From 269e4a4122685d0739f0e8e53b440111bf8a03f9 Mon Sep 17 00:00:00 2001 From: Mika Westerberg Date: Wed, 4 Sep 2013 13:37:43 +0300 Subject: [PATCH 01/45] spi/pxa2xx: check status register as well to determine if the device is off The current interrupt handler calls pm_runtime_suspended() to check if the device is suspended or not. However, runtime PM status of the device is only set to suspended once all PM runtime suspend hooks have executed. In case of Intel Lynxpoint we have the device bound to the ACPI power domain and its runtime suspend hook will put the device to D3hot (or D3cold if possible). This means that the device is powered off before its state is set to runtime suspended. While in this state the device might get an interrupt that is meant for another device (as the interrupt line is shared), and because the device is powered off accessing its registers will return 0xffffffff that the driver misinterprets as an invalid state. When this happens user will see messages like below on the console: pxa2xx-spi INT33C0:00: bad message state in interrupt handler Fix this by checking the status register for ~0 and returning IRQ_NONE in that case. Signed-off-by: Mika Westerberg Signed-off-by: Mark Brown --- drivers/spi/spi-pxa2xx.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/drivers/spi/spi-pxa2xx.c b/drivers/spi/spi-pxa2xx.c index 2eb06ee0b326..c1a50674c1e3 100644 --- a/drivers/spi/spi-pxa2xx.c +++ b/drivers/spi/spi-pxa2xx.c @@ -546,8 +546,17 @@ static irqreturn_t ssp_int(int irq, void *dev_id) if (pm_runtime_suspended(&drv_data->pdev->dev)) return IRQ_NONE; - sccr1_reg = read_SSCR1(reg); + /* + * If the device is not yet in RPM suspended state and we get an + * interrupt that is meant for another device, check if status bits + * are all set to one. That means that the device is already + * powered off. + */ status = read_SSSR(reg); + if (status == ~0) + return IRQ_NONE; + + sccr1_reg = read_SSCR1(reg); /* Ignore possible writes if we don't need to write */ if (!(sccr1_reg & SSCR1_TIE)) From 9b4003a687abcf2fe25ba9073dd07345e4529183 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Tue, 10 Sep 2013 10:19:29 +0800 Subject: [PATCH 02/45] spi: clps711x: Don't call kfree() after spi_master_put/spi_unregister_master Calling kfree() to clean up the memory obtained from spi_alloc_master() is wrong as this is done in spi_master_release() when spi_master->dev's refcount reaches zero. Signed-off-by: Axel Lin Signed-off-by: Mark Brown --- drivers/spi/spi-clps711x.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/spi/spi-clps711x.c b/drivers/spi/spi-clps711x.c index 5655acf55bfe..b597a92565dc 100644 --- a/drivers/spi/spi-clps711x.c +++ b/drivers/spi/spi-clps711x.c @@ -247,7 +247,6 @@ static int spi_clps711x_probe(struct platform_device *pdev) gpio_free(hw->chipselect[i]); spi_master_put(master); - kfree(master); return ret; } @@ -263,7 +262,6 @@ static int spi_clps711x_remove(struct platform_device *pdev) gpio_free(hw->chipselect[i]); spi_unregister_master(master); - kfree(master); return 0; } From eadf69cf5e1cdd042fb7b7beeffa538a935373cd Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Wed, 11 Sep 2013 19:15:39 +0800 Subject: [PATCH 03/45] spi: mpc512x: fix error return code in mpc512x_psc_spi_do_probe() Fix to return a negative error code from the error handling case instead of 0, as done elsewhere in this function. Signed-off-by: Wei Yongjun Signed-off-by: Mark Brown --- drivers/spi/spi-mpc512x-psc.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/spi/spi-mpc512x-psc.c b/drivers/spi/spi-mpc512x-psc.c index dbc5e999a1f5..6adf4e35816d 100644 --- a/drivers/spi/spi-mpc512x-psc.c +++ b/drivers/spi/spi-mpc512x-psc.c @@ -522,8 +522,10 @@ static int mpc512x_psc_spi_do_probe(struct device *dev, u32 regaddr, psc_num = master->bus_num; snprintf(clk_name, sizeof(clk_name), "psc%d_mclk", psc_num); clk = devm_clk_get(dev, clk_name); - if (IS_ERR(clk)) + if (IS_ERR(clk)) { + ret = PTR_ERR(clk); goto free_irq; + } ret = clk_prepare_enable(clk); if (ret) goto free_irq; From 272bfbd54f29b9361b09dbd6b25ccbd1a30163f1 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Wed, 11 Sep 2013 19:15:08 +0800 Subject: [PATCH 04/45] spi: fix return value check in dspi_probe() In case of error, the function devm_ioremap_resource() returns ERR_PTR() and never returns NULL. The NULL test in the return value check should be replaced with IS_ERR(). Signed-off-by: Wei Yongjun Signed-off-by: Mark Brown --- drivers/spi/spi-fsl-dspi.c | 10 ++-------- 1 file changed, 2 insertions(+), 8 deletions(-) diff --git a/drivers/spi/spi-fsl-dspi.c b/drivers/spi/spi-fsl-dspi.c index 6cd07d13ecab..4e44575bd87a 100644 --- a/drivers/spi/spi-fsl-dspi.c +++ b/drivers/spi/spi-fsl-dspi.c @@ -476,15 +476,9 @@ static int dspi_probe(struct platform_device *pdev) master->bus_num = bus_num; res = platform_get_resource(pdev, IORESOURCE_MEM, 0); - if (!res) { - dev_err(&pdev->dev, "can't get platform resource\n"); - ret = -EINVAL; - goto out_master_put; - } - dspi->base = devm_ioremap_resource(&pdev->dev, res); - if (!dspi->base) { - ret = -EINVAL; + if (IS_ERR(dspi->base)) { + ret = PTR_ERR(dspi->base); goto out_master_put; } From 47d06e532e95b71c0db3839ebdef3fe8812fca2c Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Tue, 10 Sep 2013 10:52:35 -0400 Subject: [PATCH 05/45] random: run random_int_secret_init() run after all late_initcalls The some platforms (e.g., ARM) initializes their clocks as late_initcalls for some unknown reason. So make sure random_int_secret_init() is run after all of the late_initcalls are run. Cc: stable@vger.kernel.org Signed-off-by: "Theodore Ts'o" --- drivers/char/random.c | 3 +-- include/linux/random.h | 1 + init/main.c | 2 ++ 3 files changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/char/random.c b/drivers/char/random.c index 0d91fe52f3f5..92e6c67e1ae6 100644 --- a/drivers/char/random.c +++ b/drivers/char/random.c @@ -1462,12 +1462,11 @@ struct ctl_table random_table[] = { static u32 random_int_secret[MD5_MESSAGE_BYTES / 4] ____cacheline_aligned; -static int __init random_int_secret_init(void) +int random_int_secret_init(void) { get_random_bytes(random_int_secret, sizeof(random_int_secret)); return 0; } -late_initcall(random_int_secret_init); /* * Get a random word for internal kernel use only. Similar to urandom but diff --git a/include/linux/random.h b/include/linux/random.h index 3b9377d6b7a5..6312dd9ba449 100644 --- a/include/linux/random.h +++ b/include/linux/random.h @@ -17,6 +17,7 @@ extern void add_interrupt_randomness(int irq, int irq_flags); extern void get_random_bytes(void *buf, int nbytes); extern void get_random_bytes_arch(void *buf, int nbytes); void generate_random_uuid(unsigned char uuid_out[16]); +extern int random_int_secret_init(void); #ifndef MODULE extern const struct file_operations random_fops, urandom_fops; diff --git a/init/main.c b/init/main.c index d03d2ec2eacf..586cd3359c02 100644 --- a/init/main.c +++ b/init/main.c @@ -75,6 +75,7 @@ #include #include #include +#include #include #include @@ -778,6 +779,7 @@ static void __init do_basic_setup(void) do_ctors(); usermodehelper_enable(); do_initcalls(); + random_int_secret_init(); } static void __init do_pre_smp_initcalls(void) From 19872d20c890073c5207d9e02bb8f14d451a11eb Mon Sep 17 00:00:00 2001 From: David Herrmann Date: Mon, 9 Sep 2013 18:33:54 +0200 Subject: [PATCH 06/45] HID: uhid: allocate static minor udev has this nice feature of creating "dead" /dev/ device-nodes if it finds a devnode: modalias. Once the node is accessed, the kernel automatically loads the module that provides the node. However, this requires udev to know the major:minor code to use for the node. This feature was introduced by: commit 578454ff7eab61d13a26b568f99a89a2c9edc881 Author: Kay Sievers Date: Thu May 20 18:07:20 2010 +0200 driver core: add devname module aliases to allow module on-demand auto-loading However, uhid uses dynamic minor numbers so this doesn't actually work. We need to load uhid to know which minor it's going to use. Hence, allocate a static minor (just like uinput does) and we're good to go. Reported-by: Tom Gundersen Signed-off-by: David Herrmann Signed-off-by: Jiri Kosina --- drivers/hid/uhid.c | 3 ++- include/linux/miscdevice.h | 1 + 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/hid/uhid.c b/drivers/hid/uhid.c index 5bf2fb785844..93b00d76374c 100644 --- a/drivers/hid/uhid.c +++ b/drivers/hid/uhid.c @@ -615,7 +615,7 @@ static const struct file_operations uhid_fops = { static struct miscdevice uhid_misc = { .fops = &uhid_fops, - .minor = MISC_DYNAMIC_MINOR, + .minor = UHID_MINOR, .name = UHID_NAME, }; @@ -634,4 +634,5 @@ module_exit(uhid_exit); MODULE_LICENSE("GPL"); MODULE_AUTHOR("David Herrmann "); MODULE_DESCRIPTION("User-space I/O driver support for HID subsystem"); +MODULE_ALIAS_MISCDEV(UHID_MINOR); MODULE_ALIAS("devname:" UHID_NAME); diff --git a/include/linux/miscdevice.h b/include/linux/miscdevice.h index 09c2300ddb37..cb358355ef43 100644 --- a/include/linux/miscdevice.h +++ b/include/linux/miscdevice.h @@ -45,6 +45,7 @@ #define MAPPER_CTRL_MINOR 236 #define LOOP_CTRL_MINOR 237 #define VHOST_NET_MINOR 238 +#define UHID_MINOR 239 #define MISC_DYNAMIC_MINOR 255 struct device; From 0158026ccced1ae312233a6a46dc0b6edce3fb6c Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Fri, 27 Sep 2013 16:21:30 +0800 Subject: [PATCH 07/45] spi/clps711x: drop clk_put for devm_clk_get in spi_clps711x_probe() devm_clk_get() is used so there is no reason to explicitly call clk_put() in probe or remove functions. Signed-off-by: Wei Yongjun Signed-off-by: Mark Brown --- drivers/spi/spi-clps711x.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/spi/spi-clps711x.c b/drivers/spi/spi-clps711x.c index b597a92565dc..6416798828e7 100644 --- a/drivers/spi/spi-clps711x.c +++ b/drivers/spi/spi-clps711x.c @@ -226,7 +226,6 @@ static int spi_clps711x_probe(struct platform_device *pdev) dev_name(&pdev->dev), hw); if (ret) { dev_err(&pdev->dev, "Can't request IRQ\n"); - clk_put(hw->spi_clk); goto clk_out; } From 5cb1327172281cadb7ee8c5fa294d7ac8e09b8db Mon Sep 17 00:00:00 2001 From: Brian Norris Date: Mon, 16 Sep 2013 17:59:20 -0700 Subject: [PATCH 08/45] mtd: nand: fix memory leak in ONFI extended parameter page This fixes a memory leak in the ONFI support code for detecting the required ECC levels from this commit: commit 6dcbe0cdd83fb5f77be4f44c9e06c535281c375a Author: Huang Shijie Date: Wed May 22 10:28:27 2013 +0800 mtd: get the ECC info from the Extended Parameter Page In the success case, we never freed the 'ep' buffer. Also, this fixes an oversight in the same commit where we (harmlessly) freed the NULL pointer. Signed-off-by: Brian Norris Acked-by: Huang Shijie Signed-off-by: David Woodhouse --- drivers/mtd/nand/nand_base.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/drivers/mtd/nand/nand_base.c b/drivers/mtd/nand/nand_base.c index 7ed4841327f2..d340b2f198c6 100644 --- a/drivers/mtd/nand/nand_base.c +++ b/drivers/mtd/nand/nand_base.c @@ -2869,10 +2869,8 @@ static int nand_flash_detect_ext_param_page(struct mtd_info *mtd, len = le16_to_cpu(p->ext_param_page_length) * 16; ep = kmalloc(len, GFP_KERNEL); - if (!ep) { - ret = -ENOMEM; - goto ext_out; - } + if (!ep) + return -ENOMEM; /* Send our own NAND_CMD_PARAM. */ chip->cmdfunc(mtd, NAND_CMD_PARAM, 0, -1); @@ -2920,7 +2918,7 @@ static int nand_flash_detect_ext_param_page(struct mtd_info *mtd, } pr_info("ONFI extended param page detected.\n"); - return 0; + ret = 0; ext_out: kfree(ep); From 2b468ef0e7959b703626b64c4d264ef822c9267a Mon Sep 17 00:00:00 2001 From: Elie De Brauwer Date: Tue, 17 Sep 2013 19:48:22 +0200 Subject: [PATCH 09/45] mtd: m25p80: Fix 4 byte addressing mode for Micron devices. According to the datasheet for Micron n25q256a (N25Q256A13ESF40F) 4-byte addressing mode should be entered as follows: To enter or exit the 4-byte address mode, the WRITE ENABLE command must be executed to set the write enable latch bit to 1. (Note: The WRITE ENABLE command must NOT be executed on the N25Q256A83ESF40x and N25Q256A83E1240x devices.) S# must be driven LOW. The effect of the command is immediate; after the command has been executed, the write enable latch bit is cleared to 0. Micron's portable way to perform this for all types of Micron flash is to first issue a write enable, then switch the addressing mode followed by a write disable to avoid leaving the flash in a write- able state. Signed-off-by: Elie De Brauwer [Brian: reworked a bit] Signed-off-by: Brian Norris Signed-off-by: David Woodhouse --- drivers/mtd/devices/m25p80.c | 17 +++++++++++++++-- 1 file changed, 15 insertions(+), 2 deletions(-) diff --git a/drivers/mtd/devices/m25p80.c b/drivers/mtd/devices/m25p80.c index 26b14f9fcac6..6bc9618af094 100644 --- a/drivers/mtd/devices/m25p80.c +++ b/drivers/mtd/devices/m25p80.c @@ -168,12 +168,25 @@ static inline int write_disable(struct m25p *flash) */ static inline int set_4byte(struct m25p *flash, u32 jedec_id, int enable) { + int status; + bool need_wren = false; + switch (JEDEC_MFR(jedec_id)) { - case CFI_MFR_MACRONIX: case CFI_MFR_ST: /* Micron, actually */ + /* Some Micron need WREN command; all will accept it */ + need_wren = true; + case CFI_MFR_MACRONIX: case 0xEF /* winbond */: + if (need_wren) + write_enable(flash); + flash->command[0] = enable ? OPCODE_EN4B : OPCODE_EX4B; - return spi_write(flash->spi, flash->command, 1); + status = spi_write(flash->spi, flash->command, 1); + + if (need_wren) + write_disable(flash); + + return status; default: /* Spansion style */ flash->command[0] = OPCODE_BRWR; From 3e2bd64d243ff0f3dc08c49ff218fed42ee9d981 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Fri, 27 Sep 2013 11:52:35 +0100 Subject: [PATCH 10/45] spi/s3c64xx: Ensure runtime PM is enabled prior to registration Otherwise we may try to start transfers immediately and then fail to runtime resume the device causing us not to have clocks enabled. Signed-off-by: Mark Brown Reviewed-by: Sylwester Nawrocki Conflicts: drivers/spi/spi-s3c64xx.c --- drivers/spi/spi-s3c64xx.