From 405c514f95898452db728df6bc79f63e44a3f2d0 Mon Sep 17 00:00:00 2001 From: Steve French Date: Sat, 23 Dec 2006 18:44:33 +0000 Subject: [PATCH 001/326] [CIFS] Update CIFS version number Signed-off-by: Steve French --- fs/cifs/CHANGES | 4 ++++ fs/cifs/cifsfs.h | 2 +- 2 files changed, 5 insertions(+), 1 deletion(-) diff --git a/fs/cifs/CHANGES b/fs/cifs/CHANGES index 3539d6ef9611..d04d2f7448d9 100644 --- a/fs/cifs/CHANGES +++ b/fs/cifs/CHANGES @@ -1,3 +1,7 @@ +Version 1.47 +------------ +Fix oops in list_del during mount caused by unaligned string. + Version 1.46 ------------ Support deep tree mounts. Better support OS/2, Win9x (DOS) time stamps. diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index a243f779b363..8aa66dcf13bd 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -100,5 +100,5 @@ extern ssize_t cifs_getxattr(struct dentry *, const char *, void *, size_t); extern ssize_t cifs_listxattr(struct dentry *, char *, size_t); extern int cifs_ioctl (struct inode * inode, struct file * filep, unsigned int command, unsigned long arg); -#define CIFS_VERSION "1.46" +#define CIFS_VERSION "1.47" #endif /* _CIFSFS_H */ From 4b95320fc4d21b0ff2f8604305dd6c851aff6096 Mon Sep 17 00:00:00 2001 From: Wang Zhenyu Date: Wed, 17 Jan 2007 11:07:54 +0800 Subject: [PATCH 002/326] [AGPGART] intel_agp: restore graphics device's pci space early in resume Currently in resuming path graphics device's pci space restore is behind host bridge, so resume function wrongly accesses graphics device's space. This makes resuming failure which crashed X. here's a patch to restore device's pci space early, which makes resuming ok with X. Signed-off-by: Wang Zhenyu Signed-off-by: Dave Jones --- drivers/char/agp/intel-agp.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/char/agp/intel-agp.c b/drivers/char/agp/intel-agp.c index ab0a9c0ad7c0..a3011de51f7c 100644 --- a/drivers/char/agp/intel-agp.c +++ b/drivers/char/agp/intel-agp.c @@ -1955,6 +1955,15 @@ static int agp_intel_resume(struct pci_dev *pdev) pci_restore_state(pdev); + /* We should restore our graphics device's config space, + * as host bridge (00:00) resumes before graphics device (02:00), + * then our access to its pci space can work right. + */ + if (intel_i810_private.i810_dev) + pci_restore_state(intel_i810_private.i810_dev); + if (intel_i830_private.i830_dev) + pci_restore_state(intel_i830_private.i830_dev); + if (bridge->driver == &intel_generic_driver) intel_configure(); else if (bridge->driver == &intel_850_driver) From 76849e3e97a95f1ef4627b10305d38a686fab112 Mon Sep 17 00:00:00 2001 From: Steve French Date: Sun, 21 Jan 2007 22:56:22 +0000 Subject: [PATCH 003/326] [CIFS] Remove 2 unneeded kzalloc casts Signed-off-by: Ahmed Darwish Signed-off-by: Steve French --- fs/cifs/misc.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index aedf683f011f..19cc294c7c70 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -71,9 +71,7 @@ sesInfoAlloc(void) { struct cifsSesInfo *ret_buf; - ret_buf = - (struct cifsSesInfo *) kzalloc(sizeof (struct cifsSesInfo), - GFP_KERNEL); + ret_buf = kzalloc(sizeof (struct cifsSesInfo), GFP_KERNEL); if (ret_buf) { write_lock(&GlobalSMBSeslock); atomic_inc(&sesInfoAllocCount); @@ -109,9 +107,7 @@ struct cifsTconInfo * tconInfoAlloc(void) { struct cifsTconInfo *ret_buf; - ret_buf = - (struct cifsTconInfo *) kzalloc(sizeof (struct cifsTconInfo), - GFP_KERNEL); + ret_buf = kzalloc(sizeof (struct cifsTconInfo), GFP_KERNEL); if (ret_buf) { write_lock(&GlobalSMBSeslock); atomic_inc(&tconInfoAllocCount); From bd2abf177b3384375c43906be551d976e4c18166 Mon Sep 17 00:00:00 2001 From: Steve French Date: Sun, 21 Jan 2007 23:19:01 +0000 Subject: [PATCH 004/326] [CIFS] cifs sprintf fix Cc: Signed-off-by: Andrew Morton Signed-off-by: Steve French --- fs/cifs/cifs_debug.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c index 96abeb738978..6017c465440e 100644 --- a/fs/cifs/cifs_debug.c +++ b/fs/cifs/cifs_debug.c @@ -143,8 +143,8 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, ses = list_entry(tmp, struct cifsSesInfo, cifsSessionList); if((ses->serverDomain == NULL) || (ses->serverOS == NULL) || (ses->serverNOS == NULL)) { - buf += sprintf("\nentry for %s not fully displayed\n\t", - ses->serverName); + buf += sprintf(buf, "\nentry for %s not fully " + "displayed\n\t", ses->serverName); } else { length = From 8e6f195af0e1f226e9b2e0256af8df46adb9d595 Mon Sep 17 00:00:00 2001 From: Steve French Date: Mon, 22 Jan 2007 01:19:30 +0000 Subject: [PATCH 005/326] [CIFS] Fix oops when Windows server sent bad domain name null terminator Fixes RedHat bug 211672 Windows sends one byte (instead of two) of null to terminate final Unicode string (domain name) in session setup response in some cases - this caused cifs to misalign some informational strings (making it hard to convert from UCS16 to UTF8). Thanks to Shaggy for his help and Akemi Yagi for debugging/testing Signed-off-by: Shirish Pargaonkar Signed-off-by: Steve French --- fs/cifs/sess.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c index bbdda99dce61..758464630893 100644 --- a/fs/cifs/sess.c +++ b/fs/cifs/sess.c @@ -182,11 +182,14 @@ static int decode_unicode_ssetup(char ** pbcc_area, int bleft, struct cifsSesInf cFYI(1,("bleft %d",bleft)); - /* word align, if bytes remaining is not even */ - if(bleft % 2) { - bleft--; - data++; - } + /* SMB header is unaligned, so cifs servers word align start of + Unicode strings */ + data++; + bleft--; /* Windows servers do not always double null terminate + their final Unicode string - in which case we + now will not attempt to decode the byte of junk + which follows it */ + words_left = bleft / 2; /* save off server operating system */ From 7386397636d49cd5f03da29432467d3e98cbad35 Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Mon, 22 Jan 2007 22:00:40 +0100 Subject: [PATCH 006/326] [Bluetooth] Missing endian swapping for L2CAP socket list The PSM value in the L2CAP socket list must be converted to host order before printing it. Signed-off-by: Marcel Holtmann --- net/bluetooth/l2cap.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/bluetooth/l2cap.c b/net/bluetooth/l2cap.c index 29a8fa4d3728..be5a6e60a3c4 100644 --- a/net/bluetooth/l2cap.c +++ b/net/bluetooth/l2cap.c @@ -2150,8 +2150,8 @@ static ssize_t l2cap_sysfs_show(struct class *dev, char *buf) str += sprintf(str, "%s %s %d %d 0x%4.4x 0x%4.4x %d %d 0x%x\n", batostr(&bt_sk(sk)->src), batostr(&bt_sk(sk)->dst), - sk->sk_state, pi->psm, pi->scid, pi->dcid, pi->imtu, - pi->omtu, pi->link_mode); + sk->sk_state, btohs(pi->psm), pi->scid, pi->dcid, + pi->imtu, pi->omtu, pi->link_mode); } read_unlock_bh(&l2cap_sk_list.lock); From 847641d7db15ac3f18b3d4aa05479812abdf397a Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Mon, 22 Jan 2007 22:00:45 +0100 Subject: [PATCH 007/326] [Bluetooth] Restrict well known PSM to privileged users The PSM values below 0x1001 of L2CAP are reserved for well known services. Restrict the possibility to bind them to privileged users. Signed-off-by: Marcel Holtmann --- net/bluetooth/l2cap.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/net/bluetooth/l2cap.c b/net/bluetooth/l2cap.c index be5a6e60a3c4..f8c25d500155 100644 --- a/net/bluetooth/l2cap.c +++ b/net/bluetooth/l2cap.c @@ -585,6 +585,12 @@ static int l2cap_sock_bind(struct socket *sock, struct sockaddr *addr, int addr_ goto done; } + if (la->l2_psm > 0 && btohs(la->l2_psm) < 0x1001 && + !capable(CAP_NET_BIND_SERVICE)) { + err = -EACCES; + goto done; + } + write_lock_bh(&l2cap_sk_list.lock); if (la->l2_psm && __l2cap_get_sock_by_addr(la->l2_psm, &la->l2_bdaddr)) { From a20f3a6d7e67a8aee571fb04634a631ba59f6e92 Mon Sep 17 00:00:00 2001 From: Ishai Rabinovitz Date: Tue, 16 Jan 2007 17:20:25 +0200 Subject: [PATCH 008/326] IB/srp: Check match_strdup() return Checks if the kmalloc in match_strdup() was successful, and bail out on looking at the token if it failed. Signed-off-by: Ishai Rabinovitz Signed-off-by: Roland Dreier --- drivers/infiniband/ulp/srp/ib_srp.c | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) diff --git a/drivers/infiniband/ulp/srp/ib_srp.c b/drivers/infiniband/ulp/srp/ib_srp.c index cdecbf5911c8..72611fd15103 100644 --- a/drivers/infiniband/ulp/srp/ib_srp.c +++ b/drivers/infiniband/ulp/srp/ib_srp.c @@ -1621,18 +1621,30 @@ static int srp_parse_options(const char *buf, struct srp_target_port *target) switch (token) { case SRP_OPT_ID_EXT: p = match_strdup(args); + if (!p) { + ret = -ENOMEM; + goto out; + } target->id_ext = cpu_to_be64(simple_strtoull(p, NULL, 16)); kfree(p); break; case SRP_OPT_IOC_GUID: p = match_strdup(args); + if (!p) { + ret = -ENOMEM; + goto out; + } target->ioc_guid = cpu_to_be64(simple_strtoull(p, NULL, 16)); kfree(p); break; case SRP_OPT_DGID: p = match_strdup(args); + if (!p) { + ret = -ENOMEM; + goto out; + } if (strlen(p) != 32) { printk(KERN_WARNING PFX "bad dest GID parameter '%s'\n", p); kfree(p); @@ -1656,6 +1668,10 @@ static int srp_parse_options(const char *buf, struct srp_target_port *target) case SRP_OPT_SERVICE_ID: p = match_strdup(args); + if (!p) { + ret = -ENOMEM; + goto out; + } target->service_id = cpu_to_be64(simple_strtoull(p, NULL, 16)); kfree(p); break; @@ -1693,6 +1709,10 @@ static int srp_parse_options(const char *buf, struct srp_target_port *target) case SRP_OPT_INITIATOR_EXT: p = match_strdup(args); + if (!p) { + ret = -ENOMEM; + goto out; + } target->initiator_ext = cpu_to_be64(simple_strtoull(p, NULL, 16)); kfree(p); break; From ce29d72cc737df3573854a4719f00385adf1c9a6 Mon Sep 17 00:00:00 2001 From: Hoang-Nam Nguyen Date: Fri, 19 Jan 2007 22:50:10 +0100 Subject: [PATCH 009/326] IB/ehca: Fix improper use of yield() with spinlock held Signed-off-by: Hoang-Nam Nguyen Signed-off-by: Roland Dreier --- drivers/infiniband/hw/ehca/ehca_cq.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/ehca/ehca_cq.c b/drivers/infiniband/hw/ehca/ehca_cq.c index 93995b658d94..6074c897f51c 100644 --- a/drivers/infiniband/hw/ehca/ehca_cq.c +++ b/drivers/infiniband/hw/ehca/ehca_cq.c @@ -344,8 +344,11 @@ int ehca_destroy_cq(struct ib_cq *cq) unsigned long flags; spin_lock_irqsave(&ehca_cq_idr_lock, flags); - while (my_cq->nr_callbacks) + while (my_cq->nr_callbacks) { + spin_unlock_irqrestore(&ehca_cq_idr_lock, flags); yield(); + spin_lock_irqsave(&ehca_cq_idr_lock, flags); + } idr_remove(&ehca_cq_idr, my_cq->token); spin_unlock_irqrestore(&ehca_cq_idr_lock, flags); From cea9ea67e9927da18af89b49bd949a8d65ba1b15 Mon Sep 17 00:00:00 2001 From: Hoang-Nam Nguyen Date: Fri, 19 Jan 2007 22:50:10 +0100 Subject: [PATCH 010/326] IB/ehca: Fix mismatched spin_unlock in irq handler The lock is taken with _irqsave and hence must be released with _irqrestore on all paths. Signed-off-by Hoang-Nam Nguyen Signed-off-by: Roland Dreier --- drivers/infiniband/hw/ehca/ehca_irq.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/ehca/ehca_irq.c b/drivers/infiniband/hw/ehca/ehca_irq.c index e7209afb4250..c069be8cbcb2 100644 --- a/drivers/infiniband/hw/ehca/ehca_irq.c +++ b/drivers/infiniband/hw/ehca/ehca_irq.c @@ -440,7 +440,8 @@ void ehca_tasklet_eq(unsigned long data) cq = idr_find(&ehca_cq_idr, token); if (cq == NULL) { - spin_unlock(&ehca_cq_idr_lock); + spin_unlock_irqrestore(&ehca_cq_idr_lock, + flags); break; } From d08d283974f96cb30d78ba24282a0a7d6709af32 Mon Sep 17 00:00:00 2001 From: Komuro Date: Sat, 2 Dec 2006 11:53:27 +0900 Subject: [PATCH 011/326] modify 3c589_cs to be SMP safe 1. EL3WINDOW is always 1 when lock is not held. 2. The second argument of el3_interrupt is 'void *dev_id', not 'struct el3_private *lp'. Signed-off-by: komurojun-mbn@nifty.com Signed-off-by: Jeff Garzik --- drivers/net/pcmcia/3c589_cs.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/net/pcmcia/3c589_cs.c b/drivers/net/pcmcia/3c589_cs.c index 342f4062de0b..461e8274ef69 100644 --- a/drivers/net/pcmcia/3c589_cs.c +++ b/drivers/net/pcmcia/3c589_cs.c @@ -606,11 +606,14 @@ static int el3_start_xmit(struct sk_buff *skb, struct net_device *dev) { kio_addr_t ioaddr = dev->base_addr; struct el3_private *priv = netdev_priv(dev); + unsigned long flags; DEBUG(3, "%s: el3_start_xmit(length = %ld) called, " "status %4.4x.\n", dev->name, (long)skb->len, inw(ioaddr + EL3_STATUS)); + spin_lock_irqsave(&priv->lock, flags); + priv->stats.tx_bytes += skb->len; /* Put out the doubleword header... */ @@ -628,6 +631,7 @@ static int el3_start_xmit(struct sk_buff *skb, struct net_device *dev) dev_kfree_skb(skb); pop_tx_status(dev); + spin_unlock_irqrestore(&priv->lock, flags); return 0; } @@ -729,14 +733,13 @@ static void media_check(unsigned long arg) if (!netif_device_present(dev)) goto reschedule; - EL3WINDOW(1); /* Check for pending interrupt with expired latency timer: with this, we can limp along even if the interrupt is blocked */ if ((inw(ioaddr + EL3_STATUS) & IntLatch) && (inb(ioaddr + EL3_TIMER) == 0xff)) { if (!lp->fast_poll) printk(KERN_WARNING "%s: interrupt(s) dropped!\n", dev->name); - el3_interrupt(dev->irq, lp); + el3_interrupt(dev->irq, dev); lp->fast_poll = HZ; } if (lp->fast_poll) { From 54b290a2ecfbfb8613fbf854f9c45667821d9b8d Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Tue, 23 Jan 2007 00:29:01 -0500 Subject: [PATCH 012/326] Note that JFFS (v1) is to be deleted, in feature-removal-schedule.txt It is already noted in Kconfig, but the listing in this file was accidentally forgotten. Signed-off-by: Jeff Garzik --- Documentation/feature-removal-schedule.txt | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index fc532395d116..0ba6af02cdaf 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -318,3 +318,10 @@ Why: /proc/acpi/button has been replaced by events to the input layer Who: Len Brown --------------------------- + +What: JFFS (version 1) +When: 2.6.21 +Why: Unmaintained for years, superceded by JFFS2 for years. +Who: Jeff Garzik + +--------------------------- From b3a242b75361936ab9a42c42c44ea35e79a9d4cd Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Thu, 11 Jan 2007 14:49:44 +0100 Subject: [PATCH 013/326] more ftape removal This patch removes some more ftape code. Signed-off-by: Adrian Bunk Signed-off-by: Jeff Garzik --- include/linux/Kbuild | 1 - include/linux/mtio.h | 146 --------------------- include/linux/qic117.h | 290 ----------------------------------------- 3 files changed, 437 deletions(-) delete mode 100644 include/linux/qic117.h diff --git a/include/linux/Kbuild b/include/linux/Kbuild index 862e483cce8a..8c634f9df39f 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -129,7 +129,6 @@ header-y += posix_types.h header-y += ppdev.h header-y += prctl.h header-y += ps2esdi.h -header-y += qic117.h header-y += qnxtypes.h header-y += quotaio_v1.h header-y += quotaio_v2.h diff --git a/include/linux/mtio.h b/include/linux/mtio.h index 8c66151821e3..6f8d2d45a8fb 100644 --- a/include/linux/mtio.h +++ b/include/linux/mtio.h @@ -10,7 +10,6 @@ #include #include -#include /* * Structures and definitions for mag tape io control commands @@ -116,32 +115,6 @@ struct mtget { #define MT_ISFTAPE_UNKNOWN 0x800000 /* obsolete */ #define MT_ISFTAPE_FLAG 0x800000 -struct mt_tape_info { - long t_type; /* device type id (mt_type) */ - char *t_name; /* descriptive name */ -}; - -#define MT_TAPE_INFO { \ - {MT_ISUNKNOWN, "Unknown type of tape device"}, \ - {MT_ISQIC02, "Generic QIC-02 tape streamer"}, \ - {MT_ISWT5150, "Wangtek 5150, QIC-150"}, \ - {MT_ISARCHIVE_5945L2, "Archive 5945L-2"}, \ - {MT_ISCMSJ500, "CMS Jumbo 500"}, \ - {MT_ISTDC3610, "Tandberg TDC 3610, QIC-24"}, \ - {MT_ISARCHIVE_VP60I, "Archive VP60i, QIC-02"}, \ - {MT_ISARCHIVE_2150L, "Archive Viper 2150L"}, \ - {MT_ISARCHIVE_2060L, "Archive Viper 2060L"}, \ - {MT_ISARCHIVESC499, "Archive SC-499 QIC-36 controller"}, \ - {MT_ISQIC02_ALL_FEATURES, "Generic QIC-02 tape, all features"}, \ - {MT_ISWT5099EEN24, "Wangtek 5099-een24, 60MB"}, \ - {MT_ISTEAC_MT2ST, "Teac MT-2ST 155mb data cassette drive"}, \ - {MT_ISEVEREX_FT40A, "Everex FT40A, QIC-40"}, \ - {MT_ISONSTREAM_SC, "OnStream SC-, DI-, DP-, or USB tape drive"}, \ - {MT_ISSCSI1, "Generic SCSI-1 tape"}, \ - {MT_ISSCSI2, "Generic SCSI-2 tape"}, \ - {0, NULL} \ -} - /* structure for MTIOCPOS - mag tape get position command */ @@ -150,130 +123,11 @@ struct mtpos { }; -/* structure for MTIOCVOLINFO, query information about the volume - * currently positioned at (zftape) - */ -struct mtvolinfo { - unsigned int mt_volno; /* vol-number */ - unsigned int mt_blksz; /* blocksize used when recording */ - unsigned int mt_rawsize; /* raw tape space consumed, in kb */ - unsigned int mt_size; /* volume size after decompression, in kb */ - unsigned int mt_cmpr:1; /* this volume has been compressed */ -}; - -/* raw access to a floppy drive, read and write an arbitrary segment. - * For ftape/zftape to support formatting etc. - */ -#define MT_FT_RD_SINGLE 0 -#define MT_FT_RD_AHEAD 1 -#define MT_FT_WR_ASYNC 0 /* start tape only when all buffers are full */ -#define MT_FT_WR_MULTI 1 /* start tape, continue until buffers are empty */ -#define MT_FT_WR_SINGLE 2 /* write a single segment and stop afterwards */ -#define MT_FT_WR_DELETE 3 /* write deleted data marks, one segment at time */ - -struct mtftseg -{ - unsigned mt_segno; /* the segment to read or write */ - unsigned mt_mode; /* modes for read/write (sync/async etc.) */ - int mt_result; /* result of r/w request, not of the ioctl */ - void __user *mt_data; /* User space buffer: must be 29kb */ -}; - -/* get tape capacity (ftape/zftape) - */ -struct mttapesize { - unsigned long mt_capacity; /* entire, uncompressed capacity - * of a cartridge - */ - unsigned long mt_used; /* what has been used so far, raw - * uncompressed amount - */ -}; - -/* possible values of the ftfmt_op field - */ -#define FTFMT_SET_PARMS 1 /* set software parms */ -#define FTFMT_GET_PARMS 2 /* get software parms */ -#define FTFMT_FORMAT_TRACK 3 /* start formatting a tape track */ -#define FTFMT_STATUS 4 /* monitor formatting a tape track */ -#define FTFMT_VERIFY 5 /* verify the given segment */ - -struct ftfmtparms { - unsigned char ft_qicstd; /* QIC-40/QIC-80/QIC-3010/QIC-3020 */ - unsigned char ft_fmtcode; /* Refer to the QIC specs */ - unsigned char ft_fhm; /* floppy head max */ - unsigned char ft_ftm; /* floppy track max */ - unsigned short ft_spt; /* segments per track */ - unsigned short ft_tpc; /* tracks per cartridge */ -}; - -struct ftfmttrack { - unsigned int ft_track; /* track to format */ - unsigned char ft_gap3; /* size of gap3, for FORMAT_TRK */ -}; - -struct ftfmtstatus { - unsigned int ft_segment; /* segment currently being formatted */ -}; - -struct ftfmtverify { - unsigned int ft_segment; /* segment to verify */ - unsigned long ft_bsm; /* bsm as result of VERIFY cmd */ -}; - -struct mtftformat { - unsigned int fmt_op; /* operation to perform */ - union fmt_arg { - struct ftfmtparms fmt_parms; /* format parameters */ - struct ftfmttrack fmt_track; /* ctrl while formatting */ - struct ftfmtstatus fmt_status; - struct ftfmtverify fmt_verify; /* for verifying */ - } fmt_arg; -}; - -struct mtftcmd { - unsigned int ft_wait_before; /* timeout to wait for drive to get ready - * before command is sent. Milliseconds - */ - qic117_cmd_t ft_cmd; /* command to send */ - unsigned char ft_parm_cnt; /* zero: no parm is sent. */ - unsigned char ft_parms[3]; /* parameter(s) to send to - * the drive. The parms are nibbles - * driver sends cmd + 2 step pulses */ - unsigned int ft_result_bits; /* if non zero, number of bits - * returned by the tape drive - */ - unsigned int ft_result; /* the result returned by the tape drive*/ - unsigned int ft_wait_after; /* timeout to wait for drive to get ready - * after command is sent. 0: don't wait */ - int ft_status; /* status returned by ready wait - * undefined if timeout was 0. - */ - int ft_error; /* error code if error status was set by - * command - */ -}; - /* mag tape io control commands */ #define MTIOCTOP _IOW('m', 1, struct mtop) /* do a mag tape op */ #define MTIOCGET _IOR('m', 2, struct mtget) /* get tape status */ #define MTIOCPOS _IOR('m', 3, struct mtpos) /* get tape position */ -/* The next two are used by the QIC-02 driver for runtime reconfiguration. - * See tpqic02.h for struct mtconfiginfo. - */ -#define MTIOCGETCONFIG _IOR('m', 4, struct mtconfiginfo) /* get tape config */ -#define MTIOCSETCONFIG _IOW('m', 5, struct mtconfiginfo) /* set tape config */ - -/* the next six are used by the floppy ftape drivers and its frontends - * sorry, but MTIOCTOP commands are write only. - */ -#define MTIOCRDFTSEG _IOWR('m', 6, struct mtftseg) /* read a segment */ -#define MTIOCWRFTSEG _IOWR('m', 7, struct mtftseg) /* write a segment */ -#define MTIOCVOLINFO _IOR('m', 8, struct mtvolinfo) /* info about volume */ -#define MTIOCGETSIZE _IOR('m', 9, struct mttapesize)/* get cartridge size*/ -#define MTIOCFTFORMAT _IOWR('m', 10, struct mtftformat) /* format ftape */ -#define MTIOCFTCMD _IOWR('m', 11, struct mtftcmd) /* send QIC-117 cmd */ /* Generic Mag Tape (device independent) status macros for examining * mt_gstat -- HP-UX compatible. diff --git a/include/linux/qic117.h b/include/linux/qic117.h deleted file mode 100644 index 07b537e5d359..000000000000 --- a/include/linux/qic117.h +++ /dev/null @@ -1,290 +0,0 @@ -#ifndef _QIC117_H -#define _QIC117_H - -/* - * Copyright (C) 1993-1996 Bas Laarhoven, - * (C) 1997 Claus-Justus Heine. - - This program is free software; you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation; either version 2, or (at your option) - any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. - - You should have received a copy of the GNU General Public License - along with this program; see the file COPYING. If not, write to - the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. - - * - * $Source: /homes/cvs/ftape-stacked/include/linux/qic117.h,v $ - * $Revision: 1.2 $ - * $Date: 1997/10/05 19:19:32 $ - * - * This file contains QIC-117 spec. related definitions for the - * QIC-40/80/3010/3020 floppy-tape driver "ftape" for Linux. - * - * These data were taken from the Quarter-Inch Cartridge - * Drive Standards, Inc. document titled: - * `Common Command Set Interface Specification for Flexible - * Disk Controller Based Minicartridge Tape Drives' - * document QIC-117 Revision J, 28 Aug 96. - * For more information, contact: - * Quarter-Inch Cartridge Drive Standards, Inc. - * 311 East Carrillo Street - * Santa Barbara, California 93101 - * Telephone (805) 963-3853 - * Fax (805) 962-1541 - * WWW http://www.qic.org - * - * Current QIC standard revisions (of interest) are: - * QIC-40-MC, Rev. M, 2 Sep 92. - * QIC-80-MC, Rev. N, 20 Mar 96. - * QIC-80-MC, Rev. K, 15 Dec 94. - * QIC-113, Rev. G, 15 Jun 95. - * QIC-117, Rev. J, 28 Aug 96. - * QIC-122, Rev. B, 6 Mar 91. - * QIC-130, Rev. C, 2 Sep 92. - * QIC-3010-MC, Rev. F, 14 Jun 95. - * QIC-3020-MC, Rev. G, 31 Aug 95. - * QIC-CRF3, Rev. B, 15 Jun 95. - * */ - -/* - * QIC-117 common command set rev. J. - * These commands are sent to the tape unit - * as number of pulses over the step line. - */ - -typedef enum { - QIC_NO_COMMAND = 0, - QIC_RESET = 1, - QIC_REPORT_NEXT_BIT = 2, - QIC_PAUSE = 3, - QIC_MICRO_STEP_PAUSE = 4, - QIC_ALTERNATE_TIMEOUT = 5, - QIC_REPORT_DRIVE_STATUS = 6, - QIC_REPORT_ERROR_CODE = 7, - QIC_REPORT_DRIVE_CONFIGURATION = 8, - QIC_REPORT_ROM_VERSION = 9, - QIC_LOGICAL_FORWARD = 10, - QIC_PHYSICAL_REVERSE = 11, - QIC_PHYSICAL_FORWARD = 12, - QIC_SEEK_HEAD_TO_TRACK = 13, - QIC_SEEK_LOAD_POINT = 14, - QIC_ENTER_FORMAT_MODE = 15, - QIC_WRITE_REFERENCE_BURST = 16, - QIC_ENTER_VERIFY_MODE = 17, - QIC_STOP_TAPE = 18, -/* commands 19-20: reserved */ - QIC_MICRO_STEP_HEAD_UP = 21, - QIC_MICRO_STEP_HEAD_DOWN = 22, - QIC_SOFT_SELECT = 23, - QIC_SOFT_DESELECT = 24, - QIC_SKIP_REVERSE = 25, - QIC_SKIP_FORWARD = 26, - QIC_SELECT_RATE = 27, -/* command 27, in ccs2: Select Rate or Format */ - QIC_ENTER_DIAGNOSTIC_1 = 28, - QIC_ENTER_DIAGNOSTIC_2 = 29, - QIC_ENTER_PRIMARY_MODE = 30, -/* command 31: vendor unique */ - QIC_REPORT_VENDOR_ID = 32, - QIC_REPORT_TAPE_STATUS = 33, - QIC_SKIP_EXTENDED_REVERSE = 34, - QIC_SKIP_EXTENDED_FORWARD = 35, - QIC_CALIBRATE_TAPE_LENGTH = 36, - QIC_REPORT_FORMAT_SEGMENTS = 37, - QIC_SET_FORMAT_SEGMENTS = 38, -/* commands 39-45: reserved */ - QIC_PHANTOM_SELECT = 46, - QIC_PHANTOM_DESELECT = 47 -} qic117_cmd_t; - -typedef enum { - discretional = 0, required, ccs1, ccs2 -} qic_compatibility; - -typedef enum { - unused, mode, motion, report -} command_types; - -struct qic117_command_table { - char *name; - __u8 mask; - __u8 state; - __u8 cmd_type; - __u8 non_intr; - __u8 level; -}; - -#define QIC117_COMMANDS {\ -/* command mask state cmd_type */\ -/* | name | | | non_intr */\ -/* | | | | | | level */\ -/* 0*/ {NULL, 0x00, 0x00, mode, 0, discretional},\ -/* 1*/ {"soft reset", 0x00, 0x00, motion, 1, required},\ -/* 2*/ {"report next bit", 0x00, 0x00, report, 0, required},\ -/* 3*/ {"pause", 0x36, 0x24, motion, 1, required},\ -/* 4*/ {"micro step pause", 0x36, 0x24, motion, 1, required},\ -/* 5*/ {"alternate command timeout", 0x00, 0x00, mode, 0, required},\ -/* 6*/ {"report drive status", 0x00, 0x00, report, 0, required},\ -/* 7*/ {"report error code", 0x01, 0x01, report, 0, required},\ -/* 8*/ {"report drive configuration",0x00, 0x00, report, 0, required},\ -/* 9*/ {"report rom version", 0x00, 0x00, report, 0, required},\ -/*10*/ {"logical forward", 0x37, 0x25, motion, 0, required},\ -/*11*/ {"physical reverse", 0x17, 0x05, motion, 0, required},\ -/*12*/ {"physical forward", 0x17, 0x05, motion, 0, required},\ -/*13*/ {"seek head to track", 0x37, 0x25, motion, 0, required},\ -/*14*/ {"seek load point", 0x17, 0x05, motion, 1, required},\ -/*15*/ {"enter format mode", 0x1f, 0x05, mode, 0, required},\ -/*16*/ {"write reference burst", 0x1f, 0x05, motion, 1, required},\ -/*17*/ {"enter verify mode", 0x37, 0x25, mode, 0, required},\ -/*18*/ {"stop tape", 0x00, 0x00, motion, 1, required},\ -/*19*/ {"reserved (19)", 0x00, 0x00, unused, 0, discretional},\ -/*20*/ {"reserved (20)", 0x00, 0x00, unused, 0, discretional},\ -/*21*/ {"micro step head up", 0x02, 0x00, motion, 0, required},\ -/*22*/ {"micro step head down", 0x02, 0x00, motion, 0, required},\ -/*23*/ {"soft select", 0x00, 0x00, mode, 0, discretional},\ -/*24*/ {"soft deselect", 0x00, 0x00, mode, 0, discretional},\ -/*25*/ {"skip segments reverse", 0x36, 0x24, motion, 1, required},\ -/*26*/ {"skip segments forward", 0x36, 0x24, motion, 1, required},\ -/*27*/ {"select rate or format", 0x03, 0x01, mode, 0, required /* [ccs2] */},\ -/*28*/ {"enter diag mode 1", 0x00, 0x00, mode, 0, discretional},\ -/*29*/ {"enter diag mode 2", 0x00, 0x00, mode, 0, discretional},\ -/*30*/ {"enter primary mode", 0x00, 0x00, mode, 0, required},\ -/*31*/ {"vendor unique (31)", 0x00, 0x00, unused, 0, discretional},\ -/*32*/ {"report vendor id", 0x00, 0x00, report, 0, required},\ -/*33*/ {"report tape status", 0x04, 0x04, report, 0, ccs1},\ -/*34*/ {"skip extended reverse", 0x36, 0x24, motion, 1, ccs1},\ -/*35*/ {"skip extended forward", 0x36, 0x24, motion, 1, ccs1},\ -/*36*/ {"calibrate tape length", 0x17, 0x05, motion, 1, ccs2},\ -/*37*/ {"report format segments", 0x17, 0x05, report, 0, ccs2},\ -/*38*/ {"set format segments", 0x17, 0x05, mode, 0, ccs2},\ -/*39*/ {"reserved (39)", 0x00, 0x00, unused, 0, discretional},\ -/*40*/ {"vendor unique (40)", 0x00, 0x00, unused, 0, discretional},\ -/*41*/ {"vendor unique (41)", 0x00, 0x00, unused, 0, discretional},\ -/*42*/ {"vendor unique (42)", 0x00, 0x00, unused, 0, discretional},\ -/*43*/ {"vendor unique (43)", 0x00, 0x00, unused, 0, discretional},\ -/*44*/ {"vendor unique (44)", 0x00, 0x00, unused, 0, discretional},\ -/*45*/ {"vendor unique (45)", 0x00, 0x00, unused, 0, discretional},\ -/*46*/ {"phantom select", 0x00, 0x00, mode, 0, discretional},\ -/*47*/ {"phantom deselect", 0x00, 0x00, mode, 0, discretional},\ -} - -/* - * Status bits returned by QIC_REPORT_DRIVE_STATUS - */ - -#define QIC_STATUS_READY 0x01 /* Drive is ready or idle. */ -#define QIC_STATUS_ERROR 0x02 /* Error detected, must read - error code to clear this */ -#define QIC_STATUS_CARTRIDGE_PRESENT 0x04 /* Tape is present */ -#define QIC_STATUS_WRITE_PROTECT 0x08 /* Tape is write protected */ -#define QIC_STATUS_NEW_CARTRIDGE 0x10 /* New cartridge inserted, must - read error status to clear. */ -#define QIC_STATUS_REFERENCED 0x20 /* Cartridge appears to have been - formatted. */ -#define QIC_STATUS_AT_BOT 0x40 /* Cartridge is at physical - beginning of tape. */ -#define QIC_STATUS_AT_EOT 0x80 /* Cartridge is at physical end - of tape. */ -/* - * Status bits returned by QIC_REPORT_DRIVE_CONFIGURATION - */ - -#define QIC_CONFIG_RATE_MASK 0x18 -#define QIC_CONFIG_RATE_SHIFT 3 -#define QIC_CONFIG_RATE_250 0 -#define QIC_CONFIG_RATE_500 2 -#define QIC_CONFIG_RATE_1000 3 -#define QIC_CONFIG_RATE_2000 1 -#define QIC_CONFIG_RATE_4000 0 /* since QIC-117 Rev. J */ - -#define QIC_CONFIG_LONG 0x40 /* Extra Length Tape Detected */ -#define QIC_CONFIG_80 0x80 /* QIC-80 detected. */ - -/* - * Status bits returned by QIC_REPORT_TAPE_STATUS - */ - -#define QIC_TAPE_STD_MASK 0x0f -#define QIC_TAPE_QIC40 0x01 -#define QIC_TAPE_QIC80 0x02 -#define QIC_TAPE_QIC3020 0x03 -#define QIC_TAPE_QIC3010 0x04 - -#define QIC_TAPE_LEN_MASK 0x70 -#define QIC_TAPE_205FT 0x10 -#define QIC_TAPE_307FT 0x20 -#define QIC_TAPE_VARIABLE 0x30 -#define QIC_TAPE_1100FT 0x40 -#define QIC_TAPE_FLEX 0x60 - -#define QIC_TAPE_WIDE 0x80 - -/* Define a value (in feet) slightly higher than - * the possible maximum tape length. - */ -#define QIC_TOP_TAPE_LEN 1500 - -/* - * Errors: List of error codes, and their severity. - */ - -typedef struct { - char *message; /* Text describing the error. */ - unsigned int fatal:1; /* Non-zero if the error is fatal. */ -} ftape_error; - -#define QIC117_ERRORS {\ - /* 0*/ { "No error", 0, },\ - /* 1*/ { "Command Received while Drive Not Ready", 0, },\ - /* 2*/ { "Cartridge Not Present or Removed", 1, },\ - /* 3*/ { "Motor Speed Error (not within 1%)", 1, },\ - /* 4*/ { "Motor Speed Fault (jammed, or gross speed error", 1, },\ - /* 5*/ { "Cartridge Write Protected", 1, },\ - /* 6*/ { "Undefined or Reserved Command Code", 1, },\ - /* 7*/ { "Illegal Track Address Specified for Seek", 1, },\ - /* 8*/ { "Illegal Command in Report Subcontext", 0, },\ - /* 9*/ { "Illegal Entry into a Diagnostic Mode", 1, },\ - /*10*/ { "Broken Tape Detected (based on hole sensor)", 1, },\ - /*11*/ { "Warning--Read Gain Setting Error", 1, },\ - /*12*/ { "Command Received While Error Status Pending (obs)", 1, },\ - /*13*/ { "Command Received While New Cartridge Pending", 1, },\ - /*14*/ { "Command Illegal or Undefined in Primary Mode", 1, },\ - /*15*/ { "Command Illegal or Undefined in Format Mode", 1, },\ - /*16*/ { "Command Illegal or Undefined in Verify Mode", 1, },\ - /*17*/ { "Logical Forward Not at Logical BOT or no Format Segments in Format Mode", 1, },\ - /*18*/ { "Logical EOT Before All Segments generated", 1, },\ - /*19*/ { "Command Illegal When Cartridge Not Referenced", 1, },\ - /*20*/ { "Self-Diagnostic Failed (cannot be cleared)", 1, },\ - /*21*/ { "Warning EEPROM Not Initialized, Defaults Set", 1, },\ - /*22*/ { "EEPROM Corrupted or Hardware Failure", 1, },\ - /*23*/ { "Motion Time-out Error", 1, },\ - /*24*/ { "Data Segment Too Long -- Logical Forward or Pause", 1, },\ - /*25*/ { "Transmit Overrun (obs)", 1, },\ - /*26*/ { "Power On Reset Occurred", 0, },\ - /*27*/ { "Software Reset Occurred", 0, },\ - /*28*/ { "Diagnostic Mode 1 Error", 1, },\ - /*29*/ { "Diagnostic Mode 2 Error", 1, },\ - /*30*/ { "Command Received During Non-Interruptible Process", 1, },\ - /*31*/ { "Rate or Format Selection Error", 1, },\ - /*32*/ { "Illegal Command While in High Speed Mode", 1, },\ - /*33*/ { "Illegal Seek Segment Value", 1, },\ - /*34*/ { "Invalid Media", 1, },\ - /*35*/ { "Head Positioning Failure", 1, },\ - /*36*/ { "Write Reference Burst Failure", 1, },\ - /*37*/ { "Prom Code Missing", 1, },\ - /*38*/ { "Invalid Format", 1, },\ - /*39*/ { "EOT/BOT System Failure", 1, },\ - /*40*/ { "Prom A Checksum Error", 1, },\ - /*41*/ { "Drive Wakeup Reset Occurred", 1, },\ - /*42*/ { "Prom B Checksum Error", 1, },\ - /*43*/ { "Illegal Entry into Format Mode", 1, },\ -} - -#endif /* _QIC117_H */ From 9f6d55d0846dd8628fd3eac15be377c4d3493c3e Mon Sep 17 00:00:00 2001 From: Kumar Gala Date: Sat, 20 Jan 2007 16:38:26 -0600 Subject: [PATCH 014/326] PHY: Export phy ethtool helpers We need to export phy_ethtool_gset and phy_ethtool_sset to allow drivers that use these functions to be built as modules. Signed-off-by: Kumar Gala Signed-off-by: Jeff Garzik --- drivers/net/phy/phy.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c index e175f3910b18..9765fa661467 100644 --- a/drivers/net/phy/phy.c +++ b/drivers/net/phy/phy.c @@ -286,6 +286,7 @@ int phy_ethtool_sset(struct phy_device *phydev, struct ethtool_cmd *cmd) return 0; } +EXPORT_SYMBOL(phy_ethtool_sset); int phy_ethtool_gset(struct phy_device *phydev, struct ethtool_cmd *cmd) { @@ -302,7 +303,7 @@ int phy_ethtool_gset(struct phy_device *phydev, struct ethtool_cmd *cmd) return 0; } - +EXPORT_SYMBOL(phy_ethtool_gset); /* Note that this function is currently incompatible with the * PHYCONTROL layer. It changes registers without regard to From 061bf3cdba753ae7b52fba8cc324d81adac77696 Mon Sep 17 00:00:00 2001 From: Thomas Klein Date: Mon, 22 Jan 2007 12:52:20 +0100 Subject: [PATCH 015/326] ehea: Fixed wrong dereferencation Not only check the pointer against 0 but also the dereferenced value Signed-off-by: Thomas Klein Signed-off-by: Jeff Garzik --- drivers/net/ehea/ehea.h | 2 +- drivers/net/ehea/ehea_main.c | 6 ++++-- 2 files changed, 5 insertions(+), 3 deletions(-) diff --git a/drivers/net/ehea/ehea.h b/drivers/net/ehea/ehea.h index 39ad9f73d1ec..be10a3a26b5b 100644 --- a/drivers/net/ehea/ehea.h +++ b/drivers/net/ehea/ehea.h @@ -39,7 +39,7 @@ #include #define DRV_NAME "ehea" -#define DRV_VERSION "EHEA_0043" +#define DRV_VERSION "EHEA_0044" #define EHEA_MSG_DEFAULT (NETIF_MSG_LINK | NETIF_MSG_TIMER \ | NETIF_MSG_RX_ERR | NETIF_MSG_TX_ERR) diff --git a/drivers/net/ehea/ehea_main.c b/drivers/net/ehea/ehea_main.c index 83fa32f72398..49f669c4b019 100644 --- a/drivers/net/ehea/ehea_main.c +++ b/drivers/net/ehea/ehea_main.c @@ -2471,14 +2471,16 @@ static int __devinit ehea_probe(struct ibmebus_dev *dev, adapter_handle = (u64*)get_property(dev->ofdev.node, "ibm,hea-handle", NULL); - if (!adapter_handle) { + if (adapter_handle) + adapter->handle = *adapter_handle; + + if (!adapter->handle) { dev_err(&dev->ofdev.dev, "failed getting handle for adapter" " '%s'\n", dev->ofdev.node->full_name); ret = -ENODEV; goto out_free_ad; } - adapter->handle = *adapter_handle; adapter->pd = EHEA_PD_ID; dev->ofdev.dev.driver_data = adapter; From 602e0d100daa9ba9bc7c17e0014a6f76b638dc80 Mon Sep 17 00:00:00 2001 From: Thomas Klein Date: Mon, 22 Jan 2007 12:52:50 +0100 Subject: [PATCH 016/326] ehea: Fixing firmware queue config issue Fix to use exactly one queue for incoming packets in all firmware configurations Signed-off-by: Thomas Klein Signed-off-by: Jeff Garzik --- drivers/net/ehea/ehea_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ehea/ehea_main.c b/drivers/net/ehea/ehea_main.c index 49f669c4b019..3e9c760cb283 100644 --- a/drivers/net/ehea/ehea_main.c +++ b/drivers/net/ehea/ehea_main.c @@ -998,7 +998,7 @@ static int ehea_configure_port(struct ehea_port *port) | EHEA_BMASK_SET(PXLY_RC_JUMBO_FRAME, 1); for (i = 0; i < port->num_def_qps; i++) - cb0->default_qpn_arr[i] = port->port_res[i].qp->init_attr.qp_nr; + cb0->default_qpn_arr[i] = port->port_res[0].qp->init_attr.qp_nr; if (netif_msg_ifup(port)) ehea_dump(cb0, sizeof(*cb0), "ehea_configure_port"); From e919b5938b11e1d48a6dcdcb2860e890a954f10d Mon Sep 17 00:00:00 2001 From: Thomas Klein Date: Mon, 22 Jan 2007 12:53:20 +0100 Subject: [PATCH 017/326] ehea: Modified initial autoneg state determination Logical partitions are not allowed to (try to) set the autonegotiation status. This patch removes the respective function call from the port setup function. Signed-off-by: Thomas Klein Signed-off-by: Jeff Garzik --- drivers/net/ehea/ehea_main.