_irq.h delete mode 100644 include/asm-v850/io.h delete mode 100644 include/asm-v850/ioctl.h delete mode 100644 include/asm-v850/ioctls.h delete mode 100644 include/asm-v850/ipcbuf.h delete mode 100644 include/asm-v850/irq.h delete mode 100644 include/asm-v850/irq_regs.h delete mode 100644 include/asm-v850/kdebug.h delete mode 100644 include/asm-v850/kmap_types.h delete mode 100644 include/asm-v850/kvm.h delete mode 100644 include/asm-v850/linkage.h delete mode 100644 include/asm-v850/local.h delete mode 100644 include/asm-v850/ma.h delete mode 100644 include/asm-v850/ma1.h delete mode 100644 include/asm-v850/machdep.h delete mode 100644 include/asm-v850/macrology.h delete mode 100644 include/asm-v850/me2.h delete mode 100644 include/asm-v850/mman.h delete mode 100644 include/asm-v850/mmu.h delete mode 100644 include/asm-v850/mmu_context.h delete mode 100644 include/asm-v850/module.h delete mode 100644 include/asm-v850/msgbuf.h delete mode 100644 include/asm-v850/mutex.h delete mode 100644 include/asm-v850/page.h delete mode 100644 include/asm-v850/param.h delete mode 100644 include/asm-v850/pci.h delete mode 100644 include/asm-v850/percpu.h delete mode 100644 include/asm-v850/pgalloc.h delete mode 100644 include/asm-v850/pgtable.h delete mode 100644 include/asm-v850/poll.h delete mode 100644 include/asm-v850/posix_types.h delete mode 100644 include/asm-v850/processor.h delete mode 100644 include/asm-v850/ptrace.h delete mode 100644 include/asm-v850/resource.h delete mode 100644 include/asm-v850/rte_cb.h delete mode 100644 include/asm-v850/rte_ma1_cb.h delete mode 100644 include/asm-v850/rte_mb_a_pci.h delete mode 100644 include/asm-v850/rte_me2_cb.h delete mode 100644 include/asm-v850/rte_nb85e_cb.h delete mode 100644 include/asm-v850/scatterlist.h delete mode 100644 include/asm-v850/sections.h delete mode 100644 include/asm-v850/segment.h delete mode 100644 include/asm-v850/semaphore.h delete mode 100644 include/asm-v850/sembuf.h delete mode 100644 include/asm-v850/serial.h delete mode 100644 include/asm-v850/setup.h delete mode 100644 include/asm-v850/shmbuf.h delete mode 100644 include/asm-v850/shmparam.h delete mode 100644 include/asm-v850/sigcontext.h delete mode 100644 include/asm-v850/siginfo.h delete mode 100644 include/asm-v850/signal.h delete mode 100644 include/asm-v850/sim.h delete mode 100644 include/asm-v850/sim85e2.h delete mode 100644 include/asm-v850/sim85e2c.h delete mode 100644 include/asm-v850/sim85e2s.h delete mode 100644 include/asm-v850/simsyscall.h delete mode 100644 include/asm-v850/socket.h delete mode 100644 include/asm-v850/sockios.h delete mode 100644 include/asm-v850/stat.h delete mode 100644 include/asm-v850/statfs.h delete mode 100644 include/asm-v850/string.h delete mode 100644 include/asm-v850/system.h delete mode 100644 include/asm-v850/teg.h delete mode 100644 include/asm-v850/termbits.h delete mode 100644 include/asm-v850/termios.h delete mode 100644 include/asm-v850/thread_info.h delete mode 100644 include/asm-v850/timex.h delete mode 100644 include/asm-v850/tlb.h delete mode 100644 include/asm-v850/tlbflush.h delete mode 100644 include/asm-v850/topology.h delete mode 100644 include/asm-v850/types.h delete mode 100644 include/asm-v850/uaccess.h delete mode 100644 include/asm-v850/ucontext.h delete mode 100644 include/asm-v850/unaligned.h delete mode 100644 include/asm-v850/unistd.h delete mode 100644 include/asm-v850/user.h delete mode 100644 include/asm-v850/v850e.h delete mode 100644 include/asm-v850/v850e2.h delete mode 100644 include/asm-v850/v850e2_cache.h delete mode 100644 include/asm-v850/v850e_cache.h delete mode 100644 include/asm-v850/v850e_intc.h delete mode 100644 include/asm-v850/v850e_timer_c.h delete mode 100644 include/asm-v850/v850e_timer_d.h delete mode 100644 include/asm-v850/v850e_uart.h delete mode 100644 include/asm-v850/v850e_uarta.h delete mode 100644 include/asm-v850/v850e_uartb.h delete mode 100644 include/asm-v850/v850e_utils.h diff --git a/MAINTAINERS b/MAINTAINERS index 7ffd78c4e277..7e5c7b0290bb 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -4131,9 +4131,6 @@ W: http://www.uclinux.org/ L: uclinux-dev@uclinux.org (subscribers-only) S: Maintained -UCLINUX FOR NEC V850 -P: Miles Bader - UCLINUX FOR RENESAS H8/300 P: Yoshinori Sato M: ysato@users.sourceforge.jp diff --git a/arch/v850/Kconfig b/arch/v850/Kconfig deleted file mode 100644 index 4379f43505ef..000000000000 --- a/arch/v850/Kconfig +++ /dev/null @@ -1,353 +0,0 @@ -############################################################################# -# -# For a description of the syntax of this configuration file, -# see Documentation/kbuild/kconfig-language.txt. -# -############################################################################# - -mainmenu "uClinux/v850 (w/o MMU) Kernel Configuration" - -config MMU - bool - default n -config ZONE_DMA - bool - default y -config RWSEM_GENERIC_SPINLOCK - bool - default y -config RWSEM_XCHGADD_ALGORITHM - bool - default n -config GENERIC_FIND_NEXT_BIT - bool - default y -config GENERIC_HWEIGHT - bool - default y -config GENERIC_CALIBRATE_DELAY - bool - default y - -config GENERIC_HARDIRQS - bool - default y - -config GENERIC_IRQ_PROBE - bool - default y - -config GENERIC_TIME - bool - default y - -config TIME_LOW_RES - bool - default y - -config ARCH_HAS_ILOG2_U32 - bool - default n - -config ARCH_HAS_ILOG2_U64 - bool - default n - -config ARCH_SUPPORTS_AOUT - def_bool y - -# Turn off some random 386 crap that can affect device config -config ISA - bool - default n -config ISAPNP - bool - default n -config EISA - bool - default n -config MCA - bool - default n - - -############################################################################# -#### v850-specific config - -# Define the architecture -config V850 - bool - default y - select HAVE_IDE - -menu "Processor type and features" - - choice - prompt "Platform" - default GDB - config V850E_SIM - bool "GDB" - config RTE_CB_MA1 - bool "RTE-V850E/MA1-CB" - config RTE_CB_NB85E - bool "RTE-V850E/NB85E-CB" - config RTE_CB_ME2 - bool "RTE-V850E/ME2-CB" - config V850E_AS85EP1 - bool "AS85EP1" - config V850E2_SIM85E2C - bool "sim85e2c" - config V850E2_SIM85E2S - bool "sim85e2s" - config V850E2_FPGA85E2C - bool "NA85E2C-FPGA" - config V850E2_ANNA - bool "Anna" - endchoice - - #### V850E processor-specific config - - # All CPUs currently supported use the v850e architecture - config V850E - bool - default y - - # The RTE-V850E/MA1-CB is the only type of V850E/MA1 platform we - # currently support - config V850E_MA1 - bool - depends on RTE_CB_MA1 - default y - # Similarly for the RTE-V850E/NB85E-CB - V850E/TEG - config V850E_TEG - bool - depends on RTE_CB_NB85E - default y - # ... and the RTE-V850E/ME2-CB - V850E/ME2 - config V850E_ME2 - bool - depends on RTE_CB_ME2 - default y - - - #### sim85e2-specific config - - config V850E2_SIM85E2 - bool - depends on V850E2_SIM85E2C || V850E2_SIM85E2S - default y - - - #### V850E2 processor-specific config - - # V850E2 processors - config V850E2 - bool - depends on V850E2_SIM85E2 || V850E2_FPGA85E2C || V850E2_ANNA - default y - - - #### RTE-CB platform-specific config - - # Boards in the RTE-x-CB series - config RTE_CB - bool - depends on RTE_CB_MA1 || RTE_CB_NB85E || RTE_CB_ME2 - default y - - config RTE_CB_MULTI - bool - # RTE_CB_NB85E can either have multi ROM support or not, but - # other platforms (currently only RTE_CB_MA1) require it. - prompt "Multi monitor ROM support" if RTE_CB_NB85E - depends on RTE_CB_MA1 || RTE_CB_NB85E - default y - - config RTE_CB_MULTI_DBTRAP - bool "Pass illegal insn trap / dbtrap to kernel" - depends on RTE_CB_MULTI - default n - - config RTE_CB_MA1_KSRAM - bool "Kernel in SRAM (limits size of kernel)" - depends on RTE_CB_MA1 && RTE_CB_MULTI - default n - - config RTE_MB_A_PCI - bool "Mother-A PCI support" - depends on RTE_CB - default y - - # The GBUS is used to talk to the RTE-MOTHER-A board - config RTE_GBUS_INT - bool - depends on RTE_MB_A_PCI - default y - - # The only PCI bus we support is on the RTE-MOTHER-A board - config PCI - bool - default RTE_MB_A_PCI - - #### Some feature-specific configs - - # Everything except for the GDB simulator uses the same interrupt controller - config V850E_INTC - bool - default !V850E_SIM - - # Everything except for the various simulators uses the "Timer D" unit - config V850E_TIMER_D - bool - default !V850E_SIM && !V850E2_SIM85E2 - - # Cache control used on some v850e1 processors - config V850E_CACHE - bool - default V850E_TEG || V850E_ME2 - - # Cache control used on v850e2 processors; I think this should - # actually apply to more, but currently only the SIM85E2S uses it - config V850E2_CACHE - bool - default V850E2_SIM85E2S - - config NO_CACHE - bool - default !V850E_CACHE && !V850E2_CACHE - - # HZ depends on the platform - config HZ - int - default 24 if V850E_SIM || V850E2_SIM85E2 - default 122 if V850E2_FPGA85E2C - default 100 - - #### Misc config - - config ROM_KERNEL - bool "Kernel in ROM" - depends on V850E2_ANNA || V850E_AS85EP1 || RTE_CB_ME2 - - # Some platforms pre-zero memory, in which case the kernel doesn't need to - config ZERO_BSS - bool - depends on !V850E2_SIM85E2C - default y - - # The crappy-ass zone allocator requires that the start of allocatable - # memory be aligned to the largest possible allocation. - config FORCE_MAX_ZONEORDER - int - default 8 if V850E2_SIM85E2C || V850E2_FPGA85E2C - - config V850E_HIGHRES_TIMER - bool "High resolution timer support" - depends on V850E_TIMER_D - config TIME_BOOTUP - bool "Time bootup" - depends on V850E_HIGHRES_TIMER - - config RESET_GUARD - bool "Reset Guard" - -source "mm/Kconfig" - -endmenu - - -############################################################################# - -source init/Kconfig - -############################################################################# - -menu "Bus options (PCI, PCMCIA, EISA, MCA, ISA)" - -# config PCI -# bool "PCI support" -# help -# Support for PCI bus. - -source "drivers/pci/Kconfig" - -source "drivers/pcmcia/Kconfig" - -source "drivers/pci/hotplug/Kconfig" - -endmenu - -menu "Executable file formats" - -source "fs/Kconfig.binfmt" - -endmenu - -source "net/Kconfig" - -############################################################################# - -source "drivers/base/Kconfig" - -source drivers/mtd/Kconfig - -source drivers/parport/Kconfig - -#source drivers/pnp/Kconfig - -source drivers/block/Kconfig - -############################################################################# - -menu "Disk device support" - -source "drivers/ide/Kconfig" - -source "drivers/scsi/Kconfig" - -endmenu - -############################################################################# - - -source "drivers/md/Kconfig" - -source "drivers/message/fusion/Kconfig" - -source "drivers/ieee1394/Kconfig" - -source "drivers/message/i2o/Kconfig" - -source "drivers/net/Kconfig" - -source "drivers/isdn/Kconfig" - -#source "drivers/telephony/Kconfig" - -# -# input before char - char/joystick depends on it. As does USB. -# -source "drivers/input/Kconfig" - -source "drivers/char/Kconfig" - -#source drivers/misc/Config.in -source "drivers/media/Kconfig" - -source "fs/Kconfig" - -source "drivers/video/Kconfig" - -source "sound/Kconfig" - -source "drivers/usb/Kconfig" - -source "arch/v850/Kconfig.debug" - -source "security/Kconfig" - -source "crypto/Kconfig" - -source "lib/Kconfig" - -############################################################################# diff --git a/arch/v850/Kconfig.debug b/arch/v850/Kconfig.debug deleted file mode 100644 index 4acfb9cca1ca..000000000000 --- a/arch/v850/Kconfig.debug +++ /dev/null @@ -1,10 +0,0 @@ -menu "Kernel hacking" - -source "lib/Kconfig.debug" - -config NO_KERNEL_MSG - bool "Suppress Kernel BUG Messages" - help - Do not output any debug BUG messages within the kernel. - -endmenu diff --git a/arch/v850/Makefile b/arch/v850/Makefile deleted file mode 100644 index 8b629df0029a..000000000000 --- a/arch/v850/Makefile +++ /dev/null @@ -1,54 +0,0 @@ -# -# arch/v850/Makefile -# -# Copyright (C) 2001,02,03,05 NEC Corporation -# Copyright (C) 2001,02,03,05 Miles Bader -# -# This file is included by the global makefile so that you can add your own -# architecture-specific flags and dependencies. Remember to do have actions -# for "archclean" and "archdep" for cleaning up and making dependencies for -# this architecture -# -# This file is subject to the terms and conditions of the GNU General Public -# License. See the file "COPYING" in the main directory of this archive -# for more details. -# - -arch_dir = arch/v850 - -KBUILD_CFLAGS += -mv850e -# r16 is a fixed pointer to the current task -KBUILD_CFLAGS += -ffixed-r16 -mno-prolog-function -KBUILD_CFLAGS += -fno-builtin -KBUILD_CFLAGS += -D__linux__ -DUTS_SYSNAME=\"uClinux\" - -# By default, build a kernel that runs on the gdb v850 simulator. -KBUILD_DEFCONFIG := sim_defconfig - -# This prevents the linker from consolidating the .gnu.linkonce.this_module -# section into .text (which the v850 default linker script for -r does for -# some reason) -LDFLAGS_MODULE += --unique=.gnu.linkonce.this_module - -OBJCOPY_FLAGS_BLOB := -I binary -O elf32-little -B v850e - - -head-y := $(arch_dir)/kernel/head.o $(arch_dir)/kernel/init_task.o -core-y += $(arch_dir)/kernel/ -libs-y += $(arch_dir)/lib/ - - -# Deal with the initial contents of the root device -ifdef ROOT_FS_IMAGE -core-y += root_fs_image.o - -# Because the kernel build-system erases all explicit .o build rules, we -# have to use an intermediate target to fool it into building for us. -# This results in it being built anew each time, but that's alright. -root_fs_image.o: root_fs_image_force - -root_fs_image_force: $(ROOT_FS_IMAGE) - $(OBJCOPY) $(OBJCOPY_FLAGS_BLOB) --rename-section .data=.root,alloc,load,readonly,data,contents $< root_fs_image.o -endif - -CLEAN_FILES += root_fs_image.o diff --git a/arch/v850/README b/arch/v850/README deleted file mode 100644 index 12f7f7a665e0..000000000000 --- a/arch/v850/README +++ /dev/null @@ -1,44 +0,0 @@ -This port to the NEC V850E processor supports the following platforms: - - "sim" - The gdb v850e simulator (CONFIG_V850E_SIM). - - "rte-ma1-cb" - The Midas labs RTE-V850E/MA1-CB and RTE-V850E/NB85E-CB evaluation - boards (CONFIG_RTE_CB_MA1 and CONFIG_RTE_CB_NB85E). This support - has only been tested when running with the Multi-debugger monitor - ROM (for the Green Hills Multi debugger). The optional NEC - Solution Gear RTE-MOTHER-A motherboard is also supported, which - allows PCI boards to be used (CONFIG_RTE_MB_A_PCI). - - "rte-me2-cb" - The Midas labs RTE-V850E/ME2-CB evaluation board (CONFIG_RTE_CB_ME2). - This has only been tested using a kernel downloaded via an ICE - connection using the Multi debugger. Support for the RTE-MOTHER-A is - present, but hasn't been tested (unlike the other Midas labs cpu - boards, the RTE-V850E/ME2-CB includes an ethernet adaptor). - - "as85ep1" - The NEC AS85EP1 V850E evaluation chip/board (CONFIG_V850E_AS85EP1). - - "anna" - The NEC `Anna' (board/chip) implementation of the V850E2 processor - (CONFIG_V850E2_ANNA). - - "sim85e2c", "sim85e2s" - The sim85e2c and sim85e2s simulators, which are verilog simulations - of the V850E2 NA85E2C/NA85E2S cpu cores (CONFIG_V850E2_SIM85E2C and - CONFIG_V850E2_SIM85E2S). - - "fpga85e2c" - A FPGA implementation of the V850E2 NA85E2C cpu core - (CONFIG_V850E2_FPGA85E2C). - -To get a default kernel configuration for a particular platform, you can -use a _defconfig make target (e.g., "make rte-me2-cb_defconfig"); -to see which default configurations are possible, look in the directory -"arch/v850/configs". - -Porting to anything with a V850E/MA1 or MA2 processor should be simple. -See the file and the files it includes for an example of -how to add platform/chip-specific support. diff --git a/arch/v850/configs/rte-ma1-cb_defconfig b/arch/v850/configs/rte-ma1-cb_defconfig deleted file mode 100644 index 1a5beda36e29..000000000000 --- a/arch/v850/configs/rte-ma1-cb_defconfig +++ /dev/null @@ -1,617 +0,0 @@ -# -# Automatically generated make config: don't edit -# Linux kernel version: 2.6.13-uc0 -# Fri Sep 2 13:54:27 2005 -# -# CONFIG_MMU is not set -# CONFIG_UID16 is not set -CONFIG_RWSEM_GENERIC_SPINLOCK=y -# CONFIG_RWSEM_XCHGADD_ALGORITHM is not set -CONFIG_GENERIC_CALIBRATE_DELAY=y -# CONFIG_ISA is not set -# CONFIG_ISAPNP is not set -# CONFIG_EISA is not set -# CONFIG_MCA is not set -CONFIG_V850=y - -# -# Processor type and features -# -# CONFIG_V850E_SIM is not set -CONFIG_RTE_CB_MA1=y -# CONFIG_RTE_CB_NB85E is not set -# CONFIG_RTE_CB_ME2 is not set -# CONFIG_V850E_AS85EP1 is not set -# CONFIG_V850E2_SIM85E2C is not set -# CONFIG_V850E2_SIM85E2S is not set -# CONFIG_V850E2_FPGA85E2C is not set -# CONFIG_V850E2_ANNA is not set -CONFIG_V850E=y -CONFIG_V850E_MA1=y -CONFIG_RTE_CB=y -CONFIG_RTE_CB_MULTI=y -CONFIG_RTE_CB_MULTI_DBTRAP=y -# CONFIG_RTE_CB_MA1_KSRAM is not set -CONFIG_RTE_MB_A_PCI=y -CONFIG_RTE_GBUS_INT=y -CONFIG_PCI=y -CONFIG_V850E_INTC=y -CONFIG_V850E_TIMER_D=y -# CONFIG_V850E_CACHE is not set -# CONFIG_V850E2_CACHE is not set -CONFIG_NO_CACHE=y -CONFIG_ZERO_BSS=y -# CONFIG_V850E_HIGHRES_TIMER is not set -# CONFIG_RESET_GUARD is not set -CONFIG_LARGE_ALLOCS=y -CONFIG_FLATMEM=y -CONFIG_FLAT_NODE_MEM_MAP=y - -# -# Code maturity level options -# -# CONFIG_EXPERIMENTAL is not set -CONFIG_CLEAN_COMPILE=y -CONFIG_BROKEN_ON_SMP=y -CONFIG_INIT_ENV_ARG_LIMIT=32 - -# -# General setup -# -CONFIG_LOCALVERSION="" -# CONFIG_BSD_PROCESS_ACCT is not set -# CONFIG_SYSCTL is not set -# CONFIG_AUDIT is not set -# CONFIG_HOTPLUG is not set -CONFIG_KOBJECT_UEVENT=y -# CONFIG_IKCONFIG is not set -CONFIG_EMBEDDED=y -# CONFIG_KALLSYMS is not set -CONFIG_PRINTK=y -CONFIG_BUG=y -# CONFIG_BASE_FULL is not set -# CONFIG_FUTEX is not set -# CONFIG_EPOLL is not set -CONFIG_CC_OPTIMIZE_FOR_SIZE=y -CONFIG_CC_ALIGN_FUNCTIONS=0 -CONFIG_CC_ALIGN_LABELS=0 -CONFIG_CC_ALIGN_LOOPS=0 -CONFIG_CC_ALIGN_JUMPS=0 -CONFIG_BASE_SMALL=1 - -# -# Loadable module support -# -CONFIG_MODULES=y -CONFIG_MODULE_UNLOAD=y -CONFIG_OBSOLETE_MODPARM=y -# CONFIG_MODULE_SRCVERSION_ALL is not set -CONFIG_KMOD=y - -# -# Bus options (PCI, PCMCIA, EISA, MCA, ISA) -# -# CONFIG_PCI_LEGACY_PROC is not set -# CONFIG_PCI_NAMES is not set -# CONFIG_PCI_DEBUG is not set - -# -# PCCARD (PCMCIA/CardBus) support -# -# CONFIG_PCCARD is not set - -# -# PCI Hotplug Support -# - -# -# Executable file formats -# -CONFIG_BINFMT_FLAT=y -# CONFIG_BINFMT_ZFLAT is not set -# CONFIG_BINFMT_SHARED_FLAT is not set -# CONFIG_BINFMT_MISC is not set - -# -# Networking -# -CONFIG_NET=y - -# -# Networking options -# -# CONFIG_PACKET is not set -# CONFIG_UNIX is not set -# CONFIG_NET_KEY is not set -CONFIG_INET=y -# CONFIG_IP_MULTICAST is not set -# CONFIG_IP_ADVANCED_ROUTER is not set -CONFIG_IP_FIB_HASH=y -# CONFIG_IP_PNP is not set -# CONFIG_NET_IPIP is not set -# CONFIG_NET_IPGRE is not set -# CONFIG_SYN_COOKIES is not set -# CONFIG_INET_AH is not set -# CONFIG_INET_ESP is not set -# CONFIG_INET_IPCOMP is not set -# CONFIG_INET_TUNNEL is not set -# CONFIG_IP_TCPDIAG is not set -# CONFIG_IP_TCPDIAG_IPV6 is not set -# CONFIG_TCP_CONG_ADVANCED is not set -CONFIG_TCP_CONG_BIC=y -# CONFIG_IPV6 is not set -# CONFIG_NETFILTER is not set -# CONFIG_BRIDGE is not set -# CONFIG_VLAN_8021Q is not set -# CONFIG_DECNET is not set -# CONFIG_LLC2 is not set -# CONFIG_IPX is not set -# CONFIG_ATALK is not set -# CONFIG_NET_SCHED is not set -# CONFIG_NET_CLS_ROUTE is not set - -# -# Network testing -# -# CONFIG_NET_PKTGEN is not set -# CONFIG_HAMRADIO is not set -# CONFIG_IRDA is not set -# CONFIG_BT is not set - -# -# Generic Driver Options -# -CONFIG_STANDALONE=y -CONFIG_PREVENT_FIRMWARE_BUILD=y -# CONFIG_FW_LOADER is not set -# CONFIG_DEBUG_DRIVER is not set - -# -# Memory Technology Devices (MTD) -# -CONFIG_MTD=y -# CONFIG_MTD_DEBUG is not set -# CONFIG_MTD_CONCAT is not set -# CONFIG_MTD_PARTITIONS is not set - -# -# User Modules And Translation Layers -# -# CONFIG_MTD_CHAR is not set -CONFIG_MTD_BLOCK=y -# CONFIG_FTL is not set -# CONFIG_NFTL is not set -# CONFIG_INFTL is not set - -# -# RAM/ROM/Flash chip drivers -# -# CONFIG_MTD_CFI is not set -# CONFIG_MTD_JEDECPROBE is not set -CONFIG_MTD_MAP_BANK_WIDTH_1=y -CONFIG_MTD_MAP_BANK_WIDTH_2=y -CONFIG_MTD_MAP_BANK_WIDTH_4=y -# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set -# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set -# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set -CONFIG_MTD_CFI_I1=y -CONFIG_MTD_CFI_I2=y -# CONFIG_MTD_CFI_I4 is not set -# CONFIG_MTD_CFI_I8 is not set -# CONFIG_MTD_RAM is not set -# CONFIG_MTD_ROM is not set -# CONFIG_MTD_ABSENT is not set - -# -# Mapping drivers for chip access -# -# CONFIG_MTD_COMPLEX_MAPPINGS is not set -# CONFIG_MTD_PLATRAM is not set - -# -# Self-contained MTD device drivers -# -# CONFIG_MTD_PMC551 is not set -CONFIG_MTD_SLRAM=y -# CONFIG_MTD_PHRAM is not set -# CONFIG_MTD_MTDRAM is not set -# CONFIG_MTD_BLKMTD is not set - -# -# Disk-On-Chip Device Drivers -# -# CONFIG_MTD_DOC2000 is not set -# CONFIG_MTD_DOC2001 is not set -# CONFIG_MTD_DOC2001PLUS is not set - -# -# NAND Flash Device Drivers -# -# CONFIG_MTD_NAND is not set - -# -# Parallel port support -# -# CONFIG_PARPORT is not set - -# -# Block devices -# -# CONFIG_BLK_DEV_FD is not set -# CONFIG_BLK_CPQ_DA is not set -# CONFIG_BLK_CPQ_CISS_DA is not set -# CONFIG_BLK_DEV_DAC960 is not set -# CONFIG_BLK_DEV_COW_COMMON is not set -# CONFIG_BLK_DEV_LOOP is not set -# CONFIG_BLK_DEV_NBD is not set -# CONFIG_BLK_DEV_SX8 is not set -# CONFIG_BLK_DEV_RAM is not set -CONFIG_BLK_DEV_RAM_COUNT=16 -CONFIG_INITRAMFS_SOURCE="" -# CONFIG_CDROM_PKTCDVD is not set - -# -# IO Schedulers -# -CONFIG_IOSCHED_NOOP=y -# CONFIG_IOSCHED_AS is not set -# CONFIG_IOSCHED_DEADLINE is not set -# CONFIG_IOSCHED_CFQ is not set -# CONFIG_ATA_OVER_ETH is not set - -# -# Disk device support -# - -# -# ATA/ATAPI/MFM/RLL support -# -# CONFIG_IDE is not set - -# -# SCSI device support -# -# CONFIG_SCSI is not set - -# -# Multi-device support (RAID and LVM) -# -# CONFIG_MD is not set - -# -# Fusion MPT device support -# -# CONFIG_FUSION is not set - -# -# IEEE 1394 (FireWire) support -# -# CONFIG_IEEE1394 is not set - -# -# I2O device support -# -# CONFIG_I2O is not set - -# -# Network device support -# -CONFIG_NETDEVICES=y -# CONFIG_DUMMY is not set -# CONFIG_BONDING is not set -# CONFIG_EQUALIZER is not set -# CONFIG_TUN is not set - -# -# ARCnet devices -# -# CONFIG_ARCNET is not set - -# -# Ethernet (10 or 100Mbit) -# -CONFIG_NET_ETHERNET=y -CONFIG_MII=y -# CONFIG_HAPPYMEAL is not set -# CONFIG_SUNGEM is not set -# CONFIG_NET_VENDOR_3COM is not set -# CONFIG_NET_VENDOR_SMC is not set - -# -# Tulip family network device support -# -# CONFIG_NET_TULIP is not set -# CONFIG_HP100 is not set -# CONFIG_NE2000 is not set -CONFIG_NET_PCI=y -# CONFIG_PCNET32 is not set -# CONFIG_AMD8111_ETH is not set -# CONFIG_ADAPTEC_STARFIRE is not set -# CONFIG_DGRS is not set -CONFIG_EEPRO100=y -# CONFIG_E100 is not set -# CONFIG_FEALNX is not set -# CONFIG_NATSEMI is not set -# CONFIG_NE2K_PCI is not set -# CONFIG_8139TOO is not set -# CONFIG_SIS900 is not set -# CONFIG_EPIC100 is not set -# CONFIG_SUNDANCE is not set -# CONFIG_TLAN is not set -# CONFIG_VIA_RHINE is not set - -# -# Ethernet (1000 Mbit) -# -# CONFIG_ACENIC is not set -# CONFIG_DL2K is not set -# CONFIG_E1000 is not set -# CONFIG_NS83820 is not set -# CONFIG_HAMACHI is not set -# CONFIG_R8169 is not set -# CONFIG_SK98LIN is not set -# CONFIG_VIA_VELOCITY is not set -# CONFIG_TIGON3 is not set -# CONFIG_BNX2 is not set - -# -# Ethernet (10000 Mbit) -# -# CONFIG_IXGB is not set -# CONFIG_S2IO is not set - -# -# Token Ring devices -# -# CONFIG_TR is not set - -# -# Wireless LAN (non-hamradio) -# -# CONFIG_NET_RADIO is not set - -# -# Wan interfaces -# -# CONFIG_WAN is not set -# CONFIG_FDDI is not set -# CONFIG_PPP is not set -# CONFIG_SLIP is not set -# CONFIG_NETPOLL is not set -# CONFIG_NET_POLL_CONTROLLER is not set - -# -# ISDN subsystem -# -# CONFIG_ISDN is not set - -# -# Input device support -# -CONFIG_INPUT=y - -# -# Userland interfaces -# -# CONFIG_INPUT_MOUSEDEV is not set -# CONFIG_INPUT_JOYDEV is not set -# CONFIG_INPUT_TSDEV is not set -# CONFIG_INPUT_EVDEV is not set -# CONFIG_INPUT_EVBUG is not set - -# -# Input Device Drivers -# -# CONFIG_INPUT_KEYBOARD is not set -# CONFIG_INPUT_MOUSE is not set -# CONFIG_INPUT_JOYSTICK is not set -# CONFIG_INPUT_TOUCHSCREEN is not set -# CONFIG_INPUT_MISC is not set - -# -# Hardware I/O ports -# -# CONFIG_SERIO is not set -# CONFIG_GAMEPORT is not set - -# -# Character devices -# -# CONFIG_VT is not set -# CONFIG_SERIAL_NONSTANDARD is not set - -# -# Serial drivers -# -# CONFIG_SERIAL_8250 is not set - -# -# Non-8250 serial port support -# -CONFIG_V850E_UART=y -CONFIG_V850E_UART_CONSOLE=y -CONFIG_SERIAL_CORE=y -CONFIG_SERIAL_CORE_CONSOLE=y -# CONFIG_SERIAL_JSM is not set -# CONFIG_UNIX98_PTYS is not set -# CONFIG_LEGACY_PTYS is not set - -# -# IPMI -# -# CONFIG_IPMI_HANDLER is not set - -# -# Watchdog Cards -# -# CONFIG_WATCHDOG is not set -# CONFIG_RTC is not set -# CONFIG_GEN_RTC is not set -# CONFIG_DTLK is not set -# CONFIG_R3964 is not set -# CONFIG_APPLICOM is not set - -# -# Ftape, the floppy tape device driver -# -# CONFIG_DRM is not set -# CONFIG_RAW_DRIVER is not set - -# -# TPM devices -# - -# -# Multimedia devices -# -# CONFIG_VIDEO_DEV is not set - -# -# Digital Video Broadcasting Devices -# -# CONFIG_DVB is not set - -# -# File systems -# -# CONFIG_EXT2_FS is not set -# CONFIG_EXT3_FS is not set -# CONFIG_JBD is not set -# CONFIG_REISERFS_FS is not set -# CONFIG_JFS_FS is not set -# CONFIG_FS_POSIX_ACL is not set - -# -# XFS support -# -# CONFIG_XFS_FS is not set -# CONFIG_MINIX_FS is not set -CONFIG_ROMFS_FS=y -# CONFIG_MAGIC_ROM_PTR is not set -CONFIG_INOTIFY=y -# CONFIG_QUOTA is not set -CONFIG_DNOTIFY=y -# CONFIG_AUTOFS_FS is not set -# CONFIG_AUTOFS4_FS is not set - -# -# CD-ROM/DVD Filesystems -# -# CONFIG_ISO9660_FS is not set -# CONFIG_UDF_FS is not set - -# -# DOS/FAT/NT Filesystems -# -# CONFIG_MSDOS_FS is not set -# CONFIG_VFAT_FS is not set -# CONFIG_NTFS_FS is not set - -# -# Pseudo filesystems -# -CONFIG_PROC_FS=y -CONFIG_SYSFS=y -# CONFIG_TMPFS is not set -# CONFIG_HUGETLB_PAGE is not set -CONFIG_RAMFS=y - -# -# Miscellaneous filesystems -# -# CONFIG_HFSPLUS_FS is not set -# CONFIG_JFFS_FS is not set -# CONFIG_JFFS2_FS is not set -# CONFIG_CRAMFS is not set -# CONFIG_VXFS_FS is not set -# CONFIG_HPFS_FS is not set -# CONFIG_QNX4FS_FS is not set -# CONFIG_SYSV_FS is not set -# CONFIG_UFS_FS is not set - -# -# Network File Systems -# -CONFIG_NFS_FS=y -CONFIG_NFS_V3=y -# CONFIG_NFS_V3_ACL is not set -# CONFIG_NFSD is not set -CONFIG_LOCKD=y -CONFIG_LOCKD_V4=y -CONFIG_NFS_COMMON=y -CONFIG_SUNRPC=y -# CONFIG_SMB_FS is not set -# CONFIG_CIFS is not set -# CONFIG_NCP_FS is not set -# CONFIG_CODA_FS is not set - -# -# Partition Types -# -# CONFIG_PARTITION_ADVANCED is not set -CONFIG_MSDOS_PARTITION=y - -# -# Native Language Support -# -# CONFIG_NLS is not set - -# -# Graphics support -# -# CONFIG_FB is not set - -# -# Sound -# -# CONFIG_SOUND is not set - -# -# USB support -# -CONFIG_USB_ARCH_HAS_HCD=y -CONFIG_USB_ARCH_HAS_OHCI=y -# CONFIG_USB is not set - -# -# USB Gadget Support -# -# CONFIG_USB_GADGET is not set - -# -# Kernel hacking -# -# CONFIG_PRINTK_TIME is not set -CONFIG_DEBUG_KERNEL=y -# CONFIG_MAGIC_SYSRQ is not set -CONFIG_LOG_BUF_SHIFT=14 -# CONFIG_SCHEDSTATS is not set -# CONFIG_DEBUG_SLAB is not set -# CONFIG_DEBUG_SPINLOCK is not set -# CONFIG_DEBUG_SPINLOCK_SLEEP is not set -# CONFIG_DEBUG_KOBJECT is not set -CONFIG_DEBUG_INFO=y -# CONFIG_DEBUG_FS is not set -# CONFIG_NO_KERNEL_MSG is not set - -# -# Security options -# -# CONFIG_KEYS is not set -# CONFIG_SECURITY is not set - -# -# Cryptographic options -# -# CONFIG_CRYPTO is not set - -# -# Hardware crypto devices -# - -# -# Library routines -# -# CONFIG_CRC_CCITT is not set -# CONFIG_CRC32 is not set -# CONFIG_LIBCRC32C is not set diff --git a/arch/v850/configs/rte-me2-cb_defconfig b/arch/v850/configs/rte-me2-cb_defconfig deleted file mode 100644 index 15e666478061..000000000000 --- a/arch/v850/configs/rte-me2-cb_defconfig +++ /dev/null @@ -1,462 +0,0 @@ -# -# Automatically generated make config: don't edit -# Linux kernel version: 2.6.13-uc0 -# Fri Sep 2 13:47:50 2005 -# -# CONFIG_MMU is not set -# CONFIG_UID16 is not set -CONFIG_RWSEM_GENERIC_SPINLOCK=y -# CONFIG_RWSEM_XCHGADD_ALGORITHM is not set -CONFIG_GENERIC_CALIBRATE_DELAY=y -# CONFIG_ISA is not set -# CONFIG_ISAPNP is not set -# CONFIG_EISA is not set -# CONFIG_MCA is not set -CONFIG_V850=y - -# -# Processor type and features -# -# CONFIG_V850E_SIM is not set -# CONFIG_RTE_CB_MA1 is not set -# CONFIG_RTE_CB_NB85E is not set -CONFIG_RTE_CB_ME2=y -# CONFIG_V850E_AS85EP1 is not set -# CONFIG_V850E2_SIM85E2C is not set -# CONFIG_V850E2_SIM85E2S is not set -# CONFIG_V850E2_FPGA85E2C is not set -# CONFIG_V850E2_ANNA is not set -CONFIG_V850E=y -CONFIG_V850E_ME2=y -CONFIG_RTE_CB=y -# CONFIG_RTE_MB_A_PCI is not set -# CONFIG_PCI is not set -CONFIG_V850E_INTC=y -CONFIG_V850E_TIMER_D=y -CONFIG_V850E_CACHE=y -# CONFIG_V850E2_CACHE is not set -# CONFIG_NO_CACHE is not set -# CONFIG_ROM_KERNEL is not set -CONFIG_ZERO_BSS=y -# CONFIG_V850E_HIGHRES_TIMER is not set -# CONFIG_RESET_GUARD is not set -CONFIG_LARGE_ALLOCS=y -CONFIG_FLATMEM=y -CONFIG_FLAT_NODE_MEM_MAP=y - -# -# Code maturity level options -# -# CONFIG_EXPERIMENTAL is not set -CONFIG_CLEAN_COMPILE=y -CONFIG_BROKEN_ON_SMP=y -CONFIG_INIT_ENV_ARG_LIMIT=32 - -# -# General setup -# -CONFIG_LOCALVERSION="" -# CONFIG_BSD_PROCESS_ACCT is not set -# CONFIG_SYSCTL is not set -# CONFIG_HOTPLUG is not set -# CONFIG_IKCONFIG is not set -CONFIG_EMBEDDED=y -# CONFIG_KALLSYMS is not set -CONFIG_PRINTK=y -CONFIG_BUG=y -# CONFIG_BASE_FULL is not set -# CONFIG_FUTEX is not set -# CONFIG_EPOLL is not set -CONFIG_CC_OPTIMIZE_FOR_SIZE=y -CONFIG_CC_ALIGN_FUNCTIONS=0 -CONFIG_CC_ALIGN_LABELS=0 -CONFIG_CC_ALIGN_LOOPS=0 -CONFIG_CC_ALIGN_JUMPS=0 -CONFIG_BASE_SMALL=1 - -# -# Loadable module support -# -CONFIG_MODULES=y -CONFIG_MODULE_UNLOAD=y -CONFIG_OBSOLETE_MODPARM=y -# CONFIG_MODULE_SRCVERSION_ALL is not set -CONFIG_KMOD=y - -# -# Bus options (PCI, PCMCIA, EISA, MCA, ISA) -# - -# -# PCCARD (PCMCIA/CardBus) support -# -# CONFIG_PCCARD is not set - -# -# PCI Hotplug Support -# - -# -# Executable file formats -# -CONFIG_BINFMT_FLAT=y -# CONFIG_BINFMT_ZFLAT is not set -# CONFIG_BINFMT_SHARED_FLAT is not set -# CONFIG_BINFMT_MISC is not set - -# -# Networking -# -# CONFIG_NET is not set - -# -# Generic Driver Options -# -CONFIG_STANDALONE=y -CONFIG_PREVENT_FIRMWARE_BUILD=y -# CONFIG_FW_LOADER is not set -# CONFIG_DEBUG_DRIVER is not set - -# -# Memory Technology Devices (MTD) -# -CONFIG_MTD=y -# CONFIG_MTD_DEBUG is not set -# CONFIG_MTD_CONCAT is not set -# CONFIG_MTD_PARTITIONS is not set - -# -# User Modules And Translation Layers -# -# CONFIG_MTD_CHAR is not set -CONFIG_MTD_BLOCK=y -# CONFIG_FTL is not set -# CONFIG_NFTL is not set -# CONFIG_INFTL is not set - -# -# RAM/ROM/Flash chip drivers -# -# CONFIG_MTD_CFI is not set -# CONFIG_MTD_JEDECPROBE is not set -CONFIG_MTD_MAP_BANK_WIDTH_1=y -CONFIG_MTD_MAP_BANK_WIDTH_2=y -CONFIG_MTD_MAP_BANK_WIDTH_4=y -# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set -# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set -# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set -CONFIG_MTD_CFI_I1=y -CONFIG_MTD_CFI_I2=y -# CONFIG_MTD_CFI_I4 is not set -# CONFIG_MTD_CFI_I8 is not set -# CONFIG_MTD_RAM is not set -# CONFIG_MTD_ROM is not set -# CONFIG_MTD_ABSENT is not set - -# -# Mapping drivers for chip access -# -# CONFIG_MTD_COMPLEX_MAPPINGS is not set -# CONFIG_MTD_PLATRAM is not set - -# -# Self-contained MTD device drivers -# -CONFIG_MTD_SLRAM=y -# CONFIG_MTD_PHRAM is not set -# CONFIG_MTD_MTDRAM is not set -# CONFIG_MTD_BLKMTD is not set - -# -# Disk-On-Chip Device Drivers -# -# CONFIG_MTD_DOC2000 is not set -# CONFIG_MTD_DOC2001 is not set -# CONFIG_MTD_DOC2001PLUS is not set - -# -# NAND Flash Device Drivers -# -# CONFIG_MTD_NAND is not set - -# -# Parallel port support -# -# CONFIG_PARPORT is not set - -# -# Block devices -# -# CONFIG_BLK_DEV_FD is not set -# CONFIG_BLK_DEV_COW_COMMON is not set -# CONFIG_BLK_DEV_LOOP is not set -# CONFIG_BLK_DEV_RAM is not set -CONFIG_BLK_DEV_RAM_COUNT=16 -CONFIG_INITRAMFS_SOURCE="" -# CONFIG_CDROM_PKTCDVD is not set - -# -# IO Schedulers -# -CONFIG_IOSCHED_NOOP=y -# CONFIG_IOSCHED_AS is not set -# CONFIG_IOSCHED_DEADLINE is not set -# CONFIG_IOSCHED_CFQ is not set - -# -# Disk device support -# - -# -# ATA/ATAPI/MFM/RLL support -# -# CONFIG_IDE is not set - -# -# SCSI device support -# -# CONFIG_SCSI is not set - -# -# Multi-device support (RAID and LVM) -# -# CONFIG_MD is not set - -# -# Fusion MPT device support -# -# CONFIG_FUSION is not set - -# -# IEEE 1394 (FireWire) support -# - -# -# I2O device support -# - -# -# Network device support -# -# CONFIG_NETPOLL is not set -# CONFIG_NET_POLL_CONTROLLER is not set - -# -# ISDN subsystem -# - -# -# Input device support -# -CONFIG_INPUT=y - -# -# Userland interfaces -# -# CONFIG_INPUT_MOUSEDEV is not set -# CONFIG_INPUT_JOYDEV is not set -# CONFIG_INPUT_TSDEV is not set -# CONFIG_INPUT_EVDEV is not set -# CONFIG_INPUT_EVBUG is not set - -# -# Input Device Drivers -# -# CONFIG_INPUT_KEYBOARD is not set -# CONFIG_INPUT_MOUSE is not set -# CONFIG_INPUT_JOYSTICK is not set -# CONFIG_INPUT_TOUCHSCREEN is not set -# CONFIG_INPUT_MISC is not set - -# -# Hardware I/O ports -# -CONFIG_SERIO=y -# CONFIG_SERIO_I8042 is not set -# CONFIG_SERIO_SERPORT is not set -# CONFIG_SERIO_LIBPS2 is not set -# CONFIG_SERIO_RAW is not set -# CONFIG_GAMEPORT is not set - -# -# Character devices -# -# CONFIG_VT is not set -# CONFIG_SERIAL_NONSTANDARD is not set - -# -# Serial drivers -# -CONFIG_SERIAL_8250=y -CONFIG_SERIAL_8250_CONSOLE=y -CONFIG_SERIAL_8250_NR_UARTS=1 -# CONFIG_SERIAL_8250_EXTENDED is not set - -# -# Non-8250 serial port support -# -# CONFIG_V850E_UART is not set -CONFIG_SERIAL_CORE=y -CONFIG_SERIAL_CORE_CONSOLE=y -# CONFIG_UNIX98_PTYS is not set -# CONFIG_LEGACY_PTYS is not set - -# -# IPMI -# -# CONFIG_IPMI_HANDLER is not set - -# -# Watchdog Cards -# -# CONFIG_WATCHDOG is not set -# CONFIG_RTC is not set -# CONFIG_GEN_RTC is not set -# CONFIG_DTLK is not set -# CONFIG_R3964 is not set - -# -# Ftape, the floppy tape device driver -# -# CONFIG_RAW_DRIVER is not set - -# -# TPM devices -# - -# -# Multimedia devices -# -# CONFIG_VIDEO_DEV is not set - -# -# Digital Video Broadcasting Devices -# - -# -# File systems -# -# CONFIG_EXT2_FS is not set -# CONFIG_EXT3_FS is not set -# CONFIG_JBD is not set -# CONFIG_REISERFS_FS is not set -# CONFIG_JFS_FS is not set -# CONFIG_FS_POSIX_ACL is not set - -# -# XFS support -# -# CONFIG_XFS_FS is not set -# CONFIG_MINIX_FS is not set -CONFIG_ROMFS_FS=y -# CONFIG_MAGIC_ROM_PTR is not set -CONFIG_INOTIFY=y -# CONFIG_QUOTA is not set -CONFIG_DNOTIFY=y -# CONFIG_AUTOFS_FS is not set -# CONFIG_AUTOFS4_FS is not set - -# -# CD-ROM/DVD Filesystems -# -# CONFIG_ISO9660_FS is not set -# CONFIG_UDF_FS is not set - -# -# DOS/FAT/NT Filesystems -# -# CONFIG_MSDOS_FS is not set -# CONFIG_VFAT_FS is not set -# CONFIG_NTFS_FS is not set - -# -# Pseudo filesystems -# -CONFIG_PROC_FS=y -CONFIG_SYSFS=y -# CONFIG_TMPFS is not set -# CONFIG_HUGETLB_PAGE is not set -CONFIG_RAMFS=y - -# -# Miscellaneous filesystems -# -# CONFIG_HFSPLUS_FS is not set -# CONFIG_JFFS_FS is not set -# CONFIG_JFFS2_FS is not set -# CONFIG_CRAMFS is not set -# CONFIG_VXFS_FS is not set -# CONFIG_HPFS_FS is not set -# CONFIG_QNX4FS_FS is not set -# CONFIG_SYSV_FS is not set -# CONFIG_UFS_FS is not set - -# -# Partition Types -# -# CONFIG_PARTITION_ADVANCED is not set -CONFIG_MSDOS_PARTITION=y - -# -# Native Language Support -# -# CONFIG_NLS is not set - -# -# Graphics support -# -# CONFIG_FB is not set - -# -# Sound -# -# CONFIG_SOUND is not set - -# -# USB support -# -# CONFIG_USB_ARCH_HAS_HCD is not set -# CONFIG_USB_ARCH_HAS_OHCI is not set - -# -# USB Gadget Support -# -# CONFIG_USB_GADGET is not set - -# -# Kernel hacking -# -# CONFIG_PRINTK_TIME is not set -CONFIG_DEBUG_KERNEL=y -# CONFIG_MAGIC_SYSRQ is not set -CONFIG_LOG_BUF_SHIFT=14 -# CONFIG_SCHEDSTATS is not set -# CONFIG_DEBUG_SLAB is not set -# CONFIG_DEBUG_SPINLOCK is not set -# CONFIG_DEBUG_SPINLOCK_SLEEP is not set -# CONFIG_DEBUG_KOBJECT is not set -CONFIG_DEBUG_INFO=y -# CONFIG_DEBUG_FS is not set -# CONFIG_NO_KERNEL_MSG is not set - -# -# Security options -# -# CONFIG_KEYS is not set -# CONFIG_SECURITY is not set - -# -# Cryptographic options -# -# CONFIG_CRYPTO is not set - -# -# Hardware crypto devices -# - -# -# Library routines -# -# CONFIG_CRC_CCITT is not set -# CONFIG_CRC32 is not set -# CONFIG_LIBCRC32C is not set diff --git a/arch/v850/configs/sim_defconfig b/arch/v850/configs/sim_defconfig deleted file mode 100644 index f31ba7398ad0..000000000000 --- a/arch/v850/configs/sim_defconfig +++ /dev/null @@ -1,451 +0,0 @@ -# -# Automatically generated make config: don't edit -# Linux kernel version: 2.6.13-uc0 -# Fri Sep 2 13:36:43 2005 -# -# CONFIG_MMU is not set -# CONFIG_UID16 is not set -CONFIG_RWSEM_GENERIC_SPINLOCK=y -# CONFIG_RWSEM_XCHGADD_ALGORITHM is not set -CONFIG_GENERIC_CALIBRATE_DELAY=y -# CONFIG_ISA is not set -# CONFIG_ISAPNP is not set -# CONFIG_EISA is not set -# CONFIG_MCA is not set -CONFIG_V850=y - -# -# Processor type and features -# -CONFIG_V850E_SIM=y -# CONFIG_RTE_CB_MA1 is not set -# CONFIG_RTE_CB_NB85E is not set -# CONFIG_RTE_CB_ME2 is not set -# CONFIG_V850E_AS85EP1 is not set -# CONFIG_V850E2_SIM85E2C is not set -# CONFIG_V850E2_SIM85E2S is not set -# CONFIG_V850E2_FPGA85E2C is not set -# CONFIG_V850E2_ANNA is not set -CONFIG_V850E=y -# CONFIG_PCI is not set -# CONFIG_V850E_INTC is not set -# CONFIG_V850E_TIMER_D is not set -# CONFIG_V850E_CACHE is not set -# CONFIG_V850E2_CACHE is not set -CONFIG_NO_CACHE=y -CONFIG_ZERO_BSS=y -# CONFIG_RESET_GUARD is not set -CONFIG_LARGE_ALLOCS=y -CONFIG_FLATMEM=y -CONFIG_FLAT_NODE_MEM_MAP=y - -# -# Code maturity level options -# -# CONFIG_EXPERIMENTAL is not set -CONFIG_CLEAN_COMPILE=y -CONFIG_BROKEN_ON_SMP=y -CONFIG_INIT_ENV_ARG_LIMIT=32 - -# -# General setup -# -CONFIG_LOCALVERSION="" -# CONFIG_BSD_PROCESS_ACCT is not set -# CONFIG_SYSCTL is not set -# CONFIG_HOTPLUG is not set -# CONFIG_IKCONFIG is not set -CONFIG_EMBEDDED=y -# CONFIG_KALLSYMS is not set -CONFIG_PRINTK=y -CONFIG_BUG=y -# CONFIG_BASE_FULL is not set -# CONFIG_FUTEX is not set -# CONFIG_EPOLL is not set -CONFIG_CC_OPTIMIZE_FOR_SIZE=y -CONFIG_CC_ALIGN_FUNCTIONS=0 -CONFIG_CC_ALIGN_LABELS=0 -CONFIG_CC_ALIGN_LOOPS=0 -CONFIG_CC_ALIGN_JUMPS=0 -CONFIG_BASE_SMALL=1 - -# -# Loadable module support -# -CONFIG_MODULES=y -CONFIG_MODULE_UNLOAD=y -CONFIG_OBSOLETE_MODPARM=y -# CONFIG_MODULE_SRCVERSION_ALL is not set -CONFIG_KMOD=y - -# -# Bus options (PCI, PCMCIA, EISA, MCA, ISA) -# - -# -# PCCARD (PCMCIA/CardBus) support -# -# CONFIG_PCCARD is not set - -# -# PCI Hotplug Support -# - -# -# Executable file formats -# -CONFIG_BINFMT_FLAT=y -# CONFIG_BINFMT_ZFLAT is not set -# CONFIG_BINFMT_SHARED_FLAT is not set -# CONFIG_BINFMT_MISC is not set - -# -# Networking -# -# CONFIG_NET is not set - -# -# Generic Driver Options -# -CONFIG_STANDALONE=y -CONFIG_PREVENT_FIRMWARE_BUILD=y -# CONFIG_FW_LOADER is not set -# CONFIG_DEBUG_DRIVER is not set - -# -# Memory Technology Devices (MTD) -# -CONFIG_MTD=y -# CONFIG_MTD_DEBUG is not set -# CONFIG_MTD_CONCAT is not set -# CONFIG_MTD_PARTITIONS is not set - -# -# User Modules And Translation Layers -# -# CONFIG_MTD_CHAR is not set -CONFIG_MTD_BLOCK=y -# CONFIG_FTL is not set -# CONFIG_NFTL is not set -# CONFIG_INFTL is not set - -# -# RAM/ROM/Flash chip drivers -# -# CONFIG_MTD_CFI is not set -# CONFIG_MTD_JEDECPROBE is not set -CONFIG_MTD_MAP_BANK_WIDTH_1=y -CONFIG_MTD_MAP_BANK_WIDTH_2=y -CONFIG_MTD_MAP_BANK_WIDTH_4=y -# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set -# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set -# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set -CONFIG_MTD_CFI_I1=y -CONFIG_MTD_CFI_I2=y -# CONFIG_MTD_CFI_I4 is not set -# CONFIG_MTD_CFI_I8 is not set -# CONFIG_MTD_RAM is not set -# CONFIG_MTD_ROM is not set -# CONFIG_MTD_ABSENT is not set - -# -# Mapping drivers for chip access -# -# CONFIG_MTD_COMPLEX_MAPPINGS is not set -# CONFIG_MTD_PLATRAM is not set - -# -# Self-contained MTD device drivers -# -CONFIG_MTD_SLRAM=y -# CONFIG_MTD_PHRAM is not set -# CONFIG_MTD_MTDRAM is not set -# CONFIG_MTD_BLKMTD is not set - -# -# Disk-On-Chip Device Drivers -# -# CONFIG_MTD_DOC2000 is not set -# CONFIG_MTD_DOC2001 is not set -# CONFIG_MTD_DOC2001PLUS is not set - -# -# NAND Flash Device Drivers -# -# CONFIG_MTD_NAND is not set - -# -# Parallel port support -# -# CONFIG_PARPORT is not set - -# -# Block devices -# -# CONFIG_BLK_DEV_FD is not set -# CONFIG_BLK_DEV_COW_COMMON is not set -# CONFIG_BLK_DEV_LOOP is not set -# CONFIG_BLK_DEV_RAM is not set -CONFIG_BLK_DEV_RAM_COUNT=16 -CONFIG_INITRAMFS_SOURCE="" -# CONFIG_CDROM_PKTCDVD is not set - -# -# IO Schedulers -# -CONFIG_IOSCHED_NOOP=y -# CONFIG_IOSCHED_AS is not set -# CONFIG_IOSCHED_DEADLINE is not set -# CONFIG_IOSCHED_CFQ is not set - -# -# Disk device support -# - -# -# ATA/ATAPI/MFM/RLL support -# -# CONFIG_IDE is not set - -# -# SCSI device support -# -# CONFIG_SCSI is not set - -# -# Multi-device support (RAID and LVM) -# -# CONFIG_MD is not set - -# -# Fusion MPT device support -# -# CONFIG_FUSION is not set - -# -# IEEE 1394 (FireWire) support -# - -# -# I2O device support -# - -# -# Network device support -# -# CONFIG_NETPOLL is not set -# CONFIG_NET_POLL_CONTROLLER is not set - -# -# ISDN subsystem -# - -# -# Input device support -# -CONFIG_INPUT=y - -# -# Userland interfaces -# -# CONFIG_INPUT_MOUSEDEV is not set -# CONFIG_INPUT_JOYDEV is not set -# CONFIG_INPUT_TSDEV is not set -# CONFIG_INPUT_EVDEV is not set -# CONFIG_INPUT_EVBUG is not set - -# -# Input Device Drivers -# -# CONFIG_INPUT_KEYBOARD is not set -# CONFIG_INPUT_MOUSE is not set -# CONFIG_INPUT_JOYSTICK is not set -# CONFIG_INPUT_TOUCHSCREEN is not set -# CONFIG_INPUT_MISC is not set - -# -# Hardware I/O ports -# -CONFIG_SERIO=y -# CONFIG_SERIO_I8042 is not set -# CONFIG_SERIO_SERPORT is not set -# CONFIG_SERIO_LIBPS2 is not set -# CONFIG_SERIO_RAW is not set -# CONFIG_GAMEPORT is not set - -# -# Character devices -# -# CONFIG_VT is not set -# CONFIG_SERIAL_NONSTANDARD is not set - -# -# Serial drivers -# -# CONFIG_SERIAL_8250 is not set - -# -# Non-8250 serial port support -# -# CONFIG_UNIX98_PTYS is not set -# CONFIG_LEGACY_PTYS is not set - -# -# IPMI -# -# CONFIG_IPMI_HANDLER is not set - -# -# Watchdog Cards -# -# CONFIG_WATCHDOG is not set -# CONFIG_RTC is not set -# CONFIG_GEN_RTC is not set -# CONFIG_DTLK is not set -# CONFIG_R3964 is not set - -# -# Ftape, the floppy tape device driver -# -# CONFIG_RAW_DRIVER is not set - -# -# TPM devices -# - -# -# Multimedia devices -# -# CONFIG_VIDEO_DEV is not set - -# -# Digital Video Broadcasting Devices -# - -# -# File systems -# -# CONFIG_EXT2_FS is not set -# CONFIG_EXT3_FS is not set -# CONFIG_JBD is not set -# CONFIG_REISERFS_FS is not set -# CONFIG_JFS_FS is not set -# CONFIG_FS_POSIX_ACL is not set - -# -# XFS support -# -# CONFIG_XFS_FS is not set -# CONFIG_MINIX_FS is not set -CONFIG_ROMFS_FS=y -# CONFIG_MAGIC_ROM_PTR is not set -CONFIG_INOTIFY=y -# CONFIG_QUOTA is not set -CONFIG_DNOTIFY=y -# CONFIG_AUTOFS_FS is not set -# CONFIG_AUTOFS4_FS is not set - -# -# CD-ROM/DVD Filesystems -# -# CONFIG_ISO9660_FS is not set -# CONFIG_UDF_FS is not set - -# -# DOS/FAT/NT Filesystems -# -# CONFIG_MSDOS_FS is not set -# CONFIG_VFAT_FS is not set -# CONFIG_NTFS_FS is not set - -# -# Pseudo filesystems -# -CONFIG_PROC_FS=y -CONFIG_SYSFS=y -# CONFIG_TMPFS is not set -# CONFIG_HUGETLB_PAGE is not set -CONFIG_RAMFS=y - -# -# Miscellaneous filesystems -# -# CONFIG_HFSPLUS_FS is not set -# CONFIG_JFFS_FS is not set -# CONFIG_JFFS2_FS is not set -# CONFIG_CRAMFS is not set -# CONFIG_VXFS_FS is not set -# CONFIG_HPFS_FS is not set -# CONFIG_QNX4FS_FS is not set -# CONFIG_SYSV_FS is not set -# CONFIG_UFS_FS is not set - -# -# Partition Types -# -# CONFIG_PARTITION_ADVANCED is not set -CONFIG_MSDOS_PARTITION=y - -# -# Native Language Support -# -# CONFIG_NLS is not set - -# -# Graphics support -# -# CONFIG_FB is not set - -# -# Sound -# -# CONFIG_SOUND is not set - -# -# USB support -# -# CONFIG_USB_ARCH_HAS_HCD is not set -# CONFIG_USB_ARCH_HAS_OHCI is not set - -# -# USB Gadget Support -# -# CONFIG_USB_GADGET is not set - -# -# Kernel hacking -# -# CONFIG_PRINTK_TIME is not set -CONFIG_DEBUG_KERNEL=y -# CONFIG_MAGIC_SYSRQ is not set -CONFIG_LOG_BUF_SHIFT=14 -# CONFIG_SCHEDSTATS is not set -# CONFIG_DEBUG_SLAB is not set -# CONFIG_DEBUG_SPINLOCK is not set -# CONFIG_DEBUG_SPINLOCK_SLEEP is not set -# CONFIG_DEBUG_KOBJECT is not set -CONFIG_DEBUG_INFO=y -# CONFIG_DEBUG_FS is not set -# CONFIG_NO_KERNEL_MSG is not set - -# -# Security options -# -# CONFIG_KEYS is not set -# CONFIG_SECURITY is not set - -# -# Cryptographic options -# -# CONFIG_CRYPTO is not set - -# -# Hardware crypto devices -# - -# -# Library routines -# -# CONFIG_CRC_CCITT is not set -# CONFIG_CRC32 is not set -# CONFIG_LIBCRC32C is not set diff --git a/arch/v850/kernel/Makefile b/arch/v850/kernel/Makefile deleted file mode 100644 index da5889c53576..000000000000 --- a/arch/v850/kernel/Makefile +++ /dev/null @@ -1,40 +0,0 @@ -# -# arch/v850/kernel/Makefile -# -# Copyright (C) 2001,02,03 NEC Electronics Corporation -# Copyright (C) 2001,02,03 Miles Bader -# -# This file is subject to the terms and conditions of the GNU General Public -# License. See the file "COPYING" in the main directory of this archive -# for more details. -# - -extra-y := head.o init_task.o vmlinux.lds - -obj-y += intv.o entry.o process.o syscalls.o time.o setup.o \ - signal.o irq.o mach.o ptrace.o bug.o -obj-$(CONFIG_MODULES) += module.o v850_ksyms.o -# chip-specific code -obj-$(CONFIG_V850E_MA1) += ma.o -obj-$(CONFIG_V850E_ME2) += me2.o -obj-$(CONFIG_V850E_TEG) += teg.o -obj-$(CONFIG_V850E_AS85EP1) += as85ep1.o -obj-$(CONFIG_V850E2_ANNA) += anna.o -# platform-specific code -obj-$(CONFIG_V850E_SIM) += sim.o simcons.o -obj-$(CONFIG_V850E2_SIM85E2) += sim85e2.o memcons.o -obj-$(CONFIG_V850E2_FPGA85E2C) += fpga85e2c.o memcons.o -obj-$(CONFIG_RTE_CB) += rte_cb.o rte_cb_leds.o -obj-$(CONFIG_RTE_CB_MA1) += rte_ma1_cb.o -obj-$(CONFIG_RTE_CB_ME2) += rte_me2_cb.o -obj-$(CONFIG_RTE_CB_NB85E) += rte_nb85e_cb.o -obj-$(CONFIG_RTE_CB_MULTI) += rte_cb_multi.o -obj-$(CONFIG_RTE_MB_A_PCI) += rte_mb_a_pci.o -obj-$(CONFIG_RTE_GBUS_INT) += gbus_int.o -# feature-specific code -obj-$(CONFIG_V850E_INTC) += v850e_intc.o -obj-$(CONFIG_V850E_TIMER_D) += v850e_timer_d.o v850e_utils.o -obj-$(CONFIG_V850E_CACHE) += v850e_cache.o -obj-$(CONFIG_V850E2_CACHE) += v850e2_cache.o -obj-$(CONFIG_V850E_HIGHRES_TIMER) += highres_timer.o -obj-$(CONFIG_PROC_FS) += procfs.o diff --git a/arch/v850/kernel/anna-rom.ld b/arch/v850/kernel/anna-rom.ld deleted file mode 100644 index 7c54e7e3f1b1..000000000000 --- a/arch/v850/kernel/anna-rom.ld +++ /dev/null @@ -1,16 +0,0 @@ -/* Linker script for the Midas labs Anna V850E2 evaluation board - (CONFIG_V850E2_ANNA), with kernel in ROM (CONFIG_ROM_KERNEL). */ - -MEMORY { - /* 8MB of flash ROM. */ - ROM : ORIGIN = 0, LENGTH = 0x00800000 - - /* 1MB of static RAM. This memory is mirrored 64 times. */ - SRAM : ORIGIN = SRAM_ADDR, LENGTH = SRAM_SIZE - /* 64MB of DRAM. */ - SDRAM : ORIGIN = SDRAM_ADDR, LENGTH = SDRAM_SIZE -} - -SECTIONS { - ROMK_SECTIONS(ROM, SRAM) -} diff --git a/arch/v850/kernel/anna.c b/arch/v850/kernel/anna.c deleted file mode 100644 index 5978a25170fb..000000000000 --- a/arch/v850/kernel/anna.c +++ /dev/null @@ -1,202 +0,0 @@ -/* - * arch/v850/kernel/anna.c -- Anna V850E2 evaluation chip/board - * - * Copyright (C) 2002,03 NEC Electronics Corporation - * Copyright (C) 2002,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include -#include - -#include "mach.h" - - -/* SRAM and SDRAM are vaguely contiguous (with a big hole in between; see - mach_reserve_bootmem for details); use both as one big area. */ -#define RAM_START SRAM_ADDR -#define RAM_END (SDRAM_ADDR + SDRAM_SIZE) - -/* The bits of this port are connected to an 8-LED bar-graph. */ -#define LEDS_PORT 0 - - -static void anna_led_tick (void); - - -void __init mach_early_init (void) -{ - ANNA_ILBEN = 0; - - V850E2_CSC(0) = 0x402F; - V850E2_CSC(1) = 0x4000; - V850E2_BPC = 0; - V850E2_BSC = 0xAAAA; - V850E2_BEC = 0; - -#if 0 - V850E2_BHC = 0xFFFF; /* icache all memory, dcache all */ -#else - V850E2_BHC = 0; /* cache no memory */ -#endif - V850E2_BCT(0) = 0xB088; - V850E2_BCT(1) = 0x0008; - V850E2_DWC(0) = 0x0027; - V850E2_DWC(1) = 0; - V850E2_BCC = 0x0006; - V850E2_ASC = 0; - V850E2_LBS = 0x0089; - V850E2_SCR(3) = 0x21A9; - V850E2_RFS(3) = 0x8121; - - v850e_intc_disable_irqs (); -} - -void __init mach_setup (char **cmdline) -{ - ANNA_PORT_PM (LEDS_PORT) = 0; /* Make all LED pins output pins. */ - mach_tick = anna_led_tick; -} - -void __init mach_get_physical_ram (unsigned long *ram_start, - unsigned long *ram_len) -{ - *ram_start = RAM_START; - *ram_len = RAM_END - RAM_START; -} - -void __init mach_reserve_bootmem () -{ - /* The space between SRAM and SDRAM is filled with duplicate - images of SRAM. Prevent the kernel from using them. */ - reserve_bootmem (SRAM_ADDR + SRAM_SIZE, - SDRAM_ADDR - (SRAM_ADDR + SRAM_SIZE), - BOOTMEM_DEFAULT); -} - -void mach_gettimeofday (struct timespec *tv) -{ - tv->tv_sec = 0; - tv->tv_nsec = 0; -} - -void __init mach_sched_init (struct irqaction *timer_action) -{ - /* Start hardware timer. */ - v850e_timer_d_configure (0, HZ); - /* Install timer interrupt handler. */ - setup_irq (IRQ_INTCMD(0), timer_action); -} - -static struct v850e_intc_irq_init irq_inits[] = { - { "IRQ", 0, NUM_MACH_IRQS, 1, 7 }, - { "PIN", IRQ_INTP(0), IRQ_INTP_NUM, 1, 4 }, - { "CCC", IRQ_INTCCC(0), IRQ_INTCCC_NUM, 1, 5 }, - { "CMD", IRQ_INTCMD(0), IRQ_INTCMD_NUM, 1, 5 }, - { "DMA", IRQ_INTDMA(0), IRQ_INTDMA_NUM, 1, 2 }, - { "DMXER", IRQ_INTDMXER,1, 1, 2 }, - { "SRE", IRQ_INTSRE(0), IRQ_INTSRE_NUM, 3, 3 }, - { "SR", IRQ_INTSR(0), IRQ_INTSR_NUM, 3, 4 }, - { "ST", IRQ_INTST(0), IRQ_INTST_NUM, 3, 5 }, - { 0 } -}; -#define NUM_IRQ_INITS (ARRAY_SIZE(irq_inits) - 1) - -static struct hw_interrupt_type hw_itypes[NUM_IRQ_INITS]; - -void __init mach_init_irqs (void) -{ - v850e_intc_init_irq_types (irq_inits, hw_itypes); -} - -void machine_restart (char *__unused) -{ -#ifdef CONFIG_RESET_GUARD - disable_reset_guard (); -#endif - asm ("jmp r0"); /* Jump to the reset vector. */ -} - -void machine_halt (void) -{ -#ifdef CONFIG_RESET_GUARD - disable_reset_guard (); -#endif - local_irq_disable (); /* Ignore all interrupts. */ - ANNA_PORT_IO(LEDS_PORT) = 0xAA; /* Note that we halted. */ - for (;;) - asm ("halt; nop; nop; nop; nop; nop"); -} - -void machine_power_off (void) -{ - machine_halt (); -} - -/* Called before configuring an on-chip UART. */ -void anna_uart_pre_configure (unsigned chan, unsigned cflags, unsigned baud) -{ - /* The Anna connects some general-purpose I/O pins on the CPU to - the RTS/CTS lines of UART 1's serial connection. I/O pins P07 - and P37 are RTS and CTS respectively. */ - if (chan == 1) { - ANNA_PORT_PM(0) &= ~0x80; /* P07 in output mode */ - ANNA_PORT_PM(3) |= 0x80; /* P37 in input mode */ - } -} - -/* Minimum and maximum bounds for the moving upper LED boundary in the - clock tick display. We can't use the last bit because it's used for - UART0's CTS output. */ -#define MIN_MAX_POS 0 -#define MAX_MAX_POS 6 - -/* There are MAX_MAX_POS^2 - MIN_MAX_POS^2 cycles in the animation, so if - we pick 6 and 0 as above, we get 49 cycles, which is when divided into - the standard 100 value for HZ, gives us an almost 1s total time. */ -#define TICKS_PER_FRAME \ - (HZ / (MAX_MAX_POS * MAX_MAX_POS - MIN_MAX_POS * MIN_MAX_POS)) - -static void anna_led_tick () -{ - static unsigned counter = 0; - - if (++counter == TICKS_PER_FRAME) { - static int pos = 0, max_pos = MAX_MAX_POS, dir = 1; - - if (dir > 0 && pos == max_pos) { - dir = -1; - if (max_pos == MIN_MAX_POS) - max_pos = MAX_MAX_POS; - else - max_pos--; - } else { - if (dir < 0 && pos == 0) - dir = 1; - - if (pos + dir <= max_pos) { - /* Each bit of port 0 has a LED. */ - clear_bit (pos, &ANNA_PORT_IO(LEDS_PORT)); - pos += dir; - set_bit (pos, &ANNA_PORT_IO(LEDS_PORT)); - } - } - - counter = 0; - } -} diff --git a/arch/v850/kernel/anna.ld b/arch/v850/kernel/anna.ld deleted file mode 100644 index df7f80f2833d..000000000000 --- a/arch/v850/kernel/anna.ld +++ /dev/null @@ -1,20 +0,0 @@ -/* Linker script for the Midas labs Anna V850E2 evaluation board - (CONFIG_V850E2_ANNA). */ - -MEMORY { - /* 256KB of internal memory (followed by one mirror). */ - iMEM0 : ORIGIN = 0, LENGTH = 0x00040000 - /* 256KB of internal memory (followed by one mirror). */ - iMEM1 : ORIGIN = 0x00040000, LENGTH = 0x00040000 - - /* 1MB of static RAM. This memory is mirrored 64 times. */ - SRAM : ORIGIN = SRAM_ADDR, LENGTH = SRAM_SIZE - /* 64MB of DRAM. */ - SDRAM : ORIGIN = SDRAM_ADDR, LENGTH = SDRAM_SIZE -} - -SECTIONS { - .intv : { INTV_CONTENTS } > iMEM0 - .sram : { RAMK_KRAM_CONTENTS } > SRAM - .root : { ROOT_FS_CONTENTS } > SDRAM -} diff --git a/arch/v850/kernel/as85ep1-rom.ld b/arch/v850/kernel/as85ep1-rom.ld deleted file mode 100644 index fe2a9a3ab525..000000000000 --- a/arch/v850/kernel/as85ep1-rom.ld +++ /dev/null @@ -1,21 +0,0 @@ -/* Linker script for the NEC AS85EP1 V850E evaluation board - (CONFIG_V850E_AS85EP1), with kernel in ROM (CONFIG_ROM_KERNEL). */ - -MEMORY { - /* 4MB of flash ROM. */ - ROM : ORIGIN = 0, LENGTH = 0x00400000 - - /* 1MB of static RAM. */ - SRAM : ORIGIN = SRAM_ADDR, LENGTH = SRAM_SIZE - - /* About 58MB of DRAM. This can actually be at one of two - positions, determined by jumper JP3; we have to use the first - position because the second is partially out of processor - instruction addressing range (though in the second position - there's actually 64MB available). */ - SDRAM : ORIGIN = SDRAM_ADDR, LENGTH = SDRAM_SIZE -} - -SECTIONS { - ROMK_SECTIONS(ROM, SRAM) -} diff --git a/arch/v850/kernel/as85ep1.c b/arch/v850/kernel/as85ep1.c deleted file mode 100644 index b525ecf3aea4..000000000000 --- a/arch/v850/kernel/as85ep1.c +++ /dev/null @@ -1,234 +0,0 @@ -/* - * arch/v850/kernel/as85ep1.c -- AS85EP1 V850E evaluation chip/board - * - * Copyright (C) 2002,03 NEC Electronics Corporation - * Copyright (C) 2002,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include -#include - -#include "mach.h" - - -/* SRAM and SDRAM are vaguely contiguous (with a big hole in between; see - mach_reserve_bootmem for details); use both as one big area. */ -#define RAM_START SRAM_ADDR -#define RAM_END (SDRAM_ADDR + SDRAM_SIZE) - -/* The bits of this port are connected to an 8-LED bar-graph. */ -#define LEDS_PORT 4 - - -static void as85ep1_led_tick (void); - -extern char _intv_copy_src_start, _intv_copy_src_end; -extern char _intv_copy_dst_start; - - -void __init mach_early_init (void) -{ -#ifndef CONFIG_ROM_KERNEL - const u32 *src; - register u32 *dst asm ("ep"); -#endif - - AS85EP1_CSC(0) = 0x0403; - AS85EP1_BCT(0) = 0xB8B8; - AS85EP1_DWC(0) = 0x0104; - AS85EP1_BCC = 0x0012; - AS85EP1_ASC = 0; - AS85EP1_LBS = 0x00A9; - - AS85EP1_PORT_PMC(6) = 0xFF; /* valid A0,A1,A20-A25 */ - AS85EP1_PORT_PMC(7) = 0x0E; /* valid CS1-CS3 */ - AS85EP1_PORT_PMC(9) = 0xFF; /* valid D16-D23 */ - AS85EP1_PORT_PMC(10) = 0xFF; /* valid D24-D31 */ - - AS85EP1_RFS(1) = 0x800c; - AS85EP1_RFS(3) = 0x800c; - AS85EP1_SCR(1) = 0x20A9; - AS85EP1_SCR(3) = 0x20A9; - -#ifndef CONFIG_ROM_KERNEL - /* The early chip we have is buggy, and writing the interrupt - vectors into low RAM may screw up, so for non-ROM kernels, we - only rely on the reset vector being downloaded, and copy the - rest of the interrupt vectors into place here. The specific bug - is that writing address N, where (N & 0x10) == 0x10, will _also_ - write to address (N - 0x10). We avoid this (effectively) by - writing in 16-byte chunks backwards from the end. */ - - AS85EP1_IRAMM = 0x3; /* "write-mode" for the internal instruction memory */ - - src = (u32 *)(((u32)&_intv_copy_src_end - 1) & ~0xF); - dst = (u32 *)&_intv_copy_dst_start - + (src - (u32 *)&_intv_copy_src_start); - do { - u32 t0 = src[0], t1 = src[1], t2 = src[2], t3 = src[3]; - dst[0] = t0; dst[1] = t1; dst[2] = t2; dst[3] = t3; - dst -= 4; - src -= 4; - } while (src > (u32 *)&_intv_copy_src_start); - - AS85EP1_IRAMM = 0x0; /* "read-mode" for the internal instruction memory */ -#endif /* !CONFIG_ROM_KERNEL */ - - v850e_intc_disable_irqs (); -} - -void __init mach_setup (char **cmdline) -{ - AS85EP1_PORT_PMC (LEDS_PORT) = 0; /* Make the LEDs port an I/O port. */ - AS85EP1_PORT_PM (LEDS_PORT) = 0; /* Make all the bits output pins. */ - mach_tick = as85ep1_led_tick; -} - -void __init mach_get_physical_ram (unsigned long *ram_start, - unsigned long *ram_len) -{ - *ram_start = RAM_START; - *ram_len = RAM_END - RAM_START; -} - -/* Convenience macros. */ -#define SRAM_END (SRAM_ADDR + SRAM_SIZE) -#define SDRAM_END (SDRAM_ADDR + SDRAM_SIZE) - -void __init mach_reserve_bootmem () -{ - if (SDRAM_ADDR < RAM_END && SDRAM_ADDR > RAM_START) - /* We can't use the space between SRAM and SDRAM, so - prevent the kernel from trying. */ - reserve_bootmem(SRAM_END, SDRAM_ADDR - SRAM_END, - BOOTMEM_DEFAULT); -} - -void mach_gettimeofday (struct timespec *tv) -{ - tv->tv_sec = 0; - tv->tv_nsec = 0; -} - -void __init mach_sched_init (struct irqaction *timer_action) -{ - /* Start hardware timer. */ - v850e_timer_d_configure (0, HZ); - /* Install timer interrupt handler. */ - setup_irq (IRQ_INTCMD(0), timer_action); -} - -static struct v850e_intc_irq_init irq_inits[] = { - { "IRQ", 0, NUM_MACH_IRQS, 1, 7 }, - { "CCC", IRQ_INTCCC(0), IRQ_INTCCC_NUM, 1, 5 }, - { "CMD", IRQ_INTCMD(0), IRQ_INTCMD_NUM, 1, 5 }, - { "SRE", IRQ_INTSRE(0), IRQ_INTSRE_NUM, 3, 3 }, - { "SR", IRQ_INTSR(0), IRQ_INTSR_NUM, 3, 4 }, - { "ST", IRQ_INTST(0), IRQ_INTST_NUM, 3, 5 }, - { 0 } -}; -#define NUM_IRQ_INITS (ARRAY_SIZE(irq_inits) - 1) - -static struct hw_interrupt_type hw_itypes[NUM_IRQ_INITS]; - -void __init mach_init_irqs (void) -{ - v850e_intc_init_irq_types (irq_inits, hw_itypes); -} - -void machine_restart (char *__unused) -{ -#ifdef CONFIG_RESET_GUARD - disable_reset_guard (); -#endif - asm ("jmp r0"); /* Jump to the reset vector. */ -} - -void machine_halt (void) -{ -#ifdef CONFIG_RESET_GUARD - disable_reset_guard (); -#endif - local_irq_disable (); /* Ignore all interrupts. */ - AS85EP1_PORT_IO (LEDS_PORT) = 0xAA; /* Note that we halted. */ - for (;;) - asm ("halt; nop; nop; nop; nop; nop"); -} - -void machine_power_off (void) -{ - machine_halt (); -} - -/* Called before configuring an on-chip UART. */ -void as85ep1_uart_pre_configure (unsigned chan, unsigned cflags, unsigned baud) -{ - /* Make the shared uart/port pins be uart pins. */ - AS85EP1_PORT_PMC(3) |= (0x5 << chan); - - /* The AS85EP1 connects some general-purpose I/O pins on the CPU to - the RTS/CTS lines of UART 1's serial connection. I/O pins P53 - and P54 are RTS and CTS respectively. */ - if (chan == 1) { - /* Put P53 & P54 in I/O port mode. */ - AS85EP1_PORT_PMC(5) &= ~0x18; - /* Make P53 an output, and P54 an input. */ - AS85EP1_PORT_PM(5) |= 0x10; - } -} - -/* Minimum and maximum bounds for the moving upper LED boundary in the - clock tick display. */ -#define MIN_MAX_POS 0 -#define MAX_MAX_POS 7 - -/* There are MAX_MAX_POS^2 - MIN_MAX_POS^2 cycles in the animation, so if - we pick 6 and 0 as above, we get 49 cycles, which is when divided into - the standard 100 value for HZ, gives us an almost 1s total time. */ -#define TICKS_PER_FRAME \ - (HZ / (MAX_MAX_POS * MAX_MAX_POS - MIN_MAX_POS * MIN_MAX_POS)) - -static void as85ep1_led_tick () -{ - static unsigned counter = 0; - - if (++counter == TICKS_PER_FRAME) { - static int pos = 0, max_pos = MAX_MAX_POS, dir = 1; - - if (dir > 0 && pos == max_pos) { - dir = -1; - if (max_pos == MIN_MAX_POS) - max_pos = MAX_MAX_POS; - else - max_pos--; - } else { - if (dir < 0 && pos == 0) - dir = 1; - - if (pos + dir <= max_pos) { - /* Each bit of port 0 has a LED. */ - set_bit (pos, &AS85EP1_PORT_IO(LEDS_PORT)); - pos += dir; - clear_bit (pos, &AS85EP1_PORT_IO(LEDS_PORT)); - } - } - - counter = 0; - } -} diff --git a/arch/v850/kernel/as85ep1.ld b/arch/v850/kernel/as85ep1.ld deleted file mode 100644 index ef2c4399063e..000000000000 --- a/arch/v850/kernel/as85ep1.ld +++ /dev/null @@ -1,49 +0,0 @@ -/* Linker script for the NEC AS85EP1 V850E evaluation board - (CONFIG_V850E_AS85EP1). */ - -MEMORY { - /* 1MB of internal instruction memory. */ - iMEM0 : ORIGIN = 0, LENGTH = 0x00100000 - - /* 1MB of static RAM. */ - SRAM : ORIGIN = SRAM_ADDR, LENGTH = SRAM_SIZE - - /* About 58MB of DRAM. This can actually be at one of two - positions, determined by jump JP3; we have to use the first - position because the second is partially out of processor - instruction addressing range (though in the second position - there's actually 64MB available). */ - SDRAM : ORIGIN = SDRAM_ADDR, LENGTH = SDRAM_SIZE -} - -SECTIONS { - .resetv : { - __intv_start = . ; - *(.intv.reset) /* Reset vector */ - } > iMEM0 - - .sram : { - RAMK_KRAM_CONTENTS - - /* We stick most of the interrupt vectors here; they'll be - copied into the proper location by the early init code (we - can't put them directly in the right place because of - hardware bugs). The vectors shouldn't need to be - relocated, so we don't have to use `> ... AT> ...' to - split the load/vm addresses (and we can't because of - problems with the loader). */ - . = ALIGN (0x10) ; - __intv_copy_src_start = . ; - *(.intv.common) /* Vectors common to all v850e proc. */ - *(.intv.mach) /* Machine-specific int. vectors. */ - . = ALIGN (0x10) ; - __intv_copy_src_end = . ; - } > SRAM - - /* Where we end up putting the vectors. */ - __intv_copy_dst_start = 0x10 ; - __intv_copy_dst_end = __intv_copy_dst_start + (__intv_copy_src_end - __intv_copy_src_start) ; - __intv_end = __intv_copy_dst_end ; - - .root : { ROOT_FS_CONTENTS } > SDRAM -} diff --git a/arch/v850/kernel/asm-offsets.c b/arch/v850/kernel/asm-offsets.c deleted file mode 100644 index 581e6986a776..000000000000 --- a/arch/v850/kernel/asm-offsets.c +++ /dev/null @@ -1,58 +0,0 @@ -/* - * This program is used to generate definitions needed by - * assembly language modules. - * - * We use the technique used in the OSF Mach kernel code: - * generate asm statements containing #defines, - * compile this file to assembler, and then extract the - * #defines from the assembly-language output. - */ - -#include -#include -#include -#include -#include -#include - -#include -#include - -int main (void) -{ - /* offsets into the task struct */ - DEFINE (TASK_STATE, offsetof (struct task_struct, state)); - DEFINE (TASK_FLAGS, offsetof (struct task_struct, flags)); - DEFINE (TASK_PTRACE, offsetof (struct task_struct, ptrace)); - DEFINE (TASK_BLOCKED, offsetof (struct task_struct, blocked)); - DEFINE (TASK_THREAD, offsetof (struct task_struct, thread)); - DEFINE (TASK_THREAD_INFO, offsetof (struct task_struct, stack)); - DEFINE (TASK_MM, offsetof (struct task_struct, mm)); - DEFINE (TASK_ACTIVE_MM, offsetof (struct task_struct, active_mm)); - DEFINE (TASK_PID, offsetof (struct task_struct, pid)); - - /* offsets into the kernel_stat struct */ - DEFINE (STAT_IRQ, offsetof (struct kernel_stat, irqs)); - - - /* signal defines */ - DEFINE (SIGSEGV, SIGSEGV); - DEFINE (SEGV_MAPERR, SEGV_MAPERR); - DEFINE (SIGTRAP, SIGTRAP); - DEFINE (SIGCHLD, SIGCHLD); - DEFINE (SIGILL, SIGILL); - DEFINE (TRAP_TRACE, TRAP_TRACE); - - /* ptrace flag bits */ - DEFINE (PT_PTRACED, PT_PTRACED); - DEFINE (PT_DTRACE, PT_DTRACE); - - /* error values */ - DEFINE (ENOSYS, ENOSYS); - - /* clone flag bits */ - DEFINE (CLONE_VFORK, CLONE_VFORK); - DEFINE (CLONE_VM, CLONE_VM); - - return 0; -} diff --git a/arch/v850/kernel/bug.c b/arch/v850/kernel/bug.c deleted file mode 100644 index c78cf750915a..000000000000 --- a/arch/v850/kernel/bug.c +++ /dev/null @@ -1,142 +0,0 @@ -/* - * arch/v850/kernel/bug.c -- Bug reporting functions - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include -#include -#include - -#include -#include -#include -#include - -/* We should use __builtin_return_address, but it doesn't work in gcc-2.90 - (which is currently our standard compiler on the v850). */ -#define ret_addr() ({ register u32 lp asm ("lp"); lp; }) -#define stack_addr() ({ register u32 sp asm ("sp"); sp; }) - -void __bug () -{ - printk (KERN_CRIT "kernel BUG at PC 0x%x (SP ~0x%x)!\n", - ret_addr() - 4, /* - 4 for `jarl' */ - stack_addr()); - machine_halt (); -} - -int bad_trap (int trap_num, struct pt_regs *regs) -{ - printk (KERN_CRIT - "unimplemented trap %d called at 0x%08lx, pid %d!\n", - trap_num, regs->pc, current->pid); - return -ENOSYS; -} - -#ifdef CONFIG_RESET_GUARD -void unexpected_reset (unsigned long ret_addr, unsigned long kmode, - struct task_struct *task, unsigned long sp) -{ - printk (KERN_CRIT - "unexpected reset in %s mode, pid %d" - " (ret_addr = 0x%lx, sp = 0x%lx)\n", - kmode ? "kernel" : "user", - task ? task->pid : -1, - ret_addr, sp); - - machine_halt (); -} -#endif /* CONFIG_RESET_GUARD */ - - - -struct spec_reg_name { - const char *name; - int gpr; -}; - -struct spec_reg_name spec_reg_names[] = { - { "sp", GPR_SP }, - { "gp", GPR_GP }, - { "tp", GPR_TP }, - { "ep", GPR_EP }, - { "lp", GPR_LP }, - { 0, 0 } -}; - -void show_regs (struct pt_regs *regs) -{ - int gpr_base, gpr_offs; - - printk (" pc 0x%08lx psw 0x%08lx kernel_mode %d\n", - regs->pc, regs->psw, regs->kernel_mode); - printk (" ctpc 0x%08lx ctpsw 0x%08lx ctbp 0x%08lx\n", - regs->ctpc, regs->ctpsw, regs->ctbp); - - for (gpr_base = 0; gpr_base < NUM_GPRS; gpr_base += 4) { - for (gpr_offs = 0; gpr_offs < 4; gpr_offs++) { - int gpr = gpr_base + gpr_offs; - long val = regs->gpr[gpr]; - struct spec_reg_name *srn; - - for (srn = spec_reg_names; srn->name; srn++) - if (srn->gpr == gpr) - break; - - if (srn->name) - printk ("%7s 0x%08lx", srn->name, val); - else - printk (" r%02d 0x%08lx", gpr, val); - } - - printk ("\n"); - } -} - -/* - * TASK is a pointer to the task whose backtrace we want to see (or NULL - * for current task), SP is the stack pointer of the first frame that - * should be shown in the back trace (or NULL if the entire call-chain of - * the task should be shown). - */ -void show_stack (struct task_struct *task, unsigned long *sp) -{ - unsigned long addr, end; - - if (sp) - addr = (unsigned long)sp; - else if (task) - addr = task_sp (task); - else - addr = stack_addr (); - - addr = addr & ~3; - end = (addr + THREAD_SIZE - 1) & THREAD_MASK; - - while (addr < end) { - printk ("%8lX: ", addr); - while (addr < end) { - printk (" %8lX", *(unsigned long *)addr); - addr += sizeof (unsigned long); - if (! (addr & 0xF)) - break; - } - printk ("\n"); - } -} - -void dump_stack () -{ - show_stack (0, 0); -} - -EXPORT_SYMBOL(dump_stack); diff --git a/arch/v850/kernel/entry.S b/arch/v850/kernel/entry.S deleted file mode 100644 index e4327a8d6bcd..000000000000 --- a/arch/v850/kernel/entry.S +++ /dev/null @@ -1,1121 +0,0 @@ -/* - * arch/v850/kernel/entry.S -- Low-level system-call handling, trap handlers, - * and context-switching - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include - -#include -#include -#include -#include -#include -#include -#include - -#include - - -/* Make a slightly more convenient alias for C_SYMBOL_NAME. */ -#define CSYM C_SYMBOL_NAME - - -/* The offset of the struct pt_regs in a state-save-frame on the stack. */ -#define PTO STATE_SAVE_PT_OFFSET - - -/* Save argument registers to the state-save-frame pointed to by EP. */ -#define SAVE_ARG_REGS \ - sst.w r6, PTO+PT_GPR(6)[ep]; \ - sst.w r7, PTO+PT_GPR(7)[ep]; \ - sst.w r8, PTO+PT_GPR(8)[ep]; \ - sst.w r9, PTO+PT_GPR(9)[ep] -/* Restore argument registers from the state-save-frame pointed to by EP. */ -#define RESTORE_ARG_REGS \ - sld.w PTO+PT_GPR(6)[ep], r6; \ - sld.w PTO+PT_GPR(7)[ep], r7; \ - sld.w PTO+PT_GPR(8)[ep], r8; \ - sld.w PTO+PT_GPR(9)[ep], r9 - -/* Save value return registers to the state-save-frame pointed to by EP. */ -#define SAVE_RVAL_REGS \ - sst.w r10, PTO+PT_GPR(10)[ep]; \ - sst.w r11, PTO+PT_GPR(11)[ep] -/* Restore value return registers from the state-save-frame pointed to by EP. */ -#define RESTORE_RVAL_REGS \ - sld.w PTO+PT_GPR(10)[ep], r10; \ - sld.w PTO+PT_GPR(11)[ep], r11 - - -#define SAVE_CALL_CLOBBERED_REGS_BEFORE_ARGS \ - sst.w r1, PTO+PT_GPR(1)[ep]; \ - sst.w r5, PTO+PT_GPR(5)[ep] -#define SAVE_CALL_CLOBBERED_REGS_AFTER_RVAL \ - sst.w r12, PTO+PT_GPR(12)[ep]; \ - sst.w r13, PTO+PT_GPR(13)[ep]; \ - sst.w r14, PTO+PT_GPR(14)[ep]; \ - sst.w r15, PTO+PT_GPR(15)[ep]; \ - sst.w r16, PTO+PT_GPR(16)[ep]; \ - sst.w r17, PTO+PT_GPR(17)[ep]; \ - sst.w r18, PTO+PT_GPR(18)[ep]; \ - sst.w r19, PTO+PT_GPR(19)[ep] -#define RESTORE_CALL_CLOBBERED_REGS_BEFORE_ARGS \ - sld.w PTO+PT_GPR(1)[ep], r1; \ - sld.w PTO+PT_GPR(5)[ep], r5 -#define RESTORE_CALL_CLOBBERED_REGS_AFTER_RVAL \ - sld.w PTO+PT_GPR(12)[ep], r12; \ - sld.w PTO+PT_GPR(13)[ep], r13; \ - sld.w PTO+PT_GPR(14)[ep], r14; \ - sld.w PTO+PT_GPR(15)[ep], r15; \ - sld.w PTO+PT_GPR(16)[ep], r16; \ - sld.w PTO+PT_GPR(17)[ep], r17; \ - sld.w PTO+PT_GPR(18)[ep], r18; \ - sld.w PTO+PT_GPR(19)[ep], r19 - -/* Save `call clobbered' registers to the state-save-frame pointed to by EP. */ -#define SAVE_CALL_CLOBBERED_REGS \ - SAVE_CALL_CLOBBERED_REGS_BEFORE_ARGS; \ - SAVE_ARG_REGS; \ - SAVE_RVAL_REGS; \ - SAVE_CALL_CLOBBERED_REGS_AFTER_RVAL -/* Restore `call clobbered' registers from the state-save-frame pointed to - by EP. */ -#define RESTORE_CALL_CLOBBERED_REGS \ - RESTORE_CALL_CLOBBERED_REGS_BEFORE_ARGS; \ - RESTORE_ARG_REGS; \ - RESTORE_RVAL_REGS; \ - RESTORE_CALL_CLOBBERED_REGS_AFTER_RVAL - -/* Save `call clobbered' registers except for the return-value registers - to the state-save-frame pointed to by EP. */ -#define SAVE_CALL_CLOBBERED_REGS_NO_RVAL \ - SAVE_CALL_CLOBBERED_REGS_BEFORE_ARGS; \ - SAVE_ARG_REGS; \ - SAVE_CALL_CLOBBERED_REGS_AFTER_RVAL -/* Restore `call clobbered' registers except for the return-value registers - from the state-save-frame pointed to by EP. */ -#define RESTORE_CALL_CLOBBERED_REGS_NO_RVAL \ - RESTORE_CALL_CLOBBERED_REGS_BEFORE_ARGS; \ - RESTORE_ARG_REGS; \ - RESTORE_CALL_CLOBBERED_REGS_AFTER_RVAL - -/* Save `call saved' registers to the state-save-frame pointed to by EP. */ -#define SAVE_CALL_SAVED_REGS \ - sst.w r2, PTO+PT_GPR(2)[ep]; \ - sst.w r20, PTO+PT_GPR(20)[ep]; \ - sst.w r21, PTO+PT_GPR(21)[ep]; \ - sst.w r22, PTO+PT_GPR(22)[ep]; \ - sst.w r23, PTO+PT_GPR(23)[ep]; \ - sst.w r24, PTO+PT_GPR(24)[ep]; \ - sst.w r25, PTO+PT_GPR(25)[ep]; \ - sst.w r26, PTO+PT_GPR(26)[ep]; \ - sst.w r27, PTO+PT_GPR(27)[ep]; \ - sst.w r28, PTO+PT_GPR(28)[ep]; \ - sst.w r29, PTO+PT_GPR(29)[ep] -/* Restore `call saved' registers from the state-save-frame pointed to by EP. */ -#define RESTORE_CALL_SAVED_REGS \ - sld.w PTO+PT_GPR(2)[ep], r2; \ - sld.w PTO+PT_GPR(20)[ep], r20; \ - sld.w PTO+PT_GPR(21)[ep], r21; \ - sld.w PTO+PT_GPR(22)[ep], r22; \ - sld.w PTO+PT_GPR(23)[ep], r23; \ - sld.w PTO+PT_GPR(24)[ep], r24; \ - sld.w PTO+PT_GPR(25)[ep], r25; \ - sld.w PTO+PT_GPR(26)[ep], r26; \ - sld.w PTO+PT_GPR(27)[ep], r27; \ - sld.w PTO+PT_GPR(28)[ep], r28; \ - sld.w PTO+PT_GPR(29)[ep], r29 - - -/* Save the PC stored in the special register SAVEREG to the state-save-frame - pointed to by EP. r19 is clobbered. */ -#define SAVE_PC(savereg) \ - stsr SR_ ## savereg, r19; \ - sst.w r19, PTO+PT_PC[ep] -/* Restore the PC from the state-save-frame pointed to by EP, to the special - register SAVEREG. LP is clobbered (it is used as a scratch register - because the POP_STATE macro restores it, and this macro is usually used - inside POP_STATE). */ -#define RESTORE_PC(savereg) \ - sld.w PTO+PT_PC[ep], lp; \ - ldsr lp, SR_ ## savereg -/* Save the PSW register stored in the special register SAVREG to the - state-save-frame pointed to by EP. r19 is clobbered. */ -#define SAVE_PSW(savereg) \ - stsr SR_ ## savereg, r19; \ - sst.w r19, PTO+PT_PSW[ep] -/* Restore the PSW register from the state-save-frame pointed to by EP, to - the special register SAVEREG. LP is clobbered (it is used as a scratch - register because the POP_STATE macro restores it, and this macro is - usually used inside POP_STATE). */ -#define RESTORE_PSW(savereg) \ - sld.w PTO+PT_PSW[ep], lp; \ - ldsr lp, SR_ ## savereg - -/* Save CTPC/CTPSW/CTBP registers to the state-save-frame pointed to by REG. - r19 is clobbered. */ -#define SAVE_CT_REGS \ - stsr SR_CTPC, r19; \ - sst.w r19, PTO+PT_CTPC[ep]; \ - stsr SR_CTPSW, r19; \ - sst.w r19, PTO+PT_CTPSW[ep]; \ - stsr SR_CTBP, r19; \ - sst.w r19, PTO+PT_CTBP[ep] -/* Restore CTPC/CTPSW/CTBP registers from the state-save-frame pointed to by EP. - LP is clobbered (it is used as a scratch register because the POP_STATE - macro restores it, and this macro is usually used inside POP_STATE). */ -#define RESTORE_CT_REGS \ - sld.w PTO+PT_CTPC[ep], lp; \ - ldsr lp, SR_CTPC; \ - sld.w PTO+PT_CTPSW[ep], lp; \ - ldsr lp, SR_CTPSW; \ - sld.w PTO+PT_CTBP[ep], lp; \ - ldsr lp, SR_CTBP - - -/* Push register state, except for the stack pointer, on the stack in the - form of a state-save-frame (plus some extra padding), in preparation for - a system call. This macro makes sure that the EP, GP, and LP - registers are saved, and TYPE identifies the set of extra registers to - be saved as well. Also copies (the new value of) SP to EP. */ -#define PUSH_STATE(type) \ - addi -STATE_SAVE_SIZE, sp, sp; /* Make room on the stack. */ \ - st.w ep, PTO+PT_GPR(GPR_EP)[sp]; \ - mov sp, ep; \ - sst.w gp, PTO+PT_GPR(GPR_GP)[ep]; \ - sst.w lp, PTO+PT_GPR(GPR_LP)[ep]; \ - type ## _STATE_SAVER -/* Pop a register state pushed by PUSH_STATE, except for the stack pointer, - from the stack. */ -#define POP_STATE(type) \ - mov sp, ep; \ - type ## _STATE_RESTORER; \ - sld.w PTO+PT_GPR(GPR_GP)[ep], gp; \ - sld.w PTO+PT_GPR(GPR_LP)[ep], lp; \ - sld.w PTO+PT_GPR(GPR_EP)[ep], ep; \ - addi STATE_SAVE_SIZE, sp, sp /* Clean up our stack space. */ - - -/* Switch to the kernel stack if necessary, and push register state on the - stack in the form of a state-save-frame. Also load the current task - pointer if switching from user mode. The stack-pointer (r3) should have - already been saved to the memory location SP_SAVE_LOC (the reason for - this is that the interrupt vectors may be beyond a 22-bit signed offset - jump from the actual interrupt handler, and this allows them to save the - stack-pointer and use that register to do an indirect jump). This macro - makes sure that `special' registers, system registers, and the stack - pointer are saved; TYPE identifies the set of extra registers to be - saved as well. SYSCALL_NUM is the register in which the system-call - number this state is for is stored (r0 if this isn't a system call). - Interrupts should already be disabled when calling this. */ -#define SAVE_STATE(type, syscall_num, sp_save_loc) \ - tst1 0, KM; /* See if already in kernel mode. */ \ - bz 1f; \ - ld.w sp_save_loc, sp; /* ... yes, use saved SP. */ \ - br 2f; \ -1: ld.w KSP, sp; /* ... no, switch to kernel stack. */ \ -2: PUSH_STATE(type); \ - ld.b KM, r19; /* Remember old kernel-mode. */ \ - sst.w r19, PTO+PT_KERNEL_MODE[ep]; \ - ld.w sp_save_loc, r19; /* Remember old SP. */ \ - sst.w r19, PTO+PT_GPR(GPR_SP)[ep]; \ - mov 1, r19; /* Now definitely in kernel-mode. */ \ - st.b r19, KM; \ - GET_CURRENT_TASK(CURRENT_TASK); /* Fetch the current task pointer. */ \ - /* Save away the syscall number. */ \ - sst.w syscall_num, PTO+PT_CUR_SYSCALL[ep] - - -/* Save register state not normally saved by PUSH_STATE for TYPE, to the - state-save-frame on the stack; also copies SP to EP. r19 may be trashed. */ -#define SAVE_EXTRA_STATE(type) \ - mov sp, ep; \ - type ## _EXTRA_STATE_SAVER -/* Restore register state not normally restored by POP_STATE for TYPE, - from the state-save-frame on the stack; also copies SP to EP. - r19 may be trashed. */ -#define RESTORE_EXTRA_STATE(type) \ - mov sp, ep; \ - type ## _EXTRA_STATE_RESTORER - -/* Save any call-clobbered registers not normally saved by PUSH_STATE for - TYPE, to the state-save-frame on the stack. - EP may be trashed, but is not guaranteed to contain a copy of SP - (unlike after most SAVE_... macros). r19 may be trashed. */ -#define SAVE_EXTRA_STATE_FOR_SCHEDULE(type) \ - type ## _SCHEDULE_EXTRA_STATE_SAVER -/* Restore any call-clobbered registers not normally restored by - POP_STATE for TYPE, to the state-save-frame on the stack. - EP may be trashed, but is not guaranteed to contain a copy of SP - (unlike after most RESTORE_... macros). r19 may be trashed. */ -#define RESTORE_EXTRA_STATE_FOR_SCHEDULE(type) \ - type ## _SCHEDULE_EXTRA_STATE_RESTORER - - -/* These are extra_state_saver/restorer values for a user trap. Note - that we save the argument registers so that restarted syscalls will - function properly (otherwise it wouldn't be necessary), and we must - _not_ restore the return-value registers (so that traps can return a - value!), but call-clobbered registers are not saved at all, as the - caller of the syscall function should have saved them. */ - -#define TRAP_RET reti -/* Traps don't save call-clobbered registers (but do still save arg regs). - We preserve PSw to keep long-term state, namely interrupt status (for traps - from kernel-mode), and the single-step flag (for user traps). */ -#define TRAP_STATE_SAVER \ - SAVE_ARG_REGS; \ - SAVE_PC(EIPC); \ - SAVE_PSW(EIPSW) -/* When traps return, they just leave call-clobbered registers (except for arg - regs) with whatever value they have from the kernel. Traps don't preserve - the PSW, but we zero EIPSW to ensure it doesn't contain anything dangerous - (in particular, the single-step flag). */ -#define TRAP_STATE_RESTORER \ - RESTORE_ARG_REGS; \ - RESTORE_PC(EIPC); \ - RESTORE_PSW(EIPSW) -/* Save registers not normally saved by traps. We need to save r12, even - though it's nominally call-clobbered, because it's used when restarting - a system call (the signal-handling path uses SAVE_EXTRA_STATE, and - expects r12 to be restored when the trap returns). */ -#define TRAP_EXTRA_STATE_SAVER \ - SAVE_RVAL_REGS; \ - sst.w r12, PTO+PT_GPR(12)[ep]; \ - SAVE_CALL_SAVED_REGS; \ - SAVE_CT_REGS -#define TRAP_EXTRA_STATE_RESTORER \ - RESTORE_RVAL_REGS; \ - sld.w PTO+PT_GPR(12)[ep], r12; \ - RESTORE_CALL_SAVED_REGS; \ - RESTORE_CT_REGS -/* Save registers prior to calling scheduler (just before trap returns). - We have to save the return-value registers to preserve the trap's return - value. Note that ..._SCHEDULE_EXTRA_STATE_SAVER, unlike most ..._SAVER - macros, is required to setup EP itself if EP is needed (this is because - in many cases, the macro is empty). */ -#define TRAP_SCHEDULE_EXTRA_STATE_SAVER \ - mov sp, ep; \ - SAVE_RVAL_REGS -/* Note that ..._SCHEDULE_EXTRA_STATE_RESTORER, unlike most ..._RESTORER - macros, is required to setup EP itself if EP is needed (this is because - in many cases, the macro is empty). */ -#define TRAP_SCHEDULE_EXTRA_STATE_RESTORER \ - mov sp, ep; \ - RESTORE_RVAL_REGS - -/* Register saving/restoring for maskable interrupts. */ -#define IRQ_RET reti -#define IRQ_STATE_SAVER \ - SAVE_CALL_CLOBBERED_REGS; \ - SAVE_PC(EIPC); \ - SAVE_PSW(EIPSW) -#define IRQ_STATE_RESTORER \ - RESTORE_CALL_CLOBBERED_REGS; \ - RESTORE_PC(EIPC); \ - RESTORE_PSW(EIPSW) -#define IRQ_EXTRA_STATE_SAVER \ - SAVE_CALL_SAVED_REGS; \ - SAVE_CT_REGS -#define IRQ_EXTRA_STATE_RESTORER \ - RESTORE_CALL_SAVED_REGS; \ - RESTORE_CT_REGS -#define IRQ_SCHEDULE_EXTRA_STATE_SAVER /* nothing */ -#define IRQ_SCHEDULE_EXTRA_STATE_RESTORER /* nothing */ - -/* Register saving/restoring for non-maskable interrupts. */ -#define NMI_RET reti -#define NMI_STATE_SAVER \ - SAVE_CALL_CLOBBERED_REGS; \ - SAVE_PC(FEPC); \ - SAVE_PSW(FEPSW); -#define NMI_STATE_RESTORER \ - RESTORE_CALL_CLOBBERED_REGS; \ - RESTORE_PC(FEPC); \ - RESTORE_PSW(FEPSW); -#define NMI_EXTRA_STATE_SAVER \ - SAVE_CALL_SAVED_REGS; \ - SAVE_CT_REGS -#define NMI_EXTRA_STATE_RESTORER \ - RESTORE_CALL_SAVED_REGS; \ - RESTORE_CT_REGS -#define NMI_SCHEDULE_EXTRA_STATE_SAVER /* nothing */ -#define NMI_SCHEDULE_EXTRA_STATE_RESTORER /* nothing */ - -/* Register saving/restoring for debug traps. */ -#define DBTRAP_RET .long 0x014607E0 /* `dbret', but gas doesn't support it. */ -#define DBTRAP_STATE_SAVER \ - SAVE_CALL_CLOBBERED_REGS; \ - SAVE_PC(DBPC); \ - SAVE_PSW(DBPSW) -#define DBTRAP_STATE_RESTORER \ - RESTORE_CALL_CLOBBERED_REGS; \ - RESTORE_PC(DBPC); \ - RESTORE_PSW(DBPSW) -#define DBTRAP_EXTRA_STATE_SAVER \ - SAVE_CALL_SAVED_REGS; \ - SAVE_CT_REGS -#define DBTRAP_EXTRA_STATE_RESTORER \ - RESTORE_CALL_SAVED_REGS; \ - RESTORE_CT_REGS -#define DBTRAP_SCHEDULE_EXTRA_STATE_SAVER /* nothing */ -#define DBTRAP_SCHEDULE_EXTRA_STATE_RESTORER /* nothing */ - -/* Register saving/restoring for a context switch. We don't need to save - too many registers, because context-switching looks like a function call - (via the function `switch_thread'), so callers will save any - call-clobbered registers themselves. We do need to save the CT regs, as - they're normally not saved during kernel entry (the kernel doesn't use - them). We save PSW so that interrupt-status state will correctly follow - each thread (mostly NMI vs. normal-IRQ/trap), though for the most part - it doesn't matter since threads are always in almost exactly the same - processor state during a context switch. The stack pointer and return - value are handled by switch_thread itself. */ -#define SWITCH_STATE_SAVER \ - SAVE_CALL_SAVED_REGS; \ - SAVE_PSW(PSW); \ - SAVE_CT_REGS -#define SWITCH_STATE_RESTORER \ - RESTORE_CALL_SAVED_REGS; \ - RESTORE_PSW(PSW); \ - RESTORE_CT_REGS - - -/* Restore register state from the state-save-frame on the stack, switch back - to the user stack if necessary, and return from the trap/interrupt. - EXTRA_STATE_RESTORER is a sequence of assembly language statements to - restore anything not restored by this macro. Only registers not saved by - the C compiler are restored (that is, R3(sp), R4(gp), R31(lp), and - anything restored by EXTRA_STATE_RESTORER). */ -#define RETURN(type) \ - ld.b PTO+PT_KERNEL_MODE[sp], r19; \ - di; /* Disable interrupts */ \ - cmp r19, r0; /* See if returning to kernel mode, */\ - bne 2f; /* ... if so, skip resched &c. */ \ - \ - /* We're returning to user mode, so check for various conditions that \ - trigger rescheduling. */ \ - GET_CURRENT_THREAD(r18); \ - ld.w TI_FLAGS[r18], r19; \ - andi _TIF_NEED_RESCHED, r19, r0; \ - bnz 3f; /* Call the scheduler. */ \ -5: andi _TIF_SIGPENDING, r19, r18; \ - ld.w TASK_PTRACE[CURRENT_TASK], r19; /* ptrace flags */ \ - or r18, r19; /* see if either is non-zero */ \ - bnz 4f; /* if so, handle them */ \ - \ -/* Return to user state. */ \ -1: st.b r0, KM; /* Now officially in user state. */ \ - \ -/* Final return. The stack-pointer fiddling is not needed when returning \ - to kernel-mode, but they don't hurt, and this way we can share the \ - (sometimes rather lengthy) POP_STATE macro. */ \ -2: POP_STATE(type); \ - st.w sp, KSP; /* Save the kernel stack pointer. */ \ - ld.w PT_GPR(GPR_SP)-PT_SIZE[sp], sp; /* Restore stack pointer. */ \ - type ## _RET; /* Return from the trap/interrupt. */ \ - \ -/* Call the scheduler before returning from a syscall/trap. */ \ -3: SAVE_EXTRA_STATE_FOR_SCHEDULE(type); /* Prepare to call scheduler. */ \ - jarl call_scheduler, lp; /* Call scheduler */ \ - di; /* The scheduler enables interrupts */\ - RESTORE_EXTRA_STATE_FOR_SCHEDULE(type); \ - GET_CURRENT_THREAD(r18); \ - ld.w TI_FLAGS[r18], r19; \ - br 5b; /* Continue with return path. */ \ - \ -/* Handle a signal or ptraced process return. \ - r18 should be non-zero if there are pending signals. */ \ -4: /* Not all registers are saved by the normal trap/interrupt entry \ - points (for instance, call-saved registers (because the normal \ - C-compiler calling sequence in the kernel makes sure they're \ - preserved), and call-clobbered registers in the case of \ - traps), but signal handlers may want to examine or change the \ - complete register state. Here we save anything not saved by \ - the normal entry sequence, so that it may be safely restored \ - (in a possibly modified form) after do_signal returns. */ \ - SAVE_EXTRA_STATE(type); /* Save state not saved by entry. */ \ - jarl handle_signal_or_ptrace_return, lp; \ - RESTORE_EXTRA_STATE(type); /* Restore extra regs. */ \ - br 1b - - -/* Jump to the appropriate function for the system call number in r12 - (r12 is not preserved), or return an error if r12 is not valid. The - LP register should point to the location where the called function - should return. [note that MAKE_SYS_CALL uses label 1] */ -#define MAKE_SYS_CALL \ - /* Figure out which function to use for this system call. */ \ - shl 2, r12; \ - /* See if the system call number is valid. */ \ - addi lo(CSYM(sys_call_table) - sys_call_table_end), r12, r0; \ - bnh 1f; \ - mov hilo(CSYM(sys_call_table)), r19; \ - add r19, r12; \ - ld.w 0[r12], r12; \ - /* Make the system call. */ \ - jmp [r12]; \ - /* The syscall number is invalid, return an error. */ \ -1: addi -ENOSYS, r0, r10; \ - jmp [lp] - - - .text - -/* - * User trap. - * - * Trap 0 system calls are also handled here. - * - * The stack-pointer (r3) should have already been saved to the memory - * location ENTRY_SP (the reason for this is that the interrupt vectors may be - * beyond a 22-bit signed offset jump from the actual interrupt handler, and - * this allows them to save the stack-pointer and use that register to do an - * indirect jump). - * - * Syscall protocol: - * Syscall number in r12, args in r6-r9 - * Return value in r10 - */ -G_ENTRY(trap): - SAVE_STATE (TRAP, r12, ENTRY_SP) // Save registers. - stsr SR_ECR, r19 // Find out which trap it was. - ei // Enable interrupts. - mov hilo(ret_from_trap), lp // where the trap should return - - // The following two shifts (1) clear out extraneous NMI data in the - // upper 16-bits, (2) convert the 0x40 - 0x5f range of trap ECR - // numbers into the (0-31) << 2 range we want, (3) set the flags. - shl 27, r19 // chop off all high bits - shr 25, r19 // scale back down and then << 2 - bnz 2f // See if not trap 0. - - // Trap 0 is a `short' system call, skip general trap table. - MAKE_SYS_CALL // Jump to the syscall function. - -2: // For other traps, use a table lookup. - mov hilo(CSYM(trap_table)), r18 - add r19, r18 - ld.w 0[r18], r18 - jmp [r18] // Jump to the trap handler. -END(trap) - -/* This is just like ret_from_trap, but first restores extra registers - saved by some wrappers. */ -L_ENTRY(restore_extra_regs_and_ret_from_trap): - RESTORE_EXTRA_STATE(TRAP) - // fall through -END(restore_extra_regs_and_ret_from_trap) - -/* Entry point used to return from a syscall/trap. */ -L_ENTRY(ret_from_trap): - RETURN(TRAP) -END(ret_from_trap) - - -/* This the initial entry point for a new child thread, with an appropriate - stack in place that makes it look that the child is in the middle of an - syscall. This function is actually `returned to' from switch_thread - (copy_thread makes ret_from_fork the return address in each new thread's - saved context). */ -C_ENTRY(ret_from_fork): - mov r10, r6 // switch_thread returns the prev task. - jarl CSYM(schedule_tail), lp // ...which is schedule_tail's arg - mov r0, r10 // Child's fork call should return 0. - br ret_from_trap // Do normal trap return. -C_END(ret_from_fork) - - -/* - * Trap 1: `long' system calls - * `Long' syscall protocol: - * Syscall number in r12, args in r6-r9, r13-r14 - * Return value in r10 - */ -L_ENTRY(syscall_long): - // Push extra arguments on the stack. Note that by default, the trap - // handler reserves enough stack space for 6 arguments, so we don't - // have to make any additional room. - st.w r13, 16[sp] // arg 5 - st.w r14, 20[sp] // arg 6 - - // Make sure r13 and r14 are preserved, in case we have to restart a - // system call because of a signal (ep has already been set by caller). - st.w r13, PTO+PT_GPR(13)[sp] - st.w r14, PTO+PT_GPR(13)[sp] - mov hilo(ret_from_long_syscall), lp - - MAKE_SYS_CALL // Jump to the syscall function. -END(syscall_long) - -/* Entry point used to return from a long syscall. Only needed to restore - r13/r14 if the general trap mechanism doesnt' do so. */ -L_ENTRY(ret_from_long_syscall): - ld.w PTO+PT_GPR(13)[sp], r13 // Restore the extra registers - ld.w PTO+PT_GPR(13)[sp], r14 - br ret_from_trap // The rest is the same as other traps -END(ret_from_long_syscall) - - -/* These syscalls need access to the struct pt_regs on the stack, so we - implement them in assembly (they're basically all wrappers anyway). */ - -L_ENTRY(sys_fork_wrapper): -#ifdef CONFIG_MMU - addi SIGCHLD, r0, r6 // Arg 0: flags - ld.w PTO+PT_GPR(GPR_SP)[sp], r7 // Arg 1: child SP (use parent's) - movea PTO, sp, r8 // Arg 2: parent context - mov r0, r9 // Arg 3/4/5: 0 - st.w r0, 16[sp] - st.w r0, 20[sp] - mov hilo(CSYM(do_fork)), r18 // Where the real work gets done - br save_extra_state_tramp // Save state and go there -#else - // fork almost works, enough to trick you into looking elsewhere :-( - addi -EINVAL, r0, r10 - jmp [lp] -#endif -END(sys_fork_wrapper) - -L_ENTRY(sys_vfork_wrapper): - addi CLONE_VFORK | CLONE_VM | SIGCHLD, r0, r6 // Arg 0: flags - ld.w PTO+PT_GPR(GPR_SP)[sp], r7 // Arg 1: child SP (use parent's) - movea PTO, sp, r8 // Arg 2: parent context - mov r0, r9 // Arg 3/4/5: 0 - st.w r0, 16[sp] - st.w r0, 20[sp] - mov hilo(CSYM(do_fork)), r18 // Where the real work gets done - br save_extra_state_tramp // Save state and go there -END(sys_vfork_wrapper) - -L_ENTRY(sys_clone_wrapper): - ld.w PTO+PT_GPR(GPR_SP)[sp], r19// parent's stack pointer - cmp r7, r0 // See if child SP arg (arg 1) is 0. - cmov z, r19, r7, r7 // ... and use the parent's if so. - movea PTO, sp, r8 // Arg 2: parent context - mov r0, r9 // Arg 3/4/5: 0 - st.w r0, 16[sp] - st.w r0, 20[sp] - mov hilo(CSYM(do_fork)), r18 // Where the real work gets done - br save_extra_state_tramp // Save state and go there -END(sys_clone_wrapper) - - -L_ENTRY(sys_execve_wrapper): - movea PTO, sp, r9 // add user context as 4th arg - jr CSYM(sys_execve) // Do real work (tail-call). -END(sys_execve_wrapper) - - -L_ENTRY(sys_sigsuspend_wrapper): - movea PTO, sp, r7 // add user context as 2nd arg - mov hilo(CSYM(sys_sigsuspend)), r18 // syscall function - jarl save_extra_state_tramp, lp // Save state and do it - br restore_extra_regs_and_ret_from_trap -END(sys_sigsuspend_wrapper) -L_ENTRY(sys_rt_sigsuspend_wrapper): - movea PTO, sp, r8 // add user context as 3rd arg - mov hilo(CSYM(sys_rt_sigsuspend)), r18 // syscall function - jarl save_extra_state_tramp, lp // Save state and do it - br restore_extra_regs_and_ret_from_trap -END(sys_rt_sigsuspend_wrapper) - -L_ENTRY(sys_sigreturn_wrapper): - movea PTO, sp, r6 // add user context as 1st arg - mov hilo(CSYM(sys_sigreturn)), r18 // syscall function - jarl save_extra_state_tramp, lp // Save state and do it - br restore_extra_regs_and_ret_from_trap -END(sys_sigreturn_wrapper) -L_ENTRY(sys_rt_sigreturn_wrapper): - movea PTO, sp, r6 // add user context as 1st arg - mov hilo(CSYM(sys_rt_sigreturn)), r18// syscall function - jarl save_extra_state_tramp, lp // Save state and do it - br restore_extra_regs_and_ret_from_trap -END(sys_rt_sigreturn_wrapper) - - -/* Save any state not saved by SAVE_STATE(TRAP), and jump to r18. - It's main purpose is to share the rather lengthy code sequence that - SAVE_STATE expands into among the above wrapper functions. */ -L_ENTRY(save_extra_state_tramp): - SAVE_EXTRA_STATE(TRAP) // Save state not saved by entry. - jmp [r18] // Do the work the caller wants -END(save_extra_state_tramp) - - -/* - * Hardware maskable interrupts. - * - * The stack-pointer (r3) should have already been saved to the memory - * location ENTRY_SP (the reason for this is that the interrupt vectors may be - * beyond a 22-bit signed offset jump from the actual interrupt handler, and - * this allows them to save the stack-pointer and use that register to do an - * indirect jump). - */ -G_ENTRY(irq): - SAVE_STATE (IRQ, r0, ENTRY_SP) // Save registers. - - stsr SR_ECR, r6 // Find out which interrupt it was. - movea PTO, sp, r7 // User regs are arg2 - - // All v850 implementations I know about encode their interrupts as - // multiples of 0x10, starting at 0x80 (after NMIs and software - // interrupts). Convert this number into a simple IRQ index for the - // rest of the kernel. We also clear the upper 16 bits, which hold - // NMI info, and don't appear to be cleared when a NMI returns. - shl 16, r6 // clear upper 16 bits - shr 20, r6 // shift back, and remove lower nibble - add -8, r6 // remove bias for irqs - - // Call the high-level interrupt handling code. - jarl CSYM(handle_irq), lp - - RETURN(IRQ) -END(irq) - - -/* - * Debug trap / illegal-instruction exception - * - * The stack-pointer (r3) should have already been saved to the memory - * location ENTRY_SP (the reason for this is that the interrupt vectors may be - * beyond a 22-bit signed offset jump from the actual interrupt handler, and - * this allows them to save the stack-pointer and use that register to do an - * indirect jump). - */ -G_ENTRY(dbtrap): - SAVE_STATE (DBTRAP, r0, ENTRY_SP)// Save registers. - - /* First see if we came from kernel mode; if so, the dbtrap - instruction has a special meaning, to set the DIR (`debug - information register') register. This is because the DIR register - can _only_ be manipulated/read while in `debug mode,' and debug - mode is only active while we're inside the dbtrap handler. The - exact functionality is: { DIR = (DIR | r6) & ~r7; return DIR; }. */ - ld.b PTO+PT_KERNEL_MODE[sp], r19 - cmp r19, r0 - bz 1f - - stsr SR_DIR, r10 - or r6, r10 - not r7, r7 - and r7, r10 - ldsr r10, SR_DIR - stsr SR_DIR, r10 // Confirm the value we set - st.w r10, PTO+PT_GPR(10)[sp] // return it - br 3f - -1: ei // Enable interrupts. - - /* The default signal type we raise. */ - mov SIGTRAP, r6 - - /* See if it's a single-step trap. */ - stsr SR_DBPSW, r19 - andi 0x0800, r19, r19 - bnz 2f - - /* Look to see if the preceding instruction was is a dbtrap or not, - to decide which signal we should use. */ - stsr SR_DBPC, r19 // PC following trapping insn - ld.hu -2[r19], r19 - ori 0xf840, r0, r20 // DBTRAP insn - cmp r19, r20 // Was this trap caused by DBTRAP? - cmov ne, SIGILL, r6, r6 // Choose signal appropriately - - /* Raise the desired signal. */ -2: mov CURRENT_TASK, r7 // Arg 1: task - jarl CSYM(send_sig), lp // tail call - -3: RETURN(DBTRAP) -END(dbtrap) - - -/* - * Hardware non-maskable interrupts. - * - * The stack-pointer (r3) should have already been saved to the memory - * location ENTRY_SP (the reason for this is that the interrupt vectors may be - * beyond a 22-bit signed offset jump from the actual interrupt handler, and - * this allows them to save the stack-pointer and use that register to do an - * indirect jump). - */ -G_ENTRY(nmi): - SAVE_STATE (NMI, r0, NMI_ENTRY_SP); /* Save registers. */ - - stsr SR_ECR, r6; /* Find out which nmi it was. */ - shr 20, r6; /* Extract NMI code in bits 20-24. */ - movea PTO, sp, r7; /* User regs are arg2. */ - - /* Non-maskable interrupts always lie right after maskable interrupts. - Call the generic IRQ handler, with two arguments, the IRQ number, - and a pointer to the user registers, to handle the specifics. - (we subtract one because the first NMI has code 1). */ - addi FIRST_NMI - 1, r6, r6 - jarl CSYM(handle_irq), lp - - RETURN(NMI) -END(nmi) - - -/* - * Trap with no handler - */ -L_ENTRY(bad_trap_wrapper): - mov r19, r6 // Arg 0: trap number - movea PTO, sp, r7 // Arg 1: user regs - jr CSYM(bad_trap) // tail call handler -END(bad_trap_wrapper) - - -/* - * Invoke the scheduler, called from the trap/irq kernel exit path. - * - * This basically just calls `schedule', but also arranges for extra - * registers to be saved for ptrace'd processes, so ptrace can modify them. - */ -L_ENTRY(call_scheduler): - ld.w TASK_PTRACE[CURRENT_TASK], r19 // See if task is ptrace'd - cmp r19, r0 - bnz 1f // ... yes, do special stuff - jr CSYM(schedule) // ... no, just tail-call scheduler - - // Save extra regs for ptrace'd task. We want to save anything - // that would otherwise only be `implicitly' saved by the normal - // compiler calling-convention. -1: mov sp, ep // Setup EP for SAVE_CALL_SAVED_REGS - SAVE_CALL_SAVED_REGS // Save call-saved registers to stack - mov lp, r20 // Save LP in a callee-saved register - - jarl CSYM(schedule), lp // Call scheduler - - mov r20, lp - mov sp, ep // We can't rely on EP after return - RESTORE_CALL_SAVED_REGS // Restore (possibly modified) regs - jmp [lp] // Return to the return path -END(call_scheduler) - - -/* - * This is an out-of-line handler for two special cases during the kernel - * trap/irq exit sequence: - * - * (1) If r18 is non-zero then a signal needs to be handled, which is - * done, and then the caller returned to. - * - * (2) If r18 is non-zero then we're returning to a ptraced process, which - * has several special cases -- single-stepping and trap tracing, both - * of which require using the `dbret' instruction to exit the kernel - * instead of the normal `reti' (this is because the CPU not correctly - * single-step after a reti). In this case, of course, this handler - * never returns to the caller. - * - * In either case, all registers should have been saved to the current - * state-save-frame on the stack, except for callee-saved registers. - * - * [These two different cases are combined merely to avoid bloating the - * macro-inlined code, not because they really make much sense together!] - */ -L_ENTRY(handle_signal_or_ptrace_return): - cmp r18, r0 // See if handling a signal - bz 1f // ... nope, go do ptrace return - - // Handle a signal - mov lp, r20 // Save link-pointer - mov r10, r21 // Save return-values (for trap) - mov r11, r22 - - movea PTO, sp, r6 // Arg 1: struct pt_regs *regs - mov r0, r7 // Arg 2: sigset_t *oldset - jarl CSYM(do_signal), lp // Handle the signal - di // sig handling enables interrupts - - mov r20, lp // Restore link-pointer - mov r21, r10 // Restore return-values (for trap) - mov r22, r11 - ld.w TASK_PTRACE[CURRENT_TASK], r19 // check ptrace flags too - cmp r19, r0 - bnz 1f // ... some set, so look more -2: jmp [lp] // ... none set, so return normally - - // ptrace return -1: ld.w PTO+PT_PSW[sp], r19 // Look at user-processes's flags - andi 0x0800, r19, r19 // See if single-step flag is set - bz 2b // ... nope, return normally - - // Return as if from a dbtrap insn - st.b r0, KM // Now officially in user state. - POP_STATE(DBTRAP) // Restore regs - st.w sp, KSP // Save the kernel stack pointer. - ld.w PT_GPR(GPR_SP)-PT_SIZE[sp], sp // Restore user stack pointer. - DBTRAP_RET // Return from the trap/interrupt. -END(handle_signal_or_ptrace_return) - - -/* - * This is where we switch between two threads. The arguments are: - * r6 -- pointer to the struct thread for the `current' process - * r7 -- pointer to the struct thread for the `new' process. - * when this function returns, it will return to the new thread. - */ -C_ENTRY(switch_thread): - // Return the previous task (r10 is not clobbered by restore below) - mov CURRENT_TASK, r10 - // First, push the current processor state on the stack - PUSH_STATE(SWITCH) - // Now save the location of the kernel stack pointer for this thread; - // since we've pushed all other state on the stack, this is enough to - // restore it all later. - st.w sp, THREAD_KSP[r6] - // Now restore the stack pointer from the new process - ld.w THREAD_KSP[r7], sp - // ... and restore all state from that - POP_STATE(SWITCH) - // Update the current task pointer - GET_CURRENT_TASK(CURRENT_TASK) - // Now return into the new thread - jmp [lp] -C_END(switch_thread) - - - .data - - .align 4 -C_DATA(trap_table): - .long bad_trap_wrapper // trap 0, doesn't use trap table. - .long syscall_long // trap 1, `long' syscall. - .long bad_trap_wrapper - .long bad_trap_wrapper - .long bad_trap_wrapper - .long bad_trap_wrapper - .long bad_trap_wrapper - .long bad_trap_wrapper - .long bad_trap_wrapper - .long bad_trap_wrapper - .long bad_trap_wrapper - .long bad_trap_wrapper - .long bad_trap_wrapper - .long bad_trap_wrapper - .long bad_trap_wrapper - .long bad_trap_wrapper -C_END(trap_table) - - - .section .rodata - - .align 4 -C_DATA(sys_call_table): - .long CSYM(sys_restart_syscall) // 0 - .long CSYM(sys_exit) - .long sys_fork_wrapper - .long CSYM(sys_read) - .long CSYM(sys_write) - .long CSYM(sys_open) // 5 - .long CSYM(sys_close) - .long CSYM(sys_waitpid) - .long CSYM(sys_creat) - .long CSYM(sys_link) - .long CSYM(sys_unlink) // 10 - .long sys_execve_wrapper - .long CSYM(sys_chdir) - .long CSYM(sys_time) - .long CSYM(sys_mknod) - .long CSYM(sys_chmod) // 15 - .long CSYM(sys_chown) - .long CSYM(sys_ni_syscall) // was: break - .long CSYM(sys_ni_syscall) // was: oldstat (aka stat) - .long CSYM(sys_lseek) - .long CSYM(sys_getpid) // 20 - .long CSYM(sys_mount) - .long CSYM(sys_oldumount) - .long CSYM(sys_setuid) - .long CSYM(sys_getuid) - .long CSYM(sys_stime) // 25 - .long CSYM(sys_ptrace) - .long CSYM(sys_alarm) - .long CSYM(sys_ni_syscall) // was: oldfstat (aka fstat) - .long CSYM(sys_pause) - .long CSYM(sys_utime) // 30 - .long CSYM(sys_ni_syscall) // was: stty - .long CSYM(sys_ni_syscall) // was: gtty - .long CSYM(sys_access) - .long CSYM(sys_nice) - .long CSYM(sys_ni_syscall) // 35, was: ftime - .long CSYM(sys_sync) - .long CSYM(sys_kill) - .long CSYM(sys_rename) - .long CSYM(sys_mkdir) - .long CSYM(sys_rmdir) // 40 - .long CSYM(sys_dup) - .long CSYM(sys_pipe) - .long CSYM(sys_times) - .long CSYM(sys_ni_syscall) // was: prof - .long CSYM(sys_brk) // 45 - .long CSYM(sys_setgid) - .long CSYM(sys_getgid) - .long CSYM(sys_signal) - .long CSYM(sys_geteuid) - .long CSYM(sys_getegid) // 50 - .long CSYM(sys_acct) - .long CSYM(sys_umount) // recycled never used phys() - .long CSYM(sys_ni_syscall) // was: lock - .long CSYM(sys_ioctl) - .long CSYM(sys_fcntl) // 55 - .long CSYM(sys_ni_syscall) // was: mpx - .long CSYM(sys_setpgid) - .long CSYM(sys_ni_syscall) // was: ulimit - .long CSYM(sys_ni_syscall) - .long CSYM(sys_umask) // 60 - .long CSYM(sys_chroot) - .long CSYM(sys_ustat) - .long CSYM(sys_dup2) - .long CSYM(sys_getppid) - .long CSYM(sys_getpgrp) // 65 - .long CSYM(sys_setsid) - .long CSYM(sys_sigaction) - .long CSYM(sys_sgetmask) - .long CSYM(sys_ssetmask) - .long CSYM(sys_setreuid) // 70 - .long CSYM(sys_setregid) - .long sys_sigsuspend_wrapper - .long CSYM(sys_sigpending) - .long CSYM(sys_sethostname) - .long CSYM(sys_setrlimit) // 75 - .long CSYM(sys_getrlimit) - .long CSYM(sys_getrusage) - .long CSYM(sys_gettimeofday) - .long CSYM(sys_settimeofday) - .long CSYM(sys_getgroups) // 80 - .long CSYM(sys_setgroups) - .long CSYM(sys_select) - .long CSYM(sys_symlink) - .long CSYM(sys_ni_syscall) // was: oldlstat (aka lstat) - .long CSYM(sys_readlink) // 85 - .long CSYM(sys_uselib) - .long CSYM(sys_swapon) - .long CSYM(sys_reboot) - .long CSYM(old_readdir) - .long CSYM(sys_mmap) // 90 - .long CSYM(sys_munmap) - .long CSYM(sys_truncate) - .long CSYM(sys_ftruncate) - .long CSYM(sys_fchmod) - .long CSYM(sys_fchown) // 95 - .long CSYM(sys_getpriority) - .long CSYM(sys_setpriority) - .long CSYM(sys_ni_syscall) // was: profil - .long CSYM(sys_statfs) - .long CSYM(sys_fstatfs) // 100 - .long CSYM(sys_ni_syscall) // i386: ioperm - .long CSYM(sys_socketcall) - .long CSYM(sys_syslog) - .long CSYM(sys_setitimer) - .long CSYM(sys_getitimer) // 105 - .long CSYM(sys_newstat) - .long CSYM(sys_newlstat) - .long CSYM(sys_newfstat) - .long CSYM(sys_ni_syscall) // was: olduname (aka uname) - .long CSYM(sys_ni_syscall) // 110, i386: iopl - .long CSYM(sys_vhangup) - .long CSYM(sys_ni_syscall) // was: idle - .long CSYM(sys_ni_syscall) // i386: vm86old - .long CSYM(sys_wait4) - .long CSYM(sys_swapoff) // 115 - .long CSYM(sys_sysinfo) - .long CSYM(sys_ipc) - .long CSYM(sys_fsync) - .long sys_sigreturn_wrapper - .long sys_clone_wrapper // 120 - .long CSYM(sys_setdomainname) - .long CSYM(sys_newuname) - .long CSYM(sys_ni_syscall) // i386: modify_ldt, m68k: cacheflush - .long CSYM(sys_adjtimex) - .long CSYM(sys_ni_syscall) // 125 - sys_mprotect - .long CSYM(sys_sigprocmask) - .long CSYM(sys_ni_syscall) // sys_create_module - .long CSYM(sys_init_module) - .long CSYM(sys_delete_module) - .long CSYM(sys_ni_syscall) // 130 - sys_get_kernel_syms - .long CSYM(sys_quotactl) - .long CSYM(sys_getpgid) - .long CSYM(sys_fchdir) - .long CSYM(sys_bdflush) - .long CSYM(sys_sysfs) // 135 - .long CSYM(sys_personality) - .long CSYM(sys_ni_syscall) // for afs_syscall - .long CSYM(sys_setfsuid) - .long CSYM(sys_setfsgid) - .long CSYM(sys_llseek) // 140 - .long CSYM(sys_getdents) - .long CSYM(sys_select) // for backward compat; remove someday - .long CSYM(sys_flock) - .long CSYM(sys_ni_syscall) // sys_msync - .long CSYM(sys_readv) // 145 - .long CSYM(sys_writev) - .long CSYM(sys_getsid) - .long CSYM(sys_fdatasync) - .long CSYM(sys_sysctl) - .long CSYM(sys_ni_syscall) // 150 - sys_mlock - .long CSYM(sys_ni_syscall) // sys_munlock - .long CSYM(sys_ni_syscall) // sys_mlockall - .long CSYM(sys_ni_syscall) // sys_munlockall - .long CSYM(sys_sched_setparam) - .long CSYM(sys_sched_getparam) // 155 - .long CSYM(sys_sched_setscheduler) - .long CSYM(sys_sched_getscheduler) - .long CSYM(sys_sched_yield) - .long CSYM(sys_sched_get_priority_max) - .long CSYM(sys_sched_get_priority_min) // 160 - .long CSYM(sys_sched_rr_get_interval) - .long CSYM(sys_nanosleep) - .long CSYM(sys_ni_syscall) // sys_mremap - .long CSYM(sys_setresuid) - .long CSYM(sys_getresuid) // 165 - .long CSYM(sys_ni_syscall) // for vm86 - .long CSYM(sys_ni_syscall) // sys_query_module - .long CSYM(sys_poll) - .long CSYM(sys_nfsservctl) - .long CSYM(sys_setresgid) // 170 - .long CSYM(sys_getresgid) - .long CSYM(sys_prctl) - .long sys_rt_sigreturn_wrapper - .long CSYM(sys_rt_sigaction) - .long CSYM(sys_rt_sigprocmask) // 175 - .long CSYM(sys_rt_sigpending) - .long CSYM(sys_rt_sigtimedwait) - .long CSYM(sys_rt_sigqueueinfo) - .long sys_rt_sigsuspend_wrapper - .long CSYM(sys_pread64) // 180 - .long CSYM(sys_pwrite64) - .long CSYM(sys_lchown) - .long CSYM(sys_getcwd) - .long CSYM(sys_capget) - .long CSYM(sys_capset) // 185 - .long CSYM(sys_sigaltstack) - .long CSYM(sys_sendfile) - .long CSYM(sys_ni_syscall) // streams1 - .long CSYM(sys_ni_syscall) // streams2 - .long sys_vfork_wrapper // 190 - .long CSYM(sys_ni_syscall) - .long CSYM(sys_mmap2) - .long CSYM(sys_truncate64) - .long CSYM(sys_ftruncate64) - .long CSYM(sys_stat64) // 195 - .long CSYM(sys_lstat64) - .long CSYM(sys_fstat64) - .long CSYM(sys_fcntl64) - .long CSYM(sys_getdents64) - .long CSYM(sys_pivot_root) // 200 - .long CSYM(sys_gettid) - .long CSYM(sys_tkill) -sys_call_table_end: -C_END(sys_call_table) diff --git a/arch/v850/kernel/fpga85e2c.c b/arch/v850/kernel/fpga85e2c.c deleted file mode 100644 index ab9cf16a85c8..000000000000 --- a/arch/v850/kernel/fpga85e2c.c +++ /dev/null @@ -1,167 +0,0 @@ -/* - * arch/v850/kernel/fpga85e2c.h -- Machine-dependent defs for - * FPGA implementation of V850E2/NA85E2C - * - * Copyright (C) 2002,03 NEC Electronics Corporation - * Copyright (C) 2002,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include - -#include "mach.h" - -extern void memcons_setup (void); - - -#define REG_DUMP_ADDR 0x220000 - - -extern struct irqaction reg_snap_action; /* fwd decl */ - - -void __init mach_early_init (void) -{ - int i; - const u32 *src; - register u32 *dst asm ("ep"); - extern u32 _intv_end, _intv_load_start; - - /* Set bus sizes: CS0 32-bit, CS1 16-bit, CS7 8-bit, - everything else 32-bit. */ - V850E2_BSC = 0x2AA6; - for (i = 2; i <= 6; i++) - CSDEV(i) = 0; /* 32 bit */ - - /* Ensure that the simulator halts on a panic, instead of going - into an infinite loop inside the panic function. */ - panic_timeout = -1; - - /* Move the interrupt vectors into their real location. Note that - any relocations there are relative to the real location, so we - don't have to fix anything up. We use a loop instead of calling - memcpy to keep this a leaf function (to avoid a function - prologue being generated). */ - dst = 0x10; /* &_intv_start + 0x10. */ - src = &_intv_load_start; - do { - u32 t0 = src[0], t1 = src[1], t2 = src[2], t3 = src[3]; - u32 t4 = src[4], t5 = src[5], t6 = src[6], t7 = src[7]; - dst[0] = t0; dst[1] = t1; dst[2] = t2; dst[3] = t3; - dst[4] = t4; dst[5] = t5; dst[6] = t6; dst[7] = t7; - dst += 8; - src += 8; - } while (dst < &_intv_end); -} - -void __init mach_setup (char **cmdline) -{ - memcons_setup (); - - /* Setup up NMI0 to copy the registers to a known memory location. - The FGPA board has a button that produces NMI0 when pressed, so - this allows us to push the button, and then look at memory to see - what's in the registers (there's no other way to easily do so). - We have to use `setup_irq' instead of `request_irq' because it's - still too early to do memory allocation. */ - setup_irq (IRQ_NMI (0), ®_snap_action); -} - -void mach_get_physical_ram (unsigned long *ram_start, unsigned long *ram_len) -{ - *ram_start = ERAM_ADDR; - *ram_len = ERAM_SIZE; -} - -void __init mach_sched_init (struct irqaction *timer_action) -{ - /* Setup up the timer interrupt. The FPGA peripheral control - registers _only_ work with single-bit writes (set1/clr1)! */ - __clear_bit (RPU_GTMC_CE_BIT, &RPU_GTMC); - __clear_bit (RPU_GTMC_CLK_BIT, &RPU_GTMC); - __set_bit (RPU_GTMC_CE_BIT, &RPU_GTMC); - - /* We use the first RPU interrupt, which occurs every 8.192ms. */ - setup_irq (IRQ_RPU (0), timer_action); -} - - -void mach_gettimeofday (struct timespec *tv) -{ - tv->tv_sec = 0; - tv->tv_nsec = 0; -} - -void machine_halt (void) __attribute__ ((noreturn)); -void machine_halt (void) -{ - for (;;) { - DWC(0) = 0x7777; - DWC(1) = 0x7777; - ASC = 0xffff; - FLGREG(0) = 1; /* Halt immediately. */ - asm ("di; halt; nop; nop; nop; nop; nop"); - } -} - -void machine_restart (char *__unused) -{ - machine_halt (); -} - -void machine_power_off (void) -{ - machine_halt (); -} - - -/* Interrupts */ - -struct v850e_intc_irq_init irq_inits[] = { - { "IRQ", 0, NUM_MACH_IRQS, 1, 7 }, - { "RPU", IRQ_RPU(0), IRQ_RPU_NUM, 1, 6 }, - { 0 } -}; -#define NUM_IRQ_INITS (ARRAY_SIZE(irq_inits) - 1) - -struct hw_interrupt_type hw_itypes[NUM_IRQ_INITS]; - -/* Initialize interrupts. */ -void __init mach_init_irqs (void) -{ - v850e_intc_init_irq_types (irq_inits, hw_itypes); -} - - -/* An interrupt handler that copies the registers to a known memory location, - for debugging purposes. */ - -static void make_reg_snap (int irq, void *dummy, struct pt_regs *regs) -{ - (*(unsigned *)REG_DUMP_ADDR)++; - (*(struct pt_regs *)(REG_DUMP_ADDR + sizeof (unsigned))) = *regs; -} - -static int reg_snap_dev_id; -static struct irqaction reg_snap_action = { - .handler = make_reg_snap, - .mask = CPU_MASK_NONE, - .name = "reg_snap", - .dev_id = ®_snap_dev_id, -}; diff --git a/arch/v850/kernel/fpga85e2c.ld b/arch/v850/kernel/fpga85e2c.ld deleted file mode 100644 index b5d4578ae411..000000000000 --- a/arch/v850/kernel/fpga85e2c.ld +++ /dev/null @@ -1,62 +0,0 @@ -/* Linker script for the FPGA implementation of the V850E2 NA85E2C cpu core - (CONFIG_V850E2_FPGA85E2C). */ - -MEMORY { - /* Reset vector. */ - RESET : ORIGIN = 0, LENGTH = 0x10 - /* Interrupt vectors. */ - INTV : ORIGIN = 0x10, LENGTH = 0x470 - /* The `window' in RAM were we're allowed to load stuff. */ - RAM_LOW : ORIGIN = 0x480, LENGTH = 0x0005FB80 - /* Some more ram above the window were we can put bss &c. */ - RAM_HIGH : ORIGIN = 0x00060000, LENGTH = 0x000A0000 - /* This is the area visible from the outside world (we can use - this only for uninitialized data). */ - VISIBLE : ORIGIN = 0x00200000, LENGTH = 0x00060000 -} - -SECTIONS { - .reset : { - __kram_start = . ; - __intv_start = . ; - *(.intv.reset) /* Reset vector */ - } > RESET - - .ram_low : { - __r0_ram = . ; /* Must be near address 0. */ - . = . + 32 ; - - TEXT_CONTENTS - DATA_CONTENTS - ROOT_FS_CONTENTS - RAMK_INIT_CONTENTS_NO_END - INITRAMFS_CONTENTS - } > RAM_LOW - - /* Where the interrupt vectors are initially loaded. */ - __intv_load_start = . ; - - .intv : { - *(.intv.common) /* Vectors common to all v850e proc. */ - *(.intv.mach) /* Machine-specific int. vectors. */ - __intv_end = . ; - } > INTV AT> RAM_LOW - - .ram_high : { - /* This is here so that when we free init memory the - load-time copy of the interrupt vectors and any empty - space at the end of the `RAM_LOW' area is freed too. */ - . = ALIGN (4096); - __init_end = . ; - - BSS_CONTENTS - __kram_end = . ; - BOOTMAP_CONTENTS - } > RAM_HIGH - - .visible : { - _memcons_output = . ; - . = . + 0x8000 ; - _memcons_output_end = . ; - } > VISIBLE -} diff --git a/arch/v850/kernel/gbus_int.c b/arch/v850/kernel/gbus_int.c deleted file mode 100644 index b2bcc251f65b..000000000000 --- a/arch/v850/kernel/gbus_int.c +++ /dev/null @@ -1,271 +0,0 @@ -/* - * arch/v850/kernel/gbus_int.c -- Midas labs GBUS interrupt support - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include -#include -#include -#include -#include - -#include - - -/* The number of shared GINT interrupts. */ -#define NUM_GINTS 4 - -/* For each GINT interrupt, how many GBUS interrupts are using it. */ -static unsigned gint_num_active_irqs[NUM_GINTS] = { 0 }; - -/* A table of GINTn interrupts we actually use. - Note that we don't use GINT0 because all the boards we support treat it - specially. */ -struct used_gint { - unsigned gint; - unsigned priority; -} used_gint[] = { - { 1, GBUS_INT_PRIORITY_HIGH }, - { 3, GBUS_INT_PRIORITY_LOW } -}; -#define NUM_USED_GINTS ARRAY_SIZE(used_gint) - -/* A table of which GINT is used by each GBUS interrupts (they are - assigned based on priority). */ -static unsigned char gbus_int_gint[IRQ_GBUS_INT_NUM]; - - -/* Interrupt enabling/disabling. */ - -/* Enable interrupt handling for interrupt IRQ. */ -void gbus_int_enable_irq (unsigned irq) -{ - unsigned gint = gbus_int_gint[irq - GBUS_INT_BASE_IRQ]; - GBUS_INT_ENABLE (GBUS_INT_IRQ_WORD(irq), gint) - |= GBUS_INT_IRQ_MASK (irq); -} - -/* Disable interrupt handling for interrupt IRQ. Note that any - interrupts received while disabled will be delivered once the - interrupt is enabled again, unless they are explicitly cleared using - `gbus_int_clear_pending_irq'. */ -void gbus_int_disable_irq (unsigned irq) -{ - unsigned gint = gbus_int_gint[irq - GBUS_INT_BASE_IRQ]; - GBUS_INT_ENABLE (GBUS_INT_IRQ_WORD(irq), gint) - &= ~GBUS_INT_IRQ_MASK (irq); -} - -/* Return true if interrupt handling for interrupt IRQ is enabled. */ -int gbus_int_irq_enabled (unsigned irq) -{ - unsigned gint = gbus_int_gint[irq - GBUS_INT_BASE_IRQ]; - return (GBUS_INT_ENABLE (GBUS_INT_IRQ_WORD(irq), gint) - & GBUS_INT_IRQ_MASK(irq)); -} - -/* Disable all GBUS irqs. */ -void gbus_int_disable_irqs () -{ - unsigned w, n; - for (w = 0; w < GBUS_INT_NUM_WORDS; w++) - for (n = 0; n < IRQ_GINT_NUM; n++) - GBUS_INT_ENABLE (w, n) = 0; -} - -/* Clear any pending interrupts for IRQ. */ -void gbus_int_clear_pending_irq (unsigned irq) -{ - GBUS_INT_CLEAR (GBUS_INT_IRQ_WORD(irq)) = GBUS_INT_IRQ_MASK (irq); -} - -/* Return true if interrupt IRQ is pending (but disabled). */ -int gbus_int_irq_pending (unsigned irq) -{ - return (GBUS_INT_STATUS (GBUS_INT_IRQ_WORD(irq)) - & GBUS_INT_IRQ_MASK(irq)); -} - - -/* Delegating interrupts. */ - -/* Handle a shared GINT interrupt by passing to the appropriate GBUS - interrupt handler. */ -static irqreturn_t gbus_int_handle_irq (int irq, void *dev_id, - struct pt_regs *regs) -{ - unsigned w; - irqreturn_t rval = IRQ_NONE; - unsigned gint = irq - IRQ_GINT (0); - - for (w = 0; w < GBUS_INT_NUM_WORDS; w++) { - unsigned status = GBUS_INT_STATUS (w); - unsigned enable = GBUS_INT_ENABLE (w, gint); - - /* Only pay attention to enabled interrupts. */ - status &= enable; - if (status) { - irq = IRQ_GBUS_INT (w * GBUS_INT_BITS_PER_WORD); - do { - /* There's an active interrupt in word - W, find out which one, and call its - handler. */ - - while (! (status & 0x1)) { - irq++; - status >>= 1; - } - status &= ~0x1; - - /* Recursively call handle_irq to handle it. */ - handle_irq (irq, regs); - rval = IRQ_HANDLED; - } while (status); - } - } - - /* Toggle the `all enable' bit back and forth, which should cause - another edge transition if there are any other interrupts - still pending, and so result in another CPU interrupt. */ - GBUS_INT_ENABLE (0, gint) &= ~0x1; - GBUS_INT_ENABLE (0, gint) |= 0x1; - - return rval; -} - - -/* Initialize GBUS interrupt sources. */ - -static void irq_nop (unsigned irq) { } - -static unsigned gbus_int_startup_irq (unsigned irq) -{ - unsigned gint = gbus_int_gint[irq - GBUS_INT_BASE_IRQ]; - - if (gint_num_active_irqs[gint] == 0) { - /* First enable the CPU interrupt. */ - int rval = - request_irq (IRQ_GINT(gint), gbus_int_handle_irq, - IRQF_DISABLED, - "gbus_int_handler", - &gint_num_active_irqs[gint]); - if (rval != 0) - return rval; - } - - gint_num_active_irqs[gint]++; - - gbus_int_clear_pending_irq (irq); - gbus_int_enable_irq (irq); - - return 0; -} - -static void gbus_int_shutdown_irq (unsigned irq) -{ - unsigned gint = gbus_int_gint[irq - GBUS_INT_BASE_IRQ]; - - gbus_int_disable_irq (irq); - - if (--gint_num_active_irqs[gint] == 0) - /* Disable the CPU interrupt. */ - free_irq (IRQ_GINT(gint), &gint_num_active_irqs[gint]); -} - -/* Initialize HW_IRQ_TYPES for INTC-controlled irqs described in array - INITS (which is terminated by an entry with the name field == 0). */ -void __init gbus_int_init_irq_types (struct gbus_int_irq_init *inits, - struct hw_interrupt_type *hw_irq_types) -{ - struct gbus_int_irq_init *init; - for (init = inits; init->name; init++) { - unsigned i; - struct hw_interrupt_type *hwit = hw_irq_types++; - - hwit->typename = init->name; - - hwit->startup = gbus_int_startup_irq; - hwit->shutdown = gbus_int_shutdown_irq; - hwit->enable = gbus_int_enable_irq; - hwit->disable = gbus_int_disable_irq; - hwit->ack = irq_nop; - hwit->end = irq_nop; - - /* Initialize kernel IRQ infrastructure for this interrupt. */ - init_irq_handlers(init->base, init->num, init->interval, hwit); - - /* Set the interrupt priorities. */ - for (i = 0; i < init->num; i++) { - unsigned j; - for (j = 0; j < NUM_USED_GINTS; j++) - if (used_gint[j].priority > init->priority) - break; - /* Wherever we stopped looking is one past the - GINT we want. */ - gbus_int_gint[init->base + i * init->interval - - GBUS_INT_BASE_IRQ] - = used_gint[j > 0 ? j - 1 : 0].gint; - } - } -} - - -/* Initialize IRQS. */ - -/* Chip interrupts (GINTn) shared among GBUS interrupts. */ -static struct hw_interrupt_type gint_hw_itypes[NUM_USED_GINTS]; - - -/* GBUS interrupts themselves. */ - -struct gbus_int_irq_init gbus_irq_inits[] __initdata = { - /* First set defaults. */ - { "GBUS_INT", IRQ_GBUS_INT(0), IRQ_GBUS_INT_NUM, 1, 6}, - { 0 } -}; -#define NUM_GBUS_IRQ_INITS (ARRAY_SIZE(gbus_irq_inits) - 1) - -static struct hw_interrupt_type gbus_hw_itypes[NUM_GBUS_IRQ_INITS]; - - -/* Initialize GBUS interrupts. */ -void __init gbus_int_init_irqs (void) -{ - unsigned i; - - /* First initialize the shared gint interrupts. */ - for (i = 0; i < NUM_USED_GINTS; i++) { - unsigned gint = used_gint[i].gint; - struct v850e_intc_irq_init gint_irq_init[2]; - - /* We initialize one GINT interrupt at a time. */ - gint_irq_init[0].name = "GINT"; - gint_irq_init[0].base = IRQ_GINT (gint); - gint_irq_init[0].num = 1; - gint_irq_init[0].interval = 1; - gint_irq_init[0].priority = used_gint[i].priority; - - gint_irq_init[1].name = 0; /* Terminate the vector. */ - - v850e_intc_init_irq_types (gint_irq_init, gint_hw_itypes); - } - - /* Then the GBUS interrupts. */ - gbus_int_disable_irqs (); - gbus_int_init_irq_types (gbus_irq_inits, gbus_hw_itypes); - /* Turn on the `all enable' bits, which are ANDed with - individual interrupt enable bits; we only want to bother with - the latter. They are the first bit in the first word of each - interrupt-enable area. */ - for (i = 0; i < NUM_USED_GINTS; i++) - GBUS_INT_ENABLE (0, used_gint[i].gint) = 0x1; -} diff --git a/arch/v850/kernel/head.S b/arch/v850/kernel/head.S deleted file mode 100644 index c490b937ef14..000000000000 --- a/arch/v850/kernel/head.S +++ /dev/null @@ -1,128 +0,0 @@ -/* - * arch/v850/kernel/head.S -- Lowest-level startup code - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include -#include -#include -#include - - -/* Make a slightly more convenient alias for C_SYMBOL_NAME. */ -#define CSYM C_SYMBOL_NAME - - - .text - - // Define `mach_early_init' as a weak symbol - .global CSYM(mach_early_init) - .weak CSYM(mach_early_init) - -C_ENTRY(start): - // Make sure interrupts are turned off, just in case - di - -#ifdef CONFIG_RESET_GUARD - // See if we got here via an unexpected reset - ld.w RESET_GUARD, r19 // Check current value of reset guard - mov RESET_GUARD_ACTIVE, r20 - cmp r19, r20 - bne 1f // Guard was not active - - // If we get here, the reset guard was active. Load up some - // interesting values as arguments, and jump to the handler. - st.w r0, RESET_GUARD // Allow further resets to succeed - mov lp, r6 // Arg 0: return address - ld.b KM, r7 // Arg 1: kernel mode - mov sp, r9 // Arg 3: stack pointer - ld.w KSP, r19 // maybe switch to kernel stack - cmp r7, r0 // see if already in kernel mode - cmov z, r19, sp, sp // and switch to kernel stack if not - GET_CURRENT_TASK(r8) // Arg 2: task pointer - jr CSYM(unexpected_reset) - -1: st.w r20, RESET_GUARD // Turn on reset guard -#endif /* CONFIG_RESET_GUARD */ - - // Setup a temporary stack for doing pre-initialization function calls. - // - // We can't use the initial kernel stack, because (1) it may be - // located in memory we're not allowed to touch, and (2) since - // it's in the data segment, calling memcpy to initialize that - // area from ROM will overwrite memcpy's return address. - mov hilo(CSYM(_init_stack_end) - 4), sp - - // See if there's a platform-specific early-initialization routine - // defined; it's a weak symbol, so it will have an address of zero if - // there's not. - mov hilo(CSYM(mach_early_init)), r6 - cmp r6, r0 - bz 3f - - // There is one, so call it. If this function is written in C, it - // should be very careful -- the stack pointer is valid, but very - // little else is (e.g., bss is not zeroed yet, and initialized data - // hasn't been). - jarl 2f, lp // first figure out return address -2: add 3f - ., lp - jmp [r6] // do call -3: - -#ifdef CONFIG_ROM_KERNEL - // Copy the data area from ROM to RAM - mov hilo(CSYM(_rom_copy_dst_start)), r6 - mov hilo(CSYM(_rom_copy_src_start)), r7 - mov hilo(CSYM(_rom_copy_dst_end)), r8 - sub r6, r8 - jarl CSYM(memcpy), lp -#endif - - // Load the initial thread's stack, and current task pointer (in r16) - mov hilo(CSYM(init_thread_union)), r19 - movea THREAD_SIZE, r19, sp - ld.w TI_TASK[r19], CURRENT_TASK - -#ifdef CONFIG_TIME_BOOTUP - /* This stuff must come after mach_early_init, because interrupts may - not work until after its been called. */ - jarl CSYM(highres_timer_reset), lp - jarl CSYM(highres_timer_start), lp -#endif - - // Kernel stack pointer save location - st.w sp, KSP - - // Assert that we're in `kernel mode' - mov 1, r19 - st.w r19, KM - -#ifdef CONFIG_ZERO_BSS - // Zero bss area, since we can't rely upon any loader to do so - mov hilo(CSYM(_sbss)), r6 - mov r0, r7 - mov hilo(CSYM(_ebss)), r8 - sub r6, r8 - jarl CSYM(memset), lp -#endif - - // What happens if the main kernel function returns (it shouldn't) - mov hilo(CSYM(machine_halt)), lp - - // Start the linux kernel. We use an indirect jump to get extra - // range, because on some platforms this initial startup code - // (and the associated platform-specific code in mach_early_init) - // are located far away from the main kernel, e.g. so that they - // can initialize RAM first and copy the kernel or something. - mov hilo(CSYM(start_kernel)), r12 - jmp [r12] -C_END(start) diff --git a/arch/v850/kernel/highres_timer.c b/arch/v850/kernel/highres_timer.c deleted file mode 100644 index b16ad1eaf966..000000000000 --- a/arch/v850/kernel/highres_timer.c +++ /dev/null @@ -1,132 +0,0 @@ -/* - * arch/v850/kernel/highres_timer.c -- High resolution timing routines - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include -#include - -#define HIGHRES_TIMER_USEC_SHIFT 12 - -/* Pre-calculated constant used for converting ticks to real time - units. We initialize it to prevent it being put into BSS. */ -static u32 highres_timer_usec_prescale = 1; - -void highres_timer_slow_tick_irq (void) __attribute__ ((noreturn)); -void highres_timer_slow_tick_irq (void) -{ - /* This is an interrupt handler, so it must be very careful to - not to trash any registers. At this point, the stack-pointer - (r3) has been saved in the chip ram location ENTRY_SP by the - interrupt vector, so we can use it as a scratch register; we - must also restore it before returning. */ - asm ("ld.w %0[r0], sp;" - "add 1, sp;" - "st.w sp, %0[r0];" - "ld.w %1[r0], sp;" /* restore pre-irq stack-pointer */ - "reti" - :: - "i" (HIGHRES_TIMER_SLOW_TICKS_ADDR), - "i" (ENTRY_SP_ADDR) - : "memory"); -} - -void highres_timer_reset (void) -{ - V850E_TIMER_D_TMD (HIGHRES_TIMER_TIMER_D_UNIT) = 0; - HIGHRES_TIMER_SLOW_TICKS = 0; -} - -void highres_timer_start (void) -{ - u32 fast_tick_rate; - - /* Start hardware timer. */ - v850e_timer_d_configure (HIGHRES_TIMER_TIMER_D_UNIT, - HIGHRES_TIMER_SLOW_TICK_RATE); - - fast_tick_rate = - (V850E_TIMER_D_BASE_FREQ - >> V850E_TIMER_D_DIVLOG2 (HIGHRES_TIMER_TIMER_D_UNIT)); - - /* The obvious way of calculating microseconds from fast ticks - is to do: - - usec = fast_ticks * 10^6 / fast_tick_rate - - However, divisions are much slower than multiplications, and - the above calculation can overflow, so we do this instead: - - usec = fast_ticks * (10^6 * 2^12 / fast_tick_rate) / 2^12 - - since we can pre-calculate (10^6 * (2^12 / fast_tick_rate)) - and use a shift for dividing by 2^12, this avoids division, - and is almost as accurate (it differs by about 2 microseconds - at the extreme value of the fast-tick counter's ranger). */ - highres_timer_usec_prescale = ((1000000 << HIGHRES_TIMER_USEC_SHIFT) - / fast_tick_rate); - - /* Enable the interrupt (which is hardwired to this use), and - give it the highest priority. */ - V850E_INTC_IC (IRQ_INTCMD (HIGHRES_TIMER_TIMER_D_UNIT)) = 0; -} - -void highres_timer_stop (void) -{ - /* Stop the timer. */ - V850E_TIMER_D_TMCD (HIGHRES_TIMER_TIMER_D_UNIT) = - V850E_TIMER_D_TMCD_CAE; - /* Disable its interrupt, just in case. */ - v850e_intc_disable_irq (IRQ_INTCMD (HIGHRES_TIMER_TIMER_D_UNIT)); -} - -inline void highres_timer_read_ticks (u32 *slow_ticks, u32 *fast_ticks) -{ - int flags; - u32 fast_ticks_1, fast_ticks_2, _slow_ticks; - - local_irq_save (flags); - fast_ticks_1 = V850E_TIMER_D_TMD (HIGHRES_TIMER_TIMER_D_UNIT); - _slow_ticks = HIGHRES_TIMER_SLOW_TICKS; - fast_ticks_2 = V850E_TIMER_D_TMD (HIGHRES_TIMER_TIMER_D_UNIT); - local_irq_restore (flags); - - if (fast_ticks_2 < fast_ticks_1) - _slow_ticks++; - - *slow_ticks = _slow_ticks; - *fast_ticks = fast_ticks_2; -} - -inline void highres_timer_ticks_to_timeval (u32 slow_ticks, u32 fast_ticks, - struct timeval *tv) -{ - unsigned long sec, sec_rem, usec; - - usec = ((fast_ticks * highres_timer_usec_prescale) - >> HIGHRES_TIMER_USEC_SHIFT); - - sec = slow_ticks / HIGHRES_TIMER_SLOW_TICK_RATE; - sec_rem = slow_ticks % HIGHRES_TIMER_SLOW_TICK_RATE; - - usec += sec_rem * (1000000 / HIGHRES_TIMER_SLOW_TICK_RATE); - - tv->tv_sec = sec; - tv->tv_usec = usec; -} - -void highres_timer_read (struct timeval *tv) -{ - u32 fast_ticks, slow_ticks; - highres_timer_read_ticks (&slow_ticks, &fast_ticks); - highres_timer_ticks_to_timeval (slow_ticks, fast_ticks, tv); -} diff --git a/arch/v850/kernel/init_task.c b/arch/v850/kernel/init_task.c deleted file mode 100644 index 44b274dff33f..000000000000 --- a/arch/v850/kernel/init_task.c +++ /dev/null @@ -1,48 +0,0 @@ -/* - * arch/v850/kernel/init_task.c -- Initial task/thread structures - * - * Copyright (C) 2002,03 NEC Electronics Corporation - * Copyright (C) 2002,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - */ - -#include -#include -#include -#include -#include -#include -#include - -#include -#include - -static struct fs_struct init_fs = INIT_FS; -static struct signal_struct init_signals = INIT_SIGNALS (init_signals); -static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand); -struct mm_struct init_mm = INIT_MM (init_mm); - -EXPORT_SYMBOL(init_mm); - -/* - * Initial task structure. - * - * All other task structs will be allocated on slabs in fork.c - */ -struct task_struct init_task = INIT_TASK (init_task); - -EXPORT_SYMBOL(init_task); - -/* - * Initial thread structure. - * - * We need to make sure that this is 8192-byte aligned due to the - * way process stacks are handled. This is done by having a special - * "init_task" linker map entry. - */ -union thread_union init_thread_union - __attribute__((__section__(".data.init_task"))) = - { INIT_THREAD_INFO(init_task) }; diff --git a/arch/v850/kernel/intv.S b/arch/v850/kernel/intv.S deleted file mode 100644 index 671e4c6150dd..000000000000 --- a/arch/v850/kernel/intv.S +++ /dev/null @@ -1,87 +0,0 @@ -/* - * arch/v850/kernel/intv.S -- Interrupt vectors - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include -#include -#include - -#ifdef CONFIG_V850E_HIGHRES_TIMER -#include -#endif - -/* Jump to an interrupt/trap handler. These handlers (defined in entry.S) - expect the stack-pointer to be saved in ENTRY_SP, so we use sp to do an - indirect jump (which avoids problems when the handler is more than a signed - 22-bit offset away). */ -#define JUMP_TO_HANDLER(name, sp_save_loc) \ - st.w sp, sp_save_loc; \ - mov hilo(name), sp; \ - jmp [sp] - - - /* Reset vector. */ - .section .intv.reset, "ax" - .org 0x0 - mov hilo(C_SYMBOL_NAME(start)), r1; - jmp [r1] - - - /* Generic interrupt vectors. */ - .section .intv.common, "ax" - .balign 0x10 - JUMP_TO_HANDLER (nmi, NMI_ENTRY_SP) // 0x10 - NMI0 - .balign 0x10 - JUMP_TO_HANDLER (nmi, NMI_ENTRY_SP) // 0x20 - NMI1 - .balign 0x10 - JUMP_TO_HANDLER (nmi, NMI_ENTRY_SP) // 0x30 - NMI2 - - .balign 0x10 - JUMP_TO_HANDLER (trap, ENTRY_SP) // 0x40 - TRAP0n - .balign 0x10 - JUMP_TO_HANDLER (trap, ENTRY_SP) // 0x50 - TRAP1n - - .balign 0x10 - JUMP_TO_HANDLER (dbtrap, ENTRY_SP) // 0x60 - Illegal op / DBTRAP insn - - - /* Hardware interrupt vectors. */ - .section .intv.mach, "ax" - .org 0x0 - -#if defined (CONFIG_V850E_HIGHRES_TIMER) && defined (IRQ_INTCMD) - - /* Interrupts before the highres timer interrupt. */ - .rept IRQ_INTCMD (HIGHRES_TIMER_TIMER_D_UNIT) - .balign 0x10 - JUMP_TO_HANDLER (irq, ENTRY_SP) - .endr - - /* The highres timer interrupt. */ - .balign 0x10 - JUMP_TO_HANDLER (C_SYMBOL_NAME (highres_timer_slow_tick_irq), ENTRY_SP) - - /* Interrupts after the highres timer interrupt. */ - .rept NUM_CPU_IRQS - IRQ_INTCMD (HIGHRES_TIMER_TIMER_D_UNIT) - 1 - .balign 0x10 - JUMP_TO_HANDLER (irq, ENTRY_SP) - .endr - -#else /* No highres timer */ - - .rept NUM_CPU_IRQS - .balign 0x10 - JUMP_TO_HANDLER (irq, ENTRY_SP) - .endr - -#endif /* Highres timer */ diff --git a/arch/v850/kernel/irq.c b/arch/v850/kernel/irq.c deleted file mode 100644 index 858c45819aab..000000000000 --- a/arch/v850/kernel/irq.c +++ /dev/null @@ -1,123 +0,0 @@ -/* - * arch/v850/kernel/irq.c -- High-level interrupt handling - * - * Copyright (C) 2001,02,03,04,05 NEC Electronics Corporation - * Copyright (C) 2001,02,03,04,05 Miles Bader - * Copyright (C) 1994-2000 Ralf Baechle - * Copyright (C) 1992 Linus Torvalds - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * This file was was derived from the mips version, arch/mips/kernel/irq.c - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include - -/* - * 'what should we do if we get a hw irq event on an illegal vector'. - * each architecture has to answer this themselves, it doesn't deserve - * a generic callback i think. - */ -void ack_bad_irq(unsigned int irq) -{ - printk("received IRQ %d with unknown interrupt type\n", irq); -} - -volatile unsigned long irq_err_count, spurious_count; - -/* - * Generic, controller-independent functions: - */ - -int show_interrupts(struct seq_file *p, void *v) -{ - int irq = *(loff_t *) v; - - if (irq == 0) { - int cpu; - seq_puts(p, " "); - for (cpu=0; cpu < 1 /*smp_num_cpus*/; cpu++) - seq_printf(p, "CPU%d ", cpu); - seq_putc(p, '\n'); - } - - if (irq < NR_IRQS) { - unsigned long flags; - struct irqaction *action; - - spin_lock_irqsave(&irq_desc[irq].lock, flags); - - action = irq_desc[irq].action; - if (action) { - int j; - int count = 0; - int num = -1; - const char *type_name = irq_desc[irq].chip->typename; - - for (j = 0; j < NR_IRQS; j++) - if (irq_desc[j].chip->typename == type_name){ - if (irq == j) - num = count; - count++; - } - - seq_printf(p, "%3d: ",irq); - seq_printf(p, "%10u ", kstat_irqs(irq)); - if (count > 1) { - int prec = (num >= 100 ? 3 : num >= 10 ? 2 : 1); - seq_printf(p, " %*s%d", 14 - prec, - type_name, num); - } else - seq_printf(p, " %14s", type_name); - - seq_printf(p, " %s", action->name); - for (action=action->next; action; action = action->next) - seq_printf(p, ", %s", action->name); - seq_putc(p, '\n'); - } - - spin_unlock_irqrestore(&irq_desc[irq].lock, flags); - } else if (irq == NR_IRQS) - seq_printf(p, "ERR: %10lu\n", irq_err_count); - - return 0; -} - -/* Handle interrupt IRQ. REGS are the registers at the time of ther - interrupt. */ -unsigned int handle_irq (int irq, struct pt_regs *regs) -{ - irq_enter(); - __do_IRQ(irq, regs); - irq_exit(); - return 1; -} - -/* Initialize irq handling for IRQs. - BASE_IRQ, BASE_IRQ+INTERVAL, ..., BASE_IRQ+NUM*INTERVAL - to IRQ_TYPE. An IRQ_TYPE of 0 means to use a generic interrupt type. */ -void __init -init_irq_handlers (int base_irq, int num, int interval, - struct hw_interrupt_type *irq_type) -{ - while (num-- > 0) { - irq_desc[base_irq].status = IRQ_DISABLED; - irq_desc[base_irq].action = NULL; - irq_desc[base_irq].depth = 1; - irq_desc[base_irq].chip = irq_type; - base_irq += interval; - } -} diff --git a/arch/v850/kernel/ma.c b/arch/v850/kernel/ma.c deleted file mode 100644 index 143774de75e1..000000000000 --- a/arch/v850/kernel/ma.c +++ /dev/null @@ -1,69 +0,0 @@ -/* - * arch/v850/kernel/ma.c -- V850E/MA series of cpu chips - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include - -#include "mach.h" - -void __init mach_sched_init (struct irqaction *timer_action) -{ - /* Start hardware timer. */ - v850e_timer_d_configure (0, HZ); - /* Install timer interrupt handler. */ - setup_irq (IRQ_INTCMD(0), timer_action); -} - -static struct v850e_intc_irq_init irq_inits[] = { - { "IRQ", 0, NUM_MACH_IRQS, 1, 7 }, - { "CMD", IRQ_INTCMD(0), IRQ_INTCMD_NUM, 1, 5 }, - { "DMA", IRQ_INTDMA(0), IRQ_INTDMA_NUM, 1, 2 }, - { "CSI", IRQ_INTCSI(0), IRQ_INTCSI_NUM, 4, 4 }, - { "SER", IRQ_INTSER(0), IRQ_INTSER_NUM, 4, 3 }, - { "SR", IRQ_INTSR(0), IRQ_INTSR_NUM, 4, 4 }, - { "ST", IRQ_INTST(0), IRQ_INTST_NUM, 4, 5 }, - { 0 } -}; -#define NUM_IRQ_INITS (ARRAY_SIZE(irq_inits) - 1) - -static struct hw_interrupt_type hw_itypes[NUM_IRQ_INITS]; - -/* Initialize MA chip interrupts. */ -void __init ma_init_irqs (void) -{ - v850e_intc_init_irq_types (irq_inits, hw_itypes); -} - -/* Called before configuring an on-chip UART. */ -void ma_uart_pre_configure (unsigned chan, unsigned cflags, unsigned baud) -{ - /* We only know about the first two UART channels (though - specific chips may have more). */ - if (chan < 2) { - unsigned bits = 0x3 << (chan * 3); - /* Specify that the relevant pins on the chip should do - serial I/O, not direct I/O. */ - MA_PORT4_PMC |= bits; - /* Specify that we're using the UART, not the CSI device. */ - MA_PORT4_PFC |= bits; - } -} diff --git a/arch/v850/kernel/mach.c b/arch/v850/kernel/mach.c deleted file mode 100644 index b9db278d2b71..000000000000 --- a/arch/v850/kernel/mach.c +++ /dev/null @@ -1,17 +0,0 @@ -/* - * arch/v850/kernel/mach.c -- Defaults for some things defined by "mach.h" - * - * Copyright (C) 2001 NEC Corporation - * Copyright (C) 2001 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include "mach.h" - -/* Called with each timer tick, if non-zero. */ -void (*mach_tick)(void) = 0; diff --git a/arch/v850/kernel/mach.h b/arch/v850/kernel/mach.h deleted file mode 100644 index 9e0e4816ec56..000000000000 --- a/arch/v850/kernel/mach.h +++ /dev/null @@ -1,56 +0,0 @@ -/* - * arch/v850/kernel/mach.h -- Machine-dependent functions used by v850 port - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_MACH_H__ -#define __V850_MACH_H__ - -#include -#include -#include -#include -#include -#include - -#include -#include -#include - -void mach_setup (char **cmdline); -void mach_gettimeofday (struct timespec *tv); -void mach_sched_init (struct irqaction *timer_action); -void mach_get_physical_ram (unsigned long *ram_start, unsigned long *ram_len); -void mach_init_irqs (void); - -/* If defined, is called very early in the kernel initialization. The - stack pointer is valid, but very little has been initialized (e.g., - bss is not zeroed yet) when this is called, so care must taken. */ -void mach_early_init (void); - -/* If defined, called after the bootmem allocator has been initialized, - to allow the platform-dependent code to reserve any areas of RAM that - the kernel shouldn't touch. */ -void mach_reserve_bootmem (void) __attribute__ ((__weak__)); - -/* Called with each timer tick, if non-zero. */ -extern void (*mach_tick) (void); - -/* The following establishes aliases for various mach_ functions to the - name by which the rest of the kernel calls them. These statements - should only have an effect in the file that defines the actual functions. */ -#define MACH_ALIAS(to, from) \ - asm (".global " macrology_stringify (C_SYMBOL_NAME (to)) ";" \ - macrology_stringify (C_SYMBOL_NAME (to)) \ - " = " macrology_stringify (C_SYMBOL_NAME (from))) -/* e.g.: MACH_ALIAS (kernel_name, arch_spec_name); */ - -#endif /* __V850_MACH_H__ */ diff --git a/arch/v850/kernel/me2.c b/arch/v850/kernel/me2.c deleted file mode 100644 index 007115dc9ce0..000000000000 --- a/arch/v850/kernel/me2.c +++ /dev/null @@ -1,73 +0,0 @@ -/* - * arch/v850/kernel/me2.c -- V850E/ME2 chip-specific support - * - * Copyright (C) 2003 NEC Corporation - * Copyright (C) 2003 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include - -#include "mach.h" - -void __init mach_sched_init (struct irqaction *timer_action) -{ - /* Start hardware timer. */ - v850e_timer_d_configure (0, HZ); - /* Install timer interrupt handler. */ - setup_irq (IRQ_INTCMD(0), timer_action); -} - -static struct v850e_intc_irq_init irq_inits[] = { - { "IRQ", 0, NUM_CPU_IRQS, 1, 7 }, - { "INTP", IRQ_INTP(0), IRQ_INTP_NUM, 1, 5 }, - { "CMD", IRQ_INTCMD(0), IRQ_INTCMD_NUM, 1, 3 }, - { "UBTIRE", IRQ_INTUBTIRE(0), IRQ_INTUBTIRE_NUM, 5, 4 }, - { "UBTIR", IRQ_INTUBTIR(0), IRQ_INTUBTIR_NUM, 5, 4 }, - { "UBTIT", IRQ_INTUBTIT(0), IRQ_INTUBTIT_NUM, 5, 4 }, - { "UBTIF", IRQ_INTUBTIF(0), IRQ_INTUBTIF_NUM, 5, 4 }, - { "UBTITO", IRQ_INTUBTITO(0), IRQ_INTUBTITO_NUM, 5, 4 }, - { 0 } -}; -#define NUM_IRQ_INITS (ARRAY_SIZE(irq_inits) - 1) - -static struct hw_interrupt_type hw_itypes[NUM_IRQ_INITS]; - -/* Initialize V850E/ME2 chip interrupts. */ -void __init me2_init_irqs (void) -{ - v850e_intc_init_irq_types (irq_inits, hw_itypes); -} - -/* Called before configuring an on-chip UART. */ -void me2_uart_pre_configure (unsigned chan, unsigned cflags, unsigned baud) -{ - if (chan == 0) { - /* Specify that the relevant pins on the chip should do - serial I/O, not direct I/O. */ - ME2_PORT1_PMC |= 0xC; - /* Specify that we're using the UART, not the CSI device. */ - ME2_PORT1_PFC |= 0xC; - } else if (chan == 1) { - /* Specify that the relevant pins on the chip should do - serial I/O, not direct I/O. */ - ME2_PORT2_PMC |= 0x6; - /* Specify that we're using the UART, not the CSI device. */ - ME2_PORT2_PFC |= 0x6; - } -} diff --git a/arch/v850/kernel/memcons.c b/arch/v850/kernel/memcons.c deleted file mode 100644 index 92f514fdcc79..000000000000 --- a/arch/v850/kernel/memcons.c +++ /dev/null @@ -1,135 +0,0 @@ -/* - * arch/v850/kernel/memcons.c -- Console I/O to a memory buffer - * - * Copyright (C) 2001,02 NEC Corporation - * Copyright (C) 2001,02 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include -#include -#include -#include - -/* If this device is enabled, the linker map should define start and - end points for its buffer. */ -extern char memcons_output[], memcons_output_end; - -/* Current offset into the buffer. */ -static unsigned long memcons_offs = 0; - -/* Spinlock protecting memcons_offs. */ -static DEFINE_SPINLOCK(memcons_lock); - - -static size_t write (const char *buf, size_t len) -{ - unsigned long flags; - char *point; - - spin_lock_irqsave (memcons_lock, flags); - - point = memcons_output + memcons_offs; - if (point + len >= &memcons_output_end) { - len = &memcons_output_end - point; - memcons_offs = 0; - } else - memcons_offs += len; - - spin_unlock_irqrestore (memcons_lock, flags); - - memcpy (point, buf, len); - - return len; -} - - -/* Low-level console. */ - -static void memcons_write (struct console *co, const char *buf, unsigned len) -{ - while (len > 0) - len -= write (buf, len); -} - -static struct tty_driver *tty_driver; - -static struct tty_driver *memcons_device (struct console *co, int *index) -{ - *index = co->index; - return tty_driver; -} - -static struct console memcons = -{ - .name = "memcons", - .write = memcons_write, - .device = memcons_device, - .flags = CON_PRINTBUFFER, - .index = -1, -}; - -void memcons_setup (void) -{ - register_console (&memcons); - printk (KERN_INFO "Console: static memory buffer (memcons)\n"); -} - -/* Higher level TTY interface. */ - -int memcons_tty_open (struct tty_struct *tty, struct file *filp) -{ - return 0; -} - -int memcons_tty_write (struct tty_struct *tty, const unsigned char *buf, int len) -{ - return write (buf, len); -} - -int memcons_tty_write_room (struct tty_struct *tty) -{ - return &memcons_output_end - (memcons_output + memcons_offs); -} - -int memcons_tty_chars_in_buffer (struct tty_struct *tty) -{ - /* We have no buffer. */ - return 0; -} - -static const struct tty_operations ops = { - .open = memcons_tty_open, - .write = memcons_tty_write, - .write_room = memcons_tty_write_room, - .chars_in_buffer = memcons_tty_chars_in_buffer, -}; - -int __init memcons_tty_init (void) -{ - int err; - struct tty_driver *driver = alloc_tty_driver(1); - if (!driver) - return -ENOMEM; - - driver->name = "memcons"; - driver->major = TTY_MAJOR; - driver->minor_start = 64; - driver->type = TTY_DRIVER_TYPE_SYSCONS; - driver->init_termios = tty_std_termios; - tty_set_operations(driver, &ops); - err = tty_register_driver(driver); - if (err) { - put_tty_driver(driver); - return err; - } - tty_driver = driver; - return 0; -} -__initcall (memcons_tty_init); diff --git a/arch/v850/kernel/module.c b/arch/v850/kernel/module.c deleted file mode 100644 index 64aeb3e37c52..000000000000 --- a/arch/v850/kernel/module.c +++ /dev/null @@ -1,237 +0,0 @@ -/* - * arch/v850/kernel/module.c -- Architecture-specific module functions - * - * Copyright (C) 2002,03 NEC Electronics Corporation - * Copyright (C) 2002,03 Miles Bader - * Copyright (C) 2001,03 Rusty Russell - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - * - * Derived in part from arch/ppc/kernel/module.c - */ - -#include -#include -#include -#include - -#if 0 -#define DEBUGP printk -#else -#define DEBUGP(fmt , ...) -#endif - -void *module_alloc (unsigned long size) -{ - return size == 0 ? 