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/spi/spi-s3c64xx.c b/drivers/spi/spi-s3c64xx.c index 512b8893893b..a80376dc3a10 100644 --- a/drivers/spi/spi-s3c64xx.c +++ b/drivers/spi/spi-s3c64xx.c @@ -1428,6 +1428,8 @@ static int s3c64xx_spi_probe(struct platform_device *pdev) S3C64XX_SPI_INT_TX_OVERRUN_EN | S3C64XX_SPI_INT_TX_UNDERRUN_EN, sdd->regs + S3C64XX_SPI_INT_EN); + pm_runtime_enable(&pdev->dev); + if (spi_register_master(master)) { dev_err(&pdev->dev, "cannot register SPI master\n"); ret = -EBUSY; @@ -1440,8 +1442,6 @@ static int s3c64xx_spi_probe(struct platform_device *pdev) mem_res, sdd->rx_dma.dmach, sdd->tx_dma.dmach); - pm_runtime_enable(&pdev->dev); - return 0; err3: From b924a69067b00d3121debae5a738fb0bcbbbb03c Mon Sep 17 00:00:00 2001 From: Chen Gang Date: Wed, 25 Sep 2013 12:14:08 +0800 Subject: [PATCH 11/45] tile: include: asm: use 'long long' instead of 'u64' for atomic64_t and its related functions atomic* value is signed value, and atomic* functions need also process signed value (parameter value, and return value), so use 'long long' instead of 'u64'. After replacement, it will also fix a bug for atomic64_add_negative(): "u64 is never less than 0". The modifications are: in vim, use "1,% s/\/long long/g" command. remove redundant '__aligned(8)'. be sure of 80 (and macro '\') columns limitation after replacement. Signed-off-by: Chen Gang Signed-off-by: Chris Metcalf [re-instated const cast] --- arch/tile/include/asm/atomic.h | 5 +++-- arch/tile/include/asm/atomic_32.h | 27 +++++++++++++++------------ arch/tile/include/asm/cmpxchg.h | 28 +++++++++++++++++----------- arch/tile/lib/atomic_32.c | 8 ++++---- 4 files changed, 39 insertions(+), 29 deletions(-) diff --git a/arch/tile/include/asm/atomic.h b/arch/tile/include/asm/atomic.h index d385eaadece7..709798460763 100644 --- a/arch/tile/include/asm/atomic.h +++ b/arch/tile/include/asm/atomic.h @@ -166,7 +166,7 @@ static inline int atomic_cmpxchg(atomic_t *v, int o, int n) * * Atomically sets @v to @i and returns old @v */ -static inline u64 atomic64_xchg(atomic64_t *v, u64 n) +static inline long long atomic64_xchg(atomic64_t *v, long long n) { return xchg64(&v->counter, n); } @@ -180,7 +180,8 @@ static inline u64 atomic64_xchg(atomic64_t *v, u64 n) * Atomically checks if @v holds @o and replaces it with @n if so. * Returns the old value at @v. */ -static inline u64 atomic64_cmpxchg(atomic64_t *v, u64 o, u64 n) +static inline long long atomic64_cmpxchg(atomic64_t *v, long long o, + long long n) { return cmpxchg64(&v->counter, o, n); } diff --git a/arch/tile/include/asm/atomic_32.h b/arch/tile/include/asm/atomic_32.h index 0d0395b1b152..1ad4a1f7d42b 100644 --- a/arch/tile/include/asm/atomic_32.h +++ b/arch/tile/include/asm/atomic_32.h @@ -80,7 +80,7 @@ static inline void atomic_set(atomic_t *v, int n) /* A 64bit atomic type */ typedef struct { - u64 __aligned(8) counter; + long long counter; } atomic64_t; #define ATOMIC64_INIT(val) { (val) } @@ -91,14 +91,14 @@ typedef struct { * * Atomically reads the value of @v. */ -static inline u64 atomic64_read(const atomic64_t *v) +static inline long long atomic64_read(const atomic64_t *v) { /* * Requires an atomic op to read both 32-bit parts consistently. * Casting away const is safe since the atomic support routines * do not write to memory if the value has not been modified. */ - return _atomic64_xchg_add((u64 *)&v->counter, 0); + return _atomic64_xchg_add((long long *)&v->counter, 0); } /** @@ -108,7 +108,7 @@ static inline u64 atomic64_read(const atomic64_t *v) * * Atomically adds @i to @v. */ -static inline void atomic64_add(u64 i, atomic64_t *v) +static inline void atomic64_add(long long i, atomic64_t *v) { _atomic64_xchg_add(&v->counter, i); } @@ -120,7 +120,7 @@ static inline void atomic64_add(u64 i, atomic64_t *v) * * Atomically adds @i to @v and returns @i + @v */ -static inline u64 atomic64_add_return(u64 i, atomic64_t *v) +static inline long long atomic64_add_return(long long i, atomic64_t *v) { smp_mb(); /* barrier for proper semantics */ return _atomic64_xchg_add(&v->counter, i) + i; @@ -135,7 +135,8 @@ static inline u64 atomic64_add_return(u64 i, atomic64_t *v) * Atomically adds @a to @v, so long as @v was not already @u. * Returns non-zero if @v was not @u, and zero otherwise. */ -static inline u64 atomic64_add_unless(atomic64_t *v, u64 a, u64 u) +static inline long long atomic64_add_unless(atomic64_t *v, long long a, + long long u) { smp_mb(); /* barrier for proper semantics */ return _atomic64_xchg_add_unless(&v->counter, a, u) != u; @@ -151,7 +152,7 @@ static inline u64 atomic64_add_unless(atomic64_t *v, u64 a, u64 u) * atomic64_set() can't be just a raw store, since it would be lost if it * fell between the load and store of one of the other atomic ops. */ -static inline void atomic64_set(atomic64_t *v, u64 n) +static inline void atomic64_set(atomic64_t *v, long long n) { _atomic64_xchg(&v->counter, n); } @@ -236,11 +237,13 @@ extern struct __get_user __atomic_xchg_add_unless(volatile int *p, extern struct __get_user __atomic_or(volatile int *p, int *lock, int n); extern struct __get_user __atomic_andn(volatile int *p, int *lock, int n); extern struct __get_user __atomic_xor(volatile int *p, int *lock, int n); -extern u64 __atomic64_cmpxchg(volatile u64 *p, int *lock, u64 o, u64 n); -extern u64 __atomic64_xchg(volatile u64 *p, int *lock, u64 n); -extern u64 __atomic64_xchg_add(volatile u64 *p, int *lock, u64 n); -extern u64 __atomic64_xchg_add_unless(volatile u64 *p, - int *lock, u64 o, u64 n); +extern long long __atomic64_cmpxchg(volatile long long *p, int *lock, + long long o, long long n); +extern long long __atomic64_xchg(volatile long long *p, int *lock, long long n); +extern long long __atomic64_xchg_add(volatile long long *p, int *lock, + long long n); +extern long long __atomic64_xchg_add_unless(volatile long long *p, + int *lock, long long o, long long n); /* Return failure from the atomic wrappers. */ struct __get_user __atomic_bad_address(int __user *addr); diff --git a/arch/tile/include/asm/cmpxchg.h b/arch/tile/include/asm/cmpxchg.h index 4001d5eab4bb..0ccda3c425be 100644 --- a/arch/tile/include/asm/cmpxchg.h +++ b/arch/tile/include/asm/cmpxchg.h @@ -35,10 +35,10 @@ int _atomic_xchg(int *ptr, int n); int _atomic_xchg_add(int *v, int i); int _atomic_xchg_add_unless(int *v, int a, int u); int _atomic_cmpxchg(int *ptr, int o, int n); -u64 _atomic64_xchg(u64 *v, u64 n); -u64 _atomic64_xchg_add(u64 *v, u64 i); -u64 _atomic64_xchg_add_unless(u64 *v, u64 a, u64 u); -u64 _atomic64_cmpxchg(u64 *v, u64 o, u64 n); +long long _atomic64_xchg(long long *v, long long n); +long long _atomic64_xchg_add(long long *v, long long i); +long long _atomic64_xchg_add_unless(long long *v, long long a, long long u); +long long _atomic64_cmpxchg(long long *v, long long o, long long n); #define xchg(ptr, n) \ ({ \ @@ -53,7 +53,8 @@ u64 _atomic64_cmpxchg(u64 *v, u64 o, u64 n); if (sizeof(*(ptr)) != 4) \ __cmpxchg_called_with_bad_pointer(); \ smp_mb(); \ - (typeof(*(ptr)))_atomic_cmpxchg((int *)ptr, (int)o, (int)n); \ + (typeof(*(ptr)))_atomic_cmpxchg((int *)ptr, (int)o, \ + (int)n); \ }) #define xchg64(ptr, n) \ @@ -61,7 +62,8 @@ u64 _atomic64_cmpxchg(u64 *v, u64 o, u64 n); if (sizeof(*(ptr)) != 8) \ __xchg_called_with_bad_pointer(); \ smp_mb(); \ - (typeof(*(ptr)))_atomic64_xchg((u64 *)(ptr), (u64)(n)); \ + (typeof(*(ptr)))_atomic64_xchg((long long *)(ptr), \ + (long long)(n)); \ }) #define cmpxchg64(ptr, o, n) \ @@ -69,7 +71,8 @@ u64 _atomic64_cmpxchg(u64 *v, u64 o, u64 n); if (sizeof(*(ptr)) != 8) \ __cmpxchg_called_with_bad_pointer(); \ smp_mb(); \ - (typeof(*(ptr)))_atomic64_cmpxchg((u64 *)ptr, (u64)o, (u64)n); \ + (typeof(*(ptr)))_atomic64_cmpxchg((long long *)ptr, \ + (long long)o, (long long)n); \ }) #else @@ -81,10 +84,11 @@ u64 _atomic64_cmpxchg(u64 *v, u64 o, u64 n); switch (sizeof(*(ptr))) { \ case 4: \ __x = (typeof(__x))(unsigned long) \ - __insn_exch4((ptr), (u32)(unsigned long)(n)); \ + __insn_exch4((ptr), \ + (u32)(unsigned long)(n)); \ break; \ case 8: \ - __x = (typeof(__x)) \ + __x = (typeof(__x)) \ __insn_exch((ptr), (unsigned long)(n)); \ break; \ default: \ @@ -103,10 +107,12 @@ u64 _atomic64_cmpxchg(u64 *v, u64 o, u64 n); switch (sizeof(*(ptr))) { \ case 4: \ __x = (typeof(__x))(unsigned long) \ - __insn_cmpexch4((ptr), (u32)(unsigned long)(n)); \ + __insn_cmpexch4((ptr), \ + (u32)(unsigned long)(n)); \ break; \ case 8: \ - __x = (typeof(__x))__insn_cmpexch((ptr), (u64)(n)); \ + __x = (typeof(__x))__insn_cmpexch((ptr), \ + (long long)(n)); \ break; \ default: \ __cmpxchg_called_with_bad_pointer(); \ diff --git a/arch/tile/lib/atomic_32.c b/arch/tile/lib/atomic_32.c index 759efa337be8..c89b211fd9e7 100644 --- a/arch/tile/lib/atomic_32.c +++ b/arch/tile/lib/atomic_32.c @@ -107,19 +107,19 @@ unsigned long _atomic_xor(volatile unsigned long *p, unsigned long mask) EXPORT_SYMBOL(_atomic_xor); -u64 _atomic64_xchg(u64 *v, u64 n) +long long _atomic64_xchg(long long *v, long long n) { return __atomic64_xchg(v, __atomic_setup(v), n); } EXPORT_SYMBOL(_atomic64_xchg); -u64 _atomic64_xchg_add(u64 *v, u64 i) +long long _atomic64_xchg_add(long long *v, long long i) { return __atomic64_xchg_add(v, __atomic_setup(v), i); } EXPORT_SYMBOL(_atomic64_xchg_add); -u64 _atomic64_xchg_add_unless(u64 *v, u64 a, u64 u) +long long _atomic64_xchg_add_unless(long long *v, long long a, long long u) { /* * Note: argument order is switched here since it is easier @@ -130,7 +130,7 @@ u64 _atomic64_xchg_add_unless(u64 *v, u64 a, u64 u) } EXPORT_SYMBOL(_atomic64_xchg_add_unless); -u64 _atomic64_cmpxchg(u64 *v, u64 o, u64 n) +long long _atomic64_cmpxchg(long long *v, long long o, long long n) { return __atomic64_cmpxchg(v, __atomic_setup(v), o, n); } From dd386e03d9753889dc3ea3eed91ad3729df08819 Mon Sep 17 00:00:00 2001 From: Henrik Austad Date: Wed, 25 Sep 2013 16:16:40 +0200 Subject: [PATCH 12/45] tile: change lock initalization in hardwall hardwall used __SPIN_LOCK_INITIALIZER directly instead of the preferred __SPIN_LOCK_UNLOCKED. This also has the benefit that it will compile when applying the preempt-rt patch series. Signed-off-by: Henrik Austad Signed-off-by: Chris Metcalf --- arch/tile/kernel/hardwall.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/tile/kernel/hardwall.c b/arch/tile/kernel/hardwall.c index df27a1fd94a3..531f4c365351 100644 --- a/arch/tile/kernel/hardwall.c +++ b/arch/tile/kernel/hardwall.c @@ -66,7 +66,7 @@ static struct hardwall_type hardwall_types[] = { 0, "udn", LIST_HEAD_INIT(hardwall_types[HARDWALL_UDN].list), - __SPIN_LOCK_INITIALIZER(hardwall_types[HARDWALL_UDN].lock), + __SPIN_LOCK_UNLOCKED(hardwall_types[HARDWALL_UDN].lock), NULL }, #ifndef __tilepro__ @@ -77,7 +77,7 @@ static struct hardwall_type hardwall_types[] = { 1, /* disabled pending hypervisor support */ "idn", LIST_HEAD_INIT(hardwall_types[HARDWALL_IDN].list), - __SPIN_LOCK_INITIALIZER(hardwall_types[HARDWALL_IDN].lock), + __SPIN_LOCK_UNLOCKED(hardwall_types[HARDWALL_IDN].lock), NULL }, { /* access to user-space IPI */ @@ -87,7 +87,7 @@ static struct hardwall_type hardwall_types[] = { 0, "ipi", LIST_HEAD_INIT(hardwall_types[HARDWALL_IPI].list), - __SPIN_LOCK_INITIALIZER(hardwall_types[HARDWALL_IPI].lock), + __SPIN_LOCK_UNLOCKED(hardwall_types[HARDWALL_IPI].lock), NULL }, #endif From 3f725c5b924e14eb00c58892d21d92100121e5ce Mon Sep 17 00:00:00 2001 From: Chris Metcalf Date: Thu, 26 Sep 2013 13:22:40 -0400 Subject: [PATCH 13/45] tile: ensure interrupts disabled for preempt_schedule_irq() When coming from a page fault (for example), interrupts might be enabled as we enter the code to return from interrupt. Cc: stable@vger.kernel.org Signed-off-by: Chris Metcalf --- arch/tile/kernel/intvec_32.S | 3 +++ arch/tile/kernel/intvec_64.S | 3 +++ 2 files changed, 6 insertions(+) diff --git a/arch/tile/kernel/intvec_32.S b/arch/tile/kernel/intvec_32.S index 088d5c141e68..2cbe6d5dd6b0 100644 --- a/arch/tile/kernel/intvec_32.S +++ b/arch/tile/kernel/intvec_32.S @@ -815,6 +815,9 @@ STD_ENTRY(interrupt_return) } bzt r28, 1f bnz r29, 1f + /* Disable interrupts explicitly for preemption. */ + IRQ_DISABLE(r20,r21) + TRACE_IRQS_OFF jal preempt_schedule_irq FEEDBACK_REENTER(interrupt_return) 1: diff --git a/arch/tile/kernel/intvec_64.S