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ehea/ehea_main.c b/drivers/net/ehea/ehea_main.c index 3e9c760cb283..d84d0950b658 100644 --- a/drivers/net/ehea/ehea_main.c +++ b/drivers/net/ehea/ehea_main.c @@ -642,6 +642,8 @@ int ehea_sense_port_attr(struct ehea_port *port) break; } + port->autoneg = 1; + /* Number of default QPs */ port->num_def_qps = cb0->num_default_qps; @@ -2334,8 +2336,6 @@ static int ehea_setup_single_port(struct ehea_port *port, INIT_LIST_HEAD(&port->mc_list->list); - ehea_set_portspeed(port, EHEA_SPEED_AUTONEG); - ret = ehea_sense_port_attr(port); if (ret) goto out; From 4e996b32e0585756c335b35980dc68852c33f297 Mon Sep 17 00:00:00 2001 From: Thomas Klein Date: Mon, 22 Jan 2007 12:53:50 +0100 Subject: [PATCH 018/326] ehea: New method to determine number of available ports Count OFDT nodes to determine the number of available ports instead of using the possibly outdated value from the hypervisor Signed-off-by: Thomas Klein Signed-off-by: Jeff Garzik --- drivers/net/ehea/ehea_main.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) diff --git a/drivers/net/ehea/ehea_main.c b/drivers/net/ehea/ehea_main.c index d84d0950b658..9486e0a8ec8b 100644 --- a/drivers/net/ehea/ehea_main.c +++ b/drivers/net/ehea/ehea_main.c @@ -2269,6 +2269,8 @@ static void ehea_tx_watchdog(struct net_device *dev) int ehea_sense_adapter_attr(struct ehea_adapter *adapter) { struct hcp_query_ehea *cb; + struct device_node *lhea_dn = NULL; + struct device_node *eth_dn = NULL; u64 hret; int ret; @@ -2285,7 +2287,18 @@ int ehea_sense_adapter_attr(struct ehea_adapter *adapter) goto out_herr; } - adapter->num_ports = cb->num_ports; + /* Determine the number of available logical ports + * by counting the child nodes of the lhea OFDT entry + */ + adapter->num_ports = 0; + lhea_dn = of_find_node_by_name(lhea_dn, "lhea"); + do { + eth_dn = of_get_next_child(lhea_dn, eth_dn); + if (eth_dn) + adapter->num_ports++; + } while ( eth_dn ); + of_node_put(lhea_dn); + adapter->max_mc_mac = cb->max_mc_mac - 1; ret = 0; From 7674a588e93c6fa1fde8e452a4c025c49037cb96 Mon Sep 17 00:00:00 2001 From: Thomas Klein Date: Mon, 22 Jan 2007 12:54:20 +0100 Subject: [PATCH 019/326] ehea: Improved logging of permission issues Disabled dump of hcall regs on some permission issues and fixed appropriate misleading logmessages Signed-off-by: Thomas Klein Signed-off-by: Jeff Garzik --- drivers/net/ehea/ehea_main.c | 16 +++++++--------- drivers/net/ehea/ehea_phyp.c | 10 ++++++++-- 2 files changed, 15 insertions(+), 11 deletions(-) diff --git a/drivers/net/ehea/ehea_main.c b/drivers/net/ehea/ehea_main.c index 9486e0a8ec8b..8281b30da32d 100644 --- a/drivers/net/ehea/ehea_main.c +++ b/drivers/net/ehea/ehea_main.c @@ -730,10 +730,7 @@ int ehea_set_portspeed(struct ehea_port *port, u32 port_speed) } } else { if (hret == H_AUTHORITY) { - ehea_info("Hypervisor denied setting port speed. Either" - " this partition is not authorized to set " - "port speed or another partition has modified" - " port speed first."); + ehea_info("Hypervisor denied setting port speed"); ret = -EPERM; } else { ret = -EIO; @@ -1487,11 +1484,12 @@ static int ehea_set_mac_addr(struct net_device *dev, void *sa) static void ehea_promiscuous_error(u64 hret, int enable) { - ehea_info("Hypervisor denied %sabling promiscuous mode.%s", - enable == 1 ? "en" : "dis", - hret != H_AUTHORITY ? "" : " Another partition owning a " - "logical port on the same physical port might have altered " - "promiscuous mode first."); + if (hret == H_AUTHORITY) + ehea_info("Hypervisor denied %sabling promiscuous mode", + enable == 1 ? "en" : "dis"); + else + ehea_error("failed %sabling promiscuous mode", + enable == 1 ? "en" : "dis"); } static void ehea_promiscuous(struct net_device *dev, int enable) diff --git a/drivers/net/ehea/ehea_phyp.c b/drivers/net/ehea/ehea_phyp.c index 0cfc2bc1a27b..37716e05e808 100644 --- a/drivers/net/ehea/ehea_phyp.c +++ b/drivers/net/ehea/ehea_phyp.c @@ -94,6 +94,7 @@ static long ehea_plpar_hcall9(unsigned long opcode, { long ret; int i, sleep_msecs; + u8 cb_cat; for (i = 0; i < 5; i++) { ret = plpar_hcall9(opcode, outs, @@ -106,7 +107,13 @@ static long ehea_plpar_hcall9(unsigned long opcode, continue; } - if (ret < H_SUCCESS) + cb_cat = EHEA_BMASK_GET(H_MEHEAPORT_CAT, arg2); + + if ((ret < H_SUCCESS) && !(((ret == H_AUTHORITY) + && (opcode == H_MODIFY_HEA_PORT)) + && (((cb_cat == H_PORT_CB4) && ((arg3 == H_PORT_CB4_JUMBO) + || (arg3 == H_PORT_CB4_SPEED))) || ((cb_cat == H_PORT_CB7) + && (arg3 == H_PORT_CB7_DUCQPN))))) ehea_error("opcode=%lx ret=%lx" " arg1=%lx arg2=%lx arg3=%lx arg4=%lx" " arg5=%lx arg6=%lx arg7=%lx arg8=%lx" @@ -120,7 +127,6 @@ static long ehea_plpar_hcall9(unsigned long opcode, outs[0], outs[1], outs[2], outs[3], outs[4], outs[5], outs[6], outs[7], outs[8]); - return ret; } From bb3a6449c18f6203e59195a98d633f5b5b57c133 Mon Sep 17 00:00:00 2001 From: Thomas Klein Date: Mon, 22 Jan 2007 12:54:50 +0100 Subject: [PATCH 020/326] ehea: Added logging off associated errors Added logging of error events associated with a specific queue pair Signed-off-by: Thomas Klein Signed-off-by: Jeff Garzik --- drivers/net/ehea/ehea_main.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/ehea/ehea_main.c b/drivers/net/ehea/ehea_main.c index 8281b30da32d..22c81ae4f9ad 100644 --- a/drivers/net/ehea/ehea_main.c +++ b/drivers/net/ehea/ehea_main.c @@ -558,12 +558,12 @@ static irqreturn_t ehea_qp_aff_irq_handler(int irq, void *param) u32 qp_token; eqe = ehea_poll_eq(port->qp_eq); - ehea_debug("eqe=%p", eqe); + while (eqe) { - ehea_debug("*eqe=%lx", *(u64*)eqe); - eqe = ehea_poll_eq(port->qp_eq); qp_token = EHEA_BMASK_GET(EHEA_EQE_QP_TOKEN, eqe->entry); - ehea_debug("next eqe=%p", eqe); + ehea_error("QP aff_err: entry=0x%lx, token=0x%x", + eqe->entry, qp_token); + eqe = ehea_poll_eq(port->qp_eq); } return IRQ_HANDLED; From 41b69c705152e93b3c6c872678dffd8a19b14d61 Mon Sep 17 00:00:00 2001 From: Thomas Klein Date: Mon, 22 Jan 2007 12:55:20 +0100 Subject: [PATCH 021/326] ehea: Fixed possible nullpointer access Fixed possible nullpointer access in event queue processing Signed-off-by: Thomas Klein Signed-off-by: Jeff Garzik --- drivers/net/ehea/ehea_main.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/net/ehea/ehea_main.c b/drivers/net/ehea/ehea_main.c index 22c81ae4f9ad..1072e69ef85d 100644 --- a/drivers/net/ehea/ehea_main.c +++ b/drivers/net/ehea/ehea_main.c @@ -575,8 +575,9 @@ static struct ehea_port *ehea_get_port(struct ehea_adapter *adapter, int i; for (i = 0; i < adapter->num_ports; i++) - if (adapter->port[i]->logical_port_id == logical_port) - return adapter->port[i]; + if (adapter->port[i]) + if (adapter->port[i]->logical_port_id == logical_port) + return adapter->port[i]; return NULL; } From 90f8b1d295e0c4a2148776e6cf801391cf07ae92 Mon Sep 17 00:00:00 2001 From: "Amit S. Kale" Date: Mon, 22 Jan 2007 06:38:05 -0800 Subject: [PATCH 022/326] NetXen: Firmware check modifications This patch is to make the driver work with multiple minor firmware versions Signed-off-by: Amit S. Kale Signed-off-by: Jeff Garzik --- drivers/net/netxen/netxen_nic.h | 7 +++---- drivers/net/netxen/netxen_nic_hw.c | 3 ++- 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/net/netxen/netxen_nic.h b/drivers/net/netxen/netxen_nic.h index 6490acf05305..663fa2fbe9f5 100644 --- a/drivers/net/netxen/netxen_nic.h +++ b/drivers/net/netxen/netxen_nic.h @@ -63,12 +63,11 @@ #include "netxen_nic_hw.h" -#define NETXEN_NIC_BUILD_NO "4" +#define NETXEN_NIC_BUILD_NO "1" #define _NETXEN_NIC_LINUX_MAJOR 3 #define _NETXEN_NIC_LINUX_MINOR 3 -#define _NETXEN_NIC_LINUX_SUBVERSION 2 -#define NETXEN_NIC_LINUX_VERSIONID "3.3.2" "-" NETXEN_NIC_BUILD_NO -#define NETXEN_NIC_FW_VERSIONID "3.3.2" +#define _NETXEN_NIC_LINUX_SUBVERSION 3 +#define NETXEN_NIC_LINUX_VERSIONID "3.3.3" "-" NETXEN_NIC_BUILD_NO #define RCV_DESC_RINGSIZE \ (sizeof(struct rcv_desc) * adapter->max_rx_desc_count) diff --git a/drivers/net/netxen/netxen_nic_hw.c b/drivers/net/netxen/netxen_nic_hw.c index c0c31d1914a7..191e2336e323 100644 --- a/drivers/net/netxen/netxen_nic_hw.c +++ b/drivers/net/netxen/netxen_nic_hw.c @@ -984,7 +984,8 @@ void netxen_nic_flash_print(struct netxen_adapter *adapter) _NETXEN_NIC_LINUX_MAJOR, fw_major); adapter->driver_mismatch = 1; } - if (fw_minor != _NETXEN_NIC_LINUX_MINOR) { + if (fw_minor != _NETXEN_NIC_LINUX_MINOR && + fw_minor != (_NETXEN_NIC_LINUX_MINOR + 1)) { printk(KERN_ERR "The mismatch in driver version and firmware " "version minor number\n" "Driver version minor number = %d \t" From 184231bdb41fc4c385ae8e115f5afaff02fe1690 Mon Sep 17 00:00:00 2001 From: "Amit S. Kale" Date: Mon, 22 Jan 2007 06:52:53 -0800 Subject: [PATCH 023/326] NetXen: Use pci_register_driver() instead of pci_module_init() in init_module This will use pci_register_driver() instead of pci_module_init(). Signed-off-by: Amit S. Kale Signed-off-by: Richard Knutsson Signed-off-by: Jeff Garzik --- drivers/net/netxen/netxen_nic.h | 2 +- drivers/net/netxen/netxen_nic_main.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/netxen/netxen_nic.h b/drivers/net/netxen/netxen_nic.h index 663fa2fbe9f5..59324b1693d6 100644 --- a/drivers/net/netxen/netxen_nic.h +++ b/drivers/net/netxen/netxen_nic.h @@ -63,7 +63,7 @@ #include "netxen_nic_hw.h" -#define NETXEN_NIC_BUILD_NO "1" +#define NETXEN_NIC_BUILD_NO "2" #define _NETXEN_NIC_LINUX_MAJOR 3 #define _NETXEN_NIC_LINUX_MINOR 3 #define _NETXEN_NIC_LINUX_SUBVERSION 3 diff --git a/drivers/net/netxen/netxen_nic_main.c b/drivers/net/netxen/netxen_nic_main.c index 8a5792fea774..96e1bee19ba0 100644 --- a/drivers/net/netxen/netxen_nic_main.c +++ b/drivers/net/netxen/netxen_nic_main.c @@ -1144,7 +1144,7 @@ static int __init netxen_init_module(void) if ((netxen_workq = create_singlethread_workqueue("netxen")) == 0) return -ENOMEM; - return pci_module_init(&netxen_driver); + return pci_register_driver(&netxen_driver); } module_init(netxen_init_module); From bc6191b10018311a301fb6ef22994177c769b868 Mon Sep 17 00:00:00 2001 From: Karsten Wiese Date: Wed, 10 Jan 2007 19:02:26 +0100 Subject: [PATCH 024/326] [ALSA] Repair snd-usb-usx2y over OHCI The previous patch 'Repair snd-usb-usx2y for usb 2.6.18' assumed urb->start_frame roll over beyond MAX_INT for both UHCI & OHCI. This isn't true until now (kernel 2.6.20). Fix this by only looking at the common between OHCI & UHCI Frame number range. This is for mainline and stable kernels >= 2.6.18. Signed-off-by: Karsten Wiese Signed-off-by: Takashi Iwai Signed-off-by: Jaroslav Kysela --- sound/usb/usx2y/usbusx2yaudio.c | 2 +- sound/usb/usx2y/usx2yhwdeppcm.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/sound/usb/usx2y/usbusx2yaudio.c b/sound/usb/usx2y/usbusx2yaudio.c index 367f8a32a665..0a352e46862f 100644 --- a/sound/usb/usx2y/usbusx2yaudio.c +++ b/sound/usb/usx2y/usbusx2yaudio.c @@ -322,7 +322,7 @@ static void i_usX2Y_urb_complete(struct urb *urb) usX2Y_error_urb_status(usX2Y, subs, urb); return; } - if (likely(urb->start_frame == usX2Y->wait_iso_frame)) + if (likely((urb->start_frame & 0xFFFF) == (usX2Y->wait_iso_frame & 0xFFFF))) subs->completed_urb = urb; else { usX2Y_error_sequence(usX2Y, subs, urb); diff --git a/sound/usb/usx2y/usx2yhwdeppcm.c b/sound/usb/usx2y/usx2yhwdeppcm.c index 8f3e35e24e72..a5e7bcd7ca2e 100644 --- a/sound/usb/usx2y/usx2yhwdeppcm.c +++ b/sound/usb/usx2y/usx2yhwdeppcm.c @@ -243,7 +243,7 @@ static void i_usX2Y_usbpcm_urb_complete(struct urb *urb) usX2Y_error_urb_status(usX2Y, subs, urb); return; } - if (likely(urb->start_frame == usX2Y->wait_iso_frame)) + if (likely((urb->start_frame & 0xFFFF) == (usX2Y->wait_iso_frame & 0xFFFF))) subs->completed_urb = urb; else { usX2Y_error_sequence(usX2Y, subs, urb); From bbe1a59b3a3d4af3869647d294618a23f6c9c6a4 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 22 Jan 2007 20:40:33 -0800 Subject: [PATCH 025/326] [PATCH] fix "kvm: add vm exit profiling" export profile_hits() on !SMP too. Cc: Ingo Molnar Cc: Avi Kivity Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/profile.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/kernel/profile.c b/kernel/profile.c index a6574a18514e..d6579d511069 100644 --- a/kernel/profile.c +++ b/kernel/profile.c @@ -331,7 +331,6 @@ void profile_hits(int type, void *__pc, unsigned int nr_hits) local_irq_restore(flags); put_cpu(); } -EXPORT_SYMBOL_GPL(profile_hits); static int __devinit profile_cpu_callback(struct notifier_block *info, unsigned long action, void *__cpu) @@ -401,6 +400,8 @@ void profile_hits(int type, void *__pc, unsigned int nr_hits) } #endif /* !CONFIG_SMP */ +EXPORT_SYMBOL_GPL(profile_hits); + void profile_tick(int type) { struct pt_regs *regs = get_irq_regs(); From 58d9ce7d751fa11c6c8ea5dcd4d63b320aae1363 Mon Sep 17 00:00:00 2001 From: Venkatesh Pallipadi Date: Mon, 22 Jan 2007 20:40:34 -0800 Subject: [PATCH 026/326] [PATCH] Revert nmi_known_cpu() check during boot option parsing Commit f2802e7f571c05f9a901b1f5bd144aa730ccc88e and its x86 version (b7471c6da94d30d3deadc55986cc38d1ff57f9ca) adds nmi_known_cpu() check while parsing boot options in x86_64 and i386. With that, "nmi_watchdog=2" stops working for me on Intel Core 2 CPU based system. The problem is, setup_nmi_watchdog is called while parsing the boot option and identify_cpu is not done yet. So, the return value of nmi_known_cpu() is not valid at this point. So revert that check. This should not have any adverse effect as the nmi_known_cpu() check is done again later in enable_lapic_nmi_watchdog(). Signed-off-by: Venkatesh Pallipadi Cc: Don Zickus Cc: Andi Kleen Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/nmi.c | 8 +------- arch/x86_64/kernel/nmi.c | 2 -- 2 files changed, 1 insertion(+), 9 deletions(-) diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c index a5e34d655965..1a6f8bb8881c 100644 --- a/arch/i386/kernel/nmi.c +++ b/arch/i386/kernel/nmi.c @@ -310,13 +310,7 @@ static int __init setup_nmi_watchdog(char *str) if ((nmi >= NMI_INVALID) || (nmi < NMI_NONE)) return 0; - /* - * If any other x86 CPU has a local APIC, then - * please test the NMI stuff there and send me the - * missing bits. Right now Intel P6/P4 and AMD K7 only. - */ - if ((nmi == NMI_LOCAL_APIC) && (nmi_known_cpu() == 0)) - return 0; /* no lapic support */ + nmi_watchdog = nmi; return 1; } diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c index 186aebbae32d..9cb42ecb7f89 100644 --- a/arch/x86_64/kernel/nmi.c +++ b/arch/x86_64/kernel/nmi.c @@ -302,8 +302,6 @@ int __init setup_nmi_watchdog(char *str) if ((nmi >= NMI_INVALID) || (nmi < NMI_NONE)) return 0; - if ((nmi == NMI_LOCAL_APIC) && (nmi_known_cpu() == 0)) - return 0; /* no lapic support */ nmi_watchdog = nmi; return 1; } From 790816dd54e22827f63c8cbb30bd623d3d45e594 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 22 Jan 2007 20:40:35 -0800 Subject: [PATCH 027/326] [PATCH] blockdev direct_io: fix signedness bug size_t is unsigned. IO errors aren't getting through. Cc: "Chen, Kenneth W" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/block_dev.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/block_dev.c b/fs/block_dev.c index 8b18e43b82fe..da020be25dd2 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -146,7 +146,7 @@ static int blk_end_aio(struct bio *bio, unsigned int bytes_done, int error) iocb->ki_nbytes = -EIO; if (atomic_dec_and_test(bio_count)) { - if (iocb->ki_nbytes < 0) + if ((long)iocb->ki_nbytes < 0) aio_complete(iocb, iocb->ki_nbytes, 0); else aio_complete(iocb, iocb->ki_left, 0); From a517b9f9fe8e57437b0b9b50e279220aaf651268 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 22 Jan 2007 20:40:36 -0800 Subject: [PATCH 028/326] [PATCH] SubmitChecklist update Sing the praises of `gcc -W'. Would have prevented that blockdev direct-IO bug. Cc: "Chen, Kenneth W" Cc: "Randy.Dunlap" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/SubmitChecklist | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/Documentation/SubmitChecklist b/Documentation/SubmitChecklist index 2270efa10153..bfbb2718a279 100644 --- a/Documentation/SubmitChecklist +++ b/Documentation/SubmitChecklist @@ -72,3 +72,7 @@ kernel patches. If the new code is substantial, addition of subsystem-specific fault injection might be appropriate. + +22: Newly-added code has been compiled with `gcc -W'. This will generate + lots of noise, but is good for finding bugs like "warning: comparison + between signed and unsigned". From 0dbe5a111382fd1320ff4b1d889e5b8c41290619 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Mon, 22 Jan 2007 20:40:36 -0800 Subject: [PATCH 029/326] [PATCH] paravirt: mark the paravirt_ops export internal The paravirt subsystem is still in flux so all exports from it are definitely internal use only. The APIs around this /will/ change. Signed-off-by: Ingo Molnar Cc: Andi Kleen Cc: Zachary Amsden Cc: Jeremy Fitzhardinge Acked-by: Rusty Russell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/paravirt.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/arch/i386/kernel/paravirt.c b/arch/i386/kernel/paravirt.c index 3dceab5828f1..e55fd05da0f5 100644 --- a/arch/i386/kernel/paravirt.c +++ b/arch/i386/kernel/paravirt.c @@ -566,4 +566,11 @@ struct paravirt_ops paravirt_ops = { .irq_enable_sysexit = native_irq_enable_sysexit, .iret = native_iret, }; -EXPORT_SYMBOL(paravirt_ops); + +/* + * NOTE: CONFIG_PARAVIRT is experimental and the paravirt_ops + * semantics are subject to change. Hence we only do this + * internal-only export of this, until it gets sorted out and + * all lowlevel CPU ops used by modules are separately exported. + */ +EXPORT_SYMBOL_GPL(paravirt_ops); From 084384754ebe6636f9e5554ad30b3143b4a26c84 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Mon, 22 Jan 2007 20:40:38 -0800 Subject: [PATCH 030/326] [PATCH] KVM: make sure there is a vcpu context loaded when destroying the mmu This makes the vmwrite errors on vm shutdown go away. Signed-off-by: Avi Kivity Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/kvm/kvm_main.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/kvm/kvm_main.c b/drivers/kvm/kvm_main.c index 67c1154960f0..be4651abe72c 100644 --- a/drivers/kvm/kvm_main.c +++ b/drivers/kvm/kvm_main.c @@ -272,7 +272,9 @@ static void kvm_free_physmem(struct kvm *kvm) static void kvm_free_vcpu(struct kvm_vcpu *vcpu) { + vcpu_load(vcpu->kvm, vcpu_slot(vcpu)); kvm_mmu_destroy(vcpu); + vcpu_put(vcpu); kvm_arch_ops->vcpu_free(vcpu); } From cccf748b810832cfab4dbb3ed4c7cf1a1ee35ad2 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Mon, 22 Jan 2007 20:40:39 -0800 Subject: [PATCH 031/326] [PATCH] KVM: fix race between mmio reads and injected interrupts The kvm mmio read path looks like: 1. guest read faults 2. kvm emulates read, calls emulator_read_emulated() 3. fails as a read requires userspace help 4. exit to userspace 5. userspace emulates read, kvm sets vcpu->mmio_read_completed 6. re-enter guest, fault again 7. kvm emulates read, calls emulator_read_emulated() 8. succeeds as vcpu->mmio_read_emulated is set 9. instruction completes and guest is resumed A problem surfaces if the userspace exit (step 5) also requests an interrupt injection. In that case, the guest does not re-execute the original instruction, but the interrupt handler. The next time an mmio read is exectued (likely for a different address), step 3 will find vcpu->mmio_read_completed set and return the value read for the original instruction. The problem manifested itself in a few annoying ways: - little squares appear randomly on console when switching virtual terminals - ne2000 fails under nfs read load - rtl8139 complains about "pci errors" even though the device model is incapable of issuing them. Fix by skipping interrupt injection if an mmio read is pending. A better fix is to avoid re-entry into the guest, and re-emulating immediately instead. However that's a bit more complex. Signed-off-by: Avi Kivity Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/kvm/svm.c | 3 ++- drivers/kvm/vmx.c | 3 ++- 2 files changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/kvm/svm.c b/drivers/kvm/svm.c index 714f6a7841cd..7397bfbbcb1c 100644 --- a/drivers/kvm/svm.c +++ b/drivers/kvm/svm.c @@ -1407,7 +1407,8 @@ static int svm_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) int r; again: - do_interrupt_requests(vcpu, kvm_run); + if (!vcpu->mmio_read_completed) + do_interrupt_requests(vcpu, kvm_run); clgi(); diff --git a/drivers/kvm/vmx.c b/drivers/kvm/vmx.c index 0aa2659f6ae5..27f2751c3baa 100644 --- a/drivers/kvm/vmx.c +++ b/drivers/kvm/vmx.c @@ -1717,7 +1717,8 @@ static int vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) vmcs_writel(HOST_GS_BASE, segment_base(gs_sel)); #endif - do_interrupt_requests(vcpu, kvm_run); + if (!vcpu->mmio_read_completed) + do_interrupt_requests(vcpu, kvm_run); if (vcpu->guest_debug.enabled) kvm_guest_debug_pre(vcpu); From 038e51de2e7ae2c8e9d8a0b15231f8509875dc33 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Mon, 22 Jan 2007 20:40:40 -0800 Subject: [PATCH 032/326] [PATCH] KVM: x86 emulator: fix bit string instructions The various bit string instructions (bts, btc, etc.) fail to adjust the address correctly if the bit address is beyond BITS_PER_LONG. This bug creeped in as the emulator originally relied on cr2 to contain the memory address; however we now decode it from the mod r/m bits, and must adjust the offset to account for large bit indices. The patch is rather large because it switches src and dst decoding around, so that the bit index is available when decoding the memory address. This fixes workloads like the FC5 installer. Signed-off-by: Avi Kivity Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/kvm/x86_emulate.c | 98 +++++++++++++++++++++------------------ 1 file changed, 52 insertions(+), 46 deletions(-) diff --git a/drivers/kvm/x86_emulate.c b/drivers/kvm/x86_emulate.c index be70795b4822..7513cddb929f 100644 --- a/drivers/kvm/x86_emulate.c +++ b/drivers/kvm/x86_emulate.c @@ -61,6 +61,7 @@ #define ModRM (1<<6) /* Destination is only written; never read. */ #define Mov (1<<7) +#define BitOp (1<<8) static u8 opcode_table[256] = { /* 0x00 - 0x07 */ @@ -148,7 +149,7 @@ static u8 opcode_table[256] = { 0, 0, ByteOp | DstMem | SrcNone | ModRM, DstMem | SrcNone | ModRM }; -static u8 twobyte_table[256] = { +static u16 twobyte_table[256] = { /* 0x00 - 0x0F */ 0, SrcMem | ModRM | DstReg, 0, 0, 0, 0, ImplicitOps, 0, 0, 0, 0, 0, 0, ImplicitOps | ModRM, 0, 0, @@ -180,16 +181,16 @@ static u8 twobyte_table[256] = { /* 0x90 - 0x9F */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 0xA0 - 0xA7 */ - 0, 0, 0, DstMem | SrcReg | ModRM, 0, 0, 0, 0, + 0, 0, 0, DstMem | SrcReg | ModRM | BitOp, 0, 0, 0, 0, /* 0xA8 - 0xAF */ - 0, 0, 0, DstMem | SrcReg | ModRM, 0, 0, 0, 0, + 0, 0, 0, DstMem | SrcReg | ModRM | BitOp, 0, 0, 0, 0, /* 0xB0 - 0xB7 */ ByteOp | DstMem | SrcReg | ModRM, DstMem | SrcReg | ModRM, 0, - DstMem | SrcReg | ModRM, + DstMem | SrcReg | ModRM | BitOp, 0, 0, ByteOp | DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem16 | ModRM | Mov, /* 0xB8 - 0xBF */ - 0, 0, DstMem | SrcImmByte | ModRM, DstMem | SrcReg | ModRM, + 0, 0, DstMem | SrcImmByte | ModRM, DstMem | SrcReg | ModRM | BitOp, 0, 0, ByteOp | DstReg | SrcMem | ModRM | Mov, DstReg | SrcMem16 | ModRM | Mov, /* 0xC0 - 0xCF */ @@ -469,7 +470,8 @@ static int read_descriptor(struct x86_emulate_ctxt *ctxt, int x86_emulate_memop(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) { - u8 b, d, sib, twobyte = 0, rex_prefix = 0; + unsigned d; + u8 b, sib, twobyte = 0, rex_prefix = 0; u8 modrm, modrm_mod = 0, modrm_reg = 0, modrm_rm = 0; unsigned long *override_base = NULL; unsigned int op_bytes, ad_bytes, lock_prefix = 0, rep_prefix = 0, i; @@ -726,46 +728,6 @@ x86_emulate_memop(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) ; } - /* Decode and fetch the destination operand: register or memory. */ - switch (d & DstMask) { - case ImplicitOps: - /* Special instructions do their own operand decoding. */ - goto special_insn; - case DstReg: - dst.type = OP_REG; - if ((d & ByteOp) - && !(twobyte_table && (b == 0xb6 || b == 0xb7))) { - dst.ptr = decode_register(modrm_reg, _regs, - (rex_prefix == 0)); - dst.val = *(u8 *) dst.ptr; - dst.bytes = 1; - } else { - dst.ptr = decode_register(modrm_reg, _regs, 0); - switch ((dst.bytes = op_bytes)) { - case 2: - dst.val = *(u16 *)dst.ptr; - break; - case 4: - dst.val = *(u32 *)dst.ptr; - break; - case 8: - dst.val = *(u64 *)dst.ptr; - break; - } - } - break; - case DstMem: - dst.type = OP_MEM; - dst.ptr = (unsigned long *)cr2; - dst.bytes = (d & ByteOp) ? 1 : op_bytes; - if (!(d & Mov) && /* optimisation - avoid slow emulated read */ - ((rc = ops->read_emulated((unsigned long)dst.ptr, - &dst.val, dst.bytes, ctxt)) != 0)) - goto done; - break; - } - dst.orig_val = dst.val; - /* * Decode and fetch the source operand: register, memory * or immediate. @@ -838,6 +800,50 @@ x86_emulate_memop(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) break; } + /* Decode and fetch the destination operand: register or memory. */ + switch (d & DstMask) { + case ImplicitOps: + /* Special instructions do their own operand decoding. */ + goto special_insn; + case DstReg: + dst.type = OP_REG; + if ((d & ByteOp) + && !(twobyte_table && (b == 0xb6 || b == 0xb7))) { + dst.ptr = decode_register(modrm_reg, _regs, + (rex_prefix == 0)); + dst.val = *(u8 *) dst.ptr; + dst.bytes = 1; + } else { + dst.ptr = decode_register(modrm_reg, _regs, 0); + switch ((dst.bytes = op_bytes)) { + case 2: + dst.val = *(u16 *)dst.ptr; + break; + case 4: + dst.val = *(u32 *)dst.ptr; + break; + case 8: + dst.val = *(u64 *)dst.ptr; + break; + } + } + break; + case DstMem: + dst.type = OP_MEM; + dst.ptr = (unsigned long *)cr2; + dst.bytes = (d & ByteOp) ? 1 : op_bytes; + if (d & BitOp) { + dst.ptr += src.val / BITS_PER_LONG; + dst.bytes = sizeof(long); + } + if (!(d & Mov) && /* optimisation - avoid slow emulated read */ + ((rc = ops->read_emulated((unsigned long)dst.ptr, + &dst.val, dst.bytes, ctxt)) != 0)) + goto done; + break; + } + dst.orig_val = dst.val; + if (twobyte) goto twobyte_insn; From fc3dffe12148b9612870eb21b24f2aecefa9ea24 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Mon, 22 Jan 2007 20:40:40 -0800 Subject: [PATCH 033/326] [PATCH] KVM: fix bogus pagefault on writable pages If a page is marked as dirty in the guest pte, set_pte_common() can set the writable bit on newly-instantiated shadow pte. This optimization avoids a write fault after the initial read fault. However, if a write fault instantiates the pte, fix_write_pf() incorrectly reports the fault as a guest page fault, and the guest oopses on what appears to be a correctly-mapped page. Fix is to detect the condition and only report a guest page fault on a user access to a kernel page. With the fix, a kvm guest can survive a whole night of running the kernel hacker's screensaver (make -j9 in a loop). Signed-off-by: Avi Kivity Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/kvm/paging_tmpl.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/kvm/paging_tmpl.h b/drivers/kvm/paging_tmpl.h index 2dbf4307ed9e..6bc41950fbb3 100644 --- a/drivers/kvm/paging_tmpl.h +++ b/drivers/kvm/paging_tmpl.h @@ -274,7 +274,7 @@ static int FNAME(fix_write_pf)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *page; if (is_writeble_pte(*shadow_ent)) - return 0; + return !user || (*shadow_ent & PT_USER_MASK); writable_shadow = *shadow_ent & PT_SHADOW_WRITABLE_MASK; if (user) { From 15c945c3d0913d73a7d57d7a0a3c4e2902598cc6 Mon Sep 17 00:00:00 2001 From: Jamie Lenehan Date: Mon, 22 Jan 2007 20:40:41 -0800 Subject: [PATCH 034/326] [PATCH] rtc-sh: act on rtc_wkalrm.enabled when setting an alarm This fixes the SH rtc driver correctly act on the "enabled" flag when setting an alarm. Signed-off-by: Jamie Lenehan Cc: David Brownell Cc: Alessandro Zummo Cc: Paul Mundt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/rtc/rtc-sh.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/drivers/rtc/rtc-sh.c b/drivers/rtc/rtc-sh.c index e9e0934380b8..198b9f22fbff 100644 --- a/drivers/rtc/rtc-sh.c +++ b/drivers/rtc/rtc-sh.c @@ -492,10 +492,10 @@ static int sh_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *wkalrm) spin_lock_irq(&rtc->lock); - /* disable alarm interrupt and clear flag */ + /* disable alarm interrupt and clear the alarm flag */ rcr1 = readb(rtc->regbase + RCR1); - rcr1 &= ~RCR1_AF; - writeb(rcr1 & ~RCR1_AIE, rtc->regbase + RCR1); + rcr1 &= ~(RCR1_AF|RCR1_AIE); + writeb(rcr1, rtc->regbase + RCR1); rtc->rearm_aie = 0; @@ -510,8 +510,10 @@ static int sh_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *wkalrm) mon += 1; sh_rtc_write_alarm_value(rtc, mon, RMONAR); - /* Restore interrupt activation status */ - writeb(rcr1, rtc->regbase + RCR1); + if (wkalrm->enabled) { + rcr1 |= RCR1_AIE; + writeb(rcr1, rtc->regbase + RCR1); + } spin_unlock_irq(&rtc->lock); From cda9205da24aeaa8fb086b0fb85cdf39571ecc3f Mon Sep 17 00:00:00 2001 From: "Chen, Kenneth W" Date: Mon, 22 Jan 2007 20:40:43 -0800 Subject: [PATCH 035/326] [PATCH] fix blk_direct_IO bio preparation For large size DIO that needs multiple bio, one full page worth of data was lost at the boundary of bio's maximum sector or segment limits. After a bio is full and got submitted. The outer while (nbytes) { ... } loop will allocate a new bio and just march on to index into next page. It just forgets about the page that bio_add_page() rejected when previous bio is full. Fix it by put the rejected page back to pvec so we pick it up again for the next bio. Signed-off-by: Ken Chen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/block_dev.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/fs/block_dev.c b/fs/block_dev.c index da020be25dd2..d9bdf2b3ade2 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -190,6 +190,12 @@ static struct page *blk_get_page(unsigned long addr, size_t count, int rw, return pvec->page[pvec->idx++]; } +/* return a page back to pvec array */ +static void blk_unget_page(struct page *page, struct pvec *pvec) +{ + pvec->page[--pvec->idx] = page; +} + static ssize_t blkdev_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, loff_t pos, unsigned long nr_segs) @@ -278,6 +284,8 @@ blkdev_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, count = min(count, nbytes); goto same_bio; } + } else { + blk_unget_page(page, &pvec); } /* bio is ready, submit it */ From 79603a35009ff39562cd5634fa1cf513eb080f27 Mon Sep 17 00:00:00 2001 From: Mark Gross Date: Mon, 22 Jan 2007 20:40:44 -0800 Subject: [PATCH 036/326] [PATCH] tlclk: bug fix + misc fixes The following patch fixes a few problems with the tlclk driver. * bug in the select_amcb1_transmit_clock * racy read sys call * racy open sys call * use of add_timer where mod_timer would be better * change to the timer data parameter use Signed-off-by: Mark Gross Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/tlclk.c | 43 ++++++++++++++++++++++++++++--------------- 1 file changed, 28 insertions(+), 15 deletions(-) diff --git a/drivers/char/tlclk.c b/drivers/char/tlclk.c index 448d5083c381..4fac2bdf6215 100644 --- a/drivers/char/tlclk.c +++ b/drivers/char/tlclk.c @@ -186,6 +186,7 @@ static int got_event; /* if events processing have been done */ static void switchover_timeout(unsigned long data); static struct timer_list switchover_timer = TIMER_INITIALIZER(switchover_timeout , 0, 0); +static unsigned long tlclk_timer_data; static struct tlclk_alarms *alarm_events; @@ -197,10 +198,19 @@ static irqreturn_t tlclk_interrupt(int irq, void *dev_id); static DECLARE_WAIT_QUEUE_HEAD(wq); +static unsigned long useflags; +static DEFINE_MUTEX(tlclk_mutex); + static int tlclk_open(struct inode *inode, struct file *filp) { int result; + if (test_and_set_bit(0, &useflags)) + return -EBUSY; + /* this legacy device is always one per system and it doesn't + * know how to handle multiple concurrent clients. + */ + /* Make sure there is no interrupt pending while * initialising interrupt handler */ inb(TLCLK_REG6); @@ -221,6 +231,7 @@ static int tlclk_open(struct inode *inode, struct file *filp) static int tlclk_release(struct inode *inode, struct file *filp) { free_irq(telclk_interrupt, tlclk_interrupt); + clear_bit(0, &useflags); return 0; } @@ -230,26 +241,25 @@ static ssize_t tlclk_read(struct file *filp, char __user *buf, size_t count, { if (count < sizeof(struct tlclk_alarms)) return -EIO; + if (mutex_lock_interruptible(&tlclk_mutex)) + return -EINTR; + wait_event_interruptible(wq, got_event); - if (copy_to_user(buf, alarm_events, sizeof(struct tlclk_alarms))) + if (copy_to_user(buf, alarm_events, sizeof(struct tlclk_alarms))) { + mutex_unlock(&tlclk_mutex); return -EFAULT; + } memset(alarm_events, 0, sizeof(struct tlclk_alarms)); got_event = 0; + mutex_unlock(&tlclk_mutex); return sizeof(struct tlclk_alarms); } -static ssize_t tlclk_write(struct file *filp, const char __user *buf, size_t count, - loff_t *f_pos) -{ - return 0; -} - static const struct file_operations tlclk_fops = { .read = tlclk_read, - .write = tlclk_write, .open = tlclk_open, .release = tlclk_release, @@ -540,7 +550,7 @@ static ssize_t store_select_amcb1_transmit_clock(struct device *d, SET_PORT_BITS(TLCLK_REG3, 0xf8, 0x7); switch (val) { case CLK_8_592MHz: - SET_PORT_BITS(TLCLK_REG0, 0xfc, 1); + SET_PORT_BITS(TLCLK_REG0, 0xfc, 2); break; case CLK_11_184MHz: SET_PORT_BITS(TLCLK_REG0, 0xfc, 0); @@ -549,7 +559,7 @@ static ssize_t store_select_amcb1_transmit_clock(struct device *d, SET_PORT_BITS(TLCLK_REG0, 0xfc, 3); break; case CLK_44_736MHz: - SET_PORT_BITS(TLCLK_REG0, 0xfc, 2); + SET_PORT_BITS(TLCLK_REG0, 0xfc, 1); break; } } else @@ -839,11 +849,13 @@ static void __exit tlclk_cleanup(void) static void switchover_timeout(unsigned long data) { - if ((data & 1)) { - if ((inb(TLCLK_REG1) & 0x08) != (data & 0x08)) + unsigned long flags = *(unsigned long *) data; + + if ((flags & 1)) { + if ((inb(TLCLK_REG1) & 0x08) != (flags & 0x08)) alarm_events->switchover_primary++; } else { - if ((inb(TLCLK_REG1) & 0x08) != (data & 0x08)) + if ((inb(TLCLK_REG1) & 0x08) != (flags & 0x08)) alarm_events->switchover_secondary++; } @@ -901,8 +913,9 @@ static irqreturn_t tlclk_interrupt(int irq, void *dev_id) /* TIMEOUT in ~10ms */ switchover_timer.expires = jiffies + msecs_to_jiffies(10); - switchover_timer.data = inb(TLCLK_REG1); - add_timer(&switchover_timer); + tlclk_timer_data = inb(TLCLK_REG1); + switchover_timer.data = (unsigned long) &tlclk_timer_data; + mod_timer(&switchover_timer, switchover_timer.expires); } else { got_event = 1; wake_up(&wq); From 30150f8d7b76f25b1127a5079528b7a17307f995 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Mon, 22 Jan 2007 20:40:45 -0800 Subject: [PATCH 037/326] [PATCH] mbind: restrict nodes to the currently allowed cpuset Currently one can specify an arbitrary node mask to mbind that includes nodes not allowed. If that is done with an interleave policy then we will go around all the nodes. Those outside of the currently allowed cpuset will be redirected to the border nodes. Interleave will then create imbalances at the borders of the cpuset. This patch restricts the nodes to the currently allowed cpuset. The RFC for this patch was discussed at http://marc.theaimsgroup.com/?t=116793842100004&r=1&w=2 Signed-off-by: Christoph Lameter Cc: Paul Jackson Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/mempolicy.