0 : vmalloc (size); -} - -void module_free (struct module *mod, void *module_region) -{ - vfree (module_region); - /* FIXME: If module_region == mod->init_region, trim exception - table entries. */ -} - -int module_finalize (const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs, - struct module *mod) -{ - return 0; -} - -/* Count how many different relocations (different symbol, different - addend) */ -static unsigned int count_relocs(const Elf32_Rela *rela, unsigned int num) -{ - unsigned int i, j, ret = 0; - - /* Sure, this is order(n^2), but it's usually short, and not - time critical */ - for (i = 0; i < num; i++) { - for (j = 0; j < i; j++) { - /* If this addend appeared before, it's - already been counted */ - if (ELF32_R_SYM(rela[i].r_info) - == ELF32_R_SYM(rela[j].r_info) - && rela[i].r_addend == rela[j].r_addend) - break; - } - if (j == i) ret++; - } - return ret; -} - -/* Get the potential trampolines size required of the init and - non-init sections */ -static unsigned long get_plt_size(const Elf32_Ehdr *hdr, - const Elf32_Shdr *sechdrs, - const char *secstrings, - int is_init) -{ - unsigned long ret = 0; - unsigned i; - - /* Everything marked ALLOC (this includes the exported - symbols) */ - for (i = 1; i < hdr->e_shnum; i++) { - /* If it's called *.init*, and we're not init, we're - not interested */ - if ((strstr(secstrings + sechdrs[i].sh_name, ".init") != 0) - != is_init) - continue; - - if (sechdrs[i].sh_type == SHT_RELA) { - DEBUGP("Found relocations in section %u\n", i); - DEBUGP("Ptr: %p. Number: %u\n", - (void *)hdr + sechdrs[i].sh_offset, - sechdrs[i].sh_size / sizeof(Elf32_Rela)); - ret += count_relocs((void *)hdr - + sechdrs[i].sh_offset, - sechdrs[i].sh_size - / sizeof(Elf32_Rela)) - * sizeof(struct v850_plt_entry); - } - } - - return ret; -} - -int module_frob_arch_sections(Elf32_Ehdr *hdr, - Elf32_Shdr *sechdrs, - char *secstrings, - struct module *me) -{ - unsigned int i; - - /* Find .plt and .pltinit sections */ - for (i = 0; i < hdr->e_shnum; i++) { - if (strcmp(secstrings + sechdrs[i].sh_name, ".init.plt") == 0) - me->arch.init_plt_section = i; - else if (strcmp(secstrings + sechdrs[i].sh_name, ".plt") == 0) - me->arch.core_plt_section = i; - } - if (!me->arch.core_plt_section || !me->arch.init_plt_section) { - printk("Module doesn't contain .plt or .plt.init sections.\n"); - return -ENOEXEC; - } - - /* Override their sizes */ - sechdrs[me->arch.core_plt_section].sh_size - = get_plt_size(hdr, sechdrs, secstrings, 0); - sechdrs[me->arch.init_plt_section].sh_size - = get_plt_size(hdr, sechdrs, secstrings, 1); - return 0; -} - -int apply_relocate (Elf32_Shdr *sechdrs, const char *strtab, - unsigned int symindex, unsigned int relsec, - struct module *mod) -{ - printk ("Barf\n"); - return -ENOEXEC; -} - -/* Set up a trampoline in the PLT to bounce us to the distant function */ -static uint32_t do_plt_call (void *location, Elf32_Addr val, - Elf32_Shdr *sechdrs, struct module *mod) -{ - struct v850_plt_entry *entry; - /* Instructions used to do the indirect jump. */ - uint32_t tramp[2]; - - /* We have to trash a register, so we assume that any control - transfer more than 21-bits away must be a function call - (so we can use a call-clobbered register). */ - tramp[0] = 0x0621 + ((val & 0xffff) << 16); /* mov sym, r1 ... */ - tramp[1] = ((val >> 16) & 0xffff) + 0x610000; /* ...; jmp r1 */ - - /* Init, or core PLT? */ - if (location >= mod->module_core - && location < mod->module_core + mod->core_size) - entry = (void *)sechdrs[mod->arch.core_plt_section].sh_addr; - else - entry = (void *)sechdrs[mod->arch.init_plt_section].sh_addr; - - /* Find this entry, or if that fails, the next avail. entry */ - while (entry->tramp[0]) - if (entry->tramp[0] == tramp[0] && entry->tramp[1] == tramp[1]) - return (uint32_t)entry; - else - entry++; - - entry->tramp[0] = tramp[0]; - entry->tramp[1] = tramp[1]; - - return (uint32_t)entry; -} - -int apply_relocate_add (Elf32_Shdr *sechdrs, const char *strtab, - unsigned int symindex, unsigned int relsec, - struct module *mod) -{ - unsigned int i; - Elf32_Rela *rela = (void *)sechdrs[relsec].sh_addr; - - DEBUGP ("Applying relocate section %u to %u\n", relsec, - sechdrs[relsec].sh_info); - - for (i = 0; i < sechdrs[relsec].sh_size / sizeof (*rela); i++) { - /* This is where to make the change */ - uint32_t *loc - = ((void *)sechdrs[sechdrs[relsec].sh_info].sh_addr - + rela[i].r_offset); - /* This is the symbol it is referring to. Note that all - undefined symbols have been resolved. */ - Elf32_Sym *sym - = ((Elf32_Sym *)sechdrs[symindex].sh_addr - + ELF32_R_SYM (rela[i].r_info)); - uint32_t val = sym->st_value + rela[i].r_addend; - - switch (ELF32_R_TYPE (rela[i].r_info)) { - case R_V850_32: - /* We write two shorts instead of a long because even - 32-bit insns only need half-word alignment, but - 32-bit data writes need to be long-word aligned. */ - val += ((uint16_t *)loc)[0]; - val += ((uint16_t *)loc)[1] << 16; - ((uint16_t *)loc)[0] = val & 0xffff; - ((uint16_t *)loc)[1] = (val >> 16) & 0xffff; - break; - - case R_V850_22_PCREL: - /* Maybe jump indirectly via a PLT table entry. */ - if ((int32_t)(val - (uint32_t)loc) > 0x1fffff - || (int32_t)(val - (uint32_t)loc) < -0x200000) - val = do_plt_call (loc, val, sechdrs, mod); - - val -= (uint32_t)loc; - - /* We write two shorts instead of a long because - even 32-bit insns only need half-word alignment, - but 32-bit data writes need to be long-word - aligned. */ - ((uint16_t *)loc)[0] = - (*(uint16_t *)loc & 0xffc0) /* opcode + reg */ - | ((val >> 16) & 0xffc03f); /* offs high */ - ((uint16_t *)loc)[1] = - (val & 0xffff); /* offs low */ - break; - - default: - printk (KERN_ERR "module %s: Unknown reloc: %u\n", - mod->name, ELF32_R_TYPE (rela[i].r_info)); - return -ENOEXEC; - } - } - - return 0; -} - -void -module_arch_cleanup(struct module *mod) -{ -} diff --git a/arch/v850/kernel/process.c b/arch/v850/kernel/process.c deleted file mode 100644 index e4a4b8e7d5a3..000000000000 --- a/arch/v850/kernel/process.c +++ /dev/null @@ -1,217 +0,0 @@ -/* - * arch/v850/kernel/process.c -- Arch-dependent process handling - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include - -void (*pm_power_off)(void) = NULL; -EXPORT_SYMBOL(pm_power_off); - -extern void ret_from_fork (void); - - -/* The idle loop. */ -static void default_idle (void) -{ - while (! need_resched ()) - asm ("halt; nop; nop; nop; nop; nop" ::: "cc"); -} - -void (*idle)(void) = default_idle; - -/* - * The idle thread. There's no useful work to be - * done, so just try to conserve power and have a - * low exit latency (ie sit in a loop waiting for - * somebody to say that they'd like to reschedule) - */ -void cpu_idle (void) -{ - /* endless idle loop with no priority at all */ - while (1) { - while (!need_resched()) - (*idle) (); - - preempt_enable_no_resched(); - schedule(); - preempt_disable(); - } -} - -/* - * This is the mechanism for creating a new kernel thread. - * - * NOTE! Only a kernel-only process (ie the swapper or direct descendants who - * haven't done an "execve()") should use this: it will work within a system - * call from a "real" process, but the process memory space will not be free'd - * until both the parent and the child have exited. - */ -int kernel_thread (int (*fn)(void *), void *arg, unsigned long flags) -{ - register mm_segment_t fs = get_fs (); - register unsigned long syscall asm (SYSCALL_NUM); - register unsigned long arg0 asm (SYSCALL_ARG0); - register unsigned long ret asm (SYSCALL_RET); - - set_fs (KERNEL_DS); - - /* Clone this thread. Note that we don't pass the clone syscall's - second argument -- it's ignored for calls from kernel mode (the - child's SP is always set to the top of the kernel stack). */ - arg0 = flags | CLONE_VM; - syscall = __NR_clone; - asm volatile ("trap " SYSCALL_SHORT_TRAP - : "=r" (ret), "=r" (syscall) - : "1" (syscall), "r" (arg0) - : SYSCALL_SHORT_CLOBBERS); - - if (ret == 0) { - /* In child thread, call FN and exit. */ - arg0 = (*fn) (arg); - syscall = __NR_exit; - asm volatile ("trap " SYSCALL_SHORT_TRAP - : "=r" (ret), "=r" (syscall) - : "1" (syscall), "r" (arg0) - : SYSCALL_SHORT_CLOBBERS); - } - - /* In parent. */ - set_fs (fs); - - return ret; -} - -void flush_thread (void) -{ - set_fs (USER_DS); -} - -int copy_thread (int nr, unsigned long clone_flags, - unsigned long stack_start, unsigned long stack_size, - struct task_struct *p, struct pt_regs *regs) -{ - /* Start pushing stuff from the top of the child's kernel stack. */ - unsigned long orig_ksp = task_tos(p); - unsigned long ksp = orig_ksp; - /* We push two `state save' stack fames (see entry.S) on the new - kernel stack: - 1) The innermost one is what switch_thread would have - pushed, and is used when we context switch to the child - thread for the first time. It's set up to return to - ret_from_fork in entry.S. - 2) The outermost one (nearest the top) is what a syscall - trap would have pushed, and is set up to return to the - same location as the parent thread, but with a return - value of 0. */ - struct pt_regs *child_switch_regs, *child_trap_regs; - - /* Trap frame. */ - ksp -= STATE_SAVE_SIZE; - child_trap_regs = (struct pt_regs *)(ksp + STATE_SAVE_PT_OFFSET); - /* Switch frame. */ - ksp -= STATE_SAVE_SIZE; - child_switch_regs = (struct pt_regs *)(ksp + STATE_SAVE_PT_OFFSET); - - /* First copy parent's register state to child. */ - *child_switch_regs = *regs; - *child_trap_regs = *regs; - - /* switch_thread returns to the restored value of the lp - register (r31), so we make that the place where we want to - jump when the child thread begins running. */ - child_switch_regs->gpr[GPR_LP] = (v850_reg_t)ret_from_fork; - - if (regs->kernel_mode) - /* Since we're returning to kernel-mode, make sure the child's - stored kernel stack pointer agrees with what the actual - stack pointer will be at that point (the trap return code - always restores the SP, even when returning to - kernel-mode). */ - child_trap_regs->gpr[GPR_SP] = orig_ksp; - else - /* Set the child's user-mode stack-pointer (the name - `stack_start' is a misnomer, it's just the initial SP - value). */ - child_trap_regs->gpr[GPR_SP] = stack_start; - - /* Thread state for the child (everything else is on the stack). */ - p->thread.ksp = ksp; - - return 0; -} - -/* - * sys_execve() executes a new program. - */ -int sys_execve (char *name, char **argv, char **envp, struct pt_regs *regs) -{ - char *filename = getname (name); - int error = PTR_ERR (filename); - - if (! IS_ERR (filename)) { - error = do_execve (filename, argv, envp, regs); - putname (filename); - } - - return error; -} - - -/* - * These bracket the sleeping functions.. - */ -#define first_sched ((unsigned long)__sched_text_start) -#define last_sched ((unsigned long)__sched_text_end) - -unsigned long get_wchan (struct task_struct *p) -{ -#if 0 /* Barf. Figure out the stack-layout later. XXX */ - unsigned long fp, pc; - int count = 0; - - if (!p || p == current || p->state == TASK_RUNNING) - return 0; - - pc = thread_saved_pc (p); - - /* This quite disgusting function walks up the stack, following - saved return address, until it something that's out of bounds - (as defined by `first_sched' and `last_sched'). It then - returns the last PC that was in-bounds. */ - do { - if (fp < stack_page + sizeof (struct task_struct) || - fp >= 8184+stack_page) - return 0; - pc = ((unsigned long *)fp)[1]; - if (pc < first_sched || pc >= last_sched) - return pc; - fp = *(unsigned long *) fp; - } while (count++ < 16); -#endif - - return 0; -} diff --git a/arch/v850/kernel/procfs.c b/arch/v850/kernel/procfs.c deleted file mode 100644 index e433cde789b4..000000000000 --- a/arch/v850/kernel/procfs.c +++ /dev/null @@ -1,67 +0,0 @@ -/* - * arch/v850/kernel/procfs.c -- Introspection functions for /proc filesystem - * - * Copyright (C) 2001,02 NEC Corporation - * Copyright (C) 2001,02 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include "mach.h" - -static int cpuinfo_print (struct seq_file *m, void *v) -{ - extern unsigned long loops_per_jiffy; - - seq_printf (m, "CPU-Family: v850\nCPU-Arch: %s\n", CPU_ARCH); - -#ifdef CPU_MODEL_LONG - seq_printf (m, "CPU-Model: %s (%s)\n", CPU_MODEL, CPU_MODEL_LONG); -#else - seq_printf (m, "CPU-Model: %s\n", CPU_MODEL); -#endif - -#ifdef CPU_CLOCK_FREQ - seq_printf (m, "CPU-Clock: %ld (%ld MHz)\n", - (long)CPU_CLOCK_FREQ, - (long)CPU_CLOCK_FREQ / 1000000); -#endif - - seq_printf (m, "BogoMips: %lu.%02lu\n", - loops_per_jiffy/(500000/HZ), - (loops_per_jiffy/(5000/HZ)) % 100); - -#ifdef PLATFORM_LONG - seq_printf (m, "Platform: %s (%s)\n", PLATFORM, PLATFORM_LONG); -#elif defined (PLATFORM) - seq_printf (m, "Platform: %s\n", PLATFORM); -#endif - - return 0; -} - -static void *cpuinfo_start (struct seq_file *m, loff_t *pos) -{ - return *pos < NR_CPUS ? ((void *) 0x12345678) : NULL; -} - -static void *cpuinfo_next (struct seq_file *m, void *v, loff_t *pos) -{ - ++*pos; - return cpuinfo_start (m, pos); -} - -static void cpuinfo_stop (struct seq_file *m, void *v) -{ -} - -const struct seq_operations cpuinfo_op = { - .start = cpuinfo_start, - .next = cpuinfo_next, - .stop = cpuinfo_stop, - .show = cpuinfo_print -}; diff --git a/arch/v850/kernel/ptrace.c b/arch/v850/kernel/ptrace.c deleted file mode 100644 index a458ac941b25..000000000000 --- a/arch/v850/kernel/ptrace.c +++ /dev/null @@ -1,235 +0,0 @@ -/* - * arch/v850/kernel/ptrace.c -- `ptrace' system call - * - * Copyright (C) 2002,03,04 NEC Electronics Corporation - * Copyright (C) 2002,03,04 Miles Bader - * - * Derived from arch/mips/kernel/ptrace.c: - * - * Copyright (C) 1992 Ross Biro - * Copyright (C) Linus Torvalds - * Copyright (C) 1994, 95, 96, 97, 98, 2000 Ralf Baechle - * Copyright (C) 1996 David S. Miller - * Kevin D. Kissell, kevink@mips.com and Carsten Langgaard, carstenl@mips.com - * Copyright (C) 1999 MIPS Technologies, Inc. - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - */ - -#include -#include -#include -#include -#include - -#include -#include -#include -#include - -/* Returns the address where the register at REG_OFFS in P is stashed away. */ -static v850_reg_t *reg_save_addr (unsigned reg_offs, struct task_struct *t) -{ - struct pt_regs *regs; - - /* Three basic cases: - - (1) A register normally saved before calling the scheduler, is - available in the kernel entry pt_regs structure at the top - of the kernel stack. The kernel trap/irq exit path takes - care to save/restore almost all registers for ptrace'd - processes. - - (2) A call-clobbered register, where the process P entered the - kernel via [syscall] trap, is not stored anywhere; that's - OK, because such registers are not expected to be preserved - when the trap returns anyway (so we don't actually bother to - test for this case). - - (3) A few registers not used at all by the kernel, and so - normally never saved except by context-switches, are in the - context switch state. */ - - if (reg_offs == PT_CTPC || reg_offs == PT_CTPSW || reg_offs == PT_CTBP) - /* Register saved during context switch. */ - regs = thread_saved_regs (t); - else - /* Register saved during kernel entry (or not available). */ - regs = task_pt_regs (t); - - return (v850_reg_t *)((char *)regs + reg_offs); -} - -/* Set the bits SET and clear the bits CLEAR in the v850e DIR - (`debug information register'). Returns the new value of DIR. */ -static inline v850_reg_t set_dir (v850_reg_t set, v850_reg_t clear) -{ - register v850_reg_t rval asm ("r10"); - register v850_reg_t arg0 asm ("r6") = set; - register v850_reg_t arg1 asm ("r7") = clear; - - /* The dbtrap handler has exactly this functionality when called - from kernel mode. 0xf840 is a `dbtrap' insn. */ - asm (".short 0xf840" : "=r" (rval) : "r" (arg0), "r" (arg1)); - - return rval; -} - -/* Makes sure hardware single-stepping is (globally) enabled. - Returns true if successful. */ -static inline int enable_single_stepping (void) -{ - static int enabled = 0; /* Remember whether we already did it. */ - if (! enabled) { - /* Turn on the SE (`single-step enable') bit, 0x100, in the - DIR (`debug information register'). This may fail if a - processor doesn't support it or something. We also try - to clear bit 0x40 (`INI'), which is necessary to use the - debug stuff on the v850e2; on the v850e, clearing 0x40 - shouldn't cause any problem. */ - v850_reg_t dir = set_dir (0x100, 0x40); - /* Make sure it really got set. */ - if (dir & 0x100) - enabled = 1; - } - return enabled; -} - -/* Try to set CHILD's single-step flag to VAL. Returns true if successful. */ -static int set_single_step (struct task_struct *t, int val) -{ - v850_reg_t *psw_addr = reg_save_addr(PT_PSW, t); - if (val) { - /* Make sure single-stepping is enabled. */ - if (! enable_single_stepping ()) - return 0; - /* Set T's single-step flag. */ - *psw_addr |= 0x800; - } else - *psw_addr &= ~0x800; - return 1; -} - -long arch_ptrace(struct task_struct *child, long request, long addr, long data) -{ - int rval; - - switch (request) { - unsigned long val; - - case PTRACE_PEEKTEXT: /* read word at location addr. */ - case PTRACE_PEEKDATA: - rval = generic_ptrace_peekdata(child, addr, data); - goto out; - - case PTRACE_POKETEXT: /* write the word at location addr. */ - case PTRACE_POKEDATA: - rval = generic_ptrace_pokedata(child, addr, data); - goto out; - - /* Read/write the word at location ADDR in the registers. */ - case PTRACE_PEEKUSR: - case PTRACE_POKEUSR: - rval = 0; - if (addr >= PT_SIZE && request == PTRACE_PEEKUSR) { - /* Special requests that don't actually correspond - to offsets in struct pt_regs. */ - if (addr == PT_TEXT_ADDR) - val = child->mm->start_code; - else if (addr == PT_DATA_ADDR) - val = child->mm->start_data; - else if (addr == PT_TEXT_LEN) - val = child->mm->end_code - - child->mm->start_code; - else - rval = -EIO; - } else if (addr >= 0 && addr < PT_SIZE && (addr & 0x3) == 0) { - v850_reg_t *reg_addr = reg_save_addr(addr, child); - if (request == PTRACE_PEEKUSR) - val = *reg_addr; - else - *reg_addr = data; - } else - rval = -EIO; - - if (rval == 0 && request == PTRACE_PEEKUSR) - rval = put_user (val, (unsigned long *)data); - goto out; - - /* Continue and stop at next (return from) syscall */ - case PTRACE_SYSCALL: - /* Restart after a signal. */ - case PTRACE_CONT: - /* Execute a single instruction. */ - case PTRACE_SINGLESTEP: - rval = -EIO; - if (!valid_signal(data)) - break; - - /* Turn CHILD's single-step flag on or off. */ - if (! set_single_step (child, request == PTRACE_SINGLESTEP)) - break; - - if (request == PTRACE_SYSCALL) - set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); - else - clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); - - child->exit_code = data; - wake_up_process(child); - rval = 0; - break; - - /* - * make the child exit. Best I can do is send it a sigkill. - * perhaps it should be put in the status that it wants to - * exit. - */ - case PTRACE_KILL: - rval = 0; - if (child->exit_state == EXIT_ZOMBIE) /* already dead */ - break; - child->exit_code = SIGKILL; - wake_up_process(child); - break; - - case PTRACE_DETACH: /* detach a process that was attached. */ - set_single_step (child, 0); /* Clear single-step flag */ - rval = ptrace_detach(child, data); - break; - - default: - rval = -EIO; - goto out; - } - out: - return rval; -} - -asmlinkage void syscall_trace(void) -{ - if (!test_thread_flag(TIF_SYSCALL_TRACE)) - return; - if (!(current->ptrace & PT_PTRACED)) - return; - /* The 0x80 provides a way for the tracing parent to distinguish - between a syscall stop and SIGTRAP delivery */ - ptrace_notify(SIGTRAP | ((current->ptrace & PT_TRACESYSGOOD) - ? 0x80 : 0)); - /* - * this isn't the same as continuing with a signal, but it will do - * for normal use. strace only continues with a signal if the - * stopping signal is not SIGTRAP. -brl - */ - if (current->exit_code) { - send_sig(current->exit_code, current, 1); - current->exit_code = 0; - } -} - -void ptrace_disable (struct task_struct *child) -{ - /* nothing to do */ -} diff --git a/arch/v850/kernel/rte_cb.c b/arch/v850/kernel/rte_cb.c deleted file mode 100644 index 43018e1edebd..000000000000 --- a/arch/v850/kernel/rte_cb.c +++ /dev/null @@ -1,193 +0,0 @@ -/* - * include/asm-v850/rte_cb.c -- Midas lab RTE-CB series of evaluation boards - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include -#include -#include -#include - -#include -#include - -#include "mach.h" - -static void led_tick (void); - -/* LED access routines. */ -extern unsigned read_leds (int pos, char *buf, int len); -extern unsigned write_leds (int pos, const char *buf, int len); - -#ifdef CONFIG_RTE_CB_MULTI -extern void multi_init (void); -#endif - - -void __init rte_cb_early_init (void) -{ - v850e_intc_disable_irqs (); - -#ifdef CONFIG_RTE_CB_MULTI - multi_init (); -#endif -} - -void __init mach_setup (char **cmdline) -{ -#ifdef CONFIG_RTE_MB_A_PCI - /* Probe for Mother-A, and print a message if we find it. */ - *(volatile unsigned long *)MB_A_SRAM_ADDR = 0xDEADBEEF; - if (*(volatile unsigned long *)MB_A_SRAM_ADDR == 0xDEADBEEF) { - *(volatile unsigned long *)MB_A_SRAM_ADDR = 0x12345678; - if (*(volatile unsigned long *)MB_A_SRAM_ADDR == 0x12345678) - printk (KERN_INFO - " NEC SolutionGear/Midas lab" - " RTE-MOTHER-A motherboard\n"); - } -#endif /* CONFIG_RTE_MB_A_PCI */ - - mach_tick = led_tick; -} - -void machine_restart (char *__unused) -{ -#ifdef CONFIG_RESET_GUARD - disable_reset_guard (); -#endif - asm ("jmp r0"); /* Jump to the reset vector. */ -} - -/* This says `HALt.' in LEDese. */ -static unsigned char halt_leds_msg[] = { 0x76, 0x77, 0x38, 0xF8 }; - -void machine_halt (void) -{ -#ifdef CONFIG_RESET_GUARD - disable_reset_guard (); -#endif - - /* Ignore all interrupts. */ - local_irq_disable (); - - /* Write a little message. */ - write_leds (0, halt_leds_msg, sizeof halt_leds_msg); - - /* Really halt. */ - for (;;) - asm ("halt; nop; nop; nop; nop; nop"); -} - -void machine_power_off (void) -{ - machine_halt (); -} - - -/* Animated LED display for timer tick. */ - -#define TICK_UPD_FREQ 6 -static int tick_frames[][10] = { - { 0x01, 0x02, 0x04, 0x08, 0x10, 0x20, -1 }, - { 0x63, 0x5c, -1 }, - { 0x5c, 0x00, -1 }, - { 0x63, 0x00, -1 }, - { -1 } -}; - -static void led_tick () -{ - static unsigned counter = 0; - - if (++counter == (HZ / TICK_UPD_FREQ)) { - /* Which frame we're currently displaying for each digit. */ - static unsigned frame_nums[LED_NUM_DIGITS] = { 0 }; - /* Display image. */ - static unsigned char image[LED_NUM_DIGITS] = { 0 }; - unsigned char prev_image[LED_NUM_DIGITS]; - int write_to_leds = 1; /* true if we should actually display */ - int digit; - - /* We check to see if the physical LEDs contains what we last - wrote to them; if not, we suppress display (this is so that - users can write to the LEDs, and not have their output - overwritten). As a special case, we start writing again if - all the LEDs are blank, or our display image is all zeros - (indicating that this is the initial update, when the actual - LEDs might contain random data). */ - read_leds (0, prev_image, LED_NUM_DIGITS); - for (digit = 0; digit < LED_NUM_DIGITS; digit++) - if (image[digit] != prev_image[digit] - && image[digit] && prev_image[digit]) - { - write_to_leds = 0; - break; - } - - /* Update display image. */ - for (digit = 0; - digit < LED_NUM_DIGITS && tick_frames[digit][0] >= 0; - digit++) - { - int frame = tick_frames[digit][frame_nums[digit]]; - if (frame < 0) { - image[digit] = tick_frames[digit][0]; - frame_nums[digit] = 1; - } else { - image[digit] = frame; - frame_nums[digit]++; - break; - } - } - - if (write_to_leds) - /* Write the display image to the physical LEDs. */ - write_leds (0, image, LED_NUM_DIGITS); - - counter = 0; - } -} - - -/* Mother-A interrupts. */ - -#ifdef CONFIG_RTE_GBUS_INT - -#define L GBUS_INT_PRIORITY_LOW -#define M GBUS_INT_PRIORITY_MEDIUM -#define H GBUS_INT_PRIORITY_HIGH - -static struct gbus_int_irq_init gbus_irq_inits[] = { -#ifdef CONFIG_RTE_MB_A_PCI - { "MB_A_LAN", IRQ_MB_A_LAN, 1, 1, L }, - { "MB_A_PCI1", IRQ_MB_A_PCI1(0), IRQ_MB_A_PCI1_NUM, 1, L }, - { "MB_A_PCI2", IRQ_MB_A_PCI2(0), IRQ_MB_A_PCI2_NUM, 1, L }, - { "MB_A_EXT", IRQ_MB_A_EXT(0), IRQ_MB_A_EXT_NUM, 1, L }, - { "MB_A_USB_OC",IRQ_MB_A_USB_OC(0), IRQ_MB_A_USB_OC_NUM, 1, L }, - { "MB_A_PCMCIA_OC",IRQ_MB_A_PCMCIA_OC, 1, 1, L }, -#endif - { 0 } -}; -#define NUM_GBUS_IRQ_INITS (ARRAY_SIZE(gbus_irq_inits) - 1) - -static struct hw_interrupt_type gbus_hw_itypes[NUM_GBUS_IRQ_INITS]; - -#endif /* CONFIG_RTE_GBUS_INT */ - - -void __init rte_cb_init_irqs (void) -{ -#ifdef CONFIG_RTE_GBUS_INT - gbus_int_init_irqs (); - gbus_int_init_irq_types (gbus_irq_inits, gbus_hw_itypes); -#endif /* CONFIG_RTE_GBUS_INT */ -} diff --git a/arch/v850/kernel/rte_cb_leds.c b/arch/v850/kernel/rte_cb_leds.c deleted file mode 100644 index aa47ab1dcd87..000000000000 --- a/arch/v850/kernel/rte_cb_leds.c +++ /dev/null @@ -1,137 +0,0 @@ -/* - * include/asm-v850/rte_cb_leds.c -- Midas lab RTE-CB board LED device support - * - * Copyright (C) 2002,03 NEC Electronics Corporation - * Copyright (C) 2002,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include -#include -#include - -#include - -#define LEDS_MINOR 169 /* Minor device number, using misc major. */ - -/* The actual LED hardware is write-only, so we hold the contents here too. */ -static unsigned char leds_image[LED_NUM_DIGITS] = { 0 }; - -/* Spinlock protecting the above leds. */ -static DEFINE_SPINLOCK(leds_lock); - -/* Common body of LED read/write functions, checks POS and LEN for - correctness, declares a variable using IMG_DECL, initialized pointing at - the POS position in the LED image buffer, and and iterates COPY_EXPR - until BUF is equal to the last buffer position; finally, sets LEN to be - the amount actually copied. IMG should be a variable declaration - (without an initializer or a terminating semicolon); POS, BUF, and LEN - should all be simple variables. */ -#define DO_LED_COPY(img_decl, pos, buf, len, copy_expr) \ -do { \ - if (pos > LED_NUM_DIGITS) \ - len = 0; \ - else { \ - if (pos + len > LED_NUM_DIGITS) \ - len = LED_NUM_DIGITS - pos; \ - \ - if (len > 0) { \ - unsigned long _flags; \ - const char *_end = buf + len; \ - img_decl = &leds_image[pos]; \ - \ - spin_lock_irqsave (leds_lock, _flags); \ - do \ - (copy_expr); \ - while (buf != _end); \ - spin_unlock_irqrestore (leds_lock, _flags); \ - } \ - } \ -} while (0) - -/* Read LEN bytes from LEDs at position POS, into BUF. - Returns actual amount read. */ -unsigned read_leds (unsigned pos, char *buf, unsigned len) -{ - DO_LED_COPY (const char *img, pos, buf, len, *buf++ = *img++); - return len; -} - -/* Write LEN bytes to LEDs at position POS, from BUF. - Returns actual amount written. */ -unsigned write_leds (unsigned pos, const char *buf, unsigned len) -{ - /* We write the actual LED values backwards, because - increasing memory addresses reflect LEDs right-to-left. */ - volatile char *led = &LED (LED_NUM_DIGITS - pos - 1); - /* We invert the value written to the hardware, because 1 = off, - and 0 = on. */ - DO_LED_COPY (char *img, pos, buf, len, - *led-- = 0xFF ^ (*img++ = *buf++)); - return len; -} - - -/* Device functions. */ - -static ssize_t leds_dev_read (struct file *file, char *buf, size_t len, - loff_t *pos) -{ - char temp_buf[LED_NUM_DIGITS]; - len = read_leds (*pos, temp_buf, len); - if (copy_to_user (buf, temp_buf, len)) - return -EFAULT; - *pos += len; - return len; -} - -static ssize_t leds_dev_write (struct file *file, const char *buf, size_t len, - loff_t *pos) -{ - char temp_buf[LED_NUM_DIGITS]; - if (copy_from_user (temp_buf, buf, min_t(size_t, len, LED_NUM_DIGITS))) - return -EFAULT; - len = write_leds (*pos, temp_buf, len); - *pos += len; - return len; -} - -static loff_t leds_dev_lseek (struct file *file, loff_t offs, int whence) -{ - if (whence == 1) - offs += file->f_pos; /* relative */ - else if (whence == 2) - offs += LED_NUM_DIGITS; /* end-relative */ - - if (offs < 0 || offs > LED_NUM_DIGITS) - return -EINVAL; - - file->f_pos = offs; - - return 0; -} - -static const struct file_operations leds_fops = { - .read = leds_dev_read, - .write = leds_dev_write, - .llseek = leds_dev_lseek -}; - -static struct miscdevice leds_miscdev = { - .name = "leds", - .minor = LEDS_MINOR, - .fops = &leds_fops -}; - -int __init leds_dev_init (void) -{ - return misc_register (&leds_miscdev); -} - -__initcall (leds_dev_init); diff --git a/arch/v850/kernel/rte_cb_multi.c b/arch/v850/kernel/rte_cb_multi.c deleted file mode 100644 index 963d55ab34cc..000000000000 --- a/arch/v850/kernel/rte_cb_multi.c +++ /dev/null @@ -1,121 +0,0 @@ -/* - * include/asm-v850/rte_multi.c -- Support for Multi debugger monitor ROM - * on Midas lab RTE-CB series of evaluation boards - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include - -#include - -#define IRQ_ADDR(irq) (0x80 + (irq) * 0x10) - -/* A table of which interrupt vectors to install, since blindly - installing all of them makes the debugger stop working. This is a - list of offsets in the interrupt vector area; each entry means to - copy that particular 16-byte vector. An entry less than zero ends - the table. */ -static long multi_intv_install_table[] = { - /* Trap vectors */ - 0x40, 0x50, - -#ifdef CONFIG_RTE_CB_MULTI_DBTRAP - /* Illegal insn / dbtrap. These are used by multi, so only handle - them if configured to do so. */ - 0x60, -#endif - - /* GINT1 - GINT3 (note, not GINT0!) */ - IRQ_ADDR (IRQ_GINT(1)), - IRQ_ADDR (IRQ_GINT(2)), - IRQ_ADDR (IRQ_GINT(3)), - - /* Timer D interrupts (up to 4 timers) */ - IRQ_ADDR (IRQ_INTCMD(0)), -#if IRQ_INTCMD_NUM > 1 - IRQ_ADDR (IRQ_INTCMD(1)), -#if IRQ_INTCMD_NUM > 2 - IRQ_ADDR (IRQ_INTCMD(2)), -#if IRQ_INTCMD_NUM > 3 - IRQ_ADDR (IRQ_INTCMD(3)), -#endif -#endif -#endif - - /* UART interrupts (up to 3 channels) */ - IRQ_ADDR (IRQ_INTSER (0)), /* err */ - IRQ_ADDR (IRQ_INTSR (0)), /* rx */ - IRQ_ADDR (IRQ_INTST (0)), /* tx */ -#if IRQ_INTSR_NUM > 1 - IRQ_ADDR (IRQ_INTSER (1)), /* err */ - IRQ_ADDR (IRQ_INTSR (1)), /* rx */ - IRQ_ADDR (IRQ_INTST (1)), /* tx */ -#if IRQ_INTSR_NUM > 2 - IRQ_ADDR (IRQ_INTSER (2)), /* err */ - IRQ_ADDR (IRQ_INTSR (2)), /* rx */ - IRQ_ADDR (IRQ_INTST (2)), /* tx */ -#endif -#endif - - -1 -}; - -/* Early initialization for kernel using Multi debugger ROM monitor. */ -void __init multi_init (void) -{ - /* We're using the Multi debugger monitor, so we have to install - the interrupt vectors. The monitor doesn't allow them to be - initially downloaded into their final destination because - it's in the monitor's scratch-RAM area. Unfortunately, Multi - also doesn't deal correctly with ELF sections where the LMA - and VMA differ -- it just ignores the LMA -- so we can't use - that feature to work around the problem. What we do instead - is just put the interrupt vectors into a normal section, and - do the necessary copying and relocation here. Since the - interrupt vector basically only contains `jr' instructions - and no-ops, it's not that hard. */ - extern unsigned long _intv_load_start, _intv_start; - register unsigned long *src = &_intv_load_start; - register unsigned long *dst = (unsigned long *)INTV_BASE; - register unsigned long jr_fixup = (char *)&_intv_start - (char *)dst; - register long *ii; - - /* Copy interrupt vectors as instructed by multi_intv_install_table. */ - for (ii = multi_intv_install_table; *ii >= 0; ii++) { - /* Copy 16-byte interrupt vector at offset *ii. */ - int boffs; - for (boffs = 0; boffs < 0x10; boffs += sizeof *src) { - /* Copy a single word, fixing up the jump offs - if it's a `jr' instruction. */ - int woffs = (*ii + boffs) / sizeof *src; - unsigned long word = src[woffs]; - - if ((word & 0xFC0) == 0x780) { - /* A `jr' insn, fix up its offset (and yes, the - weird half-word swapping is intentional). */ - unsigned short hi = word & 0xFFFF; - unsigned short lo = word >> 16; - unsigned long udisp22 - = lo + ((hi & 0x3F) << 16); - long disp22 = (long)(udisp22 << 10) >> 10; - - disp22 += jr_fixup; - - hi = ((disp22 >> 16) & 0x3F) | 0x780; - lo = disp22 & 0xFFFF; - - word = hi + (lo << 16); - } - - dst[woffs] = word; - } - } -} diff --git a/arch/v850/kernel/rte_ma1_cb-rom.ld b/arch/v850/kernel/rte_ma1_cb-rom.ld deleted file mode 100644 index 87b618f8253b..000000000000 --- a/arch/v850/kernel/rte_ma1_cb-rom.ld +++ /dev/null @@ -1,14 +0,0 @@ -/* Linker script for the Midas labs RTE-V850E/MA1-CB evaluation board - (CONFIG_RTE_CB_MA1), with kernel in ROM. */ - -MEMORY { - ROM : ORIGIN = 0x00000000, LENGTH = 0x00100000 - /* 1MB of SRAM. This memory is mirrored 4 times. */ - SRAM : ORIGIN = SRAM_ADDR, LENGTH = SRAM_SIZE - /* 32MB of SDRAM. */ - SDRAM : ORIGIN = SDRAM_ADDR, LENGTH = SDRAM_SIZE -} - -SECTIONS { - ROMK_SECTIONS(ROM, SRAM) -} diff --git a/arch/v850/kernel/rte_ma1_cb.c b/arch/v850/kernel/rte_ma1_cb.c deleted file mode 100644 index 08abf3d5f8df..000000000000 --- a/arch/v850/kernel/rte_ma1_cb.c +++ /dev/null @@ -1,107 +0,0 @@ -/* - * arch/v850/kernel/rte_ma1_cb.c -- Midas labs RTE-V850E/MA1-CB board - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include -#include - -#include -#include -#include -#include -#include - -#include "mach.h" - - -/* SRAM and SDRAM are almost contiguous (with a small hole in between; - see mach_reserve_bootmem for details), so just use both as one big area. */ -#define RAM_START SRAM_ADDR -#define RAM_END (SDRAM_ADDR + SDRAM_SIZE) - - -void __init mach_early_init (void) -{ - rte_cb_early_init (); -} - -void __init mach_get_physical_ram (unsigned long *ram_start, - unsigned long *ram_len) -{ - *ram_start = RAM_START; - *ram_len = RAM_END - RAM_START; -} - -void __init mach_reserve_bootmem () -{ -#ifdef CONFIG_RTE_CB_MULTI - /* Prevent the kernel from touching the monitor's scratch RAM. */ - reserve_bootmem(MON_SCRATCH_ADDR, MON_SCRATCH_SIZE, - BOOTMEM_DEFAULT); -#endif - - /* The space between SRAM and SDRAM is filled with duplicate - images of SRAM. Prevent the kernel from using them. */ - reserve_bootmem (SRAM_ADDR + SRAM_SIZE, - SDRAM_ADDR - (SRAM_ADDR + SRAM_SIZE), - BOOTMEM_DEFAULT); -} - -void mach_gettimeofday (struct timespec *tv) -{ - tv->tv_sec = 0; - tv->tv_nsec = 0; -} - -/* Called before configuring an on-chip UART. */ -void rte_ma1_cb_uart_pre_configure (unsigned chan, - unsigned cflags, unsigned baud) -{ - /* The RTE-MA1-CB connects some general-purpose I/O pins on the - CPU to the RTS/CTS lines of UART 0's serial connection. - I/O pins P42 and P43 are RTS and CTS respectively. */ - if (chan == 0) { - /* Put P42 & P43 in I/O port mode. */ - MA_PORT4_PMC &= ~0xC; - /* Make P42 an output, and P43 an input. */ - MA_PORT4_PM = (MA_PORT4_PM & ~0xC) | 0x8; - } - - /* Do pre-configuration for the actual UART. */ - ma_uart_pre_configure (chan, cflags, baud); -} - -void __init mach_init_irqs (void) -{ - unsigned tc; - - /* Initialize interrupts. */ - ma_init_irqs (); - rte_cb_init_irqs (); - - /* Use falling-edge-sensitivity for interrupts . */ - V850E_TIMER_C_SESC (0) &= ~0xC; - V850E_TIMER_C_SESC (1) &= ~0xF; - - /* INTP000-INTP011 are shared with `Timer C', so we have to set - up Timer C to pass them through as raw interrupts. */ - for (tc = 0; tc < 2; tc++) - /* Turn on the timer. */ - V850E_TIMER_C_TMCC0 (tc) |= V850E_TIMER_C_TMCC0_CAE; - - /* Make sure the relevant port0/port1 pins are assigned - interrupt duty. We used INTP001-INTP011 (don't screw with - INTP000 because the monitor uses it). */ - MA_PORT0_PMC |= 0x4; /* P02 (INTP001) in IRQ mode. */ - MA_PORT1_PMC |= 0x6; /* P11 (INTP010) & P12 (INTP011) in IRQ mode.*/ -} diff --git a/arch/v850/kernel/rte_ma1_cb.ld b/arch/v850/kernel/rte_ma1_cb.ld deleted file mode 100644 index c8e16d16be41..000000000000 --- a/arch/v850/kernel/rte_ma1_cb.ld +++ /dev/null @@ -1,57 +0,0 @@ -/* Linker script for the Midas labs RTE-V850E/MA1-CB evaluation board - (CONFIG_RTE_CB_MA1), with kernel in SDRAM, under Multi debugger. */ - -MEMORY { - /* 1MB of SRAM; we can't use the last 32KB, because it's used by - the monitor scratch-RAM. This memory is mirrored 4 times. */ - SRAM : ORIGIN = SRAM_ADDR, LENGTH = (SRAM_SIZE - MON_SCRATCH_SIZE) - /* Monitor scratch RAM; only the interrupt vectors should go here. */ - MRAM : ORIGIN = MON_SCRATCH_ADDR, LENGTH = MON_SCRATCH_SIZE - /* 32MB of SDRAM. */ - SDRAM : ORIGIN = SDRAM_ADDR, LENGTH = SDRAM_SIZE -} - -#ifdef CONFIG_RTE_CB_MA1_KSRAM -# define KRAM SRAM -#else -# define KRAM SDRAM -#endif - -SECTIONS { - /* We can't use RAMK_KRAM_CONTENTS because that puts the whole - kernel in a single ELF segment, and the Multi debugger (which - we use to load the kernel) appears to have bizarre problems - dealing with it. */ - - .text : { - __kram_start = . ; - TEXT_CONTENTS - } > KRAM - - .data : { - DATA_CONTENTS - BSS_CONTENTS - RAMK_INIT_CONTENTS - __kram_end = . ; - BOOTMAP_CONTENTS - - /* The address at which the interrupt vectors are initially - loaded by the loader. We can't load the interrupt vectors - directly into their target location, because the monitor - ROM for the GHS Multi debugger barfs if we try. - Unfortunately, Multi also doesn't deal correctly with ELF - sections where the LMA and VMA differ (it just ignores the - LMA), so we can't use that feature to work around the - problem! What we do instead is just put the interrupt - vectors into a normal section, and have the - `mach_early_init' function for Midas boards do the - necessary copying and relocation at runtime (this section - basically only contains `jr' instructions, so it's not - that hard). */ - . = ALIGN (0x10) ; - __intv_load_start = . ; - INTV_CONTENTS - } > KRAM - - .root ALIGN (4096) : { ROOT_FS_CONTENTS } > SDRAM -} diff --git a/arch/v850/kernel/rte_mb_a_pci.c b/arch/v850/kernel/rte_mb_a_pci.c deleted file mode 100644 index 687e367d8b64..000000000000 --- a/arch/v850/kernel/rte_mb_a_pci.c +++ /dev/null @@ -1,819 +0,0 @@ -/* - * arch/v850/kernel/mb_a_pci.c -- PCI support for Midas lab RTE-MOTHER-A board - * - * Copyright (C) 2001,02,03,05 NEC Electronics Corporation - * Copyright (C) 2001,02,03,05 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include -#include -#include -#include -#include - -#include - -/* __nomods_init is like __devinit, but is a no-op when modules are enabled. - This is used by some routines that can be called either during boot - or by a module. */ -#ifdef CONFIG_MODULES -#define __nomods_init /*nothing*/ -#else -#define __nomods_init __devinit -#endif - -/* PCI devices on the Mother-A board can only do DMA to/from the MB SRAM - (the RTE-V850E/MA1-CB cpu board doesn't support PCI access to - CPU-board memory), and since linux DMA buffers are allocated in - normal kernel memory, we basically have to copy DMA blocks around - (this is like a `bounce buffer'). When a DMA block is `mapped', we - allocate an identically sized block in MB SRAM, and if we're doing - output to the device, copy the CPU-memory block to the MB-SRAM block. - When an active block is `unmapped', we will copy the block back to - CPU memory if necessary, and then deallocate the MB SRAM block. - Ack. */ - -/* Where the motherboard SRAM is in the PCI-bus address space (the - first 512K of it is also mapped at PCI address 0). */ -#define PCI_MB_SRAM_ADDR 0x800000 - -/* Convert CPU-view MB SRAM address to/from PCI-view addresses of the - same memory. */ -#define MB_SRAM_TO_PCI(mb_sram_addr) \ - ((dma_addr_t)mb_sram_addr - MB_A_SRAM_ADDR + PCI_MB_SRAM_ADDR) -#define PCI_TO_MB_SRAM(pci_addr) \ - (void *)(pci_addr - PCI_MB_SRAM_ADDR + MB_A_SRAM_ADDR) - -static void pcibios_assign_resources (void); - -struct mb_pci_dev_irq { - unsigned dev; /* PCI device number */ - unsigned irq_base; /* First IRQ */ - unsigned query_pin; /* True if we should read the device's - Interrupt Pin info, and allocate - interrupt IRQ_BASE + PIN. */ -}; - -/* PCI interrupts are mapped statically to GBUS interrupts. */ -static struct mb_pci_dev_irq mb_pci_dev_irqs[] = { - /* Motherboard SB82558 ethernet controller */ - { 10, IRQ_MB_A_LAN, 0 }, - /* PCI slot 1 */ - { 8, IRQ_MB_A_PCI1(0), 1 }, - /* PCI slot 2 */ - { 9, IRQ_MB_A_PCI2(0), 1 } -}; -#define NUM_MB_PCI_DEV_IRQS ARRAY_SIZE(mb_pci_dev_irqs) - - -/* PCI configuration primitives. */ - -#define CONFIG_DMCFGA(bus, devfn, offs) \ - (0x80000000 \ - | ((offs) & ~0x3) \ - | ((devfn) << 8) \ - | ((bus)->number << 16)) - -static int -mb_pci_read (struct pci_bus *bus, unsigned devfn, int offs, int size, u32 *rval) -{ - u32 addr; - int flags; - - local_irq_save (flags); - - MB_A_PCI_PCICR = 0x7; - MB_A_PCI_DMCFGA = CONFIG_DMCFGA (bus, devfn, offs); - - addr = MB_A_PCI_IO_ADDR + (offs & 0x3); - - switch (size) { - case 1: *rval = *(volatile u8 *)addr; break; - case 2: *rval = *(volatile u16 *)addr; break; - case 4: *rval = *(volatile u32 *)addr; break; - } - - if (MB_A_PCI_PCISR & 0x2000) { - MB_A_PCI_PCISR = 0x2000; - *rval = ~0; - } - - MB_A_PCI_DMCFGA = 0; - - local_irq_restore (flags); - - return PCIBIOS_SUCCESSFUL; -} - -static int -mb_pci_write (struct pci_bus *bus, unsigned devfn, int offs, int size, u32 val) -{ - u32 addr; - int flags; - - local_irq_save (flags); - - MB_A_PCI_PCICR = 0x7; - MB_A_PCI_DMCFGA = CONFIG_DMCFGA (bus, devfn, offs); - - addr = MB_A_PCI_IO_ADDR + (offs & 0x3); - - switch (size) { - case 1: *(volatile u8 *)addr = val; break; - case 2: *(volatile u16 *)addr = val; break; - case 4: *(volatile u32 *)addr = val; break; - } - - if (MB_A_PCI_PCISR & 0x2000) - MB_A_PCI_PCISR = 0x2000; - - MB_A_PCI_DMCFGA = 0; - - local_irq_restore (flags); - - return PCIBIOS_SUCCESSFUL; -} - -static struct pci_ops mb_pci_config_ops = { - .read = mb_pci_read, - .write = mb_pci_write, -}; - - -/* PCI Initialization. */ - -static struct pci_bus *mb_pci_bus = 0; - -/* Do initial PCI setup. */ -static int __devinit pcibios_init (void) -{ - u32 id = MB_A_PCI_PCIHIDR; - u16 vendor = id & 0xFFFF; - u16 device = (id >> 16) & 0xFFFF; - - if (vendor == PCI_VENDOR_ID_PLX && device == PCI_DEVICE_ID_PLX_9080) { - printk (KERN_INFO - "PCI: PLX Technology PCI9080 HOST/PCI bridge\n"); - - MB_A_PCI_PCICR = 0x147; - - MB_A_PCI_PCIBAR0 = 0x007FFF00; - MB_A_PCI_PCIBAR1 = 0x0000FF00; - MB_A_PCI_PCIBAR2 = 0x00800000; - - MB_A_PCI_PCILTR = 0x20; - - MB_A_PCI_PCIPBAM |= 0x3; - - MB_A_PCI_PCISR = ~0; /* Clear errors. */ - - /* Reprogram the motherboard's IO/config address space, - as we don't support the GCS7 address space that the - default uses. */ - - /* Significant address bits used for decoding PCI GCS5 space - accesses. */ - MB_A_PCI_DMRR = ~(MB_A_PCI_MEM_SIZE - 1); - - /* I don't understand this, but the SolutionGear example code - uses such an offset, and it doesn't work without it. XXX */ -#if GCS5_SIZE == 0x00800000 -#define GCS5_CFG_OFFS 0x00800000 -#else -#define GCS5_CFG_OFFS 0 -#endif - - /* Address bit values for matching. Note that we have to give - the address from the motherboard's point of view, which is - different than the CPU's. */ - /* PCI memory space. */ - MB_A_PCI_DMLBAM = GCS5_CFG_OFFS + 0x0; - /* PCI I/O space. */ - MB_A_PCI_DMLBAI = - GCS5_CFG_OFFS + (MB_A_PCI_IO_ADDR - GCS5_ADDR); - - mb_pci_bus = pci_scan_bus (0, &mb_pci_config_ops, 0); - - pcibios_assign_resources (); - } else - printk (KERN_ERR "PCI: HOST/PCI bridge not found\n"); - - return 0; -} - -subsys_initcall (pcibios_init); - -char __devinit *pcibios_setup (char *option) -{ - /* Don't handle any options. */ - return option; -} - - -int __nomods_init pcibios_enable_device (struct pci_dev *dev, int mask) -{ - u16 cmd, old_cmd; - int idx; - struct resource *r; - - pci_read_config_word(dev, PCI_COMMAND, &cmd); - old_cmd = cmd; - for (idx = 0; idx < 6; idx++) { - r = &dev->resource[idx]; - if (!r->start && r->end) { - printk(KERN_ERR "PCI: Device %s not available because " - "of resource collisions\n", pci_name(dev)); - return -EINVAL; - } - if (r->flags & IORESOURCE_IO) - cmd |= PCI_COMMAND_IO; - if (r->flags & IORESOURCE_MEM) - cmd |= PCI_COMMAND_MEMORY; - } - if (cmd != old_cmd) { - printk("PCI: Enabling device %s (%04x -> %04x)\n", - pci_name(dev), old_cmd, cmd); - pci_write_config_word(dev, PCI_COMMAND, cmd); - } - return 0; -} - - -/* Resource allocation. */ -static void __devinit pcibios_assign_resources (void) -{ - struct pci_dev *dev = NULL; - struct resource *r; - - for_each_pci_dev(dev) { - unsigned di_num; - unsigned class = dev->class >> 8; - - if (class && class != PCI_CLASS_BRIDGE_HOST) { - unsigned r_num; - for(r_num = 0; r_num < 6; r_num++) { - r = &dev->resource[r_num]; - if (!r->start && r->end) - pci_assign_resource (dev, r_num); - } - } - - /* Assign interrupts. */ - for (di_num = 0; di_num < NUM_MB_PCI_DEV_IRQS; di_num++) { - struct mb_pci_dev_irq *di = &mb_pci_dev_irqs[di_num]; - - if (di->dev == PCI_SLOT (dev->devfn)) { - unsigned irq = di->irq_base; - - if (di->query_pin) { - /* Find out which interrupt pin - this device uses (each PCI - slot has 4). */ - u8 irq_pin; - - pci_read_config_byte (dev, - PCI_INTERRUPT_PIN, - &irq_pin); - - if (irq_pin == 0) - /* Doesn't use interrupts. */ - continue; - else - irq += irq_pin - 1; - } - - pcibios_update_irq (dev, irq); - } - } - } -} - -void __devinit pcibios_update_irq (struct pci_dev *dev, int irq) -{ - dev->irq = irq; - pci_write_config_byte (dev, PCI_INTERRUPT_LINE, irq); -} - -void __devinit -pcibios_resource_to_bus(struct pci_dev *dev, struct pci_bus_region *region, - struct resource *res) -{ - unsigned long offset = 0; - - if (res->flags & IORESOURCE_IO) { - offset = MB_A_PCI_IO_ADDR; - } else if (res->flags & IORESOURCE_MEM) { - offset = MB_A_PCI_MEM_ADDR; - } - - region->start = res->start - offset; - region->end = res->end - offset; -} - - -/* Stubs for things we don't use. */ - -/* Called after each bus is probed, but before its children are examined. */ -void pcibios_fixup_bus(struct pci_bus *b) -{ -} - -void -pcibios_align_resource (void *data, struct resource *res, - resource_size_t size, resource_size_t align) -{ -} - -void pcibios_set_master (struct pci_dev *dev) -{ -} - - -/* Mother-A SRAM memory allocation. This is a simple first-fit allocator. */ - -/* A memory free-list node. */ -struct mb_sram_free_area { - void *mem; - unsigned long size; - struct mb_sram_free_area *next; -}; - -/* The tail of the free-list, which starts out containing all the SRAM. */ -static struct mb_sram_free_area mb_sram_free_tail = { - (void *)MB_A_SRAM_ADDR, MB_A_SRAM_SIZE, 0 -}; - -/* The free-list. */ -static struct mb_sram_free_area *mb_sram_free_areas = &mb_sram_free_tail; - -/* The free-list of free free-list nodes. (:-) */ -static struct mb_sram_free_area *mb_sram_free_free_areas = 0; - -/* Spinlock protecting the above globals. */ -static DEFINE_SPINLOCK(mb_sram_lock); - -/* Allocate a memory block at least SIZE bytes long in the Mother-A SRAM - space. */ -static void *alloc_mb_sram (size_t size) -{ - struct mb_sram_free_area *prev, *fa; - unsigned long flags; - void *mem = 0; - - spin_lock_irqsave (mb_sram_lock, flags); - - /* Look for a free area that can contain SIZE bytes. */ - for (prev = 0, fa = mb_sram_free_areas; fa; prev = fa, fa = fa->next) - if (fa->size >= size) { - /* Found one! */ - mem = fa->mem; - - if (fa->size == size) { - /* In fact, it fits exactly, so remove - this node from the free-list. */ - if (prev) - prev->next = fa->next; - else - mb_sram_free_areas = fa->next; - /* Put it on the free-list-entry-free-list. */ - fa->next = mb_sram_free_free_areas; - mb_sram_free_free_areas = fa; - } else { - /* FA is bigger than SIZE, so just - reduce its size to account for this - allocation. */ - fa->mem += size; - fa->size -= size; - } - - break; - } - - spin_unlock_irqrestore (mb_sram_lock, flags); - - return mem; -} - -/* Return the memory area MEM of size SIZE to the MB SRAM free pool. */ -static void free_mb_sram (void *mem, size_t size) -{ - struct mb_sram_free_area *prev, *fa, *new_fa; - unsigned long flags; - void *end = mem + size; - - spin_lock_irqsave (mb_sram_lock, flags); - - retry: - /* Find an adjacent free-list entry. */ - for (prev = 0, fa = mb_sram_free_areas; fa; prev = fa, fa = fa->next) - if (fa->mem == end) { - /* FA is just after MEM, grow down to encompass it. */ - fa->mem = mem; - fa->size += size; - goto done; - } else if (fa->mem + fa->size == mem) { - struct mb_sram_free_area *next_fa = fa->next; - - /* FA is just before MEM, expand to encompass it. */ - fa->size += size; - - /* See if FA can now be merged with its successor. */ - if (next_fa && fa->mem + fa->size == next_fa->mem) { - /* Yup; merge NEXT_FA's info into FA. */ - fa->size += next_fa->size; - fa->next = next_fa->next; - /* Free NEXT_FA. */ - next_fa->next = mb_sram_free_free_areas; - mb_sram_free_free_areas = next_fa; - } - goto done; - } else if (fa->mem > mem) - /* We've reached the right spot in the free-list - without finding an adjacent free-area, so add - a new free area to hold mem. */ - break; - - /* Make a new free-list entry. */ - - /* First, get a free-list entry. */ - if (! mb_sram_free_free_areas) { - /* There are none, so make some. */ - void *block; - size_t block_size = sizeof (struct mb_sram_free_area) * 8; - - /* Don't hold the lock while calling kmalloc (I'm not - sure whether it would be a problem, since we use - GFP_ATOMIC, but it makes me nervous). */ - spin_unlock_irqrestore (mb_sram_lock, flags); - - block = kmalloc (block_size, GFP_ATOMIC); - if (! block) - panic ("free_mb_sram: can't allocate free-list entry"); - - /* Now get the lock back. */ - spin_lock_irqsave (mb_sram_lock, flags); - - /* Add the new free free-list entries. */ - while (block_size > 0) { - struct mb_sram_free_area *nfa = block; - nfa->next = mb_sram_free_free_areas; - mb_sram_free_free_areas = nfa; - block += sizeof *nfa; - block_size -= sizeof *nfa; - } - - /* Since we dropped the lock to call kmalloc, the - free-list could have changed, so retry from the - beginning. */ - goto retry; - } - - /* Remove NEW_FA from the free-list of free-list entries. */ - new_fa = mb_sram_free_free_areas; - mb_sram_free_free_areas = new_fa->next; - - /* NEW_FA initially holds only MEM. */ - new_fa->mem = mem; - new_fa->size = size; - - /* Insert NEW_FA in the free-list between PREV and FA. */ - new_fa->next = fa; - if (prev) - prev->next = new_fa; - else - mb_sram_free_areas = new_fa; - - done: - spin_unlock_irqrestore (mb_sram_lock, flags); -} - - -/* Maintainence of CPU -> Mother-A DMA mappings. */ - -struct dma_mapping { - void *cpu_addr; - void *mb_sram_addr; - size_t size; - struct dma_mapping *next; -}; - -/* A list of mappings from CPU addresses to MB SRAM addresses for active - DMA blocks (that have been `granted' to the PCI device). */ -static struct dma_mapping *active_dma_mappings = 0; - -/* A list of free mapping objects. */ -static struct dma_mapping *free_dma_mappings = 0; - -/* Spinlock protecting the above globals. */ -static DEFINE_SPINLOCK(dma_mappings_lock); - -static struct dma_mapping *new_dma_mapping (size_t size) -{ - unsigned long flags; - struct dma_mapping *mapping; - void *mb_sram_block = alloc_mb_sram (size); - - if (! mb_sram_block) - return 0; - - spin_lock_irqsave (dma_mappings_lock, flags); - - if (! free_dma_mappings) { - /* We're out of mapping structures, make more. */ - void *mblock; - size_t mblock_size = sizeof (struct dma_mapping) * 8; - - /* Don't hold the lock while calling kmalloc (I'm not - sure whether it would be a problem, since we use - GFP_ATOMIC, but it makes me nervous). */ - spin_unlock_irqrestore (dma_mappings_lock, flags); - - mblock = kmalloc (mblock_size, GFP_ATOMIC); - if (! mblock) { - free_mb_sram (mb_sram_block, size); - return 0; - } - - /* Get the lock back. */ - spin_lock_irqsave (dma_mappings_lock, flags); - - /* Add the new mapping structures to the free-list. */ - while (mblock_size > 0) { - struct dma_mapping *fm = mblock; - fm->next = free_dma_mappings; - free_dma_mappings = fm; - mblock += sizeof *fm; - mblock_size -= sizeof *fm; - } - } - - /* Get a mapping struct from the freelist. */ - mapping = free_dma_mappings; - free_dma_mappings = mapping->next; - - /* Initialize the mapping. Other fields should be filled in by - caller. */ - mapping->mb_sram_addr = mb_sram_block; - mapping->size = size; - - /* Add it to the list of active mappings. */ - mapping->next = active_dma_mappings; - active_dma_mappings = mapping; - - spin_unlock_irqrestore (dma_mappings_lock, flags); - - return mapping; -} - -static struct dma_mapping *find_dma_mapping (void *mb_sram_addr) -{ - unsigned long flags; - struct dma_mapping *mapping; - - spin_lock_irqsave (dma_mappings_lock, flags); - - for (mapping = active_dma_mappings; mapping; mapping = mapping->next) - if (mapping->mb_sram_addr == mb_sram_addr) { - spin_unlock_irqrestore (dma_mappings_lock, flags); - return mapping; - } - - panic ("find_dma_mapping: unmapped PCI DMA addr 0x%x", - MB_SRAM_TO_PCI (mb_sram_addr)); -} - -static struct dma_mapping *deactivate_dma_mapping (void *mb_sram_addr) -{ - unsigned long flags; - struct dma_mapping *mapping, *prev; - - spin_lock_irqsave (dma_mappings_lock, flags); - - for (prev = 0, mapping = active_dma_mappings; - mapping; - prev = mapping, mapping = mapping->next) - { - if (mapping->mb_sram_addr == mb_sram_addr) { - /* This is the MAPPING; deactivate it. */ - if (prev) - prev->next = mapping->next; - else - active_dma_mappings = mapping->next; - - spin_unlock_irqrestore (dma_mappings_lock, flags); - - return mapping; - } - } - - panic ("deactivate_dma_mapping: unmapped PCI DMA addr 0x%x", - MB_SRAM_TO_PCI (mb_sram_addr)); -} - -/* Return MAPPING to the freelist. */ -static inline void -free_dma_mapping (struct dma_mapping *mapping) -{ - unsigned long flags; - - free_mb_sram (mapping->mb_sram_addr, mapping->size); - - spin_lock_irqsave (dma_mappings_lock, flags); - - mapping->next = free_dma_mappings; - free_dma_mappings = mapping; - - spin_unlock_irqrestore (dma_mappings_lock, flags); -} - - -/* Single PCI DMA mappings. */ - -/* `Grant' to PDEV the memory block at CPU_ADDR, for doing DMA. The - 32-bit PCI bus mastering address to use is returned. the device owns - this memory until either pci_unmap_single or pci_dma_sync_single is - performed. */ -dma_addr_t -pci_map_single (struct pci_dev *pdev, void *cpu_addr, size_t size, int dir) -{ - struct dma_mapping *mapping = new_dma_mapping (size); - - if (! mapping) - return 0; - - mapping->cpu_addr = cpu_addr; - - if (dir == PCI_DMA_BIDIRECTIONAL || dir == PCI_DMA_TODEVICE) - memcpy (mapping->mb_sram_addr, cpu_addr, size); - - return MB_SRAM_TO_PCI (mapping->mb_sram_addr); -} - -/* Return to the CPU the PCI DMA memory block previously `granted' to - PDEV, at DMA_ADDR. */ -void pci_unmap_single (struct pci_dev *pdev, dma_addr_t dma_addr, size_t size, - int dir) -{ - void *mb_sram_addr = PCI_TO_MB_SRAM (dma_addr); - struct dma_mapping *mapping = deactivate_dma_mapping (mb_sram_addr); - - if (size != mapping->size) - panic ("pci_unmap_single: size (%d) doesn't match" - " size of mapping at PCI DMA addr 0x%x (%d)\n", - size, dma_addr, mapping->size); - - /* Copy back the DMA'd contents if necessary. */ - if (dir == PCI_DMA_BIDIRECTIONAL || dir == PCI_DMA_FROMDEVICE) - memcpy (mapping->cpu_addr, mb_sram_addr, size); - - /* Return mapping to the freelist. */ - free_dma_mapping (mapping); -} - -/* Make physical memory consistent for a single streaming mode DMA - translation after a transfer. - - If you perform a pci_map_single() but wish to interrogate the - buffer using the cpu, yet do not wish to teardown the PCI dma - mapping, you must call this function before doing so. At the next - point you give the PCI dma address back to the card, you must first - perform a pci_dma_sync_for_device, and then the device again owns - the buffer. */ -void -pci_dma_sync_single_for_cpu (struct pci_dev *pdev, dma_addr_t dma_addr, size_t size, - int dir) -{ - void *mb_sram_addr = PCI_TO_MB_SRAM (dma_addr); - struct dma_mapping *mapping = find_dma_mapping (mb_sram_addr); - - /* Synchronize the DMA buffer with the CPU buffer if necessary. */ - if (dir == PCI_DMA_FROMDEVICE) - memcpy (mapping->cpu_addr, mb_sram_addr, size); - else if (dir == PCI_DMA_TODEVICE) - ; /* nothing to do */ - else - panic("pci_dma_sync_single: unsupported sync dir: %d", dir); -} - -void -pci_dma_sync_single_for_device (struct pci_dev *pdev, dma_addr_t dma_addr, size_t size, - int dir) -{ - void *mb_sram_addr = PCI_TO_MB_SRAM (dma_addr); - struct dma_mapping *mapping = find_dma_mapping (mb_sram_addr); - - /* Synchronize the DMA buffer with the CPU buffer if necessary. */ - if (dir == PCI_DMA_FROMDEVICE) - ; /* nothing to do */ - else if (dir == PCI_DMA_TODEVICE) - memcpy (mb_sram_addr, mapping->cpu_addr, size); - else - panic("pci_dma_sync_single: unsupported sync dir: %d", dir); -} - - -/* Scatter-gather PCI DMA mappings. */ - -/* Do multiple DMA mappings at once. */ -int -pci_map_sg (struct pci_dev *pdev, struct scatterlist *sg, int sg_len, int dir) -{ - BUG (); - return 0; -} - -/* Unmap multiple DMA mappings at once. */ -void -pci_unmap_sg (struct pci_dev *pdev, struct scatterlist *sg, int sg_len,int dir) -{ - BUG (); -} - -/* Make physical memory consistent for a set of streaming mode DMA - translations after a transfer. The same as pci_dma_sync_single_* but - for a scatter-gather list, same rules and usage. */ - -void -pci_dma_sync_sg_for_cpu (struct pci_dev *dev, - struct scatterlist *sg, int sg_len, - int dir) -{ - BUG (); -} - -void -pci_dma_sync_sg_for_device (struct pci_dev *dev, - struct scatterlist *sg, int sg_len, - int dir) -{ - BUG (); -} - - -/* PCI mem mapping. */ - -/* Allocate and map kernel buffer using consistent mode DMA for PCI - device. Returns non-NULL cpu-view pointer to the buffer if - successful and sets *DMA_ADDR to the pci side dma address as well, - else DMA_ADDR is undefined. */ -void * -pci_alloc_consistent (struct pci_dev *pdev, size_t size, dma_addr_t *dma_addr) -{ - void *mb_sram_mem = alloc_mb_sram (size); - if (mb_sram_mem) - *dma_addr = MB_SRAM_TO_PCI (mb_sram_mem); - return mb_sram_mem; -} - -/* Free and unmap a consistent DMA buffer. CPU_ADDR and DMA_ADDR must - be values that were returned from pci_alloc_consistent. SIZE must be - the same as what as passed into pci_alloc_consistent. References to - the memory and mappings associated with CPU_ADDR or DMA_ADDR past - this call are illegal. */ -void -pci_free_consistent (struct pci_dev *pdev, size_t size, void *cpu_addr, - dma_addr_t dma_addr) -{ - void *mb_sram_mem = PCI_TO_MB_SRAM (dma_addr); - free_mb_sram (mb_sram_mem, size); -} - - -/* iomap/iomap */ - -void __iomem *pci_iomap (struct pci_dev *dev, int bar, unsigned long max) -{ - resource_size_t start = pci_resource_start (dev, bar); - resource_size_t len = pci_resource_len (dev, bar); - - if (!start || len == 0) - return 0; - - /* None of the ioremap functions actually do anything, other than - re-casting their argument, so don't bother differentiating them. */ - return ioremap (start, len); -} - -void pci_iounmap (struct pci_dev *dev, void __iomem *addr) -{ - /* nothing */ -} - - -/* symbol exports (for modules) */ - -EXPORT_SYMBOL (pci_map_single); -EXPORT_SYMBOL (pci_unmap_single); -EXPORT_SYMBOL (pci_alloc_consistent); -EXPORT_SYMBOL (pci_free_consistent); -EXPORT_SYMBOL (pci_dma_sync_single_for_cpu); -EXPORT_SYMBOL (pci_dma_sync_single_for_device); -EXPORT_SYMBOL (pci_iomap); -EXPORT_SYMBOL (pci_iounmap); diff --git a/arch/v850/kernel/rte_me2_cb.c b/arch/v850/kernel/rte_me2_cb.c deleted file mode 100644 index 46803d48dffe..000000000000 --- a/arch/v850/kernel/rte_me2_cb.c +++ /dev/null @@ -1,298 +0,0 @@ -/* - * arch/v850/kernel/rte_me2_cb.c -- Midas labs RTE-V850E/ME2-CB board - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include -#include -#include -#include -#include - -#include "mach.h" - -extern unsigned long *_intv_start; -extern unsigned long *_intv_end; - -/* LED access routines. */ -extern unsigned read_leds (int pos, char *buf, int len); -extern unsigned write_leds (int pos, const char *buf, int len); - - -/* SDRAM are almost contiguous (with a small hole in between; - see mach_reserve_bootmem for details), so just use both as one big area. */ -#define RAM_START SDRAM_ADDR -#define RAM_END (SDRAM_ADDR + SDRAM_SIZE) - - -void __init mach_get_physical_ram (unsigned long *ram_start, - unsigned long *ram_len) -{ - *ram_start = RAM_START; - *ram_len = RAM_END - RAM_START; -} - -void mach_gettimeofday (struct timespec *tv) -{ - tv->tv_sec = 0; - tv->tv_nsec = 0; -} - -/* Called before configuring an on-chip UART. */ -void rte_me2_cb_uart_pre_configure (unsigned chan, - unsigned cflags, unsigned baud) -{ - /* The RTE-V850E/ME2-CB connects some general-purpose I/O - pins on the CPU to the RTS/CTS lines of UARTB channel 0's - serial connection. - I/O pins P21 and P22 are RTS and CTS respectively. */ - if (chan == 0) { - /* Put P21 & P22 in I/O port mode. */ - ME2_PORT2_PMC &= ~0x6; - /* Make P21 and output, and P22 an input. */ - ME2_PORT2_PM = (ME2_PORT2_PM & ~0xC) | 0x4; - } - - me2_uart_pre_configure (chan, cflags, baud); -} - -void __init mach_init_irqs (void) -{ - /* Initialize interrupts. */ - me2_init_irqs (); - rte_me2_cb_init_irqs (); -} - -#ifdef CONFIG_ROM_KERNEL -/* Initialization for kernel in ROM. */ -static inline rom_kernel_init (void) -{ - /* If the kernel is in ROM, we have to copy any initialized data - from ROM into RAM. */ - extern unsigned long _data_load_start, _sdata, _edata; - register unsigned long *src = &_data_load_start; - register unsigned long *dst = &_sdata, *end = &_edata; - - while (dst != end) - *dst++ = *src++; -} -#endif /* CONFIG_ROM_KERNEL */ - -static void install_interrupt_vectors (void) -{ - unsigned long *p1, *p2; - - ME2_IRAMM = 0x03; /* V850E/ME2 iRAM write mode */ - - /* vector copy to iRAM */ - p1 = (unsigned long *)0; /* v85x vector start */ - p2 = (unsigned long *)&_intv_start; - while (p2 < (unsigned long *)&_intv_end) - *p1++ = *p2++; - - ME2_IRAMM = 0x00; /* V850E/ME2 iRAM read mode */ -} - -/* CompactFlash */ - -static void cf_power_on (void) -{ - /* CF card detected? */ - if (CB_CF_STS0 & 0x0030) - return; - - CB_CF_REG0 = 0x0002; /* reest on */ - mdelay (10); - CB_CF_REG0 = 0x0003; /* power on */ - mdelay (10); - CB_CF_REG0 = 0x0001; /* reset off */ - mdelay (10); -} - -static void cf_power_off (void) -{ - CB_CF_REG0 = 0x0003; /* power on */ - mdelay (10); - CB_CF_REG0 = 0x0002; /* reest on */ - mdelay (10); -} - -void __init mach_early_init (void) -{ - install_interrupt_vectors (); - - /* CS1 SDRAM instruction cache enable */ - v850e_cache_enable (0x04, 0x03, 0); - - rte_cb_early_init (); - - /* CompactFlash power on */ - cf_power_on (); - -#if defined (CONFIG_ROM_KERNEL) - rom_kernel_init (); -#endif -} - - -/* RTE-V850E/ME2-CB Programmable Interrupt Controller. */ - -static struct cb_pic_irq_init cb_pic_irq_inits[] = { - { "CB_EXTTM0", IRQ_CB_EXTTM0, 1, 1, 6 }, - { "CB_EXTSIO", IRQ_CB_EXTSIO, 1, 1, 6 }, - { "CB_TOVER", IRQ_CB_TOVER, 1, 1, 6 }, - { "CB_GINT0", IRQ_CB_GINT0, 1, 1, 6 }, - { "CB_USB", IRQ_CB_USB, 1, 1, 6 }, - { "CB_LANC", IRQ_CB_LANC, 1, 1, 6 }, - { "CB_USB_VBUS_ON", IRQ_CB_USB_VBUS_ON, 1, 1, 6 }, - { "CB_USB_VBUS_OFF", IRQ_CB_USB_VBUS_OFF, 1, 1, 6 }, - { "CB_EXTTM1", IRQ_CB_EXTTM1, 1, 1, 6 }, - { "CB_EXTTM2", IRQ_CB_EXTTM2, 1, 1, 6 }, - { 0 } -}; -#define NUM_CB_PIC_IRQ_INITS (ARRAY_SIZE(cb_pic_irq_inits) - 1) - -static struct hw_interrupt_type cb_pic_hw_itypes[NUM_CB_PIC_IRQ_INITS]; -static unsigned char cb_pic_active_irqs = 0; - -void __init rte_me2_cb_init_irqs (void) -{ - cb_pic_init_irq_types (cb_pic_irq_inits, cb_pic_hw_itypes); - - /* Initalize on board PIC1 (not PIC0) enable */ - CB_PIC_INT0M = 0x0000; - CB_PIC_INT1M = 0x0000; - CB_PIC_INTR = 0x0000; - CB_PIC_INTEN |= CB_PIC_INT1EN; - - ME2_PORT2_PMC |= 0x08; /* INTP23/SCK1 mode */ - ME2_PORT2_PFC &= ~0x08; /* INTP23 mode */ - ME2_INTR(2) &= ~0x08; /* INTP23 falling-edge detect */ - ME2_INTF(2) &= ~0x08; /* " */ - - rte_cb_init_irqs (); /* gbus &c */ -} - - -/* Enable interrupt handling for interrupt IRQ. */ -void cb_pic_enable_irq (unsigned irq) -{ - CB_PIC_INT1M |= 1 << (irq - CB_PIC_BASE_IRQ); -} - -void cb_pic_disable_irq (unsigned irq) -{ - CB_PIC_INT1M &= ~(1 << (irq - CB_PIC_BASE_IRQ)); -} - -void cb_pic_shutdown_irq (unsigned irq) -{ - cb_pic_disable_irq (irq); - - if (--cb_pic_active_irqs == 0) - free_irq (IRQ_CB_PIC, 0); - - CB_PIC_INT1M &= ~(1 << (irq - CB_PIC_BASE_IRQ)); -} - -static irqreturn_t cb_pic_handle_irq (int irq, void *dev_id, - struct pt_regs *regs) -{ - irqreturn_t rval = IRQ_NONE; - unsigned status = CB_PIC_INTR; - unsigned enable = CB_PIC_INT1M; - - /* Only pay attention to enabled interrupts. */ - status &= enable; - - CB_PIC_INTEN &= ~CB_PIC_INT1EN; - - if (status) { - unsigned mask = 1; - - irq = CB_PIC_BASE_IRQ; - do { - /* There's an active interrupt, find out which one, - and call its handler. */ - while (! (status & mask)) { - irq++; - mask <<= 1; - } - status &= ~mask; - - CB_PIC_INTR = mask; - - /* Recursively call handle_irq to handle it. */ - handle_irq (irq, regs); - rval = IRQ_HANDLED; - } while (status); - } - - CB_PIC_INTEN |= CB_PIC_INT1EN; - - return rval; -} - - -static void irq_nop (unsigned irq) { } - -static unsigned cb_pic_startup_irq (unsigned irq) -{ - int rval; - - if (cb_pic_active_irqs == 0) { - rval = request_irq (IRQ_CB_PIC, cb_pic_handle_irq, - IRQF_DISABLED, "cb_pic_handler", 0); - if (rval != 0) - return rval; - } - - cb_pic_active_irqs++; - - cb_pic_enable_irq (irq); - - return 0; -} - -/* Initialize HW_IRQ_TYPES for INTC-controlled irqs described in array - INITS (which is terminated by an entry with the name field == 0). */ -void __init cb_pic_init_irq_types (struct cb_pic_irq_init *inits, - struct hw_interrupt_type *hw_irq_types) -{ - struct cb_pic_irq_init *init; - for (init = inits; init->name; init++) { - struct hw_interrupt_type *hwit = hw_irq_types++; - - hwit->typename = init->name; - - hwit->startup = cb_pic_startup_irq; - hwit->shutdown = cb_pic_shutdown_irq; - hwit->enable = cb_pic_enable_irq; - hwit->disable = cb_pic_disable_irq; - hwit->ack = irq_nop; - hwit->end = irq_nop; - - /* Initialize kernel IRQ infrastructure for this interrupt. */ - init_irq_handlers(init->base, init->num, init->interval, hwit); - } -} diff --git a/arch/v850/kernel/rte_me2_cb.ld b/arch/v850/kernel/rte_me2_cb.ld deleted file mode 100644 index cf0766065ec6..000000000000 --- a/arch/v850/kernel/rte_me2_cb.ld +++ /dev/null @@ -1,30 +0,0 @@ -/* Linker script for the Midas labs RTE-V850E/ME2-CB evaluation board - (CONFIG_RTE_CB_ME2), with kernel in SDRAM. */ - -MEMORY { - /* 128Kbyte of IRAM */ - IRAM : ORIGIN = 0x00000000, LENGTH = 0x00020000 - - /* 32MB of SDRAM. */ - SDRAM : ORIGIN = SDRAM_ADDR, LENGTH = SDRAM_SIZE -} - -#define KRAM SDRAM - -SECTIONS { - .text : { - __kram_start = . ; - TEXT_CONTENTS - INTV_CONTENTS /* copy to iRAM (0x0-0x620) */ - } > KRAM - - .data : { - DATA_CONTENTS - BSS_CONTENTS - RAMK_INIT_CONTENTS - __kram_end = . ; - BOOTMAP_CONTENTS - } > KRAM - - .root ALIGN (4096) : { ROOT_FS_CONTENTS } > SDRAM -} diff --git a/arch/v850/kernel/rte_nb85e_cb-multi.ld b/arch/v850/kernel/rte_nb85e_cb-multi.ld deleted file mode 100644 index de347b4fffac..000000000000 --- a/arch/v850/kernel/rte_nb85e_cb-multi.ld +++ /dev/null @@ -1,57 +0,0 @@ -/* Linker script for the Midas labs RTE-NB85E-CB evaluation board - (CONFIG_RTE_CB_NB85E), with the Multi debugger ROM monitor . */ - -MEMORY { - /* 1MB of SRAM; we can't use the last 96KB, because it's used by - the monitor scratch-RAM. This memory is mirrored 4 times. */ - SRAM : ORIGIN = SRAM_ADDR, LENGTH = (SRAM_SIZE - MON_SCRATCH_SIZE) - /* Monitor scratch RAM; only the interrupt vectors should go here. */ - MRAM : ORIGIN = MON_SCRATCH_ADDR, LENGTH = MON_SCRATCH_SIZE - /* 16MB of SDRAM. */ - SDRAM : ORIGIN = SDRAM_ADDR, LENGTH = SDRAM_SIZE -} - -#ifdef CONFIG_RTE_CB_NB85E_KSRAM -# define KRAM SRAM -#else -# define KRAM SDRAM -#endif - -SECTIONS { - /* We can't use RAMK_KRAM_CONTENTS because that puts the whole - kernel in a single ELF segment, and the Multi debugger (which - we use to load the kernel) appears to have bizarre problems - dealing with it. */ - - .text : { - __kram_start = . ; - TEXT_CONTENTS - } > KRAM - - .data : { - DATA_CONTENTS - BSS_CONTENTS - RAMK_INIT_CONTENTS - __kram_end = . ; - BOOTMAP_CONTENTS - - /* The address at which the interrupt vectors are initially - loaded by the loader. We can't load the interrupt vectors - directly into their target location, because the monitor - ROM for the GHS Multi debugger barfs if we try. - Unfortunately, Multi also doesn't deal correctly with ELF - sections where the LMA and VMA differ (it just ignores the - LMA), so we can't use that feature to work around the - problem! What we do instead is just put the interrupt - vectors into a normal section, and have the - `mach_early_init' function for Midas boards do the - necessary copying and relocation at runtime (this section - basically only contains `jr' instructions, so it's not - that hard). */ - . = ALIGN (0x10) ; - __intv_load_start = . ; - INTV_CONTENTS - } > KRAM - - .root ALIGN (4096) : { ROOT_FS_CONTENTS } > SDRAM -} diff --git a/arch/v850/kernel/rte_nb85e_cb.c b/arch/v850/kernel/rte_nb85e_cb.c deleted file mode 100644 index b4a045da5d70..000000000000 --- a/arch/v850/kernel/rte_nb85e_cb.c +++ /dev/null @@ -1,81 +0,0 @@ -/* - * arch/v850/kernel/rte_nb85e_cb.c -- Midas labs RTE-V850E/NB85E-CB board - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include - -#include "mach.h" - -void __init mach_early_init (void) -{ - /* Configure caching; some possible settings: - - BHC = 0x0000, DCC = 0x0000 -- all caching disabled - BHC = 0x0040, DCC = 0x0000 -- SDRAM: icache only - BHC = 0x0080, DCC = 0x0C00 -- SDRAM: write-back dcache only - BHC = 0x00C0, DCC = 0x0C00 -- SDRAM: icache + write-back dcache - BHC = 0x00C0, DCC = 0x0800 -- SDRAM: icache + write-thru dcache - - We can only cache SDRAM (we can't use cache SRAM because it's in - the same memory region as the on-chip RAM and I/O space). - - Unfortunately, the dcache seems to be buggy, so we only use the - icache for now. */ - v850e_cache_enable (0x0040 /*BHC*/, 0x0003 /*ICC*/, 0x0000 /*DCC*/); - - rte_cb_early_init (); -} - -void __init mach_get_physical_ram (unsigned long *ram_start, - unsigned long *ram_len) -{ - /* We just use SDRAM here. */ - *ram_start = SDRAM_ADDR; - *ram_len = SDRAM_SIZE; -} - -void mach_gettimeofday (struct timespec *tv) -{ - tv->tv_sec = 0; - tv->tv_nsec = 0; -} - -/* Called before configuring an on-chip UART. */ -void rte_nb85e_cb_uart_pre_configure (unsigned chan, - unsigned cflags, unsigned baud) -{ - /* The RTE-NB85E-CB connects some general-purpose I/O pins on the - CPU to the RTS/CTS lines the UART's serial connection, as follows: - P00 = CTS (in), P01 = DSR (in), P02 = RTS (out), P03 = DTR (out). */ - - TEG_PORT0_PM = 0x03; /* P00 and P01 inputs, P02 and P03 outputs */ - TEG_PORT0_IO = 0x03; /* Accept input */ - - /* Do pre-configuration for the actual UART. */ - teg_uart_pre_configure (chan, cflags, baud); -} - -void __init mach_init_irqs (void) -{ - teg_init_irqs (); - rte_cb_init_irqs (); -} diff --git a/arch/v850/kernel/rte_nb85e_cb.ld b/arch/v850/kernel/rte_nb85e_cb.ld deleted file mode 100644 index b672f484f085..000000000000 --- a/arch/v850/kernel/rte_nb85e_cb.ld +++ /dev/null @@ -1,22 +0,0 @@ -/* Linker script for the Midas labs RTE-NB85E-CB evaluation board - (CONFIG_RTE_CB_NB85E). */ - -MEMORY { - LOW : ORIGIN = 0x0, LENGTH = 0x00100000 - /* 1MB of SRAM This memory is mirrored 4 times. */ - SRAM : ORIGIN = SRAM_ADDR, LENGTH = SRAM_SIZE - /* 16MB of SDRAM. */ - SDRAM : ORIGIN = SDRAM_ADDR, LENGTH = SDRAM_SIZE -} - -#ifdef CONFIG_RTE_CB_NB85E_KSRAM -# define KRAM SRAM -#else -# define KRAM SDRAM -#endif - -SECTIONS { - .intv : { INTV_CONTENTS } > LOW - .sram : { RAMK_KRAM_CONTENTS } > KRAM - .root : { ROOT_FS_CONTENTS } > SDRAM -} diff --git a/arch/v850/kernel/setup.c b/arch/v850/kernel/setup.c deleted file mode 100644 index 10335cecf7bd..000000000000 --- a/arch/v850/kernel/setup.c +++ /dev/null @@ -1,329 +0,0 @@ -/* - * arch/v850/kernel/setup.c -- Arch-dependent initialization functions - * - * Copyright (C) 2001,02,03,05,06 NEC Electronics Corporation - * Copyright (C) 2001,02,03,05,06 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include -#include /* we don't have swap, but for nr_free_pages */ -#include -#include -#include -#include -#include -#include -#include - -#include -#include - -#include "mach.h" - -/* These symbols are all defined in the linker map to delineate various - statically allocated regions of memory. */ - -extern char _intv_start, _intv_end; -/* `kram' is only used if the kernel uses part of normal user RAM. */ -extern char _kram_start __attribute__ ((__weak__)); -extern char _kram_end __attribute__ ((__weak__)); -extern char _init_start, _init_end; -extern char _bootmap; -extern char _stext, _etext, _sdata, _edata, _sbss, _ebss; -/* Many platforms use an embedded root image. */ -extern char _root_fs_image_start __attribute__ ((__weak__)); -extern char _root_fs_image_end __attribute__ ((__weak__)); - - -char __initdata command_line[COMMAND_LINE_SIZE]; - -/* Memory not used by the kernel. */ -static unsigned long total_ram_pages; - -/* System RAM. */ -static unsigned long ram_start = 0, ram_len = 0; - - -#define ADDR_TO_PAGE_UP(x) ((((unsigned long)x) + PAGE_SIZE-1) >> PAGE_SHIFT) -#define ADDR_TO_PAGE(x) (((unsigned long)x) >> PAGE_SHIFT) -#define PAGE_TO_ADDR(x) (((unsigned long)x) << PAGE_SHIFT) - -static void init_mem_alloc (unsigned long ram_start, unsigned long ram_len); - -void set_mem_root (void *addr, size_t len, char *cmd_line); - - -void __init setup_arch (char **cmdline) -{ - /* Keep a copy of command line */ - *cmdline = command_line; - memcpy (boot_command_line, command_line, COMMAND_LINE_SIZE); - boot_command_line[COMMAND_LINE_SIZE - 1] = '\0'; - - console_verbose (); - - init_mm.start_code = (unsigned long) &_stext; - init_mm.end_code = (unsigned long) &_etext; - init_mm.end_data = (unsigned long) &_edata; - init_mm.brk = (unsigned long) &_kram_end; - - /* Find out what mem this machine has. */ - mach_get_physical_ram (&ram_start, &ram_len); - /* ... and tell the kernel about it. */ - init_mem_alloc (ram_start, ram_len); - - printk (KERN_INFO "CPU: %s\nPlatform: %s\n", - CPU_MODEL_LONG, PLATFORM_LONG); - - /* do machine-specific setups. */ - mach_setup (cmdline); - -#ifdef CONFIG_MTD - if (!ROOT_DEV && &_root_fs_image_end > &_root_fs_image_start) - set_mem_root (&_root_fs_image_start, - &_root_fs_image_end - &_root_fs_image_start, - *cmdline); -#endif -} - -void __init trap_init (void) -{ -} - -#ifdef CONFIG_MTD - -/* From drivers/mtd/devices/slram.c */ -#define SLRAM_BLK_SZ 0x4000 - -/* Set the root filesystem to be the given memory region. - Some parameter may be appended to CMD_LINE. */ -void set_mem_root (void *addr, size_t len, char *cmd_line) -{ - /* Some sort of idiocy in MTD means we must supply a length that's - a multiple of SLRAM_BLK_SZ. We just round up the real length, - as the file system shouldn't attempt to access anything beyond - the end of the image anyway. */ - len = (((len - 1) + SLRAM_BLK_SZ) / SLRAM_BLK_SZ) * SLRAM_BLK_SZ; - - /* The only way to pass info to the MTD slram driver is via - the command line. */ - if (*cmd_line) { - cmd_line += strlen (cmd_line); - *cmd_line++ = ' '; - } - sprintf (cmd_line, "slram=root,0x%x,+0x%x", (u32)addr, (u32)len); - - ROOT_DEV = MKDEV (MTD_BLOCK_MAJOR, 0); -} -#endif - - -static void irq_nop (unsigned irq) { } -static unsigned irq_zero (unsigned irq) { return 0; } - -static void nmi_end (unsigned irq) -{ - if (irq != IRQ_NMI (0)) { - printk (KERN_CRIT "NMI %d is unrecoverable; restarting...", - irq - IRQ_NMI (0)); - machine_restart (0); - } -} - -static struct hw_interrupt_type nmi_irq_type = { - .typename = "NMI", - .startup = irq_zero, /* startup */ - .shutdown = irq_nop, /* shutdown */ - .enable = irq_nop, /* enable */ - .disable = irq_nop, /* disable */ - .ack = irq_nop, /* ack */ - .end = nmi_end, /* end */ -}; - -void __init init_IRQ (void) -{ - init_irq_handlers (0, NUM_MACH_IRQS, 1, 0); - init_irq_handlers (IRQ_NMI (0), NUM_NMIS, 1, &nmi_irq_type); - mach_init_irqs (); -} - - -void __init mem_init (void) -{ - max_mapnr = MAP_NR (ram_start + ram_len); - - num_physpages = ADDR_TO_PAGE (ram_len); - - total_ram_pages = free_all_bootmem (); - - printk (KERN_INFO - "Memory: %luK/%luK available" - " (%luK kernel code, %luK data)\n", - PAGE_TO_ADDR (nr_free_pages()) / 1024, - ram_len / 1024, - ((unsigned long)&_etext - (unsigned long)&_stext) / 1024, - ((unsigned long)&_ebss - (unsigned long)&_sdata) / 1024); -} - -void free_initmem (void) -{ - unsigned long ram_end = ram_start + ram_len; - unsigned long start = PAGE_ALIGN ((unsigned long)(&_init_start)); - - if (start >= ram_start && start < ram_end) { - unsigned long addr; - unsigned long end = PAGE_ALIGN ((unsigned long)(&_init_end)); - - if (end > ram_end) - end = ram_end; - - printk("Freeing unused kernel memory: %ldK freed\n", - (end - start) / 1024); - - for (addr = start; addr < end; addr += PAGE_SIZE) { - struct page *page = virt_to_page (addr); - ClearPageReserved (page); - init_page_count (page); - __free_page (page); - total_ram_pages++; - } - } -} - - -/* Initialize the `bootmem allocator'. RAM_START and RAM_LEN identify - what RAM may be used. */ -static void __init -init_bootmem_alloc (unsigned long ram_start, unsigned long ram_len) -{ - /* The part of the kernel that's in the same managed RAM space - used for general allocation. */ - unsigned long kram_start = (unsigned long)&_kram_start; - unsigned long kram_end = (unsigned long)&_kram_end; - /* End of the managed RAM space. */ - unsigned long ram_end = ram_start + ram_len; - /* Address range of the interrupt vector table. */ - unsigned long intv_start = (unsigned long)&_intv_start; - unsigned long intv_end = (unsigned long)&_intv_end; - /* True if the interrupt vectors are in the managed RAM area. */ - int intv_in_ram = (intv_end > ram_start && intv_start < ram_end); - /* True if the interrupt vectors are inside the kernel's RAM. */ - int intv_in_kram = (intv_end > kram_start && intv_start < kram_end); - /* A pointer to an optional function that reserves platform-specific - memory regions. We declare the pointer `volatile' to avoid gcc - turning the call into a static call (the problem is that since - it's a weak symbol, a static call may end up trying to reference - the location 0x0, which is not always reachable). */ - void (*volatile mrb) (void) = mach_reserve_bootmem; - /* The bootmem allocator's allocation bitmap. */ - unsigned long bootmap = (unsigned long)&_bootmap; - unsigned long bootmap_len; - - /* Round bootmap location up to next page. */ - bootmap = PAGE_TO_ADDR (ADDR_TO_PAGE_UP (bootmap)); - - /* Initialize bootmem allocator. */ - bootmap_len = init_bootmem_node (NODE_DATA (0), - ADDR_TO_PAGE (bootmap), - ADDR_TO_PAGE (PAGE_OFFSET), - ADDR_TO_PAGE (ram_end)); - - /* Now make the RAM actually allocatable (it starts out `reserved'). */ - free_bootmem (ram_start, ram_len); - - if (kram_end > kram_start) - /* Reserve the RAM part of the kernel's address space, so it - doesn't get allocated. */ - reserve_bootmem(kram_start, kram_end - kram_start, - BOOTMEM_DEFAULT); - - if (intv_in_ram && !intv_in_kram) - /* Reserve the interrupt vector space. */ - reserve_bootmem(intv_start, intv_end - intv_start, - BOOTMEM_DEFAULT); - - if (bootmap >= ram_start && bootmap < ram_end) - /* Reserve the bootmap space. */ - reserve_bootmem(bootmap, bootmap_len, - BOOTMEM_DEFAULT); - - /* Reserve the memory used by the root filesystem image if it's - in RAM. */ - if (&_root_fs_image_end > &_root_fs_image_start - && (unsigned long)&_root_fs_image_start >= ram_start - && (unsigned long)&_root_fs_image_start < ram_end) - reserve_bootmem ((unsigned long)&_root_fs_image_start, - &_root_fs_image_end - &_root_fs_image_start, - BOOTMEM_DEFAULT); - - /* Let the platform-dependent code reserve some too. */ - if (mrb) - (*mrb) (); -} - -/* Tell the kernel about what RAM it may use for memory allocation. */ -static void __init -init_mem_alloc (unsigned long ram_start, unsigned long ram_len) -{ - unsigned i; - unsigned long zones_size[MAX_NR_ZONES]; - - init_bootmem_alloc (ram_start, ram_len); - - for (i = 0; i < MAX_NR_ZONES; i++) - zones_size[i] = 0; - - /* We stuff all the memory into one area, which includes the - initial gap from PAGE_OFFSET to ram_start. */ - zones_size[ZONE_DMA] - = ADDR_TO_PAGE (ram_len + (ram_start - PAGE_OFFSET)); - - /* The allocator is very picky about the address of the first - allocatable page -- it must be at least as aligned as the - maximum allocation -- so try to detect cases where it will get - confused and signal them at compile time (this is a common - problem when porting to a new platform with ). There is a - similar runtime check in free_area_init_core. */ -#if ((PAGE_OFFSET >> PAGE_SHIFT) & ((1UL << (MAX_ORDER - 1)) - 1)) -#error MAX_ORDER is too large for given PAGE_OFFSET (use CONFIG_FORCE_MAX_ZONEORDER to change it) -#endif - NODE_DATA(0)->node_mem_map = NULL; - free_area_init_node(0, zones_size, ADDR_TO_PAGE (PAGE_OFFSET), 0); -} - - - -/* Taken from m68knommu */ -void show_mem(void) -{ - unsigned long i; - int free = 0, total = 0, reserved = 0, shared = 0; - int cached = 0; - - printk(KERN_INFO "\nMem-info:\n"); - show_free_areas(); - i = max_mapnr; - while (i-- > 0) { - total++; - if (PageReserved(mem_map+i)) - reserved++; - else if (PageSwapCache(mem_map+i)) - cached++; - else if (!page_count(mem_map+i)) - free++; - else - shared += page_count(mem_map+i) - 1; - } - printk(KERN_INFO "%d pages of RAM\n",total); - printk(KERN_INFO "%d free pages\n",free); - printk(KERN_INFO "%d reserved pages\n",reserved); - printk(KERN_INFO "%d pages shared\n",shared); - printk(KERN_INFO "%d pages swap cached\n",cached); -} diff --git a/arch/v850/kernel/signal.c b/arch/v850/kernel/signal.c deleted file mode 100644 index bf166e7e762c..000000000000 --- a/arch/v850/kernel/signal.c +++ /dev/null @@ -1,523 +0,0 @@ -/* - * arch/v850/kernel/signal.c -- Signal handling - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * Copyright (C) 1999,2000,2002 Niibe Yutaka & Kaz Kojima - * Copyright (C) 1991,1992 Linus Torvalds - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * 1997-11-28 Modified for POSIX.1b signals by Richard Henderson - * - * This file was derived from the sh version, arch/sh/kernel/signal.c - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include -#include -#include - -#define DEBUG_SIG 0 - -#define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP))) - -asmlinkage int do_signal(struct pt_regs *regs, sigset_t *oldset); - -/* - * Atomically swap in the new signal mask, and wait for a signal. - */ -asmlinkage int -sys_sigsuspend(old_sigset_t mask, struct pt_regs *regs) -{ - sigset_t saveset; - - mask &= _BLOCKABLE; - spin_lock_irq(¤t->sighand->siglock); - saveset = current->blocked; - siginitset(¤t->blocked, mask); - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - regs->gpr[GPR_RVAL] = -EINTR; - while (1) { - current->state = TASK_INTERRUPTIBLE; - schedule(); - if (do_signal(regs, &saveset)) - return -EINTR; - } -} - -asmlinkage int -sys_rt_sigsuspend(sigset_t *unewset, size_t sigsetsize, - struct pt_regs *regs) -{ - sigset_t saveset, newset; - - /* XXX: Don't preclude handling different sized sigset_t's. */ - if (sigsetsize != sizeof(sigset_t)) - return -EINVAL; - - if (copy_from_user(&newset, unewset, sizeof(newset))) - return -EFAULT; - sigdelsetmask(&newset, ~_BLOCKABLE); - spin_lock_irq(¤t->sighand->siglock); - saveset = current->blocked; - current->blocked = newset; - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - regs->gpr[GPR_RVAL] = -EINTR; - while (1) { - current->state = TASK_INTERRUPTIBLE; - schedule(); - if (do_signal(regs, &saveset)) - return -EINTR; - } -} - -asmlinkage int -sys_sigaction(int sig, const struct old_sigaction *act, - struct old_sigaction *oact) -{ - struct k_sigaction new_ka, old_ka; - int ret; - - if (act) { - old_sigset_t mask; - if (!access_ok(VERIFY_READ, act, sizeof(*act)) || - __get_user(new_ka.sa.sa_handler, &act->sa_handler) || - __get_user(new_ka.sa.sa_restorer, &act->sa_restorer)) - return -EFAULT; - __get_user(new_ka.sa.sa_flags, &act->sa_flags); - __get_user(mask, &act->sa_mask); - siginitset(&new_ka.sa.sa_mask, mask); - } - - ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL); - - if (!ret && oact) { - if (!access_ok(VERIFY_WRITE, oact, sizeof(*oact)) || - __put_user(old_ka.sa.sa_handler, &oact->sa_handler) || - __put_user(old_ka.sa.sa_restorer, &oact->sa_restorer)) - return -EFAULT; - __put_user(old_ka.sa.sa_flags, &oact->sa_flags); - __put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask); - } - - return ret; -} - -asmlinkage int -sys_sigaltstack(const stack_t *uss, stack_t *uoss, - struct pt_regs *regs) -{ - return do_sigaltstack(uss, uoss, regs->gpr[GPR_SP]); -} - - -/* - * Do a signal return; undo the signal stack. - */ - -struct sigframe -{ - struct sigcontext sc; - unsigned long extramask[_NSIG_WORDS-1]; - unsigned long tramp[2]; /* signal trampoline */ -}; - -struct rt_sigframe -{ - struct siginfo info; - struct ucontext uc; - unsigned long tramp[2]; /* signal trampoline */ -}; - -static int -restore_sigcontext(struct pt_regs *regs, struct sigcontext *sc, int *rval_p) -{ - unsigned int err = 0; - -#define COPY(x) err |= __get_user(regs->x, &sc->regs.x) - COPY(gpr[0]); COPY(gpr[1]); COPY(gpr[2]); COPY(gpr[3]); - COPY(gpr[4]); COPY(gpr[5]); COPY(gpr[6]); COPY(gpr[7]); - COPY(gpr[8]); COPY(gpr[9]); COPY(gpr[10]); COPY(gpr[11]); - COPY(gpr[12]); COPY(gpr[13]); COPY(gpr[14]); COPY(gpr[15]); - COPY(gpr[16]); COPY(gpr[17]); COPY(gpr[18]); COPY(gpr[19]); - COPY(gpr[20]); COPY(gpr[21]); COPY(gpr[22]); COPY(gpr[23]); - COPY(gpr[24]); COPY(gpr[25]); COPY(gpr[26]); COPY(gpr[27]); - COPY(gpr[28]); COPY(gpr[29]); COPY(gpr[30]); COPY(gpr[31]); - COPY(pc); COPY(psw); - COPY(ctpc); COPY(ctpsw); COPY(ctbp); -#undef COPY - - return err; -} - -asmlinkage int sys_sigreturn(struct pt_regs *regs) -{ - struct sigframe *frame = (struct sigframe *)regs->gpr[GPR_SP]; - sigset_t set; - int rval; - - if (!access_ok(VERIFY_READ, frame, sizeof(*frame))) - goto badframe; - - if (__get_user(set.sig[0], &frame->sc.oldmask) - || (_NSIG_WORDS > 1 - && __copy_from_user(&set.sig[1], &frame->extramask, - sizeof(frame->extramask)))) - goto badframe; - - sigdelsetmask(&set, ~_BLOCKABLE); - spin_lock_irq(¤t->sighand->siglock); - current->blocked = set; - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - if (restore_sigcontext(regs, &frame->sc, &rval)) - goto badframe; - return rval; - -badframe: - force_sig(SIGSEGV, current); - return 0; -} - -asmlinkage int sys_rt_sigreturn(struct pt_regs *regs) -{ - struct rt_sigframe *frame = (struct rt_sigframe *)regs->gpr[GPR_SP]; - sigset_t set; - stack_t st; - int rval; - - if (!access_ok(VERIFY_READ, frame, sizeof(*frame))) - goto badframe; - - if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set))) - goto badframe; - - sigdelsetmask(&set, ~_BLOCKABLE); - spin_lock_irq(¤t->sighand->siglock); - current->blocked = set; - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - if (restore_sigcontext(regs, &frame->uc.uc_mcontext, &rval)) - goto badframe; - - if (__copy_from_user(&st, &frame->uc.uc_stack, sizeof(st))) - goto badframe; - /* It is more difficult to avoid calling this function than to - call it and ignore errors. */ - do_sigaltstack(&st, NULL, regs->gpr[GPR_SP]); - - return rval; - -badframe: - force_sig(SIGSEGV, current); - return 0; -} - -/* - * Set up a signal frame. - */ - -static int -setup_sigcontext(struct sigcontext *sc, struct pt_regs *regs, - unsigned long mask) -{ - int err = 0; - -#define COPY(x) err |= __put_user(regs->x, &sc->regs.x) - COPY(gpr[0]); COPY(gpr[1]); COPY(gpr[2]); COPY(gpr[3]); - COPY(gpr[4]); COPY(gpr[5]); COPY(gpr[6]); COPY(gpr[7]); - COPY(gpr[8]); COPY(gpr[9]); COPY(gpr[10]); COPY(gpr[11]); - COPY(gpr[12]); COPY(gpr[13]); COPY(gpr[14]); COPY(gpr[15]); - COPY(gpr[16]); COPY(gpr[17]); COPY(gpr[18]); COPY(gpr[19]); - COPY(gpr[20]); COPY(gpr[21]); COPY(gpr[22]); COPY(gpr[23]); - COPY(gpr[24]); COPY(gpr[25]); COPY(gpr[26]); COPY(gpr[27]); - COPY(gpr[28]); COPY(gpr[29]); COPY(gpr[30]); COPY(gpr[31]); - COPY(pc); COPY(psw); - COPY(ctpc); COPY(ctpsw); COPY(ctbp); -#undef COPY - - err |= __put_user(mask, &sc->oldmask); - - return err; -} - -/* - * Determine which stack to use.. - */ -static inline void * -get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size) -{ - /* Default to using normal stack */ - unsigned long sp = regs->gpr[GPR_SP]; - - if ((ka->sa.sa_flags & SA_ONSTACK) != 0 && ! sas_ss_flags(sp)) - sp = current->sas_ss_sp + current->sas_ss_size; - - return (void *)((sp - frame_size) & -8UL); -} - -static void setup_frame(int sig, struct k_sigaction *ka, - sigset_t *set, struct pt_regs *regs) -{ - struct sigframe *frame; - int err = 0; - int signal; - - frame = get_sigframe(ka, regs, sizeof(*frame)); - - if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame))) - goto give_sigsegv; - - signal = current_thread_info()->exec_domain - && current_thread_info()->exec_domain->signal_invmap - && sig < 32 - ? current_thread_info()->exec_domain->signal_invmap[sig] - : sig; - - err |= setup_sigcontext(&frame->sc, regs, set->sig[0]); - - if (_NSIG_WORDS > 1) { - err |= __copy_to_user(frame->extramask, &set->sig[1], - sizeof(frame->extramask)); - } - - /* Set up to return from userspace. If provided, use a stub - already in userspace. */ - if (ka->sa.sa_flags & SA_RESTORER) { - regs->gpr[GPR_LP] = (unsigned long) ka->sa.sa_restorer; - } else { - /* Note, these encodings are _little endian_! */ - - /* addi __NR_sigreturn, r0, r12 */ - err |= __put_user(0x6600 | (__NR_sigreturn << 16), - frame->tramp + 0); - /* trap 0 */ - err |= __put_user(0x010007e0, - frame->tramp + 1); - - regs->gpr[GPR_LP] = (unsigned long)frame->tramp; - - flush_cache_sigtramp (regs->gpr[GPR_LP]); - } - - if (err) - goto give_sigsegv; - - /* Set up registers for signal handler. */ - regs->pc = (v850_reg_t) ka->sa.sa_handler; - regs->gpr[GPR_SP] = (v850_reg_t)frame; - /* Signal handler args: */ - regs->gpr[GPR_ARG0] = signal; /* arg 0: signum */ - regs->gpr[GPR_ARG1] = (v850_reg_t)&frame->sc;/* arg 1: sigcontext */ - - set_fs(USER_DS); - -#if DEBUG_SIG - printk("SIG deliver (%s:%d): sp=%p pc=%08lx ra=%08lx\n", - current->comm, current->pid, frame, regs->pc, ); -#endif - - return; - -give_sigsegv: - force_sigsegv(sig, current); -} - -static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info, - sigset_t *set, struct pt_regs *regs) -{ - struct rt_sigframe *frame; - int err = 0; - int signal; - - frame = get_sigframe(ka, regs, sizeof(*frame)); - - if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame))) - goto give_sigsegv; - - signal = current_thread_info()->exec_domain - && current_thread_info()->exec_domain->signal_invmap - && sig < 32 - ? current_thread_info()->exec_domain->signal_invmap[sig] - : sig; - - err |= copy_siginfo_to_user(&frame->info, info); - - /* Create the ucontext. */ - err |= __put_user(0, &frame->uc.uc_flags); - err |= __put_user(0, &frame->uc.uc_link); - err |= __put_user((void *)current->sas_ss_sp, - &frame->uc.uc_stack.ss_sp); - err |= __put_user(sas_ss_flags(regs->gpr[GPR_SP]), - &frame->uc.uc_stack.ss_flags); - err |= __put_user(current->sas_ss_size, &frame->uc.uc_stack.ss_size); - err |= setup_sigcontext(&frame->uc.uc_mcontext, - regs, set->sig[0]); - err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)); - - /* Set up to return from userspace. If provided, use a stub - already in userspace. */ - if (ka->sa.sa_flags & SA_RESTORER) { - regs->gpr[GPR_LP] = (unsigned long) ka->sa.sa_restorer; - } else { - /* Note, these encodings are _little endian_! */ - - /* addi __NR_sigreturn, r0, r12 */ - err |= __put_user(0x6600 | (__NR_sigreturn << 16), - frame->tramp + 0); - /* trap 0 */ - err |= __put_user(0x010007e0, - frame->tramp + 1); - - regs->gpr[GPR_LP] = (unsigned long)frame->tramp; - - flush_cache_sigtramp (regs->gpr[GPR_LP]); - } - - if (err) - goto give_sigsegv; - - /* Set up registers for signal handler. */ - regs->pc = (v850_reg_t) ka->sa.sa_handler; - regs->gpr[GPR_SP] = (v850_reg_t)frame; - /* Signal handler args: */ - regs->gpr[GPR_ARG0] = signal; /* arg 0: signum */ - regs->gpr[GPR_ARG1] = (v850_reg_t)&frame->info; /* arg 1: siginfo */ - regs->gpr[GPR_ARG2] = (v850_reg_t)&frame->uc; /* arg 2: ucontext */ - - set_fs(USER_DS); - -#if DEBUG_SIG - printk("SIG deliver (%s:%d): sp=%p pc=%08lx pr=%08lx\n", - current->comm, current->pid, frame, regs->pc, regs->pr); -#endif - - return; - -give_sigsegv: - force_sigsegv(sig, current); -} - -/* - * OK, we're invoking a handler - */ - -static void -handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka, - sigset_t *oldset, struct pt_regs * regs) -{ - /* Are we from a system call? */ - if (PT_REGS_SYSCALL (regs)) { - /* If so, check system call restarting.. */ - switch (regs->gpr[GPR_RVAL]) { - case -ERESTART_RESTARTBLOCK: - current_thread_info()->restart_block.fn = - do_no_restart_syscall; - /* fall through */ - case -ERESTARTNOHAND: - regs->gpr[GPR_RVAL] = -EINTR; - break; - - case -ERESTARTSYS: - if (!(ka->sa.sa_flags & SA_RESTART)) { - regs->gpr[GPR_RVAL] = -EINTR; - break; - } - /* fallthrough */ - case -ERESTARTNOINTR: - regs->gpr[12] = PT_REGS_SYSCALL (regs); - regs->pc -= 4; /* Size of `trap 0' insn. */ - } - - PT_REGS_SET_SYSCALL (regs, 0); - } - - /* Set up the stack frame */ - if (ka->sa.sa_flags & SA_SIGINFO) - setup_rt_frame(sig, ka, info, oldset, regs); - else - setup_frame(sig, ka, oldset, regs); - - spin_lock_irq(¤t->sighand->siglock); - sigorsets(¤t->blocked,¤t->blocked,&ka->sa.sa_mask); - if (!(ka->sa.sa_flags & SA_NODEFER)) - sigaddset(¤t->blocked,sig); - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); -} - -/* - * Note that 'init' is a special process: it doesn't get signals it doesn't - * want to handle. Thus you cannot kill init even with a SIGKILL even by - * mistake. - * - * Note that we go through the signals twice: once to check the signals that - * the kernel can handle, and then we build all the user-level signal handling - * stack-frames in one go after that. - */ -int do_signal(struct pt_regs *regs, sigset_t *oldset) -{ - siginfo_t info; - int signr; - struct k_sigaction ka; - - /* - * We want the common case to go fast, which - * is why we may in certain cases get here from - * kernel mode. Just return without doing anything - * if so. - */ - if (!user_mode(regs)) - return 1; - - if (!oldset) - oldset = ¤t->blocked; - - signr = get_signal_to_deliver(&info, &ka, regs, NULL); - if (signr > 0) { - /* Whee! Actually deliver the signal. */ - handle_signal(signr, &info, &ka, oldset, regs); - return 1; - } - - /* Did we come from a system call? */ - if (PT_REGS_SYSCALL (regs)) { - int rval = (int)regs->gpr[GPR_RVAL]; - /* Restart the system call - no handlers present */ - if (rval == -ERESTARTNOHAND - || rval == -ERESTARTSYS - || rval == -ERESTARTNOINTR) - { - regs->gpr[12] = PT_REGS_SYSCALL (regs); - regs->pc -= 4; /* Size of `trap 0' insn. */ - } - else if (rval == -ERESTART_RESTARTBLOCK) { - regs->gpr[12] = __NR_restart_syscall; - regs->pc -= 4; /* Size of `trap 0' insn. */ - } - } - return 0; -} diff --git a/arch/v850/kernel/sim.c b/arch/v850/kernel/sim.c deleted file mode 100644 index 467b4aa0acdd..000000000000 --- a/arch/v850/kernel/sim.c +++ /dev/null @@ -1,172 +0,0 @@ -/* - * arch/v850/kernel/sim.c -- Machine-specific stuff for GDB v850e simulator - * - * Copyright (C) 2001,02 NEC Corporation - * Copyright (C) 2001,02 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include - -#include "mach.h" - -/* The name of a file containing the root filesystem. */ -#define ROOT_FS "rootfs.image" - -extern void simcons_setup (void); -extern void simcons_poll_ttys (void); -extern void set_mem_root (void *addr, size_t len, char *cmd_line); - -static int read_file (const char *name, - unsigned long *addr, unsigned long *len, - const char **err); - -void __init mach_setup (char **cmdline) -{ - const char *err; - unsigned long root_dev_addr, root_dev_len; - - simcons_setup (); - - printk (KERN_INFO "Reading root filesystem: %s", ROOT_FS); - - if (read_file (ROOT_FS, &root_dev_addr, &root_dev_len, &err)) { - printk (" (size %luK)\n", root_dev_len / 1024); - set_mem_root ((void *)root_dev_addr, (size_t)root_dev_len, - *cmdline); - } else - printk ("...%s failed!\n", err); -} - -void mach_get_physical_ram (unsigned long *ram_start, unsigned long *ram_len) -{ - *ram_start = RAM_ADDR; - *ram_len = RAM_SIZE; -} - -void __init mach_sched_init (struct irqaction *timer_action) -{ - /* ...do magic timer initialization?... */ - mach_tick = simcons_poll_ttys; - setup_irq (0, timer_action); -} - - -static void irq_nop (unsigned irq) { } -static unsigned irq_zero (unsigned irq) { return 0; } - -static struct hw_interrupt_type sim_irq_type = { - .typename = "IRQ", - .startup = irq_zero, /* startup */ - .shutdown = irq_nop, /* shutdown */ - .enable = irq_nop, /* enable */ - .disable = irq_nop, /* disable */ - .ack = irq_nop, /* ack */ - .end = irq_nop, /* end */ -}; - -void __init mach_init_irqs (void) -{ - init_irq_handlers (0, NUM_MACH_IRQS, 1, &sim_irq_type); -} - - -void mach_gettimeofday (struct timespec *tv) -{ - long timeval[2], timezone[2]; - int rval = V850_SIM_SYSCALL (gettimeofday, timeval, timezone); - if (rval == 0) { - tv->tv_sec = timeval[0]; - tv->tv_nsec = timeval[1] * 1000; - } -} - -void machine_restart (char *__unused) -{ - V850_SIM_SYSCALL (write, 1, "RESTART\n", 8); - V850_SIM_SYSCALL (exit, 0); -} - -void machine_halt (void) -{ - V850_SIM_SYSCALL (write, 1, "HALT\n", 5); - V850_SIM_SYSCALL (exit, 0); -} - -void machine_power_off (void) -{ - V850_SIM_SYSCALL (write, 1, "POWER OFF\n", 10); - V850_SIM_SYSCALL (exit, 0); -} - - -/* Load data from a file called NAME into ram. The address and length - of the data image are returned in ADDR and LEN. */ -static int __init -read_file (const char *name, - unsigned long *addr, unsigned long *len, - const char **err) -{ - int rval, fd; - unsigned long cur, left; - /* Note this is not a normal stat buffer, it's an ad-hoc - structure defined by the simulator. */ - unsigned long stat_buf[10]; - - /* Stat the file to find out the length. */ - rval = V850_SIM_SYSCALL (stat, name, stat_buf); - if (rval < 0) { - if (err) *err = "stat"; - return 0; - } - *len = stat_buf[4]; - - /* Open the file; `0' is O_RDONLY. */ - fd = V850_SIM_SYSCALL (open, name, 0); - if (fd < 0) { - if (err) *err = "open"; - return 0; - } - - *addr = (unsigned long)alloc_bootmem(*len); - if (! *addr) { - V850_SIM_SYSCALL (close, fd); - if (err) *err = "alloc_bootmem"; - return 0; - } - - cur = *addr; - left = *len; - while (left > 0) { - int chunk = V850_SIM_SYSCALL (read, fd, cur, left); - if (chunk <= 0) - break; - cur += chunk; - left -= chunk; - } - V850_SIM_SYSCALL (close, fd); - if (left > 0) { - /* Some read failed. */ - free_bootmem (*addr, *len); - if (err) *err = "read"; - return 0; - } - - return 1; -} diff --git a/arch/v850/kernel/sim.ld b/arch/v850/kernel/sim.ld deleted file mode 100644 index 101885f3c9f0..000000000000 --- a/arch/v850/kernel/sim.ld +++ /dev/null @@ -1,13 +0,0 @@ -/* Linker script for the gdb v850e simulator (CONFIG_V850E_SIM). */ - -MEMORY { - /* Interrupt vectors. */ - INTV : ORIGIN = 0x0, LENGTH = 0xe0 - /* Main RAM. */ - RAM : ORIGIN = RAM_ADDR, LENGTH = RAM_SIZE -} - -SECTIONS { - .intv : { INTV_CONTENTS } > INTV - .ram : { RAMK_KRAM_CONTENTS } > RAM -} diff --git a/arch/v850/kernel/sim85e2.c b/arch/v850/kernel/sim85e2.c deleted file mode 100644 index 566dde5e6070..000000000000 --- a/arch/v850/kernel/sim85e2.c +++ /dev/null @@ -1,195 +0,0 @@ -/* - * arch/v850/kernel/sim85e2.c -- Machine-specific stuff for - * V850E2 RTL simulator - * - * Copyright (C) 2002,03 NEC Electronics Corporation - * Copyright (C) 2002,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include - -#include "mach.h" - - -/* There are 4 possible areas we can use: - - IRAM (1MB) is fast for instruction fetches, but slow for data - DRAM (1020KB) is fast for data, but slow for instructions - ERAM is cached, so should be fast for both insns and data - SDRAM is external DRAM, similar to ERAM -*/ - -#define INIT_MEMC_FOR_SDRAM -#define USE_SDRAM_AREA -#define KERNEL_IN_SDRAM_AREA - -#define DCACHE_MODE V850E2_CACHE_BTSC_DCM_WT -/*#define DCACHE_MODE V850E2_CACHE_BTSC_DCM_WB_ALLOC*/ - -#ifdef USE_SDRAM_AREA -#define RAM_START SDRAM_ADDR -#define RAM_END (SDRAM_ADDR + SDRAM_SIZE) -#else -/* When we use DRAM, we need to account for the fact that the end of it is - used for R0_RAM. */ -#define RAM_START DRAM_ADDR -#define RAM_END R0_RAM_ADDR -#endif - - -extern void memcons_setup (void); - - -#ifdef KERNEL_IN_SDRAM_AREA -#define EARLY_INIT_SECTION_ATTR __attribute__ ((section (".early.text"))) -#else -#define EARLY_INIT_SECTION_ATTR __init -#endif - -void EARLY_INIT_SECTION_ATTR mach_early_init (void) -{ - /* The sim85e2 simulator tracks `undefined' values, so to make - debugging easier, we begin by zeroing out all otherwise - undefined registers. This is not strictly necessary. - - The registers we zero are: - Every GPR except: - stack-pointer (r3) - task-pointer (r16) - our return addr (r31) - Every system register (SPR) that we know about except for - the PSW (SPR 5), which we zero except for the - disable-interrupts bit. - */ - - /* GPRs */ - asm volatile (" mov r0, r1 ; mov r0, r2 "); - asm volatile ("mov r0, r4 ; mov r0, r5 ; mov r0, r6 ; mov r0, r7 "); - asm volatile ("mov r0, r8 ; mov r0, r9 ; mov r0, r10; mov r0, r11"); - asm volatile ("mov r0, r12; mov r0, r13; mov r0, r14; mov r0, r15"); - asm volatile (" mov r0, r17; mov r0, r18; mov r0, r19"); - asm volatile ("mov r0, r20; mov r0, r21; mov r0, r22; mov r0, r23"); - asm volatile ("mov r0, r24; mov r0, r25; mov r0, r26; mov r0, r27"); - asm volatile ("mov r0, r28; mov r0, r29; mov r0, r30"); - - /* SPRs */ - asm volatile ("ldsr r0, 0; ldsr r0, 1; ldsr r0, 2; ldsr r0, 3"); - asm volatile ("ldsr r0, 4"); - asm volatile ("addi 0x20, r0, r1; ldsr r1, 5"); /* PSW */ - asm volatile ("ldsr r0, 16; ldsr r0, 17; ldsr r0, 18; ldsr r0, 19"); - asm volatile ("ldsr r0, 20"); - - -#ifdef INIT_MEMC_FOR_SDRAM - /* Settings for SDRAM controller. */ - V850E2_VSWC = 0x0042; - V850E2_BSC = 0x9286; - V850E2_BCT(0) = 0xb000; /* was: 0 */ - V850E2_BCT(1) = 0x000b; - V850E2_ASC = 0; - V850E2_LBS = 0xa9aa; /* was: 0xaaaa */ - V850E2_LBC(0) = 0; - V850E2_LBC(1) = 0; /* was: 0x3 */ - V850E2_BCC = 0; - V850E2_RFS(4) = 0x800a; /* was: 0xf109 */ - V850E2_SCR(4) = 0x2091; /* was: 0x20a1 */ - V850E2_RFS(3) = 0x800c; - V850E2_SCR(3) = 0x20a1; - V850E2_DWC(0) = 0; - V850E2_DWC(1) = 0; -#endif - -#if 0 -#ifdef CONFIG_V850E2_SIM85E2S - /* Turn on the caches. */ - V850E2_CACHE_BTSC = V850E2_CACHE_BTSC_ICM | DCACHE_MODE; - V850E2_BHC = 0x1010; -#elif CONFIG_V850E2_SIM85E2C - V850E2_CACHE_BTSC |= (V850E2_CACHE_BTSC_ICM | V850E2_CACHE_BTSC_DCM0); - V850E2_BUSM_BHC = 0xFFFF; -#endif -#else - V850E2_BHC = 0; -#endif - - /* Don't stop the simulator at `halt' instructions. */ - SIM85E2_NOTHAL = 1; - - /* Ensure that the simulator halts on a panic, instead of going - into an infinite loop inside the panic function. */ - panic_timeout = -1; -} - -void __init mach_setup (char **cmdline) -{ - memcons_setup (); -} - -void mach_get_physical_ram (unsigned long *ram_start, unsigned long *ram_len) -{ - *ram_start = RAM_START; - *ram_len = RAM_END - RAM_START; -} - -void __init mach_sched_init (struct irqaction *timer_action) -{ - /* The simulator actually cycles through all interrupts - periodically. We just pay attention to IRQ0, which gives us - 1/64 the rate of the periodic interrupts. */ - setup_irq (0, timer_action); -} - -void mach_gettimeofday (struct timespec *tv) -{ - tv->tv_sec = 0; - tv->tv_nsec = 0; -} - -/* Interrupts */ - -struct v850e_intc_irq_init irq_inits[] = { - { "IRQ", 0, NUM_MACH_IRQS, 1, 7 }, - { 0 } -}; -struct hw_interrupt_type hw_itypes[1]; - -/* Initialize interrupts. */ -void __init mach_init_irqs (void) -{ - v850e_intc_init_irq_types (irq_inits, hw_itypes); -} - - -void machine_halt (void) __attribute__ ((noreturn)); -void machine_halt (void) -{ - SIM85E2_SIMFIN = 0; /* Halt immediately. */ - for (;;) {} -} - -void machine_restart (char *__unused) -{ - machine_halt (); -} - -void machine_power_off (void) -{ - machine_halt (); -} - diff --git a/arch/v850/kernel/sim85e2.ld b/arch/v850/kernel/sim85e2.ld deleted file mode 100644 index 7470fd2ffb5b..000000000000 --- a/arch/v850/kernel/sim85e2.ld +++ /dev/null @@ -1,36 +0,0 @@ -/* Linker script for the sim85e2c simulator, which is a verilog simulation of - the V850E2 NA85E2C cpu core (CONFIG_V850E2_SIM85E2C). */ - -MEMORY { - /* 1MB of `instruction RAM', starting at 0. - Instruction fetches are much faster from IRAM than from DRAM. */ - IRAM : ORIGIN = IRAM_ADDR, LENGTH = IRAM_SIZE - - /* 1MB of `data RAM', below and contiguous with the I/O space. - Data fetches are much faster from DRAM than from IRAM. */ - DRAM : ORIGIN = DRAM_ADDR, LENGTH = DRAM_SIZE - - /* `external ram' (CS1 area), comes after IRAM. */ - ERAM : ORIGIN = ERAM_ADDR, LENGTH = ERAM_SIZE - - /* Dynamic RAM; uses memory controller. */ - SDRAM : ORIGIN = SDRAM_ADDR, LENGTH = SDRAM_SIZE -} - -SECTIONS { - .iram : { - INTV_CONTENTS - *arch/v850/kernel/head.o - *(.early.text) - } > IRAM - .dram : { - _memcons_output = . ; - . = . + 0x8000 ; - _memcons_output_end = . ; - } > DRAM - .sdram : { - /* We stick console output into a buffer here. */ - RAMK_KRAM_CONTENTS - ROOT_FS_CONTENTS - } > SDRAM -} diff --git a/arch/v850/kernel/simcons.c b/arch/v850/kernel/simcons.c deleted file mode 100644 index 9973596ae304..000000000000 --- a/arch/v850/kernel/simcons.c +++ /dev/null @@ -1,161 +0,0 @@ -/* - * arch/v850/kernel/simcons.c -- Console I/O for GDB v850e simulator - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include -#include -#include -#include -#include - -#include -#include -#include - - -/* Low-level console. */ - -static void simcons_write (struct console *co, const char *buf, unsigned len) -{ - V850_SIM_SYSCALL (write, 1, buf, len); -} - -static int simcons_read (struct console *co, char *buf, unsigned len) -{ - return V850_SIM_SYSCALL (read, 0, buf, len); -} - -static struct tty_driver *tty_driver; -static struct tty_driver *simcons_device (struct console *c, int *index) -{ - *index = c->index; - return tty_driver; -} - -static struct console simcons = -{ - .name = "simcons", - .write = simcons_write, - .read = simcons_read, - .device = simcons_device, - .flags = CON_PRINTBUFFER, - .index = -1, -}; - -/* Higher level TTY interface. */ - -int simcons_tty_open (struct tty_struct *tty, struct file *filp) -{ - return 0; -} - -int simcons_tty_write (struct tty_struct *tty, - const unsigned char *buf, int count) -{ - return V850_SIM_SYSCALL (write, 1, buf, count); -} - -int simcons_tty_write_room (struct tty_struct *tty) -{ - /* Completely arbitrary. */ - return 0x100000; -} - -int simcons_tty_chars_in_buffer (struct tty_struct *tty) -{ - /* We have no buffer. */ - return 0; -} - -static const struct tty_operations ops = { - .open = simcons_tty_open, - .write = simcons_tty_write, - .write_room = simcons_tty_write_room, - .chars_in_buffer = simcons_tty_chars_in_buffer, -}; - -int __init simcons_tty_init (void) -{ - struct tty_driver *driver = alloc_tty_driver(1); - int err; - if (!driver) - return -ENOMEM; - driver->name = "simcons"; - driver->major = TTY_MAJOR; - driver->minor_start = 64; - driver->type = TTY_DRIVER_TYPE_SYSCONS; - driver->init_termios = tty_std_termios; - tty_set_operations(driver, &ops); - err = tty_register_driver(driver); - if (err) { - put_tty_driver(driver); - return err; - } - tty_driver = driver; - return 0; -} -/* We use `late_initcall' instead of just `__initcall' as a workaround for - the fact that (1) simcons_tty_init can't be called before tty_init, - (2) tty_init is called via `module_init', (3) if statically linked, - module_init == device_init, and (4) there's no ordering of init lists. - We can do this easily because simcons is always statically linked, but - other tty drivers that depend on tty_init and which must use - `module_init' to declare their init routines are likely to be broken. */ -late_initcall(simcons_tty_init); - -/* Poll for input on the console, and if there's any, deliver it to the - tty driver. */ -void simcons_poll_tty (struct tty_struct *tty) -{ - char buf[32]; /* Not the nicest way to do it but I need it correct first */ - int flip = 0, send_break = 0; - struct pollfd pfd; - pfd.fd = 0; - pfd.events = POLLIN; - - if (V850_SIM_SYSCALL (poll, &pfd, 1, 0) > 0) { - if (pfd.revents & POLLIN) { - /* Real block hardware knows the transfer size before - transfer so the new tty buffering doesn't try to handle - this rather weird simulator specific case well */ - int rd = V850_SIM_SYSCALL (read, 0, buf, 32); - if (rd > 0) { - tty_insert_flip_string(tty, buf, rd); - flip = 1; - } else - send_break = 1; - } else if (pfd.revents & POLLERR) - send_break = 1; - } - - if (send_break) { - tty_insert_flip_char (tty, 0, TTY_BREAK); - flip = 1; - } - - if (flip) - tty_schedule_flip (tty); -} - -void simcons_poll_ttys (void) -{ - if (tty_driver && tty_driver->ttys[0]) - simcons_poll_tty (tty_driver->ttys[0]); -} - -void simcons_setup (void) -{ - V850_SIM_SYSCALL (make_raw, 0); - register_console (&simcons); - printk (KERN_INFO "Console: GDB V850E simulator stdio\n"); -} diff --git a/arch/v850/kernel/syscalls.c b/arch/v850/kernel/syscalls.c deleted file mode 100644 index 1a83daf8e24f..000000000000 --- a/arch/v850/kernel/syscalls.c +++ /dev/null @@ -1,196 +0,0 @@ -/* - * arch/v850/kernel/syscalls.c -- Various system-call definitions not - * defined in machine-independent code - * - * Copyright (C) 2001,02 NEC Corporation - * Copyright (C) 2001,02 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * This file was derived the ppc version, arch/ppc/kernel/syscalls.c - * ... which was derived from "arch/i386/kernel/sys_i386.c" by Gary Thomas; - * modified by Cort Dougan (cort@cs.nmt.edu) - * and Paul Mackerras (paulus@cs.anu.edu.au). - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include - -/* - * sys_ipc() is the de-multiplexer for the SysV IPC calls.. - * - * This is really horribly ugly. - */ -int -sys_ipc (uint call, int first, int second, int third, void *ptr, long fifth) -{ - int version, ret; - - version = call >> 16; /* hack for backward compatibility */ - call &= 0xffff; - - ret = -EINVAL; - switch (call) { - case SEMOP: - ret = sys_semop (first, (struct sembuf *)ptr, second); - break; - case SEMGET: - ret = sys_semget (first, second, third); - break; - case SEMCTL: - { - union semun fourth; - - if (!ptr) - break; - if ((ret = access_ok(VERIFY_READ, ptr, sizeof(long)) ? 0 : -EFAULT) - || (ret = get_user(fourth.__pad, (void **)ptr))) - break; - ret = sys_semctl (first, second, third, fourth); - break; - } - case MSGSND: - ret = sys_msgsnd (first, (struct msgbuf *) ptr, second, third); - break; - case MSGRCV: - switch (version) { - case 0: { - struct ipc_kludge tmp; - - if (!ptr) - break; - if ((ret = access_ok(VERIFY_READ, ptr, sizeof(tmp)) ? 0 : -EFAULT) - || (ret = copy_from_user(&tmp, - (struct ipc_kludge *) ptr, - sizeof (tmp)))) - break; - ret = sys_msgrcv (first, tmp.msgp, second, tmp.msgtyp, - third); - break; - } - default: - ret = sys_msgrcv (first, (struct msgbuf *) ptr, - second, fifth, third); - break; - } - break; - case MSGGET: - ret = sys_msgget ((key_t) first, second); - break; - case MSGCTL: - ret = sys_msgctl (first, second, (struct msqid_ds *) ptr); - break; - case SHMAT: - switch (version) { - default: { - ulong raddr; - - if ((ret = access_ok(VERIFY_WRITE, (ulong*) third, - sizeof(ulong)) ? 0 : -EFAULT)) - break; - ret = do_shmat (first, (char *) ptr, second, &raddr); - if (ret) - break; - ret = put_user (raddr, (ulong *) third); - break; - } - case 1: /* iBCS2 emulator entry point */ - if (!segment_eq(get_fs(), get_ds())) - break; - ret = do_shmat (first, (char *) ptr, second, - (ulong *) third); - break; - } - break; - case SHMDT: - ret = sys_shmdt ((char *)ptr); - break; - case SHMGET: - ret = sys_shmget (first, second, third); - break; - case SHMCTL: - ret = sys_shmctl (first, second, (struct shmid_ds *) ptr); - break; - } - - return ret; -} - -static inline unsigned long -do_mmap2 (unsigned long addr, size_t len, - unsigned long prot, unsigned long flags, - unsigned long fd, unsigned long pgoff) -{ - struct file * file = NULL; - int ret = -EBADF; - - flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE); - if (! (flags & MAP_ANONYMOUS)) { - if (!(file = fget (fd))) - goto out; - } - - down_write (¤t->mm->mmap_sem); - ret = do_mmap_pgoff (file, addr, len, prot, flags, pgoff); - up_write (¤t->mm->mmap_sem); - if (file) - fput (file); -out: - return ret; -} - -unsigned long sys_mmap2 (unsigned long addr, size_t len, - unsigned long prot, unsigned long flags, - unsigned long fd, unsigned long pgoff) -{ - return do_mmap2 (addr, len, prot, flags, fd, pgoff); -} - -unsigned long sys_mmap (unsigned long addr, size_t len, - unsigned long prot, unsigned long flags, - unsigned long fd, off_t offset) -{ - int err = -EINVAL; - - if (offset & ~PAGE_MASK) - goto out; - - err = do_mmap2 (addr, len, prot, flags, fd, offset >> PAGE_SHIFT); -out: - return err; -} - -/* - * Do a system call from kernel instead of calling sys_execve so we - * end up with proper pt_regs. - */ -int kernel_execve(const char *filename, char *const argv[], char *const envp[]) -{ - register char *__a __asm__ ("r6") = filename; - register void *__b __asm__ ("r7") = argv; - register void *__c __asm__ ("r8") = envp; - register unsigned long __syscall __asm__ ("r12") = __NR_execve; - register unsigned long __ret __asm__ ("r10"); - __asm__ __volatile__ ("trap 0" - : "=r" (__ret), "=r" (__syscall) - : "1" (__syscall), "r" (__a), "r" (__b), "r" (__c) - : "r1", "r5", "r11", "r13", "r14", - "r15", "r16", "r17", "r18", "r19"); - return __ret; -} diff --git a/arch/v850/kernel/teg.c b/arch/v850/kernel/teg.c deleted file mode 100644 index 699248f92aae..000000000000 --- a/arch/v850/kernel/teg.c +++ /dev/null @@ -1,62 +0,0 @@ -/* - * arch/v850/kernel/teg.c -- NB85E-TEG cpu chip - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include - -#include "mach.h" - -void __init mach_sched_init (struct irqaction *timer_action) -{ - /* Select timer interrupt instead of external pin. */ - TEG_ISS |= 0x1; - /* Start hardware timer. */ - v850e_timer_d_configure (0, HZ); - /* Install timer interrupt handler. */ - setup_irq (IRQ_INTCMD(0), timer_action); -} - -static struct v850e_intc_irq_init irq_inits[] = { - { "IRQ", 0, NUM_CPU_IRQS, 1, 7 }, - { "CMD", IRQ_INTCMD(0), IRQ_INTCMD_NUM, 1, 5 }, - { "SER", IRQ_INTSER(0), IRQ_INTSER_NUM, 1, 3 }, - { "SR", IRQ_INTSR(0), IRQ_INTSR_NUM, 1, 4 }, - { "ST", IRQ_INTST(0), IRQ_INTST_NUM, 1, 5 }, - { 0 } -}; -#define NUM_IRQ_INITS (ARRAY_SIZE(irq_inits) - 1) - -static struct hw_interrupt_type hw_itypes[NUM_IRQ_INITS]; - -/* Initialize MA chip interrupts. */ -void __init teg_init_irqs (void) -{ - v850e_intc_init_irq_types (irq_inits, hw_itypes); -} - -/* Called before configuring an on-chip UART. */ -void teg_uart_pre_configure (unsigned chan, unsigned cflags, unsigned baud) -{ - /* Enable UART I/O pins instead of external interrupt pins, and - UART interrupts instead of external pin interrupts. */ - TEG_ISS |= 0x4E; -} diff --git a/arch/v850/kernel/time.c b/arch/v850/kernel/time.c deleted file mode 100644 index d810c93fe665..000000000000 --- a/arch/v850/kernel/time.c +++ /dev/null @@ -1,106 +0,0 @@ -/* - * linux/arch/v850/kernel/time.c -- Arch-dependent timer functions - * - * Copyright (C) 1991, 1992, 1995, 2001, 2002 Linus Torvalds - * - * This file contains the v850-specific time handling details. - * Most of the stuff is located in the machine specific files. - * - * 1997-09-10 Updated NTP code according to technical memorandum Jan '96 - * "A Kernel Model for Precision Timekeeping" by Dave Mills - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include - -#include "mach.h" - -#define TICK_SIZE (tick_nsec / 1000) - -/* - * timer_interrupt() needs to keep up the real-time clock, - * as well as call the "do_timer()" routine every clocktick - */ -static irqreturn_t timer_interrupt (int irq, void *dummy, struct pt_regs *regs) -{ -#if 0 - /* last time the cmos clock got updated */ - static long last_rtc_update=0; -#endif - - /* may need to kick the hardware timer */ - if (mach_tick) - mach_tick (); - - do_timer (1); -#ifndef CONFIG_SMP - update_process_times(user_mode(regs)); -#endif - profile_tick(CPU_PROFILING, regs); -#if 0 - /* - * If we have an externally synchronized Linux clock, then update - * CMOS clock accordingly every ~11 minutes. Set_rtc_mmss() has to be - * called as close as possible to 500 ms before the new second starts. - */ - if (ntp_synced() && - xtime.tv_sec > last_rtc_update + 660 && - (xtime.tv_nsec / 1000) >= 500000 - ((unsigned) TICK_SIZE) / 2 && - (xtime.tv_nsec / 1000) <= 500000 + ((unsigned) TICK_SIZE) / 2) { - if (set_rtc_mmss (xtime.tv_sec) == 0) - last_rtc_update = xtime.tv_sec; - else - last_rtc_update = xtime.tv_sec - 600; /* do it again in 60 s */ - } -#ifdef CONFIG_HEARTBEAT - /* use power LED as a heartbeat instead -- much more useful - for debugging -- based on the version for PReP by Cort */ - /* acts like an actual heart beat -- ie thump-thump-pause... */ - if (mach_heartbeat) { - static unsigned cnt = 0, period = 0, dist = 0; - - if (cnt == 0 || cnt == dist) - mach_heartbeat ( 1 ); - else if (cnt == 7 || cnt == dist+7) - mach_heartbeat ( 0 ); - - if (++cnt > period) { - cnt = 0; - /* The hyperbolic function below modifies the heartbeat period - * length in dependency of the current (5min) load. It goes - * through the points f(0)=126, f(1)=86, f(5)=51, - * f(inf)->30. */ - period = ((672< -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include -#include - - -extern void *trap_table; -EXPORT_SYMBOL (trap_table); - -/* platform dependent support */ -EXPORT_SYMBOL (kernel_thread); -EXPORT_SYMBOL (__bug); - -/* Networking helper routines. */ -EXPORT_SYMBOL (csum_partial_copy_nocheck); -EXPORT_SYMBOL (csum_partial_copy_from_user); -EXPORT_SYMBOL (ip_compute_csum); -EXPORT_SYMBOL (ip_fast_csum); - -/* string / mem functions */ -EXPORT_SYMBOL (memset); -EXPORT_SYMBOL (memcpy); -EXPORT_SYMBOL (memmove); - -/* - * libgcc functions - functions that are used internally by the - * compiler... (prototypes are not correct though, but that - * doesn't really matter since they're not versioned). - */ -extern void __ashldi3 (void); -extern void __ashrdi3 (void); -extern void __lshrdi3 (void); -extern void __muldi3 (void); -extern void __negdi2 (void); - -EXPORT_SYMBOL (__ashldi3); -EXPORT_SYMBOL (__ashrdi3); -EXPORT_SYMBOL (__lshrdi3); -EXPORT_SYMBOL (__muldi3); -EXPORT_SYMBOL (__negdi2); diff --git a/arch/v850/kernel/v850e2_cache.c b/arch/v850/kernel/v850e2_cache.c deleted file mode 100644 index 4570312c689c..000000000000 --- a/arch/v850/kernel/v850e2_cache.c +++ /dev/null @@ -1,127 +0,0 @@ -/* - * arch/v850/kernel/v850e2_cache.c -- Cache control for V850E2 cache - * memories - * - * Copyright (C) 2003 NEC Electronics Corporation - * Copyright (C) 2003 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include - -#include - -/* Cache operations we can do. The encoding corresponds directly to the - value we need to write into the COPR register. */ -enum cache_op { - OP_SYNC_IF_DIRTY = V850E2_CACHE_COPR_CFC(0), /* 000 */ - OP_SYNC_IF_VALID = V850E2_CACHE_COPR_CFC(1), /* 001 */ - OP_SYNC_IF_VALID_AND_CLEAR = V850E2_CACHE_COPR_CFC(3), /* 011 */ - OP_WAY_CLEAR = V850E2_CACHE_COPR_CFC(4), /* 100 */ - OP_FILL = V850E2_CACHE_COPR_CFC(5), /* 101 */ - OP_CLEAR = V850E2_CACHE_COPR_CFC(6), /* 110 */ - OP_CREATE_DIRTY = V850E2_CACHE_COPR_CFC(7) /* 111 */ -}; - -/* Which cache to use. This encoding also corresponds directly to the - value we need to write into the COPR register. */ -enum cache { - ICACHE = 0, - DCACHE = V850E2_CACHE_COPR_LBSL -}; - -/* Returns ADDR rounded down to the beginning of its cache-line. */ -#define CACHE_LINE_ADDR(addr) \ - ((addr) & ~(V850E2_CACHE_LINE_SIZE - 1)) -/* Returns END_ADDR rounded up to the `limit' of its cache-line. */ -#define CACHE_LINE_END_ADDR(end_addr) \ - CACHE_LINE_ADDR(end_addr + (V850E2_CACHE_LINE_SIZE - 1)) - - -/* Low-level cache ops. */ - -/* Apply cache-op OP to all entries in CACHE. */ -static inline void cache_op_all (enum cache_op op, enum cache cache) -{ - int cmd = op | cache | V850E2_CACHE_COPR_WSLE | V850E2_CACHE_COPR_STRT; - - if (op != OP_WAY_CLEAR) { - /* The WAY_CLEAR operation does the whole way, but other - ops take begin-index and count params; we just indicate - the entire cache. */ - V850E2_CACHE_CADL = 0; - V850E2_CACHE_CADH = 0; - V850E2_CACHE_CCNT = V850E2_CACHE_WAY_SIZE - 1; - } - - V850E2_CACHE_COPR = cmd | V850E2_CACHE_COPR_WSL(0); /* way 0 */ - V850E2_CACHE_COPR = cmd | V850E2_CACHE_COPR_WSL(1); /* way 1 */ - V850E2_CACHE_COPR = cmd | V850E2_CACHE_COPR_WSL(2); /* way 2 */ - V850E2_CACHE_COPR = cmd | V850E2_CACHE_COPR_WSL(3); /* way 3 */ -} - -/* Apply cache-op OP to all entries in CACHE covering addresses ADDR - through ADDR+LEN. */ -static inline void cache_op_range (enum cache_op op, u32 addr, u32 len, - enum cache cache) -{ - u32 start = CACHE_LINE_ADDR (addr); - u32 end = CACHE_LINE_END_ADDR (addr + len); - u32 num_lines = (end - start) >> V850E2_CACHE_LINE_SIZE_BITS; - - V850E2_CACHE_CADL = start & 0xFFFF; - V850E2_CACHE_CADH = start >> 16; - V850E2_CACHE_CCNT = num_lines - 1; - - V850E2_CACHE_COPR = op | cache | V850E2_CACHE_COPR_STRT; -} - - -/* High-level ops. */ - -static void cache_exec_after_store_all (void) -{ - cache_op_all (OP_SYNC_IF_DIRTY, DCACHE); - cache_op_all (OP_WAY_CLEAR, ICACHE); -} - -static void cache_exec_after_store_range (u32 start, u32 len) -{ - cache_op_range (OP_SYNC_IF_DIRTY, start, len, DCACHE); - cache_op_range (OP_CLEAR, start, len, ICACHE); -} - - -/* Exported functions. */ - -void flush_icache (void) -{ - cache_exec_after_store_all (); -} - -void flush_icache_range (unsigned long start, unsigned long end) -{ - cache_exec_after_store_range (start, end - start); -} - -void flush_icache_page (struct vm_area_struct *vma, struct page *page) -{ - cache_exec_after_store_range (page_to_virt (page), PAGE_SIZE); -} - -void flush_icache_user_range (struct vm_area_struct *vma, struct page *page, - unsigned long addr, int len) -{ - cache_exec_after_store_range (addr, len); -} - -void flush_cache_sigtramp (unsigned long addr) -{ - /* For the exact size, see signal.c, but 16 bytes should be enough. */ - cache_exec_after_store_range (addr, 16); -} diff --git a/arch/v850/kernel/v850e_cache.c b/arch/v850/kernel/v850e_cache.c deleted file mode 100644 index ea3e51cfb259..000000000000 --- a/arch/v850/kernel/v850e_cache.c +++ /dev/null @@ -1,174 +0,0 @@ -/* - * arch/v850/kernel/v850e_cache.c -- Cache control for V850E cache memories - * - * Copyright (C) 2003 NEC Electronics Corporation - * Copyright (C) 2003 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -/* This file implements cache control for the rather simple cache used on - some V850E CPUs, specifically the NB85E/TEG CPU-core and the V850E/ME2 - CPU. V850E2 processors have their own (better) cache - implementation. */ - -#include -#include -#include - -#define WAIT_UNTIL_CLEAR(value) while (value) {} - -/* Set caching params via the BHC and DCC registers. */ -void v850e_cache_enable (u16 bhc, u16 icc, u16 dcc) -{ - unsigned long *r0_ram = (unsigned long *)R0_RAM_ADDR; - register u16 bhc_val asm ("r6") = bhc; - - /* Read the instruction cache control register (ICC) and confirm - that bits 0 and 1 (TCLR0, TCLR1) are all cleared. */ - WAIT_UNTIL_CLEAR (V850E_CACHE_ICC & 0x3); - V850E_CACHE_ICC = icc; - -#ifdef V850E_CACHE_DCC - /* Configure data-cache. */ - V850E_CACHE_DCC = dcc; -#endif /* V850E_CACHE_DCC */ - - /* Configure caching for various memory regions by writing the BHC - register. The documentation says that an instruction _cannot_ - enable/disable caching for the memory region in which the - instruction itself exists; to work around this, we store - appropriate instructions into the on-chip RAM area (which is never - cached), and briefly jump there to do the work. */ -#ifdef V850E_CACHE_WRITE_IBS - *r0_ram++ = 0xf0720760; /* st.h r0, 0xfffff072[r0] */ -#endif - *r0_ram++ = 0xf06a3760; /* st.h r6, 0xfffff06a[r0] */ - *r0_ram = 0x5640006b; /* jmp [r11] */ - - asm ("mov hilo(1f), r11; jmp [%1]; 1:;" - :: "r" (bhc_val), "r" (R0_RAM_ADDR) : "r11"); -} - -static void clear_icache (void) -{ - /* 1. Read the instruction cache control register (ICC) and confirm - that bits 0 and 1 (TCLR0, TCLR1) are all cleared. */ - WAIT_UNTIL_CLEAR (V850E_CACHE_ICC & 0x3); - - /* 2. Read the ICC register and confirm that bit 12 (LOCK0) is - cleared. Bit 13 of the ICC register is always cleared. */ - WAIT_UNTIL_CLEAR (V850E_CACHE_ICC & 0x1000); - - /* 3. Set the TCLR0 and TCLR1 bits of the ICC register as follows, - when clearing way 0 and way 1 at the same time: - (a) Set the TCLR0 and TCLR1 bits. - (b) Read the TCLR0 and TCLR1 bits to confirm that these bits - are cleared. - (c) Perform (a) and (b) above again. */ - V850E_CACHE_ICC |= 0x3; - WAIT_UNTIL_CLEAR (V850E_CACHE_ICC & 0x3); - -#ifdef V850E_CACHE_REPEAT_ICC_WRITE - /* Do it again. */ - V850E_CACHE_ICC |= 0x3; - WAIT_UNTIL_CLEAR (V850E_CACHE_ICC & 0x3); -#endif -} - -#ifdef V850E_CACHE_DCC -/* Flush or clear (or both) the data cache, depending on the value of FLAGS; - the procedure is the same for both, just the control bits used differ (and - both may be performed simultaneously). */ -static void dcache_op (unsigned short flags) -{ - /* 1. Read the data cache control register (DCC) and confirm that bits - 0, 1, 4, and 5 (DC00, DC01, DC04, DC05) are all cleared. */ - WAIT_UNTIL_CLEAR (V850E_CACHE_DCC & 0x33); - - /* 2. Clear DCC register bit 12 (DC12), bit 13 (DC13), or both - depending on the way for which tags are to be cleared. */ - V850E_CACHE_DCC &= ~0xC000; - - /* 3. Set DCC register bit 0 (DC00), bit 1 (DC01) or both depending on - the way for which tags are to be cleared. - ... - Set DCC register bit 4 (DC04), bit 5 (DC05), or both depending - on the way to be data flushed. */ - V850E_CACHE_DCC |= flags; - - /* 4. Read DCC register bit DC00, DC01 [DC04, DC05], or both depending - on the way for which tags were cleared [flushed] and confirm - that that bit is cleared. */ - WAIT_UNTIL_CLEAR (V850E_CACHE_DCC & flags); -} -#endif /* V850E_CACHE_DCC */ - -/* Flushes the contents of the dcache to memory. */ -static inline void flush_dcache (void) -{ -#ifdef V850E_CACHE_DCC - /* We only need to do something if in write-back mode. */ - if (V850E_CACHE_DCC & 0x0400) - dcache_op (0x30); -#endif /* V850E_CACHE_DCC */ -} - -/* Flushes the contents of the dcache to memory, and then clears it. */ -static inline void clear_dcache (void) -{ -#ifdef V850E_CACHE_DCC - /* We only need to do something if the dcache is enabled. */ - if (V850E_CACHE_DCC & 0x0C00) - dcache_op (0x33); -#endif /* V850E_CACHE_DCC */ -} - -/* Clears the dcache without flushing to memory first. */ -static inline void clear_dcache_no_flush (void) -{ -#ifdef V850E_CACHE_DCC - /* We only need to do something if the dcache is enabled. */ - if (V850E_CACHE_DCC & 0x0C00) - dcache_op (0x3); -#endif /* V850E_CACHE_DCC */ -} - -static inline void cache_exec_after_store (void) -{ - flush_dcache (); - clear_icache (); -} - - -/* Exported functions. */ - -void flush_icache (void) -{ - cache_exec_after_store (); -} - -void flush_icache_range (unsigned long start, unsigned long end) -{ - cache_exec_after_store (); -} - -void flush_icache_page (struct vm_area_struct *vma, struct page *page) -{ - cache_exec_after_store (); -} - -void flush_icache_user_range (struct vm_area_struct *vma, struct page *page, - unsigned long adr, int len) -{ - cache_exec_after_store (); -} - -void flush_cache_sigtramp (unsigned long addr) -{ - cache_exec_after_store (); -} diff --git a/arch/v850/kernel/v850e_intc.c b/arch/v850/kernel/v850e_intc.c deleted file mode 100644 index 8d39a52ee6d1..000000000000 --- a/arch/v850/kernel/v850e_intc.c +++ /dev/null @@ -1,104 +0,0 @@ -/* - * arch/v850/kernel/v850e_intc.c -- V850E interrupt controller (INTC) - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include -#include - -#include - -static void irq_nop (unsigned irq) { } - -static unsigned v850e_intc_irq_startup (unsigned irq) -{ - v850e_intc_clear_pending_irq (irq); - v850e_intc_enable_irq (irq); - return 0; -} - -static void v850e_intc_end_irq (unsigned irq) -{ - unsigned long psw, temp; - - /* Clear the highest-level bit in the In-service priority register - (ISPR), to allow this interrupt (or another of the same or - lesser priority) to happen again. - - The `reti' instruction normally does this automatically when the - PSW bits EP and NP are zero, but we can't always rely on reti - being used consistently to return after an interrupt (another - process can be scheduled, for instance, which can delay the - associated reti for a long time, or this process may be being - single-stepped, which uses the `dbret' instruction to return - from the kernel). - - We also set the PSW EP bit, which prevents reti from also - trying to modify the ISPR itself. */ - - /* Get PSW and disable interrupts. */ - asm volatile ("stsr psw, %0; di" : "=r" (psw)); - /* We don't want to do anything for NMIs (they don't use the ISPR). */ - if (! (psw & 0xC0)) { - /* Transition to `trap' state, so that an eventual real - reti instruction won't modify the ISPR. */ - psw |= 0x40; - /* Fake an interrupt return, which automatically clears the - appropriate bit in the ISPR. */ - asm volatile ("mov hilo(1f), %0;" - "ldsr %0, eipc; ldsr %1, eipsw;" - "reti;" - "1:" - : "=&r" (temp) : "r" (psw)); - } -} - -/* Initialize HW_IRQ_TYPES for INTC-controlled irqs described in array - INITS (which is terminated by an entry with the name field == 0). */ -void __init v850e_intc_init_irq_types (struct v850e_intc_irq_init *inits, - struct hw_interrupt_type *hw_irq_types) -{ - struct v850e_intc_irq_init *init; - for (init = inits; init->name; init++) { - unsigned i; - struct hw_interrupt_type *hwit = hw_irq_types++; - - hwit->typename = init->name; - - hwit->startup = v850e_intc_irq_startup; - hwit->shutdown = v850e_intc_disable_irq; - hwit->enable = v850e_intc_enable_irq; - hwit->disable = v850e_intc_disable_irq; - hwit->ack = irq_nop; - hwit->end = v850e_intc_end_irq; - - /* Initialize kernel IRQ infrastructure for this interrupt. */ - init_irq_handlers(init->base, init->num, init->interval, hwit); - - /* Set the interrupt priorities. */ - for (i = 0; i < init->num; i++) { - unsigned irq = init->base + i * init->interval; - - /* If the interrupt is currently enabled (all - interrupts are initially disabled), then - assume whoever enabled it has set things up - properly, and avoid messing with it. */ - if (! v850e_intc_irq_enabled (irq)) - /* This write also (1) disables the - interrupt, and (2) clears any pending - interrupts. */ - V850E_INTC_IC (irq) - = (V850E_INTC_IC_PR (init->priority) - | V850E_INTC_IC_MK); - } - } -} diff --git a/arch/v850/kernel/v850e_timer_d.c b/arch/v850/kernel/v850e_timer_d.c deleted file mode 100644 index d2a4ece2574c..000000000000 --- a/arch/v850/kernel/v850e_timer_d.c +++ /dev/null @@ -1,54 +0,0 @@ -/* - * include/asm-v850/v850e_timer_d.c -- `Timer D' component often used - * with V850E CPUs - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include - -#include -#include - -/* Start interval timer TIMER (0-3). The timer will issue the - corresponding INTCMD interrupt RATE times per second. - This function does not enable the interrupt. */ -void v850e_timer_d_configure (unsigned timer, unsigned rate) -{ - unsigned divlog2, count; - - /* Calculate params for timer. */ - if (! calc_counter_params ( - V850E_TIMER_D_BASE_FREQ, rate, - V850E_TIMER_D_TMCD_CS_MIN, V850E_TIMER_D_TMCD_CS_MAX, 16, - &divlog2, &count)) - printk (KERN_WARNING - "Cannot find interval timer %d setting suitable" - " for rate of %dHz.\n" - "Using rate of %dHz instead.\n", - timer, rate, - (V850E_TIMER_D_BASE_FREQ >> divlog2) >> 16); - - /* Do the actual hardware timer initialization: */ - - /* Enable timer. */ - V850E_TIMER_D_TMCD(timer) = V850E_TIMER_D_TMCD_CAE; - /* Set clock divider. */ - V850E_TIMER_D_TMCD(timer) - = V850E_TIMER_D_TMCD_CAE - | V850E_TIMER_D_TMCD_CS(divlog2); - /* Set timer compare register. */ - V850E_TIMER_D_CMD(timer) = count; - /* Start counting. */ - V850E_TIMER_D_TMCD(timer) - = V850E_TIMER_D_TMCD_CAE - | V850E_TIMER_D_TMCD_CS(divlog2) - | V850E_TIMER_D_TMCD_CE; -} diff --git a/arch/v850/kernel/v850e_utils.c b/arch/v850/kernel/v850e_utils.c deleted file mode 100644 index e6807ef8dee6..000000000000 --- a/arch/v850/kernel/v850e_utils.c +++ /dev/null @@ -1,62 +0,0 @@ -/* - * include/asm-v850/v850e_utils.h -- Utility functions associated with - * V850E CPUs - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include - -/* Calculate counter clock-divider and count values to attain the - desired frequency RATE from the base frequency BASE_FREQ. The - counter is expected to have a clock-divider, which can divide the - system cpu clock by a power of two value from MIN_DIVLOG2 to - MAX_DIV_LOG2, and a word-size of COUNTER_SIZE bits (the counter - counts up and resets whenever it's equal to the compare register, - generating an interrupt or whatever when it does so). The returned - values are: *DIVLOG2 -- log2 of the desired clock divider and *COUNT - -- the counter compare value to use. Returns true if it was possible - to find a reasonable value, otherwise false (and the other return - values will be set to be as good as possible). */ -int calc_counter_params (unsigned long base_freq, - unsigned long rate, - unsigned min_divlog2, unsigned max_divlog2, - unsigned counter_size, - unsigned *divlog2, unsigned *count) -{ - unsigned _divlog2; - int ok = 0; - - /* Find the lowest clock divider setting that can represent RATE. */ - for (_divlog2 = min_divlog2; _divlog2 <= max_divlog2; _divlog2++) { - /* Minimum interrupt rate possible using this divider. */ - unsigned min_int_rate - = (base_freq >> _divlog2) >> counter_size; - - if (min_int_rate <= rate) { - /* This setting is the highest resolution - setting that's slow enough enough to attain - RATE interrupts per second, so use it. */ - ok = 1; - break; - } - } - - if (_divlog2 > max_divlog2) - /* Can't find correct setting. */ - _divlog2 = max_divlog2; - - if (divlog2) - *divlog2 = _divlog2; - if (count) - *count = ((base_freq >> _divlog2) + rate/2) / rate; - - return ok; -} diff --git a/arch/v850/kernel/vmlinux.lds.S b/arch/v850/kernel/vmlinux.lds.S deleted file mode 100644 index d08cd1d27f27..000000000000 --- a/arch/v850/kernel/vmlinux.lds.S +++ /dev/null @@ -1,306 +0,0 @@ -/* - * arch/v850/vmlinux.lds.S -- kernel linker script for v850 platforms - * - * Copyright (C) 2002,03,04,05 NEC Electronics Corporation - * Copyright (C) 2002,03,04,05 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - - -#define VMLINUX_SYMBOL(_sym_) _##_sym_ -#include - -/* For most platforms, this will define useful things like RAM addr/size. */ -#include - - -/* The following macros contain the usual definitions for various data areas. - The prefix `RAMK_' is used to indicate macros suitable for kernels loaded - into RAM, and similarly `ROMK_' for ROM-resident kernels. Note that all - symbols are prefixed with an extra `_' for compatibility with the v850 - toolchain. */ - - -/* Interrupt vectors. */ -#define INTV_CONTENTS \ - . = ALIGN (0x10) ; \ - __intv_start = . ; \ - *(.intv.reset) /* Reset vector */ \ - . = __intv_start + 0x10 ; \ - *(.intv.common) /* Vectors common to all v850e proc */\ - . = __intv_start + 0x80 ; \ - *(.intv.mach) /* Machine-specific int. vectors. */ \ - __intv_end = . ; - -#define RODATA_CONTENTS \ - . = ALIGN (16) ; \ - *(.rodata) *(.rodata.*) \ - *(__vermagic) /* Kernel version magic */ \ - *(.rodata1) \ - /* PCI quirks */ \ - ___start_pci_fixups_early = . ; \ - *(.pci_fixup_early) \ - ___end_pci_fixups_early = . ; \ - ___start_pci_fixups_header = . ; \ - *(.pci_fixup_header) \ - ___end_pci_fixups_header = . ; \ - ___start_pci_fixups_final = . ; \ - *(.pci_fixup_final) \ - ___end_pci_fixups_final = . ; \ - ___start_pci_fixups_enable = . ; \ - *(.pci_fixup_enable) \ - ___end_pci_fixups_enable = . ; \ - /* Kernel symbol table: Normal symbols */ \ - ___start___ksymtab = .; \ - *(__ksymtab) \ - ___stop___ksymtab = .; \ - /* Kernel symbol table: GPL-only symbols */ \ - ___start___ksymtab_gpl = .; \ - *(__ksymtab_gpl) \ - ___stop___ksymtab_gpl = .; \ - /* Kernel symbol table: GPL-future symbols */ \ - ___start___ksymtab_gpl_future = .; \ - *(__ksymtab_gpl_future) \ - ___stop___ksymtab_gpl_future = .; \ - /* Kernel symbol table: strings */ \ - *(__ksymtab_strings) \ - /* Kernel symbol table: Normal symbols */ \ - ___start___kcrctab = .; \ - *(__kcrctab) \ - ___stop___kcrctab = .; \ - /* Kernel symbol table: GPL-only symbols */ \ - ___start___kcrctab_gpl = .; \ - *(__kcrctab_gpl) \ - ___stop___kcrctab_gpl = .; \ - /* Kernel symbol table: GPL-future symbols */ \ - ___start___kcrctab_gpl_future = .; \ - *(__kcrctab_gpl_future) \ - ___stop___kcrctab_gpl_future = .; \ - /* Built-in module parameters */ \ - . = ALIGN (4) ; \ - ___start___param = .; \ - *(__param) \ - ___stop___param = .; - - -/* Kernel text segment, and some constant data areas. */ -#define TEXT_CONTENTS \ - _text = .; \ - __stext = . ; \ - TEXT_TEXT \ - SCHED_TEXT \ - *(.exit.text) /* 2.5 convention */ \ - *(.text.exit) /* 2.4 convention */ \ - *(.text.lock) \ - *(.exitcall.exit) \ - __real_etext = . ; /* There may be data after here. */ \ - RODATA_CONTENTS \ - . = ALIGN (4) ; \ - *(.call_table_data) \ - *(.call_table_text) \ - . = ALIGN (16) ; /* Exception table. */ \ - ___start___ex_table = . ; \ - *(__ex_table) \ - ___stop___ex_table = . ; \ - . = ALIGN (4) ; \ - __etext = . ; - -/* Kernel data segment. */ -#define DATA_CONTENTS \ - __sdata = . ; \ - DATA_DATA \ - EXIT_DATA /* 2.5 convention */ \ - *(.data.exit) /* 2.4 convention */ \ - . = ALIGN (16) ; \ - *(.data.cacheline_aligned) \ - . = ALIGN (0x2000) ; \ - *(.data.init_task) \ - . = ALIGN (0x2000) ; \ - __edata = . ; - -/* Kernel BSS segment. */ -#define BSS_CONTENTS \ - __sbss = . ; \ - *(.bss) \ - *(COMMON) \ - . = ALIGN (4) ; \ - __init_stack_end = . ; \ - __ebss = . ; - -/* `initcall' tables. */ -#define INITCALL_CONTENTS \ - . = ALIGN (16) ; \ - ___setup_start = . ; \ - *(.init.setup) /* 2.5 convention */ \ - *(.setup.init) /* 2.4 convention */ \ - ___setup_end = . ; \ - ___initcall_start = . ; \ - *(.initcall.init) \ - INITCALLS \ - . = ALIGN (4) ; \ - ___initcall_end = . ; \ - ___con_initcall_start = .; \ - *(.con_initcall.init) \ - ___con_initcall_end = .; - -/* Contents of `init' section for a kernel that's loaded into RAM. */ -#define RAMK_INIT_CONTENTS \ - RAMK_INIT_CONTENTS_NO_END \ - __init_end = . ; -/* Same as RAMK_INIT_CONTENTS, but doesn't define the `__init_end' symbol. */ -#define RAMK_INIT_CONTENTS_NO_END \ - . = ALIGN (4096) ; \ - __init_start = . ; \ - __sinittext = .; \ - INIT_TEXT /* 2.5 convention */ \ - __einittext = .; \ - INIT_DATA \ - *(.text.init) /* 2.4 convention */ \ - *(.data.init) \ - INITCALL_CONTENTS \ - INITRAMFS_CONTENTS - -/* The contents of `init' section for a ROM-resident kernel which - should go into RAM. */ -#define ROMK_INIT_RAM_CONTENTS \ - . = ALIGN (4096) ; \ - __init_start = . ; \ - INIT_DATA /* 2.5 convention */ \ - *(.data.init) /* 2.4 convention */ \ - __init_end = . ; \ - . = ALIGN (4096) ; - -/* The contents of `init' section for a ROM-resident kernel which - should go into ROM. */ -#define ROMK_INIT_ROM_CONTENTS \ - _sinittext = .; \ - INIT_TEXT /* 2.5 convention */ \ - _einittext = .; \ - *(.text.init) /* 2.4 convention */ \ - INITCALL_CONTENTS \ - INITRAMFS_CONTENTS - -/* A root filesystem image, for kernels with an embedded root filesystem. */ -#define ROOT_FS_CONTENTS \ - __root_fs_image_start = . ; \ - *(.root) \ - __root_fs_image_end = . ; - -#ifdef CONFIG_BLK_DEV_INITRD -/* The initramfs archive. */ -#define INITRAMFS_CONTENTS \ - . = ALIGN (4) ; \ - ___initramfs_start = . ; \ - *(.init.ramfs) \ - ___initramfs_end = . ; -#endif - -/* Where the initial bootmap (bitmap for the boot-time memory allocator) - should be place. */ -#define BOOTMAP_CONTENTS \ - . = ALIGN (4096) ; \ - __bootmap = . ; \ - . = . + 4096 ; /* enough for 128MB. */ - -/* The contents of a `typical' kram area for a kernel in RAM. */ -#define RAMK_KRAM_CONTENTS \ - __kram_start = . ; \ - TEXT_CONTENTS \ - DATA_CONTENTS \ - BSS_CONTENTS \ - RAMK_INIT_CONTENTS \ - __kram_end = . ; \ - BOOTMAP_CONTENTS - - -/* Define output sections normally used for a ROM-resident kernel. - ROM and RAM should be appropriate memory areas to use for kernel - ROM and RAM data. This assumes that ROM starts at 0 (and thus can - hold the interrupt vectors). */ -#define ROMK_SECTIONS(ROM, RAM) \ - .rom : { \ - INTV_CONTENTS \ - TEXT_CONTENTS \ - ROMK_INIT_ROM_CONTENTS \ - ROOT_FS_CONTENTS \ - } > ROM \ - \ - __rom_copy_src_start = . ; \ - \ - .data : { \ - __kram_start = . ; \ - __rom_copy_dst_start = . ; \ - DATA_CONTENTS \ - ROMK_INIT_RAM_CONTENTS \ - __rom_copy_dst_end = . ; \ - } > RAM AT> ROM \ - \ - .bss ALIGN (4) : { \ - BSS_CONTENTS \ - __kram_end = . ; \ - BOOTMAP_CONTENTS \ - } > RAM - - -/* The 32-bit variable `jiffies' is just the lower 32-bits of `jiffies_64'. */ -_jiffies = _jiffies_64 ; - - -/* Include an appropriate platform-dependent linker-script (which - usually should use the above macros to do most of the work). */ - -#ifdef CONFIG_V850E_SIM -# include "sim.ld" -#endif - -#ifdef CONFIG_V850E2_SIM85E2 -# include "sim85e2.ld" -#endif - -#ifdef CONFIG_V850E2_FPGA85E2C -# include "fpga85e2c.ld" -#endif - -#ifdef CONFIG_V850E2_ANNA -# ifdef CONFIG_ROM_KERNEL -# include "anna-rom.ld" -# else -# include "anna.ld" -# endif -#endif - -#ifdef CONFIG_V850E_AS85EP1 -# ifdef CONFIG_ROM_KERNEL -# include "as85ep1-rom.ld" -# else -# include "as85ep1.ld" -# endif -#endif - -#ifdef CONFIG_RTE_CB_MA1 -# ifdef CONFIG_ROM_KERNEL -# include "rte_ma1_cb-rom.ld" -# else -# include "rte_ma1_cb.ld" -# endif -#endif - -#ifdef CONFIG_RTE_CB_NB85E -# ifdef CONFIG_ROM_KERNEL -# include "rte_nb85e_cb-rom.ld" -# elif defined(CONFIG_RTE_CB_MULTI) -# include "rte_nb85e_cb-multi.ld" -# else -# include "rte_nb85e_cb.ld" -# endif -#endif - -#ifdef CONFIG_RTE_CB_ME2 -# include "rte_me2_cb.ld" -#endif - diff --git a/arch/v850/lib/Makefile b/arch/v850/lib/Makefile deleted file mode 100644 index 1c78b728a117..000000000000 --- a/arch/v850/lib/Makefile +++ /dev/null @@ -1,6 +0,0 @@ -# -# arch/v850/lib/Makefile -# - -lib-y = ashrdi3.o ashldi3.o lshrdi3.o muldi3.o negdi2.o \ - checksum.o memcpy.o memset.o diff --git a/arch/v850/lib/ashldi3.c b/arch/v850/lib/ashldi3.c deleted file mode 100644 index 9e792d53f0e4..000000000000 --- a/arch/v850/lib/ashldi3.c +++ /dev/null @@ -1,62 +0,0 @@ -/* ashldi3.c extracted from gcc-2.95.2/libgcc2.c which is: */ -/* Copyright (C) 1989, 92-98, 1999 Free Software Foundation, Inc. - -This file is part of GNU CC. - -GNU CC is free software; you can redistribute it and/or modify -it under the terms of the GNU General Public License as published by -the Free Software Foundation; either version 2, or (at your option) -any later version. - -GNU CC is distributed in the hope that it will be useful, -but WITHOUT ANY WARRANTY; without even the implied warranty of -MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -GNU General Public License for more details. - -You should have received a copy of the GNU General Public License -along with GNU CC; see the file COPYING. If not, write to -the Free Software Foundation, 59 Temple Place - Suite 330, -Boston, MA 02111-1307, USA. */ - -#define BITS_PER_UNIT 8 - -typedef int SItype __attribute__ ((mode (SI))); -typedef unsigned int USItype __attribute__ ((mode (SI))); -typedef int DItype __attribute__ ((mode (DI))); -typedef int word_type __attribute__ ((mode (__word__))); - -struct DIstruct {SItype high, low;}; - -typedef union -{ - struct DIstruct s; - DItype ll; -} DIunion; - -DItype -__ashldi3 (DItype u, word_type b) -{ - DIunion w; - word_type bm; - DIunion uu; - - if (b == 0) - return u; - - uu.ll = u; - - bm = (sizeof (SItype) * BITS_PER_UNIT) - b; - if (bm <= 0) - { - w.s.low = 0; - w.s.high = (USItype)uu.s.low << -bm; - } - else - { - USItype carries = (USItype)uu.s.low >> bm; - w.s.low = (USItype)uu.s.low << b; - w.s.high = ((USItype)uu.s.high << b) | carries; - } - - return w.ll; -} diff --git a/arch/v850/lib/ashrdi3.c b/arch/v850/lib/ashrdi3.c deleted file mode 100644 index 78efb65e315a..000000000000 --- a/arch/v850/lib/ashrdi3.c +++ /dev/null @@ -1,63 +0,0 @@ -/* ashrdi3.c extracted from gcc-2.7.2/libgcc2.c which is: */ -/* Copyright (C) 1989, 1992, 1993, 1994, 1995 Free Software Foundation, Inc. - -This file is part of GNU CC. - -GNU CC is free software; you can redistribute it and/or modify -it under the terms of the GNU General Public License as published by -the Free Software Foundation; either version 2, or (at your option) -any later version. - -GNU CC is distributed in the hope that it will be useful, -but WITHOUT ANY WARRANTY; without even the implied warranty of -MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -GNU General Public License for more details. - -You should have received a copy of the GNU General Public License -along with GNU CC; see the file COPYING. If not, write to -the Free Software Foundation, 59 Temple Place - Suite 330, -Boston, MA 02111-1307, USA. */ - -#define BITS_PER_UNIT 8 - -typedef int SItype __attribute__ ((mode (SI))); -typedef unsigned int USItype __attribute__ ((mode (SI))); -typedef int DItype __attribute__ ((mode (DI))); -typedef int word_type __attribute__ ((mode (__word__))); - -struct DIstruct {SItype high, low;}; - -typedef union -{ - struct DIstruct s; - DItype ll; -} DIunion; - -DItype -__ashrdi3 (DItype u, word_type b) -{ - DIunion w; - word_type bm; - DIunion uu; - - if (b == 0) - return u; - - uu.ll = u; - - bm = (sizeof (SItype) * BITS_PER_UNIT) - b; - if (bm <= 0) - { - /* w.s.high = 1..1 or 0..0 */ - w.s.high = uu.s.high >> (sizeof (SItype) * BITS_PER_UNIT - 1); - w.s.low = uu.s.high >> -bm; - } - else - { - USItype carries = (USItype)uu.s.high << bm; - w.s.high = uu.s.high >> b; - w.s.low = ((USItype)uu.s.low >> b) | carries; - } - - return w.ll; -} diff --git a/arch/v850/lib/checksum.c b/arch/v850/lib/checksum.c deleted file mode 100644 index 042158dfe17a..000000000000 --- a/arch/v850/lib/checksum.c +++ /dev/null @@ -1,155 +0,0 @@ -/* - * INET An implementation of the TCP/IP protocol suite for the LINUX - * operating system. INET is implemented using the BSD Socket - * interface as the means of communication with the user level. - * - * MIPS specific IP/TCP/UDP checksumming routines - * - * Authors: Ralf Baechle, - * Lots of code moved from tcp.c and ip.c; see those files - * for more names. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - * - * $Id: checksum.c,v 1.1 2002/09/28 14:58:40 gerg Exp $ - */ -#include -#include -#include -#include -#include -#include - -static inline unsigned short from32to16 (unsigned long sum) -{ - unsigned int result; - /* - %0 %1 - hsw %1, %0 H L L H - add %1, %0 H L H+L+C H+L - */ - asm ("hsw %1, %0; add %1, %0" : "=&r" (result) : "r" (sum)); - return result >> 16; -} - -static inline unsigned int do_csum(const unsigned char * buff, int len) -{ - int odd, count; - unsigned int result = 0; - - if (len <= 0) - goto out; - odd = 1 & (unsigned long) buff; - if (odd) { - result = be16_to_cpu(*buff); - len--; - buff++; - } - count = len >> 1; /* nr of 16-bit words.. */ - if (count) { - if (2 & (unsigned long) buff) { - result += *(unsigned short *) buff; - count--; - len -= 2; - buff += 2; - } - count >>= 1; /* nr of 32-bit words.. */ - if (count) { - unsigned int carry = 0; - do { - unsigned int w = *(unsigned int *) buff; - count--; - buff += 4; - result += carry; - result += w; - carry = (w > result); - } while (count); - result += carry; - result = (result & 0xffff) + (result >> 16); - } - if (len & 2) { - result += *(unsigned short *) buff; - buff += 2; - } - } - if (len & 1) - result += le16_to_cpu(*buff); - result = from32to16(result); - if (odd) - result = ((result >> 8) & 0xff) | ((result & 0xff) << 8); -out: - return result; -} - -/* - * This is a version of ip_compute_csum() optimized for IP headers, - * which always checksum on 4 octet boundaries. - */ -__sum16 ip_fast_csum(const void *iph, unsigned int ihl) -{ - return (__force __sum16)~do_csum(iph,ihl*4); -} - -/* - * this routine is used for miscellaneous IP-like checksums, mainly - * in icmp.c - */ -__sum16 ip_compute_csum(const void *buff, int len) -{ - return (__force __sum16)~do_csum(buff,len); -} - -/* - * computes a partial checksum, e.g. for TCP/UDP fragments - */ -__wsum csum_partial(const void *buff, int len, __wsum sum) -{ - unsigned int result = do_csum(buff, len); - - /* add in old sum, and carry.. */ - result += (__force u32)sum; - if ((__force u32)sum > result) - result += 1; - return (__force __wsum)result; -} - -EXPORT_SYMBOL(csum_partial); - -/* - * copy while checksumming, otherwise like csum_partial - */ -__wsum csum_partial_copy_nocheck(const void *src, void *dst, - int len, __wsum sum) -{ - /* - * It's 2:30 am and I don't feel like doing it real ... - * This is lots slower than the real thing (tm) - */ - sum = csum_partial(src, len, sum); - memcpy(dst, src, len); - - return sum; -} - -/* - * Copy from userspace and compute checksum. If we catch an exception - * then zero the rest of the buffer. - */ -__wsum csum_partial_copy_from_user (const void *src, - void *dst, - int len, __wsum sum, - int *err_ptr) -{ - int missing; - - missing = copy_from_user(dst, src, len); - if (missing) { - memset(dst + len - missing, 0, missing); - *err_ptr = -EFAULT; - } - - return csum_partial(dst, len, sum); -} diff --git a/arch/v850/lib/lshrdi3.c b/arch/v850/lib/lshrdi3.c deleted file mode 100644 index 93b1cb6fdee8..000000000000 --- a/arch/v850/lib/lshrdi3.c +++ /dev/null @@ -1,62 +0,0 @@ -/* lshrdi3.c extracted from gcc-2.7.2/libgcc2.c which is: */ -/* Copyright (C) 1989, 1992, 1993, 1994, 1995 Free Software Foundation, Inc. - -This file is part of GNU CC. - -GNU CC is free software; you can redistribute it and/or modify -it under the terms of the GNU General Public License as published by -the Free Software Foundation; either version 2, or (at your option) -any later version. - -GNU CC is distributed in the hope that it will be useful, -but WITHOUT ANY WARRANTY; without even the implied warranty of -MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -GNU General Public License for more details. - -You should have received a copy of the GNU General Public License -along with GNU CC; see the file COPYING. If not, write to -the Free Software Foundation, 59 Temple Place - Suite 330, -Boston, MA 02111-1307, USA. */ - -#define BITS_PER_UNIT 8 - -typedef int SItype __attribute__ ((mode (SI))); -typedef unsigned int USItype __attribute__ ((mode (SI))); -typedef int DItype __attribute__ ((mode (DI))); -typedef int word_type __attribute__ ((mode (__word__))); - -struct DIstruct {SItype high, low;}; - -typedef union -{ - struct DIstruct s; - DItype ll; -} DIunion; - -DItype -__lshrdi3 (DItype u, word_type b) -{ - DIunion w; - word_type bm; - DIunion uu; - - if (b == 0) - return u; - - uu.ll = u; - - bm = (sizeof (SItype) * BITS_PER_UNIT) - b; - if (bm <= 0) - { - w.s.high = 0; - w.s.low = (USItype)uu.s.high >> -bm; - } - else - { - USItype carries = (USItype)uu.s.high << bm; - w.s.high = (USItype)uu.s.high >> b; - w.s.low = ((USItype)uu.s.low >> b) | carries; - } - - return w.ll; -} diff --git a/arch/v850/lib/memcpy.c b/arch/v850/lib/memcpy.c deleted file mode 100644 index 492847b3e612..000000000000 --- a/arch/v850/lib/memcpy.c +++ /dev/null @@ -1,92 +0,0 @@ -/* - * arch/v850/lib/memcpy.c -- Memory copying - * - * Copyright (C) 2001,02 NEC Corporation - * Copyright (C) 2001,02 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include -#include - -#define CHUNK_SIZE 32 /* bytes */ -#define CHUNK_ALIGNED(addr) (((unsigned long)addr & 0x3) == 0) - -/* Note that this macro uses 8 call-clobbered registers (not including - R1), which are few enough so that the following functions don't need - to spill anything to memory. It also uses R1, which is nominally - reserved for the assembler, but here it should be OK. */ -#define COPY_CHUNK(src, dst) \ - asm ("mov %0, ep;" \ - "sld.w 0[ep], r1; sld.w 4[ep], r12;" \ - "sld.w 8[ep], r13; sld.w 12[ep], r14;" \ - "sld.w 16[ep], r15; sld.w 20[ep], r17;" \ - "sld.w 24[ep], r18; sld.w 28[ep], r19;" \ - "mov %1, ep;" \ - "sst.w r1, 0[ep]; sst.w r12, 4[ep];" \ - "sst.w r13, 8[ep]; sst.w r14, 12[ep];" \ - "sst.w r15, 16[ep]; sst.w r17, 20[ep];" \ - "sst.w r18, 24[ep]; sst.w r19, 28[ep]" \ - :: "r" (src), "r" (dst) \ - : "r1", "r12", "r13", "r14", "r15", \ - "r17", "r18", "r19", "ep", "memory"); - -void *memcpy (void *dst, const void *src, __kernel_size_t size) -{ - char *_dst = dst; - const char *_src = src; - - if (size >= CHUNK_SIZE && CHUNK_ALIGNED(_src) && CHUNK_ALIGNED(_dst)) { - /* Copy large blocks efficiently. */ - unsigned count; - for (count = size / CHUNK_SIZE; count; count--) { - COPY_CHUNK (_src, _dst); - _src += CHUNK_SIZE; - _dst += CHUNK_SIZE; - } - size %= CHUNK_SIZE; - } - - if (size > 0) - do - *_dst++ = *_src++; - while (--size); - - return dst; -} - -void *memmove (void *dst, const void *src, __kernel_size_t size) -{ - if ((unsigned long)dst < (unsigned long)src - || (unsigned long)src + size < (unsigned long)dst) - return memcpy (dst, src, size); - else { - char *_dst = dst + size; - const char *_src = src + size; - - if (size >= CHUNK_SIZE - && CHUNK_ALIGNED (_src) && CHUNK_ALIGNED (_dst)) - { - /* Copy large blocks efficiently. */ - unsigned count; - for (count = size / CHUNK_SIZE; count; count--) { - _src -= CHUNK_SIZE; - _dst -= CHUNK_SIZE; - COPY_CHUNK (_src, _dst); - } - size %= CHUNK_SIZE; - } - - if (size > 0) - do - *--_dst = *--_src; - while (--size); - - return _dst; - } -} diff --git a/arch/v850/lib/memset.c b/arch/v850/lib/memset.c deleted file mode 100644 index d1b2ad821b15..000000000000 --- a/arch/v850/lib/memset.c +++ /dev/null @@ -1,68 +0,0 @@ -/* - * arch/v850/lib/memset.c -- Memory initialization - * - * Copyright (C) 2001,02,04 NEC Corporation - * Copyright (C) 2001,02,04 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#include - -void *memset (void *dst, int val, __kernel_size_t count) -{ - if (count) { - register unsigned loop; - register void *ptr asm ("ep") = dst; - - /* replicate VAL into a long. */ - val &= 0xff; - val |= val << 8; - val |= val << 16; - - /* copy initial unaligned bytes. */ - if ((long)ptr & 1) { - *(char *)ptr = val; - ptr = (void *)((char *)ptr + 1); - count--; - } - if (count > 2 && ((long)ptr & 2)) { - *(short *)ptr = val; - ptr = (void *)((short *)ptr + 1); - count -= 2; - } - - /* 32-byte copying loop. */ - for (loop = count / 32; loop; loop--) { - asm ("sst.w %0, 0[ep]; sst.w %0, 4[ep];" - "sst.w %0, 8[ep]; sst.w %0, 12[ep];" - "sst.w %0, 16[ep]; sst.w %0, 20[ep];" - "sst.w %0, 24[ep]; sst.w %0, 28[ep]" - :: "r" (val) : "memory"); - ptr += 32; - } - count %= 32; - - /* long copying loop. */ - for (loop = count / 4; loop; loop--) { - *(long *)ptr = val; - ptr = (void *)((long *)ptr + 1); - } - count %= 4; - - /* finish up with any trailing bytes. */ - if (count & 2) { - *(short *)ptr = val; - ptr = (void *)((short *)ptr + 1); - } - if (count & 1) { - *(char *)ptr = val; - } - } - - return dst; -} diff --git a/arch/v850/lib/muldi3.c b/arch/v850/lib/muldi3.c deleted file mode 100644 index 277ca25c82c8..000000000000 --- a/arch/v850/lib/muldi3.c +++ /dev/null @@ -1,61 +0,0 @@ -/* muldi3.c extracted from gcc-2.7.2.3/libgcc2.c and - gcc-2.7.2.3/longlong.h which is: */ -/* Copyright (C) 1989, 1992, 1993, 1994, 1995, 2001 Free Software Foundation, Inc. - -This file is part of GNU CC. - -GNU CC is free software; you can redistribute it and/or modify -it under the terms of the GNU General Public License as published by -the Free Software Foundation; either version 2, or (at your option) -any later version. - -GNU CC is distributed in the hope that it will be useful, -but WITHOUT ANY WARRANTY; without even the implied warranty of -MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -GNU General Public License for more details. - -You should have received a copy of the GNU General Public License -along with GNU CC; see the file COPYING. If not, write to -the Free Software Foundation, 59 Temple Place - Suite 330, -Boston, MA 02111-1307, USA. */ - -#define umul_ppmm(w1, w0, u, v) \ - __asm__ ("mulu %3, %0, %1" \ - : "=r" ((USItype)(w0)), \ - "=r" ((USItype)(w1)) \ - : "%0" ((USItype)(u)), \ - "r" ((USItype)(v))) - -#define __umulsidi3(u, v) \ - ({DIunion __w; \ - umul_ppmm (__w.s.high, __w.s.low, u, v); \ - __w.ll; }) - -typedef int SItype __attribute__ ((mode (SI))); -typedef unsigned int USItype __attribute__ ((mode (SI))); -typedef int DItype __attribute__ ((mode (DI))); -typedef int word_type __attribute__ ((mode (__word__))); - -struct DIstruct {SItype high, low;}; - -typedef union -{ - struct DIstruct s; - DItype ll; -} DIunion; - -DItype -__muldi3 (DItype u, DItype v) -{ - DIunion w; - DIunion uu, vv; - - uu.ll = u, - vv.ll = v; - - w.ll = __umulsidi3 (uu.s.low, vv.s.low); - w.s.high += ((USItype) uu.s.low * (USItype) vv.s.high - + (USItype) uu.s.high * (USItype) vv.s.low); - - return w.ll; -} diff --git a/arch/v850/lib/negdi2.c b/arch/v850/lib/negdi2.c deleted file mode 100644 index 571e04fc619a..000000000000 --- a/arch/v850/lib/negdi2.c +++ /dev/null @@ -1,25 +0,0 @@ -/* - * arch/v850/lib/negdi2.c -- 64-bit negation - * - * Copyright (C) 2001 NEC Corporation - * Copyright (C) 2001 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -typedef int DItype __attribute__ ((mode (DI))); - -DItype __negdi2 (DItype x) -{ - __asm__ __volatile__ - ("not r6, r10;" - "add 1, r10;" - "setf c, r6;" - "not r7, r11;" - "add r6, r11" - ::: "r6", "r7", "r10", "r11"); -} diff --git a/drivers/serial/Kconfig b/drivers/serial/Kconfig index 8fc7451c0049..3b4a14e355c1 100644 --- a/drivers/serial/Kconfig +++ b/drivers/serial/Kconfig @@ -942,22 +942,6 @@ config SERIAL_IP22_ZILOG_CONSOLE depends on SERIAL_IP22_ZILOG=y select SERIAL_CORE_CONSOLE -config V850E_UART - bool "NEC V850E on-chip UART support" - depends on V850E_MA1 || V850E_ME2 || V850E_TEG || V850E2_ANNA || V850E_AS85EP1 - select SERIAL_CORE - default y - -config V850E_UARTB - bool - depends on V850E_UART && V850E_ME2 - default y - -config V850E_UART_CONSOLE - bool "Use NEC V850E on-chip UART for console" - depends on V850E_UART - select SERIAL_CORE_CONSOLE - config SERIAL_SH_SCI tristate "SuperH SCI(F) serial port support" depends on SUPERH || H8300 diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig index ccb78f66c2b6..48399e134c0d 100644 --- a/drivers/watchdog/Kconfig +++ b/drivers/watchdog/Kconfig @@ -788,8 +788,6 @@ config WATCHDOG_RIO machines. The watchdog timeout period is normally one minute but can be changed with a boot-time parameter. -# V850 Architecture - # XTENSA Architecture # diff --git a/drivers/watchdog/Makefile b/drivers/watchdog/Makefile index 25b352b664d9..edd305a64e63 100644 --- a/drivers/watchdog/Makefile +++ b/drivers/watchdog/Makefile @@ -119,8 +119,6 @@ obj-$(CONFIG_SH_WDT) += shwdt.o # SPARC64 Architecture -# V850 Architecture - # XTENSA Architecture # Architecture Independant diff --git a/include/asm-v850/Kbuild b/include/asm-v850/Kbuild deleted file mode 100644 index c68e1680da01..000000000000 --- a/include/asm-v850/Kbuild +++ /dev/null @@ -1 +0,0 @@ -include include/asm-generic/Kbuild.asm diff --git a/include/asm-v850/a.out.h b/include/asm-v850/a.out.h deleted file mode 100644 index e9439a0708f6..000000000000 --- a/include/asm-v850/a.out.h +++ /dev/null @@ -1,21 +0,0 @@ -#ifndef __V850_A_OUT_H__ -#define __V850_A_OUT_H__ - -struct exec -{ - unsigned long a_info; /* Use macros N_MAGIC, etc for access */ - unsigned a_text; /* length of text, in bytes */ - unsigned a_data; /* length of data, in bytes */ - unsigned a_bss; /* length of uninitialized data area for file, in bytes */ - unsigned a_syms; /* length of symbol table data in file, in bytes */ - unsigned a_entry; /* start address */ - unsigned a_trsize; /* length of relocation info for text, in bytes */ - unsigned a_drsize; /* length of relocation info for data, in bytes */ -}; - -#define N_TRSIZE(a) ((a).a_trsize) -#define N_DRSIZE(a) ((a).a_drsize) -#define N_SYMSIZE(a) ((a).a_syms) - - -#endif /* __V850_A_OUT_H__ */ diff --git a/include/asm-v850/anna.h b/include/asm-v850/anna.h deleted file mode 100644 index cd5eaee103b0..000000000000 --- a/include/asm-v850/anna.h +++ /dev/null @@ -1,137 +0,0 @@ -/* - * include/asm-v850/anna.h -- Anna V850E2 evaluation cpu chip/board - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_ANNA_H__ -#define __V850_ANNA_H__ - -#include /* Based on V850E2 core. */ - - -#define CPU_MODEL "v850e2/anna" -#define CPU_MODEL_LONG "NEC V850E2/Anna" -#define PLATFORM "anna" -#define PLATFORM_LONG "NEC/Midas lab V850E2/Anna evaluation board" - -#define CPU_CLOCK_FREQ 200000000 /* 200MHz */ -#define SYS_CLOCK_FREQ 33300000 /* 33.3MHz */ - - -/* 1MB of static RAM. This memory is mirrored 64 times. */ -#define SRAM_ADDR 0x04000000 -#define SRAM_SIZE 0x00100000 /* 1MB */ -/* 64MB of DRAM. */ -#define SDRAM_ADDR 0x08000000 -#define SDRAM_SIZE 0x04000000 /* 64MB */ - - -/* For */ -#define PAGE_OFFSET SRAM_ADDR - -/* We use on-chip RAM, for a few miscellaneous variables that must be - accessible using a load instruction relative to R0. The Anna chip has - 128K of `dLB' ram nominally located at 0xFFF00000, but it's mirrored - every 128K, so we can use the `last mirror' (except for the portion at - the top which is overridden by I/O space). In addition, the early - sample chip we're using has lots of memory errors in the dLB ram, so we - use a specially chosen location that has at least 20 bytes of contiguous - valid memory (xxxF0020 - xxxF003F). */ -#define R0_RAM_ADDR 0xFFFF8020 - - -/* Anna specific control registers. */ -#define ANNA_ILBEN_ADDR 0xFFFFF7F2 -#define ANNA_ILBEN (*(volatile u16 *)ANNA_ILBEN_ADDR) - - -/* I/O port P0-P3. */ -/* Direct I/O. Bits 0-7 are pins Pn0-Pn7. */ -#define ANNA_PORT_IO_ADDR(n) (0xFFFFF400 + (n) * 2) -#define ANNA_PORT_IO(n) (*(volatile u8 *)ANNA_PORT_IO_ADDR(n)) -/* Port mode (for direct I/O, 0 = output, 1 = input). */ -#define ANNA_PORT_PM_ADDR(n) (0xFFFFF410 + (n) * 2) -#define ANNA_PORT_PM(n) (*(volatile u8 *)ANNA_PORT_PM_ADDR(n)) - - -/* Hardware-specific interrupt numbers (in the kernel IRQ namespace). */ -#define IRQ_INTP(n) (n) /* Pnnn (pin) interrupts 0-15 */ -#define IRQ_INTP_NUM 16 -#define IRQ_INTOV(n) (0x10 + (n)) /* 0-2 */ -#define IRQ_INTOV_NUM 2 -#define IRQ_INTCCC(n) (0x12 + (n)) -#define IRQ_INTCCC_NUM 4 -#define IRQ_INTCMD(n) (0x16 + (n)) /* interval timer interrupts 0-5 */ -#define IRQ_INTCMD_NUM 6 -#define IRQ_INTDMA(n) (0x1C + (n)) /* DMA interrupts 0-3 */ -#define IRQ_INTDMA_NUM 4 -#define IRQ_INTDMXER 0x20 -#define IRQ_INTSRE(n) (0x21 + (n)*3) /* UART 0-1 reception error */ -#define IRQ_INTSRE_NUM 2 -#define IRQ_INTSR(n) (0x22 + (n)*3) /* UART 0-1 reception completion */ -#define IRQ_INTSR_NUM 2 -#define IRQ_INTST(n) (0x23 + (n)*3) /* UART 0-1 transmission completion */ -#define IRQ_INTST_NUM 2 - -#define NUM_CPU_IRQS 64 - -#ifndef __ASSEMBLY__ -/* Initialize chip interrupts. */ -extern void anna_init_irqs (void); -#endif - - -/* Anna UART details (basically the same as the V850E/MA1, but 2 channels). */ -#define V850E_UART_NUM_CHANNELS 2 -#define V850E_UART_BASE_FREQ (SYS_CLOCK_FREQ / 2) -#define V850E_UART_CHIP_NAME "V850E2/NA85E2A" - -/* This is the UART channel that's actually connected on the board. */ -#define V850E_UART_CONSOLE_CHANNEL 1 - -/* This is a function that gets called before configuring the UART. */ -#define V850E_UART_PRE_CONFIGURE anna_uart_pre_configure -#ifndef __ASSEMBLY__ -extern void anna_uart_pre_configure (unsigned chan, - unsigned cflags, unsigned baud); -#endif - -/* This board supports RTS/CTS for the on-chip UART, but only for channel 1. */ - -/* CTS for UART channel 1 is pin P37 (bit 7 of port 3). */ -#define V850E_UART_CTS(chan) ((chan) == 1 ? !(ANNA_PORT_IO(3) & 0x80) : 1) -/* RTS for UART channel 1 is pin P07 (bit 7 of port 0). */ -#define V850E_UART_SET_RTS(chan, val) \ - do { \ - if (chan == 1) { \ - unsigned old = ANNA_PORT_IO(0); \ - if (val) \ - ANNA_PORT_IO(0) = old & ~0x80; \ - else \ - ANNA_PORT_IO(0) = old | 0x80; \ - } \ - } while (0) - - -/* Timer C details. */ -#define V850E_TIMER_C_BASE_ADDR 0xFFFFF600 - -/* Timer D details (the Anna actually has 5 of these; should change later). */ -#define V850E_TIMER_D_BASE_ADDR 0xFFFFF540 -#define V850E_TIMER_D_TMD_BASE_ADDR (V850E_TIMER_D_BASE_ADDR + 0x0) -#define V850E_TIMER_D_CMD_BASE_ADDR (V850E_TIMER_D_BASE_ADDR + 0x2) -#define V850E_TIMER_D_TMCD_BASE_ADDR (V850E_TIMER_D_BASE_ADDR + 0x4) - -#define V850E_TIMER_D_BASE_FREQ SYS_CLOCK_FREQ -#define V850E_TIMER_D_TMCD_CS_MIN 1 /* min 2^1 divider */ - - -#endif /* __V850_ANNA_H__ */ diff --git a/include/asm-v850/as85ep1.h b/include/asm-v850/as85ep1.h deleted file mode 100644 index 5a5ca9073d09..000000000000 --- a/include/asm-v850/as85ep1.h +++ /dev/null @@ -1,152 +0,0 @@ -/* - * include/asm-v850/as85ep1.h -- AS85EP1 evaluation CPU chip/board - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_AS85EP1_H__ -#define __V850_AS85EP1_H__ - -#include - - -#define CPU_MODEL "as85ep1" -#define CPU_MODEL_LONG "NEC V850E/AS85EP1" -#define PLATFORM "AS85EP1" -#define PLATFORM_LONG "NEC V850E/AS85EP1 evaluation board" - -#define CPU_CLOCK_FREQ 96000000 /* 96MHz */ -#define SYS_CLOCK_FREQ CPU_CLOCK_FREQ - - -/* 1MB of static RAM. */ -#define SRAM_ADDR 0x00400000 -#define SRAM_SIZE 0x00100000 /* 1MB */ -/* About 58MB of DRAM. This can actually be at one of two positions, - determined by jump JP3; we have to use the first position because the - second is partially out of processor instruction addressing range - (though in the second position there's actually 64MB available). */ -#define SDRAM_ADDR 0x00600000 -#define SDRAM_SIZE 0x039F8000 /* approx 58MB */ - -/* For */ -#define PAGE_OFFSET SRAM_ADDR - -/* We use on-chip RAM, for a few miscellaneous variables that must be - accessible using a load instruction relative to R0. The AS85EP1 chip - 16K of internal RAM located slightly before I/O space. */ -#define R0_RAM_ADDR 0xFFFF8000 - - -/* AS85EP1 specific control registers. */ -#define AS85EP1_CSC_ADDR(n) (0xFFFFF060 + (n) * 2) -#define AS85EP1_CSC(n) (*(volatile u16 *)AS85EP1_CSC_ADDR(n)) -#define AS85EP1_BSC_ADDR 0xFFFFF066 -#define AS85EP1_BSC (*(volatile u16 *)AS85EP1_BSC_ADDR) -#define AS85EP1_BCT_ADDR(n) (0xFFFFF480 + (n) * 2) -#define AS85EP1_BCT(n) (*(volatile u16 *)AS85EP1_BCT_ADDR(n)) -#define AS85EP1_DWC_ADDR(n) (0xFFFFF484 + (n) * 2) -#define AS85EP1_DWC(n) (*(volatile u16 *)AS85EP1_DWC_ADDR(n)) -#define AS85EP1_BCC_ADDR 0xFFFFF488 -#define AS85EP1_BCC (*(volatile u16 *)AS85EP1_BCC_ADDR) -#define AS85EP1_ASC_ADDR 0xFFFFF48A -#define AS85EP1_ASC (*(volatile u16 *)AS85EP1_ASC_ADDR) -#define AS85EP1_BCP_ADDR 0xFFFFF48C -#define AS85EP1_BCP (*(volatile u16 *)AS85EP1_BCP_ADDR) -#define AS85EP1_LBS_ADDR 0xFFFFF48E -#define AS85EP1_LBS (*(volatile u16 *)AS85EP1_LBS_ADDR) -#define AS85EP1_BMC_ADDR 0xFFFFF498 -#define AS85EP1_BMC (*(volatile u16 *)AS85EP1_BMC_ADDR) -#define AS85EP1_PRC_ADDR 0xFFFFF49A -#define AS85EP1_PRC (*(volatile u16 *)AS85EP1_PRC_ADDR) -#define AS85EP1_SCR_ADDR(n) (0xFFFFF4A0 + (n) * 4) -#define AS85EP1_SCR(n) (*(volatile u16 *)AS85EP1_SCR_ADDR(n)) -#define AS85EP1_RFS_ADDR(n) (0xFFFFF4A2 + (n) * 4) -#define AS85EP1_RFS(n) (*(volatile u16 *)AS85EP1_RFS_ADDR(n)) -#define AS85EP1_IRAMM_ADDR 0xFFFFF80A -#define AS85EP1_IRAMM (*(volatile u8 *)AS85EP1_IRAMM_ADDR) - - - -/* I/O port P0-P13. */ -/* Direct I/O. Bits 0-7 are pins Pn0-Pn7. */ -#define AS85EP1_PORT_IO_ADDR(n) (0xFFFFF400 + (n) * 2) -#define AS85EP1_PORT_IO(n) (*(volatile u8 *)AS85EP1_PORT_IO_ADDR(n)) -/* Port mode (for direct I/O, 0 = output, 1 = input). */ -#define AS85EP1_PORT_PM_ADDR(n) (0xFFFFF420 + (n) * 2) -#define AS85EP1_PORT_PM(n) (*(volatile u8 *)AS85EP1_PORT_PM_ADDR(n)) -/* Port mode control (0 = direct I/O mode, 1 = alternative I/O mode). */ -#define AS85EP1_PORT_PMC_ADDR(n) (0xFFFFF440 + (n) * 2) -#define AS85EP1_PORT_PMC(n) (*(volatile u8 *)AS85EP1_PORT_PMC_ADDR(n)) - - -/* Hardware-specific interrupt numbers (in the kernel IRQ namespace). */ -#define IRQ_INTCCC(n) (0x0C + (n)) -#define IRQ_INTCCC_NUM 8 -#define IRQ_INTCMD(n) (0x14 + (n)) /* interval timer interrupts 0-5 */ -#define IRQ_INTCMD_NUM 6 -#define IRQ_INTSRE(n) (0x1E + (n)*3) /* UART 0-1 reception error */ -#define IRQ_INTSRE_NUM 2 -#define IRQ_INTSR(n) (0x1F + (n)*3) /* UART 0-1 reception completion */ -#define IRQ_INTSR_NUM 2 -#define IRQ_INTST(n) (0x20 + (n)*3) /* UART 0-1 transmission completion */ -#define IRQ_INTST_NUM 2 - -#define NUM_CPU_IRQS 64 - -#ifndef __ASSEMBLY__ -/* Initialize chip interrupts. */ -extern void as85ep1_init_irqs (void); -#endif - - -/* AS85EP1 UART details (basically the same as the V850E/MA1, but 2 channels). */ -#define V850E_UART_NUM_CHANNELS 2 -#define V850E_UART_BASE_FREQ (SYS_CLOCK_FREQ / 4) -#define V850E_UART_CHIP_NAME "V850E/NA85E" - -/* This is a function that gets called before configuring the UART. */ -#define V850E_UART_PRE_CONFIGURE as85ep1_uart_pre_configure -#ifndef __ASSEMBLY__ -extern void as85ep1_uart_pre_configure (unsigned chan, - unsigned cflags, unsigned baud); -#endif - -/* This board supports RTS/CTS for the on-chip UART, but only for channel 1. */ - -/* CTS for UART channel 1 is pin P54 (bit 4 of port 5). */ -#define V850E_UART_CTS(chan) ((chan) == 1 ? !(AS85EP1_PORT_IO(5) & 0x10) : 1) -/* RTS for UART channel 1 is pin P53 (bit 3 of port 5). */ -#define V850E_UART_SET_RTS(chan, val) \ - do { \ - if (chan == 1) { \ - unsigned old = AS85EP1_PORT_IO(5); \ - if (val) \ - AS85EP1_PORT_IO(5) = old & ~0x8; \ - else \ - AS85EP1_PORT_IO(5) = old | 0x8; \ - } \ - } while (0) - - -/* Timer C details. */ -#define V850E_TIMER_C_BASE_ADDR 0xFFFFF600 - -/* Timer D details (the AS85EP1 actually has 5 of these; should change later). */ -#define V850E_TIMER_D_BASE_ADDR 0xFFFFF540 -#define V850E_TIMER_D_TMD_BASE_ADDR (V850E_TIMER_D_BASE_ADDR + 0x0) -#define V850E_TIMER_D_CMD_BASE_ADDR (V850E_TIMER_D_BASE_ADDR + 0x2) -#define V850E_TIMER_D_TMCD_BASE_ADDR (V850E_TIMER_D_BASE_ADDR + 0x4) - -#define V850E_TIMER_D_BASE_FREQ SYS_CLOCK_FREQ -#define V850E_TIMER_D_TMCD_CS_MIN 2 /* min 2^2 divider */ - - -#endif /* __V850_AS85EP1_H__ */ diff --git a/include/asm-v850/asm.h b/include/asm-v850/asm.h deleted file mode 100644 index bf1e785a5dde..000000000000 --- a/include/asm-v850/asm.h +++ /dev/null @@ -1,32 +0,0 @@ -/* - * include/asm-v850/asm.h -- Macros for writing assembly code - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#define G_ENTRY(name) \ - .balign 4; \ - .globl name; \ - .type name,@function; \ - name -#define G_DATA(name) \ - .globl name; \ - .type name,@object; \ - name -#define END(name) \ - .size name,.-name - -#define L_ENTRY(name) \ - .balign 4; \ - .type name,@function; \ - name -#define L_DATA(name) \ - .type name,@object; \ - name diff --git a/include/asm-v850/atomic.h b/include/asm-v850/atomic.h deleted file mode 100644 index e4e57de08f73..000000000000 --- a/include/asm-v850/atomic.h +++ /dev/null @@ -1,131 +0,0 @@ -/* - * include/asm-v850/atomic.h -- Atomic operations - * - * Copyright (C) 2001,02 NEC Corporation - * Copyright (C) 2001,02 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_ATOMIC_H__ -#define __V850_ATOMIC_H__ - - -#include - -#ifdef CONFIG_SMP -#error SMP not supported -#endif - -typedef struct { int counter; } atomic_t; - -#define ATOMIC_INIT(i) { (i) } - -#ifdef __KERNEL__ - -#define atomic_read(v) ((v)->counter) -#define atomic_set(v,i) (((v)->counter) = (i)) - -static inline int atomic_add_return (int i, volatile atomic_t *v) -{ - unsigned long flags; - int res; - - local_irq_save (flags); - res = v->counter + i; - v->counter = res; - local_irq_restore (flags); - - return res; -} - -static __inline__ int atomic_sub_return (int i, volatile atomic_t *v) -{ - unsigned long flags; - int res; - - local_irq_save (flags); - res = v->counter - i; - v->counter = res; - local_irq_restore (flags); - - return res; -} - -static __inline__ void atomic_clear_mask (unsigned long mask, unsigned long *addr) -{ - unsigned long flags; - - local_irq_save (flags); - *addr &= ~mask; - local_irq_restore (flags); -} - -#endif - -#define atomic_add(i, v) atomic_add_return ((i), (v)) -#define atomic_sub(i, v) atomic_sub_return ((i), (v)) - -#define atomic_dec_return(v) atomic_sub_return (1, (v)) -#define atomic_inc_return(v) atomic_add_return (1, (v)) -#define atomic_inc(v) atomic_inc_return (v) -#define atomic_dec(v) atomic_dec_return (v) - -/* - * atomic_inc_and_test - increment and test - * @v: pointer of type atomic_t - * - * Atomically increments @v by 1 - * and returns true if the result is zero, or false for all - * other cases. - */ -#define atomic_inc_and_test(v) (atomic_inc_return(v) == 0) - -#define atomic_sub_and_test(i,v) (atomic_sub_return ((i), (v)) == 0) -#define atomic_dec_and_test(v) (atomic_sub_return (1, (v)) == 0) -#define atomic_add_negative(i,v) (atomic_add_return ((i), (v)) < 0) - -static inline int atomic_cmpxchg(atomic_t *v, int old, int new) -{ - int ret; - unsigned long flags; - - local_irq_save(flags); - ret = v->counter; - if (likely(ret == old)) - v->counter = new; - local_irq_restore(flags); - - return ret; -} - -#define atomic_xchg(v, new) (xchg(&((v)->counter), new)) - -static inline int atomic_add_unless(atomic_t *v, int a, int u) -{ - int ret; - unsigned long flags; - - local_irq_save(flags); - ret = v->counter; - if (ret != u) - v->counter += a; - local_irq_restore(flags); - - return ret != u; -} - -#define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0) - -/* Atomic operations are already serializing on ARM */ -#define smp_mb__before_atomic_dec() barrier() -#define smp_mb__after_atomic_dec() barrier() -#define smp_mb__before_atomic_inc() barrier() -#define smp_mb__after_atomic_inc() barrier() - -#include -#endif /* __V850_ATOMIC_H__ */ diff --git a/include/asm-v850/auxvec.h b/include/asm-v850/auxvec.h deleted file mode 100644 index f493232d0224..000000000000 --- a/include/asm-v850/auxvec.h +++ /dev/null @@ -1,4 +0,0 @@ -#ifndef __V850_AUXVEC_H__ -#define __V850_AUXVEC_H__ - -#endif /* __V850_AUXVEC_H__ */ diff --git a/include/asm-v850/bitops.h b/include/asm-v850/bitops.h deleted file mode 100644 index f82f5b4a56e0..000000000000 --- a/include/asm-v850/bitops.h +++ /dev/null @@ -1,161 +0,0 @@ -/* - * include/asm-v850/bitops.h -- Bit operations - * - * Copyright (C) 2001,02,03,04,05 NEC Electronics Corporation - * Copyright (C) 2001,02,03,04,05 Miles Bader - * Copyright (C) 1992 Linus Torvalds. - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - */ - -#ifndef __V850_BITOPS_H__ -#define __V850_BITOPS_H__ - -#ifndef _LINUX_BITOPS_H -#error only can be included directly -#endif - -#include /* unlikely */ -#include /* swab32 */ -#include /* interrupt enable/disable */ - - -#ifdef __KERNEL__ - -#include - -/* - * The __ functions are not atomic - */ - -/* In the following constant-bit-op macros, a "g" constraint is used when - we really need an integer ("i" constraint). This is to avoid - warnings/errors from the compiler in the case where the associated - operand _isn't_ an integer, and shouldn't produce bogus assembly because - use of that form is protected by a guard statement that checks for - constants, and should otherwise be removed by the optimizer. This - _usually_ works -- however, __builtin_constant_p returns true for a - variable with a known constant value too, and unfortunately gcc will - happily put the variable in a register and use the register for the "g" - constraint'd asm operand. To avoid the latter problem, we add a - constant offset to the operand and subtract it back in the asm code; - forcing gcc to do arithmetic on the value is usually enough to get it - to use a real constant value. This is horrible, and ultimately - unreliable too, but it seems to work for now (hopefully gcc will offer - us more control in the future, so we can do a better job). */ - -#define __const_bit_op(op, nr, addr) \ - ({ __asm__ (op " (%0 - 0x123), %1" \ - :: "g" (((nr) & 0x7) + 0x123), \ - "m" (*((char *)(addr) + ((nr) >> 3))) \ - : "memory"); }) -#define __var_bit_op(op, nr, addr) \ - ({ int __nr = (nr); \ - __asm__ (op " %0, [%1]" \ - :: "r" (__nr & 0x7), \ - "r" ((char *)(addr) + (__nr >> 3)) \ - : "memory"); }) -#define __bit_op(op, nr, addr) \ - ((__builtin_constant_p (nr) && (unsigned)(nr) <= 0x7FFFF) \ - ? __const_bit_op (op, nr, addr) \ - : __var_bit_op (op, nr, addr)) - -#define __set_bit(nr, addr) __bit_op ("set1", nr, addr) -#define __clear_bit(nr, addr) __bit_op ("clr1", nr, addr) -#define __change_bit(nr, addr) __bit_op ("not1", nr, addr) - -/* The bit instructions used by `non-atomic' variants are actually atomic. */ -#define set_bit __set_bit -#define clear_bit __clear_bit -#define change_bit __change_bit - - -#define __const_tns_bit_op(op, nr, addr) \ - ({ int __tns_res; \ - __asm__ __volatile__ ( \ - "tst1 (%1 - 0x123), %2; setf nz, %0; " op " (%1 - 0x123), %2" \ - : "=&r" (__tns_res) \ - : "g" (((nr) & 0x7) + 0x123), \ - "m" (*((char *)(addr) + ((nr) >> 3))) \ - : "memory"); \ - __tns_res; \ - }) -#define __var_tns_bit_op(op, nr, addr) \ - ({ int __nr = (nr); \ - int __tns_res; \ - __asm__ __volatile__ ( \ - "tst1 %1, [%2]; setf nz, %0; " op " %1, [%2]" \ - : "=&r" (__tns_res) \ - : "r" (__nr & 0x7), \ - "r" ((char *)(addr) + (__nr >> 3)) \ - : "memory"); \ - __tns_res; \ - }) -#define __tns_bit_op(op, nr, addr) \ - ((__builtin_constant_p (nr) && (unsigned)(nr) <= 0x7FFFF) \ - ? __const_tns_bit_op (op, nr, addr) \ - : __var_tns_bit_op (op, nr, addr)) -#define __tns_atomic_bit_op(op, nr, addr) \ - ({ int __tns_atomic_res, __tns_atomic_flags; \ - local_irq_save (__tns_atomic_flags); \ - __tns_atomic_res = __tns_bit_op (op, nr, addr); \ - local_irq_restore (__tns_atomic_flags); \ - __tns_atomic_res; \ - }) - -#define __test_and_set_bit(nr, addr) __tns_bit_op ("set1", nr, addr) -#define test_and_set_bit(nr, addr) __tns_atomic_bit_op ("set1", nr, addr) - -#define __test_and_clear_bit(nr, addr) __tns_bit_op ("clr1", nr, addr) -#define test_and_clear_bit(nr, addr) __tns_atomic_bit_op ("clr1", nr, addr) - -#define __test_and_change_bit(nr, addr) __tns_bit_op ("not1", nr, addr) -#define test_and_change_bit(nr, addr) __tns_atomic_bit_op ("not1", nr, addr) - - -#define __const_test_bit(nr, addr) \ - ({ int __test_bit_res; \ - __asm__ __volatile__ ("tst1 (%1 - 0x123), %2; setf nz, %0" \ - : "=r" (__test_bit_res) \ - : "g" (((nr) & 0x7) + 0x123), \ - "m" (*((const char *)(addr) + ((nr) >> 3)))); \ - __test_bit_res; \ - }) -static inline int __test_bit (int nr, const void *addr) -{ - int res; - __asm__ __volatile__ ("tst1 %1, [%2]; setf nz, %0" - : "=r" (res) - : "r" (nr & 0x7), "r" (addr + (nr >> 3))); - return res; -} -#define test_bit(nr,addr) \ - ((__builtin_constant_p (nr) && (unsigned)(nr) <= 0x7FFFF) \ - ? __const_test_bit ((nr), (addr)) \ - : __test_bit ((nr), (addr))) - - -/* clear_bit doesn't provide any barrier for the compiler. */ -#define smp_mb__before_clear_bit() barrier () -#define smp_mb__after_clear_bit() barrier () - -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#define ext2_set_bit_atomic(l,n,a) test_and_set_bit(n,a) -#define ext2_clear_bit_atomic(l,n,a) test_and_clear_bit(n,a) - -#include - -#endif /* __KERNEL__ */ - -#endif /* __V850_BITOPS_H__ */ diff --git a/include/asm-v850/bug.h b/include/asm-v850/bug.h deleted file mode 100644 index b0ed2d35f3e8..000000000000 --- a/include/asm-v850/bug.h +++ /dev/null @@ -1,25 +0,0 @@ -/* - * include/asm-v850/bug.h -- Bug reporting - * - * Copyright (C) 2003 NEC Electronics Corporation - * Copyright (C) 2003 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_BUG_H__ -#define __V850_BUG_H__ - -#ifdef CONFIG_BUG -extern void __bug (void) __attribute__ ((noreturn)); -#define BUG() __bug() -#define HAVE_ARCH_BUG -#endif - -#include - -#endif /* __V850_BUG_H__ */ diff --git a/include/asm-v850/bugs.h b/include/asm-v850/bugs.h deleted file mode 100644 index 71110a65c1d7..000000000000 --- a/include/asm-v850/bugs.h +++ /dev/null @@ -1,16 +0,0 @@ -/* - * include/asm-v850e/bugs.h - * - * Copyright (C) 1994 Linus Torvalds - */ - -/* - * This is included by init/main.c to check for architecture-dependent bugs. - * - * Needs: - * void check_bugs(void); - */ - -static void check_bugs(void) -{ -} diff --git a/include/asm-v850/byteorder.h b/include/asm-v850/byteorder.h deleted file mode 100644 index a6f07530050e..000000000000 --- a/include/asm-v850/byteorder.h +++ /dev/null @@ -1,48 +0,0 @@ -/* - * include/asm-v850/byteorder.h -- Endian id and conversion ops - * - * Copyright (C) 2001 NEC Corporation - * Copyright (C) 2001 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_BYTEORDER_H__ -#define __V850_BYTEORDER_H__ - -#include -#include - -#ifdef __GNUC__ - -static __inline__ __attribute_const__ __u32 ___arch__swab32 (__u32 word) -{ - __u32 res; - __asm__ ("bsw %1, %0" : "=r" (res) : "r" (word)); - return res; -} - -static __inline__ __attribute_const__ __u16 ___arch__swab16 (__u16 half_word) -{ - __u16 res; - __asm__ ("bsh %1, %0" : "=r" (res) : "r" (half_word)); - return res; -} - -#define __arch__swab32(x) ___arch__swab32(x) -#define __arch__swab16(x) ___arch__swab16(x) - -#if !defined(__STRICT_ANSI__) || defined(__KERNEL__) -# define __BYTEORDER_HAS_U64__ -# define __SWAB_64_THRU_32__ -#endif - -#endif /* __GNUC__ */ - -#include - -#endif /* __V850_BYTEORDER_H__ */ diff --git a/include/asm-v850/cache.h b/include/asm-v850/cache.h deleted file mode 100644 index 8832c7ea3242..000000000000 --- a/include/asm-v850/cache.h +++ /dev/null @@ -1,26 +0,0 @@ -/* - * include/asm-v850/cache.h -- Cache operations - * - * Copyright (C) 2001,05 NEC Corporation - * Copyright (C) 2001,05 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_CACHE_H__ -#define __V850_CACHE_H__ - -/* All cache operations are machine-dependent. */ -#include - -#ifndef L1_CACHE_BYTES -/* This processor has no cache, so just choose an arbitrary value. */ -#define L1_CACHE_BYTES 16 -#define L1_CACHE_SHIFT 4 -#endif - -#endif /* __V850_CACHE_H__ */ diff --git a/include/asm-v850/cacheflush.h b/include/asm-v850/cacheflush.h deleted file mode 100644 index 9ece05a202ef..000000000000 --- a/include/asm-v850/cacheflush.h +++ /dev/null @@ -1,70 +0,0 @@ -/* - * include/asm-v850/cacheflush.h - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_CACHEFLUSH_H__ -#define __V850_CACHEFLUSH_H__ - -/* Somebody depends on this; sigh... */ -#include - -#include - - -/* The following are all used by the kernel in ways that only affect - systems with MMUs, so we don't need them. */ -#define flush_cache_all() ((void)0) -#define flush_cache_mm(mm) ((void)0) -#define flush_cache_dup_mm(mm) ((void)0) -#define flush_cache_range(vma, start, end) ((void)0) -#define flush_cache_page(vma, vmaddr, pfn) ((void)0) -#define flush_dcache_page(page) ((void)0) -#define flush_dcache_mmap_lock(mapping) ((void)0) -#define flush_dcache_mmap_unlock(mapping) ((void)0) -#define flush_cache_vmap(start, end) ((void)0) -#define flush_cache_vunmap(start, end) ((void)0) - -#ifdef CONFIG_NO_CACHE - -/* Some systems have no cache at all, in which case we don't need these - either. */ -#define flush_icache() ((void)0) -#define flush_icache_range(start, end) ((void)0) -#define flush_icache_page(vma,pg) ((void)0) -#define flush_icache_user_range(vma,pg,adr,len) ((void)0) -#define flush_cache_sigtramp(vaddr) ((void)0) - -#else /* !CONFIG_NO_CACHE */ - -struct page; -struct mm_struct; -struct vm_area_struct; - -/* Otherwise, somebody had better define them. */ -extern void flush_icache (void); -extern void flush_icache_range (unsigned long start, unsigned long end); -extern void flush_icache_page (struct vm_area_struct *vma, struct page *page); -extern void flush_icache_user_range (struct vm_area_struct *vma, - struct page *page, - unsigned long adr, int len); -extern void flush_cache_sigtramp (unsigned long addr); - -#endif /* CONFIG_NO_CACHE */ - -#define copy_to_user_page(vma, page, vaddr, dst, src, len) \ -do { memcpy(dst, src, len); \ - flush_icache_user_range(vma, page, vaddr, len); \ -} while (0) -#define copy_from_user_page(vma, page, vaddr, dst, src, len) \ - memcpy(dst, src, len) - -#endif /* __V850_CACHEFLUSH_H__ */ diff --git a/include/asm-v850/checksum.h b/include/asm-v850/checksum.h deleted file mode 100644 index d1dddd938262..000000000000 --- a/include/asm-v850/checksum.h +++ /dev/null @@ -1,112 +0,0 @@ -/* - * include/asm-v850/checksum.h -- Checksum ops - * - * Copyright (C) 2001,2005 NEC Corporation - * Copyright (C) 2001,2005 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_CHECKSUM_H__ -#define __V850_CHECKSUM_H__ - -/* - * computes the checksum of a memory block at buff, length len, - * and adds in "sum" (32-bit) - * - * returns a 32-bit number suitable for feeding into itself - * or csum_tcpudp_magic - * - * this function must be called with even lengths, except - * for the last fragment, which may be odd - * - * it's best to have buff aligned on a 32-bit boundary - */ -extern __wsum csum_partial(const void *buff, int len, __wsum sum); - -/* - * the same as csum_partial, but copies from src while it - * checksums - * - * here even more important to align src and dst on a 32-bit (or even - * better 64-bit) boundary - */ -extern __wsum csum_partial_copy_nocheck(const void *src, - void *dst, int len, __wsum sum); - - -/* - * the same as csum_partial_copy, but copies from user space. - * - * here even more important to align src and dst on a 32-bit (or even - * better 64-bit) boundary - */ -extern __wsum csum_partial_copy_from_user (const void *src, - void *dst, - int len, __wsum sum, - int *csum_err); - -__sum16 ip_fast_csum(const void *iph, unsigned int ihl); - -/* - * Fold a partial checksum - */ -static inline __sum16 csum_fold (__wsum sum) -{ - unsigned int result; - /* - %0 %1 - hsw %1, %0 H L L H - add %1, %0 H L H+L+C H+L - */ - asm ("hsw %1, %0; add %1, %0" : "=&r" (result) : "r" (sum)); - return (__force __sum16)(~result >> 16); -} - - -/* - * computes the checksum of the TCP/UDP pseudo-header - * returns a 16-bit checksum, already complemented - */ -static inline __wsum -csum_tcpudp_nofold (__be32 saddr, __be32 daddr, - unsigned short len, - unsigned short proto, __wsum sum) -{ - int __carry; - __asm__ ("add %2, %0;" - "setf c, %1;" - "add %1, %0;" - "add %3, %0;" - "setf c, %1;" - "add %1, %0;" - "add %4, %0;" - "setf c, %1;" - "add %1, %0" - : "=&r" (sum), "=&r" (__carry) - : "r" (daddr), "r" (saddr), - "r" ((len + proto) << 8), - "0" (sum)); - return sum; -} - -static inline __sum16 -csum_tcpudp_magic (__be32 saddr, __be32 daddr, - unsigned short len, - unsigned short proto, __wsum sum) -{ - return csum_fold (csum_tcpudp_nofold (saddr, daddr, len, proto, sum)); -} - -/* - * this routine is used for miscellaneous IP-like checksums, mainly - * in icmp.c - */ -extern __sum16 ip_compute_csum(const void *buff, int len); - - -#endif /* __V850_CHECKSUM_H__ */ diff --git a/include/asm-v850/clinkage.h b/include/asm-v850/clinkage.h deleted file mode 100644 index c389691d6f86..000000000000 --- a/include/asm-v850/clinkage.h +++ /dev/null @@ -1,26 +0,0 @@ -/* - * include/asm-v850/clinkage.h -- Macros to reflect C symbol-naming conventions - * - * Copyright (C) 2001,02 NEC Corporatione - * Copyright (C) 2001,02 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_CLINKAGE_H__ -#define __V850_CLINKAGE_H__ - -#include -#include - -#define C_SYMBOL_NAME(name) macrology_paste(_, name) -#define C_SYMBOL_STRING(name) macrology_stringify(C_SYMBOL_NAME(name)) -#define C_ENTRY(name) G_ENTRY(C_SYMBOL_NAME(name)) -#define C_DATA(name) G_DATA(C_SYMBOL_NAME(name)) -#define C_END(name) END(C_SYMBOL_NAME(name)) - -#endif /* __V850_CLINKAGE_H__ */ diff --git a/include/asm-v850/cputime.h b/include/asm-v850/cputime.h deleted file mode 100644 index 7c799c33b8a9..000000000000 --- a/include/asm-v850/cputime.h +++ /dev/null @@ -1,6 +0,0 @@ -#ifndef __V850_CPUTIME_H -#define __V850_CPUTIME_H - -#include - -#endif /* __V850_CPUTIME_H */ diff --git a/include/asm-v850/current.h b/include/asm-v850/current.h deleted file mode 100644 index 30aae5673770..000000000000 --- a/include/asm-v850/current.h +++ /dev/null @@ -1,47 +0,0 @@ -/* - * include/asm-v850/current.h -- Current task - * - * Copyright (C) 2001,02 NEC Corporation - * Copyright (C) 2001,02 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_CURRENT_H__ -#define __V850_CURRENT_H__ - -#ifndef __ASSEMBLY__ /* is not asm-safe. */ -#include -#endif - -#include - - -/* Register used to hold the current task pointer while in the kernel. - Any `call clobbered' register without a special meaning should be OK, - but check asm/v850/kernel/entry.S to be sure. */ -#define CURRENT_TASK_REGNUM 16 -#define CURRENT_TASK macrology_paste (r, CURRENT_TASK_REGNUM) - - -#ifdef __ASSEMBLY__ - -/* Put a pointer to the current task structure into REG. */ -#define GET_CURRENT_TASK(reg) \ - GET_CURRENT_THREAD(reg); \ - ld.w TI_TASK[reg], reg - -#else /* !__ASSEMBLY__ */ - -/* A pointer to the current task. */ -register struct task_struct *current \ - __asm__ (macrology_stringify (CURRENT_TASK)); - -#endif /* __ASSEMBLY__ */ - - -#endif /* _V850_CURRENT_H */ diff --git a/include/asm-v850/delay.h b/include/asm-v850/delay.h deleted file mode 100644 index 6d028e6b2354..000000000000 --- a/include/asm-v850/delay.h +++ /dev/null @@ -1,47 +0,0 @@ -/* - * include/asm-v850/delay.h -- Delay routines, using a pre-computed - * "loops_per_second" value - * - * Copyright (C) 2001,03 NEC Corporation - * Copyright (C) 2001,03 Miles Bader - * Copyright (C) 1994 Hamish Macdonald - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - */ - -#ifndef __V850_DELAY_H__ -#define __V850_DELAY_H__ - -#include - -static inline void __delay(unsigned long loops) -{ - if (loops) - __asm__ __volatile__ ("1: add -1, %0; bnz 1b" - : "=r" (loops) : "0" (loops)); -} - -/* - * Use only for very small delays ( < 1 msec). Should probably use a - * lookup table, really, as the multiplications take much too long with - * short delays. This is a "reasonable" implementation, though (and the - * first constant multiplications gets optimized away if the delay is - * a constant) - */ - -extern unsigned long loops_per_jiffy; - -static inline void udelay(unsigned long usecs) -{ - register unsigned long full_loops, part_loops; - - full_loops = ((usecs * HZ) / 1000000) * loops_per_jiffy; - usecs %= (1000000 / HZ); - part_loops = (usecs * HZ * loops_per_jiffy) / 1000000; - - __delay(full_loops + part_loops); -} - -#endif /* __V850_DELAY_H__ */ diff --git a/include/asm-v850/device.h b/include/asm-v850/device.h deleted file mode 100644 index d8f9872b0e2d..000000000000 --- a/include/asm-v850/device.h +++ /dev/null @@ -1,7 +0,0 @@ -/* - * Arch specific extensions to struct device - * - * This file is released under the GPLv2 - */ -#include - diff --git a/include/asm-v850/div64.h b/include/asm-v850/div64.h deleted file mode 100644 index 6cd978cefb28..000000000000 --- a/include/asm-v850/div64.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-v850/dma-mapping.h b/include/asm-v850/dma-mapping.h deleted file mode 100644 index 1cc42c603a1b..000000000000 --- a/include/asm-v850/dma-mapping.h +++ /dev/null @@ -1,11 +0,0 @@ -#ifndef __V850_DMA_MAPPING_H__ -#define __V850_DMA_MAPPING_H__ - - -#ifdef CONFIG_PCI -#include -#else -#include -#endif - -#endif /* __V850_DMA_MAPPING_H__ */ diff --git a/include/asm-v850/dma.h b/include/asm-v850/dma.h deleted file mode 100644 index 2369849e2d0a..000000000000 --- a/include/asm-v850/dma.h +++ /dev/null @@ -1,18 +0,0 @@ -#ifndef __V850_DMA_H__ -#define __V850_DMA_H__ - -/* What should this be? */ -#define MAX_DMA_ADDRESS 0xFFFFFFFF - -/* reserve a DMA channel */ -extern int request_dma (unsigned int dmanr, const char * device_id); -/* release it again */ -extern void free_dma (unsigned int dmanr); - -#ifdef CONFIG_PCI -extern int isa_dma_bridge_buggy; -#else -#define isa_dma_bridge_buggy (0) -#endif - -#endif /* __V850_DMA_H__ */ diff --git a/include/asm-v850/elf.h b/include/asm-v850/elf.h deleted file mode 100644 index 28f5b176ff1a..000000000000 --- a/include/asm-v850/elf.h +++ /dev/null @@ -1,99 +0,0 @@ -#ifndef __V850_ELF_H__ -#define __V850_ELF_H__ - -/* - * ELF register definitions.. - */ - -#include -#include -#include - -typedef unsigned long elf_greg_t; - -#define ELF_NGREG (sizeof (struct pt_regs) / sizeof(elf_greg_t)) -typedef elf_greg_t elf_gregset_t[ELF_NGREG]; - -typedef struct user_fpu_struct elf_fpregset_t; - -/* - * This is used to ensure we don't load something for the wrong architecture. - */ -#define elf_check_arch(x) \ - ((x)->e_machine == EM_V850 || (x)->e_machine == EM_CYGNUS_V850) - - -/* v850 relocation types. */ -#define R_V850_NONE 0 -#define R_V850_9_PCREL 1 -#define R_V850_22_PCREL 2 -#define R_V850_HI16_S 3 -#define R_V850_HI16 4 -#define R_V850_LO16 5 -#define R_V850_32 6 -#define R_V850_16 7 -#define R_V850_8 8 -#define R_V850_SDA_16_16_OFFSET 9 /* For ld.b, st.b, set1, clr1, - not1, tst1, movea, movhi */ -#define R_V850_SDA_15_16_OFFSET 10 /* For ld.w, ld.h, ld.hu, st.w, st.h */ -#define R_V850_ZDA_16_16_OFFSET 11 /* For ld.b, st.b, set1, clr1, - not1, tst1, movea, movhi */ -#define R_V850_ZDA_15_16_OFFSET 12 /* For ld.w, ld.h, ld.hu, st.w, st.h */ -#define R_V850_TDA_6_8_OFFSET 13 /* For sst.w, sld.w */ -#define R_V850_TDA_7_8_OFFSET 14 /* For sst.h, sld.h */ -#define R_V850_TDA_7_7_OFFSET 15 /* For sst.b, sld.b */ -#define R_V850_TDA_16_16_OFFSET 16 /* For set1, clr1, not1, tst1, - movea, movhi */ -#define R_V850_NUM 17 - - -/* - * These are used to set parameters in the core dumps. - */ -#define ELF_CLASS ELFCLASS32 -#ifdef __LITTLE_ENDIAN__ -#define ELF_DATA ELFDATA2LSB -#else -#define ELF_DATA ELFDATA2MSB -#endif -#define ELF_ARCH EM_V850 - -#define USE_ELF_CORE_DUMP -#define ELF_EXEC_PAGESIZE 4096 - - -#define ELF_CORE_COPY_REGS(_dest,_regs) \ - memcpy((char *) &_dest, (char *) _regs, \ - sizeof(struct pt_regs)); - -/* This yields a mask that user programs can use to figure out what - instruction set this CPU supports. This could be done in user space, - but it's not easy, and we've already done it here. */ - -#define ELF_HWCAP (0) - -/* This yields a string that ld.so will use to load implementation - specific libraries for optimization. This is more specific in - intent than poking at uname or /proc/cpuinfo. - - For the moment, we have only optimizations for the Intel generations, - but that could change... */ - -#define ELF_PLATFORM (NULL) - -#define ELF_PLAT_INIT(_r, load_addr) \ - do { \ - _r->gpr[0] = _r->gpr[1] = _r->gpr[2] = _r->gpr[3] = \ - _r->gpr[4] = _r->gpr[5] = _r->gpr[6] = _r->gpr[7] = \ - _r->gpr[8] = _r->gpr[9] = _r->gpr[10] = _r->gpr[11] = \ - _r->gpr[12] = _r->gpr[13] = _r->gpr[14] = _r->gpr[15] = \ - _r->gpr[16] = _r->gpr[17] = _r->gpr[18] = _r->gpr[19] = \ - _r->gpr[20] = _r->gpr[21] = _r->gpr[22] = _r->gpr[23] = \ - _r->gpr[24] = _r->gpr[25] = _r->gpr[26] = _r->gpr[27] = \ - _r->gpr[28] = _r->gpr[29] = _r->gpr[30] = _r->gpr[31] = \ - 0; \ - } while (0) - -#define SET_PERSONALITY(ex, ibcs2) set_personality(PER_LINUX_32BIT) - -#endif /* __V850_ELF_H__ */ diff --git a/include/asm-v850/emergency-restart.h b/include/asm-v850/emergency-restart.h deleted file mode 100644 index 108d8c48e42e..000000000000 --- a/include/asm-v850/emergency-restart.h +++ /dev/null @@ -1,6 +0,0 @@ -#ifndef _ASM_EMERGENCY_RESTART_H -#define _ASM_EMERGENCY_RESTART_H - -#include - -#endif /* _ASM_EMERGENCY_RESTART_H */ diff --git a/include/asm-v850/entry.h b/include/asm-v850/entry.h deleted file mode 100644 index d9df8ac48584..000000000000 --- a/include/asm-v850/entry.h +++ /dev/null @@ -1,113 +0,0 @@ -/* - * include/asm-v850/entry.h -- Definitions used by low-level trap handlers - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_ENTRY_H__ -#define __V850_ENTRY_H__ - - -#include -#include - - -/* These are special variables using by the kernel trap/interrupt code - to save registers in, at a time when there are no spare registers we - can use to do so, and we can't depend on the value of the stack - pointer. This means that they must be within a signed 16-bit - displacement of 0x00000000. */ - -#define KERNEL_VAR_SPACE_ADDR R0_RAM_ADDR - -#ifdef __ASSEMBLY__ -#define KERNEL_VAR(addr) addr[r0] -#else -#define KERNEL_VAR(addr) (*(volatile unsigned long *)(addr)) -#endif - -/* Kernel stack pointer, 4 bytes. */ -#define KSP_ADDR (KERNEL_VAR_SPACE_ADDR + 0) -#define KSP KERNEL_VAR (KSP_ADDR) -/* 1 if in kernel-mode, 0 if in user mode, 1 byte. */ -#define KM_ADDR (KERNEL_VAR_SPACE_ADDR + 4) -#define KM KERNEL_VAR (KM_ADDR) -/* Temporary storage for interrupt handlers, 4 bytes. */ -#define INT_SCRATCH_ADDR (KERNEL_VAR_SPACE_ADDR + 8) -#define INT_SCRATCH KERNEL_VAR (INT_SCRATCH_ADDR) -/* Where the stack-pointer is saved when jumping to various sorts of - interrupt handlers. ENTRY_SP is used by everything except NMIs, - which have their own location. Higher-priority NMIs can clobber the - value written by a lower priority NMI, since they can't be disabled, - but that's OK, because only NMI0 (the lowest-priority one) is allowed - to return. */ -#define ENTRY_SP_ADDR (KERNEL_VAR_SPACE_ADDR + 12) -#define ENTRY_SP KERNEL_VAR (ENTRY_SP_ADDR) -#define NMI_ENTRY_SP_ADDR (KERNEL_VAR_SPACE_ADDR + 16) -#define NMI_ENTRY_SP KERNEL_VAR (NMI_ENTRY_SP_ADDR) - -#ifdef CONFIG_RESET_GUARD -/* Used to detect unexpected resets (since the v850 has no MMU, any call - through a null pointer will jump to the reset vector). We detect - such resets by checking for a magic value, RESET_GUARD_ACTIVE, in - this location. Properly resetting the machine stores zero there, so - it shouldn't trigger the guard; the power-on value is uncertain, but - it's unlikely to be RESET_GUARD_ACTIVE. */ -#define RESET_GUARD_ADDR (KERNEL_VAR_SPACE_ADDR + 28) -#define RESET_GUARD KERNEL_VAR (RESET_GUARD_ADDR) -#define RESET_GUARD_ACTIVE 0xFAB4BEEF -#endif /* CONFIG_RESET_GUARD */ - -#ifdef CONFIG_V850E_HIGHRES_TIMER -#define HIGHRES_TIMER_SLOW_TICKS_ADDR (KERNEL_VAR_SPACE_ADDR + 32) -#define HIGHRES_TIMER_SLOW_TICKS KERNEL_VAR (HIGHRES_TIMER_SLOW_TICKS_ADDR) -#endif /* CONFIG_V850E_HIGHRES_TIMER */ - -#ifndef __ASSEMBLY__ - -#ifdef CONFIG_RESET_GUARD -/* Turn off reset guard, so that resetting the machine works normally. - This should be called in the various machine_halt, etc., functions. */ -static inline void disable_reset_guard (void) -{ - RESET_GUARD = 0; -} -#endif /* CONFIG_RESET_GUARD */ - -#endif /* !__ASSEMBLY__ */ - - -/* A `state save frame' is a struct pt_regs preceded by some extra space - suitable for a function call stack frame. */ - -/* Amount of room on the stack reserved for arguments and to satisfy the - C calling conventions, in addition to the space used by the struct - pt_regs that actually holds saved values. */ -#define STATE_SAVE_ARG_SPACE (6*4) /* Up to six arguments. */ - - -#ifdef __ASSEMBLY__ - -/* The size of a state save frame. */ -#define STATE_SAVE_SIZE (PT_SIZE + STATE_SAVE_ARG_SPACE) - -#else /* !__ASSEMBLY__ */ - -/* The size of a state save frame. */ -#define STATE_SAVE_SIZE (sizeof (struct pt_regs) + STATE_SAVE_ARG_SPACE) - -#endif /* __ASSEMBLY__ */ - - -/* Offset of the struct pt_regs in a state save frame. */ -#define STATE_SAVE_PT_OFFSET STATE_SAVE_ARG_SPACE - - -#endif /* __V850_ENTRY_H__ */ diff --git a/include/asm-v850/errno.h b/include/asm-v850/errno.h deleted file mode 100644 index 31c91df01205..000000000000 --- a/include/asm-v850/errno.h +++ /dev/null @@ -1,6 +0,0 @@ -#ifndef __V850_ERRNO_H__ -#define __V850_ERRNO_H__ - -#include - -#endif /* __V850_ERRNO_H__ */ diff --git a/include/asm-v850/fb.h b/include/asm-v850/fb.h deleted file mode 100644 index c7df38030992..000000000000 --- a/include/asm-v850/fb.h +++ /dev/null @@ -1,12 +0,0 @@ -#ifndef _ASM_FB_H_ -#define _ASM_FB_H_ -#include - -#define fb_pgprotect(...) do {} while (0) - -static inline int fb_is_primary_device(struct fb_info *info) -{ - return 0; -} - -#endif /* _ASM_FB_H_ */ diff --git a/include/asm-v850/fcntl.h b/include/asm-v850/fcntl.h deleted file mode 100644 index 3af4d56776dd..000000000000 --- a/include/asm-v850/fcntl.h +++ /dev/null @@ -1,11 +0,0 @@ -#ifndef __V850_FCNTL_H__ -#define __V850_FCNTL_H__ - -#define O_DIRECTORY 040000 /* must be a directory */ -#define O_NOFOLLOW 0100000 /* don't follow links */ -#define O_DIRECT 0200000 /* direct disk access hint - currently ignored */ -#define O_LARGEFILE 0400000 - -#include - -#endif /* __V850_FCNTL_H__ */ diff --git a/include/asm-v850/flat.h b/include/asm-v850/flat.h deleted file mode 100644 index 17f0ea566611..000000000000 --- a/include/asm-v850/flat.h +++ /dev/null @@ -1,133 +0,0 @@ -/* - * include/asm-v850/flat.h -- uClinux flat-format executables - * - * Copyright (C) 2002,03 NEC Electronics Corporation - * Copyright (C) 2002,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_FLAT_H__ -#define __V850_FLAT_H__ - -/* The amount by which a relocation can exceed the program image limits - without being regarded as an error. On the v850, the relocations of - some base-pointers can be offset by 0x8000 (to allow better usage of the - space offered by 16-bit signed offsets -- in most cases the offsets used - with such a base-pointer will be negative). */ - -#define flat_reloc_valid(reloc, size) ((reloc) <= (size + 0x8000)) - -#define flat_stack_align(sp) /* nothing needed */ -#define flat_argvp_envp_on_stack() 0 -#define flat_old_ram_flag(flags) (flags) -#define flat_set_persistent(relval, p) 0 - -/* We store the type of relocation in the top 4 bits of the `relval.' */ - -/* Convert a relocation entry into an address. */ -static inline unsigned long -flat_get_relocate_addr (unsigned long relval) -{ - return relval & 0x0fffffff; /* Mask out top 4-bits */ -} - -#define flat_v850_get_reloc_type(relval) ((relval) >> 28) - -#define FLAT_V850_R_32 0 /* Normal 32-bit reloc */ -#define FLAT_V850_R_HI16S_LO15 1 /* High 16-bits + signed 15-bit low field */ -#define FLAT_V850_R_HI16S_LO16 2 /* High 16-bits + signed 16-bit low field */ - -/* Extract the address to be relocated from the symbol reference at RP; - RELVAL is the raw relocation-table entry from which RP is derived. - For the v850, RP should always be half-word aligned. */ -static inline unsigned long flat_get_addr_from_rp (unsigned long *rp, - unsigned long relval, - unsigned long flags, - unsigned long *persistent) -{ - short *srp = (short *)rp; - - switch (flat_v850_get_reloc_type (relval)) - { - case FLAT_V850_R_32: - /* Simple 32-bit address. */ - return srp[0] | (srp[1] << 16); - - case FLAT_V850_R_HI16S_LO16: - /* The high and low halves of the address are in the 16 - bits at RP, and the 2nd word of the 32-bit instruction - following that, respectively. The low half is _signed_ - so we have to sign-extend it and add it to the upper - half instead of simply or-ing them together. - - Unlike most relocated address, this one is stored in - native (little-endian) byte-order to avoid problems with - trashing the low-order bit, so we have to convert to - network-byte-order before returning, as that's what the - caller expects. */ - return htonl ((srp[0] << 16) + srp[2]); - - case FLAT_V850_R_HI16S_LO15: - /* The high and low halves of the address are in the 16 - bits at RP, and the upper 15 bits of the 2nd word of the - 32-bit instruction following that, respectively. The - low half is _signed_ so we have to sign-extend it and - add it to the upper half instead of simply or-ing them - together. The lowest bit is always zero. - - Unlike most relocated address, this one is stored in - native (little-endian) byte-order to avoid problems with - trashing the low-order bit, so we have to convert to - network-byte-order before returning, as that's what the - caller expects. */ - return htonl ((srp[0] << 16) + (srp[2] & ~0x1)); - - default: - return ~0; /* bogus value */ - } -} - -/* Insert the address ADDR into the symbol reference at RP; - RELVAL is the raw relocation-table entry from which RP is derived. - For the v850, RP should always be half-word aligned. */ -static inline void flat_put_addr_at_rp (unsigned long *rp, unsigned long addr, - unsigned long relval) -{ - short *srp = (short *)rp; - - switch (flat_v850_get_reloc_type (relval)) { - case FLAT_V850_R_32: - /* Simple 32-bit address. */ - srp[0] = addr & 0xFFFF; - srp[1] = (addr >> 16); - break; - - case FLAT_V850_R_HI16S_LO16: - /* The high and low halves of the address are in the 16 - bits at RP, and the 2nd word of the 32-bit instruction - following that, respectively. The low half is _signed_ - so we must carry its sign bit to the upper half before - writing the upper half. */ - srp[0] = (addr >> 16) + ((addr >> 15) & 0x1); - srp[2] = addr & 0xFFFF; - break; - - case FLAT_V850_R_HI16S_LO15: - /* The high and low halves of the address are in the 16 - bits at RP, and the upper 15 bits of the 2nd word of the - 32-bit instruction following that, respectively. The - low half is _signed_ so we must carry its sign bit to - the upper half before writing the upper half. The - lowest bit we preserve from the existing instruction. */ - srp[0] = (addr >> 16) + ((addr >> 15) & 0x1); - srp[2] = (addr & 0xFFFE) | (srp[2] & 0x1); - break; - } -} - -#endif /* __V850_FLAT_H__ */ diff --git a/include/asm-v850/fpga85e2c.h b/include/asm-v850/fpga85e2c.h deleted file mode 100644 index 23aae666c718..000000000000 --- a/include/asm-v850/fpga85e2c.h +++ /dev/null @@ -1,82 +0,0 @@ -/* - * include/asm-v850/fpga85e2c.h -- Machine-dependent defs for - * FPGA implementation of V850E2/NA85E2C - * - * Copyright (C) 2002,03 NEC Electronics Corporation - * Copyright (C) 2002,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_FPGA85E2C_H__ -#define __V850_FPGA85E2C_H__ - -#include -#include - - -#define CPU_MODEL "v850e2/fpga85e2c" -#define CPU_MODEL_LONG "NEC V850E2/NA85E2C" -#define PLATFORM "fpga85e2c" -#define PLATFORM_LONG "NA85E2C FPGA implementation" - - -/* `external ram'. */ -#define ERAM_ADDR 0 -#define ERAM_SIZE 0x00100000 /* 1MB */ - - -/* FPGA specific control registers. */ - -/* Writing a non-zero value to FLGREG(0) will signal the controlling CPU - to stop execution. */ -#define FLGREG_ADDR(n) (0xFFE80100 + 2*(n)) -#define FLGREG(n) (*(volatile unsigned char *)FLGREG_ADDR (n)) -#define FLGREG_NUM 2 - -#define CSDEV_ADDR(n) (0xFFE80110 + 2*(n)) -#define CSDEV(n) (*(volatile unsigned char *)CSDEV_ADDR (n)) - - -/* Timer interrupts 0-3, interrupt at intervals from CLK/4096 to CLK/16384. */ -#define IRQ_RPU(n) (60 + (n)) -#define IRQ_RPU_NUM 4 - -/* For */ -#define NUM_CPU_IRQS 64 - - -/* General-purpose timer. */ -/* control/status register (can only be read/written via bit insns) */ -#define RPU_GTMC_ADDR 0xFFFFFB00 -#define RPU_GTMC (*(volatile unsigned char *)RPU_GTMC_ADDR) -#define RPU_GTMC_CE_BIT 7 /* clock enable (control) */ -#define RPU_GTMC_OV_BIT 6 /* overflow (status) */ -#define RPU_GTMC_CLK_BIT 1 /* 0 = .5 MHz CLK, 1 = 1 Mhz (control) */ -/* 32-bit count (8 least-significant bits are always zero). */ -#define RPU_GTM_ADDR 0xFFFFFB28 -#define RPU_GTM (*(volatile unsigned long *)RPU_GTMC_ADDR) - - -/* For */ -#define PAGE_OFFSET ERAM_ADDR /* minimum allocatable address */ - - -/* For */ -/* `R0 RAM', used for a few miscellaneous variables that must be accessible - using a load instruction relative to R0. The FPGA implementation - actually has no on-chip RAM, so we use part of main ram just after the - interrupt vectors. */ -#ifdef __ASSEMBLY__ -#define R0_RAM_ADDR lo(C_SYMBOL_NAME(_r0_ram)) -#else -extern char _r0_ram; -#define R0_RAM_ADDR ((unsigned long)&_r0_ram); -#endif - - -#endif /* __V850_FPGA85E2C_H__ */ diff --git a/include/asm-v850/futex.h b/include/asm-v850/futex.h deleted file mode 100644 index 6a332a9f099c..000000000000 --- a/include/asm-v850/futex.h +++ /dev/null @@ -1,6 +0,0 @@ -#ifndef _ASM_FUTEX_H -#define _ASM_FUTEX_H - -#include - -#endif diff --git a/include/asm-v850/gbus_int.h b/include/asm-v850/gbus_int.h deleted file mode 100644 index 0c4bce753c7e..000000000000 --- a/include/asm-v850/gbus_int.h +++ /dev/null @@ -1,97 +0,0 @@ -/* - * include/asm-v850/gbus_int.h -- Midas labs GBUS interrupt support - * - * Copyright (C) 2001,02 NEC Corporation - * Copyright (C) 2001,02 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_GBUS_INT_H__ -#define __V850_GBUS_INT_H__ - - -/* The GBUS interrupt interface has 32 interrupts shared among 4 - processor interrupts. The 32 GBUS interrupts are divided into two - sets of 16 each, for allocating among control registers, etc (there - are two of each control register, with bits 0-15 controlling an - interrupt each). */ - -/* The GBUS interrupts themselves. */ -#define IRQ_GBUS_INT(n) (GBUS_INT_BASE_IRQ + (n)) -#define IRQ_GBUS_INT_NUM 32 - -/* Control registers. */ -#define GBUS_INT_STATUS_ADDR(w) (GBUS_INT_BASE_ADDR + (w)*0x40) -#define GBUS_INT_STATUS(w) (*(volatile u16 *)GBUS_INT_STATUS_ADDR(w)) -#define GBUS_INT_CLEAR_ADDR(w) (GBUS_INT_BASE_ADDR + 0x10 + (w)*0x40) -#define GBUS_INT_CLEAR(w) (*(volatile u16 *)GBUS_INT_CLEAR_ADDR(w)) -#define GBUS_INT_EDGE_ADDR(w) (GBUS_INT_BASE_ADDR + 0x20 + (w)*0x40) -#define GBUS_INT_EDGE(w) (*(volatile u16 *)GBUS_INT_EDGE_ADDR(w)) -#define GBUS_INT_POLARITY_ADDR(w) (GBUS_INT_BASE_ADDR + 0x30 + (w)*0x40) -#define GBUS_INT_POLARITY(w) (*(volatile u16 *)GBUS_INT_POLARITY_ADDR(w)) -/* This allows enabling interrupt bits in word W for interrupt GINTn. */ -#define GBUS_INT_ENABLE_ADDR(w, n) \ - (GBUS_INT_BASE_ADDR + 0x100 + (w)*0x10 + (n)*0x20) -#define GBUS_INT_ENABLE(w, n) (*(volatile u16 *)GBUS_INT_ENABLE_ADDR(w, n)) - -/* Mapping between kernel interrupt numbers and hardware control regs/bits. */ -#define GBUS_INT_BITS_PER_WORD 16 -#define GBUS_INT_NUM_WORDS (IRQ_GBUS_INT_NUM / GBUS_INT_BITS_PER_WORD) -#define GBUS_INT_IRQ_WORD(irq) (((irq) - GBUS_INT_BASE_IRQ) >> 4) -#define GBUS_INT_IRQ_BIT(irq) (((irq) - GBUS_INT_BASE_IRQ) & 0xF) -#define GBUS_INT_IRQ_MASK(irq) (1 << GBUS_INT_IRQ_BIT(irq)) - - -/* Possible priorities for GBUS interrupts. */ -#define GBUS_INT_PRIORITY_HIGH 2 -#define GBUS_INT_PRIORITY_MEDIUM 4 -#define GBUS_INT_PRIORITY_LOW 6 - - -#ifndef __ASSEMBLY__ - -/* Enable interrupt handling for interrupt IRQ. */ -extern void gbus_int_enable_irq (unsigned irq); -/* Disable interrupt handling for interrupt IRQ. Note that any - interrupts received while disabled will be delivered once the - interrupt is enabled again, unless they are explicitly cleared using - `gbus_int_clear_pending_irq'. */ -extern void gbus_int_disable_irq (unsigned irq); -/* Return true if interrupt handling for interrupt IRQ is enabled. */ -extern int gbus_int_irq_enabled (unsigned irq); -/* Disable all GBUS irqs. */ -extern void gbus_int_disable_irqs (void); -/* Clear any pending interrupts for IRQ. */ -extern void gbus_int_clear_pending_irq (unsigned irq); -/* Return true if interrupt IRQ is pending (but disabled). */ -extern int gbus_int_irq_pending (unsigned irq); - - -struct gbus_int_irq_init { - const char *name; /* name of interrupt type */ - - /* Range of kernel irq numbers for this type: - BASE, BASE+INTERVAL, ..., BASE+INTERVAL*NUM */ - unsigned base, num, interval; - - unsigned priority; /* interrupt priority to assign */ -}; -struct hw_interrupt_type; /* fwd decl */ - -/* Initialize HW_IRQ_TYPES for GBUS irqs described in array - INITS (which is terminated by an entry with the name field == 0). */ -extern void gbus_int_init_irq_types (struct gbus_int_irq_init *inits, - struct hw_interrupt_type *hw_irq_types); - -/* Initialize GBUS interrupts. */ -extern void gbus_int_init_irqs (void); - -#endif /* !__ASSEMBLY__ */ - - -#endif /* __V850_GBUS_INT_H__ */ diff --git a/include/asm-v850/hardirq.h b/include/asm-v850/hardirq.h deleted file mode 100644 index 04e20127c5af..000000000000 --- a/include/asm-v850/hardirq.h +++ /dev/null @@ -1,28 +0,0 @@ -#ifndef __V850_HARDIRQ_H__ -#define __V850_HARDIRQ_H__ - -#include -#include - -#include - -typedef struct { - unsigned int __softirq_pending; -} ____cacheline_aligned irq_cpustat_t; - -#include /* Standard mappings for irq_cpustat_t above */ - -#define HARDIRQ_BITS 8 - -/* - * The hardirq mask has to be large enough to have - * space for potentially all IRQ sources in the system - * nesting on a single CPU: - */ -#if (1 << HARDIRQ_BITS) < NR_IRQS -# error HARDIRQ_BITS is too low! -#endif - -void ack_bad_irq(unsigned int irq); - -#endif /* __V850_HARDIRQ_H__ */ diff --git a/include/asm-v850/highres_timer.h b/include/asm-v850/highres_timer.h deleted file mode 100644 index 486fb49ceab6..000000000000 --- a/include/asm-v850/highres_timer.h +++ /dev/null @@ -1,44 +0,0 @@ -/* - * include/asm-v850/highres_timer.h -- High resolution timing routines - * - * Copyright (C) 2001,03 NEC Electronics Corporation - * Copyright (C) 2001,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_HIGHRES_TIMER_H__ -#define __V850_HIGHRES_TIMER_H__ - -#ifndef __ASSEMBLY__ -#include -#endif - -#include - - -/* Frequency of the `slow ticks' (one tick each time the fast-tick - counter overflows). */ -#define HIGHRES_TIMER_SLOW_TICK_RATE 25 - -/* Which timer in the V850E `Timer D' we use. */ -#define HIGHRES_TIMER_TIMER_D_UNIT 3 - - -#ifndef __ASSEMBLY__ - -extern void highres_timer_start (void), highres_timer_stop (void); -extern void highres_timer_reset (void); -extern void highres_timer_read_ticks (u32 *slow_ticks, u32 *fast_ticks); -extern void highres_timer_ticks_to_timeval (u32 slow_ticks, u32 fast_ticks, - struct timeval *tv); -extern void highres_timer_read (struct timeval *tv); - -#endif /* !__ASSEMBLY__ */ - - -#endif /* __V850_HIGHRES_TIMER_H__ */ diff --git a/include/asm-v850/hw_irq.h b/include/asm-v850/hw_irq.h deleted file mode 100644 index 043e94bb6bd8..000000000000 --- a/include/asm-v850/hw_irq.h +++ /dev/null @@ -1,4 +0,0 @@ -#ifndef __V850_HW_IRQ_H__ -#define __V850_HW_IRQ_H__ - -#endif /* __V850_HW_IRQ_H__ */ diff --git a/include/asm-v850/io.h b/include/asm-v850/io.h deleted file mode 100644 index cdad251fba9f..000000000000 --- a/include/asm-v850/io.h +++ /dev/null @@ -1,142 +0,0 @@ -/* - * include/asm-v850/io.h -- Misc I/O operations - * - * Copyright (C) 2001,02,03,04,05 NEC Electronics Corporation - * Copyright (C) 2001,02,03,04,05 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_IO_H__ -#define __V850_IO_H__ - -#define IO_SPACE_LIMIT 0xFFFFFFFF - -#define readb(addr) \ - ({ unsigned char __v = (*(volatile unsigned char *) (addr)); __v; }) -#define readw(addr) \ - ({ unsigned short __v = (*(volatile unsigned short *) (addr)); __v; }) -#define readl(addr) \ - ({ unsigned long __v = (*(volatile unsigned long *) (addr)); __v; }) - -#define readb_relaxed(a) readb(a) -#define readw_relaxed(a) readw(a) -#define readl_relaxed(a) readl(a) - -#define writeb(val, addr) \ - (void)((*(volatile unsigned char *) (addr)) = (val)) -#define writew(val, addr) \ - (void)((*(volatile unsigned short *) (addr)) = (val)) -#define writel(val, addr) \ - (void)((*(volatile unsigned int *) (addr)) = (val)) - -#define __raw_readb readb -#define __raw_readw readw -#define __raw_readl readl -#define __raw_writeb writeb -#define __raw_writew writew -#define __raw_writel writel - -#define inb(addr) readb (addr) -#define inw(addr) readw (addr) -#define inl(addr) readl (addr) -#define outb(x, addr) ((void) writeb (x, addr)) -#define outw(x, addr) ((void) writew (x, addr)) -#define outl(x, addr) ((void) writel (x, addr)) - -#define inb_p(port) inb((port)) -#define outb_p(val, port) outb((val), (port)) -#define inw_p(port) inw((port)) -#define outw_p(val, port) outw((val), (port)) -#define inl_p(port) inl((port)) -#define outl_p(val, port) outl((val), (port)) - -static inline void insb (unsigned long port, void *dst, unsigned long count) -{ - unsigned char *p = dst; - while (count--) - *p++ = inb (port); -} -static inline void insw (unsigned long port, void *dst, unsigned long count) -{ - unsigned short *p = dst; - while (count--) - *p++ = inw (port); -} -static inline void insl (unsigned long port, void *dst, unsigned long count) -{ - unsigned long *p = dst; - while (count--) - *p++ = inl (port); -} - -static inline void -outsb (unsigned long port, const void *src, unsigned long count) -{ - const unsigned char *p = src; - while (count--) - outb (*p++, port); -} -static inline void -outsw (unsigned long port, const void *src, unsigned long count) -{ - const unsigned short *p = src; - while (count--) - outw (*p++, port); -} -static inline void -outsl (unsigned long port, const void *src, unsigned long count) -{ - const unsigned long *p = src; - while (count--) - outl (*p++, port); -} - - -/* Some places try to pass in an loff_t for PHYSADDR (?!), so we cast it to - long before casting it to a pointer to avoid compiler warnings. */ -#define ioremap(physaddr, size) ((void __iomem *)(unsigned long)(physaddr)) -#define iounmap(addr) ((void)0) - -#define ioremap_nocache(physaddr, size) ioremap (physaddr, size) -#define ioremap_writethrough(physaddr, size) ioremap (physaddr, size) -#define ioremap_fullcache(physaddr, size) ioremap (physaddr, size) - -#define ioread8(addr) readb (addr) -#define ioread16(addr) readw (addr) -#define ioread32(addr) readl (addr) -#define iowrite8(val, addr) writeb (val, addr) -#define iowrite16(val, addr) writew (val, addr) -#define iowrite32(val, addr) writel (val, addr) - -#define mmiowb() - -#define page_to_phys(page) ((page - mem_map) << PAGE_SHIFT) -#if 0 -/* This is really stupid; don't define it. */ -#define page_to_bus(page) page_to_phys (page) -#endif - -/* Conversion between virtual and physical mappings. */ -#define phys_to_virt(addr) ((void *)__phys_to_virt (addr)) -#define virt_to_phys(addr) ((unsigned long)__virt_to_phys (addr)) - -#define memcpy_fromio(dst, src, len) memcpy (dst, (void *)src, len) -#define memcpy_toio(dst, src, len) memcpy ((void *)dst, src, len) - -/* - * Convert a physical pointer to a virtual kernel pointer for /dev/mem - * access - */ -#define xlate_dev_mem_ptr(p) __va(p) - -/* - * Convert a virtual cached pointer to an uncached pointer - */ -#define xlate_dev_kmem_ptr(p) p - -#endif /* __V850_IO_H__ */ diff --git a/include/asm-v850/ioctl.h b/include/asm-v850/ioctl.h deleted file mode 100644 index b279fe06dfe5..000000000000 --- a/include/asm-v850/ioctl.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-v850/ioctls.h b/include/asm-v850/ioctls.h deleted file mode 100644 index 5313abd5f388..000000000000 --- a/include/asm-v850/ioctls.h +++ /dev/null @@ -1,84 +0,0 @@ -#ifndef __V850_IOCTLS_H__ -#define __V850_IOCTLS_H__ - -#include - -/* 0x54 is just a magic number to make these relatively unique ('T') */ - -#define TCGETS 0x5401 -#define TCSETS 0x5402 -#define TCSETSW 0x5403 -#define TCSETSF 0x5404 -#define TCGETA 0x5405 -#define TCSETA 0x5406 -#define TCSETAW 0x5407 -#define TCSETAF 0x5408 -#define TCSBRK 0x5409 -#define TCXONC 0x540A -#define TCFLSH 0x540B -#define TIOCEXCL 0x540C -#define TIOCNXCL 0x540D -#define TIOCSCTTY 0x540E -#define TIOCGPGRP 0x540F -#define TIOCSPGRP 0x5410 -#define TIOCOUTQ 0x5411 -#define TIOCSTI 0x5412 -#define TIOCGWINSZ 0x5413 -#define TIOCSWINSZ 0x5414 -#define TIOCMGET 0x5415 -#define TIOCMBIS 0x5416 -#define TIOCMBIC 0x5417 -#define TIOCMSET 0x5418 -#define TIOCGSOFTCAR 0x5419 -#define TIOCSSOFTCAR 0x541A -#define FIONREAD 0x541B -#define TIOCINQ FIONREAD -#define TIOCLINUX 0x541C -#define TIOCCONS 0x541D -#define TIOCGSERIAL 0x541E -#define TIOCSSERIAL 0x541F -#define TIOCPKT 0x5420 -#define FIONBIO 0x5421 -#define TIOCNOTTY 0x5422 -#define TIOCSETD 0x5423 -#define TIOCGETD 0x5424 -#define TCSBRKP 0x5425 /* Needed for POSIX tcsendbreak() */ -#define TIOCSBRK 0x5427 /* BSD compatibility */ -#define TIOCCBRK 0x5428 /* BSD compatibility */ -#define TIOCGSID 0x5429 /* Return the session ID of FD */ -#define TCGETS2 _IOR('T',0x2A, struct termios2) -#define TCSETS2 _IOW('T',0x2B, struct termios2) -#define TCSETSW2 _IOW('T',0x2C, struct termios2) -#define TCSETSF2 _IOW('T',0x2D, struct termios2) -#define TIOCGPTN _IOR('T',0x30, unsigned int) /* Get Pty Number (of pty-mux device) */ -#define TIOCSPTLCK _IOW('T',0x31, int) /* Lock/unlock Pty */ - -#define FIONCLEX 0x5450 /* these numbers need to be adjusted. */ -#define FIOCLEX 0x5451 -#define FIOASYNC 0x5452 -#define TIOCSERCONFIG 0x5453 -#define TIOCSERGWILD 0x5454 -#define TIOCSERSWILD 0x5455 -#define TIOCGLCKTRMIOS 0x5456 -#define TIOCSLCKTRMIOS 0x5457 -#define TIOCSERGSTRUCT 0x5458 /* For debugging only */ -#define TIOCSERGETLSR 0x5459 /* Get line status register */ -#define TIOCSERGETMULTI 0x545A /* Get multiport config */ -#define TIOCSERSETMULTI 0x545B /* Set multiport config */ - -#define TIOCMIWAIT 0x545C /* wait for a change on serial input line(s) */ -#define TIOCGICOUNT 0x545D /* read serial port inline interrupt counts */ -#define FIOQSIZE 0x545E - -/* Used for packet mode */ -#define TIOCPKT_DATA 0 -#define TIOCPKT_FLUSHREAD 1 -#define TIOCPKT_FLUSHWRITE 2 -#define TIOCPKT_STOP 4 -#define TIOCPKT_START 8 -#define TIOCPKT_NOSTOP 16 -#define TIOCPKT_DOSTOP 32 - -#define TIOCSER_TEMT 0x01 /* Transmitter physically empty */ - -#endif /* __V850_IOCTLS_H__ */ diff --git a/include/asm-v850/ipcbuf.h b/include/asm-v850/ipcbuf.h deleted file mode 100644 index d8cbe9886d95..000000000000 --- a/include/asm-v850/ipcbuf.h +++ /dev/null @@ -1,29 +0,0 @@ -#ifndef __V850E_IPCBUF_H__ -#define __V850E_IPCBUF_H__ - -/* - * The user_ipc_perm structure for v850e architecture. - * Note extra padding because this structure is passed back and forth - * between kernel and user space. - * - * Pad space is left for: - * - 32-bit mode_t and seq - * - 2 miscellaneous 32-bit values - */ - -struct ipc64_perm -{ - __kernel_key_t key; - __kernel_uid32_t uid; - __kernel_gid32_t gid; - __kernel_uid32_t cuid; - __kernel_gid32_t cgid; - __kernel_mode_t mode; - unsigned short __pad1; - unsigned short seq; - unsigned short __pad2; - unsigned long __unused1; - unsigned long __unused2; -}; - -#endif /* __V850E_IPCBUF_H__ */ diff --git a/include/asm-v850/irq.h b/include/asm-v850/irq.h deleted file mode 100644 index 7d0d4cd1ce54..000000000000 --- a/include/asm-v850/irq.h +++ /dev/null @@ -1,55 +0,0 @@ -/* - * include/asm-v850/irq.h -- Machine interrupt handling - * - * Copyright (C) 2001,02,04 NEC Electronics Corporation - * Copyright (C) 2001,02,04 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_IRQ_H__ -#define __V850_IRQ_H__ - -#include - -/* Default NUM_MACH_IRQS. */ -#ifndef NUM_MACH_IRQS -#define NUM_MACH_IRQS NUM_CPU_IRQS -#endif - -/* NMIs have IRQ numbers from FIRST_NMI to FIRST_NMI+NUM_NMIS-1. */ -#define FIRST_NMI NUM_MACH_IRQS -#define IRQ_NMI(n) (FIRST_NMI + (n)) -/* v850 processors have 3 non-maskable interrupts. */ -#define NUM_NMIS 3 - -/* Includes both maskable and non-maskable irqs. */ -#define NR_IRQS (NUM_MACH_IRQS + NUM_NMIS) - - -#ifndef __ASSEMBLY__ - -struct pt_regs; -struct hw_interrupt_type; -struct irqaction; - -#define irq_canonicalize(irq) (irq) - -/* Initialize irq handling for IRQs. - BASE_IRQ, BASE_IRQ+INTERVAL, ..., BASE_IRQ+NUM*INTERVAL - to IRQ_TYPE. An IRQ_TYPE of 0 means to use a generic interrupt type. */ -extern void -init_irq_handlers (int base_irq, int num, int interval, - struct hw_interrupt_type *irq_type); - -/* Handle interrupt IRQ. REGS are the registers at the time of ther - interrupt. */ -extern unsigned int handle_irq (int irq, struct pt_regs *regs); - -#endif /* !__ASSEMBLY__ */ - -#endif /* __V850_IRQ_H__ */ diff --git a/include/asm-v850/irq_regs.h b/include/asm-v850/irq_regs.h deleted file mode 100644 index 3dd9c0b70270..000000000000 --- a/include/asm-v850/irq_regs.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-v850/kdebug.h b/include/asm-v850/kdebug.h deleted file mode 100644 index 6ece1b037665..000000000000 --- a/include/asm-v850/kdebug.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-v850/kmap_types.h b/include/asm-v850/kmap_types.h deleted file mode 100644 index 3288976b161f..000000000000 --- a/include/asm-v850/kmap_types.h +++ /dev/null @@ -1,19 +0,0 @@ -#ifndef __V850_KMAP_TYPES_H__ -#define __V850_KMAP_TYPES_H__ - -enum km_type { - KM_BOUNCE_READ, - KM_SKB_SUNRPC_DATA, - KM_SKB_DATA_SOFTIRQ, - KM_USER0, - KM_USER1, - KM_BIO_SRC_IRQ, - KM_BIO_DST_IRQ, - KM_PTE0, - KM_PTE1, - KM_IRQ0, - KM_IRQ1, - KM_TYPE_NR -}; - -#endif /* __V850_KMAP_TYPES_H__ */ diff --git a/include/asm-v850/kvm.h b/include/asm-v850/kvm.h deleted file mode 100644 index 3f729b79febc..000000000000 --- a/include/asm-v850/kvm.h +++ /dev/null @@ -1,6 +0,0 @@ -#ifndef __LINUX_KVM_V850_H -#define __LINUX_KVM_V850_H - -/* v850 does not support KVM */ - -#endif diff --git a/include/asm-v850/linkage.h b/include/asm-v850/linkage.h deleted file mode 100644 index b6185d3cfe68..000000000000 --- a/include/asm-v850/linkage.h +++ /dev/null @@ -1,8 +0,0 @@ -#ifndef __ASM_LINKAGE_H -#define __ASM_LINKAGE_H - -#ifdef __ASSEMBLY__ -#include -#endif - -#endif diff --git a/include/asm-v850/local.h b/include/asm-v850/local.h deleted file mode 100644 index 705148abe276..000000000000 --- a/include/asm-v850/local.h +++ /dev/null @@ -1,6 +0,0 @@ -#ifndef __V850_LOCAL_H__ -#define __V850_LOCAL_H__ - -#include - -#endif /* __V850_LOCAL_H__ */ diff --git a/include/asm-v850/ma.h b/include/asm-v850/ma.h deleted file mode 100644 index 89e66473a176..000000000000 --- a/include/asm-v850/ma.h +++ /dev/null @@ -1,101 +0,0 @@ -/* - * include/asm-v850/ma.h -- V850E/MA series of cpu chips - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_MA_H__ -#define __V850_MA_H__ - -/* The MA series uses the V850E cpu core. */ -#include - - -/* For */ -/* We use on-chip RAM, for a few miscellaneous variables that must be - accessible using a load instruction relative to R0. The amount - varies between chip models, but there's always at least 4K, and it - should always start at FFFFC000. */ -#define R0_RAM_ADDR 0xFFFFC000 - - -/* MA series UART details. */ -#define V850E_UART_BASE_FREQ CPU_CLOCK_FREQ - -/* This is a function that gets called before configuring the UART. */ -#define V850E_UART_PRE_CONFIGURE ma_uart_pre_configure -#ifndef __ASSEMBLY__ -extern void ma_uart_pre_configure (unsigned chan, - unsigned cflags, unsigned baud); -#endif - - -/* MA series timer C details. */ -#define V850E_TIMER_C_BASE_ADDR 0xFFFFF600 - - -/* MA series timer D details. */ -#define V850E_TIMER_D_BASE_ADDR 0xFFFFF540 -#define V850E_TIMER_D_TMD_BASE_ADDR (V850E_TIMER_D_BASE_ADDR + 0x0) -#define V850E_TIMER_D_CMD_BASE_ADDR (V850E_TIMER_D_BASE_ADDR + 0x2) -#define V850E_TIMER_D_TMCD_BASE_ADDR (V850E_TIMER_D_BASE_ADDR + 0x4) - -#define V850E_TIMER_D_BASE_FREQ CPU_CLOCK_FREQ - - -/* Port 0 */ -/* Direct I/O. Bits 0-7 are pins P00-P07. */ -#define MA_PORT0_IO_ADDR 0xFFFFF400 -#define MA_PORT0_IO (*(volatile u8 *)MA_PORT0_IO_ADDR) -/* Port mode (for direct I/O, 0 = output, 1 = input). */ -#define MA_PORT0_PM_ADDR 0xFFFFF420 -#define MA_PORT0_PM (*(volatile u8 *)MA_PORT0_PM_ADDR) -/* Port mode control (0 = direct I/O mode, 1 = alternative I/O mode). */ -#define MA_PORT0_PMC_ADDR 0xFFFFF440 -#define MA_PORT0_PMC (*(volatile u8 *)MA_PORT0_PMC_ADDR) -/* Port function control (for P04-P07, 0 = IRQ, 1 = DMARQ). */ -#define MA_PORT0_PFC_ADDR 0xFFFFF460 -#define MA_PORT0_PFC (*(volatile u8 *)MA_PORT0_PFC_ADDR) - -/* Port 1 */ -/* Direct I/O. Bits 0-3 are pins P10-P13. */ -#define MA_PORT1_IO_ADDR 0xFFFFF402 -#define MA_PORT1_IO (*(volatile u8 *)MA_PORT1_IO_ADDR) -/* Port mode (for direct I/O, 0 = output, 1 = input). */ -#define MA_PORT1_PM_ADDR 0xFFFFF420 -#define MA_PORT1_PM (*(volatile u8 *)MA_PORT1_PM_ADDR) -/* Port mode control (0 = direct I/O mode, 1 = alternative I/O mode). */ -#define MA_PORT1_PMC_ADDR 0xFFFFF442 -#define MA_PORT1_PMC (*(volatile u8 *)MA_PORT1_PMC_ADDR) - -/* Port 4 */ -/* Direct I/O. Bits 0-5 are pins P40-P45. */ -#define MA_PORT4_IO_ADDR 0xFFFFF408 -#define MA_PORT4_IO (*(volatile u8 *)MA_PORT4_IO_ADDR) -/* Port mode (for direct I/O, 0 = output, 1 = input). */ -#define MA_PORT4_PM_ADDR 0xFFFFF428 -#define MA_PORT4_PM (*(volatile u8 *)MA_PORT4_PM_ADDR) -/* Port mode control (0 = direct I/O mode, 1 = alternative I/O mode). */ -#define MA_PORT4_PMC_ADDR 0xFFFFF448 -#define MA_PORT4_PMC (*(volatile u8 *)MA_PORT4_PMC_ADDR) -/* Port function control (for serial interfaces, 0 = CSI, 1 = UART). */ -#define MA_PORT4_PFC_ADDR 0xFFFFF468 -#define MA_PORT4_PFC (*(volatile u8 *)MA_PORT4_PFC_ADDR) - - -#ifndef __ASSEMBLY__ - -/* Initialize MA chip interrupts. */ -extern void ma_init_irqs (void); - -#endif /* !__ASSEMBLY__ */ - - -#endif /* __V850_MA_H__ */ diff --git a/include/asm-v850/ma1.h b/include/asm-v850/ma1.h deleted file mode 100644 index ede1f1de2b7a..000000000000 --- a/include/asm-v850/ma1.h +++ /dev/null @@ -1,50 +0,0 @@ -/* - * include/asm-v850/ma1.h -- V850E/MA1 cpu chip - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_MA1_H__ -#define __V850_MA1_H__ - -/* Inherit more generic details from MA series. */ -#include - - -#define CPU_MODEL "v850e/ma1" -#define CPU_MODEL_LONG "NEC V850E/MA1" - - -/* Hardware-specific interrupt numbers (in the kernel IRQ namespace). */ -#define IRQ_INTOV(n) (n) /* 0-3 */ -#define IRQ_INTOV_NUM 4 -#define IRQ_INTP(n) (0x4 + (n)) /* Pnnn (pin) interrupts */ -#define IRQ_INTP_NUM 24 -#define IRQ_INTCMD(n) (0x1c + (n)) /* interval timer interrupts 0-3 */ -#define IRQ_INTCMD_NUM 4 -#define IRQ_INTDMA(n) (0x20 + (n)) /* DMA interrupts 0-3 */ -#define IRQ_INTDMA_NUM 4 -#define IRQ_INTCSI(n) (0x24 + (n)*4)/* CSI 0-2 transmit/receive completion */ -#define IRQ_INTCSI_NUM 3 -#define IRQ_INTSER(n) (0x25 + (n)*4) /* UART 0-2 reception error */ -#define IRQ_INTSER_NUM 3 -#define IRQ_INTSR(n) (0x26 + (n)*4) /* UART 0-2 reception completion */ -#define IRQ_INTSR_NUM 3 -#define IRQ_INTST(n) (0x27 + (n)*4) /* UART 0-2 transmission completion */ -#define IRQ_INTST_NUM 3 - -#define NUM_CPU_IRQS 0x30 - - -/* The MA1 has a UART with 3 channels. */ -#define V850E_UART_NUM_CHANNELS 3 - - -#endif /* __V850_MA1_H__ */ diff --git a/include/asm-v850/machdep.h b/include/asm-v850/machdep.h deleted file mode 100644 index f1e3b8b91508..000000000000 --- a/include/asm-v850/machdep.h +++ /dev/null @@ -1,60 +0,0 @@ -/* - * include/asm-v850/machdep.h -- Machine-dependent definitions - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_MACHDEP_H__ -#define __V850_MACHDEP_H__ - - -/* chips */ -#ifdef CONFIG_V850E_MA1 -#include -#endif -#ifdef CONFIG_V850E_ME2 -#include -#endif -#ifdef CONFIG_V850E_TEG -#include -#endif - -/* These are both chips _and_ platforms, so put them in the middle... */ -#ifdef CONFIG_V850E2_ANNA -#include -#endif -#ifdef CONFIG_V850E_AS85EP1 -#include -#endif - -/* platforms */ -#ifdef CONFIG_RTE_CB_MA1 -#include -#endif -#ifdef CONFIG_RTE_CB_ME2 -#include -#endif -#ifdef CONFIG_RTE_CB_NB85E -#include -#endif -#ifdef CONFIG_V850E_SIM -#include -#endif -#ifdef CONFIG_V850E2_SIM85E2C -#include -#endif -#ifdef CONFIG_V850E2_SIM85E2S -#include -#endif -#ifdef CONFIG_V850E2_FPGA85E2C -#include -#endif - -#endif /* __V850_MACHDEP_H__ */ diff --git a/include/asm-v850/macrology.h b/include/asm-v850/macrology.h deleted file mode 100644 index 37abf874832c..000000000000 --- a/include/asm-v850/macrology.h +++ /dev/null @@ -1,17 +0,0 @@ -/* - * include/asm-v850/macrology.h -- Various useful CPP macros - * - * Copyright (C) 2001 NEC Corporation - * Copyright (C) 2001 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#define macrology_paste(arg1, arg2) macrology_paste_1(arg1, arg2) -#define macrology_paste_1(arg1, arg2) arg1 ## arg2 -#define macrology_stringify(sym) macrology_stringify_1(sym) -#define macrology_stringify_1(sym) #sym diff --git a/include/asm-v850/me2.h b/include/asm-v850/me2.h deleted file mode 100644 index ac7c9ce0bdc1..000000000000 --- a/include/asm-v850/me2.h +++ /dev/null @@ -1,182 +0,0 @@ -/* - * include/asm-v850/me2.h -- V850E/ME2 cpu chip - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_ME2_H__ -#define __V850_ME2_H__ - -#include -#include - - -#define CPU_MODEL "v850e/me2" -#define CPU_MODEL_LONG "NEC V850E/ME2" - - -/* Hardware-specific interrupt numbers (in the kernel IRQ namespace). */ -#define IRQ_INTP(n) (n) /* Pnnn (pin) interrupts */ -#define IRQ_INTP_NUM 31 -#define IRQ_INTCMD(n) (0x31 + (n)) /* interval timer interrupts 0-3 */ -#define IRQ_INTCMD_NUM 4 -#define IRQ_INTDMA(n) (0x41 + (n)) /* DMA interrupts 0-3 */ -#define IRQ_INTDMA_NUM 4 -#define IRQ_INTUBTIRE(n) (0x49 + (n)*5)/* UARTB 0-1 reception error */ -#define IRQ_INTUBTIRE_NUM 2 -#define IRQ_INTUBTIR(n) (0x4a + (n)*5) /* UARTB 0-1 reception complete */ -#define IRQ_INTUBTIR_NUM 2 -#define IRQ_INTUBTIT(n) (0x4b + (n)*5) /* UARTB 0-1 transmission complete */ -#define IRQ_INTUBTIT_NUM 2 -#define IRQ_INTUBTIF(n) (0x4c + (n)*5) /* UARTB 0-1 FIFO trans. complete */ -#define IRQ_INTUBTIF_NUM 2 -#define IRQ_INTUBTITO(n) (0x4d + (n)*5) /* UARTB 0-1 reception timeout */ -#define IRQ_INTUBTITO_NUM 2 - -/* For */ -#define NUM_CPU_IRQS 0x59 /* V850E/ME2 */ - - -/* For */ -/* We use on-chip RAM, for a few miscellaneous variables that must be - accessible using a load instruction relative to R0. */ -#define R0_RAM_ADDR 0xFFFFB000 /* V850E/ME2 */ - - -/* V850E/ME2 UARTB details.*/ -#define V850E_UART_NUM_CHANNELS 2 -#define V850E_UARTB_BASE_FREQ (CPU_CLOCK_FREQ / 4) - -/* This is a function that gets called before configuring the UART. */ -#define V850E_UART_PRE_CONFIGURE me2_uart_pre_configure -#ifndef __ASSEMBLY__ -extern void me2_uart_pre_configure (unsigned chan, - unsigned cflags, unsigned baud); -#endif /* __ASSEMBLY__ */ - - -/* V850E/ME2 timer C details. */ -#define V850E_TIMER_C_BASE_ADDR 0xFFFFF600 - - -/* V850E/ME2 timer D details. */ -#define V850E_TIMER_D_BASE_ADDR 0xFFFFF540 -#define V850E_TIMER_D_TMD_BASE_ADDR (V850E_TIMER_D_BASE_ADDR + 0x0) -#define V850E_TIMER_D_CMD_BASE_ADDR (V850E_TIMER_D_BASE_ADDR + 0x2) -#define V850E_TIMER_D_TMCD_BASE_ADDR (V850E_TIMER_D_BASE_ADDR + 0x4) - -#define V850E_TIMER_D_BASE_FREQ (CPU_CLOCK_FREQ / 2) - - -/* Select iRAM mode. */ -#define ME2_IRAMM_ADDR 0xFFFFF80A -#define ME2_IRAMM (*(volatile u8*)ME2_IRAMM_ADDR) - - -/* Interrupt edge-detection configuration. INTF(n) and INTR(n) are only - valid for n == 1, 2, or 5. */ -#define ME2_INTF_ADDR(n) (0xFFFFFC00 + (n) * 0x2) -#define ME2_INTF(n) (*(volatile u8*)ME2_INTF_ADDR(n)) -#define ME2_INTR_ADDR(n) (0xFFFFFC20 + (n) * 0x2) -#define ME2_INTR(n) (*(volatile u8*)ME2_INTR_ADDR(n)) -#define ME2_INTFAL_ADDR 0xFFFFFC10 -#define ME2_INTFAL (*(volatile u8*)ME2_INTFAL_ADDR) -#define ME2_INTRAL_ADDR 0xFFFFFC30 -#define ME2_INTRAL (*(volatile u8*)ME2_INTRAL_ADDR) -#define ME2_INTFDH_ADDR 0xFFFFFC16 -#define ME2_INTFDH (*(volatile u16*)ME2_INTFDH_ADDR) -#define ME2_INTRDH_ADDR 0xFFFFFC36 -#define ME2_INTRDH (*(volatile u16*)ME2_INTRDH_ADDR) -#define ME2_SESC_ADDR(n) (0xFFFFF609 + (n) * 0x10) -#define ME2_SESC(n) (*(volatile u8*)ME2_SESC_ADDR(n)) -#define ME2_SESA10_ADDR 0xFFFFF5AD -#define ME2_SESA10 (*(volatile u8*)ME2_SESA10_ADDR) -#define ME2_SESA11_ADDR 0xFFFFF5DD -#define ME2_SESA11 (*(volatile u8*)ME2_SESA11_ADDR) - - -/* Port 1 */ -/* Direct I/O. Bits 0-3 are pins P10-P13. */ -#define ME2_PORT1_IO_ADDR 0xFFFFF402 -#define ME2_PORT1_IO (*(volatile u8 *)ME2_PORT1_IO_ADDR) -/* Port mode (for direct I/O, 0 = output, 1 = input). */ -#define ME2_PORT1_PM_ADDR 0xFFFFF422 -#define ME2_PORT1_PM (*(volatile u8 *)ME2_PORT1_PM_ADDR) -/* Port mode control (0 = direct I/O mode, 1 = alternative I/O mode). */ -#define ME2_PORT1_PMC_ADDR 0xFFFFF442 -#define ME2_PORT1_PMC (*(volatile u8 *)ME2_PORT1_PMC_ADDR) -/* Port function control (for serial interfaces, 0 = CSI30, 1 = UARTB0 ). */ -#define ME2_PORT1_PFC_ADDR 0xFFFFF462 -#define ME2_PORT1_PFC (*(volatile u8 *)ME2_PORT1_PFC_ADDR) - -/* Port 2 */ -/* Direct I/O. Bits 0-3 are pins P20-P25. */ -#define ME2_PORT2_IO_ADDR 0xFFFFF404 -#define ME2_PORT2_IO (*(volatile u8 *)ME2_PORT2_IO_ADDR) -/* Port mode (for direct I/O, 0 = output, 1 = input). */ -#define ME2_PORT2_PM_ADDR 0xFFFFF424 -#define ME2_PORT2_PM (*(volatile u8 *)ME2_PORT2_PM_ADDR) -/* Port mode control (0 = direct I/O mode, 1 = alternative I/O mode). */ -#define ME2_PORT2_PMC_ADDR 0xFFFFF444 -#define ME2_PORT2_PMC (*(volatile u8 *)ME2_PORT2_PMC_ADDR) -/* Port function control (for serial interfaces, 0 = INTP2x, 1 = UARTB1 ). */ -#define ME2_PORT2_PFC_ADDR 0xFFFFF464 -#define ME2_PORT2_PFC (*(volatile u8 *)ME2_PORT2_PFC_ADDR) - -/* Port 5 */ -/* Direct I/O. Bits 0-5 are pins P50-P55. */ -#define ME2_PORT5_IO_ADDR 0xFFFFF40A -#define ME2_PORT5_IO (*(volatile u8 *)ME2_PORT5_IO_ADDR) -/* Port mode (for direct I/O, 0 = output, 1 = input). */ -#define ME2_PORT5_PM_ADDR 0xFFFFF42A -#define ME2_PORT5_PM (*(volatile u8 *)ME2_PORT5_PM_ADDR) -/* Port mode control (0 = direct I/O mode, 1 = alternative I/O mode). */ -#define ME2_PORT5_PMC_ADDR 0xFFFFF44A -#define ME2_PORT5_PMC (*(volatile u8 *)ME2_PORT5_PMC_ADDR) -/* Port function control (). */ -#define ME2_PORT5_PFC_ADDR 0xFFFFF46A -#define ME2_PORT5_PFC (*(volatile u8 *)ME2_PORT5_PFC_ADDR) - -/* Port 6 */ -/* Direct I/O. Bits 5-7 are pins P65-P67. */ -#define ME2_PORT6_IO_ADDR 0xFFFFF40C -#define ME2_PORT6_IO (*(volatile u8 *)ME2_PORT6_IO_ADDR) -/* Port mode (for direct I/O, 0 = output, 1 = input). */ -#define ME2_PORT6_PM_ADDR 0xFFFFF42C -#define ME2_PORT6_PM (*(volatile u8 *)ME2_PORT6_PM_ADDR) -/* Port mode control (0 = direct I/O mode, 1 = alternative I/O mode). */ -#define ME2_PORT6_PMC_ADDR 0xFFFFF44C -#define ME2_PORT6_PMC (*(volatile u8 *)ME2_PORT6_PMC_ADDR) -/* Port function control (). */ -#define ME2_PORT6_PFC_ADDR 0xFFFFF46C -#define ME2_PORT6_PFC (*(volatile u8 *)ME2_PORT6_PFC_ADDR) - -/* Port 7 */ -/* Direct I/O. Bits 2-7 are pins P72-P77. */ -#define ME2_PORT7_IO_ADDR 0xFFFFF40E -#define ME2_PORT7_IO (*(volatile u8 *)ME2_PORT7_IO_ADDR) -/* Port mode (for direct I/O, 0 = output, 1 = input). */ -#define ME2_PORT7_PM_ADDR 0xFFFFF42E -#define ME2_PORT7_PM (*(volatile u8 *)ME2_PORT7_PM_ADDR) -/* Port mode control (0 = direct I/O mode, 1 = alternative I/O mode). */ -#define ME2_PORT7_PMC_ADDR 0xFFFFF44E -#define ME2_PORT7_PMC (*(volatile u8 *)ME2_PORT7_PMC_ADDR) -/* Port function control (). */ -#define ME2_PORT7_PFC_ADDR 0xFFFFF46E -#define ME2_PORT7_PFC (*(volatile u8 *)ME2_PORT7_PFC_ADDR) - - -#ifndef __ASSEMBLY__ -/* Initialize V850E/ME2 chip interrupts. */ -extern void me2_init_irqs (void); -#endif /* !__ASSEMBLY__ */ - - -#endif /* __V850_ME2_H__ */ diff --git a/include/asm-v850/mman.h b/include/asm-v850/mman.h deleted file mode 100644 index edbf6edbfb37..000000000000 --- a/include/asm-v850/mman.h +++ /dev/null @@ -1,15 +0,0 @@ -#ifndef __V850_MMAN_H__ -#define __V850_MMAN_H__ - -#include - -#define MAP_GROWSDOWN 0x0100 /* stack-like segment */ -#define MAP_DENYWRITE 0x0800 /* ETXTBSY */ -#define MAP_EXECUTABLE 0x1000 /* mark it as an executable */ -#define MAP_LOCKED 0x2000 /* pages are locked */ -#define MAP_NORESERVE 0x4000 /* don't check for reservations */ - -#define MCL_CURRENT 1 /* lock all current mappings */ -#define MCL_FUTURE 2 /* lock all future mappings */ - -#endif /* __V850_MMAN_H__ */ diff --git a/include/asm-v850/mmu.h b/include/asm-v850/mmu.h deleted file mode 100644 index 267768c66ef6..000000000000 --- a/include/asm-v850/mmu.h +++ /dev/null @@ -1,11 +0,0 @@ -/* Copyright (C) 2002, 2005, David McCullough */ - -#ifndef __V850_MMU_H__ -#define __V850_MMU_H__ - -typedef struct { - struct vm_list_struct *vmlist; - unsigned long end_brk; -} mm_context_t; - -#endif /* __V850_MMU_H__ */ diff --git a/include/asm-v850/mmu_context.h b/include/asm-v850/mmu_context.h deleted file mode 100644 index 01daacd5474e..000000000000 --- a/include/asm-v850/mmu_context.h +++ /dev/null @@ -1,13 +0,0 @@ -#ifndef __V850_MMU_CONTEXT_H__ -#define __V850_MMU_CONTEXT_H__ - -#include - -#define destroy_context(mm) ((void)0) -#define init_new_context(tsk,mm) 0 -#define switch_mm(prev,next,tsk) ((void)0) -#define deactivate_mm(tsk,mm) do { } while (0) -#define activate_mm(prev,next) ((void)0) -#define enter_lazy_tlb(mm,tsk) ((void)0) - -#endif /* __V850_MMU_CONTEXT_H__ */ diff --git a/include/asm-v850/module.h b/include/asm-v850/module.h deleted file mode 100644 index 2c2f4944f09f..000000000000 --- a/include/asm-v850/module.h +++ /dev/null @@ -1,62 +0,0 @@ -/* - * include/asm-v850/module.h -- Architecture-specific module hooks - * - * Copyright (C) 2001,02,03,04 NEC Corporation - * Copyright (C) 2001,02,03,04 Miles Bader - * Copyright (C) 2001,03 Rusty Russell - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - * - * Derived in part from include/asm-ppc/module.h - */ - -#ifndef __V850_MODULE_H__ -#define __V850_MODULE_H__ - -#define MODULE_SYMBOL_PREFIX "_" - -struct v850_plt_entry -{ - /* Indirect jump instruction sequence (6-byte mov + 2-byte jr). */ - unsigned long tramp[2]; -}; - -struct mod_arch_specific -{ - /* Indices of PLT sections within module. */ - unsigned int core_plt_section, init_plt_section; -}; - -#define Elf_Shdr Elf32_Shdr -#define Elf_Sym Elf32_Sym -#define Elf_Ehdr Elf32_Ehdr - -/* Make empty sections for module_frob_arch_sections to expand. */ -#ifdef MODULE -asm(".section .plt,\"ax\",@nobits; .align 3; .previous"); -asm(".section .init.plt,\"ax\",@nobits; .align 3; .previous"); -#endif - -/* We don't do exception tables. */ -struct exception_table_entry; -static inline const struct exception_table_entry * -search_extable(const struct exception_table_entry *first, - const struct exception_table_entry *last, - unsigned long value) -{ - return 0; -} -#define ARCH_HAS_SEARCH_EXTABLE -static inline void -sort_extable(struct exception_table_entry *start, - struct exception_table_entry *finish) -{ - /* nada */ -} -#define ARCH_HAS_SORT_EXTABLE - -#endif /* __V850_MODULE_H__ */ diff --git a/include/asm-v850/msgbuf.h b/include/asm-v850/msgbuf.h deleted file mode 100644 index ed07dbd01637..000000000000 --- a/include/asm-v850/msgbuf.h +++ /dev/null @@ -1,31 +0,0 @@ -#ifndef __V850_MSGBUF_H__ -#define __V850_MSGBUF_H__ - -/* - * The msqid64_ds structure for v850 architecture. - * Note extra padding because this structure is passed back and forth - * between kernel and user space. - * - * Pad space is left for: - * - 64-bit time_t to solve y2038 problem - * - 2 miscellaneous 32-bit values - */ - -struct msqid64_ds { - struct ipc64_perm msg_perm; - __kernel_time_t msg_stime; /* last msgsnd time */ - unsigned long __unused1; - __kernel_time_t msg_rtime; /* last msgrcv time */ - unsigned long __unused2; - __kernel_time_t msg_ctime; /* last change time */ - unsigned long __unused3; - unsigned long msg_cbytes; /* current number of bytes on queue */ - unsigned long msg_qnum; /* number of messages in queue */ - unsigned long msg_qbytes; /* max number of bytes on queue */ - __kernel_pid_t msg_lspid; /* pid of last msgsnd */ - __kernel_pid_t msg_lrpid; /* last receive pid */ - unsigned long __unused4; - unsigned long __unused5; -}; - -#endif /* __V850_MSGBUF_H__ */ diff --git a/include/asm-v850/mutex.h b/include/asm-v850/mutex.h deleted file mode 100644 index 458c1f7fbc18..000000000000 --- a/include/asm-v850/mutex.h +++ /dev/null @@ -1,9 +0,0 @@ -/* - * Pull in the generic implementation for the mutex fastpath. - * - * TODO: implement optimized primitives instead, or leave the generic - * implementation in place, or pick the atomic_xchg() based generic - * implementation. (see asm-generic/mutex-xchg.h for details) - */ - -#include diff --git a/include/asm-v850/page.h b/include/asm-v850/page.h deleted file mode 100644 index f9de35d873fa..000000000000 --- a/include/asm-v850/page.h +++ /dev/null @@ -1,124 +0,0 @@ -/* - * include/asm-v850/page.h -- VM ops - * - * Copyright (C) 2001,02,03,05 NEC Electronics Corporation - * Copyright (C) 2001,02,03,05 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_PAGE_H__ -#define __V850_PAGE_H__ - -#include - - -#define PAGE_SHIFT 12 -#define PAGE_SIZE (1UL << PAGE_SHIFT) -#define PAGE_MASK (~(PAGE_SIZE-1)) - - -/* - * PAGE_OFFSET -- the first address of the first page of memory. For archs with - * no MMU this corresponds to the first free page in physical memory (aligned - * on a page boundary). - */ -#ifndef PAGE_OFFSET -#define PAGE_OFFSET 0x0000000 -#endif - - -#ifndef __ASSEMBLY__ - -#define STRICT_MM_TYPECHECKS - -#define clear_page(page) memset ((void *)(page), 0, PAGE_SIZE) -#define copy_page(to, from) memcpy ((void *)(to), (void *)from, PAGE_SIZE) - -#define clear_user_page(addr, vaddr, page) \ - do { clear_page(addr); \ - flush_dcache_page(page); \ - } while (0) -#define copy_user_page(to, from, vaddr, page) \ - do { copy_page(to, from); \ - flush_dcache_page(page); \ - } while (0) - -#ifdef STRICT_MM_TYPECHECKS -/* - * These are used to make use of C type-checking.. - */ - -typedef struct { unsigned long pte; } pte_t; -typedef struct { unsigned long pmd; } pmd_t; -typedef struct { unsigned long pgd; } pgd_t; -typedef struct { unsigned long pgprot; } pgprot_t; -typedef struct page *pgtable_t; - -#define pte_val(x) ((x).pte) -#define pmd_val(x) ((x).pmd) -#define pgd_val(x) ((x).pgd) -#define pgprot_val(x) ((x).pgprot) - -#define __pte(x) ((pte_t) { (x) } ) -#define __pmd(x) ((pmd_t) { (x) } ) -#define __pgd(x) ((pgd_t) { (x) } ) -#define __pgprot(x) ((pgprot_t) { (x) } ) - -#else /* !STRICT_MM_TYPECHECKS */ -/* - * .. while these make it easier on the compiler - */ - -typedef unsigned long pte_t; -typedef unsigned long pmd_t; -typedef unsigned long pgd_t; -typedef unsigned long pgprot_t; - -#define pte_val(x) (x) -#define pmd_val(x) (x) -#define pgd_val(x) (x) -#define pgprot_val(x) (x) - -#define __pte(x) (x) -#define __pmd(x) (x) -#define __pgd(x) (x) -#define __pgprot(x) (x) - -#endif /* STRICT_MM_TYPECHECKS */ - -#endif /* !__ASSEMBLY__ */ - - -/* No current v850 processor has virtual memory. */ -#define __virt_to_phys(addr) (addr) -#define __phys_to_virt(addr) (addr) - -#define virt_to_pfn(kaddr) (__virt_to_phys (kaddr) >> PAGE_SHIFT) -#define pfn_to_virt(pfn) __phys_to_virt ((pfn) << PAGE_SHIFT) - -#define MAP_NR(kaddr) \ - (((unsigned long)(kaddr) - PAGE_OFFSET) >> PAGE_SHIFT) -#define virt_to_page(kaddr) (mem_map + MAP_NR (kaddr)) -#define page_to_virt(page) \ - ((((page) - mem_map) << PAGE_SHIFT) + PAGE_OFFSET) - -#define ARCH_PFN_OFFSET (PAGE_OFFSET >> PAGE_SHIFT) -#define pfn_valid(pfn) ((pfn) < max_mapnr) - -#define virt_addr_valid(kaddr) \ - (((void *)(kaddr) >= (void *)PAGE_OFFSET) && MAP_NR (kaddr) < max_mapnr) - - -#define __pa(x) __virt_to_phys ((unsigned long)(x)) -#define __va(x) ((void *)__phys_to_virt ((unsigned long)(x))) - - -#include -#include - -#endif /* __V850_PAGE_H__ */ diff --git a/include/asm-v850/param.h b/include/asm-v850/param.h deleted file mode 100644 index 4391f5fe0204..000000000000 --- a/include/asm-v850/param.h +++ /dev/null @@ -1,33 +0,0 @@ -/* - * include/asm-v850/param.h -- Varions kernel parameters - * - * Copyright (C) 2001,02 NEC Corporation - * Copyright (C) 2001,02 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_PARAM_H__ -#define __V850_PARAM_H__ - -#define EXEC_PAGESIZE 4096 - -#ifndef NOGROUP -#define NOGROUP (-1) -#endif - -#define MAXHOSTNAMELEN 64 /* max length of hostname */ - -#ifdef __KERNEL__ -# define HZ CONFIG_HZ -# define USER_HZ 100 -# define CLOCKS_PER_SEC USER_HZ -#else -# define HZ 100 -#endif - -#endif /* __V850_PARAM_H__ */ diff --git a/include/asm-v850/pci.h b/include/asm-v850/pci.h deleted file mode 100644 index de2a7d0a81cc..000000000000 --- a/include/asm-v850/pci.h +++ /dev/null @@ -1,119 +0,0 @@ -/* - * include/asm-v850/pci.h -- PCI support - * - * Copyright (C) 2001,02,05 NEC Corporation - * Copyright (C) 2001,02,05 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_PCI_H__ -#define __V850_PCI_H__ - -/* Get any platform-dependent definitions. */ -#include - -#define pcibios_scan_all_fns(a, b) 0 - -/* Generic declarations. */ - -struct scatterlist; - -extern void pcibios_set_master (struct pci_dev *dev); - -/* `Grant' to PDEV the memory block at CPU_ADDR, for doing DMA. The - 32-bit PCI bus mastering address to use is returned. the device owns - this memory until either pci_unmap_single or pci_dma_sync_single_for_cpu is - performed. */ -extern dma_addr_t -pci_map_single (struct pci_dev *pdev, void *cpu_addr, size_t size, int dir); - -/* Return to the CPU the PCI DMA memory block previously `granted' to - PDEV, at DMA_ADDR. */ -extern void -pci_unmap_single (struct pci_dev *pdev, dma_addr_t dma_addr, size_t size, - int dir); - -/* Make physical memory consistent for a single streaming mode DMA - translation after a transfer. - - If you perform a pci_map_single() but wish to interrogate the - buffer using the cpu, yet do not wish to teardown the PCI dma - mapping, you must call this function before doing so. At the next - point you give the PCI dma address back to the card, you must first - perform a pci_dma_sync_for_device, and then the device again owns - the buffer. */ -extern void -pci_dma_sync_single_for_cpu (struct pci_dev *dev, dma_addr_t dma_addr, - size_t size, int dir); - -extern void -pci_dma_sync_single_for_device (struct pci_dev *dev, dma_addr_t dma_addr, - size_t size, int dir); - - -/* Do multiple DMA mappings at once. */ -extern int -pci_map_sg (struct pci_dev *pdev, struct scatterlist *sg, int sg_len, int dir); - -/* Unmap multiple DMA mappings at once. */ -extern void -pci_unmap_sg (struct pci_dev *pdev, struct scatterlist *sg, int sg_len, - int dir); - -/* SG-list versions of pci_dma_sync functions. */ -extern void -pci_dma_sync_sg_for_cpu (struct pci_dev *dev, - struct scatterlist *sg, int sg_len, - int dir); -extern void -pci_dma_sync_sg_for_device (struct pci_dev *dev, - struct scatterlist *sg, int sg_len, - int dir); - -#define pci_map_page(dev, page, offs, size, dir) \ - pci_map_single(dev, (page_address(page) + (offs)), size, dir) -#define pci_unmap_page(dev,addr,sz,dir) \ - pci_unmap_single(dev, addr, sz, dir) - -/* Test for pci_map_single or pci_map_page having generated an error. */ -static inline int -pci_dma_mapping_error (dma_addr_t dma_addr) -{ - return dma_addr == 0; -} - -/* Allocate and map kernel buffer using consistent mode DMA for PCI - device. Returns non-NULL cpu-view pointer to the buffer if - successful and sets *DMA_ADDR to the pci side dma address as well, - else DMA_ADDR is undefined. */ -extern void * -pci_alloc_consistent (struct pci_dev *pdev, size_t size, dma_addr_t *dma_addr); - -/* Free and unmap a consistent DMA buffer. CPU_ADDR and DMA_ADDR must - be values that were returned from pci_alloc_consistent. SIZE must be - the same as what as passed into pci_alloc_consistent. References to - the memory and mappings assosciated with CPU_ADDR or DMA_ADDR past - this call are illegal. */ -extern void -pci_free_consistent (struct pci_dev *pdev, size_t size, void *cpu_addr, - dma_addr_t dma_addr); - -#ifdef CONFIG_PCI -static inline void pci_dma_burst_advice(struct pci_dev *pdev, - enum pci_dma_burst_strategy *strat, - unsigned long *strategy_parameter) -{ - *strat = PCI_DMA_BURST_INFINITY; - *strategy_parameter = ~0UL; -} -#endif - -extern void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long max); -extern void pci_iounmap (struct pci_dev *dev, void __iomem *addr); - -#endif /* __V850_PCI_H__ */ diff --git a/include/asm-v850/percpu.h b/include/asm-v850/percpu.h deleted file mode 100644 index 755ac6522b63..000000000000 --- a/include/asm-v850/percpu.h +++ /dev/null @@ -1,14 +0,0 @@ -#ifndef __V850_PERCPU_H__ -#define __V850_PERCPU_H__ - -#include - -/* This is a stupid hack to satisfy some grotty implicit include-file - dependency; basically, uses BUG_ON, which calls BUG, but - doesn't include the necessary headers to define it. In the twisted - festering mess of includes this must all be resolved somehow on other - platforms, but I haven't the faintest idea how, and don't care; here will - do, even though doesn't actually make any sense. */ -#include - -#endif /* __V850_PERCPU_H__ */ diff --git a/include/asm-v850/pgalloc.h b/include/asm-v850/pgalloc.h deleted file mode 100644 index b91eb2d02bfd..000000000000 --- a/include/asm-v850/pgalloc.h +++ /dev/null @@ -1,22 +0,0 @@ -/* - * include/asm-v850/pgalloc.h - * - * Copyright (C) 2001,02 NEC Corporation - * Copyright (C) 2001,02 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_PGALLOC_H__ -#define __V850_PGALLOC_H__ - -#include /* some crap code expects this */ - -/* ... and then, there was one. */ -#define check_pgt_cache() ((void)0) - -#endif /* __V850_PGALLOC_H__ */ diff --git a/include/asm-v850/pgtable.h b/include/asm-v850/pgtable.h deleted file mode 100644 index 1ea2a900f0f8..000000000000 --- a/include/asm-v850/pgtable.h +++ /dev/null @@ -1,59 +0,0 @@ -#ifndef __V850_PGTABLE_H__ -#define __V850_PGTABLE_H__ - -#include - -#include - - -#define pgd_present(pgd) (1) /* pages are always present on NO_MM */ -#define pgd_none(pgd) (0) -#define pgd_bad(pgd) (0) -#define pgd_clear(pgdp) ((void)0) - -#define pmd_offset(a, b) ((void *)0) - -#define kern_addr_valid(addr) (1) - - -#define __swp_type(x) (0) -#define __swp_offset(x) (0) -#define __swp_entry(typ,off) ((swp_entry_t) { ((typ) | ((off) << 7)) }) -#define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) }) -#define __swp_entry_to_pte(x) ((pte_t) { (x).val }) - -static inline int pte_file (pte_t pte) { return 0; } - - -/* These mean nothing to !CONFIG_MMU. */ -#define PAGE_NONE __pgprot(0) -#define PAGE_SHARED __pgprot(0) -#define PAGE_COPY __pgprot(0) -#define PAGE_READONLY __pgprot(0) -#define PAGE_KERNEL __pgprot(0) - - -/* - * ZERO_PAGE is a global shared page that is always zero: used - * for zero-mapped memory areas etc. When CONFIG_MMU is not defined, this - * should never actually be used, so just define it to something that's - * will hopefully cause a bus error if it is. - */ -#define ZERO_PAGE(vaddr) ((void *)0x87654321) - - -/* Some bogus code in procfs uses these; whatever. */ -#define VMALLOC_START 0 -#define VMALLOC_END (~0) - - -extern void paging_init (void); -#define swapper_pg_dir ((pgd_t *) 0) - -#define pgtable_cache_init() ((void)0) - - -extern unsigned int kobjsize(const void *objp); - - -#endif /* __V850_PGTABLE_H__ */ diff --git a/include/asm-v850/poll.h b/include/asm-v850/poll.h deleted file mode 100644 index 803cad0b9b59..000000000000 --- a/include/asm-v850/poll.h +++ /dev/null @@ -1,9 +0,0 @@ -#ifndef __V850_POLL_H__ -#define __V850_POLL_H__ - -#define POLLWRNORM POLLOUT -#define POLLWRBAND 0x0100 - -#include - -#endif /* __V850_POLL_H__ */ diff --git a/include/asm-v850/posix_types.h b/include/asm-v850/posix_types.h deleted file mode 100644 index 7f403b765390..000000000000 --- a/include/asm-v850/posix_types.h +++ /dev/null @@ -1,72 +0,0 @@ -/* - * include/asm-v850/posix_types.h -- Kernel versions of standard types - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_POSIX_TYPES_H__ -#define __V850_POSIX_TYPES_H__ - -typedef unsigned long __kernel_ino_t; -typedef unsigned long long __kernel_ino64_t; -typedef unsigned int __kernel_mode_t; -typedef unsigned int __kernel_nlink_t; -typedef long __kernel_off_t; -typedef long long __kernel_loff_t; -typedef int __kernel_pid_t; -typedef unsigned short __kernel_ipc_pid_t; -typedef unsigned int __kernel_uid_t; -typedef unsigned int __kernel_gid_t; -typedef unsigned int __kernel_size_t; -typedef int __kernel_ssize_t; -typedef int __kernel_ptrdiff_t; -typedef long __kernel_time_t; -typedef long __kernel_suseconds_t; -typedef long __kernel_clock_t; -typedef int __kernel_timer_t; -typedef int __kernel_clockid_t; -typedef int __kernel_daddr_t; -typedef char * __kernel_caddr_t; -typedef unsigned short __kernel_uid16_t; -typedef unsigned short __kernel_gid16_t; -typedef unsigned int __kernel_uid32_t; -typedef unsigned int __kernel_gid32_t; - -/* Some bogus code depends on this; we don't care. */ -typedef __kernel_uid_t __kernel_old_uid_t; -typedef unsigned int __kernel_old_dev_t; - -typedef struct { - int val[2]; -} __kernel_fsid_t; - - -#if defined(__KERNEL__) - -/* We used to include here, which seems the right thing, but - it caused nasty include-file definition order problems. Removing the - include seems to work, so fingers crossed... */ - -#undef __FD_SET -#define __FD_SET(fd, fd_set) \ - __set_bit (fd, (void *)&((__kernel_fd_set *)fd_set)->fds_bits) -#undef __FD_CLR -#define __FD_CLR(fd, fd_set) \ - __clear_bit (fd, (void *)&((__kernel_fd_set *)fd_set)->fds_bits) -#undef __FD_ISSET -#define __FD_ISSET(fd, fd_set) \ - __test_bit (fd, (void *)&((__kernel_fd_set *)fd_set)->fds_bits) -#undef __FD_ZERO -#define __FD_ZERO(fd_set) \ - memset (fd_set, 0, sizeof (*(fd_set *)fd_set)) - -#endif /* defined(__KERNEL__) */ - -#endif /* __V850_POSIX_TYPES_H__ */ diff --git a/include/asm-v850/processor.h b/include/asm-v850/processor.h deleted file mode 100644 index 979e3467f9af..000000000000 --- a/include/asm-v850/processor.h +++ /dev/null @@ -1,120 +0,0 @@ -/* - * include/asm-v850/processor.h - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_PROCESSOR_H__ -#define __V850_PROCESSOR_H__ - -#ifndef __ASSEMBLY__ /* is not asm-safe. */ -#include -#endif - -#include -#include -#include - -/* Some code expects `segment' stuff to be defined here. */ -#include - - -/* - * The only places this is used seem to be horrible bletcherous kludges, - * so we just define it to be as large as possible. - */ -#define TASK_SIZE (0xFFFFFFFF) - -/* - * This decides where the kernel will search for a free chunk of vm - * space during mmap's. We won't be using it. - */ -#define TASK_UNMAPPED_BASE 0 - - -#ifndef __ASSEMBLY__ - - -/* - * Default implementation of macro that returns current - * instruction pointer ("program counter"). - */ -#define current_text_addr() ({ __label__ _l; _l: &&_l;}) - -/* If you change this, you must change the associated assembly-languages - constants defined below, THREAD_*. */ -struct thread_struct { - /* kernel stack pointer (must be first field in structure) */ - unsigned long ksp; -}; - -#define INIT_THREAD { sizeof init_stack + (unsigned long)init_stack } - - -/* Do necessary setup to start up a newly executed thread. */ -static inline void start_thread (struct pt_regs *regs, - unsigned long pc, unsigned long usp) -{ - regs->pc = pc; - regs->gpr[GPR_SP] = usp; - regs->kernel_mode = 0; -} - -/* Free all resources held by a thread. */ -static inline void release_thread (struct task_struct *dead_task) -{ -} - -/* Prepare to copy thread state - unlazy all lazy status */ -#define prepare_to_copy(tsk) do { } while (0) - -extern int kernel_thread (int (*fn)(void *), void * arg, unsigned long flags); - -/* Free current thread data structures etc. */ -static inline void exit_thread (void) -{ -} - - -/* Return the registers saved during context-switch by the currently - not-running thread T. Note that this only includes some registers! - See entry.S for details. */ -#define thread_saved_regs(t) \ - ((struct pt_regs*)((t)->thread.ksp + STATE_SAVE_PT_OFFSET)) -/* Return saved (kernel) PC of a blocked thread. Actually, we return the - LP register, because the thread is actually blocked in switch_thread, - and we're interested in the PC it will _return_ to. */ -#define thread_saved_pc(t) (thread_saved_regs(t)->gpr[GPR_LP]) - - -unsigned long get_wchan (struct task_struct *p); - - -/* Return some info about the user process TASK. */ -#define task_tos(task) ((unsigned long)task_stack_page(task) + THREAD_SIZE) -#define task_pt_regs(task) ((struct pt_regs *)task_tos (task) - 1) -#define task_sp(task) (task_pt_regs (task)->gpr[GPR_SP]) -#define task_pc(task) (task_pt_regs (task)->pc) -/* Grotty old names for some. */ -#define KSTK_EIP(task) task_pc (task) -#define KSTK_ESP(task) task_sp (task) - - -#define cpu_relax() barrier() - - -#else /* __ASSEMBLY__ */ - -#define THREAD_KSP 0 - -#endif /* !__ASSEMBLY__ */ - - -#endif /* __V850_PROCESSOR_H__ */ diff --git a/include/asm-v850/ptrace.h b/include/asm-v850/ptrace.h deleted file mode 100644 index 4f35cf2cd641..000000000000 --- a/include/asm-v850/ptrace.h +++ /dev/null @@ -1,121 +0,0 @@ -/* - * include/asm-v850/ptrace.h -- Access to CPU registers - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_PTRACE_H__ -#define __V850_PTRACE_H__ - - -/* v850 general purpose registers with special meanings. */ -#define GPR_ZERO 0 /* constant zero */ -#define GPR_ASM 1 /* reserved for assembler */ -#define GPR_SP 3 /* stack pointer */ -#define GPR_GP 4 /* global data pointer */ -#define GPR_TP 5 /* `text pointer' */ -#define GPR_EP 30 /* `element pointer' */ -#define GPR_LP 31 /* link pointer (current return address) */ - -/* These aren't official names, but they make some code more descriptive. */ -#define GPR_ARG0 6 -#define GPR_ARG1 7 -#define GPR_ARG2 8 -#define GPR_ARG3 9 -#define GPR_RVAL0 10 -#define GPR_RVAL1 11 -#define GPR_RVAL GPR_RVAL0 - -#define NUM_GPRS 32 - -/* v850 `system' registers. */ -#define SR_EIPC 0 -#define SR_EIPSW 1 -#define SR_FEPC 2 -#define SR_FEPSW 3 -#define SR_ECR 4 -#define SR_PSW 5 -#define SR_CTPC 16 -#define SR_CTPSW 17 -#define SR_DBPC 18 -#define SR_DBPSW 19 -#define SR_CTBP 20 -#define SR_DIR 21 -#define SR_ASID 23 - - -#ifndef __ASSEMBLY__ - -typedef unsigned long v850_reg_t; - -/* How processor state is stored on the stack during a syscall/signal. - If you change this structure, change the associated assembly-language - macros below too (PT_*)! */ -struct pt_regs -{ - /* General purpose registers. */ - v850_reg_t gpr[NUM_GPRS]; - - v850_reg_t pc; /* program counter */ - v850_reg_t psw; /* program status word */ - - /* Registers used by `callt' instruction: */ - v850_reg_t ctpc; /* saved program counter */ - v850_reg_t ctpsw; /* saved psw */ - v850_reg_t ctbp; /* base pointer for callt table */ - - char kernel_mode; /* 1 if in `kernel mode', 0 if user mode */ -}; - - -#define instruction_pointer(regs) ((regs)->pc) -#define profile_pc(regs) instruction_pointer(regs) -#define user_mode(regs) (!(regs)->kernel_mode) - -/* When a struct pt_regs is used to save user state for a system call in - the kernel, the system call is stored in the space for R0 (since it's - never used otherwise, R0 being a constant 0). Non-system-calls - simply store 0 there. */ -#define PT_REGS_SYSCALL(regs) (regs)->gpr[0] -#define PT_REGS_SET_SYSCALL(regs, val) ((regs)->gpr[0] = (val)) - -#endif /* !__ASSEMBLY__ */ - - -/* The number of bytes used to store each register. */ -#define _PT_REG_SIZE 4 - -/* Offset of a general purpose register in a struct pt_regs. */ -#define PT_GPR(num) ((num) * _PT_REG_SIZE) - -/* Offsets of various special registers & fields in a struct pt_regs. */ -#define PT_PC ((NUM_GPRS + 0) * _PT_REG_SIZE) -#define PT_PSW ((NUM_GPRS + 1) * _PT_REG_SIZE) -#define PT_CTPC ((NUM_GPRS + 2) * _PT_REG_SIZE) -#define PT_CTPSW ((NUM_GPRS + 3) * _PT_REG_SIZE) -#define PT_CTBP ((NUM_GPRS + 4) * _PT_REG_SIZE) -#define PT_KERNEL_MODE ((NUM_GPRS + 5) * _PT_REG_SIZE) - -/* Where the current syscall number is stashed; obviously only valid in - the kernel! */ -#define PT_CUR_SYSCALL PT_GPR(0) - -/* Size of struct pt_regs, including alignment. */ -#define PT_SIZE ((NUM_GPRS + 6) * _PT_REG_SIZE) - - -/* These are `magic' values for PTRACE_PEEKUSR that return info about where - a process is located in memory. */ -#define PT_TEXT_ADDR (PT_SIZE + 1) -#define PT_TEXT_LEN (PT_SIZE + 2) -#define PT_DATA_ADDR (PT_SIZE + 3) - - -#endif /* __V850_PTRACE_H__ */ diff --git a/include/asm-v850/resource.h b/include/asm-v850/resource.h deleted file mode 100644 index 4b9dcd44f8d1..000000000000 --- a/include/asm-v850/resource.h +++ /dev/null @@ -1,6 +0,0 @@ -#ifndef __V850_RESOURCE_H__ -#define __V850_RESOURCE_H__ - -#include - -#endif /* __V850_RESOURCE_H__ */ diff --git a/include/asm-v850/rte_cb.h b/include/asm-v850/rte_cb.h deleted file mode 100644 index db9879f00aa7..000000000000 --- a/include/asm-v850/rte_cb.h +++ /dev/null @@ -1,78 +0,0 @@ -/* - * include/asm-v850/rte_cb.h -- Midas labs RTE-CB series of evaluation boards - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_RTE_CB_H__ -#define __V850_RTE_CB_H__ - - -/* The SRAM on the Mother-A motherboard. */ -#define MB_A_SRAM_ADDR GCS0_ADDR -#define MB_A_SRAM_SIZE 0x00200000 /* 2MB */ - - -#ifdef CONFIG_RTE_GBUS_INT -/* GBUS interrupt support. */ - -# include - -# define GBUS_INT_BASE_IRQ NUM_RTE_CB_IRQS -# define GBUS_INT_BASE_ADDR (GCS2_ADDR + 0x00006000) - -/* Some specific interrupts. */ -# define IRQ_MB_A_LAN IRQ_GBUS_INT(10) -# define IRQ_MB_A_PCI1(n) (IRQ_GBUS_INT(16) + (n)) -# define IRQ_MB_A_PCI1_NUM 4 -# define IRQ_MB_A_PCI2(n) (IRQ_GBUS_INT(20) + (n)) -# define IRQ_MB_A_PCI2_NUM 4 -# define IRQ_MB_A_EXT(n) (IRQ_GBUS_INT(24) + (n)) -# define IRQ_MB_A_EXT_NUM 4 -# define IRQ_MB_A_USB_OC(n) (IRQ_GBUS_INT(28) + (n)) -# define IRQ_MB_A_USB_OC_NUM 2 -# define IRQ_MB_A_PCMCIA_OC IRQ_GBUS_INT(30) - -/* We define NUM_MACH_IRQS to include extra interrupts from the GBUS. */ -# define NUM_MACH_IRQS (NUM_RTE_CB_IRQS + IRQ_GBUS_INT_NUM) - -#else /* !CONFIG_RTE_GBUS_INT */ - -# define NUM_MACH_IRQS NUM_RTE_CB_IRQS - -#endif /* CONFIG_RTE_GBUS_INT */ - - -#ifdef CONFIG_RTE_MB_A_PCI -/* Mother-A PCI bus support. */ - -# include - -/* These are the base addresses used for allocating device address - space. 512K of the motherboard SRAM is in the same space, so we have - to be careful not to let it be allocated. */ -# define PCIBIOS_MIN_MEM (MB_A_PCI_MEM_ADDR + 0x80000) -# define PCIBIOS_MIN_IO MB_A_PCI_IO_ADDR - -/* As we don't really support PCI DMA to cpu memory, and use bounce-buffers - instead, perversely enough, this becomes always true! */ -# define pci_dma_supported(dev, mask) 1 -# define pcibios_assign_all_busses() 1 - -#endif /* CONFIG_RTE_MB_A_PCI */ - - -#ifndef __ASSEMBLY__ -extern void rte_cb_early_init (void); -extern void rte_cb_init_irqs (void); -#endif /* !__ASSEMBLY__ */ - - -#endif /* __V850_RTE_CB_H__ */ diff --git a/include/asm-v850/rte_ma1_cb.h b/include/asm-v850/rte_ma1_cb.h deleted file mode 100644 index bd3162ab9844..000000000000 --- a/include/asm-v850/rte_ma1_cb.h +++ /dev/null @@ -1,128 +0,0 @@ -/* - * include/asm-v850/rte_ma1_cb.h -- Midas labs RTE-V850/MA1-CB board - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_RTE_MA1_CB_H__ -#define __V850_RTE_MA1_CB_H__ - -#include /* Common defs for Midas RTE-CB boards. */ - - -#define PLATFORM "rte-v850e/ma1-cb" -#define PLATFORM_LONG "Midas lab RTE-V850E/MA1-CB" - -#define CPU_CLOCK_FREQ 50000000 /* 50MHz */ - -/* 1MB of onboard SRAM. Note that the monitor ROM uses parts of this - for its own purposes, so care must be taken. Some address lines are - not decoded, so the SRAM area is mirrored every 1MB from 0x400000 to - 0x800000 (exclusive). */ -#define SRAM_ADDR 0x00400000 -#define SRAM_SIZE 0x00100000 /* 1MB */ - -/* 32MB of onbard SDRAM. */ -#define SDRAM_ADDR 0x00800000 -#define SDRAM_SIZE 0x02000000 /* 32MB */ - - -/* CPU addresses of GBUS memory spaces. */ -#define GCS0_ADDR 0x05000000 /* GCS0 - Common SRAM (2MB) */ -#define GCS0_SIZE 0x00200000 /* 2MB */ -#define GCS1_ADDR 0x06000000 /* GCS1 - Flash ROM (8MB) */ -#define GCS1_SIZE 0x00800000 /* 8MB */ -#define GCS2_ADDR 0x07900000 /* GCS2 - I/O registers */ -#define GCS2_SIZE 0x00400000 /* 4MB */ -#define GCS5_ADDR 0x04000000 /* GCS5 - PCI bus space */ -#define GCS5_SIZE 0x01000000 /* 16MB */ -#define GCS6_ADDR 0x07980000 /* GCS6 - PCI control registers */ -#define GCS6_SIZE 0x00000200 /* 512B */ - - -/* For */ -#define PAGE_OFFSET SRAM_ADDR - - -/* The GBUS GINT0 - GINT3 interrupts are connected to the INTP000 - INTP011 - pins on the CPU. These are shared among the GBUS interrupts. */ -#define IRQ_GINT(n) IRQ_INTP(n) -#define IRQ_GINT_NUM 4 - -/* Used by to derive NUM_MACH_IRQS. */ -#define NUM_RTE_CB_IRQS NUM_CPU_IRQS - - -#ifdef CONFIG_ROM_KERNEL -/* Kernel is in ROM, starting at address 0. */ - -#define INTV_BASE 0 - -#else /* !CONFIG_ROM_KERNEL */ - -#ifdef CONFIG_RTE_CB_MULTI -/* Using RAM kernel with ROM monitor for Multi debugger. */ - -/* The chip's real interrupt vectors are in ROM, but they jump to a - secondary interrupt vector table in RAM. */ -#define INTV_BASE 0x004F8000 - -/* Scratch memory used by the ROM monitor, which shouldn't be used by - linux (except for the alternate interrupt vector area, defined - above). */ -#define MON_SCRATCH_ADDR 0x004F8000 -#define MON_SCRATCH_SIZE 0x00008000 /* 32KB */ - -#else /* !CONFIG_RTE_CB_MULTI */ -/* Using RAM-kernel. Assume some sort of boot-loader got us loaded at - address 0. */ - -#define INTV_BASE 0 - -#endif /* CONFIG_RTE_CB_MULTI */ - -#endif /* CONFIG_ROM_KERNEL */ - - -/* Some misc. on-board devices. */ - -/* Seven-segment LED display (two digits). Write-only. */ -#define LED_ADDR(n) (0x07802000 + (n)) -#define LED(n) (*(volatile unsigned char *)LED_ADDR(n)) -#define LED_NUM_DIGITS 2 - - -/* Override the basic MA uart pre-initialization so that we can - initialize extra stuff. */ -#undef V850E_UART_PRE_CONFIGURE /* should be defined by */ -#define V850E_UART_PRE_CONFIGURE rte_ma1_cb_uart_pre_configure -#ifndef __ASSEMBLY__ -extern void rte_ma1_cb_uart_pre_configure (unsigned chan, - unsigned cflags, unsigned baud); -#endif - -/* This board supports RTS/CTS for the on-chip UART, but only for channel 0. */ - -/* CTS for UART channel 0 is pin P43 (bit 3 of port 4). */ -#define V850E_UART_CTS(chan) ((chan) == 0 ? !(MA_PORT4_IO & 0x8) : 1) -/* RTS for UART channel 0 is pin P42 (bit 2 of port 4). */ -#define V850E_UART_SET_RTS(chan, val) \ - do { \ - if (chan == 0) { \ - unsigned old = MA_PORT4_IO; \ - if (val) \ - MA_PORT4_IO = old & ~0x4; \ - else \ - MA_PORT4_IO = old | 0x4; \ - } \ - } while (0) - - -#endif /* __V850_RTE_MA1_CB_H__ */ diff --git a/include/asm-v850/rte_mb_a_pci.h b/include/asm-v850/rte_mb_a_pci.h deleted file mode 100644 index 41ac185ca9cd..000000000000 --- a/include/asm-v850/rte_mb_a_pci.h +++ /dev/null @@ -1,56 +0,0 @@ -/* - * include/asm-v850/mb_a_pci.h -- PCI support for Midas lab RTE-MOTHER-A board - * - * Copyright (C) 2001 NEC Corporation - * Copyright (C) 2001 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_MB_A_PCI_H__ -#define __V850_MB_A_PCI_H__ - - -#define MB_A_PCI_MEM_ADDR GCS5_ADDR -#define MB_A_PCI_MEM_SIZE (GCS5_SIZE / 2) -#define MB_A_PCI_IO_ADDR (GCS5_ADDR + MB_A_PCI_MEM_SIZE) -#define MB_A_PCI_IO_SIZE (GCS5_SIZE / 2) -#define MB_A_PCI_REG_BASE_ADDR GCS6_ADDR - -#define MB_A_PCI_PCICR_ADDR (MB_A_PCI_REG_BASE_ADDR + 0x4) -#define MB_A_PCI_PCICR (*(volatile u16 *)MB_A_PCI_PCICR_ADDR) -#define MB_A_PCI_PCISR_ADDR (MB_A_PCI_REG_BASE_ADDR + 0x6) -#define MB_A_PCI_PCISR (*(volatile u16 *)MB_A_PCI_PCISR_ADDR) -#define MB_A_PCI_PCILTR_ADDR (MB_A_PCI_REG_BASE_ADDR + 0xD) -#define MB_A_PCI_PCILTR (*(volatile u8 *)MB_A_PCI_PCILTR_ADDR) -#define MB_A_PCI_PCIBAR0_ADDR (MB_A_PCI_REG_BASE_ADDR + 0x10) -#define MB_A_PCI_PCIBAR0 (*(volatile u32 *)MB_A_PCI_PCIBAR0_ADDR) -#define MB_A_PCI_PCIBAR1_ADDR (MB_A_PCI_REG_BASE_ADDR + 0x14) -#define MB_A_PCI_PCIBAR1 (*(volatile u32 *)MB_A_PCI_PCIBAR1_ADDR) -#define MB_A_PCI_PCIBAR2_ADDR (MB_A_PCI_REG_BASE_ADDR + 0x18) -#define MB_A_PCI_PCIBAR2 (*(volatile u32 *)MB_A_PCI_PCIBAR2_ADDR) -#define MB_A_PCI_VENDOR_ID_ADDR (MB_A_PCI_REG_BASE_ADDR + 0x2C) -#define MB_A_PCI_VENDOR_ID (*(volatile u16 *)MB_A_PCI_VENDOR_ID_ADDR) -#define MB_A_PCI_DEVICE_ID_ADDR (MB_A_PCI_REG_BASE_ADDR + 0x2E) -#define MB_A_PCI_DEVICE_ID (*(volatile u16 *)MB_A_PCI_DEVICE_ID_ADDR) -#define MB_A_PCI_DMRR_ADDR (MB_A_PCI_REG_BASE_ADDR + 0x9C) -#define MB_A_PCI_DMRR (*(volatile u32 *)MB_A_PCI_DMRR_ADDR) -#define MB_A_PCI_DMLBAM_ADDR (MB_A_PCI_REG_BASE_ADDR + 0xA0) -#define MB_A_PCI_DMLBAM (*(volatile u32 *)MB_A_PCI_DMLBAM_ADDR) -#define MB_A_PCI_DMLBAI_ADDR (MB_A_PCI_REG_BASE_ADDR + 0xA4) -#define MB_A_PCI_DMLBAI (*(volatile u32 *)MB_A_PCI_DMLBAI_ADDR) -#define MB_A_PCI_PCIPBAM_ADDR (MB_A_PCI_REG_BASE_ADDR + 0xA8) -#define MB_A_PCI_PCIPBAM (*(volatile u32 *)MB_A_PCI_PCIPBAM_ADDR) -/* `PCI Configuration Address Register for Direct Master to PCI IO/CFG' */ -#define MB_A_PCI_DMCFGA_ADDR (MB_A_PCI_REG_BASE_ADDR + 0xAC) -#define MB_A_PCI_DMCFGA (*(volatile u32 *)MB_A_PCI_DMCFGA_ADDR) -/* `PCI Permanent Configuration ID Register' */ -#define MB_A_PCI_PCIHIDR_ADDR (MB_A_PCI_REG_BASE_ADDR + 0xF0) -#define MB_A_PCI_PCIHIDR (*(volatile u32 *)MB_A_PCI_PCIHIDR_ADDR) - - -#endif /* __V850_MB_A_PCI_H__ */ diff --git a/include/asm-v850/rte_me2_cb.h b/include/asm-v850/rte_me2_cb.h deleted file mode 100644 index 9922c85c85a8..000000000000 --- a/include/asm-v850/rte_me2_cb.h +++ /dev/null @@ -1,202 +0,0 @@ -/* - * include/asm-v850/rte_me2_cb.h -- Midas labs RTE-V850E/ME2-CB board - * - * Copyright (C) 2001,02,03 NEC Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_RTE_ME2_CB_H__ -#define __V850_RTE_ME2_CB_H__ - -#include /* Common defs for Midas RTE-CB boards. */ - - -#define PLATFORM "rte-v850e/me2-cb" -#define PLATFORM_LONG "Midas lab RTE-V850E/ME2-CB" - -#define CPU_CLOCK_FREQ 150000000 /* 150MHz */ -#define FIXED_BOGOMIPS 50 - -/* 32MB of onbard SDRAM. */ -#define SDRAM_ADDR 0x00800000 -#define SDRAM_SIZE 0x02000000 /* 32MB */ - - -/* CPU addresses of GBUS memory spaces. */ -#define GCS0_ADDR 0x04000000 /* GCS0 - Common SRAM (2MB) */ -#define GCS0_SIZE 0x00800000 /* 8MB */ -#define GCS1_ADDR 0x04800000 /* GCS1 - Flash ROM (8MB) */ -#define GCS1_SIZE 0x00800000 /* 8MB */ -#define GCS2_ADDR 0x07000000 /* GCS2 - I/O registers */ -#define GCS2_SIZE 0x00800000 /* 8MB */ -#define GCS5_ADDR 0x08000000 /* GCS5 - PCI bus space */ -#define GCS5_SIZE 0x02000000 /* 32MB */ -#define GCS6_ADDR 0x07800000 /* GCS6 - PCI control registers */ -#define GCS6_SIZE 0x00800000 /* 8MB */ - - -/* For */ -#define PAGE_OFFSET SDRAM_ADDR - - -#ifdef CONFIG_ROM_KERNEL -/* Kernel is in ROM, starting at address 0. */ - -#define INTV_BASE 0 -#define ROOT_FS_IMAGE_RW 0 - -#else /* !CONFIG_ROM_KERNEL */ -/* Using RAM-kernel. Assume some sort of boot-loader got us loaded at - address 0. */ - -#define INTV_BASE 0 -#define ROOT_FS_IMAGE_RW 1 - -#endif /* CONFIG_ROM_KERNEL */ - - -/* Some misc. on-board devices. */ - -/* Seven-segment LED display (four digits). */ -#define LED_ADDR(n) (0x0FE02000 + (n)) -#define LED(n) (*(volatile unsigned char *)LED_ADDR(n)) -#define LED_NUM_DIGITS 4 - - -/* On-board PIC. */ - -#define CB_PIC_BASE_ADDR 0x0FE04000 - -#define CB_PIC_INT0M_ADDR (CB_PIC_BASE_ADDR + 0x00) -#define CB_PIC_INT0M (*(volatile u16 *)CB_PIC_INT0M_ADDR) -#define CB_PIC_INT1M_ADDR (CB_PIC_BASE_ADDR + 0x10) -#define CB_PIC_INT1M (*(volatile u16 *)CB_PIC_INT1M_ADDR) -#define CB_PIC_INTR_ADDR (CB_PIC_BASE_ADDR + 0x20) -#define CB_PIC_INTR (*(volatile u16 *)CB_PIC_INTR_ADDR) -#define CB_PIC_INTEN_ADDR (CB_PIC_BASE_ADDR + 0x30) -#define CB_PIC_INTEN (*(volatile u16 *)CB_PIC_INTEN_ADDR) - -#define CB_PIC_INT0EN 0x0001 -#define CB_PIC_INT1EN 0x0002 -#define CB_PIC_INT0SEL 0x0080 - -/* The PIC interrupts themselves. */ -#define CB_PIC_BASE_IRQ NUM_CPU_IRQS -#define IRQ_CB_PIC_NUM 10 - -/* Some specific CB_PIC interrupts. */ -#define IRQ_CB_EXTTM0 (CB_PIC_BASE_IRQ + 0) -#define IRQ_CB_EXTSIO (CB_PIC_BASE_IRQ + 1) -#define IRQ_CB_TOVER (CB_PIC_BASE_IRQ + 2) -#define IRQ_CB_GINT0 (CB_PIC_BASE_IRQ + 3) -#define IRQ_CB_USB (CB_PIC_BASE_IRQ + 4) -#define IRQ_CB_LANC (CB_PIC_BASE_IRQ + 5) -#define IRQ_CB_USB_VBUS_ON (CB_PIC_BASE_IRQ + 6) -#define IRQ_CB_USB_VBUS_OFF (CB_PIC_BASE_IRQ + 7) -#define IRQ_CB_EXTTM1 (CB_PIC_BASE_IRQ + 8) -#define IRQ_CB_EXTTM2 (CB_PIC_BASE_IRQ + 9) - -/* The GBUS GINT1 - GINT3 (note, not GINT0!) interrupts are connected to - the INTP65 - INTP67 pins on the CPU. These are shared among the GBUS - interrupts. */ -#define IRQ_GINT(n) IRQ_INTP((n) + 9) /* 0 is unused! */ -#define IRQ_GINT_NUM 4 /* 0 is unused! */ - -/* The shared interrupt line from the PIC is connected to CPU pin INTP23. */ -#define IRQ_CB_PIC IRQ_INTP(4) /* P23 */ - -/* Used by to derive NUM_MACH_IRQS. */ -#define NUM_RTE_CB_IRQS (NUM_CPU_IRQS + IRQ_CB_PIC_NUM) - - -#ifndef __ASSEMBLY__ -struct cb_pic_irq_init { - const char *name; /* name of interrupt type */ - - /* Range of kernel irq numbers for this type: - BASE, BASE+INTERVAL, ..., BASE+INTERVAL*NUM */ - unsigned base, num, interval; - - unsigned priority; /* interrupt priority to assign */ -}; -struct hw_interrupt_type; /* fwd decl */ - -/* Enable interrupt handling for interrupt IRQ. */ -extern void cb_pic_enable_irq (unsigned irq); -/* Disable interrupt handling for interrupt IRQ. Note that any interrupts - received while disabled will be delivered once the interrupt is enabled - again, unless they are explicitly cleared using `cb_pic_clear_pending_irq'. */ -extern void cb_pic_disable_irq (unsigned irq); -/* Initialize HW_IRQ_TYPES for PIC irqs described in array INITS (which is - terminated by an entry with the name field == 0). */ -extern void cb_pic_init_irq_types (struct cb_pic_irq_init *inits, - struct hw_interrupt_type *hw_irq_types); -/* Initialize PIC interrupts. */ -extern void cb_pic_init_irqs (void); -#endif /* __ASSEMBLY__ */ - - -/* TL16C550C on board UART see also asm/serial.h */ -#define CB_UART_BASE 0x0FE08000 -#define CB_UART_REG_GAP 0x10 -#define CB_UART_CLOCK 0x16000000 - -/* CompactFlash setting */ -#define CB_CF_BASE 0x0FE0C000 -#define CB_CF_CCR_ADDR (CB_CF_BASE+0x200) -#define CB_CF_CCR (*(volatile u8 *)CB_CF_CCR_ADDR) -#define CB_CF_REG0_ADDR (CB_CF_BASE+0x1000) -#define CB_CF_REG0 (*(volatile u16 *)CB_CF_REG0_ADDR) -#define CB_CF_STS0_ADDR (CB_CF_BASE+0x1004) -#define CB_CF_STS0 (*(volatile u16 *)CB_CF_STS0_ADDR) -#define CB_PCATA_BASE (CB_CF_BASE+0x800) -#define CB_IDE_BASE (CB_CF_BASE+0x9F0) -#define CB_IDE_CTRL (CB_CF_BASE+0xBF6) -#define CB_IDE_REG_OFFS 0x1 - - -/* SMSC LAN91C111 setting */ -#if defined(CONFIG_SMC91111) -#define CB_LANC_BASE 0x0FE10300 -#define CONFIG_SMC16BITONLY -#define ETH0_ADDR CB_LANC_BASE -#define ETH0_IRQ IRQ_CB_LANC -#endif /* CONFIG_SMC16BITONLY */ - - -#undef V850E_UART_PRE_CONFIGURE -#define V850E_UART_PRE_CONFIGURE rte_me2_cb_uart_pre_configure -#ifndef __ASSEMBLY__ -extern void rte_me2_cb_uart_pre_configure (unsigned chan, - unsigned cflags, unsigned baud); -#endif /* __ASSEMBLY__ */ - -/* This board supports RTS/CTS for the on-chip UART, but only for channel 0. */ - -/* CTS for UART channel 0 is pin P22 (bit 2 of port 2). */ -#define V850E_UART_CTS(chan) ((chan) == 0 ? !(ME2_PORT2_IO & 0x4) : 1) -/* RTS for UART channel 0 is pin P21 (bit 1 of port 2). */ -#define V850E_UART_SET_RTS(chan, val) \ - do { \ - if (chan == 0) { \ - unsigned old = ME2_PORT2_IO; \ - if (val) \ - ME2_PORT2_IO = old & ~0x2; \ - else \ - ME2_PORT2_IO = old | 0x2; \ - } \ - } while (0) - - -#ifndef __ASSEMBLY__ -extern void rte_me2_cb_init_irqs (void); -#endif /* !__ASSEMBLY__ */ - - -#endif /* __V850_RTE_ME2_CB_H__ */ diff --git a/include/asm-v850/rte_nb85e_cb.h b/include/asm-v850/rte_nb85e_cb.h deleted file mode 100644 index f56591cad90a..000000000000 --- a/include/asm-v850/rte_nb85e_cb.h +++ /dev/null @@ -1,111 +0,0 @@ -/* - * include/asm-v850/rte_nb85e_cb.h -- Midas labs RTE-V850/NB85E-CB board - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_RTE_NB85E_CB_H__ -#define __V850_RTE_NB85E_CB_H__ - -#include /* Common defs for Midas RTE-CB boards. */ - - -#define PLATFORM "rte-v850e/nb85e-cb" -#define PLATFORM_LONG "Midas lab RTE-V850E/NB85E-CB" - -#define CPU_CLOCK_FREQ 50000000 /* 50MHz */ - -/* 1MB of onboard SRAM. Note that the monitor ROM uses parts of this - for its own purposes, so care must be taken. */ -#define SRAM_ADDR 0x03C00000 -#define SRAM_SIZE 0x00100000 /* 1MB */ - -/* 16MB of onbard SDRAM. */ -#define SDRAM_ADDR 0x01000000 -#define SDRAM_SIZE 0x01000000 /* 16MB */ - - -/* CPU addresses of GBUS memory spaces. */ -#define GCS0_ADDR 0x00400000 /* GCS0 - Common SRAM (2MB) */ -#define GCS0_SIZE 0x00400000 /* 4MB */ -#define GCS1_ADDR 0x02000000 /* GCS1 - Flash ROM (8MB) */ -#define GCS1_SIZE 0x00800000 /* 8MB */ -#define GCS2_ADDR 0x03900000 /* GCS2 - I/O registers */ -#define GCS2_SIZE 0x00080000 /* 512KB */ -#define GCS3_ADDR 0x02800000 /* GCS3 - EXT-bus: memory space */ -#define GCS3_SIZE 0x00800000 /* 8MB */ -#define GCS4_ADDR 0x03A00000 /* GCS4 - EXT-bus: I/O space */ -#define GCS4_SIZE 0x00200000 /* 2MB */ -#define GCS5_ADDR 0x00800000 /* GCS5 - PCI bus space */ -#define GCS5_SIZE 0x00800000 /* 8MB */ -#define GCS6_ADDR 0x03980000 /* GCS6 - PCI control registers */ -#define GCS6_SIZE 0x00010000 /* 64KB */ - - -/* The GBUS GINT0 - GINT3 interrupts are connected to CPU interrupts 10-12. - These are shared among the GBUS interrupts. */ -#define IRQ_GINT(n) (10 + (n)) -#define IRQ_GINT_NUM 3 - -/* Used by to derive NUM_MACH_IRQS. */ -#define NUM_RTE_CB_IRQS NUM_CPU_IRQS - - -#ifdef CONFIG_ROM_KERNEL -/* Kernel is in ROM, starting at address 0. */ - -#define INTV_BASE 0 - -#else /* !CONFIG_ROM_KERNEL */ -/* We're using the ROM monitor. */ - -/* The chip's real interrupt vectors are in ROM, but they jump to a - secondary interrupt vector table in RAM. */ -#define INTV_BASE 0x03CF8000 - -/* Scratch memory used by the ROM monitor, which shouldn't be used by - linux (except for the alternate interrupt vector area, defined - above). */ -#define MON_SCRATCH_ADDR 0x03CE8000 -#define MON_SCRATCH_SIZE 0x00018000 /* 96KB */ - -#endif /* CONFIG_ROM_KERNEL */ - - -/* Some misc. on-board devices. */ - -/* Seven-segment LED display (two digits). Write-only. */ -#define LED_ADDR(n) (0x03802000 + (n)) -#define LED(n) (*(volatile unsigned char *)LED_ADDR(n)) -#define LED_NUM_DIGITS 4 - - -/* Override the basic TEG UART pre-initialization so that we can - initialize extra stuff. */ -#undef V850E_UART_PRE_CONFIGURE /* should be defined by */ -#define V850E_UART_PRE_CONFIGURE rte_nb85e_cb_uart_pre_configure -#ifndef __ASSEMBLY__ -extern void rte_nb85e_cb_uart_pre_configure (unsigned chan, - unsigned cflags, unsigned baud); -#endif - -/* This board supports RTS/CTS for the on-chip UART. */ - -/* CTS is pin P00. */ -#define V850E_UART_CTS(chan) (! (TEG_PORT0_IO & 0x1)) -/* RTS is pin P02. */ -#define V850E_UART_SET_RTS(chan, val) \ - do { \ - unsigned old = TEG_PORT0_IO; \ - TEG_PORT0_IO = val ? (old & ~0x4) : (old | 0x4); \ - } while (0) - - -#endif /* __V850_RTE_NB85E_CB_H__ */ diff --git a/include/asm-v850/scatterlist.h b/include/asm-v850/scatterlist.h deleted file mode 100644 index 02d27b3fb061..000000000000 --- a/include/asm-v850/scatterlist.h +++ /dev/null @@ -1,31 +0,0 @@ -/* - * include/asm-v850/scatterlist.h - * - * Copyright (C) 2001,02 NEC Corporation - * Copyright (C) 2001,02 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_SCATTERLIST_H__ -#define __V850_SCATTERLIST_H__ - -#include - -struct scatterlist { -#ifdef CONFIG_DEBUG_SG - unsigned long sg_magic; -#endif - unsigned long page_link; - unsigned offset; - dma_addr_t dma_address; - unsigned length; -}; - -#define ISA_DMA_THRESHOLD (~0UL) - -#endif /* __V850_SCATTERLIST_H__ */ diff --git a/include/asm-v850/sections.h b/include/asm-v850/sections.h deleted file mode 100644 index e0238253a0d0..000000000000 --- a/include/asm-v850/sections.h +++ /dev/null @@ -1,6 +0,0 @@ -#ifndef __V850_SECTIONS_H__ -#define __V850_SECTIONS_H__ - -#include - -#endif /* __V850_SECTIONS_H__ */ diff --git a/include/asm-v850/segment.h b/include/asm-v850/segment.h deleted file mode 100644 index 5e2b15dcf3d9..000000000000 --- a/include/asm-v850/segment.h +++ /dev/null @@ -1,36 +0,0 @@ -#ifndef __V850_SEGMENT_H__ -#define __V850_SEGMENT_H__ - - -#ifndef __ASSEMBLY__ - -typedef unsigned long mm_segment_t; /* domain register */ - -#endif /* !__ASSEMBLY__ */ - - -#define __KERNEL_CS 0x0 -#define __KERNEL_DS 0x0 - -#define __USER_CS 0x1 -#define __USER_DS 0x1 - -#define KERNEL_DS __KERNEL_DS -#define KERNEL_CS __KERNEL_CS -#define USER_DS __USER_DS -#define USER_CS __USER_CS - -#define segment_eq(a,b) ((a) == (b)) - -#define get_ds() (KERNEL_DS) -#define get_fs() (USER_DS) - -#define set_fs(seg) ((void)(seg)) - - -#define copy_segments(task, mm) ((void)((void)(task), (mm))) -#define release_segments(mm) ((void)(mm)) -#define forget_segments() ((void)0) - - -#endif /* __V850_SEGMENT_H__ */ diff --git a/include/asm-v850/semaphore.h b/include/asm-v850/semaphore.h deleted file mode 100644 index d9b2034ed1d2..000000000000 --- a/include/asm-v850/semaphore.h +++ /dev/null @@ -1 +0,0 @@ -#include diff --git a/include/asm-v850/sembuf.h b/include/asm-v850/sembuf.h deleted file mode 100644 index 1622231a8b85..000000000000 --- a/include/asm-v850/sembuf.h +++ /dev/null @@ -1,25 +0,0 @@ -#ifndef __V850_SEMBUF_H__ -#define __V850_SEMBUF_H__ - -/* - * The semid64_ds structure for v850 architecture. - * Note extra padding because this structure is passed back and forth - * between kernel and user space. - * - * Pad space is left for: - * - 64-bit time_t to solve y2038 problem - * - 2 miscellaneous 32-bit values - */ - -struct semid64_ds { - struct ipc64_perm sem_perm; /* permissions .. see ipc.h */ - __kernel_time_t sem_otime; /* last semop time */ - unsigned long __unused1; - __kernel_time_t sem_ctime; /* last change time */ - unsigned long __unused2; - unsigned long sem_nsems; /* no. of semaphores in array */ - unsigned long __unused3; - unsigned long __unused4; -}; - -#endif /* __V850_SEMBUF_H__ */ diff --git a/include/asm-v850/serial.h b/include/asm-v850/serial.h deleted file mode 100644 index 36d8f4cbbf39..000000000000 --- a/include/asm-v850/serial.h +++ /dev/null @@ -1,56 +0,0 @@ -/* - * This file is subject to the terms and conditions of the GNU General Public - * License. See the file "COPYING" in the main directory of this archive - * for more details. - * - * Copyright (C) 1999 by Ralf Baechle - * Copyright (C) 1999, 2000 Silicon Graphics, Inc. - */ - -#ifdef CONFIG_RTE_CB_ME2 - -#include - -#define STD_COM_FLAGS (ASYNC_BOOT_AUTOCONF | ASYNC_SKIP_TEST) - -#define irq_cannonicalize(x) (x) -#define BASE_BAUD 250000 /* (16MHz / (16 * 38400)) * 9600 */ -#define SERIAL_PORT_DFNS \ - { 0, BASE_BAUD, CB_UART_BASE, IRQ_CB_EXTSIO, STD_COM_FLAGS }, - -/* Redefine UART register offsets. */ -#undef UART_RX -#undef UART_TX -#undef UART_DLL -#undef UART_TRG -#undef UART_DLM -#undef UART_IER -#undef UART_FCTR -#undef UART_IIR -#undef UART_FCR -#undef UART_EFR -#undef UART_LCR -#undef UART_MCR -#undef UART_LSR -#undef UART_MSR -#undef UART_SCR -#undef UART_EMSR - -#define UART_RX (0 * CB_UART_REG_GAP) -#define UART_TX (0 * CB_UART_REG_GAP) -#define UART_DLL (0 * CB_UART_REG_GAP) -#define UART_TRG (0 * CB_UART_REG_GAP) -#define UART_DLM (1 * CB_UART_REG_GAP) -#define UART_IER (1 * CB_UART_REG_GAP) -#define UART_FCTR (1 * CB_UART_REG_GAP) -#define UART_IIR (2 * CB_UART_REG_GAP) -#define UART_FCR (2 * CB_UART_REG_GAP) -#define UART_EFR (2 * CB_UART_REG_GAP) -#define UART_LCR (3 * CB_UART_REG_GAP) -#define UART_MCR (4 * CB_UART_REG_GAP) -#define UART_LSR (5 * CB_UART_REG_GAP) -#define UART_MSR (6 * CB_UART_REG_GAP) -#define UART_SCR (7 * CB_UART_REG_GAP) -#define UART_EMSR (7 * CB_UART_REG_GAP) - -#endif /* CONFIG_RTE_CB_ME2 */ diff --git a/include/asm-v850/setup.h b/include/asm-v850/setup.h deleted file mode 100644 index c48a9b97d05b..000000000000 --- a/include/asm-v850/setup.h +++ /dev/null @@ -1,6 +0,0 @@ -#ifndef _V850_SETUP_H -#define _V850_SETUP_H - -#define COMMAND_LINE_SIZE 512 - -#endif /* __SETUP_H */ diff --git a/include/asm-v850/shmbuf.h b/include/asm-v850/shmbuf.h deleted file mode 100644 index 3d085c9c418e..000000000000 --- a/include/asm-v850/shmbuf.h +++ /dev/null @@ -1,42 +0,0 @@ -#ifndef __V850_SHMBUF_H__ -#define __V850_SHMBUF_H__ - -/* - * The shmid64_ds structure for v850 architecture. - * Note extra padding because this structure is passed back and forth - * between kernel and user space. - * - * Pad space is left for: - * - 64-bit time_t to solve y2038 problem - * - 2 miscellaneous 32-bit values - */ - -struct shmid64_ds { - struct ipc64_perm shm_perm; /* operation perms */ - size_t shm_segsz; /* size of segment (bytes) */ - __kernel_time_t shm_atime; /* last attach time */ - unsigned long __unused1; - __kernel_time_t shm_dtime; /* last detach time */ - unsigned long __unused2; - __kernel_time_t shm_ctime; /* last change time */ - unsigned long __unused3; - __kernel_pid_t shm_cpid; /* pid of creator */ - __kernel_pid_t shm_lpid; /* pid of last operator */ - unsigned long shm_nattch; /* no. of current attaches */ - unsigned long __unused4; - unsigned long __unused5; -}; - -struct shminfo64 { - unsigned long shmmax; - unsigned long shmmin; - unsigned long shmmni; - unsigned long shmseg; - unsigned long shmall; - unsigned long __unused1; - unsigned long __unused2; - unsigned long __unused3; - unsigned long __unused4; -}; - -#endif /* __V850_SHMBUF_H__ */ diff --git a/include/asm-v850/shmparam.h b/include/asm-v850/shmparam.h deleted file mode 100644 index 7dcb6739073e..000000000000 --- a/include/asm-v850/shmparam.h +++ /dev/null @@ -1,6 +0,0 @@ -#ifndef __V850_SHMPARAM_H__ -#define __V850_SHMPARAM_H__ - -#define SHMLBA PAGE_SIZE /* attach addr a multiple of this */ - -#endif /* __V850_SHMPARAM_H__ */ diff --git a/include/asm-v850/sigcontext.h b/include/asm-v850/sigcontext.h deleted file mode 100644 index e0890f6f4bc9..000000000000 --- a/include/asm-v850/sigcontext.h +++ /dev/null @@ -1,25 +0,0 @@ -/* - * include/asm-v850/sigcontext.h -- Signal contexts - * - * Copyright (C) 2001 NEC Corporation - * Copyright (C) 2001 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_SIGCONTEXT_H__ -#define __V850_SIGCONTEXT_H__ - -#include - -struct sigcontext -{ - struct pt_regs regs; - unsigned long oldmask; -}; - -#endif /* __V850_SIGCONTEXT_H__ */ diff --git a/include/asm-v850/siginfo.h b/include/asm-v850/siginfo.h deleted file mode 100644 index 7eb94703dce0..000000000000 --- a/include/asm-v850/siginfo.h +++ /dev/null @@ -1,6 +0,0 @@ -#ifndef __V850_SIGINFO_H__ -#define __V850_SIGINFO_H__ - -#include - -#endif /* __V850_SIGINFO_H__ */ diff --git a/include/asm-v850/signal.h b/include/asm-v850/signal.h deleted file mode 100644 index a38df0834bbf..000000000000 --- a/include/asm-v850/signal.h +++ /dev/null @@ -1,168 +0,0 @@ -#ifndef __V850_SIGNAL_H__ -#define __V850_SIGNAL_H__ - -#include - -/* Avoid too many header ordering problems. */ -struct siginfo; - - -#ifdef __KERNEL__ - -/* Most things should be clean enough to redefine this at will, if care - is taken to make libc match. */ -#define _NSIG 64 -#define _NSIG_BPW 32 -#define _NSIG_WORDS (_NSIG / _NSIG_BPW) - -typedef unsigned long old_sigset_t; /* at least 32 bits */ - -typedef struct { - unsigned long sig[_NSIG_WORDS]; -} sigset_t; - -#else /* !