b/arch/tile/kernel/intvec_64.S index ec755d3f3734..b8fc497f2437 100644 --- a/arch/tile/kernel/intvec_64.S +++ b/arch/tile/kernel/intvec_64.S @@ -841,6 +841,9 @@ STD_ENTRY(interrupt_return) } beqzt r28, 1f bnez r29, 1f + /* Disable interrupts explicitly for preemption. */ + IRQ_DISABLE(r20,r21) + TRACE_IRQS_OFF jal preempt_schedule_irq FEEDBACK_REENTER(interrupt_return) 1: From 550dbf47819239c7ceb27bbd9879e2feb0633aab Mon Sep 17 00:00:00 2001 From: Stefan Achatz Date: Sat, 28 Sep 2013 05:57:46 +0200 Subject: [PATCH 14/45] HID: roccat: Fix "cannot create duplicate filename" problems Fixing some wrong macro stringification/concatenation. Acked-by: Greg Kroah-Hartman Signed-off-by: Stefan Achatz Signed-off-by: Jiri Kosina --- drivers/hid/hid-roccat-kone.c | 2 +- drivers/hid/hid-roccat-koneplus.c | 4 ++-- drivers/hid/hid-roccat-kovaplus.c | 4 ++-- drivers/hid/hid-roccat-pyra.c | 4 ++-- 4 files changed, 7 insertions(+), 7 deletions(-) diff --git a/drivers/hid/hid-roccat-kone.c b/drivers/hid/hid-roccat-kone.c index 602c188e9d86..6101816a7ddd 100644 --- a/drivers/hid/hid-roccat-kone.c +++ b/drivers/hid/hid-roccat-kone.c @@ -382,7 +382,7 @@ static ssize_t kone_sysfs_write_profilex(struct file *fp, } #define PROFILE_ATTR(number) \ static struct bin_attribute bin_attr_profile##number = { \ - .attr = { .name = "profile##number", .mode = 0660 }, \ + .attr = { .name = "profile" #number, .mode = 0660 }, \ .size = sizeof(struct kone_profile), \ .read = kone_sysfs_read_profilex, \ .write = kone_sysfs_write_profilex, \ diff --git a/drivers/hid/hid-roccat-koneplus.c b/drivers/hid/hid-roccat-koneplus.c index 5ddf605b6b89..5e99fcdc71b9 100644 --- a/drivers/hid/hid-roccat-koneplus.c +++ b/drivers/hid/hid-roccat-koneplus.c @@ -229,13 +229,13 @@ static ssize_t koneplus_sysfs_read_profilex_buttons(struct file *fp, #define PROFILE_ATTR(number) \ static struct bin_attribute bin_attr_profile##number##_settings = { \ - .attr = { .name = "profile##number##_settings", .mode = 0440 }, \ + .attr = { .name = "profile" #number "_settings", .mode = 0440 }, \ .size = KONEPLUS_SIZE_PROFILE_SETTINGS, \ .read = koneplus_sysfs_read_profilex_settings, \ .private = &profile_numbers[number-1], \ }; \ static struct bin_attribute bin_attr_profile##number##_buttons = { \ - .attr = { .name = "profile##number##_buttons", .mode = 0440 }, \ + .attr = { .name = "profile" #number "_buttons", .mode = 0440 }, \ .size = KONEPLUS_SIZE_PROFILE_BUTTONS, \ .read = koneplus_sysfs_read_profilex_buttons, \ .private = &profile_numbers[number-1], \ diff --git a/drivers/hid/hid-roccat-kovaplus.c b/drivers/hid/hid-roccat-kovaplus.c index 515bc03136c0..0c8e1ef0b67d 100644 --- a/drivers/hid/hid-roccat-kovaplus.c +++ b/drivers/hid/hid-roccat-kovaplus.c @@ -257,13 +257,13 @@ static ssize_t kovaplus_sysfs_read_profilex_buttons(struct file *fp, #define PROFILE_ATTR(number) \ static struct bin_attribute bin_attr_profile##number##_settings = { \ - .attr = { .name = "profile##number##_settings", .mode = 0440 }, \ + .attr = { .name = "profile" #number "_settings", .mode = 0440 }, \ .size = KOVAPLUS_SIZE_PROFILE_SETTINGS, \ .read = kovaplus_sysfs_read_profilex_settings, \ .private = &profile_numbers[number-1], \ }; \ static struct bin_attribute bin_attr_profile##number##_buttons = { \ - .attr = { .name = "profile##number##_buttons", .mode = 0440 }, \ + .attr = { .name = "profile" #number "_buttons", .mode = 0440 }, \ .size = KOVAPLUS_SIZE_PROFILE_BUTTONS, \ .read = kovaplus_sysfs_read_profilex_buttons, \ .private = &profile_numbers[number-1], \ diff --git a/drivers/hid/hid-roccat-pyra.c b/drivers/hid/hid-roccat-pyra.c index 5a6dbbeee790..1a07e07d99a0 100644 --- a/drivers/hid/hid-roccat-pyra.c +++ b/drivers/hid/hid-roccat-pyra.c @@ -225,13 +225,13 @@ static ssize_t pyra_sysfs_read_profilex_buttons(struct file *fp, #define PROFILE_ATTR(number) \ static struct bin_attribute bin_attr_profile##number##_settings = { \ - .attr = { .name = "profile##number##_settings", .mode = 0440 }, \ + .attr = { .name = "profile" #number "_settings", .mode = 0440 }, \ .size = PYRA_SIZE_PROFILE_SETTINGS, \ .read = pyra_sysfs_read_profilex_settings, \ .private = &profile_numbers[number-1], \ }; \ static struct bin_attribute bin_attr_profile##number##_buttons = { \ - .attr = { .name = "profile##number##_buttons", .mode = 0440 }, \ + .attr = { .name = "profile" #number "_buttons", .mode = 0440 }, \ .size = PYRA_SIZE_PROFILE_BUTTONS, \ .read = pyra_sysfs_read_profilex_buttons, \ .private = &profile_numbers[number-1], \ From 8f0ba630af45235a6866054704531d49e524e49a Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Wed, 11 Sep 2013 06:49:20 +0800 Subject: [PATCH 15/45] s390/3270: fix return value check in tty3270_resize_work() In case of error, the function tty3270_alloc_screen() returns ERR_PTR() and never returns NULL. The NULL test in the return value check should be replaced with IS_ERR(). Signed-off-by: Wei Yongjun Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- drivers/s390/char/tty3270.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/s390/char/tty3270.c b/drivers/s390/char/tty3270.c index a0f47c83fd62..3f4ca4e09a4c 100644 --- a/drivers/s390/char/tty3270.c +++ b/drivers/s390/char/tty3270.c @@ -810,7 +810,7 @@ static void tty3270_resize_work(struct work_struct *work) struct winsize ws; screen = tty3270_alloc_screen(tp->n_rows, tp->n_cols); - if (!screen) + if (IS_ERR(screen)) return; /* Switch to new output size */ spin_lock_bh(&tp->view.lock); From 742343551165f314c740a41b070d0fc7fbe202f7 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Mon, 2 Sep 2013 10:22:21 +0200 Subject: [PATCH 16/45] s390/kprobes: add exrl to list of prohibited opcodes "execute relative long" may have all sorts of side effects dependend on the instructions it executes. Therefore prohibit setting a kprobe on exrl just like we do for the regular execute instruction. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/kprobes.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/arch/s390/kernel/kprobes.c b/arch/s390/kernel/kprobes.c index 0ce9fb245034..d86e64eddb42 100644 --- a/arch/s390/kernel/kprobes.c +++ b/arch/s390/kernel/kprobes.c @@ -67,6 +67,11 @@ static int __kprobes is_prohibited_opcode(kprobe_opcode_t *insn) case 0xac: /* stnsm */ case 0xad: /* stosm */ return -EINVAL; + case 0xc6: + switch (insn[0] & 0x0f) { + case 0x00: /* exrl */ + return -EINVAL; + } } switch (insn[0]) { case 0x0101: /* pr */ @@ -180,7 +185,6 @@ static int __kprobes is_insn_relative_long(kprobe_opcode_t *insn) break; case 0xc6: switch (insn[0] & 0x0f) { - case 0x00: /* exrl */ case 0x02: /* pfdrl */ case 0x04: /* cghrl */ case 0x05: /* chrl */ From 22bfda6688c9b91393a07887182ec35acc9d06bb Mon Sep 17 00:00:00 2001 From: Christian Borntraeger Date: Tue, 17 Sep 2013 13:28:50 +0200 Subject: [PATCH 17/45] s390/sclp: properly detect line mode console To detect a line mode console we need a message event type for the receive mask and a command event type for the send mask. Checking for a MSG event in the sclp send mask was wrong and might result in line mode consoles not being detected. Signed-off-by: Christian Borntraeger Signed-off-by: Martin Schwidefsky --- drivers/s390/char/sclp_cmd.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/drivers/s390/char/sclp_cmd.c b/drivers/s390/char/sclp_cmd.c index 8cd34bf644b3..77df9cb00688 100644 --- a/drivers/s390/char/sclp_cmd.c +++ b/drivers/s390/char/sclp_cmd.c @@ -145,9 +145,11 @@ bool __init sclp_has_linemode(void) if (sccb->header.response_code != 0x20) return 0; - if (sccb->sclp_send_mask & (EVTYP_MSG_MASK | EVTYP_PMSGCMD_MASK)) - return 1; - return 0; + if (!(sccb->sclp_send_mask & (EVTYP_OPCMD_MASK | EVTYP_PMSGCMD_MASK))) + return 0; + if (!(sccb->sclp_receive_mask & (EVTYP_MSG_MASK | EVTYP_PMSGCMD_MASK))) + return 0; + return 1; } bool __init sclp_has_vt220(void) From 4d3b0664a0d0c0dc7501cfbd067f74226aeca510 Mon Sep 17 00:00:00 2001 From: Michael Holzheu Date: Fri, 27 Sep 2013 15:41:43 +0200 Subject: [PATCH 18/45] s390: Allow vmalloc target buffers for copy_from_oldmem() Currently copy_from_oldmem() is not able to copy to virtual memory. When using kexec pre-allocated ELF header, copy_from_oldmem() is used to copy the ELF notes information to vmalloc buffers. So fix this and use the new function copy_from_realmem() that allows copying also to vmalloc memory. Signed-off-by: Michael Holzheu Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/crash_dump.c | 42 +++++++++++++++++------------------ 1 file changed, 20 insertions(+), 22 deletions(-) diff --git a/arch/s390/kernel/crash_dump.c b/arch/s390/kernel/crash_dump.c index c84f33d51f7b..7dd21720e5b0 100644 --- a/arch/s390/kernel/crash_dump.c +++ b/arch/s390/kernel/crash_dump.c @@ -40,28 +40,26 @@ static inline void *load_real_addr(void *addr) } /* - * Copy up to one page to vmalloc or real memory + * Copy real to virtual or real memory */ -static ssize_t copy_page_real(void *buf, void *src, size_t csize) +static int copy_from_realmem(void *dest, void *src, size_t count) { - size_t size; + unsigned long size; + int rc; - if (is_vmalloc_addr(buf)) { - BUG_ON(csize >= PAGE_SIZE); - /* If buf is not page aligned, copy first part */ - size = min(roundup(__pa(buf), PAGE_SIZE) - __pa(buf), csize); - if (size) { - if (memcpy_real(load_real_addr(buf), src, size)) - return -EFAULT; - buf += size; - src += size; - } - /* Copy second part */ - size = csize - size; - return (size) ? memcpy_real(load_real_addr(buf), src, size) : 0; - } else { - return memcpy_real(buf, src, csize); - } + if (!count) + return 0; + if (!is_vmalloc_or_module_addr(dest)) + return memcpy_real(dest, src, count); + do { + size = min(count, PAGE_SIZE - (__pa(dest) & ~PAGE_MASK)); + if (memcpy_real(load_real_addr(dest), src, size)) + return -EFAULT; + count -= size; + dest += size; + src += size; + } while (count); + return 0; } /* @@ -114,7 +112,7 @@ static ssize_t copy_oldmem_page_kdump(char *buf, size_t csize, rc = copy_to_user_real((void __force __user *) buf, (void *) src, csize); else - rc = copy_page_real(buf, (void *) src, csize); + rc = copy_from_realmem(buf, (void *) src, csize); return (rc == 0) ? rc : csize; } @@ -210,7 +208,7 @@ int copy_from_oldmem(void *dest, void *src, size_t count) if (OLDMEM_BASE) { if ((unsigned long) src < OLDMEM_SIZE) { copied = min(count, OLDMEM_SIZE - (unsigned long) src); - rc = memcpy_real(dest, src + OLDMEM_BASE, copied); + rc = copy_from_realmem(dest, src + OLDMEM_BASE, copied); if (rc) return rc; } @@ -223,7 +221,7 @@ int copy_from_oldmem(void *dest, void *src, size_t count) return rc; } } - return memcpy_real(dest + copied, src + copied, count - copied); + return copy_from_realmem(dest + copied, src + copied, count - copied); } /* From dbbfe487e5f3fc00c9fe5207d63309859704d12f Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Fri, 27 Sep 2013 15:24:38 +0200 Subject: [PATCH 19/45] s390: fix system call restart after inferior call Git commit 616498813b11ffef "s390: system call path micro optimization" introduced a regression in regard to system call restarting and inferior function calls via the ptrace interface. The pointer to the system call table needs to be loaded in sysc_sigpending if do_signal returns with TIF_SYSCALl set after it restored a system call context. Cc: stable@vger.kernel.org # 3.10+ Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/entry.S | 1 + arch/s390/kernel/entry64.S | 1 + 2 files changed, 2 insertions(+) diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S index cc30d1fb000c..0dc2b6d0a1ec 100644 --- a/arch/s390/kernel/entry.S +++ b/arch/s390/kernel/entry.S @@ -266,6 +266,7 @@ sysc_sigpending: tm __TI_flags+3(%r12),_TIF_SYSCALL jno sysc_return lm %r2,%r7,__PT_R2(%r11) # load svc arguments + l %r10,__TI_sysc_table(%r12) # 31 bit system call table xr %r8,%r8 # svc 0 returns -ENOSYS clc __PT_INT_CODE+2(2,%r11),BASED(.Lnr_syscalls+2) jnl sysc_nr_ok # invalid svc number -> do svc 0 diff --git a/arch/s390/kernel/entry64.S b/arch/s390/kernel/entry64.S index 2b2188b97c6a..e5b43c97a834 100644 --- a/arch/s390/kernel/entry64.S +++ b/arch/s390/kernel/entry64.S @@ -297,6 +297,7 @@ sysc_sigpending: tm __TI_flags+7(%r12),_TIF_SYSCALL jno sysc_return lmg %r2,%r7,__PT_R2(%r11) # load svc arguments + lg %r10,__TI_sysc_table(%r12) # address of system call table lghi %r8,0 # svc 0 returns -ENOSYS llgh %r1,__PT_INT_CODE+2(%r11) # load new svc number cghi %r1,NR_syscalls From f862eefec0b68e099a9fa58d3761ffb10bad97e1 Mon Sep 17 00:00:00 2001 From: Chris Metcalf Date: Thu, 26 Sep 2013 13:24:53 -0400 Subject: [PATCH 20/45] tile: use a more conservative __my_cpu_offset in CONFIG_PREEMPT It turns out the kernel relies on barrier() to force a reload of the percpu offset value. Since we can't easily modify the definition of barrier() to include "tp" as an output register, we instead provide a definition of __my_cpu_offset as extended assembly that includes a fake stack read to hazard against barrier(), forcing gcc to know that it must reread "tp" and recompute anything based on "tp" after a barrier. This fixes observed hangs in the slub allocator when we are looping on a percpu cmpxchg_double. A similar fix for ARMv7 was made in