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/mm/mempolicy.c b/mm/mempolicy.c index da9463946556..c2aec0e1090d 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -884,6 +884,10 @@ asmlinkage long sys_mbind(unsigned long start, unsigned long len, err = get_nodes(&nodes, nmask, maxnode); if (err) return err; +#ifdef CONFIG_CPUSETS + /* Restrict the nodes to the allowed nodes in the cpuset */ + nodes_and(nodes, nodes, current->mems_allowed); +#endif return do_mbind(start, len, mode, &nodes, flags); } From de14569f94513279e3d44d9571a421e9da1759ae Mon Sep 17 00:00:00 2001 From: Vladimir Saveliev Date: Mon, 22 Jan 2007 20:40:46 -0800 Subject: [PATCH 038/326] [PATCH] resierfs: avoid tail packing if an inode was ever mmapped This patch fixes a confusion reiserfs has for a long time. On release file operation reiserfs used to try to pack file data stored in last incomplete page of some files into metadata blocks. After packing the page got cleared with clear_page_dirty. It did not take into account that the page may be mmaped into other process's address space. Recent replacement for clear_page_dirty cancel_dirty_page found the confusion with sanity check that page has to be not mapped. The patch fixes the confusion by making reiserfs avoid tail packing if an inode was ever mmapped. reiserfs_mmap and reiserfs_file_release are serialized with mutex in reiserfs specific inode. reiserfs_mmap locks the mutex and sets a bit in reiserfs specific inode flags. reiserfs_file_release checks the bit having the mutex locked. If bit is set - tail packing is avoided. This eliminates a possibility that mmapped page gets cancel_page_dirty-ed. Signed-off-by: Vladimir Saveliev Cc: Jeff Mahoney Cc: Chris Mason Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/reiserfs/file.c | 20 +++++++++++++++++++- fs/reiserfs/inode.c | 2 ++ include/linux/reiserfs_fs_i.h | 2 ++ 3 files changed, 23 insertions(+), 1 deletion(-) diff --git a/fs/reiserfs/file.c b/fs/reiserfs/file.c index 99b6f329ba23..5109f1d5e7ff 100644 --- a/fs/reiserfs/file.c +++ b/fs/reiserfs/file.c @@ -48,6 +48,11 @@ static int reiserfs_file_release(struct inode *inode, struct file *filp) } mutex_lock(&inode->i_mutex); + + mutex_lock(&(REISERFS_I(inode)->i_mmap)); + if (REISERFS_I(inode)->i_flags & i_ever_mapped) + REISERFS_I(inode)->i_flags &= ~i_pack_on_close_mask; + reiserfs_write_lock(inode->i_sb); /* freeing preallocation only involves relogging blocks that * are already in the current transaction. preallocation gets @@ -100,11 +105,24 @@ static int reiserfs_file_release(struct inode *inode, struct file *filp) err = reiserfs_truncate_file(inode, 0); } out: + mutex_unlock(&(REISERFS_I(inode)->i_mmap)); mutex_unlock(&inode->i_mutex); reiserfs_write_unlock(inode->i_sb); return err; } +static int reiserfs_file_mmap(struct file *file, struct vm_area_struct *vma) +{ + struct inode *inode; + + inode = file->f_path.dentry->d_inode; + mutex_lock(&(REISERFS_I(inode)->i_mmap)); + REISERFS_I(inode)->i_flags |= i_ever_mapped; + mutex_unlock(&(REISERFS_I(inode)->i_mmap)); + + return generic_file_mmap(file, vma); +} + static void reiserfs_vfs_truncate_file(struct inode *inode) { reiserfs_truncate_file(inode, 1); @@ -1527,7 +1545,7 @@ const struct file_operations reiserfs_file_operations = { #ifdef CONFIG_COMPAT .compat_ioctl = reiserfs_compat_ioctl, #endif - .mmap = generic_file_mmap, + .mmap = reiserfs_file_mmap, .open = generic_file_open, .release = reiserfs_file_release, .fsync = reiserfs_sync_file, diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index f3d1c4a77979..9fcbfe316977 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -1125,6 +1125,7 @@ static void init_inode(struct inode *inode, struct treepath *path) REISERFS_I(inode)->i_prealloc_count = 0; REISERFS_I(inode)->i_trans_id = 0; REISERFS_I(inode)->i_jl = NULL; + mutex_init(&(REISERFS_I(inode)->i_mmap)); reiserfs_init_acl_access(inode); reiserfs_init_acl_default(inode); reiserfs_init_xattr_rwsem(inode); @@ -1832,6 +1833,7 @@ int reiserfs_new_inode(struct reiserfs_transaction_handle *th, REISERFS_I(inode)->i_attrs = REISERFS_I(dir)->i_attrs & REISERFS_INHERIT_MASK; sd_attrs_to_i_attrs(REISERFS_I(inode)->i_attrs, inode); + mutex_init(&(REISERFS_I(inode)->i_mmap)); reiserfs_init_acl_access(inode); reiserfs_init_acl_default(inode); reiserfs_init_xattr_rwsem(inode); diff --git a/include/linux/reiserfs_fs_i.h b/include/linux/reiserfs_fs_i.h index 5b3b297aa2c5..ce3663fb0101 100644 --- a/include/linux/reiserfs_fs_i.h +++ b/include/linux/reiserfs_fs_i.h @@ -25,6 +25,7 @@ typedef enum { i_link_saved_truncate_mask = 0x0020, i_has_xattr_dir = 0x0040, i_data_log = 0x0080, + i_ever_mapped = 0x0100 } reiserfs_inode_flags; struct reiserfs_inode_info { @@ -52,6 +53,7 @@ struct reiserfs_inode_info { ** flushed */ unsigned long i_trans_id; struct reiserfs_journal_list *i_jl; + struct mutex i_mmap; #ifdef CONFIG_REISERFS_FS_POSIX_ACL struct posix_acl *i_acl_access; struct posix_acl *i_acl_default; From ea112bd5493d44967b3dc44fd078be517272b044 Mon Sep 17 00:00:00 2001 From: Horms Date: Mon, 22 Jan 2007 20:40:48 -0800 Subject: [PATCH 039/326] [PATCH] Kdump documentation update: kexec-tools update Mohan Kumar suggested making kexec-tools-testing.tar.gz a link to the latest version. I have done this and this patch updates the documentation accordingly. Signed-off-by: Simon Horman Cc: Vivek Goyal Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/kdump/kdump.txt | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) diff --git a/Documentation/kdump/kdump.txt b/Documentation/kdump/kdump.txt index 5af6676a88f0..2e5b3176de1c 100644 --- a/Documentation/kdump/kdump.txt +++ b/Documentation/kdump/kdump.txt @@ -61,7 +61,12 @@ Install kexec-tools 2) Download the kexec-tools user-space package from the following URL: -http://www.kernel.org/pub/linux/kernel/people/horms/kexec-tools/kexec-tools-testing-20061214.tar.gz +http://www.kernel.org/pub/linux/kernel/people/horms/kexec-tools/kexec-tools-testing.tar.gz + +This is a symlink to the latest version, which at the time of writing is +20061214, the only release of kexec-tools-testing so far. As other versions +are made released, the older onese will remain available at +http://www.kernel.org/pub/linux/kernel/people/horms/kexec-tools/ Note: Latest kexec-tools-testing git tree is available at @@ -71,11 +76,11 @@ http://www.kernel.org/git/?p=linux/kernel/git/horms/kexec-tools-testing.git;a=su 3) Unpack the tarball with the tar command, as follows: - tar xvpzf kexec-tools-testing-20061214.tar.gz + tar xvpzf kexec-tools-testing.tar.gz -4) Change to the kexec-tools-1.101 directory, as follows: +4) Change to the kexec-tools directory, as follows: - cd kexec-tools-testing-20061214 + cd kexec-tools-testing-VERSION 5) Configure the package, as follows: From ee8bb9eae66d3d5558b685f71b52bd8bc4ba5a62 Mon Sep 17 00:00:00 2001 From: Horms Date: Mon, 22 Jan 2007 20:40:49 -0800 Subject: [PATCH 040/326] [PATCH] Kdump documentation update: ia64 portion this patch fills in the portions for ia64 kexec. Signed-off-by: Simon Horman Cc: "Zou, Nanhai" Cc: Vivek Goyal Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/kdump/kdump.txt | 36 ++++++++++++++++++++++++++++------- 1 file changed, 29 insertions(+), 7 deletions(-) diff --git a/Documentation/kdump/kdump.txt b/Documentation/kdump/kdump.txt index 2e5b3176de1c..073306818347 100644 --- a/Documentation/kdump/kdump.txt +++ b/Documentation/kdump/kdump.txt @@ -17,7 +17,7 @@ You can use common Linux commands, such as cp and scp, to copy the memory image to a dump file on the local disk, or across the network to a remote system. -Kdump and kexec are currently supported on the x86, x86_64, ppc64 and IA64 +Kdump and kexec are currently supported on the x86, x86_64, ppc64 and ia64 architectures. When the system kernel boots, it reserves a small section of memory for @@ -229,7 +229,23 @@ Dump-capture kernel config options (Arch Dependent, ppc64) Dump-capture kernel config options (Arch Dependent, ia64) ---------------------------------------------------------- -(To be filled) + +- No specific options are required to create a dump-capture kernel + for ia64, other than those specified in the arch idependent section + above. This means that it is possible to use the system kernel + as a dump-capture kernel if desired. + + The crashkernel region can be automatically placed by the system + kernel at run time. This is done by specifying the base address as 0, + or omitting it all together. + + crashkernel=256M@0 + or + crashkernel=256M + + If the start address is specified, note that the start address of the + kernel will be aligned to 64Mb, so if the start address is not then + any space below the alignment point will be wasted. Boot into System Kernel @@ -248,6 +264,10 @@ Boot into System Kernel On ppc64, use "crashkernel=128M@32M". + On ia64, 256M@256M is a generous value that typically works. + The region may be automatically placed on ia64, see the + dump-capture kernel config option notes above. + Load the Dump-capture Kernel ============================ @@ -266,7 +286,8 @@ For x86_64: For ppc64: - Use vmlinux For ia64: - (To be filled) + - Use vmlinux or vmlinuz.gz + If you are using a uncompressed vmlinux image then use following command to load dump-capture kernel. @@ -282,18 +303,19 @@ to load dump-capture kernel. --initrd= \ --append="root= " +Please note, that --args-linux does not need to be specified for ia64. +It is planned to make this a no-op on that architecture, but for now +it should be omitted + Following are the arch specific command line options to be used while loading dump-capture kernel. -For i386 and x86_64: +For i386, x86_64 and ia64: "init 1 irqpoll maxcpus=1" For ppc64: "init 1 maxcpus=1 noirqdistrib" -For IA64 - (To be filled) - Notes on loading the dump-capture kernel: From 3a7122923e87fc5cdf8affa1845924a0def4657d Mon Sep 17 00:00:00 2001 From: Jeff Chua Date: Mon, 22 Jan 2007 20:40:50 -0800 Subject: [PATCH 041/326] [PATCH] acpi: remove "video device notify" message Seems to be some left-over debug code. Cc: Len Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/acpi/video.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/acpi/video.c b/drivers/acpi/video.c index 36b37d755dbc..3d54680d0333 100644 --- a/drivers/acpi/video.c +++ b/drivers/acpi/video.c @@ -1677,8 +1677,6 @@ static void acpi_video_device_notify(acpi_handle handle, u32 event, void *data) struct acpi_video_device *video_device = data; struct acpi_device *device = NULL; - - printk("video device notify\n"); if (!video_device) return; From ac8be955049dab828a68b9c68a75144832f8289f Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Sat, 20 Jan 2007 00:18:01 +0000 Subject: [PATCH 042/326] [MIPS] SMTC: Instant IPI replay. SMTC pseudo-interrupts between TCs are deferred and queued if the target TC is interrupt-inhibited (IXMT). In the first SMTC prototypes, these queued IPIs were serviced on return to user mode, or on entry into the kernel idle loop. The INSTANT_REPLAY option dispatches them as part of local_irq_restore() processing, which adds runtime overhead (hence the option to turn it off), but ensures that IPIs are handled promptly even under heavy I/O interrupt load. Signed-off-by: Ralf Baechle --- arch/mips/Kconfig | 14 ++++++++++ arch/mips/kernel/smtc.c | 56 ++++++++++++++++++++++--------------- include/asm-mips/irqflags.h | 22 +++++++++++++++ 3 files changed, 70 insertions(+), 22 deletions(-) diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index fd2ff0698a85..bbd386f572d9 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -1568,6 +1568,20 @@ config MIPS_MT_FPAFF depends on MIPS_MT default y +config MIPS_MT_SMTC_INSTANT_REPLAY + bool "Low-latency Dispatch of Deferred SMTC IPIs" + depends on MIPS_MT_SMTC + default y + help + SMTC pseudo-interrupts between TCs are deferred and queued + if the target TC is interrupt-inhibited (IXMT). In the first + SMTC prototypes, these queued IPIs were serviced on return + to user mode, or on entry into the kernel idle loop. The + INSTANT_REPLAY option dispatches them as part of local_irq_restore() + processing, which adds runtime overhead (hence the option to turn + it off), but ensures that IPIs are handled promptly even under + heavy I/O interrupt load. + config MIPS_VPE_LOADER_TOM bool "Load VPE program into memory hidden from linux" depends on MIPS_VPE_LOADER diff --git a/arch/mips/kernel/smtc.c b/arch/mips/kernel/smtc.c index a8b387197d5b..44238ab2fc99 100644 --- a/arch/mips/kernel/smtc.c +++ b/arch/mips/kernel/smtc.c @@ -1017,6 +1017,33 @@ void setup_cross_vpe_interrupts(void) * SMTC-specific hacks invoked from elsewhere in the kernel. */ +void smtc_ipi_replay(void) +{ + /* + * To the extent that we've ever turned interrupts off, + * we may have accumulated deferred IPIs. This is subtle. + * If we use the smtc_ipi_qdepth() macro, we'll get an + * exact number - but we'll also disable interrupts + * and create a window of failure where a new IPI gets + * queued after we test the depth but before we re-enable + * interrupts. So long as IXMT never gets set, however, + * we should be OK: If we pick up something and dispatch + * it here, that's great. If we see nothing, but concurrent + * with this operation, another TC sends us an IPI, IXMT + * is clear, and we'll handle it as a real pseudo-interrupt + * and not a pseudo-pseudo interrupt. + */ + if (IPIQ[smp_processor_id()].depth > 0) { + struct smtc_ipi *pipi; + extern void self_ipi(struct smtc_ipi *); + + while ((pipi = smtc_ipi_dq(&IPIQ[smp_processor_id()]))) { + self_ipi(pipi); + smtc_cpu_stats[smp_processor_id()].selfipis++; + } + } +} + void smtc_idle_loop_hook(void) { #ifdef SMTC_IDLE_HOOK_DEBUG @@ -1113,29 +1140,14 @@ void smtc_idle_loop_hook(void) if (pdb_msg != &id_ho_db_msg[0]) printk("CPU%d: %s", smp_processor_id(), id_ho_db_msg); #endif /* SMTC_IDLE_HOOK_DEBUG */ - /* - * To the extent that we've ever turned interrupts off, - * we may have accumulated deferred IPIs. This is subtle. - * If we use the smtc_ipi_qdepth() macro, we'll get an - * exact number - but we'll also disable interrupts - * and create a window of failure where a new IPI gets - * queued after we test the depth but before we re-enable - * interrupts. So long as IXMT never gets set, however, - * we should be OK: If we pick up something and dispatch - * it here, that's great. If we see nothing, but concurrent - * with this operation, another TC sends us an IPI, IXMT - * is clear, and we'll handle it as a real pseudo-interrupt - * and not a pseudo-pseudo interrupt. - */ - if (IPIQ[smp_processor_id()].depth > 0) { - struct smtc_ipi *pipi; - extern void self_ipi(struct smtc_ipi *); - if ((pipi = smtc_ipi_dq(&IPIQ[smp_processor_id()])) != NULL) { - self_ipi(pipi); - smtc_cpu_stats[smp_processor_id()].selfipis++; - } - } + /* + * Replay any accumulated deferred IPIs. If "Instant Replay" + * is in use, there should never be any. + */ +#ifndef CONFIG_MIPS_MT_SMTC_INSTANT_REPLAY + smtc_ipi_replay(); +#endif /* CONFIG_MIPS_MT_SMTC_INSTANT_REPLAY */ } void smtc_soft_dump(void) diff --git a/include/asm-mips/irqflags.h b/include/asm-mips/irqflags.h index 46bf5de5ac72..af3b07dfad4b 100644 --- a/include/asm-mips/irqflags.h +++ b/include/asm-mips/irqflags.h @@ -15,6 +15,27 @@ #include +/* + * CONFIG_MIPS_MT_SMTC_INSTANT_REPLAY does prompt replay of deferred IPIs, + * at the cost of branch and call overhead on each local_irq_restore() + */ + +#ifdef CONFIG_MIPS_MT_SMTC_INSTANT_REPLAY + +extern void smtc_ipi_replay(void); + +#define irq_restore_epilog(flags) \ +do { \ + if (!(flags & 0x0400)) \ + smtc_ipi_replay(); \ +} while (0) + +#else + +#define irq_restore_epilog(ignore) do { } while (0) + +#endif /* CONFIG_MIPS_MT_SMTC_INSTANT_REPLAY */ + __asm__ ( " .macro raw_local_irq_enable \n" " .set push \n" @@ -193,6 +214,7 @@ do { \ : "=r" (__tmp1) \ : "0" (flags) \ : "memory"); \ + irq_restore_epilog(flags); \ } while(0) static inline int raw_irqs_disabled_flags(unsigned long flags) From 364ca8a897eadb2f0e76b7f0ffe94168f6d83d66 Mon Sep 17 00:00:00 2001 From: Yoichi Yuasa Date: Mon, 22 Jan 2007 23:01:06 +0900 Subject: [PATCH 043/326] [MIPS] Vr41xx: Fix after GENERIC_HARDIRQS_NO__DO_IRQ change Signed-off-by: Yoichi Yuasa Signed-off-by: Ralf Baechle --- arch/mips/vr41xx/common/irq.c | 12 +++- drivers/char/vr41xx_giu.c | 114 ++++++++++++++++------------------ 2 files changed, 61 insertions(+), 65 deletions(-) diff --git a/arch/mips/vr41xx/common/irq.c b/arch/mips/vr41xx/common/irq.c index 397ba94cd7ec..16decf4ac2f4 100644 --- a/arch/mips/vr41xx/common/irq.c +++ b/arch/mips/vr41xx/common/irq.c @@ -1,7 +1,7 @@ /* * Interrupt handing routines for NEC VR4100 series. * - * Copyright (C) 2005 Yoichi Yuasa + * Copyright (C) 2005-2007 Yoichi Yuasa * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -73,13 +73,19 @@ static void irq_dispatch(unsigned int irq) if (cascade->get_irq != NULL) { unsigned int source_irq = irq; desc = irq_desc + source_irq; - desc->chip->ack(source_irq); + if (desc->chip->mask_ack) + desc->chip->mask_ack(source_irq); + else { + desc->chip->mask(source_irq); + desc->chip->ack(source_irq); + } irq = cascade->get_irq(irq); if (irq < 0) atomic_inc(&irq_err_count); else irq_dispatch(irq); - desc->chip->end(source_irq); + if (!(desc->status & IRQ_DISABLED) && desc->chip->unmask) + desc->chip->unmask(source_irq); } else do_IRQ(irq); } diff --git a/drivers/char/vr41xx_giu.c b/drivers/char/vr41xx_giu.c index a744dad9cf45..0cea8d4907df 100644 --- a/drivers/char/vr41xx_giu.c +++ b/drivers/char/vr41xx_giu.c @@ -3,7 +3,7 @@ * * Copyright (C) 2002 MontaVista Software Inc. * Author: Yoichi Yuasa - * Copyright (C) 2003-2005 Yoichi Yuasa + * Copyright (C) 2003-2007 Yoichi Yuasa * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -125,30 +125,17 @@ static inline uint16_t giu_clear(uint16_t offset, uint16_t clear) return data; } -static unsigned int startup_giuint_low_irq(unsigned int irq) +static void ack_giuint_low(unsigned int irq) { - unsigned int pin; - - pin = GPIO_PIN_OF_IRQ(irq); - giu_write(GIUINTSTATL, 1 << pin); - giu_set(GIUINTENL, 1 << pin); - - return 0; + giu_write(GIUINTSTATL, 1 << GPIO_PIN_OF_IRQ(irq)); } -static void shutdown_giuint_low_irq(unsigned int irq) +static void mask_giuint_low(unsigned int irq) { giu_clear(GIUINTENL, 1 << GPIO_PIN_OF_IRQ(irq)); } -static void enable_giuint_low_irq(unsigned int irq) -{ - giu_set(GIUINTENL, 1 << GPIO_PIN_OF_IRQ(irq)); -} - -#define disable_giuint_low_irq shutdown_giuint_low_irq - -static void ack_giuint_low_irq(unsigned int irq) +static void mask_ack_giuint_low(unsigned int irq) { unsigned int pin; @@ -157,46 +144,30 @@ static void ack_giuint_low_irq(unsigned int irq) giu_write(GIUINTSTATL, 1 << pin); } -static void end_giuint_low_irq(unsigned int irq) +static void unmask_giuint_low(unsigned int irq) { - if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS))) - giu_set(GIUINTENL, 1 << GPIO_PIN_OF_IRQ(irq)); + giu_set(GIUINTENL, 1 << GPIO_PIN_OF_IRQ(irq)); } -static struct hw_interrupt_type giuint_low_irq_type = { - .typename = "GIUINTL", - .startup = startup_giuint_low_irq, - .shutdown = shutdown_giuint_low_irq, - .enable = enable_giuint_low_irq, - .disable = disable_giuint_low_irq, - .ack = ack_giuint_low_irq, - .end = end_giuint_low_irq, +static struct irq_chip giuint_low_irq_chip = { + .name = "GIUINTL", + .ack = ack_giuint_low, + .mask = mask_giuint_low, + .mask_ack = mask_ack_giuint_low, + .unmask = unmask_giuint_low, }; -static unsigned int startup_giuint_high_irq(unsigned int irq) +static void ack_giuint_high(unsigned int irq) { - unsigned int pin; - - pin = GPIO_PIN_OF_IRQ(irq) - GIUINT_HIGH_OFFSET; - giu_write(GIUINTSTATH, 1 << pin); - giu_set(GIUINTENH, 1 << pin); - - return 0; + giu_write(GIUINTSTATH, 1 << (GPIO_PIN_OF_IRQ(irq) - GIUINT_HIGH_OFFSET)); } -static void shutdown_giuint_high_irq(unsigned int irq) +static void mask_giuint_high(unsigned int irq) { giu_clear(GIUINTENH, 1 << (GPIO_PIN_OF_IRQ(irq) - GIUINT_HIGH_OFFSET)); } -static void enable_giuint_high_irq(unsigned int irq) -{ - giu_set(GIUINTENH, 1 << (GPIO_PIN_OF_IRQ(irq) - GIUINT_HIGH_OFFSET)); -} - -#define disable_giuint_high_irq shutdown_giuint_high_irq - -static void ack_giuint_high_irq(unsigned int irq) +static void mask_ack_giuint_high(unsigned int irq) { unsigned int pin; @@ -205,20 +176,17 @@ static void ack_giuint_high_irq(unsigned int irq) giu_write(GIUINTSTATH, 1 << pin); } -static void end_giuint_high_irq(unsigned int irq) +static void unmask_giuint_high(unsigned int irq) { - if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS))) - giu_set(GIUINTENH, 1 << (GPIO_PIN_OF_IRQ(irq) - GIUINT_HIGH_OFFSET)); + giu_set(GIUINTENH, 1 << (GPIO_PIN_OF_IRQ(irq) - GIUINT_HIGH_OFFSET)); } -static struct hw_interrupt_type giuint_high_irq_type = { - .typename = "GIUINTH", - .startup = startup_giuint_high_irq, - .shutdown = shutdown_giuint_high_irq, - .enable = enable_giuint_high_irq, - .disable = disable_giuint_high_irq, - .ack = ack_giuint_high_irq, - .end = end_giuint_high_irq, +static struct irq_chip giuint_high_irq_chip = { + .name = "GIUINTH", + .ack = ack_giuint_high, + .mask = mask_giuint_high, + .mask_ack = mask_ack_giuint_high, + .unmask = unmask_giuint_high, }; static int giu_get_irq(unsigned int irq) @@ -282,9 +250,15 @@ void vr41xx_set_irq_trigger(unsigned int pin, irq_trigger_t trigger, irq_signal_ break; } } + set_irq_chip_and_handler(GIU_IRQ(pin), + &giuint_low_irq_chip, + handle_edge_irq); } else { giu_clear(GIUINTTYPL, mask); giu_clear(GIUINTHTSELL, mask); + set_irq_chip_and_handler(GIU_IRQ(pin), + &giuint_low_irq_chip, + handle_level_irq); } giu_write(GIUINTSTATL, mask); } else if (pin < GIUINT_HIGH_MAX) { @@ -311,9 +285,15 @@ void vr41xx_set_irq_trigger(unsigned int pin, irq_trigger_t trigger, irq_signal_ break; } } + set_irq_chip_and_handler(GIU_IRQ(pin), + &giuint_high_irq_chip, + handle_edge_irq); } else { giu_clear(GIUINTTYPH, mask); giu_clear(GIUINTHTSELH, mask); + set_irq_chip_and_handler(GIU_IRQ(pin), + &giuint_high_irq_chip, + handle_level_irq); } giu_write(GIUINTSTATH, mask); } @@ -617,10 +597,11 @@ static const struct file_operations gpio_fops = { static int __devinit giu_probe(struct platform_device *dev) { unsigned long start, size, flags = 0; - unsigned int nr_pins = 0; + unsigned int nr_pins = 0, trigger, i, pin; struct resource *res1, *res2 = NULL; void *base; - int retval, i; + struct irq_chip *chip; + int retval; switch (current_cpu_data.cputype) { case CPU_VR4111: @@ -688,11 +669,20 @@ static int __devinit giu_probe(struct platform_device *dev) giu_write(GIUINTENL, 0); giu_write(GIUINTENH, 0); + trigger = giu_read(GIUINTTYPH) << 16; + trigger |= giu_read(GIUINTTYPL); for (i = GIU_IRQ_BASE; i <= GIU_IRQ_LAST; i++) { - if (i < GIU_IRQ(GIUINT_HIGH_OFFSET)) - irq_desc[i].chip = &giuint_low_irq_type; + pin = GPIO_PIN_OF_IRQ(i); + if (pin < GIUINT_HIGH_OFFSET) + chip = &giuint_low_irq_chip; else - irq_desc[i].chip = &giuint_high_irq_type; + chip = &giuint_high_irq_chip; + + if (trigger & (1 << pin)) + set_irq_chip_and_handler(i, chip, handle_edge_irq); + else + set_irq_chip_and_handler(i, chip, handle_level_irq); + } return cascade_irq(GIUINT_IRQ, giu_get_irq); From 95543179f158b4891c5dc49004853ce081e8d794 Mon Sep 17 00:00:00 2001 From: Linas Vepstas Date: Tue, 23 Jan 2007 19:40:54 +0100 Subject: [PATCH 044/326] [PATCH] elevator: move clearing of unplug flag earlier A flag was recently added to the elevator code to avoid performing an unplug when reuests are being re-queued. The goal of this flag was to avoid a deep recursion that can occur when re-queueing requests after a SCSI device/host reset. See http://lkml.org/lkml/2006/5/17/254 However, that fix added the flag near the bottom of a case statement, where an earlier break (in an if statement) could transport one out of the case, without setting the flag. This patch sets the flag earlier in the case statement. I re-discovered the deep recursion recently during testing; I was told that it was a known problem, and the fix to it was in the kernel I was testing. Indeed it was ... but it didn't fix the bug. With the patch below, I no longer see the bug. Signed-off by: Linas Vepstas Signed-off-by: Jens Axboe Cc: Chris Wright Signed-off-by: Linus Torvalds --- block/elevator.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/block/elevator.c b/block/elevator.c index 536be740ba4e..f6dafa8c7c4d 100644 --- a/block/elevator.c +++ b/block/elevator.c @@ -590,6 +590,12 @@ void elv_insert(request_queue_t *q, struct request *rq, int where) */ rq->cmd_flags |= REQ_SOFTBARRIER; + /* + * Most requeues happen because of a busy condition, + * don't force unplug of the queue for that case. + */ + unplug_it = 0; + if (q->ordseq == 0) { list_add(&rq->queuelist, &q->queue_head); break; @@ -604,11 +610,6 @@ void elv_insert(request_queue_t *q, struct request *rq, int where) } list_add_tail(&rq->queuelist, pos); - /* - * most requeues happen because of a busy condition, don't - * force unplug of the queue for that case. - */ - unplug_it = 0; break; default: From 1b5180b65122666a36a1a232b7b9b38b21a9dcdd Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 23 Jan 2007 10:45:50 +0100 Subject: [PATCH 045/326] [PATCH] notifiers: fix blocking_notifier_call_chain() scalability while lock-profiling the -rt kernel i noticed weird contention during mmap-intense workloads, and the tracer showed the following gem, in one of our MM hotpaths: threaded-2771 1.... 65us : sys_munmap (sysenter_do_call) threaded-2771 1.... 66us : profile_munmap (sys_munmap) threaded-2771 1.... 66us : blocking_notifier_call_chain (profile_munmap) threaded-2771 1.... 66us : rt_down_read (blocking_notifier_call_chain) ouch! a global rw-semaphore taken in one of the most performance- sensitive codepaths of the kernel. And i dont even have oprofile enabled! All distro kernels have CONFIG_PROFILING enabled, so this scalability problem affects the majority of Linux users. The fix is to enhance blocking_notifier_call_chain() to only take the lock if there appears to be work on the call-chain. With this patch applied i get nicely saturated system, and much higher munmap performance, on SMP systems. And as a bonus this also fixes a similar scalability bottleneck in the thread-exit codepath: profile_task_exit() ... Signed-off-by: Ingo Molnar Acked-by: Peter Zijlstra Acked-by: Nick Piggin Signed-off-by: Linus Torvalds --- kernel/sys.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) diff --git a/kernel/sys.c b/kernel/sys.c index c7675c1bfdf2..6e2101dec0fc 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -323,11 +323,18 @@ EXPORT_SYMBOL_GPL(blocking_notifier_chain_unregister); int blocking_notifier_call_chain(struct blocking_notifier_head *nh, unsigned long val, void *v) { - int ret; + int ret = NOTIFY_DONE; - down_read(&nh->rwsem); - ret = notifier_call_chain(&nh->head, val, v); - up_read(&nh->rwsem); + /* + * We check the head outside the lock, but if this access is + * racy then it does not matter what the result of the test + * is, we re-check the list after having taken the lock anyway: + */ + if (rcu_dereference(nh->head)) { + down_read(&nh->rwsem); + ret = notifier_call_chain(&nh->head, val, v); + up_read(&nh->rwsem); + } return ret; } From c3ea6729feb304e0c3be74e8eca001215e78d1bc Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 23 Jan 2007 12:25:08 +0000 Subject: [PATCH 046/326] [PATCH] funsoft: ktermios fix Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- drivers/usb/serial/funsoft.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/serial/funsoft.c b/drivers/usb/serial/funsoft.c index 31501c9361b9..2bebd63d5ed1 100644 --- a/drivers/usb/serial/funsoft.c +++ b/drivers/usb/serial/funsoft.c @@ -27,7 +27,7 @@ MODULE_DEVICE_TABLE(usb, id_table); static int funsoft_ioctl(struct usb_serial_port *port, struct file *file, unsigned int cmd, unsigned long arg) { - struct termios t; + struct ktermios t; dbg("%s - port %d, cmd 0x%04x", __FUNCTION__, port->number, cmd); From 0a3c4bdc1b197a7d37fc75643a68daf45fe0a7cc Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 23 Jan 2007 12:25:08 +0000 Subject: [PATCH 047/326] [PATCH] horizon.c: missing __devinit Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- drivers/atm/horizon.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/atm/horizon.c b/drivers/atm/horizon.c index 4dc10105d610..f96446c358ba 100644 --- a/drivers/atm/horizon.c +++ b/drivers/atm/horizon.c @@ -1845,7 +1845,7 @@ static u16 __devinit read_bia (const hrz_dev * dev, u16 addr) /********** initialise a card **********/ -static int __init hrz_init (hrz_dev * dev) { +static int __devinit hrz_init (hrz_dev * dev) { int onefivefive; u16 chan; From 4384247b6910df91049f8d0bbd5c1075898ac290 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 23 Jan 2007 12:25:08 +0000 Subject: [PATCH 048/326] [PATCH] s2io bogus memset memset() after kmalloc() on size * 8 would better be on size * 8, not just size; fixed by switching to kcalloc() - it's more idiomatic anyway. Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- drivers/net/s2io.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/net/s2io.c b/drivers/net/s2io.c index 250cdbeefdfd..1dd66b8ea0fa 100644 --- a/drivers/net/s2io.c +++ b/drivers/net/s2io.c @@ -556,10 +556,9 @@ static int init_shared_mem(struct s2io_nic *nic) } } - nic->ufo_in_band_v = kmalloc((sizeof(u64) * size), GFP_KERNEL); + nic->ufo_in_band_v = kcalloc(size, sizeof(u64), GFP_KERNEL); if (!nic->ufo_in_band_v) return -ENOMEM; - memset(nic->ufo_in_band_v, 0, size); /* Allocation and initialization of RXDs in Rings */ size = 0; From 73f66ace34e3d935d1ad01208234f8871ac1f500 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 23 Jan 2007 12:27:04 +0000 Subject: [PATCH 049/326] [PATCH] fix prototype of csum_ipv6_magic() (ia64) Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- include/asm-ia64/checksum.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/include/asm-ia64/checksum.h b/include/asm-ia64/checksum.h index 2b78582cbd61..97af155057e4 100644 --- a/include/asm-ia64/checksum.h +++ b/include/asm-ia64/checksum.h @@ -72,8 +72,8 @@ static inline __sum16 csum_fold(__wsum csum) #define _HAVE_ARCH_IPV6_CSUM 1 struct in6_addr; -extern unsigned short int csum_ipv6_magic(struct in6_addr *saddr, - struct in6_addr *daddr, __u32 len, unsigned short proto, - unsigned int csum); +extern __sum16 csum_ipv6_magic(const struct in6_addr *saddr, + const struct in6_addr *daddr, __u32 len, unsigned short proto, + __wsum csum); #endif /* _ASM_IA64_CHECKSUM_H */ From f66d45e99eb7ca91822c3e3f6d7a98843c9626cb Mon Sep 17 00:00:00 2001 From: Guy Streeter Date: Tue, 23 Jan 2007 12:20:04 -0600 Subject: [PATCH 050/326] [PATCH] correct sys_shmget allocation check As written, sys_shmget will return ENOSPC when one page is still available for allocation. This patch corrects the test. Signed-off-by: Guy Streeter Signed-off-by: Linus Torvalds -- --- ipc/shm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/ipc/shm.c b/ipc/shm.c index 6d16bb6de7d2..f8e10a25ad7d 100644 --- a/ipc/shm.c +++ b/ipc/shm.c @@ -279,7 +279,7 @@ static int newseg (struct ipc_namespace *ns, key_t key, int shmflg, size_t size) if (size < SHMMIN || size > ns->shm_ctlmax) return -EINVAL; - if (ns->shm_tot + numpages >= ns->shm_ctlall) + if (ns->shm_tot + numpages > ns->shm_ctlall) return -ENOSPC; shp = ipc_rcu_alloc(sizeof(*shp)); From 3196265e97c661d9e9be04d5d64ca8ff3186536d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 23 Jan 2007 12:25:08 +0000 Subject: [PATCH 051/326] s2io bogus memset memset() after kmalloc() on size * 8 would better be on size * 8, not just size; fixed by switching to kcalloc() - it's more idiomatic anyway. Signed-off-by: Al Viro Signed-off-by: Jeff Garzik --- drivers/net/s2io.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/net/s2io.c b/drivers/net/s2io.c index 250cdbeefdfd..1dd66b8ea0fa 100644 --- a/drivers/net/s2io.c +++ b/drivers/net/s2io.c @@ -556,10 +556,9 @@ static int init_shared_mem(struct s2io_nic *nic) } } - nic->ufo_in_band_v = kmalloc((sizeof(u64) * size), GFP_KERNEL); + nic->ufo_in_band_v = kcalloc(size, sizeof(u64), GFP_KERNEL); if (!nic->ufo_in_band_v) return -ENOMEM; - memset(nic->ufo_in_band_v, 0, size); /* Allocation and initialization of RXDs in Rings */ size = 0; From d344bff9c36db17dc4765215495aaa7212c1eb6c Mon Sep 17 00:00:00 2001 From: Dale Farnsworth Date: Tue, 23 Jan 2007 09:52:25 -0700 Subject: [PATCH 052/326] mv643xx_eth: Fix race condition in mv643xx_eth_free_tx_descs mv643xx_eth: Fix race condition in mv643xx_eth_free_tx_descs This bug was found and isolated by Thibaut VARENE and Jarek Poplawski . This patch is a modification of their fixes. We acquire and release the lock for each descriptor that is freed to minimize the time the lock is held. Signed-off-by: Jeff Garzik --- drivers/net/mv643xx_eth.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c index c41ae4286eea..b3bf86422734 100644 --- a/drivers/net/mv643xx_eth.c +++ b/drivers/net/mv643xx_eth.c @@ -314,6 +314,13 @@ int mv643xx_eth_free_tx_descs(struct net_device *dev, int force) while (mp->tx_desc_count > 0) { spin_lock_irqsave(&mp->lock, flags); + + /* tx_desc_count might have changed before acquiring the lock */ + if (mp->tx_desc_count <= 0) { + spin_unlock_irqrestore(&mp->lock, flags); + return released; + } + tx_index = mp->tx_used_desc_q; desc = &mp->p_tx_desc_area[tx_index]; cmd_sts = desc->cmd_sts; @@ -332,13 +339,13 @@ int mv643xx_eth_free_tx_descs(struct net_device *dev, int force) if (skb) mp->tx_skb[tx_index] = NULL; - spin_unlock_irqrestore(&mp->lock, flags); - if (cmd_sts & ETH_ERROR_SUMMARY) { printk("%s: Error in TX\n", dev->name); mp->stats.tx_errors++; } + spin_unlock_irqrestore(&mp->lock, flags); + if (cmd_sts & ETH_TX_FIRST_DESC) dma_unmap_single(NULL, addr, count, DMA_TO_DEVICE); else From 8528b0f1de1101c6002036fd53638fb21111d0ea Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 23 Jan 2007 14:16:31 -0800 Subject: [PATCH 053/326] Clear spurious irq stat information when adding irq handler Any newly added irq handler may obviously make any old spurious irq status invalid, since the new handler may well be the thing that is supposed to handle any interrupts that came in. So just clear the statistics when adding handlers. Pointed-out-by: Alan Cox Acked-by: Thomas Gleixner Acked-by: Ingo Molnar Signed-off-by: Linus Torvalds --- kernel/irq/manage.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index b385878c6e80..8b961adc3bd2 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -315,6 +315,9 @@ int setup_irq(unsigned int irq, struct irqaction *new) /* Undo nested disables: */ desc->depth = 1; } + /* Reset broken irq detection when installing new handler */ + desc->irq_count = 0; + desc->irqs_unhandled = 0; spin_unlock_irqrestore(&desc->lock, flags); new->irq = irq; From 65ebe63420eae40fba73d3b4f79f99adc8e148b3 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 23 Jan 2007 11:38:57 -0800 Subject: [PATCH 054/326] [PATCH] email change for shemminger@osdl.org Change my email address to reflect OSDL merger. Signed-off-by: Stephen Hemminger [ The irony. Somebody still has his sign-off message hardcoded in a script or his brainstem ;^] Signed-off-by: Linus Torvalds --- MAINTAINERS | 6 +++--- drivers/net/irda/stir4200.c | 2 +- drivers/net/skge.c | 2 +- drivers/net/sky2.c | 2 +- net/ipv4/tcp_probe.c | 2 +- 5 files changed, 7 insertions(+), 7 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index b0e33617273d..d6f04a81f761 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1254,7 +1254,7 @@ S: Maintained ETHERNET BRIDGE P: Stephen Hemminger -M: shemminger@osdl.org +M: shemminger@linux-foundation.org L: bridge@osdl.org W: http://bridge.sourceforge.net/ S: Maintained @@ -2277,7 +2277,7 @@ S: Maintained NETEM NETWORK EMULATOR P: Stephen Hemminger -M: shemminger@osdl.org +M: shemminger@linux-foundation.org L: netem@osdl.org S: Maintained @@ -3081,7 +3081,7 @@ S: Maintained SKGE, SKY2 10/100/1000 GIGABIT ETHERNET DRIVERS P: Stephen Hemminger -M: shemminger@osdl.org +M: shemminger@linux-foundation.org L: netdev@vger.kernel.org S: Maintained diff --git a/drivers/net/irda/stir4200.c b/drivers/net/irda/stir4200.c index c14a74634fd5..20d306fea4cb 100644 --- a/drivers/net/irda/stir4200.c +++ b/drivers/net/irda/stir4200.c @@ -59,7 +59,7 @@ #include #include -MODULE_AUTHOR("Stephen Hemminger "); +MODULE_AUTHOR("Stephen Hemminger "); MODULE_DESCRIPTION("IrDA-USB Dongle Driver for SigmaTel STIr4200"); MODULE_LICENSE("GPL"); diff --git a/drivers/net/skge.c b/drivers/net/skge.c index deedfd5f8226..45283f3f95e4 100644 --- a/drivers/net/skge.c +++ b/drivers/net/skge.c @@ -60,7 +60,7 @@ #define LINK_HZ (HZ/2) MODULE_DESCRIPTION("SysKonnect Gigabit Ethernet driver"); -MODULE_AUTHOR("Stephen Hemminger "); +MODULE_AUTHOR("Stephen Hemminger "); MODULE_LICENSE("GPL"); MODULE_VERSION(DRV_VERSION); diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index a6601e8d423c..a2e804ddca6c 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -3691,6 +3691,6 @@ module_init(sky2_init_module); module_exit(sky2_cleanup_module); MODULE_DESCRIPTION("Marvell Yukon 2 Gigabit Ethernet driver"); -MODULE_AUTHOR("Stephen Hemminger "); +MODULE_AUTHOR("Stephen Hemminger "); MODULE_LICENSE("GPL"); MODULE_VERSION(DRV_VERSION); diff --git a/net/ipv4/tcp_probe.c b/net/ipv4/tcp_probe.c index f230eeecf092..41c157848181 100644 --- a/net/ipv4/tcp_probe.c +++ b/net/ipv4/tcp_probe.c @@ -30,7 +30,7 @@ #include -MODULE_AUTHOR("Stephen Hemminger "); +MODULE_AUTHOR("Stephen Hemminger "); MODULE_DESCRIPTION("TCP cwnd snooper"); MODULE_LICENSE("GPL"); From 99ddcc7edbb6b1d35a542f6bf0383f1f5a1aa0b0 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 23 Jan 2007 14:22:35 -0800 Subject: [PATCH 055/326] Change Linus' email address too This changes a few mentions of my email address to point to the new one, leaving things like old copyright messages alone. Signed-off-by: Linus Torvalds --- CREDITS | 2 +- Documentation/SubmittingPatches | 6 +++--- Documentation/usb/CREDITS | 2 +- README | 4 ++-- 4 files changed, 7 insertions(+), 7 deletions(-) diff --git a/CREDITS b/CREDITS index 75c5ce82720c..ae08e4c10ed4 100644 --- a/CREDITS +++ b/CREDITS @@ -3279,7 +3279,7 @@ S: Sevilla 41005 S: Spain N: Linus Torvalds -E: torvalds@osdl.org +E: torvalds@linux-foundation.org D: Original kernel hacker S: 12725 SW Millikan Way, Suite 400 S: Beaverton, Oregon 97005 diff --git a/Documentation/SubmittingPatches b/Documentation/SubmittingPatches index 302d148c2e18..b0d0043f7c46 100644 --- a/Documentation/SubmittingPatches +++ b/Documentation/SubmittingPatches @@ -134,9 +134,9 @@ Do not send more than 15 patches at once to the vger mailing lists!!! Linus Torvalds is the final arbiter of all changes accepted into the -Linux kernel. His e-mail address is . He gets -a lot of e-mail, so typically you should do your best to -avoid- sending -him e-mail. +Linux kernel. His e-mail address is . +He gets a lot of e-mail, so typically you should do your best to -avoid- +sending him e-mail. Patches which are bug fixes, are "obvious" changes, or similarly require little discussion should be sent or CC'd to Linus. Patches diff --git a/Documentation/usb/CREDITS b/Documentation/usb/CREDITS index 01e7f857ef35..27a721635f92 100644 --- a/Documentation/usb/CREDITS +++ b/Documentation/usb/CREDITS @@ -21,7 +21,7 @@ difficult to maintain, add yourself with a patch if desired. Bill Ryder Thomas Sailer Gregory P. Smith - Linus Torvalds + Linus Torvalds Roman Weissgaerber diff --git a/README b/README index c05561523029..46a66c6e76df 100644 --- a/README +++ b/README @@ -278,8 +278,8 @@ IF SOMETHING GOES WRONG: the file MAINTAINERS to see if there is a particular person associated with the part of the kernel that you are having trouble with. If there isn't anyone listed there, then the second best thing is to mail - them to me (torvalds@osdl.org), and possibly to any other relevant - mailing-list or to the newsgroup. + them to me (torvalds@linux-foundation.org), and possibly to any other + relevant mailing-list or to the newsgroup. - In all bug-reports, *please* tell what kernel you are talking about, how to duplicate the problem, and what your setup is (use your common From 419dd8378dfa32985672ab7927b4bc827f33b332 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Tue, 23 Jan 2007 20:04:13 -0300 Subject: [PATCH 056/326] V4L/DVB (5123): Buf_qbuf: fix: videobuf_queue->stream corruption and lockup We are doing ->buf_prepare(buf) before adding buf to q->stream list. This means that videobuf_qbuf() should not try to re-add a STATE_PREPARED buffer. Signed-off-by: Oleg Nesterov Signed-off-by: Mauro Carvalho Chehab --- drivers/media/video/video-buf.