__KERNEL__ */ - -/* Here we must cater to libcs that poke about in kernel headers. */ - -#define NSIG 32 -typedef unsigned long sigset_t; - -#endif /* __KERNEL__ */ - - -#define SIGHUP 1 -#define SIGINT 2 -#define SIGQUIT 3 -#define SIGILL 4 -#define SIGTRAP 5 -#define SIGABRT 6 -#define SIGIOT 6 -#define SIGBUS 7 -#define SIGFPE 8 -#define SIGKILL 9 -#define SIGUSR1 10 -#define SIGSEGV 11 -#define SIGUSR2 12 -#define SIGPIPE 13 -#define SIGALRM 14 -#define SIGTERM 15 -#define SIGSTKFLT 16 -#define SIGCHLD 17 -#define SIGCONT 18 -#define SIGSTOP 19 -#define SIGTSTP 20 -#define SIGTTIN 21 -#define SIGTTOU 22 -#define SIGURG 23 -#define SIGXCPU 24 -#define SIGXFSZ 25 -#define SIGVTALRM 26 -#define SIGPROF 27 -#define SIGWINCH 28 -#define SIGIO 29 -#define SIGPOLL SIGIO -/* -#define SIGLOST 29 -*/ -#define SIGPWR 30 -#define SIGSYS 31 -#define SIGUNUSED 31 - -/* These should not be considered constants from userland. */ -#define SIGRTMIN 32 -#define SIGRTMAX _NSIG - -/* - * SA_FLAGS values: - * - * SA_ONSTACK indicates that a registered stack_t will be used. - * SA_RESTART flag to get restarting signals (which were the default long ago) - * SA_NOCLDSTOP flag to turn off SIGCHLD when children stop. - * SA_RESETHAND clears the handler when the signal is delivered. - * SA_NOCLDWAIT flag on SIGCHLD to inhibit zombies. - * SA_NODEFER prevents the current signal from being masked in the handler. - * - * SA_ONESHOT and SA_NOMASK are the historical Linux names for the Single - * Unix names RESETHAND and NODEFER respectively. - */ -#define SA_NOCLDSTOP 0x00000001 -#define SA_NOCLDWAIT 0x00000002 -#define SA_SIGINFO 0x00000004 -#define SA_ONSTACK 0x08000000 -#define SA_RESTART 0x10000000 -#define SA_NODEFER 0x40000000 -#define SA_RESETHAND 0x80000000 - -#define SA_NOMASK SA_NODEFER -#define SA_ONESHOT SA_RESETHAND - -#define SA_RESTORER 0x04000000 - -/* - * sigaltstack controls - */ -#define SS_ONSTACK 1 -#define SS_DISABLE 2 - -#define MINSIGSTKSZ 2048 -#define SIGSTKSZ 8192 - -#include - -#ifdef __KERNEL__ - -struct old_sigaction { - __sighandler_t sa_handler; - old_sigset_t sa_mask; - unsigned long sa_flags; - void (*sa_restorer)(void); -}; - -struct sigaction { - __sighandler_t sa_handler; - unsigned long sa_flags; - void (*sa_restorer)(void); - sigset_t sa_mask; /* mask last for extensibility */ -}; - -struct k_sigaction { - struct sigaction sa; -}; - -#else /* !__KERNEL__ */ - -/* Here we must cater to libcs that poke about in kernel headers. */ - -struct sigaction { - union { - __sighandler_t _sa_handler; - void (*_sa_sigaction)(int, struct siginfo *, void *); - } _u; - sigset_t sa_mask; - unsigned long sa_flags; - void (*sa_restorer)(void); -}; - -#define sa_handler _u._sa_handler -#define sa_sigaction _u._sa_sigaction - -#endif /* __KERNEL__ */ - - -typedef struct sigaltstack { - void *ss_sp; - int ss_flags; - size_t ss_size; -} stack_t; - -#ifdef __KERNEL__ - -#include -#undef __HAVE_ARCH_SIG_BITOPS - -#define ptrace_signal_deliver(regs, cookie) do { } while (0) - -#endif /* __KERNEL__ */ - -#endif /* __V850_SIGNAL_H__ */ diff --git a/include/asm-v850/sim.h b/include/asm-v850/sim.h deleted file mode 100644 index 026932d476cd..000000000000 --- a/include/asm-v850/sim.h +++ /dev/null @@ -1,47 +0,0 @@ -/* - * include/asm-v850/sim.h -- Machine-dependent defs for GDB v850e simulator - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_SIM_H__ -#define __V850_SIM_H__ - - -#define CPU_ARCH "v850e" -#define CPU_MODEL "v850e" -#define CPU_MODEL_LONG "NEC V850E" -#define PLATFORM "gdb/v850e" -#define PLATFORM_LONG "GDB V850E simulator" - - -/* We use a weird value for RAM, not just 0, for testing purposes. - These must match the values used in the linker script. */ -#define RAM_ADDR 0x8F000000 -#define RAM_SIZE 0x03000000 - - -/* For */ -#define PAGE_OFFSET RAM_ADDR - - -/* For */ -/* `R0 RAM', used for a few miscellaneous variables that must be - accessible using a load instruction relative to R0. On real - processors, this usually is on-chip RAM, but here we just - choose an arbitrary address that meets the above constraint. */ -#define R0_RAM_ADDR 0xFFFFF000 - - -/* For */ -#define NUM_CPU_IRQS 6 - - -#endif /* __V850_SIM_H__ */ diff --git a/include/asm-v850/sim85e2.h b/include/asm-v850/sim85e2.h deleted file mode 100644 index 8b4d6974066c..000000000000 --- a/include/asm-v850/sim85e2.h +++ /dev/null @@ -1,69 +0,0 @@ -/* - * include/asm-v850/sim85e2.h -- Machine-dependent defs for - * V850E2 RTL simulator - * - * Copyright (C) 2002,03 NEC Electronics Corporation - * Copyright (C) 2002,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_SIM85E2_H__ -#define __V850_SIM85E2_H__ - - -#include /* Based on V850E2 core. */ - - -/* Various memory areas supported by the simulator. - These should match the corresponding definitions in the linker script. */ - -/* `instruction RAM'; instruction fetches are much faster from IRAM than - from DRAM. */ -#define IRAM_ADDR 0 -#define IRAM_SIZE 0x00100000 /* 1MB */ -/* `data RAM', below and contiguous with the I/O space. - Data fetches are much faster from DRAM than from IRAM. */ -#define DRAM_ADDR 0xfff00000 -#define DRAM_SIZE 0x000ff000 /* 1020KB */ -/* `external ram'. Unlike the above RAM areas, this memory is cached, - so both instruction and data fetches should be (mostly) fast -- - however, currently only write-through caching is supported, so writes - to ERAM will be slow. */ -#define ERAM_ADDR 0x00100000 -#define ERAM_SIZE 0x07f00000 /* 127MB (max) */ -/* Dynamic RAM; uses memory controller. */ -#define SDRAM_ADDR 0x10000000 -#define SDRAM_SIZE 0x01000000 /* 16MB */ - - -/* Simulator specific control registers. */ -/* NOTHAL controls whether the simulator will stop at a `halt' insn. */ -#define SIM85E2_NOTHAL_ADDR 0xffffff22 -#define SIM85E2_NOTHAL (*(volatile u8 *)SIM85E2_NOTHAL_ADDR) -/* The simulator will stop N cycles after N is written to SIMFIN. */ -#define SIM85E2_SIMFIN_ADDR 0xffffff24 -#define SIM85E2_SIMFIN (*(volatile u16 *)SIM85E2_SIMFIN_ADDR) - - -/* For */ -#define NUM_CPU_IRQS 64 - - -/* For */ -#define PAGE_OFFSET SDRAM_ADDR - - -/* For */ -/* `R0 RAM', used for a few miscellaneous variables that must be accessible - using a load instruction relative to R0. The sim85e2 simulator - actually puts 1020K of RAM from FFF00000 to FFFFF000, so we arbitarily - choose a small portion at the end of that. */ -#define R0_RAM_ADDR 0xFFFFE000 - - -#endif /* __V850_SIM85E2_H__ */ diff --git a/include/asm-v850/sim85e2c.h b/include/asm-v850/sim85e2c.h deleted file mode 100644 index eee543ff3af8..000000000000 --- a/include/asm-v850/sim85e2c.h +++ /dev/null @@ -1,26 +0,0 @@ -/* - * include/asm-v850/sim85e2c.h -- Machine-dependent defs for - * V850E2 RTL simulator - * - * Copyright (C) 2002 NEC Corporation - * Copyright (C) 2002 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_SIM85E2C_H__ -#define __V850_SIM85E2C_H__ - -/* Use generic sim85e2 settings, other than the various names. */ -#include - -#define CPU_MODEL "v850e2" -#define CPU_MODEL_LONG "NEC V850E2" -#define PLATFORM "sim85e2c" -#define PLATFORM_LONG "SIM85E2C V850E2 simulator" - -#endif /* __V850_SIM85E2C_H__ */ diff --git a/include/asm-v850/sim85e2s.h b/include/asm-v850/sim85e2s.h deleted file mode 100644 index ee066d5d3c51..000000000000 --- a/include/asm-v850/sim85e2s.h +++ /dev/null @@ -1,28 +0,0 @@ -/* - * include/asm-v850/sim85e2s.h -- Machine-dependent defs for - * V850E2 RTL simulator - * - * Copyright (C) 2003 NEC Electronics Corporation - * Copyright (C) 2003 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_SIM85E2S_H__ -#define __V850_SIM85E2S_H__ - -#include /* Use generic sim85e2 settings. */ -#if 0 -#include /* + cache */ -#endif - -#define CPU_MODEL "v850e2" -#define CPU_MODEL_LONG "NEC V850E2" -#define PLATFORM "sim85e2s" -#define PLATFORM_LONG "SIM85E2S V850E2 simulator" - -#endif /* __V850_SIM85E2S_H__ */ diff --git a/include/asm-v850/simsyscall.h b/include/asm-v850/simsyscall.h deleted file mode 100644 index 4a19d5ae9d17..000000000000 --- a/include/asm-v850/simsyscall.h +++ /dev/null @@ -1,99 +0,0 @@ -/* - * include/asm-v850/simsyscall.h -- `System calls' under the v850e emulator - * - * Copyright (C) 2001 NEC Corporation - * Copyright (C) 2001 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_SIMSYSCALL_H__ -#define __V850_SIMSYSCALL_H__ - -#define V850_SIM_SYS_exit(a...) V850_SIM_SYSCALL_1 (1 , ##a) -#define V850_SIM_SYS_fork(a...) V850_SIM_SYSCALL_0 (2 , ##a) -#define V850_SIM_SYS_read(a...) V850_SIM_SYSCALL_3 (3 , ##a) -#define V850_SIM_SYS_write(a...) V850_SIM_SYSCALL_3 (4 , ##a) -#define V850_SIM_SYS_open(a...) V850_SIM_SYSCALL_2 (5 , ##a) -#define V850_SIM_SYS_close(a...) V850_SIM_SYSCALL_1 (6 , ##a) -#define V850_SIM_SYS_wait4(a...) V850_SIM_SYSCALL_4 (7 , ##a) -/* #define V850_SIM_SYS_creat(a...) V850_SIM_SYSCALL_1 (8 , ##a) */ -/* #define V850_SIM_SYS_link(a...) V850_SIM_SYSCALL_1 (9 , ##a) */ -/* #define V850_SIM_SYS_unlink(a...) V850_SIM_SYSCALL_1 (10 , ##a) */ -#define V850_SIM_SYS_execv(a...) V850_SIM_SYSCALL_2 (11 , ##a) -/* #define V850_SIM_SYS_chdir(a...) V850_SIM_SYSCALL_1 (12 , ##a) */ -/* #define V850_SIM_SYS_mknod(a...) V850_SIM_SYSCALL_1 (14 , ##a) */ -#define V850_SIM_SYS_chmod(a...) V850_SIM_SYSCALL_2 (15 , ##a) -#define V850_SIM_SYS_chown(a...) V850_SIM_SYSCALL_2 (16 , ##a) -#define V850_SIM_SYS_lseek(a...) V850_SIM_SYSCALL_3 (19 , ##a) -/* #define V850_SIM_SYS_getpid(a...) V850_SIM_SYSCALL_1 (20 , ##a) */ -/* #define V850_SIM_SYS_isatty(a...) V850_SIM_SYSCALL_1 (21 , ##a) */ -/* #define V850_SIM_SYS_fstat(a...) V850_SIM_SYSCALL_1 (22 , ##a) */ -#define V850_SIM_SYS_time(a...) V850_SIM_SYSCALL_1 (23 , ##a) -#define V850_SIM_SYS_poll(a...) V850_SIM_SYSCALL_3 (24 , ##a) -#define V850_SIM_SYS_stat(a...) V850_SIM_SYSCALL_2 (38 , ##a) -#define V850_SIM_SYS_pipe(a...) V850_SIM_SYSCALL_1 (42 , ##a) -#define V850_SIM_SYS_times(a...) V850_SIM_SYSCALL_1 (43 , ##a) -#define V850_SIM_SYS_execve(a...) V850_SIM_SYSCALL_3 (59 , ##a) -#define V850_SIM_SYS_gettimeofday(a...) V850_SIM_SYSCALL_2 (116 , ##a) -/* #define V850_SIM_SYS_utime(a...) V850_SIM_SYSCALL_2 (201 , ##a) */ -/* #define V850_SIM_SYS_wait(a...) V850_SIM_SYSCALL_1 (202 , ##a) */ - -#define V850_SIM_SYS_make_raw(a...) V850_SIM_SYSCALL_1 (1024 , ##a) - - -#define V850_SIM_SYSCALL_0(_call) \ -({ \ - register int call __asm__ ("r6") = _call; \ - register int rval __asm__ ("r10"); \ - __asm__ __volatile__ ("trap 31" \ - : "=r" (rval) \ - : "r" (call) \ - : "r11", "memory"); \ - rval; \ -}) -#define V850_SIM_SYSCALL_1(_call, _arg0) \ -({ \ - register int call __asm__ ("r6") = _call; \ - register long arg0 __asm__ ("r7") = (long)_arg0; \ - register int rval __asm__ ("r10"); \ - __asm__ __volatile__ ("trap 31" \ - : "=r" (rval) \ - : "r" (call), "r" (arg0) \ - : "r11", "memory"); \ - rval; \ -}) -#define V850_SIM_SYSCALL_2(_call, _arg0, _arg1) \ -({ \ - register int call __asm__ ("r6") = _call; \ - register long arg0 __asm__ ("r7") = (long)_arg0; \ - register long arg1 __asm__ ("r8") = (long)_arg1; \ - register int rval __asm__ ("r10"); \ - __asm__ __volatile__ ("trap 31" \ - : "=r" (rval) \ - : "r" (call), "r" (arg0), "r" (arg1) \ - : "r11", "memory"); \ - rval; \ -}) -#define V850_SIM_SYSCALL_3(_call, _arg0, _arg1, _arg2) \ -({ \ - register int call __asm__ ("r6") = _call; \ - register long arg0 __asm__ ("r7") = (long)_arg0; \ - register long arg1 __asm__ ("r8") = (long)_arg1; \ - register long arg2 __asm__ ("r9") = (long)_arg2; \ - register int rval __asm__ ("r10"); \ - __asm__ __volatile__ ("trap 31" \ - : "=r" (rval) \ - : "r" (call), "r" (arg0), "r" (arg1), "r" (arg2)\ - : "r11", "memory"); \ - rval; \ -}) - -#define V850_SIM_SYSCALL(call, args...) \ - V850_SIM_SYS_##call (args) - -#endif /* __V850_SIMSYSCALL_H__ */ diff --git a/include/asm-v850/socket.h b/include/asm-v850/socket.h deleted file mode 100644 index e199a2bf12aa..000000000000 --- a/include/asm-v850/socket.h +++ /dev/null @@ -1,57 +0,0 @@ -#ifndef __V850_SOCKET_H__ -#define __V850_SOCKET_H__ - -#include - -/* For setsockoptions(2) */ -#define SOL_SOCKET 1 - -#define SO_DEBUG 1 -#define SO_REUSEADDR 2 -#define SO_TYPE 3 -#define SO_ERROR 4 -#define SO_DONTROUTE 5 -#define SO_BROADCAST 6 -#define SO_SNDBUF 7 -#define SO_RCVBUF 8 -#define SO_SNDBUFFORCE 32 -#define SO_RCVBUFFORCE 33 -#define SO_KEEPALIVE 9 -#define SO_OOBINLINE 10 -#define SO_NO_CHECK 11 -#define SO_PRIORITY 12 -#define SO_LINGER 13 -#define SO_BSDCOMPAT 14 -/* To add :#define SO_REUSEPORT 15 */ -#define SO_PASSCRED 16 -#define SO_PEERCRED 17 -#define SO_RCVLOWAT 18 -#define SO_SNDLOWAT 19 -#define SO_RCVTIMEO 20 -#define SO_SNDTIMEO 21 - -/* Security levels - as per NRL IPv6 - don't actually do anything */ -#define SO_SECURITY_AUTHENTICATION 22 -#define SO_SECURITY_ENCRYPTION_TRANSPORT 23 -#define SO_SECURITY_ENCRYPTION_NETWORK 24 - -#define SO_BINDTODEVICE 25 - -/* Socket filtering */ -#define SO_ATTACH_FILTER 26 -#define SO_DETACH_FILTER 27 - -#define SO_PEERNAME 28 -#define SO_TIMESTAMP 29 -#define SCM_TIMESTAMP SO_TIMESTAMP - -#define SO_ACCEPTCONN 30 - -#define SO_PEERSEC 31 -#define SO_PASSSEC 34 -#define SO_TIMESTAMPNS 35 -#define SCM_TIMESTAMPNS SO_TIMESTAMPNS - -#define SO_MARK 36 - -#endif /* __V850_SOCKET_H__ */ diff --git a/include/asm-v850/sockios.h b/include/asm-v850/sockios.h deleted file mode 100644 index 823e106e6cd0..000000000000 --- a/include/asm-v850/sockios.h +++ /dev/null @@ -1,13 +0,0 @@ -#ifndef __V850_SOCKIOS_H__ -#define __V850_SOCKIOS_H__ - -/* Socket-level I/O control calls. */ -#define FIOSETOWN 0x8901 -#define SIOCSPGRP 0x8902 -#define FIOGETOWN 0x8903 -#define SIOCGPGRP 0x8904 -#define SIOCATMARK 0x8905 -#define SIOCGSTAMP 0x8906 /* Get stamp (timeval) */ -#define SIOCGSTAMPNS 0x8907 /* Get stamp (timespec) */ - -#endif /* __V850_SOCKIOS_H__ */ diff --git a/include/asm-v850/stat.h b/include/asm-v850/stat.h deleted file mode 100644 index c68c60d06e2f..000000000000 --- a/include/asm-v850/stat.h +++ /dev/null @@ -1,73 +0,0 @@ -/* - * include/asm-v850/stat.h -- v850 stat structure - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_STAT_H__ -#define __V850_STAT_H__ - -#include - -struct stat { - unsigned int st_dev; - unsigned long st_ino; - unsigned int st_mode; - unsigned int st_nlink; - unsigned int st_uid; - unsigned int st_gid; - unsigned int st_rdev; - long st_size; - unsigned long st_blksize; - unsigned long st_blocks; - unsigned long st_atime; - unsigned long __unused1; - unsigned long st_mtime; - unsigned long __unused2; - unsigned long st_ctime; - unsigned long __unused3; - unsigned long __unused4; - unsigned long __unused5; -}; - -struct stat64 { - unsigned long long st_dev; - unsigned long __unused1; - - unsigned long long st_ino; - - unsigned int st_mode; - unsigned int st_nlink; - - unsigned int st_uid; - unsigned int st_gid; - - unsigned long long st_rdev; - unsigned long __unused3; - - long long st_size; - unsigned long st_blksize; - - unsigned long st_blocks; /* No. of 512-byte blocks allocated */ - unsigned long __unused4; /* future possible st_blocks high bits */ - - unsigned long st_atime; - unsigned long st_atime_nsec; - - unsigned long st_mtime; - unsigned long st_mtime_nsec; - - unsigned long st_ctime; - unsigned long st_ctime_nsec; - - unsigned long __unused8; -}; - -#endif /* __V850_STAT_H__ */ diff --git a/include/asm-v850/statfs.h b/include/asm-v850/statfs.h deleted file mode 100644 index ea1596607f26..000000000000 --- a/include/asm-v850/statfs.h +++ /dev/null @@ -1,6 +0,0 @@ -#ifndef __V850_STATFS_H__ -#define __V850_STATFS_H__ - -#include - -#endif /* __V850_STATFS_H__ */ diff --git a/include/asm-v850/string.h b/include/asm-v850/string.h deleted file mode 100644 index 478e234789d6..000000000000 --- a/include/asm-v850/string.h +++ /dev/null @@ -1,25 +0,0 @@ -/* - * include/asm-v850/string.h -- Architecture specific string routines - * - * Copyright (C) 2001,02 NEC Corporation - * Copyright (C) 2001,02 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_STRING_H__ -#define __V850_STRING_H__ - -#define __HAVE_ARCH_MEMCPY -#define __HAVE_ARCH_MEMSET -#define __HAVE_ARCH_MEMMOVE - -extern void *memcpy (void *, const void *, __kernel_size_t); -extern void *memset (void *, int, __kernel_size_t); -extern void *memmove (void *, const void *, __kernel_size_t); - -#endif /* __V850_STRING_H__ */ diff --git a/include/asm-v850/system.h b/include/asm-v850/system.h deleted file mode 100644 index 7daf1fdee119..000000000000 --- a/include/asm-v850/system.h +++ /dev/null @@ -1,123 +0,0 @@ -/* - * include/asm-v850/system.h -- Low-level interrupt/thread ops - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_SYSTEM_H__ -#define __V850_SYSTEM_H__ - -#include -#include - - -/* - * switch_to(n) should switch tasks to task ptr, first checking that - * ptr isn't the current task, in which case it does nothing. - */ -struct thread_struct; -extern void *switch_thread (struct thread_struct *last, - struct thread_struct *next); -#define switch_to(prev,next,last) \ - do { \ - if (prev != next) { \ - (last) = switch_thread (&prev->thread, &next->thread); \ - } \ - } while (0) - - -/* Enable/disable interrupts. */ -#define local_irq_enable() __asm__ __volatile__ ("ei") -#define local_irq_disable() __asm__ __volatile__ ("di") - -#define local_save_flags(flags) \ - __asm__ __volatile__ ("stsr %1, %0" : "=r" (flags) : "i" (SR_PSW)) -#define local_restore_flags(flags) \ - __asm__ __volatile__ ("ldsr %0, %1" :: "r" (flags), "i" (SR_PSW)) - -/* For spinlocks etc */ -#define local_irq_save(flags) \ - do { local_save_flags (flags); local_irq_disable (); } while (0) -#define local_irq_restore(flags) \ - local_restore_flags (flags); - - -static inline int irqs_disabled (void) -{ - unsigned flags; - local_save_flags (flags); - return !!(flags & 0x20); -} - - -/* - * Force strict CPU ordering. - * Not really required on v850... - */ -#define nop() __asm__ __volatile__ ("nop") -#define mb() __asm__ __volatile__ ("" ::: "memory") -#define rmb() mb () -#define wmb() mb () -#define read_barrier_depends() ((void)0) -#define set_mb(var, value) do { xchg (&var, value); } while (0) - -#define smp_mb() mb () -#define smp_rmb() rmb () -#define smp_wmb() wmb () -#define smp_read_barrier_depends() read_barrier_depends() - -#define xchg(ptr, with) \ - ((__typeof__ (*(ptr)))__xchg ((unsigned long)(with), (ptr), sizeof (*(ptr)))) - -static inline unsigned long __xchg (unsigned long with, - __volatile__ void *ptr, int size) -{ - unsigned long tmp, flags; - - local_irq_save (flags); - - switch (size) { - case 1: - tmp = *(unsigned char *)ptr; - *(unsigned char *)ptr = with; - break; - case 2: - tmp = *(unsigned short *)ptr; - *(unsigned short *)ptr = with; - break; - case 4: - tmp = *(unsigned long *)ptr; - *(unsigned long *)ptr = with; - break; - } - - local_irq_restore (flags); - - return tmp; -} - -#include - -/* - * cmpxchg_local and cmpxchg64_local are atomic wrt current CPU. Always make - * them available. - */ -#define cmpxchg_local(ptr, o, n) \ - ((__typeof__(*(ptr)))__cmpxchg_local_generic((ptr), (unsigned long)(o),\ - (unsigned long)(n), sizeof(*(ptr)))) -#define cmpxchg64_local(ptr, o, n) __cmpxchg64_local_generic((ptr), (o), (n)) - -#ifndef CONFIG_SMP -#include -#endif - -#define arch_align_stack(x) (x) - -#endif /* __V850_SYSTEM_H__ */ diff --git a/include/asm-v850/teg.h b/include/asm-v850/teg.h deleted file mode 100644 index acc8c7d95329..000000000000 --- a/include/asm-v850/teg.h +++ /dev/null @@ -1,101 +0,0 @@ -/* - * include/asm-v850/teg.h -- NB85E-TEG cpu chip - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_TEG_H__ -#define __V850_TEG_H__ - - -/* The TEG uses the V850E cpu core. */ -#include -#include - - -#define CPU_MODEL "v850e/nb85e-teg" -#define CPU_MODEL_LONG "NEC V850E/NB85E TEG" - - -/* For */ -/* We use on-chip RAM, for a few miscellaneous variables that must be - accessible using a load instruction relative to R0. On the NB85E/TEG, - There's 60KB of iRAM starting at 0xFFFF0000, however we need the base - address to be addressable by a 16-bit signed offset, so we only use the - second half of it starting from 0xFFFF8000. */ -#define R0_RAM_ADDR 0xFFFF8000 - - -/* Hardware-specific interrupt numbers (in the kernel IRQ namespace). - Some of these are parameterized even though there's only a single - interrupt, for compatibility with some generic code that works on other - processor models. */ -#define IRQ_INTCMD(n) 6 /* interval timer interrupt */ -#define IRQ_INTCMD_NUM 1 -#define IRQ_INTSER(n) 16 /* UART reception error */ -#define IRQ_INTSER_NUM 1 -#define IRQ_INTSR(n) 17 /* UART reception completion */ -#define IRQ_INTSR_NUM 1 -#define IRQ_INTST(n) 18 /* UART transmission completion */ -#define IRQ_INTST_NUM 1 - -/* For */ -#define NUM_CPU_IRQS 64 - - -/* TEG UART details. */ -#define V850E_UART_BASE_ADDR(n) (0xFFFFF600 + 0x10 * (n)) -#define V850E_UART_ASIM_ADDR(n) (V850E_UART_BASE_ADDR(n) + 0x0) -#define V850E_UART_ASIS_ADDR(n) (V850E_UART_BASE_ADDR(n) + 0x2) -#define V850E_UART_ASIF_ADDR(n) (V850E_UART_BASE_ADDR(n) + 0x4) -#define V850E_UART_CKSR_ADDR(n) (V850E_UART_BASE_ADDR(n) + 0x6) -#define V850E_UART_BRGC_ADDR(n) (V850E_UART_BASE_ADDR(n) + 0x8) -#define V850E_UART_TXB_ADDR(n) (V850E_UART_BASE_ADDR(n) + 0xA) -#define V850E_UART_RXB_ADDR(n) (V850E_UART_BASE_ADDR(n) + 0xC) -#define V850E_UART_NUM_CHANNELS 1 -#define V850E_UART_BASE_FREQ CPU_CLOCK_FREQ -/* This is a function that gets called before configuring the UART. */ -#define V850E_UART_PRE_CONFIGURE teg_uart_pre_configure -#ifndef __ASSEMBLY__ -extern void teg_uart_pre_configure (unsigned chan, - unsigned cflags, unsigned baud); -#endif - - -/* The TEG RTPU. */ -#define V850E_RTPU_BASE_ADDR 0xFFFFF210 - - -/* TEG series timer D details. */ -#define V850E_TIMER_D_BASE_ADDR 0xFFFFF210 -#define V850E_TIMER_D_TMCD_BASE_ADDR (V850E_TIMER_D_BASE_ADDR + 0x0) -#define V850E_TIMER_D_TMD_BASE_ADDR (V850E_TIMER_D_BASE_ADDR + 0x4) -#define V850E_TIMER_D_CMD_BASE_ADDR (V850E_TIMER_D_BASE_ADDR + 0x8) -#define V850E_TIMER_D_BASE_FREQ CPU_CLOCK_FREQ - - -/* `Interrupt Source Select' control register. */ -#define TEG_ISS_ADDR 0xFFFFF7FA -#define TEG_ISS (*(volatile u8 *)TEG_ISS_ADDR) - -/* Port 0 I/O register (bits 0-3 used). */ -#define TEG_PORT0_IO_ADDR 0xFFFFF7F2 -#define TEG_PORT0_IO (*(volatile u8 *)TEG_PORT0_IO_ADDR) -/* Port 0 control register (bits 0-3 control mode, 0 = output, 1 = input). */ -#define TEG_PORT0_PM_ADDR 0xFFFFF7F4 -#define TEG_PORT0_PM (*(volatile u8 *)TEG_PORT0_PM_ADDR) - - -#ifndef __ASSEMBLY__ -extern void teg_init_irqs (void); -#endif - - -#endif /* __V850_TEG_H__ */ diff --git a/include/asm-v850/termbits.h b/include/asm-v850/termbits.h deleted file mode 100644 index 295d7bf69451..000000000000 --- a/include/asm-v850/termbits.h +++ /dev/null @@ -1,200 +0,0 @@ -#ifndef __V850_TERMBITS_H__ -#define __V850_TERMBITS_H__ - -#include - -typedef unsigned char cc_t; -typedef unsigned int speed_t; -typedef unsigned int tcflag_t; - -#define NCCS 19 -struct termios { - tcflag_t c_iflag; /* input mode flags */ - tcflag_t c_oflag; /* output mode flags */ - tcflag_t c_cflag; /* control mode flags */ - tcflag_t c_lflag; /* local mode flags */ - cc_t c_line; /* line discipline */ - cc_t c_cc[NCCS]; /* control characters */ -}; - -struct termios2 { - tcflag_t c_iflag; /* input mode flags */ - tcflag_t c_oflag; /* output mode flags */ - tcflag_t c_cflag; /* control mode flags */ - tcflag_t c_lflag; /* local mode flags */ - cc_t c_line; /* line discipline */ - cc_t c_cc[NCCS]; /* control characters */ - speed_t c_ispeed; /* input speed */ - speed_t c_ospeed; /* output speed */ -}; - -struct ktermios { - tcflag_t c_iflag; /* input mode flags */ - tcflag_t c_oflag; /* output mode flags */ - tcflag_t c_cflag; /* control mode flags */ - tcflag_t c_lflag; /* local mode flags */ - cc_t c_line; /* line discipline */ - cc_t c_cc[NCCS]; /* control characters */ - speed_t c_ispeed; /* input speed */ - speed_t c_ospeed; /* output speed */ -}; - -/* c_cc characters */ -#define VINTR 0 -#define VQUIT 1 -#define VERASE 2 -#define VKILL 3 -#define VEOF 4 -#define VTIME 5 -#define VMIN 6 -#define VSWTC 7 -#define VSTART 8 -#define VSTOP 9 -#define VSUSP 10 -#define VEOL 11 -#define VREPRINT 12 -#define VDISCARD 13 -#define VWERASE 14 -#define VLNEXT 15 -#define VEOL2 16 - - -/* c_iflag bits */ -#define IGNBRK 0000001 -#define BRKINT 0000002 -#define IGNPAR 0000004 -#define PARMRK 0000010 -#define INPCK 0000020 -#define ISTRIP 0000040 -#define INLCR 0000100 -#define IGNCR 0000200 -#define ICRNL 0000400 -#define IUCLC 0001000 -#define IXON 0002000 -#define IXANY 0004000 -#define IXOFF 0010000 -#define IMAXBEL 0020000 -#define IUTF8 0040000 - -/* c_oflag bits */ -#define OPOST 0000001 -#define OLCUC 0000002 -#define ONLCR 0000004 -#define OCRNL 0000010 -#define ONOCR 0000020 -#define ONLRET 0000040 -#define OFILL 0000100 -#define OFDEL 0000200 -#define NLDLY 0000400 -#define NL0 0000000 -#define NL1 0000400 -#define CRDLY 0003000 -#define CR0 0000000 -#define CR1 0001000 -#define CR2 0002000 -#define CR3 0003000 -#define TABDLY 0014000 -#define TAB0 0000000 -#define TAB1 0004000 -#define TAB2 0010000 -#define TAB3 0014000 -#define XTABS 0014000 -#define BSDLY 0020000 -#define BS0 0000000 -#define BS1 0020000 -#define VTDLY 0040000 -#define VT0 0000000 -#define VT1 0040000 -#define FFDLY 0100000 -#define FF0 0000000 -#define FF1 0100000 - -/* c_cflag bit meaning */ -#define CBAUD 0010017 -#define B0 0000000 /* hang up */ -#define B50 0000001 -#define B75 0000002 -#define B110 0000003 -#define B134 0000004 -#define B150 0000005 -#define B200 0000006 -#define B300 0000007 -#define B600 0000010 -#define B1200 0000011 -#define B1800 0000012 -#define B2400 0000013 -#define B4800 0000014 -#define B9600 0000015 -#define B19200 0000016 -#define B38400 0000017 -#define EXTA B19200 -#define EXTB B38400 -#define CSIZE 0000060 -#define CS5 0000000 -#define CS6 0000020 -#define CS7 0000040 -#define CS8 0000060 -#define CSTOPB 0000100 -#define CREAD 0000200 -#define PARENB 0000400 -#define PARODD 0001000 -#define HUPCL 0002000 -#define CLOCAL 0004000 -#define CBAUDEX 0010000 -#define BOTHER 0010000 -#define B57600 0010001 -#define B115200 0010002 -#define B230400 0010003 -#define B460800 0010004 -#define B500000 0010005 -#define B576000 0010006 -#define B921600 0010007 -#define B1000000 0010010 -#define B1152000 0010011 -#define B1500000 0010012 -#define B2000000 0010013 -#define B2500000 0010014 -#define B3000000 0010015 -#define B3500000 0010016 -#define B4000000 0010017 -#define CIBAUD 002003600000 /* input baud rate */ -#define CMSPAR 010000000000 /* mark or space (stick) parity */ -#define CRTSCTS 020000000000 /* flow control */ - -#define IBSHIFT 16 /* Shifr from CBAUD to CIBAUD */ - -/* c_lflag bits */ -#define ISIG 0000001 -#define ICANON 0000002 -#define XCASE 0000004 -#define ECHO 0000010 -#define ECHOE 0000020 -#define ECHOK 0000040 -#define ECHONL 0000100 -#define NOFLSH 0000200 -#define TOSTOP 0000400 -#define ECHOCTL 0001000 -#define ECHOPRT 0002000 -#define ECHOKE 0004000 -#define FLUSHO 0010000 -#define PENDIN 0040000 -#define IEXTEN 0100000 - - -/* tcflow() and TCXONC use these */ -#define TCOOFF 0 -#define TCOON 1 -#define TCIOFF 2 -#define TCION 3 - -/* tcflush() and TCFLSH use these */ -#define TCIFLUSH 0 -#define TCOFLUSH 1 -#define TCIOFLUSH 2 - -/* tcsetattr uses these */ -#define TCSANOW 0 -#define TCSADRAIN 1 -#define TCSAFLUSH 2 - -#endif /* __V850_TERMBITS_H__ */ diff --git a/include/asm-v850/termios.h b/include/asm-v850/termios.h deleted file mode 100644 index fcd171838d9c..000000000000 --- a/include/asm-v850/termios.h +++ /dev/null @@ -1,90 +0,0 @@ -#ifndef __V850_TERMIOS_H__ -#define __V850_TERMIOS_H__ - -#include -#include - -struct winsize { - unsigned short ws_row; - unsigned short ws_col; - unsigned short ws_xpixel; - unsigned short ws_ypixel; -}; - -#define NCC 8 -struct termio { - unsigned short c_iflag; /* input mode flags */ - unsigned short c_oflag; /* output mode flags */ - unsigned short c_cflag; /* control mode flags */ - unsigned short c_lflag; /* local mode flags */ - unsigned char c_line; /* line discipline */ - unsigned char c_cc[NCC]; /* control characters */ -}; - -/* modem lines */ -#define TIOCM_LE 0x001 -#define TIOCM_DTR 0x002 -#define TIOCM_RTS 0x004 -#define TIOCM_ST 0x008 -#define TIOCM_SR 0x010 -#define TIOCM_CTS 0x020 -#define TIOCM_CAR 0x040 -#define TIOCM_RNG 0x080 -#define TIOCM_DSR 0x100 -#define TIOCM_CD TIOCM_CAR -#define TIOCM_RI TIOCM_RNG -#define TIOCM_OUT1 0x2000 -#define TIOCM_OUT2 0x4000 -#define TIOCM_LOOP 0x8000 - -/* ioctl (fd, TIOCSERGETLSR, &result) where result may be as below */ - -#ifdef __KERNEL__ - -/* intr=^C quit=^\ erase=del kill=^U - eof=^D vtime=\0 vmin=\1 sxtc=\0 - start=^Q stop=^S susp=^Z eol=\0 - reprint=^R discard=^U werase=^W lnext=^V - eol2=\0 -*/ -#define INIT_C_CC "\003\034\177\025\004\0\1\0\021\023\032\0\022\017\027\026\0" - -/* - * Translate a "termio" structure into a "termios". Ugh. - */ -#define SET_LOW_TERMIOS_BITS(termios, termio, x) { \ - unsigned short __tmp; \ - get_user(__tmp,&(termio)->x); \ - *(unsigned short *) &(termios)->x = __tmp; \ -} - -#define user_termio_to_kernel_termios(termios, termio) \ -({ \ - SET_LOW_TERMIOS_BITS(termios, termio, c_iflag); \ - SET_LOW_TERMIOS_BITS(termios, termio, c_oflag); \ - SET_LOW_TERMIOS_BITS(termios, termio, c_cflag); \ - SET_LOW_TERMIOS_BITS(termios, termio, c_lflag); \ - copy_from_user((termios)->c_cc, (termio)->c_cc, NCC); \ -}) - -/* - * Translate a "termios" structure into a "termio". Ugh. - */ -#define kernel_termios_to_user_termio(termio, termios) \ -({ \ - put_user((termios)->c_iflag, &(termio)->c_iflag); \ - put_user((termios)->c_oflag, &(termio)->c_oflag); \ - put_user((termios)->c_cflag, &(termio)->c_cflag); \ - put_user((termios)->c_lflag, &(termio)->c_lflag); \ - put_user((termios)->c_line, &(termio)->c_line); \ - copy_to_user((termio)->c_cc, (termios)->c_cc, NCC); \ -}) - -#define user_termios_to_kernel_termios(k, u) copy_from_user(k, u, sizeof(struct termios2)) -#define kernel_termios_to_user_termios(u, k) copy_to_user(u, k, sizeof(struct termios2)) -#define user_termios_to_kernel_termios_1(k, u) copy_from_user(k, u, sizeof(struct termios)) -#define kernel_termios_to_user_termios_1(u, k) copy_to_user(u, k, sizeof(struct termios)) - -#endif /* __KERNEL__ */ - -#endif /* __V850_TERMIOS_H__ */ diff --git a/include/asm-v850/thread_info.h b/include/asm-v850/thread_info.h deleted file mode 100644 index 1a9e6ae0c5fd..000000000000 --- a/include/asm-v850/thread_info.h +++ /dev/null @@ -1,129 +0,0 @@ -/* - * include/asm-v850/thread_info.h -- v850 low-level thread information - * - * Copyright (C) 2002 NEC Corporation - * Copyright (C) 2002 Miles Bader - * Copyright (C) 2002 David Howells (dhowells@redhat.com) - * - Incorporating suggestions made by Linus Torvalds and Dave Miller - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * This file was derived from the PPC version, include/asm-ppc/thread_info.h - * which was adapted from the i386 version by Paul Mackerras - */ - -#ifndef __V850_THREAD_INFO_H__ -#define __V850_THREAD_INFO_H__ - -#ifdef __KERNEL__ - -#ifndef __ASSEMBLY__ - -/* - * low level task data. - * If you change this, change the TI_* offsets below to match. - */ -struct thread_info { - struct task_struct *task; /* main task structure */ - struct exec_domain *exec_domain; /* execution domain */ - unsigned long flags; /* low level flags */ - int cpu; /* cpu we're on */ - int preempt_count; /* 0 => preemptable, - <0 => BUG */ - struct restart_block restart_block; -}; - -#define INIT_THREAD_INFO(tsk) \ -{ \ - .task = &tsk, \ - .exec_domain = &default_exec_domain, \ - .flags = 0, \ - .cpu = 0, \ - .preempt_count = 1, \ - .restart_block = { \ - .fn = do_no_restart_syscall, \ - }, \ -} - -#define init_thread_info (init_thread_union.thread_info) -#define init_stack (init_thread_union.stack) - -/* - * macros/functions for gaining access to the thread information structure - */ - -/* thread information allocation */ -#define alloc_thread_info(tsk) ((struct thread_info *) \ - __get_free_pages(GFP_KERNEL, 1)) -#define free_thread_info(ti) free_pages((unsigned long) (ti), 1) - -#endif /* __ASSEMBLY__ */ - - -/* - * Offsets in thread_info structure, used in assembly code - */ -#define TI_TASK 0 -#define TI_EXECDOMAIN 4 -#define TI_FLAGS 8 -#define TI_CPU 12 -#define TI_PREEMPT 16 - -#define PREEMPT_ACTIVE 0x4000000 - -/* - * thread information flag bit numbers - */ -#define TIF_SYSCALL_TRACE 0 /* syscall trace active */ -#define TIF_SIGPENDING 1 /* signal pending */ -#define TIF_NEED_RESCHED 2 /* rescheduling necessary */ -#define TIF_POLLING_NRFLAG 3 /* true if poll_idle() is polling - TIF_NEED_RESCHED */ -#define TIF_MEMDIE 4 - -/* as above, but as bit values */ -#define _TIF_SYSCALL_TRACE (1< - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_TLB_H__ -#define __V850_TLB_H__ - -#define tlb_flush(tlb) ((void)0) - -#include - -#endif /* __V850_TLB_H__ */ diff --git a/include/asm-v850/tlbflush.h b/include/asm-v850/tlbflush.h deleted file mode 100644 index c44aa64449c8..000000000000 --- a/include/asm-v850/tlbflush.h +++ /dev/null @@ -1,64 +0,0 @@ -/* - * include/asm-v850/tlbflush.h - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_TLBFLUSH_H__ -#define __V850_TLBFLUSH_H__ - -#include - - -/* - * flush all user-space atc entries. - */ -static inline void __flush_tlb(void) -{ - BUG (); -} - -static inline void __flush_tlb_one(unsigned long addr) -{ - BUG (); -} - -#define flush_tlb() __flush_tlb() - -/* - * flush all atc entries (both kernel and user-space entries). - */ -static inline void flush_tlb_all(void) -{ - BUG (); -} - -static inline void flush_tlb_mm(struct mm_struct *mm) -{ - BUG (); -} - -static inline void flush_tlb_page(struct vm_area_struct *vma, unsigned long addr) -{ - BUG (); -} - -static inline void flush_tlb_range(struct vm_area_struct *vma, - unsigned long start, unsigned long end) -{ - BUG (); -} - -static inline void flush_tlb_kernel_page(unsigned long addr) -{ - BUG (); -} - -#endif /* __V850_TLBFLUSH_H__ */ diff --git a/include/asm-v850/topology.h b/include/asm-v850/topology.h deleted file mode 100644 index 6040e41d7945..000000000000 --- a/include/asm-v850/topology.h +++ /dev/null @@ -1,6 +0,0 @@ -#ifndef __V850_TOPOLOGY_H__ -#define __V850_TOPOLOGY_H__ - -#include - -#endif /* __V850_TOPOLOGY_H__ */ diff --git a/include/asm-v850/types.h b/include/asm-v850/types.h deleted file mode 100644 index 89f735ee41dd..000000000000 --- a/include/asm-v850/types.h +++ /dev/null @@ -1,36 +0,0 @@ -#ifndef __V850_TYPES_H__ -#define __V850_TYPES_H__ - -#ifndef __ASSEMBLY__ - -/* - * This file is never included by application software unless - * explicitly requested (e.g., via linux/types.h) in which case the - * application is Linux specific so (user-) name space pollution is - * not a major issue. However, for interoperability, libraries still - * need to be careful to avoid a name clashes. - */ -#include - -typedef unsigned short umode_t; - -#endif /* !__ASSEMBLY__ */ - -/* - * These aren't exported outside the kernel to avoid name space clashes - */ -#ifdef __KERNEL__ - -#define BITS_PER_LONG 32 - -#ifndef __ASSEMBLY__ - -/* Dma addresses are 32-bits wide. */ - -typedef u32 dma_addr_t; - -#endif /* !__ASSEMBLY__ */ - -#endif /* __KERNEL__ */ - -#endif /* __V850_TYPES_H__ */ diff --git a/include/asm-v850/uaccess.h b/include/asm-v850/uaccess.h deleted file mode 100644 index 64563c409bb2..000000000000 --- a/include/asm-v850/uaccess.h +++ /dev/null @@ -1,159 +0,0 @@ -#ifndef __V850_UACCESS_H__ -#define __V850_UACCESS_H__ - -/* - * User space memory access functions - */ - -#include -#include - -#include -#include - -#define VERIFY_READ 0 -#define VERIFY_WRITE 1 - -static inline int access_ok (int type, const void *addr, unsigned long size) -{ - /* XXX I guess we should check against real ram bounds at least, and - possibly make sure ADDR is not within the kernel. - For now we just check to make sure it's not a small positive - or negative value, as that will at least catch some kinds of - error. In particular, we make sure that ADDR's not within the - interrupt vector area, which we know starts at zero, or within the - peripheral-I/O area, which is located just _before_ zero. */ - unsigned long val = (unsigned long)addr; - return val >= (0x80 + NUM_CPU_IRQS*16) && val < 0xFFFFF000; -} - -/* - * The exception table consists of pairs of addresses: the first is the - * address of an instruction that is allowed to fault, and the second is - * the address at which the program should continue. No registers are - * modified, so it is entirely up to the continuation code to figure out - * what to do. - * - * All the routines below use bits of fixup code that are out of line - * with the main instruction path. This means when everything is well, - * we don't even have to jump over them. Further, they do not intrude - * on our cache or tlb entries. - */ - -struct exception_table_entry -{ - unsigned long insn, fixup; -}; - -/* Returns 0 if exception not found and fixup otherwise. */ -extern unsigned long search_exception_table (unsigned long); - - -/* - * These are the main single-value transfer routines. They automatically - * use the right size if we just have the right pointer type. - */ - -extern int bad_user_access_length (void); - -#define __get_user(var, ptr) \ - ({ \ - int __gu_err = 0; \ - typeof(*(ptr)) __gu_val = 0; \ - switch (sizeof (*(ptr))) { \ - case 1: \ - case 2: \ - case 4: \ - __gu_val = *(ptr); \ - break; \ - case 8: \ - memcpy(&__gu_val, ptr, sizeof(__gu_val)); \ - break; \ - default: \ - __gu_val = 0; \ - __gu_err = __get_user_bad (); \ - break; \ - } \ - (var) = __gu_val; \ - __gu_err; \ - }) -#define __get_user_bad() (bad_user_access_length (), (-EFAULT)) - -#define __put_user(var, ptr) \ - ({ \ - int __pu_err = 0; \ - switch (sizeof (*(ptr))) { \ - case 1: \ - case 2: \ - case 4: \ - *(ptr) = (var); \ - break; \ - case 8: { \ - typeof(*(ptr)) __pu_val = 0; \ - memcpy(ptr, &__pu_val, sizeof(__pu_val)); \ - } \ - break; \ - default: \ - __pu_err = __put_user_bad (); \ - break; \ - } \ - __pu_err; \ - }) -#define __put_user_bad() (bad_user_access_length (), (-EFAULT)) - -#define put_user(x, ptr) __put_user(x, ptr) -#define get_user(x, ptr) __get_user(x, ptr) - -#define __copy_from_user(to, from, n) (memcpy (to, from, n), 0) -#define __copy_to_user(to, from, n) (memcpy(to, from, n), 0) - -#define __copy_to_user_inatomic __copy_to_user -#define __copy_from_user_inatomic __copy_from_user - -#define copy_from_user(to, from, n) __copy_from_user (to, from, n) -#define copy_to_user(to, from, n) __copy_to_user(to, from, n) - -#define copy_to_user_ret(to,from,n,retval) \ - ({ if (copy_to_user (to,from,n)) return retval; }) - -#define copy_from_user_ret(to,from,n,retval) \ - ({ if (copy_from_user (to,from,n)) return retval; }) - -/* - * Copy a null terminated string from userspace. - */ - -static inline long -strncpy_from_user (char *dst, const char *src, long count) -{ - char *tmp; - strncpy (dst, src, count); - for (tmp = dst; *tmp && count > 0; tmp++, count--) - ; - return tmp - dst; -} - -/* - * Return the size of a string (including the ending 0) - * - * Return 0 on exception, a value greater than N if too long - */ -static inline long strnlen_user (const char *src, long n) -{ - return strlen (src) + 1; -} - -#define strlen_user(str) strnlen_user (str, 32767) - -/* - * Zero Userspace - */ - -static inline unsigned long -clear_user (void *to, unsigned long n) -{ - memset (to, 0, n); - return 0; -} - -#endif /* __V850_UACCESS_H__ */ diff --git a/include/asm-v850/ucontext.h b/include/asm-v850/ucontext.h deleted file mode 100644 index 303c21590cff..000000000000 --- a/include/asm-v850/ucontext.h +++ /dev/null @@ -1,14 +0,0 @@ -#ifndef __V850_UCONTEXT_H__ -#define __V850_UCONTEXT_H__ - -#include - -struct ucontext { - unsigned long uc_flags; - struct ucontext *uc_link; - stack_t uc_stack; - struct sigcontext uc_mcontext; - sigset_t uc_sigmask; /* mask last for extensibility */ -}; - -#endif /* __V850_UCONTEXT_H__ */ diff --git a/include/asm-v850/unaligned.h b/include/asm-v850/unaligned.h deleted file mode 100644 index 53122b28491e..000000000000 --- a/include/asm-v850/unaligned.h +++ /dev/null @@ -1,22 +0,0 @@ -/* - * Copyright (C) 2001 NEC Corporation - * Copyright (C) 2001 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Note that some v850 chips support unaligned access, but it seems too - * annoying to use. - */ -#ifndef _ASM_V850_UNALIGNED_H -#define _ASM_V850_UNALIGNED_H - -#include -#include -#include - -#define get_unaligned __get_unaligned_le -#define put_unaligned __put_unaligned_le - -#endif /* _ASM_V850_UNALIGNED_H */ diff --git a/include/asm-v850/unistd.h b/include/asm-v850/unistd.h deleted file mode 100644 index 2241ed45ecfe..000000000000 --- a/include/asm-v850/unistd.h +++ /dev/null @@ -1,244 +0,0 @@ -/* - * include/asm-v850/unistd.h -- System call numbers and invocation mechanism - * - * Copyright (C) 2001,02,03,04 NEC Electronics Corporation - * Copyright (C) 2001,02,03,04 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_UNISTD_H__ -#define __V850_UNISTD_H__ - -#define __NR_restart_syscall 0 -#define __NR_exit 1 -#define __NR_fork 2 -#define __NR_read 3 -#define __NR_write 4 -#define __NR_open 5 -#define __NR_close 6 -#define __NR_waitpid 7 -#define __NR_creat 8 -#define __NR_link 9 -#define __NR_unlink 10 -#define __NR_execve 11 -#define __NR_chdir 12 -#define __NR_time 13 -#define __NR_mknod 14 -#define __NR_chmod 15 -#define __NR_chown 16 -#define __NR_break 17 -#define __NR_lseek 19 -#define __NR_getpid 20 -#define __NR_mount 21 -#define __NR_umount 22 -#define __NR_setuid 23 -#define __NR_getuid 24 -#define __NR_stime 25 -#define __NR_ptrace 26 -#define __NR_alarm 27 -#define __NR_pause 29 -#define __NR_utime 30 -#define __NR_stty 31 -#define __NR_gtty 32 -#define __NR_access 33 -#define __NR_nice 34 -#define __NR_ftime 35 -#define __NR_sync 36 -#define __NR_kill 37 -#define __NR_rename 38 -#define __NR_mkdir 39 -#define __NR_rmdir 40 -#define __NR_dup 41 -#define __NR_pipe 42 -#define __NR_times 43 -#define __NR_prof 44 -#define __NR_brk 45 -#define __NR_setgid 46 -#define __NR_getgid 47 -#define __NR_signal 48 -#define __NR_geteuid 49 -#define __NR_getegid 50 -#define __NR_acct 51 -#define __NR_umount2 52 -#define __NR_lock 53 -#define __NR_ioctl 54 -#define __NR_fcntl 55 -#define __NR_setpgid 57 -#define __NR_umask 60 -#define __NR_chroot 61 -#define __NR_ustat 62 -#define __NR_dup2 63 -#define __NR_getppid 64 -#define __NR_getpgrp 65 -#define __NR_setsid 66 -#define __NR_sigaction 67 -#define __NR_sgetmask 68 -#define __NR_ssetmask 69 -#define __NR_setreuid 70 -#define __NR_setregid 71 -#define __NR_sigsuspend 72 -#define __NR_sigpending 73 -#define __NR_sethostname 74 -#define __NR_setrlimit 75 -#define __NR_ugetrlimit 76 -#define __NR_getrusage 77 -#define __NR_gettimeofday 78 -#define __NR_settimeofday 79 -#define __NR_getgroups 80 -#define __NR_setgroups 81 -#define __NR_select 82 -#define __NR_symlink 83 -#define __NR_readlink 85 -#define __NR_uselib 86 -#define __NR_swapon 87 -#define __NR_reboot 88 -#define __NR_readdir 89 -#define __NR_mmap 90 -#define __NR_munmap 91 -#define __NR_truncate 92 -#define __NR_ftruncate 93 -#define __NR_fchmod 94 -#define __NR_fchown 95 -#define __NR_getpriority 96 -#define __NR_setpriority 97 -#define __NR_profil 98 -#define __NR_statfs 99 -#define __NR_fstatfs 100 -#define __NR_socketcall 102 -#define __NR_syslog 103 -#define __NR_setitimer 104 -#define __NR_getitimer 105 -#define __NR_stat 106 -#define __NR_lstat 107 -#define __NR_fstat 108 -#define __NR_vhangup 111 -#define __NR_wait4 114 -#define __NR_swapoff 115 -#define __NR_sysinfo 116 -#define __NR_ipc 117 -#define __NR_fsync 118 -#define __NR_sigreturn 119 -#define __NR_clone 120 -#define __NR_setdomainname 121 -#define __NR_uname 122 -#define __NR_cacheflush 123 -#define __NR_adjtimex 124 -#define __NR_mprotect 125 -#define __NR_sigprocmask 126 -#define __NR_create_module 127 -#define __NR_init_module 128 -#define __NR_delete_module 129 -#define __NR_get_kernel_syms 130 -#define __NR_quotactl 131 -#define __NR_getpgid 132 -#define __NR_fchdir 133 -#define __NR_bdflush 134 -#define __NR_sysfs 135 -#define __NR_personality 136 -#define __NR_afs_syscall 137 /* Syscall for Andrew File System */ -#define __NR_setfsuid 138 -#define __NR_setfsgid 139 -#define __NR__llseek 140 -#define __NR_getdents 141 -#define __NR_flock 143 -#define __NR_msync 144 -#define __NR_readv 145 -#define __NR_writev 146 -#define __NR_getsid 147 -#define __NR_fdatasync 148 -#define __NR__sysctl 149 -#define __NR_mlock 150 -#define __NR_munlock 151 -#define __NR_mlockall 152 -#define __NR_munlockall 153 -#define __NR_sched_setparam 154 -#define __NR_sched_getparam 155 -#define __NR_sched_setscheduler 156 -#define __NR_sched_getscheduler 157 -#define __NR_sched_yield 158 -#define __NR_sched_get_priority_max 159 -#define __NR_sched_get_priority_min 160 -#define __NR_sched_rr_get_interval 161 -#define __NR_nanosleep 162 -#define __NR_mremap 163 -#define __NR_setresuid 164 -#define __NR_getresuid 165 -#define __NR_query_module 167 -#define __NR_poll 168 -#define __NR_nfsservctl 169 -#define __NR_setresgid 170 -#define __NR_getresgid 171 -#define __NR_prctl 172 -#define __NR_rt_sigreturn 173 -#define __NR_rt_sigaction 174 -#define __NR_rt_sigprocmask 175 -#define __NR_rt_sigpending 176 -#define __NR_rt_sigtimedwait 177 -#define __NR_rt_sigqueueinfo 178 -#define __NR_rt_sigsuspend 179 -#define __NR_pread 180 -#define __NR_pwrite 181 -#define __NR_lchown 182 -#define __NR_getcwd 183 -#define __NR_capget 184 -#define __NR_capset 185 -#define __NR_sigaltstack 186 -#define __NR_sendfile 187 -#define __NR_getpmsg 188 /* some people actually want streams */ -#define __NR_putpmsg 189 /* some people actually want streams */ -#define __NR_vfork 190 -#define __NR_mmap2 192 -#define __NR_truncate64 193 -#define __NR_ftruncate64 194 -#define __NR_stat64 195 -#define __NR_lstat64 196 -#define __NR_fstat64 197 -#define __NR_fcntl64 198 -#define __NR_getdents64 199 -#define __NR_pivot_root 200 -#define __NR_gettid 201 -#define __NR_tkill 202 - -#ifdef __KERNEL__ - -#define __ARCH_WANT_IPC_PARSE_VERSION -#define __ARCH_WANT_OLD_READDIR -#define __ARCH_WANT_STAT64 -#define __ARCH_WANT_SYS_ALARM -#define __ARCH_WANT_SYS_GETHOSTNAME -#define __ARCH_WANT_SYS_PAUSE -#define __ARCH_WANT_SYS_SGETMASK -#define __ARCH_WANT_SYS_SIGNAL -#define __ARCH_WANT_SYS_TIME -#define __ARCH_WANT_SYS_UTIME -#define __ARCH_WANT_SYS_WAITPID -#define __ARCH_WANT_SYS_SOCKETCALL -#define __ARCH_WANT_SYS_FADVISE64 -#define __ARCH_WANT_SYS_GETPGRP -#define __ARCH_WANT_SYS_LLSEEK -#define __ARCH_WANT_SYS_NICE -#define __ARCH_WANT_SYS_OLDUMOUNT -#define __ARCH_WANT_SYS_SIGPENDING -#define __ARCH_WANT_SYS_SIGPROCMASK -#define __ARCH_WANT_SYS_RT_SIGACTION - -/* - * "Conditional" syscalls - */ -#define cond_syscall(name) \ - asm (".weak\t" C_SYMBOL_STRING(name) ";" \ - ".set\t" C_SYMBOL_STRING(name) "," C_SYMBOL_STRING(sys_ni_syscall)) -#if 0 -/* This doesn't work if there's a function prototype for NAME visible, - because the argument types probably won't match. */ -#define cond_syscall(name) \ - void name (void) __attribute__ ((weak, alias ("sys_ni_syscall"))); -#endif - -#endif /* __KERNEL__ */ -#endif /* __V850_UNISTD_H__ */ diff --git a/include/asm-v850/user.h b/include/asm-v850/user.h deleted file mode 100644 index 63cdc567d272..000000000000 --- a/include/asm-v850/user.h +++ /dev/null @@ -1,52 +0,0 @@ -#ifndef __V850_USER_H__ -#define __V850_USER_H__ - -/* Adapted from . */ - -#include -#include - -/* - * Core file format: The core file is written in such a way that gdb - * can understand it and provide useful information to the user (under - * linux we use the `trad-core' bfd, NOT the osf-core). The file contents - * are as follows: - * - * upage: 1 page consisting of a user struct that tells gdb - * what is present in the file. Directly after this is a - * copy of the task_struct, which is currently not used by gdb, - * but it may come in handy at some point. All of the registers - * are stored as part of the upage. The upage should always be - * only one page long. - * data: The data segment follows next. We use current->end_text to - * current->brk to pick up all of the user variables, plus any memory - * that may have been sbrk'ed. No attempt is made to determine if a - * page is demand-zero or if a page is totally unused, we just cover - * the entire range. All of the addresses are rounded in such a way - * that an integral number of pages is written. - * stack: We need the stack information in order to get a meaningful - * backtrace. We need to write the data from usp to - * current->start_stack, so we round each of these in order to be able - * to write an integer number of pages. - */ -struct user { - struct pt_regs regs; /* entire machine state */ - size_t u_tsize; /* text size (pages) */ - size_t u_dsize; /* data size (pages) */ - size_t u_ssize; /* stack size (pages) */ - unsigned long start_code; /* text starting address */ - unsigned long start_data; /* data starting address */ - unsigned long start_stack; /* stack starting address */ - long int signal; /* signal causing core dump */ - unsigned long u_ar0; /* help gdb find registers */ - unsigned long magic; /* identifies a core file */ - char u_comm[32]; /* user command name */ -}; - -#define NBPG PAGE_SIZE -#define UPAGES 1 -#define HOST_TEXT_START_ADDR (u.start_code) -#define HOST_DATA_START_ADDR (u.start_data) -#define HOST_STACK_END_ADDR (u.start_stack + u.u_ssize * NBPG) - -#endif /* __V850_USER_H__ */ diff --git a/include/asm-v850/v850e.h b/include/asm-v850/v850e.h deleted file mode 100644 index 5a222eb5117f..000000000000 --- a/include/asm-v850/v850e.h +++ /dev/null @@ -1,21 +0,0 @@ -/* - * include/asm-v850/v850e.h -- V850E CPU - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_V850E_H__ -#define __V850_V850E_H__ - -#include - -#define CPU_ARCH "v850e" - -#endif /* __V850_V850E_H__ */ diff --git a/include/asm-v850/v850e2.h b/include/asm-v850/v850e2.h deleted file mode 100644 index 48680408ab7e..000000000000 --- a/include/asm-v850/v850e2.h +++ /dev/null @@ -1,69 +0,0 @@ -/* - * include/asm-v850/v850e2.h -- Machine-dependent defs for V850E2 CPUs - * - * Copyright (C) 2002,03 NEC Electronics Corporation - * Copyright (C) 2002,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_V850E2_H__ -#define __V850_V850E2_H__ - -#include /* v850e-style interrupt system. */ - - -#define CPU_ARCH "v850e2" - - -/* Control registers. */ - -/* Chip area select control */ -#define V850E2_CSC_ADDR(n) (0xFFFFF060 + (n) * 2) -#define V850E2_CSC(n) (*(volatile u16 *)V850E2_CSC_ADDR(n)) -/* I/O area select control */ -#define V850E2_BPC_ADDR 0xFFFFF064 -#define V850E2_BPC (*(volatile u16 *)V850E2_BPC_ADDR) -/* Bus size configuration */ -#define V850E2_BSC_ADDR 0xFFFFF066 -#define V850E2_BSC (*(volatile u16 *)V850E2_BSC_ADDR) -/* Endian configuration */ -#define V850E2_BEC_ADDR 0xFFFFF068 -#define V850E2_BEC (*(volatile u16 *)V850E2_BEC_ADDR) -/* Cache configuration */ -#define V850E2_BHC_ADDR 0xFFFFF06A -#define V850E2_BHC (*(volatile u16 *)V850E2_BHC_ADDR) -/* NPB strobe-wait configuration */ -#define V850E2_VSWC_ADDR 0xFFFFF06E -#define V850E2_VSWC (*(volatile u16 *)V850E2_VSWC_ADDR) -/* Bus cycle type */ -#define V850E2_BCT_ADDR(n) (0xFFFFF480 + (n) * 2) -#define V850E2_BCT(n) (*(volatile u16 *)V850E2_BCT_ADDR(n)) -/* Data wait control */ -#define V850E2_DWC_ADDR(n) (0xFFFFF484 + (n) * 2) -#define V850E2_DWC(n) (*(volatile u16 *)V850E2_DWC_ADDR(n)) -/* Bus cycle control */ -#define V850E2_BCC_ADDR 0xFFFFF488 -#define V850E2_BCC (*(volatile u16 *)V850E2_BCC_ADDR) -/* Address wait control */ -#define V850E2_ASC_ADDR 0xFFFFF48A -#define V850E2_ASC (*(volatile u16 *)V850E2_ASC_ADDR) -/* Local bus sizing control */ -#define V850E2_LBS_ADDR 0xFFFFF48E -#define V850E2_LBS (*(volatile u16 *)V850E2_LBS_ADDR) -/* Line buffer control */ -#define V850E2_LBC_ADDR(n) (0xFFFFF490 + (n) * 2) -#define V850E2_LBC(n) (*(volatile u16 *)V850E2_LBC_ADDR(n)) -/* SDRAM configuration */ -#define V850E2_SCR_ADDR(n) (0xFFFFF4A0 + (n) * 4) -#define V850E2_SCR(n) (*(volatile u16 *)V850E2_SCR_ADDR(n)) -/* SDRAM refresh cycle control */ -#define V850E2_RFS_ADDR(n) (0xFFFFF4A2 + (n) * 4) -#define V850E2_RFS(n) (*(volatile u16 *)V850E2_RFS_ADDR(n)) - - -#endif /* __V850_V850E2_H__ */ diff --git a/include/asm-v850/v850e2_cache.h b/include/asm-v850/v850e2_cache.h deleted file mode 100644 index 87edf0d311d5..000000000000 --- a/include/asm-v850/v850e2_cache.h +++ /dev/null @@ -1,75 +0,0 @@ -/* - * include/asm-v850/v850e2_cache_cache.h -- Cache control for V850E2 - * cache memories - * - * Copyright (C) 2003,05 NEC Electronics Corporation - * Copyright (C) 2003,05 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_V850E2_CACHE_H__ -#define __V850_V850E2_CACHE_H__ - -#include - - -/* Cache control registers. */ - -/* Bus Transaction Control */ -#define V850E2_CACHE_BTSC_ADDR 0xFFFFF070 -#define V850E2_CACHE_BTSC (*(volatile u16 *)V850E2_CACHE_BTSC_ADDR) -#define V850E2_CACHE_BTSC_ICM 0x0001 /* icache enable */ -#define V850E2_CACHE_BTSC_DCM0 0x0004 /* dcache enable, bit 0 */ -#define V850E2_CACHE_BTSC_DCM1 0x0008 /* dcache enable, bit 1 */ -#define V850E2_CACHE_BTSC_DCM_WT /* write-through */ \ - V850E2_CACHE_BTSC_DCM0 -#ifdef CONFIG_V850E2_V850E2S -# define V850E2_CACHE_BTSC_DCM_WB_NO_ALLOC /* write-back, non-alloc */ \ - V850E2_CACHE_BTSC_DCM1 -# define V850E2_CACHE_BTSC_DCM_WB_ALLOC /* write-back, non-alloc */ \ - (V850E2_CACHE_BTSC_DCM1 | V850E2_CACHE_BTSC_DCM0) -# define V850E2_CACHE_BTSC_ISEQ 0x0010 /* icache `address sequence mode' */ -# define V850E2_CACHE_BTSC_DSEQ 0x0020 /* dcache `address sequence mode' */ -# define V850E2_CACHE_BTSC_IRFC 0x0030 -# define V850E2_CACHE_BTSC_ILCD 0x4000 -# define V850E2_CACHE_BTSC_VABE 0x8000 -#endif /* CONFIG_V850E2_V850E2S */ - -/* Cache operation start address register (low-bits). */ -#define V850E2_CACHE_CADL_ADDR 0xFFFFF074 -#define V850E2_CACHE_CADL (*(volatile u16 *)V850E2_CACHE_CADL_ADDR) -/* Cache operation start address register (high-bits). */ -#define V850E2_CACHE_CADH_ADDR 0xFFFFF076 -#define V850E2_CACHE_CADH (*(volatile u16 *)V850E2_CACHE_CADH_ADDR) -/* Cache operation count register. */ -#define V850E2_CACHE_CCNT_ADDR 0xFFFFF078 -#define V850E2_CACHE_CCNT (*(volatile u16 *)V850E2_CACHE_CCNT_ADDR) -/* Cache operation specification register. */ -#define V850E2_CACHE_COPR_ADDR 0xFFFFF07A -#define V850E2_CACHE_COPR (*(volatile u16 *)V850E2_CACHE_COPR_ADDR) -#define V850E2_CACHE_COPR_STRT 0x0001 /* start cache operation */ -#define V850E2_CACHE_COPR_LBSL 0x0100 /* 0 = icache, 1 = dcache */ -#define V850E2_CACHE_COPR_WSLE 0x0200 /* operate on cache way */ -#define V850E2_CACHE_COPR_WSL(way) ((way) * 0x0400) /* way select */ -#define V850E2_CACHE_COPR_CFC(op) ((op) * 0x1000) /* cache function code */ - - -/* Size of a cache line in bytes. */ -#define V850E2_CACHE_LINE_SIZE_BITS 4 -#define V850E2_CACHE_LINE_SIZE (1 << V850E2_CACHE_LINE_SIZE_BITS) - -/* The size of each cache `way' in lines. */ -#define V850E2_CACHE_WAY_SIZE 256 - - -/* For */ -#define L1_CACHE_BYTES V850E2_CACHE_LINE_SIZE -#define L1_CACHE_SHIFT V850E2_CACHE_LINE_SIZE_BITS - - -#endif /* __V850_V850E2_CACHE_H__ */ diff --git a/include/asm-v850/v850e_cache.h b/include/asm-v850/v850e_cache.h deleted file mode 100644 index aa7d7eb9da50..000000000000 --- a/include/asm-v850/v850e_cache.h +++ /dev/null @@ -1,48 +0,0 @@ -/* - * include/asm-v850/v850e_cache.h -- Cache control for V850E cache memories - * - * Copyright (C) 2001,03 NEC Electronics Corporation - * Copyright (C) 2001,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -/* This file implements cache control for the rather simple cache used on - some V850E CPUs, specifically the NB85E/TEG CPU-core and the V850E/ME2 - CPU. V850E2 processors have their own (better) cache - implementation. */ - -#ifndef __V850_V850E_CACHE_H__ -#define __V850_V850E_CACHE_H__ - -#include - - -/* Cache control registers. */ -#define V850E_CACHE_BHC_ADDR 0xFFFFF06A -#define V850E_CACHE_BHC (*(volatile u16 *)V850E_CACHE_BHC_ADDR) -#define V850E_CACHE_ICC_ADDR 0xFFFFF070 -#define V850E_CACHE_ICC (*(volatile u16 *)V850E_CACHE_ICC_ADDR) -#define V850E_CACHE_ISI_ADDR 0xFFFFF072 -#define V850E_CACHE_ISI (*(volatile u16 *)V850E_CACHE_ISI_ADDR) -#define V850E_CACHE_DCC_ADDR 0xFFFFF078 -#define V850E_CACHE_DCC (*(volatile u16 *)V850E_CACHE_DCC_ADDR) - -/* Size of a cache line in bytes. */ -#define V850E_CACHE_LINE_SIZE 16 - -/* For */ -#define L1_CACHE_BYTES V850E_CACHE_LINE_SIZE - - -#if defined(__KERNEL__) && !defined(__ASSEMBLY__) -/* Set caching params via the BHC, ICC, and DCC registers. */ -void v850e_cache_enable (u16 bhc, u16 icc, u16 dcc); -#endif /* __KERNEL__ && !__ASSEMBLY__ */ - - -#endif /* __V850_V850E_CACHE_H__ */ diff --git a/include/asm-v850/v850e_intc.h b/include/asm-v850/v850e_intc.h deleted file mode 100644 index 6fdf95708317..000000000000 --- a/include/asm-v850/v850e_intc.h +++ /dev/null @@ -1,133 +0,0 @@ -/* - * include/asm-v850/v850e_intc.h -- V850E CPU interrupt controller (INTC) - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_V850E_INTC_H__ -#define __V850_V850E_INTC_H__ - - -/* There are 4 16-bit `Interrupt Mask Registers' located contiguously - starting from this base. Each interrupt uses a single bit to - indicated enabled/disabled status. */ -#define V850E_INTC_IMR_BASE_ADDR 0xFFFFF100 -#define V850E_INTC_IMR_ADDR(irq) (V850E_INTC_IMR_BASE_ADDR + ((irq) >> 3)) -#define V850E_INTC_IMR_BIT(irq) ((irq) & 0x7) - -/* Each maskable interrupt has a single-byte control register at this - address. */ -#define V850E_INTC_IC_BASE_ADDR 0xFFFFF110 -#define V850E_INTC_IC_ADDR(irq) (V850E_INTC_IC_BASE_ADDR + ((irq) << 1)) -#define V850E_INTC_IC(irq) (*(volatile u8 *)V850E_INTC_IC_ADDR(irq)) -/* Encode priority PR for storing in an interrupt control register. */ -#define V850E_INTC_IC_PR(pr) (pr) -/* Interrupt disable bit in an interrupt control register. */ -#define V850E_INTC_IC_MK_BIT 6 -#define V850E_INTC_IC_MK (1 << V850E_INTC_IC_MK_BIT) -/* Interrupt pending flag in an interrupt control register. */ -#define V850E_INTC_IC_IF_BIT 7 -#define V850E_INTC_IC_IF (1 << V850E_INTC_IC_IF_BIT) - -/* The ISPR (In-service priority register) contains one bit for each interrupt - priority level, which is set to one when that level is currently being - serviced (and thus blocking any interrupts of equal or lesser level). */ -#define V850E_INTC_ISPR_ADDR 0xFFFFF1FA -#define V850E_INTC_ISPR (*(volatile u8 *)V850E_INTC_ISPR_ADDR) - - -#ifndef __ASSEMBLY__ - -/* Enable interrupt handling for interrupt IRQ. */ -static inline void v850e_intc_enable_irq (unsigned irq) -{ - __asm__ __volatile__ ("clr1 %0, [%1]" - :: "r" (V850E_INTC_IMR_BIT (irq)), - "r" (V850E_INTC_IMR_ADDR (irq)) - : "memory"); -} - -/* Disable interrupt handling for interrupt IRQ. Note that any - interrupts received while disabled will be delivered once the - interrupt is enabled again, unless they are explicitly cleared using - `v850e_intc_clear_pending_irq'. */ -static inline void v850e_intc_disable_irq (unsigned irq) -{ - __asm__ __volatile__ ("set1 %0, [%1]" - :: "r" (V850E_INTC_IMR_BIT (irq)), - "r" (V850E_INTC_IMR_ADDR (irq)) - : "memory"); -} - -/* Return true if interrupt handling for interrupt IRQ is enabled. */ -static inline int v850e_intc_irq_enabled (unsigned irq) -{ - int rval; - __asm__ __volatile__ ("tst1 %1, [%2]; setf z, %0" - : "=r" (rval) - : "r" (V850E_INTC_IMR_BIT (irq)), - "r" (V850E_INTC_IMR_ADDR (irq))); - return rval; -} - -/* Disable irqs from 0 until LIMIT. LIMIT must be a multiple of 8. */ -static inline void _v850e_intc_disable_irqs (unsigned limit) -{ - unsigned long addr; - for (addr = V850E_INTC_IMR_BASE_ADDR; limit >= 8; addr++, limit -= 8) - *(char *)addr = 0xFF; -} - -/* Disable all irqs. This is purposely a macro, because NUM_MACH_IRQS - will be only be defined later. */ -#define v850e_intc_disable_irqs() _v850e_intc_disable_irqs (NUM_MACH_IRQS) - -/* Clear any pending interrupts for IRQ. */ -static inline void v850e_intc_clear_pending_irq (unsigned irq) -{ - __asm__ __volatile__ ("clr1 %0, 0[%1]" - :: "i" (V850E_INTC_IC_IF_BIT), - "r" (V850E_INTC_IC_ADDR (irq)) - : "memory"); -} - -/* Return true if interrupt IRQ is pending (but disabled). */ -static inline int v850e_intc_irq_pending (unsigned irq) -{ - int rval; - __asm__ __volatile__ ("tst1 %1, 0[%2]; setf nz, %0" - : "=r" (rval) - : "i" (V850E_INTC_IC_IF_BIT), - "r" (V850E_INTC_IC_ADDR (irq))); - return rval; -} - - -struct v850e_intc_irq_init { - const char *name; /* name of interrupt type */ - - /* Range of kernel irq numbers for this type: - BASE, BASE+INTERVAL, ..., BASE+INTERVAL*NUM */ - unsigned base, num, interval; - - unsigned priority; /* interrupt priority to assign */ -}; -struct hw_interrupt_type; /* fwd decl */ - -/* Initialize HW_IRQ_TYPES for INTC-controlled irqs described in array - INITS (which is terminated by an entry with the name field == 0). */ -extern void v850e_intc_init_irq_types (struct v850e_intc_irq_init *inits, - struct hw_interrupt_type *hw_irq_types); - - -#endif /* !__ASSEMBLY__ */ - - -#endif /* __V850_V850E_INTC_H__ */ diff --git a/include/asm-v850/v850e_timer_c.h b/include/asm-v850/v850e_timer_c.h deleted file mode 100644 index f70575df6ea9..000000000000 --- a/include/asm-v850/v850e_timer_c.h +++ /dev/null @@ -1,48 +0,0 @@ -/* - * include/asm-v850/v850e_timer_c.h -- `Timer C' component often used - * with the V850E cpu core - * - * Copyright (C) 2001,03 NEC Electronics Corporation - * Copyright (C) 2001,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -/* NOTE: this include file currently contains only enough to allow us to - use timer C as an interrupt pass-through. */ - -#ifndef __V850_V850E_TIMER_C_H__ -#define __V850_V850E_TIMER_C_H__ - -#include -#include /* Pick up chip-specific defs. */ - - -/* Timer C (16-bit interval timers). */ - -/* Control register 0 for timer C. */ -#define V850E_TIMER_C_TMCC0_ADDR(n) (V850E_TIMER_C_BASE_ADDR + 0x6 + 0x10 *(n)) -#define V850E_TIMER_C_TMCC0(n) (*(volatile u8 *)V850E_TIMER_C_TMCC0_ADDR(n)) -#define V850E_TIMER_C_TMCC0_CAE 0x01 /* clock action enable */ -#define V850E_TIMER_C_TMCC0_CE 0x02 /* count enable */ -/* ... */ - -/* Control register 1 for timer C. */ -#define V850E_TIMER_C_TMCC1_ADDR(n) (V850E_TIMER_C_BASE_ADDR + 0x8 + 0x10 *(n)) -#define V850E_TIMER_C_TMCC1(n) (*(volatile u8 *)V850E_TIMER_C_TMCC1_ADDR(n)) -#define V850E_TIMER_C_TMCC1_CMS0 0x01 /* capture/compare mode select (ccc0) */ -#define V850E_TIMER_C_TMCC1_CMS1 0x02 /* capture/compare mode select (ccc1) */ -/* ... */ - -/* Interrupt edge-sensitivity control for timer C. */ -#define V850E_TIMER_C_SESC_ADDR(n) (V850E_TIMER_C_BASE_ADDR + 0x9 + 0x10 *(n)) -#define V850E_TIMER_C_SESC(n) (*(volatile u8 *)V850E_TIMER_C_SESC_ADDR(n)) - -/* ...etc... */ - - -#endif /* __V850_V850E_TIMER_C_H__ */ diff --git a/include/asm-v850/v850e_timer_d.h b/include/asm-v850/v850e_timer_d.h deleted file mode 100644 index 417612c5b22f..000000000000 --- a/include/asm-v850/v850e_timer_d.h +++ /dev/null @@ -1,62 +0,0 @@ -/* - * include/asm-v850/v850e_timer_d.h -- `Timer D' component often used - * with the V850E cpu core - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_V850E_TIMER_D_H__ -#define __V850_V850E_TIMER_D_H__ - -#include -#include /* Pick up chip-specific defs. */ - - -/* Timer D (16-bit interval timers). */ - -/* Count registers for timer D. */ -#define V850E_TIMER_D_TMD_ADDR(n) (V850E_TIMER_D_TMD_BASE_ADDR + 0x10 * (n)) -#define V850E_TIMER_D_TMD(n) (*(volatile u16 *)V850E_TIMER_D_TMD_ADDR(n)) - -/* Count compare registers for timer D. */ -#define V850E_TIMER_D_CMD_ADDR(n) (V850E_TIMER_D_CMD_BASE_ADDR + 0x10 * (n)) -#define V850E_TIMER_D_CMD(n) (*(volatile u16 *)V850E_TIMER_D_CMD_ADDR(n)) - -/* Control registers for timer D. */ -#define V850E_TIMER_D_TMCD_ADDR(n) (V850E_TIMER_D_TMCD_BASE_ADDR + 0x10 * (n)) -#define V850E_TIMER_D_TMCD(n) (*(volatile u8 *)V850E_TIMER_D_TMCD_ADDR(n)) -/* Control bits for timer D. */ -#define V850E_TIMER_D_TMCD_CE 0x2 /* count enable */ -#define V850E_TIMER_D_TMCD_CAE 0x1 /* clock action enable */ -/* Clock divider setting (log2). */ -#define V850E_TIMER_D_TMCD_CS(divlog2) (((divlog2) - V850E_TIMER_D_TMCD_CS_MIN) << 4) -/* Minimum clock divider setting (log2). */ -#ifndef V850E_TIMER_D_TMCD_CS_MIN /* Can be overridden by mach-specific hdrs */ -#define V850E_TIMER_D_TMCD_CS_MIN 2 /* Default is correct for the v850e/ma1 */ -#endif -/* Maximum clock divider setting (log2). */ -#define V850E_TIMER_D_TMCD_CS_MAX (V850E_TIMER_D_TMCD_CS_MIN + 7) - -/* Return the clock-divider (log2) of timer D unit N. */ -#define V850E_TIMER_D_DIVLOG2(n) \ - (((V850E_TIMER_D_TMCD(n) >> 4) & 0x7) + V850E_TIMER_D_TMCD_CS_MIN) - - -#ifndef __ASSEMBLY__ - -/* Start interval timer TIMER (0-3). The timer will issue the - corresponding INTCMD interrupt RATE times per second. This function - does not enable the interrupt. */ -extern void v850e_timer_d_configure (unsigned timer, unsigned rate); - -#endif /* !__ASSEMBLY__ */ - - -#endif /* __V850_V850E_TIMER_D_H__ */ diff --git a/include/asm-v850/v850e_uart.h b/include/asm-v850/v850e_uart.h deleted file mode 100644 index 5182fb4cc989..000000000000 --- a/include/asm-v850/v850e_uart.h +++ /dev/null @@ -1,76 +0,0 @@ -/* - * include/asm-v850/v850e_uart.h -- common V850E on-chip UART driver - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -/* There's not actually a single UART implementation used by V850E CPUs, - but rather a series of implementations that are all `close' to one - another. This file corresponds to the single driver which handles all - of them. */ - -#ifndef __V850_V850E_UART_H__ -#define __V850_V850E_UART_H__ - -#include - -#include -#include -#include /* Pick up chip-specific defs. */ - - -/* Include model-specific definitions. */ -#ifdef CONFIG_V850E_UART -# ifdef CONFIG_V850E_UARTB -# include -# else -# include /* original V850E UART */ -# endif -#endif - - -/* Optional capabilities some hardware provides. */ - -/* This UART doesn't implement RTS/CTS by default, but some platforms - implement them externally, so check to see if defined - anything. */ -#ifdef V850E_UART_CTS -#define v850e_uart_cts(n) V850E_UART_CTS(n) -#else -#define v850e_uart_cts(n) (1) -#endif - -/* Do the same for RTS. */ -#ifdef V850E_UART_SET_RTS -#define v850e_uart_set_rts(n,v) V850E_UART_SET_RTS(n,v) -#else -#define v850e_uart_set_rts(n,v) ((void)0) -#endif - - -/* This is the serial channel to use for the boot console (if desired). */ -#ifndef V850E_UART_CONSOLE_CHANNEL -# define V850E_UART_CONSOLE_CHANNEL 0 -#endif - - -#ifndef __ASSEMBLY__ - -/* Setup a console using channel 0 of the builtin uart. */ -extern void v850e_uart_cons_init (unsigned chan); - -/* Configure and turn on uart channel CHAN, using the termios `control - modes' bits in CFLAGS, and a baud-rate of BAUD. */ -void v850e_uart_configure (unsigned chan, unsigned cflags, unsigned baud); - -#endif /* !__ASSEMBLY__ */ - - -#endif /* __V850_V850E_UART_H__ */ diff --git a/include/asm-v850/v850e_uarta.h b/include/asm-v850/v850e_uarta.h deleted file mode 100644 index e483e0950725..000000000000 --- a/include/asm-v850/v850e_uarta.h +++ /dev/null @@ -1,278 +0,0 @@ -/* - * include/asm-v850/v850e_uarta.h -- original V850E on-chip UART - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -/* This is the original V850E UART implementation is called just `UART' in - the docs, but we name this header file because the - name is used for the common driver that handles both - `UART' and `UARTB' implementations. */ - -#ifndef __V850_V850E_UARTA_H__ -#define __V850_V850E_UARTA_H__ - - -/* Raw hardware interface. */ - -/* The base address of the UART control registers for channel N. - The default is the address used on the V850E/MA1. */ -#ifndef V850E_UART_BASE_ADDR -#define V850E_UART_BASE_ADDR(n) (0xFFFFFA00 + 0x10 * (n)) -#endif - -/* Addresses of specific UART control registers for channel N. - The defaults are the addresses used on the V850E/MA1; if a platform - wants to redefine any of these, it must redefine them all. */ -#ifndef V850E_UART_ASIM_ADDR -#define V850E_UART_ASIM_ADDR(n) (V850E_UART_BASE_ADDR(n) + 0x0) -#define V850E_UART_RXB_ADDR(n) (V850E_UART_BASE_ADDR(n) + 0x2) -#define V850E_UART_ASIS_ADDR(n) (V850E_UART_BASE_ADDR(n) + 0x3) -#define V850E_UART_TXB_ADDR(n) (V850E_UART_BASE_ADDR(n) + 0x4) -#define V850E_UART_ASIF_ADDR(n) (V850E_UART_BASE_ADDR(n) + 0x5) -#define V850E_UART_CKSR_ADDR(n) (V850E_UART_BASE_ADDR(n) + 0x6) -#define V850E_UART_BRGC_ADDR(n) (V850E_UART_BASE_ADDR(n) + 0x7) -#endif - -/* UART config registers. */ -#define V850E_UART_ASIM(n) (*(volatile u8 *)V850E_UART_ASIM_ADDR(n)) -/* Control bits for config registers. */ -#define V850E_UART_ASIM_CAE 0x80 /* clock enable */ -#define V850E_UART_ASIM_TXE 0x40 /* transmit enable */ -#define V850E_UART_ASIM_RXE 0x20 /* receive enable */ -#define V850E_UART_ASIM_PS_MASK 0x18 /* mask covering parity-select bits */ -#define V850E_UART_ASIM_PS_NONE 0x00 /* no parity */ -#define V850E_UART_ASIM_PS_ZERO 0x08 /* zero parity */ -#define V850E_UART_ASIM_PS_ODD 0x10 /* odd parity */ -#define V850E_UART_ASIM_PS_EVEN 0x18 /* even parity */ -#define V850E_UART_ASIM_CL_8 0x04 /* char len is 8 bits (otherwise, 7) */ -#define V850E_UART_ASIM_SL_2 0x02 /* 2 stop bits (otherwise, 1) */ -#define V850E_UART_ASIM_ISRM 0x01 /* generate INTSR interrupt on errors - (otherwise, generate INTSER) */ - -/* UART serial interface status registers. */ -#define V850E_UART_ASIS(n) (*(volatile u8 *)V850E_UART_ASIS_ADDR(n)) -/* Control bits for status registers. */ -#define V850E_UART_ASIS_PE 0x04 /* parity error */ -#define V850E_UART_ASIS_FE 0x02 /* framing error */ -#define V850E_UART_ASIS_OVE 0x01 /* overrun error */ - -/* UART serial interface transmission status registers. */ -#define V850E_UART_ASIF(n) (*(volatile u8 *)V850E_UART_ASIF_ADDR(n)) -#define V850E_UART_ASIF_TXBF 0x02 /* transmit buffer flag (data in TXB) */ -#define V850E_UART_ASIF_TXSF 0x01 /* transmit shift flag (sending data) */ - -/* UART receive buffer register. */ -#define V850E_UART_RXB(n) (*(volatile u8 *)V850E_UART_RXB_ADDR(n)) - -/* UART transmit buffer register. */ -#define V850E_UART_TXB(n) (*(volatile u8 *)V850E_UART_TXB_ADDR(n)) - -/* UART baud-rate generator control registers. */ -#define V850E_UART_CKSR(n) (*(volatile u8 *)V850E_UART_CKSR_ADDR(n)) -#define V850E_UART_CKSR_MAX 11 -#define V850E_UART_BRGC(n) (*(volatile u8 *)V850E_UART_BRGC_ADDR(n)) -#define V850E_UART_BRGC_MIN 8 - - -#ifndef V850E_UART_CKSR_MAX_FREQ -#define V850E_UART_CKSR_MAX_FREQ (25*1000*1000) -#endif - -/* Calculate the minimum value for CKSR on this processor. */ -static inline unsigned v850e_uart_cksr_min (void) -{ - int min = 0; - unsigned freq = V850E_UART_BASE_FREQ; - while (freq > V850E_UART_CKSR_MAX_FREQ) { - freq >>= 1; - min++; - } - return min; -} - - -/* Slightly abstract interface used by driver. */ - - -/* Interrupts used by the UART. */ - -/* Received when the most recently transmitted character has been sent. */ -#define V850E_UART_TX_IRQ(chan) IRQ_INTST (chan) -/* Received when a new character has been received. */ -#define V850E_UART_RX_IRQ(chan) IRQ_INTSR (chan) - - -/* UART clock generator interface. */ - -/* This type encapsulates a particular uart frequency. */ -typedef struct { - unsigned clk_divlog2; - unsigned brgen_count; -} v850e_uart_speed_t; - -/* Calculate a uart speed from BAUD for this uart. */ -static inline v850e_uart_speed_t v850e_uart_calc_speed (unsigned baud) -{ - v850e_uart_speed_t speed; - - /* Calculate the log2 clock divider and baud-rate counter values - (note that the UART divides the resulting clock by 2, so - multiply BAUD by 2 here to compensate). */ - calc_counter_params (V850E_UART_BASE_FREQ, baud * 2, - v850e_uart_cksr_min(), - V850E_UART_CKSR_MAX, 8/*bits*/, - &speed.clk_divlog2, &speed.brgen_count); - - return speed; -} - -/* Return the current speed of uart channel CHAN. */ -static inline v850e_uart_speed_t v850e_uart_speed (unsigned chan) -{ - v850e_uart_speed_t speed; - speed.clk_divlog2 = V850E_UART_CKSR (chan); - speed.brgen_count = V850E_UART_BRGC (chan); - return speed; -} - -/* Set the current speed of uart channel CHAN. */ -static inline void v850e_uart_set_speed(unsigned chan,v850e_uart_speed_t speed) -{ - V850E_UART_CKSR (chan) = speed.clk_divlog2; - V850E_UART_BRGC (chan) = speed.brgen_count; -} - -static inline int -v850e_uart_speed_eq (v850e_uart_speed_t speed1, v850e_uart_speed_t speed2) -{ - return speed1.clk_divlog2 == speed2.clk_divlog2 - && speed1.brgen_count == speed2.brgen_count; -} - -/* Minimum baud rate possible. */ -#define v850e_uart_min_baud() \ - ((V850E_UART_BASE_FREQ >> V850E_UART_CKSR_MAX) / (2 * 255) + 1) - -/* Maximum baud rate possible. The error is quite high at max, though. */ -#define v850e_uart_max_baud() \ - ((V850E_UART_BASE_FREQ >> v850e_uart_cksr_min()) / (2 *V850E_UART_BRGC_MIN)) - -/* The `maximum' clock rate the uart can used, which is wanted (though not - really used in any useful way) by the serial framework. */ -#define v850e_uart_max_clock() \ - ((V850E_UART_BASE_FREQ >> v850e_uart_cksr_min()) / 2) - - -/* UART configuration interface. */ - -/* Type of the uart config register; must be a scalar. */ -typedef u16 v850e_uart_config_t; - -/* The uart hardware config register for channel CHAN. */ -#define V850E_UART_CONFIG(chan) V850E_UART_ASIM (chan) - -/* This config bit set if the uart is enabled. */ -#define V850E_UART_CONFIG_ENABLED V850E_UART_ASIM_CAE -/* If the uart _isn't_ enabled, store this value to it to do so. */ -#define V850E_UART_CONFIG_INIT V850E_UART_ASIM_CAE -/* Store this config value to disable the uart channel completely. */ -#define V850E_UART_CONFIG_FINI 0 - -/* Setting/clearing these bits enable/disable TX/RX, respectively (but - otherwise generally leave things running). */ -#define V850E_UART_CONFIG_RX_ENABLE V850E_UART_ASIM_RXE -#define V850E_UART_CONFIG_TX_ENABLE V850E_UART_ASIM_TXE - -/* These masks define which config bits affect TX/RX modes, respectively. */ -#define V850E_UART_CONFIG_RX_BITS \ - (V850E_UART_ASIM_PS_MASK | V850E_UART_ASIM_CL_8 | V850E_UART_ASIM_ISRM) -#define V850E_UART_CONFIG_TX_BITS \ - (V850E_UART_ASIM_PS_MASK | V850E_UART_ASIM_CL_8 | V850E_UART_ASIM_SL_2) - -static inline v850e_uart_config_t v850e_uart_calc_config (unsigned cflags) -{ - v850e_uart_config_t config = 0; - - /* Figure out new configuration of control register. */ - if (cflags & CSTOPB) - /* Number of stop bits, 1 or 2. */ - config |= V850E_UART_ASIM_SL_2; - if ((cflags & CSIZE) == CS8) - /* Number of data bits, 7 or 8. */ - config |= V850E_UART_ASIM_CL_8; - if (! (cflags & PARENB)) - /* No parity check/generation. */ - config |= V850E_UART_ASIM_PS_NONE; - else if (cflags & PARODD) - /* Odd parity check/generation. */ - config |= V850E_UART_ASIM_PS_ODD; - else - /* Even parity check/generation. */ - config |= V850E_UART_ASIM_PS_EVEN; - if (cflags & CREAD) - /* Reading enabled. */ - config |= V850E_UART_ASIM_RXE; - - config |= V850E_UART_ASIM_CAE; - config |= V850E_UART_ASIM_TXE; /* Writing is always enabled. */ - config |= V850E_UART_ASIM_ISRM; /* Errors generate a read-irq. */ - - return config; -} - -/* This should delay as long as necessary for a recently written config - setting to settle, before we turn the uart back on. */ -static inline void -v850e_uart_config_delay (v850e_uart_config_t config, v850e_uart_speed_t speed) -{ - /* The UART may not be reset properly unless we wait at least 2 - `basic-clocks' until turning on the TXE/RXE bits again. - A `basic clock' is the clock used by the baud-rate generator, - i.e., the cpu clock divided by the 2^new_clk_divlog2. - The loop takes 2 insns, so loop CYCLES / 2 times. */ - register unsigned count = 1 << speed.clk_divlog2; - while (--count != 0) - /* nothing */; -} - - -/* RX/TX interface. */ - -/* Return true if all characters awaiting transmission on uart channel N - have been transmitted. */ -#define v850e_uart_xmit_done(n) \ - (! (V850E_UART_ASIF(n) & V850E_UART_ASIF_TXBF)) -/* Wait for this to be true. */ -#define v850e_uart_wait_for_xmit_done(n) \ - do { } while (! v850e_uart_xmit_done (n)) - -/* Return true if uart channel N is ready to transmit a character. */ -#define v850e_uart_xmit_ok(n) \ - (v850e_uart_xmit_done(n) && v850e_uart_cts(n)) -/* Wait for this to be true. */ -#define v850e_uart_wait_for_xmit_ok(n) \ - do { } while (! v850e_uart_xmit_ok (n)) - -/* Write character CH to uart channel CHAN. */ -#define v850e_uart_putc(chan, ch) (V850E_UART_TXB(chan) = (ch)) - -/* Return latest character read on channel CHAN. */ -#define v850e_uart_getc(chan) V850E_UART_RXB (chan) - -/* Return bit-mask of uart error status. */ -#define v850e_uart_err(chan) V850E_UART_ASIS (chan) -/* Various error bits set in the error result. */ -#define V850E_UART_ERR_OVERRUN V850E_UART_ASIS_OVE -#define V850E_UART_ERR_FRAME V850E_UART_ASIS_FE -#define V850E_UART_ERR_PARITY V850E_UART_ASIS_PE - - -#endif /* __V850_V850E_UARTA_H__ */ diff --git a/include/asm-v850/v850e_uartb.h b/include/asm-v850/v850e_uartb.h deleted file mode 100644 index 6d4767d5a835..000000000000 --- a/include/asm-v850/v850e_uartb.h +++ /dev/null @@ -1,262 +0,0 @@ -/* - * include/asm-v850/v850e_uartb.h -- V850E on-chip `UARTB' UART - * - * Copyright (C) 2001,02,03 NEC Electronics Corporation - * Copyright (C) 2001,02,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -/* The V850E UARTB is basically a superset of the original V850E UART, but - even where it's the same, the names and details have changed a bit. - It's similar enough to use the same driver (v850e_uart.c), but the - details have been abstracted slightly to do so. */ - -#ifndef __V850_V850E_UARTB_H__ -#define __V850_V850E_UARTB_H__ - - -/* Raw hardware interface. */ - -#define V850E_UARTB_BASE_ADDR(n) (0xFFFFFA00 + 0x10 * (n)) - -/* Addresses of specific UART control registers for channel N. */ -#define V850E_UARTB_CTL0_ADDR(n) (V850E_UARTB_BASE_ADDR(n) + 0x0) -#define V850E_UARTB_CTL2_ADDR(n) (V850E_UARTB_BASE_ADDR(n) + 0x2) -#define V850E_UARTB_STR_ADDR(n) (V850E_UARTB_BASE_ADDR(n) + 0x4) -#define V850E_UARTB_RX_ADDR(n) (V850E_UARTB_BASE_ADDR(n) + 0x6) -#define V850E_UARTB_RXAP_ADDR(n) (V850E_UARTB_BASE_ADDR(n) + 0x6) -#define V850E_UARTB_TX_ADDR(n) (V850E_UARTB_BASE_ADDR(n) + 0x8) -#define V850E_UARTB_FIC0_ADDR(n) (V850E_UARTB_BASE_ADDR(n) + 0xA) -#define V850E_UARTB_FIC1_ADDR(n) (V850E_UARTB_BASE_ADDR(n) + 0xB) -#define V850E_UARTB_FIC2_ADDR(n) (V850E_UARTB_BASE_ADDR(n) + 0xC) -#define V850E_UARTB_FIS0_ADDR(n) (V850E_UARTB_BASE_ADDR(n) + 0xE) -#define V850E_UARTB_FIS1_ADDR(n) (V850E_UARTB_BASE_ADDR(n) + 0xF) - -/* UARTB control register 0 (general config). */ -#define V850E_UARTB_CTL0(n) (*(volatile u8 *)V850E_UARTB_CTL0_ADDR(n)) -/* Control bits for config registers. */ -#define V850E_UARTB_CTL0_PWR 0x80 /* clock enable */ -#define V850E_UARTB_CTL0_TXE 0x40 /* transmit enable */ -#define V850E_UARTB_CTL0_RXE 0x20 /* receive enable */ -#define V850E_UARTB_CTL0_DIR 0x10 /* */ -#define V850E_UARTB_CTL0_PS1 0x08 /* parity */ -#define V850E_UARTB_CTL0_PS0 0x04 /* parity */ -#define V850E_UARTB_CTL0_CL 0x02 /* char len 1:8bit, 0:7bit */ -#define V850E_UARTB_CTL0_SL 0x01 /* stop bit 1:2bit, 0:1bit */ -#define V850E_UARTB_CTL0_PS_MASK 0x0C /* mask covering parity bits */ -#define V850E_UARTB_CTL0_PS_NONE 0x00 /* no parity */ -#define V850E_UARTB_CTL0_PS_ZERO 0x04 /* zero parity */ -#define V850E_UARTB_CTL0_PS_ODD 0x08 /* odd parity */ -#define V850E_UARTB_CTL0_PS_EVEN 0x0C /* even parity */ -#define V850E_UARTB_CTL0_CL_8 0x02 /* char len 1:8bit, 0:7bit */ -#define V850E_UARTB_CTL0_SL_2 0x01 /* stop bit 1:2bit, 0:1bit */ - -/* UARTB control register 2 (clock divider). */ -#define V850E_UARTB_CTL2(n) (*(volatile u16 *)V850E_UARTB_CTL2_ADDR(n)) -#define V850E_UARTB_CTL2_MIN 4 -#define V850E_UARTB_CTL2_MAX 0xFFFF - -/* UARTB serial interface status register. */ -#define V850E_UARTB_STR(n) (*(volatile u8 *)V850E_UARTB_STR_ADDR(n)) -/* Control bits for status registers. */ -#define V850E_UARTB_STR_TSF 0x80 /* UBTX or FIFO exist data */ -#define V850E_UARTB_STR_OVF 0x08 /* overflow error */ -#define V850E_UARTB_STR_PE 0x04 /* parity error */ -#define V850E_UARTB_STR_FE 0x02 /* framing error */ -#define V850E_UARTB_STR_OVE 0x01 /* overrun error */ - -/* UARTB receive data register. */ -#define V850E_UARTB_RX(n) (*(volatile u8 *)V850E_UARTB_RX_ADDR(n)) -#define V850E_UARTB_RXAP(n) (*(volatile u16 *)V850E_UARTB_RXAP_ADDR(n)) -/* Control bits for status registers. */ -#define V850E_UARTB_RXAP_PEF 0x0200 /* parity error */ -#define V850E_UARTB_RXAP_FEF 0x0100 /* framing error */ - -/* UARTB transmit data register. */ -#define V850E_UARTB_TX(n) (*(volatile u8 *)V850E_UARTB_TX_ADDR(n)) - -/* UARTB FIFO control register 0. */ -#define V850E_UARTB_FIC0(n) (*(volatile u8 *)V850E_UARTB_FIC0_ADDR(n)) - -/* UARTB FIFO control register 1. */ -#define V850E_UARTB_FIC1(n) (*(volatile u8 *)V850E_UARTB_FIC1_ADDR(n)) - -/* UARTB FIFO control register 2. */ -#define V850E_UARTB_FIC2(n) (*(volatile u16 *)V850E_UARTB_FIC2_ADDR(n)) - -/* UARTB FIFO status register 0. */ -#define V850E_UARTB_FIS0(n) (*(volatile u8 *)V850E_UARTB_FIS0_ADDR(n)) - -/* UARTB FIFO status register 1. */ -#define V850E_UARTB_FIS1(n) (*(volatile u8 *)V850E_UARTB_FIS1_ADDR(n)) - - -/* Slightly abstract interface used by driver. */ - - -/* Interrupts used by the UART. */ - -/* Received when the most recently transmitted character has been sent. */ -#define V850E_UART_TX_IRQ(chan) IRQ_INTUBTIT (chan) -/* Received when a new character has been received. */ -#define V850E_UART_RX_IRQ(chan) IRQ_INTUBTIR (chan) - -/* Use by serial driver for information purposes. */ -#define V850E_UART_BASE_ADDR(chan) V850E_UARTB_BASE_ADDR(chan) - - -/* UART clock generator interface. */ - -/* This type encapsulates a particular uart frequency. */ -typedef u16 v850e_uart_speed_t; - -/* Calculate a uart speed from BAUD for this uart. */ -static inline v850e_uart_speed_t v850e_uart_calc_speed (unsigned baud) -{ - v850e_uart_speed_t speed; - - /* - * V850E/ME2 UARTB baud rate is determined by the value of UBCTL2 - * fx = V850E_UARTB_BASE_FREQ = CPU_CLOCK_FREQ/4 - * baud = fx / 2*speed [ speed >= 4 ] - */ - speed = V850E_UARTB_CTL2_MIN; - while (((V850E_UARTB_BASE_FREQ / 2) / speed ) > baud) - speed++; - - return speed; -} - -/* Return the current speed of uart channel CHAN. */ -#define v850e_uart_speed(chan) V850E_UARTB_CTL2 (chan) - -/* Set the current speed of uart channel CHAN. */ -#define v850e_uart_set_speed(chan, speed) (V850E_UARTB_CTL2 (chan) = (speed)) - -/* Return true if SPEED1 and SPEED2 are the same. */ -#define v850e_uart_speed_eq(speed1, speed2) ((speed1) == (speed2)) - -/* Minimum baud rate possible. */ -#define v850e_uart_min_baud() \ - ((V850E_UARTB_BASE_FREQ / 2) / V850E_UARTB_CTL2_MAX) - -/* Maximum baud rate possible. The error is quite high at max, though. */ -#define v850e_uart_max_baud() \ - ((V850E_UARTB_BASE_FREQ / 2) / V850E_UARTB_CTL2_MIN) - -/* The `maximum' clock rate the uart can used, which is wanted (though not - really used in any useful way) by the serial framework. */ -#define v850e_uart_max_clock() \ - (V850E_UARTB_BASE_FREQ / 2) - - -/* UART configuration interface. */ - -/* Type of the uart config register; must be a scalar. */ -typedef u16 v850e_uart_config_t; - -/* The uart hardware config register for channel CHAN. */ -#define V850E_UART_CONFIG(chan) V850E_UARTB_CTL0 (chan) - -/* This config bit set if the uart is enabled. */ -#define V850E_UART_CONFIG_ENABLED V850E_UARTB_CTL0_PWR -/* If the uart _isn't_ enabled, store this value to it to do so. */ -#define V850E_UART_CONFIG_INIT V850E_UARTB_CTL0_PWR -/* Store this config value to disable the uart channel completely. */ -#define V850E_UART_CONFIG_FINI 0 - -/* Setting/clearing these bits enable/disable TX/RX, respectively (but - otherwise generally leave things running). */ -#define V850E_UART_CONFIG_RX_ENABLE V850E_UARTB_CTL0_RXE -#define V850E_UART_CONFIG_TX_ENABLE V850E_UARTB_CTL0_TXE - -/* These masks define which config bits affect TX/RX modes, respectively. */ -#define V850E_UART_CONFIG_RX_BITS \ - (V850E_UARTB_CTL0_PS_MASK | V850E_UARTB_CTL0_CL_8) -#define V850E_UART_CONFIG_TX_BITS \ - (V850E_UARTB_CTL0_PS_MASK | V850E_UARTB_CTL0_CL_8 | V850E_UARTB_CTL0_SL_2) - -static inline v850e_uart_config_t v850e_uart_calc_config (unsigned cflags) -{ - v850e_uart_config_t config = 0; - - /* Figure out new configuration of control register. */ - if (cflags & CSTOPB) - /* Number of stop bits, 1 or 2. */ - config |= V850E_UARTB_CTL0_SL_2; - if ((cflags & CSIZE) == CS8) - /* Number of data bits, 7 or 8. */ - config |= V850E_UARTB_CTL0_CL_8; - if (! (cflags & PARENB)) - /* No parity check/generation. */ - config |= V850E_UARTB_CTL0_PS_NONE; - else if (cflags & PARODD) - /* Odd parity check/generation. */ - config |= V850E_UARTB_CTL0_PS_ODD; - else - /* Even parity check/generation. */ - config |= V850E_UARTB_CTL0_PS_EVEN; - if (cflags & CREAD) - /* Reading enabled. */ - config |= V850E_UARTB_CTL0_RXE; - - config |= V850E_UARTB_CTL0_PWR; - config |= V850E_UARTB_CTL0_TXE; /* Writing is always enabled. */ - config |= V850E_UARTB_CTL0_DIR; /* LSB first. */ - - return config; -} - -/* This should delay as long as necessary for a recently written config - setting to settle, before we turn the uart back on. */ -static inline void -v850e_uart_config_delay (v850e_uart_config_t config, v850e_uart_speed_t speed) -{ - /* The UART may not be reset properly unless we wait at least 2 - `basic-clocks' until turning on the TXE/RXE bits again. - A `basic clock' is the clock used by the baud-rate generator, - i.e., the cpu clock divided by the 2^new_clk_divlog2. - The loop takes 2 insns, so loop CYCLES / 2 times. */ - register unsigned count = 1 << speed; - while (--count != 0) - /* nothing */; -} - - -/* RX/TX interface. */ - -/* Return true if all characters awaiting transmission on uart channel N - have been transmitted. */ -#define v850e_uart_xmit_done(n) \ - (! (V850E_UARTB_STR(n) & V850E_UARTB_STR_TSF)) -/* Wait for this to be true. */ -#define v850e_uart_wait_for_xmit_done(n) \ - do { } while (! v850e_uart_xmit_done (n)) - -/* Return true if uart channel N is ready to transmit a character. */ -#define v850e_uart_xmit_ok(n) \ - (v850e_uart_xmit_done(n) && v850e_uart_cts(n)) -/* Wait for this to be true. */ -#define v850e_uart_wait_for_xmit_ok(n) \ - do { } while (! v850e_uart_xmit_ok (n)) - -/* Write character CH to uart channel CHAN. */ -#define v850e_uart_putc(chan, ch) (V850E_UARTB_TX(chan) = (ch)) - -/* Return latest character read on channel CHAN. */ -#define v850e_uart_getc(chan) V850E_UARTB_RX (chan) - -/* Return bit-mask of uart error status. */ -#define v850e_uart_err(chan) V850E_UARTB_STR (chan) -/* Various error bits set in the error result. */ -#define V850E_UART_ERR_OVERRUN V850E_UARTB_STR_OVE -#define V850E_UART_ERR_FRAME V850E_UARTB_STR_FE -#define V850E_UART_ERR_PARITY V850E_UARTB_STR_PE - - -#endif /* __V850_V850E_UARTB_H__ */ diff --git a/include/asm-v850/v850e_utils.h b/include/asm-v850/v850e_utils.h deleted file mode 100644 index 52eb72822d3d..000000000000 --- a/include/asm-v850/v850e_utils.h +++ /dev/null @@ -1,35 +0,0 @@ -/* - * include/asm-v850/v850e_utils.h -- Utility functions associated with - * V850E CPUs - * - * Copyright (C) 2001,03 NEC Electronics Corporation - * Copyright (C) 2001,03 Miles Bader - * - * This file is subject to the terms and conditions of the GNU General - * Public License. See the file COPYING in the main directory of this - * archive for more details. - * - * Written by Miles Bader - */ - -#ifndef __V850_V850E_UTILS_H__ -#define __V850_V850E_UTILS_H__ - -/* Calculate counter clock-divider and count values to attain the - desired frequency RATE from the base frequency BASE_FREQ. The - counter is expected to have a clock-divider, which can divide the - system cpu clock by a power of two value from MIN_DIVLOG2 to - MAX_DIV_LOG2, and a word-size of COUNTER_SIZE bits (the counter - counts up and resets whenever it's equal to the compare register, - generating an interrupt or whatever when it does so). The returned - values are: *DIVLOG2 -- log2 of the desired clock divider and *COUNT - -- the counter compare value to use. Returns true if it was possible - to find a reasonable value, otherwise false (and the other return - values will be set to be as good as possible). */ -extern int calc_counter_params (unsigned long base_freq, - unsigned long rate, - unsigned min_divlog2, unsigned max_divlog2, - unsigned counter_size, - unsigned *divlog2, unsigned *count); - -#endif /* __V850_V850E_UTILS_H__ */ diff --git a/include/linux/audit.h b/include/linux/audit.h index 8b82974bdc12..6272a395d43c 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -286,7 +286,6 @@ #define AUDIT_ARCH_SHEL64 (EM_SH|__AUDIT_ARCH_64BIT|__AUDIT_ARCH_LE) #define AUDIT_ARCH_SPARC (EM_SPARC) #define AUDIT_ARCH_SPARC64 (EM_SPARCV9|__AUDIT_ARCH_64BIT) -#define AUDIT_ARCH_V850 (EM_V850|__AUDIT_ARCH_LE) #define AUDIT_ARCH_X86_64 (EM_X86_64|__AUDIT_ARCH_64BIT|__AUDIT_ARCH_LE) #define AUDIT_PERM_EXEC 1 diff --git a/include/linux/module.h b/include/linux/module.h index fce15ebd0e1c..68e09557c951 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -23,7 +23,7 @@ /* Not Yet Implemented */ #define MODULE_SUPPORTED_DEVICE(name) -/* v850 toolchain uses a `_' prefix for all user symbols */ +/* some toolchains uses a `_' prefix for all user symbols */ #ifndef MODULE_SYMBOL_PREFIX #define MODULE_SYMBOL_PREFIX "" #endif diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h index f3a1c0e45021..3b2f6c04855e 100644 --- a/include/linux/serial_core.h +++ b/include/linux/serial_core.h @@ -59,9 +59,6 @@ #define PORT_SUNZILOG 38 #define PORT_SUNSAB 39 -/* NEC v850. */ -#define PORT_V850E_UART 40 - /* DEC */ #define PORT_DZ 46 #define PORT_ZS 47 diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 0522f368f9d7..4394dadff813 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -443,7 +443,7 @@ asmlinkage long sys_newuname(struct new_utsname __user *name); asmlinkage long sys_getrlimit(unsigned int resource, struct rlimit __user *rlim); -#if defined(COMPAT_RLIM_OLD_INFINITY) || !(defined(CONFIG_IA64) || defined(CONFIG_V850)) +#if defined(COMPAT_RLIM_OLD_INFINITY) || !(defined(CONFIG_IA64)) asmlinkage long sys_old_getrlimit(unsigned int resource, struct rlimit __user *rlim); #endif asmlinkage long sys_setrlimit(unsigned int resource, diff --git a/scripts/genksyms/genksyms.c b/scripts/genksyms/genksyms.c index dca5e0dd09bf..4f8a3007e457 100644 --- a/scripts/genksyms/genksyms.c +++ b/scripts/genksyms/genksyms.c @@ -520,8 +520,7 @@ int main(int argc, char **argv) genksyms_usage(); return 1; } - if ((strcmp(arch, "v850") == 0) || (strcmp(arch, "h8300") == 0) - || (strcmp(arch, "blackfin") == 0)) + if ((strcmp(arch, "h8300") == 0) || (strcmp(arch, "blackfin") == 0)) mod_prefix = "_"; { extern int yydebug; diff --git a/scripts/mod/file2alias.c b/scripts/mod/file2alias.c index 1fcaf3284a6a..4fa1f3ad2513 100644 --- a/scripts/mod/file2alias.c +++ b/scripts/mod/file2alias.c @@ -623,7 +623,7 @@ static int do_i2c_entry(const char *filename, struct i2c_device_id *id, return 1; } -/* Ignore any prefix, eg. v850 prepends _ */ +/* Ignore any prefix, eg. some architectures prepend _ */ static inline int sym_is(const char *symbol, const char *name) { const char *match; diff --git a/scripts/mod/mk_elfconfig.c b/scripts/mod/mk_elfconfig.c index db3881f14c2d..6a96d47bd1e6 100644 --- a/scripts/mod/mk_elfconfig.c +++ b/scripts/mod/mk_elfconfig.c @@ -55,7 +55,7 @@ main(int argc, char **argv) else exit(1); - if ((strcmp(argv[1], "v850") == 0) || (strcmp(argv[1], "h8300") == 0) + if ((strcmp(argv[1], "h8300") == 0) || (strcmp(argv[1], "blackfin") == 0)) printf("#define MODULE_SYMBOL_PREFIX \"_\"\n"); else From 82736f4d1d2b7063b829cc93171a6e5aea8a9c49 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= Date: Wed, 23 Jul 2008 21:28:54 -0700 Subject: [PATCH 245/857] generic irqs: handle failure of irqchip->set_type in setup_irq MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit set_type returns an int indicating success or failure, but up to now setup_irq ignores that. In my case this resulted in a machine hang: gpio-keys requested IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING, but arm/ns9xxx can only trigger on one direction so set_type didn't touch the configuration which happens do default to a level sensitiveness and returned -EINVAL. setup_irq ignored that and unmasked the irq. This resulted in an endless triggering of the gpio-key interrupt service routine which effectively killed the machine. With this patch applied setup_irq propagates the error to the caller. Note that before in the case chip && !chip->set_type && !chip->name a NULL pointer was feed to printk. This is fixed, too. Signed-off-by: Uwe Kleine-König Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/irq/manage.c | 64 +++++++++++++++++++++++++++++---------------- 1 file changed, 42 insertions(+), 22 deletions(-) diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 3cfc0fefb5ee..5bc6e5ecc493 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -308,6 +308,30 @@ void compat_irq_chip_set_default_handler(struct irq_desc *desc) desc->handle_irq = NULL; } +static int __irq_set_trigger(struct irq_chip *chip, unsigned int irq, + unsigned long flags) +{ + int ret; + + if (!chip || !chip->set_type) { + /* + * IRQF_TRIGGER_* but the PIC does not support multiple + * flow-types? + */ + pr_warning("No set_type function for IRQ %d (%s)\n", irq, + chip ? (chip->name ? : "unknown") : "unknown"); + return 0; + } + + ret = chip->set_type(irq, flags & IRQF_TRIGGER_MASK); + + if (ret) + pr_err("setting flow type for irq %u failed (%pF)\n", + irq, chip->set_type); + + return ret; +} + /* * Internal function to register an irqaction - typically used to * allocate special interrupts that are part of the architecture. @@ -319,6 +343,7 @@ int setup_irq(unsigned int irq, struct irqaction *new) const char *old_name = NULL; unsigned long flags; int shared = 0; + int ret; if (irq >= NR_IRQS) return -EINVAL; @@ -376,36 +401,24 @@ int setup_irq(unsigned int irq, struct irqaction *new) shared = 1; } - *p = new; - - /* Exclude IRQ from balancing */ - if (new->flags & IRQF_NOBALANCING) - desc->status |= IRQ_NO_BALANCING; - if (!shared) { irq_chip_set_defaults(desc->chip); + /* Setup the type (level, edge polarity) if configured: */ + if (new->flags & IRQF_TRIGGER_MASK) { + ret = __irq_set_trigger(desc->chip, irq, new->flags); + + if (ret) { + spin_unlock_irqrestore(&desc->lock, flags); + return ret; + } + } else + compat_irq_chip_set_default_handler(desc); #if defined(CONFIG_IRQ_PER_CPU) if (new->flags & IRQF_PERCPU) desc->status |= IRQ_PER_CPU; #endif - /* Setup the type (level, edge polarity) if configured: */ - if (new->flags & IRQF_TRIGGER_MASK) { - if (desc->chip->set_type) - desc->chip->set_type(irq, - new->flags & IRQF_TRIGGER_MASK); - else - /* - * IRQF_TRIGGER_* but the PIC does not support - * multiple flow-types? - */ - printk(KERN_WARNING "No IRQF_TRIGGER set_type " - "function for IRQ %d (%s)\n", irq, - desc->chip->name); - } else - compat_irq_chip_set_default_handler(desc); - desc->status &= ~(IRQ_AUTODETECT | IRQ_WAITING | IRQ_INPROGRESS | IRQ_SPURIOUS_DISABLED); @@ -423,6 +436,13 @@ int setup_irq(unsigned int irq, struct irqaction *new) /* Set default affinity mask once everything is setup */ irq_select_affinity(irq); } + + *p = new; + + /* Exclude IRQ from balancing */ + if (new->flags & IRQF_NOBALANCING) + desc->status |= IRQ_NO_BALANCING; + /* Reset broken irq detection when installing new handler */ desc->irq_count = 0; desc->irqs_unhandled = 0; From 5aa0769d089125e63f8dc23e0283e559e1790493 Mon Sep 17 00:00:00 2001 From: Hans-Christian Egtvedt Date: Wed, 23 Jul 2008 21:28:55 -0700 Subject: [PATCH 246/857] atmel_pwm: set up only one PWM clock when allocating a clock This patch will only setup one clock, if free, and return this clock to the caller. The previous solution would setup both clocks with the same prescaler and divider and return PWM_CPR_CLKB, thus taking both clocks in the same call without the caller knowing. Signed-off-by: Hans-Christian Egtvedt Cc: David Brownell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/misc/atmel_pwm.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/misc/atmel_pwm.c b/drivers/misc/atmel_pwm.c index 5b5a14dab3d3..6aa5294dfec4 100644 --- a/drivers/misc/atmel_pwm.c +++ b/drivers/misc/atmel_pwm.c @@ -211,8 +211,7 @@ int pwm_clk_alloc(unsigned prescale, unsigned div) if ((mr & 0xffff) == 0) { mr |= val; ret = PWM_CPR_CLKA; - } - if ((mr & (0xffff << 16)) == 0) { + } else if ((mr & (0xffff << 16)) == 0) { mr |= val << 16; ret = PWM_CPR_CLKB; } From 6cbb2e711128b505209f7c910018aac77335c887 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:28:55 -0700 Subject: [PATCH 247/857] checkpatch: Version: 0.20 Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 6971bf078d13..66f060ecb81f 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -9,7 +9,7 @@ use strict; my $P = $0; $P =~ s@.*/@@g; -my $V = '0.19'; +my $V = '0.20'; use Getopt::Long qw(:config no_auto_abbrev); From fee61c47d15270bdea699a8a3dd867f0825c3541 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:28:56 -0700 Subject: [PATCH 248/857] checkpatch: return is not a function -- parentheses for casts are ok too Casts require parentheses so it is possible to have something like this: return (int)(*a); This miss trips the complexity function. Ensure that the two separate parenthesised sections are not coelesced. Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 1 + 1 file changed, 1 insertion(+) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 66f060ecb81f..83ae37b38621 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -1670,6 +1670,7 @@ sub process { my $value = $2; # Flatten any parentheses and braces + $value =~ s/\)\(/\) \(/g; while ($value =~ s/\([^\(\)]*\)/1/) { } From c8cb2ca37ed51aa1f3b20e3eff1e72df1c400f70 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:28:57 -0700 Subject: [PATCH 249/857] checkpatch: types: some types may also be identifiers Some types such as typedefs may overlap real identifiers. Be more targetted about when a type can really exist. Where it cannot let it be an identifier. This prevents false reporting of the minus '-' in unary context in the following: foo[bar->bool - 1]; Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 83ae37b38621..5420db6502fc 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -171,6 +171,7 @@ our @modifierList = ( sub build_types { my $mods = "(?: \n" . join("|\n ", @modifierList) . "\n)"; my $all = "(?: \n" . join("|\n ", @typeList) . "\n)"; + $Modifier = qr{(?:$Attribute|$Sparse|$mods)}; $NonptrType = qr{ (?:const\s+)? (?:$mods\s+)? @@ -178,15 +179,14 @@ sub build_types { (?:typeof|__typeof__)\s*\(\s*\**\s*$Ident\s*\)| (?:${all}\b) ) - (?:\s+$Sparse|\s+const)* + (?:\s+$Modifier|\s+const)* }x; $Type = qr{ $NonptrType (?:\s*\*+\s*const|\s*\*+|(?:\s*\[\s*\])+)? - (?:\s+$Inline|\s+$Sparse|\s+$Attribute|\s+$mods)* + (?:\s+$Inline|\s+$Modifier)* }x; $Declare = qr{(?:$Storage\s+)?$Type}; - $Modifier = qr{(?:$Attribute|$Sparse|$mods)}; } build_types(); @@ -715,7 +715,7 @@ sub annotate_values { $av_preprocessor = 0; } - } elsif ($cur =~ /^($Type)/) { + } elsif ($cur =~ /^($Type)\s*(?:$Ident|,|\))/) { print "DECLARE($1)\n" if ($dbg_values > 1); $type = 'T'; @@ -800,8 +800,9 @@ sub annotate_values { print "PAREN('$1')\n" if ($dbg_values > 1); } - } elsif ($cur =~ /^($Ident)\(/o) { + } elsif ($cur =~ /^($Ident)\s*\(/o) { print "FUNC($1)\n" if ($dbg_values > 1); + $type = 'V'; $av_pending = 'V'; } elsif ($cur =~ /^($Ident|$Constant)/o) { From f3db6639fee577f6ed92c0a1fc881e748c47ec48 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Wed, 23 Jul 2008 21:28:57 -0700 Subject: [PATCH 250/857] checkpatch: add a checkpatch warning for new uses of __initcall(). [apw@shadowen.org: generalise pattern and add tests] Signed-off-by: Michael Ellerman Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 5420db6502fc..cf70f123f574 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -2108,6 +2108,10 @@ sub process { if ($line =~ /\bsimple_(strto.*?)\s*\(/) { WARN("consider using strict_$1 in preference to simple_$1\n" . $herecurr); } +# check for __initcall(), use device_initcall() explicitly please + if ($line =~ /^.\s*__initcall\s*\(/) { + WARN("please use device_initcall() instead of __initcall()\n" . $herecurr); + } # use of NR_CPUS is usually wrong # ignore definitions of NR_CPUS and usage to define arrays as likely right From d3ddcf471ea90d7ff711dbaa371ef379ed625ec0 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:28:58 -0700 Subject: [PATCH 251/857] checkpatch: possible types: __asm__ is never a type We are false matching __asm__ as a type, and then tripping the external function checks. Squash. Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index cf70f123f574..fd597a4b5dad 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -846,7 +846,7 @@ sub possible { if ($possible !~ /^(?:$Storage|$Type|DEFINE_\S+)$/ && $possible ne 'goto' && $possible ne 'return' && $possible ne 'case' && $possible ne 'else' && - $possible ne 'asm' && + $possible ne 'asm' && $possible ne '__asm__' && $possible !~ /^(typedef|struct|enum)\b/) { # Check for modifiers. $possible =~ s/\s*$Storage\s*//g; From beae6332493a40116dba24928154621f2e88b9a9 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:28:59 -0700 Subject: [PATCH 252/857] checkpatch: comment detection: ignore macro continuation when detecting associated comments When looking for an associated comment they may be suffixed by a macro continuation. Ignore this. Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index fd597a4b5dad..94250d1a3a47 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -631,7 +631,7 @@ sub ctx_locate_comment { my ($first_line, $end_line) = @_; # Catch a comment on the end of the line itself. - my ($current_comment) = ($rawlines[$end_line - 1] =~ m@.*(/\*.*\*/)\s*$@); + my ($current_comment) = ($rawlines[$end_line - 1] =~ m@.*(/\*.*\*/)\s*(?:\\\s*)?$@); return $current_comment if (defined $current_comment); # Look through the context and try and figure out if there is a From 6ef9b297f6e8850da3be9c9ff5f00385c0977004 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:28:59 -0700 Subject: [PATCH 253/857] checkpatch: types: unary -- goto introduces unary context When we see a goto we enter unary context. For example: goto *h; Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 94250d1a3a47..b2b0648ee14a 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -780,7 +780,7 @@ sub annotate_values { $av_pending = 'N'; $type = 'N'; - } elsif ($cur =~/^(return|case|else)/o) { + } elsif ($cur =~/^(return|case|else|goto)/o) { print "KEYWORD($1)\n" if ($dbg_values > 1); $type = 'N'; From a3bb97a7aba36055d476896ed6393ab35a119d5b Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:29:00 -0700 Subject: [PATCH 254/857] checkpatch: macros: fix statement counting block end detection We are incorrectly counting the lines in a block while accumulating the trailing lines in a macro statement, leading to false positives. Fix end of block handling and general counting for negative context lines. Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 23 +++++++++++------------ 1 file changed, 11 insertions(+), 12 deletions(-) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index b2b0648ee14a..add86862325b 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -470,7 +470,9 @@ sub ctx_statement_block { } $off++; } + # We are truly at the end, so shuffle to the next line. if ($off == $len) { + $loff = $len + 1; $line++; $remain--; } @@ -1793,30 +1795,26 @@ sub process { $lines[$ln - 1] =~ /^(?:-|..*\\$)/) { $ctx .= $rawlines[$ln - 1] . "\n"; + $cnt-- if ($lines[$ln - 1] !~ /^-/); $ln++; - $cnt--; } $ctx .= $rawlines[$ln - 1]; ($dstat, $dcond, $ln, $cnt, $off) = ctx_statement_block($linenr, $ln - $linenr + 1, 0); #print "dstat<$dstat> dcond<$dcond> cnt<$cnt> off<$off>\n"; - #print "LINE<$lines[$ln]> len<" . length($lines[$ln]) . "\n"; + #print "LINE<$lines[$ln-1]> len<" . length($lines[$ln-1]) . "\n"; # Extract the remainder of the define (if any) and # rip off surrounding spaces, and trailing \'s. $rest = ''; - if (defined $lines[$ln - 1] && - $off > length($lines[$ln - 1])) - { + while ($off != 0 || ($cnt > 0 && $rest =~ /(?:^|\\)\s*$/)) { + #print "ADDING $off <" . substr($lines[$ln - 1], $off) . ">\n"; + if ($off != 0 || $lines[$ln - 1] !~ /^-/) { + $rest .= substr($lines[$ln - 1], $off) . "\n"; + $cnt--; + } $ln++; - $cnt--; - $off = 0; - } - while ($cnt > 0) { - $rest .= substr($lines[$ln - 1], $off) . "\n"; - $ln++; - $cnt--; $off = 0; } $rest =~ s/\\\n.//g; @@ -1847,6 +1845,7 @@ sub process { DEFINE_PER_CPU| __typeof__\( }x; + #print "REST<$rest>\n"; if ($rest ne '') { if ($rest !~ /while\s*\(/ && $dstat !~ /$exceptions/) From 548596d523d83dff5a670beb84be0daf4c3bcd16 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:29:01 -0700 Subject: [PATCH 255/857] checkpatch: trailing statement indent: fix end of statement location Fix end of statement location. Where the last line of the statement is replaced we are miss reporting the newly added replacement an incorrectly indented trailing statement for the negative context. We are also incorrectly reporting negative statements generally. Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index add86862325b..89177c349f93 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -1249,17 +1249,22 @@ sub process { my $pre_ctx = "$1$2"; my ($level, @ctx) = ctx_statement_level($linenr, $realcnt, 0); - my $ctx_ln = $linenr + $#ctx + 1; my $ctx_cnt = $realcnt - $#ctx - 1; my $ctx = join("\n", @ctx); - ##warn "realcnt<$realcnt> ctx_cnt<$ctx_cnt>\n"; + my $ctx_ln = $linenr; + my $ctx_skip = $realcnt; - # Skip over any removed lines in the context following statement. - while (defined($lines[$ctx_ln - 1]) && $lines[$ctx_ln - 1] =~ /^-/) { + while ($ctx_skip > $ctx_cnt || ($ctx_skip == $ctx_cnt && + defined $lines[$ctx_ln - 1] && + $lines[$ctx_ln - 1] =~ /^-/)) { + ##print "SKIP<$ctx_skip> CNT<$ctx_cnt>\n"; + $ctx_skip-- if (!defined $lines[$ctx_ln - 1] || $lines[$ctx_ln - 1] !~ /^-/); $ctx_ln++; } - ##warn "pre<$pre_ctx>\nline<$line>\nctx<$ctx>\nnext<$lines[$ctx_ln - 1]>\n"; + + ##print "realcnt<$realcnt> ctx_cnt<$ctx_cnt>\n"; + ##print "pre<$pre_ctx>\nline<$line>\nctx<$ctx>\nnext<$lines[$ctx_ln - 1]>\n"; if ($ctx !~ /{\s*/ && defined($lines[$ctx_ln -1]) && $lines[$ctx_ln - 1] =~ /^\+\s*{/) { ERROR("that open brace { should be on the previous line\n" . From f4c014c0dede10cc0a8463e748892e738e190699 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:29:01 -0700 Subject: [PATCH 256/857] checkpatch: allow printk strings to exceed 80 characters to maintain their searchability Allow printk strings to break the 80 character width limits, thus keeping them complete and searchable. Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 89177c349f93..614999f29aa0 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -1138,7 +1138,9 @@ sub process { } #80 column limit if ($line =~ /^\+/ && $prevrawline !~ /\/\*\*/ && - $rawline !~ /^.\s*\*\s*\@$Ident\s/ && $length > 80) + $rawline !~ /^.\s*\*\s*\@$Ident\s/ && + $line !~ /^\+\s*printk\s*\(\s*(?:KERN_\S+\s*)?"[X\t]*"\s*(?:,|\)\s*;)\s*$/ && + $length > 80) { WARN("line over 80 characters\n" . $herecurr); } From e2a763c20b89890d2153551b1af6962b135de4c0 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:29:02 -0700 Subject: [PATCH 257/857] checkpatch: switch -- report trailing statements on case and default Report trailing statements on case and default lines. Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 614999f29aa0..5f71b3050253 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -1244,6 +1244,10 @@ sub process { ERROR("switch and case should be at the same indent\n$hereline$err"); } } + if ($line =~ /^.\s*(?:case\s*.*|default\s*):/g && + $line !~ /\G(?:\s*{)?(?:\s*$;*)(?:\s*\\)?\s*$/g) { + ERROR("trailing statements should be on next line\n" . $herecurr); + } # if/while/etc brace do not go on next line, unless defining a do while loop, # or if that brace on the next line is for something else From 8d31cfcecf67563d70cd68616cb8fb4384f24b51 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:29:02 -0700 Subject: [PATCH 258/857] checkpatch: check spacing for square brackets Check on the spacing before square brackets. We should only allow spaces there if this is part of a type definition or an initialialiser. Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 5f71b3050253..e7c8ab1b54bd 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -1435,6 +1435,17 @@ sub process { ERROR("open brace '{' following $1 go on the same line\n" . $hereprev); } +# check for spacing round square brackets; allowed: +# 1. with a type on the left -- int [] a; +# 2. at the beginning of a line for slice initialisers -- [0..10] = 5, + while ($line =~ /(.*?\s)\[/g) { + my ($where, $prefix) = ($-[1], $1); + if ($prefix !~ /$Type\s+$/ && + ($where != 0 || $prefix !~ /^.\s+$/)) { + ERROR("space prohibited before open square bracket '['\n" . $herecurr); + } + } + # check for spaces between functions and their parentheses. while ($line =~ /($Ident)\s+\(/g) { my $name = $1; From 53210168feeff9a3c780bd42f69936d4c12381d5 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:29:03 -0700 Subject: [PATCH 259/857] checkpatch: toughen trailing if statement checks and extend them to while and for Extend the trailing statement checks to report a trailing semi-colon ';' as we really want it on the next line and indented so it is really really obvious. Also extend the tests to include while and for. Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index e7c8ab1b54bd..8616baee0aeb 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -1269,8 +1269,8 @@ sub process { $ctx_ln++; } - ##print "realcnt<$realcnt> ctx_cnt<$ctx_cnt>\n"; - ##print "pre<$pre_ctx>\nline<$line>\nctx<$ctx>\nnext<$lines[$ctx_ln - 1]>\n"; + #print "realcnt<$realcnt> ctx_cnt<$ctx_cnt>\n"; + #print "pre<$pre_ctx>\nline<$line>\nctx<$ctx>\nnext<$lines[$ctx_ln - 1]>\n"; if ($ctx !~ /{\s*/ && defined($lines[$ctx_ln -1]) && $lines[$ctx_ln - 1] =~ /^\+\s*{/) { ERROR("that open brace { should be on the previous line\n" . @@ -1713,7 +1713,7 @@ sub process { } # Check for illegal assignment in if conditional. - if ($line =~ /\bif\s*\(/) { + if ($line =~ /\b(?:if|while|for)\s*\(/ && $line !~ /^.\s*#/) { my ($s, $c) = ($stat, $cond); if ($c =~ /\bif\s*\(.*[^<>!=]=[^=].*/) { @@ -1725,8 +1725,8 @@ sub process { substr($s, 0, length($c), ''); $s =~ s/\n.*//g; $s =~ s/$;//g; # Remove any comments - if (length($c) && $s !~ /^\s*({|;|)\s*\\*\s*$/ && - $c !~ /^.\s*\#\s*if/) + if (length($c) && $s !~ /^\s*{?\s*\\*\s*$/ && + $c !~ /}\s*while\s*/) { ERROR("trailing statements should be on next line\n" . $herecurr); } From f5fe35dd95549b1b419cdeb2ec3fe61fda94fa93 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:29:03 -0700 Subject: [PATCH 260/857] checkpatch: condition/loop indent checks Check to see if the block/statement which a condition or loop introduces is indented correctly. Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 59 ++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 55 insertions(+), 4 deletions(-) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 8616baee0aeb..13d7a330b5dd 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -1167,10 +1167,10 @@ sub process { } # Check for potential 'bare' types - my ($stat, $cond); + my ($stat, $cond, $line_nr_next, $remain_next); if ($realcnt && $line =~ /.\s*\S/) { - ($stat, $cond) = ctx_statement_block($linenr, - $realcnt, 0); + ($stat, $cond, $line_nr_next, $remain_next) = + ctx_statement_block($linenr, $realcnt, 0); $stat =~ s/\n./\n /g; $cond =~ s/\n./\n /g; @@ -1712,7 +1712,8 @@ sub process { ERROR("space required before the open parenthesis '('\n" . $herecurr); } -# Check for illegal assignment in if conditional. +# Check for illegal assignment in if conditional -- and check for trailing +# statements after the conditional. if ($line =~ /\b(?:if|while|for)\s*\(/ && $line !~ /^.\s*#/) { my ($s, $c) = ($stat, $cond); @@ -1732,6 +1733,56 @@ sub process { } } +# Check relative indent for conditionals and blocks. + if ($line =~ /\b(?:(?:if|while|for)\s*\(|do\b)/ && $line !~ /^.\s*#/ && $line !~ /\}\s*while\s*/) { + my ($s, $c) = ($stat, $cond); + + substr($s, 0, length($c), ''); + + # Make sure we remove the line prefixes as we have + # none on the first line, and are going to readd them + # where necessary. + $s =~ s/\n./\n/gs; + + # We want to check the first line inside the block + # starting at the end of the conditional, so remove: + # 1) any blank line termination + # 2) any opening brace { on end of the line + # 3) any do (...) { + my $continuation = 0; + my $check = 0; + $s =~ s/^.*\bdo\b//; + $s =~ s/^\s*{//; + if ($s =~ s/^\s*\\//) { + $continuation = 1; + } + if ($s =~ s/^\s*\n//) { + $check = 1; + } + + # Also ignore a loop construct at the end of a + # preprocessor statement. + if (($prevline =~ /^.\s*#\s*define\s/ || + $prevline =~ /\\\s*$/) && $continuation == 0) { + $check = 0; + } + + # Ignore the current line if its is a preprocessor + # line. + if ($s =~ /^\s*#\s*/) { + $check = 0; + } + + my (undef, $sindent) = line_stats("+" . $s); + + ##print "line<$line> prevline<$prevline> indent<$indent> sindent<$sindent> check<$check> continuation<$continuation> s<$s>\n"; + + if ($check && (($sindent % 8) != 0 || + ($sindent <= $indent && $s ne ''))) { + WARN("suspect code indent for conditional statements\n" . $herecurr); + } + } + # Check for bitwise tests written as boolean if ($line =~ / (?: From 4c432a8f0134504814aa8dcce6cc57c89d175604 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Wed, 23 Jul 2008 21:29:04 -0700 Subject: [PATCH 261/857] checkpatch: usb_free_urb() can take NULL usb_free_urb() can take a NULL, so let's check and warn about that. Signed-off-by: Greg Kroah-Hartman Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 13d7a330b5dd..a4e8087a0cae 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -2078,6 +2078,13 @@ sub process { WARN("kfree(NULL) is safe this check is probabally not required\n" . $hereprev); } } +# check for needless usb_free_urb() checks + if ($prevline =~ /\bif\s*\(([^\)]*)\)/) { + my $expr = $1; + if ($line =~ /\busb_free_urb\(\Q$expr\E\);/) { + WARN("usb_free_urb(NULL) is safe this check is probably not required\n" . $hereprev); + } + } # warn about #ifdefs in C files # if ($line =~ /^.\s*\#\s*if(|n)def/ && ($realfile =~ /\.c$/)) { From 3c232147a7d5b0418b0a0bae0e5b9a62fb81f4f2 Mon Sep 17 00:00:00 2001 From: Wolfram Sang Date: Wed, 23 Jul 2008 21:29:05 -0700 Subject: [PATCH 262/857] checkpatch: correct spelling in kfree checks Correct spelling in the kfree reports. Signed-off-by: Wolfram Sang Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index a4e8087a0cae..3961e759a256 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -2075,7 +2075,7 @@ sub process { if ($prevline =~ /\bif\s*\(([^\)]*)\)/) { my $expr = $1; if ($line =~ /\bkfree\(\Q$expr\E\);/) { - WARN("kfree(NULL) is safe this check is probabally not required\n" . $hereprev); + WARN("kfree(NULL) is safe this check is probably not required\n" . $hereprev); } } # check for needless usb_free_urb() checks From 389a2fe57ffc59a649bea39db4d7e6d2eff2b562 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:29:05 -0700 Subject: [PATCH 263/857] checkpatch: allow for type modifiers on multiple declarations Allow for type modifiers mid declaration on multiple declarations: struct mxser_mstatus ms, __user *msu = argp; Reported by Jiri Slaby. Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 3961e759a256..bcfb8ef00fee 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -721,6 +721,10 @@ sub annotate_values { print "DECLARE($1)\n" if ($dbg_values > 1); $type = 'T'; + } elsif ($cur =~ /^($Modifier)\s*/) { + print "MODIFIER($1)\n" if ($dbg_values > 1); + $type = 'T'; + } elsif ($cur =~ /^(\#\s*define\s*$Ident)(\(?)/o) { print "DEFINE($1,$2)\n" if ($dbg_values > 1); $av_preprocessor = 1; From 7429c6903e3628fc2cfea65ec7e13bac030c7bfe Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:29:06 -0700 Subject: [PATCH 264/857] checkpatch: improve type matcher debug Improve type matcher debug so we can see what it does match. As part of this move us to to using the common debug framework. Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index bcfb8ef00fee..077a2ca33043 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -17,7 +17,6 @@ my $quiet = 0; my $tree = 1; my $chk_signoff = 1; my $chk_patch = 1; -my $tst_type = 0; my $tst_only; my $emacs = 0; my $terse = 0; @@ -44,7 +43,6 @@ GetOptions( 'summary-file!' => \$summary_file, 'debug=s' => \%debug, - 'test-type!' => \$tst_type, 'test-only=s' => \$tst_only, ) or exit; @@ -67,6 +65,7 @@ if ($#ARGV < 0) { my $dbg_values = 0; my $dbg_possible = 0; +my $dbg_type = 0; for my $key (keys %debug) { eval "\${dbg_$key} = '$debug{$key}';" } @@ -1307,8 +1306,12 @@ sub process { if ($line=~/^[^\+]/) {next;} # TEST: allow direct testing of the type matcher. - if ($tst_type && $line =~ /^.$Declare$/) { - ERROR("TEST: is type $Declare\n" . $herecurr); + if ($dbg_type) { + if ($line =~ /^.\s*$Declare\s*$/) { + ERROR("TEST: is type\n" . $herecurr); + } elsif ($dbg_type > 1 && $line =~ /^.+($Declare)/) { + ERROR("TEST: is not type ($1 is)\n". $herecurr); + } next; } From d2172eb5bd4b7d06577113ec40635083619ca54a Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:29:07 -0700 Subject: [PATCH 265/857] checkpatch: possible modifiers are not being correctly matched Although we are finding the added modifier in the declaration below we are not correctly matching it as a type. Fix the declaration. static void __ref *vmem_alloc_pages(unsigned int order) { } Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 077a2ca33043..53ec3946670d 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -168,12 +168,11 @@ our @modifierList = ( ); sub build_types { - my $mods = "(?: \n" . join("|\n ", @modifierList) . "\n)"; - my $all = "(?: \n" . join("|\n ", @typeList) . "\n)"; + my $mods = "(?x: \n" . join("|\n ", @modifierList) . "\n)"; + my $all = "(?x: \n" . join("|\n ", @typeList) . "\n)"; $Modifier = qr{(?:$Attribute|$Sparse|$mods)}; $NonptrType = qr{ - (?:const\s+)? - (?:$mods\s+)? + (?:$Modifier\s+|const\s+)* (?: (?:typeof|__typeof__)\s*\(\s*\**\s*$Ident\s*\)| (?:${all}\b) From b8f96a31f38c8e9fc75f0a89c6815e7cbc402858 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:29:07 -0700 Subject: [PATCH 266/857] checkpatch: macro complexity checks are meaningless in linker scripts Exclude vmlinux.lds.h from the macro complexity checks. They will never apply sanely here. Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 53ec3946670d..775f2b146aa1 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -1860,7 +1860,8 @@ sub process { # multi-statement macros should be enclosed in a do while loop, grab the # first statement and ensure its the whole macro if its not enclosed # in a known good container - if ($line =~ /^.\s*\#\s*define\s*$Ident(\()?/) { + if ($realfile !~ m@/vmlinux.lds.h$@ && + $line =~ /^.\s*\#\s*define\s*$Ident(\()?/) { my $ln = $linenr; my $cnt = $realcnt; my ($off, $dstat, $dcond, $rest); From 8ea3eb9a20f39d5afa52900a34092b4b5f6b55cb Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:29:08 -0700 Subject: [PATCH 267/857] checkpatch: handle return types of pointers to functions Make sure we correctly mark the return type of the pointer to a function declaration. const void *(*sb_tag)(struct sysfs_tag_info *info); Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 775f2b146aa1..6d07b6778c91 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -715,7 +715,7 @@ sub annotate_values { $av_preprocessor = 0; } - } elsif ($cur =~ /^($Type)\s*(?:$Ident|,|\))/) { + } elsif ($cur =~ /^($Type)\s*(?:$Ident|,|\)|\()/) { print "DECLARE($1)\n" if ($dbg_values > 1); $type = 'T'; From 0221f55c142b0ac8baf6f0b6c4e1ec89f0c98e96 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:29:08 -0700 Subject: [PATCH 268/857] checkpatch: possible types -- known modifiers cannot be types Ensure we do not inadvertantly load known modifiers up as possible types. Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 6d07b6778c91..9c209165f252 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -847,7 +847,7 @@ sub possible { my ($possible, $line) = @_; print "CHECK<$possible> ($line)\n" if ($dbg_possible > 1); - if ($possible !~ /^(?:$Storage|$Type|DEFINE_\S+)$/ && + if ($possible !~ /^(?:$Modifier|$Storage|$Type|DEFINE_\S+)$/ && $possible ne 'goto' && $possible ne 'return' && $possible ne 'case' && $possible ne 'else' && $possible ne 'asm' && $possible ne '__asm__' && From d2506586586c59f5db0e2ce00d5d31ccec6260b8 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:29:09 -0700 Subject: [PATCH 269/857] checkpatch: possible modifiers -- handle multiple modifiers and trailing Add support for multiple modifiers such as: int __one __two foo; Also handle trailing known modifiers when defecting modifiers: int __one foo __read_mostly; Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 9c209165f252..8a3b0fd67ad7 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -859,8 +859,10 @@ sub possible { } elsif ($possible =~ /\s/) { $possible =~ s/\s*$Type\s*//g; - warn "MODIFIER: $possible ($line)\n" if ($dbg_possible); - push(@modifierList, $possible); + for my $modifier (split(' ', $possible)) { + warn "MODIFIER: $modifier ($possible) ($line)\n" if ($dbg_possible); + push(@modifierList, $modifier); + } } else { warn "POSSIBLE: $possible ($line)\n" if ($dbg_possible); @@ -1186,7 +1188,7 @@ sub process { } elsif ($s =~ /^.\s*$Ident\s*\(/s) { # declarations always start with types - } elsif ($prev_values eq 'E' && $s =~ /^.\s*(?:$Storage\s+)?(?:$Inline\s+)?(?:const\s+)?((?:\s*$Ident)+)\b(?:\s+$Sparse)?\s*\**\s*(?:$Ident|\(\*[^\)]*\))\s*(?:;|=|,|\()/s) { + } elsif ($prev_values eq 'E' && $s =~ /^.\s*(?:$Storage\s+)?(?:$Inline\s+)?(?:const\s+)?((?:\s*$Ident)+?)\b(?:\s+$Sparse)?\s*\**\s*(?:$Ident|\(\*[^\)]*\))(?:\s*$Modifier)?\s*(?:;|=|,|\()/s) { my $type = $1; $type =~ s/\s+/ /g; possible($type, "A:" . $s); From 1f65f947a6a875e1fe7867dc08e981c4101d435d Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:29:10 -0700 Subject: [PATCH 270/857] checkpatch: add checks for question mark and colon spacing Add checks for the question mark colon operator spacing, and also check the other uses of colon. Colon means a number of things: - it introduces the else part of the ?: operator, - it terminates a goto label, - it terminates the case value, - it separates the identifier from the bit size on bit fields, and - it is used to introduce option types in asm(). Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 81 ++++++++++++++++++++++++++++++++++++++----- 1 file changed, 72 insertions(+), 9 deletions(-) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 8a3b0fd67ad7..88027f237cdf 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -689,17 +689,20 @@ sub cat_vet { my $av_preprocessor = 0; my $av_pending; my @av_paren_type; +my $av_pend_colon; sub annotate_reset { $av_preprocessor = 0; $av_pending = '_'; @av_paren_type = ('E'); + $av_pend_colon = 'O'; } sub annotate_values { my ($stream, $type) = @_; my $res; + my $var = '_' x length($stream); my $cur = $stream; print "$stream\n" if ($dbg_values > 1); @@ -784,7 +787,12 @@ sub annotate_values { $av_pending = 'N'; $type = 'N'; - } elsif ($cur =~/^(return|case|else|goto)/o) { + } elsif ($cur =~/^(case)/o) { + print "CASE($1)\n" if ($dbg_values > 1); + $av_pend_colon = 'C'; + $type = 'N'; + + } elsif ($cur =~/^(return|else|goto)/o) { print "KEYWORD($1)\n" if ($dbg_values > 1); $type = 'N'; @@ -809,6 +817,15 @@ sub annotate_values { $type = 'V'; $av_pending = 'V'; + } elsif ($cur =~ /^($Ident\s*):/) { + if ($type eq 'E') { + $av_pend_colon = 'L'; + } elsif ($type eq 'T') { + $av_pend_colon = 'B'; + } + print "IDENT_COLON($1,$type>$av_pend_colon)\n" if ($dbg_values > 1); + $type = 'V'; + } elsif ($cur =~ /^($Ident|$Constant)/o) { print "IDENT($1)\n" if ($dbg_values > 1); $type = 'V'; @@ -820,8 +837,24 @@ sub annotate_values { } elsif ($cur =~/^(;|{|})/) { print "END($1)\n" if ($dbg_values > 1); $type = 'E'; + $av_pend_colon = 'O'; - } elsif ($cur =~ /^(;|\?|:|\[)/o) { + } elsif ($cur =~ /^(\?)/o) { + print "QUESTION($1)\n" if ($dbg_values > 1); + $type = 'N'; + + } elsif ($cur =~ /^(:)/o) { + print "COLON($1,$av_pend_colon)\n" if ($dbg_values > 1); + + substr($var, length($res), 1, $av_pend_colon); + if ($av_pend_colon eq 'C' || $av_pend_colon eq 'L') { + $type = 'E'; + } else { + $type = 'N'; + } + $av_pend_colon = 'O'; + + } elsif ($cur =~ /^(;|\[)/o) { print "CLOSE($1)\n" if ($dbg_values > 1); $type = 'N'; @@ -840,7 +873,7 @@ sub annotate_values { } } - return $res; + return ($res, $var); } sub possible { @@ -1294,12 +1327,14 @@ sub process { # Track the 'values' across context and added lines. my $opline = $line; $opline =~ s/^./ /; - my $curr_values = annotate_values($opline . "\n", $prev_values); + my ($curr_values, $curr_vars) = + annotate_values($opline . "\n", $prev_values); $curr_values = $prev_values . $curr_values; if ($dbg_values) { my $outline = $opline; $outline =~ s/\t/ /g; print "$linenr > .$outline\n"; print "$linenr > $curr_values\n"; + print "$linenr > $curr_vars\n"; } $prev_values = substr($curr_values, -1); @@ -1490,7 +1525,8 @@ sub process { <<=|>>=|<=|>=|==|!=| \+=|-=|\*=|\/=|%=|\^=|\|=|&=| =>|->|<<|>>|<|>|=|!|~| - &&|\|\||,|\^|\+\+|--|&|\||\+|-|\*|\/|% + &&|\|\||,|\^|\+\+|--|&|\||\+|-|\*|\/|%| + \?|: }x; my @elements = split(/($ops|;)/, $opline); my $off = 0; @@ -1554,6 +1590,9 @@ sub process { # print "UNARY: <$op_left$op_type $is_unary $a:$op:$c> <$ca:$op:$cc> <$unary_ctx>\n"; #} + # Get the full operator variant. + my $opv = $op . substr($curr_vars, $off, 1); + # Ignore operators passed as parameters. if ($op_type ne 'V' && $ca =~ /\s$/ && $cc =~ /^\s*,/) { @@ -1571,8 +1610,10 @@ sub process { # // is a comment } elsif ($op eq '//') { - # -> should have no spaces - } elsif ($op eq '->') { + # No spaces for: + # -> + # : when part of a bitfield + } elsif ($op eq '->' || $opv eq ':B') { if ($ctx =~ /Wx.|.xW/) { ERROR("spaces prohibited around that '$op' $at\n" . $hereptr); } @@ -1628,11 +1669,33 @@ sub process { $hereptr); } + # A colon needs no spaces before when it is + # terminating a case value or a label. + } elsif ($opv eq ':C' || $opv eq ':L') { + if ($ctx =~ /Wx./) { + ERROR("space prohibited before that '$op' $at\n" . $hereptr); + } + # All the others need spaces both sides. } elsif ($ctx !~ /[EWC]x[CWE]/) { + my $ok = 0; + # Ignore email addresses - if (!($op eq '<' && $cb =~ /$;\S+\@\S+>/) && - !($op eq '>' && $cb =~ /<\S+\@\S+$;/)) { + if (($op eq '<' && + $cc =~ /^\S+\@\S+>/) || + ($op eq '>' && + $ca =~ /<\S+\@\S+$/)) + { + $ok = 1; + } + + # Ignore ?: + if (($opv eq ':O' && $ca =~ /\?$/) || + ($op eq '?' && $cc =~ /^:/)) { + $ok = 1; + } + + if ($ok == 0) { ERROR("spaces required around that '$op' $at\n" . $hereptr); } } From 74048ed811152a995a88945ba9e0dded34adfff4 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:29:10 -0700 Subject: [PATCH 271/857] checkpatch: variants -- move the main unary/binary operators to use variants Now that we have a variants system, move to using that to carry the unary/binary designation for +, -, &, and *. Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 36 ++++++++++++++++++------------------ 1 file changed, 18 insertions(+), 18 deletions(-) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 88027f237cdf..8afa88aaed96 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -858,6 +858,19 @@ sub annotate_values { print "CLOSE($1)\n" if ($dbg_values > 1); $type = 'N'; + } elsif ($cur =~ /^(-(?![->])|\+(?!\+)|\*|\&(?!\&))/o) { + my $variant; + + print "OPV($1)\n" if ($dbg_values > 1); + if ($type eq 'V') { + $variant = 'B'; + } else { + $variant = 'U'; + } + + substr($var, length($res), 1, $variant); + $type = 'N'; + } elsif ($cur =~ /^($Operators)/o) { print "OP($1)\n" if ($dbg_values > 1); if ($1 ne '++' && $1 ne '--') { @@ -1573,22 +1586,8 @@ sub process { my $ptr = substr($blank, 0, $off) . "^"; my $hereptr = "$hereline$ptr\n"; - # Classify operators into binary, unary, or - # definitions (* only) where they have more - # than one mode. + # Pull out the value of this operator. my $op_type = substr($curr_values, $off + 1, 1); - my $op_left = substr($curr_values, $off, 1); - my $is_unary; - if ($op_type eq 'T') { - $is_unary = 2; - } elsif ($op_left eq 'V') { - $is_unary = 0; - } else { - $is_unary = 1; - } - #if ($op eq '-' || $op eq '&' || $op eq '*') { - # print "UNARY: <$op_left$op_type $is_unary $a:$op:$c> <$ca:$op:$cc> <$unary_ctx>\n"; - #} # Get the full operator variant. my $opv = $op . substr($curr_vars, $off, 1); @@ -1625,18 +1624,19 @@ sub process { } # '*' as part of a type definition -- reported already. - } elsif ($op eq '*' && $is_unary == 2) { + } elsif ($opv eq '*_') { #warn "'*' is part of type\n"; # unary operators should have a space before and # none after. May be left adjacent to another # unary operator, or a cast } elsif ($op eq '!' || $op eq '~' || - ($is_unary && ($op eq '*' || $op eq '-' || $op eq '&'))) { + $opv eq '*U' || $opv eq '-U' || + $opv eq '&U') { if ($ctx !~ /[WEBC]x./ && $ca !~ /(?:\)|!|~|\*|-|\&|\||\+\+|\-\-|\{)$/) { ERROR("space required before that '$op' $at\n" . $hereptr); } - if ($op eq '*' && $cc =~/\s*const\b/) { + if ($op eq '*' && $cc =~/\s*const\b/) { # A unary '*' may be const } elsif ($ctx =~ /.xW/) { From 292f1a9b342d763f94ea3915726a48905be4acd1 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:29:11 -0700 Subject: [PATCH 272/857] checkpatch: complex macros need to ignore comments Ensure we ignore comments in complex macro detection else we incorrectly report this: #define PFM_GROUP_PERM_ANY -1 /* any user/group */ Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 1 + 1 file changed, 1 insertion(+) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 8afa88aaed96..96a762be5748 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -1972,6 +1972,7 @@ sub process { } else { $dstat =~ s/^.\s*\#\s*define\s+$Ident\s*//; } + $dstat =~ s/$;//g; $dstat =~ s/\\\n.//g; $dstat =~ s/^\s*//s; $dstat =~ s/\s*$//s; From 234fff6515a11cf3e67c793146689da426787fea Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:29:12 -0700 Subject: [PATCH 273/857] checkpatch: types cannot start mid word for pointer tests When checking spacing for pointer checks the type cannot start in the middle of a word, ie. this is not 'int * bar': x = fooint * bar; Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 96a762be5748..022ee557b681 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -1435,11 +1435,11 @@ sub process { ERROR("\"(foo $1 )\" should be \"(foo $1)\"\n" . $herecurr); - } elsif ($line =~ m{$NonptrType(\*+)(?:\s+(?:$Attribute|$Sparse))?\s+[A-Za-z\d_]+}) { + } elsif ($line =~ m{\b$NonptrType(\*+)(?:\s+(?:$Attribute|$Sparse))?\s+[A-Za-z\d_]+}) { ERROR("\"foo$1 bar\" should be \"foo $1bar\"\n" . $herecurr); - } elsif ($line =~ m{$NonptrType\s+(\*+)(?!\s+(?:$Attribute|$Sparse))\s+[A-Za-z\d_]+}) { + } elsif ($line =~ m{\b$NonptrType\s+(\*+)(?!\s+(?:$Attribute|$Sparse))\s+[A-Za-z\d_]+}) { ERROR("\"foo $1 bar\" should be \"foo $1bar\"\n" . $herecurr); } From 33cba0657393a75e18e1781e3e13613303f18124 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Wed, 23 Jul 2008 21:29:12 -0700 Subject: [PATCH 274/857] checkpatch: version 0.21 Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/checkpatch.pl | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 022ee557b681..bc6779398229 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -9,7 +9,7 @@ use strict; my $P = $0; $P =~ s@.*/@@g; -my $V = '0.20'; +my $V = '0.21'; use Getopt::Long qw(:config no_auto_abbrev); From 7102ed519a08b70eadc8fea9d8765d2d990241d1 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 23 Jul 2008 21:29:13 -0700 Subject: [PATCH 275/857] remove the OSS trident driver SOUND_TRIDENT was the last PCI OSS driver, and since there's already an ALSA driver for the same hardware we can remove it. [muli@il.ibm.com: update CREDITS] Signed-off-by: Adrian Bunk Signed-off-by: Muli Ben-Yehuda Signed-off-by: Muli Ben-Yehuda Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- CREDITS | 8 + MAINTAINERS | 6 - sound/oss/Kconfig | 41 - sound/oss/Makefile | 1 - sound/oss/trident.c | 4654 ------------------------------------------- sound/oss/trident.h | 358 ---- 6 files changed, 8 insertions(+), 5060 deletions(-) delete mode 100644 sound/oss/trident.c delete mode 100644 sound/oss/trident.h diff --git a/CREDITS b/CREDITS index 077b147388bd..c62dcb3b7e26 100644 --- a/CREDITS +++ b/CREDITS @@ -317,6 +317,14 @@ S: 2322 37th Ave SW S: Seattle, Washington 98126-2010 S: USA +N: Muli Ben-Yehuda +E: mulix@mulix.org +E: muli@il.ibm.com +W: http://www.mulix.org +D: trident OSS sound driver, x86-64 dma-ops and Calgary IOMMU, +D: KVM and Xen bits and other misc. hackery. +S: Haifa, Israel + N: Johannes Berg E: johannes@sipsolutions.net W: http://johannes.sipsolutions.net/ diff --git a/MAINTAINERS b/MAINTAINERS index 7e5c7b0290bb..5ecb97e13e59 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -4080,12 +4080,6 @@ W: http://www.prosec.rub.de/tpm/ L: tpmdd-devel@lists.sourceforge.net S: Maintained -TRIDENT 4DWAVE/SIS 7018 PCI AUDIO CORE -P: Muli Ben-Yehuda -M: mulix@mulix.org -L: linux-kernel@vger.kernel.org -S: Maintained - TRIVIAL PATCHES P: Jesper Juhl M: trivial@kernel.org diff --git a/sound/oss/Kconfig b/sound/oss/Kconfig index 33940139844b..d4fafb6eec6c 100644 --- a/sound/oss/Kconfig +++ b/sound/oss/Kconfig @@ -35,47 +35,6 @@ config SOUND_AU1550_AC97 tristate "Au1550/Au1200 AC97 Sound" depends on SOC_AU1550 || SOC_AU1200 -config SOUND_TRIDENT - tristate "Trident 4DWave DX/NX, SiS 7018 or ALi 5451 PCI Audio Core" - depends on PCI - ---help--- - Say Y or M if you have a PCI sound card utilizing the Trident - 4DWave-DX/NX chipset or your mother board chipset has SiS 7018 - or ALi 5451 built-in. The SiS 7018 PCI Audio Core is embedded - in SiS960 Super South Bridge and SiS540/630 Single Chipset. - The ALi 5451 PCI Audio Core is embedded in ALi M1535, M1535D, - M1535+ or M1535D+ South Bridge. - - Use lspci -n to find out if your sound card or chipset uses - Trident 4DWave or SiS 7018. PCI ID 1023:2000 or 1023:2001 stands - for Trident 4Dwave. PCI ID 1039:7018 stands for SiS7018. PCI ID - 10B9:5451 stands for ALi5451. - - This driver supports S/PDIF in/out (record/playback) for ALi 5451 - embedded in ALi M1535+ and M1535D+. Note that they aren't all - enabled by default; you can enable them by saying Y to "/proc file - system support" and "Sysctl support", and after the /proc file - system has been mounted, executing the command - - command what is enabled - - echo 0>/proc/ALi5451 pcm out is also set to S/PDIF out. (Default). - - echo 1>/proc/ALi5451 use S/PDIF out to output pcm data. - - echo 2>/proc/ALi5451 use S/PDIF out to output non-pcm data. - (AC3...). - - echo 3>/proc/ALi5451 record from Ac97 in(MIC, Line in...). - (Default). - - echo 4>/proc/ALi5451 no matter Ac97 settings, record from S/PDIF - in. - - - This driver differs slightly from OSS/Free, so PLEASE READ the - comments at the top of . - config SOUND_MSNDCLAS tristate "Support for Turtle Beach MultiSound Classic, Tahiti, Monterey" depends on (m || !STANDALONE) && ISA diff --git a/sound/oss/Makefile b/sound/oss/Makefile index 1f86299fae40..3a141474fb73 100644 --- a/sound/oss/Makefile +++ b/sound/oss/Makefile @@ -29,7 +29,6 @@ obj-$(CONFIG_SOUND_MSNDCLAS) += msnd.o msnd_classic.o obj-$(CONFIG_SOUND_MSNDPIN) += msnd.o msnd_pinnacle.o obj-$(CONFIG_SOUND_VWSND) += vwsnd.o obj-$(CONFIG_SOUND_AU1550_AC97) += au1550_ac97.o ac97_codec.o -obj-$(CONFIG_SOUND_TRIDENT) += trident.o ac97_codec.o obj-$(CONFIG_SOUND_BCM_CS4297A) += swarm_cs4297a.o obj-$(CONFIG_SOUND_WM97XX) += ac97_plugin_wm97xx.o diff --git a/sound/oss/trident.c b/sound/oss/trident.c deleted file mode 100644 index f43f91ef86c7..000000000000 --- a/sound/oss/trident.c +++ /dev/null @@ -1,4654 +0,0 @@ -/* - * OSS driver for Linux 2.