June in change 509eb76ebf97. Cc: stable@vger.kernel.org Signed-off-by: Chris Metcalf --- arch/tile/include/asm/percpu.h | 34 +++++++++++++++++++++++++++++++--- 1 file changed, 31 insertions(+), 3 deletions(-) diff --git a/arch/tile/include/asm/percpu.h b/arch/tile/include/asm/percpu.h index 63294f5a8efb..4f7ae39fa202 100644 --- a/arch/tile/include/asm/percpu.h +++ b/arch/tile/include/asm/percpu.h @@ -15,9 +15,37 @@ #ifndef _ASM_TILE_PERCPU_H #define _ASM_TILE_PERCPU_H -register unsigned long __my_cpu_offset __asm__("tp"); -#define __my_cpu_offset __my_cpu_offset -#define set_my_cpu_offset(tp) (__my_cpu_offset = (tp)) +register unsigned long my_cpu_offset_reg asm("tp"); + +#ifdef CONFIG_PREEMPT +/* + * For full preemption, we can't just use the register variable + * directly, since we need barrier() to hazard against it, causing the + * compiler to reload anything computed from a previous "tp" value. + * But we also don't want to use volatile asm, since we'd like the + * compiler to be able to cache the value across multiple percpu reads. + * So we use a fake stack read as a hazard against barrier(). + * The 'U' constraint is like 'm' but disallows postincrement. + */ +static inline unsigned long __my_cpu_offset(void) +{ + unsigned long tp; + register unsigned long *sp asm("sp"); + asm("move %0, tp" : "=r" (tp) : "U" (*sp)); + return tp; +} +#define __my_cpu_offset __my_cpu_offset() +#else +/* + * We don't need to hazard against barrier() since "tp" doesn't ever + * change with PREEMPT_NONE, and with PREEMPT_VOLUNTARY it only + * changes at function call points, at which we are already re-reading + * the value of "tp" due to "my_cpu_offset_reg" being a global variable. + */ +#define __my_cpu_offset my_cpu_offset_reg +#endif + +#define set_my_cpu_offset(tp) (my_cpu_offset_reg = (tp)) #include From 0cc96a745059e8daf1d7d6a26672f0ad9056e989 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Fri, 27 Sep 2013 11:26:30 +0300 Subject: [PATCH 21/45] arch: tile: re-use kbasename() helper kbasename() returns the filename part of a pathname. Signed-off-by: Andy Shevchenko --- arch/tile/kernel/stack.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) diff --git a/arch/tile/kernel/stack.c b/arch/tile/kernel/stack.c index 362284af3afd..c93977a62116 100644 --- a/arch/tile/kernel/stack.c +++ b/arch/tile/kernel/stack.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include #include @@ -332,21 +333,18 @@ static void describe_addr(struct KBacktraceIterator *kbt, } if (vma->vm_file) { - char *s; p = d_path(&vma->vm_file->f_path, buf, bufsize); if (IS_ERR(p)) p = "?"; - s = strrchr(p, '/'); - if (s) - p = s+1; + name = kbasename(p); } else { - p = "anon"; + name = "anon"; } /* Generate a string description of the vma info. */ - namelen = strlen(p); + namelen = strlen(name); remaining = (bufsize - 1) - namelen; - memmove(buf, p, namelen); + memmove(buf, name, namelen); snprintf(buf + namelen, remaining, "[%lx+%lx] ", vma->vm_start, vma->vm_end - vma->vm_start); } From 0f5a24c6602063e014ee48892ebf56093241106e Mon Sep 17 00:00:00 2001 From: Manoj Chourasia Date: Tue, 1 Oct 2013 15:39:00 +0530 Subject: [PATCH 22/45] HID: hidraw: close underlying device at removal of last reader Even though device exist bit is set the underlying HW device should be closed when the last reader of the device is closed i.e. open count drops to zero. Signed-off-by: Manoj Chourasia Reported-by: mika.westerberg@linux.intel.com Tested-by: mika.westerberg@linux.intel.com Signed-off-by: Jiri Kosina --- drivers/hid/hidraw.c | 21 ++++++++++++++------- 1 file changed, 14 insertions(+), 7 deletions(-) diff --git a/drivers/hid/hidraw.c b/drivers/hid/hidraw.c index 8918dd12bb69..6a6dd5cd7833 100644 --- a/drivers/hid/hidraw.c +++ b/drivers/hid/hidraw.c @@ -308,18 +308,25 @@ static int hidraw_fasync(int fd, struct file *file, int on) static void drop_ref(struct hidraw *hidraw, int exists_bit) { if (exists_bit) { - hid_hw_close(hidraw->hid); hidraw->exist = 0; - if (hidraw->open) + if (hidraw->open) { + hid_hw_close(hidraw->hid); wake_up_interruptible(&hidraw->wait); + } } else { --hidraw->open; } - - if (!hidraw->open && !hidraw->exist) { - device_destroy(hidraw_class, MKDEV(hidraw_major, hidraw->minor)); - hidraw_table[hidraw->minor] = NULL; - kfree(hidraw); + if (!hidraw->open) { + if (!hidraw->exist) { + device_destroy(hidraw_class, + MKDEV(hidraw_major, hidraw->minor)); + hidraw_table[hidraw->minor] = NULL; + kfree(hidraw); + } else { + /* close device for last reader */ + hid_hw_power(hidraw->hid, PM_HINT_NORMAL); + hid_hw_close(hidraw->hid); + } } } From 7da7cbbbeb60e0939fecdf9723b388136c175e5c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Anders=20F=2E=20U=2E=20Ki=C3=A6r?= Date: Tue, 1 Oct 2013 19:22:05 +0200 Subject: [PATCH 23/45] HID: add Holtek USB ID 04d9:a081 SHARKOON DarkGlider MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Added id, bindings and comments for Holtek USB ID 04d9:a081 SHARKOON DarkGlider Gaming mouse to use the same corrections of the report descriptor as Holtek 04d9:a04a. As the mouse exceed HID_MAX_USAGES at the same offsets in the reported descriptor. Tested on the hardware. Signed-off-by: Anders F. U. Kiær Signed-off-by: Jiri Kosina --- drivers/hid/Kconfig | 1 + drivers/hid/hid-core.c | 1 + drivers/hid/hid-holtek-mouse.c | 4 ++++ drivers/hid/hid-ids.h | 1 + 4 files changed, 7 insertions(+) diff --git a/drivers/hid/Kconfig b/drivers/hid/Kconfig index 71b70e3a7a71..c91d547191dd 100644 --- a/drivers/hid/Kconfig +++ b/drivers/hid/Kconfig @@ -241,6 +241,7 @@ config HID_HOLTEK - Sharkoon Drakonia / Perixx MX-2000 gaming mice - Tracer Sniper TRM-503 / NOVA Gaming Slider X200 / Zalman ZM-GM1 + - SHARKOON DarkGlider Gaming mouse config HOLTEK_FF bool "Holtek On Line Grip force feedback support" diff --git a/drivers/hid/hid-core.c b/drivers/hid/hid-core.c index b8470b1a10fe..5a8c01112a23 100644 --- a/drivers/hid/hid-core.c +++ b/drivers/hid/hid-core.c @@ -1715,6 +1715,7 @@ static const struct hid_device_id hid_have_special_driver[] = { { HID_USB_DEVICE(USB_VENDOR_ID_HOLTEK_ALT, USB_DEVICE_ID_HOLTEK_ALT_KEYBOARD) }, { HID_USB_DEVICE(USB_VENDOR_ID_HOLTEK_ALT, USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A04A) }, { HID_USB_DEVICE(USB_VENDOR_ID_HOLTEK_ALT, USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A067) }, + { HID_USB_DEVICE(USB_VENDOR_ID_HOLTEK_ALT, USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A081) }, { HID_USB_DEVICE(USB_VENDOR_ID_HUION, USB_DEVICE_ID_HUION_580) }, { HID_USB_DEVICE(USB_VENDOR_ID_JESS2, USB_DEVICE_ID_JESS2_COLOR_RUMBLE_PAD) }, { HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_ION, USB_DEVICE_ID_ICADE) }, diff --git a/drivers/hid/hid-holtek-mouse.c b/drivers/hid/hid-holtek-mouse.c index 7e6db3cf46f9..e696566cde46 100644 --- a/drivers/hid/hid-holtek-mouse.c +++ b/drivers/hid/hid-holtek-mouse.c @@ -27,6 +27,7 @@ * - USB ID 04d9:a067, sold as Sharkoon Drakonia and Perixx MX-2000 * - USB ID 04d9:a04a, sold as Tracer Sniper TRM-503, NOVA Gaming Slider X200 * and Zalman ZM-GM1 + * - USB ID 04d9:a081, sold as SHARKOON DarkGlider Gaming mouse */ static __u8 *holtek_mouse_report_fixup(struct hid_device *hdev, __u8 *rdesc, @@ -46,6 +47,7 @@ static __u8 *holtek_mouse_report_fixup(struct hid_device *hdev, __u8 *rdesc, } break; case USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A04A: + case USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A081: if (*rsize >= 113 && rdesc[106] == 0xff && rdesc[107] == 0x7f && rdesc[111] == 0xff && rdesc[112] == 0x7f) { hid_info(hdev, "Fixing up report descriptor\n"); @@ -63,6 +65,8 @@ static const struct hid_device_id holtek_mouse_devices[] = { USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A067) }, { HID_USB_DEVICE(USB_VENDOR_ID_HOLTEK_ALT, USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A04A) }, + { HID_USB_DEVICE(USB_VENDOR_ID_HOLTEK_ALT, + USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A081) }, { } }; MODULE_DEVICE_TABLE(hid, holtek_mouse_devices); diff --git a/drivers/hid/hid-ids.h b/drivers/hid/hid-ids.h index e60e8d530697..9cbc7ab07dfa 100644 --- a/drivers/hid/hid-ids.h +++ b/drivers/hid/hid-ids.h @@ -450,6 +450,7 @@ #define USB_DEVICE_ID_HOLTEK_ALT_KEYBOARD 0xa055 #define USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A067 0xa067 #define USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A04A 0xa04a +#define USB_DEVICE_ID_HOLTEK_ALT_MOUSE_A081 0xa081 #define USB_VENDOR_ID_IMATION 0x0718 #define USB_DEVICE_ID_DISC_STAKKA 0xd000 From 19514fc665ffbce624785f76ee7ad0ea6378a527 Mon Sep 17 00:00:00 2001 From: Robert Richter Date: Wed, 17 Jul 2013 18:05:06 +0200 Subject: [PATCH 24/45] arm, kbuild: make "make install" not depend on vmlinux Install targets (install, zinstall, uinstall) on arm have a dependency to vmlinux. This may cause parts of the kernel to be rebuilt during installation. We must avoid this since this may run as root. Install targets "ABSOLUTELY MUST NOT MODIFY THE SOURCE TREE." as Linus emphasized this in: http://lkml.org/lkml/2013/7/10/600 So on arm and maybe other archs we need the same as for x86: 1648e4f8 x86, kbuild: make "make install" not depend on vmlinux This patch fixes this for arm. Dependencies are removed and instead a check to install.sh is added for the files that are needed. This issue was uncovered by this build error where the -j option is used in conjunction with install targets: $ make $ make zinstall ... DEPMOD Usage: .../scripts/depmod.sh /sbin/depmod (INSTALL_MOD_PATH and INSTALL_PATH variables set, so no root perms required in this case.) The problem is that zinstall on arm due to its dependency to vmlinux does a prepare/prepare3 and finally does a forced rewrite of kernel.release even if it exists already. Rebuilding kernel.release removes it first and then recreates it. This might race with another parallel make job running depmod. So this patch should fix this one too. Also quoting $(KERNELRELEASE) arg for install.sh as this messes argument order in case it is empty (which is the case if the kernel was not built yet). Signed-off-by: Robert Richter Signed-off-by: Robert Richter Acked-by: Michal Marek . Acked-by: Linus Torvalds Signed-off-by: "Yann E. MORIN" Signed-off-by: Michal Marek --- arch/arm/Makefile | 9 +++++++-- arch/arm/boot/Makefile | 16 ++++++++-------- arch/arm/boot/install.sh | 14 ++++++++++++++ 3 files changed, 29 insertions(+), 10 deletions(-) diff --git a/arch/arm/Makefile b/arch/arm/Makefile index a37a50f575a2..db50b626be98 100644 --- a/arch/arm/Makefile +++ b/arch/arm/Makefile @@ -296,10 +296,15 @@ archprepare: # Convert bzImage to zImage bzImage: zImage -zImage Image xipImage bootpImage uImage: vmlinux +BOOT_TARGETS = zImage Image xipImage bootpImage uImage +INSTALL_TARGETS = zinstall uinstall install + +PHONY += bzImage $(BOOT_TARGETS) $(INSTALL_TARGETS) + +$(BOOT_TARGETS): vmlinux $(Q)$(MAKE) $(build)=$(boot) MACHINE=$(MACHINE) $(boot)/$@ -zinstall uinstall install: vmlinux +$(INSTALL_TARGETS): $(Q)$(MAKE) $(build)=$(boot) MACHINE=$(MACHINE) $@ %.dtb: | scripts diff --git a/arch/arm/boot/Makefile b/arch/arm/boot/Makefile index 84aa2caf07ed..ec2f8065f955 100644 --- a/arch/arm/boot/Makefile +++ b/arch/arm/boot/Makefile @@ -95,24 +95,24 @@ initrd: @test "$(INITRD)" != "" || \ (echo You must specify INITRD; exit -1) -install: $(obj)/Image - $(CONFIG_SHELL) $(srctree)/$(src)/install.sh $(KERNELRELEASE) \ +install: + $(CONFIG_SHELL) $(srctree)/$(src)/install.sh "$(KERNELRELEASE)" \ $(obj)/Image System.map "$(INSTALL_PATH)" -zinstall: $(obj)/zImage - $(CONFIG_SHELL) $(srctree)/$(src)/install.sh $(KERNELRELEASE) \ +zinstall: + $(CONFIG_SHELL) $(srctree)/$(src)/install.sh "$(KERNELRELEASE)" \ $(obj)/zImage System.map "$(INSTALL_PATH)" -uinstall: $(obj)/uImage - $(CONFIG_SHELL) $(srctree)/$(src)/install.sh $(KERNELRELEASE) \ +uinstall: + $(CONFIG_SHELL) $(srctree)/$(src)/install.sh "$(KERNELRELEASE)" \ $(obj)/uImage System.map "$(INSTALL_PATH)" zi: - $(CONFIG_SHELL) $(srctree)/$(src)/install.sh $(KERNELRELEASE) \ + $(CONFIG_SHELL) $(srctree)/$(src)/install.sh "$(KERNELRELEASE)" \ $(obj)/zImage System.map "$(INSTALL_PATH)" i: - $(CONFIG_SHELL) $(srctree)/$(src)/install.sh $(KERNELRELEASE) \ + $(CONFIG_SHELL) $(srctree)/$(src)/install.sh "$(KERNELRELEASE)" \ $(obj)/Image System.map "$(INSTALL_PATH)" subdir- := bootp compressed dts diff --git a/arch/arm/boot/install.sh b/arch/arm/boot/install.sh index 06ea7d42ce8e..2a45092a40e3 100644 --- a/arch/arm/boot/install.sh +++ b/arch/arm/boot/install.sh @@ -20,6 +20,20 @@ # $4 - default install path (blank if root directory) # +verify () { + if [ ! -f "$1" ]; then + echo "" 1>&2 + echo " *** Missing file: $1" 1>&2 + echo ' *** You need to run "make" before "make install".' 