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/media/video/video-buf.c b/drivers/media/video/video-buf.c index 635d102c86f0..6504a5866849 100644 --- a/drivers/media/video/video-buf.c +++ b/drivers/media/video/video-buf.c @@ -700,6 +700,7 @@ videobuf_qbuf(struct videobuf_queue *q, goto done; } if (buf->state == STATE_QUEUED || + buf->state == STATE_PREPARED || buf->state == STATE_ACTIVE) { dprintk(1,"qbuf: buffer is already queued or active.\n"); goto done; From e0e8f1c8220c43bdf25cfb5622f6ab6947027fb1 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Wed, 10 Jan 2007 22:06:32 -0800 Subject: [PATCH 057/326] [IPSEC] flow: Fix potential memory leak When old flow cache entries that are not at the head of their chain trigger a transient security error they get unlinked along with all the entries preceding them in the chain. The preceding entries are not freed correctly. This patch fixes this by simply leaving the entry around. It's based on a suggestion by Venkat Yekkirala. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- net/core/flow.c | 22 ++++++++-------------- 1 file changed, 8 insertions(+), 14 deletions(-) diff --git a/net/core/flow.c b/net/core/flow.c index d137f971f97d..5d25697920b1 100644 --- a/net/core/flow.c +++ b/net/core/flow.c @@ -231,22 +231,16 @@ void *flow_cache_lookup(struct flowi *key, u16 family, u8 dir, err = resolver(key, family, dir, &obj, &obj_ref); - if (fle) { - if (err) { - /* Force security policy check on next lookup */ - *head = fle->next; - flow_entry_kill(cpu, fle); - } else { - fle->genid = atomic_read(&flow_cache_genid); + if (fle && !err) { + fle->genid = atomic_read(&flow_cache_genid); - if (fle->object) - atomic_dec(fle->object_ref); + if (fle->object) + atomic_dec(fle->object_ref); - fle->object = obj; - fle->object_ref = obj_ref; - if (obj) - atomic_inc(fle->object_ref); - } + fle->object = obj; + fle->object_ref = obj_ref; + if (obj) + atomic_inc(fle->object_ref); } local_bh_enable(); From d88ae4cc97b24783ee4480697fbdcc02ab4133a6 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Sun, 14 Jan 2007 21:48:40 -0800 Subject: [PATCH 058/326] [IPV6] MCAST: Fix joining all-node multicast group on device initialization. Join all-node multicast group after assignment of dev->ip6_ptr because it must be assigned when ipv6_dev_mc_inc() is called. This fixes Bug#7817, reported by . Closes: 7817 Signed-off-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller --- net/ipv6/addrconf.c | 6 ++++++ net/ipv6/mcast.c | 6 ------ 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 171e5b55d7d6..2a7e4618f526 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -341,6 +341,7 @@ void in6_dev_finish_destroy(struct inet6_dev *idev) static struct inet6_dev * ipv6_add_dev(struct net_device *dev) { struct inet6_dev *ndev; + struct in6_addr maddr; ASSERT_RTNL(); @@ -425,6 +426,11 @@ static struct inet6_dev * ipv6_add_dev(struct net_device *dev) #endif /* protected by rtnl_lock */ rcu_assign_pointer(dev->ip6_ptr, ndev); + + /* Join all-node multicast group */ + ipv6_addr_all_nodes(&maddr); + ipv6_dev_mc_inc(dev, &maddr); + return ndev; } diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index a1c231a04ac2..882cde4b4047 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -2258,8 +2258,6 @@ void ipv6_mc_up(struct inet6_dev *idev) void ipv6_mc_init_dev(struct inet6_dev *idev) { - struct in6_addr maddr; - write_lock_bh(&idev->lock); rwlock_init(&idev->mc_lock); idev->mc_gq_running = 0; @@ -2275,10 +2273,6 @@ void ipv6_mc_init_dev(struct inet6_dev *idev) idev->mc_maxdelay = IGMP6_UNSOLICITED_IVAL; idev->mc_v1_seen = 0; write_unlock_bh(&idev->lock); - - /* Add all-nodes address. */ - ipv6_addr_all_nodes(&maddr); - ipv6_dev_mc_inc(idev->dev, &maddr); } /* From 334c85569b8adeaa820c0f2fab3c8f0a9dc8b92e Mon Sep 17 00:00:00 2001 From: Venkat Yekkirala Date: Mon, 15 Jan 2007 16:38:45 -0800 Subject: [PATCH 059/326] [SELINUX]: increment flow cache genid Currently, old flow cache entries remain valid even after a reload of SELinux policy. This patch increments the flow cache generation id on policy (re)loads so that flow cache entries are revalidated as needed. Thanks to Herbet Xu for pointing this out. See: http://marc.theaimsgroup.com/?l=linux-netdev&m=116841378704536&w=2 There's also a general issue as well as a solution proposed by David Miller for when flow_cache_genid wraps. I might be submitting a separate patch for that later. I request that this be applied to 2.6.20 since it's a security relevant fix. Signed-off-by: Venkat Yekkirala Signed-off-by: David S. Miller --- security/selinux/ss/services.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c index 3eb1fa9f0de1..ff0393317f39 100644 --- a/security/selinux/ss/services.c +++ b/security/selinux/ss/services.c @@ -1299,6 +1299,7 @@ int security_load_policy(void *data, size_t len) avc_ss_reset(seqno); selnl_notify_policyload(seqno); selinux_netlbl_cache_invalidate(); + atomic_inc(&flow_cache_genid); return 0; } @@ -1354,6 +1355,7 @@ int security_load_policy(void *data, size_t len) avc_ss_reset(seqno); selnl_notify_policyload(seqno); selinux_netlbl_cache_invalidate(); + atomic_inc(&flow_cache_genid); return 0; @@ -1853,6 +1855,7 @@ int security_set_bools(int len, int *values) if (!rc) { avc_ss_reset(seqno); selnl_notify_policyload(seqno); + atomic_inc(&flow_cache_genid); } return rc; } From c54ea3b95ac504ed81e0ec3acfaa26d0f55bdfa4 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Mon, 15 Jan 2007 17:16:03 -0800 Subject: [PATCH 060/326] [NETFILTER]: ctnetlink: fix leak in ctnetlink_create_conntrack error path Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/ipv4/netfilter/ip_conntrack_netlink.c | 2 +- net/netfilter/nf_conntrack_netlink.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/net/ipv4/netfilter/ip_conntrack_netlink.c b/net/ipv4/netfilter/ip_conntrack_netlink.c index 5fcf91d617cd..6f31fad9be13 100644 --- a/net/ipv4/netfilter/ip_conntrack_netlink.c +++ b/net/ipv4/netfilter/ip_conntrack_netlink.c @@ -959,7 +959,7 @@ ctnetlink_create_conntrack(struct nfattr *cda[], if (cda[CTA_PROTOINFO-1]) { err = ctnetlink_change_protoinfo(ct, cda); if (err < 0) - return err; + goto err; } #if defined(CONFIG_IP_NF_CONNTRACK_MARK) diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index bd1d2de75e45..811e3e782f0f 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -981,7 +981,7 @@ ctnetlink_create_conntrack(struct nfattr *cda[], if (cda[CTA_PROTOINFO-1]) { err = ctnetlink_change_protoinfo(ct, cda); if (err < 0) - return err; + goto err; } #if defined(CONFIG_NF_CONNTRACK_MARK) From 16d807988ffaf9b7cbb1966955aa8f738c32e740 Mon Sep 17 00:00:00 2001 From: Mikael Pettersson Date: Mon, 15 Jan 2007 17:17:31 -0800 Subject: [PATCH 061/326] [NETFILTER]: fix xt_state compile failure In file included from net/netfilter/xt_state.c:13: include/net/netfilter/nf_conntrack_compat.h: In function 'nf_ct_l3proto_try_module_get': include/net/netfilter/nf_conntrack_compat.h:70: error: 'PF_INET' undeclared (first use in this function) include/net/netfilter/nf_conntrack_compat.h:70: error: (Each undeclared identifier is reported only once include/net/netfilter/nf_conntrack_compat.h:70: error: for each function it appears in.) include/net/netfilter/nf_conntrack_compat.h:71: warning: control reaches end of non-void function make[2]: *** [net/netfilter/xt_state.o] Error 1 make[1]: *** [net/netfilter] Error 2 make: *** [net] Error 2 A simple fix is to have nf_conntrack_compat.h #include . Signed-off-by: Mikael Pettersson Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/net/netfilter/nf_conntrack_compat.h | 1 + 1 file changed, 1 insertion(+) diff --git a/include/net/netfilter/nf_conntrack_compat.h b/include/net/netfilter/nf_conntrack_compat.h index b9ce5c80d9d5..6f84c1f7fcd4 100644 --- a/include/net/netfilter/nf_conntrack_compat.h +++ b/include/net/netfilter/nf_conntrack_compat.h @@ -6,6 +6,7 @@ #if defined(CONFIG_IP_NF_CONNTRACK) || defined(CONFIG_IP_NF_CONNTRACK_MODULE) #include +#include #ifdef CONFIG_IP_NF_CONNTRACK_MARK static inline u_int32_t *nf_ct_get_mark(const struct sk_buff *skb, From ebdfcad4dc2a6851f75fac0a3315046cbd9c4410 Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Mon, 15 Jan 2007 19:12:31 -0800 Subject: [PATCH 062/326] [SCTP]: Set correct error cause value for missing parameters sctp_process_missing_param() needs to use the SCTP_ERROR_MISS_PARAM error cause value. Signed-off-by: Vlad Yasevich Signed-off-by: Sridhar Samudrala Signed-off-by: David S. Miller --- net/sctp/sm_make_chunk.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 167d888d1df2..ea0f8fac3f01 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -1562,7 +1562,7 @@ static int sctp_process_missing_param(const struct sctp_association *asoc, if (*errp) { report.num_missing = htonl(1); report.type = paramtype; - sctp_init_cause(*errp, SCTP_ERROR_INV_PARAM, + sctp_init_cause(*errp, SCTP_ERROR_MISS_PARAM, &report, sizeof(report)); } From d023f629451ace6f37eb5d2cf29ddd24497c91dc Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Mon, 15 Jan 2007 19:15:45 -0800 Subject: [PATCH 063/326] [SCTP]: Verify some mandatory parameters. Verify init_tag and a_rwnd mandatory parameters in INIT and INIT-ACK chunks. Signed-off-by: Vlad Yasevich Signed-off-by: Sridhar Samudrala Signed-off-by: David S. Miller --- net/sctp/sm_make_chunk.c | 4 +++- net/sctp/sm_statefuns.c | 19 ------------------- 2 files changed, 3 insertions(+), 20 deletions(-) diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index ea0f8fac3f01..0b1ddb1005ac 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -1775,7 +1775,9 @@ int sctp_verify_init(const struct sctp_association *asoc, /* Verify stream values are non-zero. */ if ((0 == peer_init->init_hdr.num_outbound_streams) || - (0 == peer_init->init_hdr.num_inbound_streams)) { + (0 == peer_init->init_hdr.num_inbound_streams) || + (0 == peer_init->init_hdr.init_tag) || + (SCTP_DEFAULT_MINWINDOW > ntohl(peer_init->init_hdr.a_rwnd))) { sctp_process_inv_mandatory(asoc, chunk, errp); return 0; diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index aa51d190bfb2..2c165dc9fb71 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -440,7 +440,6 @@ sctp_disposition_t sctp_sf_do_5_1C_ack(const struct sctp_endpoint *ep, { struct sctp_chunk *chunk = arg; sctp_init_chunk_t *initchunk; - __u32 init_tag; struct sctp_chunk *err_chunk; struct sctp_packet *packet; sctp_error_t error; @@ -462,24 +461,6 @@ sctp_disposition_t sctp_sf_do_5_1C_ack(const struct sctp_endpoint *ep, /* Grab the INIT header. */ chunk->subh.init_hdr = (sctp_inithdr_t *) chunk->skb->data; - init_tag = ntohl(chunk->subh.init_hdr->init_tag); - - /* Verification Tag: 3.3.3 - * If the value of the Initiate Tag in a received INIT ACK - * chunk is found to be 0, the receiver MUST treat it as an - * error and close the association by transmitting an ABORT. - */ - if (!init_tag) { - struct sctp_chunk *reply = sctp_make_abort(asoc, chunk, 0); - if (!reply) - goto nomem; - - sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(reply)); - return sctp_stop_t1_and_abort(commands, SCTP_ERROR_INV_PARAM, - ECONNREFUSED, asoc, - chunk->transport); - } - /* Verify the INIT chunk before processing it. */ err_chunk = NULL; if (!sctp_verify_init(asoc, chunk->chunk_hdr->type, From 610ab73ac4cc8912fc253bbdc6d1f74bad3c8e3a Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Mon, 15 Jan 2007 19:18:30 -0800 Subject: [PATCH 064/326] [SCTP]: Correctly handle unexpected INIT-ACK chunk. Consider the chunk as Out-of-the-Blue if we don't have an endpoint. Otherwise discard it as before. Signed-off-by: Vlad Yasevich Signed-off-by: Sridhar Samudrala Signed-off-by: David S. Miller --- include/net/sctp/sm.h | 1 + net/sctp/sm_statefuns.c | 22 ++++++++++++++++++++++ net/sctp/sm_statetable.c | 2 +- 3 files changed, 24 insertions(+), 1 deletion(-) diff --git a/include/net/sctp/sm.h b/include/net/sctp/sm.h index 3269ed1cc222..73cb9943c8a8 100644 --- a/include/net/sctp/sm.h +++ b/include/net/sctp/sm.h @@ -134,6 +134,7 @@ sctp_state_fn_t sctp_sf_violation; sctp_state_fn_t sctp_sf_discard_chunk; sctp_state_fn_t sctp_sf_do_5_2_1_siminit; sctp_state_fn_t sctp_sf_do_5_2_2_dupinit; +sctp_state_fn_t sctp_sf_do_5_2_3_initack; sctp_state_fn_t sctp_sf_do_5_2_4_dupcook; sctp_state_fn_t sctp_sf_unk_chunk; sctp_state_fn_t sctp_sf_do_8_5_1_E_sa; diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index 2c165dc9fb71..fce1f602cde2 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -1534,6 +1534,28 @@ sctp_disposition_t sctp_sf_do_5_2_2_dupinit(const struct sctp_endpoint *ep, } +/* + * Unexpected INIT-ACK handler. + * + * Section 5.2.3 + * If an INIT ACK received by an endpoint in any state other than the + * COOKIE-WAIT state, the endpoint should discard the INIT ACK chunk. + * An unexpected INIT ACK usually indicates the processing of an old or + * duplicated INIT chunk. +*/ +sctp_disposition_t sctp_sf_do_5_2_3_initack(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, sctp_cmd_seq_t *commands) +{ + /* Per the above section, we'll discard the chunk if we have an + * endpoint. If this is an OOTB INIT-ACK, treat it as such. + */ + if (ep == sctp_sk((sctp_get_ctl_sock()))->ep) + return sctp_sf_ootb(ep, asoc, type, arg, commands); + else + return sctp_sf_discard_chunk(ep, asoc, type, arg, commands); +} /* Unexpected COOKIE-ECHO handler for peer restart (Table 2, action 'A') * diff --git a/net/sctp/sm_statetable.c b/net/sctp/sm_statetable.c index 733dd87b3a7d..5f6cc7aa661b 100644 --- a/net/sctp/sm_statetable.c +++ b/net/sctp/sm_statetable.c @@ -152,7 +152,7 @@ const sctp_sm_table_entry_t *sctp_sm_lookup_event(sctp_event_t event_type, /* SCTP_STATE_EMPTY */ \ TYPE_SCTP_FUNC(sctp_sf_ootb), \ /* SCTP_STATE_CLOSED */ \ - TYPE_SCTP_FUNC(sctp_sf_discard_chunk), \ + TYPE_SCTP_FUNC(sctp_sf_do_5_2_3_initack), \ /* SCTP_STATE_COOKIE_WAIT */ \ TYPE_SCTP_FUNC(sctp_sf_do_5_1C_ack), \ /* SCTP_STATE_COOKIE_ECHOED */ \ From 732ba35e759112be5cecd79d4351084edf88dba7 Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Mon, 15 Jan 2007 19:20:21 -0800 Subject: [PATCH 065/326] [SCTP]: Fix SACK sequence during shutdown Currently, when association enters SHUTDOWN state,the implementation will SACK any DATA first and then transmit the SHUTDOWN chunk. This is against the order required by 2960bis spec. SHUTDOWN must always be first, followed by SACK. This change forces this order and also enables bundling. Signed-off-by: Vlad Yasevich Signed-off-by: Sridhar Samudrala Signed-off-by: David S. Miller --- net/sctp/sm_sideeffect.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c index 7bbc6156e455..8bd30976cdee 100644 --- a/net/sctp/sm_sideeffect.c +++ b/net/sctp/sm_sideeffect.c @@ -217,7 +217,7 @@ static int sctp_gen_sack(struct sctp_association *asoc, int force, asoc->peer.sack_needed = 0; - error = sctp_outq_tail(&asoc->outqueue, sack); + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(sack)); /* Stop the SACK timer. */ sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, From 9d0f7d29f38d4f7bf66d38d448a7d23e0bed9074 Mon Sep 17 00:00:00 2001 From: Andrew Hendry Date: Mon, 15 Jan 2007 19:29:31 -0800 Subject: [PATCH 066/326] [X.25]: Add missing sock_put in x25_receive_data __x25_find_socket does a sock_hold. This adds a missing sock_put in x25_receive_data. Signed-off-by: Andrew Hendry Signed-off-by: David S. Miller --- net/x25/x25_dev.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/x25/x25_dev.c b/net/x25/x25_dev.c index 47b68a301677..328d80f000ad 100644 --- a/net/x25/x25_dev.c +++ b/net/x25/x25_dev.c @@ -56,6 +56,7 @@ static int x25_receive_data(struct sk_buff *skb, struct x25_neigh *nb) sk_add_backlog(sk, skb); } bh_unlock_sock(sk); + sock_put(sk); return queued; } From 3958fb34ef18529c1e4a3eca44b7aaf94d4f4697 Mon Sep 17 00:00:00 2001 From: Samuel Ortiz Date: Mon, 15 Jan 2007 19:37:25 -0800 Subject: [PATCH 067/326] [IrDA]: irda-usb TX path optimization (was Re: IrDA spams logfiles - since 2.6.19) Since we stop using dev_alloc_skb on the IrDA TX frame, we constantly run into the case of the skb headroom being 0, and thus we call skb_cow for every IrDA TX frame. This patch uses a local buffer and memcpy the skb to it, saving us a kmalloc for each of those IrDA TX frames. Signed-off-by: Samuel Ortiz Signed-off-by: David S. Miller --- drivers/net/irda/irda-usb.c | 43 +++++++++++++++++-------------------- drivers/net/irda/irda-usb.h | 1 + 2 files changed, 21 insertions(+), 23 deletions(-) diff --git a/drivers/net/irda/irda-usb.c b/drivers/net/irda/irda-usb.c index 3ca1082ec776..80cbf3f80432 100644 --- a/drivers/net/irda/irda-usb.c +++ b/drivers/net/irda/irda-usb.c @@ -441,25 +441,13 @@ static int irda_usb_hard_xmit(struct sk_buff *skb, struct net_device *netdev) goto drop; } - /* Make sure there is room for IrDA-USB header. The actual - * allocation will be done lower in skb_push(). - * Also, we don't use directly skb_cow(), because it require - * headroom >= 16, which force unnecessary copies - Jean II */ - if (skb_headroom(skb) < self->header_length) { - IRDA_DEBUG(0, "%s(), Insuficient skb headroom.\n", __FUNCTION__); - if (skb_cow(skb, self->header_length)) { - IRDA_WARNING("%s(), failed skb_cow() !!!\n", __FUNCTION__); - goto drop; - } - } + memcpy(self->tx_buff + self->header_length, skb->data, skb->len); /* Change setting for next frame */ - if (self->capability & IUC_STIR421X) { __u8 turnaround_time; - __u8* frame; + __u8* frame = self->tx_buff; turnaround_time = get_turnaround_time( skb ); - frame= skb_push(skb, self->header_length); irda_usb_build_header(self, frame, 0); frame[2] = turnaround_time; if ((skb->len != 0) && @@ -472,17 +460,17 @@ static int irda_usb_hard_xmit(struct sk_buff *skb, struct net_device *netdev) frame[1] = 0; } } else { - irda_usb_build_header(self, skb_push(skb, self->header_length), 0); + irda_usb_build_header(self, self->tx_buff, 0); } /* FIXME: Make macro out of this one */ ((struct irda_skb_cb *)skb->cb)->context = self; - usb_fill_bulk_urb(urb, self->usbdev, + usb_fill_bulk_urb(urb, self->usbdev, usb_sndbulkpipe(self->usbdev, self->bulk_out_ep), - skb->data, IRDA_SKB_MAX_MTU, + self->tx_buff, skb->len + self->header_length, write_bulk_callback, skb); - urb->transfer_buffer_length = skb->len; + /* This flag (URB_ZERO_PACKET) indicates that what we send is not * a continuous stream of data but separate packets. * In this case, the USB layer will insert an empty USB frame (TD) @@ -1455,6 +1443,9 @@ static inline void irda_usb_close(struct irda_usb_cb *self) /* Remove the speed buffer */ kfree(self->speed_buff); self->speed_buff = NULL; + + kfree(self->tx_buff); + self->tx_buff = NULL; } /********************** USB CONFIG SUBROUTINES **********************/ @@ -1753,9 +1744,14 @@ static int irda_usb_probe(struct usb_interface *intf, memset(self->speed_buff, 0, IRDA_USB_SPEED_MTU); + self->tx_buff = kzalloc(IRDA_SKB_MAX_MTU + self->header_length, + GFP_KERNEL); + if (self->tx_buff == NULL) + goto err_out_4; + ret = irda_usb_open(self); if (ret) - goto err_out_4; + goto err_out_5; IRDA_MESSAGE("IrDA: Registered device %s\n", net->name); usb_set_intfdata(intf, self); @@ -1766,14 +1762,14 @@ static int irda_usb_probe(struct usb_interface *intf, self->needspatch = (ret < 0); if (self->needspatch) { IRDA_ERROR("STIR421X: Couldn't upload patch\n"); - goto err_out_5; + goto err_out_6; } /* replace IrDA class descriptor with what patched device is now reporting */ irda_desc = irda_usb_find_class_desc (self->usbintf); if (irda_desc == NULL) { ret = -ENODEV; - goto err_out_5; + goto err_out_6; } if (self->irda_desc) kfree (self->irda_desc); @@ -1782,9 +1778,10 @@ static int irda_usb_probe(struct usb_interface *intf, } return 0; - -err_out_5: +err_out_6: unregister_netdev(self->netdev); +err_out_5: + kfree(self->tx_buff); err_out_4: kfree(self->speed_buff); err_out_3: diff --git a/drivers/net/irda/irda-usb.h b/drivers/net/irda/irda-usb.h index 6b2271f18e77..e846c38224a3 100644 --- a/drivers/net/irda/irda-usb.h +++ b/drivers/net/irda/irda-usb.h @@ -156,6 +156,7 @@ struct irda_usb_cb { struct irlap_cb *irlap; /* The link layer we are binded to */ struct qos_info qos; char *speed_buff; /* Buffer for speed changes */ + char *tx_buff; struct timeval stamp; struct timeval now; From 8f1adb5f27d352c776ac34648cc277d1f8199dba Mon Sep 17 00:00:00 2001 From: Samuel Ortiz Date: Mon, 15 Jan 2007 19:40:34 -0800 Subject: [PATCH 068/326] [IrDA]: Removed incorrect IRDA_ASSERT() With USB2.0 bulk out MTU can be 512 bytes, so checking it only for 64 bytes is incorrect. Signed-off-by: Samuel Ortiz Signed-off-by: David S. Miller --- drivers/net/irda/irda-usb.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/net/irda/irda-usb.c b/drivers/net/irda/irda-usb.c index 80cbf3f80432..340ee99652eb 100644 --- a/drivers/net/irda/irda-usb.c +++ b/drivers/net/irda/irda-usb.c @@ -1515,8 +1515,6 @@ static inline int irda_usb_parse_endpoints(struct irda_usb_cb *self, struct usb_ IRDA_DEBUG(0, "%s(), And our endpoints are : in=%02X, out=%02X (%d), int=%02X\n", __FUNCTION__, self->bulk_in_ep, self->bulk_out_ep, self->bulk_out_mtu, self->bulk_int_ep); - /* Should be 8, 16, 32 or 64 bytes */ - IRDA_ASSERT(self->bulk_out_mtu == 64, ;); return((self->bulk_in_ep != 0) && (self->bulk_out_ep != 0)); } From a6c7ab55dda3e16ab5a3cf6f39585aee5876ac3a Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Tue, 16 Jan 2007 16:52:02 -0800 Subject: [PATCH 069/326] [IPSEC]: Policy list disorder The recent hashing introduced an off-by-one bug in policy list insertion. Instead of adding after the last entry with a lesser or equal priority, we're adding after the successor of that entry. This patch fixes this and also adds a warning if we detect a duplicate entry in the policy list. This should never happen due to this if clause. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- net/xfrm/xfrm_policy.c | 16 +++++----------- 1 file changed, 5 insertions(+), 11 deletions(-) diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index bebd40e5a62e..b7e537fe2d75 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -650,19 +650,18 @@ int xfrm_policy_insert(int dir, struct xfrm_policy *policy, int excl) struct xfrm_policy *pol; struct xfrm_policy *delpol; struct hlist_head *chain; - struct hlist_node *entry, *newpos, *last; + struct hlist_node *entry, *newpos; struct dst_entry *gc_list; write_lock_bh(&xfrm_policy_lock); chain = policy_hash_bysel(&policy->selector, policy->family, dir); delpol = NULL; newpos = NULL; - last = NULL; hlist_for_each_entry(pol, entry, chain, bydst) { - if (!delpol && - pol->type == policy->type && + if (pol->type == policy->type && !selector_cmp(&pol->selector, &policy->selector) && - xfrm_sec_ctx_match(pol->security, policy->security)) { + xfrm_sec_ctx_match(pol->security, policy->security) && + !WARN_ON(delpol)) { if (excl) { write_unlock_bh(&xfrm_policy_lock); return -EEXIST; @@ -671,17 +670,12 @@ int xfrm_policy_insert(int dir, struct xfrm_policy *policy, int excl) if (policy->priority > pol->priority) continue; } else if (policy->priority >= pol->priority) { - last = &pol->bydst; + newpos = &pol->bydst; continue; } - if (!newpos) - newpos = &pol->bydst; if (delpol) break; - last = &pol->bydst; } - if (!newpos) - newpos = last; if (newpos) hlist_add_after(newpos, &policy->bydst); else From fb7e2399ec17f1004c0e0ccfd17439f8759ede01 Mon Sep 17 00:00:00 2001 From: Masayuki Nakagawa Date: Tue, 23 Jan 2007 20:15:06 -0800 Subject: [PATCH 070/326] [TCP]: skb is unexpectedly freed. I encountered a kernel panic with my test program, which is a very simple IPv6 client-server program. The server side sets IPV6_RECVPKTINFO on a listening socket, and the client side just sends a message to the server. Then the kernel panic occurs on the server. (If you need the test program, please let me know. I can provide it.) This problem happens because a skb is forcibly freed in tcp_rcv_state_process(). When a socket in listening state(TCP_LISTEN) receives a syn packet, then tcp_v6_conn_request() will be called from tcp_rcv_state_process(). If the tcp_v6_conn_request() successfully returns, the skb would be discarded by __kfree_skb(). However, in case of a listening socket which was already set IPV6_RECVPKTINFO, an address of the skb will be stored in treq->pktopts and a ref count of the skb will be incremented in tcp_v6_conn_request(). But, even if the skb is still in use, the skb will be freed. Then someone still using the freed skb will cause the kernel panic. I suggest to use kfree_skb() instead of __kfree_skb(). Signed-off-by: Masayuki Nakagawa Signed-off-by: David S. Miller --- net/ipv4/tcp_input.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index c701f6abbfc1..5c16e24a6061 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -4420,9 +4420,11 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb, * But, this leaves one open to an easy denial of * service attack, and SYN cookies can't defend * against this problem. So, we drop the data - * in the interest of security over speed. + * in the interest of security over speed unless + * it's still in use. */ - goto discard; + kfree_skb(skb); + return 0; } goto discard; From 778a43fd626b710faca32038afc7460f314ba82a Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 23 Jan 2007 21:16:40 -0800 Subject: [PATCH 071/326] [IRDA] vlsi_ir.{h,c}: remove kernel 2.4 code This patch removes kernel 2.4 compatibility code. Signed-off-by: Adrian Bunk Acked-by: Samuel Ortiz Signed-off-by: David S. Miller --- drivers/net/irda/vlsi_ir.c | 16 ++++++++-------- drivers/net/irda/vlsi_ir.h | 33 --------------------------------- 2 files changed, 8 insertions(+), 41 deletions(-) diff --git a/drivers/net/irda/vlsi_ir.c b/drivers/net/irda/vlsi_ir.c index 18c68193bf14..e2b1af618450 100644 --- a/drivers/net/irda/vlsi_ir.c +++ b/drivers/net/irda/vlsi_ir.c @@ -166,7 +166,7 @@ static void vlsi_proc_pdev(struct seq_file *seq, struct pci_dev *pdev) unsigned i; seq_printf(seq, "\n%s (vid/did: %04x/%04x)\n", - PCIDEV_NAME(pdev), (int)pdev->vendor, (int)pdev->device); + pci_name(pdev), (int)pdev->vendor, (int)pdev->device); seq_printf(seq, "pci-power-state: %u\n", (unsigned) pdev->current_state); seq_printf(seq, "resources: irq=%u / io=0x%04x / dma_mask=0x%016Lx\n", pdev->irq, (unsigned)pci_resource_start(pdev, 0), (unsigned long long)pdev->dma_mask); @@ -1401,7 +1401,7 @@ static void vlsi_tx_timeout(struct net_device *ndev) if (vlsi_start_hw(idev)) IRDA_ERROR("%s: failed to restart hw - %s(%s) unusable!\n", - __FUNCTION__, PCIDEV_NAME(idev->pdev), ndev->name); + __FUNCTION__, pci_name(idev->pdev), ndev->name); else netif_start_queue(ndev); } @@ -1643,7 +1643,7 @@ vlsi_irda_probe(struct pci_dev *pdev, const struct pci_device_id *id) pdev->current_state = 0; /* hw must be running now */ IRDA_MESSAGE("%s: IrDA PCI controller %s detected\n", - drivername, PCIDEV_NAME(pdev)); + drivername, pci_name(pdev)); if ( !pci_resource_start(pdev,0) || !(pci_resource_flags(pdev,0) & IORESOURCE_IO) ) { @@ -1728,7 +1728,7 @@ static void __devexit vlsi_irda_remove(struct pci_dev *pdev) pci_set_drvdata(pdev, NULL); - IRDA_MESSAGE("%s: %s removed\n", drivername, PCIDEV_NAME(pdev)); + IRDA_MESSAGE("%s: %s removed\n", drivername, pci_name(pdev)); } #ifdef CONFIG_PM @@ -1748,7 +1748,7 @@ static int vlsi_irda_suspend(struct pci_dev *pdev, pm_message_t state) if (!ndev) { IRDA_ERROR("%s - %s: no netdevice \n", - __FUNCTION__, PCIDEV_NAME(pdev)); + __FUNCTION__, pci_name(pdev)); return 0; } idev = ndev->priv; @@ -1759,7 +1759,7 @@ static int vlsi_irda_suspend(struct pci_dev *pdev, pm_message_t state) pdev->current_state = state.event; } else - IRDA_ERROR("%s - %s: invalid suspend request %u -> %u\n", __FUNCTION__, PCIDEV_NAME(pdev), pdev->current_state, state.event); + IRDA_ERROR("%s - %s: invalid suspend request %u -> %u\n", __FUNCTION__, pci_name(pdev), pdev->current_state, state.event); up(&idev->sem); return 0; } @@ -1787,7 +1787,7 @@ static int vlsi_irda_resume(struct pci_dev *pdev) if (!ndev) { IRDA_ERROR("%s - %s: no netdevice \n", - __FUNCTION__, PCIDEV_NAME(pdev)); + __FUNCTION__, pci_name(pdev)); return 0; } idev = ndev->priv; @@ -1795,7 +1795,7 @@ static int vlsi_irda_resume(struct pci_dev *pdev) if (pdev->current_state == 0) { up(&idev->sem); IRDA_WARNING("%s - %s: already resumed\n", - __FUNCTION__, PCIDEV_NAME(pdev)); + __FUNCTION__, pci_name(pdev)); return 0; } diff --git a/drivers/net/irda/vlsi_ir.h b/drivers/net/irda/vlsi_ir.h index c37f0bc4c7f9..2d3b773d8e35 100644 --- a/drivers/net/irda/vlsi_ir.h +++ b/drivers/net/irda/vlsi_ir.h @@ -41,39 +41,6 @@ #define PCI_CLASS_SUBCLASS_MASK 0xffff #endif -/* in recent 2.5 interrupt handlers have non-void return value */ -#ifndef IRQ_RETVAL -typedef void irqreturn_t; -#define IRQ_NONE -#define IRQ_HANDLED -#define IRQ_RETVAL(x) -#endif - -/* some stuff need to check kernelversion. Not all 2.5 stuff was present - * in early 2.5.x - the test is merely to separate 2.4 from 2.5 - */ -#include - -#if LINUX_VERSION_CODE < KERNEL_VERSION(2,5,0) - -/* PDE() introduced in 2.5.4 */ -#ifdef CONFIG_PROC_FS -#define PDE(inode) ((inode)->i_private) -#endif - -/* irda crc16 calculation exported in 2.5.42 */ -#define irda_calc_crc16(fcs,buf,len) (GOOD_FCS) - -/* we use this for unified pci device name access */ -#define PCIDEV_NAME(pdev) ((pdev)->name) - -#else /* 2.5 or later */ - -/* whatever we get from the associated struct device - bus:slot:dev.fn id */ -#define PCIDEV_NAME(pdev) (pci_name(pdev)) - -#endif - /* ================================================================ */ /* non-standard PCI registers */ From 2748e5dec7ca8a3804852c7c4171f9156384d15c Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 23 Jan 2007 22:00:13 -0800 Subject: [PATCH 072/326] [NETFILTER]: Fix iptables ABI breakage on (at least) CRIS With the introduction of x_tables we accidentally broke compatibility by defining IPT_TABLE_MAXNAMELEN to XT_FUNCTION_MAXNAMELEN instead of XT_TABLE_MAXNAMELEN, which is two bytes larger. On most architectures it doesn't really matter since we don't have any tables with names that long in the kernel and the structure layout didn't change because of alignment requirements of following members. On CRIS however (and other architectures that don't align data) this changed the structure layout and thus broke compatibility with old iptables binaries. Changing it back will break compatibility with binaries compiled against recent kernels again, but since the breakage has only been there for three releases this seems like the better choice. Spotted by Jonas Berlin . Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter_ipv4/ip_tables.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/netfilter_ipv4/ip_tables.h b/include/linux/netfilter_ipv4/ip_tables.h index 4f06dad0bde9..98d566c5e32a 100644 --- a/include/linux/netfilter_ipv4/ip_tables.h +++ b/include/linux/netfilter_ipv4/ip_tables.h @@ -28,7 +28,7 @@ #include #define IPT_FUNCTION_MAXNAMELEN XT_FUNCTION_MAXNAMELEN -#define IPT_TABLE_MAXNAMELEN XT_FUNCTION_MAXNAMELEN +#define IPT_TABLE_MAXNAMELEN XT_TABLE_MAXNAMELEN #define ipt_match xt_match #define ipt_target xt_target #define ipt_table xt_table From bf1c1ee88008fd639ebb5c74f0555fd414369bdc Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 23 Jan 2007 22:04:35 -0800 Subject: [PATCH 073/326] [NET]: Process include/linux/if_{addr,link}.h with unifdef After commit d3dcc077bf88806201093f86325ec656e4dbfbce, include/linux/if_{addr,link}.h should be processed with unifdef. Signed-off-by: Adrian Bunk Signed-off-by: David S. Miller --- include/linux/Kbuild | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/include/linux/Kbuild b/include/linux/Kbuild index 8c634f9df39f..157db77a7170 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -69,7 +69,6 @@ header-y += hysdn_if.h header-y += i2c-dev.h header-y += i8k.h header-y += icmp.h -header-y += if_addr.h header-y += if_arcnet.h header-y += if_arp.h header-y += if_bonding.h @@ -79,7 +78,6 @@ header-y += if_fddi.h header-y += if.h header-y += if_hippi.h header-y += if_infiniband.h -header-y += if_link.h header-y += if_packet.h header-y += if_plip.h header-y += if_ppp.h @@ -213,6 +211,7 @@ unifdef-y += hpet.h unifdef-y += i2c.h unifdef-y += i2o-dev.h unifdef-y += icmpv6.h +unifdef-y += if_addr.h unifdef-y += if_bridge.h unifdef-y += if_ec.h unifdef-y += if_eql.h @@ -220,6 +219,7 @@ unifdef-y += if_ether.h unifdef-y += if_fddi.h unifdef-y += if_frad.h unifdef-y += if_ltalk.h +unifdef-y += if_link.h unifdef-y += if_pppox.h unifdef-y += if_shaper.h unifdef-y += if_tr.h From 52d570aabe921663a987b2e4bae2bdc411cee480 Mon Sep 17 00:00:00 2001 From: Jarek Poplawski Date: Tue, 23 Jan 2007 22:07:12 -0800 Subject: [PATCH 074/326] [TCP]: rare bad TCP checksum with 2.6.19 The patch "Replace CHECKSUM_HW by CHECKSUM_PARTIAL/CHECKSUM_COMPLETE" changed to unconditional copying of ip_summed field from collapsed skb. This patch reverts this change. The majority of substantial work including heavy testing and diagnosing by: Michael Tokarev Possible reasons pointed by: Herbert Xu and Patrick McHardy. Signed-off-by: Jarek Poplawski Acked-by: Herbert Xu Signed-off-by: David S. Miller --- net/ipv4/tcp_output.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 32c1a972fa31..23e32c806916 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -1650,7 +1650,8 @@ static void tcp_retrans_try_collapse(struct sock *sk, struct sk_buff *skb, int m memcpy(skb_put(skb, next_skb_size), next_skb->data, next_skb_size); - skb->ip_summed = next_skb->ip_summed; + if (next_skb->ip_summed == CHECKSUM_PARTIAL) + skb->ip_summed = CHECKSUM_PARTIAL; if (skb->ip_summed != CHECKSUM_PARTIAL) skb->csum = csum_block_add(skb->csum, next_skb->csum, skb_size); From 6a2b9ce0a383059492c93682bc094cce0f705fff Mon Sep 17 00:00:00 2001 From: Noriaki TAKAMIYA Date: Tue, 23 Jan 2007 22:09:41 -0800 Subject: [PATCH 075/326] [IPV6]: Fixed the size of the netlink message notified by inet6_rt_notify(). I think the return value of rt6_nlmsg_size() should includes the amount of RTA_METRICS. Signed-off-by: Noriaki TAKAMIYA Acked-by: Thomas Graf Signed-off-by: David S. Miller --- net/ipv6/route.