[46].x for - * - * Trident 4D-Wave - * SiS 7018 - * ALi 5451 - * Tvia/IGST CyberPro 5050 - * - * Driver: Alan Cox - * - * Built from: - * Low level code: from ALSA - * Framework: Thomas Sailer - * Extended by: Zach Brown - * - * Hacked up by: - * Aaron Holtzman - * Ollie Lho SiS 7018 Audio Core Support - * Ching-Ling Lee ALi 5451 Audio Core Support - * Matt Wu ALi 5451 Audio Core Support - * Peter Wächtler CyberPro5050 support - * Muli Ben-Yehuda - * - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. - * - * History - * v0.14.10j - * January 3 2004 Eugene Teo - * minor cleanup to use pr_debug instead of TRDBG since it is already - * defined in linux/kernel.h. - * v0.14.10i - * December 29 2003 Muli Ben-Yehuda - * major cleanup for 2.6, fix a few error patch buglets - * with returning without properly cleaning up first, - * get rid of lock_kernel(). - * v0.14.10h - * Sept 10 2002 Pascal Schmidt - * added support for ALi 5451 joystick port - * v0.14.10g - * Sept 05 2002 Alan Cox - * adapt to new pci joystick attachment interface - * v0.14.10f - * July 24 2002 Muli Ben-Yehuda - * patch from Eric Lemar (via Ian Soboroff): in suspend and resume, - * fix wrong cast from pci_dev* to struct trident_card*. - * v0.14.10e - * July 19 2002 Muli Ben-Yehuda - * rewrite the DMA buffer allocation/deallcoation functions, to make it - * modular and fix a bug where we would call free_pages on memory - * obtained with pci_alloc_consistent. Also remove unnecessary #ifdef - * CONFIG_PROC_FS and various other cleanups. - * v0.14.10d - * July 19 2002 Muli Ben-Yehuda - * made several printk(KERN_NOTICE...) into TRDBG(...), to avoid spamming - * my syslog with hundreds of messages. - * v0.14.10c - * July 16 2002 Muli Ben-Yehuda - * Cleaned up Lei Hu's 0.4.10 driver to conform to Documentation/CodingStyle - * and the coding style used in the rest of the file. - * v0.14.10b - * June 23 2002 Muli Ben-Yehuda - * add a missing unlock_set_fmt, remove a superflous lock/unlock pair - * with nothing in between. - * v0.14.10a - * June 21 2002 Muli Ben-Yehuda - * use a debug macro instead of #ifdef CONFIG_DEBUG, trim to 80 columns - * per line, use 'do {} while (0)' in statement macros. - * v0.14.10 - * June 6 2002 Lei Hu - * rewrite the part to read/write registers of audio codec for Ali5451 - * v0.14.9e - * January 2 2002 Vojtech Pavlik added gameport - * support to avoid resource conflict with pcigame.c - * v0.14.9d - * October 8 2001 Arnaldo Carvalho de Melo - * use set_current_state, properly release resources on failure in - * trident_probe, get rid of check_region - * v0.14.9c - * August 10 2001 Peter Wächtler - * added support for Tvia (formerly Integraphics/IGST) CyberPro5050 - * this chip is often found in settop boxes (combined video+audio) - * v0.14.9b - * Switch to static inline not extern inline (gcc 3) - * v0.14.9a - * Aug 6 2001 Alan Cox - * 0.14.9 crashed on rmmod due to a timer/bh left running. Simplified - * the existing logic (the BH doesn't help as ac97 is lock_irqsave) - * and used del_timer_sync to clean up - * Fixed a problem where the ALi change broke my generic card - * v0.14.9 - * Jul 10 2001 Matt Wu - * Add H/W Volume Control - * v0.14.8a - * July 7 2001 Alan Cox - * Moved Matt Wu's ac97 register cache into the card structure - * v0.14.8 - * Apr 30 2001 Matt Wu - * Set EBUF1 and EBUF2 to still mode - * Add dc97/ac97 reset function - * Fix power management: ali_restore_regs - * unreleased - * Mar 09 2001 Matt Wu - * Add cache for ac97 access - * v0.14.7 - * Feb 06 2001 Matt Wu - * Fix ac97 initialization - * Fix bug: an extra tail will be played when playing - * Jan 05 2001 Matt Wu - * Implement multi-channels and S/PDIF in support for ALi 1535+ - * v0.14.6 - * Nov 1 2000 Ching-Ling Lee - * Fix the bug of memory leak when switching 5.1-channels to 2 channels. - * Add lock protection into dynamic changing format of data. - * Oct 18 2000 Ching-Ling Lee - * 5.1-channels support for ALi - * June 28 2000 Ching-Ling Lee - * S/PDIF out/in(playback/record) support for ALi 1535+, using /proc to be selected by user - * Simple Power Management support for ALi - * v0.14.5 May 23 2000 Ollie Lho - * Misc bug fix from the Net - * v0.14.4 May 20 2000 Aaron Holtzman - * Fix kfree'd memory access in release - * Fix race in open while looking for a free virtual channel slot - * remove open_wait wq (which appears to be unused) - * v0.14.3 May 10 2000 Ollie Lho - * fixed a small bug in trident_update_ptr, xmms 1.0.1 no longer uses 100% CPU - * v0.14.2 Mar 29 2000 Ching-Ling Lee - * Add clear to silence advance in trident_update_ptr - * fix invalid data of the end of the sound - * v0.14.1 Mar 24 2000 Ching-Ling Lee - * ALi 5451 support added, playback and recording O.K. - * ALi 5451 originally developed and structured based on sonicvibes, and - * suggested to merge into this file by Alan Cox. - * v0.14 Mar 15 2000 Ollie Lho - * 5.1 channel output support with channel binding. What's the Matrix ? - * v0.13.1 Mar 10 2000 Ollie Lho - * few minor bugs on dual codec support, needs more testing - * v0.13 Mar 03 2000 Ollie Lho - * new pci_* for 2.4 kernel, back ported to 2.2 - * v0.12 Feb 23 2000 Ollie Lho - * Preliminary Recording support - * v0.11.2 Feb 19 2000 Ollie Lho - * removed incomplete full-dulplex support - * v0.11.1 Jan 28 2000 Ollie Lho - * small bug in setting sample rate for 4d-nx (reported by Aaron) - * v0.11 Jan 27 2000 Ollie Lho - * DMA bug, scheduler latency, second try - * v0.10 Jan 24 2000 Ollie Lho - * DMA bug fixed, found kernel scheduling problem - * v0.09 Jan 20 2000 Ollie Lho - * Clean up of channel register access routine (prepare for channel binding) - * v0.08 Jan 14 2000 Ollie Lho - * Isolation of AC97 codec code - * v0.07 Jan 13 2000 Ollie Lho - * Get rid of ugly old low level access routines (e.g. CHRegs.lp****) - * v0.06 Jan 11 2000 Ollie Lho - * Preliminary support for dual (more ?) AC97 codecs - * v0.05 Jan 08 2000 Luca Montecchiani - * adapt to 2.3.x new __setup/__init call - * v0.04 Dec 31 1999 Ollie Lho - * Multiple Open, using Middle Loop Interrupt to smooth playback - * v0.03 Dec 24 1999 Ollie Lho - * mem leak in prog_dmabuf and dealloc_dmabuf removed - * v0.02 Dec 15 1999 Ollie Lho - * SiS 7018 support added, playback O.K. - * v0.01 Alan Cox et. al. - * Initial Release in kernel 2.3.30, does not work - * - * ToDo - * Clean up of low level channel register access code. (done) - * Fix the bug on dma buffer management in update_ptr, read/write, drain_dac (done) - * Dual AC97 codecs support (done) - * Recording support (done) - * Mmap support - * "Channel Binding" ioctl extension (done) - * new pci device driver interface for 2.4 kernel (done) - * - * Lock order (high->low) - * lock - hardware lock - * open_mutex - guard opens - * sem - guard dmabuf, write re-entry etc - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include - -#if defined(CONFIG_ALPHA_NAUTILUS) || defined(CONFIG_ALPHA_GENERIC) -#include -#endif - -#include "trident.h" - -#define DRIVER_VERSION "0.14.10j-2.6" - -#if defined(CONFIG_GAMEPORT) || (defined(MODULE) && defined(CONFIG_GAMEPORT_MODULE)) -#define SUPPORT_JOYSTICK 1 -#endif - -/* magic numbers to protect our data structures */ -#define TRIDENT_CARD_MAGIC 0x5072696E /* "Prin" */ -#define TRIDENT_STATE_MAGIC 0x63657373 /* "cess" */ - -#define TRIDENT_DMA_MASK 0x3fffffff /* DMA buffer mask for pci_alloc_consist */ -#define ALI_DMA_MASK 0x7fffffff /* ALI Tridents have 31-bit DMA. Wow. */ - -#define NR_HW_CH 32 - -/* maximum number of AC97 codecs connected, AC97 2.0 defined 4, but 7018 and 4D-NX only - have 2 SDATA_IN lines (currently) */ -#define NR_AC97 2 - -/* minor number of /dev/swmodem (temporary, experimental) */ -#define SND_DEV_SWMODEM 7 - -static const unsigned ali_multi_channels_5_1[] = { - /*ALI_SURR_LEFT_CHANNEL, ALI_SURR_RIGHT_CHANNEL, */ - ALI_CENTER_CHANNEL, - ALI_LEF_CHANNEL, - ALI_SURR_LEFT_CHANNEL, - ALI_SURR_RIGHT_CHANNEL -}; - -static const unsigned sample_size[] = { 1, 2, 2, 4 }; -static const unsigned sample_shift[] = { 0, 1, 1, 2 }; - -static const char invalid_magic[] = KERN_CRIT "trident: invalid magic value in %s\n"; - -enum { - TRIDENT_4D_DX = 0, - TRIDENT_4D_NX, - SIS_7018, - ALI_5451, - CYBER5050 -}; - -static char *card_names[] = { - "Trident 4DWave DX", - "Trident 4DWave NX", - "SiS 7018 PCI Audio", - "ALi Audio Accelerator", - "Tvia/IGST CyberPro 5050" -}; - -static struct pci_device_id trident_pci_tbl[] = { - {PCI_DEVICE(PCI_VENDOR_ID_TRIDENT, PCI_DEVICE_ID_TRIDENT_4DWAVE_DX), - PCI_CLASS_MULTIMEDIA_AUDIO << 8, 0xffff00, TRIDENT_4D_DX}, - {PCI_DEVICE(PCI_VENDOR_ID_TRIDENT, PCI_DEVICE_ID_TRIDENT_4DWAVE_NX), - 0, 0, TRIDENT_4D_NX}, - {PCI_DEVICE(PCI_VENDOR_ID_SI, PCI_DEVICE_ID_SI_7018), 0, 0, SIS_7018}, - {PCI_DEVICE(PCI_VENDOR_ID_ALI, PCI_DEVICE_ID_ALI_5451), 0, 0, ALI_5451}, - {PCI_DEVICE(PCI_VENDOR_ID_INTERG, PCI_DEVICE_ID_INTERG_5050), - 0, 0, CYBER5050}, - {0,} -}; - -MODULE_DEVICE_TABLE(pci, trident_pci_tbl); - -/* "software" or virtual channel, an instance of opened /dev/dsp */ -struct trident_state { - unsigned int magic; - struct trident_card *card; /* Card info */ - - /* file mode */ - mode_t open_mode; - - /* virtual channel number */ - int virt; - - struct dmabuf { - /* wave sample stuff */ - unsigned int rate; - unsigned char fmt, enable; - - /* hardware channel */ - struct trident_channel *channel; - - /* OSS buffer management stuff */ - void *rawbuf; - dma_addr_t dma_handle; - unsigned buforder; - unsigned numfrag; - unsigned fragshift; - - /* our buffer acts like a circular ring */ - unsigned hwptr; /* where dma last started, updated by update_ptr */ - unsigned swptr; /* where driver last clear/filled, updated by read/write */ - int count; /* bytes to be comsumed or been generated by dma machine */ - unsigned total_bytes; /* total bytes dmaed by hardware */ - - unsigned error; /* number of over/underruns */ - /* put process on wait queue when no more space in buffer */ - wait_queue_head_t wait; - - /* redundant, but makes calculations easier */ - unsigned fragsize; - unsigned dmasize; - unsigned fragsamples; - - /* OSS stuff */ - unsigned mapped:1; - unsigned ready:1; - unsigned endcleared:1; - unsigned update_flag; - unsigned ossfragshift; - int ossmaxfrags; - unsigned subdivision; - - } dmabuf; - - /* 5.1 channels */ - struct trident_state *other_states[4]; - int multi_channels_adjust_count; - unsigned chans_num; - unsigned long fmt_flag; - /* Guard against mmap/write/read races */ - struct mutex sem; - -}; - -/* hardware channels */ -struct trident_channel { - int num; /* channel number */ - u32 lba; /* Loop Begine Address, where dma buffer starts */ - u32 eso; /* End Sample Offset, wehre dma buffer ends */ - /* (in the unit of samples) */ - u32 delta; /* delta value, sample rate / 48k for playback, */ - /* 48k/sample rate for recording */ - u16 attribute; /* control where PCM data go and come */ - u16 fm_vol; - u32 control; /* signed/unsigned, 8/16 bits, mono/stereo */ -}; - -struct trident_pcm_bank_address { - u32 start; - u32 stop; - u32 aint; - u32 aint_en; -}; - -static struct trident_pcm_bank_address bank_a_addrs = { - T4D_START_A, - T4D_STOP_A, - T4D_AINT_A, - T4D_AINTEN_A -}; - -static struct trident_pcm_bank_address bank_b_addrs = { - T4D_START_B, - T4D_STOP_B, - T4D_AINT_B, - T4D_AINTEN_B -}; - -struct trident_pcm_bank { - /* register addresses to control bank operations */ - struct trident_pcm_bank_address *addresses; - /* each bank has 32 channels */ - u32 bitmap; /* channel allocation bitmap */ - struct trident_channel channels[32]; -}; - -struct trident_card { - unsigned int magic; - - /* We keep trident cards in a linked list */ - struct trident_card *next; - - /* single open lock mechanism, only used for recording */ - struct mutex open_mutex; - - /* The trident has a certain amount of cross channel interaction - so we use a single per card lock */ - spinlock_t lock; - - /* PCI device stuff */ - struct pci_dev *pci_dev; - u16 pci_id; - u8 revision; - - /* soundcore stuff */ - int dev_audio; - - /* structures for abstraction of hardware facilities, codecs, */ - /* banks and channels */ - struct ac97_codec *ac97_codec[NR_AC97]; - struct trident_pcm_bank banks[NR_BANKS]; - struct trident_state *states[NR_HW_CH]; - - /* hardware resources */ - unsigned long iobase; - u32 irq; - - /* Function support */ - struct trident_channel *(*alloc_pcm_channel) (struct trident_card *); - struct trident_channel *(*alloc_rec_pcm_channel) (struct trident_card *); - void (*free_pcm_channel) (struct trident_card *, unsigned int chan); - void (*address_interrupt) (struct trident_card *); - - /* Added by Matt Wu 01-05-2001 for spdif in */ - int multi_channel_use_count; - int rec_channel_use_count; - u16 mixer_regs[64][NR_AC97]; /* Made card local by Alan */ - int mixer_regs_ready; - - /* Added for hardware volume control */ - int hwvolctl; - struct timer_list timer; - - /* Game port support */ - struct gameport *gameport; -}; - -enum dmabuf_mode { - DM_PLAYBACK = 0, - DM_RECORD -}; - -/* table to map from CHANNELMASK to channel attribute for SiS 7018 */ -static u16 mask2attr[] = { - PCM_LR, PCM_LR, SURR_LR, CENTER_LFE, - HSET, MIC, MODEM_LINE1, MODEM_LINE2, - I2S_LR, SPDIF_LR -}; - -/* table to map from channel attribute to CHANNELMASK for SiS 7018 */ -static int attr2mask[] = { - DSP_BIND_MODEM1, DSP_BIND_MODEM2, DSP_BIND_FRONT, DSP_BIND_HANDSET, - DSP_BIND_I2S, DSP_BIND_CENTER_LFE, DSP_BIND_SURR, DSP_BIND_SPDIF -}; - -/* Added by Matt Wu 01-05-2001 for spdif in */ -static int ali_close_multi_channels(void); -static void ali_delay(struct trident_card *card, int interval); -static void ali_detect_spdif_rate(struct trident_card *card); - -static void ali_ac97_write(struct ac97_codec *codec, u8 reg, u16 val); -static u16 ali_ac97_read(struct ac97_codec *codec, u8 reg); - -static struct trident_card *devs; - -static void trident_ac97_set(struct ac97_codec *codec, u8 reg, u16 val); -static u16 trident_ac97_get(struct ac97_codec *codec, u8 reg); - -static int trident_open_mixdev(struct inode *inode, struct file *file); -static int trident_ioctl_mixdev(struct inode *inode, struct file *file, - unsigned int cmd, unsigned long arg); - -static void ali_ac97_set(struct trident_card *card, int secondary, u8 reg, u16 val); -static u16 ali_ac97_get(struct trident_card *card, int secondary, u8 reg); -static void ali_set_spdif_out_rate(struct trident_card *card, unsigned int rate); -static void ali_enable_special_channel(struct trident_state *stat); -static struct trident_channel *ali_alloc_rec_pcm_channel(struct trident_card *card); -static struct trident_channel *ali_alloc_pcm_channel(struct trident_card *card); -static void ali_free_pcm_channel(struct trident_card *card, unsigned int channel); -static int ali_setup_multi_channels(struct trident_card *card, int chan_nums); -static unsigned int ali_get_spdif_in_rate(struct trident_card *card); -static void ali_setup_spdif_in(struct trident_card *card); -static void ali_disable_spdif_in(struct trident_card *card); -static void ali_disable_special_channel(struct trident_card *card, int ch); -static void ali_setup_spdif_out(struct trident_card *card, int flag); -static int ali_write_5_1(struct trident_state *state, - const char __user *buffer, - int cnt_for_multi_channel, unsigned int *copy_count, - unsigned int *state_cnt); -static int ali_allocate_other_states_resources(struct trident_state *state, - int chan_nums); -static void ali_free_other_states_resources(struct trident_state *state); - -#define seek_offset(dma_ptr, buffer, cnt, offset, copy_count) do { \ - (dma_ptr) += (offset); \ - (buffer) += (offset); \ - (cnt) -= (offset); \ - (copy_count) += (offset); \ -} while (0) - -static inline int lock_set_fmt(struct trident_state* state) -{ - if (test_and_set_bit(0, &state->fmt_flag)) - return -EFAULT; - - return 0; -} - -static inline void unlock_set_fmt(struct trident_state* state) -{ - clear_bit(0, &state->fmt_flag); -} - -static int -trident_enable_loop_interrupts(struct trident_card *card) -{ - u32 global_control; - - global_control = inl(TRID_REG(card, T4D_LFO_GC_CIR)); - - switch (card->pci_id) { - case PCI_DEVICE_ID_SI_7018: - global_control |= (ENDLP_IE | MIDLP_IE | BANK_B_EN); - break; - case PCI_DEVICE_ID_ALI_5451: - case PCI_DEVICE_ID_TRIDENT_4DWAVE_DX: - case PCI_DEVICE_ID_TRIDENT_4DWAVE_NX: - case PCI_DEVICE_ID_INTERG_5050: - global_control |= (ENDLP_IE | MIDLP_IE); - break; - default: - return 0; - } - - outl(global_control, TRID_REG(card, T4D_LFO_GC_CIR)); - - pr_debug("trident: Enable Loop Interrupts, globctl = 0x%08X\n", - inl(TRID_REG(card, T4D_LFO_GC_CIR))); - - return 1; -} - -static int -trident_disable_loop_interrupts(struct trident_card *card) -{ - u32 global_control; - - global_control = inl(TRID_REG(card, T4D_LFO_GC_CIR)); - global_control &= ~(ENDLP_IE | MIDLP_IE); - outl(global_control, TRID_REG(card, T4D_LFO_GC_CIR)); - - pr_debug("trident: Disabled Loop Interrupts, globctl = 0x%08X\n", - global_control); - - return 1; -} - -static void -trident_enable_voice_irq(struct trident_card *card, unsigned int channel) -{ - unsigned int mask = 1 << (channel & 0x1f); - struct trident_pcm_bank *bank = &card->banks[channel >> 5]; - u32 reg, addr = bank->addresses->aint_en; - - reg = inl(TRID_REG(card, addr)); - reg |= mask; - outl(reg, TRID_REG(card, addr)); - -#ifdef DEBUG - reg = inl(TRID_REG(card, addr)); - pr_debug("trident: enabled IRQ on channel %d, %s = 0x%08x(addr:%X)\n", - channel, addr == T4D_AINTEN_B ? "AINTEN_B" : "AINTEN_A", - reg, addr); -#endif /* DEBUG */ -} - -static void -trident_disable_voice_irq(struct trident_card *card, unsigned int channel) -{ - unsigned int mask = 1 << (channel & 0x1f); - struct trident_pcm_bank *bank = &card->banks[channel >> 5]; - u32 reg, addr = bank->addresses->aint_en; - - reg = inl(TRID_REG(card, addr)); - reg &= ~mask; - outl(reg, TRID_REG(card, addr)); - - /* Ack the channel in case the interrupt was set before we disable it. */ - outl(mask, TRID_REG(card, bank->addresses->aint)); - -#ifdef DEBUG - reg = inl(TRID_REG(card, addr)); - pr_debug("trident: disabled IRQ on channel %d, %s = 0x%08x(addr:%X)\n", - channel, addr == T4D_AINTEN_B ? "AINTEN_B" : "AINTEN_A", - reg, addr); -#endif /* DEBUG */ -} - -static void -trident_start_voice(struct trident_card *card, unsigned int channel) -{ - unsigned int mask = 1 << (channel & 0x1f); - struct trident_pcm_bank *bank = &card->banks[channel >> 5]; - u32 addr = bank->addresses->start; - -#ifdef DEBUG - u32 reg; -#endif /* DEBUG */ - - outl(mask, TRID_REG(card, addr)); - -#ifdef DEBUG - reg = inl(TRID_REG(card, addr)); - pr_debug("trident: start voice on channel %d, %s = 0x%08x(addr:%X)\n", - channel, addr == T4D_START_B ? "START_B" : "START_A", - reg, addr); -#endif /* DEBUG */ -} - -static void -trident_stop_voice(struct trident_card *card, unsigned int channel) -{ - unsigned int mask = 1 << (channel & 0x1f); - struct trident_pcm_bank *bank = &card->banks[channel >> 5]; - u32 addr = bank->addresses->stop; - -#ifdef DEBUG - u32 reg; -#endif /* DEBUG */ - - outl(mask, TRID_REG(card, addr)); - -#ifdef DEBUG - reg = inl(TRID_REG(card, addr)); - pr_debug("trident: stop voice on channel %d, %s = 0x%08x(addr:%X)\n", - channel, addr == T4D_STOP_B ? "STOP_B" : "STOP_A", - reg, addr); -#endif /* DEBUG */ -} - -static u32 -trident_get_interrupt_mask(struct trident_card *card, unsigned int channel) -{ - struct trident_pcm_bank *bank = &card->banks[channel]; - u32 addr = bank->addresses->aint; - return inl(TRID_REG(card, addr)); -} - -static int -trident_check_channel_interrupt(struct trident_card *card, unsigned int channel) -{ - unsigned int mask = 1 << (channel & 0x1f); - u32 reg = trident_get_interrupt_mask(card, channel >> 5); - -#ifdef DEBUG - if (reg & mask) - pr_debug("trident: channel %d has interrupt, %s = 0x%08x\n", - channel, reg == T4D_AINT_B ? "AINT_B" : "AINT_A", - reg); -#endif /* DEBUG */ - return (reg & mask) ? 1 : 0; -} - -static void -trident_ack_channel_interrupt(struct trident_card *card, unsigned int channel) -{ - unsigned int mask = 1 << (channel & 0x1f); - struct trident_pcm_bank *bank = &card->banks[channel >> 5]; - u32 reg, addr = bank->addresses->aint; - - reg = inl(TRID_REG(card, addr)); - reg &= mask; - outl(reg, TRID_REG(card, addr)); - -#ifdef DEBUG - reg = inl(TRID_REG(card, T4D_AINT_B)); - pr_debug("trident: Ack channel %d interrupt, AINT_B = 0x%08x\n", - channel, reg); -#endif /* DEBUG */ -} - -static struct trident_channel * -trident_alloc_pcm_channel(struct trident_card *card) -{ - struct trident_pcm_bank *bank; - int idx; - - bank = &card->banks[BANK_B]; - - for (idx = 31; idx >= 0; idx--) { - if (!(bank->bitmap & (1 << idx))) { - struct trident_channel *channel = &bank->channels[idx]; - bank->bitmap |= 1 << idx; - channel->num = idx + 32; - return channel; - } - } - - /* no more free channels available */ - printk(KERN_ERR "trident: no more channels available on Bank B.\n"); - return NULL; -} - -static void -trident_free_pcm_channel(struct trident_card *card, unsigned int channel) -{ - int bank; - unsigned char b; - - if (channel < 31 || channel > 63) - return; - - if (card->pci_id == PCI_DEVICE_ID_TRIDENT_4DWAVE_DX || - card->pci_id == PCI_DEVICE_ID_TRIDENT_4DWAVE_NX) { - b = inb(TRID_REG(card, T4D_REC_CH)); - if ((b & ~0x80) == channel) - outb(0x0, TRID_REG(card, T4D_REC_CH)); - } - - bank = channel >> 5; - channel = channel & 0x1f; - - card->banks[bank].bitmap &= ~(1 << (channel)); -} - -static struct trident_channel * -cyber_alloc_pcm_channel(struct trident_card *card) -{ - struct trident_pcm_bank *bank; - int idx; - - /* The cyberpro 5050 has only 32 voices and one bank */ - /* .. at least they are not documented (if you want to call that - * crap documentation), perhaps broken ? */ - - bank = &card->banks[BANK_A]; - - for (idx = 31; idx >= 0; idx--) { - if (!(bank->bitmap & (1 << idx))) { - struct trident_channel *channel = &bank->channels[idx]; - bank->bitmap |= 1 << idx; - channel->num = idx; - return channel; - } - } - - /* no more free channels available */ - printk(KERN_ERR "cyberpro5050: no more channels available on Bank A.\n"); - return NULL; -} - -static void -cyber_free_pcm_channel(struct trident_card *card, unsigned int channel) -{ - if (channel > 31) - return; - card->banks[BANK_A].bitmap &= ~(1 << (channel)); -} - -static inline void -cyber_outidx(int port, int idx, int data) -{ - outb(idx, port); - outb(data, port + 1); -} - -static inline int -cyber_inidx(int port, int idx) -{ - outb(idx, port); - return inb(port + 1); -} - -static int -cyber_init_ritual(struct trident_card *card) -{ - /* some black magic, taken from SDK samples */ - /* remove this and nothing will work */ - int portDat; - int ret = 0; - unsigned long flags; - - /* - * Keep interrupts off for the configure - we don't want to - * clash with another cyberpro config event - */ - - spin_lock_irqsave(&card->lock, flags); - portDat = cyber_inidx(CYBER_PORT_AUDIO, CYBER_IDX_AUDIO_ENABLE); - /* enable, if it was disabled */ - if ((portDat & CYBER_BMSK_AUENZ) != CYBER_BMSK_AUENZ_ENABLE) { - printk(KERN_INFO "cyberpro5050: enabling audio controller\n"); - cyber_outidx(CYBER_PORT_AUDIO, CYBER_IDX_AUDIO_ENABLE, - portDat | CYBER_BMSK_AUENZ_ENABLE); - /* check again if hardware is enabled now */ - portDat = cyber_inidx(CYBER_PORT_AUDIO, CYBER_IDX_AUDIO_ENABLE); - } - if ((portDat & CYBER_BMSK_AUENZ) != CYBER_BMSK_AUENZ_ENABLE) { - printk(KERN_ERR "cyberpro5050: initAudioAccess: no success\n"); - ret = -1; - } else { - cyber_outidx(CYBER_PORT_AUDIO, CYBER_IDX_IRQ_ENABLE, - CYBER_BMSK_AUDIO_INT_ENABLE); - cyber_outidx(CYBER_PORT_AUDIO, 0xbf, 0x01); - cyber_outidx(CYBER_PORT_AUDIO, 0xba, 0x20); - cyber_outidx(CYBER_PORT_AUDIO, 0xbb, 0x08); - cyber_outidx(CYBER_PORT_AUDIO, 0xbf, 0x02); - cyber_outidx(CYBER_PORT_AUDIO, 0xb3, 0x06); - cyber_outidx(CYBER_PORT_AUDIO, 0xbf, 0x00); - } - spin_unlock_irqrestore(&card->lock, flags); - return ret; -} - -/* called with spin lock held */ - -static int -trident_load_channel_registers(struct trident_card *card, u32 * data, - unsigned int channel) -{ - int i; - - if (channel > 63) - return 0; - - /* select hardware channel to write */ - outb(channel, TRID_REG(card, T4D_LFO_GC_CIR)); - - /* Output the channel registers, but don't write register - three to an ALI chip. */ - for (i = 0; i < CHANNEL_REGS; i++) { - if (i == 3 && card->pci_id == PCI_DEVICE_ID_ALI_5451) - continue; - outl(data[i], TRID_REG(card, CHANNEL_START + 4 * i)); - } - if (card->pci_id == PCI_DEVICE_ID_ALI_5451 || - card->pci_id == PCI_DEVICE_ID_INTERG_5050) { - outl(ALI_EMOD_Still, TRID_REG(card, ALI_EBUF1)); - outl(ALI_EMOD_Still, TRID_REG(card, ALI_EBUF2)); - } - return 1; -} - -/* called with spin lock held */ -static int -trident_write_voice_regs(struct trident_state *state) -{ - unsigned int data[CHANNEL_REGS + 1]; - struct trident_channel *channel; - - channel = state->dmabuf.channel; - - data[1] = channel->lba; - data[4] = channel->control; - - switch (state->card->pci_id) { - case PCI_DEVICE_ID_ALI_5451: - data[0] = 0; /* Current Sample Offset */ - data[2] = (channel->eso << 16) | (channel->delta & 0xffff); - data[3] = 0; - break; - case PCI_DEVICE_ID_SI_7018: - case PCI_DEVICE_ID_INTERG_5050: - data[0] = 0; /* Current Sample Offset */ - data[2] = (channel->eso << 16) | (channel->delta & 0xffff); - data[3] = (channel->attribute << 16) | (channel->fm_vol & 0xffff); - break; - case PCI_DEVICE_ID_TRIDENT_4DWAVE_DX: - data[0] = 0; /* Current Sample Offset */ - data[2] = (channel->eso << 16) | (channel->delta & 0xffff); - data[3] = channel->fm_vol & 0xffff; - break; - case PCI_DEVICE_ID_TRIDENT_4DWAVE_NX: - data[0] = (channel->delta << 24); - data[2] = ((channel->delta << 16) & 0xff000000) | - (channel->eso & 0x00ffffff); - data[3] = channel->fm_vol & 0xffff; - break; - default: - return 0; - } - - return trident_load_channel_registers(state->card, data, channel->num); -} - -static int -compute_rate_play(u32 rate) -{ - int delta; - /* We special case 44100 and 8000 since rounding with the equation - does not give us an accurate enough value. For 11025 and 22050 - the equation gives us the best answer. All other frequencies will - also use the equation. JDW */ - if (rate == 44100) - delta = 0xeb3; - else if (rate == 8000) - delta = 0x2ab; - else if (rate == 48000) - delta = 0x1000; - else - delta = (((rate << 12) + rate) / 48000) & 0x0000ffff; - return delta; -} - -static int -compute_rate_rec(u32 rate) -{ - int delta; - - if (rate == 44100) - delta = 0x116a; - else if (rate == 8000) - delta = 0x6000; - else if (rate == 48000) - delta = 0x1000; - else - delta = ((48000 << 12) / rate) & 0x0000ffff; - - return delta; -} - -/* set playback sample rate */ -static unsigned int -trident_set_dac_rate(struct trident_state *state, unsigned int rate) -{ - struct dmabuf *dmabuf = &state->dmabuf; - - if (rate > 48000) - rate = 48000; - if (rate < 4000) - rate = 4000; - - dmabuf->rate = rate; - dmabuf->channel->delta = compute_rate_play(rate); - - trident_write_voice_regs(state); - - pr_debug("trident: called trident_set_dac_rate : rate = %d\n", rate); - - return rate; -} - -/* set recording sample rate */ -static unsigned int -trident_set_adc_rate(struct trident_state *state, unsigned int rate) -{ - struct dmabuf *dmabuf = &state->dmabuf; - - if (rate > 48000) - rate = 48000; - if (rate < 4000) - rate = 4000; - - dmabuf->rate = rate; - dmabuf->channel->delta = compute_rate_rec(rate); - - trident_write_voice_regs(state); - - pr_debug("trident: called trident_set_adc_rate : rate = %d\n", rate); - - return rate; -} - -/* prepare channel attributes for playback */ -static void -trident_play_setup(struct trident_state *state) -{ - struct dmabuf *dmabuf = &state->dmabuf; - struct trident_channel *channel = dmabuf->channel; - - channel->lba = dmabuf->dma_handle; - channel->delta = compute_rate_play(dmabuf->rate); - - channel->eso = dmabuf->dmasize >> sample_shift[dmabuf->fmt]; - channel->eso -= 1; - - if (state->card->pci_id != PCI_DEVICE_ID_SI_7018) { - channel->attribute = 0; - if (state->card->pci_id == PCI_DEVICE_ID_ALI_5451) { - if ((channel->num == ALI_SPDIF_IN_CHANNEL) || - (channel->num == ALI_PCM_IN_CHANNEL)) - ali_disable_special_channel(state->card, channel->num); - else if ((inl(TRID_REG(state->card, ALI_GLOBAL_CONTROL)) - & ALI_SPDIF_OUT_CH_ENABLE) - && (channel->num == ALI_SPDIF_OUT_CHANNEL)) { - ali_set_spdif_out_rate(state->card, - state->dmabuf.rate); - state->dmabuf.channel->delta = 0x1000; - } - } - } - - channel->fm_vol = 0x0; - - channel->control = CHANNEL_LOOP; - if (dmabuf->fmt & TRIDENT_FMT_16BIT) { - /* 16-bits */ - channel->control |= CHANNEL_16BITS; - /* signed */ - channel->control |= CHANNEL_SIGNED; - } - if (dmabuf->fmt & TRIDENT_FMT_STEREO) - /* stereo */ - channel->control |= CHANNEL_STEREO; - - pr_debug("trident: trident_play_setup, LBA = 0x%08x, Delta = 0x%08x, " - "ESO = 0x%08x, Control = 0x%08x\n", channel->lba, - channel->delta, channel->eso, channel->control); - - trident_write_voice_regs(state); -} - -/* prepare channel attributes for recording */ -static void -trident_rec_setup(struct trident_state *state) -{ - u16 w; - u8 bval; - - struct trident_card *card = state->card; - struct dmabuf *dmabuf = &state->dmabuf; - struct trident_channel *channel = dmabuf->channel; - unsigned int rate; - - /* Enable AC-97 ADC (capture) */ - switch (card->pci_id) { - case PCI_DEVICE_ID_ALI_5451: - ali_enable_special_channel(state); - break; - case PCI_DEVICE_ID_SI_7018: - /* for 7018, the ac97 is always in playback/record (duplex) mode */ - break; - case PCI_DEVICE_ID_TRIDENT_4DWAVE_DX: - w = inb(TRID_REG(card, DX_ACR2_AC97_COM_STAT)); - outb(w | 0x48, TRID_REG(card, DX_ACR2_AC97_COM_STAT)); - /* enable and set record channel */ - outb(0x80 | channel->num, TRID_REG(card, T4D_REC_CH)); - break; - case PCI_DEVICE_ID_TRIDENT_4DWAVE_NX: - w = inw(TRID_REG(card, T4D_MISCINT)); - outw(w | 0x1000, TRID_REG(card, T4D_MISCINT)); - /* enable and set record channel */ - outb(0x80 | channel->num, TRID_REG(card, T4D_REC_CH)); - break; - case PCI_DEVICE_ID_INTERG_5050: - /* don't know yet, using special channel 22 in GC1(0xd4)? */ - break; - default: - return; - } - - channel->lba = dmabuf->dma_handle; - channel->delta = compute_rate_rec(dmabuf->rate); - if ((card->pci_id == PCI_DEVICE_ID_ALI_5451) && - (channel->num == ALI_SPDIF_IN_CHANNEL)) { - rate = ali_get_spdif_in_rate(card); - if (rate == 0) { - printk(KERN_WARNING "trident: ALi 5451 " - "S/PDIF input setup error!\n"); - rate = 48000; - } - bval = inb(TRID_REG(card, ALI_SPDIF_CTRL)); - if (bval & 0x10) { - outb(bval, TRID_REG(card, ALI_SPDIF_CTRL)); - printk(KERN_WARNING "trident: cleared ALi " - "5451 S/PDIF parity error flag.\n"); - } - - if (rate != 48000) - channel->delta = ((rate << 12) / dmabuf->rate) & 0x0000ffff; - } - - channel->eso = dmabuf->dmasize >> sample_shift[dmabuf->fmt]; - channel->eso -= 1; - - if (state->card->pci_id != PCI_DEVICE_ID_SI_7018) { - channel->attribute = 0; - } - - channel->fm_vol = 0x0; - - channel->control = CHANNEL_LOOP; - if (dmabuf->fmt & TRIDENT_FMT_16BIT) { - /* 16-bits */ - channel->control |= CHANNEL_16BITS; - /* signed */ - channel->control |= CHANNEL_SIGNED; - } - if (dmabuf->fmt & TRIDENT_FMT_STEREO) - /* stereo */ - channel->control |= CHANNEL_STEREO; - - pr_debug("trident: trident_rec_setup, LBA = 0x%08x, Delat = 0x%08x, " - "ESO = 0x%08x, Control = 0x%08x\n", channel->lba, - channel->delta, channel->eso, channel->control); - - trident_write_voice_regs(state); -} - -/* get current playback/recording dma buffer pointer (byte offset from LBA), - called with spinlock held! */ -static inline unsigned -trident_get_dma_addr(struct trident_state *state) -{ - struct dmabuf *dmabuf = &state->dmabuf; - u32 cso; - - if (!dmabuf->enable) - return 0; - - outb(dmabuf->channel->num, TRID_REG(state->card, T4D_LFO_GC_CIR)); - - switch (state->card->pci_id) { - case PCI_DEVICE_ID_ALI_5451: - case PCI_DEVICE_ID_SI_7018: - case PCI_DEVICE_ID_TRIDENT_4DWAVE_DX: - case PCI_DEVICE_ID_INTERG_5050: - /* 16 bits ESO, CSO for 7018 and DX */ - cso = inw(TRID_REG(state->card, CH_DX_CSO_ALPHA_FMS + 2)); - break; - case PCI_DEVICE_ID_TRIDENT_4DWAVE_NX: - /* 24 bits ESO, CSO for NX */ - cso = inl(TRID_REG(state->card, CH_NX_DELTA_CSO)) & 0x00ffffff; - break; - default: - return 0; - } - - pr_debug("trident: trident_get_dma_addr: chip reported channel: %d, " - "cso = 0x%04x\n", dmabuf->channel->num, cso); - - /* ESO and CSO are in units of Samples, convert to byte offset */ - cso <<= sample_shift[dmabuf->fmt]; - - return (cso % dmabuf->dmasize); -} - -/* Stop recording (lock held) */ -static inline void -__stop_adc(struct trident_state *state) -{ - struct dmabuf *dmabuf = &state->dmabuf; - unsigned int chan_num = dmabuf->channel->num; - struct trident_card *card = state->card; - - dmabuf->enable &= ~ADC_RUNNING; - trident_stop_voice(card, chan_num); - trident_disable_voice_irq(card, chan_num); -} - -static void -stop_adc(struct trident_state *state) -{ - struct trident_card *card = state->card; - unsigned long flags; - - spin_lock_irqsave(&card->lock, flags); - __stop_adc(state); - spin_unlock_irqrestore(&card->lock, flags); -} - -static void -start_adc(struct trident_state *state) -{ - struct dmabuf *dmabuf = &state->dmabuf; - unsigned int chan_num = dmabuf->channel->num; - struct trident_card *card = state->card; - unsigned long flags; - - spin_lock_irqsave(&card->lock, flags); - if ((dmabuf->mapped || - dmabuf->count < (signed) dmabuf->dmasize) && - dmabuf->ready) { - dmabuf->enable |= ADC_RUNNING; - trident_enable_voice_irq(card, chan_num); - trident_start_voice(card, chan_num); - } - spin_unlock_irqrestore(&card->lock, flags); -} - -/* stop playback (lock held) */ -static inline void -__stop_dac(struct trident_state *state) -{ - struct dmabuf *dmabuf = &state->dmabuf; - unsigned int chan_num = dmabuf->channel->num; - struct trident_card *card = state->card; - - dmabuf->enable &= ~DAC_RUNNING; - trident_stop_voice(card, chan_num); - if (state->chans_num == 6) { - trident_stop_voice(card, state->other_states[0]-> - dmabuf.channel->num); - trident_stop_voice(card, state->other_states[1]-> - dmabuf.channel->num); - trident_stop_voice(card, state->other_states[2]-> - dmabuf.channel->num); - trident_stop_voice(card, state->other_states[3]-> - dmabuf.channel->num); - } - trident_disable_voice_irq(card, chan_num); -} - -static void -stop_dac(struct trident_state *state) -{ - struct trident_card *card = state->card; - unsigned long flags; - - spin_lock_irqsave(&card->lock, flags); - __stop_dac(state); - spin_unlock_irqrestore(&card->lock, flags); -} - -static void -start_dac(struct trident_state *state) -{ - struct dmabuf *dmabuf = &state->dmabuf; - unsigned int chan_num = dmabuf->channel->num; - struct trident_card *card = state->card; - unsigned long flags; - - spin_lock_irqsave(&card->lock, flags); - if ((dmabuf->mapped || dmabuf->count > 0) && dmabuf->ready) { - dmabuf->enable |= DAC_RUNNING; - trident_enable_voice_irq(card, chan_num); - trident_start_voice(card, chan_num); - if (state->chans_num == 6) { - trident_start_voice(card, state->other_states[0]-> - dmabuf.channel->num); - trident_start_voice(card, state->other_states[1]-> - dmabuf.channel->num); - trident_start_voice(card, state->other_states[2]-> - dmabuf.channel->num); - trident_start_voice(card, state->other_states[3]-> - dmabuf.channel->num); - } - } - spin_unlock_irqrestore(&card->lock, flags); -} - -#define DMABUF_DEFAULTORDER (15-PAGE_SHIFT) -#define DMABUF_MINORDER 1 - -/* alloc a DMA buffer of with a buffer of this order */ -static int -alloc_dmabuf(struct dmabuf *dmabuf, struct pci_dev *pci_dev, int order) -{ - void *rawbuf = NULL; - struct page *page, *pend; - - if (!(rawbuf = pci_alloc_consistent(pci_dev, PAGE_SIZE << order, - &dmabuf->dma_handle))) - return -ENOMEM; - - pr_debug("trident: allocated %ld (order = %d) bytes at %p\n", - PAGE_SIZE << order, order, rawbuf); - - dmabuf->ready = dmabuf->mapped = 0; - dmabuf->rawbuf = rawbuf; - dmabuf->buforder = order; - - /* now mark the pages as reserved; otherwise */ - /* remap_pfn_range doesn't do what we want */ - pend = virt_to_page(rawbuf + (PAGE_SIZE << order) - 1); - for (page = virt_to_page(rawbuf); page <= pend; page++) - SetPageReserved(page); - - return 0; -} - -/* allocate the main DMA buffer, playback and recording buffer should be */ -/* allocated separately */ -static int -alloc_main_dmabuf(struct trident_state *state) -{ - struct dmabuf *dmabuf = &state->dmabuf; - int order; - int ret = -ENOMEM; - - /* alloc as big a chunk as we can, FIXME: is this necessary ?? */ - for (order = DMABUF_DEFAULTORDER; order >= DMABUF_MINORDER; order--) { - if (!(ret = alloc_dmabuf(dmabuf, state->card->pci_dev, order))) - return 0; - /* else try again */ - } - return ret; -} - -/* deallocate a DMA buffer */ -static void -dealloc_dmabuf(struct dmabuf *dmabuf, struct pci_dev *pci_dev) -{ - struct page *page, *pend; - - if (dmabuf->rawbuf) { - /* undo marking the pages as reserved */ - pend = virt_to_page(dmabuf->rawbuf + (PAGE_SIZE << dmabuf->buforder) - 1); - for (page = virt_to_page(dmabuf->rawbuf); page <= pend; page++) - ClearPageReserved(page); - pci_free_consistent(pci_dev, PAGE_SIZE << dmabuf->buforder, - dmabuf->rawbuf, dmabuf->dma_handle); - dmabuf->rawbuf = NULL; - } - dmabuf->mapped = dmabuf->ready = 0; -} - -static int -prog_dmabuf(struct trident_state *state, enum dmabuf_mode rec) -{ - struct dmabuf *dmabuf = &state->dmabuf; - unsigned bytepersec; - struct trident_state *s = state; - unsigned bufsize, dma_nums; - unsigned long flags; - int ret, i, order; - - if ((ret = lock_set_fmt(state)) < 0) - return ret; - - if (state->chans_num == 6) - dma_nums = 5; - else - dma_nums = 1; - - for (i = 0; i < dma_nums; i++) { - if (i > 0) { - s = state->other_states[i - 1]; - dmabuf = &s->dmabuf; - dmabuf->fmt = state->dmabuf.fmt; - dmabuf->rate = state->dmabuf.rate; - } - - spin_lock_irqsave(&s->card->lock, flags); - dmabuf->hwptr = dmabuf->swptr = dmabuf->total_bytes = 0; - dmabuf->count = dmabuf->error = 0; - spin_unlock_irqrestore(&s->card->lock, flags); - - /* allocate DMA buffer if not allocated yet */ - if (!dmabuf->rawbuf) { - if (i == 0) { - if ((ret = alloc_main_dmabuf(state))) { - unlock_set_fmt(state); - return ret; - } - } else { - ret = -ENOMEM; - order = state->dmabuf.buforder - 1; - if (order >= DMABUF_MINORDER) { - ret = alloc_dmabuf(dmabuf, - state->card->pci_dev, - order); - } - if (ret) { - /* release the main DMA buffer */ - dealloc_dmabuf(&state->dmabuf, state->card->pci_dev); - /* release the auxiliary DMA buffers */ - for (i -= 2; i >= 0; i--) - dealloc_dmabuf(&state->other_states[i]->dmabuf, - state->card->pci_dev); - unlock_set_fmt(state); - return ret; - } - } - } - /* FIXME: figure out all this OSS fragment stuff */ - bytepersec = dmabuf->rate << sample_shift[dmabuf->fmt]; - bufsize = PAGE_SIZE << dmabuf->buforder; - if (dmabuf->ossfragshift) { - if ((1000 << dmabuf->ossfragshift) < bytepersec) - dmabuf->fragshift = ld2(bytepersec / 1000); - else - dmabuf->fragshift = dmabuf->ossfragshift; - } else { - /* lets hand out reasonable big ass buffers by default */ - dmabuf->fragshift = (dmabuf->buforder + PAGE_SHIFT - 2); - } - dmabuf->numfrag = bufsize >> dmabuf->fragshift; - while (dmabuf->numfrag < 4 && dmabuf->fragshift > 3) { - dmabuf->fragshift--; - dmabuf->numfrag = bufsize >> dmabuf->fragshift; - } - dmabuf->fragsize = 1 << dmabuf->fragshift; - if (dmabuf->ossmaxfrags >= 4 && dmabuf->ossmaxfrags < dmabuf->numfrag) - dmabuf->numfrag = dmabuf->ossmaxfrags; - dmabuf->fragsamples = dmabuf->fragsize >> sample_shift[dmabuf->fmt]; - dmabuf->dmasize = dmabuf->numfrag << dmabuf->fragshift; - - memset(dmabuf->rawbuf, (dmabuf->fmt & TRIDENT_FMT_16BIT) ? 0 : 0x80, - dmabuf->dmasize); - - spin_lock_irqsave(&s->card->lock, flags); - if (rec == DM_RECORD) - trident_rec_setup(s); - else /* DM_PLAYBACK */ - trident_play_setup(s); - - spin_unlock_irqrestore(&s->card->lock, flags); - - /* set the ready flag for the dma buffer */ - dmabuf->ready = 1; - - pr_debug("trident: prog_dmabuf(%d), sample rate = %d, " - "format = %d, numfrag = %d, fragsize = %d " - "dmasize = %d\n", dmabuf->channel->num, - dmabuf->rate, dmabuf->fmt, dmabuf->numfrag, - dmabuf->fragsize, dmabuf->dmasize); - } - unlock_set_fmt(state); - return 0; -} - - -static inline int prog_dmabuf_record(struct trident_state* state) -{ - return prog_dmabuf(state, DM_RECORD); -} - -static inline int prog_dmabuf_playback(struct trident_state* state) -{ - return prog_dmabuf(state, DM_PLAYBACK); -} - -/* we are doing quantum mechanics here, the buffer can only be empty, half or full filled i.e. - |------------|------------| or |xxxxxxxxxxxx|------------| or |xxxxxxxxxxxx|xxxxxxxxxxxx| - but we almost always get this - |xxxxxx------|------------| or |xxxxxxxxxxxx|xxxxx-------| - so we have to clear the tail space to "silence" - |xxxxxx000000|------------| or |xxxxxxxxxxxx|xxxxxx000000| -*/ -static void -trident_clear_tail(struct trident_state *state) -{ - struct dmabuf *dmabuf = &state->dmabuf; - unsigned swptr; - unsigned char silence = (dmabuf->fmt & TRIDENT_FMT_16BIT) ? 0 : 0x80; - unsigned int len; - unsigned long flags; - - spin_lock_irqsave(&state->card->lock, flags); - swptr = dmabuf->swptr; - spin_unlock_irqrestore(&state->card->lock, flags); - - if (swptr == 0 || swptr == dmabuf->dmasize / 2 || - swptr == dmabuf->dmasize) - return; - - if (swptr < dmabuf->dmasize / 2) - len = dmabuf->dmasize / 2 - swptr; - else - len = dmabuf->dmasize - swptr; - - memset(dmabuf->rawbuf + swptr, silence, len); - if (state->card->pci_id != PCI_DEVICE_ID_ALI_5451) { - spin_lock_irqsave(&state->card->lock, flags); - dmabuf->swptr += len; - dmabuf->count += len; - spin_unlock_irqrestore(&state->card->lock, flags); - } - - /* restart the dma machine in case it is halted */ - start_dac(state); -} - -static int -drain_dac(struct trident_state *state, int nonblock) -{ - DECLARE_WAITQUEUE(wait, current); - struct dmabuf *dmabuf = &state->dmabuf; - unsigned long flags; - unsigned long tmo; - int count; - unsigned long diff = 0; - - if (dmabuf->mapped || !dmabuf->ready) - return 0; - - add_wait_queue(&dmabuf->wait, &wait); - for (;;) { - /* It seems that we have to set the current state to TASK_INTERRUPTIBLE - every time to make the process really go to sleep */ - set_current_state(TASK_INTERRUPTIBLE); - - spin_lock_irqsave(&state->card->lock, flags); - count = dmabuf->count; - spin_unlock_irqrestore(&state->card->lock, flags); - - if (count <= 0) - break; - - if (signal_pending(current)) - break; - - if (nonblock) { - remove_wait_queue(&dmabuf->wait, &wait); - set_current_state(TASK_RUNNING); - return -EBUSY; - } - - /* No matter how much data is left in the buffer, we have to wait until - CSO == ESO/2 or CSO == ESO when address engine interrupts */ - if (state->card->pci_id == PCI_DEVICE_ID_ALI_5451 || - state->card->pci_id == PCI_DEVICE_ID_INTERG_5050) { - diff = dmabuf->swptr - trident_get_dma_addr(state) + dmabuf->dmasize; - diff = diff % (dmabuf->dmasize); - tmo = (diff * HZ) / dmabuf->rate; - } else { - tmo = (dmabuf->dmasize * HZ) / dmabuf->rate; - } - tmo >>= sample_shift[dmabuf->fmt]; - if (!schedule_timeout(tmo ? tmo : 1) && tmo) { - break; - } - } - remove_wait_queue(&dmabuf->wait, &wait); - set_current_state(TASK_RUNNING); - if (signal_pending(current)) - return -ERESTARTSYS; - - return 0; -} - -/* update buffer manangement pointers, especially, */ -/* dmabuf->count and dmabuf->hwptr */ -static void -trident_update_ptr(struct trident_state *state) -{ - struct dmabuf *dmabuf = &state->dmabuf; - unsigned hwptr, swptr; - int clear_cnt = 0; - int diff; - unsigned char silence; - unsigned half_dmasize; - - /* update hardware pointer */ - hwptr = trident_get_dma_addr(state); - diff = (dmabuf->dmasize + hwptr - dmabuf->hwptr) % dmabuf->dmasize; - dmabuf->hwptr = hwptr; - dmabuf->total_bytes += diff; - - /* error handling and process wake up for ADC */ - if (dmabuf->enable == ADC_RUNNING) { - if (dmabuf->mapped) { - dmabuf->count -= diff; - if (dmabuf->count >= (signed) dmabuf->fragsize) - wake_up(&dmabuf->wait); - } else { - dmabuf->count += diff; - - if (dmabuf->count < 0 || - dmabuf->count > dmabuf->dmasize) { - /* buffer underrun or buffer overrun, */ - /* we have no way to recover it here, just */ - /* stop the machine and let the process */ - /* force hwptr and swptr to sync */ - __stop_adc(state); - dmabuf->error++; - } - if (dmabuf->count < (signed) dmabuf->dmasize / 2) - wake_up(&dmabuf->wait); - } - } - - /* error handling and process wake up for DAC */ - if (dmabuf->enable == DAC_RUNNING) { - if (dmabuf->mapped) { - dmabuf->count += diff; - if (dmabuf->count >= (signed) dmabuf->fragsize) - wake_up(&dmabuf->wait); - } else { - dmabuf->count -= diff; - - if (dmabuf->count < 0 || - dmabuf->count > dmabuf->dmasize) { - /* buffer underrun or buffer overrun, we have no way to recover - it here, just stop the machine and let the process force hwptr - and swptr to sync */ - __stop_dac(state); - dmabuf->error++; - } else if (!dmabuf->endcleared) { - swptr = dmabuf->swptr; - silence = (dmabuf->fmt & TRIDENT_FMT_16BIT ? 0 : 0x80); - if (dmabuf->update_flag & ALI_ADDRESS_INT_UPDATE) { - /* We must clear end data of 1/2 dmabuf if needed. - According to 1/2 algorithm of Address Engine Interrupt, - check the validation of the data of half dmasize. */ - half_dmasize = dmabuf->dmasize / 2; - if ((diff = hwptr - half_dmasize) < 0) - diff = hwptr; - if ((dmabuf->count + diff) < half_dmasize) { - //there is invalid data in the end of half buffer - if ((clear_cnt = half_dmasize - swptr) < 0) - clear_cnt += half_dmasize; - //clear the invalid data - memset(dmabuf->rawbuf + swptr, silence, clear_cnt); - if (state->chans_num == 6) { - clear_cnt = clear_cnt / 2; - swptr = swptr / 2; - memset(state->other_states[0]->dmabuf.rawbuf + swptr, - silence, clear_cnt); - memset(state->other_states[1]->dmabuf.rawbuf + swptr, - silence, clear_cnt); - memset(state->other_states[2]->dmabuf.rawbuf + swptr, - silence, clear_cnt); - memset(state->other_states[3]->dmabuf.rawbuf + swptr, - silence, clear_cnt); - } - dmabuf->endcleared = 1; - } - } else if (dmabuf->count < (signed) dmabuf->fragsize) { - clear_cnt = dmabuf->fragsize; - if ((swptr + clear_cnt) > dmabuf->dmasize) - clear_cnt = dmabuf->dmasize - swptr; - memset(dmabuf->rawbuf + swptr, silence, clear_cnt); - if (state->chans_num == 6) { - clear_cnt = clear_cnt / 2; - swptr = swptr / 2; - memset(state->other_states[0]->dmabuf.rawbuf + swptr, - silence, clear_cnt); - memset(state->other_states[1]->dmabuf.rawbuf + swptr, - silence, clear_cnt); - memset(state->other_states[2]->dmabuf.rawbuf + swptr, - silence, clear_cnt); - memset(state->other_states[3]->dmabuf.rawbuf + swptr, - silence, clear_cnt); - } - dmabuf->endcleared = 1; - } - } - /* trident_update_ptr is called by interrupt handler or by process via - ioctl/poll, we only wake up the waiting process when we have more - than 1/2 buffer free (always true for interrupt handler) */ - if (dmabuf->count < (signed) dmabuf->dmasize / 2) - wake_up(&dmabuf->wait); - } - } - dmabuf->update_flag &= ~ALI_ADDRESS_INT_UPDATE; -} - -static void -trident_address_interrupt(struct trident_card *card) -{ - int i; - struct trident_state *state; - unsigned int channel; - - /* Update the pointers for all channels we are running. */ - /* FIXME: should read interrupt status only once */ - for (i = 0; i < NR_HW_CH; i++) { - channel = 63 - i; - if (trident_check_channel_interrupt(card, channel)) { - trident_ack_channel_interrupt(card, channel); - if ((state = card->states[i]) != NULL) { - trident_update_ptr(state); - } else { - printk(KERN_WARNING "trident: spurious channel " - "irq %d.\n", channel); - trident_stop_voice(card, channel); - trident_disable_voice_irq(card, channel); - } - } - } -} - -static void -ali_hwvol_control(struct trident_card *card, int opt) -{ - u16 dwTemp, volume[2], mute, diff, *pVol[2]; - - dwTemp = ali_ac97_read(card->ac97_codec[0], 0x02); - mute = dwTemp & 0x8000; - volume[0] = dwTemp & 0x001f; - volume[1] = (dwTemp & 0x1f00) >> 8; - if (volume[0] < volume[1]) { - pVol[0] = &volume[0]; - pVol[1] = &volume[1]; - } else { - pVol[1] = &volume[0]; - pVol[0] = &volume[1]; - } - diff = *(pVol[1]) - *(pVol[0]); - - if (opt == 1) { // MUTE - dwTemp ^= 0x8000; - ali_ac97_write(card->ac97_codec[0], - 0x02, dwTemp); - } else if (opt == 2) { // Down - if (mute) - return; - if (*(pVol[1]) < 0x001f) { - (*pVol[1])++; - *(pVol[0]) = *(pVol[1]) - diff; - } - dwTemp &= 0xe0e0; - dwTemp |= (volume[0]) | (volume[1] << 8); - ali_ac97_write(card->ac97_codec[0], 0x02, dwTemp); - card->ac97_codec[0]->mixer_state[0] = ((32 - volume[0]) * 25 / 8) | - (((32 - volume[1]) * 25 / 8) << 8); - } else if (opt == 4) { // Up - if (mute) - return; - if (*(pVol[0]) > 0) { - (*pVol[0])--; - *(pVol[1]) = *(pVol[0]) + diff; - } - dwTemp &= 0xe0e0; - dwTemp |= (volume[0]) | (volume[1] << 8); - ali_ac97_write(card->ac97_codec[0], 0x02, dwTemp); - card->ac97_codec[0]->mixer_state[0] = ((32 - volume[0]) * 25 / 8) | - (((32 - volume[1]) * 25 / 8) << 8); - } else { - /* Nothing needs doing */ - } -} - -/* - * Re-enable reporting of vol change after 0.1 seconds - */ - -static void -ali_timeout(unsigned long ptr) -{ - struct trident_card *card = (struct trident_card *) ptr; - u16 temp = 0; - - /* Enable GPIO IRQ (MISCINT bit 18h) */ - temp = inw(TRID_REG(card, T4D_MISCINT + 2)); - temp |= 0x0004; - outw(temp, TRID_REG(card, T4D_MISCINT + 2)); -} - -/* - * Set up the timer to clear the vol change notification - */ - -static void -ali_set_timer(struct trident_card *card) -{ - /* Add Timer Routine to Enable GPIO IRQ */ - del_timer(&card->timer); /* Never queue twice */ - card->timer.function = ali_timeout; - card->timer.data = (unsigned long) card; - card->timer.expires = jiffies + HZ / 10; - add_timer(&card->timer); -} - -/* - * Process a GPIO event - */ - -static void -ali_queue_task(struct trident_card *card, int opt) -{ - u16 temp; - - /* Disable GPIO IRQ (MISCINT bit 18h) */ - temp = inw(TRID_REG(card, T4D_MISCINT + 2)); - temp &= (u16) (~0x0004); - outw(temp, TRID_REG(card, T4D_MISCINT + 2)); - - /* Adjust the volume */ - ali_hwvol_control(card, opt); - - /* Set the timer for 1/10th sec */ - ali_set_timer(card); -} - -static void -cyber_address_interrupt(struct trident_card *card) -{ - int i, irq_status; - struct trident_state *state; - unsigned int channel; - - /* Update the pointers for all channels we are running. */ - /* FIXED: read interrupt status only once */ - irq_status = inl(TRID_REG(card, T4D_AINT_A)); - - pr_debug("cyber_address_interrupt: irq_status 0x%X\n", irq_status); - - for (i = 0; i < NR_HW_CH; i++) { - channel = 31 - i; - if (irq_status & (1 << channel)) { - /* clear bit by writing a 1, zeroes are ignored */ - outl((1 << channel), TRID_REG(card, T4D_AINT_A)); - - pr_debug("cyber_interrupt: channel %d\n", channel); - - if ((state = card->states[i]) != NULL) { - trident_update_ptr(state); - } else { - printk(KERN_WARNING "cyber5050: spurious " - "channel irq %d.\n", channel); - trident_stop_voice(card, channel); - trident_disable_voice_irq(card, channel); - } - } - } -} - -static irqreturn_t -trident_interrupt(int irq, void *dev_id) -{ - struct trident_card *card = (struct trident_card *) dev_id; - u32 event; - u32 gpio; - - spin_lock(&card->lock); - event = inl(TRID_REG(card, T4D_MISCINT)); - - pr_debug("trident: trident_interrupt called, MISCINT = 0x%08x\n", - event); - - if (event & ADDRESS_IRQ) { - card->address_interrupt(card); - } - - if (card->pci_id == PCI_DEVICE_ID_ALI_5451) { - /* GPIO IRQ (H/W Volume Control) */ - event = inl(TRID_REG(card, T4D_MISCINT)); - if (event & (1 << 25)) { - gpio = inl(TRID_REG(card, ALI_GPIO)); - if (!timer_pending(&card->timer)) - ali_queue_task(card, gpio & 0x07); - } - event = inl(TRID_REG(card, T4D_MISCINT)); - outl(event | (ST_TARGET_REACHED | MIXER_OVERFLOW | MIXER_UNDERFLOW), - TRID_REG(card, T4D_MISCINT)); - spin_unlock(&card->lock); - return IRQ_HANDLED; - } - - /* manually clear interrupt status, bad hardware design, blame T^2 */ - outl((ST_TARGET_REACHED | MIXER_OVERFLOW | MIXER_UNDERFLOW), - TRID_REG(card, T4D_MISCINT)); - spin_unlock(&card->lock); - return IRQ_HANDLED; -} - -/* in this loop, dmabuf.count signifies the amount of data that is waiting */ -/* to be copied to the user's buffer. it is filled by the dma machine and */ -/* drained by this loop. */ -static ssize_t -trident_read(struct file *file, char __user *buffer, size_t count, loff_t * ppos) -{ - struct trident_state *state = (struct trident_state *)file->private_data; - struct dmabuf *dmabuf = &state->dmabuf; - ssize_t ret = 0; - unsigned long flags; - unsigned swptr; - int cnt; - - pr_debug("trident: trident_read called, count = %zd\n", count); - - VALIDATE_STATE(state); - - if (dmabuf->mapped) - return -ENXIO; - if (!access_ok(VERIFY_WRITE, buffer, count)) - return -EFAULT; - - mutex_lock(&state->sem); - if (!dmabuf->ready && (ret = prog_dmabuf_record(state))) - goto out; - - while (count > 0) { - spin_lock_irqsave(&state->card->lock, flags); - if (dmabuf->count > (signed) dmabuf->dmasize) { - /* buffer overrun, we are recovering from */ - /* sleep_on_timeout, resync hwptr and swptr, */ - /* make process flush the buffer */ - dmabuf->count = dmabuf->dmasize; - dmabuf->swptr = dmabuf->hwptr; - } - swptr = dmabuf->swptr; - cnt = dmabuf->dmasize - swptr; - if (dmabuf->count < cnt) - cnt = dmabuf->count; - spin_unlock_irqrestore(&state->card->lock, flags); - - if (cnt > count) - cnt = count; - if (cnt <= 0) { - unsigned long tmo; - /* buffer is empty, start the dma machine and */ - /* wait for data to be recorded */ - start_adc(state); - if (file->f_flags & O_NONBLOCK) { - if (!ret) - ret = -EAGAIN; - goto out; - } - - mutex_unlock(&state->sem); - /* No matter how much space left in the buffer, */ - /* we have to wait until CSO == ESO/2 or CSO == ESO */ - /* when address engine interrupts */ - tmo = (dmabuf->dmasize * HZ) / (dmabuf->rate * 2); - tmo >>= sample_shift[dmabuf->fmt]; - /* There are two situations when sleep_on_timeout returns, one is when - the interrupt is serviced correctly and the process is waked up by - ISR ON TIME. Another is when timeout is expired, which means that - either interrupt is NOT serviced correctly (pending interrupt) or it - is TOO LATE for the process to be scheduled to run (scheduler latency) - which results in a (potential) buffer overrun. And worse, there is - NOTHING we can do to prevent it. */ - if (!interruptible_sleep_on_timeout(&dmabuf->wait, tmo)) { - pr_debug(KERN_ERR "trident: recording schedule timeout, " - "dmasz %u fragsz %u count %i hwptr %u swptr %u\n", - dmabuf->dmasize, dmabuf->fragsize, dmabuf->count, - dmabuf->hwptr, dmabuf->swptr); - - /* a buffer overrun, we delay the recovery until next time the - while loop begin and we REALLY have space to record */ - } - if (signal_pending(current)) { - if (!ret) - ret = -ERESTARTSYS; - goto out; - } - mutex_lock(&state->sem); - if (dmabuf->mapped) { - if (!ret) - ret = -ENXIO; - goto out; - } - continue; - } - - if (copy_to_user(buffer, dmabuf->rawbuf + swptr, cnt)) { - if (!ret) - ret = -EFAULT; - goto out; - } - - swptr = (swptr + cnt) % dmabuf->dmasize; - - spin_lock_irqsave(&state->card->lock, flags); - dmabuf->swptr = swptr; - dmabuf->count -= cnt; - spin_unlock_irqrestore(&state->card->lock, flags); - - count -= cnt; - buffer += cnt; - ret += cnt; - start_adc(state); - } -out: - mutex_unlock(&state->sem); - return ret; -} - -/* in this loop, dmabuf.count signifies the amount of data that is waiting to be dma to - the soundcard. it is drained by the dma machine and filled by this loop. */ - -static ssize_t -trident_write(struct file *file, const char __user *buffer, size_t count, loff_t * ppos) -{ - struct trident_state *state = (struct trident_state *)file->private_data; - struct dmabuf *dmabuf = &state->dmabuf; - ssize_t ret; - unsigned long flags; - unsigned swptr; - int cnt; - unsigned int state_cnt; - unsigned int copy_count; - int lret; /* for lock_set_fmt */ - - pr_debug("trident: trident_write called, count = %zd\n", count); - - VALIDATE_STATE(state); - - /* - * Guard against an mmap or ioctl while writing - */ - - mutex_lock(&state->sem); - - if (dmabuf->mapped) { - ret = -ENXIO; - goto out; - } - if (!dmabuf->ready && (ret = prog_dmabuf_playback(state))) - goto out; - - if (!access_ok(VERIFY_READ, buffer, count)) { - ret = -EFAULT; - goto out; - } - - ret = 0; - - while (count > 0) { - spin_lock_irqsave(&state->card->lock, flags); - if (dmabuf->count < 0) { - /* buffer underrun, we are recovering from */ - /* sleep_on_timeout, resync hwptr and swptr */ - dmabuf->count = 0; - dmabuf->swptr = dmabuf->hwptr; - } - swptr = dmabuf->swptr; - cnt = dmabuf->dmasize - swptr; - if (dmabuf->count + cnt > dmabuf->dmasize) - cnt = dmabuf->dmasize - dmabuf->count; - spin_unlock_irqrestore(&state->card->lock, flags); - - if (cnt > count) - cnt = count; - if (cnt <= 0) { - unsigned long tmo; - /* buffer is full, start the dma machine and */ - /* wait for data to be played */ - start_dac(state); - if (file->f_flags & O_NONBLOCK) { - if (!ret) - ret = -EAGAIN; - goto out; - } - /* No matter how much data left in the buffer, */ - /* we have to wait until CSO == ESO/2 or CSO == ESO */ - /* when address engine interrupts */ - lock_set_fmt(state); - tmo = (dmabuf->dmasize * HZ) / (dmabuf->rate * 2); - tmo >>= sample_shift[dmabuf->fmt]; - unlock_set_fmt(state); - mutex_unlock(&state->sem); - - /* There are two situations when sleep_on_timeout */ - /* returns, one is when the interrupt is serviced */ - /* correctly and the process is waked up by ISR */ - /* ON TIME. Another is when timeout is expired, which */ - /* means that either interrupt is NOT serviced */ - /* correctly (pending interrupt) or it is TOO LATE */ - /* for the process to be scheduled to run */ - /* (scheduler latency) which results in a (potential) */ - /* buffer underrun. And worse, there is NOTHING we */ - /* can do to prevent it. */ - if (!interruptible_sleep_on_timeout(&dmabuf->wait, tmo)) { - pr_debug(KERN_ERR "trident: playback schedule " - "timeout, dmasz %u fragsz %u count %i " - "hwptr %u swptr %u\n", dmabuf->dmasize, - dmabuf->fragsize, dmabuf->count, - dmabuf->hwptr, dmabuf->swptr); - - /* a buffer underrun, we delay the recovery */ - /* until next time the while loop begin and */ - /* we REALLY have data to play */ - } - if (signal_pending(current)) { - if (!ret) - ret = -ERESTARTSYS; - goto out_nolock; - } - mutex_lock(&state->sem); - if (dmabuf->mapped) { - if (!ret) - ret = -ENXIO; - goto out; - } - continue; - } - if ((lret = lock_set_fmt(state)) < 0) { - ret = lret; - goto out; - } - - if (state->chans_num == 6) { - copy_count = 0; - state_cnt = 0; - if (ali_write_5_1(state, buffer, cnt, ©_count, - &state_cnt) == -EFAULT) { - if (state_cnt) { - swptr = (swptr + state_cnt) % dmabuf->dmasize; - spin_lock_irqsave(&state->card->lock, flags); - dmabuf->swptr = swptr; - dmabuf->count += state_cnt; - dmabuf->endcleared = 0; - spin_unlock_irqrestore(&state->card->lock, flags); - } - ret += copy_count; - if (!ret) - ret = -EFAULT; - unlock_set_fmt(state); - goto out; - } - } else { - if (copy_from_user(dmabuf->rawbuf + swptr, - buffer, cnt)) { - if (!ret) - ret = -EFAULT; - unlock_set_fmt(state); - goto out; - } - state_cnt = cnt; - } - unlock_set_fmt(state); - - swptr = (swptr + state_cnt) % dmabuf->dmasize; - - spin_lock_irqsave(&state->card->lock, flags); - dmabuf->swptr = swptr; - dmabuf->count += state_cnt; - dmabuf->endcleared = 0; - spin_unlock_irqrestore(&state->card->lock, flags); - - count -= cnt; - buffer += cnt; - ret += cnt; - start_dac(state); - } -out: - mutex_unlock(&state->sem); -out_nolock: - return ret; -} - -/* No kernel lock - we have our own spinlock */ -static unsigned int -trident_poll(struct file *file, struct poll_table_struct *wait) -{ - struct trident_state *state = (struct trident_state *)file->private_data; - struct dmabuf *dmabuf = &state->dmabuf; - unsigned long flags; - unsigned int mask = 0; - - VALIDATE_STATE(state); - - /* - * Guard against a parallel poll and write causing multiple - * prog_dmabuf events - */ - - mutex_lock(&state->sem); - - if (file->f_mode & FMODE_WRITE) { - if (!dmabuf->ready && prog_dmabuf_playback(state)) { - mutex_unlock(&state->sem); - return 0; - } - poll_wait(file, &dmabuf->wait, wait); - } - if (file->f_mode & FMODE_READ) { - if (!dmabuf->ready && prog_dmabuf_record(state)) { - mutex_unlock(&state->sem); - return 0; - } - poll_wait(file, &dmabuf->wait, wait); - } - - mutex_unlock(&state->sem); - - spin_lock_irqsave(&state->card->lock, flags); - trident_update_ptr(state); - if (file->f_mode & FMODE_READ) { - if (dmabuf->count >= (signed) dmabuf->fragsize) - mask |= POLLIN | POLLRDNORM; - } - if (file->f_mode & FMODE_WRITE) { - if (dmabuf->mapped) { - if (dmabuf->count >= (signed) dmabuf->fragsize) - mask |= POLLOUT | POLLWRNORM; - } else { - if ((signed) dmabuf->dmasize >= dmabuf->count + - (signed) dmabuf->fragsize) - mask |= POLLOUT | POLLWRNORM; - } - } - spin_unlock_irqrestore(&state->card->lock, flags); - - return mask; -} - -static int -trident_mmap(struct file *file, struct vm_area_struct *vma) -{ - struct trident_state *state = (struct trident_state *)file->private_data; - struct dmabuf *dmabuf = &state->dmabuf; - int ret = -EINVAL; - unsigned long size; - - VALIDATE_STATE(state); - - /* - * Lock against poll read write or mmap creating buffers. Also lock - * a read or write against an mmap. - */ - - mutex_lock(&state->sem); - - if (vma->vm_flags & VM_WRITE) { - if ((ret = prog_dmabuf_playback(state)) != 0) - goto out; - } else if (vma->vm_flags & VM_READ) { - if ((ret = prog_dmabuf_record(state)) != 0) - goto out; - } else - goto out; - - ret = -EINVAL; - if (vma->vm_pgoff != 0) - goto out; - size = vma->vm_end - vma->vm_start; - if (size > (PAGE_SIZE << dmabuf->buforder)) - goto out; - ret = -EAGAIN; - if (remap_pfn_range(vma, vma->vm_start, - virt_to_phys(dmabuf->rawbuf) >> PAGE_SHIFT, - size, vma->vm_page_prot)) - goto out; - dmabuf->mapped = 1; - ret = 0; -out: - mutex_unlock(&state->sem); - return ret; -} - -static int -trident_ioctl(struct inode *inode, struct file *file, - unsigned int cmd, unsigned long arg) -{ - struct trident_state *state = (struct trident_state *)file->private_data; - struct dmabuf *dmabuf = &state->dmabuf; - unsigned long flags; - audio_buf_info abinfo; - count_info cinfo; - int val, mapped, ret = 0; - struct trident_card *card = state->card; - void __user *argp = (void __user *)arg; - int __user *p = argp; - - VALIDATE_STATE(state); - - - mapped = ((file->f_mode & (FMODE_WRITE | FMODE_READ)) && dmabuf->mapped); - - pr_debug("trident: trident_ioctl, command = %2d, arg = 0x%08x\n", - _IOC_NR(cmd), arg ? *p : 0); - - switch (cmd) { - case OSS_GETVERSION: - ret = put_user(SOUND_VERSION, p); - break; - - case SNDCTL_DSP_RESET: - /* FIXME: spin_lock ? */ - if (file->f_mode & FMODE_WRITE) { - stop_dac(state); - synchronize_irq(card->irq); - dmabuf->ready = 0; - dmabuf->swptr = dmabuf->hwptr = 0; - dmabuf->count = dmabuf->total_bytes = 0; - } - if (file->f_mode & FMODE_READ) { - stop_adc(state); - synchronize_irq(card->irq); - dmabuf->ready = 0; - dmabuf->swptr = dmabuf->hwptr = 0; - dmabuf->count = dmabuf->total_bytes = 0; - } - break; - - case SNDCTL_DSP_SYNC: - if (file->f_mode & FMODE_WRITE) - ret = drain_dac(state, file->f_flags & O_NONBLOCK); - break; - - case SNDCTL_DSP_SPEED: /* set smaple rate */ - if (get_user(val, p)) { - ret = -EFAULT; - break; - } - if (val >= 0) { - if (file->f_mode & FMODE_WRITE) { - stop_dac(state); - dmabuf->ready = 0; - spin_lock_irqsave(&state->card->lock, flags); - trident_set_dac_rate(state, val); - spin_unlock_irqrestore(&state->card->lock, flags); - } - if (file->f_mode & FMODE_READ) { - stop_adc(state); - dmabuf->ready = 0; - spin_lock_irqsave(&state->card->lock, flags); - trident_set_adc_rate(state, val); - spin_unlock_irqrestore(&state->card->lock, flags); - } - } - ret = put_user(dmabuf->rate, p); - break; - - case SNDCTL_DSP_STEREO: /* set stereo or mono channel */ - if (get_user(val, p)) { - ret = -EFAULT; - break; - } - if ((ret = lock_set_fmt(state)) < 0) - return ret; - - if (file->f_mode & FMODE_WRITE) { - stop_dac(state); - dmabuf->ready = 0; - if (val) - dmabuf->fmt |= TRIDENT_FMT_STEREO; - else - dmabuf->fmt &= ~TRIDENT_FMT_STEREO; - } - if (file->f_mode & FMODE_READ) { - stop_adc(state); - dmabuf->ready = 0; - if (val) - dmabuf->fmt |= TRIDENT_FMT_STEREO; - else - dmabuf->fmt &= ~TRIDENT_FMT_STEREO; - } - unlock_set_fmt(state); - break; - - case SNDCTL_DSP_GETBLKSIZE: - if (file->f_mode & FMODE_WRITE) { - if ((val = prog_dmabuf_playback(state))) - ret = val; - else - ret = put_user(dmabuf->fragsize, p); - break; - } - if (file->f_mode & FMODE_READ) { - if ((val = prog_dmabuf_record(state))) - ret = val; - else - ret = put_user(dmabuf->fragsize, p); - break; - } - /* neither READ nor WRITE? is this even possible? */ - ret = -EINVAL; - break; - - - case SNDCTL_DSP_GETFMTS: /* Returns a mask of supported sample format */ - ret = put_user(AFMT_S16_LE | AFMT_U16_LE | AFMT_S8 | - AFMT_U8, p); - break; - - case SNDCTL_DSP_SETFMT: /* Select sample format */ - if (get_user(val, p)) { - ret = -EFAULT; - break; - } - if ((ret = lock_set_fmt(state)) < 0) - return ret; - - if (val != AFMT_QUERY) { - if (file->f_mode & FMODE_WRITE) { - stop_dac(state); - dmabuf->ready = 0; - if (val == AFMT_S16_LE) - dmabuf->fmt |= TRIDENT_FMT_16BIT; - else - dmabuf->fmt &= ~TRIDENT_FMT_16BIT; - } - if (file->f_mode & FMODE_READ) { - stop_adc(state); - dmabuf->ready = 0; - if (val == AFMT_S16_LE) - dmabuf->fmt |= TRIDENT_FMT_16BIT; - else - dmabuf->fmt &= ~TRIDENT_FMT_16BIT; - } - } - unlock_set_fmt(state); - ret = put_user((dmabuf->fmt & TRIDENT_FMT_16BIT) ? AFMT_S16_LE : - AFMT_U8, p); - break; - - case SNDCTL_DSP_CHANNELS: - if (get_user(val, p)) { - ret = -EFAULT; - break; - } - if (val != 0) { - if ((ret = lock_set_fmt(state)) < 0) - return ret; - - if (file->f_mode & FMODE_WRITE) { - stop_dac(state); - dmabuf->ready = 0; - - //prevent from memory leak - if ((state->chans_num > 2) && (state->chans_num != val)) { - ali_free_other_states_resources(state); - state->chans_num = 1; - } - - if (val >= 2) { - - dmabuf->fmt |= TRIDENT_FMT_STEREO; - if ((val == 6) && (state->card->pci_id == PCI_DEVICE_ID_ALI_5451)) { - if (card->rec_channel_use_count > 0) { - printk(KERN_ERR "trident: Record is " - "working on the card!\n"); - ret = -EBUSY; - unlock_set_fmt(state); - break; - } - - ret = ali_setup_multi_channels(state->card, 6); - if (ret < 0) { - unlock_set_fmt(state); - break; - } - mutex_lock(&state->card->open_mutex); - ret = ali_allocate_other_states_resources(state, 6); - if (ret < 0) { - mutex_unlock(&state->card->open_mutex); - unlock_set_fmt(state); - break; - } - state->card->multi_channel_use_count++; - mutex_unlock(&state->card->open_mutex); - } else - val = 2; /*yield to 2-channels */ - } else - dmabuf->fmt &= ~TRIDENT_FMT_STEREO; - state->chans_num = val; - } - if (file->f_mode & FMODE_READ) { - stop_adc(state); - dmabuf->ready = 0; - if (val >= 2) { - if (!((file->f_mode & FMODE_WRITE) && - (val == 6))) - val = 2; - dmabuf->fmt |= TRIDENT_FMT_STEREO; - } else - dmabuf->fmt &= ~TRIDENT_FMT_STEREO; - state->chans_num = val; - } - unlock_set_fmt(state); - } - ret = put_user(val, p); - break; - - case SNDCTL_DSP_POST: - /* Cause the working fragment to be output */ - break; - - case SNDCTL_DSP_SUBDIVIDE: - if (dmabuf->subdivision) { - ret = -EINVAL; - break; - } - if (get_user(val, p)) { - ret = -EFAULT; - break; - } - if (val != 1 && val != 2 && val != 4) { - ret = -EINVAL; - break; - } - dmabuf->subdivision = val; - break; - - case SNDCTL_DSP_SETFRAGMENT: - if (get_user(val, p)) { - ret = -EFAULT; - break; - } - - dmabuf->ossfragshift = val & 0xffff; - dmabuf->ossmaxfrags = (val >> 16) & 0xffff; - if (dmabuf->ossfragshift < 4) - dmabuf->ossfragshift = 4; - if (dmabuf->ossfragshift > 15) - dmabuf->ossfragshift = 15; - if (dmabuf->ossmaxfrags < 4) - dmabuf->ossmaxfrags = 4; - - break; - - case SNDCTL_DSP_GETOSPACE: - if (!(file->f_mode & FMODE_WRITE)) { - ret = -EINVAL; - break; - } - if (!dmabuf->ready && (val = prog_dmabuf_playback(state)) != 0) { - ret = val; - break; - } - spin_lock_irqsave(&state->card->lock, flags); - trident_update_ptr(state); - abinfo.fragsize = dmabuf->fragsize; - abinfo.bytes = dmabuf->dmasize - dmabuf->count; - abinfo.fragstotal = dmabuf->numfrag; - abinfo.fragments = abinfo.bytes >> dmabuf->fragshift; - spin_unlock_irqrestore(&state->card->lock, flags); - ret = copy_to_user(argp, &abinfo, sizeof (abinfo)) ? - -EFAULT : 0; - break; - - case SNDCTL_DSP_GETISPACE: - if (!(file->f_mode & FMODE_READ)) { - ret = -EINVAL; - break; - } - if (!dmabuf->ready && (val = prog_dmabuf_record(state)) != 0) { - ret = val; - break; - } - spin_lock_irqsave(&state->card->lock, flags); - trident_update_ptr(state); - abinfo.fragsize = dmabuf->fragsize; - abinfo.bytes = dmabuf->count; - abinfo.fragstotal = dmabuf->numfrag; - abinfo.fragments = abinfo.bytes >> dmabuf->fragshift; - spin_unlock_irqrestore(&state->card->lock, flags); - ret = copy_to_user(argp, &abinfo, sizeof (abinfo)) ? - -EFAULT : 0; - break; - - case SNDCTL_DSP_NONBLOCK: - file->f_flags |= O_NONBLOCK; - break; - - case SNDCTL_DSP_GETCAPS: - ret = put_user(DSP_CAP_REALTIME | DSP_CAP_TRIGGER | - DSP_CAP_MMAP | DSP_CAP_BIND, p); - break; - - case SNDCTL_DSP_GETTRIGGER: - val = 0; - if ((file->f_mode & FMODE_READ) && dmabuf->enable) - val |= PCM_ENABLE_INPUT; - if ((file->f_mode & FMODE_WRITE) && dmabuf->enable) - val |= PCM_ENABLE_OUTPUT; - ret = put_user(val, p); - break; - - case SNDCTL_DSP_SETTRIGGER: - if (get_user(val, p)) { - ret = -EFAULT; - break; - } - if (file->f_mode & FMODE_READ) { - if (val & PCM_ENABLE_INPUT) { - if (!dmabuf->ready && - (ret = prog_dmabuf_record(state))) - break; - start_adc(state); - } else - stop_adc(state); - } - if (file->f_mode & FMODE_WRITE) { - if (val & PCM_ENABLE_OUTPUT) { - if (!dmabuf->ready && - (ret = prog_dmabuf_playback(state))) - break; - start_dac(state); - } else - stop_dac(state); - } - break; - - case SNDCTL_DSP_GETIPTR: - if (!(file->f_mode & FMODE_READ)) { - ret = -EINVAL; - break; - } - if (!dmabuf->ready && (val = prog_dmabuf_record(state)) - != 0) { - ret = val; - break; - } - spin_lock_irqsave(&state->card->lock, flags); - trident_update_ptr(state); - cinfo.bytes = dmabuf->total_bytes; - cinfo.blocks = dmabuf->count >> dmabuf->fragshift; - cinfo.ptr = dmabuf->hwptr; - if (dmabuf->mapped) - dmabuf->count &= dmabuf->fragsize - 1; - spin_unlock_irqrestore(&state->card->lock, flags); - ret = copy_to_user(argp, &cinfo, sizeof (cinfo)) ? - -EFAULT : 0; - break; - - case SNDCTL_DSP_GETOPTR: - if (!(file->f_mode & FMODE_WRITE)) { - ret = -EINVAL; - break; - } - if (!dmabuf->ready && (val = prog_dmabuf_playback(state)) - != 0) { - ret = val; - break; - } - - spin_lock_irqsave(&state->card->lock, flags); - trident_update_ptr(state); - cinfo.bytes = dmabuf->total_bytes; - cinfo.blocks = dmabuf->count >> dmabuf->fragshift; - cinfo.ptr = dmabuf->hwptr; - if (dmabuf->mapped) - dmabuf->count &= dmabuf->fragsize - 1; - spin_unlock_irqrestore(&state->card->lock, flags); - ret = copy_to_user(argp, &cinfo, sizeof (cinfo)) ? - -EFAULT : 0; - break; - - case SNDCTL_DSP_SETDUPLEX: - ret = -EINVAL; - break; - - case SNDCTL_DSP_GETODELAY: - if (!(file->f_mode & FMODE_WRITE)) { - ret = -EINVAL; - break; - } - if (!dmabuf->ready && (val = prog_dmabuf_playback(state)) != 0) { - ret = val; - break; - } - spin_lock_irqsave(&state->card->lock, flags); - trident_update_ptr(state); - val = dmabuf->count; - spin_unlock_irqrestore(&state->card->lock, flags); - ret = put_user(val, p); - break; - - case SOUND_PCM_READ_RATE: - ret = put_user(dmabuf->rate, p); - break; - - case SOUND_PCM_READ_CHANNELS: - ret = put_user((dmabuf->fmt & TRIDENT_FMT_STEREO) ? 2 : 1, - p); - break; - - case SOUND_PCM_READ_BITS: - ret = put_user((dmabuf->fmt & TRIDENT_FMT_16BIT) ? AFMT_S16_LE : - AFMT_U8, p); - break; - - case SNDCTL_DSP_GETCHANNELMASK: - ret = put_user(DSP_BIND_FRONT | DSP_BIND_SURR | - DSP_BIND_CENTER_LFE, p); - break; - - case SNDCTL_DSP_BIND_CHANNEL: - if (state->card->pci_id != PCI_DEVICE_ID_SI_7018) { - ret = -EINVAL; - break; - } - - if (get_user(val, p)) { - ret = -EFAULT; - break; - } - if (val == DSP_BIND_QUERY) { - val = dmabuf->channel->attribute | 0x3c00; - val = attr2mask[val >> 8]; - } else { - dmabuf->ready = 0; - if (file->f_mode & FMODE_READ) - dmabuf->channel->attribute = (CHANNEL_REC | - SRC_ENABLE); - if (file->f_mode & FMODE_WRITE) - dmabuf->channel->attribute = (CHANNEL_SPC_PB | - SRC_ENABLE); - dmabuf->channel->attribute |= mask2attr[ffs(val)]; - } - ret = put_user(val, p); - break; - - case SNDCTL_DSP_MAPINBUF: - case SNDCTL_DSP_MAPOUTBUF: - case SNDCTL_DSP_SETSYNCRO: - case SOUND_PCM_WRITE_FILTER: - case SOUND_PCM_READ_FILTER: - default: - ret = -EINVAL; - break; - - } - return ret; -} - -static int -trident_open(struct inode *inode, struct file *file) -{ - int i = 0; - int minor = iminor(inode); - struct trident_card *card = devs; - struct trident_state *state = NULL; - struct dmabuf *dmabuf = NULL; - unsigned long flags; - - /* Added by Matt Wu 01-05-2001 */ - /* TODO: there's some redundacy here wrt the check below */ - /* for multi_use_count > 0. Should we return -EBUSY or find */ - /* a different card? for now, don't break current behaviour */ - /* -- mulix */ - if (file->f_mode & FMODE_READ) { - if (card->pci_id == PCI_DEVICE_ID_ALI_5451) { - if (card->multi_channel_use_count > 0) - return -EBUSY; - } - } - - /* find an available virtual channel (instance of /dev/dsp) */ - while (card != NULL) { - mutex_lock(&card->open_mutex); - if (file->f_mode & FMODE_READ) { - /* Skip opens on cards that are in 6 channel mode */ - if (card->multi_channel_use_count > 0) { - mutex_unlock(&card->open_mutex); - card = card->next; - continue; - } - } - for (i = 0; i < NR_HW_CH; i++) { - if (card->states[i] == NULL) { - state = card->states[i] = kzalloc(sizeof(*state), GFP_KERNEL); - if (state == NULL) { - mutex_unlock(&card->open_mutex); - return -ENOMEM; - } - mutex_init(&state->sem); - dmabuf = &state->dmabuf; - goto found_virt; - } - } - mutex_unlock(&card->open_mutex); - card = card->next; - } - /* no more virtual channel avaiable */ - if (!state) { - return -ENODEV; - } - found_virt: - /* found a free virtual channel, allocate hardware channels */ - if (file->f_mode & FMODE_READ) - dmabuf->channel = card->alloc_rec_pcm_channel(card); - else - dmabuf->channel = card->alloc_pcm_channel(card); - - if (dmabuf->channel == NULL) { - kfree(card->states[i]); - card->states[i] = NULL; - return -ENODEV; - } - - /* initialize the virtual channel */ - state->virt = i; - state->card = card; - state->magic = TRIDENT_STATE_MAGIC; - init_waitqueue_head(&dmabuf->wait); - file->private_data = state; - - /* set default sample format. According to OSS Programmer's */ - /* Guide /dev/dsp should be default to unsigned 8-bits, mono, */ - /* with sample rate 8kHz and /dev/dspW will accept 16-bits sample */ - if (file->f_mode & FMODE_WRITE) { - dmabuf->fmt &= ~TRIDENT_FMT_MASK; - if ((minor & 0x0f) == SND_DEV_DSP16) - dmabuf->fmt |= TRIDENT_FMT_16BIT; - dmabuf->ossfragshift = 0; - dmabuf->ossmaxfrags = 0; - dmabuf->subdivision = 0; - if (card->pci_id == PCI_DEVICE_ID_SI_7018) { - /* set default channel attribute to normal playback */ - dmabuf->channel->attribute = CHANNEL_PB; - } - spin_lock_irqsave(&card->lock, flags); - trident_set_dac_rate(state, 8000); - spin_unlock_irqrestore(&card->lock, flags); - } - - if (file->f_mode & FMODE_READ) { - /* FIXME: Trident 4d can only record in signed 16-bits stereo, */ - /* 48kHz sample, to be dealed with in trident_set_adc_rate() ?? */ - dmabuf->fmt &= ~TRIDENT_FMT_MASK; - if ((minor & 0x0f) == SND_DEV_DSP16) - dmabuf->fmt |= TRIDENT_FMT_16BIT; - dmabuf->ossfragshift = 0; - dmabuf->ossmaxfrags = 0; - dmabuf->subdivision = 0; - if (card->pci_id == PCI_DEVICE_ID_SI_7018) { - /* set default channel attribute to 0x8a80, record from - PCM L/R FIFO and mono = (left + right + 1)/2 */ - dmabuf->channel->attribute = (CHANNEL_REC | PCM_LR | - MONO_MIX); - } - spin_lock_irqsave(&card->lock, flags); - trident_set_adc_rate(state, 8000); - spin_unlock_irqrestore(&card->lock, flags); - - /* Added by Matt Wu 01-05-2001 */ - if (card->pci_id == PCI_DEVICE_ID_ALI_5451) - card->rec_channel_use_count++; - } - - state->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE); - mutex_unlock(&card->open_mutex); - - pr_debug("trident: open virtual channel %d, hard channel %d\n", - state->virt, dmabuf->channel->num); - - return nonseekable_open(inode, file); -} - -static int -trident_release(struct inode *inode, struct file *file) -{ - struct trident_state *state = (struct trident_state *)file->private_data; - struct trident_card *card; - struct dmabuf *dmabuf; - - VALIDATE_STATE(state); - - card = state->card; - dmabuf = &state->dmabuf; - - if (file->f_mode & FMODE_WRITE) { - trident_clear_tail(state); - drain_dac(state, file->f_flags & O_NONBLOCK); - } - - pr_debug("trident: closing virtual channel %d, hard channel %d\n", - state->virt, dmabuf->channel->num); - - /* stop DMA state machine and free DMA buffers/channels */ - mutex_lock(&card->open_mutex); - - if (file->f_mode & FMODE_WRITE) { - stop_dac(state); - dealloc_dmabuf(&state->dmabuf, state->card->pci_dev); - state->card->free_pcm_channel(state->card, dmabuf->channel->num); - - /* Added by Matt Wu */ - if (card->pci_id == PCI_DEVICE_ID_ALI_5451) { - if (state->chans_num > 2) { - if (card->multi_channel_use_count-- < 0) - card->multi_channel_use_count = 0; - if (card->multi_channel_use_count == 0) - ali_close_multi_channels(); - ali_free_other_states_resources(state); - } - } - } - if (file->f_mode & FMODE_READ) { - stop_adc(state); - dealloc_dmabuf(&state->dmabuf, state->card->pci_dev); - state->card->free_pcm_channel(state->card, dmabuf->channel->num); - - /* Added by Matt Wu */ - if (card->pci_id == PCI_DEVICE_ID_ALI_5451) { - if (card->rec_channel_use_count-- < 0) - card->rec_channel_use_count = 0; - } - } - - card->states[state->virt] = NULL; - kfree(state); - - /* we're covered by the open_mutex */ - mutex_unlock(&card->open_mutex); - - return 0; -} - -static const struct file_operations trident_audio_fops = { - .owner = THIS_MODULE, - .llseek = no_llseek, - .read = trident_read, - .write = trident_write, - .poll = trident_poll, - .ioctl = trident_ioctl, - .mmap = trident_mmap, - .open = trident_open, - .release = trident_release, -}; - -/* trident specific AC97 functions */ -/* Write AC97 codec registers */ -static void -trident_ac97_set(struct ac97_codec *codec, u8 reg, u16 val) -{ - struct trident_card *card = (struct trident_card *)codec->private_data; - unsigned int address, mask, busy; - unsigned short count = 0xffff; - unsigned long flags; - u32 data; - - data = ((u32) val) << 16; - - switch (card->pci_id) { - default: - case PCI_DEVICE_ID_SI_7018: - address = SI_AC97_WRITE; - mask = SI_AC97_BUSY_WRITE | SI_AC97_AUDIO_BUSY; - if (codec->id) - mask |= SI_AC97_SECONDARY; - busy = SI_AC97_BUSY_WRITE; - break; - case PCI_DEVICE_ID_TRIDENT_4DWAVE_DX: - address = DX_ACR0_AC97_W; - mask = busy = DX_AC97_BUSY_WRITE; - break; - case PCI_DEVICE_ID_TRIDENT_4DWAVE_NX: - address = NX_ACR1_AC97_W; - mask = NX_AC97_BUSY_WRITE; - if (codec->id) - mask |= NX_AC97_WRITE_SECONDARY; - busy = NX_AC97_BUSY_WRITE; - break; - case PCI_DEVICE_ID_INTERG_5050: - address = SI_AC97_WRITE; - mask = busy = SI_AC97_BUSY_WRITE; - if (codec->id) - mask |= SI_AC97_SECONDARY; - break; - } - - spin_lock_irqsave(&card->lock, flags); - do { - if ((inw(TRID_REG(card, address)) & busy) == 0) - break; - } while (--count); - - data |= (mask | (reg & AC97_REG_ADDR)); - - if (count == 0) { - printk(KERN_ERR "trident: AC97 CODEC write timed out.\n"); - spin_unlock_irqrestore(&card->lock, flags); - return; - } - - outl(data, TRID_REG(card, address)); - spin_unlock_irqrestore(&card->lock, flags); -} - -/* Read AC97 codec registers */ -static u16 -trident_ac97_get(struct ac97_codec *codec, u8 reg) -{ - struct trident_card *card = (struct trident_card *)codec->private_data; - unsigned int address, mask, busy; - unsigned short count = 0xffff; - unsigned long flags; - u32 data; - - switch (card->pci_id) { - default: - case PCI_DEVICE_ID_SI_7018: - address = SI_AC97_READ; - mask = SI_AC97_BUSY_READ | SI_AC97_AUDIO_BUSY; - if (codec->id) - mask |= SI_AC97_SECONDARY; - busy = SI_AC97_BUSY_READ; - break; - case PCI_DEVICE_ID_TRIDENT_4DWAVE_DX: - address = DX_ACR1_AC97_R; - mask = busy = DX_AC97_BUSY_READ; - break; - case PCI_DEVICE_ID_TRIDENT_4DWAVE_NX: - if (codec->id) - address = NX_ACR3_AC97_R_SECONDARY; - else - address = NX_ACR2_AC97_R_PRIMARY; - mask = NX_AC97_BUSY_READ; - busy = NX_AC97_BUSY_READ | NX_AC97_BUSY_DATA; - break; - case PCI_DEVICE_ID_INTERG_5050: - address = SI_AC97_READ; - mask = busy = SI_AC97_BUSY_READ; - if (codec->id) - mask |= SI_AC97_SECONDARY; - break; - } - - data = (mask | (reg & AC97_REG_ADDR)); - - spin_lock_irqsave(&card->lock, flags); - outl(data, TRID_REG(card, address)); - do { - data = inl(TRID_REG(card, address)); - if ((data & busy) == 0) - break; - } while (--count); - spin_unlock_irqrestore(&card->lock, flags); - - if (count == 0) { - printk(KERN_ERR "trident: AC97 CODEC read timed out.\n"); - data = 0; - } - return ((u16) (data >> 16)); -} - -/* rewrite ac97 read and write mixer register by hulei for ALI*/ -static int -acquirecodecaccess(struct trident_card *card) -{ - u16 wsemamask = 0x6000; /* bit 14..13 */ - u16 wsemabits; - u16 wcontrol; - int block = 0; - int ncount = 25; - while (1) { - wcontrol = inw(TRID_REG(card, ALI_AC97_WRITE)); - wsemabits = wcontrol & wsemamask; - - if (wsemabits == 0x4000) - return 1; /* 0x4000 is audio ,then success */ - if (ncount-- < 0) - break; - if (wsemabits == 0) { - unlock: - outl(((u32) (wcontrol & 0x1eff) | 0x00004000), - TRID_REG(card, ALI_AC97_WRITE)); - continue; - } - udelay(20); - } - if (!block) { - pr_debug("accesscodecsemaphore: try unlock\n"); - block = 1; - goto unlock; - } - return 0; -} - -static void -releasecodecaccess(struct trident_card *card) -{ - unsigned long wcontrol; - wcontrol = inl(TRID_REG(card, ALI_AC97_WRITE)); - outl((wcontrol & 0xffff1eff), TRID_REG(card, ALI_AC97_WRITE)); -} - -static int -waitforstimertick(struct trident_card *card) -{ - unsigned long chk1, chk2; - unsigned int wcount = 0xffff; - chk1 = inl(TRID_REG(card, ALI_STIMER)); - - while (1) { - chk2 = inl(TRID_REG(card, ALI_STIMER)); - if ((wcount > 0) && chk1 != chk2) - return 1; - if (wcount <= 0) - break; - udelay(50); - } - return 0; -} - -/* Read AC97 codec registers for ALi*/ -static u16 -ali_ac97_get(struct trident_card *card, int secondary, u8 reg) -{ - unsigned int address, mask; - unsigned int ncount; - unsigned long aud_reg; - u32 data; - u16 wcontrol; - unsigned long flags; - - BUG_ON(!card); - - address = ALI_AC97_READ; - if (card->revision == ALI_5451_V02) { - address = ALI_AC97_WRITE; - } - mask = ALI_AC97_READ_ACTION | ALI_AC97_AUDIO_BUSY; - if (secondary) - mask |= ALI_AC97_SECONDARY; - - spin_lock_irqsave(&card->lock, flags); - - if (!acquirecodecaccess(card)) - printk(KERN_ERR "access codec fail\n"); - - wcontrol = inw(TRID_REG(card, ALI_AC97_WRITE)); - wcontrol &= 0xfe00; - wcontrol |= (0x8000 | reg); - outw(wcontrol, TRID_REG(card, ALI_AC97_WRITE)); - - data = (mask | (reg & AC97_REG_ADDR)); - - if (!waitforstimertick(card)) { - printk(KERN_ERR "ali_ac97_read: BIT_CLOCK is dead\n"); - goto releasecodec; - } - - udelay(20); - - ncount = 10; - - while (1) { - if ((inw(TRID_REG(card, ALI_AC97_WRITE)) & ALI_AC97_BUSY_READ) - != 0) - break; - if (ncount <= 0) - break; - if (ncount-- == 1) { - pr_debug("ali_ac97_read :try clear busy flag\n"); - aud_reg = inl(TRID_REG(card, ALI_AC97_WRITE)); - outl((aud_reg & 0xffff7fff), - TRID_REG(card, ALI_AC97_WRITE)); - } - udelay(10); - } - - data = inl(TRID_REG(card, address)); - - spin_unlock_irqrestore(&card->lock, flags); - - return ((u16) (data >> 16)); - - releasecodec: - releasecodecaccess(card); - spin_unlock_irqrestore(&card->lock, flags); - printk(KERN_ERR "ali_ac97_read: AC97 CODEC read timed out.\n"); - return 0; -} - -/* Write AC97 codec registers for hulei*/ -static void -ali_ac97_set(struct trident_card *card, int secondary, u8 reg, u16 val) -{ - unsigned int address, mask; - unsigned int ncount; - u32 data; - u16 wcontrol; - unsigned long flags; - - data = ((u32) val) << 16; - - BUG_ON(!card); - - address = ALI_AC97_WRITE; - mask = ALI_AC97_WRITE_ACTION | ALI_AC97_AUDIO_BUSY; - if (secondary) - mask |= ALI_AC97_SECONDARY; - if (card->revision == ALI_5451_V02) - mask |= ALI_AC97_WRITE_MIXER_REGISTER; - - spin_lock_irqsave(&card->lock, flags); - if (!acquirecodecaccess(card)) - printk(KERN_ERR "ali_ac97_write: access codec fail\n"); - - wcontrol = inw(TRID_REG(card, ALI_AC97_WRITE)); - wcontrol &= 0xff00; - wcontrol |= (0x8100 | reg); /* bit 8=1: (ali1535 )reserved/ */ - /* ali1535+ write */ - outl((data | wcontrol), TRID_REG(card, ALI_AC97_WRITE)); - - if (!waitforstimertick(card)) { - printk(KERN_ERR "BIT_CLOCK is dead\n"); - goto releasecodec; - } - - ncount = 10; - while (1) { - wcontrol = inw(TRID_REG(card, ALI_AC97_WRITE)); - if (!(wcontrol & 0x8000)) - break; - if (ncount <= 0) - break; - if (ncount-- == 1) { - pr_debug("ali_ac97_set :try clear busy flag!!\n"); - outw(wcontrol & 0x7fff, - TRID_REG(card, ALI_AC97_WRITE)); - } - udelay(10); - } - - releasecodec: - releasecodecaccess(card); - spin_unlock_irqrestore(&card->lock, flags); - return; -} - -static void -ali_enable_special_channel(struct trident_state *stat) -{ - struct trident_card *card = stat->card; - unsigned long s_channels; - - s_channels = inl(TRID_REG(card, ALI_GLOBAL_CONTROL)); - s_channels |= (1 << stat->dmabuf.channel->num); - outl(s_channels, TRID_REG(card, ALI_GLOBAL_CONTROL)); -} - -static u16 -ali_ac97_read(struct ac97_codec *codec, u8 reg) -{ - int id; - u16 data; - struct trident_card *card = NULL; - - /* Added by Matt Wu */ - BUG_ON(!codec); - - card = (struct trident_card *) codec->private_data; - - if (!card->mixer_regs_ready) - return ali_ac97_get(card, codec->id, reg); - - /* - * FIXME: need to stop this caching some registers - */ - if (codec->id) - id = 1; - else - id = 0; - - data = card->mixer_regs[reg / 2][id]; - return data; -} - -static void -ali_ac97_write(struct ac97_codec *codec, u8 reg, u16 val) -{ - int id; - struct trident_card *card; - - /* Added by Matt Wu */ - BUG_ON(!codec); - - card = (struct trident_card *) codec->private_data; - - if (!card->mixer_regs_ready) { - ali_ac97_set(card, codec->id, reg, val); - return; - } - - if (codec->id) - id = 1; - else - id = 0; - - card->mixer_regs[reg / 2][id] = val; - ali_ac97_set(card, codec->id, reg, val); -} - -/* -flag: ALI_SPDIF_OUT_TO_SPDIF_OUT - ALI_PCM_TO_SPDIF_OUT -*/ - -static void -ali_setup_spdif_out(struct trident_card *card, int flag) -{ - unsigned long spdif; - unsigned char ch; - - char temp; - struct pci_dev *pci_dev = NULL; - - pci_dev = pci_get_device(PCI_VENDOR_ID_AL, PCI_DEVICE_ID_AL_M1533, - pci_dev); - if (pci_dev == NULL) - return; - pci_read_config_byte(pci_dev, 0x61, &temp); - temp |= 0x40; - pci_write_config_byte(pci_dev, 0x61, temp); - pci_read_config_byte(pci_dev, 0x7d, &temp); - temp |= 0x01; - pci_write_config_byte(pci_dev, 0x7d, temp); - pci_read_config_byte(pci_dev, 0x7e, &temp); - temp &= (~0x20); - temp |= 0x10; - pci_write_config_byte(pci_dev, 0x7e, temp); - - pci_dev_put(pci_dev); - - ch = inb(TRID_REG(card, ALI_SCTRL)); - outb(ch | ALI_SPDIF_OUT_ENABLE, TRID_REG(card, ALI_SCTRL)); - ch = inb(TRID_REG(card, ALI_SPDIF_CTRL)); - outb(ch & ALI_SPDIF_OUT_CH_STATUS, TRID_REG(card, ALI_SPDIF_CTRL)); - - if (flag & ALI_SPDIF_OUT_TO_SPDIF_OUT) { - spdif = inw(TRID_REG(card, ALI_GLOBAL_CONTROL)); - spdif |= ALI_SPDIF_OUT_CH_ENABLE; - spdif &= ALI_SPDIF_OUT_SEL_SPDIF; - outw(spdif, TRID_REG(card, ALI_GLOBAL_CONTROL)); - spdif = inw(TRID_REG(card, ALI_SPDIF_CS)); - if (flag & ALI_SPDIF_OUT_NON_PCM) - spdif |= 0x0002; - else - spdif &= (~0x0002); - outw(spdif, TRID_REG(card, ALI_SPDIF_CS)); - } else { - spdif = inw(TRID_REG(card, ALI_GLOBAL_CONTROL)); - spdif |= ALI_SPDIF_OUT_SEL_PCM; - outw(spdif, TRID_REG(card, ALI_GLOBAL_CONTROL)); - } -} - -static void -ali_disable_special_channel(struct trident_card *card, int ch) -{ - unsigned long sc; - - sc = inl(TRID_REG(card, ALI_GLOBAL_CONTROL)); - sc &= ~(1 << ch); - outl(sc, TRID_REG(card, ALI_GLOBAL_CONTROL)); -} - -static void -ali_disable_spdif_in(struct trident_card *card) -{ - unsigned long spdif; - - spdif = inl(TRID_REG(card, ALI_GLOBAL_CONTROL)); - spdif &= (~ALI_SPDIF_IN_SUPPORT); - outl(spdif, TRID_REG(card, ALI_GLOBAL_CONTROL)); - - ali_disable_special_channel(card, ALI_SPDIF_IN_CHANNEL); -} - -static void -ali_setup_spdif_in(struct trident_card *card) -{ - unsigned long spdif; - - //Set SPDIF IN Supported - spdif = inl(TRID_REG(card, ALI_GLOBAL_CONTROL)); - spdif |= ALI_SPDIF_IN_SUPPORT; - outl(spdif, TRID_REG(card, ALI_GLOBAL_CONTROL)); - - //Set SPDIF IN Rec - spdif = inl(TRID_REG(card, ALI_GLOBAL_CONTROL)); - spdif |= ALI_SPDIF_IN_CH_ENABLE; - outl(spdif, TRID_REG(card, ALI_GLOBAL_CONTROL)); - - spdif = inb(TRID_REG(card, ALI_SPDIF_CTRL)); - spdif |= ALI_SPDIF_IN_CH_STATUS; - outb(spdif, TRID_REG(card, ALI_SPDIF_CTRL)); -/* - spdif = inb(TRID_REG(card, ALI_SPDIF_CTRL)); - spdif |= ALI_SPDIF_IN_FUNC_ENABLE; - outb(spdif, TRID_REG(card, ALI_SPDIF_CTRL)); -*/ -} - -static void -ali_delay(struct trident_card *card, int interval) -{ - unsigned long begintimer, currenttimer; - - begintimer = inl(TRID_REG(card, ALI_STIMER)); - currenttimer = inl(TRID_REG(card, ALI_STIMER)); - - while (currenttimer < begintimer + interval) - currenttimer = inl(TRID_REG(card, ALI_STIMER)); -} - -static void -ali_detect_spdif_rate(struct trident_card *card) -{ - u16 wval = 0; - u16 count = 0; - u8 bval = 0, R1 = 0, R2 = 0; - - bval = inb(TRID_REG(card, ALI_SPDIF_CTRL)); - bval |= 0x02; - outb(bval, TRID_REG(card, ALI_SPDIF_CTRL)); - - bval = inb(TRID_REG(card, ALI_SPDIF_CTRL + 1)); - bval |= 0x1F; - outb(bval, TRID_REG(card, ALI_SPDIF_CTRL + 1)); - - while (((R1 < 0x0B) || (R1 > 0x0E)) && (R1 != 0x12) && - count <= 50000) { - count++; - - ali_delay(card, 6); - - bval = inb(TRID_REG(card, ALI_SPDIF_CTRL + 1)); - R1 = bval & 0x1F; - } - - if (count > 50000) { - printk(KERN_WARNING "trident: Error in " - "ali_detect_spdif_rate!\n"); - return; - } - - count = 0; - - while (count <= 50000) { - count++; - - ali_delay(card, 6); - - bval = inb(TRID_REG(card, ALI_SPDIF_CTRL + 1)); - R2 = bval & 0x1F; - - if (R2 != R1) - R1 = R2; - else - break; - } - - if (count > 50000) { - printk(KERN_WARNING "trident: Error in " - "ali_detect_spdif_rate!\n"); - return; - } - - switch (R2) { - case 0x0b: - case 0x0c: - case 0x0d: - case 0x0e: - wval = inw(TRID_REG(card, ALI_SPDIF_CTRL + 2)); - wval &= 0xE0F0; - wval |= (u16) 0x09 << 8 | (u16) 0x05; - outw(wval, TRID_REG(card, ALI_SPDIF_CTRL + 2)); - - bval = inb(TRID_REG(card, ALI_SPDIF_CS + 3)) & 0xF0; - outb(bval | 0x02, TRID_REG(card, ALI_SPDIF_CS + 3)); - break; - - case 0x12: - wval = inw(TRID_REG(card, ALI_SPDIF_CTRL + 2)); - wval &= 0xE0F0; - wval |= (u16) 0x0E << 8 | (u16) 0x08; - outw(wval, TRID_REG(card, ALI_SPDIF_CTRL + 2)); - - bval = inb(TRID_REG(card, ALI_SPDIF_CS + 3)) & 0xF0; - outb(bval | 0x03, TRID_REG(card, ALI_SPDIF_CS + 3)); - break; - - default: - break; - } - -} - -static unsigned int -ali_get_spdif_in_rate(struct trident_card *card) -{ - u32 dwRate = 0; - u8 bval = 0; - - ali_detect_spdif_rate(card); - - bval = inb(TRID_REG(card, ALI_SPDIF_CTRL)); - bval &= 0x7F; - bval |= 0x40; - outb(bval, TRID_REG(card, ALI_SPDIF_CTRL)); - - bval = inb(TRID_REG(card, ALI_SPDIF_CS + 3)); - bval &= 0x0F; - - switch (bval) { - case 0: - dwRate = 44100; - break; - case 1: - dwRate = 48000; - break; - case 2: - dwRate = 32000; - break; - default: - // Error occurs - break; - } - - return dwRate; - -} - -static int -ali_close_multi_channels(void) -{ - char temp = 0; - struct pci_dev *pci_dev = NULL; - - pci_dev = pci_get_device(PCI_VENDOR_ID_AL, PCI_DEVICE_ID_AL_M1533, - pci_dev); - if (pci_dev == NULL) - return -1; - - pci_read_config_byte(pci_dev, 0x59, &temp); - temp &= ~0x80; - pci_write_config_byte(pci_dev, 0x59, temp); - - pci_dev_put(pci_dev); - - pci_dev = pci_get_device(PCI_VENDOR_ID_AL, PCI_DEVICE_ID_AL_M7101, - NULL); - if (pci_dev == NULL) - return -1; - - pci_read_config_byte(pci_dev, 0xB8, &temp); - temp &= ~0x20; - pci_write_config_byte(pci_dev, 0xB8, temp); - - pci_dev_put(pci_dev); - - return 0; -} - -static int -ali_setup_multi_channels(struct trident_card *card, int chan_nums) -{ - unsigned long dwValue; - char temp = 0; - struct pci_dev *pci_dev = NULL; - - pci_dev = pci_get_device(PCI_VENDOR_ID_AL, PCI_DEVICE_ID_AL_M1533, - pci_dev); - if (pci_dev == NULL) - return -1; - pci_read_config_byte(pci_dev, 0x59, &temp); - temp |= 0x80; - pci_write_config_byte(pci_dev, 0x59, temp); - - pci_dev_put(pci_dev); - - pci_dev = pci_get_device(PCI_VENDOR_ID_AL, PCI_DEVICE_ID_AL_M7101, - NULL); - if (pci_dev == NULL) - return -1; - pci_read_config_byte(pci_dev, (int) 0xB8, &temp); - temp |= 0x20; - pci_write_config_byte(pci_dev, (int) 0xB8, (u8) temp); - - pci_dev_put(pci_dev); - - if (chan_nums == 6) { - dwValue = inl(TRID_REG(card, ALI_SCTRL)) | 0x000f0000; - outl(dwValue, TRID_REG(card, ALI_SCTRL)); - mdelay(4); - dwValue = inl(TRID_REG(card, ALI_SCTRL)); - if (dwValue & 0x2000000) { - ali_ac97_write(card->ac97_codec[0], 0x02, 8080); - ali_ac97_write(card->ac97_codec[0], 0x36, 0); - ali_ac97_write(card->ac97_codec[0], 0x38, 0); - /* - * On a board with a single codec you won't get the - * surround. On other boards configure it. - */ - if (card->ac97_codec[1] != NULL) { - ali_ac97_write(card->ac97_codec[1], 0x36, 0); - ali_ac97_write(card->ac97_codec[1], 0x38, 0); - ali_ac97_write(card->ac97_codec[1], 0x02, 0x0606); - ali_ac97_write(card->ac97_codec[1], 0x18, 0x0303); - ali_ac97_write(card->ac97_codec[1], 0x74, 0x3); - } - return 1; - } - } - return -EINVAL; -} - -static void -ali_free_pcm_channel(struct trident_card *card, unsigned int channel) -{ - int bank; - - if (channel > 31) - return; - - bank = channel >> 5; - channel = channel & 0x1f; - - card->banks[bank].bitmap &= ~(1 << (channel)); -} - -static int -ali_allocate_other_states_resources(struct trident_state *state, int chan_nums) -{ - struct trident_card *card = state->card; - struct trident_state *s; - int i, state_count = 0; - struct trident_pcm_bank *bank; - struct trident_channel *channel; - unsigned long num; - - bank = &card->banks[BANK_A]; - - if (chan_nums != 6) - return 0; - - for (i = 0; (i < ALI_CHANNELS) && (state_count != 4); i++) { - if (card->states[i]) - continue; - - num = ali_multi_channels_5_1[state_count]; - if (!(bank->bitmap & (1 << num))) { - bank->bitmap |= 1 << num; - channel = &bank->channels[num]; - channel->num = num; - } else { - state_count--; - for (; state_count >= 0; state_count--) { - kfree(state->other_states[state_count]); - num = ali_multi_channels_5_1[state_count]; - ali_free_pcm_channel(card, num); - } - return -EBUSY; - } - s = card->states[i] = kzalloc(sizeof(*state), GFP_KERNEL); - if (!s) { - num = ali_multi_channels_5_1[state_count]; - ali_free_pcm_channel(card, num); - state_count--; - for (; state_count >= 0; state_count--) { - num = ali_multi_channels_5_1[state_count]; - ali_free_pcm_channel(card, num); - kfree(state->other_states[state_count]); - } - return -ENOMEM; - } - - s->dmabuf.channel = channel; - s->dmabuf.ossfragshift = s->dmabuf.ossmaxfrags = - s->dmabuf.subdivision = 0; - init_waitqueue_head(&s->dmabuf.wait); - s->magic = card->magic; - s->card = card; - s->virt = i; - ali_enable_special_channel(s); - state->other_states[state_count++] = s; - } - - if (state_count != 4) { - state_count--; - for (; state_count >= 0; state_count--) { - kfree(state->other_states[state_count]); - num = ali_multi_channels_5_1[state_count]; - ali_free_pcm_channel(card, num); - } - return -EBUSY; - } - return 0; -} - -#ifdef CONFIG_PM -/* save registers for ALi Power Management */ -static struct ali_saved_registers { - unsigned long global_regs[ALI_GLOBAL_REGS]; - unsigned long channel_regs[ALI_CHANNELS][ALI_CHANNEL_REGS]; - unsigned mixer_regs[ALI_MIXER_REGS]; -} ali_registers; - -static void -ali_save_regs(struct trident_card *card) -{ - unsigned long flags; - int i, j; - - spin_lock_irqsave(&card->lock, flags); - - ali_registers.global_regs[0x2c] = inl(TRID_REG(card, T4D_MISCINT)); - //ali_registers.global_regs[0x20] = inl(TRID_REG(card,T4D_START_A)); - ali_registers.global_regs[0x21] = inl(TRID_REG(card, T4D_STOP_A)); - - //disable all IRQ bits - outl(ALI_DISABLE_ALL_IRQ, TRID_REG(card, T4D_MISCINT)); - - for (i = 1; i < ALI_MIXER_REGS; i++) - ali_registers.mixer_regs[i] = ali_ac97_read(card->ac97_codec[0], - i * 2); - - for (i = 0; i < ALI_GLOBAL_REGS; i++) { - if ((i * 4 == T4D_MISCINT) || (i * 4 == T4D_STOP_A)) - continue; - ali_registers.global_regs[i] = inl(TRID_REG(card, i * 4)); - } - - for (i = 0; i < ALI_CHANNELS; i++) { - outb(i, TRID_REG(card, T4D_LFO_GC_CIR)); - for (j = 0; j < ALI_CHANNEL_REGS; j++) - ali_registers.channel_regs[i][j] = inl(TRID_REG(card, - j * 4 + 0xe0)); - } - - //Stop all HW channel - outl(ALI_STOP_ALL_CHANNELS, TRID_REG(card, T4D_STOP_A)); - - spin_unlock_irqrestore(&card->lock, flags); -} - -static void -ali_restore_regs(struct trident_card *card) -{ - unsigned long flags; - int i, j; - - spin_lock_irqsave(&card->lock, flags); - - for (i = 1; i < ALI_MIXER_REGS; i++) - ali_ac97_write(card->ac97_codec[0], i * 2, - ali_registers.mixer_regs[i]); - - for (i = 0; i < ALI_CHANNELS; i++) { - outb(i, TRID_REG(card, T4D_LFO_GC_CIR)); - for (j = 0; j < ALI_CHANNEL_REGS; j++) - outl(ali_registers.channel_regs[i][j], - TRID_REG(card, j * 4 + 0xe0)); - } - - for (i = 0; i < ALI_GLOBAL_REGS; i++) { - if ((i * 4 == T4D_MISCINT) || (i * 4 == T4D_STOP_A) || - (i * 4 == T4D_START_A)) - continue; - outl(ali_registers.global_regs[i], TRID_REG(card, i * 4)); - } - - //start HW channel - outl(ali_registers.global_regs[0x20], TRID_REG(card, T4D_START_A)); - //restore IRQ enable bits - outl(ali_registers.global_regs[0x2c], TRID_REG(card, T4D_MISCINT)); - - spin_unlock_irqrestore(&card->lock, flags); -} - -static int -trident_suspend(struct pci_dev *dev, pm_message_t unused) -{ - struct trident_card *card = pci_get_drvdata(dev); - - if (card->pci_id == PCI_DEVICE_ID_ALI_5451) { - ali_save_regs(card); - } - return 0; -} - -static int -trident_resume(struct pci_dev *dev) -{ - struct trident_card *card = pci_get_drvdata(dev); - - if (card->pci_id == PCI_DEVICE_ID_ALI_5451) { - ali_restore_regs(card); - } - return 0; -} -#endif - -static struct trident_channel * -ali_alloc_pcm_channel(struct trident_card *card) -{ - struct trident_pcm_bank *bank; - int idx; - - bank = &card->banks[BANK_A]; - - if (inl(TRID_REG(card, ALI_GLOBAL_CONTROL)) & - (ALI_SPDIF_OUT_CH_ENABLE)) { - idx = ALI_SPDIF_OUT_CHANNEL; - if (!(bank->bitmap & (1 << idx))) { - struct trident_channel *channel = &bank->channels[idx]; - bank->bitmap |= 1 << idx; - channel->num = idx; - return channel; - } - } - - for (idx = ALI_PCM_OUT_CHANNEL_FIRST; idx <= ALI_PCM_OUT_CHANNEL_LAST; - idx++) { - if (!(bank->bitmap & (1 << idx))) { - struct trident_channel *channel = &bank->channels[idx]; - bank->bitmap |= 1 << idx; - channel->num = idx; - return channel; - } - } - - /* no more free channels avaliable */ -#if 0 - printk(KERN_ERR "ali: no more channels available on Bank A.\n"); -#endif /* 0 */ - return NULL; -} - -static struct trident_channel * -ali_alloc_rec_pcm_channel(struct trident_card *card) -{ - struct trident_pcm_bank *bank; - int idx; - - if (inl(TRID_REG(card, ALI_GLOBAL_CONTROL)) & ALI_SPDIF_IN_SUPPORT) - idx = ALI_SPDIF_IN_CHANNEL; - else - idx = ALI_PCM_IN_CHANNEL; - - bank = &card->banks[BANK_A]; - - if (!(bank->bitmap & (1 << idx))) { - struct trident_channel *channel = &bank->channels[idx]; - bank->bitmap |= 1 << idx; - channel->num = idx; - return channel; - } - - /* no free recordable channels avaliable */ -#if 0 - printk(KERN_ERR "ali: no recordable channels available on Bank A.\n"); -#endif /* 0 */ - return NULL; -} - -static void -ali_set_spdif_out_rate(struct trident_card *card, unsigned int rate) -{ - unsigned char ch_st_sel; - unsigned short status_rate; - - switch (rate) { - case 44100: - status_rate = 0; - break; - case 32000: - status_rate = 0x300; - break; - case 48000: - default: - status_rate = 0x200; - break; - } - - /* select spdif_out */ - ch_st_sel = inb(TRID_REG(card, ALI_SPDIF_CTRL)) & ALI_SPDIF_OUT_CH_STATUS; - - ch_st_sel |= 0x80; /* select right */ - outb(ch_st_sel, TRID_REG(card, ALI_SPDIF_CTRL)); - outb(status_rate | 0x20, TRID_REG(card, ALI_SPDIF_CS + 2)); - - ch_st_sel &= (~0x80); /* select left */ - outb(ch_st_sel, TRID_REG(card, ALI_SPDIF_CTRL)); - outw(status_rate | 0x10, TRID_REG(card, ALI_SPDIF_CS + 2)); -} - -static void -ali_address_interrupt(struct trident_card *card) -{ - int i, channel; - struct trident_state *state; - u32 mask, channel_mask; - - mask = trident_get_interrupt_mask(card, 0); - for (i = 0; i < NR_HW_CH; i++) { - if ((state = card->states[i]) == NULL) - continue; - channel = state->dmabuf.channel->num; - if ((channel_mask = 1 << channel) & mask) { - mask &= ~channel_mask; - trident_ack_channel_interrupt(card, channel); - udelay(100); - state->dmabuf.update_flag |= ALI_ADDRESS_INT_UPDATE; - trident_update_ptr(state); - } - } - if (mask) { - for (i = 0; i < NR_HW_CH; i++) { - if (mask & (1 << i)) { - printk("ali: spurious channel irq %d.\n", i); - trident_ack_channel_interrupt(card, i); - trident_stop_voice(card, i); - trident_disable_voice_irq(card, i); - } - } - } -} - -/* Updating the values of counters of other_states' DMAs without lock -protection is no harm because all DMAs of multi-channels and interrupt -depend on a master state's DMA, and changing the counters of the master -state DMA is protected by a spinlock. -*/ -static int -ali_write_5_1(struct trident_state *state, const char __user *buf, - int cnt_for_multi_channel, unsigned int *copy_count, - unsigned int *state_cnt) -{ - - struct dmabuf *dmabuf = &state->dmabuf; - struct dmabuf *dmabuf_temp; - const char __user *buffer = buf; - unsigned swptr, other_dma_nums, sample_s; - unsigned int i, loop; - - other_dma_nums = 4; - sample_s = sample_size[dmabuf->fmt] >> 1; - swptr = dmabuf->swptr; - - if ((i = state->multi_channels_adjust_count) > 0) { - if (i == 1) { - if (copy_from_user(dmabuf->rawbuf + swptr, - buffer, sample_s)) - return -EFAULT; - seek_offset(swptr, buffer, cnt_for_multi_channel, - sample_s, *copy_count); - i--; - (*state_cnt) += sample_s; - state->multi_channels_adjust_count++; - } else - i = i - (state->chans_num - other_dma_nums); - for (; (i < other_dma_nums) && (cnt_for_multi_channel > 0); i++) { - dmabuf_temp = &state->other_states[i]->dmabuf; - if (copy_from_user(dmabuf_temp->rawbuf + dmabuf_temp->swptr, - buffer, sample_s)) - return -EFAULT; - seek_offset(dmabuf_temp->swptr, buffer, cnt_for_multi_channel, - sample_s, *copy_count); - } - if (cnt_for_multi_channel == 0) - state->multi_channels_adjust_count += i; - } - if (cnt_for_multi_channel > 0) { - loop = cnt_for_multi_channel / (state->chans_num * sample_s); - for (i = 0; i < loop; i++) { - if (copy_from_user(dmabuf->rawbuf + swptr, buffer, - sample_s * 2)) - return -EFAULT; - seek_offset(swptr, buffer, cnt_for_multi_channel, - sample_s * 2, *copy_count); - (*state_cnt) += (sample_s * 2); - - dmabuf_temp = &state->other_states[0]->dmabuf; - if (copy_from_user(dmabuf_temp->rawbuf + dmabuf_temp->swptr, - buffer, sample_s)) - return -EFAULT; - seek_offset(dmabuf_temp->swptr, buffer, cnt_for_multi_channel, - sample_s, *copy_count); - - dmabuf_temp = &state->other_states[1]->dmabuf; - if (copy_from_user(dmabuf_temp->rawbuf + dmabuf_temp->swptr, - buffer, sample_s)) - return -EFAULT; - seek_offset(dmabuf_temp->swptr, buffer, cnt_for_multi_channel, - sample_s, *copy_count); - - dmabuf_temp = &state->other_states[2]->dmabuf; - if (copy_from_user(dmabuf_temp->rawbuf + dmabuf_temp->swptr, - buffer, sample_s)) - return -EFAULT; - seek_offset(dmabuf_temp->swptr, buffer, cnt_for_multi_channel, - sample_s, *copy_count); - - dmabuf_temp = &state->other_states[3]->dmabuf; - if (copy_from_user(dmabuf_temp->rawbuf + dmabuf_temp->swptr, - buffer, sample_s)) - return -EFAULT; - seek_offset(dmabuf_temp->swptr, buffer, cnt_for_multi_channel, - sample_s, *copy_count); - } - - if (cnt_for_multi_channel > 0) { - state->multi_channels_adjust_count = cnt_for_multi_channel / sample_s; - - if (copy_from_user(dmabuf->rawbuf + swptr, buffer, sample_s)) - return -EFAULT; - seek_offset(swptr, buffer, cnt_for_multi_channel, - sample_s, *copy_count); - (*state_cnt) += sample_s; - - if (cnt_for_multi_channel > 0) { - if (copy_from_user(dmabuf->rawbuf + swptr, - buffer, sample_s)) - return -EFAULT; - seek_offset(swptr, buffer, cnt_for_multi_channel, - sample_s, *copy_count); - (*state_cnt) += sample_s; - - if (cnt_for_multi_channel > 0) { - int diff = state->chans_num - other_dma_nums; - loop = state->multi_channels_adjust_count - diff; - for (i = 0; i < loop; i++) { - dmabuf_temp = &state->other_states[i]->dmabuf; - if (copy_from_user(dmabuf_temp->rawbuf + - dmabuf_temp->swptr, - buffer, sample_s)) - return -EFAULT; - seek_offset(dmabuf_temp->swptr, buffer, - cnt_for_multi_channel, - sample_s, *copy_count); - } - } - } - } else - state->multi_channels_adjust_count = 0; - } - for (i = 0; i < other_dma_nums; i++) { - dmabuf_temp = &state->other_states[i]->dmabuf; - dmabuf_temp->swptr = dmabuf_temp->swptr % dmabuf_temp->dmasize; - } - return *state_cnt; -} - -static void -ali_free_other_states_resources(struct trident_state *state) -{ - int i; - struct trident_card *card = state->card; - struct trident_state *s; - unsigned other_states_count; - - other_states_count = state->chans_num - 2; /* except PCM L/R channels */ - for (i = 0; i < other_states_count; i++) { - s = state->other_states[i]; - dealloc_dmabuf(&s->dmabuf, card->pci_dev); - ali_disable_special_channel(s->card, s->dmabuf.channel->num); - state->card->free_pcm_channel(s->card, s->dmabuf.channel->num); - card->states[s->virt] = NULL; - kfree(s); - } -} - -static struct proc_dir_entry *res; - -static int -ali_write_proc(struct file *file, const char __user *buffer, unsigned long count, void *data) -{ - struct trident_card *card = (struct trident_card *) data; - unsigned long flags; - char c; - - if (count < 0) - return -EINVAL; - if (count == 0) - return 0; - if (get_user(c, buffer)) - return -EFAULT; - - spin_lock_irqsave(&card->lock, flags); - switch (c) { - case '0': - ali_setup_spdif_out(card, ALI_PCM_TO_SPDIF_OUT); - ali_disable_special_channel(card, ALI_SPDIF_OUT_CHANNEL); - break; - case '1': - ali_setup_spdif_out(card, ALI_SPDIF_OUT_TO_SPDIF_OUT | - ALI_SPDIF_OUT_PCM); - break; - case '2': - ali_setup_spdif_out(card, ALI_SPDIF_OUT_TO_SPDIF_OUT | - ALI_SPDIF_OUT_NON_PCM); - break; - case '3': - ali_disable_spdif_in(card); //default - break; - case '4': - ali_setup_spdif_in(card); - break; - } - spin_unlock_irqrestore(&card->lock, flags); - - return count; -} - -/* OSS /dev/mixer file operation methods */ -static int -trident_open_mixdev(struct inode *inode, struct file *file) -{ - int i = 0; - int minor = iminor(inode); - struct trident_card *card = devs; - - for (card = devs; card != NULL; card = card->next) - for (i = 0; i < NR_AC97; i++) - if (card->ac97_codec[i] != NULL && - card->ac97_codec[i]->dev_mixer == minor) - goto match; - - if (!card) { - return -ENODEV; - } - match: - file->private_data = card->ac97_codec[i]; - - return nonseekable_open(inode, file); -} - -static int -trident_ioctl_mixdev(struct inode *inode, struct file *file, unsigned int cmd, - unsigned long arg) -{ - struct ac97_codec *codec = (struct ac97_codec *) file->private_data; - - return codec->mixer_ioctl(codec, cmd, arg); -} - -static const struct file_operations trident_mixer_fops = { - .owner = THIS_MODULE, - .llseek = no_llseek, - .ioctl = trident_ioctl_mixdev, - .open = trident_open_mixdev, -}; - -static int -ali_reset_5451(struct trident_card *card) -{ - struct pci_dev *pci_dev = NULL; - unsigned int dwVal; - unsigned short wCount, wReg; - - pci_dev = pci_get_device(PCI_VENDOR_ID_AL, PCI_DEVICE_ID_AL_M1533, - pci_dev); - if (pci_dev == NULL) - return -1; - - pci_read_config_dword(pci_dev, 0x7c, &dwVal); - pci_write_config_dword(pci_dev, 0x7c, dwVal | 0x08000000); - udelay(5000); - pci_read_config_dword(pci_dev, 0x7c, &dwVal); - pci_write_config_dword(pci_dev, 0x7c, dwVal & 0xf7ffffff); - udelay(5000); - pci_dev_put(pci_dev); - - pci_dev = card->pci_dev; - if (pci_dev == NULL) - return -1; - - pci_read_config_dword(pci_dev, 0x44, &dwVal); - pci_write_config_dword(pci_dev, 0x44, dwVal | 0x000c0000); - udelay(500); - pci_read_config_dword(pci_dev, 0x44, &dwVal); - pci_write_config_dword(pci_dev, 0x44, dwVal & 0xfffbffff); - udelay(5000); - - /* TODO: recognize if we have a PM capable codec and only do this */ - /* if the codec is PM capable */ - wCount = 2000; - while (wCount--) { - wReg = ali_ac97_get(card, 0, AC97_POWER_CONTROL); - if ((wReg & 0x000f) == 0x000f) - return 0; - udelay(5000); - } - /* This is non fatal if you have a non PM capable codec.. */ - return 0; -} - -/* AC97 codec initialisation. */ -static int __devinit -trident_ac97_init(struct trident_card *card) -{ - int num_ac97 = 0; - unsigned long ready_2nd = 0; - struct ac97_codec *codec; - int i = 0; - - /* initialize controller side of AC link, and find out if secondary codes - really exist */ - switch (card->pci_id) { - case PCI_DEVICE_ID_ALI_5451: - if (ali_reset_5451(card)) { - printk(KERN_ERR "trident_ac97_init: error " - "resetting 5451.\n"); - return -1; - } - outl(0x80000001, TRID_REG(card, ALI_GLOBAL_CONTROL)); - outl(0x00000000, TRID_REG(card, T4D_AINTEN_A)); - outl(0xffffffff, TRID_REG(card, T4D_AINT_A)); - outl(0x00000000, TRID_REG(card, T4D_MUSICVOL_WAVEVOL)); - outb(0x10, TRID_REG(card, ALI_MPUR2)); - ready_2nd = inl(TRID_REG(card, ALI_SCTRL)); - ready_2nd &= 0x3fff; - outl(ready_2nd | PCMOUT | 0x8000, TRID_REG(card, ALI_SCTRL)); - ready_2nd = inl(TRID_REG(card, ALI_SCTRL)); - ready_2nd &= SI_AC97_SECONDARY_READY; - if (card->revision < ALI_5451_V02) - ready_2nd = 0; - break; - case PCI_DEVICE_ID_SI_7018: - /* disable AC97 GPIO interrupt */ - outl(0x00, TRID_REG(card, SI_AC97_GPIO)); - /* when power up the AC link is in cold reset mode so stop it */ - outl(PCMOUT | SURROUT | CENTEROUT | LFEOUT | SECONDARY_ID, - TRID_REG(card, SI_SERIAL_INTF_CTRL)); - /* it take a long time to recover from a cold reset */ - /* (especially when you have more than one codec) */ - udelay(2000); - ready_2nd = inl(TRID_REG(card, SI_SERIAL_INTF_CTRL)); - ready_2nd &= SI_AC97_SECONDARY_READY; - break; - case PCI_DEVICE_ID_TRIDENT_4DWAVE_DX: - /* playback on */ - outl(DX_AC97_PLAYBACK, TRID_REG(card, DX_ACR2_AC97_COM_STAT)); - break; - case PCI_DEVICE_ID_TRIDENT_4DWAVE_NX: - /* enable AC97 Output Slot 3,4 (PCM Left/Right Playback) */ - outl(NX_AC97_PCM_OUTPUT, TRID_REG(card, NX_ACR0_AC97_COM_STAT)); - ready_2nd = inl(TRID_REG(card, NX_ACR0_AC97_COM_STAT)); - ready_2nd &= NX_AC97_SECONDARY_READY; - break; - case PCI_DEVICE_ID_INTERG_5050: - /* disable AC97 GPIO interrupt */ - outl(0x00, TRID_REG(card, SI_AC97_GPIO)); - /* when power up, the AC link is in cold reset mode, so stop it */ - outl(PCMOUT | SURROUT | CENTEROUT | LFEOUT, - TRID_REG(card, SI_SERIAL_INTF_CTRL)); - /* it take a long time to recover from a cold reset (especially */ - /* when you have more than one codec) */ - udelay(2000); - ready_2nd = inl(TRID_REG(card, SI_SERIAL_INTF_CTRL)); - ready_2nd &= SI_AC97_SECONDARY_READY; - break; - } - - for (num_ac97 = 0; num_ac97 < NR_AC97; num_ac97++) { - if ((codec = ac97_alloc_codec()) == NULL) - return -ENOMEM; - - /* initialize some basic codec information, other fields */ - /* will be filled in ac97_probe_codec */ - codec->private_data = card; - codec->id = num_ac97; - - if (card->pci_id == PCI_DEVICE_ID_ALI_5451) { - codec->codec_read = ali_ac97_read; - codec->codec_write = ali_ac97_write; - } else { - codec->codec_read = trident_ac97_get; - codec->codec_write = trident_ac97_set; - } - - if (ac97_probe_codec(codec) == 0) - break; - - codec->dev_mixer = register_sound_mixer(&trident_mixer_fops, -1); - if (codec->dev_mixer < 0) { - printk(KERN_ERR "trident: couldn't register mixer!\n"); - ac97_release_codec(codec); - break; - } - - card->ac97_codec[num_ac97] = codec; - - /* if there is no secondary codec at all, don't probe any more */ - if (!ready_2nd) - break; - } - - if (card->pci_id == PCI_DEVICE_ID_ALI_5451) { - for (num_ac97 = 0; num_ac97 < NR_AC97; num_ac97++) { - if (card->ac97_codec[num_ac97] == NULL) - break; - for (i = 0; i < 64; i++) { - u16 reg = ali_ac97_get(card, num_ac97, i * 2); - card->mixer_regs[i][num_ac97] = reg; - } - } - } - return num_ac97 + 1; -} - -#ifdef SUPPORT_JOYSTICK -/* Gameport functions for the cards ADC gameport */ - -static unsigned char trident_game_read(struct gameport *gameport) -{ - struct trident_card *card = gameport->port_data; - - return inb(TRID_REG(card, T4D_GAME_LEG)); -} - -static void trident_game_trigger(struct gameport *gameport) -{ - struct trident_card *card = gameport->port_data; - - outb(0xff, TRID_REG(card, T4D_GAME_LEG)); -} - -static int trident_game_cooked_read(struct gameport *gameport, - int *axes, int *buttons) -{ - struct trident_card *card = gameport->port_data; - int i; - - *buttons = (~inb(TRID_REG(card, T4D_GAME_LEG)) >> 4) & 0xf; - - for (i = 0; i < 4; i++) { - axes[i] = inw(TRID_REG(card, T4D_GAME_AXD) + i * sizeof (u16)); - if (axes[i] == 0xffff) - axes[i] = -1; - } - - return 0; -} - -static int trident_game_open(struct gameport *gameport, int mode) -{ - struct trident_card *card = gameport->port_data; - - switch (mode) { - case GAMEPORT_MODE_COOKED: - outb(0x80, TRID_REG(card, T4D_GAME_CR)); - msleep(20); - return 0; - case GAMEPORT_MODE_RAW: - outb(0x00, TRID_REG(card, T4D_GAME_CR)); - return 0; - default: - return -1; - } - - return 0; -} - -static int __devinit trident_register_gameport(struct trident_card *card) -{ - struct gameport *gp; - - card->gameport = gp = gameport_allocate_port(); - if (!gp) { - printk(KERN_ERR "trident: can not allocate memory for gameport\n"); - return -ENOMEM; - } - - gameport_set_name(gp, "Trident 4DWave"); - gameport_set_phys(gp, "pci%s/gameport0", pci_name(card->pci_dev)); - gp->read = trident_game_read; - gp->trigger = trident_game_trigger; - gp->cooked_read = trident_game_cooked_read; - gp->open = trident_game_open; - gp->fuzz = 64; - gp->port_data = card; - - gameport_register_port(gp); - - return 0; -} - -static inline void trident_unregister_gameport(struct trident_card *card) -{ - if (card->gameport) - gameport_unregister_port(card->gameport); -} - -#else -static inline int trident_register_gameport(struct trident_card *card) { return -ENOSYS; } -static inline void trident_unregister_gameport(struct trident_card *card) { } -#endif /* SUPPORT_JOYSTICK */ - -/* install the driver, we do not allocate hardware channel nor DMA buffer */ -/* now, they are defered until "ACCESS" time (in prog_dmabuf called by */ -/* open/read/write/ioctl/mmap) */ -static int __devinit -trident_probe(struct pci_dev *pci_dev, const struct pci_device_id *pci_id) -{ - unsigned long iobase; - struct trident_card *card; - u8 bits; - u8 revision; - int i = 0; - u16 temp; - struct pci_dev *pci_dev_m1533 = NULL; - int rc = -ENODEV; - u64 dma_mask; - - if (pci_enable_device(pci_dev)) - goto out; - - if (pci_dev->device == PCI_DEVICE_ID_ALI_5451) - dma_mask = ALI_DMA_MASK; - else - dma_mask = TRIDENT_DMA_MASK; - if (pci_set_dma_mask(pci_dev, dma_mask)) { - printk(KERN_ERR "trident: architecture does not support" - " %s PCI busmaster DMA\n", - pci_dev->device == PCI_DEVICE_ID_ALI_5451 ? - "32-bit" : "30-bit"); - goto out; - } - pci_read_config_byte(pci_dev, PCI_CLASS_REVISION, &revision); - - if (pci_id->device == PCI_DEVICE_ID_INTERG_5050) - iobase = pci_resource_start(pci_dev, 1); - else - iobase = pci_resource_start(pci_dev, 0); - - if (!request_region(iobase, 256, card_names[pci_id->driver_data])) { - printk(KERN_ERR "trident: can't allocate I/O space at " - "0x%4.4lx\n", iobase); - goto out; - } - - rc = -ENOMEM; - if ((card = kzalloc(sizeof(*card), GFP_KERNEL)) == NULL) { - printk(KERN_ERR "trident: out of memory\n"); - goto out_release_region; - } - - init_timer(&card->timer); - card->iobase = iobase; - card->pci_dev = pci_dev_get(pci_dev); - card->pci_id = pci_id->device; - card->revision = revision; - card->irq = pci_dev->irq; - card->next = devs; - card->magic = TRIDENT_CARD_MAGIC; - card->banks[BANK_A].addresses = &bank_a_addrs; - card->banks[BANK_A].bitmap = 0UL; - card->banks[BANK_B].addresses = &bank_b_addrs; - card->banks[BANK_B].bitmap = 0UL; - - mutex_init(&card->open_mutex); - spin_lock_init(&card->lock); - init_timer(&card->timer); - - devs = card; - - pci_set_master(pci_dev); - - printk(KERN_INFO "trident: %s found at IO 0x%04lx, IRQ %d\n", - card_names[pci_id->driver_data], card->iobase, card->irq); - - if (card->pci_id == PCI_DEVICE_ID_ALI_5451) { - /* ALi channel Management */ - card->alloc_pcm_channel = ali_alloc_pcm_channel; - card->alloc_rec_pcm_channel = ali_alloc_rec_pcm_channel; - card->free_pcm_channel = ali_free_pcm_channel; - - card->address_interrupt = ali_address_interrupt; - - /* Added by Matt Wu 01-05-2001 for spdif in */ - card->multi_channel_use_count = 0; - card->rec_channel_use_count = 0; - - /* ALi SPDIF OUT function */ - if (card->revision == ALI_5451_V02) { - ali_setup_spdif_out(card, ALI_PCM_TO_SPDIF_OUT); - res = create_proc_entry("ALi5451", 0, NULL); - if (res) { - res->write_proc = ali_write_proc; - res->data = card; - } - } - - /* Add H/W Volume Control By Matt Wu Jul. 06, 2001 */ - card->hwvolctl = 0; - pci_dev_m1533 = pci_get_device(PCI_VENDOR_ID_AL, - PCI_DEVICE_ID_AL_M1533, - pci_dev_m1533); - rc = -ENODEV; - if (pci_dev_m1533 == NULL) - goto out_proc_fs; - pci_read_config_byte(pci_dev_m1533, 0x63, &bits); - if (bits & (1 << 5)) - card->hwvolctl = 1; - if (card->hwvolctl) { - /* Clear m1533 pci cfg 78h bit 30 to zero, which makes - GPIO11/12/13 work as ACGP_UP/DOWN/MUTE. */ - pci_read_config_byte(pci_dev_m1533, 0x7b, &bits); - bits &= 0xbf; /*clear bit 6 */ - pci_write_config_byte(pci_dev_m1533, 0x7b, bits); - } - pci_dev_put(pci_dev_m1533); - - } else if (card->pci_id == PCI_DEVICE_ID_INTERG_5050) { - card->alloc_pcm_channel = cyber_alloc_pcm_channel; - card->alloc_rec_pcm_channel = cyber_alloc_pcm_channel; - card->free_pcm_channel = cyber_free_pcm_channel; - card->address_interrupt = cyber_address_interrupt; - cyber_init_ritual(card); - } else { - card->alloc_pcm_channel = trident_alloc_pcm_channel; - card->alloc_rec_pcm_channel = trident_alloc_pcm_channel; - card->free_pcm_channel = trident_free_pcm_channel; - card->address_interrupt = trident_address_interrupt; - } - - /* claim our irq */ - rc = -ENODEV; - if (request_irq(card->irq, &trident_interrupt, IRQF_SHARED, - card_names[pci_id->driver_data], card)) { - printk(KERN_ERR "trident: unable to allocate irq %d\n", - card->irq); - goto out_proc_fs; - } - /* register /dev/dsp */ - if ((card->dev_audio = register_sound_dsp(&trident_audio_fops, -1)) < 0) { - printk(KERN_ERR "trident: couldn't register DSP device!\n"); - goto out_free_irq; - } - card->mixer_regs_ready = 0; - /* initialize AC97 codec and register /dev/mixer */ - if (trident_ac97_init(card) <= 0) { - /* unregister audio devices */ - for (i = 0; i < NR_AC97; i++) { - if (card->ac97_codec[i] != NULL) { - struct ac97_codec* codec = card->ac97_codec[i]; - unregister_sound_mixer(codec->dev_mixer); - ac97_release_codec(codec); - } - } - goto out_unregister_sound_dsp; - } - card->mixer_regs_ready = 1; - outl(0x00, TRID_REG(card, T4D_MUSICVOL_WAVEVOL)); - - if (card->pci_id == PCI_DEVICE_ID_ALI_5451) { - /* Add H/W Volume Control By Matt Wu Jul. 06, 2001 */ - if (card->hwvolctl) { - /* Enable GPIO IRQ (MISCINT bit 18h) */ - temp = inw(TRID_REG(card, T4D_MISCINT + 2)); - temp |= 0x0004; - outw(temp, TRID_REG(card, T4D_MISCINT + 2)); - - /* Enable H/W Volume Control GLOVAL CONTROL bit 0 */ - temp = inw(TRID_REG(card, ALI_GLOBAL_CONTROL)); - temp |= 0x0001; - outw(temp, TRID_REG(card, ALI_GLOBAL_CONTROL)); - - } - if (card->revision == ALI_5451_V02) - ali_close_multi_channels(); - /* edited by HMSEO for GT sound */ -#if defined(CONFIG_ALPHA_NAUTILUS) || defined(CONFIG_ALPHA_GENERIC) - { - u16 ac97_data; - extern struct hwrpb_struct *hwrpb; - - if ((hwrpb->sys_type) == 201) { - printk(KERN_INFO "trident: Running on Alpha system " - "type Nautilus\n"); - ac97_data = ali_ac97_get(card, 0, AC97_POWER_CONTROL); - ali_ac97_set(card, 0, AC97_POWER_CONTROL, - ac97_data | ALI_EAPD_POWER_DOWN); - } - } -#endif /* CONFIG_ALPHA_NAUTILUS || CONFIG_ALPHA_GENERIC */ - /* edited by HMSEO for GT sound */ - } - rc = 0; - pci_set_drvdata(pci_dev, card); - - /* Enable Address Engine Interrupts */ - trident_enable_loop_interrupts(card); - - /* Register gameport */ - trident_register_gameport(card); - -out: - return rc; - -out_unregister_sound_dsp: - unregister_sound_dsp(card->dev_audio); -out_free_irq: - free_irq(card->irq, card); -out_proc_fs: - pci_dev_put(card->pci_dev); - if (res) { - remove_proc_entry("ALi5451", NULL); - res = NULL; - } - kfree(card); - devs = NULL; -out_release_region: - release_region(iobase, 256); - return rc; -} - -static void __devexit -trident_remove(struct pci_dev *pci_dev) -{ - int i; - struct trident_card *card = pci_get_drvdata(pci_dev); - - /* - * Kill running timers before unload. We can't have them - * going off after rmmod! - */ - if (card->hwvolctl) - del_timer_sync(&card->timer); - - /* ALi S/PDIF and Power Management */ - if (card->pci_id == PCI_DEVICE_ID_ALI_5451) { - ali_setup_spdif_out(card, ALI_PCM_TO_SPDIF_OUT); - ali_disable_special_channel(card, ALI_SPDIF_OUT_CHANNEL); - ali_disable_spdif_in(card); - remove_proc_entry("ALi5451", NULL); - } - - /* Unregister gameport */ - trident_unregister_gameport(card); - - /* Kill interrupts, and SP/DIF */ - trident_disable_loop_interrupts(card); - - /* free hardware resources */ - free_irq(card->irq, card); - release_region(card->iobase, 256); - - /* unregister audio devices */ - for (i = 0; i < NR_AC97; i++) - if (card->ac97_codec[i] != NULL) { - unregister_sound_mixer(card->ac97_codec[i]->dev_mixer); - ac97_release_codec(card->ac97_codec[i]); - } - unregister_sound_dsp(card->dev_audio); - - pci_set_drvdata(pci_dev, NULL); - pci_dev_put(card->pci_dev); - kfree(card); -} - -MODULE_AUTHOR("Alan Cox, Aaron Holtzman, Ollie Lho, Ching Ling Lee, Muli Ben-Yehuda"); -MODULE_DESCRIPTION("Trident 4DWave/SiS 7018/ALi 5451 and Tvia/IGST CyberPro5050 PCI " - "Audio Driver"); -MODULE_LICENSE("GPL"); - -#define TRIDENT_MODULE_NAME "trident" - -static struct pci_driver trident_pci_driver = { - .name = TRIDENT_MODULE_NAME, - .id_table = trident_pci_tbl, - .probe = trident_probe, - .remove = __devexit_p(trident_remove), -#ifdef CONFIG_PM - .suspend = trident_suspend, - .resume = trident_resume -#endif -}; - -static int __init -trident_init_module(void) -{ - printk(KERN_INFO "Trident 4DWave/SiS 7018/ALi 5451,Tvia CyberPro " - "5050 PCI Audio, version " DRIVER_VERSION ", " __TIME__ " " - __DATE__ "\n"); - - return pci_register_driver(&trident_pci_driver); -} - -static void __exit -trident_cleanup_module(void) -{ - pci_unregister_driver(&trident_pci_driver); -} - -module_init(trident_init_module); -module_exit(trident_cleanup_module); diff --git a/sound/oss/trident.h b/sound/oss/trident.h deleted file mode 100644 index ff30a1d7c2f1..000000000000 --- a/sound/oss/trident.h +++ /dev/null @@ -1,358 +0,0 @@ -#ifndef __TRID4DWAVE_H -#define __TRID4DWAVE_H - -/* - * audio@tridentmicro.com - * Fri Feb 19 15:55:28 MST 1999 - * Definitions for Trident 4DWave DX/NX chips - * - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. - * - */ - -/* PCI vendor and device ID */ -#ifndef PCI_VENDOR_ID_TRIDENT -#define PCI_VENDOR_ID_TRIDENT 0x1023 -#endif - -#ifndef PCI_VENDOR_ID_SI -#define PCI_VENDOR_ID_SI 0x1039 -#endif - -#ifndef PCI_VENDOR_ID_ALI -#define PCI_VENDOR_ID_ALI 0x10b9 -#endif - -#ifndef PCI_DEVICE_ID_TRIDENT_4DWAVE_DX -#define PCI_DEVICE_ID_TRIDENT_4DWAVE_DX 0x2000 -#endif - -#ifndef PCI_DEVICE_ID_TRIDENT_4DWAVE_NX -#define PCI_DEVICE_ID_TRIDENT_4DWAVE_NX 0x2001 -#endif - -#ifndef PCI_DEVICE_ID_SI_7018 -#define PCI_DEVICE_ID_SI_7018 0x7018 -#endif - -#ifndef PCI_DEVICE_ID_ALI_5451 -#define PCI_DEVICE_ID_ALI_5451 0x5451 -#endif - -#ifndef PCI_DEVICE_ID_ALI_1533 -#define PCI_DEVICE_ID_ALI_1533 0x1533 -#endif - -#define CHANNEL_REGS 5 -#define CHANNEL_START 0xe0 // The first bytes of the contiguous register space. - -#define BANK_A 0 -#define BANK_B 1 -#define NR_BANKS 2 - -#define TRIDENT_FMT_STEREO 0x01 -#define TRIDENT_FMT_16BIT 0x02 -#define TRIDENT_FMT_MASK 0x03 - -#define DAC_RUNNING 0x01 -#define ADC_RUNNING 0x02 - -/* Register Addresses */ - -/* operational registers common to DX, NX, 7018 */ -enum trident_op_registers { - T4D_GAME_CR = 0x30, T4D_GAME_LEG = 0x31, - T4D_GAME_AXD = 0x34, - T4D_REC_CH = 0x70, - T4D_START_A = 0x80, T4D_STOP_A = 0x84, - T4D_DLY_A = 0x88, T4D_SIGN_CSO_A = 0x8c, - T4D_CSPF_A = 0x90, T4D_CEBC_A = 0x94, - T4D_AINT_A = 0x98, T4D_EINT_A = 0x9c, - T4D_LFO_GC_CIR = 0xa0, T4D_AINTEN_A = 0xa4, - T4D_MUSICVOL_WAVEVOL = 0xa8, T4D_SBDELTA_DELTA_R = 0xac, - T4D_MISCINT = 0xb0, T4D_START_B = 0xb4, - T4D_STOP_B = 0xb8, T4D_CSPF_B = 0xbc, - T4D_SBBL_SBCL = 0xc0, T4D_SBCTRL_SBE2R_SBDD = 0xc4, - T4D_STIMER = 0xc8, T4D_LFO_B_I2S_DELTA = 0xcc, - T4D_AINT_B = 0xd8, T4D_AINTEN_B = 0xdc, - ALI_MPUR2 = 0x22, ALI_GPIO = 0x7c, - ALI_EBUF1 = 0xf4, - ALI_EBUF2 = 0xf8 -}; - -enum ali_op_registers { - ALI_SCTRL = 0x48, - ALI_GLOBAL_CONTROL = 0xd4, - ALI_STIMER = 0xc8, - ALI_SPDIF_CS = 0x70, - ALI_SPDIF_CTRL = 0x74 -}; - -enum ali_registers_number { - ALI_GLOBAL_REGS = 56, - ALI_CHANNEL_REGS = 8, - ALI_MIXER_REGS = 20 -}; - -enum ali_sctrl_control_bit { - ALI_SPDIF_OUT_ENABLE = 0x20 -}; - -enum ali_global_control_bit { - ALI_SPDIF_OUT_SEL_PCM = 0x00000400, - ALI_SPDIF_IN_SUPPORT = 0x00000800, - ALI_SPDIF_OUT_CH_ENABLE = 0x00008000, - ALI_SPDIF_IN_CH_ENABLE = 0x00080000, - ALI_PCM_IN_DISABLE = 0x7fffffff, - ALI_PCM_IN_ENABLE = 0x80000000, - ALI_SPDIF_IN_CH_DISABLE = 0xfff7ffff, - ALI_SPDIF_OUT_CH_DISABLE = 0xffff7fff, - ALI_SPDIF_OUT_SEL_SPDIF = 0xfffffbff - -}; - -enum ali_spdif_control_bit { - ALI_SPDIF_IN_FUNC_ENABLE = 0x02, - ALI_SPDIF_IN_CH_STATUS = 0x40, - ALI_SPDIF_OUT_CH_STATUS = 0xbf - -}; - -enum ali_control_all { - ALI_DISABLE_ALL_IRQ = 0, - ALI_CHANNELS = 32, - ALI_STOP_ALL_CHANNELS = 0xffffffff, - ALI_MULTI_CHANNELS_START_STOP = 0x07800000 -}; - -enum ali_EMOD_control_bit { - ALI_EMOD_DEC = 0x00000000, - ALI_EMOD_INC = 0x10000000, - ALI_EMOD_Delay = 0x20000000, - ALI_EMOD_Still = 0x30000000 -}; - -enum ali_pcm_in_channel_num { - ALI_NORMAL_CHANNEL = 0, - ALI_SPDIF_OUT_CHANNEL = 15, - ALI_SPDIF_IN_CHANNEL = 19, - ALI_LEF_CHANNEL = 23, - ALI_CENTER_CHANNEL = 24, - ALI_SURR_RIGHT_CHANNEL = 25, - ALI_SURR_LEFT_CHANNEL = 26, - ALI_PCM_IN_CHANNEL = 31 -}; - -enum ali_pcm_out_channel_num { - ALI_PCM_OUT_CHANNEL_FIRST = 0, - ALI_PCM_OUT_CHANNEL_LAST = 31 -}; - -enum ali_ac97_power_control_bit { - ALI_EAPD_POWER_DOWN = 0x8000 -}; - -enum ali_update_ptr_flags { - ALI_ADDRESS_INT_UPDATE = 0x01 -}; - -enum ali_revision { - ALI_5451_V02 = 0x02 -}; - -enum ali_spdif_out_control { - ALI_PCM_TO_SPDIF_OUT = 0, - ALI_SPDIF_OUT_TO_SPDIF_OUT = 1, - ALI_SPDIF_OUT_PCM = 0, - ALI_SPDIF_OUT_NON_PCM = 2 -}; - -/* S/PDIF Operational Registers for 4D-NX */ -enum nx_spdif_registers { - NX_SPCTRL_SPCSO = 0x24, NX_SPLBA = 0x28, - NX_SPESO = 0x2c, NX_SPCSTATUS = 0x64 -}; - -/* OP registers to access each hardware channel */ -enum channel_registers { - CH_DX_CSO_ALPHA_FMS = 0xe0, CH_DX_ESO_DELTA = 0xe8, - CH_DX_FMC_RVOL_CVOL = 0xec, - CH_NX_DELTA_CSO = 0xe0, CH_NX_DELTA_ESO = 0xe8, - CH_NX_ALPHA_FMS_FMC_RVOL_CVOL = 0xec, - CH_LBA = 0xe4, - CH_GVSEL_PAN_VOL_CTRL_EC = 0xf0 -}; - -/* registers to read/write/control AC97 codec */ -enum dx_ac97_registers { - DX_ACR0_AC97_W = 0x40, DX_ACR1_AC97_R = 0x44, - DX_ACR2_AC97_COM_STAT = 0x48 -}; - -enum nx_ac97_registers { - NX_ACR0_AC97_COM_STAT = 0x40, NX_ACR1_AC97_W = 0x44, - NX_ACR2_AC97_R_PRIMARY = 0x48, NX_ACR3_AC97_R_SECONDARY = 0x4c -}; - -enum si_ac97_registers { - SI_AC97_WRITE = 0x40, SI_AC97_READ = 0x44, - SI_SERIAL_INTF_CTRL = 0x48, SI_AC97_GPIO = 0x4c -}; - -enum ali_ac97_registers { - ALI_AC97_WRITE = 0x40, ALI_AC97_READ = 0x44 -}; - -/* Bit mask for operational registers */ -#define AC97_REG_ADDR 0x000000ff - -enum ali_ac97_bits { - ALI_AC97_BUSY_WRITE = 0x8000, ALI_AC97_BUSY_READ = 0x8000, - ALI_AC97_WRITE_ACTION = 0x8000, ALI_AC97_READ_ACTION = 0x8000, - ALI_AC97_AUDIO_BUSY = 0x4000, ALI_AC97_SECONDARY = 0x0080, - ALI_AC97_READ_MIXER_REGISTER = 0xfeff, - ALI_AC97_WRITE_MIXER_REGISTER = 0x0100 -}; - -enum sis7018_ac97_bits { - SI_AC97_BUSY_WRITE = 0x8000, SI_AC97_BUSY_READ = 0x8000, - SI_AC97_AUDIO_BUSY = 0x4000, SI_AC97_MODEM_BUSY = 0x2000, - SI_AC97_SECONDARY = 0x0080 -}; - -enum trident_dx_ac97_bits { - DX_AC97_BUSY_WRITE = 0x8000, DX_AC97_BUSY_READ = 0x8000, - DX_AC97_READY = 0x0010, DX_AC97_RECORD = 0x0008, - DX_AC97_PLAYBACK = 0x0002 -}; - -enum trident_nx_ac97_bits { - /* ACR1-3 */ - NX_AC97_BUSY_WRITE = 0x0800, NX_AC97_BUSY_READ = 0x0800, - NX_AC97_BUSY_DATA = 0x0400, NX_AC97_WRITE_SECONDARY = 0x0100, - /* ACR0 */ - NX_AC97_SECONDARY_READY = 0x0040, NX_AC97_SECONDARY_RECORD = 0x0020, - NX_AC97_SURROUND_OUTPUT = 0x0010, - NX_AC97_PRIMARY_READY = 0x0008, NX_AC97_PRIMARY_RECORD = 0x0004, - NX_AC97_PCM_OUTPUT = 0x0002, - NX_AC97_WARM_RESET = 0x0001 -}; - -enum serial_intf_ctrl_bits { - WARM_REST = 0x00000001, COLD_RESET = 0x00000002, - I2S_CLOCK = 0x00000004, PCM_SEC_AC97= 0x00000008, - AC97_DBL_RATE = 0x00000010, SPDIF_EN = 0x00000020, - I2S_OUTPUT_EN = 0x00000040, I2S_INPUT_EN = 0x00000080, - PCMIN = 0x00000100, LINE1IN = 0x00000200, - MICIN = 0x00000400, LINE2IN = 0x00000800, - HEAD_SET_IN = 0x00001000, GPIOIN = 0x00002000, - /* 7018 spec says id = 01 but the demo board routed to 10 - SECONDARY_ID= 0x00004000, */ - SECONDARY_ID= 0x00004000, - PCMOUT = 0x00010000, SURROUT = 0x00020000, - CENTEROUT = 0x00040000, LFEOUT = 0x00080000, - LINE1OUT = 0x00100000, LINE2OUT = 0x00200000, - GPIOOUT = 0x00400000, - SI_AC97_PRIMARY_READY = 0x01000000, - SI_AC97_SECONDARY_READY = 0x02000000, -}; - -enum global_control_bits { - CHANNLE_IDX = 0x0000003f, PB_RESET = 0x00000100, - PAUSE_ENG = 0x00000200, - OVERRUN_IE = 0x00000400, UNDERRUN_IE = 0x00000800, - ENDLP_IE = 0x00001000, MIDLP_IE = 0x00002000, - ETOG_IE = 0x00004000, - EDROP_IE = 0x00008000, BANK_B_EN = 0x00010000 -}; - -enum channel_control_bits { - CHANNEL_LOOP = 0x00001000, CHANNEL_SIGNED = 0x00002000, - CHANNEL_STEREO = 0x00004000, CHANNEL_16BITS = 0x00008000, -}; - -enum channel_attribute { - /* playback/record select */ - CHANNEL_PB = 0x0000, CHANNEL_SPC_PB = 0x4000, - CHANNEL_REC = 0x8000, CHANNEL_REC_PB = 0xc000, - /* playback destination/record source select */ - MODEM_LINE1 = 0x0000, MODEM_LINE2 = 0x0400, - PCM_LR = 0x0800, HSET = 0x0c00, - I2S_LR = 0x1000, CENTER_LFE = 0x1400, - SURR_LR = 0x1800, SPDIF_LR = 0x1c00, - MIC = 0x1400, - /* mist stuff */ - MONO_LEFT = 0x0000, MONO_RIGHT = 0x0100, - MONO_MIX = 0x0200, SRC_ENABLE = 0x0080, -}; - -enum miscint_bits { - PB_UNDERRUN_IRO = 0x00000001, REC_OVERRUN_IRQ = 0x00000002, - SB_IRQ = 0x00000004, MPU401_IRQ = 0x00000008, - OPL3_IRQ = 0x00000010, ADDRESS_IRQ = 0x00000020, - ENVELOPE_IRQ = 0x00000040, ST_IRQ = 0x00000080, - PB_UNDERRUN = 0x00000100, REC_OVERRUN = 0x00000200, - MIXER_UNDERFLOW = 0x00000400, MIXER_OVERFLOW = 0x00000800, - ST_TARGET_REACHED = 0x00008000, PB_24K_MODE = 0x00010000, - ST_IRQ_EN = 0x00800000, ACGPIO_IRQ = 0x01000000 -}; - -#define TRID_REG( trident, x ) ( (trident) -> iobase + (x) ) - -#define CYBER_PORT_AUDIO 0x3CE -#define CYBER_IDX_AUDIO_ENABLE 0x7B -#define CYBER_BMSK_AUDIO_INT_ENABLE 0x09 -#define CYBER_BMSK_AUENZ 0x01 -#define CYBER_BMSK_AUENZ_ENABLE 0x00 -#define CYBER_IDX_IRQ_ENABLE 0x12 - -#define VALIDATE_MAGIC(FOO,MAG) \ -({ \ - if (!(FOO) || (FOO)->magic != MAG) { \ - printk(invalid_magic,__func__); \ - return -ENXIO; \ - } \ -}) - -#define VALIDATE_STATE(a) VALIDATE_MAGIC(a,TRIDENT_STATE_MAGIC) -#define VALIDATE_CARD(a) VALIDATE_MAGIC(a,TRIDENT_CARD_MAGIC) - -static inline unsigned ld2(unsigned int x) -{ - unsigned r = 0; - - if (x >= 0x10000) { - x >>= 16; - r += 16; - } - if (x >= 0x100) { - x >>= 8; - r += 8; - } - if (x >= 0x10) { - x >>= 4; - r += 4; - } - if (x >= 4) { - x >>= 2; - r += 2; - } - if (x >= 2) - r++; - return r; -} - -#endif /* __TRID4DWAVE_H */ From 76a6f3dc9a7108785c145a298f82c72f9208fe17 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 23 Jul 2008 21:29:15 -0700 Subject: [PATCH 276/857] CONFIG_SOUND_WM97XX: remove stale makefile line The driver is gone for a long time. Reported-by: Robert P. J. Day Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- sound/oss/Makefile | 2 -- 1 file changed, 2 deletions(-) diff --git a/sound/oss/Makefile b/sound/oss/Makefile index 3a141474fb73..c611514f7ff1 100644 --- a/sound/oss/Makefile +++ b/sound/oss/Makefile @@ -31,8 +31,6 @@ obj-$(CONFIG_SOUND_VWSND) += vwsnd.o obj-$(CONFIG_SOUND_AU1550_AC97) += au1550_ac97.o ac97_codec.o obj-$(CONFIG_SOUND_BCM_CS4297A) += swarm_cs4297a.o -obj-$(CONFIG_SOUND_WM97XX) += ac97_plugin_wm97xx.o - obj-$(CONFIG_DMASOUND) += dmasound/ # Declare multi-part drivers. From 6e2c10a12a2170856f5582d62d583cbcd1cb5eaf Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Wed, 23 Jul 2008 21:29:15 -0700 Subject: [PATCH 277/857] binfmt_misc: use simple_read_from_buffer() Signed-off-by: Akinobu Mita Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/binfmt_misc.c | 20 +++----------------- 1 file changed, 3 insertions(+), 17 deletions(-) diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c index 7191306367c5..756205314c24 100644 --- a/fs/binfmt_misc.c +++ b/fs/binfmt_misc.c @@ -27,6 +27,7 @@ #include #include #include +#include #include @@ -535,31 +536,16 @@ static ssize_t bm_entry_read(struct file * file, char __user * buf, size_t nbytes, loff_t *ppos) { Node *e = file->f_path.dentry->d_inode->i_private; - loff_t pos = *ppos; ssize_t res; char *page; - int len; if (!(page = (char*) __get_free_page(GFP_KERNEL))) return -ENOMEM; entry_status(e, page); - len = strlen(page); - res = -EINVAL; - if (pos < 0) - goto out; - res = 0; - if (pos >= len) - goto out; - if (len < pos + nbytes) - nbytes = len - pos; - res = -EFAULT; - if (copy_to_user(buf, page + pos, nbytes)) - goto out; - *ppos = pos + nbytes; - res = nbytes; -out: + res = simple_read_from_buffer(buf, nbytes, ppos, page, strlen(page)); + free_page((unsigned long) page); return res; } From a677a039be7243357d93502bff2b40850c942e2d Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 23 Jul 2008 21:29:17 -0700 Subject: [PATCH 278/857] flag parameters: socket and socketpair This patch adds support for flag values which are ORed to the type passwd to socket and socketpair. The additional code is minimal. The flag values in this implementation can and must match the O_* flags. This avoids overhead in the conversion. The internal functions sock_alloc_fd and sock_map_fd get a new parameters and all callers are changed. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #include #include #include #include #include #define PORT 57392 /* For Linux these must be the same. */ #define SOCK_CLOEXEC O_CLOEXEC int main (void) { int fd; fd = socket (PF_INET, SOCK_STREAM, 0); if (fd == -1) { puts ("socket(0) failed"); return 1; } int coe = fcntl (fd, F_GETFD); if (coe == -1) { puts ("fcntl failed"); return 1; } if (coe & FD_CLOEXEC) { puts ("socket(0) set close-on-exec flag"); return 1; } close (fd); fd = socket (PF_INET, SOCK_STREAM|SOCK_CLOEXEC, 0); if (fd == -1) { puts ("socket(SOCK_CLOEXEC) failed"); return 1; } coe = fcntl (fd, F_GETFD); if (coe == -1) { puts ("fcntl failed"); return 1; } if ((coe & FD_CLOEXEC) == 0) { puts ("socket(SOCK_CLOEXEC) does not set close-on-exec flag"); return 1; } close (fd); int fds[2]; if (socketpair (PF_UNIX, SOCK_STREAM, 0, fds) == -1) { puts ("socketpair(0) failed"); return 1; } for (int i = 0; i < 2; ++i) { coe = fcntl (fds[i], F_GETFD); if (coe == -1) { puts ("fcntl failed"); return 1; } if (coe & FD_CLOEXEC) { printf ("socketpair(0) set close-on-exec flag for fds[%d]\n", i); return 1; } close (fds[i]); } if (socketpair (PF_UNIX, SOCK_STREAM|SOCK_CLOEXEC, 0, fds) == -1) { puts ("socketpair(SOCK_CLOEXEC) failed"); return 1; } for (int i = 0; i < 2; ++i) { coe = fcntl (fds[i], F_GETFD); if (coe == -1) { puts ("fcntl failed"); return 1; } if ((coe & FD_CLOEXEC) == 0) { printf ("socketpair(SOCK_CLOEXEC) does not set close-on-exec flag for fds[%d]\n", i); return 1; } close (fds[i]); } puts ("OK"); return 0; } ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Signed-off-by: Ulrich Drepper Acked-by: Davide Libenzi Cc: Michael Kerrisk Cc: "David S. Miller" Cc: Ralf Baechle Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-mips/socket.h | 7 +++++++ include/linux/net.h | 9 ++++++++- net/9p/trans_fd.c | 2 +- net/sctp/socket.c | 2 +- net/socket.c | 28 ++++++++++++++++++++-------- 5 files changed, 37 insertions(+), 11 deletions(-) diff --git a/include/asm-mips/socket.h b/include/asm-mips/socket.h index 63f60254d308..facc2d7a87ca 100644 --- a/include/asm-mips/socket.h +++ b/include/asm-mips/socket.h @@ -102,6 +102,13 @@ enum sock_type { }; #define SOCK_MAX (SOCK_PACKET + 1) +/* Mask which covers at least up to SOCK_MASK-1. The + * * remaining bits are used as flags. */ +#define SOCK_TYPE_MASK 0xf + +/* Flags for socket, socketpair, paccept */ +#define SOCK_CLOEXEC O_CLOEXEC +#define SOCK_NONBLOCK O_NONBLOCK #define ARCH_HAS_SOCKET_TYPES 1 diff --git a/include/linux/net.h b/include/linux/net.h index 150a48c68d52..8b5383c45b45 100644 --- a/include/linux/net.h +++ b/include/linux/net.h @@ -20,6 +20,7 @@ #include #include +#include /* For O_CLOEXEC */ #include struct poll_table_struct; @@ -94,6 +95,12 @@ enum sock_type { }; #define SOCK_MAX (SOCK_PACKET + 1) +/* Mask which covers at least up to SOCK_MASK-1. The + * remaining bits are used as flags. */ +#define SOCK_TYPE_MASK 0xf + +/* Flags for socket, socketpair, paccept */ +#define SOCK_CLOEXEC O_CLOEXEC #endif /* ARCH_HAS_SOCKET_TYPES */ @@ -208,7 +215,7 @@ extern int sock_sendmsg(struct socket *sock, struct msghdr *msg, size_t len); extern int sock_recvmsg(struct socket *sock, struct msghdr *msg, size_t size, int flags); -extern int sock_map_fd(struct socket *sock); +extern int sock_map_fd(struct socket *sock, int flags); extern struct socket *sockfd_lookup(int fd, int *err); #define sockfd_put(sock) fput(sock->file) extern int net_ratelimit(void); diff --git a/net/9p/trans_fd.c b/net/9p/trans_fd.c index 4507f744f44e..cdf137af7adc 100644 --- a/net/9p/trans_fd.c +++ b/net/9p/trans_fd.c @@ -1285,7 +1285,7 @@ static int p9_socket_open(struct p9_trans *trans, struct socket *csocket) int fd, ret; csocket->sk->sk_allocation = GFP_NOIO; - fd = sock_map_fd(csocket); + fd = sock_map_fd(csocket, 0); if (fd < 0) { P9_EPRINTK(KERN_ERR, "p9_socket_open: failed to map fd\n"); return fd; diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 79bece16aede..dbb79adf8f3c 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -3910,7 +3910,7 @@ static int sctp_getsockopt_peeloff(struct sock *sk, int len, char __user *optval goto out; /* Map the socket to an unused fd that can be returned to the user. */ - retval = sock_map_fd(newsock); + retval = sock_map_fd(newsock, 0); if (retval < 0) { sock_release(newsock); goto out; diff --git a/net/socket.c b/net/socket.c index 1ba57d888981..64601f900352 100644 --- a/net/socket.c +++ b/net/socket.c @@ -349,11 +349,11 @@ static struct dentry_operations sockfs_dentry_operations = { * but we take care of internal coherence yet. */ -static int sock_alloc_fd(struct file **filep) +static int sock_alloc_fd(struct file **filep, int flags) { int fd; - fd = get_unused_fd(); + fd = get_unused_fd_flags(flags); if (likely(fd >= 0)) { struct file *file = get_empty_filp(); @@ -396,10 +396,10 @@ static int sock_attach_fd(struct socket *sock, struct file *file) return 0; } -int sock_map_fd(struct socket *sock) +int sock_map_fd(struct socket *sock, int flags) { struct file *newfile; - int fd = sock_alloc_fd(&newfile); + int fd = sock_alloc_fd(&newfile, flags); if (likely(fd >= 0)) { int err = sock_attach_fd(sock, newfile); @@ -1218,12 +1218,18 @@ asmlinkage long sys_socket(int family, int type, int protocol) { int retval; struct socket *sock; + int flags; + + flags = type & ~SOCK_TYPE_MASK; + if (flags & ~SOCK_CLOEXEC) + return -EINVAL; + type &= SOCK_TYPE_MASK; retval = sock_create(family, type, protocol, &sock); if (retval < 0) goto out; - retval = sock_map_fd(sock); + retval = sock_map_fd(sock, flags & O_CLOEXEC); if (retval < 0) goto out_release; @@ -1246,6 +1252,12 @@ asmlinkage long sys_socketpair(int family, int type, int protocol, struct socket *sock1, *sock2; int fd1, fd2, err; struct file *newfile1, *newfile2; + int flags; + + flags = type & ~SOCK_TYPE_MASK; + if (flags & ~SOCK_CLOEXEC) + return -EINVAL; + type &= SOCK_TYPE_MASK; /* * Obtain the first socket and check if the underlying protocol @@ -1264,13 +1276,13 @@ asmlinkage long sys_socketpair(int family, int type, int protocol, if (err < 0) goto out_release_both; - fd1 = sock_alloc_fd(&newfile1); + fd1 = sock_alloc_fd(&newfile1, flags & O_CLOEXEC); if (unlikely(fd1 < 0)) { err = fd1; goto out_release_both; } - fd2 = sock_alloc_fd(&newfile2); + fd2 = sock_alloc_fd(&newfile2, flags & O_CLOEXEC); if (unlikely(fd2 < 0)) { err = fd2; put_filp(newfile1); @@ -1426,7 +1438,7 @@ asmlinkage long sys_accept(int fd, struct sockaddr __user *upeer_sockaddr, */ __module_get(newsock->ops->owner); - newfd = sock_alloc_fd(&newfile); + newfd = sock_alloc_fd(&newfile, 0); if (unlikely(newfd < 0)) { err = newfd; sock_release(newsock); From aaca0bdca573f3f51ea03139f9c7289541e7bca3 Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 23 Jul 2008 21:29:20 -0700 Subject: [PATCH 279/857] flag parameters: paccept This patch is by far the most complex in the series. It adds a new syscall paccept. This syscall differs from accept in that it adds (at the userlevel) two additional parameters: - a signal mask - a flags value The flags parameter can be used to set flag like SOCK_CLOEXEC. This is imlpemented here as well. Some people argued that this is a property which should be inherited from the file desriptor for the server but this is against POSIX. Additionally, we really want the signal mask parameter as well (similar to pselect, ppoll, etc). So an interface change in inevitable. The flag value is the same as for socket and socketpair. I think diverging here will only create confusion. Similar to the filesystem interfaces where the use of the O_* constants differs, it is acceptable here. The signal mask is handled as for pselect etc. The mask is temporarily installed for the thread and removed before the call returns. I modeled the code after pselect. If there is a problem it's likely also in pselect. For architectures which use socketcall I maintained this interface instead of adding a system call. The symmetry shouldn't be broken. The following test must be adjusted for architectures other than x86 and x86-64 and in case the syscall numbers changed. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #include #include #include #include #include #include #include #include #include #ifndef __NR_paccept # ifdef __x86_64__ # define __NR_paccept 288 # elif defined __i386__ # define SYS_PACCEPT 18 # define USE_SOCKETCALL 1 # else # error "need __NR_paccept" # endif #endif #ifdef USE_SOCKETCALL # define paccept(fd, addr, addrlen, mask, flags) \ ({ long args[6] = { \ (long) fd, (long) addr, (long) addrlen, (long) mask, 8, (long) flags }; \ syscall (__NR_socketcall, SYS_PACCEPT, args); }) #else # define paccept(fd, addr, addrlen, mask, flags) \ syscall (__NR_paccept, fd, addr, addrlen, mask, 8, flags) #endif #define PORT 57392 #define SOCK_CLOEXEC O_CLOEXEC static pthread_barrier_t b; static void * tf (void *arg) { pthread_barrier_wait (&b); int s = socket (AF_INET, SOCK_STREAM, 0); struct sockaddr_in sin; sin.sin_family = AF_INET; sin.sin_addr.s_addr = htonl (INADDR_LOOPBACK); sin.sin_port = htons (PORT); connect (s, (const struct sockaddr *) &sin, sizeof (sin)); close (s); pthread_barrier_wait (&b); s = socket (AF_INET, SOCK_STREAM, 0); sin.sin_port = htons (PORT); connect (s, (const struct sockaddr *) &sin, sizeof (sin)); close (s); pthread_barrier_wait (&b); pthread_barrier_wait (&b); sleep (2); pthread_kill ((pthread_t) arg, SIGUSR1); return NULL; } static void handler (int s) { } int main (void) { pthread_barrier_init (&b, NULL, 2); struct sockaddr_in sin; pthread_t th; if (pthread_create (&th, NULL, tf, (void *) pthread_self ()) != 0) { puts ("pthread_create failed"); return 1; } int s = socket (AF_INET, SOCK_STREAM, 0); int reuse = 1; setsockopt (s, SOL_SOCKET, SO_REUSEADDR, &reuse, sizeof (reuse)); sin.sin_family = AF_INET; sin.sin_addr.s_addr = htonl (INADDR_LOOPBACK); sin.sin_port = htons (PORT); bind (s, (struct sockaddr *) &sin, sizeof (sin)); listen (s, SOMAXCONN); pthread_barrier_wait (&b); int s2 = paccept (s, NULL, 0, NULL, 0); if (s2 < 0) { puts ("paccept(0) failed"); return 1; } int coe = fcntl (s2, F_GETFD); if (coe & FD_CLOEXEC) { puts ("paccept(0) set close-on-exec-flag"); return 1; } close (s2); pthread_barrier_wait (&b); s2 = paccept (s, NULL, 0, NULL, SOCK_CLOEXEC); if (s2 < 0) { puts ("paccept(SOCK_CLOEXEC) failed"); return 1; } coe = fcntl (s2, F_GETFD); if ((coe & FD_CLOEXEC) == 0) { puts ("paccept(SOCK_CLOEXEC) does not set close-on-exec flag"); return 1; } close (s2); pthread_barrier_wait (&b); struct sigaction sa; sa.sa_handler = handler; sa.sa_flags = 0; sigemptyset (&sa.sa_mask); sigaction (SIGUSR1, &sa, NULL); sigset_t ss; pthread_sigmask (SIG_SETMASK, NULL, &ss); sigaddset (&ss, SIGUSR1); pthread_sigmask (SIG_SETMASK, &ss, NULL); sigdelset (&ss, SIGUSR1); alarm (4); pthread_barrier_wait (&b); errno = 0 ; s2 = paccept (s, NULL, 0, &ss, 0); if (s2 != -1 || errno != EINTR) { puts ("paccept did not fail with EINTR"); return 1; } close (s); puts ("OK"); return 0; } ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ [akpm@linux-foundation.org: make it compile] [akpm@linux-foundation.org: add sys_ni stub] Signed-off-by: Ulrich Drepper Acked-by: Davide Libenzi Cc: Michael Kerrisk Cc: Cc: "David S. Miller" Cc: Roland McGrath Cc: Kyle McMartin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-alpha/socket.h | 5 +++ include/asm-parisc/socket.h | 5 +++ include/asm-x86/unistd_64.h | 2 + include/linux/net.h | 3 ++ include/linux/syscalls.h | 2 + kernel/sys_ni.c | 1 + net/compat.c | 52 ++++++++++++++++++++++-- net/socket.c | 81 +++++++++++++++++++++++++++++++++---- 8 files changed, 139 insertions(+), 12 deletions(-) diff --git a/include/asm-alpha/socket.h b/include/asm-alpha/socket.h index 08c979319929..a1057c2d95e7 100644 --- a/include/asm-alpha/socket.h +++ b/include/asm-alpha/socket.h @@ -62,4 +62,9 @@ #define SO_MARK 36 +/* O_NONBLOCK clashes with the bits used for socket types. Therefore we + * have to define SOCK_NONBLOCK to a different value here. + */ +#define SOCK_NONBLOCK 0x40000000 + #endif /* _ASM_SOCKET_H */ diff --git a/include/asm-parisc/socket.h b/include/asm-parisc/socket.h index 69a7a0d30b02..fba402c95ac2 100644 --- a/include/asm-parisc/socket.h +++ b/include/asm-parisc/socket.h @@ -54,4 +54,9 @@ #define SO_MARK 0x401f +/* O_NONBLOCK clashes with the bits used for socket types. Therefore we + * have to define SOCK_NONBLOCK to a different value here. + */ +#define SOCK_NONBLOCK 0x40000000 + #endif /* _ASM_SOCKET_H */ diff --git a/include/asm-x86/unistd_64.h b/include/asm-x86/unistd_64.h index 9c1a4a3470d9..e323994a370f 100644 --- a/include/asm-x86/unistd_64.h +++ b/include/asm-x86/unistd_64.h @@ -639,6 +639,8 @@ __SYSCALL(__NR_fallocate, sys_fallocate) __SYSCALL(__NR_timerfd_settime, sys_timerfd_settime) #define __NR_timerfd_gettime 287 __SYSCALL(__NR_timerfd_gettime, sys_timerfd_gettime) +#define __NR_paccept 288 +__SYSCALL(__NR_paccept, sys_paccept) #ifndef __NO_STUBS diff --git a/include/linux/net.h b/include/linux/net.h index 8b5383c45b45..3a9b06d4d0fe 100644 --- a/include/linux/net.h +++ b/include/linux/net.h @@ -47,6 +47,7 @@ struct net; #define SYS_GETSOCKOPT 15 /* sys_getsockopt(2) */ #define SYS_SENDMSG 16 /* sys_sendmsg(2) */ #define SYS_RECVMSG 17 /* sys_recvmsg(2) */ +#define SYS_PACCEPT 18 /* sys_paccept(2) */ typedef enum { SS_FREE = 0, /* not allocated */ @@ -219,6 +220,8 @@ extern int sock_map_fd(struct socket *sock, int flags); extern struct socket *sockfd_lookup(int fd, int *err); #define sockfd_put(sock) fput(sock->file) extern int net_ratelimit(void); +extern long do_accept(int fd, struct sockaddr __user *upeer_sockaddr, + int __user *upeer_addrlen, int flags); #define net_random() random32() #define net_srandom(seed) srandom32((__force u32)seed) diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 4394dadff813..2a2a40af6b2c 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -409,6 +409,8 @@ asmlinkage long sys_getsockopt(int fd, int level, int optname, asmlinkage long sys_bind(int, struct sockaddr __user *, int); asmlinkage long sys_connect(int, struct sockaddr __user *, int); asmlinkage long sys_accept(int, struct sockaddr __user *, int __user *); +asmlinkage long sys_paccept(int, struct sockaddr __user *, int __user *, + const sigset_t *, size_t, int); asmlinkage long sys_getsockname(int, struct sockaddr __user *, int __user *); asmlinkage long sys_getpeername(int, struct sockaddr __user *, int __user *); asmlinkage long sys_send(int, void __user *, size_t, unsigned); diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c index 0fea0ee12da9..2f0b8a2e600f 100644 --- a/kernel/sys_ni.c +++ b/kernel/sys_ni.c @@ -31,6 +31,7 @@ cond_syscall(sys_socketpair); cond_syscall(sys_bind); cond_syscall(sys_listen); cond_syscall(sys_accept); +cond_syscall(sys_paccept); cond_syscall(sys_connect); cond_syscall(sys_getsockname); cond_syscall(sys_getpeername); diff --git a/net/compat.c b/net/compat.c index 6e1b03b51933..67fb6a3834a3 100644 --- a/net/compat.c +++ b/net/compat.c @@ -722,9 +722,10 @@ EXPORT_SYMBOL(compat_mc_getsockopt); /* Argument list sizes for compat_sys_socketcall */ #define AL(x) ((x) * sizeof(u32)) -static unsigned char nas[18]={AL(0),AL(3),AL(3),AL(3),AL(2),AL(3), +static unsigned char nas[19]={AL(0),AL(3),AL(3),AL(3),AL(2),AL(3), AL(3),AL(3),AL(4),AL(4),AL(4),AL(6), - AL(6),AL(2),AL(5),AL(5),AL(3),AL(3)}; + AL(6),AL(2),AL(5),AL(5),AL(3),AL(3), + AL(6)}; #undef AL asmlinkage long compat_sys_sendmsg(int fd, struct compat_msghdr __user *msg, unsigned flags) @@ -737,13 +738,52 @@ asmlinkage long compat_sys_recvmsg(int fd, struct compat_msghdr __user *msg, uns return sys_recvmsg(fd, (struct msghdr __user *)msg, flags | MSG_CMSG_COMPAT); } +asmlinkage long compat_sys_paccept(int fd, struct sockaddr __user *upeer_sockaddr, + int __user *upeer_addrlen, + const compat_sigset_t __user *sigmask, + compat_size_t sigsetsize, int flags) +{ + compat_sigset_t ss32; + sigset_t ksigmask, sigsaved; + int ret; + + if (sigmask) { + if (sigsetsize != sizeof(compat_sigset_t)) + return -EINVAL; + if (copy_from_user(&ss32, sigmask, sizeof(ss32))) + return -EFAULT; + sigset_from_compat(&ksigmask, &ss32); + + sigdelsetmask(&ksigmask, sigmask(SIGKILL)|sigmask(SIGSTOP)); + sigprocmask(SIG_SETMASK, &ksigmask, &sigsaved); + } + + ret = do_accept(fd, upeer_sockaddr, upeer_addrlen, flags); + + if (ret == -ERESTARTNOHAND) { + /* + * Don't restore the signal mask yet. Let do_signal() deliver + * the signal on the way back to userspace, before the signal + * mask is restored. + */ + if (sigmask) { + memcpy(¤t->saved_sigmask, &sigsaved, + sizeof(sigsaved)); + set_restore_sigmask(); + } + } else if (sigmask) + sigprocmask(SIG_SETMASK, &sigsaved, NULL); + + return ret; +} + asmlinkage long compat_sys_socketcall(int call, u32 __user *args) { int ret; u32 a[6]; u32 a0, a1; - if (call < SYS_SOCKET || call > SYS_RECVMSG) + if (call < SYS_SOCKET || call > SYS_PACCEPT) return -EINVAL; if (copy_from_user(a, args, nas[call])) return -EFAULT; @@ -764,7 +804,7 @@ asmlinkage long compat_sys_socketcall(int call, u32 __user *args) ret = sys_listen(a0, a1); break; case SYS_ACCEPT: - ret = sys_accept(a0, compat_ptr(a1), compat_ptr(a[2])); + ret = do_accept(a0, compat_ptr(a1), compat_ptr(a[2]), 0); break; case SYS_GETSOCKNAME: ret = sys_getsockname(a0, compat_ptr(a1), compat_ptr(a[2])); @@ -804,6 +844,10 @@ asmlinkage long compat_sys_socketcall(int call, u32 __user *args) case SYS_RECVMSG: ret = compat_sys_recvmsg(a0, compat_ptr(a1), a[2]); break; + case SYS_PACCEPT: + ret = compat_sys_paccept(a0, compat_ptr(a1), compat_ptr(a[2]), + compat_ptr(a[3]), a[4], a[5]); + break; default: ret = -EINVAL; break; diff --git a/net/socket.c b/net/socket.c index 64601f900352..a0ce8ad72252 100644 --- a/net/socket.c +++ b/net/socket.c @@ -63,6 +63,7 @@ #include #include #include +#include #include #include #include @@ -1225,6 +1226,9 @@ asmlinkage long sys_socket(int family, int type, int protocol) return -EINVAL; type &= SOCK_TYPE_MASK; + if (SOCK_NONBLOCK != O_NONBLOCK && (flags & SOCK_NONBLOCK)) + flags = (flags & ~SOCK_NONBLOCK) | O_NONBLOCK; + retval = sock_create(family, type, protocol, &sock); if (retval < 0) goto out; @@ -1259,6 +1263,9 @@ asmlinkage long sys_socketpair(int family, int type, int protocol, return -EINVAL; type &= SOCK_TYPE_MASK; + if (SOCK_NONBLOCK != O_NONBLOCK && (flags & SOCK_NONBLOCK)) + flags = (flags & ~SOCK_NONBLOCK) | O_NONBLOCK; + /* * Obtain the first socket and check if the underlying protocol * supports the socketpair call. @@ -1413,14 +1420,20 @@ asmlinkage long sys_listen(int fd, int backlog) * clean when we restucture accept also. */ -asmlinkage long sys_accept(int fd, struct sockaddr __user *upeer_sockaddr, - int __user *upeer_addrlen) +long do_accept(int fd, struct sockaddr __user *upeer_sockaddr, + int __user *upeer_addrlen, int flags) { struct socket *sock, *newsock; struct file *newfile; int err, len, newfd, fput_needed; struct sockaddr_storage address; + if (flags & ~SOCK_CLOEXEC) + return -EINVAL; + + if (SOCK_NONBLOCK != O_NONBLOCK && (flags & SOCK_NONBLOCK)) + flags = (flags & ~SOCK_NONBLOCK) | O_NONBLOCK; + sock = sockfd_lookup_light(fd, &err, &fput_needed); if (!sock) goto out; @@ -1438,7 +1451,7 @@ asmlinkage long sys_accept(int fd, struct sockaddr __user *upeer_sockaddr, */ __module_get(newsock->ops->owner); - newfd = sock_alloc_fd(&newfile, 0); + newfd = sock_alloc_fd(&newfile, flags & O_CLOEXEC); if (unlikely(newfd < 0)) { err = newfd; sock_release(newsock); @@ -1491,6 +1504,50 @@ asmlinkage long sys_accept(int fd, struct sockaddr __user *upeer_sockaddr, goto out_put; } +asmlinkage long sys_paccept(int fd, struct sockaddr __user *upeer_sockaddr, + int __user *upeer_addrlen, + const sigset_t __user *sigmask, + size_t sigsetsize, int flags) +{ + sigset_t ksigmask, sigsaved; + int ret; + + if (sigmask) { + /* XXX: Don't preclude handling different sized sigset_t's. */ + if (sigsetsize != sizeof(sigset_t)) + return -EINVAL; + if (copy_from_user(&ksigmask, sigmask, sizeof(ksigmask))) + return -EFAULT; + + sigdelsetmask(&ksigmask, sigmask(SIGKILL)|sigmask(SIGSTOP)); + sigprocmask(SIG_SETMASK, &ksigmask, &sigsaved); + } + + ret = do_accept(fd, upeer_sockaddr, upeer_addrlen, flags); + + if (ret < 0 && signal_pending(current)) { + /* + * Don't restore the signal mask yet. Let do_signal() deliver + * the signal on the way back to userspace, before the signal + * mask is restored. + */ + if (sigmask) { + memcpy(¤t->saved_sigmask, &sigsaved, + sizeof(sigsaved)); + set_restore_sigmask(); + } + } else if (sigmask) + sigprocmask(SIG_SETMASK, &sigsaved, NULL); + + return ret; +} + +asmlinkage long sys_accept(int fd, struct sockaddr __user *upeer_sockaddr, + int __user *upeer_addrlen) +{ + return do_accept(fd, upeer_sockaddr, upeer_addrlen, 0); +} + /* * Attempt to connect to a socket with the server address. The address * is in user space so we verify it is OK and move it to kernel space. @@ -2011,10 +2068,11 @@ asmlinkage long sys_recvmsg(int fd, struct msghdr __user *msg, /* Argument list sizes for sys_socketcall */ #define AL(x) ((x) * sizeof(unsigned long)) -static const unsigned char nargs[18]={ +static const unsigned char nargs[19]={ AL(0),AL(3),AL(3),AL(3),AL(2),AL(3), AL(3),AL(3),AL(4),AL(4),AL(4),AL(6), - AL(6),AL(2),AL(5),AL(5),AL(3),AL(3) + AL(6),AL(2),AL(5),AL(5),AL(3),AL(3), + AL(6) }; #undef AL @@ -2033,7 +2091,7 @@ asmlinkage long sys_socketcall(int call, unsigned long __user *args) unsigned long a0, a1; int err; - if (call < 1 || call > SYS_RECVMSG) + if (call < 1 || call > SYS_PACCEPT) return -EINVAL; /* copy_from_user should be SMP safe. */ @@ -2062,8 +2120,8 @@ asmlinkage long sys_socketcall(int call, unsigned long __user *args) break; case SYS_ACCEPT: err = - sys_accept(a0, (struct sockaddr __user *)a1, - (int __user *)a[2]); + do_accept(a0, (struct sockaddr __user *)a1, + (int __user *)a[2], 0); break; case SYS_GETSOCKNAME: err = @@ -2110,6 +2168,13 @@ asmlinkage long sys_socketcall(int call, unsigned long __user *args) case SYS_RECVMSG: err = sys_recvmsg(a0, (struct msghdr __user *)a1, a[2]); break; + case SYS_PACCEPT: + err = + sys_paccept(a0, (struct sockaddr __user *)a1, + (int __user *)a[2], + (const sigset_t __user *) a[3], + a[4], a[5]); + break; default: err = -EINVAL; break; From c019bbc612f6633ede7ed67725cbf68de45ae8a4 Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 23 Jul 2008 21:29:21 -0700 Subject: [PATCH 280/857] flag parameters: paccept w/out set_restore_sigmask Some platforms do not have support to restore the signal mask in the return path from a syscall. For those platforms syscalls like pselect are not defined at all. This is, I think, not a good choice for paccept() since paccept() adds more value on top of accept() than just the signal mask handling. Therefore this patch defines a scaled down version of the sys_paccept function for those platforms. It returns -EINVAL in case the signal mask is non-NULL but behaves the same otherwise. Note that I explicitly included . I saw that it is currently included but indirectly two levels down. There is too much risk in relying on this. The header might change and then suddenly the function definition would change without anyone immediately noticing. Signed-off-by: Ulrich Drepper Cc: Davide Libenzi Cc: Michael Kerrisk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/net.h | 3 +++ net/socket.c | 17 +++++++++++++++++ 2 files changed, 20 insertions(+) diff --git a/include/linux/net.h b/include/linux/net.h index 3a9b06d4d0fe..39a23af059b4 100644 --- a/include/linux/net.h +++ b/include/linux/net.h @@ -102,6 +102,9 @@ enum sock_type { /* Flags for socket, socketpair, paccept */ #define SOCK_CLOEXEC O_CLOEXEC +#ifndef SOCK_NONBLOCK +#define SOCK_NONBLOCK O_NONBLOCK +#endif #endif /* ARCH_HAS_SOCKET_TYPES */ diff --git a/net/socket.c b/net/socket.c index a0ce8ad72252..d163adff95bf 100644 --- a/net/socket.c +++ b/net/socket.c @@ -69,6 +69,7 @@ #include #include #include +#include #include #include #include @@ -1504,6 +1505,7 @@ long do_accept(int fd, struct sockaddr __user *upeer_sockaddr, goto out_put; } +#ifdef HAVE_SET_RESTORE_SIGMASK asmlinkage long sys_paccept(int fd, struct sockaddr __user *upeer_sockaddr, int __user *upeer_addrlen, const sigset_t __user *sigmask, @@ -1541,6 +1543,21 @@ asmlinkage long sys_paccept(int fd, struct sockaddr __user *upeer_sockaddr, return ret; } +#else +asmlinkage long sys_paccept(int fd, struct sockaddr __user *upeer_sockaddr, + int __user *upeer_addrlen, + const sigset_t __user *sigmask, + size_t sigsetsize, int flags) +{ + /* The platform does not support restoring the signal mask in the + * return path. So we do not allow using paccept() with a signal + * mask. */ + if (sigmask) + return -EINVAL; + + return do_accept(fd, upeer_sockaddr, upeer_addrlen, flags); +} +#endif asmlinkage long sys_accept(int fd, struct sockaddr __user *upeer_sockaddr, int __user *upeer_addrlen) From 7d9dbca34240ebb6ff88d8a29c6c7bffd098f0c1 Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 23 Jul 2008 21:29:22 -0700 Subject: [PATCH 281/857] flag parameters: anon_inode_getfd extension This patch just extends the anon_inode_getfd interface to take an additional parameter with a flag value. The flag value is passed on to get_unused_fd_flags in anticipation for a use with the O_CLOEXEC flag. No actual semantic changes here, the changed callers all pass 0 for now. [akpm@linux-foundation.org: KVM fix] Signed-off-by: Ulrich Drepper Acked-by: Davide Libenzi Cc: Michael Kerrisk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/anon_inodes.c | 9 +++++---- fs/eventfd.c | 2 +- fs/eventpoll.c | 2 +- fs/signalfd.c | 3 ++- fs/timerfd.c | 2 +- include/linux/anon_inodes.h | 2 +- virt/kvm/kvm_main.c | 4 ++-- 7 files changed, 13 insertions(+), 11 deletions(-) diff --git a/fs/anon_inodes.c b/fs/anon_inodes.c index 977ef208c051..1a4eee620b0d 100644 --- a/fs/anon_inodes.c +++ b/fs/anon_inodes.c @@ -58,8 +58,9 @@ static struct dentry_operations anon_inodefs_dentry_operations = { * of the file * * @name: [in] name of the "class" of the new file - * @fops [in] file operations for the new file - * @priv [in] private data for the new file (will be file's private_data) + * @fops: [in] file operations for the new file + * @priv: [in] private data for the new file (will be file's private_data) + * @flags: [in] flags * * Creates a new file by hooking it on a single inode. This is useful for files * that do not need to have a full-fledged inode in order to operate correctly. @@ -68,7 +69,7 @@ static struct dentry_operations anon_inodefs_dentry_operations = { * setup. Returns new descriptor or -error. */ int anon_inode_getfd(const char *name, const struct file_operations *fops, - void *priv) + void *priv, int flags) { struct qstr this; struct dentry *dentry; @@ -78,7 +79,7 @@ int anon_inode_getfd(const char *name, const struct file_operations *fops, if (IS_ERR(anon_inode_inode)) return -ENODEV; - error = get_unused_fd(); + error = get_unused_fd_flags(flags); if (error < 0) return error; fd = error; diff --git a/fs/eventfd.c b/fs/eventfd.c index 343942deeec1..6094265ca409 100644 --- a/fs/eventfd.c +++ b/fs/eventfd.c @@ -214,7 +214,7 @@ asmlinkage long sys_eventfd(unsigned int count) * When we call this, the initialization must be complete, since * anon_inode_getfd() will install the fd. */ - fd = anon_inode_getfd("[eventfd]", &eventfd_fops, ctx); + fd = anon_inode_getfd("[eventfd]", &eventfd_fops, ctx, 0); if (fd < 0) kfree(ctx); return fd; diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 990c01d2d66b..9392dd968125 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -1068,7 +1068,7 @@ asmlinkage long sys_epoll_create(int size) * Creates all the items needed to setup an eventpoll file. That is, * a file structure and a free file descriptor. */ - fd = anon_inode_getfd("[eventpoll]", &eventpoll_fops, ep); + fd = anon_inode_getfd("[eventpoll]", &eventpoll_fops, ep, 0); if (fd < 0) ep_free(ep); diff --git a/fs/signalfd.c b/fs/signalfd.c index 619725644c75..ddb328b74bde 100644 --- a/fs/signalfd.c +++ b/fs/signalfd.c @@ -227,7 +227,8 @@ asmlinkage long sys_signalfd(int ufd, sigset_t __user *user_mask, size_t sizemas * When we call this, the initialization must be complete, since * anon_inode_getfd() will install the fd. */ - ufd = anon_inode_getfd("[signalfd]", &signalfd_fops, ctx); + ufd = anon_inode_getfd("[signalfd]", &signalfd_fops, ctx, + 0); if (ufd < 0) kfree(ctx); } else { diff --git a/fs/timerfd.c b/fs/timerfd.c index d87d354ec424..77c2bc92cbee 100644 --- a/fs/timerfd.c +++ b/fs/timerfd.c @@ -198,7 +198,7 @@ asmlinkage long sys_timerfd_create(int clockid, int flags) ctx->clockid = clockid; hrtimer_init(&ctx->tmr, clockid, HRTIMER_MODE_ABS); - ufd = anon_inode_getfd("[timerfd]", &timerfd_fops, ctx); + ufd = anon_inode_getfd("[timerfd]", &timerfd_fops, ctx, 0); if (ufd < 0) kfree(ctx); diff --git a/include/linux/anon_inodes.h b/include/linux/anon_inodes.h index 6129e58ca7c9..e0a0cdc2da43 100644 --- a/include/linux/anon_inodes.h +++ b/include/linux/anon_inodes.h @@ -9,7 +9,7 @@ #define _LINUX_ANON_INODES_H int anon_inode_getfd(const char *name, const struct file_operations *fops, - void *priv); + void *priv, int flags); #endif /* _LINUX_ANON_INODES_H */ diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 904d7b7bd780..a845890b6800 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -902,7 +902,7 @@ static const struct file_operations kvm_vcpu_fops = { */ static int create_vcpu_fd(struct kvm_vcpu *vcpu) { - int fd = anon_inode_getfd("kvm-vcpu", &kvm_vcpu_fops, vcpu); + int fd = anon_inode_getfd("kvm-vcpu", &kvm_vcpu_fops, vcpu, 0); if (fd < 0) kvm_put_kvm(vcpu->kvm); return fd; @@ -1261,7 +1261,7 @@ static int kvm_dev_ioctl_create_vm(void) kvm = kvm_create_vm(); if (IS_ERR(kvm)) return PTR_ERR(kvm); - fd = anon_inode_getfd("kvm-vm", &kvm_vm_fops, kvm); + fd = anon_inode_getfd("kvm-vm", &kvm_vm_fops, kvm, 0); if (fd < 0) kvm_put_kvm(kvm); From 9deb27baedb79759c3ab9435a7d8b841842d56e9 Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 23 Jul 2008 21:29:24 -0700 Subject: [PATCH 282/857] flag parameters: signalfd This patch adds the new signalfd4 syscall. It extends the old signalfd syscall by one parameter which is meant to hold a flag value. In this patch the only flag support is SFD_CLOEXEC which causes the close-on-exec flag for the returned file descriptor to be set. A new name SFD_CLOEXEC is introduced which in this implementation must have the same value as O_CLOEXEC. The following test must be adjusted for architectures other than x86 and x86-64 and in case the syscall numbers changed. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #include #include #include #include #include #ifndef __NR_signalfd4 # ifdef __x86_64__ # define __NR_signalfd4 289 # elif defined __i386__ # define __NR_signalfd4 327 # else # error "need __NR_signalfd4" # endif #endif #define SFD_CLOEXEC O_CLOEXEC int main (void) { sigset_t ss; sigemptyset (&ss); sigaddset (&ss, SIGUSR1); int fd = syscall (__NR_signalfd4, -1, &ss, 8, 0); if (fd == -1) { puts ("signalfd4(0) failed"); return 1; } int coe = fcntl (fd, F_GETFD); if (coe == -1) { puts ("fcntl failed"); return 1; } if (coe & FD_CLOEXEC) { puts ("signalfd4(0) set close-on-exec flag"); return 1; } close (fd); fd = syscall (__NR_signalfd4, -1, &ss, 8, SFD_CLOEXEC); if (fd == -1) { puts ("signalfd4(SFD_CLOEXEC) failed"); return 1; } coe = fcntl (fd, F_GETFD); if (coe == -1) { puts ("fcntl failed"); return 1; } if ((coe & FD_CLOEXEC) == 0) { puts ("signalfd4(SFD_CLOEXEC) does not set close-on-exec flag"); return 1; } close (fd); puts ("OK"); return 0; } ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ [akpm@linux-foundation.org: add sys_ni stub] Signed-off-by: Ulrich Drepper Acked-by: Davide Libenzi Cc: Michael Kerrisk Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/ia32/ia32entry.S | 1 + arch/x86/kernel/syscall_table_32.S | 1 + fs/compat.c | 14 ++++++++++---- fs/signalfd.c | 14 ++++++++++++-- include/asm-x86/unistd_32.h | 1 + include/asm-x86/unistd_64.h | 2 ++ include/linux/signalfd.h | 5 +++++ include/linux/syscalls.h | 1 + kernel/sys_ni.c | 1 + 9 files changed, 34 insertions(+), 6 deletions(-) diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S index 021d71bc69b5..c308128b9251 100644 --- a/arch/x86/ia32/ia32entry.S +++ b/arch/x86/ia32/ia32entry.S @@ -826,4 +826,5 @@ ia32_sys_call_table: .quad sys32_fallocate .quad compat_sys_timerfd_settime /* 325 */ .quad compat_sys_timerfd_gettime + .quad compat_sys_signalfd4 ia32_syscall_end: diff --git a/arch/x86/kernel/syscall_table_32.S b/arch/x86/kernel/syscall_table_32.S index adff5562f5fd..c12a36c9fd51 100644 --- a/arch/x86/kernel/syscall_table_32.S +++ b/arch/x86/kernel/syscall_table_32.S @@ -326,3 +326,4 @@ ENTRY(sys_call_table) .long sys_fallocate .long sys_timerfd_settime /* 325 */ .long sys_timerfd_gettime + .long sys_signalfd4 diff --git a/fs/compat.c b/fs/compat.c index b46604281766..106eba28ec5a 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -2131,9 +2131,9 @@ asmlinkage long compat_sys_epoll_pwait(int epfd, #ifdef CONFIG_SIGNALFD -asmlinkage long compat_sys_signalfd(int ufd, - const compat_sigset_t __user *sigmask, - compat_size_t sigsetsize) +asmlinkage long compat_sys_signalfd4(int ufd, + const compat_sigset_t __user *sigmask, + compat_size_t sigsetsize, int flags) { compat_sigset_t ss32; sigset_t tmp; @@ -2148,9 +2148,15 @@ asmlinkage long compat_sys_signalfd(int ufd, if (copy_to_user(ksigmask, &tmp, sizeof(sigset_t))) return -EFAULT; - return sys_signalfd(ufd, ksigmask, sizeof(sigset_t)); + return sys_signalfd4(ufd, ksigmask, sizeof(sigset_t), flags); } +asmlinkage long compat_sys_signalfd(int ufd, + const compat_sigset_t __user *sigmask, + compat_size_t sigsetsize) +{ + return compat_sys_signalfd4(ufd, sigmask, sigsetsize, 0); +} #endif /* CONFIG_SIGNALFD */ #ifdef CONFIG_TIMERFD diff --git a/fs/signalfd.c b/fs/signalfd.c index ddb328b74bde..c8609fa51a13 100644 --- a/fs/signalfd.c +++ b/fs/signalfd.c @@ -205,11 +205,15 @@ static const struct file_operations signalfd_fops = { .read = signalfd_read, }; -asmlinkage long sys_signalfd(int ufd, sigset_t __user *user_mask, size_t sizemask) +asmlinkage long sys_signalfd4(int ufd, sigset_t __user *user_mask, + size_t sizemask, int flags) { sigset_t sigmask; struct signalfd_ctx *ctx; + if (flags & ~SFD_CLOEXEC) + return -EINVAL; + if (sizemask != sizeof(sigset_t) || copy_from_user(&sigmask, user_mask, sizeof(sigmask))) return -EINVAL; @@ -228,7 +232,7 @@ asmlinkage long sys_signalfd(int ufd, sigset_t __user *user_mask, size_t sizemas * anon_inode_getfd() will install the fd. */ ufd = anon_inode_getfd("[signalfd]", &signalfd_fops, ctx, - 0); + flags & O_CLOEXEC); if (ufd < 0) kfree(ctx); } else { @@ -250,3 +254,9 @@ asmlinkage long sys_signalfd(int ufd, sigset_t __user *user_mask, size_t sizemas return ufd; } + +asmlinkage long sys_signalfd(int ufd, sigset_t __user *user_mask, + size_t sizemask) +{ + return sys_signalfd4(ufd, user_mask, sizemask, 0); +} diff --git a/include/asm-x86/unistd_32.h b/include/asm-x86/unistd_32.h index 8317d94771d3..c310371f5613 100644 --- a/include/asm-x86/unistd_32.h +++ b/include/asm-x86/unistd_32.h @@ -332,6 +332,7 @@ #define __NR_fallocate 324 #define __NR_timerfd_settime 325 #define __NR_timerfd_gettime 326 +#define __NR_signalfd4 327 #ifdef __KERNEL__ diff --git a/include/asm-x86/unistd_64.h b/include/asm-x86/unistd_64.h index e323994a370f..e0a9b45b2346 100644 --- a/include/asm-x86/unistd_64.h +++ b/include/asm-x86/unistd_64.h @@ -641,6 +641,8 @@ __SYSCALL(__NR_timerfd_settime, sys_timerfd_settime) __SYSCALL(__NR_timerfd_gettime, sys_timerfd_gettime) #define __NR_paccept 288 __SYSCALL(__NR_paccept, sys_paccept) +#define __NR_signalfd4 289 +__SYSCALL(__NR_signalfd4, sys_signalfd4) #ifndef __NO_STUBS diff --git a/include/linux/signalfd.h b/include/linux/signalfd.h index ea037f28df91..8b3f7b7420a1 100644 --- a/include/linux/signalfd.h +++ b/include/linux/signalfd.h @@ -8,6 +8,11 @@ #ifndef _LINUX_SIGNALFD_H #define _LINUX_SIGNALFD_H +/* For O_CLOEXEC */ +#include + +/* Flags for signalfd4. */ +#define SFD_CLOEXEC O_CLOEXEC struct signalfd_siginfo { __u32 ssi_signo; diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 2a2a40af6b2c..1c2707797845 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -610,6 +610,7 @@ asmlinkage long sys_set_robust_list(struct robust_list_head __user *head, size_t len); asmlinkage long sys_getcpu(unsigned __user *cpu, unsigned __user *node, struct getcpu_cache __user *cache); asmlinkage long sys_signalfd(int ufd, sigset_t __user *user_mask, size_t sizemask); +asmlinkage long sys_signalfd4(int ufd, sigset_t __user *user_mask, size_t sizemask, int flags); asmlinkage long sys_timerfd_create(int clockid, int flags); asmlinkage long sys_timerfd_settime(int ufd, int flags, const struct itimerspec __user *utmr, diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c index 2f0b8a2e600f..8627c89ae9e8 100644 --- a/kernel/sys_ni.c +++ b/kernel/sys_ni.c @@ -156,6 +156,7 @@ cond_syscall(sys_ioprio_get); /* New file descriptors */ cond_syscall(sys_signalfd); +cond_syscall(sys_signalfd4); cond_syscall(compat_sys_signalfd); cond_syscall(sys_timerfd_create); cond_syscall(sys_timerfd_settime); From b087498eb5605673b0f260a7620d91818cd72304 Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 23 Jul 2008 21:29:25 -0700 Subject: [PATCH 283/857] flag parameters: eventfd This patch adds the new eventfd2 syscall. It extends the old eventfd syscall by one parameter which is meant to hold a flag value. In this patch the only flag support is EFD_CLOEXEC which causes the close-on-exec flag for the returned file descriptor to be set. A new name EFD_CLOEXEC is introduced which in this implementation must have the same value as O_CLOEXEC. The following test must be adjusted for architectures other than x86 and x86-64 and in case the syscall numbers changed. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #include #include #include #include #ifndef __NR_eventfd2 # ifdef __x86_64__ # define __NR_eventfd2 290 # elif defined __i386__ # define __NR_eventfd2 328 # else # error "need __NR_eventfd2" # endif #endif #define EFD_CLOEXEC O_CLOEXEC int main (void) { int fd = syscall (__NR_eventfd2, 1, 0); if (fd == -1) { puts ("eventfd2(0) failed"); return 1; } int coe = fcntl (fd, F_GETFD); if (coe == -1) { puts ("fcntl failed"); return 1; } if (coe & FD_CLOEXEC) { puts ("eventfd2(0) sets close-on-exec flag"); return 1; } close (fd); fd = syscall (__NR_eventfd2, 1, EFD_CLOEXEC); if (fd == -1) { puts ("eventfd2(EFD_CLOEXEC) failed"); return 1; } coe = fcntl (fd, F_GETFD); if (coe == -1) { puts ("fcntl failed"); return 1; } if ((coe & FD_CLOEXEC) == 0) { puts ("eventfd2(EFD_CLOEXEC) does not set close-on-exec flag"); return 1; } close (fd); puts ("OK"); return 0; } ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ [akpm@linux-foundation.org: add sys_ni stub] Signed-off-by: Ulrich Drepper Acked-by: Davide Libenzi Cc: Michael Kerrisk Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/ia32/ia32entry.S | 1 + arch/x86/kernel/syscall_table_32.S | 1 + fs/eventfd.c | 13 +++++++++++-- include/asm-x86/unistd_32.h | 1 + include/asm-x86/unistd_64.h | 2 ++ include/linux/eventfd.h | 6 ++++++ include/linux/syscalls.h | 1 + kernel/sys_ni.c | 1 + 8 files changed, 24 insertions(+), 2 deletions(-) diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S index c308128b9251..cf0eb31745ca 100644 --- a/arch/x86/ia32/ia32entry.S +++ b/arch/x86/ia32/ia32entry.S @@ -827,4 +827,5 @@ ia32_sys_call_table: .quad compat_sys_timerfd_settime /* 325 */ .quad compat_sys_timerfd_gettime .quad compat_sys_signalfd4 + .quad sys_eventfd2 ia32_syscall_end: diff --git a/arch/x86/kernel/syscall_table_32.S b/arch/x86/kernel/syscall_table_32.S index c12a36c9fd51..cf112cb11c37 100644 --- a/arch/x86/kernel/syscall_table_32.S +++ b/arch/x86/kernel/syscall_table_32.S @@ -327,3 +327,4 @@ ENTRY(sys_call_table) .long sys_timerfd_settime /* 325 */ .long sys_timerfd_gettime .long sys_signalfd4 + .long sys_eventfd2 diff --git a/fs/eventfd.c b/fs/eventfd.c index 6094265ca409..bd420e6478ad 100644 --- a/fs/eventfd.c +++ b/fs/eventfd.c @@ -198,11 +198,14 @@ struct file *eventfd_fget(int fd) return file; } -asmlinkage long sys_eventfd(unsigned int count) +asmlinkage long sys_eventfd2(unsigned int count, int flags) { int fd; struct eventfd_ctx *ctx; + if (flags & ~EFD_CLOEXEC) + return -EINVAL; + ctx = kmalloc(sizeof(*ctx), GFP_KERNEL); if (!ctx) return -ENOMEM; @@ -214,9 +217,15 @@ asmlinkage long sys_eventfd(unsigned int count) * When we call this, the initialization must be complete, since * anon_inode_getfd() will install the fd. */ - fd = anon_inode_getfd("[eventfd]", &eventfd_fops, ctx, 0); + fd = anon_inode_getfd("[eventfd]", &eventfd_fops, ctx, + flags & O_CLOEXEC); if (fd < 0) kfree(ctx); return fd; } +asmlinkage long sys_eventfd(unsigned int count) +{ + return sys_eventfd2(count, 0); +} + diff --git a/include/asm-x86/unistd_32.h b/include/asm-x86/unistd_32.h index c310371f5613..edbd8723c939 100644 --- a/include/asm-x86/unistd_32.h +++ b/include/asm-x86/unistd_32.h @@ -333,6 +333,7 @@ #define __NR_timerfd_settime 325 #define __NR_timerfd_gettime 326 #define __NR_signalfd4 327 +#define __NR_eventfd2 328 #ifdef __KERNEL__ diff --git a/include/asm-x86/unistd_64.h b/include/asm-x86/unistd_64.h index e0a9b45b2346..fb059a6feeb1 100644 --- a/include/asm-x86/unistd_64.h +++ b/include/asm-x86/unistd_64.h @@ -643,6 +643,8 @@ __SYSCALL(__NR_timerfd_gettime, sys_timerfd_gettime) __SYSCALL(__NR_paccept, sys_paccept) #define __NR_signalfd4 289 __SYSCALL(__NR_signalfd4, sys_signalfd4) +#define __NR_eventfd2 290 +__SYSCALL(__NR_eventfd2, sys_eventfd2) #ifndef __NO_STUBS diff --git a/include/linux/eventfd.h b/include/linux/eventfd.h index a701399b7fed..a6c0eaedb1b0 100644 --- a/include/linux/eventfd.h +++ b/include/linux/eventfd.h @@ -10,6 +10,12 @@ #ifdef CONFIG_EVENTFD +/* For O_CLOEXEC */ +#include + +/* Flags for eventfd2. */ +#define EFD_CLOEXEC O_CLOEXEC + struct file *eventfd_fget(int fd); int eventfd_signal(struct file *file, int n); diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 1c2707797845..9ab09926a7f2 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -617,6 +617,7 @@ asmlinkage long sys_timerfd_settime(int ufd, int flags, struct itimerspec __user *otmr); asmlinkage long sys_timerfd_gettime(int ufd, struct itimerspec __user *otmr); asmlinkage long sys_eventfd(unsigned int count); +asmlinkage long sys_eventfd2(unsigned int count, int flags); asmlinkage long sys_fallocate(int fd, int mode, loff_t offset, loff_t len); int kernel_execve(const char *filename, char *const argv[], char *const envp[]); diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c index 8627c89ae9e8..2a361ccdc7ca 100644 --- a/kernel/sys_ni.c +++ b/kernel/sys_ni.c @@ -164,3 +164,4 @@ cond_syscall(sys_timerfd_gettime); cond_syscall(compat_sys_timerfd_settime); cond_syscall(compat_sys_timerfd_gettime); cond_syscall(sys_eventfd); +cond_syscall(sys_eventfd2); From 11fcb6c14676023d0bd437841f5dcd670e7990a0 Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 23 Jul 2008 21:29:26 -0700 Subject: [PATCH 284/857] flag parameters: timerfd_create The timerfd_create syscall already has a flags parameter. It just is unused so far. This patch changes this by introducing the TFD_CLOEXEC flag to set the close-on-exec flag for the returned file descriptor. A new name TFD_CLOEXEC is introduced which in this implementation must have the same value as O_CLOEXEC. The following test must be adjusted for architectures other than x86 and x86-64 and in case the syscall numbers changed. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #include #include #include #include #include #ifndef __NR_timerfd_create # ifdef __x86_64__ # define __NR_timerfd_create 283 # elif defined __i386__ # define __NR_timerfd_create 322 # else # error "need __NR_timerfd_create" # endif #endif #define TFD_CLOEXEC O_CLOEXEC int main (void) { int fd = syscall (__NR_timerfd_create, CLOCK_REALTIME, 0); if (fd == -1) { puts ("timerfd_create(0) failed"); return 1; } int coe = fcntl (fd, F_GETFD); if (coe == -1) { puts ("fcntl failed"); return 1; } if (coe & FD_CLOEXEC) { puts ("timerfd_create(0) set close-on-exec flag"); return 1; } close (fd); fd = syscall (__NR_timerfd_create, CLOCK_REALTIME, TFD_CLOEXEC); if (fd == -1) { puts ("timerfd_create(TFD_CLOEXEC) failed"); return 1; } coe = fcntl (fd, F_GETFD); if (coe == -1) { puts ("fcntl failed"); return 1; } if ((coe & FD_CLOEXEC) == 0) { puts ("timerfd_create(TFD_CLOEXEC) set close-on-exec flag"); return 1; } close (fd); puts ("OK"); return 0; } ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Signed-off-by: Ulrich Drepper Acked-by: Davide Libenzi Cc: Michael Kerrisk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/timerfd.c | 5 +++-- include/linux/timerfd.h | 5 +++++ 2 files changed, 8 insertions(+), 2 deletions(-) diff --git a/fs/timerfd.c b/fs/timerfd.c index 77c2bc92cbee..c6ef5e33cb34 100644 --- a/fs/timerfd.c +++ b/fs/timerfd.c @@ -184,7 +184,7 @@ asmlinkage long sys_timerfd_create(int clockid, int flags) int ufd; struct timerfd_ctx *ctx; - if (flags) + if (flags & ~TFD_CLOEXEC) return -EINVAL; if (clockid != CLOCK_MONOTONIC && clockid != CLOCK_REALTIME) @@ -198,7 +198,8 @@ asmlinkage long sys_timerfd_create(int clockid, int flags) ctx->clockid = clockid; hrtimer_init(&ctx->tmr, clockid, HRTIMER_MODE_ABS); - ufd = anon_inode_getfd("[timerfd]", &timerfd_fops, ctx, 0); + ufd = anon_inode_getfd("[timerfd]", &timerfd_fops, ctx, + flags & O_CLOEXEC); if (ufd < 0) kfree(ctx); diff --git a/include/linux/timerfd.h b/include/linux/timerfd.h index cf2b10d75731..96ed97dff00f 100644 --- a/include/linux/timerfd.h +++ b/include/linux/timerfd.h @@ -8,9 +8,14 @@ #ifndef _LINUX_TIMERFD_H #define _LINUX_TIMERFD_H +/* For O_CLOEXEC */ +#include +/* Flags for timerfd_settime. */ #define TFD_TIMER_ABSTIME (1 << 0) +/* Flags for timerfd_create. */ +#define TFD_CLOEXEC O_CLOEXEC #endif /* _LINUX_TIMERFD_H */ From a0998b50c3f0b8fdd265c63e0032f86ebe377dbf Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 23 Jul 2008 21:29:27 -0700 Subject: [PATCH 285/857] flag parameters: epoll_create This patch adds the new epoll_create2 syscall. It extends the old epoll_create syscall by one parameter which is meant to hold a flag value. In this patch the only flag support is EPOLL_CLOEXEC which causes the close-on-exec flag for the returned file descriptor to be set. A new name EPOLL_CLOEXEC is introduced which in this implementation must have the same value as O_CLOEXEC. The following test must be adjusted for architectures other than x86 and x86-64 and in case the syscall numbers changed. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #include #include #include #include #include #ifndef __NR_epoll_create2 # ifdef __x86_64__ # define __NR_epoll_create2 291 # elif defined __i386__ # define __NR_epoll_create2 329 # else # error "need __NR_epoll_create2" # endif #endif #define EPOLL_CLOEXEC O_CLOEXEC int main (void) { int fd = syscall (__NR_epoll_create2, 1, 0); if (fd == -1) { puts ("epoll_create2(0) failed"); return 1; } int coe = fcntl (fd, F_GETFD); if (coe == -1) { puts ("fcntl failed"); return 1; } if (coe & FD_CLOEXEC) { puts ("epoll_create2(0) set close-on-exec flag"); return 1; } close (fd); fd = syscall (__NR_epoll_create2, 1, EPOLL_CLOEXEC); if (fd == -1) { puts ("epoll_create2(EPOLL_CLOEXEC) failed"); return 1; } coe = fcntl (fd, F_GETFD); if (coe == -1) { puts ("fcntl failed"); return 1; } if ((coe & FD_CLOEXEC) == 0) { puts ("epoll_create2(EPOLL_CLOEXEC) set close-on-exec flag"); return 1; } close (fd); puts ("OK"); return 0; } ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Signed-off-by: Ulrich Drepper Acked-by: Davide Libenzi Cc: Michael Kerrisk Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/ia32/ia32entry.S | 1 + arch/x86/kernel/syscall_table_32.S | 1 + fs/eventpoll.c | 13 +++++++++++-- include/asm-x86/unistd_32.h | 1 + include/asm-x86/unistd_64.h | 2 ++ include/linux/eventpoll.h | 4 ++++ include/linux/syscalls.h | 1 + 7 files changed, 21 insertions(+), 2 deletions(-) diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S index cf0eb31745ca..04366f08f424 100644 --- a/arch/x86/ia32/ia32entry.S +++ b/arch/x86/ia32/ia32entry.S @@ -828,4 +828,5 @@ ia32_sys_call_table: .quad compat_sys_timerfd_gettime .quad compat_sys_signalfd4 .quad sys_eventfd2 + .quad sys_epoll_create2 ia32_syscall_end: diff --git a/arch/x86/kernel/syscall_table_32.S b/arch/x86/kernel/syscall_table_32.S index cf112cb11c37..4d7007ca263d 100644 --- a/arch/x86/kernel/syscall_table_32.S +++ b/arch/x86/kernel/syscall_table_32.S @@ -328,3 +328,4 @@ ENTRY(sys_call_table) .long sys_timerfd_gettime .long sys_signalfd4 .long sys_eventfd2 + .long sys_epoll_create2 diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 9392dd968125..3fd4014f3c5a 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -1046,11 +1046,14 @@ static int ep_poll(struct eventpoll *ep, struct epoll_event __user *events, * RB tree. With the current implementation, the "size" parameter is ignored * (besides sanity checks). */ -asmlinkage long sys_epoll_create(int size) +asmlinkage long sys_epoll_create2(int size, int flags) { int error, fd = -1; struct eventpoll *ep; + if (flags & ~EPOLL_CLOEXEC) + return -EINVAL; + DNPRINTK(3, (KERN_INFO "[%p] eventpoll: sys_epoll_create(%d)\n", current, size)); @@ -1068,7 +1071,8 @@ asmlinkage long sys_epoll_create(int size) * Creates all the items needed to setup an eventpoll file. That is, * a file structure and a free file descriptor. */ - fd = anon_inode_getfd("[eventpoll]", &eventpoll_fops, ep, 0); + fd = anon_inode_getfd("[eventpoll]", &eventpoll_fops, ep, + flags & O_CLOEXEC); if (fd < 0) ep_free(ep); @@ -1079,6 +1083,11 @@ asmlinkage long sys_epoll_create(int size) return fd; } +asmlinkage long sys_epoll_create(int size) +{ + return sys_epoll_create2(size, 0); +} + /* * The following function implements the controller interface for * the eventpoll file that enables the insertion/removal/change of diff --git a/include/asm-x86/unistd_32.h b/include/asm-x86/unistd_32.h index edbd8723c939..a37d6b0c4e1e 100644 --- a/include/asm-x86/unistd_32.h +++ b/include/asm-x86/unistd_32.h @@ -334,6 +334,7 @@ #define __NR_timerfd_gettime 326 #define __NR_signalfd4 327 #define __NR_eventfd2 328 +#define __NR_epoll_create2 329 #ifdef __KERNEL__ diff --git a/include/asm-x86/unistd_64.h b/include/asm-x86/unistd_64.h index fb059a6feeb1..a1a4a5b6e5ee 100644 --- a/include/asm-x86/unistd_64.h +++ b/include/asm-x86/unistd_64.h @@ -645,6 +645,8 @@ __SYSCALL(__NR_paccept, sys_paccept) __SYSCALL(__NR_signalfd4, sys_signalfd4) #define __NR_eventfd2 290 __SYSCALL(__NR_eventfd2, sys_eventfd2) +#define __NR_epoll_create2 291 +__SYSCALL(__NR_epoll_create2, sys_epoll_create2) #ifndef __NO_STUBS diff --git a/include/linux/eventpoll.h b/include/linux/eventpoll.h index cf79853967ff..1cfaa40059c8 100644 --- a/include/linux/eventpoll.h +++ b/include/linux/eventpoll.h @@ -14,8 +14,12 @@ #ifndef _LINUX_EVENTPOLL_H #define _LINUX_EVENTPOLL_H +/* For O_CLOEXEC */ +#include #include +/* Flags for epoll_create2. */ +#define EPOLL_CLOEXEC O_CLOEXEC /* Valid opcodes to issue to sys_epoll_ctl() */ #define EPOLL_CTL_ADD 1 diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 9ab09926a7f2..85953240f28c 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -430,6 +430,7 @@ asmlinkage long sys_poll(struct pollfd __user *ufds, unsigned int nfds, asmlinkage long sys_select(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp, struct timeval __user *tvp); asmlinkage long sys_epoll_create(int size); +asmlinkage long sys_epoll_create2(int size, int flags); asmlinkage long sys_epoll_ctl(int epfd, int op, int fd, struct epoll_event __user *event); asmlinkage long sys_epoll_wait(int epfd, struct epoll_event __user *events, From 336dd1f70ff62d7dd8655228caed4c5bfc818c56 Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 23 Jul 2008 21:29:29 -0700 Subject: [PATCH 286/857] flag parameters: dup2 This patch adds the new dup3 syscall. It extends the old dup2 syscall by one parameter which is meant to hold a flag value. Support for the O_CLOEXEC flag is added in this patch. The following test must be adjusted for architectures other than x86 and x86-64 and in case the syscall numbers changed. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #include #include #include #include #include #ifndef __NR_dup3 # ifdef __x86_64__ # define __NR_dup3 292 # elif defined __i386__ # define __NR_dup3 330 # else # error "need __NR_dup3" # endif #endif int main (void) { int fd = syscall (__NR_dup3, 1, 4, 0); if (fd == -1) { puts ("dup3(0) failed"); return 1; } int coe = fcntl (fd, F_GETFD); if (coe == -1) { puts ("fcntl failed"); return 1; } if (coe & FD_CLOEXEC) { puts ("dup3(0) set close-on-exec flag"); return 1; } close (fd); fd = syscall (__NR_dup3, 1, 4, O_CLOEXEC); if (fd == -1) { puts ("dup3(O_CLOEXEC) failed"); return 1; } coe = fcntl (fd, F_GETFD); if (coe == -1) { puts ("fcntl failed"); return 1; } if ((coe & FD_CLOEXEC) == 0) { puts ("dup3(O_CLOEXEC) set close-on-exec flag"); return 1; } close (fd); puts ("OK"); return 0; } ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Signed-off-by: Ulrich Drepper Acked-by: Davide Libenzi Cc: Michael Kerrisk Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/ia32/ia32entry.S | 1 + arch/x86/kernel/syscall_table_32.S | 1 + fs/fcntl.c | 15 +++++++++++++-- include/asm-x86/unistd_32.h | 1 + include/asm-x86/unistd_64.h | 2 ++ include/linux/syscalls.h | 1 + 6 files changed, 19 insertions(+), 2 deletions(-) diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S index 04366f08f424..5614a8f7bed4 100644 --- a/arch/x86/ia32/ia32entry.S +++ b/arch/x86/ia32/ia32entry.S @@ -829,4 +829,5 @@ ia32_sys_call_table: .quad compat_sys_signalfd4 .quad sys_eventfd2 .quad sys_epoll_create2 + .quad sys_dup3 /* 330 */ ia32_syscall_end: diff --git a/arch/x86/kernel/syscall_table_32.S b/arch/x86/kernel/syscall_table_32.S index 4d7007ca263d..24a3f1ea6a0e 100644 --- a/arch/x86/kernel/syscall_table_32.S +++ b/arch/x86/kernel/syscall_table_32.S @@ -329,3 +329,4 @@ ENTRY(sys_call_table) .long sys_signalfd4 .long sys_eventfd2 .long sys_epoll_create2 + .long sys_dup3 /* 330 */ diff --git a/fs/fcntl.c b/fs/fcntl.c index 330a7d782591..9679fcbdeaa0 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -125,13 +125,16 @@ static int dupfd(struct file *file, unsigned int start, int cloexec) return fd; } -asmlinkage long sys_dup2(unsigned int oldfd, unsigned int newfd) +asmlinkage long sys_dup3(unsigned int oldfd, unsigned int newfd, int flags) { int err = -EBADF; struct file * file, *tofree; struct files_struct * files = current->files; struct fdtable *fdt; + if ((flags & ~O_CLOEXEC) != 0) + return -EINVAL; + spin_lock(&files->file_lock); if (!(file = fcheck(oldfd))) goto out_unlock; @@ -163,7 +166,10 @@ asmlinkage long sys_dup2(unsigned int oldfd, unsigned int newfd) rcu_assign_pointer(fdt->fd[newfd], file); FD_SET(newfd, fdt->open_fds); - FD_CLR(newfd, fdt->close_on_exec); + if (flags & O_CLOEXEC) + FD_SET(newfd, fdt->close_on_exec); + else + FD_CLR(newfd, fdt->close_on_exec); spin_unlock(&files->file_lock); if (tofree) @@ -181,6 +187,11 @@ asmlinkage long sys_dup2(unsigned int oldfd, unsigned int newfd) goto out; } +asmlinkage long sys_dup2(unsigned int oldfd, unsigned int newfd) +{ + return sys_dup3(oldfd, newfd, 0); +} + asmlinkage long sys_dup(unsigned int fildes) { int ret = -EBADF; diff --git a/include/asm-x86/unistd_32.h b/include/asm-x86/unistd_32.h index a37d6b0c4e1e..a1f6383bf695 100644 --- a/include/asm-x86/unistd_32.h +++ b/include/asm-x86/unistd_32.h @@ -335,6 +335,7 @@ #define __NR_signalfd4 327 #define __NR_eventfd2 328 #define __NR_epoll_create2 329 +#define __NR_dup3 330 #ifdef __KERNEL__ diff --git a/include/asm-x86/unistd_64.h b/include/asm-x86/unistd_64.h index a1a4a5b6e5ee..f0fb2bd40cdb 100644 --- a/include/asm-x86/unistd_64.h +++ b/include/asm-x86/unistd_64.h @@ -647,6 +647,8 @@ __SYSCALL(__NR_signalfd4, sys_signalfd4) __SYSCALL(__NR_eventfd2, sys_eventfd2) #define __NR_epoll_create2 291 __SYSCALL(__NR_epoll_create2, sys_epoll_create2) +#define __NR_dup3 292 +__SYSCALL(__NR_dup3, sys_dup3) #ifndef __NO_STUBS diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 85953240f28c..034d3358549e 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -305,6 +305,7 @@ asmlinkage long sys_fcntl64(unsigned int fd, #endif asmlinkage long sys_dup(unsigned int fildes); asmlinkage long sys_dup2(unsigned int oldfd, unsigned int newfd); +asmlinkage long sys_dup3(unsigned int oldfd, unsigned int newfd, int flags); asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int on); asmlinkage long sys_ioctl(unsigned int fd, unsigned int cmd, unsigned long arg); From ed8cae8ba01348bfd83333f4648dd807b04d7f08 Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 23 Jul 2008 21:29:30 -0700 Subject: [PATCH 287/857] flag parameters: pipe This patch introduces the new syscall pipe2 which is like pipe but it also takes an additional parameter which takes a flag value. This patch implements the handling of O_CLOEXEC for the flag. I did not add support for the new syscall for the architectures which have a special sys_pipe implementation. I think the maintainers of those archs have the chance to go with the unified implementation but that's up to them. The implementation introduces do_pipe_flags. I did that instead of changing all callers of do_pipe because some of the callers are written in assembler. I would probably screw up changing the assembly code. To avoid breaking code do_pipe is now a small wrapper around do_pipe_flags. Once all callers are changed over to do_pipe_flags the old do_pipe function can be removed. The following test must be adjusted for architectures other than x86 and x86-64 and in case the syscall numbers changed. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #include #include #include #include #ifndef __NR_pipe2 # ifdef __x86_64__ # define __NR_pipe2 293 # elif defined __i386__ # define __NR_pipe2 331 # else # error "need __NR_pipe2" # endif #endif int main (void) { int fd[2]; if (syscall (__NR_pipe2, fd, 0) != 0) { puts ("pipe2(0) failed"); return 1; } for (int i = 0; i < 2; ++i) { int coe = fcntl (fd[i], F_GETFD); if (coe == -1) { puts ("fcntl failed"); return 1; } if (coe & FD_CLOEXEC) { printf ("pipe2(0) set close-on-exit for fd[%d]\n", i); return 1; } } close (fd[0]); close (fd[1]); if (syscall (__NR_pipe2, fd, O_CLOEXEC) != 0) { puts ("pipe2(O_CLOEXEC) failed"); return 1; } for (int i = 0; i < 2; ++i) { int coe = fcntl (fd[i], F_GETFD); if (coe == -1) { puts ("fcntl failed"); return 1; } if ((coe & FD_CLOEXEC) == 0) { printf ("pipe2(O_CLOEXEC) does not set close-on-exit for fd[%d]\n", i); return 1; } } close (fd[0]); close (fd[1]); puts ("OK"); return 0; } ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Signed-off-by: Ulrich Drepper Acked-by: Davide Libenzi Cc: Michael Kerrisk Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/ia32/sys_ia32.c | 2 +- arch/ia64/kernel/sys_ia64.c | 2 +- arch/mips/kernel/syscall.c | 2 +- arch/parisc/hpux/sys_hpux.c | 2 +- arch/sh/kernel/sys_sh32.c | 2 +- arch/sparc/kernel/sys_sparc.c | 2 +- arch/sparc64/kernel/sys_sparc.c | 2 +- arch/x86/ia32/ia32entry.S | 1 + arch/x86/ia32/sys_ia32.c | 2 +- arch/x86/kernel/syscall_table_32.S | 1 + arch/xtensa/kernel/syscall.c | 2 +- fs/pipe.c | 23 ++++++++++++++++++----- include/asm-x86/unistd_32.h | 1 + include/asm-x86/unistd_64.h | 2 ++ include/linux/fs.h | 1 + 15 files changed, 33 insertions(+), 14 deletions(-) diff --git a/arch/ia64/ia32/sys_ia32.c b/arch/ia64/ia32/sys_ia32.c index 7e028ceb93ba..465116aecb85 100644 --- a/arch/ia64/ia32/sys_ia32.c +++ b/arch/ia64/ia32/sys_ia32.c @@ -1139,7 +1139,7 @@ sys32_pipe (int __user *fd) int retval; int fds[2]; - retval = do_pipe(fds); + retval = do_pipe_flags(fds, 0); if (retval) goto out; if (copy_to_user(fd, fds, sizeof(fds))) diff --git a/arch/ia64/kernel/sys_ia64.c b/arch/ia64/kernel/sys_ia64.c index 1eda194b9559..bcbb6d8792d3 100644 --- a/arch/ia64/kernel/sys_ia64.c +++ b/arch/ia64/kernel/sys_ia64.c @@ -160,7 +160,7 @@ sys_pipe (void) int fd[2]; int retval; - retval = do_pipe(fd); + retval = do_pipe_flags(fd, 0); if (retval) goto out; retval = fd[0]; diff --git a/arch/mips/kernel/syscall.c b/arch/mips/kernel/syscall.c index 3523c8d12eda..343015a2f418 100644 --- a/arch/mips/kernel/syscall.c +++ b/arch/mips/kernel/syscall.c @@ -52,7 +52,7 @@ asmlinkage int sysm_pipe(nabi_no_regargs volatile struct pt_regs regs) int fd[2]; int error, res; - error = do_pipe(fd); + error = do_pipe_flags(fd, 0); if (error) { res = error; goto out; diff --git a/arch/parisc/hpux/sys_hpux.c b/arch/parisc/hpux/sys_hpux.c index 0c5b9dabb475..be255ebb609c 100644 --- a/arch/parisc/hpux/sys_hpux.c +++ b/arch/parisc/hpux/sys_hpux.c @@ -448,7 +448,7 @@ int hpux_pipe(int *kstack_fildes) int error; lock_kernel(); - error = do_pipe(kstack_fildes); + error = do_pipe_flags(kstack_fildes, 0); unlock_kernel(); return error; } diff --git a/arch/sh/kernel/sys_sh32.c b/arch/sh/kernel/sys_sh32.c index 125e493ead82..f0aa5c398656 100644 --- a/arch/sh/kernel/sys_sh32.c +++ b/arch/sh/kernel/sys_sh32.c @@ -29,7 +29,7 @@ asmlinkage int sys_pipe(unsigned long r4, unsigned long r5, int fd[2]; int error; - error = do_pipe(fd); + error = do_pipe_flags(fd, 0); if (!error) { regs->regs[1] = fd[1]; return fd[0]; diff --git a/arch/sparc/kernel/sys_sparc.c b/arch/sparc/kernel/sys_sparc.c index 3c6b49a53ae8..4d73421559c3 100644 --- a/arch/sparc/kernel/sys_sparc.c +++ b/arch/sparc/kernel/sys_sparc.c @@ -97,7 +97,7 @@ asmlinkage int sparc_pipe(struct pt_regs *regs) int fd[2]; int error; - error = do_pipe(fd); + error = do_pipe_flags(fd, 0); if (error) goto out; regs->u_regs[UREG_I1] = fd[1]; diff --git a/arch/sparc64/kernel/sys_sparc.c b/arch/sparc64/kernel/sys_sparc.c index e1f4eba2e576..39749e32dc7e 100644 --- a/arch/sparc64/kernel/sys_sparc.c +++ b/arch/sparc64/kernel/sys_sparc.c @@ -418,7 +418,7 @@ asmlinkage long sparc_pipe(struct pt_regs *regs) int fd[2]; int error; - error = do_pipe(fd); + error = do_pipe_flags(fd, 0); if (error) goto out; regs->u_regs[UREG_I1] = fd[1]; diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S index 5614a8f7bed4..18808b164570 100644 --- a/arch/x86/ia32/ia32entry.S +++ b/arch/x86/ia32/ia32entry.S @@ -830,4 +830,5 @@ ia32_sys_call_table: .quad sys_eventfd2 .quad sys_epoll_create2 .quad sys_dup3 /* 330 */ + .quad sys_pipe2 ia32_syscall_end: diff --git a/arch/x86/ia32/sys_ia32.c b/arch/x86/ia32/sys_ia32.c index f00afdf61e67..d3c64088b981 100644 --- a/arch/x86/ia32/sys_ia32.c +++ b/arch/x86/ia32/sys_ia32.c @@ -238,7 +238,7 @@ asmlinkage long sys32_pipe(int __user *fd) int retval; int fds[2]; - retval = do_pipe(fds); + retval = do_pipe_flags(fds, 0); if (retval) goto out; if (copy_to_user(fd, fds, sizeof(fds))) diff --git a/arch/x86/kernel/syscall_table_32.S b/arch/x86/kernel/syscall_table_32.S index 24a3f1ea6a0e..66154769d52f 100644 --- a/arch/x86/kernel/syscall_table_32.S +++ b/arch/x86/kernel/syscall_table_32.S @@ -330,3 +330,4 @@ ENTRY(sys_call_table) .long sys_eventfd2 .long sys_epoll_create2 .long sys_dup3 /* 330 */ + .long sys_pipe2 diff --git a/arch/xtensa/kernel/syscall.c b/arch/xtensa/kernel/syscall.c index f3e16efcd47a..ac15ecbdf919 100644 --- a/arch/xtensa/kernel/syscall.c +++ b/arch/xtensa/kernel/syscall.c @@ -49,7 +49,7 @@ asmlinkage long xtensa_pipe(int __user *userfds) int fd[2]; int error; - error = do_pipe(fd); + error = do_pipe_flags(fd, 0); if (!error) { if (copy_to_user(userfds, fd, 2 * sizeof(int))) error = -EFAULT; diff --git a/fs/pipe.c b/fs/pipe.c index 700f4e0d9572..68e82061070c 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -1027,12 +1027,15 @@ struct file *create_read_pipe(struct file *wrf) return f; } -int do_pipe(int *fd) +int do_pipe_flags(int *fd, int flags) { struct file *fw, *fr; int error; int fdw, fdr; + if (flags & ~O_CLOEXEC) + return -EINVAL; + fw = create_write_pipe(); if (IS_ERR(fw)) return PTR_ERR(fw); @@ -1041,12 +1044,12 @@ int do_pipe(int *fd) if (IS_ERR(fr)) goto err_write_pipe; - error = get_unused_fd(); + error = get_unused_fd_flags(flags); if (error < 0) goto err_read_pipe; fdr = error; - error = get_unused_fd(); + error = get_unused_fd_flags(flags); if (error < 0) goto err_fdr; fdw = error; @@ -1074,16 +1077,21 @@ int do_pipe(int *fd) return error; } +int do_pipe(int *fd) +{ + return do_pipe_flags(fd, 0); +} + /* * sys_pipe() is the normal C calling standard for creating * a pipe. It's not the way Unix traditionally does this, though. */ -asmlinkage long __weak sys_pipe(int __user *fildes) +asmlinkage long __weak sys_pipe2(int __user *fildes, int flags) { int fd[2]; int error; - error = do_pipe(fd); + error = do_pipe_flags(fd, flags); if (!error) { if (copy_to_user(fildes, fd, sizeof(fd))) { sys_close(fd[0]); @@ -1094,6 +1102,11 @@ asmlinkage long __weak sys_pipe(int __user *fildes) return error; } +asmlinkage long __weak sys_pipe(int __user *fildes) +{ + return sys_pipe2(fildes, 0); +} + /* * pipefs should _never_ be mounted by userland - too much of security hassle, * no real gain from having the whole whorehouse mounted. So we don't need diff --git a/include/asm-x86/unistd_32.h b/include/asm-x86/unistd_32.h index a1f6383bf695..748a05c77da4 100644 --- a/include/asm-x86/unistd_32.h +++ b/include/asm-x86/unistd_32.h @@ -336,6 +336,7 @@ #define __NR_eventfd2 328 #define __NR_epoll_create2 329 #define __NR_dup3 330 +#define __NR_pipe2 331 #ifdef __KERNEL__ diff --git a/include/asm-x86/unistd_64.h b/include/asm-x86/unistd_64.h index f0fb2bd40cdb..d2284b43ad58 100644 --- a/include/asm-x86/unistd_64.h +++ b/include/asm-x86/unistd_64.h @@ -649,6 +649,8 @@ __SYSCALL(__NR_eventfd2, sys_eventfd2) __SYSCALL(__NR_epoll_create2, sys_epoll_create2) #define __NR_dup3 292 __SYSCALL(__NR_dup3, sys_dup3) +#define __NR_pipe2 293 +__SYSCALL(__NR_pipe2, sys_pipe2) #ifndef __NO_STUBS diff --git a/include/linux/fs.h b/include/linux/fs.h index e5e6a244096c..0e80cd717d32 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1777,6 +1777,7 @@ static inline void allow_write_access(struct file *file) atomic_inc(&file->f_path.dentry->d_inode->i_writecount); } extern int do_pipe(int *); +extern int do_pipe_flags(int *, int); extern struct file *create_read_pipe(struct file *f); extern struct file *create_write_pipe(void); extern void free_write_pipe(struct file *); From 4006553b06306b34054529477b06b68a1c66249b Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 23 Jul 2008 21:29:32 -0700 Subject: [PATCH 288/857] flag parameters: inotify_init This patch introduces the new syscall inotify_init1 (note: the 1 stands for the one parameter the syscall takes, as opposed to no parameter before). The values accepted for this parameter are function-specific and defined in the inotify.h header. Here the values must match the O_* flags, though. In this patch CLOEXEC support is introduced. The following test must be adjusted for architectures other than x86 and x86-64 and in case the syscall numbers changed. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #include #include #include #include #ifndef __NR_inotify_init1 # ifdef __x86_64__ # define __NR_inotify_init1 294 # elif defined __i386__ # define __NR_inotify_init1 332 # else # error "need __NR_inotify_init1" # endif #endif #define IN_CLOEXEC O_CLOEXEC int main (void) { int fd; fd = syscall (__NR_inotify_init1, 0); if (fd == -1) { puts ("inotify_init1(0) failed"); return 1; } int coe = fcntl (fd, F_GETFD); if (coe == -1) { puts ("fcntl failed"); return 1; } if (coe & FD_CLOEXEC) { puts ("inotify_init1(0) set close-on-exit"); return 1; } close (fd); fd = syscall (__NR_inotify_init1, IN_CLOEXEC); if (fd == -1) { puts ("inotify_init1(IN_CLOEXEC) failed"); return 1; } coe = fcntl (fd, F_GETFD); if (coe == -1) { puts ("fcntl failed"); return 1; } if ((coe & FD_CLOEXEC) == 0) { puts ("inotify_init1(O_CLOEXEC) does not set close-on-exit"); return 1; } close (fd); puts ("OK"); return 0; } ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ [akpm@linux-foundation.org: add sys_ni stub] Signed-off-by: Ulrich Drepper Acked-by: Davide Libenzi Cc: Michael Kerrisk Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/ia32/ia32entry.S | 1 + arch/x86/kernel/syscall_table_32.S | 1 + fs/inotify_user.c | 12 ++++++++++-- include/asm-x86/unistd_32.h | 1 + include/asm-x86/unistd_64.h | 2 ++ include/linux/inotify.h | 5 +++++ include/linux/syscalls.h | 1 + kernel/sys_ni.c | 1 + 8 files changed, 22 insertions(+), 2 deletions(-) diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S index 18808b164570..4541073dd837 100644 --- a/arch/x86/ia32/ia32entry.S +++ b/arch/x86/ia32/ia32entry.S @@ -831,4 +831,5 @@ ia32_sys_call_table: .quad sys_epoll_create2 .quad sys_dup3 /* 330 */ .quad sys_pipe2 + .quad sys_inotify_init1 ia32_syscall_end: diff --git a/arch/x86/kernel/syscall_table_32.S b/arch/x86/kernel/syscall_table_32.S index 66154769d52f..f59aba5ff0f0 100644 --- a/arch/x86/kernel/syscall_table_32.S +++ b/arch/x86/kernel/syscall_table_32.S @@ -331,3 +331,4 @@ ENTRY(sys_call_table) .long sys_epoll_create2 .long sys_dup3 /* 330 */ .long sys_pipe2 + .long sys_inotify_init1 diff --git a/fs/inotify_user.c b/fs/inotify_user.c index 6676c06bb7c1..851005998cd4 100644 --- a/fs/inotify_user.c +++ b/fs/inotify_user.c @@ -566,7 +566,7 @@ static const struct inotify_operations inotify_user_ops = { .destroy_watch = free_inotify_user_watch, }; -asmlinkage long sys_inotify_init(void) +asmlinkage long sys_inotify_init1(int flags) { struct inotify_device *dev; struct inotify_handle *ih; @@ -574,7 +574,10 @@ asmlinkage long sys_inotify_init(void) struct file *filp; int fd, ret; - fd = get_unused_fd(); + if (flags & ~IN_CLOEXEC) + return -EINVAL; + + fd = get_unused_fd_flags(flags & O_CLOEXEC); if (fd < 0) return fd; @@ -638,6 +641,11 @@ asmlinkage long sys_inotify_init(void) return ret; } +asmlinkage long sys_inotify_init(void) +{ + return sys_inotify_init1(0); +} + asmlinkage long sys_inotify_add_watch(int fd, const char __user *path, u32 mask) { struct inode *inode; diff --git a/include/asm-x86/unistd_32.h b/include/asm-x86/unistd_32.h index 748a05c77da4..b3daf503ab93 100644 --- a/include/asm-x86/unistd_32.h +++ b/include/asm-x86/unistd_32.h @@ -337,6 +337,7 @@ #define __NR_epoll_create2 329 #define __NR_dup3 330 #define __NR_pipe2 331 +#define __NR_inotify_init1 332 #ifdef __KERNEL__ diff --git a/include/asm-x86/unistd_64.h b/include/asm-x86/unistd_64.h index d2284b43ad58..c8cb88d70c6b 100644 --- a/include/asm-x86/unistd_64.h +++ b/include/asm-x86/unistd_64.h @@ -651,6 +651,8 @@ __SYSCALL(__NR_epoll_create2, sys_epoll_create2) __SYSCALL(__NR_dup3, sys_dup3) #define __NR_pipe2 293 __SYSCALL(__NR_pipe2, sys_pipe2) +#define __NR_inotify_init1 294 +__SYSCALL(__NR_inotify_init1, sys_inotify_init1) #ifndef __NO_STUBS diff --git a/include/linux/inotify.h b/include/linux/inotify.h index 742b917e7d1b..72ef82120512 100644 --- a/include/linux/inotify.h +++ b/include/linux/inotify.h @@ -7,6 +7,8 @@ #ifndef _LINUX_INOTIFY_H #define _LINUX_INOTIFY_H +/* For O_CLOEXEC */ +#include #include /* @@ -63,6 +65,9 @@ struct inotify_event { IN_MOVED_TO | IN_DELETE | IN_CREATE | IN_DELETE_SELF | \ IN_MOVE_SELF) +/* Flags for sys_inotify_init1. */ +#define IN_CLOEXEC O_CLOEXEC + #ifdef __KERNEL__ #include diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 034d3358549e..93a7e7f017a6 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -547,6 +547,7 @@ asmlinkage long sys_get_mempolicy(int __user *policy, unsigned long addr, unsigned long flags); asmlinkage long sys_inotify_init(void); +asmlinkage long sys_inotify_init1(int flags); asmlinkage long sys_inotify_add_watch(int fd, const char __user *path, u32 mask); asmlinkage long sys_inotify_rm_watch(int fd, u32 wd); diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c index 2a361ccdc7ca..bd66ac5406f3 100644 --- a/kernel/sys_ni.c +++ b/kernel/sys_ni.c @@ -96,6 +96,7 @@ cond_syscall(sys_keyctl); cond_syscall(compat_sys_keyctl); cond_syscall(compat_sys_socketcall); cond_syscall(sys_inotify_init); +cond_syscall(sys_inotify_init1); cond_syscall(sys_inotify_add_watch); cond_syscall(sys_inotify_rm_watch); cond_syscall(sys_migrate_pages); From 99829b832997d907c30669bfd17da32151e18f04 Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 23 Jul 2008 21:29:33 -0700 Subject: [PATCH 289/857] flag parameters: NONBLOCK in anon_inode_getfd Building on the previous change to anon_inode_getfd, this patch introduces support for handling of O_NONBLOCK in addition to the already supported O_CLOEXEC. Following patches will take advantage of this support. As can be seen, the additional support for supporting this functionality is minimal. Signed-off-by: Ulrich Drepper Acked-by: Davide Libenzi Cc: Michael Kerrisk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/anon_inodes.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/anon_inodes.c b/fs/anon_inodes.c index 1a4eee620b0d..3662dd44896b 100644 --- a/fs/anon_inodes.c +++ b/fs/anon_inodes.c @@ -116,7 +116,7 @@ int anon_inode_getfd(const char *name, const struct file_operations *fops, file->f_mapping = anon_inode_inode->i_mapping; file->f_pos = 0; - file->f_flags = O_RDWR; + file->f_flags = O_RDWR | (flags & O_NONBLOCK); file->f_version = 0; file->private_data = priv; From 77d2720059618b9b6e827a8b73831eb6c6fad63c Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 23 Jul 2008 21:29:35 -0700 Subject: [PATCH 290/857] flag parameters: NONBLOCK in socket and socketpair This patch introduces support for the SOCK_NONBLOCK flag in socket, socketpair, and paccept. To do this the internal function sock_attach_fd gets an additional parameter which it uses to set the appropriate flag for the file descriptor. Given that in modern, scalable programs almost all socket connections are non-blocking and the minimal additional cost for the new functionality I see no reason not to add this code. The following test must be adjusted for architectures other than x86 and x86-64 and in case the syscall numbers changed. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #include #include #include #include #include #include #include #ifndef __NR_paccept # ifdef __x86_64__ # define __NR_paccept 288 # elif defined __i386__ # define SYS_PACCEPT 18 # define USE_SOCKETCALL 1 # else # error "need __NR_paccept" # endif #endif #ifdef USE_SOCKETCALL # define paccept(fd, addr, addrlen, mask, flags) \ ({ long args[6] = { \ (long) fd, (long) addr, (long) addrlen, (long) mask, 8, (long) flags }; \ syscall (__NR_socketcall, SYS_PACCEPT, args); }) #else # define paccept(fd, addr, addrlen, mask, flags) \ syscall (__NR_paccept, fd, addr, addrlen, mask, 8, flags) #endif #define PORT 57392 #define SOCK_NONBLOCK O_NONBLOCK static pthread_barrier_t b; static void * tf (void *arg) { pthread_barrier_wait (&b); int s = socket (AF_INET, SOCK_STREAM, 0); struct sockaddr_in sin; sin.sin_family = AF_INET; sin.sin_addr.s_addr = htonl (INADDR_LOOPBACK); sin.sin_port = htons (PORT); connect (s, (const struct sockaddr *) &sin, sizeof (sin)); close (s); pthread_barrier_wait (&b); pthread_barrier_wait (&b); s = socket (AF_INET, SOCK_STREAM, 0); sin.sin_port = htons (PORT); connect (s, (const struct sockaddr *) &sin, sizeof (sin)); close (s); pthread_barrier_wait (&b); return NULL; } int main (void) { int fd; fd = socket (PF_INET, SOCK_STREAM, 0); if (fd == -1) { puts ("socket(0) failed"); return 1; } int fl = fcntl (fd, F_GETFL); if (fl == -1) { puts ("fcntl failed"); return 1; } if (fl & O_NONBLOCK) { puts ("socket(0) set non-blocking mode"); return 1; } close (fd); fd = socket (PF_INET, SOCK_STREAM|SOCK_NONBLOCK, 0); if (fd == -1) { puts ("socket(SOCK_NONBLOCK) failed"); return 1; } fl = fcntl (fd, F_GETFL); if (fl == -1) { puts ("fcntl failed"); return 1; } if ((fl & O_NONBLOCK) == 0) { puts ("socket(SOCK_NONBLOCK) does not set non-blocking mode"); return 1; } close (fd); int fds[2]; if (socketpair (PF_UNIX, SOCK_STREAM, 0, fds) == -1) { puts ("socketpair(0) failed"); return 1; } for (int i = 0; i < 2; ++i) { fl = fcntl (fds[i], F_GETFL); if (fl == -1) { puts ("fcntl failed"); return 1; } if (fl & O_NONBLOCK) { printf ("socketpair(0) set non-blocking mode for fds[%d]\n", i); return 1; } close (fds[i]); } if (socketpair (PF_UNIX, SOCK_STREAM|SOCK_NONBLOCK, 0, fds) == -1) { puts ("socketpair(SOCK_NONBLOCK) failed"); return 1; } for (int i = 0; i < 2; ++i) { fl = fcntl (fds[i], F_GETFL); if (fl == -1) { puts ("fcntl failed"); return 1; } if ((fl & O_NONBLOCK) == 0) { printf ("socketpair(SOCK_NONBLOCK) does not set non-blocking mode for fds[%d]\n", i); return 1; } close (fds[i]); } pthread_barrier_init (&b, NULL, 2); struct sockaddr_in sin; pthread_t th; if (pthread_create (&th, NULL, tf, NULL) != 0) { puts ("pthread_create failed"); return 1; } int s = socket (AF_INET, SOCK_STREAM, 0); int reuse = 1; setsockopt (s, SOL_SOCKET, SO_REUSEADDR, &reuse, sizeof (reuse)); sin.sin_family = AF_INET; sin.sin_addr.s_addr = htonl (INADDR_LOOPBACK); sin.sin_port = htons (PORT); bind (s, (struct sockaddr *) &sin, sizeof (sin)); listen (s, SOMAXCONN); pthread_barrier_wait (&b); int s2 = paccept (s, NULL, 0, NULL, 0); if (s2 < 0) { puts ("paccept(0) failed"); return 1; } fl = fcntl (s2, F_GETFL); if (fl & O_NONBLOCK) { puts ("paccept(0) set non-blocking mode"); return 1; } close (s2); close (s); pthread_barrier_wait (&b); s = socket (AF_INET, SOCK_STREAM, 0); sin.sin_port = htons (PORT); setsockopt (s, SOL_SOCKET, SO_REUSEADDR, &reuse, sizeof (reuse)); bind (s, (struct sockaddr *) &sin, sizeof (sin)); listen (s, SOMAXCONN); pthread_barrier_wait (&b); s2 = paccept (s, NULL, 0, NULL, SOCK_NONBLOCK); if (s2 < 0) { puts ("paccept(SOCK_NONBLOCK) failed"); return 1; } fl = fcntl (s2, F_GETFL); if ((fl & O_NONBLOCK) == 0) { puts ("paccept(SOCK_NONBLOCK) does not set non-blocking mode"); return 1; } close (s2); close (s); pthread_barrier_wait (&b); puts ("OK"); return 0; } ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Signed-off-by: Ulrich Drepper Acked-by: Davide Libenzi Cc: Michael Kerrisk Cc: "David S. Miller" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/net.h | 2 +- net/socket.c | 20 ++++++++++---------- 2 files changed, 11 insertions(+), 11 deletions(-) diff --git a/include/linux/net.h b/include/linux/net.h index 39a23af059b4..2f999fbb188d 100644 --- a/include/linux/net.h +++ b/include/linux/net.h @@ -20,7 +20,7 @@ #include #include -#include /* For O_CLOEXEC */ +#include /* For O_CLOEXEC and O_NONBLOCK */ #include struct poll_table_struct; diff --git a/net/socket.c b/net/socket.c index d163adff95bf..31105f9048a8 100644 --- a/net/socket.c +++ b/net/socket.c @@ -369,7 +369,7 @@ static int sock_alloc_fd(struct file **filep, int flags) return fd; } -static int sock_attach_fd(struct socket *sock, struct file *file) +static int sock_attach_fd(struct socket *sock, struct file *file, int flags) { struct dentry *dentry; struct qstr name = { .name = "" }; @@ -391,7 +391,7 @@ static int sock_attach_fd(struct socket *sock, struct file *file) init_file(file, sock_mnt, dentry, FMODE_READ | FMODE_WRITE, &socket_file_ops); SOCK_INODE(sock)->i_fop = &socket_file_ops; - file->f_flags = O_RDWR; + file->f_flags = O_RDWR | (flags & O_NONBLOCK); file->f_pos = 0; file->private_data = sock; @@ -404,7 +404,7 @@ int sock_map_fd(struct socket *sock, int flags) int fd = sock_alloc_fd(&newfile, flags); if (likely(fd >= 0)) { - int err = sock_attach_fd(sock, newfile); + int err = sock_attach_fd(sock, newfile, flags); if (unlikely(err < 0)) { put_filp(newfile); @@ -1223,7 +1223,7 @@ asmlinkage long sys_socket(int family, int type, int protocol) int flags; flags = type & ~SOCK_TYPE_MASK; - if (flags & ~SOCK_CLOEXEC) + if (flags & ~(SOCK_CLOEXEC | SOCK_NONBLOCK)) return -EINVAL; type &= SOCK_TYPE_MASK; @@ -1234,7 +1234,7 @@ asmlinkage long sys_socket(int family, int type, int protocol) if (retval < 0) goto out; - retval = sock_map_fd(sock, flags & O_CLOEXEC); + retval = sock_map_fd(sock, flags & (O_CLOEXEC | O_NONBLOCK)); if (retval < 0) goto out_release; @@ -1260,7 +1260,7 @@ asmlinkage long sys_socketpair(int family, int type, int protocol, int flags; flags = type & ~SOCK_TYPE_MASK; - if (flags & ~SOCK_CLOEXEC) + if (flags & ~(SOCK_CLOEXEC | SOCK_NONBLOCK)) return -EINVAL; type &= SOCK_TYPE_MASK; @@ -1298,12 +1298,12 @@ asmlinkage long sys_socketpair(int family, int type, int protocol, goto out_release_both; } - err = sock_attach_fd(sock1, newfile1); + err = sock_attach_fd(sock1, newfile1, flags & O_NONBLOCK); if (unlikely(err < 0)) { goto out_fd2; } - err = sock_attach_fd(sock2, newfile2); + err = sock_attach_fd(sock2, newfile2, flags & O_NONBLOCK); if (unlikely(err < 0)) { fput(newfile1); goto out_fd1; @@ -1429,7 +1429,7 @@ long do_accept(int fd, struct sockaddr __user *upeer_sockaddr, int err, len, newfd, fput_needed; struct sockaddr_storage address; - if (flags & ~SOCK_CLOEXEC) + if (flags & ~(SOCK_CLOEXEC | SOCK_NONBLOCK)) return -EINVAL; if (SOCK_NONBLOCK != O_NONBLOCK && (flags & SOCK_NONBLOCK)) @@ -1459,7 +1459,7 @@ long do_accept(int fd, struct sockaddr __user *upeer_sockaddr, goto out_put; } - err = sock_attach_fd(newsock, newfile); + err = sock_attach_fd(newsock, newfile, flags & O_NONBLOCK); if (err < 0) goto out_fd_simple; From 5fb5e04926a54bc1c22bba7ca166840f4476196f Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 23 Jul 2008 21:29:37 -0700 Subject: [PATCH 291/857] flag parameters: NONBLOCK in signalfd This patch adds support for the SFD_NONBLOCK flag to signalfd4. The additional changes needed are minimal. The following test must be adjusted for architectures other than x86 and x86-64 and in case the syscall numbers changed. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #include #include #include #include #include #ifndef __NR_signalfd4 # ifdef __x86_64__ # define __NR_signalfd4 289 # elif defined __i386__ # define __NR_signalfd4 327 # else # error "need __NR_signalfd4" # endif #endif #define SFD_NONBLOCK O_NONBLOCK int main (void) { sigset_t ss; sigemptyset (&ss); sigaddset (&ss, SIGUSR1); int fd = syscall (__NR_signalfd4, -1, &ss, 8, 0); if (fd == -1) { puts ("signalfd4(0) failed"); return 1; } int fl = fcntl (fd, F_GETFL); if (fl == -1) { puts ("fcntl failed"); return 1; } if (fl & O_NONBLOCK) { puts ("signalfd4(0) set non-blocking mode"); return 1; } close (fd); fd = syscall (__NR_signalfd4, -1, &ss, 8, SFD_NONBLOCK); if (fd == -1) { puts ("signalfd4(SFD_NONBLOCK) failed"); return 1; } fl = fcntl (fd, F_GETFL); if (fl == -1) { puts ("fcntl failed"); return 1; } if ((fl & O_NONBLOCK) == 0) { puts ("signalfd4(SFD_NONBLOCK) does not set non-blocking mode"); return 1; } close (fd); puts ("OK"); return 0; } ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Signed-off-by: Ulrich Drepper Acked-by: Davide Libenzi Cc: Michael Kerrisk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/signalfd.c | 4 ++-- include/linux/signalfd.h | 3 ++- 2 files changed, 4 insertions(+), 3 deletions(-) diff --git a/fs/signalfd.c b/fs/signalfd.c index c8609fa51a13..5441a4bca772 100644 --- a/fs/signalfd.c +++ b/fs/signalfd.c @@ -211,7 +211,7 @@ asmlinkage long sys_signalfd4(int ufd, sigset_t __user *user_mask, sigset_t sigmask; struct signalfd_ctx *ctx; - if (flags & ~SFD_CLOEXEC) + if (flags & ~(SFD_CLOEXEC | SFD_NONBLOCK)) return -EINVAL; if (sizemask != sizeof(sigset_t) || @@ -232,7 +232,7 @@ asmlinkage long sys_signalfd4(int ufd, sigset_t __user *user_mask, * anon_inode_getfd() will install the fd. */ ufd = anon_inode_getfd("[signalfd]", &signalfd_fops, ctx, - flags & O_CLOEXEC); + flags & (O_CLOEXEC | O_NONBLOCK)); if (ufd < 0) kfree(ctx); } else { diff --git a/include/linux/signalfd.h b/include/linux/signalfd.h index 8b3f7b7420a1..bef0c46d4713 100644 --- a/include/linux/signalfd.h +++ b/include/linux/signalfd.h @@ -8,11 +8,12 @@ #ifndef _LINUX_SIGNALFD_H #define _LINUX_SIGNALFD_H -/* For O_CLOEXEC */ +/* For O_CLOEXEC and O_NONBLOCK */ #include /* Flags for signalfd4. */ #define SFD_CLOEXEC O_CLOEXEC +#define SFD_NONBLOCK O_NONBLOCK struct signalfd_siginfo { __u32 ssi_signo; From e7d476dfdf0bcfed478a207aecfdc84f81efecaf Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 23 Jul 2008 21:29:38 -0700 Subject: [PATCH 292/857] flag parameters: NONBLOCK in eventfd This patch adds support for the EFD_NONBLOCK flag to eventfd2. The additional changes needed are minimal. The following test must be adjusted for architectures other than x86 and x86-64 and in case the syscall numbers changed. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #include #include #include #include #ifndef __NR_eventfd2 # ifdef __x86_64__ # define __NR_eventfd2 290 # elif defined __i386__ # define __NR_eventfd2 328 # else # error "need __NR_eventfd2" # endif #endif #define EFD_NONBLOCK O_NONBLOCK int main (void) { int fd = syscall (__NR_eventfd2, 1, 0); if (fd == -1) { puts ("eventfd2(0) failed"); return 1; } int fl = fcntl (fd, F_GETFL); if (fl == -1) { puts ("fcntl failed"); return 1; } if (fl & O_NONBLOCK) { puts ("eventfd2(0) sets non-blocking mode"); return 1; } close (fd); fd = syscall (__NR_eventfd2, 1, EFD_NONBLOCK); if (fd == -1) { puts ("eventfd2(EFD_NONBLOCK) failed"); return 1; } fl = fcntl (fd, F_GETFL); if (fl == -1) { puts ("fcntl failed"); return 1; } if ((fl & O_NONBLOCK) == 0) { puts ("eventfd2(EFD_NONBLOCK) does not set non-blocking mode"); return 1; } close (fd); puts ("OK"); return 0; } ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Signed-off-by: Ulrich Drepper Acked-by: Davide Libenzi Cc: Michael Kerrisk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/eventfd.c | 4 ++-- include/linux/eventfd.h | 3 ++- 2 files changed, 4 insertions(+), 3 deletions(-) diff --git a/fs/eventfd.c b/fs/eventfd.c index bd420e6478ad..3ed4466177a7 100644 --- a/fs/eventfd.c +++ b/fs/eventfd.c @@ -203,7 +203,7 @@ asmlinkage long sys_eventfd2(unsigned int count, int flags) int fd; struct eventfd_ctx *ctx; - if (flags & ~EFD_CLOEXEC) + if (flags & ~(EFD_CLOEXEC | EFD_NONBLOCK)) return -EINVAL; ctx = kmalloc(sizeof(*ctx), GFP_KERNEL); @@ -218,7 +218,7 @@ asmlinkage long sys_eventfd2(unsigned int count, int flags) * anon_inode_getfd() will install the fd. */ fd = anon_inode_getfd("[eventfd]", &eventfd_fops, ctx, - flags & O_CLOEXEC); + flags & (O_CLOEXEC | O_NONBLOCK)); if (fd < 0) kfree(ctx); return fd; diff --git a/include/linux/eventfd.h b/include/linux/eventfd.h index a6c0eaedb1b0..a667637b54e3 100644 --- a/include/linux/eventfd.h +++ b/include/linux/eventfd.h @@ -10,11 +10,12 @@ #ifdef CONFIG_EVENTFD -/* For O_CLOEXEC */ +/* For O_CLOEXEC and O_NONBLOCK */ #include /* Flags for eventfd2. */ #define EFD_CLOEXEC O_CLOEXEC +#define EFD_NONBLOCK O_NONBLOCK struct file *eventfd_fget(int fd); int eventfd_signal(struct file *file, int n); From 6b1ef0e60d42f2fdaec26baee8327eb156347b4f Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 23 Jul 2008 21:29:39 -0700 Subject: [PATCH 293/857] flag parameters: NONBLOCK in timerfd_create This patch adds support for the TFD_NONBLOCK flag to timerfd_create. The additional changes needed are minimal. The following test must be adjusted for architectures other than x86 and x86-64 and in case the syscall numbers changed. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #include #include #include #include #include #ifndef __NR_timerfd_create # ifdef __x86_64__ # define __NR_timerfd_create 283 # elif defined __i386__ # define __NR_timerfd_create 322 # else # error "need __NR_timerfd_create" # endif #endif #define TFD_NONBLOCK O_NONBLOCK int main (void) { int fd = syscall (__NR_timerfd_create, CLOCK_REALTIME, 0); if (fd == -1) { puts ("timerfd_create(0) failed"); return 1; } int fl = fcntl (fd, F_GETFL); if (fl == -1) { puts ("fcntl failed"); return 1; } if (fl & O_NONBLOCK) { puts ("timerfd_create(0) set non-blocking mode"); return 1; } close (fd); fd = syscall (__NR_timerfd_create, CLOCK_REALTIME, TFD_NONBLOCK); if (fd == -1) { puts ("timerfd_create(TFD_NONBLOCK) failed"); return 1; } fl = fcntl (fd, F_GETFL); if (fl == -1) { puts ("fcntl failed"); return 1; } if ((fl & O_NONBLOCK) == 0) { puts ("timerfd_create(TFD_NONBLOCK) set non-blocking mode"); return 1; } close (fd); puts ("OK"); return 0; } ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Signed-off-by: Ulrich Drepper Acked-by: Davide Libenzi Cc: Michael Kerrisk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/timerfd.c | 4 ++-- include/linux/timerfd.h | 3 ++- 2 files changed, 4 insertions(+), 3 deletions(-) diff --git a/fs/timerfd.c b/fs/timerfd.c index c6ef5e33cb34..75d44efe346c 100644 --- a/fs/timerfd.c +++ b/fs/timerfd.c @@ -184,7 +184,7 @@ asmlinkage long sys_timerfd_create(int clockid, int flags) int ufd; struct timerfd_ctx *ctx; - if (flags & ~TFD_CLOEXEC) + if (flags & ~(TFD_CLOEXEC | TFD_NONBLOCK)) return -EINVAL; if (clockid != CLOCK_MONOTONIC && clockid != CLOCK_REALTIME) @@ -199,7 +199,7 @@ asmlinkage long sys_timerfd_create(int clockid, int flags) hrtimer_init(&ctx->tmr, clockid, HRTIMER_MODE_ABS); ufd = anon_inode_getfd("[timerfd]", &timerfd_fops, ctx, - flags & O_CLOEXEC); + flags & (O_CLOEXEC | O_NONBLOCK)); if (ufd < 0) kfree(ctx); diff --git a/include/linux/timerfd.h b/include/linux/timerfd.h index 96ed97dff00f..86cb0501d3e2 100644 --- a/include/linux/timerfd.h +++ b/include/linux/timerfd.h @@ -8,7 +8,7 @@ #ifndef _LINUX_TIMERFD_H #define _LINUX_TIMERFD_H -/* For O_CLOEXEC */ +/* For O_CLOEXEC and O_NONBLOCK */ #include /* Flags for timerfd_settime. */ @@ -16,6 +16,7 @@ /* Flags for timerfd_create. */ #define TFD_CLOEXEC O_CLOEXEC +#define TFD_NONBLOCK O_NONBLOCK #endif /* _LINUX_TIMERFD_H */ From be61a86d7237dd80510615f38ae21d6e1e98660c Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 23 Jul 2008 21:29:40 -0700 Subject: [PATCH 294/857] flag parameters: NONBLOCK in pipe This patch adds O_NONBLOCK support to pipe2. It is minimally more involved than the patches for eventfd et.al but still trivial. The interfaces of the create_write_pipe and create_read_pipe helper functions were changed and the one other caller as well. The following test must be adjusted for architectures other than x86 and x86-64 and in case the syscall numbers changed. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #include #include #include #include #ifndef __NR_pipe2 # ifdef __x86_64__ # define __NR_pipe2 293 # elif defined __i386__ # define __NR_pipe2 331 # else # error "need __NR_pipe2" # endif #endif int main (void) { int fds[2]; if (syscall (__NR_pipe2, fds, 0) == -1) { puts ("pipe2(0) failed"); return 1; } for (int i = 0; i < 2; ++i) { int fl = fcntl (fds[i], F_GETFL); if (fl == -1) { puts ("fcntl failed"); return 1; } if (fl & O_NONBLOCK) { printf ("pipe2(0) set non-blocking mode for fds[%d]\n", i); return 1; } close (fds[i]); } if (syscall (__NR_pipe2, fds, O_NONBLOCK) == -1) { puts ("pipe2(O_NONBLOCK) failed"); return 1; } for (int i = 0; i < 2; ++i) { int fl = fcntl (fds[i], F_GETFL); if (fl == -1) { puts ("fcntl failed"); return 1; } if ((fl & O_NONBLOCK) == 0) { printf ("pipe2(O_NONBLOCK) does not set non-blocking mode for fds[%d]\n", i); return 1; } close (fds[i]); } puts ("OK"); return 0; } ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Signed-off-by: Ulrich Drepper Acked-by: Davide Libenzi Cc: Michael Kerrisk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/pipe.c | 14 +++++++------- include/linux/fs.h | 4 ++-- kernel/kmod.c | 4 ++-- 3 files changed, 11 insertions(+), 11 deletions(-) diff --git a/fs/pipe.c b/fs/pipe.c index 68e82061070c..10c4e9aa5c49 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -950,7 +950,7 @@ static struct inode * get_pipe_inode(void) return NULL; } -struct file *create_write_pipe(void) +struct file *create_write_pipe(int flags) { int err; struct inode *inode; @@ -983,7 +983,7 @@ struct file *create_write_pipe(void) goto err_dentry; f->f_mapping = inode->i_mapping; - f->f_flags = O_WRONLY; + f->f_flags = O_WRONLY | (flags & O_NONBLOCK); f->f_version = 0; return f; @@ -1007,7 +1007,7 @@ void free_write_pipe(struct file *f) put_filp(f); } -struct file *create_read_pipe(struct file *wrf) +struct file *create_read_pipe(struct file *wrf, int flags) { struct file *f = get_empty_filp(); if (!f) @@ -1019,7 +1019,7 @@ struct file *create_read_pipe(struct file *wrf) f->f_mapping = wrf->f_path.dentry->d_inode->i_mapping; f->f_pos = 0; - f->f_flags = O_RDONLY; + f->f_flags = O_RDONLY | (flags & O_NONBLOCK); f->f_op = &read_pipe_fops; f->f_mode = FMODE_READ; f->f_version = 0; @@ -1033,13 +1033,13 @@ int do_pipe_flags(int *fd, int flags) int error; int fdw, fdr; - if (flags & ~O_CLOEXEC) + if (flags & ~(O_CLOEXEC | O_NONBLOCK)) return -EINVAL; - fw = create_write_pipe(); + fw = create_write_pipe(flags); if (IS_ERR(fw)) return PTR_ERR(fw); - fr = create_read_pipe(fw); + fr = create_read_pipe(fw, flags); error = PTR_ERR(fr); if (IS_ERR(fr)) goto err_write_pipe; diff --git a/include/linux/fs.h b/include/linux/fs.h index 0e80cd717d32..4b86f806014c 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1778,8 +1778,8 @@ static inline void allow_write_access(struct file *file) } extern int do_pipe(int *); extern int do_pipe_flags(int *, int); -extern struct file *create_read_pipe(struct file *f); -extern struct file *create_write_pipe(void); +extern struct file *create_read_pipe(struct file *f, int flags); +extern struct file *create_write_pipe(int flags); extern void free_write_pipe(struct file *); extern struct file *do_filp_open(int dfd, const char *pathname, diff --git a/kernel/kmod.c b/kernel/kmod.c index 90d7af1c1655..2989f67c4446 100644 --- a/kernel/kmod.c +++ b/kernel/kmod.c @@ -417,12 +417,12 @@ int call_usermodehelper_stdinpipe(struct subprocess_info *sub_info, { struct file *f; - f = create_write_pipe(); + f = create_write_pipe(0); if (IS_ERR(f)) return PTR_ERR(f); *filp = f; - f = create_read_pipe(f); + f = create_read_pipe(f, 0); if (IS_ERR(f)) { free_write_pipe(*filp); return PTR_ERR(f); From 510df2dd482496083e1c3b1a8c9b6afd5fa4c7d7 Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 23 Jul 2008 21:29:41 -0700 Subject: [PATCH 295/857] flag parameters: NONBLOCK in inotify_init This patch adds non-blocking support for inotify_init1. The additional changes needed are minimal. The following test must be adjusted for architectures other than x86 and x86-64 and in case the syscall numbers changed. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #include #include #include #include #ifndef __NR_inotify_init1 # ifdef __x86_64__ # define __NR_inotify_init1 294 # elif defined __i386__ # define __NR_inotify_init1 332 # else # error "need __NR_inotify_init1" # endif #endif #define IN_NONBLOCK O_NONBLOCK int main (void) { int fd = syscall (__NR_inotify_init1, 0); if (fd == -1) { puts ("inotify_init1(0) failed"); return 1; } int fl = fcntl (fd, F_GETFL); if (fl == -1) { puts ("fcntl failed"); return 1; } if (fl & O_NONBLOCK) { puts ("inotify_init1(0) set non-blocking mode"); return 1; } close (fd); fd = syscall (__NR_inotify_init1, IN_NONBLOCK); if (fd == -1) { puts ("inotify_init1(IN_NONBLOCK) failed"); return 1; } fl = fcntl (fd, F_GETFL); if (fl == -1) { puts ("fcntl failed"); return 1; } if ((fl & O_NONBLOCK) == 0) { puts ("inotify_init1(IN_NONBLOCK) set non-blocking mode"); return 1; } close (fd); puts ("OK"); return 0; } ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Signed-off-by: Ulrich Drepper Acked-by: Davide Libenzi Cc: Michael Kerrisk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/inotify_user.c | 4 ++-- include/linux/inotify.h | 3 ++- 2 files changed, 4 insertions(+), 3 deletions(-) diff --git a/fs/inotify_user.c b/fs/inotify_user.c index 851005998cd4..dc7e1f619748 100644 --- a/fs/inotify_user.c +++ b/fs/inotify_user.c @@ -574,7 +574,7 @@ asmlinkage long sys_inotify_init1(int flags) struct file *filp; int fd, ret; - if (flags & ~IN_CLOEXEC) + if (flags & ~(IN_CLOEXEC | IN_NONBLOCK)) return -EINVAL; fd = get_unused_fd_flags(flags & O_CLOEXEC); @@ -613,7 +613,7 @@ asmlinkage long sys_inotify_init1(int flags) filp->f_path.dentry = dget(inotify_mnt->mnt_root); filp->f_mapping = filp->f_path.dentry->d_inode->i_mapping; filp->f_mode = FMODE_READ; - filp->f_flags = O_RDONLY; + filp->f_flags = O_RDONLY | (flags & O_NONBLOCK); filp->private_data = dev; INIT_LIST_HEAD(&dev->events); diff --git a/include/linux/inotify.h b/include/linux/inotify.h index 72ef82120512..bd578578a8b9 100644 --- a/include/linux/inotify.h +++ b/include/linux/inotify.h @@ -7,7 +7,7 @@ #ifndef _LINUX_INOTIFY_H #define _LINUX_INOTIFY_H -/* For O_CLOEXEC */ +/* For O_CLOEXEC and O_NONBLOCK */ #include #include @@ -67,6 +67,7 @@ struct inotify_event { /* Flags for sys_inotify_init1. */ #define IN_CLOEXEC O_CLOEXEC +#define IN_NONBLOCK O_NONBLOCK #ifdef __KERNEL__ From e38b36f325153eaadd1c2a7abc5762079233e540 Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 23 Jul 2008 21:29:42 -0700 Subject: [PATCH 296/857] flag parameters: check magic constants This patch adds test that ensure the boundary conditions for the various constants introduced in the previous patches is met. No code is generated. [akpm@linux-foundation.org: fix alpha] Signed-off-by: Ulrich Drepper Acked-by: Davide Libenzi Cc: Michael Kerrisk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/eventfd.c | 4 ++++ fs/eventpoll.c | 3 +++ fs/inotify_user.c | 4 ++++ fs/signalfd.c | 4 ++++ fs/timerfd.c | 4 ++++ net/socket.c | 6 ++++++ 6 files changed, 25 insertions(+) diff --git a/fs/eventfd.c b/fs/eventfd.c index 3ed4466177a7..08bf558d0408 100644 --- a/fs/eventfd.c +++ b/fs/eventfd.c @@ -203,6 +203,10 @@ asmlinkage long sys_eventfd2(unsigned int count, int flags) int fd; struct eventfd_ctx *ctx; + /* Check the EFD_* constants for consistency. */ + BUILD_BUG_ON(EFD_CLOEXEC != O_CLOEXEC); + BUILD_BUG_ON(EFD_NONBLOCK != O_NONBLOCK); + if (flags & ~(EFD_CLOEXEC | EFD_NONBLOCK)) return -EINVAL; diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 3fd4014f3c5a..2fdad4204044 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -1051,6 +1051,9 @@ asmlinkage long sys_epoll_create2(int size, int flags) int error, fd = -1; struct eventpoll *ep; + /* Check the EPOLL_* constant for consistency. */ + BUILD_BUG_ON(EPOLL_CLOEXEC != O_CLOEXEC); + if (flags & ~EPOLL_CLOEXEC) return -EINVAL; diff --git a/fs/inotify_user.c b/fs/inotify_user.c index dc7e1f619748..fe79c25d95dc 100644 --- a/fs/inotify_user.c +++ b/fs/inotify_user.c @@ -574,6 +574,10 @@ asmlinkage long sys_inotify_init1(int flags) struct file *filp; int fd, ret; + /* Check the IN_* constants for consistency. */ + BUILD_BUG_ON(IN_CLOEXEC != O_CLOEXEC); + BUILD_BUG_ON(IN_NONBLOCK != O_NONBLOCK); + if (flags & ~(IN_CLOEXEC | IN_NONBLOCK)) return -EINVAL; diff --git a/fs/signalfd.c b/fs/signalfd.c index 5441a4bca772..9c39bc7f8431 100644 --- a/fs/signalfd.c +++ b/fs/signalfd.c @@ -211,6 +211,10 @@ asmlinkage long sys_signalfd4(int ufd, sigset_t __user *user_mask, sigset_t sigmask; struct signalfd_ctx *ctx; + /* Check the SFD_* constants for consistency. */ + BUILD_BUG_ON(SFD_CLOEXEC != O_CLOEXEC); + BUILD_BUG_ON(SFD_NONBLOCK != O_NONBLOCK); + if (flags & ~(SFD_CLOEXEC | SFD_NONBLOCK)) return -EINVAL; diff --git a/fs/timerfd.c b/fs/timerfd.c index 75d44efe346c..c502c60e4f54 100644 --- a/fs/timerfd.c +++ b/fs/timerfd.c @@ -184,6 +184,10 @@ asmlinkage long sys_timerfd_create(int clockid, int flags) int ufd; struct timerfd_ctx *ctx; + /* Check the TFD_* constants for consistency. */ + BUILD_BUG_ON(TFD_CLOEXEC != O_CLOEXEC); + BUILD_BUG_ON(TFD_NONBLOCK != O_NONBLOCK); + if (flags & ~(TFD_CLOEXEC | TFD_NONBLOCK)) return -EINVAL; if (clockid != CLOCK_MONOTONIC && diff --git a/net/socket.c b/net/socket.c index 31105f9048a8..1310a82cbba7 100644 --- a/net/socket.c +++ b/net/socket.c @@ -1222,6 +1222,12 @@ asmlinkage long sys_socket(int family, int type, int protocol) struct socket *sock; int flags; + /* Check the SOCK_* constants for consistency. */ + BUILD_BUG_ON(SOCK_CLOEXEC != O_CLOEXEC); + BUILD_BUG_ON((SOCK_MAX | SOCK_TYPE_MASK) != SOCK_TYPE_MASK); + BUILD_BUG_ON(SOCK_CLOEXEC & SOCK_TYPE_MASK); + BUILD_BUG_ON(SOCK_NONBLOCK & SOCK_TYPE_MASK); + flags = type & ~SOCK_TYPE_MASK; if (flags & ~(SOCK_CLOEXEC | SOCK_NONBLOCK)) return -EINVAL; From 9fe5ad9c8cef9ad5873d8ee55d1cf00d9b607df0 Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 23 Jul 2008 21:29:43 -0700 Subject: [PATCH 297/857] flag parameters add-on: remove epoll_create size param Remove the size parameter from the new epoll_create syscall and renames the syscall itself. The updated test program follows. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ #include #include #include #include #include #ifndef __NR_epoll_create2 # ifdef __x86_64__ # define __NR_epoll_create2 291 # elif defined __i386__ # define __NR_epoll_create2 329 # else # error "need __NR_epoll_create2" # endif #endif #define EPOLL_CLOEXEC O_CLOEXEC int main (void) { int fd = syscall (__NR_epoll_create2, 0); if (fd == -1) { puts ("epoll_create2(0) failed"); return 1; } int coe = fcntl (fd, F_GETFD); if (coe == -1) { puts ("fcntl failed"); return 1; } if (coe & FD_CLOEXEC) { puts ("epoll_create2(0) set close-on-exec flag"); return 1; } close (fd); fd = syscall (__NR_epoll_create2, EPOLL_CLOEXEC); if (fd == -1) { puts ("epoll_create2(EPOLL_CLOEXEC) failed"); return 1; } coe = fcntl (fd, F_GETFD); if (coe == -1) { puts ("fcntl failed"); return 1; } if ((coe & FD_CLOEXEC) == 0) { puts ("epoll_create2(EPOLL_CLOEXEC) set close-on-exec flag"); return 1; } close (fd); puts ("OK"); return 0; } ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Signed-off-by: Ulrich Drepper Acked-by: Davide Libenzi Cc: Michael Kerrisk Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/ia32/ia32entry.S | 2 +- arch/x86/kernel/syscall_table_32.S | 2 +- fs/eventpoll.c | 18 ++++++++++-------- include/asm-x86/unistd_32.h | 2 +- include/asm-x86/unistd_64.h | 4 ++-- include/linux/eventpoll.h | 2 +- include/linux/syscalls.h | 2 +- 7 files changed, 17 insertions(+), 15 deletions(-) diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S index 4541073dd837..e4bd1793a5e4 100644 --- a/arch/x86/ia32/ia32entry.S +++ b/arch/x86/ia32/ia32entry.S @@ -828,7 +828,7 @@ ia32_sys_call_table: .quad compat_sys_timerfd_gettime .quad compat_sys_signalfd4 .quad sys_eventfd2 - .quad sys_epoll_create2 + .quad sys_epoll_create1 .quad sys_dup3 /* 330 */ .quad sys_pipe2 .quad sys_inotify_init1 diff --git a/arch/x86/kernel/syscall_table_32.S b/arch/x86/kernel/syscall_table_32.S index f59aba5ff0f0..d44395ff34c3 100644 --- a/arch/x86/kernel/syscall_table_32.S +++ b/arch/x86/kernel/syscall_table_32.S @@ -328,7 +328,7 @@ ENTRY(sys_call_table) .long sys_timerfd_gettime .long sys_signalfd4 .long sys_eventfd2 - .long sys_epoll_create2 + .long sys_epoll_create1 .long sys_dup3 /* 330 */ .long sys_pipe2 .long sys_inotify_init1 diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 2fdad4204044..0c87474f7917 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -1046,7 +1046,7 @@ static int ep_poll(struct eventpoll *ep, struct epoll_event __user *events, * RB tree. With the current implementation, the "size" parameter is ignored * (besides sanity checks). */ -asmlinkage long sys_epoll_create2(int size, int flags) +asmlinkage long sys_epoll_create1(int flags) { int error, fd = -1; struct eventpoll *ep; @@ -1058,14 +1058,13 @@ asmlinkage long sys_epoll_create2(int size, int flags) return -EINVAL; DNPRINTK(3, (KERN_INFO "[%p] eventpoll: sys_epoll_create(%d)\n", - current, size)); + current, flags)); /* - * Sanity check on the size parameter, and create the internal data - * structure ( "struct eventpoll" ). + * Create the internal data structure ( "struct eventpoll" ). */ - error = -EINVAL; - if (size <= 0 || (error = ep_alloc(&ep)) < 0) { + error = ep_alloc(&ep); + if (error < 0) { fd = error; goto error_return; } @@ -1081,14 +1080,17 @@ asmlinkage long sys_epoll_create2(int size, int flags) error_return: DNPRINTK(3, (KERN_INFO "[%p] eventpoll: sys_epoll_create(%d) = %d\n", - current, size, fd)); + current, flags, fd)); return fd; } asmlinkage long sys_epoll_create(int size) { - return sys_epoll_create2(size, 0); + if (size < 0) + return -EINVAL; + + return sys_epoll_create1(0); } /* diff --git a/include/asm-x86/unistd_32.h b/include/asm-x86/unistd_32.h index b3daf503ab93..d7394673b772 100644 --- a/include/asm-x86/unistd_32.h +++ b/include/asm-x86/unistd_32.h @@ -334,7 +334,7 @@ #define __NR_timerfd_gettime 326 #define __NR_signalfd4 327 #define __NR_eventfd2 328 -#define __NR_epoll_create2 329 +#define __NR_epoll_create1 329 #define __NR_dup3 330 #define __NR_pipe2 331 #define __NR_inotify_init1 332 diff --git a/include/asm-x86/unistd_64.h b/include/asm-x86/unistd_64.h index c8cb88d70c6b..3a341d791792 100644 --- a/include/asm-x86/unistd_64.h +++ b/include/asm-x86/unistd_64.h @@ -645,8 +645,8 @@ __SYSCALL(__NR_paccept, sys_paccept) __SYSCALL(__NR_signalfd4, sys_signalfd4) #define __NR_eventfd2 290 __SYSCALL(__NR_eventfd2, sys_eventfd2) -#define __NR_epoll_create2 291 -__SYSCALL(__NR_epoll_create2, sys_epoll_create2) +#define __NR_epoll_create1 291 +__SYSCALL(__NR_epoll_create1, sys_epoll_create1) #define __NR_dup3 292 __SYSCALL(__NR_dup3, sys_dup3) #define __NR_pipe2 293 diff --git a/include/linux/eventpoll.h b/include/linux/eventpoll.h index 1cfaa40059c8..f1e1d3c47125 100644 --- a/include/linux/eventpoll.h +++ b/include/linux/eventpoll.h @@ -18,7 +18,7 @@ #include #include -/* Flags for epoll_create2. */ +/* Flags for epoll_create1. */ #define EPOLL_CLOEXEC O_CLOEXEC /* Valid opcodes to issue to sys_epoll_ctl() */ diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 93a7e7f017a6..06f2bf76c030 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -431,7 +431,7 @@ asmlinkage long sys_poll(struct pollfd __user *ufds, unsigned int nfds, asmlinkage long sys_select(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp, struct timeval __user *tvp); asmlinkage long sys_epoll_create(int size); -asmlinkage long sys_epoll_create2(int size, int flags); +asmlinkage long sys_epoll_create1(int flags); asmlinkage long sys_epoll_ctl(int epfd, int op, int fd, struct epoll_event __user *event); asmlinkage long sys_epoll_wait(int epfd, struct epoll_event __user *events, From 920519c1c31ca46ef6caab1a4be102ed0dfb5fbc Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Wed, 23 Jul 2008 21:29:44 -0700 Subject: [PATCH 298/857] serial/8250_gsc.c: add MODULE_LICENSE This patch adds the missing MODULE_LICENSE("GPL"). Signed-off-by: Adrian Bunk Acked-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/serial/8250_gsc.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/serial/8250_gsc.c b/drivers/serial/8250_gsc.c index 4eb7437a404a..0416ad3bc127 100644 --- a/drivers/serial/8250_gsc.c +++ b/drivers/serial/8250_gsc.c @@ -119,3 +119,5 @@ int __init probe_serial_gsc(void) } module_init(probe_serial_gsc); + +MODULE_LICENSE("GPL"); From 7500b1f602aad75901774a67a687ee985d85893f Mon Sep 17 00:00:00 2001 From: Aristeu Rozanski Date: Wed, 23 Jul 2008 21:29:45 -0700 Subject: [PATCH 299/857] 8250: fix break handling for Intel 82571 Intel 82571 has a "Serial Over LAN" feature that doesn't properly implements the receiving of break characters. When a break is received, it doesn't set UART_LSR_DR and unless another character is received, the break won't be received by the application. Signed-off-by: Aristeu Rozanski Acked-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/serial/8250.c | 17 ++++++++++++++--- 1 file changed, 14 insertions(+), 3 deletions(-) diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c index 27f34a9f9cb7..a97f1ae11f78 100644 --- a/drivers/serial/8250.c +++ b/drivers/serial/8250.c @@ -1293,7 +1293,18 @@ receive_chars(struct uart_8250_port *up, unsigned int *status) char flag; do { - ch = serial_inp(up, UART_RX); + if (likely(lsr & UART_LSR_DR)) + ch = serial_inp(up, UART_RX); + else + /* + * Intel 82571 has a Serial Over Lan device that will + * set UART_LSR_BI without setting UART_LSR_DR when + * it receives a break. To avoid reading from the + * receive buffer without UART_LSR_DR bit set, we + * just force the read character to be 0 + */ + ch = 0; + flag = TTY_NORMAL; up->port.icount.rx++; @@ -1342,7 +1353,7 @@ receive_chars(struct uart_8250_port *up, unsigned int *status) ignore_char: lsr = serial_inp(up, UART_LSR); - } while ((lsr & UART_LSR_DR) && (max_count-- > 0)); + } while ((lsr & (UART_LSR_DR | UART_LSR_BI)) && (max_count-- > 0)); spin_unlock(&up->port.lock); tty_flip_buffer_push(tty); spin_lock(&up->port.lock); @@ -1425,7 +1436,7 @@ serial8250_handle_port(struct uart_8250_port *up) DEBUG_INTR("status = %x...", status); - if (status & UART_LSR_DR) + if (status & (UART_LSR_DR | UART_LSR_BI)) receive_chars(up, &status); check_modem_status(up); if (status & UART_LSR_THRE) From b76c5a0717094f0a900d9afd8e36f7ad8dbba587 Mon Sep 17 00:00:00 2001 From: "Catalin(ux) M BOIE" Date: Wed, 23 Jul 2008 21:29:46 -0700 Subject: [PATCH 300/857] serial: add support for a no-name 4 ports multiserial card It is a no-name PCI card. I found no reference to a producer so I used "UNKNOWN_0x1584" as the name. Full lspci: 01:07.0 0780: 10b5:9050 (rev 01) Subsystem: 10b5:1584 Control: I/O+ Mem+ BusMaster- SpecCycle- MemWINV- VGASnoop- \ ParErr- Stepping- SERR+ FastB2B- Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- \ DEVSEL=medium >TAbort- SERR- Acked-by: Alan Cox Acked-by: Russell King Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/serial/8250_pci.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/drivers/serial/8250_pci.c b/drivers/serial/8250_pci.c index 1b36087665a2..c2f23933155b 100644 --- a/drivers/serial/8250_pci.c +++ b/drivers/serial/8250_pci.c @@ -767,6 +767,9 @@ pci_default_setup(struct serial_private *priv, struct pciserial_board *board, #define PCI_SUBDEVICE_ID_POCTAL232 0x0308 #define PCI_SUBDEVICE_ID_POCTAL422 0x0408 +/* Unknown vendors/cards - this should not be in linux/pci_ids.h */ +#define PCI_SUBDEVICE_ID_UNKNOWN_0x1584 0x1584 + /* * Master list of serial port init/setup/exit quirks. * This does not describe the general nature of the port. @@ -880,6 +883,15 @@ static struct pci_serial_quirk pci_serial_quirks[] __refdata = { .setup = pci_default_setup, .exit = __devexit_p(pci_plx9050_exit), }, + { + .vendor = PCI_VENDOR_ID_PLX, + .device = PCI_DEVICE_ID_PLX_9050, + .subvendor = PCI_VENDOR_ID_PLX, + .subdevice = PCI_SUBDEVICE_ID_UNKNOWN_0x1584, + .init = pci_plx9050_init, + .setup = pci_default_setup, + .exit = __devexit_p(pci_plx9050_exit), + }, { .vendor = PCI_VENDOR_ID_PLX, .device = PCI_DEVICE_ID_PLX_ROMULUS, @@ -2197,6 +2209,11 @@ static struct pci_device_id serial_pci_tbl[] = { { PCI_VENDOR_ID_PLX, PCI_DEVICE_ID_PLX_1077, PCI_ANY_ID, PCI_ANY_ID, 0, 0, pbn_b2_4_921600 }, + /* Unknown card - subdevice 0x1584 */ + { PCI_VENDOR_ID_PLX, PCI_DEVICE_ID_PLX_9050, + PCI_VENDOR_ID_PLX, + PCI_SUBDEVICE_ID_UNKNOWN_0x1584, 0, 0, + pbn_b0_4_115200 }, { PCI_VENDOR_ID_PLX, PCI_DEVICE_ID_PLX_9050, PCI_SUBVENDOR_ID_KEYSPAN, PCI_SUBDEVICE_ID_KEYSPAN_SX2, 0, 0, From 377135912806ddc87d56d64fafa685f4063c45f1 Mon Sep 17 00:00:00 2001 From: "Maciej W. Rozycki" Date: Wed, 23 Jul 2008 21:29:48 -0700 Subject: [PATCH 301/857] serial: Z85C30: avoid a hang at console switch-over Changes to the generic console support code that happened a while ago introduced a scenario where the initial console is used in parallel with the final console during a brief period when switching between the two is in progress. During that time a message about the switch-over is printed. With some combinations of chips, firmware and drivers, such as the Zilog Z85C30 SCC used with the DECstation, a hang may happen because the firmware used for the initial console may not expect the state of the chip after it has been initialised by the driver. This is not a bug in the firmware, as some registers it would have to examine are write-only. This is a workaround for the Z85C30 which reuses the power-management callback to keep the transmitter of the line associated with the console enabled. It reflects the consensus reached in a discussion a while ago. Signed-off-by: Maciej W. Rozycki Cc: Jiri Slaby Cc: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/serial/zs.c | 21 +++++++++++++++++++-- 1 file changed, 19 insertions(+), 2 deletions(-) diff --git a/drivers/serial/zs.c b/drivers/serial/zs.c index bd45b6230fd8..9e6a873f8203 100644 --- a/drivers/serial/zs.c +++ b/drivers/serial/zs.c @@ -787,7 +787,6 @@ static int zs_startup(struct uart_port *uport) zport->regs[1] &= ~RxINT_MASK; zport->regs[1] |= RxINT_ALL | TxINT_ENAB | EXT_INT_ENAB; zport->regs[3] |= RxENABLE; - zport->regs[5] |= TxENAB; zport->regs[15] |= BRKIE; write_zsreg(zport, R1, zport->regs[1]); write_zsreg(zport, R3, zport->regs[3]); @@ -814,7 +813,6 @@ static void zs_shutdown(struct uart_port *uport) spin_lock_irqsave(&scc->zlock, flags); - zport->regs[5] &= ~TxENAB; zport->regs[3] &= ~RxENABLE; write_zsreg(zport, R5, zport->regs[5]); write_zsreg(zport, R3, zport->regs[3]); @@ -959,6 +957,23 @@ static void zs_set_termios(struct uart_port *uport, struct ktermios *termios, spin_unlock_irqrestore(&scc->zlock, flags); } +/* + * Hack alert! + * Required solely so that the initial PROM-based console + * works undisturbed in parallel with this one. + */ +static void zs_pm(struct uart_port *uport, unsigned int state, + unsigned int oldstate) +{ + struct zs_port *zport = to_zport(uport); + + if (state < 3) + zport->regs[5] |= TxENAB; + else + zport->regs[5] &= ~TxENAB; + write_zsreg(zport, R5, zport->regs[5]); +} + static const char *zs_type(struct uart_port *uport) { @@ -1041,6 +1056,7 @@ static struct uart_ops zs_ops = { .startup = zs_startup, .shutdown = zs_shutdown, .set_termios = zs_set_termios, + .pm = zs_pm, .type = zs_type, .release_port = zs_release_port, .request_port = zs_request_port, @@ -1190,6 +1206,7 @@ static int __init zs_console_setup(struct console *co, char *options) return ret; zs_reset(zport); + zs_pm(uport, 0, -1); if (options) uart_parse_options(options, &baud, &parity, &bits, &flow); From e9a8f4d1de12633bfb71b5fee47745b32877b7b5 Mon Sep 17 00:00:00 2001 From: "Maciej W. Rozycki" Date: Wed, 23 Jul 2008 21:29:49 -0700 Subject: [PATCH 302/857] serial: DZ11: avoid a hang at console switch-over Changes to the generic console support code that happened a while ago introduced a scenario where the initial console is used in parallel with the final console during a brief period when switching between the two is in progress. During that time a message about the switch-over is printed. With some combinations of chips, firmware and drivers, such as the DEC DZ11 clone used with the DECstation, a hang may happen because the firmware used for the initial console may not expect the state of the chip after it has been initialised by the driver. This is a workaround for the DZ11 which reuses the power-management callback to keep the transmitter of the line associated with the console enabled. It reflects the consensus reached in a discussion a while ago. Signed-off-by: Maciej W. Rozycki Cc: Jiri Slaby Cc: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/serial/dz.c | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+) diff --git a/drivers/serial/dz.c b/drivers/serial/dz.c index a81d2c2ff8a2..6042b87797a1 100644 --- a/drivers/serial/dz.c +++ b/drivers/serial/dz.c @@ -642,6 +642,26 @@ static void dz_set_termios(struct uart_port *uport, struct ktermios *termios, spin_unlock_irqrestore(&dport->port.lock, flags); } +/* + * Hack alert! + * Required solely so that the initial PROM-based console + * works undisturbed in parallel with this one. + */ +static void dz_pm(struct uart_port *uport, unsigned int state, + unsigned int oldstate) +{ + struct dz_port *dport = to_dport(uport); + unsigned long flags; + + spin_lock_irqsave(&dport->port.lock, flags); + if (state < 3) + dz_start_tx(&dport->port); + else + dz_stop_tx(&dport->port); + spin_unlock_irqrestore(&dport->port.lock, flags); +} + + static const char *dz_type(struct uart_port *uport) { return "DZ"; @@ -738,6 +758,7 @@ static struct uart_ops dz_ops = { .startup = dz_startup, .shutdown = dz_shutdown, .set_termios = dz_set_termios, + .pm = dz_pm, .type = dz_type, .release_port = dz_release_port, .request_port = dz_request_port, @@ -861,7 +882,10 @@ static int __init dz_console_setup(struct console *co, char *options) if (ret) return ret; + spin_lock_init(&dport->port.lock); /* For dz_pm(). */ + dz_reset(dport); + dz_pm(uport, 0, -1); if (options) uart_parse_options(options, &baud, &parity, &bits, &flow); From ae2d4c396e19f45918ed6e0900b031538d009823 Mon Sep 17 00:00:00 2001 From: Nye Liu Date: Wed, 23 Jul 2008 21:29:50 -0700 Subject: [PATCH 303/857] cpm1: don't send break on TX_STOP, don't interrupt RX/TX when adjusting termios parameters Before setting STOP_TX, set _brkcr to 0 so the SMC does not send a break character. The driver appears to properly re-initialize _brkcr when the SMC is restarted. Do not interrupt RX/TX when the termios is being adjusted; it results in corrupted characters appearing on the line. Cc: Vitaly Bordug Cc: Scott Wood Cc: Paul Mackerras Cc: Kumar Gala Cc: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/serial/cpm_uart/cpm_uart_core.c | 21 ++++++++++++++------- 1 file changed, 14 insertions(+), 7 deletions(-) diff --git a/drivers/serial/cpm_uart/cpm_uart_core.c b/drivers/serial/cpm_uart/cpm_uart_core.c index 1ff80de177db..a4f86927a74b 100644 --- a/drivers/serial/cpm_uart/cpm_uart_core.c +++ b/drivers/serial/cpm_uart/cpm_uart_core.c @@ -435,10 +435,13 @@ static void cpm_uart_shutdown(struct uart_port *port) } /* Shut them really down and reinit buffer descriptors */ - if (IS_SMC(pinfo)) + if (IS_SMC(pinfo)) { + out_be16(&pinfo->smcup->smc_brkcr, 0); cpm_line_cr_cmd(pinfo, CPM_CR_STOP_TX); - else + } else { + out_be16(&pinfo->sccup->scc_brkcr, 0); cpm_line_cr_cmd(pinfo, CPM_CR_GRA_STOP_TX); + } cpm_uart_initbd(pinfo); } @@ -554,9 +557,11 @@ static void cpm_uart_set_termios(struct uart_port *port, * enables, because we want to put them back if they were * present. */ - prev_mode = in_be16(&smcp->smc_smcmr); - out_be16(&smcp->smc_smcmr, smcr_mk_clen(bits) | cval | SMCMR_SM_UART); - setbits16(&smcp->smc_smcmr, (prev_mode & (SMCMR_REN | SMCMR_TEN))); + prev_mode = in_be16(&smcp->smc_smcmr) & (SMCMR_REN | SMCMR_TEN); + /* Output in *one* operation, so we don't interrupt RX/TX if they + * were already enabled. */ + out_be16(&smcp->smc_smcmr, smcr_mk_clen(bits) | cval | + SMCMR_SM_UART | prev_mode); } else { out_be16(&sccp->scc_psmr, (sbits << 12) | scval); } @@ -1198,12 +1203,14 @@ static int __init cpm_uart_console_setup(struct console *co, char *options) udbg_putc = NULL; #endif - cpm_line_cr_cmd(pinfo, CPM_CR_STOP_TX); - if (IS_SMC(pinfo)) { + out_be16(&pinfo->smcup->smc_brkcr, 0); + cpm_line_cr_cmd(pinfo, CPM_CR_STOP_TX); clrbits8(&pinfo->smcp->smc_smcm, SMCM_RX | SMCM_TX); clrbits16(&pinfo->smcp->smc_smcmr, SMCMR_REN | SMCMR_TEN); } else { + out_be16(&pinfo->sccup->scc_brkcr, 0); + cpm_line_cr_cmd(pinfo, CPM_CR_GRA_STOP_TX); clrbits16(&pinfo->sccp->scc_sccm, UART_SCCM_TX | UART_SCCM_RX); clrbits32(&pinfo->sccp->scc_gsmrl, SCC_GSMRL_ENR | SCC_GSMRL_ENT); } From 708d8cefd0f6d8dc13027f899e865ccfa5f63871 Mon Sep 17 00:00:00 2001 From: Andre Haupt Date: Wed, 23 Jul 2008 21:29:51 -0700 Subject: [PATCH 304/857] stallion: removed unused variable Signed-off-by: Andre Haupt Acked-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/stallion.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/char/stallion.c b/drivers/char/stallion.c index b976248e1072..19db1eb87c26 100644 --- a/drivers/char/stallion.c +++ b/drivers/char/stallion.c @@ -1256,7 +1256,6 @@ static int stl_tiocmset(struct tty_struct *tty, struct file *file, static int stl_ioctl(struct tty_struct *tty, struct file *file, unsigned int cmd, unsigned long arg) { struct stlport *portp; - unsigned int ival; int rc; void __user *argp = (void __user *)arg; From a61f5345eba34772a71523227de890a28410f320 Mon Sep 17 00:00:00 2001 From: Chen Gong Date: Wed, 23 Jul 2008 21:29:52 -0700 Subject: [PATCH 305/857] spi: spi_mpc83xx clockrate fixes This updates the SPI clock rate calculations for the spi_mpc83xx driver. Some boundary conditions were wrong, and in several cases divide-by-16 wasn't always needed Signed-off-by: Chen Gong Signed-off-by: David Brownell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/spi/spi_mpc83xx.c | 29 ++++++++++++++++------------- 1 file changed, 16 insertions(+), 13 deletions(-) diff --git a/drivers/spi/spi_mpc83xx.c b/drivers/spi/spi_mpc83xx.c index 6832da6f7109..070c6219e2d6 100644 --- a/drivers/spi/spi_mpc83xx.c +++ b/drivers/spi/spi_mpc83xx.c @@ -266,21 +266,24 @@ int mpc83xx_spi_setup_transfer(struct spi_device *spi, struct spi_transfer *t) cs->hw_mode |= SPMODE_LEN(bits_per_word); - if ((mpc83xx_spi->spibrg / hz) >= 64) { - pm = mpc83xx_spi->spibrg / (hz * 64) - 1; - if (pm > 0x0f) { - dev_err(&spi->dev, "Requested speed is too " - "low: %d Hz. Will use %d Hz instead.\n", - hz, mpc83xx_spi->spibrg / 1024); - pm = 0x0f; + if ((mpc83xx_spi->spibrg / hz) > 64) { + pm = mpc83xx_spi->spibrg / (hz * 64); + if (pm > 16) { + cs->hw_mode |= SPMODE_DIV16; + pm /= 16; + if (pm > 16) { + dev_err(&spi->dev, "Requested speed is too " + "low: %d Hz. Will use %d Hz instead.\n", + hz, mpc83xx_spi->spibrg / 1024); + pm = 16; + } } - cs->hw_mode |= SPMODE_PM(pm) | SPMODE_DIV16; - } else { + } else pm = mpc83xx_spi->spibrg / (hz * 4); - if (pm) - pm--; - cs->hw_mode |= SPMODE_PM(pm); - } + if (pm) + pm--; + + cs->hw_mode |= SPMODE_PM(pm); regval = mpc83xx_spi_read_reg(&mpc83xx_spi->base->mode); if (cs->hw_mode != regval) { unsigned long flags; From 166a375b657b7af494f4ce3f72c4d2002180da44 Mon Sep 17 00:00:00 2001 From: Roel Kluin <12o3l@tiscali.nl> Date: Wed, 23 Jul 2008 21:29:53 -0700 Subject: [PATCH 306/857] xilinx_spi: test below 0 on unsigned irq in xilinx_spi_probe() xilinx_spi->irq is unsigned, so the test fails Signed-off-by: Roel Kluin <12o3l@tiscali.nl> Cc: David Brownell Cc: Andrei Konovalov Cc: Yuri Frolov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/spi/xilinx_spi.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/spi/xilinx_spi.c b/drivers/spi/xilinx_spi.c index 113a0468ffcb..68d6f4988fb5 100644 --- a/drivers/spi/xilinx_spi.c +++ b/drivers/spi/xilinx_spi.c @@ -353,11 +353,12 @@ static int __init xilinx_spi_probe(struct platform_device *dev) goto put_master; } - xspi->irq = platform_get_irq(dev, 0); - if (xspi->irq < 0) { + ret = platform_get_irq(dev, 0); + if (ret < 0) { ret = -ENXIO; goto unmap_io; } + xspi->irq = ret; master->bus_num = pdata->bus_num; master->num_chipselect = pdata->num_chipselect; From 6291fe2abce4689d6ee7cbaea16692c79bf0d01b Mon Sep 17 00:00:00 2001 From: "Robert P. J. Day" Date: Wed, 23 Jul 2008 21:29:53 -0700 Subject: [PATCH 307/857] SPI Kconfig simplifications Use "if SPI_MASTER" to remove numerous dependencies. [dbrownell@users.sourceforge.net: remove a couple now-needless EXPERIMENTAL dependencies too] Signed-off-by: Robert P. J. Day Signed-off-by: David Brownell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/spi/Kconfig | 45 +++++++++++++++++++++++---------------------- 1 file changed, 23 insertions(+), 22 deletions(-) diff --git a/drivers/spi/Kconfig b/drivers/spi/Kconfig index 66ec5d8808de..2303521b4f09 100644 --- a/drivers/spi/Kconfig +++ b/drivers/spi/Kconfig @@ -49,25 +49,26 @@ config SPI_MASTER controller and the protocol drivers for the SPI slave chips that are connected. +if SPI_MASTER + comment "SPI Master Controller Drivers" - depends on SPI_MASTER config SPI_ATMEL tristate "Atmel SPI Controller" - depends on (ARCH_AT91 || AVR32) && SPI_MASTER + depends on (ARCH_AT91 || AVR32) help This selects a driver for the Atmel SPI Controller, present on many AT32 (AVR32) and AT91 (ARM) chips. config SPI_BFIN tristate "SPI controller driver for ADI Blackfin5xx" - depends on SPI_MASTER && BLACKFIN + depends on BLACKFIN help This is the SPI controller master driver for Blackfin 5xx processor. config SPI_AU1550 tristate "Au1550/Au12x0 SPI Controller" - depends on SPI_MASTER && (SOC_AU1550 || SOC_AU1200) && EXPERIMENTAL + depends on (SOC_AU1550 || SOC_AU1200) && EXPERIMENTAL select SPI_BITBANG help If you say yes to this option, support will be included for the @@ -78,7 +79,6 @@ config SPI_AU1550 config SPI_BITBANG tristate "Bitbanging SPI master" - depends on SPI_MASTER && EXPERIMENTAL help With a few GPIO pins, your system can bitbang the SPI protocol. Select this to get SPI support through I/O pins (GPIO, parallel @@ -92,7 +92,7 @@ config SPI_BITBANG config SPI_BUTTERFLY tristate "Parallel port adapter for AVR Butterfly (DEVELOPMENT)" - depends on SPI_MASTER && PARPORT && EXPERIMENTAL + depends on PARPORT select SPI_BITBANG help This uses a custom parallel port cable to connect to an AVR @@ -102,14 +102,14 @@ config SPI_BUTTERFLY config SPI_IMX tristate "Freescale iMX SPI controller" - depends on SPI_MASTER && ARCH_IMX && EXPERIMENTAL + depends on ARCH_IMX && EXPERIMENTAL help This enables using the Freescale iMX SPI controller in master mode. config SPI_LM70_LLP tristate "Parallel port adapter for LM70 eval board (DEVELOPMENT)" - depends on SPI_MASTER && PARPORT && EXPERIMENTAL + depends on PARPORT && EXPERIMENTAL select SPI_BITBANG help This driver supports the NS LM70 LLP Evaluation Board, @@ -118,14 +118,14 @@ config SPI_LM70_LLP config SPI_MPC52xx_PSC tristate "Freescale MPC52xx PSC SPI controller" - depends on SPI_MASTER && PPC_MPC52xx && EXPERIMENTAL + depends on PPC_MPC52xx && EXPERIMENTAL help This enables using the Freescale MPC52xx Programmable Serial Controller in master SPI mode. config SPI_MPC83xx tristate "Freescale MPC83xx/QUICC Engine SPI controller" - depends on SPI_MASTER && (PPC_83xx || QUICC_ENGINE) && EXPERIMENTAL + depends on (PPC_83xx || QUICC_ENGINE) && EXPERIMENTAL help This enables using the Freescale MPC83xx and QUICC Engine SPI controllers in master mode. @@ -137,21 +137,21 @@ config SPI_MPC83xx config SPI_OMAP_UWIRE tristate "OMAP1 MicroWire" - depends on SPI_MASTER && ARCH_OMAP1 + depends on ARCH_OMAP1 select SPI_BITBANG help This hooks up to the MicroWire controller on OMAP1 chips. config SPI_OMAP24XX tristate "McSPI driver for OMAP24xx/OMAP34xx" - depends on SPI_MASTER && (ARCH_OMAP24XX || ARCH_OMAP34XX) + depends on ARCH_OMAP24XX || ARCH_OMAP34XX help SPI master controller for OMAP24xx/OMAP34xx Multichannel SPI (McSPI) modules. config SPI_PXA2XX tristate "PXA2xx SSP SPI master" - depends on SPI_MASTER && ARCH_PXA && EXPERIMENTAL + depends on ARCH_PXA && EXPERIMENTAL select PXA_SSP help This enables using a PXA2xx SSP port as a SPI master controller. @@ -160,14 +160,14 @@ config SPI_PXA2XX config SPI_S3C24XX tristate "Samsung S3C24XX series SPI" - depends on SPI_MASTER && ARCH_S3C2410 && EXPERIMENTAL + depends on ARCH_S3C2410 && EXPERIMENTAL select SPI_BITBANG help SPI driver for Samsung S3C24XX series ARM SoCs config SPI_S3C24XX_GPIO tristate "Samsung S3C24XX series SPI by GPIO" - depends on SPI_MASTER && ARCH_S3C2410 && EXPERIMENTAL + depends on ARCH_S3C2410 && EXPERIMENTAL select SPI_BITBANG help SPI driver for Samsung S3C24XX series ARM SoCs using @@ -177,20 +177,20 @@ config SPI_S3C24XX_GPIO config SPI_SH_SCI tristate "SuperH SCI SPI controller" - depends on SPI_MASTER && SUPERH + depends on SUPERH select SPI_BITBANG help SPI driver for SuperH SCI blocks. config SPI_TXX9 tristate "Toshiba TXx9 SPI controller" - depends on SPI_MASTER && GENERIC_GPIO && CPU_TX49XX + depends on GENERIC_GPIO && CPU_TX49XX help SPI driver for Toshiba TXx9 MIPS SoCs config SPI_XILINX tristate "Xilinx SPI controller" - depends on SPI_MASTER && XILINX_VIRTEX && EXPERIMENTAL + depends on XILINX_VIRTEX && EXPERIMENTAL select SPI_BITBANG help This exposes the SPI controller IP from the Xilinx EDK. @@ -207,11 +207,10 @@ config SPI_XILINX # being probably the most widely used ones. # comment "SPI Protocol Masters" - depends on SPI_MASTER config SPI_AT25 tristate "SPI EEPROMs from most vendors" - depends on SPI_MASTER && SYSFS + depends on SYSFS help Enable this driver to get read/write support to most SPI EEPROMs, after you configure the board init code to know about each eeprom @@ -222,7 +221,7 @@ config SPI_AT25 config SPI_SPIDEV tristate "User mode SPI device driver support" - depends on SPI_MASTER && EXPERIMENTAL + depends on EXPERIMENTAL help This supports user mode SPI protocol drivers. @@ -231,7 +230,7 @@ config SPI_SPIDEV config SPI_TLE62X0 tristate "Infineon TLE62X0 (for power switching)" - depends on SPI_MASTER && SYSFS + depends on SYSFS help SPI driver for Infineon TLE62X0 series line driver chips, such as the TLE6220, TLE6230 and TLE6240. This provides a @@ -242,6 +241,8 @@ config SPI_TLE62X0 # Add new SPI protocol masters in alphabetical order above this line # +endif # SPI_MASTER + # (slave support would go here) endif # SPI From 102eb97564c73ea73645b38599c5cbe6f54b030c Mon Sep 17 00:00:00 2001 From: Grant Likely Date: Wed, 23 Jul 2008 21:29:55 -0700 Subject: [PATCH 308/857] spi: make spi_board_info.modalias a char array Currently, 'modalias' in the spi_device structure is a 'const char *'. The spi_new_device() function fills in the modalias value from a passed in spi_board_info data block. Since it is a pointer copy, the new spi_device remains dependent on the spi_board_info structure after the new spi_device is registered (no other fields in spi_device directly depend on the spi_board_info structure; all of the other data is copied). This causes a problem when dynamically propulating the list of attached SPI devices. For example, in arch/powerpc, the list of SPI devices can be populated from data in the device tree. With the current code, the device tree adapter must kmalloc() a new spi_board_info structure for each new SPI device it finds in the device tree, and there is no simple mechanism in place for keeping track of these allocations. This patch changes modalias from a 'const char *' to a fixed char array. By copying the modalias string instead of referencing it, the dependency on the spi_board_info structure is eliminated and an outside caller does not need to maintain a separate spi_board_info allocation for each device. If searched through the code to the best of my ability for any references to modalias which may be affected by this change and haven't found anything. It has been tested with the lite5200b platform in arch/powerpc. [dbrownell@users.sourceforge.net: cope with linux-next changes: KOBJ_NAME_LEN obliterated, etc] Signed-off-by: Grant Likely Signed-off-by: David Brownell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/spi/spi.c | 4 +++- include/linux/spi/spi.h | 2 +- 2 files changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c index 1771b2456bfa..ecca4a6a6f94 100644 --- a/drivers/spi/spi.c +++ b/drivers/spi/spi.c @@ -218,6 +218,8 @@ struct spi_device *spi_new_device(struct spi_master *master, if (!spi_master_get(master)) return NULL; + WARN_ON(strlen(chip->modalias) >= sizeof(proxy->modalias)); + proxy = kzalloc(sizeof *proxy, GFP_KERNEL); if (!proxy) { dev_err(dev, "can't alloc dev for cs%d\n", @@ -229,7 +231,7 @@ struct spi_device *spi_new_device(struct spi_master *master, proxy->max_speed_hz = chip->max_speed_hz; proxy->mode = chip->mode; proxy->irq = chip->irq; - proxy->modalias = chip->modalias; + strlcpy(proxy->modalias, chip->modalias, sizeof(proxy->modalias)); snprintf(proxy->dev.bus_id, sizeof proxy->dev.bus_id, "%s.%u", master->dev.bus_id, diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h index b9a76c972084..a9cc29d46653 100644 --- a/include/linux/spi/spi.h +++ b/include/linux/spi/spi.h @@ -82,7 +82,7 @@ struct spi_device { int irq; void *controller_state; void *controller_data; - const char *modalias; + char modalias[32]; /* * likely need more hooks for more protocol options affecting how From 4ef754b7d7971a704d5b1b4608839da1bae37e5e Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Wed, 23 Jul 2008 21:29:55 -0700 Subject: [PATCH 309/857] spidev: BKL removal Another step to removing ->ioctl and to removing the BKL [dbrownell@users.sourceforge.net: take final step; BKL not needed] Signed-off-by: Alan Cox Signed-off-by: David Brownell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/spi/spidev.c | 19 +++++++++++++------ 1 file changed, 13 insertions(+), 6 deletions(-) diff --git a/drivers/spi/spidev.c b/drivers/spi/spidev.c index 2833fd772a24..e5e0cfed5e3b 100644 --- a/drivers/spi/spidev.c +++ b/drivers/spi/spidev.c @@ -228,7 +228,6 @@ static int spidev_message(struct spidev_data *spidev, * We walk the array of user-provided transfers, using each one * to initialize a kernel version of the same transfer. */ - mutex_lock(&spidev->buf_lock); buf = spidev->buffer; total = 0; for (n = n_xfers, k_tmp = k_xfers, u_tmp = u_xfers; @@ -296,14 +295,12 @@ static int spidev_message(struct spidev_data *spidev, status = total; done: - mutex_unlock(&spidev->buf_lock); kfree(k_xfers); return status; } -static int -spidev_ioctl(struct inode *inode, struct file *filp, - unsigned int cmd, unsigned long arg) +static long +spidev_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { int err = 0; int retval = 0; @@ -341,6 +338,14 @@ spidev_ioctl(struct inode *inode, struct file *filp, if (spi == NULL) return -ESHUTDOWN; + /* use the buffer lock here for triple duty: + * - prevent I/O (from us) so calling spi_setup() is safe; + * - prevent concurrent SPI_IOC_WR_* from morphing + * data fields while SPI_IOC_RD_* reads them; + * - SPI_IOC_MESSAGE needs the buffer locked "normally". + */ + mutex_lock(&spidev->buf_lock); + switch (cmd) { /* read requests */ case SPI_IOC_RD_MODE: @@ -456,6 +461,8 @@ spidev_ioctl(struct inode *inode, struct file *filp, kfree(ioc); break; } + + mutex_unlock(&spidev->buf_lock); spi_dev_put(spi); return retval; } @@ -533,7 +540,7 @@ static struct file_operations spidev_fops = { */ .write = spidev_write, .read = spidev_read, - .ioctl = spidev_ioctl, + .unlocked_ioctl = spidev_ioctl, .open = spidev_open, .release = spidev_release, }; From 3a93a159c61e38a12f7ecbb3a25cf3f012abcf7a Mon Sep 17 00:00:00 2001 From: Manuel Lauss Date: Wed, 23 Jul 2008 21:29:56 -0700 Subject: [PATCH 310/857] spi: au1550_spi: proper platform device Remove the Au1550 resource table and instead extract MMIO/IRQ/DMA resources from platform resource information like any well-behaved platform driver. Signed-off-by: Manuel Lauss Signed-off-by: Jan Nikitenko Signed-off-by: David Brownell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/spi/au1550_spi.c | 138 ++++++++++++---------- include/asm-mips/mach-au1x00/au1550_spi.h | 1 - 2 files changed, 74 insertions(+), 65 deletions(-) diff --git a/drivers/spi/au1550_spi.c b/drivers/spi/au1550_spi.c index 072c4a595334..3860dd2fa5d9 100644 --- a/drivers/spi/au1550_spi.c +++ b/drivers/spi/au1550_spi.c @@ -26,6 +26,7 @@ #include #include #include +#include #include #include #include @@ -81,6 +82,7 @@ struct au1550_spi { struct spi_master *master; struct device *dev; struct au1550_spi_info *pdata; + struct resource *ioarea; }; @@ -96,6 +98,8 @@ static dbdev_tab_t au1550_spi_mem_dbdev = .dev_intpolarity = 0 }; +static int ddma_memid; /* id to above mem dma device */ + static void au1550_spi_bits_handlers_set(struct au1550_spi *hw, int bpw); @@ -732,6 +736,7 @@ static int __init au1550_spi_probe(struct platform_device *pdev) { struct au1550_spi *hw; struct spi_master *master; + struct resource *r; int err = 0; master = spi_alloc_master(&pdev->dev, sizeof(struct au1550_spi)); @@ -753,6 +758,52 @@ static int __init au1550_spi_probe(struct platform_device *pdev) goto err_no_pdata; } + r = platform_get_resource(pdev, IORESOURCE_IRQ, 0); + if (!r) { + dev_err(&pdev->dev, "no IRQ\n"); + err = -ENODEV; + goto err_no_iores; + } + hw->irq = r->start; + + hw->usedma = 0; + r = platform_get_resource(pdev, IORESOURCE_DMA, 0); + if (r) { + hw->dma_tx_id = r->start; + r = platform_get_resource(pdev, IORESOURCE_DMA, 1); + if (r) { + hw->dma_rx_id = r->start; + if (usedma && ddma_memid) { + if (pdev->dev.dma_mask == NULL) + dev_warn(&pdev->dev, "no dma mask\n"); + else + hw->usedma = 1; + } + } + } + + r = platform_get_resource(pdev, IORESOURCE_MEM, 0); + if (!r) { + dev_err(&pdev->dev, "no mmio resource\n"); + err = -ENODEV; + goto err_no_iores; + } + + hw->ioarea = request_mem_region(r->start, sizeof(psc_spi_t), + pdev->name); + if (!hw->ioarea) { + dev_err(&pdev->dev, "Cannot reserve iomem region\n"); + err = -ENXIO; + goto err_no_iores; + } + + hw->regs = (psc_spi_t __iomem *)ioremap(r->start, sizeof(psc_spi_t)); + if (!hw->regs) { + dev_err(&pdev->dev, "cannot ioremap\n"); + err = -ENXIO; + goto err_ioremap; + } + platform_set_drvdata(pdev, hw); init_completion(&hw->master_done); @@ -763,66 +814,8 @@ static int __init au1550_spi_probe(struct platform_device *pdev) hw->bitbang.master->setup = au1550_spi_setup; hw->bitbang.txrx_bufs = au1550_spi_txrx_bufs; - switch (hw->pdata->bus_num) { - case 0: - hw->irq = AU1550_PSC0_INT; - hw->regs = (volatile psc_spi_t *)PSC0_BASE_ADDR; - hw->dma_rx_id = DSCR_CMD0_PSC0_RX; - hw->dma_tx_id = DSCR_CMD0_PSC0_TX; - break; - case 1: - hw->irq = AU1550_PSC1_INT; - hw->regs = (volatile psc_spi_t *)PSC1_BASE_ADDR; - hw->dma_rx_id = DSCR_CMD0_PSC1_RX; - hw->dma_tx_id = DSCR_CMD0_PSC1_TX; - break; - case 2: - hw->irq = AU1550_PSC2_INT; - hw->regs = (volatile psc_spi_t *)PSC2_BASE_ADDR; - hw->dma_rx_id = DSCR_CMD0_PSC2_RX; - hw->dma_tx_id = DSCR_CMD0_PSC2_TX; - break; - case 3: - hw->irq = AU1550_PSC3_INT; - hw->regs = (volatile psc_spi_t *)PSC3_BASE_ADDR; - hw->dma_rx_id = DSCR_CMD0_PSC3_RX; - hw->dma_tx_id = DSCR_CMD0_PSC3_TX; - break; - default: - dev_err(&pdev->dev, "Wrong bus_num of SPI\n"); - err = -ENOENT; - goto err_no_pdata; - } - - if (request_mem_region((unsigned long)hw->regs, sizeof(psc_spi_t), - pdev->name) == NULL) { - dev_err(&pdev->dev, "Cannot reserve iomem region\n"); - err = -ENXIO; - goto err_no_iores; - } - - - if (usedma) { - if (pdev->dev.dma_mask == NULL) - dev_warn(&pdev->dev, "no dma mask\n"); - else - hw->usedma = 1; - } - if (hw->usedma) { - /* - * create memory device with 8 bits dev_devwidth - * needed for proper byte ordering to spi fifo - */ - int memid = au1xxx_ddma_add_device(&au1550_spi_mem_dbdev); - if (!memid) { - dev_err(&pdev->dev, - "Cannot create dma 8 bit mem device\n"); - err = -ENXIO; - goto err_dma_add_dev; - } - - hw->dma_tx_ch = au1xxx_dbdma_chan_alloc(memid, + hw->dma_tx_ch = au1xxx_dbdma_chan_alloc(ddma_memid, hw->dma_tx_id, NULL, (void *)hw); if (hw->dma_tx_ch == 0) { dev_err(&pdev->dev, @@ -841,7 +834,7 @@ static int __init au1550_spi_probe(struct platform_device *pdev) hw->dma_rx_ch = au1xxx_dbdma_chan_alloc(hw->dma_rx_id, - memid, NULL, (void *)hw); + ddma_memid, NULL, (void *)hw); if (hw->dma_rx_ch == 0) { dev_err(&pdev->dev, "Cannot allocate rx dma channel\n"); @@ -874,7 +867,7 @@ static int __init au1550_spi_probe(struct platform_device *pdev) goto err_no_irq; } - master->bus_num = hw->pdata->bus_num; + master->bus_num = pdev->id; master->num_chipselect = hw->pdata->num_chipselect; /* @@ -924,8 +917,11 @@ static int __init au1550_spi_probe(struct platform_device *pdev) au1xxx_dbdma_chan_free(hw->dma_tx_ch); err_no_txdma: -err_dma_add_dev: - release_mem_region((unsigned long)hw->regs, sizeof(psc_spi_t)); + iounmap((void __iomem *)hw->regs); + +err_ioremap: + release_resource(hw->ioarea); + kfree(hw->ioarea); err_no_iores: err_no_pdata: @@ -944,7 +940,9 @@ static int __exit au1550_spi_remove(struct platform_device *pdev) spi_bitbang_stop(&hw->bitbang); free_irq(hw->irq, hw); - release_mem_region((unsigned long)hw->regs, sizeof(psc_spi_t)); + iounmap((void __iomem *)hw->regs); + release_resource(hw->ioarea); + kfree(hw->ioarea); if (hw->usedma) { au1550_spi_dma_rxtmp_free(hw); @@ -971,12 +969,24 @@ static struct platform_driver au1550_spi_drv = { static int __init au1550_spi_init(void) { + /* + * create memory device with 8 bits dev_devwidth + * needed for proper byte ordering to spi fifo + */ + if (usedma) { + ddma_memid = au1xxx_ddma_add_device(&au1550_spi_mem_dbdev); + if (!ddma_memid) + printk(KERN_ERR "au1550-spi: cannot add memory" + "dbdma device\n"); + } return platform_driver_probe(&au1550_spi_drv, au1550_spi_probe); } module_init(au1550_spi_init); static void __exit au1550_spi_exit(void) { + if (usedma && ddma_memid) + au1xxx_ddma_del_device(ddma_memid); platform_driver_unregister(&au1550_spi_drv); } module_exit(au1550_spi_exit); diff --git a/include/asm-mips/mach-au1x00/au1550_spi.h b/include/asm-mips/mach-au1x00/au1550_spi.h index 40e6c489833a..08e1958e9410 100644 --- a/include/asm-mips/mach-au1x00/au1550_spi.h +++ b/include/asm-mips/mach-au1x00/au1550_spi.h @@ -6,7 +6,6 @@ #define _AU1550_SPI_H_ struct au1550_spi_info { - s16 bus_num; /* defines which PSC and IRQ to use */ u32 mainclk_hz; /* main input clock frequency of PSC */ u16 num_chipselect; /* number of chipselects supported */ void (*activate_cs)(struct au1550_spi_info *spi, int cs, int polarity); From bbe48ecc7f6559318cfc6c023da225a0b0e14ab3 Mon Sep 17 00:00:00 2001 From: Jan Nikitenko Date: Wed, 23 Jul 2008 21:29:57 -0700 Subject: [PATCH 311/857] spi: au1550_spi: improve pio transfer mode Improve PIO transfer mode of au1550 spi controller by continuing of spi transfer, instead of aborting transfer when transmit underflow interrupt occurrs. Verified by oscilloscope that the spi clock pauses on trasmit underflow, so transfer continuation is perfectly valid even though au1550 datasheet says that on tx underflow zeroes will be transfered. Also make some error messages more specific. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by: Jan Nikitenko Signed-off-by: David Brownell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/spi/au1550_spi.c | 69 ++++++++++++++++++++++++++++------------ 1 file changed, 49 insertions(+), 20 deletions(-) diff --git a/drivers/spi/au1550_spi.c b/drivers/spi/au1550_spi.c index 3860dd2fa5d9..9149689c79d9 100644 --- a/drivers/spi/au1550_spi.c +++ b/drivers/spi/au1550_spi.c @@ -484,9 +484,13 @@ static irqreturn_t au1550_spi_dma_irq_callback(struct au1550_spi *hw) au1xxx_dbdma_reset(hw->dma_tx_ch); au1550_spi_reset_fifos(hw); - dev_err(hw->dev, - "Unexpected SPI error: event=0x%x stat=0x%x!\n", - evnt, stat); + if (evnt == PSC_SPIEVNT_RO) + dev_err(hw->dev, + "dma transfer: receive FIFO overflow!\n"); + else + dev_err(hw->dev, + "dma transfer: unexpected SPI error " + "(event=0x%x stat=0x%x)!\n", evnt, stat); complete(&hw->master_done); return IRQ_HANDLED; @@ -596,17 +600,17 @@ static irqreturn_t au1550_spi_pio_irq_callback(struct au1550_spi *hw) if ((evnt & (PSC_SPIEVNT_MM | PSC_SPIEVNT_RO | PSC_SPIEVNT_RU | PSC_SPIEVNT_TO - | PSC_SPIEVNT_TU | PSC_SPIEVNT_SD)) + | PSC_SPIEVNT_SD)) != 0) { - dev_err(hw->dev, - "Unexpected SPI error: event=0x%x stat=0x%x!\n", - evnt, stat); /* * due to an error we consider transfer as done, * so mask all events until before next transfer start */ au1550_spi_mask_ack_all(hw); au1550_spi_reset_fifos(hw); + dev_err(hw->dev, + "pio transfer: unexpected SPI error " + "(event=0x%x stat=0x%x)!\n", evnt, stat); complete(&hw->master_done); return IRQ_HANDLED; } @@ -620,27 +624,50 @@ static irqreturn_t au1550_spi_pio_irq_callback(struct au1550_spi *hw) stat = hw->regs->psc_spistat; au_sync(); - if ((stat & PSC_SPISTAT_RE) == 0 && hw->rx_count < hw->len) { + /* + * Take care to not let the Rx FIFO overflow. + * + * We only write a byte if we have read one at least. Initially, + * the write fifo is full, so we should read from the read fifo + * first. + * In case we miss a word from the read fifo, we should get a + * RO event and should back out. + */ + if (!(stat & PSC_SPISTAT_RE) && hw->rx_count < hw->len) { hw->rx_word(hw); - /* ack the receive request event */ - hw->regs->psc_spievent = PSC_SPIEVNT_RR; - au_sync(); busy = 1; - } - if ((stat & PSC_SPISTAT_TF) == 0 && hw->tx_count < hw->len) { - hw->tx_word(hw); - /* ack the transmit request event */ - hw->regs->psc_spievent = PSC_SPIEVNT_TR; - au_sync(); - busy = 1; + if (!(stat & PSC_SPISTAT_TF) && hw->tx_count < hw->len) + hw->tx_word(hw); } } while (busy); - evnt = hw->regs->psc_spievent; + hw->regs->psc_spievent = PSC_SPIEVNT_RR | PSC_SPIEVNT_TR; au_sync(); - if (hw->rx_count >= hw->len || (evnt & PSC_SPIEVNT_MD) != 0) { + /* + * Restart the SPI transmission in case of a transmit underflow. + * This seems to work despite the notes in the Au1550 data book + * of Figure 8-4 with flowchart for SPI master operation: + * + * """Note 1: An XFR Error Interrupt occurs, unless masked, + * for any of the following events: Tx FIFO Underflow, + * Rx FIFO Overflow, or Multiple-master Error + * Note 2: In case of a Tx Underflow Error, all zeroes are + * transmitted.""" + * + * By simply restarting the spi transfer on Tx Underflow Error, + * we assume that spi transfer was paused instead of zeroes + * transmittion mentioned in the Note 2 of Au1550 data book. + */ + if (evnt & PSC_SPIEVNT_TU) { + hw->regs->psc_spievent = PSC_SPIEVNT_TU | PSC_SPIEVNT_MD; + au_sync(); + hw->regs->psc_spipcr = PSC_SPIPCR_MS; + au_sync(); + } + + if (hw->rx_count >= hw->len) { /* transfer completed successfully */ au1550_spi_mask_ack_all(hw); complete(&hw->master_done); @@ -729,6 +756,8 @@ static void __init au1550_spi_setup_psc_as_spi(struct au1550_spi *hw) stat = hw->regs->psc_spistat; au_sync(); } while ((stat & PSC_SPISTAT_DR) == 0); + + au1550_spi_reset_fifos(hw); } From e0426e6a09954d205da2d674a3d368d2715e3afd Mon Sep 17 00:00:00 2001 From: Jiri Slaby Date: Wed, 23 Jul 2008 21:29:58 -0700 Subject: [PATCH 312/857] vt: hold console_sem across sysfs operations Hold console sem while creating/destroying sysfs files. Serialisation is so far done by BKL held in tty release_dev and chrdev_open, but no other locks are held in open path. Signed-off-by: Jiri Slaby Cc: Alan Cox Cc: Aristeu Rozanski Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/vt.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/char/vt.c b/drivers/char/vt.c index e32a076d5f1f..ab53a1d4d895 100644 --- a/drivers/char/vt.c +++ b/drivers/char/vt.c @@ -2749,8 +2749,8 @@ static int con_open(struct tty_struct *tty, struct file *filp) tty->termios->c_iflag |= IUTF8; else tty->termios->c_iflag &= ~IUTF8; - release_console_sem(); vcs_make_sysfs(tty); + release_console_sem(); return ret; } } @@ -2775,8 +2775,8 @@ static void con_close(struct tty_struct *tty, struct file *filp) if (vc) vc->vc_tty = NULL; tty->driver_data = NULL; - release_console_sem(); vcs_remove_sysfs(tty); + release_console_sem(); mutex_unlock(&tty_mutex); /* * tty_mutex is released, but we still hold BKL, so there is From f700d6e5e5549cb9349d22043f4bd153792c621f Mon Sep 17 00:00:00 2001 From: Stefano Stabellini Date: Wed, 23 Jul 2008 21:29:59 -0700 Subject: [PATCH 313/857] vt: do not update when the console is blanked vt.c DO_UPDATE macro checks if the console is visible but doesn't check if the console is blanked. In fact updating fbcon while the console is blanked is not only unnecessary but can even cause screen corruption. Therefore I am adding a simple check on console_blanked in DO_UPDATE. Signed-off-by: Stefano Stabellini Cc: Krzysztof Helt Cc: "Antonino A. Daplas" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/vt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/char/vt.c b/drivers/char/vt.c index ab53a1d4d895..cb8c90da3934 100644 --- a/drivers/char/vt.c +++ b/drivers/char/vt.c @@ -261,7 +261,7 @@ static void notify_update(struct vc_data *vc) #ifdef VT_BUF_VRAM_ONLY #define DO_UPDATE(vc) 0 #else -#define DO_UPDATE(vc) CON_IS_VISIBLE(vc) +#define DO_UPDATE(vc) (CON_IS_VISIBLE(vc) && !console_blanked) #endif static inline unsigned short *screenpos(struct vc_data *vc, int offset, int viewed) From 0293902a4d66fab27d0ddcc0766e05dae68f004e Mon Sep 17 00:00:00 2001 From: Wang Chen Date: Wed, 23 Jul 2008 21:30:01 -0700 Subject: [PATCH 314/857] I2O: handle sysfs_create_link() failures Compile warning: ignoring return value of `sysfs_create_link', declared with attribute warn_unused_result. If sysfs_create_link failed, take care of the return value and do some error handle after the failure. Since sysfs_remove_link() will check whether a link exists, when removing the link in error path, we don't need to care whether a link was created. Signed-off-by: Wang Chen Cc: Markus Lidel Cc: Jens Axboe Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/message/i2o/device.c | 54 ++++++++++++++++++++++++++++-------- 1 file changed, 42 insertions(+), 12 deletions(-) diff --git a/drivers/message/i2o/device.c b/drivers/message/i2o/device.c index 489d7c5c4965..8774c670e668 100644 --- a/drivers/message/i2o/device.c +++ b/drivers/message/i2o/device.c @@ -243,29 +243,41 @@ static int i2o_device_add(struct i2o_controller *c, i2o_lct_entry *entry) /* create user entries for this device */ tmp = i2o_iop_find_device(i2o_dev->iop, i2o_dev->lct_data.user_tid); - if (tmp && (tmp != i2o_dev)) - sysfs_create_link(&i2o_dev->device.kobj, &tmp->device.kobj, - "user"); + if (tmp && (tmp != i2o_dev)) { + rc = sysfs_create_link(&i2o_dev->device.kobj, + &tmp->device.kobj, "user"); + if (rc) + goto unreg_dev; + } /* create user entries refering to this device */ list_for_each_entry(tmp, &c->devices, list) if ((tmp->lct_data.user_tid == i2o_dev->lct_data.tid) - && (tmp != i2o_dev)) - sysfs_create_link(&tmp->device.kobj, - &i2o_dev->device.kobj, "user"); + && (tmp != i2o_dev)) { + rc = sysfs_create_link(&tmp->device.kobj, + &i2o_dev->device.kobj, "user"); + if (rc) + goto rmlink1; + } /* create parent entries for this device */ tmp = i2o_iop_find_device(i2o_dev->iop, i2o_dev->lct_data.parent_tid); - if (tmp && (tmp != i2o_dev)) - sysfs_create_link(&i2o_dev->device.kobj, &tmp->device.kobj, - "parent"); + if (tmp && (tmp != i2o_dev)) { + rc = sysfs_create_link(&i2o_dev->device.kobj, + &tmp->device.kobj, "parent"); + if (rc) + goto rmlink1; + } /* create parent entries refering to this device */ list_for_each_entry(tmp, &c->devices, list) if ((tmp->lct_data.parent_tid == i2o_dev->lct_data.tid) - && (tmp != i2o_dev)) - sysfs_create_link(&tmp->device.kobj, - &i2o_dev->device.kobj, "parent"); + && (tmp != i2o_dev)) { + rc = sysfs_create_link(&tmp->device.kobj, + &i2o_dev->device.kobj, "parent"); + if (rc) + goto rmlink2; + } i2o_driver_notify_device_add_all(i2o_dev); @@ -273,6 +285,24 @@ static int i2o_device_add(struct i2o_controller *c, i2o_lct_entry *entry) return 0; +rmlink2: + /* If link creating failed halfway, we loop whole list to cleanup. + * And we don't care wrong removing of link, because sysfs_remove_link + * will take care of it. + */ + list_for_each_entry(tmp, &c->devices, list) { + if (tmp->lct_data.parent_tid == i2o_dev->lct_data.tid) + sysfs_remove_link(&tmp->device.kobj, "parent"); + } + sysfs_remove_link(&i2o_dev->device.kobj, "parent"); +rmlink1: + list_for_each_entry(tmp, &c->devices, list) + if (tmp->lct_data.user_tid == i2o_dev->lct_data.tid) + sysfs_remove_link(&tmp->device.kobj, "user"); + sysfs_remove_link(&i2o_dev->device.kobj, "user"); +unreg_dev: + list_del(&i2o_dev->list); + device_unregister(&i2o_dev->device); err: kfree(i2o_dev); return rc; From 746f1e558bc52b9693c1a1ecdab60f8392e5ff18 Mon Sep 17 00:00:00 2001 From: Michael Halcrow Date: Wed, 23 Jul 2008 21:30:02 -0700 Subject: [PATCH 315/857] eCryptfs: Privileged kthread for lower file opens eCryptfs would really like to have read-write access to all files in the lower filesystem. Right now, the persistent lower file may be opened read-only if the attempt to open it read-write fails. One way to keep from having to do that is to have a privileged kthread that can open the lower persistent file on behalf of the user opening the eCryptfs file; this patch implements this functionality. This patch will properly allow a less-privileged user to open the eCryptfs file, followed by a more-privileged user opening the eCryptfs file, with the first user only being able to read and the second user being able to both read and write. eCryptfs currently does this wrong; it will wind up calling vfs_write() on a file that was opened read-only. This is fixed in this patch. Signed-off-by: Michael Halcrow Cc: Dave Kleikamp Cc: Serge Hallyn Cc: Eric Sandeen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ecryptfs/Makefile | 2 +- fs/ecryptfs/ecryptfs_kernel.h | 19 ++++ fs/ecryptfs/file.c | 7 ++ fs/ecryptfs/kthread.c | 203 ++++++++++++++++++++++++++++++++++ fs/ecryptfs/main.c | 42 +++---- 5 files changed, 251 insertions(+), 22 deletions(-) create mode 100644 fs/ecryptfs/kthread.c diff --git a/fs/ecryptfs/Makefile b/fs/ecryptfs/Makefile index 1e34a7fd4884..b4755a85996e 100644 --- a/fs/ecryptfs/Makefile +++ b/fs/ecryptfs/Makefile @@ -4,4 +4,4 @@ obj-$(CONFIG_ECRYPT_FS) += ecryptfs.o -ecryptfs-objs := dentry.o file.o inode.o main.o super.o mmap.o read_write.o crypto.o keystore.o messaging.o netlink.o miscdev.o debug.o +ecryptfs-objs := dentry.o file.o inode.o main.o super.o mmap.o read_write.o crypto.o keystore.o messaging.o netlink.o miscdev.o kthread.o debug.o diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h index c15c25745e05..b4a0cccfdd7c 100644 --- a/fs/ecryptfs/ecryptfs_kernel.h +++ b/fs/ecryptfs/ecryptfs_kernel.h @@ -559,6 +559,20 @@ extern struct kmem_cache *ecryptfs_key_record_cache; extern struct kmem_cache *ecryptfs_key_sig_cache; extern struct kmem_cache *ecryptfs_global_auth_tok_cache; extern struct kmem_cache *ecryptfs_key_tfm_cache; +extern struct kmem_cache *ecryptfs_open_req_cache; + +struct ecryptfs_open_req { +#define ECRYPTFS_REQ_PROCESSED 0x00000001 +#define ECRYPTFS_REQ_DROPPED 0x00000002 +#define ECRYPTFS_REQ_ZOMBIE 0x00000004 + u32 flags; + struct file **lower_file; + struct dentry *lower_dentry; + struct vfsmount *lower_mnt; + wait_queue_head_t wait; + struct mutex mux; + struct list_head kthread_ctl_list; +}; int ecryptfs_interpose(struct dentry *hidden_dentry, struct dentry *this_dentry, struct super_block *sb, @@ -690,5 +704,10 @@ void ecryptfs_msg_ctx_alloc_to_free(struct ecryptfs_msg_ctx *msg_ctx); int ecryptfs_spawn_daemon(struct ecryptfs_daemon **daemon, uid_t euid, struct user_namespace *user_ns, struct pid *pid); +int ecryptfs_init_kthread(void); +void ecryptfs_destroy_kthread(void); +int ecryptfs_privileged_open(struct file **lower_file, + struct dentry *lower_dentry, + struct vfsmount *lower_mnt); #endif /* #ifndef ECRYPTFS_KERNEL_H */ diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c index 24749bf0668f..f0be29051528 100644 --- a/fs/ecryptfs/file.c +++ b/fs/ecryptfs/file.c @@ -192,6 +192,13 @@ static int ecryptfs_open(struct inode *inode, struct file *file) | ECRYPTFS_ENCRYPTED); } mutex_unlock(&crypt_stat->cs_mutex); + if ((ecryptfs_inode_to_private(inode)->lower_file->f_flags & O_RDONLY) + && !(file->f_flags & O_RDONLY)) { + rc = -EPERM; + printk(KERN_WARNING "%s: Lower persistent file is RO; eCryptfs " + "file must hence be opened RO\n", __func__); + goto out; + } ecryptfs_set_file_lower( file, ecryptfs_inode_to_private(inode)->lower_file); if (S_ISDIR(ecryptfs_dentry->d_inode->i_mode)) { diff --git a/fs/ecryptfs/kthread.c b/fs/ecryptfs/kthread.c new file mode 100644 index 000000000000..c440c6b58b2d --- /dev/null +++ b/fs/ecryptfs/kthread.c @@ -0,0 +1,203 @@ +/** + * eCryptfs: Linux filesystem encryption layer + * + * Copyright (C) 2008 International Business Machines Corp. + * Author(s): Michael A. Halcrow + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation; either version 2 of the + * License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA + * 02111-1307, USA. + */ + +#include +#include +#include +#include +#include "ecryptfs_kernel.h" + +struct kmem_cache *ecryptfs_open_req_cache; + +static struct ecryptfs_kthread_ctl { +#define ECRYPTFS_KTHREAD_ZOMBIE 0x00000001 + u32 flags; + struct mutex mux; + struct list_head req_list; + wait_queue_head_t wait; +} ecryptfs_kthread_ctl; + +static struct task_struct *ecryptfs_kthread; + +/** + * ecryptfs_threadfn + * @ignored: ignored + * + * The eCryptfs kernel thread that has the responsibility of getting + * the lower persistent file with RW permissions. + * + * Returns zero on success; non-zero otherwise + */ +static int ecryptfs_threadfn(void *ignored) +{ + set_freezable(); + while (1) { + struct ecryptfs_open_req *req; + + wait_event_freezable( + ecryptfs_kthread_ctl.wait, + (!list_empty(&ecryptfs_kthread_ctl.req_list) + || kthread_should_stop())); + mutex_lock(&ecryptfs_kthread_ctl.mux); + if (ecryptfs_kthread_ctl.flags & ECRYPTFS_KTHREAD_ZOMBIE) { + mutex_unlock(&ecryptfs_kthread_ctl.mux); + goto out; + } + while (!list_empty(&ecryptfs_kthread_ctl.req_list)) { + req = list_first_entry(&ecryptfs_kthread_ctl.req_list, + struct ecryptfs_open_req, + kthread_ctl_list); + mutex_lock(&req->mux); + list_del(&req->kthread_ctl_list); + if (!