1>&2 + echo "" 1>&2 + exit 1 + fi +} + +# Make sure the files actually exist +verify "$2" +verify "$3" + # User may have a custom install script if [ -x ~/bin/${INSTALLKERNEL} ]; then exec ~/bin/${INSTALLKERNEL} "$@"; fi if [ -x /sbin/${INSTALLKERNEL} ]; then exec /sbin/${INSTALLKERNEL} "$@"; fi From 268d76430d1b68c340687357ffd18b4b12d02269 Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Thu, 3 Oct 2013 02:15:50 -0700 Subject: [PATCH 25/45] spi/hspi: fixup Runtime PM enable timing 3e00a09d2fbd64f0ad98e7c8c29dbf9e038fc746 (spi/hspi: Convert to core runtime PM) enabled master->auto_runtime_pm. Then, pm_runtime_enable() is required *before* spi_register_master() calling. This patch fixed it up. Kernel will hang up with "spi_master spi0: Failed to power device: -13" message without this patch. Signed-off-by: Kuninori Morimoto Signed-off-by: Mark Brown --- drivers/spi/spi-sh-hspi.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/spi/spi-sh-hspi.c b/drivers/spi/spi-sh-hspi.c index 0b68cb592fa4..e488a90a98b8 100644 --- a/drivers/spi/spi-sh-hspi.c +++ b/drivers/spi/spi-sh-hspi.c @@ -296,6 +296,8 @@ static int hspi_probe(struct platform_device *pdev) goto error1; } + pm_runtime_enable(&pdev->dev); + master->num_chipselect = 1; master->bus_num = pdev->id; master->setup = hspi_setup; @@ -309,8 +311,6 @@ static int hspi_probe(struct platform_device *pdev) goto error1; } - pm_runtime_enable(&pdev->dev); - return 0; error1: From 9886167d20c0720dcfb01e62cdff4d906b226f43 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 3 Oct 2013 16:02:23 +0200 Subject: [PATCH 26/45] perf: Fix perf_pmu_migrate_context While auditing the list_entry usage due to a trinity bug I found that perf_pmu_migrate_context violates the rules for perf_event::event_entry. The problem is that perf_event::event_entry is a RCU list element, and hence we must wait for a full RCU grace period before re-using the element after deletion. Therefore the usage in perf_pmu_migrate_context() which re-uses the entry immediately is broken. For now introduce another list_head into perf_event for this specific usage. This doesn't actually fix the trinity report because that never goes through this code. Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/n/tip-mkj72lxagw1z8fvjm648iznw@git.kernel.org Signed-off-by: Ingo Molnar --- include/linux/perf_event.h | 24 +++++++++++++++++++++++- kernel/events/core.c | 6 +++--- 2 files changed, 26 insertions(+), 4 deletions(-) diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 866e85c5eb94..c8ba627c1d60 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -294,9 +294,31 @@ struct ring_buffer; */ struct perf_event { #ifdef CONFIG_PERF_EVENTS - struct list_head group_entry; + /* + * entry onto perf_event_context::event_list; + * modifications require ctx->lock + * RCU safe iterations. + */ struct list_head event_entry; + + /* + * XXX: group_entry and sibling_list should be mutually exclusive; + * either you're a sibling on a group, or you're the group leader. + * Rework the code to always use the same list element. + * + * Locked for modification by both ctx->mutex and ctx->lock; holding + * either sufficies for read. + */ + struct list_head group_entry; struct list_head sibling_list; + + /* + * We need storage to track the entries in perf_pmu_migrate_context; we + * cannot use the event_entry because of RCU and we want to keep the + * group in tact which avoids us using the other two entries. + */ + struct list_head migrate_entry; + struct hlist_node hlist_entry; int nr_siblings; int group_flags; diff --git a/kernel/events/core.c b/kernel/events/core.c index cb4238e85b38..d49a9d29334c 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -7234,15 +7234,15 @@ void perf_pmu_migrate_context(struct pmu *pmu, int src_cpu, int dst_cpu) perf_remove_from_context(event); unaccount_event_cpu(event, src_cpu); put_ctx(src_ctx); - list_add(&event->event_entry, &events); + list_add(&event->migrate_entry, &events); } mutex_unlock(&src_ctx->mutex); synchronize_rcu(); mutex_lock(&dst_ctx->mutex); - list_for_each_entry_safe(event, tmp, &events, event_entry) { - list_del(&event->event_entry); + list_for_each_entry_safe(event, tmp, &events, migrate_entry) { + list_del(&event->migrate_entry); if (event->state >= PERF_EVENT_STATE_OFF) event->state = PERF_EVENT_STATE_INACTIVE; account_event_cpu(event, dst_cpu); From d8b11a0cbd1c66ce283eb9dabe0498dfa6483f32 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 3 Oct 2013 16:00:14 +0200 Subject: [PATCH 27/45] perf/x86: Clean up cap_user_time* setting Currently the cap_user_time_zero capability has different tests than cap_user_time; even though they expose the exact same data. Switch from CONSTANT && NONSTOP to sched_clock_stable to also deal with multi cabinet machines and drop the tsc_disabled() check.. non of this will work sanely without tsc anyway. Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/n/tip-nmgn0j0muo1r4c94vlfh23xy@git.kernel.org Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/perf_event.c | 11 +++-------- 1 file changed, 3 insertions(+), 8 deletions(-) diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 897783b3302a..9d8449158cf9 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -1888,10 +1888,7 @@ void arch_perf_update_userpage(struct perf_event_mmap_page *userpg, u64 now) userpg->cap_user_rdpmc = x86_pmu.attr_rdpmc; userpg->pmc_width = x86_pmu.cntval_bits; - if (!boot_cpu_has(X86_FEATURE_CONSTANT_TSC)) - return; - - if (!boot_cpu_has(X86_FEATURE_NONSTOP_TSC)) + if (!sched_clock_stable) return; userpg->cap_user_time = 1; @@ -1899,10 +1896,8 @@ void arch_perf_update_userpage(struct perf_event_mmap_page *userpg, u64 now) userpg->time_shift = CYC2NS_SCALE_FACTOR; userpg->time_offset = this_cpu_read(cyc2ns_offset) - now; - if (sched_clock_stable && !check_tsc_disabled()) { - userpg->cap_user_time_zero = 1; - userpg->time_zero = this_cpu_read(cyc2ns_offset); - } + userpg->cap_user_time_zero = 1; + userpg->time_zero = this_cpu_read(cyc2ns_offset); } /* From 6adb0b0ae26fcc35cfec068d71f13863faac5b44 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Sun, 22 Sep 2013 19:44:59 -0600 Subject: [PATCH 28/45] perf tools: Add default handler for mmap2 events Commands that do not implement an mmap2 handler should at least not die with a segfault when processing files with MMAP2 events. Signed-off-by: David Ahern Link: http://lkml.kernel.org/r/1379900700-5186-5-git-send-email-dsahern@gmail.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/session.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c index 70ffa41518f3..37c4718f6be1 100644 --- a/tools/perf/util/session.c +++ b/tools/perf/util/session.c @@ -256,6 +256,8 @@ void perf_tool__fill_defaults(struct perf_tool *tool) tool->sample = process_event_sample_stub; if (tool->mmap == NULL) tool->mmap = process_event_stub; + if (tool->mmap2 == NULL) + tool->mmap2 = process_event_stub; if (tool->comm == NULL) tool->comm = process_event_stub; if (tool->fork == NULL) From d20a47e70b57a77f4d1b0dba98a668e14c2f9bf9 Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Mon, 30 Sep 2013 18:01:11 +0900 Subject: [PATCH 29/45] perf stat: Set child_pid after perf_evlist__prepare_workload() The commit acf2892270dc ("perf stat: Use perf_evlist__prepare/ start_workload()") converted to use the function but forgot to update child_pid. Fix it. Signed-off-by: Namhyung Kim Cc: David Ahern Cc: Ingo Molnar Cc: Jiri Olsa Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1380531671-28076-1-git-send-email-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/builtin-stat.c | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index f686d5ff594e..5098f144b92d 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -457,6 +457,7 @@ static int __run_perf_stat(int argc, const char **argv) perror("failed to prepare workload"); return -1; } + child_pid = evsel_list->workload.pid; } if (group) From 47a92b828639dcb1a6033cd0a441099df828b7a9 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Wed, 2 Oct 2013 18:25:26 -0300 Subject: [PATCH 30/45] perf tools: Fix libaudit test MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In ubuntu systems the libaudit test was always failing due to the newline in the printf call not being escaped, which somehow didn't prevented the test from working as expected on other systems, such as fedora18. Fix it by removing the newline, as this is just a test, that program is just a compile test. The error messages, obtained using 'make V=1': CHK libaudit : In function ‘main’: :5:9: error: missing terminating " character [-Werror] :5:2: error: missing terminating " character :6:1: error: missing terminating " character [-Werror] :6:1: error: missing terminating " character :7:2: error: expected expression before ‘return’ :8:1: error: expected ‘;’ before ‘}’ token cc1: all warnings being treated as errors config/Makefile:241: No libaudit.h found, disables 'trace' tool, please install audit-libs-devel or libaudit-dev After this change the test works as expected in all systems tested and the 'trace' tool is built when the needed devel packages are installed. Cc: Adrian Hunter Cc: David Ahern Cc: Frederic Weisbecker Cc: Jiri Olsa Cc: Mike Galbraith Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/n/tip-0trw8qs9hafeopc0vj1sicay@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/config/feature-tests.mak | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/perf/config/feature-tests.mak b/tools/perf/config/feature-tests.mak index d5a8dd44945f..f79305739ecc 100644 --- a/tools/perf/config/feature-tests.mak +++ b/tools/perf/config/feature-tests.mak @@ -219,7 +219,7 @@ define SOURCE_LIBAUDIT int main(void) { - printf(\"error message: %s\n\", audit_errno_to_name(0)); + printf(\"error message: %s\", audit_errno_to_name(0)); return audit_open(); } endef From e08cfd4bda7683cdbe6971c26cf23e2afdb1e7a8 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Mon, 30 Sep 2013 18:21:44 +0900 Subject: [PATCH 31/45] perf probe: Fix to find line information for probe list Fix to find the correct (as much as possible) line information for listing probes. Without this fix, perf probe --list action will show incorrect line information as below; probe:getname_flags (on getname_flags@ksrc/linux-3/fs/namei.c) probe:getname_flags_1 (on getname:-89@x86/include/asm/current.h) probe:getname_flags_2 (on user_path_at_empty:-2054@x86/include/asm/current.h) The minus line number is obviously wrong, and current.h is not related to the probe point. Deeper investigation discovered that there were 2 issues related to this bug, and minor typos too. The 1st issue is the rack of considering about nested inlined functions, which causes the wrong (relative) line number. The 2nd issue is that the dwarf line info is not correct at those points. It points 14th line of current.h. Since it seems that the line info includes somewhat unreliable information, this fixes perf to try to find correct line information from both of debuginfo and line info as below. 1) Probe address is the entry of a function instance In this case, the line is set as the function declared line. 2) Probe address is the entry of an expanded inline function block In this case, the line is set as the function call-site line. This means that the line number is relative from the entry line of caller function (which can be an inlined function if nested) 3) Probe address is inside a function instance or an expanded inline function block In this case, perf probe queries the line number from lineinfo and verify the function declared file is same as the file name queried from lineinfo. If the file name is different, it is a failure case. The probe address is shown as symbol+offset. 