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 8c3d56871b50..5f0043c30b70 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2017,6 +2017,7 @@ static inline size_t rt6_nlmsg_size(void) + nla_total_size(4) /* RTA_IIF */ + nla_total_size(4) /* RTA_OIF */ + nla_total_size(4) /* RTA_PRIORITY */ + + RTAX_MAX * nla_total_size(4) /* RTA_METRICS */ + nla_total_size(sizeof(struct rta_cacheinfo)); } From 6fd8bb881509c6bdc3469b3ed16ec25a3b7cad0e Mon Sep 17 00:00:00 2001 From: Masahide NAKAMURA Date: Tue, 23 Jan 2007 22:17:23 -0800 Subject: [PATCH 076/326] [IP] TUNNEL: Fix to be built with user application. include/linux/if_tunnel.h is broken for user application because it was changed to use __be32 which is required to include linux/types.h in advance but didn't. (This issue is found when building MIPL2 daemon. We are not sure this is the last header to be fixed about __be32.) Signed-off-by: Masahide NAKAMURA Signed-off-by: TAKAMIYA Noriaki Signed-off-by: David S. Miller --- include/linux/if_tunnel.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/include/linux/if_tunnel.h b/include/linux/if_tunnel.h index 8de079ba1107..660b5010c2d9 100644 --- a/include/linux/if_tunnel.h +++ b/include/linux/if_tunnel.h @@ -1,6 +1,8 @@ #ifndef _IF_TUNNEL_H_ #define _IF_TUNNEL_H_ +#include + #define SIOCGETTUNNEL (SIOCDEVPRIVATE + 0) #define SIOCADDTUNNEL (SIOCDEVPRIVATE + 1) #define SIOCDELTUNNEL (SIOCDEVPRIVATE + 2) From 1e5c11fc89ef6663aaa14db1e9e27477f07c24e0 Mon Sep 17 00:00:00 2001 From: Brian Haley Date: Tue, 23 Jan 2007 22:32:23 -0800 Subject: [PATCH 077/326] [SCTP]: Fix compiler warning. > --- a/net/sctp/sm_statefuns.c > +++ b/net/sctp/sm_statefuns.c > @@ -462,24 +461,6 @@ sctp_disposition_t sctp_sf_do_5_1C_ack(const struct sctp_endpoint *ep, > - if (!init_tag) { > - struct sctp_chunk *reply = sctp_make_abort(asoc, chunk, 0); > - if (!reply) > - goto nomem; This introduced a compiler warning, easily fixed. Signed-off-by: Brian Haley Signed-off-by: David S. Miller --- net/sctp/sm_statefuns.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index fce1f602cde2..fbbc9e6a3b78 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -531,9 +531,6 @@ sctp_disposition_t sctp_sf_do_5_1C_ack(const struct sctp_endpoint *ep, SCTP_CHUNK(err_chunk)); return SCTP_DISPOSITION_CONSUME; - -nomem: - return SCTP_DISPOSITION_NOMEM; } /* From 82490c0937cb455e7e4150455ff52e89a9fc5ab8 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 23 Jan 2007 15:13:39 +0900 Subject: [PATCH 078/326] ahci: make ULi M5288 ignore interface fatal error bit As with JMicron controllers, ULi M5288 sets interface fatal error bit on device error including ATAPI CC. This makes libata hardreset the port on ATAPI CC thus making it impossible to use. Ignore interface fatal error bit on ULi M5288. This fixes bugzilla bug #7837. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/ata/ahci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c index b517d2493551..bd241767caea 100644 --- a/drivers/ata/ahci.c +++ b/drivers/ata/ahci.c @@ -361,7 +361,7 @@ static const struct pci_device_id ahci_pci_tbl[] = { { PCI_VDEVICE(INTEL, 0x27c1), board_ahci }, /* ICH7 */ { PCI_VDEVICE(INTEL, 0x27c5), board_ahci }, /* ICH7M */ { PCI_VDEVICE(INTEL, 0x27c3), board_ahci }, /* ICH7R */ - { PCI_VDEVICE(AL, 0x5288), board_ahci }, /* ULi M5288 */ + { PCI_VDEVICE(AL, 0x5288), board_ahci_ign_iferr }, /* ULi M5288 */ { PCI_VDEVICE(INTEL, 0x2681), board_ahci }, /* ESB2 */ { PCI_VDEVICE(INTEL, 0x2682), board_ahci }, /* ESB2 */ { PCI_VDEVICE(INTEL, 0x2683), board_ahci }, /* ESB2 */ From f740d1689d91415cfc749d17138a11ed03b7d38b Mon Sep 17 00:00:00 2001 From: Robert Hancock Date: Tue, 23 Jan 2007 20:09:02 -0600 Subject: [PATCH 079/326] sata_nv: don't rely on NV_INT_DEV indication with ADMA Several people reported issues with certain drive commands timing out on sata_nv controllers running in ADMA mode. The commands in question were non-DMA-mapped commands, usually FLUSH CACHE or FLUSH CACHE EXT. From experimentation it appears that the NV_INT_DEV indication isn't always set when a legitimate command completion interrupt is received on a legacy-mode command, at least not on these controllers in ADMA mode. When a command is pending on the port, force the flag on always in the irq_stat value before calling nv_host_intr so that the drive busy state is always checked by ata_host_intr. This also fixes some questionable code in nv_host_intr which called ata_check_status when a command was pending and ata_host_intr returned "unhandled". If the device interrupted at just the wrong time this could cause interrupts to be lost. Signed-off-by: Robert Hancock Signed-off-by: Jeff Garzik --- drivers/ata/sata_nv.c | 14 ++++++-------- 1 file changed, 6 insertions(+), 8 deletions(-) diff --git a/drivers/ata/sata_nv.c b/drivers/ata/sata_nv.c index f6d498e1cf80..f7a963eb1f02 100644 --- a/drivers/ata/sata_nv.c +++ b/drivers/ata/sata_nv.c @@ -700,7 +700,6 @@ static void nv_adma_check_cpb(struct ata_port *ap, int cpb_num, int force_err) static int nv_host_intr(struct ata_port *ap, u8 irq_stat) { struct ata_queued_cmd *qc = ata_qc_from_tag(ap, ap->active_tag); - int handled; /* freeze if hotplugged */ if (unlikely(irq_stat & (NV_INT_ADDED | NV_INT_REMOVED))) { @@ -719,13 +718,7 @@ static int nv_host_intr(struct ata_port *ap, u8 irq_stat) } /* handle interrupt */ - handled = ata_host_intr(ap, qc); - if (unlikely(!handled)) { - /* spurious, clear it */ - ata_check_status(ap); - } - - return 1; + return ata_host_intr(ap, qc); } static irqreturn_t nv_adma_interrupt(int irq, void *dev_instance) @@ -752,6 +745,11 @@ static irqreturn_t nv_adma_interrupt(int irq, void *dev_instance) if (pp->flags & NV_ADMA_PORT_REGISTER_MODE) { u8 irq_stat = readb(host->mmio_base + NV_INT_STATUS_CK804) >> (NV_INT_PORT_SHIFT * i); + if(ata_tag_valid(ap->active_tag)) + /** NV_INT_DEV indication seems unreliable at times + at least in ADMA mode. Force it on always when a + command is active, to prevent losing interrupts. */ + irq_stat |= NV_INT_DEV; handled += nv_host_intr(ap, irq_stat); continue; } From 07c53dac4904206a50dd7c87adabbb1acff903fb Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Sun, 21 Jan 2007 02:10:11 +0900 Subject: [PATCH 080/326] ahci: don't enter slumber on power down Some ATA/ATAPI devices act weirdly after the link is put into slumber mode. Some hang completely requiring physical power removal while others fail to wake up till the link is hardreset a couple of times. The addition of slumber on power down was never driven by real need. It just followed what ahci spec said literally. The spec itself seems faulty in that it doesn't consider devices (not controllers) which don't support link powersaving mode. Theory never matches reality when it comes to dark allys of cheap ATA/ATAPI world. It's just unrealistic to expect vendors to test rarely used link powersaving feature rigorously. This patch makes ahci more friendly to the coldness of reality. This shouldn't have any negative effect - when suspend operation succeeds, we power off the whole machine; otherwise, we wake up everything. I can't see any reason to be so elaborate with powering down the link in the first place. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/ata/ahci.c | 39 +++++++++++---------------------------- 1 file changed, 11 insertions(+), 28 deletions(-) diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c index bd241767caea..e3c7b312287a 100644 --- a/drivers/ata/ahci.c +++ b/drivers/ata/ahci.c @@ -586,35 +586,18 @@ static void ahci_power_down(void __iomem *port_mmio, u32 cap) { u32 cmd, scontrol; + if (!(cap & HOST_CAP_SSS)) + return; + + /* put device into listen mode, first set PxSCTL.DET to 0 */ + scontrol = readl(port_mmio + PORT_SCR_CTL); + scontrol &= ~0xf; + writel(scontrol, port_mmio + PORT_SCR_CTL); + + /* then set PxCMD.SUD to 0 */ cmd = readl(port_mmio + PORT_CMD) & ~PORT_CMD_ICC_MASK; - - if (cap & HOST_CAP_SSC) { - /* enable transitions to slumber mode */ - scontrol = readl(port_mmio + PORT_SCR_CTL); - if ((scontrol & 0x0f00) > 0x100) { - scontrol &= ~0xf00; - writel(scontrol, port_mmio + PORT_SCR_CTL); - } - - /* put device into slumber mode */ - writel(cmd | PORT_CMD_ICC_SLUMBER, port_mmio + PORT_CMD); - - /* wait for the transition to complete */ - ata_wait_register(port_mmio + PORT_CMD, PORT_CMD_ICC_SLUMBER, - PORT_CMD_ICC_SLUMBER, 1, 50); - } - - /* put device into listen mode */ - if (cap & HOST_CAP_SSS) { - /* first set PxSCTL.DET to 0 */ - scontrol = readl(port_mmio + PORT_SCR_CTL); - scontrol &= ~0xf; - writel(scontrol, port_mmio + PORT_SCR_CTL); - - /* then set PxCMD.SUD to 0 */ - cmd &= ~PORT_CMD_SPIN_UP; - writel(cmd, port_mmio + PORT_CMD); - } + cmd &= ~PORT_CMD_SPIN_UP; + writel(cmd, port_mmio + PORT_CMD); } static void ahci_init_port(void __iomem *port_mmio, u32 cap, From 7a801184fa480e11e6431f184a5bdf31f63326fb Mon Sep 17 00:00:00 2001 From: Brian King Date: Wed, 17 Jan 2007 12:32:12 -0600 Subject: [PATCH 081/326] libata: Fixup n_elem initialization Fixup the inialization of qc->n_elem. It currently gets initialized to 1 for commands that do not transfer any data. Fix this by initializing n_elem to 0 and only setting to 1 in ata_scsi_qc_new when there is data to transfer. This fixes some problems seen with SATA devices attached to ipr adapters. Signed-off-by: Brian King Signed-off-by: Jeff Garzik --- drivers/ata/libata-scsi.c | 2 +- include/linux/libata.h | 1 + 2 files changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c index 836947da5b14..7cc5a4a910a4 100644 --- a/drivers/ata/libata-scsi.c +++ b/drivers/ata/libata-scsi.c @@ -372,7 +372,7 @@ struct ata_queued_cmd *ata_scsi_qc_new(struct ata_device *dev, if (cmd->use_sg) { qc->__sg = (struct scatterlist *) cmd->request_buffer; qc->n_elem = cmd->use_sg; - } else { + } else if (cmd->request_bufflen) { qc->__sg = &qc->sgent; qc->n_elem = 1; } diff --git a/include/linux/libata.h b/include/linux/libata.h index e53a13ba7f78..65d318746ac6 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -1149,6 +1149,7 @@ static inline void ata_qc_reinit(struct ata_queued_cmd *qc) qc->cursect = qc->cursg = qc->cursg_ofs = 0; qc->nsect = 0; qc->nbytes = qc->curbytes = 0; + qc->n_elem = 0; qc->err_mask = 0; ata_tf_init(qc->dev, &qc->tf); From d0f29485686d9d1c4f31240953a742d5dd4fdb72 Mon Sep 17 00:00:00 2001 From: Brian King Date: Wed, 17 Jan 2007 12:32:28 -0600 Subject: [PATCH 082/326] libata: Initialize qc->pad_len Initialize qc->pad_len for each new command. This ensures that pad_len is not set to a stale value for zero data length commands. Signed-off-by: Brian King Signed-off-by: Jeff Garzik --- include/linux/libata.h | 1 + 1 file changed, 1 insertion(+) diff --git a/include/linux/libata.h b/include/linux/libata.h index 65d318746ac6..f7f268e38749 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -1151,6 +1151,7 @@ static inline void ata_qc_reinit(struct ata_queued_cmd *qc) qc->nbytes = qc->curbytes = 0; qc->n_elem = 0; qc->err_mask = 0; + qc->pad_len = 0; ata_tf_init(qc->dev, &qc->tf); From 63ea9c1710f05a7309c272750bbf40134c628f36 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Wed, 20 Dec 2006 16:11:57 +0100 Subject: [PATCH 083/326] [POWERPC] PS3: Fix uniprocessor kernel build Allow to build a uniprocessor kernel for PS3. Signed-off-by: Geert Uytterhoeven Signed-off-by: Paul Mackerras --- arch/powerpc/platforms/ps3/Makefile | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/powerpc/platforms/ps3/Makefile b/arch/powerpc/platforms/ps3/Makefile index 3757cfabc8ce..1994904f580f 100644 --- a/arch/powerpc/platforms/ps3/Makefile +++ b/arch/powerpc/platforms/ps3/Makefile @@ -1,4 +1,5 @@ -obj-y += setup.o mm.o smp.o time.o hvcall.o htab.o repository.o +obj-y += setup.o mm.o time.o hvcall.o htab.o repository.o obj-y += interrupt.o exports.o os-area.o +obj-$(CONFIG_SMP) += smp.o obj-$(CONFIG_SPU_BASE) += spu.o From ded84bcb246780137ecaa3f6d137ac1b0f06fd08 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Thu, 21 Dec 2006 13:57:16 +0100 Subject: [PATCH 084/326] [POWERPC] ps3_free_io_irq: Fix inverted error check ps3_free_io_irq: Fix inverted error check after calling lv1_destruct_io_irq_outlet(). Signed-off-by: Geert Uytterhoeven Signed-off-by: Paul Mackerras --- arch/powerpc/platforms/ps3/interrupt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/powerpc/platforms/ps3/interrupt.c b/arch/powerpc/platforms/ps3/interrupt.c index 056c1e4141ba..6f5de438b980 100644 --- a/arch/powerpc/platforms/ps3/interrupt.c +++ b/arch/powerpc/platforms/ps3/interrupt.c @@ -71,7 +71,7 @@ int ps3_free_io_irq(unsigned int virq) result = lv1_destruct_io_irq_outlet(virq_to_hw(virq)); - if (!result) + if (result) pr_debug("%s:%d: lv1_destruct_io_irq_outlet failed: %s\n", __func__, __LINE__, ps3_result(result)); From fb1d50418ef09a6a43b3928ff78d433914ea954d Mon Sep 17 00:00:00 2001 From: Andrew Victor Date: Tue, 9 Jan 2007 08:44:41 +0100 Subject: [PATCH 085/326] [ARM] 4084/1: Remove CONFIG_DEBUG_WAITQ Remove the legacy CONFIG_DEBUG_WAITQ from the SAM9260-EK and SAM9261-EK default configuration files. Signed-off-by: Andrew Victor Signed-off-by: Russell King --- arch/arm/configs/at91sam9260ek_defconfig | 1 - arch/arm/configs/at91sam9261ek_defconfig | 1 - 2 files changed, 2 deletions(-) diff --git a/arch/arm/configs/at91sam9260ek_defconfig b/arch/arm/configs/at91sam9260ek_defconfig index 79049206dfa5..46b0c734aeb9 100644 --- a/arch/arm/configs/at91sam9260ek_defconfig +++ b/arch/arm/configs/at91sam9260ek_defconfig @@ -923,7 +923,6 @@ CONFIG_FORCED_INLINING=y # CONFIG_HEADERS_CHECK is not set # CONFIG_RCU_TORTURE_TEST is not set CONFIG_DEBUG_USER=y -# CONFIG_DEBUG_WAITQ is not set # CONFIG_DEBUG_ERRORS is not set CONFIG_DEBUG_LL=y # CONFIG_DEBUG_ICEDCC is not set diff --git a/arch/arm/configs/at91sam9261ek_defconfig b/arch/arm/configs/at91sam9261ek_defconfig index 784ad7c0186d..fcd8fa091e9d 100644 --- a/arch/arm/configs/at91sam9261ek_defconfig +++ b/arch/arm/configs/at91sam9261ek_defconfig @@ -1079,7 +1079,6 @@ CONFIG_FORCED_INLINING=y # CONFIG_HEADERS_CHECK is not set # CONFIG_RCU_TORTURE_TEST is not set CONFIG_DEBUG_USER=y -# CONFIG_DEBUG_WAITQ is not set # CONFIG_DEBUG_ERRORS is not set CONFIG_DEBUG_LL=y # CONFIG_DEBUG_ICEDCC is not set From 410f4eae4b33631ace70d84470218f3db302afac Mon Sep 17 00:00:00 2001 From: Andrew Victor Date: Tue, 9 Jan 2007 08:51:43 +0100 Subject: [PATCH 086/326] [ARM] 4085/1: AT91: Header fixes. Fix two typo's where AT01_* was used instead of AT91_*. [Patch from Wojtek Kaniewski] Fix definition of AT91_SMC_EXNWMODE for the SAM9 processors. [Patch from Wu Xuan] Signed-off-by: Andrew Victor Signed-off-by: Russell King --- include/asm-arm/arch-at91rm9200/at91_rstc.h | 2 +- include/asm-arm/arch-at91rm9200/at91sam9261_matrix.h | 2 +- include/asm-arm/arch-at91rm9200/at91sam926x_mc.h | 8 ++++---- 3 files changed, 6 insertions(+), 6 deletions(-) diff --git a/include/asm-arm/arch-at91rm9200/at91_rstc.h b/include/asm-arm/arch-at91rm9200/at91_rstc.h index ccdc52da973d..237d3c40b318 100644 --- a/include/asm-arm/arch-at91rm9200/at91_rstc.h +++ b/include/asm-arm/arch-at91rm9200/at91_rstc.h @@ -17,7 +17,7 @@ #define AT91_RSTC_PROCRST (1 << 0) /* Processor Reset */ #define AT91_RSTC_PERRST (1 << 2) /* Peripheral Reset */ #define AT91_RSTC_EXTRST (1 << 3) /* External Reset */ -#define AT01_RSTC_KEY (0xff << 24) /* KEY Password */ +#define AT91_RSTC_KEY (0xff << 24) /* KEY Password */ #define AT91_RSTC_SR (AT91_RSTC + 0x04) /* Reset Controller Status Register */ #define AT91_RSTC_URSTS (1 << 0) /* User Reset Status */ diff --git a/include/asm-arm/arch-at91rm9200/at91sam9261_matrix.h b/include/asm-arm/arch-at91rm9200/at91sam9261_matrix.h index 270a5dcdf1cd..ab25f9e94d69 100644 --- a/include/asm-arm/arch-at91rm9200/at91sam9261_matrix.h +++ b/include/asm-arm/arch-at91rm9200/at91sam9261_matrix.h @@ -15,7 +15,7 @@ #define AT91_MATRIX_MCFG (AT91_MATRIX + 0x00) /* Master Configuration Register */ #define AT91_MATRIX_RCB0 (1 << 0) /* Remap Command for AHB Master 0 (ARM926EJ-S Instruction Master) */ -#define AT01_MATRIX_RCB1 (1 << 1) /* Remap Command for AHB Master 1 (ARM926EJ-S Data Master) */ +#define AT91_MATRIX_RCB1 (1 << 1) /* Remap Command for AHB Master 1 (ARM926EJ-S Data Master) */ #define AT91_MATRIX_SCFG0 (AT91_MATRIX + 0x04) /* Slave Configuration Register 0 */ #define AT91_MATRIX_SCFG1 (AT91_MATRIX + 0x08) /* Slave Configuration Register 1 */ diff --git a/include/asm-arm/arch-at91rm9200/at91sam926x_mc.h b/include/asm-arm/arch-at91rm9200/at91sam926x_mc.h index 7d94968b5d57..8ac797e39a14 100644 --- a/include/asm-arm/arch-at91rm9200/at91sam926x_mc.h +++ b/include/asm-arm/arch-at91rm9200/at91sam926x_mc.h @@ -110,10 +110,10 @@ #define AT91_SMC_MODE(n) (AT91_SMC + 0x0c + ((n)*0x10)) /* Mode Register for CS n */ #define AT91_SMC_READMODE (1 << 0) /* Read Mode */ #define AT91_SMC_WRITEMODE (1 << 1) /* Write Mode */ -#define AT91_SMC_EXNWMODE (3 << 5) /* NWAIT Mode */ -#define AT91_SMC_EXNWMODE_DISABLE (0 << 5) -#define AT91_SMC_EXNWMODE_FROZEN (2 << 5) -#define AT91_SMC_EXNWMODE_READY (3 << 5) +#define AT91_SMC_EXNWMODE (3 << 4) /* NWAIT Mode */ +#define AT91_SMC_EXNWMODE_DISABLE (0 << 4) +#define AT91_SMC_EXNWMODE_FROZEN (2 << 4) +#define AT91_SMC_EXNWMODE_READY (3 << 4) #define AT91_SMC_BAT (1 << 8) /* Byte Access Type */ #define AT91_SMC_BAT_SELECT (0 << 8) #define AT91_SMC_BAT_WRITE (1 << 8) From a14d527306dc7dbc38e4607c3cc3a50a600fc98b Mon Sep 17 00:00:00 2001 From: Andrew Victor Date: Tue, 9 Jan 2007 09:03:42 +0100 Subject: [PATCH 087/326] [ARM] 4086/1: AT91: Whitespace cleanup A couple of whitespace cleanups, mainly in the AT91 header files. Signed-off-by: Andrew Victor Signed-off-by: Russell King --- arch/arm/mach-at91rm9200/at91rm9200_devices.c | 2 +- drivers/serial/atmel_serial.c | 4 ++-- drivers/serial/atmel_serial.h | 2 +- include/asm-arm/arch-at91rm9200/at91_ecc.h | 8 ++++---- include/asm-arm/arch-at91rm9200/at91_pmc.h | 2 +- include/asm-arm/arch-at91rm9200/at91_rtc.h | 16 ++++++++-------- include/asm-arm/arch-at91rm9200/at91rm9200.h | 2 +- .../asm-arm/arch-at91rm9200/at91sam9260_matrix.h | 2 +- .../asm-arm/arch-at91rm9200/at91sam9261_matrix.h | 4 ++-- include/asm-arm/arch-at91rm9200/at91sam926x_mc.h | 8 ++++---- 10 files changed, 25 insertions(+), 25 deletions(-) diff --git a/arch/arm/mach-at91rm9200/at91rm9200_devices.c b/arch/arm/mach-at91rm9200/at91rm9200_devices.c index 4641b99db0ee..57fac7203fe4 100644 --- a/arch/arm/mach-at91rm9200/at91rm9200_devices.c +++ b/arch/arm/mach-at91rm9200/at91rm9200_devices.c @@ -272,7 +272,7 @@ void __init at91_add_device_cf(struct at91_cf_data *data) at91_set_A_periph(AT91_PIN_PC12, 0); /* NCS6/CFCE2 */ /* nWAIT is _not_ a default setting */ - at91_set_A_periph(AT91_PIN_PC6, 1); /* nWAIT */ + at91_set_A_periph(AT91_PIN_PC6, 1); /* nWAIT */ cf_data = *data; platform_device_register(&at91rm9200_cf_device); diff --git a/drivers/serial/atmel_serial.c b/drivers/serial/atmel_serial.c index ed7f7209ea59..1f9222c2e656 100644 --- a/drivers/serial/atmel_serial.c +++ b/drivers/serial/atmel_serial.c @@ -689,9 +689,9 @@ static void __devinit atmel_init_port(struct atmel_uart_port *atmel_port, struct struct atmel_uart_data *data = pdev->dev.platform_data; port->iotype = UPIO_MEM; - port->flags = UPF_BOOT_AUTOCONF; + port->flags = UPF_BOOT_AUTOCONF; port->ops = &atmel_pops; - port->fifosize = 1; + port->fifosize = 1; port->line = pdev->id; port->dev = &pdev->dev; diff --git a/drivers/serial/atmel_serial.h b/drivers/serial/atmel_serial.h index fe1763b2a6d5..11b44360e108 100644 --- a/drivers/serial/atmel_serial.h +++ b/drivers/serial/atmel_serial.h @@ -106,7 +106,7 @@ #define ATMEL_US_CSR 0x14 /* Channel Status Register */ #define ATMEL_US_RHR 0x18 /* Receiver Holding Register */ #define ATMEL_US_THR 0x1c /* Transmitter Holding Register */ -#define ATMEL_US_SYNH (1 << 15) /* Transmit/Receive Sync [SAM9 only] */ +#define ATMEL_US_SYNH (1 << 15) /* Transmit/Receive Sync [AT91SAM9261 only] */ #define ATMEL_US_BRGR 0x20 /* Baud Rate Generator Register */ #define ATMEL_US_CD (0xffff << 0) /* Clock Divider */ diff --git a/include/asm-arm/arch-at91rm9200/at91_ecc.h b/include/asm-arm/arch-at91rm9200/at91_ecc.h index fddf256a98d3..5c564ede5c5d 100644 --- a/include/asm-arm/arch-at91rm9200/at91_ecc.h +++ b/include/asm-arm/arch-at91rm9200/at91_ecc.h @@ -14,7 +14,7 @@ #define AT91_ECC_H #define AT91_ECC_CR (AT91_ECC + 0x00) /* Control register */ -#define AT91_ECC_RST (1 << 0) /* Reset parity */ +#define AT91_ECC_RST (1 << 0) /* Reset parity */ #define AT91_ECC_MR (AT91_ECC + 0x04) /* Mode register */ #define AT91_ECC_PAGESIZE (3 << 0) /* Page Size */ @@ -23,16 +23,16 @@ #define AT91_ECC_PAGESIZE_2112 (2) #define AT91_ECC_PAGESIZE_4224 (3) -#define AT91_ECC_SR (AT91_ECC + 0x08) /* Status register */ +#define AT91_ECC_SR (AT91_ECC + 0x08) /* Status register */ #define AT91_ECC_RECERR (1 << 0) /* Recoverable Error */ #define AT91_ECC_ECCERR (1 << 1) /* ECC Single Bit Error */ #define AT91_ECC_MULERR (1 << 2) /* Multiple Errors */ -#define AT91_ECC_PR (AT91_ECC + 0x0c) /* Parity register */ +#define AT91_ECC_PR (AT91_ECC + 0x0c) /* Parity register */ #define AT91_ECC_BITADDR (0xf << 0) /* Bit Error Address */ #define AT91_ECC_WORDADDR (0xfff << 4) /* Word Error Address */ -#define AT91_ECC_NPR (AT91_ECC + 0x10) /* NParity register */ +#define AT91_ECC_NPR (AT91_ECC + 0x10) /* NParity register */ #define AT91_ECC_NPARITY (0xffff << 0) /* NParity */ #endif diff --git a/include/asm-arm/arch-at91rm9200/at91_pmc.h b/include/asm-arm/arch-at91rm9200/at91_pmc.h index de8c3da74a01..c3b489d09b6c 100644 --- a/include/asm-arm/arch-at91rm9200/at91_pmc.h +++ b/include/asm-arm/arch-at91rm9200/at91_pmc.h @@ -61,7 +61,7 @@ #define AT91_PMC_CSS_PLLA (2 << 0) #define AT91_PMC_CSS_PLLB (3 << 0) #define AT91_PMC_PRES (7 << 2) /* Master Clock Prescaler */ -#define AT91_PMC_PRES_1 (0 << 2) +#define AT91_PMC_PRES_1 (0 << 2) #define AT91_PMC_PRES_2 (1 << 2) #define AT91_PMC_PRES_4 (2 << 2) #define AT91_PMC_PRES_8 (3 << 2) diff --git a/include/asm-arm/arch-at91rm9200/at91_rtc.h b/include/asm-arm/arch-at91rm9200/at91_rtc.h index 6e5065d56260..095fe0883102 100644 --- a/include/asm-arm/arch-at91rm9200/at91_rtc.h +++ b/include/asm-arm/arch-at91rm9200/at91_rtc.h @@ -21,21 +21,21 @@ #define AT91_RTC_UPDCAL (1 << 1) /* Update Request Calendar Register */ #define AT91_RTC_TIMEVSEL (3 << 8) /* Time Event Selection */ #define AT91_RTC_TIMEVSEL_MINUTE (0 << 8) -#define AT91_RTC_TIMEVSEL_HOUR (1 << 8) -#define AT91_RTC_TIMEVSEL_DAY24 (2 << 8) -#define AT91_RTC_TIMEVSEL_DAY12 (3 << 8) +#define AT91_RTC_TIMEVSEL_HOUR (1 << 8) +#define AT91_RTC_TIMEVSEL_DAY24 (2 << 8) +#define AT91_RTC_TIMEVSEL_DAY12 (3 << 8) #define AT91_RTC_CALEVSEL (3 << 16) /* Calendar Event Selection */ -#define AT91_RTC_CALEVSEL_WEEK (0 << 16) -#define AT91_RTC_CALEVSEL_MONTH (1 << 16) -#define AT91_RTC_CALEVSEL_YEAR (2 << 16) +#define AT91_RTC_CALEVSEL_WEEK (0 << 16) +#define AT91_RTC_CALEVSEL_MONTH (1 << 16) +#define AT91_RTC_CALEVSEL_YEAR (2 << 16) #define AT91_RTC_MR (AT91_RTC + 0x04) /* Mode Register */ -#define AT91_RTC_HRMOD (1 << 0) /* 12/24 Hour Mode */ +#define AT91_RTC_HRMOD (1 << 0) /* 12/24 Hour Mode */ #define AT91_RTC_TIMR (AT91_RTC + 0x08) /* Time Register */ #define AT91_RTC_SEC (0x7f << 0) /* Current Second */ #define AT91_RTC_MIN (0x7f << 8) /* Current Minute */ -#define AT91_RTC_HOUR (0x3f << 16) /* Current Hour */ +#define AT91_RTC_HOUR (0x3f << 16) /* Current Hour */ #define AT91_RTC_AMPM (1 << 22) /* Ante Meridiem Post Meridiem Indicator */ #define AT91_RTC_CALR (AT91_RTC + 0x0c) /* Calendar Register */ diff --git a/include/asm-arm/arch-at91rm9200/at91rm9200.h b/include/asm-arm/arch-at91rm9200/at91rm9200.h index 4d51177efddd..c569b6a21a42 100644 --- a/include/asm-arm/arch-at91rm9200/at91rm9200.h +++ b/include/asm-arm/arch-at91rm9200/at91rm9200.h @@ -274,7 +274,7 @@ #define AT91_PD19_TPK7 (1 << 19) /* B: ETM Trace Packet Port 7 */ #define AT91_PD20_NPCS3 (1 << 20) /* A: SPI Peripheral Chip Select 3 */ #define AT91_PD20_TPK8 (1 << 20) /* B: ETM Trace Packet Port 8 */ -#define AT91_PD21_RTS0 (1 << 21) /* A: USART Ready To Send 0 */ +#define AT91_PD21_RTS0 (1 << 21) /* A: USART Ready To Send 0 */ #define AT91_PD21_TPK9 (1 << 21) /* B: ETM Trace Packet Port 9 */ #define AT91_PD22_RTS1 (1 << 22) /* A: USART Ready To Send 1 */ #define AT91_PD22_TPK10 (1 << 22) /* B: ETM Trace Packet Port 10 */ diff --git a/include/asm-arm/arch-at91rm9200/at91sam9260_matrix.h b/include/asm-arm/arch-at91rm9200/at91sam9260_matrix.h index 746d973705bf..78f6b4917b8b 100644 --- a/include/asm-arm/arch-at91rm9200/at91sam9260_matrix.h +++ b/include/asm-arm/arch-at91rm9200/at91sam9260_matrix.h @@ -58,7 +58,7 @@ #define AT91_MATRIX_RCB1 (1 << 1) /* Remap Command for AHB Master 1 (ARM926EJ-S Data Master) */ #define AT91_MATRIX_EBICSA (AT91_MATRIX + 0x11C) /* EBI Chip Select Assignment Register */ -#define AT91_MATRIX_CS1A (1 << 1) /* Chip Select 1 Assignment */ +#define AT91_MATRIX_CS1A (1 << 1) /* Chip Select 1 Assignment */ #define AT91_MATRIX_CS1A_SMC (0 << 1) #define AT91_MATRIX_CS1A_SDRAMC (1 << 1) #define AT91_MATRIX_CS3A (1 << 3) /* Chip Select 3 Assignment */ diff --git a/include/asm-arm/arch-at91rm9200/at91sam9261_matrix.h b/include/asm-arm/arch-at91rm9200/at91sam9261_matrix.h index ab25f9e94d69..ec88efabbe6c 100644 --- a/include/asm-arm/arch-at91rm9200/at91sam9261_matrix.h +++ b/include/asm-arm/arch-at91rm9200/at91sam9261_matrix.h @@ -43,8 +43,8 @@ #define AT91_MATRIX_EBICSA (AT91_MATRIX + 0x30) /* EBI Chip Select Assignment Register */ #define AT91_MATRIX_CS1A (1 << 1) /* Chip Select 1 Assignment */ -#define AT91_MATRIX_CS1A_SMC (0 << 1) -#define AT91_MATRIX_CS1A_SDRAMC (1 << 1) +#define AT91_MATRIX_CS1A_SMC (0 << 1) +#define AT91_MATRIX_CS1A_SDRAMC (1 << 1) #define AT91_MATRIX_CS3A (1 << 3) /* Chip Select 3 Assignment */ #define AT91_MATRIX_CS3A_SMC (0 << 3) #define AT91_MATRIX_CS3A_SMC_SMARTMEDIA (1 << 3) diff --git a/include/asm-arm/arch-at91rm9200/at91sam926x_mc.h b/include/asm-arm/arch-at91rm9200/at91sam926x_mc.h index 8ac797e39a14..972e7531c7f4 100644 --- a/include/asm-arm/arch-at91rm9200/at91sam926x_mc.h +++ b/include/asm-arm/arch-at91rm9200/at91sam926x_mc.h @@ -33,14 +33,14 @@ #define AT91_SDRAMC_NC_9 (1 << 0) #define AT91_SDRAMC_NC_10 (2 << 0) #define AT91_SDRAMC_NC_11 (3 << 0) -#define AT91_SDRAMC_NR (3 << 2) /* Number of Row Bits */ +#define AT91_SDRAMC_NR (3 << 2) /* Number of Row Bits */ #define AT91_SDRAMC_NR_11 (0 << 2) #define AT91_SDRAMC_NR_12 (1 << 2) #define AT91_SDRAMC_NR_13 (2 << 2) -#define AT91_SDRAMC_NB (1 << 4) /* Number of Banks */ +#define AT91_SDRAMC_NB (1 << 4) /* Number of Banks */ #define AT91_SDRAMC_NB_2 (0 << 4) -#define AT91_SDRAMC_NB_4 (1 << 4) -#define AT91_SDRAMC_CAS (3 << 5) /* CAS Latency */ +#define AT91_SDRAMC_NB_4 (1 << 4) +#define AT91_SDRAMC_CAS (3 << 5) /* CAS Latency */ #define AT91_SDRAMC_CAS_1 (1 << 5) #define AT91_SDRAMC_CAS_2 (2 << 5) #define AT91_SDRAMC_CAS_3 (3 << 5) From e9d10a16ea76f42ea9ce58508e4af7c52acf1e40 Mon Sep 17 00:00:00 2001 From: Andrew Victor Date: Tue, 9 Jan 2007 12:47:51 +0100 Subject: [PATCH 088/326] [ARM] 4087/1: AT91: CPU reset for SAM9x processors This patch implements CPU and peripheral reset on AT91SAM9260 and AT91SAM9261. Original patch from Wojtek Kaniewski. Signed-off-by: Andrew Victor Signed-off-by: Russell King --- arch/arm/mach-at91rm9200/at91sam9260.c | 3 ++- arch/arm/mach-at91rm9200/at91sam9261.c | 3 ++- 2 files changed, 4 insertions(+), 2 deletions(-) diff --git a/arch/arm/mach-at91rm9200/at91sam9260.c b/arch/arm/mach-at91rm9200/at91sam9260.c index 203f073a53e6..b14871adc300 100644 --- a/arch/arm/mach-at91rm9200/at91sam9260.c +++ b/arch/arm/mach-at91rm9200/at91sam9260.c @@ -16,6 +16,7 @@ #include #include #include +#include #include "generic.h" #include "clock.h" @@ -212,7 +213,7 @@ static struct at91_gpio_bank at91sam9260_gpio[] = { static void at91sam9260_reset(void) { -#warning "Implement CPU reset" + at91_sys_write(AT91_RSTC_CR, (0xA5 << 24) | AT91_RSTC_PROCRST | AT91_RSTC_PERRST); } diff --git a/arch/arm/mach-at91rm9200/at91sam9261.c b/arch/arm/mach-at91rm9200/at91sam9261.c index 5a82f35da2e9..d242bb885c6d 100644 --- a/arch/arm/mach-at91rm9200/at91sam9261.c +++ b/arch/arm/mach-at91rm9200/at91sam9261.c @@ -16,6 +16,7 @@ #include #include #include +#include #include "generic.h" #include "clock.h" @@ -207,7 +208,7 @@ static struct at91_gpio_bank at91sam9261_gpio[] = { static void at91sam9261_reset(void) { -#warning "Implement CPU reset" + at91_sys_write(AT91_RSTC_CR, (0xA5 << 24) | AT91_RSTC_PROCRST | AT91_RSTC_PERRST); } From 9b938166907558e664d8fa413e6233a36669e0c0 Mon Sep 17 00:00:00 2001 From: Andrew Victor Date: Tue, 9 Jan 2007 13:20:54 +0100 Subject: [PATCH 089/326] [ARM] 4088/1: AT91: Unbalanced IRQ in serial driver suspend/resume This patch fixes the unbalanced calls to enable_irq_wake() and disable_irq_wake() in the AT91 (and AVR32) serial driver. It should resolve these kernel messages: Unbalanced IRQ x wake disable BUG: warning at kernel/irq/manage.c:167/set_irq_wake() Original patch from Marc Pignat. Signed-off-by: Andrew Victor Signed-off-by: Russell King --- drivers/serial/atmel_serial.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/serial/atmel_serial.c b/drivers/serial/atmel_serial.c index 1f9222c2e656..881f886b91c6 100644 --- a/drivers/serial/atmel_serial.c +++ b/drivers/serial/atmel_serial.c @@ -890,7 +890,6 @@ static int atmel_serial_suspend(struct platform_device *pdev, pm_message_t state if (device_may_wakeup(&pdev->dev) && !at91_suspend_entering_slow_clock()) enable_irq_wake(port->irq); else { - disable_irq_wake(port->irq); uart_suspend_port(&atmel_uart, port); atmel_port->suspended = 1; } @@ -907,6 +906,8 @@ static int atmel_serial_resume(struct platform_device *pdev) uart_resume_port(&atmel_uart, port); atmel_port->suspended = 0; } + else + disable_irq_wake(port->irq); return 0; } From 3ea163e44c041b9e1d7314998dfbfd4fbc6eea20 Mon Sep 17 00:00:00 2001 From: Andrew Victor Date: Tue, 9 Jan 2007 13:47:29 +0100 Subject: [PATCH 090/326] [ARM] 4089/1: AT91: GPIO wake IRQ cleanup Cleanup of at91 platform level gpio wake and suspend/resume logic. The GPIO core now delegates wakeups to the parent AIC by refcounting, and delegates clock management to the clock API. This makes these system modules more independent of each other, which is cleaner and will also help with the AT91SAM9263 (where some GPIO controllers share the same irq and clock). Original patch by David Brownell. Signed-off-by: Andrew Victor Signed-off-by: Russell King --- arch/arm/mach-at91rm9200/gpio.c | 39 +++++++++++---------------------- 1 file changed, 13 insertions(+), 26 deletions(-) diff --git a/arch/arm/mach-at91rm9200/gpio.c b/arch/arm/mach-at91rm9200/gpio.c index 3f188508c391..af22659c8a28 100644 --- a/arch/arm/mach-at91rm9200/gpio.c +++ b/arch/arm/mach-at91rm9200/gpio.c @@ -20,7 +20,6 @@ #include #include #include -#include #include #include "generic.h" @@ -224,17 +223,17 @@ static u32 backups[MAX_GPIO_BANKS]; static int gpio_irq_set_wake(unsigned pin, unsigned state) { unsigned mask = pin_to_mask(pin); + unsigned bank = (pin - PIN_BASE) / 32; - pin -= PIN_BASE; - pin /= 32; - - if (unlikely(pin >= MAX_GPIO_BANKS)) + if (unlikely(bank >= MAX_GPIO_BANKS)) return -EINVAL; if (state) - wakeups[pin] |= mask; + wakeups[bank] |= mask; else - wakeups[pin] &= ~mask; + wakeups[bank] &= ~mask; + + set_irq_wake(gpio[bank].id, state); return 0; } @@ -246,29 +245,15 @@ void at91_gpio_suspend(void) for (i = 0; i < gpio_banks; i++) { u32 pio = gpio[i].offset; - /* - * Note: drivers should have disabled GPIO interrupts that - * aren't supposed to be wakeup sources. - * But that is not much good on ARM..... disable_irq() does - * not update the hardware immediately, so the hardware mask - * (IMR) has the wrong value (not current, too much is - * permitted). - * - * Our workaround is to disable all non-wakeup IRQs ... - * which is exactly what correct drivers asked for in the - * first place! - */ backups[i] = at91_sys_read(pio + PIO_IMR); at91_sys_write(pio + PIO_IDR, backups[i]); at91_sys_write(pio + PIO_IER, wakeups[i]); - if (!wakeups[i]) { - disable_irq_wake(gpio[i].id); - at91_sys_write(AT91_PMC_PCDR, 1 << gpio[i].id); - } else { - enable_irq_wake(gpio[i].id); + if (!wakeups[i]) + clk_disable(gpio[i].clock); + else { #ifdef CONFIG_PM_DEBUG - printk(KERN_DEBUG "GPIO-%c may wake for %08x\n", "ABCD"[i], wakeups[i]); + printk(KERN_DEBUG "GPIO-%c may wake for %08x\n", 'A'+i, wakeups[i]); #endif } } @@ -281,9 +266,11 @@ void at91_gpio_resume(void) for (i = 0; i < gpio_banks; i++) { u32 pio = gpio[i].offset; + if (!wakeups[i]) + clk_enable(gpio[i].clock); + at91_sys_write(pio + PIO_IDR, wakeups[i]); at91_sys_write(pio + PIO_IER, backups[i]); - at91_sys_write(AT91_PMC_PCER, 1 << gpio[i].id); } } From 5225cd8079484ed27cd52040e8584616d1ef1e2c Mon Sep 17 00:00:00 2001 From: Pavel Pisa Date: Fri, 12 Jan 2007 09:57:22 +0100 Subject: [PATCH 091/326] [ARM] 4092/1: i.MX/MX1 CPU Frequency scaling latency definition The transition latency has to be defined and reasonably small to allow on-demand and conservative governors. The value has been defined according to manual. The imx_set_target() protected against seen out of range requests now. Signed-off-by: Pavel Pisa Acked-by: Sascha Hauer Signed-off-by: Russell King --- arch/arm/mach-imx/cpufreq.c | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) diff --git a/arch/arm/mach-imx/cpufreq.c b/arch/arm/mach-imx/cpufreq.c index ac5f99895660..4f66e90db74f 100644 --- a/arch/arm/mach-imx/cpufreq.c +++ b/arch/arm/mach-imx/cpufreq.c @@ -184,6 +184,17 @@ static int imx_set_target(struct cpufreq_policy *policy, long sysclk; unsigned int bclk_div = 1; + /* + * Some governors do not respects CPU and policy lower limits + * which leads to bad things (division by zero etc), ensure + * that such things do not happen. + */ + if(target_freq < policy->cpuinfo.min_freq) + target_freq = policy->cpuinfo.min_freq; + + if(target_freq < policy->min) + target_freq = policy->min; + freq = target_freq * 1000; pr_debug(KERN_DEBUG "imx: requested frequency %ld Hz, mpctl0 at boot 0x%08x\n", @@ -258,7 +269,8 @@ static int __init imx_cpufreq_driver_init(struct cpufreq_policy *policy) policy->governor = CPUFREQ_DEFAULT_GOVERNOR; policy->cpuinfo.min_freq = 8000; policy->cpuinfo.max_freq = 200000; - policy->cpuinfo.transition_latency = CPUFREQ_ETERNAL; + /* Manual states, that PLL stabilizes in two CLK32 periods */ + policy->cpuinfo.transition_latency = 4 * 1000000000LL / CLK32; return 0; } From 6c3c5bb3c68b932ece9f92b9d201196d537cb99c Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Tue, 16 Jan 2007 12:33:35 +0100 Subject: [PATCH 092/326] [ARM] 4095/1: S3C24XX: Fix GPIO set for Bank A GPIO bank A can only be output or a special function, and the regs-gpio.h header has mistakenly got this as input or output. The mistake is carried on into the gpio.c s3c2410_gpio_cfgpin() call which will set the wrong value if S3C2410_GPIO_OUTPUT is passed. Signed-off-by: Ben Dooks Signed-off-by: Russell King --- arch/arm/mach-s3c2410/gpio.c | 1 + include/asm-arm/arch-s3c2410/regs-gpio.h | 4 ++-- 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/arch/arm/mach-s3c2410/gpio.c b/arch/arm/mach-s3c2410/gpio.c index ba346546150b..d1740643a135 100644 --- a/arch/arm/mach-s3c2410/gpio.c +++ b/arch/arm/mach-s3c2410/gpio.c @@ -57,6 +57,7 @@ void s3c2410_gpio_cfgpin(unsigned int pin, unsigned int function) case S3C2410_GPIO_SFN2: case S3C2410_GPIO_SFN3: if (pin < S3C2410_GPIO_BANKB) { + function -= 1; function &= 1; function <<= S3C2410_GPIO_OFFSET(pin); } else { diff --git a/include/asm-arm/arch-s3c2410/regs-gpio.h b/include/asm-arm/arch-s3c2410/regs-gpio.h index b2893e32a236..eae91694edcd 100644 --- a/include/asm-arm/arch-s3c2410/regs-gpio.h +++ b/include/asm-arm/arch-s3c2410/regs-gpio.h @@ -52,10 +52,10 @@ /* general configuration options */ #define S3C2410_GPIO_LEAVE (0xFFFFFFFF) -#define S3C2410_GPIO_INPUT (0xFFFFFFF0) +#define S3C2410_GPIO_INPUT (0xFFFFFFF0) /* not available on A */ #define S3C2410_GPIO_OUTPUT (0xFFFFFFF1) #define S3C2410_GPIO_IRQ (0xFFFFFFF2) /* not available for all */ -#define S3C2410_GPIO_SFN2 (0xFFFFFFF2) /* not available on A */ +#define S3C2410_GPIO_SFN2 (0xFFFFFFF2) /* bank A => addr/cs/nand */ #define S3C2410_GPIO_SFN3 (0xFFFFFFF3) /* not available on A */ /* register address for the GPIO registers. From 30f0e0f4153918a22d802b415980974514139389 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Tue, 16 Jan 2007 12:56:58 +0100 Subject: [PATCH 093/326] [ARM] 4096/1: S3C24XX: change return code form s3c2410_gpio_getcfg() The s3c2410_gpio_getcfg() currently returns a value which is dependant on the GPIO no passed in. Now we have more generic constants it is sensible to use those as return codes so that any function dealing with >1 GPIO does not need to do it's own number processing. Since this function is only currently used in pm.c, it is easy to fixup (and correct pm.c to use the generic constants) Signed-off-by: Ben Dooks Signed-off-by: Russell King --- arch/arm/mach-s3c2410/gpio.c | 11 +++++++---- arch/arm/mach-s3c2410/pm.c | 7 +++---- 2 files changed, 10 insertions(+), 8 deletions(-) diff --git a/arch/arm/mach-s3c2410/gpio.c b/arch/arm/mach-s3c2410/gpio.c index d1740643a135..f6fb215bb48c 100644 --- a/arch/arm/mach-s3c2410/gpio.c +++ b/arch/arm/mach-s3c2410/gpio.c @@ -84,15 +84,18 @@ EXPORT_SYMBOL(s3c2410_gpio_cfgpin); unsigned int s3c2410_gpio_getcfg(unsigned int pin) { void __iomem *base = S3C24XX_GPIO_BASE(pin); - unsigned long mask; + unsigned long val = __raw_readl(base); if (pin < S3C2410_GPIO_BANKB) { - mask = 1 << S3C2410_GPIO_OFFSET(pin); + val >>= S3C2410_GPIO_OFFSET(pin); + val &= 1; + val += 1; } else { - mask = 3 << S3C2410_GPIO_OFFSET(pin)*2; + val >>= S3C2410_GPIO_OFFSET(pin)*2; + val &= 3; } - return __raw_readl(base) & mask; + return val | S3C2410_GPIO_INPUT; } EXPORT_SYMBOL(s3c2410_gpio_getcfg); diff --git a/arch/arm/mach-s3c2410/pm.c b/arch/arm/mach-s3c2410/pm.c index 00834097eb82..ebf294dd31da 100644 --- a/arch/arm/mach-s3c2410/pm.c +++ b/arch/arm/mach-s3c2410/pm.c @@ -451,15 +451,14 @@ static void s3c2410_pm_check_resume_pin(unsigned int pin, unsigned int irqoffs) irqstate = s3c_irqwake_eintmask & (1L<>= S3C2410_GPIO_OFFSET(pin)*2; if (!irqstate) { - if (pinstate == 0x02) + if (pinstate == S3C2410_GPIO_IRQ) DBG("Leaving IRQ %d (pin %d) enabled\n", irq, pin); } else { - if (pinstate == 0x02) { + if (pinstate == S3C2410_GPIO_IRQ) { DBG("Disabling IRQ %d (pin %d)\n", irq, pin); - s3c2410_gpio_cfgpin(pin, 0x00); + s3c2410_gpio_cfgpin(pin, S3C2410_GPIO_INPUT); } } } From 204ecae4e10c235e6987cb7b2809a665511ab174 Mon Sep 17 00:00:00 2001 From: Russell King Date: Tue, 16 Jan 2007 14:01:47 +0000 Subject: [PATCH 094/326] [ARM] Fix show_mem() for discontigmem show_mem() was assuming incorrectly that the mem_map for any node started at PFN 0. This is obviously wrong; fix it to take account of node_start_pfn. Signed-off-by: Russell King --- arch/arm/mm/init.