(req->flags & ECRYPTFS_REQ_ZOMBIE)) { + dget(req->lower_dentry); + mntget(req->lower_mnt); + (*req->lower_file) = dentry_open( + req->lower_dentry, req->lower_mnt, + (O_RDWR | O_LARGEFILE)); + req->flags |= ECRYPTFS_REQ_PROCESSED; + } + wake_up(&req->wait); + mutex_unlock(&req->mux); + } + mutex_unlock(&ecryptfs_kthread_ctl.mux); + } +out: + return 0; +} + +int ecryptfs_init_kthread(void) +{ + int rc = 0; + + mutex_init(&ecryptfs_kthread_ctl.mux); + init_waitqueue_head(&ecryptfs_kthread_ctl.wait); + INIT_LIST_HEAD(&ecryptfs_kthread_ctl.req_list); + ecryptfs_kthread = kthread_run(&ecryptfs_threadfn, NULL, + "ecryptfs-kthread"); + if (IS_ERR(ecryptfs_kthread)) { + rc = PTR_ERR(ecryptfs_kthread); + printk(KERN_ERR "%s: Failed to create kernel thread; rc = [%d]" + "\n", __func__, rc); + } + return rc; +} + +void ecryptfs_destroy_kthread(void) +{ + struct ecryptfs_open_req *req; + + mutex_lock(&ecryptfs_kthread_ctl.mux); + ecryptfs_kthread_ctl.flags |= ECRYPTFS_KTHREAD_ZOMBIE; + list_for_each_entry(req, &ecryptfs_kthread_ctl.req_list, + kthread_ctl_list) { + mutex_lock(&req->mux); + req->flags |= ECRYPTFS_REQ_ZOMBIE; + wake_up(&req->wait); + mutex_unlock(&req->mux); + } + mutex_unlock(&ecryptfs_kthread_ctl.mux); + kthread_stop(ecryptfs_kthread); + wake_up(&ecryptfs_kthread_ctl.wait); +} + +/** + * ecryptfs_privileged_open + * @lower_file: Result of dentry_open by root on lower dentry + * @lower_dentry: Lower dentry for file to open + * @lower_mnt: Lower vfsmount for file to open + * + * This function gets a r/w file opened againt the lower dentry. + * + * Returns zero on success; non-zero otherwise + */ +int ecryptfs_privileged_open(struct file **lower_file, + struct dentry *lower_dentry, + struct vfsmount *lower_mnt) +{ + struct ecryptfs_open_req *req; + int rc = 0; + + /* Corresponding dput() and mntput() are done when the + * persistent file is fput() when the eCryptfs inode is + * destroyed. */ + dget(lower_dentry); + mntget(lower_mnt); + (*lower_file) = dentry_open(lower_dentry, lower_mnt, + (O_RDWR | O_LARGEFILE)); + if (!IS_ERR(*lower_file)) + goto out; + req = kmem_cache_alloc(ecryptfs_open_req_cache, GFP_KERNEL); + if (!req) { + rc = -ENOMEM; + goto out; + } + mutex_init(&req->mux); + req->lower_file = lower_file; + req->lower_dentry = lower_dentry; + req->lower_mnt = lower_mnt; + init_waitqueue_head(&req->wait); + req->flags = 0; + mutex_lock(&ecryptfs_kthread_ctl.mux); + if (ecryptfs_kthread_ctl.flags & ECRYPTFS_KTHREAD_ZOMBIE) { + rc = -EIO; + mutex_unlock(&ecryptfs_kthread_ctl.mux); + printk(KERN_ERR "%s: We are in the middle of shutting down; " + "aborting privileged request to open lower file\n", + __func__); + goto out_free; + } + list_add_tail(&req->kthread_ctl_list, &ecryptfs_kthread_ctl.req_list); + mutex_unlock(&ecryptfs_kthread_ctl.mux); + wake_up(&ecryptfs_kthread_ctl.wait); + wait_event(req->wait, (req->flags != 0)); + mutex_lock(&req->mux); + BUG_ON(req->flags == 0); + if (req->flags & ECRYPTFS_REQ_DROPPED + || req->flags & ECRYPTFS_REQ_ZOMBIE) { + rc = -EIO; + printk(KERN_WARNING "%s: Privileged open request dropped\n", + __func__); + goto out_unlock; + } + if (IS_ERR(*req->lower_file)) { + rc = PTR_ERR(*req->lower_file); + dget(lower_dentry); + mntget(lower_mnt); + (*lower_file) = dentry_open(lower_dentry, lower_mnt, + (O_RDONLY | O_LARGEFILE)); + if (IS_ERR(*lower_file)) { + rc = PTR_ERR(*req->lower_file); + (*lower_file) = NULL; + printk(KERN_WARNING "%s: Error attempting privileged " + "open of lower file with either RW or RO " + "perms; rc = [%d]. Giving up.\n", + __func__, rc); + } + } +out_unlock: + mutex_unlock(&req->mux); +out_free: + kmem_cache_free(ecryptfs_open_req_cache, req); +out: + return rc; +} diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index d603631601eb..f36ab2feea28 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -130,26 +130,12 @@ static int ecryptfs_init_persistent_file(struct dentry *ecryptfs_dentry) ecryptfs_dentry_to_lower_mnt(ecryptfs_dentry); lower_dentry = ecryptfs_dentry_to_lower(ecryptfs_dentry); - /* Corresponding dput() and mntput() are done when the - * persistent file is fput() when the eCryptfs inode - * is destroyed. */ - dget(lower_dentry); - mntget(lower_mnt); - inode_info->lower_file = dentry_open(lower_dentry, - lower_mnt, - (O_RDWR | O_LARGEFILE)); - if (IS_ERR(inode_info->lower_file)) { - dget(lower_dentry); - mntget(lower_mnt); - inode_info->lower_file = dentry_open(lower_dentry, - lower_mnt, - (O_RDONLY - | O_LARGEFILE)); - } - if (IS_ERR(inode_info->lower_file)) { + rc = ecryptfs_privileged_open(&inode_info->lower_file, + lower_dentry, lower_mnt); + if (rc || IS_ERR(inode_info->lower_file)) { printk(KERN_ERR "Error opening lower persistent file " - "for lower_dentry [0x%p] and lower_mnt [0x%p]\n", - lower_dentry, lower_mnt); + "for lower_dentry [0x%p] and lower_mnt [0x%p]; " + "rc = [%d]\n", lower_dentry, lower_mnt, rc); rc = PTR_ERR(inode_info->lower_file); inode_info->lower_file = NULL; } @@ -679,6 +665,11 @@ static struct ecryptfs_cache_info { .name = "ecryptfs_key_tfm_cache", .size = sizeof(struct ecryptfs_key_tfm), }, + { + .cache = &ecryptfs_open_req_cache, + .name = "ecryptfs_open_req_cache", + .size = sizeof(struct ecryptfs_open_req), + }, }; static void ecryptfs_free_kmem_caches(void) @@ -795,11 +786,17 @@ static int __init ecryptfs_init(void) printk(KERN_ERR "sysfs registration failed\n"); goto out_unregister_filesystem; } + rc = ecryptfs_init_kthread(); + if (rc) { + printk(KERN_ERR "%s: kthread initialization failed; " + "rc = [%d]\n", __func__, rc); + goto out_do_sysfs_unregistration; + } rc = ecryptfs_init_messaging(ecryptfs_transport); if (rc) { - ecryptfs_printk(KERN_ERR, "Failure occured while attempting to " + printk(KERN_ERR "Failure occured while attempting to " "initialize the eCryptfs netlink socket\n"); - goto out_do_sysfs_unregistration; + goto out_destroy_kthread; } rc = ecryptfs_init_crypto(); if (rc) { @@ -814,6 +811,8 @@ static int __init ecryptfs_init(void) goto out; out_release_messaging: ecryptfs_release_messaging(ecryptfs_transport); +out_destroy_kthread: + ecryptfs_destroy_kthread(); out_do_sysfs_unregistration: do_sysfs_unregistration(); out_unregister_filesystem: @@ -833,6 +832,7 @@ static void __exit ecryptfs_exit(void) printk(KERN_ERR "Failure whilst attempting to destroy crypto; " "rc = [%d]\n", rc); ecryptfs_release_messaging(ecryptfs_transport); + ecryptfs_destroy_kthread(); do_sysfs_unregistration(); unregister_filesystem(&ecryptfs_fs_type); ecryptfs_free_kmem_caches(); From 6c4c17b073cd4a5a61bc04329561632870bb21fc Mon Sep 17 00:00:00 2001 From: Tyler Hicks Date: Wed, 23 Jul 2008 21:30:04 -0700 Subject: [PATCH 316/857] ecryptfs: discard ecryptfsd registration messages in miscdev The userspace eCryptfs daemon sends HELO and QUIT messages to the kernel for per-user daemon (un)registration. These messages are required when netlink is used as the transport, but (un)registration is handled by opening and closing the device file when miscdev is the transport. These messages should be discarded in the miscdev transport so that a daemon isn't registered twice. Signed-off-by: Tyler Hicks Cc: Michael Halcrow Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ecryptfs/miscdev.c | 59 ------------------------------------------- 1 file changed, 59 deletions(-) diff --git a/fs/ecryptfs/miscdev.c b/fs/ecryptfs/miscdev.c index 09a4522f65e6..b484792a0996 100644 --- a/fs/ecryptfs/miscdev.c +++ b/fs/ecryptfs/miscdev.c @@ -357,46 +357,6 @@ ecryptfs_miscdev_read(struct file *file, char __user *buf, size_t count, return rc; } -/** - * ecryptfs_miscdev_helo - * @euid: effective user id of miscdevess sending helo packet - * @user_ns: The namespace in which @euid applies - * @pid: miscdevess id of miscdevess sending helo packet - * - * Returns zero on success; non-zero otherwise - */ -static int ecryptfs_miscdev_helo(uid_t euid, struct user_namespace *user_ns, - struct pid *pid) -{ - int rc; - - rc = ecryptfs_process_helo(ECRYPTFS_TRANSPORT_MISCDEV, euid, user_ns, - pid); - if (rc) - printk(KERN_WARNING "Error processing HELO; rc = [%d]\n", rc); - return rc; -} - -/** - * ecryptfs_miscdev_quit - * @euid: effective user id of miscdevess sending quit packet - * @user_ns: The namespace in which @euid applies - * @pid: miscdevess id of miscdevess sending quit packet - * - * Returns zero on success; non-zero otherwise - */ -static int ecryptfs_miscdev_quit(uid_t euid, struct user_namespace *user_ns, - struct pid *pid) -{ - int rc; - - rc = ecryptfs_process_quit(euid, user_ns, pid); - if (rc) - printk(KERN_WARNING - "Error processing QUIT message; rc = [%d]\n", rc); - return rc; -} - /** * ecryptfs_miscdev_response - miscdevess response to message previously sent to daemon * @data: Bytes comprising struct ecryptfs_message @@ -512,26 +472,7 @@ ecryptfs_miscdev_write(struct file *file, const char __user *buf, __func__, rc); break; case ECRYPTFS_MSG_HELO: - rc = ecryptfs_miscdev_helo(current->euid, - current->nsproxy->user_ns, - task_pid(current)); - if (rc) { - printk(KERN_ERR "%s: Error attempting to process " - "helo from pid [0x%p]; rc = [%d]\n", __func__, - task_pid(current), rc); - goto out_free; - } - break; case ECRYPTFS_MSG_QUIT: - rc = ecryptfs_miscdev_quit(current->euid, - current->nsproxy->user_ns, - task_pid(current)); - if (rc) { - printk(KERN_ERR "%s: Error attempting to process " - "quit from pid [0x%p]; rc = [%d]\n", __func__, - task_pid(current), rc); - goto out_free; - } break; default: ecryptfs_printk(KERN_WARNING, "Dropping miscdev " From 982363c97f8cad7aea4c3d2cfebffc1cc2d2f166 Mon Sep 17 00:00:00 2001 From: Eric Sandeen Date: Wed, 23 Jul 2008 21:30:04 -0700 Subject: [PATCH 317/857] ecryptfs: propagate key errors up at mount time Mounting with invalid key signatures should probably fail, if they were specifically requested but not available. Also fix case checks in process_request_key_err() for the right sign of the errnos, as spotted by Jan Tluka. Signed-off-by: Eric Sandeen Reviewed-by: Jan Tluka Acked-by: Michael Halcrow Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ecryptfs/keystore.c | 9 ++++----- fs/ecryptfs/main.c | 4 ++-- 2 files changed, 6 insertions(+), 7 deletions(-) diff --git a/fs/ecryptfs/keystore.c b/fs/ecryptfs/keystore.c index e82b457180be..f5b76a331b9c 100644 --- a/fs/ecryptfs/keystore.c +++ b/fs/ecryptfs/keystore.c @@ -44,15 +44,15 @@ static int process_request_key_err(long err_code) int rc = 0; switch (err_code) { - case ENOKEY: + case -ENOKEY: ecryptfs_printk(KERN_WARNING, "No key\n"); rc = -ENOENT; break; - case EKEYEXPIRED: + case -EKEYEXPIRED: ecryptfs_printk(KERN_WARNING, "Key expired\n"); rc = -ETIME; break; - case EKEYREVOKED: + case -EKEYREVOKED: ecryptfs_printk(KERN_WARNING, "Key revoked\n"); rc = -EINVAL; break; @@ -963,8 +963,7 @@ int ecryptfs_keyring_auth_tok_for_sig(struct key **auth_tok_key, if (!(*auth_tok_key) || IS_ERR(*auth_tok_key)) { printk(KERN_ERR "Could not find key with description: [%s]\n", sig); - process_request_key_err(PTR_ERR(*auth_tok_key)); - rc = -EINVAL; + rc = process_request_key_err(PTR_ERR(*auth_tok_key)); goto out; } (*auth_tok) = ecryptfs_get_key_payload_data(*auth_tok_key); diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index f36ab2feea28..8876fe7c76e2 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -248,10 +248,11 @@ static int ecryptfs_init_global_auth_toks( "session keyring for sig specified in mount " "option: [%s]\n", global_auth_tok->sig); global_auth_tok->flags |= ECRYPTFS_AUTH_TOK_INVALID; - rc = 0; + goto out; } else global_auth_tok->flags &= ~ECRYPTFS_AUTH_TOK_INVALID; } +out: return rc; } @@ -416,7 +417,6 @@ static int ecryptfs_parse_options(struct super_block *sb, char *options) printk(KERN_WARNING "One or more global auth toks could not " "properly register; rc = [%d]\n", rc); } - rc = 0; out: return rc; } From 8f2368095e25018838e1bf145041f58270ccd32e Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 23 Jul 2008 21:30:05 -0700 Subject: [PATCH 318/857] ecryptfs: string copy cleanup Clean up overcomplicated string copy, which also gets rid of this bogus warning: fs/ecryptfs/main.c: In function 'ecryptfs_parse_options': include/asm/arch/string_32.h:75: warning: array subscript is above array bounds Signed-off-by: Miklos Szeredi Cc: Michael Halcrow Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ecryptfs/main.c | 18 ++++++------------ 1 file changed, 6 insertions(+), 12 deletions(-) diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index 8876fe7c76e2..10475d93ff53 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -301,7 +301,6 @@ static int ecryptfs_parse_options(struct super_block *sb, char *options) char *cipher_name_dst; char *cipher_name_src; char *cipher_key_bytes_src; - int cipher_name_len; if (!options) { rc = -EINVAL; @@ -382,17 +381,12 @@ static int ecryptfs_parse_options(struct super_block *sb, char *options) goto out; } if (!cipher_name_set) { - cipher_name_len = strlen(ECRYPTFS_DEFAULT_CIPHER); - if (unlikely(cipher_name_len - >= ECRYPTFS_MAX_CIPHER_NAME_SIZE)) { - rc = -EINVAL; - BUG(); - goto out; - } - memcpy(mount_crypt_stat->global_default_cipher_name, - ECRYPTFS_DEFAULT_CIPHER, cipher_name_len); - mount_crypt_stat->global_default_cipher_name[cipher_name_len] - = '\0'; + int cipher_name_len = strlen(ECRYPTFS_DEFAULT_CIPHER); + + BUG_ON(cipher_name_len >= ECRYPTFS_MAX_CIPHER_NAME_SIZE); + + strcpy(mount_crypt_stat->global_default_cipher_name, + ECRYPTFS_DEFAULT_CIPHER); } if (!cipher_key_bytes_set) { mount_crypt_stat->global_default_cipher_key_size = 0; From 29335c6a41568d4708d4ec3b9187f9b6d302e5ea Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Wed, 23 Jul 2008 21:30:06 -0700 Subject: [PATCH 319/857] ecryptfs: crypto.c use unaligned byteorder helpers Fixes the following sparse warnings: fs/ecryptfs/crypto.c:1036:8: warning: cast to restricted __be32 fs/ecryptfs/crypto.c:1038:8: warning: cast to restricted __be32 fs/ecryptfs/crypto.c:1077:10: warning: cast to restricted __be32 fs/ecryptfs/crypto.c:1103:6: warning: incorrect type in assignment (different base types) fs/ecryptfs/crypto.c:1105:6: warning: incorrect type in assignment (different base types) fs/ecryptfs/crypto.c:1124:8: warning: incorrect type in assignment (different base types) fs/ecryptfs/crypto.c:1241:21: warning: incorrect type in assignment (different base types) fs/ecryptfs/crypto.c:1244:30: warning: incorrect type in assignment (different base types) fs/ecryptfs/crypto.c:1414:23: warning: cast to restricted __be32 fs/ecryptfs/crypto.c:1417:32: warning: cast to restricted __be16 Signed-off-by: Harvey Harrison Cc: Michael Halcrow Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ecryptfs/crypto.c | 37 ++++++++++++++----------------------- 1 file changed, 14 insertions(+), 23 deletions(-) diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index e2832bc7869a..7b99917ffadc 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -33,6 +33,7 @@ #include #include #include +#include #include "ecryptfs_kernel.h" static int @@ -1032,10 +1033,8 @@ static int contains_ecryptfs_marker(char *data) { u32 m_1, m_2; - memcpy(&m_1, data, 4); - m_1 = be32_to_cpu(m_1); - memcpy(&m_2, (data + 4), 4); - m_2 = be32_to_cpu(m_2); + m_1 = get_unaligned_be32(data); + m_2 = get_unaligned_be32(data + 4); if ((m_1 ^ MAGIC_ECRYPTFS_MARKER) == m_2) return 1; ecryptfs_printk(KERN_DEBUG, "m_1 = [0x%.8x]; m_2 = [0x%.8x]; " @@ -1073,8 +1072,7 @@ static int ecryptfs_process_flags(struct ecryptfs_crypt_stat *crypt_stat, int i; u32 flags; - memcpy(&flags, page_virt, 4); - flags = be32_to_cpu(flags); + flags = get_unaligned_be32(page_virt); for (i = 0; i < ((sizeof(ecryptfs_flag_map) / sizeof(struct ecryptfs_flag_map_elem))); i++) if (flags & ecryptfs_flag_map[i].file_flag) { @@ -1100,11 +1098,9 @@ static void write_ecryptfs_marker(char *page_virt, size_t *written) get_random_bytes(&m_1, (MAGIC_ECRYPTFS_MARKER_SIZE_BYTES / 2)); m_2 = (m_1 ^ MAGIC_ECRYPTFS_MARKER); - m_1 = cpu_to_be32(m_1); - memcpy(page_virt, &m_1, (MAGIC_ECRYPTFS_MARKER_SIZE_BYTES / 2)); - m_2 = cpu_to_be32(m_2); - memcpy(page_virt + (MAGIC_ECRYPTFS_MARKER_SIZE_BYTES / 2), &m_2, - (MAGIC_ECRYPTFS_MARKER_SIZE_BYTES / 2)); + put_unaligned_be32(m_1, page_virt); + page_virt += (MAGIC_ECRYPTFS_MARKER_SIZE_BYTES / 2); + put_unaligned_be32(m_2, page_virt); (*written) = MAGIC_ECRYPTFS_MARKER_SIZE_BYTES; } @@ -1121,8 +1117,7 @@ write_ecryptfs_flags(char *page_virt, struct ecryptfs_crypt_stat *crypt_stat, flags |= ecryptfs_flag_map[i].file_flag; /* Version is in top 8 bits of the 32-bit flag vector */ flags |= ((((u8)crypt_stat->file_version) << 24) & 0xFF000000); - flags = cpu_to_be32(flags); - memcpy(page_virt, &flags, 4); + put_unaligned_be32(flags, page_virt); (*written) = 4; } @@ -1238,11 +1233,9 @@ ecryptfs_write_header_metadata(char *virt, num_header_extents_at_front = (u16)(crypt_stat->num_header_bytes_at_front / crypt_stat->extent_size); - header_extent_size = cpu_to_be32(header_extent_size); - memcpy(virt, &header_extent_size, 4); + put_unaligned_be32(header_extent_size, virt); virt += 4; - num_header_extents_at_front = cpu_to_be16(num_header_extents_at_front); - memcpy(virt, &num_header_extents_at_front, 2); + put_unaligned_be16(num_header_extents_at_front, virt); (*written) = 6; } @@ -1410,15 +1403,13 @@ static int parse_header_metadata(struct ecryptfs_crypt_stat *crypt_stat, u32 header_extent_size; u16 num_header_extents_at_front; - memcpy(&header_extent_size, virt, sizeof(u32)); - header_extent_size = be32_to_cpu(header_extent_size); - virt += sizeof(u32); - memcpy(&num_header_extents_at_front, virt, sizeof(u16)); - num_header_extents_at_front = be16_to_cpu(num_header_extents_at_front); + header_extent_size = get_unaligned_be32(virt); + virt += sizeof(__be32); + num_header_extents_at_front = get_unaligned_be16(virt); crypt_stat->num_header_bytes_at_front = (((size_t)num_header_extents_at_front * (size_t)header_extent_size)); - (*bytes_read) = (sizeof(u32) + sizeof(u16)); + (*bytes_read) = (sizeof(__be32) + sizeof(__be16)); if ((validate_header_size == ECRYPTFS_VALIDATE_HEADER_SIZE) && (crypt_stat->num_header_bytes_at_front < ECRYPTFS_MINIMUM_HEADER_EXTENT_SIZE)) { From 0a688ad713949643e201431d3f4a4ceddfeb70ca Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Wed, 23 Jul 2008 21:30:07 -0700 Subject: [PATCH 320/857] ecryptfs: inode.c mmap.c use unaligned byteorder helpers Fixe sparse warnings: fs/ecryptfs/inode.c:368:15: warning: cast to restricted __be64 fs/ecryptfs/mmap.c:385:12: warning: incorrect type in assignment (different base types) fs/ecryptfs/mmap.c:385:12: expected unsigned long long [unsigned] [assigned] [usertype] file_size fs/ecryptfs/mmap.c:385:12: got restricted __be64 [usertype] fs/ecryptfs/mmap.c:428:12: warning: incorrect type in assignment (different base types) fs/ecryptfs/mmap.c:428:12: expected unsigned long long [unsigned] [assigned] [usertype] file_size fs/ecryptfs/mmap.c:428:12: got restricted __be64 [usertype] Signed-off-by: Harvey Harrison Cc: Michael Halcrow Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ecryptfs/inode.c | 4 ++-- fs/ecryptfs/mmap.c | 11 +++-------- 2 files changed, 5 insertions(+), 10 deletions(-) diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index c92cc1c00aae..7315547193ea 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -31,6 +31,7 @@ #include #include #include +#include #include "ecryptfs_kernel.h" static struct dentry *lock_parent(struct dentry *dentry) @@ -364,8 +365,7 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry, else file_size = i_size_read(lower_dentry->d_inode); } else { - memcpy(&file_size, page_virt, sizeof(file_size)); - file_size = be64_to_cpu(file_size); + file_size = get_unaligned_be64(page_virt); } i_size_write(dentry->d_inode, (loff_t)file_size); kmem_cache_free(ecryptfs_header_cache_2, page_virt); diff --git a/fs/ecryptfs/mmap.c b/fs/ecryptfs/mmap.c index 2b6fe1e6e8ba..245c2dc02d5c 100644 --- a/fs/ecryptfs/mmap.c +++ b/fs/ecryptfs/mmap.c @@ -32,6 +32,7 @@ #include #include #include +#include #include "ecryptfs_kernel.h" /** @@ -372,7 +373,6 @@ static int ecryptfs_prepare_write(struct file *file, struct page *page, */ static int ecryptfs_write_inode_size_to_header(struct inode *ecryptfs_inode) { - u64 file_size; char *file_size_virt; int rc; @@ -381,9 +381,7 @@ static int ecryptfs_write_inode_size_to_header(struct inode *ecryptfs_inode) rc = -ENOMEM; goto out; } - file_size = (u64)i_size_read(ecryptfs_inode); - file_size = cpu_to_be64(file_size); - memcpy(file_size_virt, &file_size, sizeof(u64)); + put_unaligned_be64(i_size_read(ecryptfs_inode), file_size_virt); rc = ecryptfs_write_lower(ecryptfs_inode, file_size_virt, 0, sizeof(u64)); kfree(file_size_virt); @@ -403,7 +401,6 @@ static int ecryptfs_write_inode_size_to_xattr(struct inode *ecryptfs_inode) struct dentry *lower_dentry = ecryptfs_inode_to_private(ecryptfs_inode)->lower_file->f_dentry; struct inode *lower_inode = lower_dentry->d_inode; - u64 file_size; int rc; if (!lower_inode->i_op->getxattr || !lower_inode->i_op->setxattr) { @@ -424,9 +421,7 @@ static int ecryptfs_write_inode_size_to_xattr(struct inode *ecryptfs_inode) xattr_virt, PAGE_CACHE_SIZE); if (size < 0) size = 8; - file_size = (u64)i_size_read(ecryptfs_inode); - file_size = cpu_to_be64(file_size); - memcpy(xattr_virt, &file_size, sizeof(u64)); + put_unaligned_be64(i_size_read(ecryptfs_inode), xattr_virt); rc = lower_inode->i_op->setxattr(lower_dentry, ECRYPTFS_XATTR_NAME, xattr_virt, size, 0); mutex_unlock(&lower_inode->i_mutex); From 72b55fffd631a89e5be6fe1b4f2565bc4cd90deb Mon Sep 17 00:00:00 2001 From: Michael Halcrow Date: Wed, 23 Jul 2008 21:30:07 -0700 Subject: [PATCH 321/857] eCryptfs: do not try to open device files on mknod When creating device nodes, eCryptfs needs to delay actually opening the lower persistent file until an application tries to open. Device handles may not be backed by anything when they first come into existence. [Valdis.Kletnieks@vt.edu: build fix] Signed-off-by: Michael Halcrow Cc: Signed-off-by: Linus Torvalds --- fs/ecryptfs/ecryptfs_kernel.h | 6 +++++- fs/ecryptfs/file.c | 14 ++++++++++++++ fs/ecryptfs/inode.c | 6 ++++-- fs/ecryptfs/main.c | 29 +++++++++++++++++++---------- 4 files changed, 42 insertions(+), 13 deletions(-) diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h index b4a0cccfdd7c..b0727f91454e 100644 --- a/fs/ecryptfs/ecryptfs_kernel.h +++ b/fs/ecryptfs/ecryptfs_kernel.h @@ -235,6 +235,7 @@ struct ecryptfs_crypt_stat { #define ECRYPTFS_METADATA_IN_XATTR 0x00000100 #define ECRYPTFS_VIEW_AS_ENCRYPTED 0x00000200 #define ECRYPTFS_KEY_SET 0x00000400 +#define ECRYPTFS_DELAY_PERSISTENT 0x00000800 u32 flags; unsigned int file_version; size_t iv_bytes; @@ -574,9 +575,11 @@ struct ecryptfs_open_req { struct list_head kthread_ctl_list; }; +#define ECRYPTFS_INTERPOSE_FLAG_D_ADD 0x00000001 +#define ECRYPTFS_INTERPOSE_FLAG_DELAY_PERSISTENT_FILE 0x00000002 int ecryptfs_interpose(struct dentry *hidden_dentry, struct dentry *this_dentry, struct super_block *sb, - int flag); + u32 flags); int ecryptfs_fill_zeros(struct file *file, loff_t new_length); int ecryptfs_decode_filename(struct ecryptfs_crypt_stat *crypt_stat, const char *name, int length, @@ -709,5 +712,6 @@ void ecryptfs_destroy_kthread(void); int ecryptfs_privileged_open(struct file **lower_file, struct dentry *lower_dentry, struct vfsmount *lower_mnt); +int ecryptfs_init_persistent_file(struct dentry *ecryptfs_dentry); #endif /* #ifndef ECRYPTFS_KERNEL_H */ diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c index f0be29051528..2c2d60df3f60 100644 --- a/fs/ecryptfs/file.c +++ b/fs/ecryptfs/file.c @@ -199,6 +199,20 @@ static int ecryptfs_open(struct inode *inode, struct file *file) "file must hence be opened RO\n", __func__); goto out; } + if (!ecryptfs_inode_to_private(inode)->lower_file) { + BUG_ON(!(crypt_stat->flags & ECRYPTFS_DELAY_PERSISTENT)); + mutex_lock(&crypt_stat->cs_mutex); + crypt_stat->flags &= ~(ECRYPTFS_ENCRYPTED); + mutex_unlock(&crypt_stat->cs_mutex); + rc = ecryptfs_init_persistent_file(ecryptfs_dentry); + if (rc) { + printk(KERN_ERR "%s: Error attempting to initialize " + "the persistent file for the dentry with name " + "[%s]; rc = [%d]\n", __func__, + ecryptfs_dentry->d_name.name, rc); + goto out; + } + } ecryptfs_set_file_lower( file, ecryptfs_inode_to_private(inode)->lower_file); if (S_ISDIR(ecryptfs_dentry->d_inode->i_mode)) { diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index 7315547193ea..26090878c930 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -308,7 +308,8 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry, d_add(dentry, NULL); goto out; } - rc = ecryptfs_interpose(lower_dentry, dentry, dir->i_sb, 1); + rc = ecryptfs_interpose(lower_dentry, dentry, dir->i_sb, + ECRYPTFS_INTERPOSE_FLAG_D_ADD); if (rc) { ecryptfs_printk(KERN_ERR, "Error interposing\n"); goto out_dput; @@ -537,7 +538,8 @@ ecryptfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev) rc = vfs_mknod(lower_dir_dentry->d_inode, lower_dentry, mode, dev); if (rc || !lower_dentry->d_inode) goto out; - rc = ecryptfs_interpose(lower_dentry, dentry, dir->i_sb, 0); + rc = ecryptfs_interpose(lower_dentry, dentry, dir->i_sb, + ECRYPTFS_INTERPOSE_FLAG_DELAY_PERSISTENT_FILE); if (rc) goto out; fsstack_copy_attr_times(dir, lower_dir_dentry->d_inode); diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index 10475d93ff53..ee4f84b20410 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -117,7 +117,7 @@ void __ecryptfs_printk(const char *fmt, ...) * * Returns zero on success; non-zero otherwise */ -static int ecryptfs_init_persistent_file(struct dentry *ecryptfs_dentry) +int ecryptfs_init_persistent_file(struct dentry *ecryptfs_dentry) { struct ecryptfs_inode_info *inode_info = ecryptfs_inode_to_private(ecryptfs_dentry->d_inode); @@ -149,14 +149,14 @@ static int ecryptfs_init_persistent_file(struct dentry *ecryptfs_dentry) * @lower_dentry: Existing dentry in the lower filesystem * @dentry: ecryptfs' dentry * @sb: ecryptfs's super_block - * @flag: If set to true, then d_add is called, else d_instantiate is called + * @flags: flags to govern behavior of interpose procedure * * Interposes upper and lower dentries. * * Returns zero on success; non-zero otherwise */ int ecryptfs_interpose(struct dentry *lower_dentry, struct dentry *dentry, - struct super_block *sb, int flag) + struct super_block *sb, u32 flags) { struct inode *lower_inode; struct inode *inode; @@ -193,7 +193,7 @@ int ecryptfs_interpose(struct dentry *lower_dentry, struct dentry *dentry, init_special_inode(inode, lower_inode->i_mode, lower_inode->i_rdev); dentry->d_op = &ecryptfs_dops; - if (flag) + if (flags & ECRYPTFS_INTERPOSE_FLAG_D_ADD) d_add(dentry, inode); else d_instantiate(dentry, inode); @@ -201,12 +201,21 @@ int ecryptfs_interpose(struct dentry *lower_dentry, struct dentry *dentry, /* This size will be overwritten for real files w/ headers and * other metadata */ fsstack_copy_inode_size(inode, lower_inode); - rc = ecryptfs_init_persistent_file(dentry); - if (rc) { - printk(KERN_ERR "%s: Error attempting to initialize the " - "persistent file for the dentry with name [%s]; " - "rc = [%d]\n", __func__, dentry->d_name.name, rc); - goto out; + if (!(flags & ECRYPTFS_INTERPOSE_FLAG_DELAY_PERSISTENT_FILE)) { + rc = ecryptfs_init_persistent_file(dentry); + if (rc) { + printk(KERN_ERR "%s: Error attempting to initialize " + "the persistent file for the dentry with name " + "[%s]; rc = [%d]\n", __func__, + dentry->d_name.name, rc); + goto out; + } + } else { + struct ecryptfs_inode_info *inode_info = + ecryptfs_inode_to_private(dentry->d_inode); + + inode_info->lower_file = NULL; + inode_info->crypt_stat.flags |= ECRYPTFS_DELAY_PERSISTENT; } out: return rc; From 391b52f98cf2e9bff227dad8bf9ea206fec43fa4 Mon Sep 17 00:00:00 2001 From: Michael Halcrow Date: Wed, 23 Jul 2008 21:30:08 -0700 Subject: [PATCH 322/857] eCryptfs: Make all persistent file opens delayed There is no good reason to immediately open the lower file, and that can cause problems with files that the user does not intend to immediately open, such as device nodes. This patch removes the persistent file open from the interpose step and pushes that to the locations where eCryptfs really does need the lower persistent file, such as just before reading or writing the metadata stored in the lower file header. Two functions are jumping to out_dput when they should just be jumping to out on error paths. This patch also fixes these. Signed-off-by: Michael Halcrow Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ecryptfs/ecryptfs_kernel.h | 2 -- fs/ecryptfs/file.c | 4 ---- fs/ecryptfs/inode.c | 27 +++++++++++++++++++++++---- fs/ecryptfs/main.c | 16 ---------------- 4 files changed, 23 insertions(+), 26 deletions(-) diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h index b0727f91454e..b73fb752c5f8 100644 --- a/fs/ecryptfs/ecryptfs_kernel.h +++ b/fs/ecryptfs/ecryptfs_kernel.h @@ -235,7 +235,6 @@ struct ecryptfs_crypt_stat { #define ECRYPTFS_METADATA_IN_XATTR 0x00000100 #define ECRYPTFS_VIEW_AS_ENCRYPTED 0x00000200 #define ECRYPTFS_KEY_SET 0x00000400 -#define ECRYPTFS_DELAY_PERSISTENT 0x00000800 u32 flags; unsigned int file_version; size_t iv_bytes; @@ -576,7 +575,6 @@ struct ecryptfs_open_req { }; #define ECRYPTFS_INTERPOSE_FLAG_D_ADD 0x00000001 -#define ECRYPTFS_INTERPOSE_FLAG_DELAY_PERSISTENT_FILE 0x00000002 int ecryptfs_interpose(struct dentry *hidden_dentry, struct dentry *this_dentry, struct super_block *sb, u32 flags); diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c index 2c2d60df3f60..9244d653743e 100644 --- a/fs/ecryptfs/file.c +++ b/fs/ecryptfs/file.c @@ -200,10 +200,6 @@ static int ecryptfs_open(struct inode *inode, struct file *file) goto out; } if (!ecryptfs_inode_to_private(inode)->lower_file) { - BUG_ON(!(crypt_stat->flags & ECRYPTFS_DELAY_PERSISTENT)); - mutex_lock(&crypt_stat->cs_mutex); - crypt_stat->flags &= ~(ECRYPTFS_ENCRYPTED); - mutex_unlock(&crypt_stat->cs_mutex); rc = ecryptfs_init_persistent_file(ecryptfs_dentry); if (rc) { printk(KERN_ERR "%s: Error attempting to initialize " diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index 26090878c930..d755455e3bff 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -189,6 +189,16 @@ static int ecryptfs_initialize_file(struct dentry *ecryptfs_dentry) "context; rc = [%d]\n", rc); goto out; } + if (!ecryptfs_inode_to_private(ecryptfs_dentry->d_inode)->lower_file) { + rc = ecryptfs_init_persistent_file(ecryptfs_dentry); + if (rc) { + printk(KERN_ERR "%s: Error attempting to initialize " + "the persistent file for the dentry with name " + "[%s]; rc = [%d]\n", __func__, + ecryptfs_dentry->d_name.name, rc); + goto out; + } + } rc = ecryptfs_write_metadata(ecryptfs_dentry); if (rc) { printk(KERN_ERR "Error writing headers; rc = [%d]\n", rc); @@ -312,7 +322,7 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry, ECRYPTFS_INTERPOSE_FLAG_D_ADD); if (rc) { ecryptfs_printk(KERN_ERR, "Error interposing\n"); - goto out_dput; + goto out; } if (S_ISDIR(lower_inode->i_mode)) { ecryptfs_printk(KERN_DEBUG, "Is a directory; returning\n"); @@ -338,11 +348,21 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry, rc = -ENOMEM; ecryptfs_printk(KERN_ERR, "Cannot ecryptfs_kmalloc a page\n"); - goto out_dput; + goto out; } crypt_stat = &ecryptfs_inode_to_private(dentry->d_inode)->crypt_stat; if (!(crypt_stat->flags & ECRYPTFS_POLICY_APPLIED)) ecryptfs_set_default_sizes(crypt_stat); + if (!ecryptfs_inode_to_private(dentry->d_inode)->lower_file) { + rc = ecryptfs_init_persistent_file(dentry); + if (rc) { + printk(KERN_ERR "%s: Error attempting to initialize " + "the persistent file for the dentry with name " + "[%s]; rc = [%d]\n", __func__, + dentry->d_name.name, rc); + goto out; + } + } rc = ecryptfs_read_and_validate_header_region(page_virt, dentry->d_inode); if (rc) { @@ -538,8 +558,7 @@ ecryptfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev) rc = vfs_mknod(lower_dir_dentry->d_inode, lower_dentry, mode, dev); if (rc || !lower_dentry->d_inode) goto out; - rc = ecryptfs_interpose(lower_dentry, dentry, dir->i_sb, - ECRYPTFS_INTERPOSE_FLAG_DELAY_PERSISTENT_FILE); + rc = ecryptfs_interpose(lower_dentry, dentry, dir->i_sb, 0); if (rc) goto out; fsstack_copy_attr_times(dir, lower_dir_dentry->d_inode); diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index ee4f84b20410..6f403cfba14f 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -201,22 +201,6 @@ int ecryptfs_interpose(struct dentry *lower_dentry, struct dentry *dentry, /* This size will be overwritten for real files w/ headers and * other metadata */ fsstack_copy_inode_size(inode, lower_inode); - if (!(flags & ECRYPTFS_INTERPOSE_FLAG_DELAY_PERSISTENT_FILE)) { - rc = ecryptfs_init_persistent_file(dentry); - if (rc) { - printk(KERN_ERR "%s: Error attempting to initialize " - "the persistent file for the dentry with name " - "[%s]; rc = [%d]\n", __func__, - dentry->d_name.name, rc); - goto out; - } - } else { - struct ecryptfs_inode_info *inode_info = - ecryptfs_inode_to_private(dentry->d_inode); - - inode_info->lower_file = NULL; - inode_info->crypt_stat.flags |= ECRYPTFS_DELAY_PERSISTENT; - } out: return rc; } From 5f6f4f28b6ba543beef8bad91aa6f69c7ffeee51 Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Wed, 23 Jul 2008 21:30:09 -0700 Subject: [PATCH 323/857] autofs4: don't make expiring dentry negative Correct the error of making a positive dentry negative after it has been instantiated. The code that makes this error attempts to re-use the dentry from a concurrent expire and mount to resolve a race and the dentry used for the lookup must be negative for mounts to trigger in the required cases. The fact is that the dentry doesn't need to be re-used because all that is needed is to preserve the flag that indicates an expire is still incomplete at the time of the mount request. This change uses the the dentry to check the flag and wait for the expire to complete then discards it instead of attempting to re-use it. Signed-off-by: Ian Kent Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/autofs_i.h | 6 +-- fs/autofs4/inode.c | 6 +-- fs/autofs4/root.c | 118 ++++++++++++++++-------------------------- 3 files changed, 52 insertions(+), 78 deletions(-) diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h index c3d352d7fa93..69b1497b0029 100644 --- a/fs/autofs4/autofs_i.h +++ b/fs/autofs4/autofs_i.h @@ -52,7 +52,7 @@ struct autofs_info { int flags; - struct list_head rehash; + struct list_head expiring; struct autofs_sb_info *sbi; unsigned long last_used; @@ -112,8 +112,8 @@ struct autofs_sb_info { struct mutex wq_mutex; spinlock_t fs_lock; struct autofs_wait_queue *queues; /* Wait queue pointer */ - spinlock_t rehash_lock; - struct list_head rehash_list; + spinlock_t lookup_lock; + struct list_head expiring_list; }; static inline struct autofs_sb_info *autofs4_sbi(struct super_block *sb) diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c index 2fdcf5e1d236..94bfc154d7a2 100644 --- a/fs/autofs4/inode.c +++ b/fs/autofs4/inode.c @@ -47,7 +47,7 @@ struct autofs_info *autofs4_init_ino(struct autofs_info *ino, ino->dentry = NULL; ino->size = 0; - INIT_LIST_HEAD(&ino->rehash); + INIT_LIST_HEAD(&ino->expiring); ino->last_used = jiffies; atomic_set(&ino->count, 0); @@ -338,8 +338,8 @@ int autofs4_fill_super(struct super_block *s, void *data, int silent) mutex_init(&sbi->wq_mutex); spin_lock_init(&sbi->fs_lock); sbi->queues = NULL; - spin_lock_init(&sbi->rehash_lock); - INIT_LIST_HEAD(&sbi->rehash_list); + spin_lock_init(&sbi->lookup_lock); + INIT_LIST_HEAD(&sbi->expiring_list); s->s_blocksize = 1024; s->s_blocksize_bits = 10; s->s_magic = AUTOFS_SUPER_MAGIC; diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index edf5b6bddb52..9ead2279df4f 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -493,10 +493,10 @@ void autofs4_dentry_release(struct dentry *de) struct autofs_sb_info *sbi = autofs4_sbi(de->d_sb); if (sbi) { - spin_lock(&sbi->rehash_lock); - if (!list_empty(&inf->rehash)) - list_del(&inf->rehash); - spin_unlock(&sbi->rehash_lock); + spin_lock(&sbi->lookup_lock); + if (!list_empty(&inf->expiring)) + list_del(&inf->expiring); + spin_unlock(&sbi->lookup_lock); } inf->dentry = NULL; @@ -518,7 +518,7 @@ static struct dentry_operations autofs4_dentry_operations = { .d_release = autofs4_dentry_release, }; -static struct dentry *autofs4_lookup_unhashed(struct autofs_sb_info *sbi, struct dentry *parent, struct qstr *name) +static struct dentry *autofs4_lookup_expiring(struct autofs_sb_info *sbi, struct dentry *parent, struct qstr *name) { unsigned int len = name->len; unsigned int hash = name->hash; @@ -526,14 +526,14 @@ static struct dentry *autofs4_lookup_unhashed(struct autofs_sb_info *sbi, struct struct list_head *p, *head; spin_lock(&dcache_lock); - spin_lock(&sbi->rehash_lock); - head = &sbi->rehash_list; + spin_lock(&sbi->lookup_lock); + head = &sbi->expiring_list; list_for_each(p, head) { struct autofs_info *ino; struct dentry *dentry; struct qstr *qstr; - ino = list_entry(p, struct autofs_info, rehash); + ino = list_entry(p, struct autofs_info, expiring); dentry = ino->dentry; spin_lock(&dentry->d_lock); @@ -555,33 +555,16 @@ static struct dentry *autofs4_lookup_unhashed(struct autofs_sb_info *sbi, struct goto next; if (d_unhashed(dentry)) { - struct inode *inode = dentry->d_inode; - - ino = autofs4_dentry_ino(dentry); - list_del_init(&ino->rehash); dget(dentry); - /* - * Make the rehashed dentry negative so the VFS - * behaves as it should. - */ - if (inode) { - dentry->d_inode = NULL; - list_del_init(&dentry->d_alias); - spin_unlock(&dentry->d_lock); - spin_unlock(&sbi->rehash_lock); - spin_unlock(&dcache_lock); - iput(inode); - return dentry; - } spin_unlock(&dentry->d_lock); - spin_unlock(&sbi->rehash_lock); + spin_unlock(&sbi->lookup_lock); spin_unlock(&dcache_lock); return dentry; } next: spin_unlock(&dentry->d_lock); } - spin_unlock(&sbi->rehash_lock); + spin_unlock(&sbi->lookup_lock); spin_unlock(&dcache_lock); return NULL; @@ -591,7 +574,7 @@ static struct dentry *autofs4_lookup_unhashed(struct autofs_sb_info *sbi, struct static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) { struct autofs_sb_info *sbi; - struct dentry *unhashed; + struct dentry *expiring; int oz_mode; DPRINTK("name = %.*s", @@ -607,44 +590,44 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s DPRINTK("pid = %u, pgrp = %u, catatonic = %d, oz_mode = %d", current->pid, task_pgrp_nr(current), sbi->catatonic, oz_mode); - unhashed = autofs4_lookup_unhashed(sbi, dentry->d_parent, &dentry->d_name); - if (!unhashed) { - /* - * Mark the dentry incomplete but don't hash it. We do this - * to serialize our inode creation operations (symlink and - * mkdir) which prevents deadlock during the callback to - * the daemon. Subsequent user space lookups for the same - * dentry are placed on the wait queue while the daemon - * itself is allowed passage unresticted so the create - * operation itself can then hash the dentry. Finally, - * we check for the hashed dentry and return the newly - * hashed dentry. - */ - dentry->d_op = &autofs4_root_dentry_operations; - - dentry->d_fsdata = NULL; - d_instantiate(dentry, NULL); - } else { - struct autofs_info *ino = autofs4_dentry_ino(unhashed); - DPRINTK("rehash %p with %p", dentry, unhashed); + expiring = autofs4_lookup_expiring(sbi, dentry->d_parent, &dentry->d_name); + if (expiring) { + struct autofs_info *ino = autofs4_dentry_ino(expiring); /* * If we are racing with expire the request might not * be quite complete but the directory has been removed * so it must have been successful, so just wait for it. - * We need to ensure the AUTOFS_INF_EXPIRING flag is clear - * before continuing as revalidate may fail when calling - * try_to_fill_dentry (returning EAGAIN) if we don't. */ while (ino && (ino->flags & AUTOFS_INF_EXPIRING)) { DPRINTK("wait for incomplete expire %p name=%.*s", - unhashed, unhashed->d_name.len, - unhashed->d_name.name); - autofs4_wait(sbi, unhashed, NFY_NONE); + expiring, expiring->d_name.len, + expiring->d_name.name); + autofs4_wait(sbi, expiring, NFY_NONE); DPRINTK("request completed"); } - dentry = unhashed; + spin_lock(&sbi->lookup_lock); + if (!list_empty(&ino->expiring)) + list_del_init(&ino->expiring); + spin_unlock(&sbi->lookup_lock); + dput(expiring); } + /* + * Mark the dentry incomplete but don't hash it. We do this + * to serialize our inode creation operations (symlink and + * mkdir) which prevents deadlock during the callback to + * the daemon. Subsequent user space lookups for the same + * dentry are placed on the wait queue while the daemon + * itself is allowed passage unresticted so the create + * operation itself can then hash the dentry. Finally, + * we check for the hashed dentry and return the newly + * hashed dentry. + */ + dentry->d_op = &autofs4_root_dentry_operations; + + dentry->d_fsdata = NULL; + d_instantiate(dentry, NULL); + if (!oz_mode) { spin_lock(&dentry->d_lock); dentry->d_flags |= DCACHE_AUTOFS_PENDING; @@ -668,8 +651,6 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s if (sigismember (sigset, SIGKILL) || sigismember (sigset, SIGQUIT) || sigismember (sigset, SIGINT)) { - if (unhashed) - dput(unhashed); return ERR_PTR(-ERESTARTNOINTR); } } @@ -699,15 +680,9 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s else dentry = ERR_PTR(-ENOENT); - if (unhashed) - dput(unhashed); - return dentry; } - if (unhashed) - return dentry; - return NULL; } @@ -769,9 +744,8 @@ static int autofs4_dir_symlink(struct inode *dir, * that the file no longer exists. However, doing that means that the * VFS layer can turn the dentry into a negative dentry. We don't want * this, because the unlink is probably the result of an expire. - * We simply d_drop it and add it to a rehash candidates list in the - * super block, which allows the dentry lookup to reuse it retaining - * the flags, such as expire in progress, in case we're racing with expire. + * We simply d_drop it and add it to a expiring list in the super block, + * which allows the dentry lookup to check for an incomplete expire. * * If a process is blocked on the dentry waiting for the expire to finish, * it will invalidate the dentry and try to mount with a new one. @@ -801,9 +775,9 @@ static int autofs4_dir_unlink(struct inode *dir, struct dentry *dentry) dir->i_mtime = CURRENT_TIME; spin_lock(&dcache_lock); - spin_lock(&sbi->rehash_lock); - list_add(&ino->rehash, &sbi->rehash_list); - spin_unlock(&sbi->rehash_lock); + spin_lock(&sbi->lookup_lock); + list_add(&ino->expiring, &sbi->expiring_list); + spin_unlock(&sbi->lookup_lock); spin_lock(&dentry->d_lock); __d_drop(dentry); spin_unlock(&dentry->d_lock); @@ -829,9 +803,9 @@ static int autofs4_dir_rmdir(struct inode *dir, struct dentry *dentry) spin_unlock(&dcache_lock); return -ENOTEMPTY; } - spin_lock(&sbi->rehash_lock); - list_add(&ino->rehash, &sbi->rehash_list); - spin_unlock(&sbi->rehash_lock); + spin_lock(&sbi->lookup_lock); + list_add(&ino->expiring, &sbi->expiring_list); + spin_unlock(&sbi->lookup_lock); spin_lock(&dentry->d_lock); __d_drop(dentry); spin_unlock(&dentry->d_lock); From caf7da3d5d4d9dd873eb52d025d8cc63b89f1fdb Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Wed, 23 Jul 2008 21:30:11 -0700 Subject: [PATCH 324/857] autofs4: revert - redo lookup in ttfd This patch series enables the use of a single dentry for lookups prior to the dentry being hashed and so we no longer need to redo the lookup. This patch reverts the patch of commit 033790449ba9c4dcf8478a87693d33df625c23b5. Signed-off-by: Ian Kent Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/root.c | 21 --------------------- 1 file changed, 21 deletions(-) diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index 9ead2279df4f..53dabe8d5b8b 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -242,7 +242,6 @@ static int try_to_fill_dentry(struct dentry *dentry, int flags) { struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb); struct autofs_info *ino = autofs4_dentry_ino(dentry); - struct dentry *new; int status; /* Block on any pending expiry here; invalidate the dentry @@ -320,26 +319,6 @@ static int try_to_fill_dentry(struct dentry *dentry, int flags) dentry->d_flags &= ~DCACHE_AUTOFS_PENDING; spin_unlock(&dentry->d_lock); - /* - * The dentry that is passed in from lookup may not be the one - * we end up using, as mkdir can create a new one. If this - * happens, and another process tries the lookup at the same time, - * it will set the PENDING flag on this new dentry, but add itself - * to our waitq. Then, if after the lookup succeeds, the first - * process that requested the mount performs another lookup of the - * same directory, it will show up as still pending! So, we need - * to redo the lookup here and clear pending on that dentry. - */ - if (d_unhashed(dentry)) { - new = d_lookup(dentry->d_parent, &dentry->d_name); - if (new) { - spin_lock(&new->d_lock); - new->d_flags &= ~DCACHE_AUTOFS_PENDING; - spin_unlock(&new->d_lock); - dput(new); - } - } - return 0; } From 2576737873dc1d9ea461a5955a5f6779b569a350 Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Wed, 23 Jul 2008 21:30:12 -0700 Subject: [PATCH 325/857] autofs4: use look aside list for lookups A while ago a patch to resolve a deadlock during directory creation was merged. This delayed the hashing of lookup dentrys until the ->mkdir() (or ->symlink()) operation completed to ensure we always went through ->lookup() instead of also having processes go through ->revalidate() so our VFS locking remained consistent. Now we are seeing a couple of side affects of that change in situations with heavy mount activity. Two cases have been identified: 1) When a mount request is triggered, due to the delayed hashing, the directory created by user space for the mount point doesn't have the DCACHE_AUTOFS_PENDING flag set. In the case of an autofs multi-mount where a tree of mount point directories are created this can lead to the path walk continuing rather than the dentry being sent to the wait queue to wait for request completion. This is because, if the pending flag isn't set, the criteria for deciding this is a mount in progress fails to hold, namely that the dentry is not a mount point and has no subdirectories. 2) A mount request dentry is initially created negative and unhashed. It remains this way until the ->mkdir() callback completes. Since it is unhashed a fresh dentry is used when the user space mount request creates the mount point directory. This leaves the original dentry negative and unhashed. But revalidate has no way to tell the VFS that the dentry has changed, other than to force another ->lookup() by returning false, which is at best wastefull and at worst not possible. This results in an -ENOENT return from the original path walk when in fact the mount succeeded. To resolve this we need to ensure that the same dentry is used in all calls to ->lookup() during the course of a mount request. This patch achieves that by adding the initial dentry to a look aside list and removes it at ->mkdir() or ->symlink() completion (or when the dentry is released), since these are the only create operations autofs4 supports. Signed-off-by: Ian Kent Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/autofs_i.h | 2 + fs/autofs4/inode.c | 25 ++++--- fs/autofs4/root.c | 169 ++++++++++++++++++++++++++++++++++-------- 3 files changed, 156 insertions(+), 40 deletions(-) diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h index 69b1497b0029..2dce2334737d 100644 --- a/fs/autofs4/autofs_i.h +++ b/fs/autofs4/autofs_i.h @@ -52,6 +52,7 @@ struct autofs_info { int flags; + struct list_head active; struct list_head expiring; struct autofs_sb_info *sbi; @@ -113,6 +114,7 @@ struct autofs_sb_info { spinlock_t fs_lock; struct autofs_wait_queue *queues; /* Wait queue pointer */ spinlock_t lookup_lock; + struct list_head active_list; struct list_head expiring_list; }; diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c index 94bfc154d7a2..e3e70994ab46 100644 --- a/fs/autofs4/inode.c +++ b/fs/autofs4/inode.c @@ -24,8 +24,10 @@ static void ino_lnkfree(struct autofs_info *ino) { - kfree(ino->u.symlink); - ino->u.symlink = NULL; + if (ino->u.symlink) { + kfree(ino->u.symlink); + ino->u.symlink = NULL; + } } struct autofs_info *autofs4_init_ino(struct autofs_info *ino, @@ -41,16 +43,18 @@ struct autofs_info *autofs4_init_ino(struct autofs_info *ino, if (ino == NULL) return NULL; - ino->flags = 0; + if (!reinit) { + ino->flags = 0; + ino->inode = NULL; + ino->dentry = NULL; + ino->size = 0; + INIT_LIST_HEAD(&ino->active); + INIT_LIST_HEAD(&ino->expiring); + atomic_set(&ino->count, 0); + } + ino->mode = mode; - ino->inode = NULL; - ino->dentry = NULL; - ino->size = 0; - - INIT_LIST_HEAD(&ino->expiring); - ino->last_used = jiffies; - atomic_set(&ino->count, 0); ino->sbi = sbi; @@ -339,6 +343,7 @@ int autofs4_fill_super(struct super_block *s, void *data, int silent) spin_lock_init(&sbi->fs_lock); sbi->queues = NULL; spin_lock_init(&sbi->lookup_lock); + INIT_LIST_HEAD(&sbi->active_list); INIT_LIST_HEAD(&sbi->expiring_list); s->s_blocksize = 1024; s->s_blocksize_bits = 10; diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index 53dabe8d5b8b..dbb70d5a4882 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -473,6 +473,8 @@ void autofs4_dentry_release(struct dentry *de) if (sbi) { spin_lock(&sbi->lookup_lock); + if (!list_empty(&inf->active)) + list_del(&inf->active); if (!list_empty(&inf->expiring)) list_del(&inf->expiring); spin_unlock(&sbi->lookup_lock); @@ -497,6 +499,58 @@ static struct dentry_operations autofs4_dentry_operations = { .d_release = autofs4_dentry_release, }; +static struct dentry *autofs4_lookup_active(struct autofs_sb_info *sbi, struct dentry *parent, struct qstr *name) +{ + unsigned int len = name->len; + unsigned int hash = name->hash; + const unsigned char *str = name->name; + struct list_head *p, *head; + + spin_lock(&dcache_lock); + spin_lock(&sbi->lookup_lock); + head = &sbi->active_list; + list_for_each(p, head) { + struct autofs_info *ino; + struct dentry *dentry; + struct qstr *qstr; + + ino = list_entry(p, struct autofs_info, active); + dentry = ino->dentry; + + spin_lock(&dentry->d_lock); + + /* Already gone? */ + if (atomic_read(&dentry->d_count) == 0) + goto next; + + qstr = &dentry->d_name; + + if (dentry->d_name.hash != hash) + goto next; + if (dentry->d_parent != parent) + goto next; + + if (qstr->len != len) + goto next; + if (memcmp(qstr->name, str, len)) + goto next; + + if (d_unhashed(dentry)) { + dget(dentry); + spin_unlock(&dentry->d_lock); + spin_unlock(&sbi->lookup_lock); + spin_unlock(&dcache_lock); + return dentry; + } +next: + spin_unlock(&dentry->d_lock); + } + spin_unlock(&sbi->lookup_lock); + spin_unlock(&dcache_lock); + + return NULL; +} + static struct dentry *autofs4_lookup_expiring(struct autofs_sb_info *sbi, struct dentry *parent, struct qstr *name) { unsigned int len = name->len; @@ -553,7 +607,8 @@ static struct dentry *autofs4_lookup_expiring(struct autofs_sb_info *sbi, struct static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) { struct autofs_sb_info *sbi; - struct dentry *expiring; + struct autofs_info *ino; + struct dentry *expiring, *unhashed; int oz_mode; DPRINTK("name = %.*s", @@ -571,12 +626,12 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s expiring = autofs4_lookup_expiring(sbi, dentry->d_parent, &dentry->d_name); if (expiring) { - struct autofs_info *ino = autofs4_dentry_ino(expiring); /* * If we are racing with expire the request might not * be quite complete but the directory has been removed * so it must have been successful, so just wait for it. */ + ino = autofs4_dentry_ino(expiring); while (ino && (ino->flags & AUTOFS_INF_EXPIRING)) { DPRINTK("wait for incomplete expire %p name=%.*s", expiring, expiring->d_name.len, @@ -591,21 +646,41 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s dput(expiring); } - /* - * Mark the dentry incomplete but don't hash it. We do this - * to serialize our inode creation operations (symlink and - * mkdir) which prevents deadlock during the callback to - * the daemon. Subsequent user space lookups for the same - * dentry are placed on the wait queue while the daemon - * itself is allowed passage unresticted so the create - * operation itself can then hash the dentry. Finally, - * we check for the hashed dentry and return the newly - * hashed dentry. - */ - dentry->d_op = &autofs4_root_dentry_operations; + unhashed = autofs4_lookup_active(sbi, dentry->d_parent, &dentry->d_name); + if (unhashed) + dentry = unhashed; + else { + /* + * Mark the dentry incomplete but don't hash it. We do this + * to serialize our inode creation operations (symlink and + * mkdir) which prevents deadlock during the callback to + * the daemon. Subsequent user space lookups for the same + * dentry are placed on the wait queue while the daemon + * itself is allowed passage unresticted so the create + * operation itself can then hash the dentry. Finally, + * we check for the hashed dentry and return the newly + * hashed dentry. + */ + dentry->d_op = &autofs4_root_dentry_operations; - dentry->d_fsdata = NULL; - d_instantiate(dentry, NULL); + /* + * And we need to ensure that the same dentry is used for + * all following lookup calls until it is hashed so that + * the dentry flags are persistent throughout the request. + */ + ino = autofs4_init_ino(NULL, sbi, 0555); + if (!ino) + return ERR_PTR(-ENOMEM); + + dentry->d_fsdata = ino; + ino->dentry = dentry; + + spin_lock(&sbi->lookup_lock); + list_add(&ino->active, &sbi->active_list); + spin_unlock(&sbi->lookup_lock); + + d_instantiate(dentry, NULL); + } if (!oz_mode) { spin_lock(&dentry->d_lock); @@ -630,12 +705,16 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s if (sigismember (sigset, SIGKILL) || sigismember (sigset, SIGQUIT) || sigismember (sigset, SIGINT)) { + if (unhashed) + dput(unhashed); return ERR_PTR(-ERESTARTNOINTR); } } - spin_lock(&dentry->d_lock); - dentry->d_flags &= ~DCACHE_AUTOFS_PENDING; - spin_unlock(&dentry->d_lock); + if (!oz_mode) { + spin_lock(&dentry->d_lock); + dentry->d_flags &= ~DCACHE_AUTOFS_PENDING; + spin_unlock(&dentry->d_lock); + } } /* @@ -659,9 +738,15 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s else dentry = ERR_PTR(-ENOENT); + if (unhashed) + dput(unhashed); + return dentry; } + if (unhashed) + return unhashed; + return NULL; } @@ -682,20 +767,30 @@ static int autofs4_dir_symlink(struct inode *dir, return -EACCES; ino = autofs4_init_ino(ino, sbi, S_IFLNK | 0555); - if (ino == NULL) - return -ENOSPC; + if (!ino) + return -ENOMEM; - ino->size = strlen(symname); - ino->u.symlink = cp = kmalloc(ino->size + 1, GFP_KERNEL); + spin_lock(&sbi->lookup_lock); + if (!list_empty(&ino->active)) + list_del_init(&ino->active); + spin_unlock(&sbi->lookup_lock); - if (cp == NULL) { - kfree(ino); - return -ENOSPC; + cp = kmalloc(ino->size + 1, GFP_KERNEL); + if (!cp) { + if (!dentry->d_fsdata) + kfree(ino); + return -ENOMEM; } strcpy(cp, symname); inode = autofs4_get_inode(dir->i_sb, ino); + if (!inode) { + kfree(cp); + if (!dentry->d_fsdata) + kfree(ino); + return -ENOMEM; + } d_add(dentry, inode); if (dir == dir->i_sb->s_root->d_inode) @@ -711,6 +806,8 @@ static int autofs4_dir_symlink(struct inode *dir, atomic_inc(&p_ino->count); ino->inode = inode; + ino->size = strlen(symname); + ino->u.symlink = cp; dir->i_mtime = CURRENT_TIME; return 0; @@ -755,7 +852,8 @@ static int autofs4_dir_unlink(struct inode *dir, struct dentry *dentry) spin_lock(&dcache_lock); spin_lock(&sbi->lookup_lock); - list_add(&ino->expiring, &sbi->expiring_list); + if (list_empty(&ino->expiring)) + list_add(&ino->expiring, &sbi->expiring_list); spin_unlock(&sbi->lookup_lock); spin_lock(&dentry->d_lock); __d_drop(dentry); @@ -783,7 +881,8 @@ static int autofs4_dir_rmdir(struct inode *dir, struct dentry *dentry) return -ENOTEMPTY; } spin_lock(&sbi->lookup_lock); - list_add(&ino->expiring, &sbi->expiring_list); + if (list_empty(&ino->expiring)) + list_add(&ino->expiring, &sbi->expiring_list); spin_unlock(&sbi->lookup_lock); spin_lock(&dentry->d_lock); __d_drop(dentry); @@ -819,10 +918,20 @@ static int autofs4_dir_mkdir(struct inode *dir, struct dentry *dentry, int mode) dentry, dentry->d_name.len, dentry->d_name.name); ino = autofs4_init_ino(ino, sbi, S_IFDIR | 0555); - if (ino == NULL) - return -ENOSPC; + if (!ino) + return -ENOMEM; + + spin_lock(&sbi->lookup_lock); + if (!list_empty(&ino->active)) + list_del_init(&ino->active); + spin_unlock(&sbi->lookup_lock); inode = autofs4_get_inode(dir->i_sb, ino); + if (!inode) { + if (!dentry->d_fsdata) + kfree(ino); + return -ENOMEM; + } d_add(dentry, inode); if (dir == dir->i_sb->s_root->d_inode) From ef581a742874ebc4c28d24b374c78b762144ebdc Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Wed, 23 Jul 2008 21:30:13 -0700 Subject: [PATCH 326/857] autofs4: fix symlink name allocation The length of the symlink name has been moved but it needs to be set before allocating space for it in the dentry info struct. This corrects a mistake in a recent patch. Signed-off-by: Ian Kent Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/root.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index dbb70d5a4882..324290c68275 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -775,6 +775,7 @@ static int autofs4_dir_symlink(struct inode *dir, list_del_init(&ino->active); spin_unlock(&sbi->lookup_lock); + ino->size = strlen(symname); cp = kmalloc(ino->size + 1, GFP_KERNEL); if (!cp) { if (!dentry->d_fsdata) @@ -806,7 +807,6 @@ static int autofs4_dir_symlink(struct inode *dir, atomic_inc(&p_ino->count); ino->inode = inode; - ino->size = strlen(symname); ino->u.symlink = cp; dir->i_mtime = CURRENT_TIME; From c432c2586a0811c7d0030d78f0993568bc889a6f Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Wed, 23 Jul 2008 21:30:14 -0700 Subject: [PATCH 327/857] autofs4: don't release directory mutex if called in oz_mode Since we now delay hashing of dentrys until the ->mkdir() call, droping and re-taking the directory mutex within the ->lookup() function when we are being called by user space is not needed. This can lead to a race when other processes are attempting to access the same directory during mount point directory creation. In this case we need to hang onto the mutex to ensure we don't get user processes trying to create a mount request for a newly created dentry after the mount point entry has already been created. This ensures that when we need to check a dentry passed to autofs4_wait(), if it is hashed, it is always the mount point dentry and not a new dentry created by another lookup during directory creation. Signed-off-by: Ian Kent Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/root.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index 324290c68275..1e901e5ea013 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -686,12 +686,11 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s spin_lock(&dentry->d_lock); dentry->d_flags |= DCACHE_AUTOFS_PENDING; spin_unlock(&dentry->d_lock); - } - - if (dentry->d_op && dentry->d_op->d_revalidate) { - mutex_unlock(&dir->i_mutex); - (dentry->d_op->d_revalidate)(dentry, nd); - mutex_lock(&dir->i_mutex); + if (dentry->d_op && dentry->d_op->d_revalidate) { + mutex_unlock(&dir->i_mutex); + (dentry->d_op->d_revalidate)(dentry, nd); + mutex_lock(&dir->i_mutex); + } } /* From 6d5cb926fa0162b1e62f37c117cc7ce763cfcbb9 Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Wed, 23 Jul 2008 21:30:15 -0700 Subject: [PATCH 328/857] autofs4: use lookup intent flags to trigger mounts When an open(2) call is made on an autofs mount point directory that already exists and the O_DIRECTORY flag is not used the needed mount callback to the daemon is not done. This leads to the path walk continuing resulting in a callback to the daemon with an incorrect key. open(2) is called without O_DIRECTORY by the "find" utility but this should be handled properly anyway. This happens because autofs needs to use the lookup flags to decide when to callback to the daemon to perform a mount to prevent mount storms. For example, an autofs indirect mount map that has the "browse" option will have the mount point directories are pre-created and the stat(2) call made by a color ls against each directory will cause all these directories to be mounted. It is unfortunate we need to resort to this but mount maps can be quite large. Additionally, if a user manually umounts an autofs indirect mount the directory isn't removed which also leads to this situation. To resolve this autofs needs to use the lookup intent flags to enable it to make this decision. This patch adds this check and triggers a call back if any of the lookup intent flags are set as all these calls warrant a mount attempt be requested. I know that external VFS code which uses the lookup flags is something that the VFS would like to eliminate but I have no choice as I can't see any other way to do this. A VFS dentry or inode operation callback which returns the lookup "type" (requires a definition) would be sufficient. But this change is needed now and I'm not aware of the form that coming VFS changes will take so I'm not willing to propose anything along these lines. If anyone can provide an alternate method I would be happy to use it. [akpm@linux-foundation.org: fix build for concurrent VFS changes] Signed-off-by: Ian Kent Cc: Al Viro Cc: Jeff Moyer Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/root.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index 1e901e5ea013..87352654ff4e 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -31,6 +31,9 @@ static int autofs4_root_readdir(struct file * filp, void * dirent, filldir_t fil static struct dentry *autofs4_lookup(struct inode *,struct dentry *, struct nameidata *); static void *autofs4_follow_link(struct dentry *, struct nameidata *); +#define TRIGGER_FLAGS (LOOKUP_CONTINUE | LOOKUP_DIRECTORY) +#define TRIGGER_INTENTS (LOOKUP_OPEN | LOOKUP_CREATE) + const struct file_operations autofs4_root_operations = { .open = dcache_dir_open, .release = dcache_dir_close, @@ -291,7 +294,7 @@ static int try_to_fill_dentry(struct dentry *dentry, int flags) return status; } /* Trigger mount for path component or follow link */ - } else if (flags & (LOOKUP_CONTINUE | LOOKUP_DIRECTORY) || + } else if (flags & (TRIGGER_FLAGS | TRIGGER_INTENTS) || current->link_count) { DPRINTK("waiting for mount name=%.*s", dentry->d_name.len, dentry->d_name.name); @@ -336,7 +339,7 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd) nd->flags); /* If it's our master or we shouldn't trigger a mount we're done */ - lookup_type = nd->flags & (LOOKUP_CONTINUE | LOOKUP_DIRECTORY); + lookup_type = nd->flags & (TRIGGER_FLAGS | TRIGGER_INTENTS); if (oz_mode || !lookup_type) goto done; From 70b52a0a5005ce6a0ceec56e97222437a0ba7506 Mon Sep 17 00:00:00 2001 From: Jeff Moyer Date: Wed, 23 Jul 2008 21:30:16 -0700 Subject: [PATCH 329/857] autofs4: use struct qstr in waitq.c The autofs_wait_queue already contains all of the fields of the struct qstr, so change it into a qstr. This patch, from Jeff Moyer, has been modified a liitle by myself. Signed-off-by: Jeff Moyer Signed-off-by: Ian Kent Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/autofs_i.h | 4 +- fs/autofs4/waitq.c | 86 ++++++++++++++++++++++--------------------- 2 files changed, 46 insertions(+), 44 deletions(-) diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h index 2dce2334737d..da8882ff31e6 100644 --- a/fs/autofs4/autofs_i.h +++ b/fs/autofs4/autofs_i.h @@ -75,9 +75,7 @@ struct autofs_wait_queue { struct autofs_wait_queue *next; autofs_wqt_t wait_queue_token; /* We use the following to see what we are waiting for */ - unsigned int hash; - unsigned int len; - char *name; + struct qstr name; u32 dev; u64 ino; uid_t uid; diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c index 75e5955c3f6d..5208cfb1df4e 100644 --- a/fs/autofs4/waitq.c +++ b/fs/autofs4/waitq.c @@ -36,8 +36,10 @@ void autofs4_catatonic_mode(struct autofs_sb_info *sbi) while (wq) { nwq = wq->next; wq->status = -ENOENT; /* Magic is gone - report failure */ - kfree(wq->name); - wq->name = NULL; + if (wq->name.name) { + kfree(wq->name.name); + wq->name.name = NULL; + } wake_up_interruptible(&wq->queue); wq = nwq; } @@ -92,7 +94,7 @@ static void autofs4_notify_daemon(struct autofs_sb_info *sbi, size_t pktsz; DPRINTK("wait id = 0x%08lx, name = %.*s, type=%d", - wq->wait_queue_token, wq->len, wq->name, type); + wq->wait_queue_token, wq->name.len, wq->name.name, type); memset(&pkt,0,sizeof pkt); /* For security reasons */ @@ -107,9 +109,9 @@ static void autofs4_notify_daemon(struct autofs_sb_info *sbi, pktsz = sizeof(*mp); mp->wait_queue_token = wq->wait_queue_token; - mp->len = wq->len; - memcpy(mp->name, wq->name, wq->len); - mp->name[wq->len] = '\0'; + mp->len = wq->name.len; + memcpy(mp->name, wq->name.name, wq->name.len); + mp->name[wq->name.len] = '\0'; break; } case autofs_ptype_expire_multi: @@ -119,9 +121,9 @@ static void autofs4_notify_daemon(struct autofs_sb_info *sbi, pktsz = sizeof(*ep); ep->wait_queue_token = wq->wait_queue_token; - ep->len = wq->len; - memcpy(ep->name, wq->name, wq->len); - ep->name[wq->len] = '\0'; + ep->len = wq->name.len; + memcpy(ep->name, wq->name.name, wq->name.len); + ep->name[wq->name.len] = '\0'; break; } /* @@ -138,9 +140,9 @@ static void autofs4_notify_daemon(struct autofs_sb_info *sbi, pktsz = sizeof(*packet); packet->wait_queue_token = wq->wait_queue_token; - packet->len = wq->len; - memcpy(packet->name, wq->name, wq->len); - packet->name[wq->len] = '\0'; + packet->len = wq->name.len; + memcpy(packet->name, wq->name.name, wq->name.len); + packet->name[wq->name.len] = '\0'; packet->dev = wq->dev; packet->ino = wq->ino; packet->uid = wq->uid; @@ -191,15 +193,15 @@ static int autofs4_getpath(struct autofs_sb_info *sbi, } static struct autofs_wait_queue * -autofs4_find_wait(struct autofs_sb_info *sbi, - char *name, unsigned int hash, unsigned int len) +autofs4_find_wait(struct autofs_sb_info *sbi, struct qstr *qstr) { struct autofs_wait_queue *wq; for (wq = sbi->queues; wq; wq = wq->next) { - if (wq->hash == hash && - wq->len == len && - wq->name && !memcmp(wq->name, name, len)) + if (wq->name.hash == qstr->hash && + wq->name.len == qstr->len && + wq->name.name && + !memcmp(wq->name.name, qstr->name, qstr->len)) break; } return wq; @@ -210,9 +212,8 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, { struct autofs_info *ino; struct autofs_wait_queue *wq; + struct qstr qstr; char *name; - unsigned int len = 0; - unsigned int hash = 0; int status, type; /* In catatonic mode, we don't wait for nobody */ @@ -225,22 +226,23 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, /* If this is a direct mount request create a dummy name */ if (IS_ROOT(dentry) && (sbi->type & AUTOFS_TYPE_DIRECT)) - len = sprintf(name, "%p", dentry); + qstr.len = sprintf(name, "%p", dentry); else { - len = autofs4_getpath(sbi, dentry, &name); - if (!len) { + qstr.len = autofs4_getpath(sbi, dentry, &name); + if (!qstr.len) { kfree(name); return -ENOENT; } } - hash = full_name_hash(name, len); + qstr.name = name; + qstr.hash = full_name_hash(name, qstr.len); if (mutex_lock_interruptible(&sbi->wq_mutex)) { - kfree(name); + kfree(qstr.name); return -EINTR; } - wq = autofs4_find_wait(sbi, name, hash, len); + wq = autofs4_find_wait(sbi, &qstr); ino = autofs4_dentry_ino(dentry); if (!wq && ino && notify == NFY_NONE) { /* @@ -254,10 +256,10 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, mutex_unlock(&sbi->wq_mutex); schedule_timeout_interruptible(HZ/10); if (mutex_lock_interruptible(&sbi->wq_mutex)) { - kfree(name); + kfree(qstr.name); return -EINTR; } - wq = autofs4_find_wait(sbi, name, hash, len); + wq = autofs4_find_wait(sbi, &qstr); if (wq) break; } @@ -268,7 +270,7 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, * return status of the wait. */ if (!wq) { - kfree(name); + kfree(qstr.name); mutex_unlock(&sbi->wq_mutex); return 0; } @@ -278,7 +280,7 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, /* Create a new wait queue */ wq = kmalloc(sizeof(struct autofs_wait_queue),GFP_KERNEL); if (!wq) { - kfree(name); + kfree(qstr.name); mutex_unlock(&sbi->wq_mutex); return -ENOMEM; } @@ -289,9 +291,7 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, wq->next = sbi->queues; sbi->queues = wq; init_waitqueue_head(&wq->queue); - wq->hash = hash; - wq->name = name; - wq->len = len; + memcpy(&wq->name, &qstr, sizeof(struct qstr)); wq->dev = autofs4_get_dev(sbi); wq->ino = autofs4_get_ino(sbi); wq->uid = current->uid; @@ -319,16 +319,18 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, } DPRINTK("new wait id = 0x%08lx, name = %.*s, nfy=%d\n", - (unsigned long) wq->wait_queue_token, wq->len, wq->name, notify); + (unsigned long) wq->wait_queue_token, wq->name.len, + wq->name.name, notify); /* autofs4_notify_daemon() may block */ autofs4_notify_daemon(sbi, wq, type); } else { atomic_inc(&wq->wait_ctr); mutex_unlock(&sbi->wq_mutex); - kfree(name); + kfree(qstr.name); DPRINTK("existing wait id = 0x%08lx, name = %.*s, nfy=%d", - (unsigned long) wq->wait_queue_token, wq->len, wq->name, notify); + (unsigned long) wq->wait_queue_token, wq->name.len, + wq->name.name, notify); } /* wq->name is NULL if and only if the lock is already released */ @@ -336,11 +338,13 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, if (sbi->catatonic) { /* We might have slept, so check again for catatonic mode */ wq->status = -ENOENT; - kfree(wq->name); - wq->name = NULL; + if (wq->name.name) { + kfree(wq->name.name); + wq->name.name = NULL; + } } - if (wq->name) { + if (wq->name.name) { /* Block all but "shutdown" signals while waiting */ sigset_t oldset; unsigned long irqflags; @@ -351,7 +355,7 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, recalc_sigpending(); spin_unlock_irqrestore(¤t->sighand->siglock, irqflags); - wait_event_interruptible(wq->queue, wq->name == NULL); + wait_event_interruptible(wq->queue, wq->name.name == NULL); spin_lock_irqsave(¤t->sighand->siglock, irqflags); current->blocked = oldset; @@ -388,8 +392,8 @@ int autofs4_wait_release(struct autofs_sb_info *sbi, autofs_wqt_t wait_queue_tok *wql = wq->next; /* Unlink from chain */ mutex_unlock(&sbi->wq_mutex); - kfree(wq->name); - wq->name = NULL; /* Do not wait on this queue */ + kfree(wq->name.name); + wq->name.name = NULL; /* Do not wait on this queue */ wq->status = status; From 5a11d4d0ee1ff284271f7265929d07ea4a1168a6 Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Wed, 23 Jul 2008 21:30:17 -0700 Subject: [PATCH 330/857] autofs4: fix waitq locking The autofs4_catatonic_mode() function accesses the wait queue without any locking but can be called at any time. This could lead to a possible double free of the name field of the wait and a double fput of the daemon communication pipe or an fput of a NULL file pointer. Signed-off-by: Ian Kent Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/inode.c | 4 ++-- fs/autofs4/waitq.c | 23 ++++++++++++----------- 2 files changed, 14 insertions(+), 13 deletions(-) diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c index e3e70994ab46..7bb3e5ba0537 100644 --- a/fs/autofs4/inode.c +++ b/fs/autofs4/inode.c @@ -163,8 +163,8 @@ void autofs4_kill_sb(struct super_block *sb) if (!sbi) goto out_kill_sb; - if (!sbi->catatonic) - autofs4_catatonic_mode(sbi); /* Free wait queues, close pipe */ + /* Free wait queues, close pipe */ + autofs4_catatonic_mode(sbi); /* Clean up and release dangling references */ autofs4_force_release(sbi); diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c index 5208cfb1df4e..55aac10cf328 100644 --- a/fs/autofs4/waitq.c +++ b/fs/autofs4/waitq.c @@ -28,6 +28,12 @@ void autofs4_catatonic_mode(struct autofs_sb_info *sbi) { struct autofs_wait_queue *wq, *nwq; + mutex_lock(&sbi->wq_mutex); + if (sbi->catatonic) { + mutex_unlock(&sbi->wq_mutex); + return; + } + DPRINTK("entering catatonic mode"); sbi->catatonic = 1; @@ -45,6 +51,8 @@ void autofs4_catatonic_mode(struct autofs_sb_info *sbi) } fput(sbi->pipe); /* Close the pipe */ sbi->pipe = NULL; + sbi->pipefd = -1; + mutex_unlock(&sbi->wq_mutex); } static int autofs4_write(struct file *file, const void *addr, int bytes) @@ -333,17 +341,10 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, wq->name.name, notify); } - /* wq->name is NULL if and only if the lock is already released */ - - if (sbi->catatonic) { - /* We might have slept, so check again for catatonic mode */ - wq->status = -ENOENT; - if (wq->name.name) { - kfree(wq->name.name); - wq->name.name = NULL; - } - } - + /* + * wq->name.name is NULL iff the lock is already released + * or the mount has been made catatonic. + */ if (wq->name.name) { /* Block all but "shutdown" signals while waiting */ sigset_t oldset; From a1362fe92f1bde687b3a9e93d6b8d105d0a84f74 Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Wed, 23 Jul 2008 21:30:19 -0700 Subject: [PATCH 331/857] autofs4: fix pending mount race Close a race between a pending mount that is about to finish and a new lookup for the same directory. Process P1 triggers a mount of directory foo. It sets DCACHE_AUTOFS_PENDING in the ->lookup routine, creates a waitq entry for 'foo', and calls out to the daemon to perform the mount. The autofs daemon will then create the directory 'foo', using a new dentry that will be hashed in the dcache. Before the mount completes, another process, P2, tries to walk into the 'foo' directory. The vfs path walking code finds an entry for 'foo' and calls the revalidate method. Revalidate finds that the entry is not PENDING (because PENDING was never set on the dentry created by the mkdir), but it does find the directory is empty. Revalidate calls try_to_fill_dentry, which sets the PENDING flag and then calls into the autofs4 wait code to trigger or wait for a mount of 'foo'. The wait code finds the entry for 'foo' and goes to sleep waiting for the completion of the mount. Yet another process, P3, tries to walk into the 'foo' directory. This process again finds a dentry in the dcache for 'foo', and calls into the autofs revalidate code. The revalidate code finds that the PENDING flag is set, and so calls try_to_fill_dentry. a) try_to_fill_dentry sets the PENDING flag redundantly for this dentry, then calls into the autofs4 wait code. b) the autofs4 wait code takes the waitq mutex and searches for an entry for 'foo' Between a and b, P1 is woken up because the mount completed. P1 takes the wait queue mutex, clears the PENDING flag from the dentry, and removes the waitqueue entry for 'foo' from the list. When it releases the waitq mutex, P3 (eventually) acquires it. At this time, it looks for an existing waitq for 'foo', finds none, and so creates a new one and calls out to the daemon to mount the 'foo' directory. Now, the reason that three processes are required to trigger this race is that, because the PENDING flag is not set on the dentry created by mkdir, the window for the race would be way to slim for it to ever occur. Basically, between the testing of d_mountpoint(dentry) and the taking of the waitq mutex, the mount would have to complete and the daemon would have to be woken up, and that in turn would have to wake up P1. This is simply impossible. Add the third process, though, and it becomes slightly more likely. Signed-off-by: Jeff Moyer Signed-off-by: Ian Kent Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/waitq.c | 135 ++++++++++++++++++++++++++++++++------------- 1 file changed, 97 insertions(+), 38 deletions(-) diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c index 55aac10cf328..cd3b2a671696 100644 --- a/fs/autofs4/waitq.c +++ b/fs/autofs4/waitq.c @@ -215,19 +215,106 @@ autofs4_find_wait(struct autofs_sb_info *sbi, struct qstr *qstr) return wq; } +/* + * Check if we have a valid request. + * Returns + * 1 if the request should continue. + * In this case we can return an autofs_wait_queue entry if one is + * found or NULL to idicate a new wait needs to be created. + * 0 or a negative errno if the request shouldn't continue. + */ +static int validate_request(struct autofs_wait_queue **wait, + struct autofs_sb_info *sbi, + struct qstr *qstr, + struct dentry*dentry, enum autofs_notify notify) +{ + struct autofs_wait_queue *wq; + struct autofs_info *ino; + + /* Wait in progress, continue; */ + wq = autofs4_find_wait(sbi, qstr); + if (wq) { + *wait = wq; + return 1; + } + + *wait = NULL; + + /* If we don't yet have any info this is a new request */ + ino = autofs4_dentry_ino(dentry); + if (!ino) + return 1; + + /* + * If we've been asked to wait on an existing expire (NFY_NONE) + * but there is no wait in the queue ... + */ + if (notify == NFY_NONE) { + /* + * Either we've betean the pending expire to post it's + * wait or it finished while we waited on the mutex. + * So we need to wait till either, the wait appears + * or the expire finishes. + */ + + while (ino->flags & AUTOFS_INF_EXPIRING) { + mutex_unlock(&sbi->wq_mutex); + schedule_timeout_interruptible(HZ/10); + if (mutex_lock_interruptible(&sbi->wq_mutex)) + return -EINTR; + + wq = autofs4_find_wait(sbi, qstr); + if (wq) { + *wait = wq; + return 1; + } + } + + /* + * Not ideal but the status has already gone. Of the two + * cases where we wait on NFY_NONE neither depend on the + * return status of the wait. + */ + return 0; + } + + /* + * If we've been asked to trigger a mount and the request + * completed while we waited on the mutex ... + */ + if (notify == NFY_MOUNT) { + /* + * If the dentry isn't hashed just go ahead and try the + * mount again with a new wait (not much else we can do). + */ + if (!d_unhashed(dentry)) { + /* + * But if the dentry is hashed, that means that we + * got here through the revalidate path. Thus, we + * need to check if the dentry has been mounted + * while we waited on the wq_mutex. If it has, + * simply return success. + */ + if (d_mountpoint(dentry)) + return 0; + } + } + + return 1; +} + int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, enum autofs_notify notify) { - struct autofs_info *ino; struct autofs_wait_queue *wq; struct qstr qstr; char *name; - int status, type; + int status, ret, type; /* In catatonic mode, we don't wait for nobody */ if (sbi->catatonic) return -ENOENT; - + name = kmalloc(NAME_MAX + 1, GFP_KERNEL); if (!name) return -ENOMEM; @@ -245,43 +332,15 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, qstr.name = name; qstr.hash = full_name_hash(name, qstr.len); - if (mutex_lock_interruptible(&sbi->wq_mutex)) { - kfree(qstr.name); + if (mutex_lock_interruptible(&sbi->wq_mutex)) return -EINTR; - } - wq = autofs4_find_wait(sbi, &qstr); - ino = autofs4_dentry_ino(dentry); - if (!wq && ino && notify == NFY_NONE) { - /* - * Either we've betean the pending expire to post it's - * wait or it finished while we waited on the mutex. - * So we need to wait till either, the wait appears - * or the expire finishes. - */ - - while (ino->flags & AUTOFS_INF_EXPIRING) { + ret = validate_request(&wq, sbi, &qstr, dentry, notify); + if (ret <= 0) { + if (ret == 0) mutex_unlock(&sbi->wq_mutex); - schedule_timeout_interruptible(HZ/10); - if (mutex_lock_interruptible(&sbi->wq_mutex)) { - kfree(qstr.name); - return -EINTR; - } - wq = autofs4_find_wait(sbi, &qstr); - if (wq) - break; - } - - /* - * Not ideal but the status has already gone. Of the two - * cases where we wait on NFY_NONE neither depend on the - * return status of the wait. - */ - if (!wq) { - kfree(qstr.name); - mutex_unlock(&sbi->wq_mutex); - return 0; - } + kfree(qstr.name); + return ret; } if (!wq) { @@ -392,9 +451,9 @@ int autofs4_wait_release(struct autofs_sb_info *sbi, autofs_wqt_t wait_queue_tok } *wql = wq->next; /* Unlink from chain */ - mutex_unlock(&sbi->wq_mutex); kfree(wq->name.name); wq->name.name = NULL; /* Do not wait on this queue */ + mutex_unlock(&sbi->wq_mutex); wq->status = status; From f4c7da02615bebcaf89f15a8d055922f515160b8 Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Wed, 23 Jul 2008 21:30:19 -0700 Subject: [PATCH 332/857] autofs4: add missing kfree It see that the patch tittled "autofs4 - fix pending mount race" is missing a change that I had recently made. It's missing a kfree for the case mutex_lock_interruptible() fails to aquire the wait queue mutex. Signed-off-by: Ian Kent Cc: Jeff Moyer Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/waitq.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c index cd3b2a671696..1132cc2a0310 100644 --- a/fs/autofs4/waitq.c +++ b/fs/autofs4/waitq.c @@ -332,8 +332,10 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, qstr.name = name; qstr.hash = full_name_hash(name, qstr.len); - if (mutex_lock_interruptible(&sbi->wq_mutex)) + if (mutex_lock_interruptible(&sbi->wq_mutex)) { + kfree(qstr.name); return -EINTR; + } ret = validate_request(&wq, sbi, &qstr, dentry, notify); if (ret <= 0) { From e64be33ccaceaca67c84237dff8805b861398eab Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Wed, 23 Jul 2008 21:30:20 -0700 Subject: [PATCH 333/857] autofs4: check kernel communication pipe is valid for write It is possible for an autofs mount to become catatonic (and for the daemon communication pipe to become NULL) after a wait has been initiallized but before the request has been sent to the daemon. We need to check for this before sending the request packet. Signed-off-by: Ian Kent Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/waitq.c | 16 ++++++++++++++-- 1 file changed, 14 insertions(+), 2 deletions(-) diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c index 1132cc2a0310..dd2914d7ad7f 100644 --- a/fs/autofs4/waitq.c +++ b/fs/autofs4/waitq.c @@ -99,6 +99,7 @@ static void autofs4_notify_daemon(struct autofs_sb_info *sbi, union autofs_packet_union v4_pkt; union autofs_v5_packet_union v5_pkt; } pkt; + struct file *pipe = NULL; size_t pktsz; DPRINTK("wait id = 0x%08lx, name = %.*s, type=%d", @@ -164,8 +165,19 @@ static void autofs4_notify_daemon(struct autofs_sb_info *sbi, return; } - if (autofs4_write(sbi->pipe, &pkt, pktsz)) - autofs4_catatonic_mode(sbi); + /* Check if we have become catatonic */ + mutex_lock(&sbi->wq_mutex); + if (!sbi->catatonic) { + pipe = sbi->pipe; + get_file(pipe); + } + mutex_unlock(&sbi->wq_mutex); + + if (pipe) { + if (autofs4_write(pipe, &pkt, pktsz)) + autofs4_catatonic_mode(sbi); + fput(pipe); + } } static int autofs4_getpath(struct autofs_sb_info *sbi, From 296f7bf78bc5c7a4d772aea580ce800d14040d1a Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Wed, 23 Jul 2008 21:30:21 -0700 Subject: [PATCH 334/857] autofs4: fix waitq memory leak If an autofs mount becomes catatonic before autofs4_wait_release() is called the wait queue counter will not be decremented down to zero and the entry will never be freed. There are also races decrementing the wait counter in the wait release function. To deal with this the counter needs to be updated while holding the wait queue mutex and waiters need to be woken up unconditionally when the wait is removed from the queue to ensure we eventually free the wait. Signed-off-by: Ian Kent Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/autofs_i.h | 2 +- fs/autofs4/waitq.c | 18 +++++++++--------- 2 files changed, 10 insertions(+), 10 deletions(-) diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h index da8882ff31e6..058e1800caec 100644 --- a/fs/autofs4/autofs_i.h +++ b/fs/autofs4/autofs_i.h @@ -84,7 +84,7 @@ struct autofs_wait_queue { pid_t tgid; /* This is for status reporting upon return */ int status; - atomic_t wait_ctr; + unsigned int wait_ctr; }; #define AUTOFS_SBI_MAGIC 0x6d4a556d diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c index dd2914d7ad7f..3458dbc8fff0 100644 --- a/fs/autofs4/waitq.c +++ b/fs/autofs4/waitq.c @@ -46,6 +46,7 @@ void autofs4_catatonic_mode(struct autofs_sb_info *sbi) kfree(wq->name.name); wq->name.name = NULL; } + wq->wait_ctr--; wake_up_interruptible(&wq->queue); wq = nwq; } @@ -380,7 +381,7 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, wq->pid = current->pid; wq->tgid = current->tgid; wq->status = -EINTR; /* Status return if interrupted */ - atomic_set(&wq->wait_ctr, 2); + wq->wait_ctr = 2; mutex_unlock(&sbi->wq_mutex); if (sbi->version < 5) { @@ -406,7 +407,7 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, /* autofs4_notify_daemon() may block */ autofs4_notify_daemon(sbi, wq, type); } else { - atomic_inc(&wq->wait_ctr); + wq->wait_ctr++; mutex_unlock(&sbi->wq_mutex); kfree(qstr.name); DPRINTK("existing wait id = 0x%08lx, name = %.*s, nfy=%d", @@ -442,8 +443,10 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, status = wq->status; /* Are we the last process to need status? */ - if (atomic_dec_and_test(&wq->wait_ctr)) + mutex_lock(&sbi->wq_mutex); + if (!--wq->wait_ctr) kfree(wq); + mutex_unlock(&sbi->wq_mutex); return status; } @@ -467,14 +470,11 @@ int autofs4_wait_release(struct autofs_sb_info *sbi, autofs_wqt_t wait_queue_tok *wql = wq->next; /* Unlink from chain */ kfree(wq->name.name); wq->name.name = NULL; /* Do not wait on this queue */ - mutex_unlock(&sbi->wq_mutex); - wq->status = status; - - if (atomic_dec_and_test(&wq->wait_ctr)) /* Is anyone still waiting for this guy? */ + wake_up_interruptible(&wq->queue); + if (!--wq->wait_ctr) kfree(wq); - else - wake_up_interruptible(&wq->queue); + mutex_unlock(&sbi->wq_mutex); return 0; } From eb3b176796b0e53fd26fce86847231542eb0d198 Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Wed, 23 Jul 2008 21:30:22 -0700 Subject: [PATCH 335/857] autofs4: detect invalid direct mount requests autofs v5 direct and offset mounts within an autofs filesystem are triggered by existing autofs triger mounts so the mount point dentry must be positive. If the mount point dentry is negative then the trigger doesn't exist so we can return fail immediately. Signed-off-by: Ian Kent Cc: Jeff Moyer Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/waitq.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c index 3458dbc8fff0..bcb6c5265467 100644 --- a/fs/autofs4/waitq.c +++ b/fs/autofs4/waitq.c @@ -328,6 +328,10 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, if (sbi->catatonic) return -ENOENT; + if (!dentry->d_inode && + (sbi->type & (AUTOFS_TYPE_DIRECT | AUTOFS_TYPE_OFFSET))) + return -ENOENT; + name = kmalloc(NAME_MAX + 1, GFP_KERNEL); if (!name) return -ENOMEM; From c72305b5472522299bb6f45b736080128eb1c822 Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Wed, 23 Jul 2008 21:30:23 -0700 Subject: [PATCH 336/857] autofs4: indirect dentry must almost always be positive We have been seeing mount requests comming to the automount daemon for keys of the form "/" which are lookups for invalid map keys. But we can check for this in the kernel module and return a fail immediately, without having to send a request to the daemon. It is possible to recognise these requests are invalid based on whether the request dentry is negative and its relation to the autofs file system root. For example, given the indirect multi-mount map entry: idm1 \ /mm1 :/ /mm2 :/ For a request to mount idm1, IS_ROOT((idm1)->d_parent) will be always be true and the dentry may be negative. But directories idm1/mm1 and idm1/mm2 will always be created as part of the mount request for idm1. So any mount request within idm1 itself must have a positive dentry otherwise the map key is invalid. In version 4 these multi-mount entries are all mounted and umounted as a single request and in version 5 the directories idm1/mm1 and idm1/mm2 are created and an autofs fs mounted on them to act as a mount trigger so the above is also true. This also holds true for the autofs version 4 pseudo direct mount feature. When this feature is used without the "--ghost" option automount(8) will create internal submounts as we go down the map key paths which are essentially normal indirect mounts for which the above holds. If the "--ghost" option is given the directories for map keys are created at daemon startup so valid map entries correspond to postive dentries in the autofs fs. autofs version 5 direct mount maps are similar except that the IS_ROOT check is not needed. This has been addressed in a previous patch tittled "autofs4 - detect invalid direct mount requests". For example, given the direct multi-mount map entry: /test/dm1 \ /mm1 :/ /mm2 :/ An autofs fs is mounted on /test/dm1 as a trigger mount and when a mount is triggered for /test/dm1, the multi-mount offset directories /test/dm1/mm1 and /test/dm1/mm2 are created and an autofs fs is mounted on them to act as mount triggers. So valid direct mount requests must always have a positive dentry if they correspond to a valid map entry. Signed-off-by: Ian Kent Acked-by: Jeff Moyer Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/waitq.c | 17 ++++++++++++++--- 1 file changed, 14 insertions(+), 3 deletions(-) diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c index bcb6c5265467..35216d18d8b5 100644 --- a/fs/autofs4/waitq.c +++ b/fs/autofs4/waitq.c @@ -328,9 +328,20 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, if (sbi->catatonic) return -ENOENT; - if (!dentry->d_inode && - (sbi->type & (AUTOFS_TYPE_DIRECT | AUTOFS_TYPE_OFFSET))) - return -ENOENT; + if (!dentry->d_inode) { + /* + * A wait for a negative dentry is invalid for certain + * cases. A direct or offset mount "always" has its mount + * point directory created and so the request dentry must + * be positive or the map key doesn't exist. The situation + * is very similar for indirect mounts except only dentrys + * in the root of the autofs file system may be negative. + */ + if (sbi->type & (AUTOFS_TYPE_DIRECT|AUTOFS_TYPE_OFFSET)) + return -ENOENT; + else if (!IS_ROOT(dentry->d_parent)) + return -ENOENT; + } name = kmalloc(NAME_MAX + 1, GFP_KERNEL); if (!name) From ff9cd499d6258952385cb2f12e9a3c0908fd5786 Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Wed, 23 Jul 2008 21:30:24 -0700 Subject: [PATCH 337/857] autofs4: cleanup redundant readir code The mount triggering functionality of readdir and related functions is no longer used (and is quite broken as well). The unused portions have been removed. Signed-off-by: Ian Kent Reviewed-by: Jeff Moyer Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/root.c | 149 +++++----------------------------------------- 1 file changed, 16 insertions(+), 133 deletions(-) diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index 87352654ff4e..51c873ca8e8d 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -25,8 +25,6 @@ static int autofs4_dir_rmdir(struct inode *,struct dentry *); static int autofs4_dir_mkdir(struct inode *,struct dentry *,int); static int autofs4_root_ioctl(struct inode *, struct file *,unsigned int,unsigned long); static int autofs4_dir_open(struct inode *inode, struct file *file); -static int autofs4_dir_close(struct inode *inode, struct file *file); -static int autofs4_dir_readdir(struct file * filp, void * dirent, filldir_t filldir); static int autofs4_root_readdir(struct file * filp, void * dirent, filldir_t filldir); static struct dentry *autofs4_lookup(struct inode *,struct dentry *, struct nameidata *); static void *autofs4_follow_link(struct dentry *, struct nameidata *); @@ -44,9 +42,9 @@ const struct file_operations autofs4_root_operations = { const struct file_operations autofs4_dir_operations = { .open = autofs4_dir_open, - .release = autofs4_dir_close, + .release = dcache_dir_close, .read = generic_read_dir, - .readdir = autofs4_dir_readdir, + .readdir = dcache_readdir, }; const struct inode_operations autofs4_indirect_root_inode_operations = { @@ -98,17 +96,7 @@ static int autofs4_root_readdir(struct file *file, void *dirent, static int autofs4_dir_open(struct inode *inode, struct file *file) { struct dentry *dentry = file->f_path.dentry; - struct vfsmount *mnt = file->f_path.mnt; struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb); - struct dentry *cursor; - int status; - - status = dcache_dir_open(inode, file); - if (status) - goto out; - - cursor = file->private_data; - cursor->d_fsdata = NULL; DPRINTK("file=%p dentry=%p %.*s", file, dentry, dentry->d_name.len, dentry->d_name.name); @@ -116,129 +104,24 @@ static int autofs4_dir_open(struct inode *inode, struct file *file) if (autofs4_oz_mode(sbi)) goto out; - if (autofs4_ispending(dentry)) { - DPRINTK("dentry busy"); - dcache_dir_close(inode, file); - status = -EBUSY; - goto out; - } - - status = -ENOENT; - if (!d_mountpoint(dentry) && dentry->d_op && dentry->d_op->d_revalidate) { - struct nameidata nd; - int empty, ret; - - /* In case there are stale directory dentrys from a failed mount */ - spin_lock(&dcache_lock); - empty = list_empty(&dentry->d_subdirs); + /* + * An empty directory in an autofs file system is always a + * mount point. The daemon must have failed to mount this + * during lookup so it doesn't exist. This can happen, for + * example, if user space returns an incorrect status for a + * mount request. Otherwise we're doing a readdir on the + * autofs file system so just let the libfs routines handle + * it. + */ + spin_lock(&dcache_lock); + if (!d_mountpoint(dentry) && __simple_empty(dentry)) { spin_unlock(&dcache_lock); - - if (!empty) - d_invalidate(dentry); - - nd.flags = LOOKUP_DIRECTORY; - ret = (dentry->d_op->d_revalidate)(dentry, &nd); - - if (ret <= 0) { - if (ret < 0) - status = ret; - dcache_dir_close(inode, file); - goto out; - } + return -ENOENT; } + spin_unlock(&dcache_lock); - if (d_mountpoint(dentry)) { - struct file *fp = NULL; - struct path fp_path = { .dentry = dentry, .mnt = mnt }; - - path_get(&fp_path); - - if (!autofs4_follow_mount(&fp_path.mnt, &fp_path.dentry)) { - path_put(&fp_path); - dcache_dir_close(inode, file); - goto out; - } - - fp = dentry_open(fp_path.dentry, fp_path.mnt, file->f_flags); - status = PTR_ERR(fp); - if (IS_ERR(fp)) { - dcache_dir_close(inode, file); - goto out; - } - cursor->d_fsdata = fp; - } - return 0; out: - return status; -} - -static int autofs4_dir_close(struct inode *inode, struct file *file) -{ - struct dentry *dentry = file->f_path.dentry; - struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb); - struct dentry *cursor = file->private_data; - int status = 0; - - DPRINTK("file=%p dentry=%p %.*s", - file, dentry, dentry->d_name.len, dentry->d_name.name); - - if (autofs4_oz_mode(sbi)) - goto out; - - if (autofs4_ispending(dentry)) { - DPRINTK("dentry busy"); - status = -EBUSY; - goto out; - } - - if (d_mountpoint(dentry)) { - struct file *fp = cursor->d_fsdata; - if (!fp) { - status = -ENOENT; - goto out; - } - filp_close(fp, current->files); - } -out: - dcache_dir_close(inode, file); - return status; -} - -static int autofs4_dir_readdir(struct file *file, void *dirent, filldir_t filldir) -{ - struct dentry *dentry = file->f_path.dentry; - struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb); - struct dentry *cursor = file->private_data; - int status; - - DPRINTK("file=%p dentry=%p %.*s", - file, dentry, dentry->d_name.len, dentry->d_name.name); - - if (autofs4_oz_mode(sbi)) - goto out; - - if (autofs4_ispending(dentry)) { - DPRINTK("dentry busy"); - return -EBUSY; - } - - if (d_mountpoint(dentry)) { - struct file *fp = cursor->d_fsdata; - - if (!fp) - return -ENOENT; - - if (!fp->f_op || !fp->f_op->readdir) - goto out; - - status = vfs_readdir(fp, filldir, dirent); - file->f_pos = fp->f_pos; - if (status) - autofs4_copy_atime(file, fp); - return status; - } -out: - return dcache_readdir(file, dirent, filldir); + return dcache_dir_open(inode, file); } static int try_to_fill_dentry(struct dentry *dentry, int flags) From 26e81b3142f1ba497d4cd0365c13661684b784ce Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Wed, 23 Jul 2008 21:30:25 -0700 Subject: [PATCH 338/857] autofs4: fix pending checks There are two cases for which a dentry that has a pending mount request does not wait for completion. One is via autofs4_revalidate() and the other via autofs4_follow_link(). In revalidate, after the mount point directory is created, but before the mount is done, the check in try_to_fill_dentry() can can fail to send the dentry to the wait queue since the dentry is positive and the lookup flags may contain only LOOKUP_FOLLOW. Although we don't trigger a mount for the LOOKUP_FOLLOW flag, if ther's one pending we might as well wait and use the mounted dentry for the lookup. In autofs4_follow_link() the dentry is not checked to see if it is pending so it may fail to call try_to_fill_dentry() and not wait for mount completion. A dentry that is pending must always be sent to the wait queue. Signed-off-by: Ian Kent Reviewed-by: Jeff Moyer Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/root.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index 51c873ca8e8d..61d1dca16884 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -177,7 +177,8 @@ static int try_to_fill_dentry(struct dentry *dentry, int flags) return status; } /* Trigger mount for path component or follow link */ - } else if (flags & (TRIGGER_FLAGS | TRIGGER_INTENTS) || + } else if (dentry->d_flags & DCACHE_AUTOFS_PENDING || + flags & (TRIGGER_FLAGS | TRIGGER_INTENTS) || current->link_count) { DPRINTK("waiting for mount name=%.*s", dentry->d_name.len, dentry->d_name.name); @@ -223,7 +224,8 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd) /* If it's our master or we shouldn't trigger a mount we're done */ lookup_type = nd->flags & (TRIGGER_FLAGS | TRIGGER_INTENTS); - if (oz_mode || !lookup_type) + if (oz_mode || + !(lookup_type || dentry->d_flags & DCACHE_AUTOFS_PENDING)) goto done; /* If an expire request is pending wait for it. */ @@ -242,7 +244,8 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd) * don't try to mount it again. */ spin_lock(&dcache_lock); - if (!d_mountpoint(dentry) && __simple_empty(dentry)) { + if (dentry->d_flags & DCACHE_AUTOFS_PENDING || + (!d_mountpoint(dentry) && __simple_empty(dentry))) { spin_unlock(&dcache_lock); status = try_to_fill_dentry(dentry, 0); From 97e7449a7ad883bf9f516fc970778d75999c7843 Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Wed, 23 Jul 2008 21:30:26 -0700 Subject: [PATCH 339/857] autofs4: fix indirect mount pending expire race The selection of a dentry for expiration and the setting of the AUTOFS_INF_EXPIRING flag isn't done atomically which can lead to lookups walking into an expiring mount. What happens is that an expire is initiated by the daemon and a dentry is selected for expire but, since there is no lock held between the selection and setting of the expiring flag, a process may find the flag clear and continue walking into the mount tree at the same time the daemon attempts the expire it. Signed-off-by: Ian Kent Reviewed-by: Jeff Moyer Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/autofs_i.h | 10 +++------- fs/autofs4/expire.c | 46 ++++++++++++++++++++++++++++++++----------- fs/autofs4/root.c | 32 +++++++++++++++++++++++++----- 3 files changed, 65 insertions(+), 23 deletions(-) diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h index 058e1800caec..5d90ed3b4b43 100644 --- a/fs/autofs4/autofs_i.h +++ b/fs/autofs4/autofs_i.h @@ -138,18 +138,14 @@ static inline int autofs4_oz_mode(struct autofs_sb_info *sbi) { static inline int autofs4_ispending(struct dentry *dentry) { struct autofs_info *inf = autofs4_dentry_ino(dentry); - int pending = 0; if (dentry->d_flags & DCACHE_AUTOFS_PENDING) return 1; - if (inf) { - spin_lock(&inf->sbi->fs_lock); - pending = inf->flags & AUTOFS_INF_EXPIRING; - spin_unlock(&inf->sbi->fs_lock); - } + if (inf->flags & AUTOFS_INF_EXPIRING) + return 1; - return pending; + return 0; } static inline void autofs4_copy_atime(struct file *src, struct file *dst) diff --git a/fs/autofs4/expire.c b/fs/autofs4/expire.c index 894fee54d4d8..19f5bea2704f 100644 --- a/fs/autofs4/expire.c +++ b/fs/autofs4/expire.c @@ -292,6 +292,8 @@ static struct dentry *autofs4_expire_indirect(struct super_block *sb, struct list_head *next; int do_now = how & AUTOFS_EXP_IMMEDIATE; int exp_leaves = how & AUTOFS_EXP_LEAVES; + struct autofs_info *ino; + unsigned int ino_count; if (!root) return NULL; @@ -316,6 +318,9 @@ static struct dentry *autofs4_expire_indirect(struct super_block *sb, dentry = dget(dentry); spin_unlock(&dcache_lock); + spin_lock(&sbi->fs_lock); + ino = autofs4_dentry_ino(dentry); + /* * Case 1: (i) indirect mount or top level pseudo direct mount * (autofs-4.1). @@ -326,6 +331,11 @@ static struct dentry *autofs4_expire_indirect(struct super_block *sb, DPRINTK("checking mountpoint %p %.*s", dentry, (int)dentry->d_name.len, dentry->d_name.name); + /* Path walk currently on this dentry? */ + ino_count = atomic_read(&ino->count) + 2; + if (atomic_read(&dentry->d_count) > ino_count) + goto next; + /* Can we umount this guy */ if (autofs4_mount_busy(mnt, dentry)) goto next; @@ -343,23 +353,25 @@ static struct dentry *autofs4_expire_indirect(struct super_block *sb, /* Case 2: tree mount, expire iff entire tree is not busy */ if (!exp_leaves) { - /* Lock the tree as we must expire as a whole */ - spin_lock(&sbi->fs_lock); - if (!autofs4_tree_busy(mnt, dentry, timeout, do_now)) { - struct autofs_info *inf = autofs4_dentry_ino(dentry); + /* Path walk currently on this dentry? */ + ino_count = atomic_read(&ino->count) + 1; + if (atomic_read(&dentry->d_count) > ino_count) + goto next; - /* Set this flag early to catch sys_chdir and the like */ - inf->flags |= AUTOFS_INF_EXPIRING; - spin_unlock(&sbi->fs_lock); + if (!autofs4_tree_busy(mnt, dentry, timeout, do_now)) { expired = dentry; goto found; } - spin_unlock(&sbi->fs_lock); /* * Case 3: pseudo direct mount, expire individual leaves * (autofs-4.1). */ } else { + /* Path walk currently on this dentry? */ + ino_count = atomic_read(&ino->count) + 1; + if (atomic_read(&dentry->d_count) > ino_count) + goto next; + expired = autofs4_check_leaves(mnt, dentry, timeout, do_now); if (expired) { dput(dentry); @@ -367,6 +379,7 @@ static struct dentry *autofs4_expire_indirect(struct super_block *sb, } } next: + spin_unlock(&sbi->fs_lock); dput(dentry); spin_lock(&dcache_lock); next = next->next; @@ -377,6 +390,9 @@ static struct dentry *autofs4_expire_indirect(struct super_block *sb, found: DPRINTK("returning %p %.*s", expired, (int)expired->d_name.len, expired->d_name.name); + ino = autofs4_dentry_ino(expired); + ino->flags |= AUTOFS_INF_EXPIRING; + spin_unlock(&sbi->fs_lock); spin_lock(&dcache_lock); list_move(&expired->d_parent->d_subdirs, &expired->d_u.d_child); spin_unlock(&dcache_lock); @@ -390,7 +406,9 @@ int autofs4_expire_run(struct super_block *sb, struct autofs_packet_expire __user *pkt_p) { struct autofs_packet_expire pkt; + struct autofs_info *ino; struct dentry *dentry; + int ret = 0; memset(&pkt,0,sizeof pkt); @@ -406,9 +424,14 @@ int autofs4_expire_run(struct super_block *sb, dput(dentry); if ( copy_to_user(pkt_p, &pkt, sizeof(struct autofs_packet_expire)) ) - return -EFAULT; + ret = -EFAULT; - return 0; + spin_lock(&sbi->fs_lock); + ino = autofs4_dentry_ino(dentry); + ino->flags &= ~AUTOFS_INF_EXPIRING; + spin_unlock(&sbi->fs_lock); + + return ret; } /* Call repeatedly until it returns -EAGAIN, meaning there's nothing @@ -433,9 +456,10 @@ int autofs4_expire_multi(struct super_block *sb, struct vfsmount *mnt, /* This is synchronous because it makes the daemon a little easier */ - ino->flags |= AUTOFS_INF_EXPIRING; ret = autofs4_wait(sbi, dentry, NFY_EXPIRE); + spin_lock(&sbi->fs_lock); ino->flags &= ~AUTOFS_INF_EXPIRING; + spin_unlock(&sbi->fs_lock); dput(dentry); } diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index 61d1dca16884..1c2579de1f2e 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -133,7 +133,10 @@ static int try_to_fill_dentry(struct dentry *dentry, int flags) /* Block on any pending expiry here; invalidate the dentry when expiration is done to trigger mount request with a new dentry */ - if (ino && (ino->flags & AUTOFS_INF_EXPIRING)) { + spin_lock(&sbi->fs_lock); + if (ino->flags & AUTOFS_INF_EXPIRING) { + spin_unlock(&sbi->fs_lock); + DPRINTK("waiting for expire %p name=%.*s", dentry, dentry->d_name.len, dentry->d_name.name); @@ -149,8 +152,11 @@ static int try_to_fill_dentry(struct dentry *dentry, int flags) status = d_invalidate(dentry); if (status != -EBUSY) return -EAGAIN; - } + goto cont; + } + spin_unlock(&sbi->fs_lock); +cont: DPRINTK("dentry=%p %.*s ino=%p", dentry, dentry->d_name.len, dentry->d_name.name, dentry->d_inode); @@ -229,15 +235,21 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd) goto done; /* If an expire request is pending wait for it. */ - if (ino && (ino->flags & AUTOFS_INF_EXPIRING)) { + spin_lock(&sbi->fs_lock); + if (ino->flags & AUTOFS_INF_EXPIRING) { + spin_unlock(&sbi->fs_lock); + DPRINTK("waiting for active request %p name=%.*s", dentry, dentry->d_name.len, dentry->d_name.name); status = autofs4_wait(sbi, dentry, NFY_NONE); DPRINTK("request done status=%d", status); - } + goto cont; + } + spin_unlock(&sbi->fs_lock); +cont: /* * If the dentry contains directories then it is an * autofs multi-mount with no root mount offset. So @@ -292,8 +304,11 @@ static int autofs4_revalidate(struct dentry *dentry, struct nameidata *nd) int status = 1; /* Pending dentry */ + spin_lock(&sbi->fs_lock); if (autofs4_ispending(dentry)) { /* The daemon never causes a mount to trigger */ + spin_unlock(&sbi->fs_lock); + if (oz_mode) return 1; @@ -316,6 +331,7 @@ static int autofs4_revalidate(struct dentry *dentry, struct nameidata *nd) return status; } + spin_unlock(&sbi->fs_lock); /* Negative dentry.. invalidate if "old" */ if (dentry->d_inode == NULL) @@ -329,6 +345,7 @@ static int autofs4_revalidate(struct dentry *dentry, struct nameidata *nd) DPRINTK("dentry=%p %.*s, emptydir", dentry, dentry->d_name.len, dentry->d_name.name); spin_unlock(&dcache_lock); + /* The daemon never causes a mount to trigger */ if (oz_mode) return 1; @@ -521,13 +538,18 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s * so it must have been successful, so just wait for it. */ ino = autofs4_dentry_ino(expiring); - while (ino && (ino->flags & AUTOFS_INF_EXPIRING)) { + spin_lock(&sbi->fs_lock); + if (ino->flags & AUTOFS_INF_EXPIRING) { + spin_unlock(&sbi->fs_lock); DPRINTK("wait for incomplete expire %p name=%.*s", expiring, expiring->d_name.len, expiring->d_name.name); autofs4_wait(sbi, expiring, NFY_NONE); DPRINTK("request completed"); + goto cont; } + spin_unlock(&sbi->fs_lock); +cont: spin_lock(&sbi->lookup_lock); if (!list_empty(&ino->expiring)) list_del_init(&ino->expiring); From 6e60a9ab5f5d314735467752f623072f5b75157a Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Wed, 23 Jul 2008 21:30:27 -0700 Subject: [PATCH 340/857] autofs4: fix direct mount pending expire race For direct and offset type mounts that are covered by another mount we cannot check the AUTOFS_INF_EXPIRING flag during a path walk which leads to lookups walking into an expiring mount while it is being expired. For example, for the direct multi-mount map entry with a couple of offsets: /race/mm1 / :/ /om1 :/ /om2 :/ an autofs trigger mount is mounted on /race/mm1 and when accessed it is over mounted and trigger mounts made for /race/mm1/om1 and /race/mm1/om2. So it isn't possible for path walks to see the expiring flag at all and they happily walk into the file system while it is expiring. When expiring these mounts follow_down() must stop at the autofs mount and all processes must block in the ->follow_link() method (except the daemon) until the expire is complete. This is done by decrementing the d_mounted field of the autofs trigger mount root dentry until the expire is completed. In ->follow_link() all processes wait on the expire and the mount following is completed for the daemon until the expire is complete. Signed-off-by: Ian Kent Cc: Jeff Moyer Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/autofs_i.h | 3 ++ fs/autofs4/expire.c | 16 ++++++++-- fs/autofs4/root.c | 72 +++++++++++++++++++++++++++++-------------- 3 files changed, 65 insertions(+), 26 deletions(-) diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h index 5d90ed3b4b43..4b40cbc71e9b 100644 --- a/fs/autofs4/autofs_i.h +++ b/fs/autofs4/autofs_i.h @@ -52,6 +52,8 @@ struct autofs_info { int flags; + struct completion expire_complete; + struct list_head active; struct list_head expiring; @@ -69,6 +71,7 @@ struct autofs_info { }; #define AUTOFS_INF_EXPIRING (1<<0) /* dentry is in the process of expiring */ +#define AUTOFS_INF_MOUNTPOINT (1<<1) /* mountpoint status for direct expire */ struct autofs_wait_queue { wait_queue_head_t queue; diff --git a/fs/autofs4/expire.c b/fs/autofs4/expire.c index 19f5bea2704f..705b9f057fb3 100644 --- a/fs/autofs4/expire.c +++ b/fs/autofs4/expire.c @@ -259,13 +259,15 @@ static struct dentry *autofs4_expire_direct(struct super_block *sb, now = jiffies; timeout = sbi->exp_timeout; - /* Lock the tree as we must expire as a whole */ spin_lock(&sbi->fs_lock); if (!autofs4_direct_busy(mnt, root, timeout, do_now)) { struct autofs_info *ino = autofs4_dentry_ino(root); - - /* Set this flag early to catch sys_chdir and the like */ + if (d_mountpoint(root)) { + ino->flags |= AUTOFS_INF_MOUNTPOINT; + root->d_mounted--; + } ino->flags |= AUTOFS_INF_EXPIRING; + init_completion(&ino->expire_complete); spin_unlock(&sbi->fs_lock); return root; } @@ -392,6 +394,7 @@ static struct dentry *autofs4_expire_indirect(struct super_block *sb, expired, (int)expired->d_name.len, expired->d_name.name); ino = autofs4_dentry_ino(expired); ino->flags |= AUTOFS_INF_EXPIRING; + init_completion(&ino->expire_complete); spin_unlock(&sbi->fs_lock); spin_lock(&dcache_lock); list_move(&expired->d_parent->d_subdirs, &expired->d_u.d_child); @@ -429,6 +432,7 @@ int autofs4_expire_run(struct super_block *sb, spin_lock(&sbi->fs_lock); ino = autofs4_dentry_ino(dentry); ino->flags &= ~AUTOFS_INF_EXPIRING; + complete_all(&ino->expire_complete); spin_unlock(&sbi->fs_lock); return ret; @@ -457,8 +461,14 @@ int autofs4_expire_multi(struct super_block *sb, struct vfsmount *mnt, /* This is synchronous because it makes the daemon a little easier */ ret = autofs4_wait(sbi, dentry, NFY_EXPIRE); + spin_lock(&sbi->fs_lock); + if (ino->flags & AUTOFS_INF_MOUNTPOINT) { + sb->s_root->d_mounted++; + ino->flags &= ~AUTOFS_INF_MOUNTPOINT; + } ino->flags &= ~AUTOFS_INF_EXPIRING; + complete_all(&ino->expire_complete); spin_unlock(&sbi->fs_lock); dput(dentry); } diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index 1c2579de1f2e..adbd8559e870 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -141,6 +141,7 @@ static int try_to_fill_dentry(struct dentry *dentry, int flags) dentry, dentry->d_name.len, dentry->d_name.name); status = autofs4_wait(sbi, dentry, NFY_NONE); + wait_for_completion(&ino->expire_complete); DPRINTK("expire done status=%d", status); @@ -227,14 +228,32 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd) DPRINTK("dentry=%p %.*s oz_mode=%d nd->flags=%d", dentry, dentry->d_name.len, dentry->d_name.name, oz_mode, nd->flags); - - /* If it's our master or we shouldn't trigger a mount we're done */ - lookup_type = nd->flags & (TRIGGER_FLAGS | TRIGGER_INTENTS); - if (oz_mode || - !(lookup_type || dentry->d_flags & DCACHE_AUTOFS_PENDING)) + /* + * For an expire of a covered direct or offset mount we need + * to beeak out of follow_down() at the autofs mount trigger + * (d_mounted--), so we can see the expiring flag, and manage + * the blocking and following here until the expire is completed. + */ + if (oz_mode) { + spin_lock(&sbi->fs_lock); + if (ino->flags & AUTOFS_INF_EXPIRING) { + spin_unlock(&sbi->fs_lock); + /* Follow down to our covering mount. */ + if (!follow_down(&nd->path.mnt, &nd->path.dentry)) + goto done; + /* + * We shouldn't need to do this but we have no way + * of knowing what may have been done so try a follow + * just in case. + */ + autofs4_follow_mount(&nd->path.mnt, &nd->path.dentry); + goto done; + } + spin_unlock(&sbi->fs_lock); goto done; + } - /* If an expire request is pending wait for it. */ + /* If an expire request is pending everyone must wait. */ spin_lock(&sbi->fs_lock); if (ino->flags & AUTOFS_INF_EXPIRING) { spin_unlock(&sbi->fs_lock); @@ -243,6 +262,7 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd) dentry, dentry->d_name.len, dentry->d_name.name); status = autofs4_wait(sbi, dentry, NFY_NONE); + wait_for_completion(&ino->expire_complete); DPRINTK("request done status=%d", status); @@ -250,10 +270,15 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd) } spin_unlock(&sbi->fs_lock); cont: + /* We trigger a mount for almost all flags */ + lookup_type = nd->flags & (TRIGGER_FLAGS | TRIGGER_INTENTS); + if (!(lookup_type || dentry->d_flags & DCACHE_AUTOFS_PENDING)) + goto done; + /* - * If the dentry contains directories then it is an - * autofs multi-mount with no root mount offset. So - * don't try to mount it again. + * If the dentry contains directories then it is an autofs + * multi-mount with no root mount offset. So don't try to + * mount it again. */ spin_lock(&dcache_lock); if (dentry->d_flags & DCACHE_AUTOFS_PENDING || @@ -264,22 +289,22 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd) if (status) goto out_error; - /* - * The mount succeeded but if there is no root mount - * it must be an autofs multi-mount with no root offset - * so we don't need to follow the mount. - */ - if (d_mountpoint(dentry)) { - if (!autofs4_follow_mount(&nd->path.mnt, - &nd->path.dentry)) { - status = -ENOENT; - goto out_error; - } - } - - goto done; + goto follow; } spin_unlock(&dcache_lock); +follow: + /* + * If there is no root mount it must be an autofs + * multi-mount with no root offset so we don't need + * to follow it. + */ + if (d_mountpoint(dentry)) { + if (!autofs4_follow_mount(&nd->path.mnt, + &nd->path.dentry)) { + status = -ENOENT; + goto out_error; + } + } done: return NULL; @@ -545,6 +570,7 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s expiring, expiring->d_name.len, expiring->d_name.name); autofs4_wait(sbi, expiring, NFY_NONE); + wait_for_completion(&ino->expire_complete); DPRINTK("request completed"); goto cont; } From ec6e8c7d3f9073336ec7b2eed3fcda6f922087c3 Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Wed, 23 Jul 2008 21:30:28 -0700 Subject: [PATCH 341/857] autofs4: fix direct mount pending expire race - correction Appologies, somehow I seem to have sent an out dated version of this patch. Here is an additional patch that brings the patch up to date. Signed-off-by: Ian Kent Cc: Jeff Moyer Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/root.c | 10 ++-------- 1 file changed, 2 insertions(+), 8 deletions(-) diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index adbd8559e870..e062ee5a3ed5 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -241,13 +241,7 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd) /* Follow down to our covering mount. */ if (!follow_down(&nd->path.mnt, &nd->path.dentry)) goto done; - /* - * We shouldn't need to do this but we have no way - * of knowing what may have been done so try a follow - * just in case. - */ - autofs4_follow_mount(&nd->path.mnt, &nd->path.dentry); - goto done; + goto follow; } spin_unlock(&sbi->fs_lock); goto done; @@ -273,7 +267,7 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd) /* We trigger a mount for almost all flags */ lookup_type = nd->flags & (TRIGGER_FLAGS | TRIGGER_INTENTS); if (!(lookup_type || dentry->d_flags & DCACHE_AUTOFS_PENDING)) - goto done; + goto follow; /* * If the dentry contains directories then it is an autofs From 06a3598552dc3b2b30eb18bd53bbac2a901489d7 Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Wed, 23 Jul 2008 21:30:28 -0700 Subject: [PATCH 342/857] autofs4: reorganize expire pending wait function calls This patch re-orgnirzes the checking for and waiting on active expires and elininates redundant checks. Signed-off-by: Ian Kent Cc: Jeff Moyer Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/autofs_i.h | 1 + fs/autofs4/expire.c | 29 +++++++++++++++++ fs/autofs4/root.c | 75 ++++++------------------------------------- 3 files changed, 40 insertions(+), 65 deletions(-) diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h index 4b40cbc71e9b..69a2f5c92319 100644 --- a/fs/autofs4/autofs_i.h +++ b/fs/autofs4/autofs_i.h @@ -163,6 +163,7 @@ void autofs4_free_ino(struct autofs_info *); /* Expiration */ int is_autofs4_dentry(struct dentry *); +int autofs4_expire_wait(struct dentry *dentry); int autofs4_expire_run(struct super_block *, struct vfsmount *, struct autofs_sb_info *, struct autofs_packet_expire __user *); diff --git a/fs/autofs4/expire.c b/fs/autofs4/expire.c index 705b9f057fb3..cdabb796ff01 100644 --- a/fs/autofs4/expire.c +++ b/fs/autofs4/expire.c @@ -402,6 +402,35 @@ static struct dentry *autofs4_expire_indirect(struct super_block *sb, return expired; } +int autofs4_expire_wait(struct dentry *dentry) +{ + struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb); + struct autofs_info *ino = autofs4_dentry_ino(dentry); + int status; + + /* Block on any pending expire */ + spin_lock(&sbi->fs_lock); + if (ino->flags & AUTOFS_INF_EXPIRING) { + spin_unlock(&sbi->fs_lock); + + DPRINTK("waiting for expire %p name=%.*s", + dentry, dentry->d_name.len, dentry->d_name.name); + + status = autofs4_wait(sbi, dentry, NFY_NONE); + wait_for_completion(&ino->expire_complete); + + DPRINTK("expire done status=%d", status); + + if (d_unhashed(dentry)) + return -EAGAIN; + + return status; + } + spin_unlock(&sbi->fs_lock); + + return 0; +} + /* Perform an expiry operation */ int autofs4_expire_run(struct super_block *sb, struct vfsmount *mnt, diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index e062ee5a3ed5..ae22bde0bbd7 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -130,34 +130,6 @@ static int try_to_fill_dentry(struct dentry *dentry, int flags) struct autofs_info *ino = autofs4_dentry_ino(dentry); int status; - /* Block on any pending expiry here; invalidate the dentry - when expiration is done to trigger mount request with a new - dentry */ - spin_lock(&sbi->fs_lock); - if (ino->flags & AUTOFS_INF_EXPIRING) { - spin_unlock(&sbi->fs_lock); - - DPRINTK("waiting for expire %p name=%.*s", - dentry, dentry->d_name.len, dentry->d_name.name); - - status = autofs4_wait(sbi, dentry, NFY_NONE); - wait_for_completion(&ino->expire_complete); - - DPRINTK("expire done status=%d", status); - - /* - * If the directory still exists the mount request must - * continue otherwise it can't be followed at the right - * time during the walk. - */ - status = d_invalidate(dentry); - if (status != -EBUSY) - return -EAGAIN; - - goto cont; - } - spin_unlock(&sbi->fs_lock); -cont: DPRINTK("dentry=%p %.*s ino=%p", dentry, dentry->d_name.len, dentry->d_name.name, dentry->d_inode); @@ -248,22 +220,8 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd) } /* If an expire request is pending everyone must wait. */ - spin_lock(&sbi->fs_lock); - if (ino->flags & AUTOFS_INF_EXPIRING) { - spin_unlock(&sbi->fs_lock); + autofs4_expire_wait(dentry); - DPRINTK("waiting for active request %p name=%.*s", - dentry, dentry->d_name.len, dentry->d_name.name); - - status = autofs4_wait(sbi, dentry, NFY_NONE); - wait_for_completion(&ino->expire_complete); - - DPRINTK("request done status=%d", status); - - goto cont; - } - spin_unlock(&sbi->fs_lock); -cont: /* We trigger a mount for almost all flags */ lookup_type = nd->flags & (TRIGGER_FLAGS | TRIGGER_INTENTS); if (!(lookup_type || dentry->d_flags & DCACHE_AUTOFS_PENDING)) @@ -331,6 +289,14 @@ static int autofs4_revalidate(struct dentry *dentry, struct nameidata *nd) if (oz_mode) return 1; + /* + * If the directory has gone away due to an expire + * we have been called as ->d_revalidate() and so + * we need to return false and proceed to ->lookup(). + */ + if (autofs4_expire_wait(dentry) == -EAGAIN) + return 0; + /* * A zero status is success otherwise we have a * negative error code. @@ -339,15 +305,6 @@ static int autofs4_revalidate(struct dentry *dentry, struct nameidata *nd) if (status == 0) return 1; - /* - * A status of EAGAIN here means that the dentry has gone - * away while waiting for an expire to complete. If we are - * racing with expire lookup will wait for it so this must - * be a revalidate and we need to send it to lookup. - */ - if (status == -EAGAIN) - return 0; - return status; } spin_unlock(&sbi->fs_lock); @@ -557,19 +514,7 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s * so it must have been successful, so just wait for it. */ ino = autofs4_dentry_ino(expiring); - spin_lock(&sbi->fs_lock); - if (ino->flags & AUTOFS_INF_EXPIRING) { - spin_unlock(&sbi->fs_lock); - DPRINTK("wait for incomplete expire %p name=%.*s", - expiring, expiring->d_name.len, - expiring->d_name.name); - autofs4_wait(sbi, expiring, NFY_NONE); - wait_for_completion(&ino->expire_complete); - DPRINTK("request completed"); - goto cont; - } - spin_unlock(&sbi->fs_lock); -cont: + autofs4_expire_wait(expiring); spin_lock(&sbi->lookup_lock); if (!list_empty(&ino->expiring)) list_del_init(&ino->expiring); From aa55ddf340c9fa3f303ee16bbf35887e42c50304 Mon Sep 17 00:00:00 2001 From: Ian Kent Date: Wed, 23 Jul 2008 21:30:29 -0700 Subject: [PATCH 343/857] autofs4: remove unused ioctls The ioctls AUTOFS_IOC_TOGGLEREGHOST and AUTOFS_IOC_ASKREGHOST were added several years ago but what they were intended for has never been implemented (as far as I'm aware noone uses them) so remove them. Signed-off-by: Ian Kent Reviewed-by: Jeff Moyer Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/autofs4/root.c | 68 +--------------------------------------- fs/compat_ioctl.c | 2 -- include/linux/auto_fs4.h | 2 -- 3 files changed, 1 insertion(+), 71 deletions(-) diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index ae22bde0bbd7..bcfb2dc0a61b 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -25,7 +25,6 @@ static int autofs4_dir_rmdir(struct inode *,struct dentry *); static int autofs4_dir_mkdir(struct inode *,struct dentry *,int); static int autofs4_root_ioctl(struct inode *, struct file *,unsigned int,unsigned long); static int autofs4_dir_open(struct inode *inode, struct file *file); -static int autofs4_root_readdir(struct file * filp, void * dirent, filldir_t filldir); static struct dentry *autofs4_lookup(struct inode *,struct dentry *, struct nameidata *); static void *autofs4_follow_link(struct dentry *, struct nameidata *); @@ -36,7 +35,7 @@ const struct file_operations autofs4_root_operations = { .open = dcache_dir_open, .release = dcache_dir_close, .read = generic_read_dir, - .readdir = autofs4_root_readdir, + .readdir = dcache_readdir, .ioctl = autofs4_root_ioctl, }; @@ -71,28 +70,6 @@ const struct inode_operations autofs4_dir_inode_operations = { .rmdir = autofs4_dir_rmdir, }; -static int autofs4_root_readdir(struct file *file, void *dirent, - filldir_t filldir) -{ - struct autofs_sb_info *sbi = autofs4_sbi(file->f_path.dentry->d_sb); - int oz_mode = autofs4_oz_mode(sbi); - - DPRINTK("called, filp->f_pos = %lld", file->f_pos); - - /* - * Don't set reghost flag if: - * 1) f_pos is larger than zero -- we've already been here. - * 2) we haven't even enabled reghosting in the 1st place. - * 3) this is the daemon doing a readdir - */ - if (oz_mode && file->f_pos == 0 && sbi->reghost_enabled) - sbi->needs_reghost = 1; - - DPRINTK("needs_reghost = %d", sbi->needs_reghost); - - return dcache_readdir(file, dirent, filldir); -} - static int autofs4_dir_open(struct inode *inode, struct file *file) { struct dentry *dentry = file->f_path.dentry; @@ -858,44 +835,6 @@ static inline int autofs4_get_protosubver(struct autofs_sb_info *sbi, int __user return put_user(sbi->sub_version, p); } -/* - * Tells the daemon whether we need to reghost or not. Also, clears - * the reghost_needed flag. - */ -static inline int autofs4_ask_reghost(struct autofs_sb_info *sbi, int __user *p) -{ - int status; - - DPRINTK("returning %d", sbi->needs_reghost); - - status = put_user(sbi->needs_reghost, p); - if (status) - return status; - - sbi->needs_reghost = 0; - return 0; -} - -/* - * Enable / Disable reghosting ioctl() operation - */ -static inline int autofs4_toggle_reghost(struct autofs_sb_info *sbi, int __user *p) -{ - int status; - int val; - - status = get_user(val, p); - - DPRINTK("reghost = %d", val); - - if (status) - return status; - - /* turn on/off reghosting, with the val */ - sbi->reghost_enabled = val; - return 0; -} - /* * Tells the daemon whether it can umount the autofs mount. */ @@ -960,11 +899,6 @@ static int autofs4_root_ioctl(struct inode *inode, struct file *filp, case AUTOFS_IOC_SETTIMEOUT: return autofs4_get_set_timeout(sbi, p); - case AUTOFS_IOC_TOGGLEREGHOST: - return autofs4_toggle_reghost(sbi, p); - case AUTOFS_IOC_ASKREGHOST: - return autofs4_ask_reghost(sbi, p); - case AUTOFS_IOC_ASKUMOUNT: return autofs4_ask_umount(filp->f_path.mnt, p); diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c index 7b3a03c7c6a9..18e2c548161d 100644 --- a/fs/compat_ioctl.c +++ b/fs/compat_ioctl.c @@ -2297,8 +2297,6 @@ COMPATIBLE_IOCTL(AUTOFS_IOC_PROTOVER) COMPATIBLE_IOCTL(AUTOFS_IOC_EXPIRE) COMPATIBLE_IOCTL(AUTOFS_IOC_EXPIRE_MULTI) COMPATIBLE_IOCTL(AUTOFS_IOC_PROTOSUBVER) -COMPATIBLE_IOCTL(AUTOFS_IOC_ASKREGHOST) -COMPATIBLE_IOCTL(AUTOFS_IOC_TOGGLEREGHOST) COMPATIBLE_IOCTL(AUTOFS_IOC_ASKUMOUNT) /* Raw devices */ COMPATIBLE_IOCTL(RAW_SETBIND) diff --git a/include/linux/auto_fs4.h b/include/linux/auto_fs4.h index 31a29541b504..b785c6f8644d 100644 --- a/include/linux/auto_fs4.h +++ b/include/linux/auto_fs4.h @@ -98,8 +98,6 @@ union autofs_v5_packet_union { #define AUTOFS_IOC_EXPIRE_INDIRECT AUTOFS_IOC_EXPIRE_MULTI #define AUTOFS_IOC_EXPIRE_DIRECT AUTOFS_IOC_EXPIRE_MULTI #define AUTOFS_IOC_PROTOSUBVER _IOR(0x93,0x67,int) -#define AUTOFS_IOC_ASKREGHOST _IOR(0x93,0x68,int) -#define AUTOFS_IOC_TOGGLEREGHOST _IOR(0x93,0x69,int) #define AUTOFS_IOC_ASKUMOUNT _IOR(0x93,0x70,int) From 35aa64f3a117a16c466f688f52ac3847b3b572e8 Mon Sep 17 00:00:00 2001 From: "Maciej W. Rozycki" Date: Wed, 23 Jul 2008 21:30:29 -0700 Subject: [PATCH 344/857] rtc: m41t80: sort header inclusions for readability Sort the header inclusions for readability. No functional changes. Signed-off-by: Maciej W. Rozycki Cc: Alessandro Zummo Cc: Alexander Bigga Cc: David Brownell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/rtc/rtc-m41t80.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/drivers/rtc/rtc-m41t80.c b/drivers/rtc/rtc-m41t80.c index 0a19c06019be..4b260664547d 100644 --- a/drivers/rtc/rtc-m41t80.c +++ b/drivers/rtc/rtc-m41t80.c @@ -13,21 +13,21 @@ * */ -#include +#include +#include #include #include +#include +#include #include #include #include -#include -#include -#include #ifdef CONFIG_RTC_DRV_M41T80_WDT -#include -#include -#include #include #include +#include +#include +#include #endif #define M41T80_REG_SSEC 0 From 4c228db0b30fa12d65ae7461ce29ed1f4da12c5b Mon Sep 17 00:00:00 2001 From: "Maciej W. Rozycki" Date: Wed, 23 Jul 2008 21:30:32 -0700 Subject: [PATCH 345/857] rtc: m41t80: use pr_info() as appropriate Replace printk(KERN_INFO ...) calls with appropriate pr_info(...) equivalents. Signed-off-by: Maciej W. Rozycki Cc: Alessandro Zummo Cc: Alexander Bigga Cc: David Brownell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/rtc/rtc-m41t80.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/rtc/rtc-m41t80.c b/drivers/rtc/rtc-m41t80.c index 4b260664547d..24bc1689fc74 100644 --- a/drivers/rtc/rtc-m41t80.c +++ b/drivers/rtc/rtc-m41t80.c @@ -631,14 +631,12 @@ static int wdt_ioctl(struct inode *inode, struct file *file, unsigned int cmd, return -EFAULT; if (rv & WDIOS_DISABLECARD) { - printk(KERN_INFO - "rtc-m41t80: disable watchdog\n"); + pr_info("rtc-m41t80: disable watchdog\n"); wdt_disable(); } if (rv & WDIOS_ENABLECARD) { - printk(KERN_INFO - "rtc-m41t80: enable watchdog\n"); + pr_info("rtc-m41t80: enable watchdog\n"); wdt_ping(); } From 53f1b1433da7eac2607a4a0898a221a4485fd732 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Wed, 23 Jul 2008 21:30:32 -0700 Subject: [PATCH 346/857] rtc: push the BKL down into the driver ioctl method For now just wrap the main logic, but this driver is a prime candidate for someone wanting to eliminate the lock entirely [lizf@cn.fujitsu.com: fix build failure] Signed-off-by: Alan Cox Signed-off-by: Li Zefan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/rtc.c | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) diff --git a/drivers/char/rtc.c b/drivers/char/rtc.c index fa92a8af5a5a..d1569a0d0506 100644 --- a/drivers/char/rtc.c +++ b/drivers/char/rtc.c @@ -78,9 +78,10 @@ #include #include #include +#include +#include #include -#include #include #ifdef CONFIG_X86 @@ -144,8 +145,7 @@ static DEFINE_TIMER(rtc_irq_timer, rtc_dropped_irq, 0, 0); static ssize_t rtc_read(struct file *file, char __user *buf, size_t count, loff_t *ppos); -static int rtc_ioctl(struct inode *inode, struct file *file, - unsigned int cmd, unsigned long arg); +static long rtc_ioctl(struct file *file, unsigned int cmd, unsigned long arg); #ifdef RTC_IRQ static unsigned int rtc_poll(struct file *file, poll_table *wait); @@ -719,10 +719,13 @@ static int rtc_do_ioctl(unsigned int cmd, unsigned long arg, int kernel) &wtime, sizeof wtime) ? -EFAULT : 0; } -static int rtc_ioctl(struct inode *inode, struct file *file, unsigned int cmd, - unsigned long arg) +static long rtc_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { - return rtc_do_ioctl(cmd, arg, 0); + long ret; + lock_kernel(); + ret = rtc_do_ioctl(cmd, arg, 0); + unlock_kernel(); + return ret; } /* @@ -915,7 +918,7 @@ static const struct file_operations rtc_fops = { #ifdef RTC_IRQ .poll = rtc_poll, #endif - .ioctl = rtc_ioctl, + .unlocked_ioctl = rtc_ioctl, .open = rtc_open, .release = rtc_release, .fasync = rtc_fasync, From 5ad31a575157147b43fa84ef1e21471661653878 Mon Sep 17 00:00:00 2001 From: David Brownell Date: Wed, 23 Jul 2008 21:30:33 -0700 Subject: [PATCH 347/857] rtc: remove BKL for ioctl() Remove implicit use of BKL in ioctl() from the RTC framework. Instead, the rtc->ops_lock is used. That's the same lock that already protects the RTC operations when they're issued through the exported rtc_*() calls in drivers/rtc/interface.c ... making this a bugfix, not just a cleanup, since both ioctl calls and set_alarm() need to update IRQ enable flags and that implies a common lock (which RTC drivers as a rule do not provide on their own). A new comment at the declaration of "struct rtc_class_ops" summarizes current locking rules. It's not clear to me that the exceptions listed there should exist ... if not, those are pre-existing problems which can be fixed in a patch that doesn't relate to BKL removal. Signed-off-by: David Brownell Cc: Alan Cox Cc: Jonathan Corbet Acked-by: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/rtc/rtc-dev.c | 58 ++++++++++++++++++++++++++++--------------- include/linux/rtc.h | 17 +++++++++++++ 2 files changed, 55 insertions(+), 20 deletions(-) diff --git a/drivers/rtc/rtc-dev.c b/drivers/rtc/rtc-dev.c index 0114a78b7cbb..0a870b7e5c32 100644 --- a/drivers/rtc/rtc-dev.c +++ b/drivers/rtc/rtc-dev.c @@ -209,7 +209,7 @@ static unsigned int rtc_dev_poll(struct file *file, poll_table *wait) return (data != 0) ? (POLLIN | POLLRDNORM) : 0; } -static int rtc_dev_ioctl(struct inode *inode, struct file *file, +static long rtc_dev_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { int err = 0; @@ -219,6 +219,10 @@ static int rtc_dev_ioctl(struct inode *inode, struct file *file, struct rtc_wkalrm alarm; void __user *uarg = (void __user *) arg; + err = mutex_lock_interruptible(&rtc->ops_lock); + if (err) + return -EBUSY; + /* check that the calling task has appropriate permissions * for certain ioctls. doing this check here is useful * to avoid duplicate code in each driver. @@ -227,26 +231,31 @@ static int rtc_dev_ioctl(struct inode *inode, struct file *file, case RTC_EPOCH_SET: case RTC_SET_TIME: if (!capable(CAP_SYS_TIME)) - return -EACCES; + err = -EACCES; break; case RTC_IRQP_SET: if (arg > rtc->max_user_freq && !capable(CAP_SYS_RESOURCE)) - return -EACCES; + err = -EACCES; break; case RTC_PIE_ON: if (rtc->irq_freq > rtc->max_user_freq && !capable(CAP_SYS_RESOURCE)) - return -EACCES; + err = -EACCES; break; } + if (err) + goto done; + /* try the driver's ioctl interface */ if (ops->ioctl) { err = ops->ioctl(rtc->dev.parent, cmd, arg); - if (err != -ENOIOCTLCMD) + if (err != -ENOIOCTLCMD) { + mutex_unlock(&rtc->ops_lock); return err; + } } /* if the driver does not provide the ioctl interface @@ -265,15 +274,19 @@ static int rtc_dev_ioctl(struct inode *inode, struct file *file, switch (cmd) { case RTC_ALM_READ: + mutex_unlock(&rtc->ops_lock); + err = rtc_read_alarm(rtc, &alarm); if (err < 0) return err; if (copy_to_user(uarg, &alarm.time, sizeof(tm))) - return -EFAULT; - break; + err = -EFAULT; + return err; case RTC_ALM_SET: + mutex_unlock(&rtc->ops_lock); + if (copy_from_user(&alarm.time, uarg, sizeof(tm))) return -EFAULT; @@ -321,24 +334,26 @@ static int rtc_dev_ioctl(struct inode *inode, struct file *file, } } - err = rtc_set_alarm(rtc, &alarm); - break; + return rtc_set_alarm(rtc, &alarm); case RTC_RD_TIME: + mutex_unlock(&rtc->ops_lock); + err = rtc_read_time(rtc, &tm); if (err < 0) return err; if (copy_to_user(uarg, &tm, sizeof(tm))) - return -EFAULT; - break; + err = -EFAULT; + return err; case RTC_SET_TIME: + mutex_unlock(&rtc->ops_lock); + if (copy_from_user(&tm, uarg, sizeof(tm))) return -EFAULT; - err = rtc_set_time(rtc, &tm); - break; + return rtc_set_time(rtc, &tm); case RTC_PIE_ON: err = rtc_irq_set_state(rtc, NULL, 1); @@ -376,34 +391,37 @@ static int rtc_dev_ioctl(struct inode *inode, struct file *file, break; #endif case RTC_WKALM_SET: + mutex_unlock(&rtc->ops_lock); if (copy_from_user(&alarm, uarg, sizeof(alarm))) return -EFAULT; - err = rtc_set_alarm(rtc, &alarm); - break; + return rtc_set_alarm(rtc, &alarm); case RTC_WKALM_RD: + mutex_unlock(&rtc->ops_lock); err = rtc_read_alarm(rtc, &alarm); if (err < 0) return err; if (copy_to_user(uarg, &alarm, sizeof(alarm))) - return -EFAULT; - break; + err = -EFAULT; + return err; #ifdef CONFIG_RTC_INTF_DEV_UIE_EMUL case RTC_UIE_OFF: clear_uie(rtc); - return 0; + break; case RTC_UIE_ON: - return set_uie(rtc); + err = set_uie(rtc); #endif default: err = -ENOTTY; break; } +done: + mutex_unlock(&rtc->ops_lock); return err; } @@ -432,7 +450,7 @@ static const struct file_operations rtc_dev_fops = { .llseek = no_llseek, .read = rtc_dev_read, .poll = rtc_dev_poll, - .ioctl = rtc_dev_ioctl, + .unlocked_ioctl = rtc_dev_ioctl, .open = rtc_dev_open, .release = rtc_dev_release, .fasync = rtc_dev_fasync, diff --git a/include/linux/rtc.h b/include/linux/rtc.h index f2d0d1527721..b01fe004cb5e 100644 --- a/include/linux/rtc.h +++ b/include/linux/rtc.h @@ -115,6 +115,23 @@ extern void rtc_time_to_tm(unsigned long time, struct rtc_time *tm); extern struct class *rtc_class; +/* + * For these RTC methods the device parameter is the physical device + * on whatever bus holds the hardware (I2C, Platform, SPI, etc), which + * was passed to rtc_device_register(). Its driver_data normally holds + * device state, including the rtc_device pointer for the RTC. + * + * Most of these methods are called with rtc_device.ops_lock held, + * through the rtc_*(struct rtc_device *, ...) calls. + * + * The (current) exceptions are mostly filesystem hooks: + * - the proc() hook for procfs + * - non-ioctl() chardev hooks: open(), release(), read_callback() + * - periodic irq calls: irq_set_state(), irq_set_freq() + * + * REVISIT those periodic irq calls *do* have ops_lock when they're + * issued through ioctl() ... + */ struct rtc_class_ops { int (*open)(struct device *); void (*release)(struct device *); From 8fc2c767b06067b417c565c4e75731e68ed41fd8 Mon Sep 17 00:00:00 2001 From: "Kim B. Heino" Date: Wed, 23 Jul 2008 21:30:34 -0700 Subject: [PATCH 348/857] rtc: add support for ST M41T94 SPI RTC This patch adds kernel driver for M41T94 RTC chip connected via SPI. I've tested it on two different AT91-based hardwares. This is third revision of the patch: some comments made by Alessandro Zummo fixed. Revision two added support for century bit and fixes. Signed-off-by: Kim B. Heino Signed-off-by: Alessandro Zummo Cc: David Brownell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/rtc/Kconfig | 9 ++ drivers/rtc/Makefile | 1 + drivers/rtc/rtc-m41t94.c | 173 +++++++++++++++++++++++++++++++++++++++ 3 files changed, 183 insertions(+) create mode 100644 drivers/rtc/rtc-m41t94.c diff --git a/drivers/rtc/Kconfig b/drivers/rtc/Kconfig index fc85bf2e4a97..beffb834c440 100644 --- a/drivers/rtc/Kconfig +++ b/drivers/rtc/Kconfig @@ -273,6 +273,15 @@ comment "SPI RTC drivers" if SPI_MASTER +config RTC_DRV_M41T94 + tristate "ST M41T94" + help + If you say yes here you will get support for the + ST M41T94 SPI RTC chip. + + This driver can also be built as a module. If so, the module + will be called rtc-m41t94. + config RTC_DRV_MAX6902 tristate "Maxim MAX6902" help diff --git a/drivers/rtc/Makefile b/drivers/rtc/Makefile index b5d9d67df887..b0e1af54f800 100644 --- a/drivers/rtc/Makefile +++ b/drivers/rtc/Makefile @@ -34,6 +34,7 @@ obj-$(CONFIG_RTC_DRV_EP93XX) += rtc-ep93xx.o obj-$(CONFIG_RTC_DRV_FM3130) += rtc-fm3130.o obj-$(CONFIG_RTC_DRV_ISL1208) += rtc-isl1208.o obj-$(CONFIG_RTC_DRV_M41T80) += rtc-m41t80.o +obj-$(CONFIG_RTC_DRV_M41T94) += rtc-m41t94.o obj-$(CONFIG_RTC_DRV_M48T59) += rtc-m48t59.o obj-$(CONFIG_RTC_DRV_M48T86) += rtc-m48t86.o obj-$(CONFIG_RTC_DRV_MAX6900) += rtc-max6900.o diff --git a/drivers/rtc/rtc-m41t94.c b/drivers/rtc/rtc-m41t94.c new file mode 100644 index 000000000000..9b19499c829e --- /dev/null +++ b/drivers/rtc/rtc-m41t94.c @@ -0,0 +1,173 @@ +/* + * Driver for ST M41T94 SPI RTC + * + * Copyright (C) 2008 Kim B. Heino + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include +#include +#include + +#define M41T94_REG_SECONDS 0x01 +#define M41T94_REG_MINUTES 0x02 +#define M41T94_REG_HOURS 0x03 +#define M41T94_REG_WDAY 0x04 +#define M41T94_REG_DAY 0x05 +#define M41T94_REG_MONTH 0x06 +#define M41T94_REG_YEAR 0x07 +#define M41T94_REG_HT 0x0c + +#define M41T94_BIT_HALT 0x40 +#define M41T94_BIT_STOP 0x80 +#define M41T94_BIT_CB 0x40 +#define M41T94_BIT_CEB 0x80 + +static int m41t94_set_time(struct device *dev, struct rtc_time *tm) +{ + struct spi_device *spi = to_spi_device(dev); + u8 buf[8]; /* write cmd + 7 registers */ + + dev_dbg(dev, "%s secs=%d, mins=%d, " + "hours=%d, mday=%d, mon=%d, year=%d, wday=%d\n", + "write", tm->tm_sec, tm->tm_min, + tm->tm_hour, tm->tm_mday, + tm->tm_mon, tm->tm_year, tm->tm_wday); + + buf[0] = 0x80 | M41T94_REG_SECONDS; /* write time + date */ + buf[M41T94_REG_SECONDS] = BIN2BCD(tm->tm_sec); + buf[M41T94_REG_MINUTES] = BIN2BCD(tm->tm_min); + buf[M41T94_REG_HOURS] = BIN2BCD(tm->tm_hour); + buf[M41T94_REG_WDAY] = BIN2BCD(tm->tm_wday + 1); + buf[M41T94_REG_DAY] = BIN2BCD(tm->tm_mday); + buf[M41T94_REG_MONTH] = BIN2BCD(tm->tm_mon + 1); + + buf[M41T94_REG_HOURS] |= M41T94_BIT_CEB; + if (tm->tm_year >= 100) + buf[M41T94_REG_HOURS] |= M41T94_BIT_CB; + buf[M41T94_REG_YEAR] = BIN2BCD(tm->tm_year % 100); + + return spi_write(spi, buf, 8); +} + +static int m41t94_read_time(struct device *dev, struct rtc_time *tm) +{ + struct spi_device *spi = to_spi_device(dev); + u8 buf[2]; + int ret, hour; + + /* clear halt update bit */ + ret = spi_w8r8(spi, M41T94_REG_HT); + if (ret < 0) + return ret; + if (ret & M41T94_BIT_HALT) { + buf[0] = 0x80 | M41T94_REG_HT; + buf[1] = ret & ~M41T94_BIT_HALT; + spi_write(spi, buf, 2); + } + + /* clear stop bit */ + ret = spi_w8r8(spi, M41T94_REG_SECONDS); + if (ret < 0) + return ret; + if (ret & M41T94_BIT_STOP) { + buf[0] = 0x80 | M41T94_REG_SECONDS; + buf[1] = ret & ~M41T94_BIT_STOP; + spi_write(spi, buf, 2); + } + + tm->tm_sec = BCD2BIN(spi_w8r8(spi, M41T94_REG_SECONDS)); + tm->tm_min = BCD2BIN(spi_w8r8(spi, M41T94_REG_MINUTES)); + hour = spi_w8r8(spi, M41T94_REG_HOURS); + tm->tm_hour = BCD2BIN(hour & 0x3f); + tm->tm_wday = BCD2BIN(spi_w8r8(spi, M41T94_REG_WDAY)) - 1; + tm->tm_mday = BCD2BIN(spi_w8r8(spi, M41T94_REG_DAY)); + tm->tm_mon = BCD2BIN(spi_w8r8(spi, M41T94_REG_MONTH)) - 1; + tm->tm_year = BCD2BIN(spi_w8r8(spi, M41T94_REG_YEAR)); + if ((hour & M41T94_BIT_CB) || !(hour & M41T94_BIT_CEB)) + tm->tm_year += 100; + + dev_dbg(dev, "%s secs=%d, mins=%d, " + "hours=%d, mday=%d, mon=%d, year=%d, wday=%d\n", + "read", tm->tm_sec, tm->tm_min, + tm->tm_hour, tm->tm_mday, + tm->tm_mon, tm->tm_year, tm->tm_wday); + + /* initial clock setting can be undefined */ + return rtc_valid_tm(tm); +} + +static const struct rtc_class_ops m41t94_rtc_ops = { + .read_time = m41t94_read_time, + .set_time = m41t94_set_time, +}; + +static struct spi_driver m41t94_driver; + +static int __devinit m41t94_probe(struct spi_device *spi) +{ + struct rtc_device *rtc; + int res; + + spi->bits_per_word = 8; + spi_setup(spi); + + res = spi_w8r8(spi, M41T94_REG_SECONDS); + if (res < 0) { + dev_err(&spi->dev, "not found.