4) Probe address is not in the any function instance This is a failure case, the probe address is shown as symbol+offset. With this fix, perf probe -l shows correct probe lines as below; probe:getname_flags (on getname_flags@ksrc/linux-3/fs/namei.c) probe:getname_flags_1 (on getname:2@ksrc/linux-3/fs/namei.c) probe:getname_flags_2 (on user_path_at_empty:4@ksrc/linux-3/fs/namei.c) Changes at v2: - Fix typos in the function comments. (Thanks to Namhyung Kim) - Use die_find_top_inlinefunc instead of die_find_inlinefunc_next. Signed-off-by: Masami Hiramatsu Cc: Ingo Molnar Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20130930092144.1693.11058.stgit@udc4-manage.rcp.hitachi.co.jp Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/dwarf-aux.c | 27 +++++++++++++++---- tools/perf/util/dwarf-aux.h | 6 ++++- tools/perf/util/probe-finder.c | 49 +++++++++++++++++++++++----------- 3 files changed, 60 insertions(+), 22 deletions(-) diff --git a/tools/perf/util/dwarf-aux.c b/tools/perf/util/dwarf-aux.c index e23bde19d590..7defd77105d0 100644 --- a/tools/perf/util/dwarf-aux.c +++ b/tools/perf/util/dwarf-aux.c @@ -426,7 +426,7 @@ static int __die_search_func_cb(Dwarf_Die *fn_die, void *data) * @die_mem: a buffer for result DIE * * Search a non-inlined function DIE which includes @addr. Stores the - * DIE to @die_mem and returns it if found. Returns NULl if failed. + * DIE to @die_mem and returns it if found. Returns NULL if failed. */ Dwarf_Die *die_find_realfunc(Dwarf_Die *cu_die, Dwarf_Addr addr, Dwarf_Die *die_mem) @@ -454,15 +454,32 @@ static int __die_find_inline_cb(Dwarf_Die *die_mem, void *data) } /** - * die_find_inlinefunc - Search an inlined function at given address - * @cu_die: a CU DIE which including @addr + * die_find_top_inlinefunc - Search the top inlined function at given address + * @sp_die: a subprogram DIE which including @addr * @addr: target address * @die_mem: a buffer for result DIE * * Search an inlined function DIE which includes @addr. Stores the - * DIE to @die_mem and returns it if found. Returns NULl if failed. + * DIE to @die_mem and returns it if found. Returns NULL if failed. + * Even if several inlined functions are expanded recursively, this + * doesn't trace it down, and returns the topmost one. + */ +Dwarf_Die *die_find_top_inlinefunc(Dwarf_Die *sp_die, Dwarf_Addr addr, + Dwarf_Die *die_mem) +{ + return die_find_child(sp_die, __die_find_inline_cb, &addr, die_mem); +} + +/** + * die_find_inlinefunc - Search an inlined function at given address + * @sp_die: a subprogram DIE which including @addr + * @addr: target address + * @die_mem: a buffer for result DIE + * + * Search an inlined function DIE which includes @addr. Stores the + * DIE to @die_mem and returns it if found. Returns NULL if failed. * If several inlined functions are expanded recursively, this trace - * it and returns deepest one. + * it down and returns deepest one. */ Dwarf_Die *die_find_inlinefunc(Dwarf_Die *sp_die, Dwarf_Addr addr, Dwarf_Die *die_mem) diff --git a/tools/perf/util/dwarf-aux.h b/tools/perf/util/dwarf-aux.h index 8658d41697d2..b4fe90c6cb2d 100644 --- a/tools/perf/util/dwarf-aux.h +++ b/tools/perf/util/dwarf-aux.h @@ -79,7 +79,11 @@ extern Dwarf_Die *die_find_child(Dwarf_Die *rt_die, extern Dwarf_Die *die_find_realfunc(Dwarf_Die *cu_die, Dwarf_Addr addr, Dwarf_Die *die_mem); -/* Search an inlined function including given address */ +/* Search the top inlined function including given address */ +extern Dwarf_Die *die_find_top_inlinefunc(Dwarf_Die *sp_die, Dwarf_Addr addr, + Dwarf_Die *die_mem); + +/* Search the deepest inlined function including given address */ extern Dwarf_Die *die_find_inlinefunc(Dwarf_Die *sp_die, Dwarf_Addr addr, Dwarf_Die *die_mem); diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c index 371476cb8ddc..c09e0a9fdf4c 100644 --- a/tools/perf/util/probe-finder.c +++ b/tools/perf/util/probe-finder.c @@ -1327,8 +1327,8 @@ int debuginfo__find_probe_point(struct debuginfo *self, unsigned long addr, struct perf_probe_point *ppt) { Dwarf_Die cudie, spdie, indie; - Dwarf_Addr _addr, baseaddr; - const char *fname = NULL, *func = NULL, *tmp; + Dwarf_Addr _addr = 0, baseaddr = 0; + const char *fname = NULL, *func = NULL, *basefunc = NULL, *tmp; int baseline = 0, lineno = 0, ret = 0; /* Adjust address with bias */ @@ -1349,27 +1349,36 @@ int debuginfo__find_probe_point(struct debuginfo *self, unsigned long addr, /* Find a corresponding function (name, baseline and baseaddr) */ if (die_find_realfunc(&cudie, (Dwarf_Addr)addr, &spdie)) { /* Get function entry information */ - tmp = dwarf_diename(&spdie); - if (!tmp || + func = basefunc = dwarf_diename(&spdie); + if (!func || dwarf_entrypc(&spdie, &baseaddr) != 0 || - dwarf_decl_line(&spdie, &baseline) != 0) + dwarf_decl_line(&spdie, &baseline) != 0) { + lineno = 0; goto post; - func = tmp; + } - if (addr == (unsigned long)baseaddr) + if (addr == (unsigned long)baseaddr) { /* Function entry - Relative line number is 0 */ lineno = baseline; - else if (die_find_inlinefunc(&spdie, (Dwarf_Addr)addr, - &indie)) { + fname = dwarf_decl_file(&spdie); + goto post; + } + + /* Track down the inline functions step by step */ + while (die_find_top_inlinefunc(&spdie, (Dwarf_Addr)addr, + &indie)) { + /* There is an inline function */ if (dwarf_entrypc(&indie, &_addr) == 0 && - _addr == addr) + _addr == addr) { /* * addr is at an inline function entry. * In this case, lineno should be the call-site - * line number. + * line number. (overwrite lineinfo) */ lineno = die_get_call_lineno(&indie); - else { + fname = die_get_call_file(&indie); + break; + } else { /* * addr is in an inline function body. * Since lineno points one of the lines @@ -1377,19 +1386,27 @@ int debuginfo__find_probe_point(struct debuginfo *self, unsigned long addr, * be the entry line of the inline function. */ tmp = dwarf_diename(&indie); - if (tmp && - dwarf_decl_line(&spdie, &baseline) == 0) - func = tmp; + if (!tmp || + dwarf_decl_line(&indie, &baseline) != 0) + break; + func = tmp; + spdie = indie; } } + /* Verify the lineno and baseline are in a same file */ + tmp = dwarf_decl_file(&spdie); + if (!tmp || strcmp(tmp, fname) != 0) + lineno = 0; } post: /* Make a relative line number or an offset */ if (lineno) ppt->line = lineno - baseline; - else if (func) + else if (basefunc) { ppt->offset = addr - (unsigned long)baseaddr; + func = basefunc; + } /* Duplicate strings */ if (func) { From 027a7e86a17fb2d172f4eb17cd6b0e9f139224fd Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Fri, 4 Oct 2013 10:24:17 -0300 Subject: [PATCH 32/45] perf tools: Fix installation of libexec components Doing a fresh install on a user home directory needs to first make sure that the ~/libexec/perf-core/ directory is present so that 'perf-archive' like scripts, 'perf test' attr config files and 'perf script' scripts can be installed. Cc: Adrian Hunter Cc: David Ahern Cc: Frederic Weisbecker Cc: Jiri Olsa Cc: Mike Galbraith Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/n/tip-z7ryi3r1b9dn9smbfnab0fdc@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/Makefile | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/perf/Makefile b/tools/perf/Makefile index 3a0ff7fb71b6..64c043b7a438 100644 --- a/tools/perf/Makefile +++ b/tools/perf/Makefile @@ -770,6 +770,7 @@ check: $(OUTPUT)common-cmds.h install-bin: all $(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(bindir_SQ)' $(INSTALL) $(OUTPUT)perf '$(DESTDIR_SQ)$(bindir_SQ)' + $(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)' $(INSTALL) $(OUTPUT)perf-archive -t '$(DESTDIR_SQ)$(perfexec_instdir_SQ)' ifndef NO_LIBPERL $(INSTALL) -d -m 755 '$(DESTDIR_SQ)$(perfexec_instdir_SQ)/scripts/perl/Perf-Trace-Util/lib/Perf/Trace' From b314e5cfd11fd78545ce6c2be42646254390c1aa Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Mon, 30 Sep 2013 17:19:48 +0900 Subject: [PATCH 33/45] perf session: Fix infinite loop on invalid perf.data file perf-record updates the header in the perf.data file at termination. Without this update perf-report (and other processing built-ins) it caused an infinite loop when perf report (or something like) called. This is because the algorithm in __perf_session__process_events() depends on the data_size which is read from file header. Use file size directly instead in this case to do the best-effort processing. Signed-off-by: Namhyung Kim Tested-by: David Ahern Tested-by: Sonny Rao Acked-by: Ingo Molnar Cc: David Ahern Cc: Ingo Molnar Cc: Jiri Olsa Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Sonny Rao Link: http://lkml.kernel.org/r/1380529188-27193-1-git-send-email-namhyung@kernel.org Signed-off-by: David Ahern [ Reworded warning as per Ingo Molnar suggestion, replaces 'perf.data' with session->filename, to precisely identify the data file involved ] Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/header.c | 12 ++++++++++++ tools/perf/util/session.c | 2 +- 2 files changed, 13 insertions(+), 1 deletion(-) diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c index ce69901176d8..c3e5a3b817ab 100644 --- a/tools/perf/util/header.c +++ b/tools/perf/util/header.c @@ -2768,6 +2768,18 @@ int perf_session__read_header(struct perf_session *session) if (perf_file_header__read(&f_header, header, fd) < 0) return -EINVAL; + /* + * Sanity check that perf.data was written cleanly; data size is + * initialized to 0 and updated only if the on_exit function is run. + * If data size is still 0 then the file contains only partial + * information. Just warn user and process it as much as it can. + */ + if (f_header.data.size == 0) { + pr_warning("WARNING: The %s file's data size field is 0 which is unexpected.\n" + "Was the 'perf record' command properly terminated?\n", + session->filename); + } + nr_attrs = f_header.attrs.size / f_header.attr_size; lseek(fd, f_header.attrs.offset, SEEK_SET); diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c index 37c4718f6be1..568b750c01f6 100644 --- a/tools/perf/util/session.c +++ b/tools/perf/util/session.c @@ -1312,7 +1312,7 @@ int __perf_session__process_events(struct perf_session *session, file_offset = page_offset; head = data_offset - page_offset; - if (data_offset + data_size < file_size) + if (data_size && (data_offset + data_size < file_size)) file_size = data_offset + data_size; progress_next = file_size / 16; From de8cc234a517b61c9a8450a34408dbae53aa171f Mon Sep 17 00:00:00 2001 From: Sachin Kamat Date: Tue, 10 Sep 2013 17:06:26 +0530 Subject: [PATCH 34/45] spi: atmel: Fix incorrect error path 'irq' was not released when clk_prepare_enable failed. Fix it. Signed-off-by: Sachin Kamat Acked-by: Nicolas Ferre Signed-off-by: Mark Brown --- drivers/spi/spi-atmel.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/spi/spi-atmel.c b/drivers/spi/spi-atmel.c index fd7cc566095a..d4ac60b4a56e 100644 --- a/drivers/spi/spi-atmel.c +++ b/drivers/spi/spi-atmel.c @@ -1583,7 +1583,7 @@ static int atmel_spi_probe(struct platform_device *pdev) /* Initialize the hardware */ ret = clk_prepare_enable(clk); if (ret) - goto out_unmap_regs; + goto out_free_irq; spi_writel(as, CR, SPI_BIT(SWRST)); spi_writel(as, CR, SPI_BIT(SWRST)); /* AT91SAM9263 Rev B workaround */ if (as->caps.has_wdrbt) { @@ -1614,6 +1614,7 @@ static int atmel_spi_probe(struct platform_device *pdev) spi_writel(as, CR, SPI_BIT(SWRST)); spi_writel(as, CR, SPI_BIT(SWRST)); /* AT91SAM9263 Rev B workaround */ clk_disable_unprepare(clk); +out_free_irq: free_irq(irq, master); out_unmap_regs: iounmap(as->regs); From f50f9aabf32db7414551ffdfdccc71be5f3d6e7d Mon Sep 17 00:00:00 2001 From: David Herrmann Date: Wed, 2 Oct 2013 13:47:28 +0200 Subject: [PATCH 35/45] HID: wiimote: fix FF deadlock The input core has an internal spinlock that is acquired during event injection via input_event() and friends but also held during FF callbacks. That means, there is no way to share a lock between event-injection and FF handling. Unfortunately, this is what is required for wiimote state tracking and what we do with state.lock and input->lock. This deadlock can be triggered when using continuous data reporting and FF on a wiimote device at the same time. I takes me at least 30m of stress-testing to trigger it but users reported considerably shorter times (http://bpaste.net/show/132504/) when using some gaming-console emulators. The real problem is that we have two copies of internal state, one in the wiimote objects and the other in the input device. As the input-lock is not supposed to be accessed from outside of input-core, we have no other chance than offloading FF handling into a worker. This actually works pretty nice and also allows to implictly merge fast rumble changes into a single request. Due to the 3-layered workers (rumble+queue+l2cap) this might reduce FF responsiveness. Initial tests were fine so lets fix the race first and if it turns out to be too slow we can always handle FF out-of-band and skip the queue-worker. Cc: # 3.11+ Reported-by: Thomas Schneider Signed-off-by: David Herrmann Signed-off-by: Jiri Kosina --- drivers/hid/hid-wiimote-modules.c | 40 ++++++++++++++++++++++--------- drivers/hid/hid-wiimote.h | 4 +++- 2 files changed, 32 insertions(+), 12 deletions(-) diff --git a/drivers/hid/hid-wiimote-modules.c b/drivers/hid/hid-wiimote-modules.c index 2e7d644dba18..71adf9e60b13 100644 --- a/drivers/hid/hid-wiimote-modules.c +++ b/drivers/hid/hid-wiimote-modules.c @@ -119,12 +119,22 @@ static const struct wiimod_ops wiimod_keys = { * the rumble motor, this flag shouldn't be set. */ +/* used by wiimod_rumble and wiipro_rumble */ +static void wiimod_rumble_worker(struct work_struct *work) +{ + struct wiimote_data *wdata = container_of(work, struct wiimote_data, + rumble_worker); + + spin_lock_irq(&wdata->state.lock); + wiiproto_req_rumble(wdata, wdata->state.cache_rumble); + spin_unlock_irq(&wdata->state.lock); +} + static int wiimod_rumble_play(struct input_dev *dev, void *data, struct ff_effect *eff) { struct wiimote_data *wdata = input_get_drvdata(dev); __u8 value; - unsigned long flags; /* * The wiimote supports only a single rumble motor so if any magnitude @@ -137,9 +147,10 @@ static int wiimod_rumble_play(struct input_dev *dev, void *data, else value = 0; - spin_lock_irqsave(&wdata->state.lock, flags); - wiiproto_req_rumble(wdata, value); - spin_unlock_irqrestore(&wdata->state.lock, flags); + /* Locking state.lock here might deadlock with input_event() calls. + * schedule_work acts as barrier. Merging multiple changes is fine. */ + wdata->state.cache_rumble = value; + schedule_work(&wdata->rumble_worker); return 0; } @@ -147,6 +158,8 @@ static int wiimod_rumble_play(struct input_dev *dev, void *data, static int wiimod_rumble_probe(const struct wiimod_ops *ops, struct wiimote_data *wdata) { + INIT_WORK(&wdata->rumble_worker, wiimod_rumble_worker); + set_bit(FF_RUMBLE, wdata->input->ffbit); if (input_ff_create_memless(wdata->input, NULL, wiimod_rumble_play)) return -ENOMEM; @@ -159,6 +172,8 @@ static void wiimod_rumble_remove(const struct wiimod_ops *ops, { unsigned long