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c index b5814b4b6f35..7760193e74cc 100644 --- a/arch/arm/mm/init.c +++ b/arch/arm/mm/init.c @@ -52,15 +52,18 @@ void show_mem(void) printk("Free swap: %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10)); for_each_online_node(node) { + pg_data_t *n = NODE_DATA(node); + struct page *map = n->node_mem_map - n->node_start_pfn; + for_each_nodebank (i,mi,node) { unsigned int pfn1, pfn2; struct page *page, *end; - pfn1 = mi->bank[i].start >> PAGE_SHIFT; - pfn2 = (mi->bank[i].size + mi->bank[i].start) >> PAGE_SHIFT; + pfn1 = __phys_to_pfn(mi->bank[i].start); + pfn2 = __phys_to_pfn(mi->bank[i].size + mi->bank[i].start); - page = NODE_MEM_MAP(node) + pfn1; - end = NODE_MEM_MAP(node) + pfn2; + page = map + pfn1; + end = map + pfn2; do { total++; From 87b865776d2510fc98cdf789c59b8980432ef93a Mon Sep 17 00:00:00 2001 From: Russell King Date: Tue, 16 Jan 2007 16:53:41 +0000 Subject: [PATCH 095/326] [ARM] Update mach-types Signed-off-by: Russell King --- arch/arm/tools/mach-types | 25 ++++++++++++++++++++++++- 1 file changed, 24 insertions(+), 1 deletion(-) diff --git a/arch/arm/tools/mach-types b/arch/arm/tools/mach-types index 8bcb838e5444..bd78058b7178 100644 --- a/arch/arm/tools/mach-types +++ b/arch/arm/tools/mach-types @@ -12,7 +12,7 @@ # # http://www.arm.linux.org.uk/developer/machines/?action=new # -# Last update: Thu Dec 7 17:19:20 2006 +# Last update: Tue Jan 16 16:52:56 2007 # # machine_is_xxx CONFIG_xxxx MACH_TYPE_xxx number # @@ -1219,3 +1219,26 @@ zevio_1020 MACH_ZEVIO_1020 ZEVIO_1020 1207 hitrack MACH_HITRACK HITRACK 1208 syme1 MACH_SYME1 SYME1 1209 syhl1 MACH_SYHL1 SYHL1 1210 +empca400 MACH_EMPCA400 EMPCA400 1211 +em7210 MACH_EM7210 EM7210 1212 +htchermes MACH_HTCHERMES HTCHERMES 1213 +eti_c1 MACH_ETI_C1 ETI_C1 1214 +mach_dep2410 MACH_MACH_DEP2410 MACH_DEP2410 1215 +ac100 MACH_AC100 AC100 1216 +sneetch MACH_SNEETCH SNEETCH 1217 +studentmate MACH_STUDENTMATE STUDENTMATE 1218 +zir2410 MACH_ZIR2410 ZIR2410 1219 +zir2413 MACH_ZIR2413 ZIR2413 1220 +dlonip3 MACH_DLONIP3 DLONIP3 1221 +instream MACH_INSTREAM INSTREAM 1222 +ambarella MACH_AMBARELLA AMBARELLA 1223 +nevis MACH_NEVIS NEVIS 1224 +htc_trinity MACH_HTC_TRINITY HTC_TRINITY 1225 +ql202b MACH_QL202B QL202B 1226 +vpac270 MACH_VPAC270 VPAC270 1227 +rd129 MACH_RD129 RD129 1228 +htcwizard MACH_HTCWIZARD HTCWIZARD 1229 +xscale_treo680 MACH_XSCALE_TREO680 XSCALE_TREO680 1230 +tecon_tmezon MACH_TECON_TMEZON TECON_TMEZON 1231 +zylonite MACH_ZYLONITE ZYLONITE 1233 +gene1270 MACH_GENE1270 GENE1270 1234 From 7f215abc69302dc027f024fe656e4841063e8fe8 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Thu, 18 Jan 2007 20:36:00 +0100 Subject: [PATCH 096/326] [ARM] 4100/1: iop3xx: fix cpu mask for iop333 cosmetic fix so iop333 is not reported as ixp46x iop333 cpuid = 0x69054210 Signed-off-by: Dan Williams Signed-off-by: Russell King --- arch/arm/mm/proc-xscale.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/mm/proc-xscale.S b/arch/arm/mm/proc-xscale.S index 490e11b34231..d29fe927ee9e 100644 --- a/arch/arm/mm/proc-xscale.S +++ b/arch/arm/mm/proc-xscale.S @@ -708,7 +708,7 @@ __8032x_proc_info: .type __8033x_proc_info,#object __8033x_proc_info: .long 0x69054010 - .long 0xffffff30 + .long 0xfffffd30 .long PMD_TYPE_SECT | \ PMD_SECT_BUFFERABLE | \ PMD_SECT_CACHEABLE | \ From d28122a5877cc40350fa801353fd5a9350563ec3 Mon Sep 17 00:00:00 2001 From: Russell King Date: Mon, 22 Jan 2007 18:59:42 +0000 Subject: [PATCH 097/326] [ARM] Fix AMBA serial drivers for non-first serial ports Using console=ttyAM1 or console=ttyAMA1 resulted in an oops during boot due to trying to drive the console before that port had been registered. Fix this by checking whether the port is present before allowing console setup to proceed. Signed-off-by: Russell King --- drivers/serial/amba-pl010.c | 2 ++ drivers/serial/amba-pl011.c | 2 ++ 2 files changed, 4 insertions(+) diff --git a/drivers/serial/amba-pl010.c b/drivers/serial/amba-pl010.c index 61db6973755a..f69bd097166e 100644 --- a/drivers/serial/amba-pl010.c +++ b/drivers/serial/amba-pl010.c @@ -589,6 +589,8 @@ static int __init pl010_console_setup(struct console *co, char *options) */ if (co->index >= UART_NR) co->index = 0; + if (!amba_ports[co->index]) + return -ENODEV; port = &amba_ports[co->index]->port; if (options) diff --git a/drivers/serial/amba-pl011.c b/drivers/serial/amba-pl011.c index 9a3b374b2a08..44639e71372a 100644 --- a/drivers/serial/amba-pl011.c +++ b/drivers/serial/amba-pl011.c @@ -661,6 +661,8 @@ static int __init pl011_console_setup(struct console *co, char *options) if (co->index >= UART_NR) co->index = 0; uap = amba_ports[co->index]; + if (!uap) + return -ENODEV; uap->port.uartclk = clk_get_rate(uap->clk); From d4e1c889c1ec547371227558e1da5f2f50c7dd5e Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Sun, 21 Jan 2007 20:08:33 +0100 Subject: [PATCH 098/326] [ARM] 4102/1: Allow for PHYS_OFFSET on any valid 2MiB address This patchs allows the offset to the first page of physical memory to be on any 2MB boundary whereas the previous code could only handle psysical offset to any 16MB boundary (0xNN000000) or any 1MB boundary below 0x01000000 (e.g. 0x00N00000). The problem is a consequence of the orr one-byte syntax, so we fix this and we can place the first bank of memory at 0x28e00000. I have also included an explicit check that disallow compilation when PHYS_OFFSET is not on a 2MiB boundary. head.S would be the proper place to have this at since this is the first file that attempts to use PHYS_OFFSET during compile. Signed-off-by: Linus Walleij Signed-off-by: Russell King --- arch/arm/kernel/head.S | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/arch/arm/kernel/head.S b/arch/arm/kernel/head.S index d994561816a1..cf495a3084b3 100644 --- a/arch/arm/kernel/head.S +++ b/arch/arm/kernel/head.S @@ -22,6 +22,10 @@ #include #include +#if (PHYS_OFFSET & 0x001fffff) +#error "PHYS_OFFSET must be at an even 2MiB boundary!" +#endif + #define KERNEL_RAM_VADDR (PAGE_OFFSET + TEXT_OFFSET) #define KERNEL_RAM_PADDR (PHYS_OFFSET + TEXT_OFFSET) @@ -251,7 +255,8 @@ __create_page_tables: * Then map first 1MB of ram in case it contains our boot params. */ add r0, r4, #PAGE_OFFSET >> 18 - orr r6, r7, #PHYS_OFFSET + orr r6, r7, #(PHYS_OFFSET & 0xff000000) + orr r6, r6, #(PHYS_OFFSET & 0x00e00000) str r6, [r0] #ifdef CONFIG_XIP_KERNEL From 7baced8a5923ce13d3d42d50a042a869092ab4e5 Mon Sep 17 00:00:00 2001 From: Matt Reimer Date: Wed, 24 Jan 2007 01:30:37 +0100 Subject: [PATCH 099/326] [ARM] 4106/1: S3C2410: typo fixes in register definitions The Trcd* bits of the S3C24xx BANKCON6 and BANKCON7 registers are misspelled in include/asm-arm/arch-s3c2410/regs-mem.h as Trdc*. Signed-off-by: Matt Reimer Acked-by: Ben Dooks Signed-off-by: Russell King --- include/asm-arm/arch-s3c2410/regs-mem.h | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/include/asm-arm/arch-s3c2410/regs-mem.h b/include/asm-arm/arch-s3c2410/regs-mem.h index 375dca50364e..e4d82341f7ba 100644 --- a/include/asm-arm/arch-s3c2410/regs-mem.h +++ b/include/asm-arm/arch-s3c2410/regs-mem.h @@ -133,10 +133,10 @@ #define S3C2410_BANKCON_SDRAM (0x3 << 15) /* next bits only for EDO DRAM in 6,7 */ -#define S3C2400_BANKCON_EDO_Trdc1 (0x00 << 4) -#define S3C2400_BANKCON_EDO_Trdc2 (0x01 << 4) -#define S3C2400_BANKCON_EDO_Trdc3 (0x02 << 4) -#define S3C2400_BANKCON_EDO_Trdc4 (0x03 << 4) +#define S3C2400_BANKCON_EDO_Trcd1 (0x00 << 4) +#define S3C2400_BANKCON_EDO_Trcd2 (0x01 << 4) +#define S3C2400_BANKCON_EDO_Trcd3 (0x02 << 4) +#define S3C2400_BANKCON_EDO_Trcd4 (0x03 << 4) /* CAS pulse width */ #define S3C2400_BANKCON_EDO_PULSE1 (0x00 << 3) @@ -153,9 +153,9 @@ #define S3C2400_BANKCON_EDO_SCANb11 (0x03 << 0) /* next bits only for SDRAM in 6,7 */ -#define S3C2410_BANKCON_Trdc2 (0x00 << 2) -#define S3C2410_BANKCON_Trdc3 (0x01 << 2) -#define S3C2410_BANKCON_Trdc4 (0x02 << 2) +#define S3C2410_BANKCON_Trcd2 (0x00 << 2) +#define S3C2410_BANKCON_Trcd3 (0x01 << 2) +#define S3C2410_BANKCON_Trcd4 (0x02 << 2) /* control column address select */ #define S3C2410_BANKCON_SCANb8 (0x00 << 0) From 48c35b2d245fffedadce62769aafea8ecf493d19 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Tue, 23 Jan 2007 21:30:14 +0300 Subject: [PATCH 100/326] [MIPS] There is no __GNUC_MAJOR__ Gcc major version number is in __GNUC__. As side effect fix checking with sparse if sparse was built with gcc 4.1 and mips cross-compiler is 3.4. Sparse will inherit version 4.1, __GNUC__ won't be filtered from "-dM -E -xc" output, sparse will pick only new major, effectively becoming gcc version 3.1 which is unsupported. Signed-off-by: Alexey Dobriyan Signed-off-by: Ralf Baechle --- arch/mips/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/mips/Makefile b/arch/mips/Makefile index d1b026a0337d..c68b5d3e5d18 100644 --- a/arch/mips/Makefile +++ b/arch/mips/Makefile @@ -623,7 +623,7 @@ LDFLAGS += -m $(ld-emul) ifdef CONFIG_MIPS CHECKFLAGS += $(shell $(CC) $(CFLAGS) -dM -E -xc /dev/null | \ - egrep -vw '__GNUC_(MAJOR|MINOR|PATCHLEVEL)__' | \ + egrep -vw '__GNUC_(|MINOR_|PATCHLEVEL_)_' | \ sed -e 's/^\#define /-D/' -e "s/ /='/" -e "s/$$/'/") ifdef CONFIG_64BIT CHECKFLAGS += -m64 From 89c07fd14fe857c223b042a857a08c3ea46b92eb Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Wed, 24 Jan 2007 12:31:47 +0000 Subject: [PATCH 101/326] [MIPS] Fix APM build Definitions for TIF_FREEZE and _TIF_FREEZE were missing. Signed-off-by: Ralf Baechle --- include/asm-mips/thread_info.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/include/asm-mips/thread_info.h b/include/asm-mips/thread_info.h index e475c45ea263..fbcda8204473 100644 --- a/include/asm-mips/thread_info.h +++ b/include/asm-mips/thread_info.h @@ -118,6 +118,7 @@ register struct thread_info *__current_thread_info __asm__("$28"); #define TIF_USEDFPU 16 /* FPU was used by this task this quantum (SMP) */ #define TIF_POLLING_NRFLAG 17 /* true if poll_idle() is polling TIF_NEED_RESCHED */ #define TIF_MEMDIE 18 +#define TIF_FREEZE 19 #define TIF_SYSCALL_TRACE 31 /* syscall trace active */ #define _TIF_SYSCALL_TRACE (1< Date: Fri, 19 Jan 2007 14:35:14 +0000 Subject: [PATCH 102/326] [MIPS] SMTC: Fix TLB sizing bug for TLB of 64 >= entries Signed-off-by: Ralf Baechle --- arch/mips/kernel/smtc.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/arch/mips/kernel/smtc.c b/arch/mips/kernel/smtc.c index 44238ab2fc99..c37e83b173e9 100644 --- a/arch/mips/kernel/smtc.c +++ b/arch/mips/kernel/smtc.c @@ -270,9 +270,12 @@ void smtc_configure_tlb(void) * of their initialization in smtc_cpu_setup(). */ - tlbsiz = tlbsiz & 0x3f; /* MIPS32 limits TLB indices to 64 */ - cpu_data[0].tlbsize = tlbsiz; + /* MIPS32 limits TLB indices to 64 */ + if (tlbsiz > 64) + tlbsiz = 64; + cpu_data[0].tlbsize = current_cpu_data.tlbsize = tlbsiz; smtc_status |= SMTC_TLB_SHARED; + local_flush_tlb_all(); printk("TLB of %d entry pairs shared by %d VPEs\n", tlbsiz, vpes); From ec43c01420fc1da8bf0b19f0ceb24d7d3c7f47f3 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Wed, 24 Jan 2007 19:23:21 +0000 Subject: [PATCH 103/326] [MIPS] SMTC: Fix module build by exporting symbol Signed-off-by: Ralf Baechle --- arch/mips/kernel/smtc.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/mips/kernel/smtc.c b/arch/mips/kernel/smtc.c index c37e83b173e9..6a857bf030b0 100644 --- a/arch/mips/kernel/smtc.c +++ b/arch/mips/kernel/smtc.c @@ -4,6 +4,7 @@ #include #include #include +#include #include #include @@ -1047,6 +1048,8 @@ void smtc_ipi_replay(void) } } +EXPORT_SYMBOL(smtc_ipi_replay); + void smtc_idle_loop_hook(void) { #ifdef SMTC_IDLE_HOOK_DEBUG From 3f3183709feb35c30ca5f649224d01921870f81c Mon Sep 17 00:00:00 2001 From: Yoichi Yuasa Date: Wed, 24 Jan 2007 22:22:06 +0900 Subject: [PATCH 104/326] [MIPS] vr41xx: need one more nop with mtc0_tlbw_hazard() NEC VR4111 and VR4121 need one more nop with mtc0_tlbw_hazard(). Signed-off-by: Yoichi Yuasa Signed-off-by: Ralf Baechle --- include/asm-mips/hazards.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/asm-mips/hazards.h b/include/asm-mips/hazards.h index 0fe02945feba..50073157a617 100644 --- a/include/asm-mips/hazards.h +++ b/include/asm-mips/hazards.h @@ -157,7 +157,7 @@ ASMMACRO(back_to_back_c0_hazard, * processors. */ ASMMACRO(mtc0_tlbw_hazard, - nop + nop; nop ) ASMMACRO(tlbw_use_hazard, nop; nop; nop From 80aac59ed5b3e5ae2132bc620ff5607929c6a304 Mon Sep 17 00:00:00 2001 From: Thiemo Seufer Date: Wed, 24 Jan 2007 01:29:15 +0000 Subject: [PATCH 105/326] [MIPS] Fix reported amount of freed memory - it's in kB not bytes While at it, change message on DEC for consistency. Signed-off-by: Thiemo Seufer Signed-off-by: Ralf Baechle --- arch/mips/dec/prom/memory.c | 2 +- arch/mips/mm/init.c | 3 ++- 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/arch/mips/dec/prom/memory.c b/arch/mips/dec/prom/memory.c index 3027ce782797..3aa01d268f2d 100644 --- a/arch/mips/dec/prom/memory.c +++ b/arch/mips/dec/prom/memory.c @@ -122,7 +122,7 @@ unsigned long __init prom_free_prom_memory(void) addr += PAGE_SIZE; } - printk("Freeing unused PROM memory: %ldk freed\n", + printk("Freeing unused PROM memory: %ldkb freed\n", (end - PAGE_SIZE) >> 10); return end - PAGE_SIZE; diff --git a/arch/mips/mm/init.c b/arch/mips/mm/init.c index 30245c09d025..49065c133ebf 100644 --- a/arch/mips/mm/init.c +++ b/arch/mips/mm/init.c @@ -501,7 +501,8 @@ void free_initmem(void) freed = prom_free_prom_memory(); if (freed) - printk(KERN_INFO "Freeing firmware memory: %ldk freed\n",freed); + printk(KERN_INFO "Freeing firmware memory: %ldkb freed\n", + freed >> 10); free_init_pages("unused kernel memory", __pa_symbol(&__init_begin), From 9cfdf6f15a2a462b6b328b4c60b252cfc04ec03e Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Wed, 24 Jan 2007 19:13:08 +0000 Subject: [PATCH 106/326] [MIPS] VPE loader: Initialize lists before they're actually being used ... kspd which due to makefile order happens to be initialized before the vpe loader causes references to vpecontrol lists before they're actually been initialized. Signed-off-by: Ralf Baechle --- arch/mips/kernel/vpe.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/arch/mips/kernel/vpe.c b/arch/mips/kernel/vpe.c index 666bef484dcb..458fccf87c54 100644 --- a/arch/mips/kernel/vpe.c +++ b/arch/mips/kernel/vpe.c @@ -139,13 +139,16 @@ struct tc { struct list_head list; }; -struct vpecontrol_ { +struct { /* Virtual processing elements */ struct list_head vpe_list; /* Thread contexts */ struct list_head tc_list; -} vpecontrol; +} vpecontrol = { + .vpe_list = LIST_HEAD_INIT(vpecontrol.vpe_list), + .tc_list = LIST_HEAD_INIT(vpecontrol.tc_list) +}; static void release_progmem(void *ptr); /* static __attribute_used__ void dump_vpe(struct vpe * v); */ @@ -1388,8 +1391,6 @@ static int __init vpe_module_init(void) /* dump_mtregs(); */ - INIT_LIST_HEAD(&vpecontrol.vpe_list); - INIT_LIST_HEAD(&vpecontrol.tc_list); val = read_c0_mvpconf0(); for (i = 0; i < ((val & MVPCONF0_PTC) + 1); i++) { From 66218da212bf141532d678a699f5789c78145ab1 Mon Sep 17 00:00:00 2001 From: Atsushi Nemoto Date: Wed, 24 Jan 2007 15:43:34 +0900 Subject: [PATCH 107/326] [MIPS] Fix wrong checksum calculation on 64-bit MIPS The commit 8e3d8433d8c22ca6c42cba4a67d300c39aae7822 ([NET]: MIPS checksum annotations and cleanups) broke 64-bit MIPS. The problem is the commit replaces some unsigned long with __be32. On 64bit MIPS, a __be32 (i.e. unsigned int) value is represented as a sign-extented 32-bit value in a 64-bit argument register. So the address 192.168.0.1 (0xc0a80001) is passed as 0xffffffffc0a80001 to csum_tcpudp_nofold() but the asm code in the function expects 0x00000000c0a80001, therefore it returns a wrong checksum. Explicit cast to unsigned long is needed to drop high 32bit. Signed-off-by: Atsushi Nemoto Signed-off-by: Ralf Baechle --- include/asm-mips/checksum.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/include/asm-mips/checksum.h b/include/asm-mips/checksum.h index 24cdcc6eaab8..20a81e1548f5 100644 --- a/include/asm-mips/checksum.h +++ b/include/asm-mips/checksum.h @@ -159,7 +159,8 @@ static inline __wsum csum_tcpudp_nofold(__be32 saddr, #endif " .set pop" : "=r" (sum) - : "0" (daddr), "r"(saddr), + : "0" ((__force unsigned long)daddr), + "r" ((__force unsigned long)saddr), #ifdef __MIPSEL__ "r" ((proto + len) << 8), #else From bde8f00ce64d9824a4f227c8594e335a1a10d044 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 24 Jan 2007 11:54:53 -0800 Subject: [PATCH 108/326] [PATCH] NFS: Fix Oops in rpc_call_sync() Fix the Oops in http://bugzilla.linux-nfs.org/show_bug.cgi?id=138 We shouldn't be calling rpc_release_task() for tasks that are not active. Signed-off-by: Trond Myklebust Signed-off-by: Linus Torvalds --- include/linux/sunrpc/sched.h | 1 - net/sunrpc/clnt.c | 8 +++----- net/sunrpc/sched.c | 3 ++- 3 files changed, 5 insertions(+), 7 deletions(-) diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h index 97c761652581..8b6ce60ea057 100644 --- a/include/linux/sunrpc/sched.h +++ b/include/linux/sunrpc/sched.h @@ -250,7 +250,6 @@ void rpc_init_task(struct rpc_task *task, struct rpc_clnt *clnt, int flags, const struct rpc_call_ops *ops, void *data); void rpc_put_task(struct rpc_task *); -void rpc_release_task(struct rpc_task *); void rpc_exit_task(struct rpc_task *); void rpc_release_calldata(const struct rpc_call_ops *, void *); void rpc_killall_tasks(struct rpc_clnt *); diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index aba528b9ae76..16c9fbc1db69 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -490,16 +490,14 @@ int rpc_call_sync(struct rpc_clnt *clnt, struct rpc_message *msg, int flags) /* Set up the call info struct and execute the task */ status = task->tk_status; - if (status != 0) { - rpc_release_task(task); + if (status != 0) goto out; - } atomic_inc(&task->tk_count); status = rpc_execute(task); if (status == 0) status = task->tk_status; - rpc_put_task(task); out: + rpc_put_task(task); rpc_restore_sigmask(&oldset); return status; } @@ -537,7 +535,7 @@ rpc_call_async(struct rpc_clnt *clnt, struct rpc_message *msg, int flags, if (status == 0) rpc_execute(task); else - rpc_release_task(task); + rpc_put_task(task); rpc_restore_sigmask(&oldset); return status; diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index 79bc4cdf5d48..fc083f0b3544 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -42,6 +42,7 @@ static mempool_t *rpc_buffer_mempool __read_mostly; static void __rpc_default_timer(struct rpc_task *task); static void rpciod_killall(void); static void rpc_async_schedule(struct work_struct *); +static void rpc_release_task(struct rpc_task *task); /* * RPC tasks sit here while waiting for conditions to improve. @@ -896,7 +897,7 @@ void rpc_put_task(struct rpc_task *task) } EXPORT_SYMBOL(rpc_put_task); -void rpc_release_task(struct rpc_task *task) +static void rpc_release_task(struct rpc_task *task) { #ifdef RPC_DEBUG BUG_ON(task->tk_magic != RPC_TASK_MAGIC_ID); From 717d44e849219781ced028a40fcc59d3e1f49e4c Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 24 Jan 2007 11:54:55 -0800 Subject: [PATCH 109/326] [PATCH] NFS: Fix races in nfs_revalidate_mapping() Prevent the call to invalidate_inode_pages2() from racing with file writes by taking the inode->i_mutex across the page cache flush and invalidate. Signed-off-by: Trond Myklebust Signed-off-by: Linus Torvalds --- fs/nfs/dir.c | 2 +- fs/nfs/inode.c | 97 +++++++++++++++++++++++++++++------------- fs/nfs/symlink.c | 4 +- include/linux/nfs_fs.h | 1 + 4 files changed, 72 insertions(+), 32 deletions(-) diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index dee3d6c0f194..d9ba8cb0ee75 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -532,7 +532,7 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir) lock_kernel(); - res = nfs_revalidate_mapping(inode, filp->f_mapping); + res = nfs_revalidate_mapping_nolock(inode, filp->f_mapping); if (res < 0) { unlock_kernel(); return res; diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 63e470279309..d83498282837 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -665,49 +665,86 @@ int nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) return __nfs_revalidate_inode(server, inode); } +static int nfs_invalidate_mapping_nolock(struct inode *inode, struct address_space *mapping) +{ + struct nfs_inode *nfsi = NFS_I(inode); + + if (mapping->nrpages != 0) { + int ret = invalidate_inode_pages2(mapping); + if (ret < 0) + return ret; + } + spin_lock(&inode->i_lock); + nfsi->cache_validity &= ~NFS_INO_INVALID_DATA; + if (S_ISDIR(inode->i_mode)) { + memset(nfsi->cookieverf, 0, sizeof(nfsi->cookieverf)); + /* This ensures we revalidate child dentries */ + nfsi->cache_change_attribute = jiffies; + } + spin_unlock(&inode->i_lock); + nfs_inc_stats(inode, NFSIOS_DATAINVALIDATE); + dfprintk(PAGECACHE, "NFS: (%s/%Ld) data cache invalidated\n", + inode->i_sb->s_id, (long long)NFS_FILEID(inode)); + return 0; +} + +static int nfs_invalidate_mapping(struct inode *inode, struct address_space *mapping) +{ + int ret = 0; + + mutex_lock(&inode->i_mutex); + if (NFS_I(inode)->cache_validity & NFS_INO_INVALID_DATA) { + ret = nfs_sync_mapping(mapping); + if (ret == 0) + ret = nfs_invalidate_mapping_nolock(inode, mapping); + } + mutex_unlock(&inode->i_mutex); + return ret; +} + +/** + * nfs_revalidate_mapping_nolock - Revalidate the pagecache + * @inode - pointer to host inode + * @mapping - pointer to mapping + */ +int nfs_revalidate_mapping_nolock(struct inode *inode, struct address_space *mapping) +{ + struct nfs_inode *nfsi = NFS_I(inode); + int ret = 0; + + if ((nfsi->cache_validity & NFS_INO_REVAL_PAGECACHE) + || nfs_attribute_timeout(inode) || NFS_STALE(inode)) { + ret = __nfs_revalidate_inode(NFS_SERVER(inode), inode); + if (ret < 0) + goto out; + } + if (nfsi->cache_validity & NFS_INO_INVALID_DATA) + ret = nfs_invalidate_mapping_nolock(inode, mapping); +out: + return ret; +} + /** * nfs_revalidate_mapping - Revalidate the pagecache * @inode - pointer to host inode * @mapping - pointer to mapping + * + * This version of the function will take the inode->i_mutex and attempt to + * flush out all dirty data if it needs to invalidate the page cache. */ int nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping) { struct nfs_inode *nfsi = NFS_I(inode); int ret = 0; - if (NFS_STALE(inode)) - ret = -ESTALE; if ((nfsi->cache_validity & NFS_INO_REVAL_PAGECACHE) - || nfs_attribute_timeout(inode)) + || nfs_attribute_timeout(inode) || NFS_STALE(inode)) { ret = __nfs_revalidate_inode(NFS_SERVER(inode), inode); - if (ret < 0) - goto out; - - if (nfsi->cache_validity & NFS_INO_INVALID_DATA) { - if (mapping->nrpages != 0) { - if (S_ISREG(inode->i_mode)) { - ret = nfs_sync_mapping(mapping); - if (ret < 0) - goto out; - } - ret = invalidate_inode_pages2(mapping); - if (ret < 0) - goto out; - } - spin_lock(&inode->i_lock); - nfsi->cache_validity &= ~NFS_INO_INVALID_DATA; - if (S_ISDIR(inode->i_mode)) { - memset(nfsi->cookieverf, 0, sizeof(nfsi->cookieverf)); - /* This ensures we revalidate child dentries */ - nfsi->cache_change_attribute = jiffies; - } - spin_unlock(&inode->i_lock); - - nfs_inc_stats(inode, NFSIOS_DATAINVALIDATE); - dfprintk(PAGECACHE, "NFS: (%s/%Ld) data cache invalidated\n", - inode->i_sb->s_id, - (long long)NFS_FILEID(inode)); + if (ret < 0) + goto out; } + if (nfsi->cache_validity & NFS_INO_INVALID_DATA) + ret = nfs_invalidate_mapping(inode, mapping); out: return ret; } diff --git a/fs/nfs/symlink.c b/fs/nfs/symlink.c index 6c686112cc03..525c136c7d8c 100644 --- a/fs/nfs/symlink.c +++ b/fs/nfs/symlink.c @@ -50,7 +50,9 @@ static void *nfs_follow_link(struct dentry *dentry, struct nameidata *nd) { struct inode *inode = dentry->d_inode; struct page *page; - void *err = ERR_PTR(nfs_revalidate_mapping(inode, inode->i_mapping)); + void *err; + + err = ERR_PTR(nfs_revalidate_mapping_nolock(inode, inode->i_mapping)); if (err) goto read_failed; page = read_cache_page(&inode->i_data, 0, diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 04963063e620..c5d4084773e8 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -308,6 +308,7 @@ extern int nfs_attribute_timeout(struct inode *inode); extern int nfs_revalidate_inode(struct nfs_server *server, struct inode *inode); extern int __nfs_revalidate_inode(struct nfs_server *, struct inode *); extern int nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping); +extern int nfs_revalidate_mapping_nolock(struct inode *inode, struct address_space *mapping); extern int nfs_setattr(struct dentry *, struct iattr *); extern void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr); extern void nfs_begin_attr_update(struct inode *); From 6640e69731b42fd5e3d2b26201c8b34fc897a0ee Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Wed, 24 Jan 2007 14:42:04 -0800 Subject: [PATCH 110/326] [IPV4]: Fix the fib trie iterator to work with a single entry routing tables In a kernel with trie routing enabled I had a simple routing setup with only a single route to the outside world and no default route. "ip route table list main" showed my the route just fine but /proc/net/route was an empty file. What was going on? Thinking it was a bug in something I did and I looked deeper. Eventually I setup a second route and everything looked correct, huh? Finally I realized that the it was just the iterator pair in fib_trie_get_first, fib_trie_get_next just could not handle a routing table with a single entry. So to save myself and others further confusion, here is a simple fix for the fib proc iterator so it works even when there is only a single route in a routing table. Signed-off-by: Eric W. Biederman Signed-off-by: Robert Olsson Signed-off-by: David S. Miller --- net/ipv4/fib_trie.c | 21 ++++++++++++++++----- 1 file changed, 16 insertions(+), 5 deletions(-) diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index cfb249cc0a58..13307c04d5a1 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -1989,6 +1989,10 @@ static struct node *fib_trie_get_next(struct fib_trie_iter *iter) unsigned cindex = iter->index; struct tnode *p; + /* A single entry routing table */ + if (!tn) + return NULL; + pr_debug("get_next iter={node=%p index=%d depth=%d}\n", iter->tnode, iter->index, iter->depth); rescan: @@ -2037,11 +2041,18 @@ static struct node *fib_trie_get_first(struct fib_trie_iter *iter, if(!iter) return NULL; - if (n && IS_TNODE(n)) { - iter->tnode = (struct tnode *) n; - iter->trie = t; - iter->index = 0; - iter->depth = 1; + if (n) { + if (IS_TNODE(n)) { + iter->tnode = (struct tnode *) n; + iter->trie = t; + iter->index = 0; + iter->depth = 1; + } else { + iter->tnode = NULL; + iter->trie = t; + iter->index = 0; + iter->depth = 0; + } return n; } return NULL; From dbcb5855d108b7fa20ab42567a5412ce9dcd776a Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 24 Jan 2007 15:21:02 -0800 Subject: [PATCH 111/326] [AF_PACKET]: Fix BPF handling. This fixes a bug introduced by: commit fda9ef5d679b07c9d9097aaf6ef7f069d794a8f9 Author: Dmitry Mishin Date: Thu Aug 31 15:28:39 2006 -0700 [NET]: Fix sk->sk_filter field access sk_run_filter() returns either 0 or an unsigned 32-bit length which says how much of the packet to retain. If that 32-bit unsigned integer is larger than the packet, this is fine we just leave the packet unchanged. The above commit caused all filter return values which were negative when interpreted as a signed integer to indicate a packet drop, which is wrong. Based upon a report and initial patch by Raivis Bucis. Signed-off-by: David S. Miller --- net/packet/af_packet.c | 30 +++++++++++++++--------------- 1 file changed, 15 insertions(+), 15 deletions(-) diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index da73e8a8c18d..594c078c5ebc 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -428,24 +428,18 @@ static int packet_sendmsg_spkt(struct kiocb *iocb, struct socket *sock, } #endif -static inline int run_filter(struct sk_buff *skb, struct sock *sk, - unsigned *snaplen) +static inline unsigned int run_filter(struct sk_buff *skb, struct sock *sk, + unsigned int res) { struct sk_filter *filter; - int err = 0; rcu_read_lock_bh(); filter = rcu_dereference(sk->sk_filter); - if (filter != NULL) { - err = sk_run_filter(skb, filter->insns, filter->len); - if (!err) - err = -EPERM; - else if (*snaplen > err) - *snaplen = err; - } + if (filter != NULL) + res = sk_run_filter(skb, filter->insns, filter->len); rcu_read_unlock_bh(); - return err; + return res; } /* @@ -467,7 +461,7 @@ static int packet_rcv(struct sk_buff *skb, struct net_device *dev, struct packet struct packet_sock *po; u8 * skb_head = skb->data; int skb_len = skb->len; - unsigned snaplen; + unsigned int snaplen, res; if (skb->pkt_type == PACKET_LOOPBACK) goto drop; @@ -495,8 +489,11 @@ static int packet_rcv(struct sk_buff *skb, struct net_device *dev, struct packet snaplen = skb->len; - if (run_filter(skb, sk, &snaplen) < 0) + res = run_filter(skb, sk, snaplen); + if (!res) goto drop_n_restore; + if (snaplen > res) + snaplen = res; if (atomic_read(&sk->sk_rmem_alloc) + skb->truesize >= (unsigned)sk->sk_rcvbuf) @@ -568,7 +565,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe struct tpacket_hdr *h; u8 * skb_head = skb->data; int skb_len = skb->len; - unsigned snaplen; + unsigned int snaplen, res; unsigned long status = TP_STATUS_LOSING|TP_STATUS_USER; unsigned short macoff, netoff; struct sk_buff *copy_skb = NULL; @@ -592,8 +589,11 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe snaplen = skb->len; - if (run_filter(skb, sk, &snaplen) < 0) + res = run_filter(skb, sk, snaplen); + if (!res) goto drop_n_restore; + if (snaplen > res) + snaplen = res; if (sk->sk_type == SOCK_DGRAM) { macoff = netoff = TPACKET_ALIGN(TPACKET_HDRLEN) + 16; From 6a40da02be1e1d74eee653c6d181934d35cbca7d Mon Sep 17 00:00:00 2001 From: Alan Date: Wed, 24 Jan 2007 11:49:03 +0000 Subject: [PATCH 112/326] libata cmd64x: whack into a shape that looks like the documentation Signed-off-by: Alan Cox Signed-off-by: Jeff Garzik --- drivers/ata/pata_cmd64x.c | 23 ++++++++++++++++------- 1 file changed, 16 insertions(+), 7 deletions(-) diff --git a/drivers/ata/pata_cmd64x.c b/drivers/ata/pata_cmd64x.c index 15841a563694..449162cbf93e 100644 --- a/drivers/ata/pata_cmd64x.c +++ b/drivers/ata/pata_cmd64x.c @@ -197,7 +197,7 @@ static void cmd64x_set_piomode(struct ata_port *ap, struct ata_device *adev) static void cmd64x_set_dmamode(struct ata_port *ap, struct ata_device *adev) { static const u8 udma_data[] = { - 0x31, 0x21, 0x11, 0x25, 0x15, 0x05 + 0x30, 0x20, 0x10, 0x20, 0x10, 0x00 }; static const u8 mwdma_data[] = { 0x30, 0x20, 0x10 @@ -213,12 +213,21 @@ static void cmd64x_set_dmamode(struct ata_port *ap, struct ata_device *adev) pci_read_config_byte(pdev, pciD, ®D); pci_read_config_byte(pdev, pciU, ®U); - regD &= ~(0x20 << shift); - regU &= ~(0x35 << shift); + /* DMA bits off */ + regD &= ~(0x20 << adev->devno); + /* DMA control bits */ + regU &= ~(0x30 << shift); + /* DMA timing bits */ + regU &= ~(0x05 << adev->devno); - if (adev->dma_mode >= XFER_UDMA_0) + if (adev->dma_mode >= XFER_UDMA_0) { + /* Merge thge timing value */ regU |= udma_data[adev->dma_mode - XFER_UDMA_0] << shift; - else + /* Merge the control bits */ + regU |= 1 << adev->devno; /* UDMA on */ + if (adev->dma_mode > 2) /* 15nS timing */ + regU |= 4 << adev->devno; + } else regD |= mwdma_data[adev->dma_mode - XFER_MW_DMA_0] << shift; regD |= 0x20 << adev->devno; @@ -239,8 +248,8 @@ static void cmd648_bmdma_stop(struct ata_queued_cmd *qc) struct ata_port *ap = qc->ap; struct pci_dev *pdev = to_pci_dev(ap->host->dev); u8 dma_intr; - int dma_reg = ap->port_no ? ARTTIM23_INTR_CH1 : CFR_INTR_CH0; - int dma_mask = ap->port_no ? ARTTIM2 : CFR; + int dma_mask = ap->port_no ? ARTTIM23_INTR_CH1 : CFR_INTR_CH0; + int dma_reg = ap->port_no ? ARTTIM2 : CFR; ata_bmdma_stop(qc); From a52865c239b1bc4f62e387509b1ad3415e476ee5 Mon Sep 17 00:00:00 2001 From: Alan Date: Wed, 24 Jan 2007 11:51:38 +0000 Subject: [PATCH 113/326] libata hpt3xn: Hopefully sort out the DPLL logic versus the vendor code Rather than ending up with two layers of negation jut rename the variable and lose one. Signed-off-by: Alan Cox Signed-off-by: Jeff Garzik --- drivers/ata/pata_hpt3x2n.