\n"); + return res; + } + + rtc = rtc_device_register(m41t94_driver.driver.name, + &spi->dev, &m41t94_rtc_ops, THIS_MODULE); + if (IS_ERR(rtc)) + return PTR_ERR(rtc); + + dev_set_drvdata(&spi->dev, rtc); + + return 0; +} + +static int __devexit m41t94_remove(struct spi_device *spi) +{ + struct rtc_device *rtc = platform_get_drvdata(spi); + + if (rtc) + rtc_device_unregister(rtc); + + return 0; +} + +static struct spi_driver m41t94_driver = { + .driver = { + .name = "rtc-m41t94", + .bus = &spi_bus_type, + .owner = THIS_MODULE, + }, + .probe = m41t94_probe, + .remove = __devexit_p(m41t94_remove), +}; + +static __init int m41t94_init(void) +{ + return spi_register_driver(&m41t94_driver); +} + +module_init(m41t94_init); + +static __exit void m41t94_exit(void) +{ + spi_unregister_driver(&m41t94_driver); +} + +module_exit(m41t94_exit); + +MODULE_AUTHOR("Kim B. Heino "); +MODULE_DESCRIPTION("Driver for ST M41T94 SPI RTC"); +MODULE_LICENSE("GPL"); From 53e84b672c1a8190af2b376c35c7a39cf1214f59 Mon Sep 17 00:00:00 2001 From: David Brownell Date: Wed, 23 Jul 2008 21:30:36 -0700 Subject: [PATCH 349/857] rtc: ds1305/ds1306 driver Support the Dallas/Maxim DS1305 and DS1306 RTC chips. These use SPI, and support alarms, NVRAM, and a trickle charger for use when their backup power supply is a supercap or rechargeable cell. This basic driver doesn't yet support suspend/resume or wakealarms. Signed-off-by: David Brownell Cc: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/rtc/Kconfig | 10 + drivers/rtc/Makefile | 1 + drivers/rtc/rtc-ds1305.c | 847 +++++++++++++++++++++++++++++++++++++ include/linux/spi/ds1305.h | 35 ++ 4 files changed, 893 insertions(+) create mode 100644 drivers/rtc/rtc-ds1305.c create mode 100644 include/linux/spi/ds1305.h diff --git a/drivers/rtc/Kconfig b/drivers/rtc/Kconfig index beffb834c440..90ab73825401 100644 --- a/drivers/rtc/Kconfig +++ b/drivers/rtc/Kconfig @@ -282,6 +282,16 @@ config RTC_DRV_M41T94 This driver can also be built as a module. If so, the module will be called rtc-m41t94. +config RTC_DRV_DS1305 + tristate "Dallas/Maxim DS1305/DS1306" + help + Select this driver to get support for the Dallas/Maxim DS1305 + and DS1306 real time clock chips. These support a trickle + charger, alarms, and NVRAM in addition to the clock. + + This driver can also be built as a module. If so, the module + will be called rtc-ds1305. + config RTC_DRV_MAX6902 tristate "Maxim MAX6902" help diff --git a/drivers/rtc/Makefile b/drivers/rtc/Makefile index b0e1af54f800..18622ef84cab 100644 --- a/drivers/rtc/Makefile +++ b/drivers/rtc/Makefile @@ -24,6 +24,7 @@ obj-$(CONFIG_RTC_DRV_BFIN) += rtc-bfin.o obj-$(CONFIG_RTC_DRV_CMOS) += rtc-cmos.o obj-$(CONFIG_RTC_DRV_DS1216) += rtc-ds1216.o obj-$(CONFIG_RTC_DRV_DS1302) += rtc-ds1302.o +obj-$(CONFIG_RTC_DRV_DS1305) += rtc-ds1305.o obj-$(CONFIG_RTC_DRV_DS1307) += rtc-ds1307.o obj-$(CONFIG_RTC_DRV_DS1374) += rtc-ds1374.o obj-$(CONFIG_RTC_DRV_DS1511) += rtc-ds1511.o diff --git a/drivers/rtc/rtc-ds1305.c b/drivers/rtc/rtc-ds1305.c new file mode 100644 index 000000000000..b91d02a3ace9 --- /dev/null +++ b/drivers/rtc/rtc-ds1305.c @@ -0,0 +1,847 @@ +/* + * rtc-ds1305.c -- driver for DS1305 and DS1306 SPI RTC chips + * + * Copyright (C) 2008 David Brownell + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + */ +#include +#include +#include +#include +#include + +#include +#include + + +/* + * Registers ... mask DS1305_WRITE into register address to write, + * otherwise you're reading it. All non-bitmask values are BCD. + */ +#define DS1305_WRITE 0x80 + + +/* RTC date/time ... the main special cases are that we: + * - Need fancy "hours" encoding in 12hour mode + * - Don't rely on the "day-of-week" field (or tm_wday) + * - Are a 21st-century clock (2000 <= year < 2100) + */ +#define DS1305_RTC_LEN 7 /* bytes for RTC regs */ + +#define DS1305_SEC 0x00 /* register addresses */ +#define DS1305_MIN 0x01 +#define DS1305_HOUR 0x02 +# define DS1305_HR_12 0x40 /* set == 12 hr mode */ +# define DS1305_HR_PM 0x20 /* set == PM (12hr mode) */ +#define DS1305_WDAY 0x03 +#define DS1305_MDAY 0x04 +#define DS1305_MON 0x05 +#define DS1305_YEAR 0x06 + + +/* The two alarms have only sec/min/hour/wday fields (ALM_LEN). + * DS1305_ALM_DISABLE disables a match field (some combos are bad). + * + * NOTE that since we don't use WDAY, we limit ourselves to alarms + * only one day into the future (vs potentially up to a week). + * + * NOTE ALSO that while we could generate once-a-second IRQs (UIE), we + * don't currently support them. We'd either need to do it only when + * no alarm is pending (not the standard model), or to use the second + * alarm (implying that this is a DS1305 not DS1306, *and* that either + * it's wired up a second IRQ we know, or that INTCN is set) + */ +#define DS1305_ALM_LEN 4 /* bytes for ALM regs */ +#define DS1305_ALM_DISABLE 0x80 + +#define DS1305_ALM0(r) (0x07 + (r)) /* register addresses */ +#define DS1305_ALM1(r) (0x0b + (r)) + + +/* three control registers */ +#define DS1305_CONTROL_LEN 3 /* bytes of control regs */ + +#define DS1305_CONTROL 0x0f /* register addresses */ +# define DS1305_nEOSC 0x80 /* low enables oscillator */ +# define DS1305_WP 0x40 /* write protect */ +# define DS1305_INTCN 0x04 /* clear == only int0 used */ +# define DS1306_1HZ 0x04 /* enable 1Hz output */ +# define DS1305_AEI1 0x02 /* enable ALM1 IRQ */ +# define DS1305_AEI0 0x01 /* enable ALM0 IRQ */ +#define DS1305_STATUS 0x10 +/* status has just AEIx bits, mirrored as IRQFx */ +#define DS1305_TRICKLE 0x11 +/* trickle bits are defined in */ + +/* a bunch of NVRAM */ +#define DS1305_NVRAM_LEN 96 /* bytes of NVRAM */ + +#define DS1305_NVRAM 0x20 /* register addresses */ + + +struct ds1305 { + struct spi_device *spi; + struct rtc_device *rtc; + + struct work_struct work; + + unsigned long flags; +#define FLAG_EXITING 0 + + bool hr12; + u8 ctrl[DS1305_CONTROL_LEN]; +}; + + +/*----------------------------------------------------------------------*/ + +/* + * Utilities ... tolerate 12-hour AM/PM notation in case of non-Linux + * software (like a bootloader) which may require it. + */ + +static unsigned bcd2hour(u8 bcd) +{ + if (bcd & DS1305_HR_12) { + unsigned hour = 0; + + bcd &= ~DS1305_HR_12; + if (bcd & DS1305_HR_PM) { + hour = 12; + bcd &= ~DS1305_HR_PM; + } + hour += BCD2BIN(bcd); + return hour - 1; + } + return BCD2BIN(bcd); +} + +static u8 hour2bcd(bool hr12, int hour) +{ + if (hr12) { + hour++; + if (hour <= 12) + return DS1305_HR_12 | BIN2BCD(hour); + hour -= 12; + return DS1305_HR_12 | DS1305_HR_PM | BIN2BCD(hour); + } + return BIN2BCD(hour); +} + +/*----------------------------------------------------------------------*/ + +/* + * Interface to RTC framework + */ + +#ifdef CONFIG_RTC_INTF_DEV + +/* + * Context: caller holds rtc->ops_lock (to protect ds1305->ctrl) + */ +static int ds1305_ioctl(struct device *dev, unsigned cmd, unsigned long arg) +{ + struct ds1305 *ds1305 = dev_get_drvdata(dev); + u8 buf[2]; + int status = -ENOIOCTLCMD; + + buf[0] = DS1305_WRITE | DS1305_CONTROL; + buf[1] = ds1305->ctrl[0]; + + switch (cmd) { + case RTC_AIE_OFF: + status = 0; + if (!(buf[1] & DS1305_AEI0)) + goto done; + buf[1] &= ~DS1305_AEI0; + break; + + case RTC_AIE_ON: + status = 0; + if (ds1305->ctrl[0] & DS1305_AEI0) + goto done; + buf[1] |= DS1305_AEI0; + break; + } + if (status == 0) { + status = spi_write_then_read(ds1305->spi, buf, sizeof buf, + NULL, 0); + if (status >= 0) + ds1305->ctrl[0] = buf[1]; + } + +done: + return status; +} + +#else +#define ds1305_ioctl NULL +#endif + +/* + * Get/set of date and time is pretty normal. + */ + +static int ds1305_get_time(struct device *dev, struct rtc_time *time) +{ + struct ds1305 *ds1305 = dev_get_drvdata(dev); + u8 addr = DS1305_SEC; + u8 buf[DS1305_RTC_LEN]; + int status; + + /* Use write-then-read to get all the date/time registers + * since dma from stack is nonportable + */ + status = spi_write_then_read(ds1305->spi, &addr, sizeof addr, + buf, sizeof buf); + if (status < 0) + return status; + + dev_vdbg(dev, "%s: %02x %02x %02x, %02x %02x %02x %02x\n", + "read", buf[0], buf[1], buf[2], buf[3], + buf[4], buf[5], buf[6]); + + /* Decode the registers */ + time->tm_sec = BCD2BIN(buf[DS1305_SEC]); + time->tm_min = BCD2BIN(buf[DS1305_MIN]); + time->tm_hour = bcd2hour(buf[DS1305_HOUR]); + time->tm_wday = buf[DS1305_WDAY] - 1; + time->tm_mday = BCD2BIN(buf[DS1305_MDAY]); + time->tm_mon = BCD2BIN(buf[DS1305_MON]) - 1; + time->tm_year = BCD2BIN(buf[DS1305_YEAR]) + 100; + + dev_vdbg(dev, "%s secs=%d, mins=%d, " + "hours=%d, mday=%d, mon=%d, year=%d, wday=%d\n", + "read", time->tm_sec, time->tm_min, + time->tm_hour, time->tm_mday, + time->tm_mon, time->tm_year, time->tm_wday); + + /* Time may not be set */ + return rtc_valid_tm(time); +} + +static int ds1305_set_time(struct device *dev, struct rtc_time *time) +{ + struct ds1305 *ds1305 = dev_get_drvdata(dev); + u8 buf[1 + DS1305_RTC_LEN]; + u8 *bp = buf; + + dev_vdbg(dev, "%s secs=%d, mins=%d, " + "hours=%d, mday=%d, mon=%d, year=%d, wday=%d\n", + "write", time->tm_sec, time->tm_min, + time->tm_hour, time->tm_mday, + time->tm_mon, time->tm_year, time->tm_wday); + + /* Write registers starting at the first time/date address. */ + *bp++ = DS1305_WRITE | DS1305_SEC; + + *bp++ = BIN2BCD(time->tm_sec); + *bp++ = BIN2BCD(time->tm_min); + *bp++ = hour2bcd(ds1305->hr12, time->tm_hour); + *bp++ = (time->tm_wday < 7) ? (time->tm_wday + 1) : 1; + *bp++ = BIN2BCD(time->tm_mday); + *bp++ = BIN2BCD(time->tm_mon + 1); + *bp++ = BIN2BCD(time->tm_year - 100); + + dev_dbg(dev, "%s: %02x %02x %02x, %02x %02x %02x %02x\n", + "write", buf[1], buf[2], buf[3], + buf[4], buf[5], buf[6], buf[7]); + + /* use write-then-read since dma from stack is nonportable */ + return spi_write_then_read(ds1305->spi, buf, sizeof buf, + NULL, 0); +} + +/* + * Get/set of alarm is a bit funky: + * + * - First there's the inherent raciness of getting the (partitioned) + * status of an alarm that could trigger while we're reading parts + * of that status. + * + * - Second there's its limited range (we could increase it a bit by + * relying on WDAY), which means it will easily roll over. + * + * - Third there's the choice of two alarms and alarm signals. + * Here we use ALM0 and expect that nINT0 (open drain) is used; + * that's the only real option for DS1306 runtime alarms, and is + * natural on DS1305. + * + * - Fourth, there's also ALM1, and a second interrupt signal: + * + On DS1305 ALM1 uses nINT1 (when INTCN=1) else nINT0; + * + On DS1306 ALM1 only uses INT1 (an active high pulse) + * and it won't work when VCC1 is active. + * + * So to be most general, we should probably set both alarms to the + * same value, letting ALM1 be the wakeup event source on DS1306 + * and handling several wiring options on DS1305. + * + * - Fifth, we support the polled mode (as well as possible; why not?) + * even when no interrupt line is wired to an IRQ. + */ + +/* + * Context: caller holds rtc->ops_lock (to protect ds1305->ctrl) + */ +static int ds1305_get_alarm(struct device *dev, struct rtc_wkalrm *alm) +{ + struct ds1305 *ds1305 = dev_get_drvdata(dev); + struct spi_device *spi = ds1305->spi; + u8 addr; + int status; + u8 buf[DS1305_ALM_LEN]; + + /* Refresh control register cache BEFORE reading ALM0 registers, + * since reading alarm registers acks any pending IRQ. That + * makes returning "pending" status a bit of a lie, but that bit + * of EFI status is at best fragile anyway (given IRQ handlers). + */ + addr = DS1305_CONTROL; + status = spi_write_then_read(spi, &addr, sizeof addr, + ds1305->ctrl, sizeof ds1305->ctrl); + if (status < 0) + return status; + + alm->enabled = !!(ds1305->ctrl[0] & DS1305_AEI0); + alm->pending = !!(ds1305->ctrl[1] & DS1305_AEI0); + + /* get and check ALM0 registers */ + addr = DS1305_ALM0(DS1305_SEC); + status = spi_write_then_read(spi, &addr, sizeof addr, + buf, sizeof buf); + if (status < 0) + return status; + + dev_vdbg(dev, "%s: %02x %02x %02x %02x\n", + "alm0 read", buf[DS1305_SEC], buf[DS1305_MIN], + buf[DS1305_HOUR], buf[DS1305_WDAY]); + + if ((DS1305_ALM_DISABLE & buf[DS1305_SEC]) + || (DS1305_ALM_DISABLE & buf[DS1305_MIN]) + || (DS1305_ALM_DISABLE & buf[DS1305_HOUR])) + return -EIO; + + /* Stuff these values into alm->time and let RTC framework code + * fill in the rest ... and also handle rollover to tomorrow when + * that's needed. + */ + alm->time.tm_sec = BCD2BIN(buf[DS1305_SEC]); + alm->time.tm_min = BCD2BIN(buf[DS1305_MIN]); + alm->time.tm_hour = bcd2hour(buf[DS1305_HOUR]); + alm->time.tm_mday = -1; + alm->time.tm_mon = -1; + alm->time.tm_year = -1; + /* next three fields are unused by Linux */ + alm->time.tm_wday = -1; + alm->time.tm_mday = -1; + alm->time.tm_isdst = -1; + + return 0; +} + +/* + * Context: caller holds rtc->ops_lock (to protect ds1305->ctrl) + */ +static int ds1305_set_alarm(struct device *dev, struct rtc_wkalrm *alm) +{ + struct ds1305 *ds1305 = dev_get_drvdata(dev); + struct spi_device *spi = ds1305->spi; + unsigned long now, later; + struct rtc_time tm; + int status; + u8 buf[1 + DS1305_ALM_LEN]; + + /* convert desired alarm to time_t */ + status = rtc_tm_to_time(&alm->time, &later); + if (status < 0) + return status; + + /* Read current time as time_t */ + status = ds1305_get_time(dev, &tm); + if (status < 0) + return status; + status = rtc_tm_to_time(&tm, &now); + if (status < 0) + return status; + + /* make sure alarm fires within the next 24 hours */ + if (later <= now) + return -EINVAL; + if ((later - now) > 24 * 60 * 60) + return -EDOM; + + /* disable alarm if needed */ + if (ds1305->ctrl[0] & DS1305_AEI0) { + ds1305->ctrl[0] &= ~DS1305_AEI0; + + buf[0] = DS1305_WRITE | DS1305_CONTROL; + buf[1] = ds1305->ctrl[0]; + status = spi_write_then_read(ds1305->spi, buf, 2, NULL, 0); + if (status < 0) + return status; + } + + /* write alarm */ + buf[0] = DS1305_WRITE | DS1305_ALM0(DS1305_SEC); + buf[1 + DS1305_SEC] = BIN2BCD(alm->time.tm_sec); + buf[1 + DS1305_MIN] = BIN2BCD(alm->time.tm_min); + buf[1 + DS1305_HOUR] = hour2bcd(ds1305->hr12, alm->time.tm_hour); + buf[1 + DS1305_WDAY] = DS1305_ALM_DISABLE; + + dev_dbg(dev, "%s: %02x %02x %02x %02x\n", + "alm0 write", buf[1 + DS1305_SEC], buf[1 + DS1305_MIN], + buf[1 + DS1305_HOUR], buf[1 + DS1305_WDAY]); + + status = spi_write_then_read(spi, buf, sizeof buf, NULL, 0); + if (status < 0) + return status; + + /* enable alarm if requested */ + if (alm->enabled) { + ds1305->ctrl[0] |= DS1305_AEI0; + + buf[0] = DS1305_WRITE | DS1305_CONTROL; + buf[1] = ds1305->ctrl[0]; + status = spi_write_then_read(ds1305->spi, buf, 2, NULL, 0); + } + + return status; +} + +#ifdef CONFIG_PROC_FS + +static int ds1305_proc(struct device *dev, struct seq_file *seq) +{ + struct ds1305 *ds1305 = dev_get_drvdata(dev); + char *diodes = "no"; + char *resistors = ""; + + /* ctrl[2] is treated as read-only; no locking needed */ + if ((ds1305->ctrl[2] & 0xf0) == DS1305_TRICKLE_MAGIC) { + switch (ds1305->ctrl[2] & 0x0c) { + case DS1305_TRICKLE_DS2: + diodes = "2 diodes, "; + break; + case DS1305_TRICKLE_DS1: + diodes = "1 diode, "; + break; + default: + goto done; + } + switch (ds1305->ctrl[2] & 0x03) { + case DS1305_TRICKLE_2K: + resistors = "2k Ohm"; + break; + case DS1305_TRICKLE_4K: + resistors = "4k Ohm"; + break; + case DS1305_TRICKLE_8K: + resistors = "8k Ohm"; + break; + default: + diodes = "no"; + break; + } + } + +done: + return seq_printf(seq, + "trickle_charge\t: %s%s\n", + diodes, resistors); +} + +#else +#define ds1305_proc NULL +#endif + +static const struct rtc_class_ops ds1305_ops = { + .ioctl = ds1305_ioctl, + .read_time = ds1305_get_time, + .set_time = ds1305_set_time, + .read_alarm = ds1305_get_alarm, + .set_alarm = ds1305_set_alarm, + .proc = ds1305_proc, +}; + +static void ds1305_work(struct work_struct *work) +{ + struct ds1305 *ds1305 = container_of(work, struct ds1305, work); + struct mutex *lock = &ds1305->rtc->ops_lock; + struct spi_device *spi = ds1305->spi; + u8 buf[3]; + int status; + + /* lock to protect ds1305->ctrl */ + mutex_lock(lock); + + /* Disable the IRQ, and clear its status ... for now, we "know" + * that if more than one alarm is active, they're in sync. + * Note that reading ALM data registers also clears IRQ status. + */ + ds1305->ctrl[0] &= ~(DS1305_AEI1 | DS1305_AEI0); + ds1305->ctrl[1] = 0; + + buf[0] = DS1305_WRITE | DS1305_CONTROL; + buf[1] = ds1305->ctrl[0]; + buf[2] = 0; + + status = spi_write_then_read(spi, buf, sizeof buf, + NULL, 0); + if (status < 0) + dev_dbg(&spi->dev, "clear irq --> %d\n", status); + + mutex_unlock(lock); + + if (!test_bit(FLAG_EXITING, &ds1305->flags)) + enable_irq(spi->irq); + + /* rtc_update_irq() requires an IRQ-disabled context */ + local_irq_disable(); + rtc_update_irq(ds1305->rtc, 1, RTC_AF | RTC_IRQF); + local_irq_enable(); +} + +/* + * This "real" IRQ handler hands off to a workqueue mostly to allow + * mutex locking for ds1305->ctrl ... unlike I2C, we could issue async + * I/O requests in IRQ context (to clear the IRQ status). + */ +static irqreturn_t ds1305_irq(int irq, void *p) +{ + struct ds1305 *ds1305 = p; + + disable_irq(irq); + schedule_work(&ds1305->work); + return IRQ_HANDLED; +} + +/*----------------------------------------------------------------------*/ + +/* + * Interface for NVRAM + */ + +static void msg_init(struct spi_message *m, struct spi_transfer *x, + u8 *addr, size_t count, char *tx, char *rx) +{ + spi_message_init(m); + memset(x, 0, 2 * sizeof(*x)); + + x->tx_buf = addr; + x->len = 1; + spi_message_add_tail(x, m); + + x++; + + x->tx_buf = tx; + x->rx_buf = rx; + x->len = count; + spi_message_add_tail(x, m); +} + +static ssize_t +ds1305_nvram_read(struct kobject *kobj, struct bin_attribute *attr, + char *buf, loff_t off, size_t count) +{ + struct spi_device *spi; + u8 addr; + struct spi_message m; + struct spi_transfer x[2]; + int status; + + spi = container_of(kobj, struct spi_device, dev.kobj); + + if (unlikely(off >= DS1305_NVRAM_LEN)) + return 0; + if (count >= DS1305_NVRAM_LEN) + count = DS1305_NVRAM_LEN; + if ((off + count) > DS1305_NVRAM_LEN) + count = DS1305_NVRAM_LEN - off; + if (unlikely(!count)) + return count; + + addr = DS1305_NVRAM + off; + msg_init(&m, x, &addr, count, NULL, buf); + + status = spi_sync(spi, &m); + if (status < 0) + dev_err(&spi->dev, "nvram %s error %d\n", "read", status); + return (status < 0) ? status : count; +} + +static ssize_t +ds1305_nvram_write(struct kobject *kobj, struct bin_attribute *attr, + char *buf, loff_t off, size_t count) +{ + struct spi_device *spi; + u8 addr; + struct spi_message m; + struct spi_transfer x[2]; + int status; + + spi = container_of(kobj, struct spi_device, dev.kobj); + + if (unlikely(off >= DS1305_NVRAM_LEN)) + return -EFBIG; + if (count >= DS1305_NVRAM_LEN) + count = DS1305_NVRAM_LEN; + if ((off + count) > DS1305_NVRAM_LEN) + count = DS1305_NVRAM_LEN - off; + if (unlikely(!count)) + return count; + + addr = (DS1305_WRITE | DS1305_NVRAM) + off; + msg_init(&m, x, &addr, count, buf, NULL); + + status = spi_sync(spi, &m); + if (status < 0) + dev_err(&spi->dev, "nvram %s error %d\n", "write", status); + return (status < 0) ? status : count; +} + +static struct bin_attribute nvram = { + .attr.name = "nvram", + .attr.mode = S_IRUGO | S_IWUSR, + .attr.owner = THIS_MODULE, + .read = ds1305_nvram_read, + .write = ds1305_nvram_write, + .size = DS1305_NVRAM_LEN, +}; + +/*----------------------------------------------------------------------*/ + +/* + * Interface to SPI stack + */ + +static int __devinit ds1305_probe(struct spi_device *spi) +{ + struct ds1305 *ds1305; + struct rtc_device *rtc; + int status; + u8 addr, value; + struct ds1305_platform_data *pdata = spi->dev.platform_data; + bool write_ctrl = false; + + /* Sanity check board setup data. This may be hooked up + * in 3wire mode, but we don't care. Note that unless + * there's an inverter in place, this needs SPI_CS_HIGH! + */ + if ((spi->bits_per_word && spi->bits_per_word != 8) + || (spi->max_speed_hz > 2000000) + || !(spi->mode & SPI_CPHA)) + return -EINVAL; + + /* set up driver data */ + ds1305 = kzalloc(sizeof *ds1305, GFP_KERNEL); + if (!ds1305) + return -ENOMEM; + ds1305->spi = spi; + spi_set_drvdata(spi, ds1305); + + /* read and cache control registers */ + addr = DS1305_CONTROL; + status = spi_write_then_read(spi, &addr, sizeof addr, + ds1305->ctrl, sizeof ds1305->ctrl); + if (status < 0) { + dev_dbg(&spi->dev, "can't %s, %d\n", + "read", status); + goto fail0; + } + + dev_dbg(&spi->dev, "ctrl %s: %02x %02x %02x\n", + "read", ds1305->ctrl[0], + ds1305->ctrl[1], ds1305->ctrl[2]); + + /* Sanity check register values ... partially compensating for the + * fact that SPI has no device handshake. A pullup on MISO would + * make these tests fail; but not all systems will have one. If + * some register is neither 0x00 nor 0xff, a chip is likely there. + */ + if ((ds1305->ctrl[0] & 0x38) != 0 || (ds1305->ctrl[1] & 0xfc) != 0) { + dev_dbg(&spi->dev, "RTC chip is not present\n"); + status = -ENODEV; + goto fail0; + } + if (ds1305->ctrl[2] == 0) + dev_dbg(&spi->dev, "chip may not be present\n"); + + /* enable writes if needed ... if we were paranoid it would + * make sense to enable them only when absolutely necessary. + */ + if (ds1305->ctrl[0] & DS1305_WP) { + u8 buf[2]; + + ds1305->ctrl[0] &= ~DS1305_WP; + + buf[0] = DS1305_WRITE | DS1305_CONTROL; + buf[1] = ds1305->ctrl[0]; + status = spi_write_then_read(spi, buf, sizeof buf, NULL, 0); + + dev_dbg(&spi->dev, "clear WP --> %d\n", status); + if (status < 0) + goto fail0; + } + + /* on DS1305, maybe start oscillator; like most low power + * oscillators, it may take a second to stabilize + */ + if (ds1305->ctrl[0] & DS1305_nEOSC) { + ds1305->ctrl[0] &= ~DS1305_nEOSC; + write_ctrl = true; + dev_warn(&spi->dev, "SET TIME!\n"); + } + + /* ack any pending IRQs */ + if (ds1305->ctrl[1]) { + ds1305->ctrl[1] = 0; + write_ctrl = true; + } + + /* this may need one-time (re)init */ + if (pdata) { + /* maybe enable trickle charge */ + if (((ds1305->ctrl[2] & 0xf0) != DS1305_TRICKLE_MAGIC)) { + ds1305->ctrl[2] = DS1305_TRICKLE_MAGIC + | pdata->trickle; + write_ctrl = true; + } + + /* on DS1306, configure 1 Hz signal */ + if (pdata->is_ds1306) { + if (pdata->en_1hz) { + if (!(ds1305->ctrl[0] & DS1306_1HZ)) { + ds1305->ctrl[0] |= DS1306_1HZ; + write_ctrl = true; + } + } else { + if (ds1305->ctrl[0] & DS1306_1HZ) { + ds1305->ctrl[0] &= ~DS1306_1HZ; + write_ctrl = true; + } + } + } + } + + if (write_ctrl) { + u8 buf[4]; + + buf[0] = DS1305_WRITE | DS1305_CONTROL; + buf[1] = ds1305->ctrl[0]; + buf[2] = ds1305->ctrl[1]; + buf[3] = ds1305->ctrl[2]; + status = spi_write_then_read(spi, buf, sizeof buf, NULL, 0); + if (status < 0) { + dev_dbg(&spi->dev, "can't %s, %d\n", + "write", status); + goto fail0; + } + + dev_dbg(&spi->dev, "ctrl %s: %02x %02x %02x\n", + "write", ds1305->ctrl[0], + ds1305->ctrl[1], ds1305->ctrl[2]); + } + + /* see if non-Linux software set up AM/PM mode */ + addr = DS1305_HOUR; + status = spi_write_then_read(spi, &addr, sizeof addr, + &value, sizeof value); + if (status < 0) { + dev_dbg(&spi->dev, "read HOUR --> %d\n", status); + goto fail0; + } + + ds1305->hr12 = (DS1305_HR_12 & value) != 0; + if (ds1305->hr12) + dev_dbg(&spi->dev, "AM/PM\n"); + + /* register RTC ... from here on, ds1305->ctrl needs locking */ + rtc = rtc_device_register("ds1305", &spi->dev, + &ds1305_ops, THIS_MODULE); + if (IS_ERR(rtc)) { + status = PTR_ERR(rtc); + dev_dbg(&spi->dev, "register rtc --> %d\n", status); + goto fail0; + } + ds1305->rtc = rtc; + + /* Maybe set up alarm IRQ; be ready to handle it triggering right + * away. NOTE that we don't share this. The signal is active low, + * and we can't ack it before a SPI message delay. We temporarily + * disable the IRQ until it's acked, which lets us work with more + * IRQ trigger modes (not all IRQ controllers can do falling edge). + */ + if (spi->irq) { + INIT_WORK(&ds1305->work, ds1305_work); + status = request_irq(spi->irq, ds1305_irq, + 0, dev_name(&rtc->dev), ds1305); + if (status < 0) { + dev_dbg(&spi->dev, "request_irq %d --> %d\n", + spi->irq, status); + goto fail1; + } + } + + /* export NVRAM */ + status = sysfs_create_bin_file(&spi->dev.kobj, &nvram); + if (status < 0) { + dev_dbg(&spi->dev, "register nvram --> %d\n", status); + goto fail2; + } + + return 0; + +fail2: + free_irq(spi->irq, ds1305); +fail1: + rtc_device_unregister(rtc); +fail0: + kfree(ds1305); + return status; +} + +static int __devexit ds1305_remove(struct spi_device *spi) +{ + struct ds1305 *ds1305 = spi_get_drvdata(spi); + + sysfs_remove_bin_file(&spi->dev.kobj, &nvram); + + /* carefully shut down irq and workqueue, if present */ + if (spi->irq) { + set_bit(FLAG_EXITING, &ds1305->flags); + free_irq(spi->irq, ds1305); + flush_scheduled_work(); + } + + rtc_device_unregister(ds1305->rtc); + spi_set_drvdata(spi, NULL); + kfree(ds1305); + return 0; +} + +static struct spi_driver ds1305_driver = { + .driver.name = "rtc-ds1305", + .driver.owner = THIS_MODULE, + .probe = ds1305_probe, + .remove = __devexit_p(ds1305_remove), + /* REVISIT add suspend/resume */ +}; + +static int __init ds1305_init(void) +{ + return spi_register_driver(&ds1305_driver); +} +module_init(ds1305_init); + +static void __exit ds1305_exit(void) +{ + spi_unregister_driver(&ds1305_driver); +} +module_exit(ds1305_exit); + +MODULE_DESCRIPTION("RTC driver for DS1305 and DS1306 chips"); +MODULE_LICENSE("GPL"); diff --git a/include/linux/spi/ds1305.h b/include/linux/spi/ds1305.h new file mode 100644 index 000000000000..287ec830eab7 --- /dev/null +++ b/include/linux/spi/ds1305.h @@ -0,0 +1,35 @@ +#ifndef __LINUX_SPI_DS1305_H +#define __LINUX_SPI_DS1305_H + +/* + * One-time configuration for ds1305 and ds1306 RTC chips. + * + * Put a pointer to this in spi_board_info.platform_data if you want to + * be sure that Linux (re)initializes this as needed ... after losing + * backup power, and potentially on the first boot. + */ +struct ds1305_platform_data { + + /* Trickle charge configuration: it's OK to leave out the MAGIC + * bitmask; mask in either DS1 or DS2, and then one of 2K/4k/8K. + */ +#define DS1305_TRICKLE_MAGIC 0xa0 +#define DS1305_TRICKLE_DS2 0x08 /* two diodes */ +#define DS1305_TRICKLE_DS1 0x04 /* one diode */ +#define DS1305_TRICKLE_2K 0x01 /* 2 KOhm resistance */ +#define DS1305_TRICKLE_4K 0x02 /* 4 KOhm resistance */ +#define DS1305_TRICKLE_8K 0x03 /* 8 KOhm resistance */ + u8 trickle; + + /* set only on ds1306 parts */ + bool is_ds1306; + + /* ds1306 only: enable 1 Hz output */ + bool en_1hz; + + /* REVISIT: the driver currently expects nINT0 to be wired + * as the alarm IRQ. ALM1 may also need to be set up ... + */ +}; + +#endif /* __LINUX_SPI_DS1305_H */ From d3de851a445123f24ad8ece18662014b5e8a8b4e Mon Sep 17 00:00:00 2001 From: David Brownell Date: Wed, 23 Jul 2008 21:30:37 -0700 Subject: [PATCH 350/857] rtc: BCD codeshrink This updates to define the key routines as constant functions, which the macros will then call. Newer code can now call bcd2bin() instead of SCREAMING BCD2BIN() TO THE FOUR WINDS. This lets each driver shrink their codespace by using N function calls to a single (global) copy of those routines, instead of N inlined copies of these functions per driver. These routines aren't used in speed-critical code. Almost all callers are in the RTC framework. Typical per-driver savings is near 300 bytes. Signed-off-by: David Brownell Acked-by: Adrian Bunk Cc: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/bcd.h | 9 +++++++-- lib/Makefile | 2 +- lib/bcd.c | 14 ++++++++++++++ 3 files changed, 22 insertions(+), 3 deletions(-) create mode 100644 lib/bcd.c diff --git a/include/linux/bcd.h b/include/linux/bcd.h index c545308125b0..7ac518e3c152 100644 --- a/include/linux/bcd.h +++ b/include/linux/bcd.h @@ -10,8 +10,13 @@ #ifndef _BCD_H #define _BCD_H -#define BCD2BIN(val) (((val) & 0x0f) + ((val)>>4)*10) -#define BIN2BCD(val) ((((val)/10)<<4) + (val)%10) +#include + +unsigned bcd2bin(unsigned char val) __attribute_const__; +unsigned char bin2bcd(unsigned val) __attribute_const__; + +#define BCD2BIN(val) bcd2bin(val) +#define BIN2BCD(val) bin2bcd(val) /* backwards compat */ #define BCD_TO_BIN(val) ((val)=BCD2BIN(val)) diff --git a/lib/Makefile b/lib/Makefile index 818c4d455518..9085ad6fa53d 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -18,7 +18,7 @@ lib-$(CONFIG_SMP) += cpumask.o lib-y += kobject.o kref.o klist.o -obj-y += div64.o sort.o parser.o halfmd4.o debug_locks.o random32.o \ +obj-y += bcd.o div64.o sort.o parser.o halfmd4.o debug_locks.o random32.o \ bust_spinlocks.o hexdump.o kasprintf.o bitmap.o scatterlist.o ifeq ($(CONFIG_DEBUG_KOBJECT),y) diff --git a/lib/bcd.c b/lib/bcd.c new file mode 100644 index 000000000000..d74257fd0fe7 --- /dev/null +++ b/lib/bcd.c @@ -0,0 +1,14 @@ +#include +#include + +unsigned bcd2bin(unsigned char val) +{ + return (val & 0x0f) + (val >> 4) * 10; +} +EXPORT_SYMBOL(bcd2bin); + +unsigned char bin2bcd(unsigned val) +{ + return ((val / 10) << 4) + val % 10; +} +EXPORT_SYMBOL(bin2bcd); From 71fc822455ccb63a66be0b6e97a415aceb0062c6 Mon Sep 17 00:00:00 2001 From: David Brownell Date: Wed, 23 Jul 2008 21:30:38 -0700 Subject: [PATCH 351/857] rtc: rtc-omap footprint shrinkage Shrink the runtime footprint of the OMAP1 RTC driver a bunch by removing some old hacks and switching to platform_driver_probe(). Signed-off-by: David Brownell Cc: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/rtc/rtc-omap.c | 21 ++++----------------- 1 file changed, 4 insertions(+), 17 deletions(-) diff --git a/drivers/rtc/rtc-omap.c b/drivers/rtc/rtc-omap.c index eb23d8423f42..8876605d4d4b 100644 --- a/drivers/rtc/rtc-omap.c +++ b/drivers/rtc/rtc-omap.c @@ -92,18 +92,6 @@ #define rtc_write(val, addr) omap_writeb(val, OMAP_RTC_BASE + (addr)) -/* platform_bus isn't hotpluggable, so for static linkage it'd be safe - * to get rid of probe() and remove() code ... too bad the driver struct - * remembers probe(), that's about 25% of the runtime footprint!! - */ -#ifndef MODULE -#undef __devexit -#undef __devexit_p -#define __devexit __exit -#define __devexit_p __exit_p -#endif - - /* we rely on the rtc framework to handle locking (rtc->ops_lock), * so the only other requirement is that register accesses which * require BUSY to be clear are made with IRQs locally disabled @@ -324,7 +312,7 @@ static struct rtc_class_ops omap_rtc_ops = { static int omap_rtc_alarm; static int omap_rtc_timer; -static int __devinit omap_rtc_probe(struct platform_device *pdev) +static int __init omap_rtc_probe(struct platform_device *pdev) { struct resource *res, *mem; struct rtc_device *rtc; @@ -440,7 +428,7 @@ static int __devinit omap_rtc_probe(struct platform_device *pdev) return -EIO; } -static int __devexit omap_rtc_remove(struct platform_device *pdev) +static int __exit omap_rtc_remove(struct platform_device *pdev) { struct rtc_device *rtc = platform_get_drvdata(pdev);; @@ -498,8 +486,7 @@ static void omap_rtc_shutdown(struct platform_device *pdev) MODULE_ALIAS("platform:omap_rtc"); static struct platform_driver omap_rtc_driver = { - .probe = omap_rtc_probe, - .remove = __devexit_p(omap_rtc_remove), + .remove = __exit_p(omap_rtc_remove), .suspend = omap_rtc_suspend, .resume = omap_rtc_resume, .shutdown = omap_rtc_shutdown, @@ -511,7 +498,7 @@ static struct platform_driver omap_rtc_driver = { static int __init rtc_init(void) { - return platform_driver_register(&omap_rtc_driver); + return platform_driver_probe(&omap_rtc_driver, omap_rtc_probe); } module_init(rtc_init); From 02bb584f3b1cfc8188522a4d2c8881b65073a4f1 Mon Sep 17 00:00:00 2001 From: Wolfram Sang Date: Wed, 23 Jul 2008 21:30:39 -0700 Subject: [PATCH 352/857] rtc: convert the PCF8583 driver to the new I2C style framework with device_ids Convert the PCF8583 driver to the new I2C style framework with device_ids Signed-off-by: Juergen Beisert Signed-off-by: Wolfram Sang Signed-off-by: Alessandro Zummo Cc: David Brownell Acked-by: Jean Delvare Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/rtc/rtc-pcf8583.c | 135 ++++++++++++-------------------------- 1 file changed, 42 insertions(+), 93 deletions(-) diff --git a/drivers/rtc/rtc-pcf8583.c b/drivers/rtc/rtc-pcf8583.c index 3d09d8f0b1f0..d388c662bf4b 100644 --- a/drivers/rtc/rtc-pcf8583.c +++ b/drivers/rtc/rtc-pcf8583.c @@ -2,6 +2,7 @@ * drivers/rtc/rtc-pcf8583.c * * Copyright (C) 2000 Russell King + * Copyright (C) 2008 Wolfram Sang & Juergen Beisert, Pengutronix * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -14,7 +15,6 @@ #include #include #include -#include #include #include #include @@ -27,7 +27,6 @@ struct rtc_mem { }; struct pcf8583 { - struct i2c_client client; struct rtc_device *rtc; unsigned char ctrl; }; @@ -40,10 +39,6 @@ struct pcf8583 { #define CTRL_ALARM 0x02 #define CTRL_TIMER 0x01 -static const unsigned short normal_i2c[] = { 0x50, I2C_CLIENT_END }; - -/* Module parameters */ -I2C_CLIENT_INSMOD; static struct i2c_driver pcf8583_driver; @@ -269,107 +264,61 @@ static const struct rtc_class_ops pcf8583_rtc_ops = { .set_time = pcf8583_rtc_set_time, }; -static int pcf8583_probe(struct i2c_adapter *adap, int addr, int kind); - -static int pcf8583_attach(struct i2c_adapter *adap) +static int pcf8583_probe(struct i2c_client *client, + const struct i2c_device_id *id) { - return i2c_probe(adap, &addr_data, pcf8583_probe); + struct pcf8583 *pcf8583; + int err; + + if (!i2c_check_functionality(client->adapter, I2C_FUNC_I2C)) + return -ENODEV; + + pcf8583 = kzalloc(sizeof(struct pcf8583), GFP_KERNEL); + if (!pcf8583) + return -ENOMEM; + + pcf8583->rtc = rtc_device_register(pcf8583_driver.driver.name, + &client->dev, &pcf8583_rtc_ops, THIS_MODULE); + + if (IS_ERR(pcf8583->rtc)) { + err = PTR_ERR(pcf8583->rtc); + goto exit_kfree; + } + + i2c_set_clientdata(client, pcf8583); + return 0; + +exit_kfree: + kfree(pcf8583); + return err; } -static int pcf8583_detach(struct i2c_client *client) +static int __devexit pcf8583_remove(struct i2c_client *client) { - int err; - struct pcf8583 *pcf = i2c_get_clientdata(client); - struct rtc_device *rtc = pcf->rtc; + struct pcf8583 *pcf8583 = i2c_get_clientdata(client); - if (rtc) - rtc_device_unregister(rtc); - - if ((err = i2c_detach_client(client))) - return err; - - kfree(pcf); + if (pcf8583->rtc) + rtc_device_unregister(pcf8583->rtc); + kfree(pcf8583); return 0; } +static const struct i2c_device_id pcf8583_id[] = { + { "pcf8583", 0 }, + { } +}; +MODULE_DEVICE_TABLE(i2c, pcf8583_id); + static struct i2c_driver pcf8583_driver = { .driver = { .name = "pcf8583", + .owner = THIS_MODULE, }, - .id = I2C_DRIVERID_PCF8583, - .attach_adapter = pcf8583_attach, - .detach_client = pcf8583_detach, + .probe = pcf8583_probe, + .remove = __devexit_p(pcf8583_remove), + .id_table = pcf8583_id, }; -static int pcf8583_probe(struct i2c_adapter *adap, int addr, int kind) -{ - struct pcf8583 *pcf; - struct i2c_client *client; - struct rtc_device *rtc; - unsigned char buf[1], ad[1] = { 0 }; - int err; - struct i2c_msg msgs[2] = { - { - .addr = addr, - .flags = 0, - .len = 1, - .buf = ad, - }, { - .addr = addr, - .flags = I2C_M_RD, - .len = 1, - .buf = buf, - } - }; - - if (!i2c_check_functionality(adap, I2C_FUNC_I2C)) - return 0; - - pcf = kzalloc(sizeof(*pcf), GFP_KERNEL); - if (!pcf) - return -ENOMEM; - - client = &pcf->client; - - client->addr = addr; - client->adapter = adap; - client->driver = &pcf8583_driver; - - strlcpy(client->name, pcf8583_driver.driver.name, I2C_NAME_SIZE); - - if (i2c_transfer(client->adapter, msgs, 2) != 2) { - err = -EIO; - goto exit_kfree; - } - - err = i2c_attach_client(client); - - if (err) - goto exit_kfree; - - rtc = rtc_device_register(pcf8583_driver.driver.name, &client->dev, - &pcf8583_rtc_ops, THIS_MODULE); - - if (IS_ERR(rtc)) { - err = PTR_ERR(rtc); - goto exit_detach; - } - - pcf->rtc = rtc; - i2c_set_clientdata(client, pcf); - set_ctrl(client, buf[0]); - - return 0; - -exit_detach: - i2c_detach_client(client); - -exit_kfree: - kfree(pcf); - - return err; -} - static __init int pcf8583_init(void) { return i2c_add_driver(&pcf8583_driver); From c68d07b2da54c941bb36c9d6d35fe8f263ee10ef Mon Sep 17 00:00:00 2001 From: "Carlos R. Mafra" Date: Wed, 23 Jul 2008 21:30:40 -0700 Subject: [PATCH 353/857] rtc: remove and clarify unneeded externs When CONFIG_HPET_EMULATE_RTC is defined the external declaration of hpet_rtc_interrupt is redundant due to the inclusion of hpet.h. When !CONFIG_HPET_EMULATE_RTC we make it clear that hpet_rtc_interrupt is not used by defining it to return zero. Signed-off-by: Carlos R. Mafra Cc: Ingo Molnar Cc: Thomas Gleixner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/rtc.c | 2 -- drivers/rtc/rtc-cmos.c | 5 ++++- 2 files changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/char/rtc.c b/drivers/char/rtc.c index d1569a0d0506..dbefbb30ed44 100644 --- a/drivers/char/rtc.c +++ b/drivers/char/rtc.c @@ -121,8 +121,6 @@ static irqreturn_t hpet_rtc_interrupt(int irq, void *dev_id) return 0; } #endif -#else -extern irqreturn_t hpet_rtc_interrupt(int irq, void *dev_id); #endif /* diff --git a/drivers/rtc/rtc-cmos.c b/drivers/rtc/rtc-cmos.c index d7bb9bac71df..94b89a2d9c2e 100644 --- a/drivers/rtc/rtc-cmos.c +++ b/drivers/rtc/rtc-cmos.c @@ -52,7 +52,10 @@ #define hpet_rtc_timer_init() do { } while (0) #define hpet_register_irq_handler(h) 0 #define hpet_unregister_irq_handler(h) do { } while (0) -extern irqreturn_t hpet_rtc_interrupt(int irq, void *dev_id); +static irqreturn_t hpet_rtc_interrupt(int irq, void *dev_id) +{ + return 0; +} #endif struct cmos_rtc { From 35d3fdd5f304c06654c940921fc045c60df34693 Mon Sep 17 00:00:00 2001 From: David Brownell Date: Wed, 23 Jul 2008 21:30:43 -0700 Subject: [PATCH 354/857] rtc-cmos: improve HPET IRQ glue Resolve http://bugzilla.kernel.org/show_bug.cgi?id=11051 and other bugs related to the way the HPET glue code in rtc-cmos was incomplete and inconsistent: * Switch the approach so that the basic driver code flow isn't changed by having HPET ... instead, just have HPET shadow the RTC_CONTROL irq enables and RTC_FREQ_SELECT data. It's only coping with IRQ thievery, after all. * Do that consistently (!!) to avoid problems when the HPET code is out of sync with the real RTC intent. Examples include: - cmos_procfs(), which now reports correct data - cmos_irq_set_state() ... also removing the previous PIE_{ON,OFF} ioctl support so only one code path manages "periodic" IRQs - cmos_do_shutdown() ... currently a "just in case" change. - cmos_suspend() and cmos_resume() ... also handling a bug that was specific to HPET's IRQ thievery, where the alarm wasn't disabled after waking the system * Always call that HPET code under the RTC spinlock (it doesn't do its own locking) Also clean up the HPET glue: * Add some comments explaining what's going on. * Switch to having just one #ifdef for the HPET glue, and inline functions (not #defines) to avoid some compiler warnings. * Have the probe message also report when HPET IRQs are involved This still leaves various holes in the HPET glue, like the emulated update IRQs being out of sync with the RTC, alarms never using day or month matches, and many extra IRQs (at 64 Hz). [akpm@linux-foundation.org: fix build] Signed-off-by: David Brownell Cc: Tomas Janousek Cc: Bernhard Walle Cc: Carlos R. Mafra Acked-by: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/rtc/rtc-cmos.c | 193 ++++++++++++++++++++++++++++------------- 1 file changed, 131 insertions(+), 62 deletions(-) diff --git a/drivers/rtc/rtc-cmos.c b/drivers/rtc/rtc-cmos.c index 94b89a2d9c2e..e9984650ea95 100644 --- a/drivers/rtc/rtc-cmos.c +++ b/drivers/rtc/rtc-cmos.c @@ -36,28 +36,9 @@ #include #include -#ifdef CONFIG_HPET_EMULATE_RTC -#include -#endif - /* this is for "generic access to PC-style RTC" using CMOS_READ/CMOS_WRITE */ #include -#ifndef CONFIG_HPET_EMULATE_RTC -#define is_hpet_enabled() 0 -#define hpet_set_alarm_time(hrs, min, sec) do { } while (0) -#define hpet_set_periodic_freq(arg) 0 -#define hpet_mask_rtc_irq_bit(arg) do { } while (0) -#define hpet_set_rtc_irq_bit(arg) do { } while (0) -#define hpet_rtc_timer_init() do { } while (0) -#define hpet_register_irq_handler(h) 0 -#define hpet_unregister_irq_handler(h) do { } while (0) -static irqreturn_t hpet_rtc_interrupt(int irq, void *dev_id) -{ - return 0; -} -#endif - struct cmos_rtc { struct rtc_device *rtc; struct device *dev; @@ -96,6 +77,72 @@ static inline int is_intr(u8 rtc_intr) /*----------------------------------------------------------------*/ +/* Much modern x86 hardware has HPETs (10+ MHz timers) which, because + * many BIOS programmers don't set up "sane mode" IRQ routing, are mostly + * used in a broken "legacy replacement" mode. The breakage includes + * HPET #1 hijacking the IRQ for this RTC, and being unavailable for + * other (better) use. + * + * When that broken mode is in use, platform glue provides a partial + * emulation of hardware RTC IRQ facilities using HPET #1. We don't + * want to use HPET for anything except those IRQs though... + */ +#ifdef CONFIG_HPET_EMULATE_RTC +#include +#else + +static inline int is_hpet_enabled(void) +{ + return 0; +} + +static inline int hpet_mask_rtc_irq_bit(unsigned long mask) +{ + return 0; +} + +static inline int hpet_set_rtc_irq_bit(unsigned long mask) +{ + return 0; +} + +static inline int +hpet_set_alarm_time(unsigned char hrs, unsigned char min, unsigned char sec) +{ + return 0; +} + +static inline int hpet_set_periodic_freq(unsigned long freq) +{ + return 0; +} + +static inline int hpet_rtc_dropped_irq(void) +{ + return 0; +} + +static inline int hpet_rtc_timer_init(void) +{ + return 0; +} + +extern irq_handler_t hpet_rtc_interrupt; + +static inline int hpet_register_irq_handler(irq_handler_t handler) +{ + return 0; +} + +static inline int hpet_unregister_irq_handler(irq_handler_t handler) +{ + return 0; +} + +#endif + +/*----------------------------------------------------------------*/ + static int cmos_read_time(struct device *dev, struct rtc_time *t) { /* REVISIT: if the clock has a "century" register, use @@ -216,13 +263,14 @@ static int cmos_set_alarm(struct device *dev, struct rtc_wkalrm *t) sec = t->time.tm_sec; sec = (sec < 60) ? BIN2BCD(sec) : 0xff; - hpet_set_alarm_time(t->time.tm_hour, t->time.tm_min, t->time.tm_sec); spin_lock_irq(&rtc_lock); /* next rtc irq must not be from previous alarm setting */ rtc_control = CMOS_READ(RTC_CONTROL); rtc_control &= ~RTC_AIE; CMOS_WRITE(rtc_control, RTC_CONTROL); + hpet_mask_rtc_irq_bit(RTC_AIE); + rtc_intr = CMOS_READ(RTC_INTR_FLAGS); rtc_intr &= (rtc_control & RTC_IRQMASK) | RTC_IRQF; if (is_intr(rtc_intr)) @@ -240,9 +288,16 @@ static int cmos_set_alarm(struct device *dev, struct rtc_wkalrm *t) CMOS_WRITE(mon, cmos->mon_alrm); } + /* FIXME the HPET alarm glue currently ignores day_alrm + * and mon_alrm ... + */ + hpet_set_alarm_time(t->time.tm_hour, t->time.tm_min, t->time.tm_sec); + if (t->enabled) { rtc_control |= RTC_AIE; CMOS_WRITE(rtc_control, RTC_CONTROL); + hpet_set_rtc_irq_bit(RTC_AIE); + rtc_intr = CMOS_READ(RTC_INTR_FLAGS); rtc_intr &= (rtc_control & RTC_IRQMASK) | RTC_IRQF; if (is_intr(rtc_intr)) @@ -270,8 +325,8 @@ static int cmos_irq_set_freq(struct device *dev, int freq) f = 16 - f; spin_lock_irqsave(&rtc_lock, flags); - if (!hpet_set_periodic_freq(freq)) - CMOS_WRITE(RTC_REF_CLCK_32KHZ | f, RTC_FREQ_SELECT); + hpet_set_periodic_freq(freq); + CMOS_WRITE(RTC_REF_CLCK_32KHZ | f, RTC_FREQ_SELECT); spin_unlock_irqrestore(&rtc_lock, flags); return 0; @@ -289,11 +344,13 @@ static int cmos_irq_set_state(struct device *dev, int enabled) spin_lock_irqsave(&rtc_lock, flags); rtc_control = CMOS_READ(RTC_CONTROL); - if (enabled) + if (enabled) { rtc_control |= RTC_PIE; - else + hpet_set_rtc_irq_bit(RTC_PIE); + } else { rtc_control &= ~RTC_PIE; - + hpet_mask_rtc_irq_bit(RTC_PIE); + } CMOS_WRITE(rtc_control, RTC_CONTROL); rtc_intr = CMOS_READ(RTC_INTR_FLAGS); @@ -319,11 +376,10 @@ cmos_rtc_ioctl(struct device *dev, unsigned int cmd, unsigned long arg) case RTC_AIE_ON: case RTC_UIE_OFF: case RTC_UIE_ON: - case RTC_PIE_OFF: - case RTC_PIE_ON: if (!is_valid_irq(cmos->irq)) return -EINVAL; break; + /* PIE ON/OFF is handled by cmos_irq_set_state() */ default: return -ENOIOCTLCMD; } @@ -347,17 +403,8 @@ cmos_rtc_ioctl(struct device *dev, unsigned int cmd, unsigned long arg) rtc_control |= RTC_UIE; hpet_set_rtc_irq_bit(RTC_UIE); break; - case RTC_PIE_OFF: /* periodic off */ - rtc_control &= ~RTC_PIE; - hpet_mask_rtc_irq_bit(RTC_PIE); - break; - case RTC_PIE_ON: /* periodic on */ - rtc_control |= RTC_PIE; - hpet_set_rtc_irq_bit(RTC_PIE); - break; } - if (!is_hpet_enabled()) - CMOS_WRITE(rtc_control, RTC_CONTROL); + CMOS_WRITE(rtc_control, RTC_CONTROL); rtc_intr = CMOS_READ(RTC_INTR_FLAGS); rtc_intr &= (rtc_control & RTC_IRQMASK) | RTC_IRQF; @@ -505,18 +552,19 @@ static irqreturn_t cmos_interrupt(int irq, void *p) u8 rtc_control; spin_lock(&rtc_lock); - /* - * In this case it is HPET RTC interrupt handler - * calling us, with the interrupt information - * passed as arg1, instead of irq. + + /* When the HPET interrupt handler calls us, the interrupt + * status is passed as arg1 instead of the irq number. But + * always clear irq status, even when HPET is in the way. + * + * Note that HPET and RTC are almost certainly out of phase, + * giving different IRQ status ... */ + irqstat = CMOS_READ(RTC_INTR_FLAGS); + rtc_control = CMOS_READ(RTC_CONTROL); if (is_hpet_enabled()) irqstat = (unsigned long)irq & 0xF0; - else { - irqstat = CMOS_READ(RTC_INTR_FLAGS); - rtc_control = CMOS_READ(RTC_CONTROL); - irqstat &= (rtc_control & RTC_IRQMASK) | RTC_IRQF; - } + irqstat &= (rtc_control & RTC_IRQMASK) | RTC_IRQF; /* All Linux RTC alarms should be treated as if they were oneshot. * Similar code may be needed in system wakeup paths, in case the @@ -526,6 +574,8 @@ static irqreturn_t cmos_interrupt(int irq, void *p) rtc_control = CMOS_READ(RTC_CONTROL); rtc_control &= ~RTC_AIE; CMOS_WRITE(rtc_control, RTC_CONTROL); + hpet_mask_rtc_irq_bit(RTC_AIE); + CMOS_READ(RTC_INTR_FLAGS); } spin_unlock(&rtc_lock); @@ -632,8 +682,8 @@ cmos_do_probe(struct device *dev, struct resource *ports, int rtc_irq) * do something about other clock frequencies. */ cmos_rtc.rtc->irq_freq = 1024; - if (!hpet_set_periodic_freq(cmos_rtc.rtc->irq_freq)) - CMOS_WRITE(RTC_REF_CLCK_32KHZ | 0x06, RTC_FREQ_SELECT); + hpet_set_periodic_freq(cmos_rtc.rtc->irq_freq); + CMOS_WRITE(RTC_REF_CLCK_32KHZ | 0x06, RTC_FREQ_SELECT); /* disable irqs. * @@ -643,6 +693,8 @@ cmos_do_probe(struct device *dev, struct resource *ports, int rtc_irq) rtc_control = CMOS_READ(RTC_CONTROL); rtc_control &= ~(RTC_PIE | RTC_AIE | RTC_UIE); CMOS_WRITE(rtc_control, RTC_CONTROL); + hpet_mask_rtc_irq_bit(RTC_PIE | RTC_AIE | RTC_UIE); + CMOS_READ(RTC_INTR_FLAGS); spin_unlock_irq(&rtc_lock); @@ -690,7 +742,7 @@ cmos_do_probe(struct device *dev, struct resource *ports, int rtc_irq) goto cleanup2; } - pr_info("%s: alarms up to one %s%s\n", + pr_info("%s: alarms up to one %s%s%s\n", cmos_rtc.rtc->dev.bus_id, is_valid_irq(rtc_irq) ? (cmos_rtc.mon_alrm @@ -698,8 +750,8 @@ cmos_do_probe(struct device *dev, struct resource *ports, int rtc_irq) : (cmos_rtc.day_alrm ? "month" : "day")) : "no", - cmos_rtc.century ? ", y3k" : "" - ); + cmos_rtc.century ? ", y3k" : "", + is_hpet_enabled() ? ", hpet irqs" : ""); return 0; @@ -720,8 +772,10 @@ static void cmos_do_shutdown(void) spin_lock_irq(&rtc_lock); rtc_control = CMOS_READ(RTC_CONTROL); - rtc_control &= ~(RTC_PIE|RTC_AIE|RTC_UIE); + rtc_control &= ~RTC_IRQMASK; CMOS_WRITE(rtc_control, RTC_CONTROL); + hpet_mask_rtc_irq_bit(RTC_IRQMASK); + CMOS_READ(RTC_INTR_FLAGS); spin_unlock_irq(&rtc_lock); } @@ -764,12 +818,16 @@ static int cmos_suspend(struct device *dev, pm_message_t mesg) cmos->suspend_ctrl = tmp = CMOS_READ(RTC_CONTROL); if (tmp & (RTC_PIE|RTC_AIE|RTC_UIE)) { unsigned char irqstat; + unsigned char mask; if (do_wake) - tmp &= ~(RTC_PIE|RTC_UIE); + mask = RTC_IRQMASK & ~RTC_AIE; else - tmp &= ~(RTC_PIE|RTC_AIE|RTC_UIE); + mask = RTC_IRQMASK; + tmp &= ~mask; CMOS_WRITE(tmp, RTC_CONTROL); + hpet_mask_rtc_irq_bit(mask); + irqstat = CMOS_READ(RTC_INTR_FLAGS); irqstat &= (tmp & RTC_IRQMASK) | RTC_IRQF; if (is_intr(irqstat)) @@ -799,7 +857,8 @@ static int cmos_resume(struct device *dev) unsigned char tmp = cmos->suspend_ctrl; /* re-enable any irqs previously active */ - if (tmp & (RTC_PIE|RTC_AIE|RTC_UIE)) { + if (tmp & RTC_IRQMASK) { + unsigned char mask; if (cmos->enabled_wake) { if (cmos->wake_off) @@ -810,18 +869,28 @@ static int cmos_resume(struct device *dev) } spin_lock_irq(&rtc_lock); - CMOS_WRITE(tmp, RTC_CONTROL); - tmp = CMOS_READ(RTC_INTR_FLAGS); - tmp &= (cmos->suspend_ctrl & RTC_IRQMASK) | RTC_IRQF; - if (is_intr(tmp)) - rtc_update_irq(cmos->rtc, 1, tmp); + do { + CMOS_WRITE(tmp, RTC_CONTROL); + hpet_set_rtc_irq_bit(tmp & RTC_IRQMASK); + + mask = CMOS_READ(RTC_INTR_FLAGS); + mask &= (tmp & RTC_IRQMASK) | RTC_IRQF; + if (!is_intr(mask)) + break; + + /* force one-shot behavior if HPET blocked + * the wake alarm's irq + */ + rtc_update_irq(cmos->rtc, 1, mask); + tmp &= ~RTC_AIE; + hpet_mask_rtc_irq_bit(RTC_AIE); + } while (mask & RTC_AIE); spin_unlock_irq(&rtc_lock); } pr_debug("%s: resume, ctrl %02x\n", cmos_rtc.rtc->dev.bus_id, - cmos->suspend_ctrl); - + tmp); return 0; } From 4cd0c5c40b64ef9fd94fb8382dade2fd28f2b935 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Wed, 23 Jul 2008 21:30:44 -0700 Subject: [PATCH 355/857] rtc: rtc-s3c: add __devexit and __devinit markers Add the relevant __devinit and __devexit attributes to the rtc-s3c driver. Signed-off-by: Ben Dooks Acked-by: Alessandro Zummo Cc: David Brownell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/rtc/rtc-s3c.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/rtc/rtc-s3c.c b/drivers/rtc/rtc-s3c.c index fed86e507fdf..b81ba7020d92 100644 --- a/drivers/rtc/rtc-s3c.c +++ b/drivers/rtc/rtc-s3c.c @@ -430,7 +430,7 @@ static void s3c_rtc_enable(struct platform_device *pdev, int en) } } -static int s3c_rtc_remove(struct platform_device *dev) +static int __devexit s3c_rtc_remove(struct platform_device *dev) { struct rtc_device *rtc = platform_get_drvdata(dev); @@ -447,7 +447,7 @@ static int s3c_rtc_remove(struct platform_device *dev) return 0; } -static int s3c_rtc_probe(struct platform_device *pdev) +static int __devinit s3c_rtc_probe(struct platform_device *pdev) { struct rtc_device *rtc; struct resource *res; @@ -560,7 +560,7 @@ static int s3c_rtc_resume(struct platform_device *pdev) static struct platform_driver s3c2410_rtcdrv = { .probe = s3c_rtc_probe, - .remove = s3c_rtc_remove, + .remove = __devexit_p(s3c_rtc_remove), .suspend = s3c_rtc_suspend, .resume = s3c_rtc_resume, .driver = { From 773be7ee97c11fbb6b8a912a58b268dbe8a6a3fe Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Wed, 23 Jul 2008 21:30:45 -0700 Subject: [PATCH 356/857] rtc: rtc-s3c: update IRQ handling The rtc-s3c.c driver has been using its own ioctl() handling to deal with alarm and periodic interrupts to handle what should now be done with the rtc core code. Change to using the .irq_set_freq and .irq_set_state driver entries and remove the .ioctl handling. Signed-off-by: Ben Dooks Acked-by: Alessandro Zummo Cc: David Brownell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/rtc/rtc-s3c.c | 83 +++++++++---------------------------------- 1 file changed, 16 insertions(+), 67 deletions(-) diff --git a/drivers/rtc/rtc-s3c.c b/drivers/rtc/rtc-s3c.c index b81ba7020d92..54b1ebb01502 100644 --- a/drivers/rtc/rtc-s3c.c +++ b/drivers/rtc/rtc-s3c.c @@ -36,10 +36,8 @@ static struct resource *s3c_rtc_mem; static void __iomem *s3c_rtc_base; static int s3c_rtc_alarmno = NO_IRQ; static int s3c_rtc_tickno = NO_IRQ; -static int s3c_rtc_freq = 1; static DEFINE_SPINLOCK(s3c_rtc_pie_lock); -static unsigned int tick_count; /* IRQ Handlers */ @@ -55,7 +53,7 @@ static irqreturn_t s3c_rtc_tickirq(int irq, void *id) { struct rtc_device *rdev = id; - rtc_update_irq(rdev, tick_count++, RTC_PF | RTC_IRQF); + rtc_update_irq(rdev, 1, RTC_PF | RTC_IRQF); return IRQ_HANDLED; } @@ -74,35 +72,37 @@ static void s3c_rtc_setaie(int to) writeb(tmp, s3c_rtc_base + S3C2410_RTCALM); } -static void s3c_rtc_setpie(int to) +static int s3c_rtc_setpie(struct device *dev, int enabled) { unsigned int tmp; - pr_debug("%s: pie=%d\n", __func__, to); + pr_debug("%s: pie=%d\n", __func__, enabled); spin_lock_irq(&s3c_rtc_pie_lock); tmp = readb(s3c_rtc_base + S3C2410_TICNT) & ~S3C2410_TICNT_ENABLE; - if (to) + if (enabled) tmp |= S3C2410_TICNT_ENABLE; writeb(tmp, s3c_rtc_base + S3C2410_TICNT); spin_unlock_irq(&s3c_rtc_pie_lock); + + return 0; } -static void s3c_rtc_setfreq(int freq) +static int s3c_rtc_setfreq(struct device *dev, int freq) { unsigned int tmp; spin_lock_irq(&s3c_rtc_pie_lock); + tmp = readb(s3c_rtc_base + S3C2410_TICNT) & S3C2410_TICNT_ENABLE; - - s3c_rtc_freq = freq; - tmp |= (128 / freq)-1; writeb(tmp, s3c_rtc_base + S3C2410_TICNT); spin_unlock_irq(&s3c_rtc_pie_lock); + + return 0; } /* Time read/write */ @@ -267,12 +267,7 @@ static int s3c_rtc_setalarm(struct device *dev, struct rtc_wkalrm *alrm) writeb(alrm_en, base + S3C2410_RTCALM); - if (0) { - alrm_en = readb(base + S3C2410_RTCALM); - alrm_en &= ~S3C2410_RTCALM_ALMEN; - writeb(alrm_en, base + S3C2410_RTCALM); - disable_irq_wake(s3c_rtc_alarmno); - } + s3c_rtc_setaie(alrm->enabled); if (alrm->enabled) enable_irq_wake(s3c_rtc_alarmno); @@ -282,59 +277,12 @@ static int s3c_rtc_setalarm(struct device *dev, struct rtc_wkalrm *alrm) return 0; } -static int s3c_rtc_ioctl(struct device *dev, - unsigned int cmd, unsigned long arg) -{ - unsigned int ret = -ENOIOCTLCMD; - - switch (cmd) { - case RTC_AIE_OFF: - case RTC_AIE_ON: - s3c_rtc_setaie((cmd == RTC_AIE_ON) ? 1 : 0); - ret = 0; - break; - - case RTC_PIE_OFF: - case RTC_PIE_ON: - tick_count = 0; - s3c_rtc_setpie((cmd == RTC_PIE_ON) ? 1 : 0); - ret = 0; - break; - - case RTC_IRQP_READ: - ret = put_user(s3c_rtc_freq, (unsigned long __user *)arg); - break; - - case RTC_IRQP_SET: - if (!is_power_of_2(arg)) { - ret = -EINVAL; - goto exit; - } - - pr_debug("s3c2410_rtc: setting frequency %ld\n", arg); - - s3c_rtc_setfreq(arg); - ret = 0; - break; - - case RTC_UIE_ON: - case RTC_UIE_OFF: - ret = -EINVAL; - } - - exit: - return ret; -} - static int s3c_rtc_proc(struct device *dev, struct seq_file *seq) { unsigned int ticnt = readb(s3c_rtc_base + S3C2410_TICNT); seq_printf(seq, "periodic_IRQ\t: %s\n", (ticnt & S3C2410_TICNT_ENABLE) ? "yes" : "no" ); - - seq_printf(seq, "periodic_freq\t: %d\n", s3c_rtc_freq); - return 0; } @@ -374,7 +322,7 @@ static void s3c_rtc_release(struct device *dev) /* do not clear AIE here, it may be needed for wake */ - s3c_rtc_setpie(0); + s3c_rtc_setpie(dev, 0); free_irq(s3c_rtc_alarmno, rtc_dev); free_irq(s3c_rtc_tickno, rtc_dev); } @@ -382,11 +330,12 @@ static void s3c_rtc_release(struct device *dev) static const struct rtc_class_ops s3c_rtcops = { .open = s3c_rtc_open, .release = s3c_rtc_release, - .ioctl = s3c_rtc_ioctl, .read_time = s3c_rtc_gettime, .set_time = s3c_rtc_settime, .read_alarm = s3c_rtc_getalarm, .set_alarm = s3c_rtc_setalarm, + .irq_set_freq = s3c_rtc_setfreq, + .irq_set_state = s3c_rtc_setpie, .proc = s3c_rtc_proc, }; @@ -437,7 +386,7 @@ static int __devexit s3c_rtc_remove(struct platform_device *dev) platform_set_drvdata(dev, NULL); rtc_device_unregister(rtc); - s3c_rtc_setpie(0); + s3c_rtc_setpie(&dev->dev, 0); s3c_rtc_setaie(0); iounmap(s3c_rtc_base); @@ -504,7 +453,7 @@ static int __devinit s3c_rtc_probe(struct platform_device *pdev) pr_debug("s3c2410_rtc: RTCCON=%02x\n", readb(s3c_rtc_base + S3C2410_RTCCON)); - s3c_rtc_setfreq(s3c_rtc_freq); + s3c_rtc_setfreq(&pdev->dev, 1); /* register RTC and exit */ From 449321b39f6c6ebfa15d6da24f134240bd51db29 Mon Sep 17 00:00:00 2001 From: David Brownell Date: Wed, 23 Jul 2008 21:30:46 -0700 Subject: [PATCH 357/857] rtc-at91rm9200: avoid spurious irqs This fixes kernel http://bugzilla.kernel.org/show_bug.cgi?id=11112 (bogus RTC update IRQs reported) for rtc-at91rm9200 by scrubbing old IRQ status before enabling IRQs. It also removes nonfunctional periodic IRQ support from this driver; only update IRQs are reported, or provided by the hardware. I suspect some other RTCs probably have versions of #11112; it's easy to overlook, since most non-RTC drivers don't care about spurious IRQs: they're not reported to userspace. Signed-off-by: David Brownell Report-by: W Unruh Cc: Andrew Victor Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/rtc/rtc-at91rm9200.c | 16 ++++++---------- 1 file changed, 6 insertions(+), 10 deletions(-) diff --git a/drivers/rtc/rtc-at91rm9200.c b/drivers/rtc/rtc-at91rm9200.c index 9c3db934cc24..cd32d05db773 100644 --- a/drivers/rtc/rtc-at91rm9200.c +++ b/drivers/rtc/rtc-at91rm9200.c @@ -171,8 +171,10 @@ static int at91_rtc_setalarm(struct device *dev, struct rtc_wkalrm *alrm) | BIN2BCD(tm.tm_mday) << 24 | AT91_RTC_DATEEN | AT91_RTC_MTHEN); - if (alrm->enabled) + if (alrm->enabled) { + at91_sys_write(AT91_RTC_SCCR, AT91_RTC_ALARM); at91_sys_write(AT91_RTC_IER, AT91_RTC_ALARM); + } pr_debug("%s(): %4d-%02d-%02d %02d:%02d:%02d\n", __func__, at91_alarm_year, tm.tm_mon, tm.tm_mday, tm.tm_hour, @@ -191,28 +193,22 @@ static int at91_rtc_ioctl(struct device *dev, unsigned int cmd, pr_debug("%s(): cmd=%08x, arg=%08lx.\n", __func__, cmd, arg); + /* important: scrub old status before enabling IRQs */ switch (cmd) { case RTC_AIE_OFF: /* alarm off */ at91_sys_write(AT91_RTC_IDR, AT91_RTC_ALARM); break; case RTC_AIE_ON: /* alarm on */ + at91_sys_write(AT91_RTC_SCCR, AT91_RTC_ALARM); at91_sys_write(AT91_RTC_IER, AT91_RTC_ALARM); break; case RTC_UIE_OFF: /* update off */ - case RTC_PIE_OFF: /* periodic off */ at91_sys_write(AT91_RTC_IDR, AT91_RTC_SECEV); break; case RTC_UIE_ON: /* update on */ - case RTC_PIE_ON: /* periodic on */ + at91_sys_write(AT91_RTC_SCCR, AT91_RTC_SECEV); at91_sys_write(AT91_RTC_IER, AT91_RTC_SECEV); break; - case RTC_IRQP_READ: /* read periodic alarm frequency */ - ret = put_user(AT91_RTC_FREQ, (unsigned long *) arg); - break; - case RTC_IRQP_SET: /* set periodic alarm frequency */ - if (arg != AT91_RTC_FREQ) - ret = -EINVAL; - break; default: ret = -ENOIOCTLCMD; break; From 7e2a31da854dcf8324012a83a31b40bc11e52589 Mon Sep 17 00:00:00 2001 From: David Brownell Date: Wed, 23 Jul 2008 21:30:47 -0700 Subject: [PATCH 358/857] rtc-cmos: avoid spurious irqs This fixes kernel http://bugzilla.kernel.org/show_bug.cgi?id=11112 (bogus RTC update IRQs reported) for rtc-cmos, in two ways: - When HPET is stealing the IRQs, use the first IRQ to grab the seconds counter which will be monitored (instead of using whatever was previously in that memory); - In sane IRQ handling modes, scrub out old IRQ status before enabling IRQs. That latter is done by tightening up IRQ handling for rtc-cmos everywhere, also ensuring that when HPET is used it's the only thing triggering IRQ reports to userspace; net object shrink. Also fix a bogus HPET message related to its RTC emulation. Signed-off-by: David Brownell Report-by: W Unruh Cc: Andrew Victor Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/kernel/hpet.c | 10 ++- drivers/rtc/rtc-cmos.c | 140 +++++++++++++++++++---------------------- 2 files changed, 70 insertions(+), 80 deletions(-) diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index 0ea6a19bfdfe..ad2b15a1334d 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -468,7 +468,7 @@ void hpet_disable(void) #define RTC_NUM_INTS 1 static unsigned long hpet_rtc_flags; -static unsigned long hpet_prev_update_sec; +static int hpet_prev_update_sec; static struct rtc_time hpet_alarm_time; static unsigned long hpet_pie_count; static unsigned long hpet_t1_cmp; @@ -575,6 +575,9 @@ int hpet_set_rtc_irq_bit(unsigned long bit_mask) hpet_rtc_flags |= bit_mask; + if ((bit_mask & RTC_UIE) && !(oldbits & RTC_UIE)) + hpet_prev_update_sec = -1; + if (!oldbits) hpet_rtc_timer_init(); @@ -652,7 +655,7 @@ static void hpet_rtc_timer_reinit(void) if (hpet_rtc_flags & RTC_PIE) hpet_pie_count += lost_ints; if (printk_ratelimit()) - printk(KERN_WARNING "rtc: lost %d interrupts\n", + printk(KERN_WARNING "hpet1: lost %d rtc interrupts\n", lost_ints); } } @@ -670,7 +673,8 @@ irqreturn_t hpet_rtc_interrupt(int irq, void *dev_id) if (hpet_rtc_flags & RTC_UIE && curr_time.tm_sec != hpet_prev_update_sec) { - rtc_int_flag = RTC_UF; + if (hpet_prev_update_sec >= 0) + rtc_int_flag = RTC_UF; hpet_prev_update_sec = curr_time.tm_sec; } diff --git a/drivers/rtc/rtc-cmos.c b/drivers/rtc/rtc-cmos.c index e9984650ea95..6ea349aba3ba 100644 --- a/drivers/rtc/rtc-cmos.c +++ b/drivers/rtc/rtc-cmos.c @@ -235,11 +235,56 @@ static int cmos_read_alarm(struct device *dev, struct rtc_wkalrm *t) return 0; } +static void cmos_checkintr(struct cmos_rtc *cmos, unsigned char rtc_control) +{ + unsigned char rtc_intr; + + /* NOTE after changing RTC_xIE bits we always read INTR_FLAGS; + * allegedly some older rtcs need that to handle irqs properly + */ + rtc_intr = CMOS_READ(RTC_INTR_FLAGS); + + if (is_hpet_enabled()) + return; + + rtc_intr &= (rtc_control & RTC_IRQMASK) | RTC_IRQF; + if (is_intr(rtc_intr)) + rtc_update_irq(cmos->rtc, 1, rtc_intr); +} + +static void cmos_irq_enable(struct cmos_rtc *cmos, unsigned char mask) +{ + unsigned char rtc_control; + + /* flush any pending IRQ status, notably for update irqs, + * before we enable new IRQs + */ + rtc_control = CMOS_READ(RTC_CONTROL); + cmos_checkintr(cmos, rtc_control); + + rtc_control |= mask; + CMOS_WRITE(rtc_control, RTC_CONTROL); + hpet_set_rtc_irq_bit(mask); + + cmos_checkintr(cmos, rtc_control); +} + +static void cmos_irq_disable(struct cmos_rtc *cmos, unsigned char mask) +{ + unsigned char rtc_control; + + rtc_control = CMOS_READ(RTC_CONTROL); + rtc_control &= ~mask; + CMOS_WRITE(rtc_control, RTC_CONTROL); + hpet_mask_rtc_irq_bit(mask); + + cmos_checkintr(cmos, rtc_control); +} + static int cmos_set_alarm(struct device *dev, struct rtc_wkalrm *t) { struct cmos_rtc *cmos = dev_get_drvdata(dev); unsigned char mon, mday, hrs, min, sec; - unsigned char rtc_control, rtc_intr; if (!is_valid_irq(cmos->irq)) return -EIO; @@ -266,15 +311,7 @@ static int cmos_set_alarm(struct device *dev, struct rtc_wkalrm *t) spin_lock_irq(&rtc_lock); /* next rtc irq must not be from previous alarm setting */ - rtc_control = CMOS_READ(RTC_CONTROL); - rtc_control &= ~RTC_AIE; - CMOS_WRITE(rtc_control, RTC_CONTROL); - hpet_mask_rtc_irq_bit(RTC_AIE); - - rtc_intr = CMOS_READ(RTC_INTR_FLAGS); - rtc_intr &= (rtc_control & RTC_IRQMASK) | RTC_IRQF; - if (is_intr(rtc_intr)) - rtc_update_irq(cmos->rtc, 1, rtc_intr); + cmos_irq_disable(cmos, RTC_AIE); /* update alarm */ CMOS_WRITE(hrs, RTC_HOURS_ALARM); @@ -293,16 +330,8 @@ static int cmos_set_alarm(struct device *dev, struct rtc_wkalrm *t) */ hpet_set_alarm_time(t->time.tm_hour, t->time.tm_min, t->time.tm_sec); - if (t->enabled) { - rtc_control |= RTC_AIE; - CMOS_WRITE(rtc_control, RTC_CONTROL); - hpet_set_rtc_irq_bit(RTC_AIE); - - rtc_intr = CMOS_READ(RTC_INTR_FLAGS); - rtc_intr &= (rtc_control & RTC_IRQMASK) | RTC_IRQF; - if (is_intr(rtc_intr)) - rtc_update_irq(cmos->rtc, 1, rtc_intr); - } + if (t->enabled) + cmos_irq_enable(cmos, RTC_AIE); spin_unlock_irq(&rtc_lock); @@ -335,28 +364,17 @@ static int cmos_irq_set_freq(struct device *dev, int freq) static int cmos_irq_set_state(struct device *dev, int enabled) { struct cmos_rtc *cmos = dev_get_drvdata(dev); - unsigned char rtc_control, rtc_intr; unsigned long flags; if (!is_valid_irq(cmos->irq)) return -ENXIO; spin_lock_irqsave(&rtc_lock, flags); - rtc_control = CMOS_READ(RTC_CONTROL); - if (enabled) { - rtc_control |= RTC_PIE; - hpet_set_rtc_irq_bit(RTC_PIE); - } else { - rtc_control &= ~RTC_PIE; - hpet_mask_rtc_irq_bit(RTC_PIE); - } - CMOS_WRITE(rtc_control, RTC_CONTROL); - - rtc_intr = CMOS_READ(RTC_INTR_FLAGS); - rtc_intr &= (rtc_control & RTC_IRQMASK) | RTC_IRQF; - if (is_intr(rtc_intr)) - rtc_update_irq(cmos->rtc, 1, rtc_intr); + if (enabled) + cmos_irq_enable(cmos, RTC_PIE); + else + cmos_irq_disable(cmos, RTC_PIE); spin_unlock_irqrestore(&rtc_lock, flags); return 0; @@ -368,7 +386,6 @@ static int cmos_rtc_ioctl(struct device *dev, unsigned int cmd, unsigned long arg) { struct cmos_rtc *cmos = dev_get_drvdata(dev); - unsigned char rtc_control, rtc_intr; unsigned long flags; switch (cmd) { @@ -385,32 +402,20 @@ cmos_rtc_ioctl(struct device *dev, unsigned int cmd, unsigned long arg) } spin_lock_irqsave(&rtc_lock, flags); - rtc_control = CMOS_READ(RTC_CONTROL); switch (cmd) { case RTC_AIE_OFF: /* alarm off */ - rtc_control &= ~RTC_AIE; - hpet_mask_rtc_irq_bit(RTC_AIE); + cmos_irq_disable(cmos, RTC_AIE); break; case RTC_AIE_ON: /* alarm on */ - rtc_control |= RTC_AIE; - hpet_set_rtc_irq_bit(RTC_AIE); + cmos_irq_enable(cmos, RTC_AIE); break; case RTC_UIE_OFF: /* update off */ - rtc_control &= ~RTC_UIE; - hpet_mask_rtc_irq_bit(RTC_UIE); + cmos_irq_disable(cmos, RTC_UIE); break; case RTC_UIE_ON: /* update on */ - rtc_control |= RTC_UIE; - hpet_set_rtc_irq_bit(RTC_UIE); + cmos_irq_enable(cmos, RTC_UIE); break; } - CMOS_WRITE(rtc_control, RTC_CONTROL); - - rtc_intr = CMOS_READ(RTC_INTR_FLAGS); - rtc_intr &= (rtc_control & RTC_IRQMASK) | RTC_IRQF; - if (is_intr(rtc_intr)) - rtc_update_irq(cmos->rtc, 1, rtc_intr); - spin_unlock_irqrestore(&rtc_lock, flags); return 0; } @@ -571,7 +576,6 @@ static irqreturn_t cmos_interrupt(int irq, void *p) * alarm woke the system. */ if (irqstat & RTC_AIE) { - rtc_control = CMOS_READ(RTC_CONTROL); rtc_control &= ~RTC_AIE; CMOS_WRITE(rtc_control, RTC_CONTROL); hpet_mask_rtc_irq_bit(RTC_AIE); @@ -685,17 +689,10 @@ cmos_do_probe(struct device *dev, struct resource *ports, int rtc_irq) hpet_set_periodic_freq(cmos_rtc.rtc->irq_freq); CMOS_WRITE(RTC_REF_CLCK_32KHZ | 0x06, RTC_FREQ_SELECT); - /* disable irqs. - * - * NOTE after changing RTC_xIE bits we always read INTR_FLAGS; - * allegedly some older rtcs need that to handle irqs properly - */ - rtc_control = CMOS_READ(RTC_CONTROL); - rtc_control &= ~(RTC_PIE | RTC_AIE | RTC_UIE); - CMOS_WRITE(rtc_control, RTC_CONTROL); - hpet_mask_rtc_irq_bit(RTC_PIE | RTC_AIE | RTC_UIE); + /* disable irqs */ + cmos_irq_disable(&cmos_rtc, RTC_PIE | RTC_AIE | RTC_UIE); - CMOS_READ(RTC_INTR_FLAGS); + rtc_control = CMOS_READ(RTC_CONTROL); spin_unlock_irq(&rtc_lock); @@ -768,15 +765,8 @@ cmos_do_probe(struct device *dev, struct resource *ports, int rtc_irq) static void cmos_do_shutdown(void) { - unsigned char rtc_control; - spin_lock_irq(&rtc_lock); - rtc_control = CMOS_READ(RTC_CONTROL); - rtc_control &= ~RTC_IRQMASK; - CMOS_WRITE(rtc_control, RTC_CONTROL); - hpet_mask_rtc_irq_bit(RTC_IRQMASK); - - CMOS_READ(RTC_INTR_FLAGS); + cmos_irq_disable(&cmos_rtc, RTC_IRQMASK); spin_unlock_irq(&rtc_lock); } @@ -817,7 +807,6 @@ static int cmos_suspend(struct device *dev, pm_message_t mesg) spin_lock_irq(&rtc_lock); cmos->suspend_ctrl = tmp = CMOS_READ(RTC_CONTROL); if (tmp & (RTC_PIE|RTC_AIE|RTC_UIE)) { - unsigned char irqstat; unsigned char mask; if (do_wake) @@ -828,10 +817,7 @@ static int cmos_suspend(struct device *dev, pm_message_t mesg) CMOS_WRITE(tmp, RTC_CONTROL); hpet_mask_rtc_irq_bit(mask); - irqstat = CMOS_READ(RTC_INTR_FLAGS); - irqstat &= (tmp & RTC_IRQMASK) | RTC_IRQF; - if (is_intr(irqstat)) - rtc_update_irq(cmos->rtc, 1, irqstat); + cmos_checkintr(cmos, tmp); } spin_unlock_irq(&rtc_lock); @@ -875,7 +861,7 @@ static int cmos_resume(struct device *dev) mask = CMOS_READ(RTC_INTR_FLAGS); mask &= (tmp & RTC_IRQMASK) | RTC_IRQF; - if (!is_intr(mask)) + if (!is_hpet_enabled() || !is_intr(mask)) break; /* force one-shot behavior if HPET blocked From 4cad4431fcd872a1b2efc093b0db6df943f5a898 Mon Sep 17 00:00:00 2001 From: Yoichi Yuasa Date: Wed, 23 Jul 2008 21:30:48 -0700 Subject: [PATCH 359/857] rtc-vr41xx: add irq_set_freq() and irq_set_state() Implement the ioctls RTC_PIE_ON, RTC_PIE_OFF, RTC_IRQP_SET and RTC_IRQP_READ in the standard RTC way. Thanks Dave for noticing it. Signed-off-by: Yoichi Yuasa Cc: David Brownell Cc: Ralf Baechle Cc: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/rtc/rtc-vr41xx.c | 65 +++++++++++++++++++++------------------- 1 file changed, 34 insertions(+), 31 deletions(-) diff --git a/drivers/rtc/rtc-vr41xx.c b/drivers/rtc/rtc-vr41xx.c index be9c70d0b193..884b635f028b 100644 --- a/drivers/rtc/rtc-vr41xx.c +++ b/drivers/rtc/rtc-vr41xx.c @@ -1,7 +1,7 @@ /* * Driver for NEC VR4100 series Real Time Clock unit. * - * Copyright (C) 2003-2006 Yoichi Yuasa + * Copyright (C) 2003-2008 Yoichi Yuasa * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -34,7 +34,7 @@ MODULE_AUTHOR("Yoichi Yuasa "); MODULE_DESCRIPTION("NEC VR4100 series RTC driver"); -MODULE_LICENSE("GPL"); +MODULE_LICENSE("GPL v2"); /* RTC 1 registers */ #define ETIMELREG 0x00 @@ -82,7 +82,6 @@ static unsigned long epoch = 1970; /* Jan 1 1970 00:00:00 */ static DEFINE_SPINLOCK(rtc_lock); static char rtc_name[] = "RTC"; -static unsigned long periodic_frequency; static unsigned long periodic_count; static unsigned int alarm_enabled; static int aie_irq = -1; @@ -207,10 +206,37 @@ static int vr41xx_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *wkalrm) return 0; } -static int vr41xx_rtc_ioctl(struct device *dev, unsigned int cmd, unsigned long arg) +static int vr41xx_rtc_irq_set_freq(struct device *dev, int freq) { unsigned long count; + count = RTC_FREQUENCY; + do_div(count, freq); + + periodic_count = count; + + spin_lock_irq(&rtc_lock); + + rtc1_write(RTCL1LREG, count); + rtc1_write(RTCL1HREG, count >> 16); + + spin_unlock_irq(&rtc_lock); + + return 0; +} + +static int vr41xx_rtc_irq_set_state(struct device *dev, int enabled) +{ + if (enabled) + enable_irq(pie_irq); + else + disable_irq(pie_irq); + + return 0; +} + +static int vr41xx_rtc_ioctl(struct device *dev, unsigned int cmd, unsigned long arg) +{ switch (cmd) { case RTC_AIE_ON: spin_lock_irq(&rtc_lock); @@ -230,33 +256,6 @@ static int vr41xx_rtc_ioctl(struct device *dev, unsigned int cmd, unsigned long alarm_enabled = 0; } - spin_unlock_irq(&rtc_lock); - break; - case RTC_PIE_ON: - enable_irq(pie_irq); - break; - case RTC_PIE_OFF: - disable_irq(pie_irq); - break; - case RTC_IRQP_READ: - return put_user(periodic_frequency, (unsigned long __user *)arg); - break; - case RTC_IRQP_SET: - if (arg > MAX_PERIODIC_RATE) - return -EINVAL; - - periodic_frequency = arg; - - count = RTC_FREQUENCY; - do_div(count, arg); - - periodic_count = count; - - spin_lock_irq(&rtc_lock); - - rtc1_write(RTCL1LREG, count); - rtc1_write(RTCL1HREG, count >> 16); - spin_unlock_irq(&rtc_lock); break; case RTC_EPOCH_READ: @@ -309,6 +308,8 @@ static const struct rtc_class_ops vr41xx_rtc_ops = { .set_time = vr41xx_rtc_set_time, .read_alarm = vr41xx_rtc_read_alarm, .set_alarm = vr41xx_rtc_set_alarm, + .irq_set_freq = vr41xx_rtc_irq_set_freq, + .irq_set_state = vr41xx_rtc_irq_set_state, }; static int __devinit rtc_probe(struct platform_device *pdev) @@ -346,6 +347,8 @@ static int __devinit rtc_probe(struct platform_device *pdev) goto err_iounmap_all; } + rtc->max_user_freq = MAX_PERIODIC_RATE; + spin_lock_irq(&rtc_lock); rtc1_write(ECMPLREG, 0); From 2ece5f43b041b96fa2a05107a10a6b0ea0c03a3b Mon Sep 17 00:00:00 2001 From: Sebastian Siewior Date: Wed, 23 Jul 2008 21:30:49 -0700 Subject: [PATCH 360/857] fbdev: add the carmine FB driver Basic FB driver for the carmine chip. The driver registers two FB devices for the two possible screens. The DRAM settings can be be switched via Kconfig (between eval board and custom). Signed-off-by: Sebastian Siewior Cc: "Antonino A. Daplas" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/Kconfig | 26 ++ drivers/video/Makefile | 1 + drivers/video/carminefb.c | 790 +++++++++++++++++++++++++++++++++ drivers/video/carminefb.h | 64 +++ drivers/video/carminefb_regs.h | 159 +++++++ 5 files changed, 1040 insertions(+) create mode 100644 drivers/video/carminefb.c create mode 100644 drivers/video/carminefb.h create mode 100644 drivers/video/carminefb_regs.h diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig index 9b887ef64ff1..7072d2c5a040 100644 --- a/drivers/video/Kconfig +++ b/drivers/video/Kconfig @@ -1658,6 +1658,32 @@ config FB_PM3 similar boards, 3DLabs Permedia3 Create!, Appian Jeronimo 2000 and maybe other boards. +config FB_CARMINE + tristate "Fujitsu carmine frame buffer support" + depends on FB && PCI + select FB_CFB_FILLRECT + select FB_CFB_COPYAREA + select FB_CFB_IMAGEBLIT + help + This is the frame buffer device driver for the Fujitsu Carmine chip. + The driver provides two independent frame buffer devices. + +choice + depends on FB_CARMINE + prompt "DRAM timing" + default FB_CARMINE_DRAM_EVAL + +config FB_CARMINE_DRAM_EVAL + bool "Eval board timings" + help + Use timings which work on the eval card. + +config CARMINE_DRAM_CUSTOM + bool "Custom board timings" + help + Use custom board timings. +endchoice + config FB_AU1100 bool "Au1100 LCD Driver" depends on (FB = y) && MIPS && SOC_AU1100 diff --git a/drivers/video/Makefile b/drivers/video/Makefile index 04bca35403ff..7ee85c0d2e5f 100644 --- a/drivers/video/Makefile +++ b/drivers/video/Makefile @@ -117,6 +117,7 @@ obj-$(CONFIG_FB_SM501) += sm501fb.o obj-$(CONFIG_FB_XILINX) += xilinxfb.o obj-$(CONFIG_FB_OMAP) += omap/ obj-$(CONFIG_XEN_FBDEV_FRONTEND) += xen-fbfront.o +obj-$(CONFIG_FB_CARMINE) += carminefb.o # Platform or fallback drivers go here obj-$(CONFIG_FB_UVESA) += uvesafb.o diff --git a/drivers/video/carminefb.c b/drivers/video/carminefb.c new file mode 100644 index 000000000000..e15bb447440a --- /dev/null +++ b/drivers/video/carminefb.c @@ -0,0 +1,790 @@ +/* + * Frame buffer driver for the Carmine GPU. + * + * The driver configures the GPU as follows + * - FB0 is display 0 with unique memory area + * - FB1 is display 1 with unique memory area + * - both display use 32 bit colors + */ +#include +#include +#include +#include +#include + +#include "carminefb.h" +#include "carminefb_regs.h" + +#if !defined(__LITTLE_ENDIAN) && !defined(__BIG_ENDIAN) +#error "The endianness of the target host has not been defined." +#endif + +/* + * The initial video mode can be supplied via two different ways: + * - as a string that is passed to fb_find_mode() (module option fb_mode_str) + * - as an integer that picks the video mode from carmine_modedb[] (module + * option fb_mode) + * + * If nothing is used than the initial video mode will be the + * CARMINEFB_DEFAULT_VIDEO_MODE member of the carmine_modedb[]. + */ +#define CARMINEFB_DEFAULT_VIDEO_MODE 1 + +static unsigned int fb_mode = CARMINEFB_DEFAULT_VIDEO_MODE; +module_param(fb_mode, uint, 444); +MODULE_PARM_DESC(fb_mode, "Initial video mode as integer."); + +static char *fb_mode_str; +module_param(fb_mode_str, charp, 444); +MODULE_PARM_DESC(fb_mode_str, "Initial video mode in characters."); + +/* + * Carminefb displays: + * 0b000 None + * 0b001 Display 0 + * 0b010 Display 1 + */ +static int fb_displays = CARMINE_USE_DISPLAY0 | CARMINE_USE_DISPLAY1; +module_param(fb_displays, int, 444); +MODULE_PARM_DESC(fb_displays, "Bit mode, which displays are used"); + +struct carmine_hw { + void __iomem *v_regs; + void __iomem *screen_mem; + struct fb_info *fb[MAX_DISPLAY]; +}; + +struct carmine_resolution { + u32 htp; + u32 hsp; + u32 hsw; + u32 hdp; + u32 vtr; + u32 vsp; + u32 vsw; + u32 vdp; + u32 disp_mode; +}; + +struct carmine_fb { + void __iomem *display_reg; + void __iomem *screen_base; + u32 smem_offset; + u32 cur_mode; + u32 new_mode; + struct carmine_resolution *res; + u32 pseudo_palette[16]; +}; + +static struct fb_fix_screeninfo carminefb_fix __devinitdata = { + .id = "Carmine", + .type = FB_TYPE_PACKED_PIXELS, + .visual = FB_VISUAL_TRUECOLOR, + .accel = FB_ACCEL_NONE, +}; + +static const struct fb_videomode carmine_modedb[] = { + { + .name = "640x480", + .xres = 640, + .yres = 480, + }, { + .name = "800x600", + .xres = 800, + .yres = 600, + }, +}; + +static struct carmine_resolution car_modes[] = { + { + /* 640x480 */ + .htp = 800, + .hsp = 672, + .hsw = 96, + .hdp = 640, + .vtr = 525, + .vsp = 490, + .vsw = 2, + .vdp = 480, + .disp_mode = 0x1400, + }, + { + /* 800x600 */ + .htp = 1060, + .hsp = 864, + .hsw = 72, + .hdp = 800, + .vtr = 628, + .vsp = 601, + .vsw = 2, + .vdp = 600, + .disp_mode = 0x0d00, + } +}; + +static int carmine_find_mode(const struct fb_var_screeninfo *var) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(car_modes); i++) + if (car_modes[i].hdp == var->xres && + car_modes[i].vdp == var->yres) + return i; + return -EINVAL; +} + +static void c_set_disp_reg(const struct carmine_fb *par, + u32 offset, u32 val) +{ + writel(val, par->display_reg + offset); +} + +static u32 c_get_disp_reg(const struct carmine_fb *par, + u32 offset) +{ + return readl(par->display_reg + offset); +} + +static void c_set_hw_reg(const struct carmine_hw *hw, + u32 offset, u32 val) +{ + writel(val, hw->v_regs + offset); +} + +static u32 c_get_hw_reg(const struct carmine_hw *hw, + u32 offset) +{ + return readl(hw->v_regs + offset); +} + +static int carmine_setcolreg(unsigned regno, unsigned red, unsigned green, + unsigned blue, unsigned transp, struct fb_info *info) +{ + if (regno >= 16) + return 1; + + red >>= 8; + green >>= 8; + blue >>= 8; + transp >>= 8; + + ((u32 *)info->pseudo_palette)[regno] = be32_to_cpu(transp << 24 | + red << 0 | green << 8 | blue << 16); + return 0; +} + +static int carmine_check_var(struct fb_var_screeninfo *var, + struct fb_info *info) +{ + int ret; + + ret = carmine_find_mode(var); + if (ret < 0) + return ret; + + if (var->grayscale || var->rotate || var->nonstd) + return -EINVAL; + + var->xres_virtual = var->xres; + var->yres_virtual = var->yres; + + var->bits_per_pixel = 32; + +#ifdef __BIG_ENDIAN + var->transp.offset = 24; + var->red.offset = 0; + var->green.offset = 8; + var->blue.offset = 16; +#else + var->transp.offset = 24; + var->red.offset = 16; + var->green.offset = 8; + var->blue.offset = 0; +#endif + + var->red.length = 8; + var->green.length = 8; + var->blue.length = 8; + var->transp.length = 8; + + var->red.msb_right = 0; + var->green.msb_right = 0; + var->blue.msb_right = 0; + var->transp.msb_right = 0; + return 0; +} + +static void carmine_init_display_param(struct carmine_fb *par) +{ + u32 width; + u32 height; + u32 param; + u32 window_size; + u32 soffset = par->smem_offset; + + c_set_disp_reg(par, CARMINE_DISP_REG_C_TRANS, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_MLMR_TRANS, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_CURSOR_MODE, + CARMINE_CURSOR0_PRIORITY_MASK | + CARMINE_CURSOR1_PRIORITY_MASK | + CARMINE_CURSOR_CUTZ_MASK); + + /* Set default cursor position */ + c_set_disp_reg(par, CARMINE_DISP_REG_CUR1_POS, 0 << 16 | 0); + c_set_disp_reg(par, CARMINE_DISP_REG_CUR2_POS, 0 << 16 | 0); + + /* Set default display mode */ + c_set_disp_reg(par, CARMINE_DISP_REG_L0_EXT_MODE, CARMINE_WINDOW_MODE | + CARMINE_EXT_CMODE_DIRECT24_RGBA); + c_set_disp_reg(par, CARMINE_DISP_REG_L1_EXT_MODE, + CARMINE_EXT_CMODE_DIRECT24_RGBA); + c_set_disp_reg(par, CARMINE_DISP_REG_L2_EXT_MODE, CARMINE_EXTEND_MODE | + CARMINE_EXT_CMODE_DIRECT24_RGBA); + c_set_disp_reg(par, CARMINE_DISP_REG_L3_EXT_MODE, CARMINE_EXTEND_MODE | + CARMINE_EXT_CMODE_DIRECT24_RGBA); + c_set_disp_reg(par, CARMINE_DISP_REG_L4_EXT_MODE, CARMINE_EXTEND_MODE | + CARMINE_EXT_CMODE_DIRECT24_RGBA); + c_set_disp_reg(par, CARMINE_DISP_REG_L5_EXT_MODE, CARMINE_EXTEND_MODE | + CARMINE_EXT_CMODE_DIRECT24_RGBA); + c_set_disp_reg(par, CARMINE_DISP_REG_L6_EXT_MODE, CARMINE_EXTEND_MODE | + CARMINE_EXT_CMODE_DIRECT24_RGBA); + c_set_disp_reg(par, CARMINE_DISP_REG_L7_EXT_MODE, CARMINE_EXTEND_MODE | + CARMINE_EXT_CMODE_DIRECT24_RGBA); + + /* Set default frame size to layer mode register */ + width = par->res->hdp * 4 / CARMINE_DISP_WIDTH_UNIT; + width = width << CARMINE_DISP_WIDTH_SHIFT; + + height = par->res->vdp - 1; + param = width | height; + + c_set_disp_reg(par, CARMINE_DISP_REG_L0_MODE_W_H, param); + c_set_disp_reg(par, CARMINE_DISP_REG_L1_WIDTH, width); + c_set_disp_reg(par, CARMINE_DISP_REG_L2_MODE_W_H, param); + c_set_disp_reg(par, CARMINE_DISP_REG_L3_MODE_W_H, param); + c_set_disp_reg(par, CARMINE_DISP_REG_L4_MODE_W_H, param); + c_set_disp_reg(par, CARMINE_DISP_REG_L5_MODE_W_H, param); + c_set_disp_reg(par, CARMINE_DISP_REG_L6_MODE_W_H, param); + c_set_disp_reg(par, CARMINE_DISP_REG_L7_MODE_W_H, param); + + /* Set default pos and size */ + window_size = (par->res->vdp - 1) << CARMINE_DISP_WIN_H_SHIFT; + window_size |= par->res->hdp; + + c_set_disp_reg(par, CARMINE_DISP_REG_L0_WIN_POS, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L0_WIN_SIZE, window_size); + c_set_disp_reg(par, CARMINE_DISP_REG_L1_WIN_POS, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L1_WIN_SIZE, window_size); + c_set_disp_reg(par, CARMINE_DISP_REG_L2_WIN_POS, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L2_WIN_SIZE, window_size); + c_set_disp_reg(par, CARMINE_DISP_REG_L3_WIN_POS, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L3_WIN_SIZE, window_size); + c_set_disp_reg(par, CARMINE_DISP_REG_L4_WIN_POS, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L4_WIN_SIZE, window_size); + c_set_disp_reg(par, CARMINE_DISP_REG_L5_WIN_POS, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L5_WIN_SIZE, window_size); + c_set_disp_reg(par, CARMINE_DISP_REG_L6_WIN_POS, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L6_WIN_SIZE, window_size); + c_set_disp_reg(par, CARMINE_DISP_REG_L7_WIN_POS, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L7_WIN_SIZE, window_size); + + /* Set default origin address */ + c_set_disp_reg(par, CARMINE_DISP_REG_L0_ORG_ADR, soffset); + c_set_disp_reg(par, CARMINE_DISP_REG_L1_ORG_ADR, soffset); + c_set_disp_reg(par, CARMINE_DISP_REG_L2_ORG_ADR1, soffset); + c_set_disp_reg(par, CARMINE_DISP_REG_L3_ORG_ADR1, soffset); + c_set_disp_reg(par, CARMINE_DISP_REG_L4_ORG_ADR1, soffset); + c_set_disp_reg(par, CARMINE_DISP_REG_L5_ORG_ADR1, soffset); + c_set_disp_reg(par, CARMINE_DISP_REG_L6_ORG_ADR1, soffset); + c_set_disp_reg(par, CARMINE_DISP_REG_L7_ORG_ADR1, soffset); + + /* Set default display address */ + c_set_disp_reg(par, CARMINE_DISP_REG_L0_DISP_ADR, soffset); + c_set_disp_reg(par, CARMINE_DISP_REG_L2_DISP_ADR1, soffset); + c_set_disp_reg(par, CARMINE_DISP_REG_L3_DISP_ADR1, soffset); + c_set_disp_reg(par, CARMINE_DISP_REG_L4_DISP_ADR1, soffset); + c_set_disp_reg(par, CARMINE_DISP_REG_L5_DISP_ADR1, soffset); + c_set_disp_reg(par, CARMINE_DISP_REG_L6_DISP_ADR0, soffset); + c_set_disp_reg(par, CARMINE_DISP_REG_L7_DISP_ADR0, soffset); + + /* Set default display position */ + c_set_disp_reg(par, CARMINE_DISP_REG_L0_DISP_POS, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L2_DISP_POS, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L3_DISP_POS, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L4_DISP_POS, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L5_DISP_POS, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L6_DISP_POS, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L7_DISP_POS, 0); + + /* Set default blend mode */ + c_set_disp_reg(par, CARMINE_DISP_REG_BLEND_MODE_L0, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_BLEND_MODE_L1, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_BLEND_MODE_L2, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_BLEND_MODE_L3, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_BLEND_MODE_L4, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_BLEND_MODE_L5, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_BLEND_MODE_L6, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_BLEND_MODE_L7, 0); + + /* default transparency mode */ + c_set_disp_reg(par, CARMINE_DISP_REG_L0_TRANS, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L1_TRANS, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L2_TRANS, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L3_TRANS, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L4_TRANS, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L5_TRANS, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L6_TRANS, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L7_TRANS, 0); + + /* Set default read skip parameter */ + c_set_disp_reg(par, CARMINE_DISP_REG_L0RM, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L2RM, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L3RM, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L4RM, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L5RM, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L6RM, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L7RM, 0); + + c_set_disp_reg(par, CARMINE_DISP_REG_L0PX, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L2PX, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L3PX, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L4PX, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L5PX, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L6PX, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L7PX, 0); + + c_set_disp_reg(par, CARMINE_DISP_REG_L0PY, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L2PY, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L3PY, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L4PY, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L5PY, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L6PY, 0); + c_set_disp_reg(par, CARMINE_DISP_REG_L7PY, 0); +} + +static void set_display_parameters(struct carmine_fb *par) +{ + u32 mode; + u32 hdp, vdp, htp, hsp, hsw, vtr, vsp, vsw; + + /* + * display timing. Parameters are decreased by one because hardware + * spec is 0 to (n - 1) + * */ + hdp = par->res->hdp - 1; + vdp = par->res->vdp - 1; + htp = par->res->htp - 1; + hsp = par->res->hsp - 1; + hsw = par->res->hsw - 1; + vtr = par->res->vtr - 1; + vsp = par->res->vsp - 1; + vsw = par->res->vsw - 1; + + c_set_disp_reg(par, CARMINE_DISP_REG_H_TOTAL, + htp << CARMINE_DISP_HTP_SHIFT); + c_set_disp_reg(par, CARMINE_DISP_REG_H_PERIOD, + (hdp << CARMINE_DISP_HDB_SHIFT) | hdp); + c_set_disp_reg(par, CARMINE_DISP_REG_V_H_W_H_POS, + (vsw << CARMINE_DISP_VSW_SHIFT) | + (hsw << CARMINE_DISP_HSW_SHIFT) | + (hsp)); + c_set_disp_reg(par, CARMINE_DISP_REG_V_TOTAL, + vtr << CARMINE_DISP_VTR_SHIFT); + c_set_disp_reg(par, CARMINE_DISP_REG_V_PERIOD_POS, + (vdp << CARMINE_DISP_VDP_SHIFT) | vsp); + + /* clock */ + mode = c_get_disp_reg(par, CARMINE_DISP_REG_DCM1); + mode = (mode & ~CARMINE_DISP_DCM_MASK) | + (par->res->disp_mode & CARMINE_DISP_DCM_MASK); + /* enable video output and layer 0 */ + mode |= CARMINE_DEN | CARMINE_L0E; + c_set_disp_reg(par, CARMINE_DISP_REG_DCM1, mode); +} + +static int carmine_set_par(struct fb_info *info) +{ + struct carmine_fb *par = info->par; + int ret; + + ret = carmine_find_mode(&info->var); + if (ret < 0) + return ret; + + par->new_mode = ret; + if (par->cur_mode != par->new_mode) { + + par->cur_mode = par->new_mode; + par->res = &car_modes[par->new_mode]; + + carmine_init_display_param(par); + set_display_parameters(par); + } + + info->fix.line_length = info->var.xres * info->var.bits_per_pixel / 8; + return 0; +} + +static int init_hardware(struct carmine_hw *hw) +{ + u32 flags; + u32 loops; + u32 ret; + + /* Initalize Carmine */ + /* Sets internal clock */ + c_set_hw_reg(hw, CARMINE_CTL_REG + CARMINE_CTL_REG_CLOCK_ENABLE, + CARMINE_DFLT_IP_CLOCK_ENABLE); + + /* Video signal output is turned off */ + c_set_hw_reg(hw, CARMINE_DISP0_REG + CARMINE_DISP_REG_DCM1, 0); + c_set_hw_reg(hw, CARMINE_DISP1_REG + CARMINE_DISP_REG_DCM1, 0); + + /* Software reset */ + c_set_hw_reg(hw, CARMINE_CTL_REG + CARMINE_CTL_REG_SOFTWARE_RESET, 1); + c_set_hw_reg(hw, CARMINE_CTL_REG + CARMINE_CTL_REG_SOFTWARE_RESET, 0); + + /* I/O mode settings */ + flags = CARMINE_DFLT_IP_DCTL_IO_CONT1 << 16 | + CARMINE_DFLT_IP_DCTL_IO_CONT0; + c_set_hw_reg(hw, CARMINE_DCTL_REG + CARMINE_DCTL_REG_IOCONT1_IOCONT0, + flags); + + /* DRAM initial sequence */ + flags = CARMINE_DFLT_IP_DCTL_MODE << 16 | CARMINE_DFLT_IP_DCTL_ADD; + c_set_hw_reg(hw, CARMINE_DCTL_REG + CARMINE_DCTL_REG_MODE_ADD, + flags); + + flags = CARMINE_DFLT_IP_DCTL_SET_TIME1 << 16 | + CARMINE_DFLT_IP_DCTL_EMODE; + c_set_hw_reg(hw, CARMINE_DCTL_REG + CARMINE_DCTL_REG_SETTIME1_EMODE, + flags); + + flags = CARMINE_DFLT_IP_DCTL_REFRESH << 16 | + CARMINE_DFLT_IP_DCTL_SET_TIME2; + c_set_hw_reg(hw, CARMINE_DCTL_REG + CARMINE_DCTL_REG_REFRESH_SETTIME2, + flags); + + flags = CARMINE_DFLT_IP_DCTL_RESERVE2 << 16 | + CARMINE_DFLT_IP_DCTL_FIFO_DEPTH; + c_set_hw_reg(hw, CARMINE_DCTL_REG + CARMINE_DCTL_REG_RSV2_RSV1, flags); + + flags = CARMINE_DFLT_IP_DCTL_DDRIF2 << 16 | CARMINE_DFLT_IP_DCTL_DDRIF1; + c_set_hw_reg(hw, CARMINE_DCTL_REG + CARMINE_DCTL_REG_DDRIF2_DDRIF1, + flags); + + flags = CARMINE_DFLT_IP_DCTL_RESERVE0 << 16 | + CARMINE_DFLT_IP_DCTL_STATES; + c_set_hw_reg(hw, CARMINE_DCTL_REG + CARMINE_DCTL_REG_RSV0_STATES, + flags); + + /* Executes DLL reset */ + if (CARMINE_DCTL_DLL_RESET) { + for (loops = 0; loops < CARMINE_DCTL_INIT_WAIT_LIMIT; loops++) { + + ret = c_get_hw_reg(hw, CARMINE_DCTL_REG + + CARMINE_DCTL_REG_RSV0_STATES); + ret &= CARMINE_DCTL_REG_STATES_MASK; + if (!ret) + break; + + mdelay(CARMINE_DCTL_INIT_WAIT_INTERVAL); + } + + if (loops >= CARMINE_DCTL_INIT_WAIT_LIMIT) { + printk(KERN_ERR "DRAM init failed\n"); + return -EIO; + } + } + + flags = CARMINE_DFLT_IP_DCTL_MODE_AFT_RST << 16 | + CARMINE_DFLT_IP_DCTL_ADD; + c_set_hw_reg(hw, CARMINE_DCTL_REG + CARMINE_DCTL_REG_MODE_ADD, flags); + + flags = CARMINE_DFLT_IP_DCTL_RESERVE0 << 16 | + CARMINE_DFLT_IP_DCTL_STATES_AFT_RST; + c_set_hw_reg(hw, CARMINE_DCTL_REG + CARMINE_DCTL_REG_RSV0_STATES, + flags); + + /* Initialize the write back register */ + c_set_hw_reg(hw, CARMINE_WB_REG + CARMINE_WB_REG_WBM, + CARMINE_WB_REG_WBM_DEFAULT); + + /* Initialize the Kottos registers */ + c_set_hw_reg(hw, CARMINE_GRAPH_REG + CARMINE_GRAPH_REG_VRINTM, 0); + c_set_hw_reg(hw, CARMINE_GRAPH_REG + CARMINE_GRAPH_REG_VRERRM, 0); + + /* Set DC offsets */ + c_set_hw_reg(hw, CARMINE_GRAPH_REG + CARMINE_GRAPH_REG_DC_OFFSET_PX, 0); + c_set_hw_reg(hw, CARMINE_GRAPH_REG + CARMINE_GRAPH_REG_DC_OFFSET_PY, 0); + c_set_hw_reg(hw, CARMINE_GRAPH_REG + CARMINE_GRAPH_REG_DC_OFFSET_LX, 0); + c_set_hw_reg(hw, CARMINE_GRAPH_REG + CARMINE_GRAPH_REG_DC_OFFSET_LY, 0); + c_set_hw_reg(hw, CARMINE_GRAPH_REG + CARMINE_GRAPH_REG_DC_OFFSET_TX, 0); + c_set_hw_reg(hw, CARMINE_GRAPH_REG + CARMINE_GRAPH_REG_DC_OFFSET_TY, 0); + return 0; +} + +static struct fb_ops carminefb_ops = { + .owner = THIS_MODULE, + .fb_fillrect = cfb_fillrect, + .fb_copyarea = cfb_copyarea, + .fb_imageblit = cfb_imageblit, + + .fb_check_var = carmine_check_var, + .fb_set_par = carmine_set_par, + .fb_setcolreg = carmine_setcolreg, +}; + +static int alloc_carmine_fb(void __iomem *regs, void __iomem *smem_base, + int smem_offset, struct device *device, struct fb_info **rinfo) +{ + int ret; + struct fb_info *info; + struct carmine_fb *par; + + info = framebuffer_alloc(sizeof *par, device); + if (!info) + return -ENOMEM; + + par = info->par; + par->display_reg = regs; + par->smem_offset = smem_offset; + + info->screen_base = smem_base + smem_offset; + info->screen_size = CARMINE_DISPLAY_MEM; + info->fbops = &carminefb_ops; + + info->fix = carminefb_fix; + info->pseudo_palette = par->pseudo_palette; + info->flags = FBINFO_DEFAULT; + + ret = fb_alloc_cmap(&info->cmap, 256, 1); + if (ret < 0) + goto err_free_fb; + + if (fb_mode > ARRAY_SIZE(carmine_modedb)) + fb_mode = CARMINEFB_DEFAULT_VIDEO_MODE; + + par->cur_mode = par->new_mode = ~0; + + ret = fb_find_mode(&info->var, info, fb_mode_str, carmine_modedb, + ARRAY_SIZE(carmine_modedb), + &carmine_modedb[fb_mode], 32); + if (!ret || ret == 4) { + ret = -EINVAL; + goto err_dealloc_cmap; + } + + fb_videomode_to_modelist(carmine_modedb, ARRAY_SIZE(carmine_modedb), + &info->modelist); + + ret = register_framebuffer(info); + if (ret < 0) + goto err_dealloc_cmap; + + printk(KERN_INFO "fb%d: %s frame buffer device\n", info->node, + info->fix.id); + + *rinfo = info; + return 0; + +err_dealloc_cmap: + fb_dealloc_cmap(&info->cmap); +err_free_fb: + framebuffer_release(info); + return ret; +} + +static void cleanup_fb_device(struct fb_info *info) +{ + if (info) { + unregister_framebuffer(info); + fb_dealloc_cmap(&info->cmap); + framebuffer_release(info); + } +} + +static int __devinit carminefb_probe(struct pci_dev *dev, + const struct pci_device_id *ent) +{ + struct carmine_hw *hw; + struct device *device = &dev->dev; + struct fb_info *info; + int ret; + + ret = pci_enable_device(dev); + if (ret) + return ret; + + ret = -ENOMEM; + hw = kzalloc(sizeof *hw, GFP_KERNEL); + if (!hw) + goto err_enable_pci; + + carminefb_fix.mmio_start = pci_resource_start(dev, CARMINE_CONFIG_BAR); + carminefb_fix.mmio_len = pci_resource_len(dev, CARMINE_CONFIG_BAR); + + if (!request_mem_region(carminefb_fix.mmio_start, + carminefb_fix.mmio_len, + "carminefb regbase")) { + printk(KERN_ERR "carminefb: Can't reserve regbase.\n"); + ret = -EBUSY; + goto err_free_hw; + } + hw->v_regs = ioremap_nocache(carminefb_fix.mmio_start, + carminefb_fix.mmio_len); + if (!hw->v_regs) { + printk(KERN_ERR "carminefb: Can't remap %s register.\n", + carminefb_fix.id); + goto err_free_reg_mmio; + } + + carminefb_fix.smem_start = pci_resource_start(dev, CARMINE_MEMORY_BAR); + carminefb_fix.smem_len = pci_resource_len(dev, CARMINE_MEMORY_BAR); + + /* The memory area tends to be very large (256 MiB). Remap only what + * is required for that largest resolution to avoid remaps at run + * time + */ + if (carminefb_fix.smem_len > CARMINE_TOTAL_DIPLAY_MEM) + carminefb_fix.smem_len = CARMINE_TOTAL_DIPLAY_MEM; + + else if (carminefb_fix.smem_len < CARMINE_TOTAL_DIPLAY_MEM) { + printk(KERN_ERR "carminefb: Memory bar is only %d bytes, %d " + "are required.", carminefb_fix.smem_len, + CARMINE_TOTAL_DIPLAY_MEM); + goto err_free_reg_mmio; + } + + if (!request_mem_region(carminefb_fix.smem_start, + carminefb_fix.smem_len, "carminefb smem")) { + printk(KERN_ERR "carminefb: Can't reserve smem.\n"); + goto err_unmap_vregs; + } + + hw->screen_mem = ioremap_nocache(carminefb_fix.smem_start, + carminefb_fix.smem_len); + if (!hw->screen_mem) { + printk(KERN_ERR "carmine: Can't ioremap smem area.\n"); + release_mem_region(carminefb_fix.smem_start, + carminefb_fix.smem_len); + goto err_reg_smem; + } + + ret = init_hardware(hw); + if (ret) + goto err_unmap_screen; + + info = NULL; + if (fb_displays & CARMINE_USE_DISPLAY0) { + ret = alloc_carmine_fb(hw->v_regs + CARMINE_DISP0_REG, + hw->screen_mem, CARMINE_DISPLAY_MEM * 0, + device, &info); + if (ret) + goto err_deinit_hw; + } + + hw->fb[0] = info; + + info = NULL; + if (fb_displays & CARMINE_USE_DISPLAY1) { + ret = alloc_carmine_fb(hw->v_regs + CARMINE_DISP1_REG, + hw->screen_mem, CARMINE_DISPLAY_MEM * 1, + device, &info); + if (ret) + goto err_cleanup_fb0; + } + + hw->fb[1] = info; + info = NULL; + + pci_set_drvdata(dev, hw); + return 0; + +err_cleanup_fb0: + cleanup_fb_device(hw->fb[0]); +err_deinit_hw: + /* disable clock, etc */ + c_set_hw_reg(hw, CARMINE_CTL_REG + CARMINE_CTL_REG_CLOCK_ENABLE, 0); +err_unmap_screen: + iounmap(hw->screen_mem); +err_reg_smem: + release_mem_region(carminefb_fix.mmio_start, carminefb_fix.mmio_len); +err_unmap_vregs: + iounmap(hw->v_regs); +err_free_reg_mmio: + release_mem_region(carminefb_fix.mmio_start, carminefb_fix.mmio_len); +err_free_hw: + kfree(hw); +err_enable_pci: + pci_disable_device(dev); + return ret; +} + +static void __devexit carminefb_remove(struct pci_dev *dev) +{ + struct carmine_hw *hw = pci_get_drvdata(dev); + struct fb_fix_screeninfo fix; + int i; + + /* in case we use only fb1 and not fb1 */ + if (hw->fb[0]) + fix = hw->fb[0]->fix; + else + fix = hw->fb[1]->fix; + + /* deactivate display(s) and switch clocks */ + c_set_hw_reg(hw, CARMINE_DISP0_REG + CARMINE_DISP_REG_DCM1, 0); + c_set_hw_reg(hw, CARMINE_DISP1_REG + CARMINE_DISP_REG_DCM1, 0); + c_set_hw_reg(hw, CARMINE_CTL_REG + CARMINE_CTL_REG_CLOCK_ENABLE, 0); + + for (i = 0; i < MAX_DISPLAY; i++) + cleanup_fb_device(hw->fb[i]); + + iounmap(hw->screen_mem); + release_mem_region(fix.smem_start, fix.smem_len); + iounmap(hw->v_regs); + release_mem_region(fix.mmio_start, fix.mmio_len); + + pci_set_drvdata(dev, NULL); + pci_disable_device(dev); + kfree(hw); +} + +#define PCI_VENDOR_ID_FUJITU_LIMITED 0x10cf +static struct pci_device_id carmine_devices[] __devinitdata = { +{ + PCI_DEVICE(PCI_VENDOR_ID_FUJITU_LIMITED, 0x202b)}, + {0, 0, 0, 0, 0, 0, 0} +}; + +MODULE_DEVICE_TABLE(pci, carmine_devices); + +static struct pci_driver carmine_pci_driver = { + .name = "carminefb", + .id_table = carmine_devices, + .probe = carminefb_probe, + .remove = __devexit_p(carminefb_remove), +}; + +static int __init carminefb_init(void) +{ + if (!(fb_displays & + (CARMINE_USE_DISPLAY0 | CARMINE_USE_DISPLAY1))) { + printk(KERN_ERR "If you disable both displays than you don't " + "need the driver at all\n"); + return -EINVAL; + } + return pci_register_driver(&carmine_pci_driver); +} +module_init(carminefb_init); + +static void __exit carminefb_cleanup(void) +{ + pci_unregister_driver(&carmine_pci_driver); +} +module_exit(carminefb_cleanup); + +MODULE_AUTHOR("Sebastian Siewior "); +MODULE_DESCRIPTION("Framebuffer driver for Fujitsu Carmine based devices"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/video/carminefb.h b/drivers/video/carminefb.h new file mode 100644 index 000000000000..05306de0c6b6 --- /dev/null +++ b/drivers/video/carminefb.h @@ -0,0 +1,64 @@ +#ifndef CARMINE_CARMINE_H +#define CARMINE_CARMINE_H + +#define CARMINE_MEMORY_BAR 2 +#define CARMINE_CONFIG_BAR 3 + +#define MAX_DISPLAY 2 +#define CARMINE_DISPLAY_MEM (800 * 600 * 4) +#define CARMINE_TOTAL_DIPLAY_MEM (CARMINE_DISPLAY_MEM * MAX_DISPLAY) + +#define CARMINE_USE_DISPLAY0 (1 << 0) +#define CARMINE_USE_DISPLAY1 (1 << 1) + +/* + * This values work on the eval card. Custom boards may use different timings, + * here an example :) + */ + +/* DRAM initialization values */ +#ifdef CONFIG_FB_CARMINE_DRAM_EVAL + +#define CARMINE_DFLT_IP_CLOCK_ENABLE (0x03ff) +#define CARMINE_DFLT_IP_DCTL_ADD (0x05c3) +#define CARMINE_DFLT_IP_DCTL_MODE (0x0121) +#define CARMINE_DFLT_IP_DCTL_EMODE (0x8000) +#define CARMINE_DFLT_IP_DCTL_SET_TIME1 (0x4749) +#define CARMINE_DFLT_IP_DCTL_SET_TIME2 (0x2a22) +#define CARMINE_DFLT_IP_DCTL_REFRESH (0x0042) +#define CARMINE_DFLT_IP_DCTL_STATES (0x0003) +#define CARMINE_DFLT_IP_DCTL_RESERVE0 (0x0020) +#define CARMINE_DFLT_IP_DCTL_FIFO_DEPTH (0x000f) +#define CARMINE_DFLT_IP_DCTL_RESERVE2 (0x0000) +#define CARMINE_DFLT_IP_DCTL_DDRIF1 (0x6646) +#define CARMINE_DFLT_IP_DCTL_DDRIF2 (0x0055) +#define CARMINE_DFLT_IP_DCTL_MODE_AFT_RST (0x0021) +#define CARMINE_DFLT_IP_DCTL_STATES_AFT_RST (0x0002) +#define CARMINE_DFLT_IP_DCTL_IO_CONT0 (0x0555) +#define CARMINE_DFLT_IP_DCTL_IO_CONT1 (0x0555) +#define CARMINE_DCTL_DLL_RESET (1) +#endif + +#ifdef CONFIG_CARMINE_DRAM_CUSTOM + +#define CARMINE_DFLT_IP_CLOCK_ENABLE (0x03ff) +#define CARMINE_DFLT_IP_DCTL_ADD (0x03b2) +#define CARMINE_DFLT_IP_DCTL_MODE (0x0161) +#define CARMINE_DFLT_IP_DCTL_EMODE (0x8000) +#define CARMINE_DFLT_IP_DCTL_SET_TIME1 (0x2628) +#define CARMINE_DFLT_IP_DCTL_SET_TIME2 (0x1a09) +#define CARMINE_DFLT_IP_DCTL_REFRESH (0x00fe) +#define CARMINE_DFLT_IP_DCTL_STATES (0x0003) +#define CARMINE_DFLT_IP_DCTL_RESERVE0 (0x0020) +#define CARMINE_DFLT_IP_DCTL_FIFO_DEPTH (0x000f) +#define CARMINE_DFLT_IP_DCTL_RESERVE2 (0x0000) +#define CARMINE_DFLT_IP_DCTL_DDRIF1 (0x0646) +#define CARMINE_DFLT_IP_DCTL_DDRIF2 (0x55aa) +#define CARMINE_DFLT_IP_DCTL_MODE_AFT_RST (0x0061) +#define CARMINE_DFLT_IP_DCTL_STATES_AFT_RST (0x0002) +#define CARMINE_DFLT_IP_DCTL_IO_CONT0 (0x0555) +#define CARMINE_DFLT_IP_DCTL_IO_CONT1 (0x0555) +#define CARMINE_DCTL_DLL_RESET (1) +#endif + +#endif diff --git a/drivers/video/carminefb_regs.h b/drivers/video/carminefb_regs.h new file mode 100644 index 000000000000..045215600b73 --- /dev/null +++ b/drivers/video/carminefb_regs.h @@ -0,0 +1,159 @@ +#ifndef _CARMINEFB_REGS_H +#define _CARMINEFB_REGS_H + +#define CARMINE_OVERLAY_EXT_MODE (0x00000002) +#define CARMINE_GRAPH_REG (0x00000000) +#define CARMINE_DISP0_REG (0x00100000) +#define CARMINE_DISP1_REG (0x00140000) +#define CARMINE_WB_REG (0x00180000) +#define CARMINE_DCTL_REG (0x00300000) +#define CARMINE_CTL_REG (0x00400000) +#define CARMINE_WINDOW_MODE (0x00000001) +#define CARMINE_EXTEND_MODE (CARMINE_WINDOW_MODE | \ + CARMINE_OVERLAY_EXT_MODE) +#define CARMINE_L0E (1 << 16) +#define CARMINE_L2E (1 << 18) +#define CARMINE_DEN (1 << 31) + +#define CARMINE_EXT_CMODE_DIRECT24_RGBA (0xC0000000) +#define CARMINE_DCTL_REG_MODE_ADD (0x00) +#define CARMINE_DCTL_REG_SETTIME1_EMODE (0x04) +#define CARMINE_DCTL_REG_REFRESH_SETTIME2 (0x08) +#define CARMINE_DCTL_REG_RSV0_STATES (0x0C) +#define CARMINE_DCTL_REG_RSV2_RSV1 (0x10) +#define CARMINE_DCTL_REG_DDRIF2_DDRIF1 (0x14) +#define CARMINE_DCTL_REG_IOCONT1_IOCONT0 (0x24) +#define CARMINE_DCTL_REG_STATES_MASK (0x000F) +#define CARMINE_DCTL_INIT_WAIT_INTERVAL (1) +#define CARMINE_DCTL_INIT_WAIT_LIMIT (5000) +#define CARMINE_WB_REG_WBM_DEFAULT (0x0001c020) +#define CARMINE_DISP_REG_L0RM (0x1880) +#define CARMINE_DISP_REG_L0PX (0x1884) +#define CARMINE_DISP_REG_L0PY (0x1888) +#define CARMINE_DISP_REG_L2RM (0x18A0) +#define CARMINE_DISP_REG_L2PX (0x18A4) +#define CARMINE_DISP_REG_L2PY (0x18A8) +#define CARMINE_DISP_REG_L3RM (0x18B0) +#define CARMINE_DISP_REG_L3PX (0x18B4) +#define CARMINE_DISP_REG_L3PY (0x18B8) +#define CARMINE_DISP_REG_L4RM (0x18C0) +#define CARMINE_DISP_REG_L4PX (0x18C4) +#define CARMINE_DISP_REG_L4PY (0x18C8) +#define CARMINE_DISP_REG_L5RM (0x18D0) +#define CARMINE_DISP_REG_L5PX (0x18D4) +#define CARMINE_DISP_REG_L5PY (0x18D8) +#define CARMINE_DISP_REG_L6RM (0x1924) +#define CARMINE_DISP_REG_L6PX (0x1928) +#define CARMINE_DISP_REG_L6PY (0x192C) +#define CARMINE_DISP_REG_L7RM (0x1964) +#define CARMINE_DISP_REG_L7PX (0x1968) +#define CARMINE_DISP_REG_L7PY (0x196C) +#define CARMINE_WB_REG_WBM (0x0004) +#define CARMINE_DISP_HTP_SHIFT (16) +#define CARMINE_DISP_HDB_SHIFT (16) +#define CARMINE_DISP_HSW_SHIFT (16) +#define CARMINE_DISP_VSW_SHIFT (24) +#define CARMINE_DISP_VTR_SHIFT (16) +#define CARMINE_DISP_VDP_SHIFT (16) +#define CARMINE_CURSOR_CUTZ_MASK (0x00000100) +#define CARMINE_CURSOR0_PRIORITY_MASK (0x00010000) +#define CARMINE_CURSOR1_PRIORITY_MASK (0x00020000) +#define CARMINE_DISP_WIDTH_SHIFT (16) +#define CARMINE_DISP_WIN_H_SHIFT (16) +#define CARMINE_DISP_REG_H_TOTAL (0x0004) +#define CARMINE_DISP_REG_H_PERIOD (0x0008) +#define CARMINE_DISP_REG_V_H_W_H_POS (0x000C) +#define CARMINE_DISP_REG_V_TOTAL (0x0010) +#define CARMINE_DISP_REG_V_PERIOD_POS (0x0014) +#define CARMINE_DISP_REG_L0_MODE_W_H (0x0020) +#define CARMINE_DISP_REG_L0_ORG_ADR (0x0024) +#define CARMINE_DISP_REG_L0_DISP_ADR (0x0028) +#define CARMINE_DISP_REG_L0_DISP_POS (0x002C) +#define CARMINE_DISP_REG_L1_WIDTH (0x0030) +#define CARMINE_DISP_REG_L1_ORG_ADR (0x0034) +#define CARMINE_DISP_REG_L2_MODE_W_H (0x0040) +#define CARMINE_DISP_REG_L2_ORG_ADR1 (0x0044) +#define CARMINE_DISP_REG_L2_DISP_ADR1 (0x0048) +#define CARMINE_DISP_REG_L2_DISP_POS (0x0054) +#define CARMINE_DISP_REG_L3_MODE_W_H (0x0058) +#define CARMINE_DISP_REG_L3_ORG_ADR1 (0x005C) +#define CARMINE_DISP_REG_L3_DISP_ADR1 (0x0060) +#define CARMINE_DISP_REG_L3_DISP_POS (0x006C) +#define CARMINE_DISP_REG_L4_MODE_W_H (0x0070) +#define CARMINE_DISP_REG_L4_ORG_ADR1 (0x0074) +#define CARMINE_DISP_REG_L4_DISP_ADR1 (0x0078) +#define CARMINE_DISP_REG_L4_DISP_POS (0x0084) +#define CARMINE_DISP_REG_L5_MODE_W_H (0x0088) +#define CARMINE_DISP_REG_L5_ORG_ADR1 (0x008C) +#define CARMINE_DISP_REG_L5_DISP_ADR1 (0x0090) +#define CARMINE_DISP_REG_L5_DISP_POS (0x009C) +#define CARMINE_DISP_REG_CURSOR_MODE (0x00A0) +#define CARMINE_DISP_REG_CUR1_POS (0x00A8) +#define CARMINE_DISP_REG_CUR2_POS (0x00B0) +#define CARMINE_DISP_REG_C_TRANS (0x00BC) +#define CARMINE_DISP_REG_MLMR_TRANS (0x00C0) +#define CARMINE_DISP_REG_L0_EXT_MODE (0x0110) +#define CARMINE_DISP_REG_L0_WIN_POS (0x0114) +#define CARMINE_DISP_REG_L0_WIN_SIZE (0x0118) +#define CARMINE_DISP_REG_L1_EXT_MODE (0x0120) +#define CARMINE_DISP_REG_L1_WIN_POS (0x0124) +#define CARMINE_DISP_REG_L1_WIN_SIZE (0x0128) +#define CARMINE_DISP_REG_L2_EXT_MODE (0x0130) +#define CARMINE_DISP_REG_L2_WIN_POS (0x0134) +#define CARMINE_DISP_REG_L2_WIN_SIZE (0x0138) +#define CARMINE_DISP_REG_L3_EXT_MODE (0x0140) +#define CARMINE_DISP_REG_L3_WIN_POS (0x0144) +#define CARMINE_DISP_REG_L3_WIN_SIZE (0x0148) +#define CARMINE_DISP_REG_L4_EXT_MODE (0x0150) +#define CARMINE_DISP_REG_L4_WIN_POS (0x0154) +#define CARMINE_DISP_REG_L4_WIN_SIZE (0x0158) +#define CARMINE_DISP_REG_L5_EXT_MODE (0x0160) +#define CARMINE_DISP_REG_L5_WIN_POS (0x0164) +#define CARMINE_DISP_REG_L5_WIN_SIZE (0x0168) +#define CARMINE_DISP_REG_L6_EXT_MODE (0x1918) +#define CARMINE_DISP_REG_L6_WIN_POS (0x191c) +#define CARMINE_DISP_REG_L6_WIN_SIZE (0x1920) +#define CARMINE_DISP_REG_L7_EXT_MODE (0x1958) +#define CARMINE_DISP_REG_L7_WIN_POS (0x195c) +#define CARMINE_DISP_REG_L7_WIN_SIZE (0x1960) +#define CARMINE_DISP_REG_BLEND_MODE_L0 (0x00B4) +#define CARMINE_DISP_REG_BLEND_MODE_L1 (0x0188) +#define CARMINE_DISP_REG_BLEND_MODE_L2 (0x018C) +#define CARMINE_DISP_REG_BLEND_MODE_L3 (0x0190) +#define CARMINE_DISP_REG_BLEND_MODE_L4 (0x0194) +#define CARMINE_DISP_REG_BLEND_MODE_L5 (0x0198) +#define CARMINE_DISP_REG_BLEND_MODE_L6 (0x1990) +#define CARMINE_DISP_REG_BLEND_MODE_L7 (0x1994) +#define CARMINE_DISP_REG_L0_TRANS (0x01A0) +#define CARMINE_DISP_REG_L1_TRANS (0x01A4) +#define CARMINE_DISP_REG_L2_TRANS (0x01A8) +#define CARMINE_DISP_REG_L3_TRANS (0x01AC) +#define CARMINE_DISP_REG_L4_TRANS (0x01B0) +#define CARMINE_DISP_REG_L5_TRANS (0x01B4) +#define CARMINE_DISP_REG_L6_TRANS (0x1998) +#define CARMINE_DISP_REG_L7_TRANS (0x199c) +#define CARMINE_EXTEND_MODE_MASK (0x00000003) +#define CARMINE_DISP_DCM_MASK (0x0000FFFF) +#define CARMINE_DISP_REG_DCM1 (0x0100) +#define CARMINE_DISP_WIDTH_UNIT (64) +#define CARMINE_DISP_REG_L6_MODE_W_H (0x1900) +#define CARMINE_DISP_REG_L6_ORG_ADR1 (0x1904) +#define CARMINE_DISP_REG_L6_DISP_ADR0 (0x1908) +#define CARMINE_DISP_REG_L6_DISP_POS (0x1914) +#define CARMINE_DISP_REG_L7_MODE_W_H (0x1940) +#define CARMINE_DISP_REG_L7_ORG_ADR1 (0x1944) +#define CARMINE_DISP_REG_L7_DISP_ADR0 (0x1948) +#define CARMINE_DISP_REG_L7_DISP_POS (0x1954) +#define CARMINE_CTL_REG_CLOCK_ENABLE (0x000C) +#define CARMINE_CTL_REG_SOFTWARE_RESET (0x0010) +#define CARMINE_CTL_REG_IST_MASK_ALL (0x07FFFFFF) +#define CARMINE_GRAPH_REG_VRINTM (0x00028064) +#define CARMINE_GRAPH_REG_VRERRM (0x0002806C) +#define CARMINE_GRAPH_REG_DC_OFFSET_PX (0x0004005C) +#define CARMINE_GRAPH_REG_DC_OFFSET_PY (0x00040060) +#define CARMINE_GRAPH_REG_DC_OFFSET_LX (0x00040064) +#define CARMINE_GRAPH_REG_DC_OFFSET_LY (0x00040068) +#define CARMINE_GRAPH_REG_DC_OFFSET_TX (0x0004006C) +#define CARMINE_GRAPH_REG_DC_OFFSET_TY (0x00040070) + +#endif From 306fa6f60a2870b7a9827a64e1b45cd35a9549aa Mon Sep 17 00:00:00 2001 From: Krzysztof Helt Date: Wed, 23 Jul 2008 21:30:50 -0700 Subject: [PATCH 361/857] tridentfb: replace macros with functions This patch replaces macros with static functions and puts tridentfb_par pointer as the first argument of these functions. These is a step toward multihead support. Additionally, bogus TRIDENT_MMIO define is removed as the driver supports graphics cards only through the mmio mode. Signed-off-by: Krzysztof Helt Cc: "Antonino A. Daplas" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/tridentfb.c | 526 ++++++++++++++++++++------------------ 1 file changed, 283 insertions(+), 243 deletions(-) diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c index beefab2992c0..3e8a1ef892ce 100644 --- a/drivers/video/tridentfb.c +++ b/drivers/video/tridentfb.c @@ -137,28 +137,34 @@ static int iscyber(int id) #define CRT 0x3D0 /* CRTC registers offset for color display */ -#ifndef TRIDENT_MMIO - #define TRIDENT_MMIO 1 -#endif - -#if TRIDENT_MMIO - #define t_outb(val, reg) writeb(val,((struct tridentfb_par *)(fb_info.par))->io_virt + reg) - #define t_inb(reg) readb(((struct tridentfb_par*)(fb_info.par))->io_virt + reg) -#else - #define t_outb(val, reg) outb(val, reg) - #define t_inb(reg) inb(reg) -#endif +static inline void t_outb(struct tridentfb_par *p, u8 val, u16 reg) +{ + fb_writeb(val, p->io_virt + reg); +} +static inline u8 t_inb(struct tridentfb_par *p, u16 reg) +{ + return fb_readb(p->io_virt + reg); +} static struct accel_switch { - void (*init_accel) (int, int); - void (*wait_engine) (void); - void (*fill_rect) (u32, u32, u32, u32, u32, u32); - void (*copy_rect) (u32, u32, u32, u32, u32, u32); + void (*init_accel) (struct tridentfb_par *, int, int); + void (*wait_engine) (struct tridentfb_par *); + void (*fill_rect) + (struct tridentfb_par *par, u32, u32, u32, u32, u32, u32); + void (*copy_rect) + (struct tridentfb_par *par, u32, u32, u32, u32, u32, u32); } *acc; -#define writemmr(r, v) writel(v, ((struct tridentfb_par *)fb_info.par)->io_virt + r) -#define readmmr(r) readl(((struct tridentfb_par *)fb_info.par)->io_virt + r) +static inline void writemmr(struct tridentfb_par *par, u16 r, u32 v) +{ + fb_writel(v, par->io_virt + r); +} + +static inline u32 readmmr(struct tridentfb_par *par, u16 r) +{ + return fb_readl(par->io_virt + r); +} /* * Blade specific acceleration. @@ -176,7 +182,7 @@ static struct accel_switch { #define ROP_S 0xCC -static void blade_init_accel(int pitch, int bpp) +static void blade_init_accel(struct tridentfb_par *par, int pitch, int bpp) { int v1 = (pitch >> 3) << 20; int tmp = 0, v2; @@ -196,33 +202,35 @@ static void blade_init_accel(int pitch, int bpp) break; } v2 = v1 | (tmp << 29); - writemmr(0x21C0, v2); - writemmr(0x21C4, v2); - writemmr(0x21B8, v2); - writemmr(0x21BC, v2); - writemmr(0x21D0, v1); - writemmr(0x21D4, v1); - writemmr(0x21C8, v1); - writemmr(0x21CC, v1); - writemmr(0x216C, 0); + writemmr(par, 0x21C0, v2); + writemmr(par, 0x21C4, v2); + writemmr(par, 0x21B8, v2); + writemmr(par, 0x21BC, v2); + writemmr(par, 0x21D0, v1); + writemmr(par, 0x21D4, v1); + writemmr(par, 0x21C8, v1); + writemmr(par, 0x21CC, v1); + writemmr(par, 0x216C, 0); } -static void blade_wait_engine(void) +static void blade_wait_engine(struct tridentfb_par *par) { - while (readmmr(STA) & 0xFA800000) ; + while (readmmr(par, STA) & 0xFA800000) ; } -static void blade_fill_rect(u32 x, u32 y, u32 w, u32 h, u32 c, u32 rop) +static void blade_fill_rect(struct tridentfb_par *par, + u32 x, u32 y, u32 w, u32 h, u32 c, u32 rop) { - writemmr(CLR, c); - writemmr(ROP, rop ? 0x66 : ROP_S); - writemmr(CMD, 0x20000000 | 1 << 19 | 1 << 4 | 2 << 2); + writemmr(par, CLR, c); + writemmr(par, ROP, rop ? 0x66 : ROP_S); + writemmr(par, CMD, 0x20000000 | 1 << 19 | 1 << 4 | 2 << 2); - writemmr(DR1, point(x, y)); - writemmr(DR2, point(x + w - 1, y + h - 1)); + writemmr(par, DR1, point(x, y)); + writemmr(par, DR2, point(x + w - 1, y + h - 1)); } -static void blade_copy_rect(u32 x1, u32 y1, u32 x2, u32 y2, u32 w, u32 h) +static void blade_copy_rect(struct tridentfb_par *par, + u32 x1, u32 y1, u32 x2, u32 y2, u32 w, u32 h) { u32 s1, s2, d1, d2; int direction = 2; @@ -234,13 +242,13 @@ static void blade_copy_rect(u32 x1, u32 y1, u32 x2, u32 y2, u32 w, u32 h) if ((y1 > y2) || ((y1 == y2) && (x1 > x2))) direction = 0; - writemmr(ROP, ROP_S); - writemmr(CMD, 0xE0000000 | 1 << 19 | 1 << 4 | 1 << 2 | direction); + writemmr(par, ROP, ROP_S); + writemmr(par, CMD, 0xE0000000 | 1 << 19 | 1 << 4 | 1 << 2 | direction); - writemmr(SR1, direction ? s2 : s1); - writemmr(SR2, direction ? s1 : s2); - writemmr(DR1, direction ? d2 : d1); - writemmr(DR2, direction ? d1 : d2); + writemmr(par, SR1, direction ? s2 : s1); + writemmr(par, SR2, direction ? s1 : s2); + writemmr(par, DR1, direction ? d2 : d1); + writemmr(par, DR2, direction ? d1 : d2); } static struct accel_switch accel_blade = { @@ -257,7 +265,7 @@ static struct accel_switch accel_blade = { #define ROP_P 0xF0 #define masked_point(x, y) ((y & 0xffff)<<16|(x & 0xffff)) -static void xp_init_accel(int pitch, int bpp) +static void xp_init_accel(struct tridentfb_par *par, int pitch, int bpp) { int tmp = 0, v1; unsigned char x = 0; @@ -293,7 +301,7 @@ static void xp_init_accel(int pitch, int bpp) break; } - t_outb(x, 0x2125); + t_outb(par, x, 0x2125); eng_oper = x | 0x40; @@ -313,12 +321,12 @@ static void xp_init_accel(int pitch, int bpp) v1 = pitch << tmp; - writemmr(0x2154, v1); - writemmr(0x2150, v1); - t_outb(3, 0x2126); + writemmr(par, 0x2154, v1); + writemmr(par, 0x2150, v1); + t_outb(par, 3, 0x2126); } -static void xp_wait_engine(void) +static void xp_wait_engine(struct tridentfb_par *par) { int busy; int count, timeout; @@ -326,7 +334,7 @@ static void xp_wait_engine(void) count = 0; timeout = 0; for (;;) { - busy = t_inb(STA) & 0x80; + busy = t_inb(par, STA) & 0x80; if (busy != 0x80) return; count++; @@ -336,25 +344,27 @@ static void xp_wait_engine(void) timeout++; if (timeout == 8) { /* Reset engine */ - t_outb(0x00, 0x2120); + t_outb(par, 0x00, 0x2120); return; } } } } -static void xp_fill_rect(u32 x, u32 y, u32 w, u32 h, u32 c, u32 rop) +static void xp_fill_rect(struct tridentfb_par *par, + u32 x, u32 y, u32 w, u32 h, u32 c, u32 rop) { - writemmr(0x2127, ROP_P); - writemmr(0x2158, c); - writemmr(0x2128, 0x4000); - writemmr(0x2140, masked_point(h, w)); - writemmr(0x2138, masked_point(y, x)); - t_outb(0x01, 0x2124); - t_outb(eng_oper, 0x2125); + writemmr(par, 0x2127, ROP_P); + writemmr(par, 0x2158, c); + writemmr(par, 0x2128, 0x4000); + writemmr(par, 0x2140, masked_point(h, w)); + writemmr(par, 0x2138, masked_point(y, x)); + t_outb(par, 0x01, 0x2124); + t_outb(par, eng_oper, 0x2125); } -static void xp_copy_rect(u32 x1, u32 y1, u32 x2, u32 y2, u32 w, u32 h) +static void xp_copy_rect(struct tridentfb_par *par, + u32 x1, u32 y1, u32 x2, u32 y2, u32 w, u32 h) { int direction; u32 x1_tmp, x2_tmp, y1_tmp, y2_tmp; @@ -379,12 +389,12 @@ static void xp_copy_rect(u32 x1, u32 y1, u32 x2, u32 y2, u32 w, u32 h) y2_tmp = y2; } - writemmr(0x2128, direction); - t_outb(ROP_S, 0x2127); - writemmr(0x213C, masked_point(y1_tmp, x1_tmp)); - writemmr(0x2138, masked_point(y2_tmp, x2_tmp)); - writemmr(0x2140, masked_point(h, w)); - t_outb(0x01, 0x2124); + writemmr(par, 0x2128, direction); + t_outb(par, ROP_S, 0x2127); + writemmr(par, 0x213C, masked_point(y1_tmp, x1_tmp)); + writemmr(par, 0x2138, masked_point(y2_tmp, x2_tmp)); + writemmr(par, 0x2140, masked_point(h, w)); + t_outb(par, 0x01, 0x2124); } static struct accel_switch accel_xp = { @@ -397,7 +407,7 @@ static struct accel_switch accel_xp = { /* * Image specific acceleration functions */ -static void image_init_accel(int pitch, int bpp) +static void image_init_accel(struct tridentfb_par *par, int pitch, int bpp) { int tmp = 0; switch (bpp) { @@ -415,40 +425,42 @@ static void image_init_accel(int pitch, int bpp) tmp = 2; break; } - writemmr(0x2120, 0xF0000000); - writemmr(0x2120, 0x40000000 | tmp); - writemmr(0x2120, 0x80000000); - writemmr(0x2144, 0x00000000); - writemmr(0x2148, 0x00000000); - writemmr(0x2150, 0x00000000); - writemmr(0x2154, 0x00000000); - writemmr(0x2120, 0x60000000 | (pitch << 16) | pitch); - writemmr(0x216C, 0x00000000); - writemmr(0x2170, 0x00000000); - writemmr(0x217C, 0x00000000); - writemmr(0x2120, 0x10000000); - writemmr(0x2130, (2047 << 16) | 2047); + writemmr(par, 0x2120, 0xF0000000); + writemmr(par, 0x2120, 0x40000000 | tmp); + writemmr(par, 0x2120, 0x80000000); + writemmr(par, 0x2144, 0x00000000); + writemmr(par, 0x2148, 0x00000000); + writemmr(par, 0x2150, 0x00000000); + writemmr(par, 0x2154, 0x00000000); + writemmr(par, 0x2120, 0x60000000 | (pitch << 16) | pitch); + writemmr(par, 0x216C, 0x00000000); + writemmr(par, 0x2170, 0x00000000); + writemmr(par, 0x217C, 0x00000000); + writemmr(par, 0x2120, 0x10000000); + writemmr(par, 0x2130, (2047 << 16) | 2047); } -static void image_wait_engine(void) +static void image_wait_engine(struct tridentfb_par *par) { - while (readmmr(0x2164) & 0xF0000000) ; + while (readmmr(par, 0x2164) & 0xF0000000) ; } -static void image_fill_rect(u32 x, u32 y, u32 w, u32 h, u32 c, u32 rop) +static void image_fill_rect(struct tridentfb_par *par, + u32 x, u32 y, u32 w, u32 h, u32 c, u32 rop) { - writemmr(0x2120, 0x80000000); - writemmr(0x2120, 0x90000000 | ROP_S); + writemmr(par, 0x2120, 0x80000000); + writemmr(par, 0x2120, 0x90000000 | ROP_S); - writemmr(0x2144, c); + writemmr(par, 0x2144, c); - writemmr(DR1, point(x, y)); - writemmr(DR2, point(x + w - 1, y + h - 1)); + writemmr(par, DR1, point(x, y)); + writemmr(par, DR2, point(x + w - 1, y + h - 1)); - writemmr(0x2124, 0x80000000 | 3 << 22 | 1 << 10 | 1 << 9); + writemmr(par, 0x2124, 0x80000000 | 3 << 22 | 1 << 10 | 1 << 9); } -static void image_copy_rect(u32 x1, u32 y1, u32 x2, u32 y2, u32 w, u32 h) +static void image_copy_rect(struct tridentfb_par *par, + u32 x1, u32 y1, u32 x2, u32 y2, u32 w, u32 h) { u32 s1, s2, d1, d2; int direction = 2; @@ -460,14 +472,15 @@ static void image_copy_rect(u32 x1, u32 y1, u32 x2, u32 y2, u32 w, u32 h) if ((y1 > y2) || ((y1 == y2) && (x1 > x2))) direction = 0; - writemmr(0x2120, 0x80000000); - writemmr(0x2120, 0x90000000 | ROP_S); + writemmr(par, 0x2120, 0x80000000); + writemmr(par, 0x2120, 0x90000000 | ROP_S); - writemmr(SR1, direction ? s2 : s1); - writemmr(SR2, direction ? s1 : s2); - writemmr(DR1, direction ? d2 : d1); - writemmr(DR2, direction ? d1 : d2); - writemmr(0x2124, 0x80000000 | 1 << 22 | 1 << 10 | 1 << 7 | direction); + writemmr(par, SR1, direction ? s2 : s1); + writemmr(par, SR2, direction ? s1 : s2); + writemmr(par, DR1, direction ? d2 : d1); + writemmr(par, DR2, direction ? d1 : d2); + writemmr(par, 0x2124, + 0x80000000 | 1 << 22 | 1 << 10 | 1 << 7 | direction); } static struct accel_switch accel_image = { @@ -484,6 +497,7 @@ static struct accel_switch accel_image = { static void tridentfb_fillrect(struct fb_info *info, const struct fb_fillrect *fr) { + struct tridentfb_par *par = info->par; int bpp = info->var.bits_per_pixel; int col = 0; @@ -502,14 +516,18 @@ static void tridentfb_fillrect(struct fb_info *info, break; } - acc->fill_rect(fr->dx, fr->dy, fr->width, fr->height, col, fr->rop); - acc->wait_engine(); + acc->fill_rect(par, fr->dx, fr->dy, fr->width, + fr->height, col, fr->rop); + acc->wait_engine(par); } static void tridentfb_copyarea(struct fb_info *info, const struct fb_copyarea *ca) { - acc->copy_rect(ca->sx, ca->sy, ca->dx, ca->dy, ca->width, ca->height); - acc->wait_engine(); + struct tridentfb_par *par = info->par; + + acc->copy_rect(par, ca->sx, ca->sy, ca->dx, ca->dy, + ca->width, ca->height); + acc->wait_engine(par); } #else /* !CONFIG_FB_TRIDENT_ACCEL */ #define tridentfb_fillrect cfb_fillrect @@ -521,49 +539,51 @@ static void tridentfb_copyarea(struct fb_info *info, * Hardware access functions */ -static inline unsigned char read3X4(int reg) +static inline unsigned char read3X4(struct tridentfb_par *par, int reg) { - struct tridentfb_par *par = (struct tridentfb_par *)fb_info.par; writeb(reg, par->io_virt + CRT + 4); return readb(par->io_virt + CRT + 5); } -static inline void write3X4(int reg, unsigned char val) +static inline void write3X4(struct tridentfb_par *par, int reg, + unsigned char val) { - struct tridentfb_par *par = (struct tridentfb_par *)fb_info.par; writeb(reg, par->io_virt + CRT + 4); writeb(val, par->io_virt + CRT + 5); } -static inline unsigned char read3C4(int reg) +static inline unsigned char read3C4(struct tridentfb_par *par, int reg) { - t_outb(reg, 0x3C4); - return t_inb(0x3C5); + t_outb(par, reg, 0x3C4); + return t_inb(par, 0x3C5); } -static inline void write3C4(int reg, unsigned char val) +static inline void write3C4(struct tridentfb_par *par, int reg, + unsigned char val) { - t_outb(reg, 0x3C4); - t_outb(val, 0x3C5); + t_outb(par, reg, 0x3C4); + t_outb(par, val, 0x3C5); } -static inline unsigned char read3CE(int reg) +static inline unsigned char read3CE(struct tridentfb_par *par, int reg) { - t_outb(reg, 0x3CE); - return t_inb(0x3CF); + t_outb(par, reg, 0x3CE); + return t_inb(par, 0x3CF); } -static inline void writeAttr(int reg, unsigned char val) +static inline void writeAttr(struct tridentfb_par *par, int reg, + unsigned char val) { - readb(((struct tridentfb_par *)fb_info.par)->io_virt + CRT + 0x0A); /* flip-flop to index */ - t_outb(reg, 0x3C0); - t_outb(val, 0x3C0); + fb_readb(par->io_virt + CRT + 0x0A); /* flip-flop to index */ + t_outb(par, reg, 0x3C0); + t_outb(par, val, 0x3C0); } -static inline void write3CE(int reg, unsigned char val) +static inline void write3CE(struct tridentfb_par *par, int reg, + unsigned char val) { - t_outb(reg, 0x3CE); - t_outb(val, 0x3CF); + t_outb(par, reg, 0x3CE); + t_outb(par, val, 0x3CF); } static void enable_mmio(void) @@ -581,32 +601,35 @@ static void enable_mmio(void) outb(inb(0x3D5) | 0x01, 0x3D5); } -static void disable_mmio(void) +static void disable_mmio(struct tridentfb_par *par) { /* Goto New Mode */ - t_outb(0x0B, 0x3C4); - t_inb(0x3C5); + t_outb(par, 0x0B, 0x3C4); + t_inb(par, 0x3C5); /* Unprotect registers */ - t_outb(NewMode1, 0x3C4); - t_outb(0x80, 0x3C5); + t_outb(par, NewMode1, 0x3C4); + t_outb(par, 0x80, 0x3C5); /* Disable MMIO */ - t_outb(PCIReg, 0x3D4); - t_outb(t_inb(0x3D5) & ~0x01, 0x3D5); + t_outb(par, PCIReg, 0x3D4); + t_outb(par, t_inb(par, 0x3D5) & ~0x01, 0x3D5); } -#define crtc_unlock() write3X4(CRTVSyncEnd, read3X4(CRTVSyncEnd) & 0x7F) +static void crtc_unlock(struct tridentfb_par *par) +{ + write3X4(par, CRTVSyncEnd, read3X4(par, CRTVSyncEnd) & 0x7F); +} /* Return flat panel's maximum x resolution */ -static int __devinit get_nativex(void) +static int __devinit get_nativex(struct tridentfb_par *par) { int x, y, tmp; if (nativex) return nativex; - tmp = (read3CE(VertStretch) >> 4) & 3; + tmp = (read3CE(par, VertStretch) >> 4) & 3; switch (tmp) { case 0: @@ -632,44 +655,45 @@ static int __devinit get_nativex(void) } /* Set pitch */ -static void set_lwidth(int width) +static void set_lwidth(struct tridentfb_par *par, int width) { - write3X4(Offset, width & 0xFF); - write3X4(AddColReg, - (read3X4(AddColReg) & 0xCF) | ((width & 0x300) >> 4)); + write3X4(par, Offset, width & 0xFF); + write3X4(par, AddColReg, + (read3X4(par, AddColReg) & 0xCF) | ((width & 0x300) >> 4)); } /* For resolutions smaller than FP resolution stretch */ -static void screen_stretch(void) +static void screen_stretch(struct tridentfb_par *par) { if (chip_id != CYBERBLADEXPAi1) - write3CE(BiosReg, 0); + write3CE(par, BiosReg, 0); else - write3CE(BiosReg, 8); - write3CE(VertStretch, (read3CE(VertStretch) & 0x7C) | 1); - write3CE(HorStretch, (read3CE(HorStretch) & 0x7C) | 1); + write3CE(par, BiosReg, 8); + write3CE(par, VertStretch, (read3CE(par, VertStretch) & 0x7C) | 1); + write3CE(par, HorStretch, (read3CE(par, HorStretch) & 0x7C) | 1); } /* For resolutions smaller than FP resolution center */ -static void screen_center(void) +static void screen_center(struct tridentfb_par *par) { - write3CE(VertStretch, (read3CE(VertStretch) & 0x7C) | 0x80); - write3CE(HorStretch, (read3CE(HorStretch) & 0x7C) | 0x80); + write3CE(par, VertStretch, (read3CE(par, VertStretch) & 0x7C) | 0x80); + write3CE(par, HorStretch, (read3CE(par, HorStretch) & 0x7C) | 0x80); } /* Address of first shown pixel in display memory */ -static void set_screen_start(int base) +static void set_screen_start(struct tridentfb_par *par, int base) { - write3X4(StartAddrLow, base & 0xFF); - write3X4(StartAddrHigh, (base & 0xFF00) >> 8); - write3X4(CRTCModuleTest, - (read3X4(CRTCModuleTest) & 0xDF) | ((base & 0x10000) >> 11)); - write3X4(CRTHiOrd, - (read3X4(CRTHiOrd) & 0xF8) | ((base & 0xE0000) >> 17)); + u8 tmp; + write3X4(par, StartAddrLow, base & 0xFF); + write3X4(par, StartAddrHigh, (base & 0xFF00) >> 8); + tmp = read3X4(par, CRTCModuleTest) & 0xDF; + write3X4(par, CRTCModuleTest, tmp | ((base & 0x10000) >> 11)); + tmp = read3X4(par, CRTHiOrd) & 0xF8; + write3X4(par, CRTHiOrd, tmp | ((base & 0xE0000) >> 17)); } /* Set dotclock frequency */ -static void set_vclk(unsigned long freq) +static void set_vclk(struct tridentfb_par *par, unsigned long freq) { int m, n, k; unsigned long f, fi, d, di; @@ -690,8 +714,8 @@ static void set_vclk(unsigned long freq) break; } if (chip3D) { - write3C4(ClockHigh, hi); - write3C4(ClockLow, lo); + write3C4(par, ClockHigh, hi); + write3C4(par, ClockLow, lo); } else { outb(lo, 0x43C8); outb(hi, 0x43C9); @@ -700,9 +724,9 @@ static void set_vclk(unsigned long freq) } /* Set number of lines for flat panels*/ -static void set_number_of_lines(int lines) +static void set_number_of_lines(struct tridentfb_par *par, int lines) { - int tmp = read3CE(CyberEnhance) & 0x8F; + int tmp = read3CE(par, CyberEnhance) & 0x8F; if (lines > 1024) tmp |= 0x50; else if (lines > 768) @@ -711,24 +735,24 @@ static void set_number_of_lines(int lines) tmp |= 0x20; else if (lines > 480) tmp |= 0x10; - write3CE(CyberEnhance, tmp); + write3CE(par, CyberEnhance, tmp); } /* * If we see that FP is active we assume we have one. * Otherwise we have a CRT display.User can override. */ -static unsigned int __devinit get_displaytype(void) +static unsigned int __devinit get_displaytype(struct tridentfb_par *par) { if (fp) return DISPLAY_FP; if (crt || !chipcyber) return DISPLAY_CRT; - return (read3CE(FPConfig) & 0x10) ? DISPLAY_FP : DISPLAY_CRT; + return (read3CE(par, FPConfig) & 0x10) ? DISPLAY_FP : DISPLAY_CRT; } /* Try detecting the video memory size */ -static unsigned int __devinit get_memsize(void) +static unsigned int __devinit get_memsize(struct tridentfb_par *par) { unsigned char tmp, tmp2; unsigned int k; @@ -742,7 +766,7 @@ static unsigned int __devinit get_memsize(void) k = 2560 * Kb; break; default: - tmp = read3X4(SPR) & 0x0F; + tmp = read3X4(par, SPR) & 0x0F; switch (tmp) { case 0x01: @@ -774,7 +798,7 @@ static unsigned int __devinit get_memsize(void) break; case 0x0E: /* XP */ - tmp2 = read3C4(0xC1); + tmp2 = read3C4(par, 0xC1); switch (tmp2) { case 0x00: k = 20 * Mb; @@ -862,6 +886,7 @@ static int tridentfb_check_var(struct fb_var_screeninfo *var, static int tridentfb_pan_display(struct fb_var_screeninfo *var, struct fb_info *info) { + struct tridentfb_par *par = info->par; unsigned int offset; debug("enter\n"); @@ -869,13 +894,20 @@ static int tridentfb_pan_display(struct fb_var_screeninfo *var, * var->bits_per_pixel / 32; info->var.xoffset = var->xoffset; info->var.yoffset = var->yoffset; - set_screen_start(offset); + set_screen_start(par, offset); debug("exit\n"); return 0; } -#define shadowmode_on() write3CE(CyberControl, read3CE(CyberControl) | 0x81) -#define shadowmode_off() write3CE(CyberControl, read3CE(CyberControl) & 0x7E) +static void shadowmode_on(struct tridentfb_par *par) +{ + write3CE(par, CyberControl, read3CE(par, CyberControl) | 0x81); +} + +static void shadowmode_off(struct tridentfb_par *par) +{ + write3CE(par, CyberControl, read3CE(par, CyberControl) & 0x7E); +} /* Set the hardware to the requested video mode */ static int tridentfb_set_par(struct fb_info *info) @@ -905,8 +937,8 @@ static int tridentfb_set_par(struct fb_info *info) vblankstart = var->yres; vblankend = vtotal + 2; - crtc_unlock(); - write3CE(CyberControl, 8); + crtc_unlock(par); + write3CE(par, CyberControl, 8); if (flatpanel && var->xres < nativex) { /* @@ -914,35 +946,36 @@ static int tridentfb_set_par(struct fb_info *info) * than requested resolution decide whether * we stretch or center */ - t_outb(0xEB, 0x3C2); + t_outb(par, 0xEB, 0x3C2); - shadowmode_on(); + shadowmode_on(par); if (center) - screen_center(); + screen_center(par); else if (stretch) - screen_stretch(); + screen_stretch(par); } else { - t_outb(0x2B, 0x3C2); - write3CE(CyberControl, 8); + t_outb(par, 0x2B, 0x3C2); + write3CE(par, CyberControl, 8); } /* vertical timing values */ - write3X4(CRTVTotal, vtotal & 0xFF); - write3X4(CRTVDispEnd, vdispend & 0xFF); - write3X4(CRTVSyncStart, vsyncstart & 0xFF); - write3X4(CRTVSyncEnd, (vsyncend & 0x0F)); - write3X4(CRTVBlankStart, vblankstart & 0xFF); - write3X4(CRTVBlankEnd, 0 /* p->vblankend & 0xFF */ ); + write3X4(par, CRTVTotal, vtotal & 0xFF); + write3X4(par, CRTVDispEnd, vdispend & 0xFF); + write3X4(par, CRTVSyncStart, vsyncstart & 0xFF); + write3X4(par, CRTVSyncEnd, (vsyncend & 0x0F)); + write3X4(par, CRTVBlankStart, vblankstart & 0xFF); + write3X4(par, CRTVBlankEnd, 0 /* p->vblankend & 0xFF */); /* horizontal timing values */ - write3X4(CRTHTotal, htotal & 0xFF); - write3X4(CRTHDispEnd, hdispend & 0xFF); - write3X4(CRTHSyncStart, hsyncstart & 0xFF); - write3X4(CRTHSyncEnd, (hsyncend & 0x1F) | ((hblankend & 0x20) << 2)); - write3X4(CRTHBlankStart, hblankstart & 0xFF); - write3X4(CRTHBlankEnd, 0 /* (p->hblankend & 0x1F) */ ); + write3X4(par, CRTHTotal, htotal & 0xFF); + write3X4(par, CRTHDispEnd, hdispend & 0xFF); + write3X4(par, CRTHSyncStart, hsyncstart & 0xFF); + write3X4(par, CRTHSyncEnd, + (hsyncend & 0x1F) | ((hblankend & 0x20) << 2)); + write3X4(par, CRTHBlankStart, hblankstart & 0xFF); + write3X4(par, CRTHBlankEnd, 0 /* (p->hblankend & 0x1F) */); /* higher bits of vertical timing values */ tmp = 0x10; @@ -954,38 +987,40 @@ static int tridentfb_set_par(struct fb_info *info) if (vtotal & 0x200) tmp |= 0x20; if (vdispend & 0x200) tmp |= 0x40; if (vsyncstart & 0x200) tmp |= 0x80; - write3X4(CRTOverflow, tmp); + write3X4(par, CRTOverflow, tmp); - tmp = read3X4(CRTHiOrd) | 0x08; /* line compare bit 10 */ + tmp = read3X4(par, CRTHiOrd) | 0x08; /* line compare bit 10 */ if (vtotal & 0x400) tmp |= 0x80; if (vblankstart & 0x400) tmp |= 0x40; if (vsyncstart & 0x400) tmp |= 0x20; if (vdispend & 0x400) tmp |= 0x10; - write3X4(CRTHiOrd, tmp); + write3X4(par, CRTHiOrd, tmp); tmp = 0; if (htotal & 0x800) tmp |= 0x800 >> 11; if (hblankstart & 0x800) tmp |= 0x800 >> 7; - write3X4(HorizOverflow, tmp); + write3X4(par, HorizOverflow, tmp); tmp = 0x40; if (vblankstart & 0x200) tmp |= 0x20; //FIXME if (info->var.vmode & FB_VMODE_DOUBLE) tmp |= 0x80; /* double scan for 200 line modes */ - write3X4(CRTMaxScanLine, tmp); + write3X4(par, CRTMaxScanLine, tmp); - write3X4(CRTLineCompare, 0xFF); - write3X4(CRTPRowScan, 0); - write3X4(CRTModeControl, 0xC3); + write3X4(par, CRTLineCompare, 0xFF); + write3X4(par, CRTPRowScan, 0); + write3X4(par, CRTModeControl, 0xC3); - write3X4(LinearAddReg, 0x20); /* enable linear addressing */ + write3X4(par, LinearAddReg, 0x20); /* enable linear addressing */ tmp = (info->var.vmode & FB_VMODE_INTERLACED) ? 0x84 : 0x80; - write3X4(CRTCModuleTest, tmp); /* enable access extended memory */ + /* enable access extended memory */ + write3X4(par, CRTCModuleTest, tmp); - write3X4(GraphEngReg, 0x80); /* enable GE for text acceleration */ + /* enable GE for text acceleration */ + write3X4(par, GraphEngReg, 0x80); #ifdef CONFIG_FB_TRIDENT_ACCEL - acc->init_accel(info->var.xres, bpp); + acc->init_accel(par, info->var.xres, bpp); #endif switch (bpp) { @@ -1003,49 +1038,52 @@ static int tridentfb_set_par(struct fb_info *info) break; } - write3X4(PixelBusReg, tmp); + write3X4(par, PixelBusReg, tmp); tmp = 0x10; if (chipcyber) tmp |= 0x20; - write3X4(DRAMControl, tmp); /* both IO, linear enable */ + write3X4(par, DRAMControl, tmp); /* both IO, linear enable */ - write3X4(InterfaceSel, read3X4(InterfaceSel) | 0x40); - write3X4(Performance, 0x92); - write3X4(PCIReg, 0x07); /* MMIO & PCI read and write burst enable */ + write3X4(par, InterfaceSel, read3X4(par, InterfaceSel) | 0x40); + write3X4(par, Performance, 0x92); + /* MMIO & PCI read and write burst enable */ + write3X4(par, PCIReg, 0x07); /* convert from picoseconds to kHz */ vclk = PICOS2KHZ(info->var.pixclock); if (bpp == 32) vclk *= 2; - set_vclk(vclk); + set_vclk(par, vclk); - write3C4(0, 3); - write3C4(1, 1); /* set char clock 8 dots wide */ - write3C4(2, 0x0F); /* enable 4 maps because needed in chain4 mode */ - write3C4(3, 0); - write3C4(4, 0x0E); /* memory mode enable bitmaps ?? */ + write3C4(par, 0, 3); + write3C4(par, 1, 1); /* set char clock 8 dots wide */ + /* enable 4 maps because needed in chain4 mode */ + write3C4(par, 2, 0x0F); + write3C4(par, 3, 0); + write3C4(par, 4, 0x0E); /* memory mode enable bitmaps ?? */ - write3CE(MiscExtFunc, (bpp == 32) ? 0x1A : 0x12); /* divide clock by 2 if 32bpp */ - /* chain4 mode display and CPU path */ - write3CE(0x5, 0x40); /* no CGA compat, allow 256 col */ - write3CE(0x6, 0x05); /* graphics mode */ - write3CE(0x7, 0x0F); /* planes? */ + /* divide clock by 2 if 32bpp chain4 mode display and CPU path */ + write3CE(par, MiscExtFunc, (bpp == 32) ? 0x1A : 0x12); + write3CE(par, 0x5, 0x40); /* no CGA compat, allow 256 col */ + write3CE(par, 0x6, 0x05); /* graphics mode */ + write3CE(par, 0x7, 0x0F); /* planes? */ if (chip_id == CYBERBLADEXPAi1) { /* This fixes snow-effect in 32 bpp */ - write3X4(CRTHSyncStart, 0x84); + write3X4(par, CRTHSyncStart, 0x84); } - writeAttr(0x10, 0x41); /* graphics mode and support 256 color modes */ - writeAttr(0x12, 0x0F); /* planes */ - writeAttr(0x13, 0); /* horizontal pel panning */ + /* graphics mode and support 256 color modes */ + writeAttr(par, 0x10, 0x41); + writeAttr(par, 0x12, 0x0F); /* planes */ + writeAttr(par, 0x13, 0); /* horizontal pel panning */ /* colors */ for (tmp = 0; tmp < 0x10; tmp++) - writeAttr(tmp, tmp); - readb(par->io_virt + CRT + 0x0A); /* flip-flop to index */ - t_outb(0x20, 0x3C0); /* enable attr */ + writeAttr(par, tmp, tmp); + fb_readb(par->io_virt + CRT + 0x0A); /* flip-flop to index */ + t_outb(par, 0x20, 0x3C0); /* enable attr */ switch (bpp) { case 8: @@ -1063,17 +1101,17 @@ static int tridentfb_set_par(struct fb_info *info) break; } - t_inb(0x3C8); - t_inb(0x3C6); - t_inb(0x3C6); - t_inb(0x3C6); - t_inb(0x3C6); - t_outb(tmp, 0x3C6); - t_inb(0x3C8); + t_inb(par, 0x3C8); + t_inb(par, 0x3C6); + t_inb(par, 0x3C6); + t_inb(par, 0x3C6); + t_inb(par, 0x3C6); + t_outb(par, tmp, 0x3C6); + t_inb(par, 0x3C8); if (flatpanel) - set_number_of_lines(info->var.yres); - set_lwidth(info->var.xres * bpp / (4 * 16)); + set_number_of_lines(par, info->var.yres); + set_lwidth(par, info->var.xres * bpp / (4 * 16)); info->fix.visual = (bpp == 8) ? FB_VISUAL_PSEUDOCOLOR : FB_VISUAL_TRUECOLOR; info->fix.line_length = info->var.xres * (bpp >> 3); info->cmap.len = (bpp == 8) ? 256 : 16; @@ -1087,17 +1125,18 @@ static int tridentfb_setcolreg(unsigned regno, unsigned red, unsigned green, struct fb_info *info) { int bpp = info->var.bits_per_pixel; + struct tridentfb_par *par = info->par; if (regno >= info->cmap.len) return 1; if (bpp == 8) { - t_outb(0xFF, 0x3C6); - t_outb(regno, 0x3C8); + t_outb(par, 0xFF, 0x3C6); + t_outb(par, regno, 0x3C8); - t_outb(red >> 10, 0x3C9); - t_outb(green >> 10, 0x3C9); - t_outb(blue >> 10, 0x3C9); + t_outb(par, red >> 10, 0x3C9); + t_outb(par, green >> 10, 0x3C9); + t_outb(par, blue >> 10, 0x3C9); } else if (regno < 16) { if (bpp == 16) { /* RGB 565 */ @@ -1123,13 +1162,14 @@ static int tridentfb_setcolreg(unsigned regno, unsigned red, unsigned green, static int tridentfb_blank(int blank_mode, struct fb_info *info) { unsigned char PMCont, DPMSCont; + struct tridentfb_par *par = info->par; debug("enter\n"); if (flatpanel) return 0; - t_outb(0x04, 0x83C8); /* Read DPMS Control */ - PMCont = t_inb(0x83C6) & 0xFC; - DPMSCont = read3CE(PowerStatus) & 0xFC; + t_outb(par, 0x04, 0x83C8); /* Read DPMS Control */ + PMCont = t_inb(par, 0x83C6) & 0xFC; + DPMSCont = read3CE(par, PowerStatus) & 0xFC; switch (blank_mode) { case FB_BLANK_UNBLANK: /* Screen: On, HSync: On, VSync: On */ @@ -1155,9 +1195,9 @@ static int tridentfb_blank(int blank_mode, struct fb_info *info) break; } - write3CE(PowerStatus, DPMSCont); - t_outb(4, 0x83C8); - t_outb(PMCont, 0x83C6); + write3CE(par, PowerStatus, DPMSCont); + t_outb(par, 4, 0x83C8); + t_outb(par, PMCont, 0x83C6); debug("exit\n"); @@ -1265,11 +1305,11 @@ static int __devinit trident_pci_probe(struct pci_dev * dev, /* setup framebuffer memory */ tridentfb_fix.smem_start = pci_resource_start(dev, 0); - tridentfb_fix.smem_len = get_memsize(); + tridentfb_fix.smem_len = get_memsize(&default_par); if (!request_mem_region(tridentfb_fix.smem_start, tridentfb_fix.smem_len, "tridentfb")) { debug("request_mem_region failed!\n"); - disable_mmio(); + disable_mmio(fb_info.par); err = -1; goto out_unmap1; } @@ -1284,10 +1324,10 @@ static int __devinit trident_pci_probe(struct pci_dev * dev, } output("%s board found\n", pci_name(dev)); - displaytype = get_displaytype(); + displaytype = get_displaytype(&default_par); if (flatpanel) - nativex = get_nativex(); + nativex = get_nativex(&default_par); fb_info.fix = tridentfb_fix; fb_info.fbops = &tridentfb_ops; @@ -1330,7 +1370,7 @@ static int __devinit trident_pci_probe(struct pci_dev * dev, if (fb_info.screen_base) iounmap(fb_info.screen_base); release_mem_region(tridentfb_fix.smem_start, tridentfb_fix.smem_len); - disable_mmio(); + disable_mmio(fb_info.par); out_unmap1: if (default_par.io_virt) iounmap(default_par.io_virt); From e09ed099d0169ac3a22b17cfeece0fa54a9e43eb Mon Sep 17 00:00:00 2001 From: Krzysztof Helt Date: Wed, 23 Jul 2008 21:30:51 -0700 Subject: [PATCH 362/857] tridentfb: convert fb_info into allocated one This patch converts fb_info structure from global variable to allocatable one. The global default_par is moved into function variable. Signed-off-by: Krzysztof Helt Cc: "Antonino A. Daplas" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/tridentfb.c | 82 ++++++++++++++++++++++----------------- 1 file changed, 46 insertions(+), 36 deletions(-) diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c index 3e8a1ef892ce..cb37e10734b6 100644 --- a/drivers/video/tridentfb.c +++ b/drivers/video/tridentfb.c @@ -33,10 +33,7 @@ struct tridentfb_par { static unsigned char eng_oper; /* engine operation... */ static struct fb_ops tridentfb_ops; -static struct tridentfb_par default_par; - /* FIXME:kmalloc these 3 instead */ -static struct fb_info fb_info; static u32 pseudo_pal[16]; static struct fb_var_screeninfo default_var; @@ -1217,16 +1214,23 @@ static struct fb_ops tridentfb_ops = { .fb_imageblit = cfb_imageblit, }; -static int __devinit trident_pci_probe(struct pci_dev * dev, - const struct pci_device_id * id) +static int __devinit trident_pci_probe(struct pci_dev *dev, + const struct pci_device_id *id) { int err; unsigned char revision; + struct fb_info *info; + struct tridentfb_par *default_par; err = pci_enable_device(dev); if (err) return err; + info = framebuffer_alloc(sizeof(struct tridentfb_par), &dev->dev); + if (!info) + return -ENOMEM; + default_par = info->par; + chip_id = id->device; if (chip_id == CYBERBLADEi1) @@ -1282,8 +1286,6 @@ static int __devinit trident_pci_probe(struct pci_dev * dev, /* acceleration is on by default for 3D chips */ defaultaccel = chip3D && !noaccel; - fb_info.par = &default_par; - /* setup MMIO region */ tridentfb_fix.mmio_start = pci_resource_start(dev, 1); tridentfb_fix.mmio_len = chip3D ? 0x20000 : 0x10000; @@ -1293,9 +1295,10 @@ static int __devinit trident_pci_probe(struct pci_dev * dev, return -1; } - default_par.io_virt = ioremap_nocache(tridentfb_fix.mmio_start, tridentfb_fix.mmio_len); + default_par->io_virt = ioremap_nocache(tridentfb_fix.mmio_start, + tridentfb_fix.mmio_len); - if (!default_par.io_virt) { + if (!default_par->io_virt) { debug("ioremap failed\n"); err = -1; goto out_unmap1; @@ -1305,46 +1308,46 @@ static int __devinit trident_pci_probe(struct pci_dev * dev, /* setup framebuffer memory */ tridentfb_fix.smem_start = pci_resource_start(dev, 0); - tridentfb_fix.smem_len = get_memsize(&default_par); + tridentfb_fix.smem_len = get_memsize(default_par); if (!request_mem_region(tridentfb_fix.smem_start, tridentfb_fix.smem_len, "tridentfb")) { debug("request_mem_region failed!\n"); - disable_mmio(fb_info.par); + disable_mmio(info->par); err = -1; goto out_unmap1; } - fb_info.screen_base = ioremap_nocache(tridentfb_fix.smem_start, - tridentfb_fix.smem_len); + info->screen_base = ioremap_nocache(tridentfb_fix.smem_start, + tridentfb_fix.smem_len); - if (!fb_info.screen_base) { + if (!info->screen_base) { debug("ioremap failed\n"); err = -1; goto out_unmap2; } output("%s board found\n", pci_name(dev)); - displaytype = get_displaytype(&default_par); + displaytype = get_displaytype(default_par); if (flatpanel) - nativex = get_nativex(&default_par); + nativex = get_nativex(default_par); - fb_info.fix = tridentfb_fix; - fb_info.fbops = &tridentfb_ops; + info->fix = tridentfb_fix; + info->fbops = &tridentfb_ops; - fb_info.flags = FBINFO_DEFAULT | FBINFO_HWACCEL_YPAN; + info->flags = FBINFO_DEFAULT | FBINFO_HWACCEL_YPAN; #ifdef CONFIG_FB_TRIDENT_ACCEL - fb_info.flags |= FBINFO_HWACCEL_COPYAREA | FBINFO_HWACCEL_FILLRECT; + info->flags |= FBINFO_HWACCEL_COPYAREA | FBINFO_HWACCEL_FILLRECT; #endif - fb_info.pseudo_palette = pseudo_pal; + info->pseudo_palette = pseudo_pal; - if (!fb_find_mode(&default_var, &fb_info, + if (!fb_find_mode(&default_var, info, mode_option, NULL, 0, NULL, bpp)) { err = -EINVAL; goto out_unmap2; } - err = fb_alloc_cmap(&fb_info.cmap, 256, 0); + err = fb_alloc_cmap(&info->cmap, 256, 0); if (err < 0) goto out_unmap2; @@ -1353,39 +1356,46 @@ static int __devinit trident_pci_probe(struct pci_dev * dev, else default_var.accel_flags &= ~FB_ACCELF_TEXT; default_var.activate |= FB_ACTIVATE_NOW; - fb_info.var = default_var; - fb_info.device = &dev->dev; - if (register_framebuffer(&fb_info) < 0) { + info->var = default_var; + info->device = &dev->dev; + if (register_framebuffer(info) < 0) { printk(KERN_ERR "tridentfb: could not register Trident framebuffer\n"); - fb_dealloc_cmap(&fb_info.cmap); + fb_dealloc_cmap(&info->cmap); err = -EINVAL; goto out_unmap2; } output("fb%d: %s frame buffer device %dx%d-%dbpp\n", - fb_info.node, fb_info.fix.id, default_var.xres, + info->node, info->fix.id, default_var.xres, default_var.yres, default_var.bits_per_pixel); + + pci_set_drvdata(dev, info); return 0; out_unmap2: - if (fb_info.screen_base) - iounmap(fb_info.screen_base); + if (info->screen_base) + iounmap(info->screen_base); release_mem_region(tridentfb_fix.smem_start, tridentfb_fix.smem_len); - disable_mmio(fb_info.par); + disable_mmio(info->par); out_unmap1: - if (default_par.io_virt) - iounmap(default_par.io_virt); + if (default_par->io_virt) + iounmap(default_par->io_virt); release_mem_region(tridentfb_fix.mmio_start, tridentfb_fix.mmio_len); + framebuffer_release(info); return err; } static void __devexit trident_pci_remove(struct pci_dev *dev) { - struct tridentfb_par *par = (struct tridentfb_par*)fb_info.par; - unregister_framebuffer(&fb_info); + struct fb_info *info = pci_get_drvdata(dev); + struct tridentfb_par *par = info->par; + + unregister_framebuffer(info); iounmap(par->io_virt); - iounmap(fb_info.screen_base); + iounmap(info->screen_base); release_mem_region(tridentfb_fix.smem_start, tridentfb_fix.smem_len); release_mem_region(tridentfb_fix.mmio_start, tridentfb_fix.mmio_len); + pci_set_drvdata(dev, NULL); + framebuffer_release(info); } /* List of boards that we are trying to support */ From ea8ee55c12f77cbbb6e067f91e0cd794baa692ab Mon Sep 17 00:00:00 2001 From: Krzysztof Helt Date: Wed, 23 Jul 2008 21:30:51 -0700 Subject: [PATCH 363/857] tridentfb: move global pseudo palette into structure This patch moves pseudo palette int tridentfb_par structure and removes global default_var. Signed-off-by: Krzysztof Helt Cc: "Antonino A. Daplas" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/tridentfb.c | 21 +++++++-------------- 1 file changed, 7 insertions(+), 14 deletions(-) diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c index cb37e10734b6..0f6e4054c996 100644 --- a/drivers/video/tridentfb.c +++ b/drivers/video/tridentfb.c @@ -28,16 +28,12 @@ struct tridentfb_par { void __iomem *io_virt; /* iospace virtual memory address */ + u32 pseudo_pal[16]; }; static unsigned char eng_oper; /* engine operation... */ static struct fb_ops tridentfb_ops; -/* FIXME:kmalloc these 3 instead */ -static u32 pseudo_pal[16]; - -static struct fb_var_screeninfo default_var; - static struct fb_fix_screeninfo tridentfb_fix = { .id = "Trident", .type = FB_TYPE_PACKED_PIXELS, @@ -1340,9 +1336,7 @@ static int __devinit trident_pci_probe(struct pci_dev *dev, #ifdef CONFIG_FB_TRIDENT_ACCEL info->flags |= FBINFO_HWACCEL_COPYAREA | FBINFO_HWACCEL_FILLRECT; #endif - info->pseudo_palette = pseudo_pal; - - if (!fb_find_mode(&default_var, info, + if (!fb_find_mode(&info->var, info, mode_option, NULL, 0, NULL, bpp)) { err = -EINVAL; goto out_unmap2; @@ -1352,11 +1346,10 @@ static int __devinit trident_pci_probe(struct pci_dev *dev, goto out_unmap2; if (defaultaccel && acc) - default_var.accel_flags |= FB_ACCELF_TEXT; + info->var.accel_flags |= FB_ACCELF_TEXT; else - default_var.accel_flags &= ~FB_ACCELF_TEXT; - default_var.activate |= FB_ACTIVATE_NOW; - info->var = default_var; + info->var.accel_flags &= ~FB_ACCELF_TEXT; + info->var.activate |= FB_ACTIVATE_NOW; info->device = &dev->dev; if (register_framebuffer(info) < 0) { printk(KERN_ERR "tridentfb: could not register Trident framebuffer\n"); @@ -1365,8 +1358,8 @@ static int __devinit trident_pci_probe(struct pci_dev *dev, goto out_unmap2; } output("fb%d: %s frame buffer device %dx%d-%dbpp\n", - info->node, info->fix.id, default_var.xres, - default_var.yres, default_var.bits_per_pixel); + info->node, info->fix.id, info->var.xres, + info->var.yres, info->var.bits_per_pixel); pci_set_drvdata(dev, info); return 0; From 122e8ad3cbf172043ea93f2db8e107fa9f9b0192 Mon Sep 17 00:00:00 2001 From: Krzysztof Helt Date: Wed, 23 Jul 2008 21:30:52 -0700 Subject: [PATCH 364/857] tridentfb: move global chip_id into structure This patch moves the chip_id into tridentfb_par structure and removes global chip_id related constants. It also bumps version of the driver to 0.7.9 Signed-off-by: Krzysztof Helt Cc: "Antonino A. Daplas" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/tridentfb.c | 27 +++++++++++++-------------- 1 file changed, 13 insertions(+), 14 deletions(-) diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c index 0f6e4054c996..dfe52b424c97 100644 --- a/drivers/video/tridentfb.c +++ b/drivers/video/tridentfb.c @@ -24,11 +24,12 @@ #include #include