flags; + cancel_work_sync(&wdata->rumble_worker); + spin_lock_irqsave(&wdata->state.lock, flags); wiiproto_req_rumble(wdata, 0); spin_unlock_irqrestore(&wdata->state.lock, flags); @@ -1731,7 +1746,6 @@ static int wiimod_pro_play(struct input_dev *dev, void *data, { struct wiimote_data *wdata = input_get_drvdata(dev); __u8 value; - unsigned long flags; /* * The wiimote supports only a single rumble motor so if any magnitude @@ -1744,9 +1758,10 @@ static int wiimod_pro_play(struct input_dev *dev, void *data, else value = 0; - spin_lock_irqsave(&wdata->state.lock, flags); - wiiproto_req_rumble(wdata, value); - spin_unlock_irqrestore(&wdata->state.lock, flags); + /* Locking state.lock here might deadlock with input_event() calls. + * schedule_work acts as barrier. Merging multiple changes is fine. */ + wdata->state.cache_rumble = value; + schedule_work(&wdata->rumble_worker); return 0; } @@ -1756,6 +1771,8 @@ static int wiimod_pro_probe(const struct wiimod_ops *ops, { int ret, i; + INIT_WORK(&wdata->rumble_worker, wiimod_rumble_worker); + wdata->extension.input = input_allocate_device(); if (!wdata->extension.input) return -ENOMEM; @@ -1817,12 +1834,13 @@ static void wiimod_pro_remove(const struct wiimod_ops *ops, if (!wdata->extension.input) return; + input_unregister_device(wdata->extension.input); + wdata->extension.input = NULL; + cancel_work_sync(&wdata->rumble_worker); + spin_lock_irqsave(&wdata->state.lock, flags); wiiproto_req_rumble(wdata, 0); spin_unlock_irqrestore(&wdata->state.lock, flags); - - input_unregister_device(wdata->extension.input); - wdata->extension.input = NULL; } static const struct wiimod_ops wiimod_pro = { diff --git a/drivers/hid/hid-wiimote.h b/drivers/hid/hid-wiimote.h index f1474f372c0b..75db0c400037 100644 --- a/drivers/hid/hid-wiimote.h +++ b/drivers/hid/hid-wiimote.h @@ -133,13 +133,15 @@ struct wiimote_state { __u8 *cmd_read_buf; __u8 cmd_read_size; - /* calibration data */ + /* calibration/cache data */ __u16 calib_bboard[4][3]; + __u8 cache_rumble; }; struct wiimote_data { struct hid_device *hdev; struct input_dev *input; + struct work_struct rumble_worker; struct led_classdev *leds[4]; struct input_dev *accel; struct input_dev *ir; From 8b5ede69d24db939f52b47effff2f6fe1e83e08b Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Tue, 8 Oct 2013 08:08:24 +1100 Subject: [PATCH 36/45] powerpc/irq: Don't switch to irq stack from softirq stack MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit irq_exit() is now called on the irq stack, which can trigger a switch to the softirq stack from the irq stack. If an interrupt happens at that point, we will not properly detect the re-entrancy and clobber the original return context on the irq stack. This fixes it. The side effect is to prevent all nesting from softirq stack to irq stack even in the "safe" case but it's simpler that way and matches what x86_64 does. Reported-by: Cédric Le Goater Tested-by: Cédric Le Goater Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Linus Torvalds --- arch/powerpc/kernel/irq.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c index 57d286a78f86..c7cb8c232d2f 100644 --- a/arch/powerpc/kernel/irq.c +++ b/arch/powerpc/kernel/irq.c @@ -495,14 +495,15 @@ void __do_irq(struct pt_regs *regs) void do_IRQ(struct pt_regs *regs) { struct pt_regs *old_regs = set_irq_regs(regs); - struct thread_info *curtp, *irqtp; + struct thread_info *curtp, *irqtp, *sirqtp; /* Switch to the irq stack to handle this */ curtp = current_thread_info(); irqtp = hardirq_ctx[raw_smp_processor_id()]; + sirqtp = softirq_ctx[raw_smp_processor_id()]; /* Already there ? */ - if (unlikely(curtp == irqtp)) { + if (unlikely(curtp == irqtp || curtp == sirqtp)) { __do_irq(regs); set_irq_regs(old_regs); return; From 25f2bd7f5add608c1d1405938f39c96927b275ca Mon Sep 17 00:00:00 2001 From: Henrik Rydberg Date: Wed, 2 Oct 2013 19:15:03 +0200 Subject: [PATCH 37/45] hwmon: (applesmc) Always read until end of data The crash reported and investigated in commit 5f4513 turned out to be caused by a change to the read interface on newer (2012) SMCs. Tests by Chris show that simply reading the data valid line is enough for the problem to go away. Additional tests show that the newer SMCs no longer wait for the number of requested bytes, but start sending data right away. Apparently the number of bytes to read is no longer specified as before, but instead found out by reading until end of data. Failure to read until end of data confuses the state machine, which eventually causes the crash. As a remedy, assuming bit0 is the read valid line, make sure there is nothing more to read before leaving the read function. Tested to resolve the original problem, and runtested on MBA3,1, MBP4,1, MBP8,2, MBP10,1, MBP10,2. The patch seems to have no effect on machines before 2012. Tested-by: Chris Murphy Cc: stable@vger.kernel.org Signed-off-by: Henrik Rydberg Signed-off-by: Guenter Roeck --- drivers/hwmon/applesmc.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/drivers/hwmon/applesmc.c b/drivers/hwmon/applesmc.c index 98814d12a604..3288f13d2d87 100644 --- a/drivers/hwmon/applesmc.c +++ b/drivers/hwmon/applesmc.c @@ -230,6 +230,7 @@ static int send_argument(const char *key) static int read_smc(u8 cmd, const char *key, u8 *buffer, u8 len) { + u8 status, data = 0; int i; if (send_command(cmd) || send_argument(key)) { @@ -237,6 +238,7 @@ static int read_smc(u8 cmd, const char *key, u8 *buffer, u8 len) return -EIO; } + /* This has no effect on newer (2012) SMCs */ if (send_byte(len, APPLESMC_DATA_PORT)) { pr_warn("%.4s: read len fail\n", key); return -EIO; @@ -250,6 +252,17 @@ static int read_smc(u8 cmd, const char *key, u8 *buffer, u8 len) buffer[i] = inb(APPLESMC_DATA_PORT); } + /* Read the data port until bit0 is cleared */ + for (i = 0; i < 16; i++) { + udelay(APPLESMC_MIN_WAIT); + status = inb(APPLESMC_CMD_PORT); + if (!(status & 0x01)) + break; + data = inb(APPLESMC_DATA_PORT); + } + if (i) + pr_warn("flushed %d bytes, last value is: %d\n", i, data); + return 0; } From d0d538b9d1239ed6ed318a2304ba5a763f835089 Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Wed, 9 Oct 2013 19:13:19 +0300 Subject: [PATCH 38/45] KVM: nVMX: fix shadow on EPT 72f857950f6f19 broke shadow on EPT. This patch reverts it and fixes PAE on nEPT (which reverted commit fixed) in other way. Shadow on EPT is now broken because while L1 builds shadow page table for L2 (which is PAE while L2 is in real mode) it never loads L2's GUEST_PDPTR[0-3]. They do not need to be loaded because without nested virtualization HW does this during guest entry if EPT is disabled, but in our case L0 emulates L2's vmentry while EPT is enables, so we cannot rely on vmcs12->guest_pdptr[0-3] to contain up-to-date values and need to re-read PDPTEs from L2 memory. This is what kvm_set_cr3() is doing, but by clearing cache bits during L2 vmentry we drop values that kvm_set_cr3() read from memory. So why the same code does not work for PAE on nEPT? kvm_set_cr3() reads pdptes into vcpu->arch.walk_mmu->pdptrs[]. walk_mmu points to vcpu->arch.nested_mmu while nested guest is running, but ept_load_pdptrs() uses vcpu->arch.mmu which contain incorrect values. Fix that by using walk_mmu in ept_(load|save)_pdptrs. Signed-off-by: Gleb Natapov Reviewed-by: Paolo Bonzini Tested-by: Paolo Bonzini Signed-off-by: Paolo Bonzini --- arch/x86/kvm/vmx.c | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 3b8e7459dd4d..2b2fce1b2009 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -3255,25 +3255,29 @@ static void vmx_decache_cr4_guest_bits(struct kvm_vcpu *vcpu) static void ept_load_pdptrs(struct kvm_vcpu *vcpu) { + struct kvm_mmu *mmu = vcpu->arch.walk_mmu; + if (!test_bit(VCPU_EXREG_PDPTR, (unsigned long *)&vcpu->arch.regs_dirty)) return; if (is_paging(vcpu) && is_pae(vcpu) && !is_long_mode(vcpu)) { - vmcs_write64(GUEST_PDPTR0, vcpu->arch.mmu.pdptrs[0]); - vmcs_write64(GUEST_PDPTR1, vcpu->arch.mmu.pdptrs[1]); - vmcs_write64(GUEST_PDPTR2, vcpu->arch.mmu.pdptrs[2]); - vmcs_write64(GUEST_PDPTR3, vcpu->arch.mmu.pdptrs[3]); + vmcs_write64(GUEST_PDPTR0, mmu->pdptrs[0]); + vmcs_write64(GUEST_PDPTR1, mmu->pdptrs[1]); + vmcs_write64(GUEST_PDPTR2, mmu->pdptrs[2]); + vmcs_write64(GUEST_PDPTR3, mmu->pdptrs[3]); } } static void ept_save_pdptrs(struct kvm_vcpu *vcpu) { + struct kvm_mmu *mmu = vcpu->arch.walk_mmu; + if (is_paging(vcpu) && is_pae(vcpu) && !is_long_mode(vcpu)) { - vcpu->arch.mmu.pdptrs[0] = vmcs_read64(GUEST_PDPTR0); - vcpu->arch.mmu.pdptrs[1] = vmcs_read64(GUEST_PDPTR1); - vcpu->arch.mmu.pdptrs[2] = vmcs_read64(GUEST_PDPTR2); - vcpu->arch.mmu.pdptrs[3] = vmcs_read64(GUEST_PDPTR3); + mmu->pdptrs[0] = vmcs_read64(GUEST_PDPTR0); + mmu->pdptrs[1] = vmcs_read64(GUEST_PDPTR1); + mmu->pdptrs[2] = vmcs_read64(GUEST_PDPTR2); + mmu->pdptrs[3] = vmcs_read64(GUEST_PDPTR3); } __set_bit(VCPU_EXREG_PDPTR, @@ -7777,10 +7781,6 @@ static void prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12) vmcs_write64(GUEST_PDPTR1, vmcs12->guest_pdptr1); vmcs_write64(GUEST_PDPTR2, vmcs12->guest_pdptr2); vmcs_write64(GUEST_PDPTR3, vmcs12->guest_pdptr3); - __clear_bit(VCPU_EXREG_PDPTR, - (unsigned long *)&vcpu->arch.regs_avail); - __clear_bit(VCPU_EXREG_PDPTR, - (unsigned long *)&vcpu->arch.regs_dirty); } kvm_register_write(vcpu, VCPU_REGS_RSP, vmcs12->guest_rsp); From cfc860253abd73e1681696c08ea268d33285a2c4 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Sat, 21 Sep 2013 09:53:28 +1000 Subject: [PATCH 39/45] KVM: PPC: Book3S HV: Fix typo in saving DSCR This fixes a typo in the code that saves the guest DSCR (Data Stream Control Register) into the kvm_vcpu_arch struct on guest exit. The effect of the typo was that the DSCR value was saved in the wrong place, so changes to the DSCR by the guest didn't persist across guest exit and entry, and some host kernel memory got corrupted. Cc: stable@vger.kernel.org [v3.1+] Signed-off-by: Paul Mackerras Acked-by: Alexander Graf Signed-off-by: Paolo Bonzini --- arch/powerpc/kvm/book3s_hv_rmhandlers.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S index 294b7af28cdd..c71103b8a748 100644 --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S @@ -1066,7 +1066,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_ARCH_206) BEGIN_FTR_SECTION mfspr r8, SPRN_DSCR ld r7, HSTATE_DSCR(r13) - std r8, VCPU_DSCR(r7) + std r8, VCPU_DSCR(r9) mtspr SPRN_DSCR, r7 END_FTR_SECTION_IFSET(CPU_FTR_ARCH_206) From 40fde70d0df993d1a652d6cc69f8b4e967656170 Mon Sep 17 00:00:00 2001 From: Bharat Bhushan Date: Wed, 7 Aug 2013 15:33:46 +0530 Subject: [PATCH 40/45] kvm: ppc: booke: check range page invalidation progress on page setup When the MM code is invalidating a range of pages, it calls the KVM kvm_mmu_notifier_invalidate_range_start() notifier function, which calls kvm_unmap_hva_range(), which arranges to flush all the TLBs for guest pages. However, the Linux PTEs for the range being flushed are still valid at that point. We are not supposed to establish any new references to pages in the range until the ...range_end() notifier gets called. The PPC-specific KVM code doesn't get any explicit notification of that; instead, we are supposed to use mmu_notifier_retry() to test whether we are or have been inside a range flush notifier pair while we have been referencing a page. This patch calls the mmu_notifier_retry() while mapping the guest page to ensure we are not referencing a page when in range invalidation. This call is inside a region locked with kvm->mmu_lock, which is the same lock that is called by the KVM MMU notifier functions, thus ensuring that no new notification can proceed while we are in the locked region. Signed-off-by: Bharat Bhushan Acked-by: Alexander Graf [Backported to 3.12 - Paolo] Reviewed-by: Bharat Bhushan Signed-off-by: Paolo Bonzini --- arch/powerpc/kvm/e500_mmu_host.c | 18 +++++++++++++++++- 1 file changed, 17 insertions(+), 1 deletion(-) diff --git a/arch/powerpc/kvm/e500_mmu_host.c b/arch/powerpc/kvm/e500_mmu_host.c index 1c6a9d729df4..c65593abae8e 100644 --- a/arch/powerpc/kvm/e500_mmu_host.c +++ b/arch/powerpc/kvm/e500_mmu_host.c @@ -332,6 +332,13 @@ static inline int kvmppc_e500_shadow_map(struct kvmppc_vcpu_e500 *vcpu_e500, unsigned long hva; int pfnmap = 0; int tsize = BOOK3E_PAGESZ_4K; + int ret = 0; + unsigned long mmu_seq; + struct kvm *kvm = vcpu_e500->vcpu.kvm; + + /* used to check for invalidations in progress */ + mmu_seq = kvm->mmu_notifier_seq; + smp_rmb(); /* * Translate guest physical to true physical, acquiring @@ -449,6 +456,12 @@ static inline int kvmppc_e500_shadow_map(struct kvmppc_vcpu_e500 *vcpu_e500, gvaddr &= ~((tsize_pages << PAGE_SHIFT) - 1); } + spin_lock(&kvm->mmu_lock); + if (mmu_notifier_retry(kvm, mmu_seq)) { + ret = -EAGAIN; + goto out; + } + kvmppc_e500_ref_setup(ref, gtlbe, pfn); kvmppc_e500_setup_stlbe(&vcpu_e500->vcpu, gtlbe, tsize, @@ -457,10 +470,13 @@ static inline int kvmppc_e500_shadow_map(struct kvmppc_vcpu_e500 *vcpu_e500, /* Clear i-cache for new pages */ kvmppc_mmu_flush_icache(pfn); +out: + spin_unlock(&kvm->mmu_lock); + /* Drop refcount on page, so that mmu notifiers can clear it */ kvm_release_pfn_clean(pfn); - return 0; + return ret; } /* XXX only