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/ata/pata_hpt3x2n.c b/drivers/ata/pata_hpt3x2n.c index f6817b4093a4..886fab9aa62c 100644 --- a/drivers/ata/pata_hpt3x2n.c +++ b/drivers/ata/pata_hpt3x2n.c @@ -25,7 +25,7 @@ #include #define DRV_NAME "pata_hpt3x2n" -#define DRV_VERSION "0.3" +#define DRV_VERSION "0.3.2" enum { HPT_PCI_FAST = (1 << 31), @@ -297,11 +297,11 @@ static int hpt3x2n_pair_idle(struct ata_port *ap) return 0; } -static int hpt3x2n_use_dpll(struct ata_port *ap, int reading) +static int hpt3x2n_use_dpll(struct ata_port *ap, int writing) { long flags = (long)ap->host->private_data; /* See if we should use the DPLL */ - if (reading == 0) + if (writing) return USE_DPLL; /* Needed for write */ if (flags & PCI66) return USE_DPLL; /* Needed at 66Mhz */ From b229a7b0aed808f2ef6a5e9dbf78b0f17cefb4d0 Mon Sep 17 00:00:00 2001 From: Alan Date: Wed, 24 Jan 2007 11:47:07 +0000 Subject: [PATCH 114/326] libata: set_mode, Fix the FIXME When set_mode() changed ->set_mode didn't adapt. This makes the needed changes and removes the relevant FIXME case. Signed-off-by: Alan Cox Signed-off-by: Jeff Garzik --- drivers/ata/ata_generic.c | 6 ++++-- drivers/ata/libata-core.c | 14 ++------------ drivers/ata/pata_it821x.c | 4 +++- drivers/ata/pata_ixp4xx_cf.c | 5 +++-- drivers/ata/pata_legacy.c | 4 +++- drivers/ata/pata_rz1000.c | 6 ++++-- include/linux/libata.h | 4 ++-- 7 files changed, 21 insertions(+), 22 deletions(-) diff --git a/drivers/ata/ata_generic.c b/drivers/ata/ata_generic.c index 908751d27e76..24af56081b5d 100644 --- a/drivers/ata/ata_generic.c +++ b/drivers/ata/ata_generic.c @@ -64,6 +64,7 @@ static void generic_error_handler(struct ata_port *ap) /** * generic_set_mode - mode setting * @ap: interface to set up + * @unused: returned device on error * * Use a non standard set_mode function. We don't want to be tuned. * The BIOS configured everything. Our job is not to fiddle. We @@ -71,7 +72,7 @@ static void generic_error_handler(struct ata_port *ap) * and respect them. */ -static void generic_set_mode(struct ata_port *ap) +static int generic_set_mode(struct ata_port *ap, struct ata_device **unused) { int dma_enabled = 0; int i; @@ -82,7 +83,7 @@ static void generic_set_mode(struct ata_port *ap) for (i = 0; i < ATA_MAX_DEVICES; i++) { struct ata_device *dev = &ap->device[i]; - if (ata_dev_enabled(dev)) { + if (ata_dev_ready(dev)) { /* We don't really care */ dev->pio_mode = XFER_PIO_0; dev->dma_mode = XFER_MW_DMA_0; @@ -99,6 +100,7 @@ static void generic_set_mode(struct ata_port *ap) } } } + return 0; } static struct scsi_host_template generic_sht = { diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 0d51d13b16bf..a388a8df0043 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -2431,18 +2431,8 @@ int ata_set_mode(struct ata_port *ap, struct ata_device **r_failed_dev) int i, rc = 0, used_dma = 0, found = 0; /* has private set_mode? */ - if (ap->ops->set_mode) { - /* FIXME: make ->set_mode handle no device case and - * return error code and failing device on failure. - */ - for (i = 0; i < ATA_MAX_DEVICES; i++) { - if (ata_dev_ready(&ap->device[i])) { - ap->ops->set_mode(ap); - break; - } - } - return 0; - } + if (ap->ops->set_mode) + return ap->ops->set_mode(ap, r_failed_dev); /* step 1: calculate xfer_mask */ for (i = 0; i < ATA_MAX_DEVICES; i++) { diff --git a/drivers/ata/pata_it821x.c b/drivers/ata/pata_it821x.c index 0b56ff3d1cfe..e8afd486434a 100644 --- a/drivers/ata/pata_it821x.c +++ b/drivers/ata/pata_it821x.c @@ -476,6 +476,7 @@ static unsigned int it821x_passthru_qc_issue_prot(struct ata_queued_cmd *qc) /** * it821x_smart_set_mode - mode setting * @ap: interface to set up + * @unused: device that failed (error only) * * Use a non standard set_mode function. We don't want to be tuned. * The BIOS configured everything. Our job is not to fiddle. We @@ -483,7 +484,7 @@ static unsigned int it821x_passthru_qc_issue_prot(struct ata_queued_cmd *qc) * and respect them. */ -static void it821x_smart_set_mode(struct ata_port *ap) +static int it821x_smart_set_mode(struct ata_port *ap, struct ata_device **unused) { int dma_enabled = 0; int i; @@ -512,6 +513,7 @@ static void it821x_smart_set_mode(struct ata_port *ap) } } } + return 0; } /** diff --git a/drivers/ata/pata_ixp4xx_cf.c b/drivers/ata/pata_ixp4xx_cf.c index cb8924109f59..23b8aab3ebd8 100644 --- a/drivers/ata/pata_ixp4xx_cf.c +++ b/drivers/ata/pata_ixp4xx_cf.c @@ -23,9 +23,9 @@ #include #define DRV_NAME "pata_ixp4xx_cf" -#define DRV_VERSION "0.1.1" +#define DRV_VERSION "0.1.1ac1" -static void ixp4xx_set_mode(struct ata_port *ap) +static int ixp4xx_set_mode(struct ata_port *ap, struct ata_device *adev) { int i; @@ -38,6 +38,7 @@ static void ixp4xx_set_mode(struct ata_port *ap) dev->flags |= ATA_DFLAG_PIO; } } + return 0; } static void ixp4xx_phy_reset(struct ata_port *ap) diff --git a/drivers/ata/pata_legacy.c b/drivers/ata/pata_legacy.c index e7bf9d89c8ee..581cb33c6f45 100644 --- a/drivers/ata/pata_legacy.c +++ b/drivers/ata/pata_legacy.c @@ -96,6 +96,7 @@ static int pio_mask = 0x1F; /* PIO range for autospeed devices */ /** * legacy_set_mode - mode setting * @ap: IDE interface + * @unused: Device that failed when error is returned * * Use a non standard set_mode function. We don't want to be tuned. * @@ -105,7 +106,7 @@ static int pio_mask = 0x1F; /* PIO range for autospeed devices */ * expand on this as per hdparm in the base kernel. */ -static void legacy_set_mode(struct ata_port *ap) +static int legacy_set_mode(struct ata_port *ap, struct ata_device **unused) { int i; @@ -118,6 +119,7 @@ static void legacy_set_mode(struct ata_port *ap) dev->flags |= ATA_DFLAG_PIO; } } + return 0; } static struct scsi_host_template legacy_sht = { diff --git a/drivers/ata/pata_rz1000.c b/drivers/ata/pata_rz1000.c index adf4cc134f25..cec0729225e1 100644 --- a/drivers/ata/pata_rz1000.c +++ b/drivers/ata/pata_rz1000.c @@ -52,19 +52,20 @@ static void rz1000_error_handler(struct ata_port *ap) /** * rz1000_set_mode - mode setting function * @ap: ATA interface + * @unused: returned device on set_mode failure * * Use a non standard set_mode function. We don't want to be tuned. We * would prefer to be BIOS generic but for the fact our hardware is * whacked out. */ -static void rz1000_set_mode(struct ata_port *ap) +static int rz1000_set_mode(struct ata_port *ap, struct ata_device **unused) { int i; for (i = 0; i < ATA_MAX_DEVICES; i++) { struct ata_device *dev = &ap->device[i]; - if (ata_dev_enabled(dev)) { + if (ata_dev_ready(dev)) { /* We don't really care */ dev->pio_mode = XFER_PIO_0; dev->xfer_mode = XFER_PIO_0; @@ -72,6 +73,7 @@ static void rz1000_set_mode(struct ata_port *ap) dev->flags |= ATA_DFLAG_PIO; } } + return 0; } diff --git a/include/linux/libata.h b/include/linux/libata.h index f7f268e38749..df76fc467816 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -612,11 +612,11 @@ struct ata_port_operations { void (*dev_select)(struct ata_port *ap, unsigned int device); void (*phy_reset) (struct ata_port *ap); /* obsolete */ - void (*set_mode) (struct ata_port *ap); + int (*set_mode) (struct ata_port *ap, struct ata_device **r_failed_dev); void (*post_set_mode) (struct ata_port *ap); - int (*check_atapi_dma) (struct ata_queued_cmd *qc); + int (*check_atapi_dma) (struct ata_queued_cmd *qc); void (*bmdma_setup) (struct ata_queued_cmd *qc); void (*bmdma_start) (struct ata_queued_cmd *qc); From 99abfeafb5f2eea1bb481330ff37343e1133c924 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 24 Jan 2007 18:19:28 -0800 Subject: [PATCH 115/326] Linux 2.6.20-rc6 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 477f52e3c7f6..9e1adac8aa28 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 2 PATCHLEVEL = 6 SUBLEVEL = 20 -EXTRAVERSION =-rc5 +EXTRAVERSION =-rc6 NAME = Homicidal Dwarf Hamster # *DOCUMENTATION* From 412489af76b5c0e4029d4406d93554c22a88fc73 Mon Sep 17 00:00:00 2001 From: Catalin Marinas Date: Thu, 25 Jan 2007 14:16:47 +0100 Subject: [PATCH 116/326] [ARM] 4112/1: Only ioremap to supersections if DOMAIN_IO is zero Supersections do not have a field for the domain and it is always 0. This patch prevents the creation of supersections during ioremap when DOMAIN_IO is not zero (i.e. !defined(CONFIG_IO_36)). Signed-off-by: Catalin Marinas Signed-off-by: Russell King --- arch/arm/mm/ioremap.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/arm/mm/ioremap.c b/arch/arm/mm/ioremap.c index 251685fe73a8..0ac615c0f798 100644 --- a/arch/arm/mm/ioremap.c +++ b/arch/arm/mm/ioremap.c @@ -300,7 +300,8 @@ __ioremap_pfn(unsigned long pfn, unsigned long offset, size_t size, addr = (unsigned long)area->addr; #ifndef CONFIG_SMP - if ((((cpu_architecture() >= CPU_ARCH_ARMv6) && (get_cr() & CR_XP)) || + if (DOMAIN_IO == 0 && + (((cpu_architecture() >= CPU_ARCH_ARMv6) && (get_cr() & CR_XP)) || cpu_is_xsc3()) && !((__pfn_to_phys(pfn) | size | addr) & ~SUPERSECTION_MASK)) { area->flags |= VM_ARM_SECTION_MAPPING; From c6428464894889e110418928e6b37dc2eb4cee56 Mon Sep 17 00:00:00 2001 From: Catalin Marinas Date: Wed, 24 Jan 2007 18:47:08 +0100 Subject: [PATCH 117/326] [ARM] 4111/1: Allow VFP to work with thread migration on SMP The current lazy saving of the VFP registers is no longer possible with thread migration on SMP. This patch implements a per-CPU vfp-state pointer and the saving of the VFP registers at every context switch. The registers restoring is still performed in a lazy way. Signed-off-by: Catalin Marinas Signed-off-by: Russell King --- arch/arm/vfp/entry.S | 1 + arch/arm/vfp/vfp.h | 4 ++++ arch/arm/vfp/vfphw.S | 26 ++++++++++++++++++++++++-- arch/arm/vfp/vfpmodule.c | 30 ++++++++++++++++++++++++++---- include/asm-arm/fpstate.h | 3 +++ 5 files changed, 58 insertions(+), 6 deletions(-) diff --git a/arch/arm/vfp/entry.S b/arch/arm/vfp/entry.S index 7b595547c1c8..ca2a5ad19ea6 100644 --- a/arch/arm/vfp/entry.S +++ b/arch/arm/vfp/entry.S @@ -25,6 +25,7 @@ do_vfp: enable_irq ldr r4, .LCvfp + ldr r11, [r10, #TI_CPU] @ CPU number add r10, r10, #TI_VFPSTATE @ r10 = workspace ldr pc, [r4] @ call VFP entry point diff --git a/arch/arm/vfp/vfp.h b/arch/arm/vfp/vfp.h index f2797896e6d5..54a2ad6d9ca2 100644 --- a/arch/arm/vfp/vfp.h +++ b/arch/arm/vfp/vfp.h @@ -370,3 +370,7 @@ struct op { u32 (* const fn)(int dd, int dn, int dm, u32 fpscr); u32 flags; }; + +#ifdef CONFIG_SMP +extern void vfp_save_state(void *location, u32 fpexc); +#endif diff --git a/arch/arm/vfp/vfphw.S b/arch/arm/vfp/vfphw.S index e51e6679c402..d4b7b229631d 100644 --- a/arch/arm/vfp/vfphw.S +++ b/arch/arm/vfp/vfphw.S @@ -65,6 +65,7 @@ @ r2 = faulted PC+4 @ r9 = successful return @ r10 = vfp_state union +@ r11 = CPU number @ lr = failure return .globl vfp_support_entry @@ -79,7 +80,7 @@ vfp_support_entry: DBGSTR1 "enable %x", r10 ldr r3, last_VFP_context_address orr r1, r1, #FPEXC_ENABLE @ user FPEXC has the enable bit set - ldr r4, [r3] @ last_VFP_context pointer + ldr r4, [r3, r11, lsl #2] @ last_VFP_context pointer bic r5, r1, #FPEXC_EXCEPTION @ make sure exceptions are disabled cmp r4, r10 beq check_for_exception @ we are returning to the same @@ -91,7 +92,9 @@ vfp_support_entry: @ exceptions, so we can get at the @ rest of it +#ifndef CONFIG_SMP @ Save out the current registers to the old thread state + @ No need for SMP since this is not done lazily DBGSTR1 "save old state %p", r4 cmp r4, #0 @@ -105,10 +108,11 @@ vfp_support_entry: stmia r4, {r1, r5, r6, r8} @ save FPEXC, FPSCR, FPINST, FPINST2 @ and point r4 at the word at the @ start of the register dump +#endif no_old_VFP_process: DBGSTR1 "load state %p", r10 - str r10, [r3] @ update the last_VFP_context pointer + str r10, [r3, r11, lsl #2] @ update the last_VFP_context pointer @ Load the saved state back into the VFP VFPFLDMIA r10 @ reload the working registers while @ FPEXC is in a safe state @@ -162,6 +166,24 @@ process_exception: @ required. If not, the user code will @ retry the faulted instruction +#ifdef CONFIG_SMP + .globl vfp_save_state + .type vfp_save_state, %function +vfp_save_state: + @ Save the current VFP state + @ r0 - save location + @ r1 - FPEXC + DBGSTR1 "save VFP state %p", r0 + VFPFMRX r2, FPSCR @ current status + VFPFMRX r3, FPINST @ FPINST (always there, rev0 onwards) + tst r1, #FPEXC_FPV2 @ is there an FPINST2 to read? + VFPFMRX r12, FPINST2, NE @ FPINST2 if needed - avoids reading + @ nonexistant reg on rev0 + VFPFSTMIA r0 @ save the working registers + stmia r0, {r1, r2, r3, r12} @ save FPEXC, FPSCR, FPINST, FPINST2 + mov pc, lr +#endif + last_VFP_context_address: .word last_VFP_context diff --git a/arch/arm/vfp/vfpmodule.c b/arch/arm/vfp/vfpmodule.c index 490d9d18a7d1..f1e5951dc721 100644 --- a/arch/arm/vfp/vfpmodule.c +++ b/arch/arm/vfp/vfpmodule.c @@ -28,7 +28,7 @@ void vfp_testing_entry(void); void vfp_support_entry(void); void (*vfp_vector)(void) = vfp_testing_entry; -union vfp_state *last_VFP_context; +union vfp_state *last_VFP_context[NR_CPUS]; /* * Dual-use variable. @@ -41,13 +41,35 @@ static int vfp_notifier(struct notifier_block *self, unsigned long cmd, void *v) { struct thread_info *thread = v; union vfp_state *vfp; + __u32 cpu = thread->cpu; if (likely(cmd == THREAD_NOTIFY_SWITCH)) { + u32 fpexc = fmrx(FPEXC); + +#ifdef CONFIG_SMP + /* + * On SMP, if VFP is enabled, save the old state in + * case the thread migrates to a different CPU. The + * restoring is done lazily. + */ + if ((fpexc & FPEXC_ENABLE) && last_VFP_context[cpu]) { + vfp_save_state(last_VFP_context[cpu], fpexc); + last_VFP_context[cpu]->hard.cpu = cpu; + } + /* + * Thread migration, just force the reloading of the + * state on the new CPU in case the VFP registers + * contain stale data. + */ + if (thread->vfpstate.hard.cpu != cpu) + last_VFP_context[cpu] = NULL; +#endif + /* * Always disable VFP so we can lazily save/restore the * old state. */ - fmxr(FPEXC, fmrx(FPEXC) & ~FPEXC_ENABLE); + fmxr(FPEXC, fpexc & ~FPEXC_ENABLE); return NOTIFY_DONE; } @@ -68,8 +90,8 @@ static int vfp_notifier(struct notifier_block *self, unsigned long cmd, void *v) } /* flush and release case: Per-thread VFP cleanup. */ - if (last_VFP_context == vfp) - last_VFP_context = NULL; + if (last_VFP_context[cpu] == vfp) + last_VFP_context[cpu] = NULL; return NOTIFY_DONE; } diff --git a/include/asm-arm/fpstate.h b/include/asm-arm/fpstate.h index 6af4e6bd1290..f31cda5a55ee 100644 --- a/include/asm-arm/fpstate.h +++ b/include/asm-arm/fpstate.h @@ -35,6 +35,9 @@ struct vfp_hard_struct { */ __u32 fpinst; __u32 fpinst2; +#ifdef CONFIG_SMP + __u32 cpu; +#endif }; union vfp_state { From 767fe7877c2928b5633992ee60a49ad4516dc2af Mon Sep 17 00:00:00 2001 From: Jiri Kosina Date: Wed, 24 Jan 2007 23:05:07 +0100 Subject: [PATCH 118/326] HID: fix memleaking of collection hid_free_device() doesn't free device->collection (but it does free device->rdesc and device itself). This imposes memory leak. Fix it. Signed-off-by: Jiri Kosina --- drivers/hid/hid-core.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/hid/hid-core.c b/drivers/hid/hid-core.c index b8cf50fcd64d..49f18f5b2514 100644 --- a/drivers/hid/hid-core.c +++ b/drivers/hid/hid-core.c @@ -543,6 +543,7 @@ void hid_free_device(struct hid_device *device) } kfree(device->rdesc); + kfree(device->collection); kfree(device); } EXPORT_SYMBOL_GPL(hid_free_device); From ad2905f06512c70c44a7efd178536ad197c48528 Mon Sep 17 00:00:00 2001 From: Jiri Kosina Date: Tue, 23 Jan 2007 14:02:53 +0100 Subject: [PATCH 119/326] USB HID: fix hid_blacklist clash for 0x08ca/0x0010 commit d8c8a393166d6283003fb111d0b4a40931c0eda4 introduced a clash in hid_blacklist for 0x08ca/0x0010 (GTCO vs. AIPTEK). As the vendor of GTCO device doesn't seem to be interested in supporting their legacy HW with this conflicting ids, it is OK to remove it. Signed-off-by: Jiri Kosina --- drivers/usb/input/hid-core.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c index ea3636d96e1b..b86480457a2c 100644 --- a/drivers/usb/input/hid-core.c +++ b/drivers/usb/input/hid-core.c @@ -582,7 +582,6 @@ void usbhid_init_reports(struct hid_device *hid) } #define USB_VENDOR_ID_GTCO 0x078c -#define USB_VENDOR_ID_GTCO_IPANEL_1 0x08ca #define USB_VENDOR_ID_GTCO_IPANEL_2 0x5543 #define USB_DEVICE_ID_GTCO_90 0x0090 #define USB_DEVICE_ID_GTCO_100 0x0100 @@ -629,7 +628,6 @@ void usbhid_init_reports(struct hid_device *hid) #define USB_DEVICE_ID_GTCO_1004 0x1004 #define USB_DEVICE_ID_GTCO_1005 0x1005 #define USB_DEVICE_ID_GTCO_1006 0x1006 -#define USB_DEVICE_ID_GTCO_10 0x0010 #define USB_DEVICE_ID_GTCO_8 0x0008 #define USB_DEVICE_ID_GTCO_d 0x000d @@ -883,7 +881,6 @@ static const struct hid_blacklist { { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_1004, HID_QUIRK_IGNORE }, { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_1005, HID_QUIRK_IGNORE }, { USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_1006, HID_QUIRK_IGNORE }, - { USB_VENDOR_ID_GTCO_IPANEL_1, USB_DEVICE_ID_GTCO_10, HID_QUIRK_IGNORE }, { USB_VENDOR_ID_GTCO_IPANEL_2, USB_DEVICE_ID_GTCO_8, HID_QUIRK_IGNORE }, { USB_VENDOR_ID_GTCO_IPANEL_2, USB_DEVICE_ID_GTCO_d, HID_QUIRK_IGNORE }, { USB_VENDOR_ID_IMATION, USB_DEVICE_ID_DISC_STAKKA, HID_QUIRK_IGNORE }, From 8eda232e226b97c8f2fcbc5f672135247c1ee348 Mon Sep 17 00:00:00 2001 From: Simon Bennett Date: Wed, 24 Jan 2007 10:27:20 +0100 Subject: [PATCH 120/326] HID: fix hid-input mapping for Firefly Mini Remote Control Patch adds entries to the HID consumer page for the Firefly Mini IR remote control Signed-off-by: Simon Bennett Signed-off-by: Jiri Kosina --- drivers/hid/hid-input.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/hid/hid-input.c b/drivers/hid/hid-input.c index 9cf591a1bda3..981fcf01d41b 100644 --- a/drivers/hid/hid-input.c +++ b/drivers/hid/hid-input.c @@ -431,6 +431,7 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel case 0x040: map_key_clear(KEY_MENU); break; case 0x045: map_key_clear(KEY_RADIO); break; + case 0x083: map_key_clear(KEY_LAST); break; case 0x088: map_key_clear(KEY_PC); break; case 0x089: map_key_clear(KEY_TV); break; case 0x08a: map_key_clear(KEY_WWW); break; @@ -448,6 +449,7 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel case 0x096: map_key_clear(KEY_TAPE); break; case 0x097: map_key_clear(KEY_TV2); break; case 0x098: map_key_clear(KEY_SAT); break; + case 0x09a: map_key_clear(KEY_PVR); break; case 0x09c: map_key_clear(KEY_CHANNELUP); break; case 0x09d: map_key_clear(KEY_CHANNELDOWN); break; From db3ccdac261e015023cfd922840170f14c9cdc09 Mon Sep 17 00:00:00 2001 From: Baruch Even Date: Thu, 25 Jan 2007 13:35:06 -0800 Subject: [PATCH 121/326] [TCP]: Fix sorting of SACK blocks. The sorting of SACK blocks actually munges them rather than sort, causing the TCP stack to ignore some SACK information and breaking the assumption of ordered SACK blocks after sorting. The sort takes the data from a second buffer which isn't moved causing subsequent data moves to occur from the wrong location. The fix is to use a temporary buffer as a normal sort does. Signed-off-By: Baruch Even Signed-off-by: David S. Miller --- net/ipv4/tcp_input.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 5c16e24a6061..c26076fb890e 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -1011,10 +1011,11 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_ for (j = 0; j < i; j++){ if (after(ntohl(sp[j].start_seq), ntohl(sp[j+1].start_seq))){ - sp[j].start_seq = htonl(tp->recv_sack_cache[j+1].start_seq); - sp[j].end_seq = htonl(tp->recv_sack_cache[j+1].end_seq); - sp[j+1].start_seq = htonl(tp->recv_sack_cache[j].start_seq); - sp[j+1].end_seq = htonl(tp->recv_sack_cache[j].end_seq); + struct tcp_sack_block_wire tmp; + + tmp = sp[j]; + sp[j] = sp[j+1]; + sp[j+1] = tmp; } } From 17234246eb82898cf98e3c29e81d941c738e0587 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 25 Jan 2007 20:46:59 +0900 Subject: [PATCH 122/326] sata_via: don't diddle with ATA_NIEN in ->freeze vt6420 completely loses its ability to raise IRQ for ATAPI devices if ATA_NIEN is diddled with in ->freeze. Further investigation is necessary to determine whether this problem is shared on other controllers but it doesn't seem to be at this point. Make vt6420's ->freeze only clear IRQ to fix this problem. This makes vt6420 relatively more prone to IRQ storms but the controller is way too braindamaged to worry about that anyway. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/ata/sata_via.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/drivers/ata/sata_via.c b/drivers/ata/sata_via.c index 88f0565c8883..55b01236a6f0 100644 --- a/drivers/ata/sata_via.c +++ b/drivers/ata/sata_via.c @@ -74,6 +74,7 @@ enum { static int svia_init_one (struct pci_dev *pdev, const struct pci_device_id *ent); static u32 svia_scr_read (struct ata_port *ap, unsigned int sc_reg); static void svia_scr_write (struct ata_port *ap, unsigned int sc_reg, u32 val); +static void svia_noop_freeze(struct ata_port *ap); static void vt6420_error_handler(struct ata_port *ap); static const struct pci_device_id svia_pci_tbl[] = { @@ -128,7 +129,7 @@ static const struct ata_port_operations vt6420_sata_ops = { .qc_issue = ata_qc_issue_prot, .data_xfer = ata_pio_data_xfer, - .freeze = ata_bmdma_freeze, + .freeze = svia_noop_freeze, .thaw = ata_bmdma_thaw, .error_handler = vt6420_error_handler, .post_internal_cmd = ata_bmdma_post_internal_cmd, @@ -204,6 +205,15 @@ static void svia_scr_write (struct ata_port *ap, unsigned int sc_reg, u32 val) outl(val, ap->ioaddr.scr_addr + (4 * sc_reg)); } +static void svia_noop_freeze(struct ata_port *ap) +{ + /* Some VIA controllers choke if ATA_NIEN is manipulated in + * certain way. Leave it alone and just clear pending IRQ. + */ + ata_chk_status(ap); + ap->ops->irq_clear(ap); +} + /** * vt6420_prereset - prereset for vt6420 * @ap: target ATA port From 0291f95fdb5fcd91cc077aafabea2c5b109fa8a8 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 25 Jan 2007 19:16:28 +0900 Subject: [PATCH 123/326] ahci: improve and limit spurious interrupt messages, take#3 We're still seeing a lot of issues with NCQ implementation in drive firmwares. Sprious FISes during NCQ command phase occur on many drives and some of them seem potentially dangerous (at least to me). Until we find the solution, spurious messages can give us more info. Improve and limit them such that more info can be reported while not disturbing users too much. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/ata/ahci.c | 57 +++++++++++++++++++++++++++++++++++++++------- 1 file changed, 49 insertions(+), 8 deletions(-) diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c index e3c7b312287a..2fe5a58195de 100644 --- a/drivers/ata/ahci.c +++ b/drivers/ata/ahci.c @@ -75,6 +75,7 @@ enum { AHCI_CMD_CLR_BUSY = (1 << 10), RX_FIS_D2H_REG = 0x40, /* offset of D2H Register FIS data */ + RX_FIS_SDB = 0x58, /* offset of SDB FIS data */ RX_FIS_UNK = 0x60, /* offset of Unknown FIS data */ board_ahci = 0, @@ -202,6 +203,10 @@ struct ahci_port_priv { dma_addr_t cmd_tbl_dma; void *rx_fis; dma_addr_t rx_fis_dma; + /* for NCQ spurious interrupt analysis */ + int ncq_saw_spurious_sdb_cnt; + unsigned int ncq_saw_d2h:1; + unsigned int ncq_saw_dmas:1; }; static u32 ahci_scr_read (struct ata_port *ap, unsigned int sc_reg); @@ -1109,8 +1114,9 @@ static void ahci_host_intr(struct ata_port *ap) void __iomem *mmio = ap->host->mmio_base; void __iomem *port_mmio = ahci_port_base(mmio, ap->port_no); struct ata_eh_info *ehi = &ap->eh_info; + struct ahci_port_priv *pp = ap->private_data; u32 status, qc_active; - int rc; + int rc, known_irq = 0; status = readl(port_mmio + PORT_IRQ_STAT); writel(status, port_mmio + PORT_IRQ_STAT); @@ -1137,17 +1143,52 @@ static void ahci_host_intr(struct ata_port *ap) /* hmmm... a spurious interupt */ - /* some devices send D2H reg with I bit set during NCQ command phase */ - if (ap->sactive && (status & PORT_IRQ_D2H_REG_FIS)) + /* if !NCQ, ignore. No modern ATA device has broken HSM + * implementation for non-NCQ commands. + */ + if (!ap->sactive) return; - /* ignore interim PIO setup fis interrupts */ - if (ata_tag_valid(ap->active_tag) && (status & PORT_IRQ_PIOS_FIS)) - return; + if (status & PORT_IRQ_D2H_REG_FIS) { + if (!pp->ncq_saw_d2h) + ata_port_printk(ap, KERN_INFO, + "D2H reg with I during NCQ, " + "this message won't be printed again\n"); + pp->ncq_saw_d2h = 1; + known_irq = 1; + } - if (ata_ratelimit()) + if (status & PORT_IRQ_DMAS_FIS) { + if (!pp->ncq_saw_dmas) + ata_port_printk(ap, KERN_INFO, + "DMAS FIS during NCQ, " + "this message won't be printed again\n"); + pp->ncq_saw_dmas = 1; + known_irq = 1; + } + + if (status & PORT_IRQ_SDB_FIS && + pp->ncq_saw_spurious_sdb_cnt < 10) { + /* SDB FIS containing spurious completions might be + * dangerous, we need to know more about them. Print + * more of it. + */ + const u32 *f = pp->rx_fis + RX_FIS_SDB; + + ata_port_printk(ap, KERN_INFO, "Spurious SDB FIS during NCQ " + "issue=0x%x SAct=0x%x FIS=%08x:%08x%s\n", + readl(port_mmio + PORT_CMD_ISSUE), + readl(port_mmio + PORT_SCR_ACT), f[0], f[1], + pp->ncq_saw_spurious_sdb_cnt < 10 ? + "" : ", shutting up"); + + pp->ncq_saw_spurious_sdb_cnt++; + known_irq = 1; + } + + if (!known_irq) ata_port_printk(ap, KERN_INFO, "spurious interrupt " - "(irq_stat 0x%x active_tag %d sactive 0x%x)\n", + "(irq_stat 0x%x active_tag 0x%x sactive 0x%x)\n", status, ap->active_tag, ap->sactive); } From b2a8bbe67d73631c71492fd60b757fc50a87f182 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 25 Jan 2007 19:40:05 +0900 Subject: [PATCH 124/326] libata: implement ATA_FLAG_IGN_SIMPLEX and use it in sata_uli Some uli controllers have stuck SIMPLEX bit which can't be cleared with ata_pci_clear_simplex(), but the controller is capable of doing DMAs on both channels simultaneously. Implement ATA_FLAG_IGN_SIMPLEX which makes libata ignore the simplex bit and use it in sata_uli. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/ata/libata-sff.c | 12 ++++++++---- drivers/ata/sata_uli.c | 3 ++- include/linux/libata.h | 1 + 3 files changed, 11 insertions(+), 5 deletions(-) diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c index 623cec914c9b..114fa81deb83 100644 --- a/drivers/ata/libata-sff.c +++ b/drivers/ata/libata-sff.c @@ -870,7 +870,8 @@ ata_pci_init_native_mode(struct pci_dev *pdev, struct ata_port_info **port, int pci_resource_start(pdev, 1) | ATA_PCI_CTL_OFS; bmdma = pci_resource_start(pdev, 4); if (bmdma) { - if (inb(bmdma + 2) & 0x80) + if ((!(port[p]->flags & ATA_FLAG_IGN_SIMPLEX)) && + (inb(bmdma + 2) & 0x80)) probe_ent->_host_flags |= ATA_HOST_SIMPLEX; probe_ent->port[p].bmdma_addr = bmdma; } @@ -886,7 +887,8 @@ ata_pci_init_native_mode(struct pci_dev *pdev, struct ata_port_info **port, int bmdma = pci_resource_start(pdev, 4); if (bmdma) { bmdma += 8; - if(inb(bmdma + 2) & 0x80) + if ((!(port[p]->flags & ATA_FLAG_IGN_SIMPLEX)) && + (inb(bmdma + 2) & 0x80)) probe_ent->_host_flags |= ATA_HOST_SIMPLEX; probe_ent->port[p].bmdma_addr = bmdma; } @@ -920,7 +922,8 @@ static struct ata_probe_ent *ata_pci_init_legacy_port(struct pci_dev *pdev, probe_ent->port[0].ctl_addr = ATA_PRIMARY_CTL; if (bmdma) { probe_ent->port[0].bmdma_addr = bmdma; - if (inb(bmdma + 2) & 0x80) + if ((!(port[0]->flags & ATA_FLAG_IGN_SIMPLEX)) && + (inb(bmdma + 2) & 0x80)) probe_ent->_host_flags |= ATA_HOST_SIMPLEX; } ata_std_ports(&probe_ent->port[0]); @@ -937,7 +940,8 @@ static struct ata_probe_ent *ata_pci_init_legacy_port(struct pci_dev *pdev, probe_ent->port[1].ctl_addr = ATA_SECONDARY_CTL; if (bmdma) { probe_ent->port[1].bmdma_addr = bmdma + 8; - if (inb(bmdma + 10) & 0x80) + if ((!(port[1]->flags & ATA_FLAG_IGN_SIMPLEX)) && + (inb(bmdma + 10) & 0x80)) probe_ent->_host_flags |= ATA_HOST_SIMPLEX; } ata_std_ports(&probe_ent->port[1]); diff --git a/drivers/ata/sata_uli.c b/drivers/ata/sata_uli.c index 5c603ca3a50a..a43aec62d505 100644 --- a/drivers/ata/sata_uli.c +++ b/drivers/ata/sata_uli.c @@ -128,7 +128,8 @@ static const struct ata_port_operations uli_ops = { static struct ata_port_info uli_port_info = { .sht = &uli_sht, - .flags = ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY, + .flags = ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY | + ATA_FLAG_IGN_SIMPLEX, .pio_mask = 0x1f, /* pio0-4 */ .udma_mask = 0x7f, /* udma0-6 */ .port_ops = &uli_ops, diff --git a/include/linux/libata.h b/include/linux/libata.h index df76fc467816..22aa69e20905 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -177,6 +177,7 @@ enum { * Register FIS clearing BSY */ ATA_FLAG_DEBUGMSG = (1 << 13), ATA_FLAG_SETXFER_POLLING= (1 << 14), /* use polling for SETXFER */ + ATA_FLAG_IGN_SIMPLEX = (1 << 15), /* ignore SIMPLEX */ /* The following flag belongs to ap->pflags but is kept in * ap->flags because it's referenced in many LLDs and will be From 61dd08c6c8d2b4ede530e43c01fa72f789ef65b1 Mon Sep 17 00:00:00 2001 From: Alan Date: Thu, 25 Jan 2007 15:09:05 +0000 Subject: [PATCH 125/326] libata-sff: Don't call bmdma_stop on non DMA capable controllers Fixes bogus accesses to ports 0-15 with a non DMA capable controller. This I think should go in for 2.6.20 Signed-off-by: Alan Cox Signed-off-by: Jeff Garzik --- drivers/ata/libata-sff.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c index 114fa81deb83..942aeba2940a 100644 --- a/drivers/ata/libata-sff.c +++ b/drivers/ata/libata-sff.c @@ -827,7 +827,8 @@ void ata_bmdma_error_handler(struct ata_port *ap) */ void ata_bmdma_post_internal_cmd(struct ata_queued_cmd *qc) { - ata_bmdma_stop(qc); + if (qc->ap->ioaddr.bmdma_addr) + ata_bmdma_stop(qc); } #ifdef CONFIG_PCI From 24cb230b587cf3aad8794b150682d8d8303a2120 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Thu, 25 Jan 2007 15:49:56 -0800 Subject: [PATCH 126/326] [BNX2]: Fix 2nd port's MAC address. On the 5709, we need to add the proper offset to calculate the shared memory base address of the 2nd port correctly. Otherwise, the 2nd port's MAC address and other information will be the same as the 1st port. Update version to 1.5.4. Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/bnx2.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c index ca5acc4736df..953808efe551 100644 --- a/drivers/net/bnx2.c +++ b/drivers/net/bnx2.c @@ -57,8 +57,8 @@ #define DRV_MODULE_NAME "bnx2" #define PFX DRV_MODULE_NAME ": " -#define DRV_MODULE_VERSION "1.5.3" -#define DRV_MODULE_RELDATE "January 8, 2007" +#define DRV_MODULE_VERSION "1.5.4" +#define DRV_MODULE_RELDATE "January 24, 2007" #define RUN_AT(x) (jiffies + (x)) @@ -5845,9 +5845,11 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev) reg = REG_RD_IND(bp, BNX2_SHM_HDR_SIGNATURE); if ((reg & BNX2_SHM_HDR_SIGNATURE_SIG_MASK) == - BNX2_SHM_HDR_SIGNATURE_SIG) - bp->shmem_base = REG_RD_IND(bp, BNX2_SHM_HDR_ADDR_0); - else + BNX2_SHM_HDR_SIGNATURE_SIG) { + u32 off = PCI_FUNC(pdev->devfn) << 2; + + bp->shmem_base = REG_RD_IND(bp, BNX2_SHM_HDR_ADDR_0 + off); + } else bp->shmem_base = HOST_VIEW_SHMEM_BASE; /* Get the permanent MAC address. First we need to make sure the From 95743deb34de29defbb98ad477700aaf344f93f3 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Thu, 25 Jan 2007 15:51:51 -0800 Subject: [PATCH 127/326] [DECNET]: Handle a failure in neigh_parms_alloc (take 2) While enhancing the neighbour code to handle multiple network namespaces I noticed that decnet is assuming neigh_parms_alloc will allways succeed, which is clearly wrong. So handle the failure. Signed-off-by: Eric W. Biederman Acked-by: Steven Whitehouse Signed-off-by: David S. Miller --- net/decnet/dn_dev.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/net/decnet/dn_dev.c b/net/decnet/dn_dev.c index fc6f3c023a54..ed083ab455b7 100644 --- a/net/decnet/dn_dev.c +++ b/net/decnet/dn_dev.c @@ -1145,16 +1145,23 @@ struct dn_dev *dn_dev_create(struct net_device *dev, int *err) init_timer(&dn_db->timer); dn_db->uptime = jiffies; + + dn_db->neigh_parms = neigh_parms_alloc(dev, &dn_neigh_table); + if (!dn_db->neigh_parms) { + dev->dn_ptr = NULL; + kfree(dn_db); + return NULL; + } + if (dn_db->parms.up) { if (dn_db->parms.up(dev) < 0) { + neigh_parms_release(&dn_neigh_table, dn_db->neigh_parms); dev->dn_ptr = NULL; kfree(dn_db); return NULL; } } - dn_db->neigh_parms = neigh_parms_alloc(dev, &dn_neigh_table); - dn_dev_sysctl_register(dev, &dn_db->parms); dn_dev_set_timer(dev); From 5ad0d383ddbf0d2fce43b8aac267a6c299fd2dff Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Thu, 25 Jan 2007 17:19:51 -0800 Subject: [PATCH 128/326] [PATCH] x86_64: fix put_user for 64-bit constant On x86-64, a put_user call using a 64-bit pointer and a constant value that is > 0xffffffff will produce code that doesn't assemble. This patch fixes the asm construct to use the Z constraint for 32-bit constants. Signed-off-by: Roland McGrath Signed-off-by: Linus Torvalds --- include/asm-x86_64/uaccess.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/asm-x86_64/uaccess.h b/include/asm-x86_64/uaccess.h index d5dbc87274f8..c0eac519840b 100644 --- a/include/asm-x86_64/uaccess.h +++ b/include/asm-x86_64/uaccess.h @@ -157,7 +157,7 @@ do { \ case 1: __put_user_asm(x,ptr,retval,"b","b","iq",-EFAULT); break;\ case 2: __put_user_asm(x,ptr,retval,"w","w","ir",-EFAULT); break;\ case 4: __put_user_asm(x,ptr,retval,"l","k","ir",-EFAULT); break;\ - case 8: __put_user_asm(x,ptr,retval,"q","","ir",-EFAULT); break;\ + case 8: __put_user_asm(x,ptr,retval,"q","","Zr",-EFAULT); break;\ default: __put_user_bad(); \ } \ } while (0) From d5e76b0a280f71b20bdd20d1c1b4d6812ceb8c3a Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Thu, 25 Jan 2007 19:30:36 -0800 Subject: [PATCH 129/326] [AF_PACKET]: Check device down state before hard header callbacks. If the device is down, invoking the device hard header callbacks is not legal, so check it early. Based upon a shaper OOPS report from Frederik Deweerdt. Signed-off-by: David S. Miller --- net/packet/af_packet.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 594c078c5ebc..6dc01bdeb76b 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -359,6 +359,10 @@ static int packet_sendmsg_spkt(struct kiocb *iocb, struct socket *sock, if (dev == NULL) goto out_unlock; + err = -ENETDOWN; + if (!(dev->flags & IFF_UP)) + goto out_unlock; + /* * You may not queue a frame bigger than the mtu. This is the lowest level * raw protocol and you must do your own fragmentation at this level. @@ -407,10 +411,6 @@ static int packet_sendmsg_spkt(struct kiocb *iocb, struct socket *sock, if (err) goto out_free; - err = -ENETDOWN; - if (!(dev->flags & IFF_UP)) - goto out_free; - /* * Now send it */ @@ -738,6 +738,10 @@ static int packet_sendmsg(struct kiocb *iocb, struct socket *sock, if (sock->type == SOCK_RAW) reserve = dev->hard_header_len; + err = -ENETDOWN; + if (!(dev->flags & IFF_UP)) + goto out_unlock; + err = -EMSGSIZE; if (len > dev->mtu+reserve) goto out_unlock; @@ -770,10 +774,6 @@ static int packet_sendmsg(struct kiocb *iocb, struct socket *sock, skb->dev = dev; skb->priority = sk->sk_priority; - err = -ENETDOWN; - if (!(dev->flags & IFF_UP)) - goto out_free; - /* * Now send it */ From e89862f4c5b3c4ac9afcd8cb1365d2f1e16ddc3b Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 26 Jan 2007 01:04:55 -0800 Subject: [PATCH 130/326] [TCP]: Restore SKB socket owner setting in tcp_transmit_skb(). Revert 931731123a103cfb3f70ac4b7abfc71d94ba1f03 We can't elide the skb_set_owner_w() here because things like certain netfilter targets (such as owner MATCH) need a socket to be set on the SKB for correct operation. Thanks to Jan Engelhardt and other netfilter list members for pointing this out. Signed-off-by: David S. Miller --- include/net/inet6_connection_sock.h | 2 +- include/net/inet_connection_sock.h | 3 +-- include/net/ip.h | 2 +- net/dccp/output.c | 4 ++-- net/ipv4/ip_output.c | 3 ++- net/ipv4/tcp_output.c | 3 ++- net/ipv6/inet6_connection_sock.c | 3 ++- net/sctp/protocol.c | 2 +- 8 files changed, 12 insertions(+), 10 deletions(-) diff --git a/include/net/inet6_connection_sock.h b/include/net/inet6_connection_sock.h index 16aa96a6a53b..f13ddc2543b1 100644 --- a/include/net/inet6_connection_sock.h +++ b/include/net/inet6_connection_sock.h @@ -38,5 +38,5 @@ extern void inet6_csk_reqsk_queue_hash_add(struct sock *sk, extern void inet6_csk_addr2sockaddr(struct sock *sk, struct sockaddr *uaddr); -extern int inet6_csk_xmit(struct sk_buff *skb, struct sock *sk, int ipfragok); +extern int inet6_csk_xmit(struct sk_buff *skb, int ipfragok); #endif /* _INET6_CONNECTION_SOCK_H */ diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index bf16d98d372c..133cf30d2d79 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -37,8 +37,7 @@ struct tcp_congestion_ops; * (i.e. things that depend on the address family) */ struct inet_connection_sock_af_ops { - int (*queue_xmit)(struct sk_buff *skb, struct sock *sk, - int ipfragok); + int (*queue_xmit)(struct sk_buff *skb, int ipfragok); void (*send_check)(struct sock *sk, int len, struct sk_buff *skb); int (*rebuild_header)(struct sock *sk); diff --git a/include/net/ip.h b/include/net/ip.h index 053f02b5cb89..e79c3e3aa4f6 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -97,7 +97,7 @@ extern int ip_mc_output(struct sk_buff *skb); extern int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)); extern int ip_do_nat(struct sk_buff *skb); extern void ip_send_check(struct iphdr *ip); -extern int ip_queue_xmit(struct sk_buff *skb, struct sock *sk, int ipfragok); +extern int ip_queue_xmit(struct sk_buff *skb, int ipfragok); extern void ip_init(void); extern int ip_append_data(struct sock *sk, int getfrag(void *from, char *to, int offset, int len, diff --git a/net/dccp/output.c b/net/dccp/output.c index 824569659083..3435542e9652 100644 --- a/net/dccp/output.c +++ b/net/dccp/output.c @@ -124,7 +124,7 @@ static int dccp_transmit_skb(struct sock *sk, struct sk_buff *skb) DCCP_INC_STATS(DCCP_MIB_OUTSEGS); memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); - err = icsk->icsk_af_ops->queue_xmit(skb, sk, 0); + err = icsk->icsk_af_ops->queue_xmit(skb, 0); return net_xmit_eval(err); } return -ENOBUFS; @@ -396,7 +396,7 @@ int dccp_send_reset(struct sock *sk, enum dccp_reset_codes code) code); if (skb != NULL) { memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); - err = inet_csk(sk)->icsk_af_ops->queue_xmit(skb, sk, 0); + err = inet_csk(sk)->icsk_af_ops->queue_xmit(skb, 0); return net_xmit_eval(err); } } diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index f071f84808fa..a0f2008584bc 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -281,8 +281,9 @@ int ip_output(struct sk_buff *skb) !