map the one-one case, for now use TLB0 */ From 61875f30daf60305712e25b209ef41ced2635bad Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Sat, 21 Sep 2013 13:58:22 -0400 Subject: [PATCH 41/45] random: allow architectures to optionally define random_get_entropy() Allow architectures which have a disabled get_cycles() function to provide a random_get_entropy() function which provides a fine-grained, rapidly changing counter that can be used by the /dev/random driver. For example, an architecture might have a rapidly changing register used to control random TLB cache eviction, or DRAM refresh that doesn't meet the requirements of get_cycles(), but which is good enough for the needs of the random driver. Signed-off-by: "Theodore Ts'o" Cc: stable@vger.kernel.org --- drivers/char/random.c | 8 ++++---- include/linux/timex.h | 14 ++++++++++++++ 2 files changed, 18 insertions(+), 4 deletions(-) diff --git a/drivers/char/random.c b/drivers/char/random.c index 92e6c67e1ae6..2d5daf9b58e9 100644 --- a/drivers/char/random.c +++ b/drivers/char/random.c @@ -643,7 +643,7 @@ struct timer_rand_state { */ void add_device_randomness(const void *buf, unsigned int size) { - unsigned long time = get_cycles() ^ jiffies; + unsigned long time = random_get_entropy() ^ jiffies; mix_pool_bytes(&input_pool, buf, size, NULL); mix_pool_bytes(&input_pool, &time, sizeof(time), NULL); @@ -680,7 +680,7 @@ static void add_timer_randomness(struct timer_rand_state *state, unsigned num) goto out; sample.jiffies = jiffies; - sample.cycles = get_cycles(); + sample.cycles = random_get_entropy(); sample.num = num; mix_pool_bytes(&input_pool, &sample, sizeof(sample), NULL); @@ -747,7 +747,7 @@ void add_interrupt_randomness(int irq, int irq_flags) struct fast_pool *fast_pool = &__get_cpu_var(irq_randomness); struct pt_regs *regs = get_irq_regs(); unsigned long now = jiffies; - __u32 input[4], cycles = get_cycles(); + __u32 input[4], cycles = random_get_entropy(); input[0] = cycles ^ jiffies; input[1] = irq; @@ -1485,7 +1485,7 @@ unsigned int get_random_int(void) hash = get_cpu_var(get_random_int_hash); - hash[0] += current->pid + jiffies + get_cycles(); + hash[0] += current->pid + jiffies + random_get_entropy(); md5_transform(hash, random_int_secret); ret = hash[0]; put_cpu_var(get_random_int_hash); diff --git a/include/linux/timex.h b/include/linux/timex.h index b3726e61368e..da4c32dbb2aa 100644 --- a/include/linux/timex.h +++ b/include/linux/timex.h @@ -64,6 +64,20 @@ #include +#ifndef random_get_entropy +/* + * The random_get_entropy() function is used by the /dev/random driver + * in order to extract entropy via the relative unpredictability of + * when an interrupt takes places versus a high speed, fine-grained + * timing source or cycle counter. Since it will be occurred on every + * single interrupt, it must have a very low cost/overhead. + * + * By default we use get_cycles() for this purpose, but individual + * architectures may override this in their asm/timex.h header file. + */ +#define random_get_entropy() get_cycles() +#endif + /* * SHIFT_PLL is used as a dampening factor to define how much we * adjust the frequency correction for a given offset in PLL mode. From 0821f20d49d19f49e53b05ef6727e010c4b1f11d Mon Sep 17 00:00:00 2001 From: "Mingarelli, Thomas" Date: Fri, 9 Aug 2013 16:31:09 +0000 Subject: [PATCH 42/45] watchdog: hpwdt: Patch to ignore auxilary iLO devices This patch is to prevent hpwdt from loading on any auxilary iLO devices defined after the initial (or main) iLO device. All auxilary iLO devices will have a subsystem device ID set to 0x1979 in order for hpwdt to differentiate between the two types. Signed-off-by: Thomas Mingarelli Tested-by: Lisa Mitchell Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/hpwdt.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/watchdog/hpwdt.c b/drivers/watchdog/hpwdt.c index 5be5e3d14f79..19f3c3fc65f4 100644 --- a/drivers/watchdog/hpwdt.c +++ b/drivers/watchdog/hpwdt.c @@ -802,6 +802,12 @@ static int hpwdt_init_one(struct pci_dev *dev, return -ENODEV; } + /* + * Ignore all auxilary iLO devices with the following PCI ID + */ + if (dev->subsystem_device == 0x1979) + return -ENODEV; + if (pci_enable_device(dev)) { dev_warn(&dev->dev, "Not possible to enable PCI Device: 0x%x:0x%x.\n", From 0ff18e3734c248fa516b4be942b1196832ee936c Mon Sep 17 00:00:00 2001 From: Fengguang Wu Date: Wed, 9 Oct 2013 09:26:21 +0800 Subject: [PATCH 43/45] kobject: show debug info on delayed kobject release Useful for locating buggy drivers on kernel oops. It may add dozens of new lines to boot dmesg. DEBUG_KOBJECT_RELEASE is hopefully only enabled in debug kernels (like maybe the Fedora rawhide one, or at developers), so being a bit more verbose is likely ok. Acked-by: Russell King Cc: Greg Kroah-Hartman Signed-off-by: Fengguang Wu Signed-off-by: Linus Torvalds --- lib/kobject.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/kobject.c b/lib/kobject.c index 669bf190d4fb..084f7b18d0c0 100644 --- a/lib/kobject.c +++ b/lib/kobject.c @@ -592,7 +592,7 @@ static void kobject_release(struct kref *kref) { struct kobject *kobj = container_of(kref, struct kobject, kref); #ifdef CONFIG_DEBUG_KOBJECT_RELEASE - pr_debug("kobject: '%s' (%p): %s, parent %p (delayed)\n", + pr_info("kobject: '%s' (%p): %s, parent %p (delayed)\n", kobject_name(kobj), kobj, __func__, kobj->parent); INIT_DELAYED_WORK(&kobj->release, kobject_delayed_cleanup); schedule_delayed_work(&kobj->release, HZ); From 2fe80d3bbf1c8bd9efc5b8154207c8dd104e7306 Mon Sep 17 00:00:00 2001 From: Kent Overstreet Date: Thu, 10 Oct 2013 17:31:15 -0700 Subject: [PATCH 44/45] bcache: Fix a null ptr deref regression Commit c0f04d88e46d ("bcache: Fix flushes in writeback mode") was fixing a reported data corruption bug, but it seems some last minute refactoring or rebasing introduced a null pointer deref. Signed-off-by: Kent Overstreet Cc: linux-stable # >= v3.10 Reported-by: Gabriel de Perthuis Signed-off-by: Linus Torvalds --- drivers/md/bcache/request.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/md/bcache/request.c b/drivers/md/bcache/request.c index 71eb233b9ace..b6a74bcbb08f 100644 --- a/drivers/md/bcache/request.c +++ b/drivers/md/bcache/request.c @@ -996,6 +996,7 @@ static void request_write(struct cached_dev *dc, struct search *s) closure_bio_submit(bio, cl, s->d); } else { bch_writeback_add(dc); + s->op.cache_bio = bio; if (bio->bi_rw & REQ_FLUSH) { /* Also need to send a flush to the backing device */ @@ -1008,8 +1009,6 @@ static void request_write(struct cached_dev *dc, struct search *s) flush->bi_private = cl; closure_bio_submit(flush, cl, s->d); - } else { - s->op.cache_bio = bio; } } out: From 3f0116c3238a96bc18ad4b4acefe4e7be32fa861 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 10 Oct 2013 10:16:30 +0200 Subject: [PATCH 45/45] compiler/gcc4: Add quirk for 'asm goto' miscompilation bug Fengguang Wu, Oleg Nesterov and Peter Zijlstra tracked down a kernel crash to a GCC bug: GCC miscompiles certain 'asm goto' constructs, as outlined here: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58670 Implement a workaround suggested by Jakub Jelinek. Reported-and-tested-by: Fengguang Wu Reported-by: Oleg Nesterov Reported-by: Peter Zijlstra Suggested-by: Jakub Jelinek Reviewed-by: Richard Henderson Cc: Linus Torvalds Cc: Andrew Morton Cc: Signed-off-by: Ingo Molnar --- arch/arm/include/asm/jump_label.h | 2 +- arch/mips/include/asm/jump_label.h | 2 +- arch/powerpc/include/asm/jump_label.h | 2 +- arch/s390/include/asm/jump_label.h | 2 +- arch/sparc/include/asm/jump_label.h | 2 +- arch/x86/include/asm/cpufeature.h | 6 +++--- arch/x86/include/asm/jump_label.h | 2 +- arch/x86/include/asm/mutex_64.h | 4 ++-- include/linux/compiler-gcc4.h | 15 +++++++++++++++ 9 files changed, 26 insertions(+), 11 deletions(-) diff --git a/arch/arm/include/asm/jump_label.h b/arch/arm/include/asm/jump_label.h index bfc198c75913..863c892b4aaa 100644 --- a/arch/arm/include/asm/jump_label.h +++ b/arch/arm/include/asm/jump_label.h @@ -16,7 +16,7 @@ static __always_inline bool arch_static_branch(struct static_key *key) { - asm goto("1:\n\t" + asm_volatile_goto("1:\n\t" JUMP_LABEL_NOP "\n\t" ".pushsection __jump_table, \"aw\"\n\t" ".word 1b, %l[l_yes], %c0\n\t" diff --git a/arch/mips/include/asm/jump_label.h b/arch/mips/include/asm/jump_label.h index 4d6d77ed9b9d..e194f957ca8c 100644 --- a/arch/mips/include/asm/jump_label.h +++ b/arch/mips/include/asm/jump_label.h @@ -22,7 +22,7 @@ static __always_inline bool arch_static_branch(struct static_key *key) { - asm goto("1:\tnop\n\t" + asm_volatile_goto("1:\tnop\n\t" "nop\n\t" ".pushsection __jump_table, \"aw\"\n\t" WORD_INSN " 1b, %l[l_yes], %0\n\t" diff --git a/arch/powerpc/include/asm/jump_label.h b/arch/powerpc/include/asm/jump_label.h index ae098c438f00..f016bb699b5f 100644 --- a/arch/powerpc/include/asm/jump_label.h +++ b/arch/powerpc/include/asm/jump_label.h @@ -19,7 +19,7 @@ static __always_inline bool arch_static_branch(struct static_key *key) { - asm goto("1:\n\t" + asm_volatile_goto("1:\n\t" "nop\n\t" ".pushsection __jump_table, \"aw\"\n\t" JUMP_ENTRY_TYPE "1b, %l[l_yes], %c0\n\t" diff --git a/arch/s390/include/asm/jump_label.h b/arch/s390/include/asm/jump_label.h index 6c32190dc73e..346b1c85ffb4 100644 --- a/arch/s390/include/asm/jump_label.h +++ b/arch/s390/include/asm/jump_label.h @@ -15,7 +15,7 @@ static __always_inline bool arch_static_branch(struct static_key *key) { - asm goto("0: brcl 0,0\n" + asm_volatile_goto("0: brcl 0,0\n" ".pushsection __jump_table, \"aw\"\n" ASM_ALIGN "\n" ASM_PTR " 0b, %l[label], %0\n" diff --git a/arch/sparc/include/asm/jump_label.h b/arch/sparc/include/asm/jump_label.h index 5080d16a832f..ec2e2e2aba7d 100644 --- a/arch/sparc/include/asm/jump_label.h +++ b/arch/sparc/include/asm/jump_label.h @@ -9,7 +9,7 @@ static __always_inline bool arch_static_branch(struct static_key *key) { - asm goto("1:\n\t" + asm_volatile_goto("1:\n\t" "nop\n\t" "nop\n\t" ".pushsection __jump_table, \"aw\"\n\t" diff --git a/arch/x86/include/asm/cpufeature.h b/arch/x86/include/asm/cpufeature.h index d3f5c63078d8..89270b4318db 100644 --- a/arch/x86/include/asm/cpufeature.h +++ b/arch/x86/include/asm/cpufeature.h @@ -374,7 +374,7 @@ static __always_inline __pure bool __static_cpu_has(u16 bit) * Catch too early usage of this before alternatives * have run. */ - asm goto("1: jmp %l[t_warn]\n" + asm_volatile_goto("1: jmp %l[t_warn]\n" "2:\n" ".section .altinstructions,\"a\"\n" " .long 1b - .\n" @@ -388,7 +388,7 @@ static __always_inline __pure bool __static_cpu_has(u16 bit) #endif - asm goto("1: jmp %l[t_no]\n" + asm_volatile_goto("1: jmp %l[t_no]\n" "2:\n" ".section .altinstructions,\"a\"\n" " .long 1b - .\n" @@ -453,7 +453,7 @@ static __always_inline __pure bool _static_cpu_has_safe(u16 bit) * have. Thus, we force the jump to the widest, 4-byte, signed relative * offset even though the last would often fit in less bytes. */ - asm goto("1: .byte 0xe9\n .long %l[t_dynamic] - 2f\n" + asm_volatile_goto("1: .byte 0xe9\n .long %l[t_dynamic] - 2f\n" "2:\n" ".section .altinstructions,\"a\"\n" " .long 1b - .\n" /* src offset */ diff --git a/arch/x86/include/asm/jump_label.h b/arch/x86/include/asm/jump_label.h index 64507f35800c..6a2cefb4395a 100644 --- a/arch/x86/include/asm/jump_label.h +++ b/arch/x86/include/asm/jump_label.h @@ -18,7 +18,7 @@ static __always_inline bool arch_static_branch(struct static_key *key) { - asm goto("1:" + asm_volatile_goto("1:" ".byte " __stringify(STATIC_KEY_INIT_NOP) "\n\t" ".pushsection __jump_table, \"aw\" \n\t" _ASM_ALIGN "\n\t" diff --git a/arch/x86/include/asm/mutex_64.h b/arch/x86/include/asm/mutex_64.h index e7e6751648ed..07537a44216e 100644 --- a/arch/x86/include/asm/mutex_64.h +++ b/arch/x86/include/asm/mutex_64.h @@ -20,7 +20,7 @@ static inline void __mutex_fastpath_lock(atomic_t *v, void (*fail_fn)(atomic_t *)) { - asm volatile goto(LOCK_PREFIX " decl %0\n" + asm_volatile_goto(LOCK_PREFIX " decl %0\n" " jns %l[exit]\n" : : "m" (v->counter) : "memory", "cc" @@ -75,7 +75,7 @@ static inline int __mutex_fastpath_lock_retval(atomic_t *count) static inline void __mutex_fastpath_unlock(atomic_t *v, void (*fail_fn)(atomic_t *)) { - asm volatile goto(LOCK_PREFIX " incl %0\n" + asm_volatile_goto(LOCK_PREFIX " incl %0\n" " jg %l[exit]\n" : : "m" (v->counter) : "memory", "cc" diff --git a/include/linux/compiler-gcc4.h b/include/linux/compiler-gcc4.h index 842de225055f..ded429966c1f 100644 --- a/include/linux/compiler-gcc4.h +++ b/include/linux/compiler-gcc4.h @@ -65,6 +65,21 @@ #define __visible __attribute__((externally_visible)) #endif +/* + * GCC 'asm goto' miscompiles certain code sequences: + * + * http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58670 + * + * Work it around via a compiler barrier quirk suggested by Jakub Jelinek. + * Fixed in GCC 4.8.2 and later versions. + * + * (asm goto is automatically volatile - the naming reflects this.) + */ +#if GCC_VERSION <= 40801 +# define asm_volatile_goto(x...) do { asm goto(x); asm (""); } while (0) +#else +# define asm_volatile_goto(x...) do { asm goto(x); } while (0) +#endif #ifdef CONFIG_ARCH_USE_BUILTIN_BSWAP #if GCC_VERSION >= 40400