(IPCB(skb)->flags & IPSKB_REROUTED)); } -int ip_queue_xmit(struct sk_buff *skb, struct sock *sk, int ipfragok) +int ip_queue_xmit(struct sk_buff *skb, int ipfragok) { + struct sock *sk = skb->sk; struct inet_sock *inet = inet_sk(sk); struct ip_options *opt = inet->opt; struct rtable *rt; diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 23e32c806916..975f4472af29 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -467,6 +467,7 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it, th = (struct tcphdr *) skb_push(skb, tcp_header_size); skb->h.th = th; + skb_set_owner_w(skb, sk); /* Build TCP header and checksum it. */ th->source = inet->sport; @@ -540,7 +541,7 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it, if (after(tcb->end_seq, tp->snd_nxt) || tcb->seq == tcb->end_seq) TCP_INC_STATS(TCP_MIB_OUTSEGS); - err = icsk->icsk_af_ops->queue_xmit(skb, sk, 0); + err = icsk->icsk_af_ops->queue_xmit(skb, 0); if (likely(err <= 0)) return err; diff --git a/net/ipv6/inet6_connection_sock.c b/net/ipv6/inet6_connection_sock.c index c700302ad51a..116f94a49071 100644 --- a/net/ipv6/inet6_connection_sock.c +++ b/net/ipv6/inet6_connection_sock.c @@ -139,8 +139,9 @@ void inet6_csk_addr2sockaddr(struct sock *sk, struct sockaddr * uaddr) EXPORT_SYMBOL_GPL(inet6_csk_addr2sockaddr); -int inet6_csk_xmit(struct sk_buff *skb, struct sock *sk, int ipfragok) +int inet6_csk_xmit(struct sk_buff *skb, int ipfragok) { + struct sock *sk = skb->sk; struct inet_sock *inet = inet_sk(sk); struct ipv6_pinfo *np = inet6_sk(sk); struct flowi fl; diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index 225f39b5d595..0ef48126b117 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -804,7 +804,7 @@ static inline int sctp_v4_xmit(struct sk_buff *skb, NIPQUAD(((struct rtable *)skb->dst)->rt_dst)); SCTP_INC_STATS(SCTP_MIB_OUTSCTPPACKS); - return ip_queue_xmit(skb, skb->sk, ipfragok); + return ip_queue_xmit(skb, ipfragok); } static struct sctp_af sctp_ipv4_specific; From c72c6b2a291bb6c61b1546d116784a79e15a6c29 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Fri, 26 Jan 2007 01:06:47 -0800 Subject: [PATCH 131/326] [NETFILTER]: nf_nat: fix ICMP translation with statically linked conntrack When nf_nat/nf_conntrack_ipv4 are linked statically, nf_nat is initialized before nf_conntrack_ipv4, which makes the nf_ct_l3proto_find_get(AF_INET) call during nf_nat initialization return the generic l3proto instead of the AF_INET specific one. This breaks ICMP error translation since the generic protocol always initializes the IPs in the tuple to 0. Change the linking order and put nf_conntrack_ipv4 first. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/ipv4/netfilter/Makefile | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index 15e741aeb291..16d177b71bf8 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -4,6 +4,14 @@ # objects for the standalone - connection tracking / NAT ip_conntrack-objs := ip_conntrack_standalone.o ip_conntrack_core.o ip_conntrack_proto_generic.o ip_conntrack_proto_tcp.o ip_conntrack_proto_udp.o ip_conntrack_proto_icmp.o +# objects for l3 independent conntrack +nf_conntrack_ipv4-objs := nf_conntrack_l3proto_ipv4.o nf_conntrack_proto_icmp.o +ifeq ($(CONFIG_NF_CONNTRACK_PROC_COMPAT),y) +ifeq ($(CONFIG_PROC_FS),y) +nf_conntrack_ipv4-objs += nf_conntrack_l3proto_ipv4_compat.o +endif +endif + ip_nat-objs := ip_nat_core.o ip_nat_helper.o ip_nat_proto_unknown.o ip_nat_proto_tcp.o ip_nat_proto_udp.o ip_nat_proto_icmp.o nf_nat-objs := nf_nat_core.o nf_nat_helper.o nf_nat_proto_unknown.o nf_nat_proto_tcp.o nf_nat_proto_udp.o nf_nat_proto_icmp.o ifneq ($(CONFIG_NF_NAT),) @@ -20,6 +28,8 @@ ip_nat_h323-objs := ip_nat_helper_h323.o # connection tracking obj-$(CONFIG_IP_NF_CONNTRACK) += ip_conntrack.o +obj-$(CONFIG_NF_CONNTRACK_IPV4) += nf_conntrack_ipv4.o + obj-$(CONFIG_IP_NF_NAT) += ip_nat.o obj-$(CONFIG_NF_NAT) += nf_nat.o @@ -106,13 +116,3 @@ obj-$(CONFIG_IP_NF_ARPFILTER) += arptable_filter.o obj-$(CONFIG_IP_NF_QUEUE) += ip_queue.o -# objects for l3 independent conntrack -nf_conntrack_ipv4-objs := nf_conntrack_l3proto_ipv4.o nf_conntrack_proto_icmp.o -ifeq ($(CONFIG_NF_CONNTRACK_PROC_COMPAT),y) -ifeq ($(CONFIG_PROC_FS),y) -nf_conntrack_ipv4-objs += nf_conntrack_l3proto_ipv4_compat.o -endif -endif - -# l3 independent conntrack -obj-$(CONFIG_NF_CONNTRACK_IPV4) += nf_conntrack_ipv4.o From a46bf7d5a81b350cd204b82bd25ee6ffbc2967d4 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Fri, 26 Jan 2007 01:07:30 -0800 Subject: [PATCH 132/326] [NETFILTER]: nf_nat_pptp: fix expectation removal When removing the expectation for the opposite direction, the PPTP NAT helper initializes the tuple for lookup with the addresses of the opposite direction, which makes the lookup fail. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/ipv4/netfilter/nf_nat_pptp.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/ipv4/netfilter/nf_nat_pptp.c b/net/ipv4/netfilter/nf_nat_pptp.c index 0ae45b79a4eb..5df4fcae3ab6 100644 --- a/net/ipv4/netfilter/nf_nat_pptp.c +++ b/net/ipv4/netfilter/nf_nat_pptp.c @@ -72,9 +72,9 @@ static void pptp_nat_expected(struct nf_conn *ct, DEBUGP("we are PAC->PNS\n"); /* build tuple for PNS->PAC */ t.src.l3num = AF_INET; - t.src.u3.ip = master->tuplehash[exp->dir].tuple.src.u3.ip; + t.src.u3.ip = master->tuplehash[!exp->dir].tuple.src.u3.ip; t.src.u.gre.key = nat_pptp_info->pns_call_id; - t.dst.u3.ip = master->tuplehash[exp->dir].tuple.dst.u3.ip; + t.dst.u3.ip = master->tuplehash[!exp->dir].tuple.dst.u3.ip; t.dst.u.gre.key = nat_pptp_info->pac_call_id; t.dst.protonum = IPPROTO_GRE; } From 7399072a7348d025e7bcb5eb5d5e9be941d490b7 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Fri, 26 Jan 2007 01:07:59 -0800 Subject: [PATCH 133/326] [NETFILTER]: nf_conntrack_pptp: fix NAT setup of expected GRE connections When an expected connection arrives, the NAT helper should be called to set up NAT similar to the master connection. The PPTP conntrack helper incorrectly checks whether the _expected_ connection has NAT setup before calling the NAT helper (which is never the case), instead of checkeing whether the _master_ connection is NATed. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/netfilter/nf_conntrack_pptp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/netfilter/nf_conntrack_pptp.c b/net/netfilter/nf_conntrack_pptp.c index f0ff00e0d052..c59df3bc2bbd 100644 --- a/net/netfilter/nf_conntrack_pptp.c +++ b/net/netfilter/nf_conntrack_pptp.c @@ -113,7 +113,7 @@ static void pptp_expectfn(struct nf_conn *ct, rcu_read_lock(); nf_nat_pptp_expectfn = rcu_dereference(nf_nat_pptp_hook_expectfn); - if (nf_nat_pptp_expectfn && ct->status & IPS_NAT_MASK) + if (nf_nat_pptp_expectfn && ct->master->status & IPS_NAT_MASK) nf_nat_pptp_expectfn(ct, exp); else { struct nf_conntrack_tuple inv_t; From 0cd78989a2d98a62017a5842f81b61e1b425cd0d Mon Sep 17 00:00:00 2001 From: Haavard Skinnemoen Date: Fri, 26 Jan 2007 13:12:25 +0100 Subject: [PATCH 134/326] [AVR32] Export clear_page symbol Add missing EXPORT_SYMBOL(clear_page), allowing ext3 to be compiled as a module. Signed-off-by: Haavard Skinnemoen --- arch/avr32/kernel/avr32_ksyms.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/avr32/kernel/avr32_ksyms.c b/arch/avr32/kernel/avr32_ksyms.c index 7c4c76114bba..80f55f8dbf1c 100644 --- a/arch/avr32/kernel/avr32_ksyms.c +++ b/arch/avr32/kernel/avr32_ksyms.c @@ -29,6 +29,7 @@ EXPORT_SYMBOL(__avr32_asr64); */ EXPORT_SYMBOL(memset); EXPORT_SYMBOL(memcpy); +EXPORT_SYMBOL(clear_page); /* * Userspace access stuff. From 4fe4f4ace5a667122f8dd04302e19e46d99e3594 Mon Sep 17 00:00:00 2001 From: Haavard Skinnemoen Date: Fri, 26 Jan 2007 13:19:48 +0100 Subject: [PATCH 135/326] [AVR32] Update ATSTK1000 defconfig: Enable macb by default Enable the Atmel MACB ethernet driver by default on ATSTK1000. Signed-off-by: Haavard Skinnemoen --- arch/avr32/configs/atstk1002_defconfig | 39 ++++++++++++++++++-------- 1 file changed, 27 insertions(+), 12 deletions(-) diff --git a/arch/avr32/configs/atstk1002_defconfig b/arch/avr32/configs/atstk1002_defconfig index ae92a14ef9a0..77dace9d54bc 100644 --- a/arch/avr32/configs/atstk1002_defconfig +++ b/arch/avr32/configs/atstk1002_defconfig @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.19-rc2 -# Fri Oct 20 11:52:37 2006 +# Linux kernel version: 2.6.20-rc6 +# Fri Jan 26 13:12:59 2007 # CONFIG_AVR32=y CONFIG_GENERIC_HARDIRQS=y @@ -9,6 +9,8 @@ CONFIG_HARDIRQS_SW_RESEND=y CONFIG_GENERIC_IRQ_PROBE=y CONFIG_RWSEM_GENERIC_SPINLOCK=y CONFIG_GENERIC_TIME=y +# CONFIG_ARCH_HAS_ILOG2_U32 is not set +# CONFIG_ARCH_HAS_ILOG2_U64 is not set CONFIG_GENERIC_HWEIGHT=y CONFIG_GENERIC_CALIBRATE_DELAY=y CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config" @@ -36,6 +38,7 @@ CONFIG_TASK_DELAY_ACCT=y # CONFIG_UTS_NS is not set CONFIG_AUDIT=y # CONFIG_IKCONFIG is not set +CONFIG_SYSFS_DEPRECATED=y CONFIG_RELAY=y CONFIG_INITRAMFS_SOURCE="" CONFIG_CC_OPTIMIZE_FOR_SIZE=y @@ -75,7 +78,9 @@ CONFIG_MODULE_UNLOAD=y # Block layer # CONFIG_BLOCK=y +# CONFIG_LBD is not set # CONFIG_BLK_DEV_IO_TRACE is not set +# CONFIG_LSF is not set # # IO Schedulers @@ -125,6 +130,7 @@ CONFIG_SPLIT_PTLOCK_CPUS=4 # CONFIG_OWNERSHIP_TRACE is not set # CONFIG_HZ_100 is not set CONFIG_HZ_250=y +# CONFIG_HZ_300 is not set # CONFIG_HZ_1000 is not set CONFIG_HZ=250 CONFIG_CMDLINE="" @@ -182,6 +188,7 @@ CONFIG_INET_TCP_DIAG=y # CONFIG_TCP_CONG_ADVANCED is not set CONFIG_TCP_CONG_CUBIC=y CONFIG_DEFAULT_TCP_CONG="cubic" +# CONFIG_TCP_MD5SIG is not set # CONFIG_IPV6 is not set # CONFIG_INET6_XFRM_TUNNEL is not set # CONFIG_INET6_TUNNEL is not set @@ -260,6 +267,7 @@ CONFIG_MTD_CMDLINE_PARTS=y # User Modules And Translation Layers # CONFIG_MTD_CHAR=y +CONFIG_MTD_BLKDEVS=y CONFIG_MTD_BLOCK=y # CONFIG_FTL is not set # CONFIG_NFTL is not set @@ -355,7 +363,6 @@ CONFIG_BLK_DEV_INITRD=y # # Misc devices # -# CONFIG_SGI_IOC4 is not set # CONFIG_TIFM_CORE is not set # @@ -405,11 +412,14 @@ CONFIG_TUN=m # # PHY device support # +# CONFIG_PHYLIB is not set # # Ethernet (10 or 100Mbit) # -# CONFIG_NET_ETHERNET is not set +CONFIG_NET_ETHERNET=y +CONFIG_MII=y +CONFIG_MACB=y # # Ethernet (1000 Mbit) @@ -505,10 +515,6 @@ CONFIG_UNIX98_PTYS=y # CONFIG_GEN_RTC is not set # CONFIG_DTLK is not set # CONFIG_R3964 is not set - -# -# Ftape, the floppy tape device driver -# # CONFIG_RAW_DRIVER is not set # @@ -620,6 +626,10 @@ CONFIG_UNIX98_PTYS=y # DMA Devices # +# +# Virtualization +# + # # File systems # @@ -683,7 +693,6 @@ CONFIG_CONFIGFS_FS=m # CONFIG_BEFS_FS is not set # CONFIG_BFS_FS is not set # CONFIG_EFS_FS is not set -# CONFIG_JFFS_FS is not set CONFIG_JFFS2_FS=y CONFIG_JFFS2_FS_DEBUG=0 CONFIG_JFFS2_FS_WRITEBUFFER=y @@ -762,6 +771,11 @@ CONFIG_NLS_ISO8859_1=m # CONFIG_NLS_KOI8_U is not set CONFIG_NLS_UTF8=m +# +# Distributed Lock Manager +# +# CONFIG_DLM is not set + # # Kernel hacking # @@ -770,6 +784,8 @@ CONFIG_TRACE_IRQFLAGS_SUPPORT=y CONFIG_ENABLE_MUST_CHECK=y CONFIG_MAGIC_SYSRQ=y # CONFIG_UNUSED_SYMBOLS is not set +CONFIG_DEBUG_FS=y +# CONFIG_HEADERS_CHECK is not set CONFIG_DEBUG_KERNEL=y CONFIG_LOG_BUF_SHIFT=14 CONFIG_DETECT_SOFTLOCKUP=y @@ -785,13 +801,10 @@ CONFIG_DETECT_SOFTLOCKUP=y # CONFIG_DEBUG_KOBJECT is not set CONFIG_DEBUG_BUGVERBOSE=y # CONFIG_DEBUG_INFO is not set -CONFIG_DEBUG_FS=y # CONFIG_DEBUG_VM is not set # CONFIG_DEBUG_LIST is not set CONFIG_FRAME_POINTER=y -# CONFIG_UNWIND_INFO is not set CONFIG_FORCED_INLINING=y -# CONFIG_HEADERS_CHECK is not set # CONFIG_RCU_TORTURE_TEST is not set # CONFIG_KPROBES is not set @@ -809,6 +822,7 @@ CONFIG_FORCED_INLINING=y # # Library routines # +CONFIG_BITREVERSE=y CONFIG_CRC_CCITT=m # CONFIG_CRC16 is not set CONFIG_CRC32=y @@ -817,3 +831,4 @@ CONFIG_AUDIT_GENERIC=y CONFIG_ZLIB_INFLATE=y CONFIG_ZLIB_DEFLATE=y CONFIG_PLIST=y +CONFIG_IOMAP_COPY=y From ecdfc9787fe527491baefc22dce8b2dbd5b2908d Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Fri, 26 Jan 2007 12:47:06 -0800 Subject: [PATCH 136/326] Resurrect 'try_to_free_buffers()' VM hackery It's not pretty, but it appears that ext3 with data=journal will clean pages without ever actually telling the VM that they are clean. This, in turn, will result in the VM (and balance_dirty_pages() in particular) to never realize that the pages got cleaned, and wait forever for an event that already happened. Technically, this seems to be a problem with ext3 itself, but it used to be hidden by 'try_to_free_buffers()' noticing this situation on its own, and just working around the filesystem problem. This commit re-instates that hack, in order to avoid a regression for the 2.6.20 release. This fixes bugzilla 7844: http://bugzilla.kernel.org/show_bug.cgi?id=7844 Peter Zijlstra points out that we should probably retain the debugging code that this removes from cancel_dirty_page(), and I agree, but for the imminent release we might as well just silence the warning too (since it's not a new bug: anything that triggers that warning has been around forever). Acked-by: Randy Dunlap Acked-by: Jens Axboe Acked-by: Peter Zijlstra Cc: Andrew Morton Signed-off-by: Linus Torvalds --- fs/buffer.c | 15 ++++++++++++++- mm/truncate.c | 21 ++++++++++++++------- 2 files changed, 28 insertions(+), 8 deletions(-) diff --git a/fs/buffer.c b/fs/buffer.c index 3b116078b4c3..460f1c43238e 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -2834,7 +2834,7 @@ int try_to_free_buffers(struct page *page) int ret = 0; BUG_ON(!PageLocked(page)); - if (PageDirty(page) || PageWriteback(page)) + if (PageWriteback(page)) return 0; if (mapping == NULL) { /* can this still happen? */ @@ -2845,6 +2845,19 @@ int try_to_free_buffers(struct page *page) spin_lock(&mapping->private_lock); ret = drop_buffers(page, &buffers_to_free); spin_unlock(&mapping->private_lock); + + /* + * If the filesystem writes its buffers by hand (eg ext3) + * then we can have clean buffers against a dirty page. We + * clean the page here; otherwise the VM will never notice + * that the filesystem did any IO at all. + * + * Also, during truncate, discard_buffer will have marked all + * the page's buffers clean. We discover that here and clean + * the page also. + */ + if (ret) + cancel_dirty_page(page, PAGE_CACHE_SIZE); out: if (buffers_to_free) { struct buffer_head *bh = buffers_to_free; diff --git a/mm/truncate.c b/mm/truncate.c index 6c79ca4a1ca7..3262740aa059 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -51,15 +51,22 @@ static inline void truncate_partial_page(struct page *page, unsigned partial) do_invalidatepage(page, partial); } +/* + * This cancels just the dirty bit on the kernel page itself, it + * does NOT actually remove dirty bits on any mmap's that may be + * around. It also leaves the page tagged dirty, so any sync + * activity will still find it on the dirty lists, and in particular, + * clear_page_dirty_for_io() will still look at the dirty bits in + * the VM. + * + * Doing this should *normally* only ever be done when a page + * is truncated, and is not actually mapped anywhere at all. However, + * fs/buffer.c does this when it notices that somebody has cleaned + * out all the buffers on a page without actually doing it through + * the VM. Can you say "ext3 is horribly ugly"? Tought you could. + */ void cancel_dirty_page(struct page *page, unsigned int account_size) { - /* If we're cancelling the page, it had better not be mapped any more */ - if (page_mapped(page)) { - static unsigned int warncount; - - WARN_ON(++warncount < 5); - } - if (TestClearPageDirty(page)) { struct address_space *mapping = page->mapping; if (mapping && mapping_cap_account_dirty(mapping)) { From 4b89eed93e0fa40a63e3d7b1796ec1337ea7a3aa Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Fri, 26 Jan 2007 12:53:20 -0800 Subject: [PATCH 137/326] Write back inode data pages even when the inode itself is locked In __writeback_single_inode(), when we find a locked inode and we're not doing a data-integrity sync, we used to just skip writing entirely, since we didn't want to wait for the inode to unlock. However, there's really no reason to skip writing the data pages, which are likely to be the the bulk of the dirty state anyway (and the main reason why writeback was started for the non-data-integrity case, of course!) Acked-by: Nick Piggin Cc: Andrew Morton , Cc: Peter Zijlstra Cc: Hugh Dickins Cc: David Howells Signed-off-by: Linus Torvalds --- fs/fs-writeback.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index c403b66ec83c..a4b142a6a2c7 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -251,8 +251,19 @@ __writeback_single_inode(struct inode *inode, struct writeback_control *wbc) WARN_ON(inode->i_state & I_WILL_FREE); if ((wbc->sync_mode != WB_SYNC_ALL) && (inode->i_state & I_LOCK)) { + struct address_space *mapping = inode->i_mapping; + int ret; + list_move(&inode->i_list, &inode->i_sb->s_dirty); - return 0; + + /* + * Even if we don't actually write the inode itself here, + * we can at least start some of the data writeout.. + */ + spin_unlock(&inode_lock); + ret = do_writepages(mapping, wbc); + spin_lock(&inode_lock); + return ret; } /* From bce66ca4a2f695509e1b021311eb4de1e4fdf3e4 Mon Sep 17 00:00:00 2001 From: Leonard Norrgard Date: Fri, 26 Jan 2007 00:56:38 -0800 Subject: [PATCH 138/326] [PATCH] KVM: SVM: Fix SVM idt confusion There's an obvious typo in svm_{get,set}_idt, causing it to access the ldt instead. Because these functions are only called for save/load on AMD, the bug does not impact normal operation. With the fix, save/load works as expected on AMD hosts. Signed-off-by: Uri Lublin Signed-off-by: Avi Kivity Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/kvm/svm.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/kvm/svm.c b/drivers/kvm/svm.c index 7397bfbbcb1c..717aabb012cc 100644 --- a/drivers/kvm/svm.c +++ b/drivers/kvm/svm.c @@ -680,14 +680,14 @@ static void svm_get_cs_db_l_bits(struct kvm_vcpu *vcpu, int *db, int *l) static void svm_get_idt(struct kvm_vcpu *vcpu, struct descriptor_table *dt) { - dt->limit = vcpu->svm->vmcb->save.ldtr.limit; - dt->base = vcpu->svm->vmcb->save.ldtr.base; + dt->limit = vcpu->svm->vmcb->save.idtr.limit; + dt->base = vcpu->svm->vmcb->save.idtr.base; } static void svm_set_idt(struct kvm_vcpu *vcpu, struct descriptor_table *dt) { - vcpu->svm->vmcb->save.ldtr.limit = dt->limit; - vcpu->svm->vmcb->save.ldtr.base = dt->base ; + vcpu->svm->vmcb->save.idtr.limit = dt->limit; + vcpu->svm->vmcb->save.idtr.base = dt->base ; } static void svm_get_gdt(struct kvm_vcpu *vcpu, struct descriptor_table *dt) From 6f00e68f210c0407dd666743ce61ae543cfd509d Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Fri, 26 Jan 2007 00:56:40 -0800 Subject: [PATCH 139/326] [PATCH] KVM: Emulate IA32_MISC_ENABLE msr This allows netbsd 3.1 i386 to get further along installing. Signed-off-by: Avi Kivity Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/kvm/kvm.h | 1 + drivers/kvm/kvm_main.c | 17 ++++++++++++++++- 2 files changed, 17 insertions(+), 1 deletion(-) diff --git a/drivers/kvm/kvm.h b/drivers/kvm/kvm.h index 91e0c75aca8f..2db1ca4c6800 100644 --- a/drivers/kvm/kvm.h +++ b/drivers/kvm/kvm.h @@ -242,6 +242,7 @@ struct kvm_vcpu { u64 pdptrs[4]; /* pae */ u64 shadow_efer; u64 apic_base; + u64 ia32_misc_enable_msr; int nmsrs; struct vmx_msr_entry *guest_msrs; struct vmx_msr_entry *host_msrs; diff --git a/drivers/kvm/kvm_main.c b/drivers/kvm/kvm_main.c index be4651abe72c..b10972ed0c9f 100644 --- a/drivers/kvm/kvm_main.c +++ b/drivers/kvm/kvm_main.c @@ -1226,6 +1226,9 @@ int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata) case MSR_IA32_APICBASE: data = vcpu->apic_base; break; + case MSR_IA32_MISC_ENABLE: + data = vcpu->ia32_misc_enable_msr; + break; #ifdef CONFIG_X86_64 case MSR_EFER: data = vcpu->shadow_efer; @@ -1297,6 +1300,9 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data) case MSR_IA32_APICBASE: vcpu->apic_base = data; break; + case MSR_IA32_MISC_ENABLE: + vcpu->ia32_misc_enable_msr = data; + break; default: printk(KERN_ERR "kvm: unhandled wrmsr: 0x%x\n", msr); return 1; @@ -1600,6 +1606,10 @@ static u32 msrs_to_save[] = { static unsigned num_msrs_to_save; +static u32 emulated_msrs[] = { + MSR_IA32_MISC_ENABLE, +}; + static __init void kvm_init_msr_list(void) { u32 dummy[2]; @@ -1925,7 +1935,7 @@ static long kvm_dev_ioctl(struct file *filp, if (copy_from_user(&msr_list, user_msr_list, sizeof msr_list)) goto out; n = msr_list.nmsrs; - msr_list.nmsrs = num_msrs_to_save; + msr_list.nmsrs = num_msrs_to_save + ARRAY_SIZE(emulated_msrs); if (copy_to_user(user_msr_list, &msr_list, sizeof msr_list)) goto out; r = -E2BIG; @@ -1935,6 +1945,11 @@ static long kvm_dev_ioctl(struct file *filp, if (copy_to_user(user_msr_list->indices, &msrs_to_save, num_msrs_to_save * sizeof(u32))) goto out; + if (copy_to_user(user_msr_list->indices + + num_msrs_to_save * sizeof(u32), + &emulated_msrs, + ARRAY_SIZE(emulated_msrs) * sizeof(u32))) + goto out; r = 0; break; } From 7993ba43db1c07245ada067791f91dbf018095ac Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Fri, 26 Jan 2007 00:56:41 -0800 Subject: [PATCH 140/326] [PATCH] KVM: MMU: Perform access checks in walk_addr() Check pte permission bits in walk_addr(), instead of scattering the checks all over the code. This has the following benefits: 1. We no longer set the accessed bit for accessed which fail permission checks. 2. Setting the accessed bit is simplified. 3. Under some circumstances, we used to pretend a page fault was fixed when it would actually fail the access checks. This caused an unnecessary vmexit. 4. The error code for guest page faults is now correct. The fix helps netbsd further along booting, and allows kvm to pass the new mmu testsuite. Signed-off-by: Avi Kivity Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/kvm/mmu.c | 10 ------ drivers/kvm/paging_tmpl.h | 68 ++++++++++++++++++++++----------------- 2 files changed, 38 insertions(+), 40 deletions(-) diff --git a/drivers/kvm/mmu.c b/drivers/kvm/mmu.c index c6f972914f08..a05d0609d918 100644 --- a/drivers/kvm/mmu.c +++ b/drivers/kvm/mmu.c @@ -992,16 +992,6 @@ static inline int fix_read_pf(u64 *shadow_ent) return 0; } -static int may_access(u64 pte, int write, int user) -{ - - if (user && !(pte & PT_USER_MASK)) - return 0; - if (write && !(pte & PT_WRITABLE_MASK)) - return 0; - return 1; -} - static void paging_free(struct kvm_vcpu *vcpu) { nonpaging_free(vcpu); diff --git a/drivers/kvm/paging_tmpl.h b/drivers/kvm/paging_tmpl.h index 6bc41950fbb3..afcd2a8f45bb 100644 --- a/drivers/kvm/paging_tmpl.h +++ b/drivers/kvm/paging_tmpl.h @@ -63,13 +63,15 @@ struct guest_walker { pt_element_t *ptep; pt_element_t inherited_ar; gfn_t gfn; + u32 error_code; }; /* * Fetch a guest pte for a guest virtual address */ -static void FNAME(walk_addr)(struct guest_walker *walker, - struct kvm_vcpu *vcpu, gva_t addr) +static int FNAME(walk_addr)(struct guest_walker *walker, + struct kvm_vcpu *vcpu, gva_t addr, + int write_fault, int user_fault) { hpa_t hpa; struct kvm_memory_slot *slot; @@ -86,7 +88,7 @@ static void FNAME(walk_addr)(struct guest_walker *walker, walker->ptep = &vcpu->pdptrs[(addr >> 30) & 3]; root = *walker->ptep; if (!(root & PT_PRESENT_MASK)) - return; + goto not_present; --walker->level; } #endif @@ -111,11 +113,18 @@ static void FNAME(walk_addr)(struct guest_walker *walker, ASSERT(((unsigned long)walker->table & PAGE_MASK) == ((unsigned long)ptep & PAGE_MASK)); - if (is_present_pte(*ptep) && !(*ptep & PT_ACCESSED_MASK)) - *ptep |= PT_ACCESSED_MASK; - if (!is_present_pte(*ptep)) - break; + goto not_present; + + if (write_fault && !is_writeble_pte(*ptep)) + if (user_fault || is_write_protection(vcpu)) + goto access_error; + + if (user_fault && !(*ptep & PT_USER_MASK)) + goto access_error; + + if (!(*ptep & PT_ACCESSED_MASK)) + *ptep |= PT_ACCESSED_MASK; /* avoid rmw */ if (walker->level == PT_PAGE_TABLE_LEVEL) { walker->gfn = (*ptep & PT_BASE_ADDR_MASK) @@ -146,6 +155,21 @@ static void FNAME(walk_addr)(struct guest_walker *walker, } walker->ptep = ptep; pgprintk("%s: pte %llx\n", __FUNCTION__, (u64)*ptep); + return 1; + +not_present: + walker->error_code = 0; + goto err; + +access_error: + walker->error_code = PFERR_PRESENT_MASK; + +err: + if (write_fault) + walker->error_code |= PFERR_WRITE_MASK; + if (user_fault) + walker->error_code |= PFERR_USER_MASK; + return 0; } static void FNAME(release_walker)(struct guest_walker *walker) @@ -347,7 +371,6 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, u32 error_code) { int write_fault = error_code & PFERR_WRITE_MASK; - int pte_present = error_code & PFERR_PRESENT_MASK; int user_fault = error_code & PFERR_USER_MASK; struct guest_walker walker; u64 *shadow_pte; @@ -365,19 +388,19 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, /* * Look up the shadow pte for the faulting address. */ - FNAME(walk_addr)(&walker, vcpu, addr); - shadow_pte = FNAME(fetch)(vcpu, addr, &walker); + r = FNAME(walk_addr)(&walker, vcpu, addr, write_fault, user_fault); /* * The page is not mapped by the guest. Let the guest handle it. */ - if (!shadow_pte) { - pgprintk("%s: not mapped\n", __FUNCTION__); - inject_page_fault(vcpu, addr, error_code); + if (!r) { + pgprintk("%s: guest page fault\n", __FUNCTION__); + inject_page_fault(vcpu, addr, walker.error_code); FNAME(release_walker)(&walker); return 0; } + shadow_pte = FNAME(fetch)(vcpu, addr, &walker); pgprintk("%s: shadow pte %p %llx\n", __FUNCTION__, shadow_pte, *shadow_pte); @@ -399,22 +422,7 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, * mmio: emulate if accessible, otherwise its a guest fault. */ if (is_io_pte(*shadow_pte)) { - if (may_access(*shadow_pte, write_fault, user_fault)) - return 1; - pgprintk("%s: io work, no access\n", __FUNCTION__); - inject_page_fault(vcpu, addr, - error_code | PFERR_PRESENT_MASK); - kvm_mmu_audit(vcpu, "post page fault (io)"); - return 0; - } - - /* - * pte not present, guest page fault. - */ - if (pte_present && !fixed && !write_pt) { - inject_page_fault(vcpu, addr, error_code); - kvm_mmu_audit(vcpu, "post page fault (guest)"); - return 0; + return 1; } ++kvm_stat.pf_fixed; @@ -429,7 +437,7 @@ static gpa_t FNAME(gva_to_gpa)(struct kvm_vcpu *vcpu, gva_t vaddr) pt_element_t guest_pte; gpa_t gpa; - FNAME(walk_addr)(&walker, vcpu, vaddr); + FNAME(walk_addr)(&walker, vcpu, vaddr, 0, 0); guest_pte = *walker.ptep; FNAME(release_walker)(&walker); From 73b1087e6176a34c01eea3db269848f72fad72c1 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Fri, 26 Jan 2007 00:56:41 -0800 Subject: [PATCH 141/326] [PATCH] KVM: MMU: Report nx faults to the guest With the recent guest page fault change, we perform access checks on our own instead of relying on the cpu. This means we have to perform the nx checks as well. Software like the google toolbar on windows appears to rely on this somehow. Signed-off-by: Avi Kivity Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/kvm/mmu.c | 6 ++++++ drivers/kvm/paging_tmpl.h | 15 ++++++++++++--- 2 files changed, 18 insertions(+), 3 deletions(-) diff --git a/drivers/kvm/mmu.c b/drivers/kvm/mmu.c index a05d0609d918..22c426cd8cb2 100644 --- a/drivers/kvm/mmu.c +++ b/drivers/kvm/mmu.c @@ -143,6 +143,7 @@ static int dbg = 1; #define PFERR_PRESENT_MASK (1U << 0) #define PFERR_WRITE_MASK (1U << 1) #define PFERR_USER_MASK (1U << 2) +#define PFERR_FETCH_MASK (1U << 4) #define PT64_ROOT_LEVEL 4 #define PT32_ROOT_LEVEL 2 @@ -168,6 +169,11 @@ static int is_cpuid_PSE36(void) return 1; } +static int is_nx(struct kvm_vcpu *vcpu) +{ + return vcpu->shadow_efer & EFER_NX; +} + static int is_present_pte(unsigned long pte) { return pte & PT_PRESENT_MASK; diff --git a/drivers/kvm/paging_tmpl.h b/drivers/kvm/paging_tmpl.h index afcd2a8f45bb..149fa45fd9a5 100644 --- a/drivers/kvm/paging_tmpl.h +++ b/drivers/kvm/paging_tmpl.h @@ -71,7 +71,7 @@ struct guest_walker { */ static int FNAME(walk_addr)(struct guest_walker *walker, struct kvm_vcpu *vcpu, gva_t addr, - int write_fault, int user_fault) + int write_fault, int user_fault, int fetch_fault) { hpa_t hpa; struct kvm_memory_slot *slot; @@ -123,6 +123,11 @@ static int FNAME(walk_addr)(struct guest_walker *walker, if (user_fault && !(*ptep & PT_USER_MASK)) goto access_error; +#if PTTYPE == 64 + if (fetch_fault && is_nx(vcpu) && (*ptep & PT64_NX_MASK)) + goto access_error; +#endif + if (!(*ptep & PT_ACCESSED_MASK)) *ptep |= PT_ACCESSED_MASK; /* avoid rmw */ @@ -169,6 +174,8 @@ static int FNAME(walk_addr)(struct guest_walker *walker, walker->error_code |= PFERR_WRITE_MASK; if (user_fault) walker->error_code |= PFERR_USER_MASK; + if (fetch_fault) + walker->error_code |= PFERR_FETCH_MASK; return 0; } @@ -372,6 +379,7 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, { int write_fault = error_code & PFERR_WRITE_MASK; int user_fault = error_code & PFERR_USER_MASK; + int fetch_fault = error_code & PFERR_FETCH_MASK; struct guest_walker walker; u64 *shadow_pte; int fixed; @@ -388,7 +396,8 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, /* * Look up the shadow pte for the faulting address. */ - r = FNAME(walk_addr)(&walker, vcpu, addr, write_fault, user_fault); + r = FNAME(walk_addr)(&walker, vcpu, addr, write_fault, user_fault, + fetch_fault); /* * The page is not mapped by the guest. Let the guest handle it. @@ -437,7 +446,7 @@ static gpa_t FNAME(gva_to_gpa)(struct kvm_vcpu *vcpu, gva_t vaddr) pt_element_t guest_pte; gpa_t gpa; - FNAME(walk_addr)(&walker, vcpu, vaddr, 0, 0); + FNAME(walk_addr)(&walker, vcpu, vaddr, 0, 0, 0); guest_pte = *walker.ptep; FNAME(release_walker)(&walker); From 46fe4ddd9dbb15305ab9b458e6cfa4dd47ac3e47 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Fri, 26 Jan 2007 00:56:42 -0800 Subject: [PATCH 142/326] [PATCH] KVM: SVM: Propagate cpu shutdown events to userspace This patch implements forwarding of SHUTDOWN intercepts from the guest on to userspace on AMD SVM. A SHUTDOWN event occurs when the guest produces a triple fault (e.g. on reboot). This also fixes the bug that a guest reboot actually causes a host reboot under some circumstances. Signed-off-by: Joerg Roedel Signed-off-by: Avi Kivity Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/kvm/svm.c | 15 +++++++++++++++ include/linux/kvm.h | 1 + 2 files changed, 16 insertions(+) diff --git a/drivers/kvm/svm.c b/drivers/kvm/svm.c index 717aabb012cc..9c70ff65e6b7 100644 --- a/drivers/kvm/svm.c +++ b/drivers/kvm/svm.c @@ -502,6 +502,7 @@ static void init_vmcb(struct vmcb *vmcb) (1ULL << INTERCEPT_IOIO_PROT) | (1ULL << INTERCEPT_MSR_PROT) | (1ULL << INTERCEPT_TASK_SWITCH) | + (1ULL << INTERCEPT_SHUTDOWN) | (1ULL << INTERCEPT_VMRUN) | (1ULL << INTERCEPT_VMMCALL) | (1ULL << INTERCEPT_VMLOAD) | @@ -892,6 +893,19 @@ static int pf_interception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) return 0; } +static int shutdown_interception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + /* + * VMCB is undefined after a SHUTDOWN intercept + * so reinitialize it. + */ + memset(vcpu->svm->vmcb, 0, PAGE_SIZE); + init_vmcb(vcpu->svm->vmcb); + + kvm_run->exit_reason = KVM_EXIT_SHUTDOWN; + return 0; +} + static int io_get_override(struct kvm_vcpu *vcpu, struct vmcb_seg **seg, int *addr_override) @@ -1249,6 +1263,7 @@ static int (*svm_exit_handlers[])(struct kvm_vcpu *vcpu, [SVM_EXIT_IOIO] = io_interception, [SVM_EXIT_MSR] = msr_interception, [SVM_EXIT_TASK_SWITCH] = task_switch_interception, + [SVM_EXIT_SHUTDOWN] = shutdown_interception, [SVM_EXIT_VMRUN] = invalid_op_interception, [SVM_EXIT_VMMCALL] = invalid_op_interception, [SVM_EXIT_VMLOAD] = invalid_op_interception, diff --git a/include/linux/kvm.h b/include/linux/kvm.h index bc8b4616bad7..1be148f0fce4 100644 --- a/include/linux/kvm.h +++ b/include/linux/kvm.h @@ -46,6 +46,7 @@ enum kvm_exit_reason { KVM_EXIT_HLT = 5, KVM_EXIT_MMIO = 6, KVM_EXIT_IRQ_WINDOW_OPEN = 7, + KVM_EXIT_SHUTDOWN = 8, }; /* for KVM_RUN */ From 8736b9270c2f8993ca44c30f64d4c6d25e379687 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Fri, 26 Jan 2007 00:56:43 -0800 Subject: [PATCH 143/326] [PATCH] S3C24XX: fix passing spi chipselect to select routine It turns out that the spi chipselect was not being passed to the set_cs routine if one was specified in the platform data. As part of the fix, change to using a set_cs field in the controller state, and put a default gpio routine in if the data passed does not specify it. Also remove the //#define DEBUG Signed-off-by: Ben Dooks Signed-off-by: David Brownell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/spi/spi_s3c24xx.c | 28 ++++++++++++++-------------- 1 file changed, 14 insertions(+), 14 deletions(-) diff --git a/drivers/spi/spi_s3c24xx.c b/drivers/spi/spi_s3c24xx.c index 8ca08713528e..651379c51ae6 100644 --- a/drivers/spi/spi_s3c24xx.c +++ b/drivers/spi/spi_s3c24xx.c @@ -10,9 +10,6 @@ * */ - -//#define DEBUG - #include #include #include @@ -44,6 +41,9 @@ struct s3c24xx_spi { int len; int count; + int (*set_cs)(struct s3c2410_spi_info *spi, + int cs, int pol); + /* data buffers */ const unsigned char *tx; unsigned char *rx; @@ -64,6 +64,11 @@ static inline struct s3c24xx_spi *to_hw(struct spi_device *sdev) return spi_master_get_devdata(sdev->master); } +static void s3c24xx_spi_gpiocs(struct s3c2410_spi_info *spi, int cs, int pol) +{ + s3c2410_gpio_setpin(spi->pin_cs, pol); +} + static void s3c24xx_spi_chipsel(struct spi_device *spi, int value) { struct s3c24xx_spi *hw = to_hw(spi); @@ -72,10 +77,7 @@ static void s3c24xx_spi_chipsel(struct spi_device *spi, int value) switch (value) { case BITBANG_CS_INACTIVE: - if (hw->pdata->set_cs) - hw->pdata->set_cs(hw->pdata, value, cspol); - else - s3c2410_gpio_setpin(hw->pdata->pin_cs, cspol ^ 1); + hw->pdata->set_cs(hw->pdata, spi->chip_select, cspol^1); break; case BITBANG_CS_ACTIVE: @@ -96,14 +98,9 @@ static void s3c24xx_spi_chipsel(struct spi_device *spi, int value) /* write new configration */ writeb(spcon, hw->regs + S3C2410_SPCON); - - if (hw->pdata->set_cs) - hw->pdata->set_cs(hw->pdata, value, cspol); - else - s3c2410_gpio_setpin(hw->pdata->pin_cs, cspol); + hw->pdata->set_cs(hw->pdata, spi->chip_select, cspol); break; - } } @@ -330,9 +327,12 @@ static int s3c24xx_spi_probe(struct platform_device *pdev) /* setup any gpio we can */ if (!hw->pdata->set_cs) { + hw->set_cs = s3c24xx_spi_gpiocs; + s3c2410_gpio_setpin(hw->pdata->pin_cs, 1); s3c2410_gpio_cfgpin(hw->pdata->pin_cs, S3C2410_GPIO_OUTPUT); - } + } else + hw->set_cs = hw->pdata->set_cs; /* register our spi controller */ From 7f6ee1adc75bf31d1b76814338f76a88e653cb60 Mon Sep 17 00:00:00 2001 From: Justin Clacherty Date: Fri, 26 Jan 2007 00:56:44 -0800 Subject: [PATCH 144/326] [PATCH] spi: fix error setting the spi mode in pxa2xx_spi.c Currently the spi mode can be set to the wrong mode if you are switching from any mode other than mode 0. This is because the mode is set using a bitwise or on uncleared bits. The following patch clears the mode bits before setting the new mode. I've also modified it to use the appropriate defines from pxa-regs.h for readability. Signed-off-by: Justin Clacherty Signed-off-by: David Brownell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/spi/pxa2xx_spi.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/spi/pxa2xx_spi.c b/drivers/spi/pxa2xx_spi.c index 6ed3f1da9296..8b41f9cc2560 100644 --- a/drivers/spi/pxa2xx_spi.c +++ b/drivers/spi/pxa2xx_spi.c @@ -1169,8 +1169,9 @@ static int setup(struct spi_device *spi) spi->bits_per_word - 16 : spi->bits_per_word) | SSCR0_SSE | (spi->bits_per_word > 16 ? SSCR0_EDSS : 0); - chip->cr1 |= (((spi->mode & SPI_CPHA) != 0) << 4) - | (((spi->mode & SPI_CPOL) != 0) << 3); + chip->cr1 &= ~(SSCR1_SPO | SSCR1_SPH); + chip->cr1 |= (((spi->mode & SPI_CPHA) != 0) ? SSCR1_SPH : 0) + | (((spi->mode & SPI_CPOL) != 0) ? SSCR1_SPO : 0); /* NOTE: PXA25x_SSP _could_ use external clocking ... */ if (drv_data->ssp_type != PXA25x_SSP) From a1f3bb9ae4497a2ed3eac773fd7798ac33a0371f Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Fri, 26 Jan 2007 00:56:46 -0800 Subject: [PATCH 145/326] [PATCH] Fix CONFIG_COMPAT_VDSO I wouldn't mind if CONFIG_COMPAT_VDSO went away entirely. But if it's there, it should work properly. Currently it's quite haphazard: both real vma and fixmap are mapped, both are put in the two different AT_* slots, sysenter returns to the vma address rather than the fixmap address, and core dumps yet are another story. This patch makes CONFIG_COMPAT_VDSO disable the real vma and use the fixmap area consistently. This makes it actually compatible with what the old vdso implementation did. Signed-off-by: Roland McGrath Cc: Ingo Molnar Cc: Paul Mackerras Cc: Benjamin Herrenschmidt Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/entry.S | 4 ++++ arch/i386/kernel/sysenter.c | 2 ++ include/asm-i386/elf.h | 7 +++---- include/asm-i386/fixmap.h | 2 ++ include/asm-i386/page.h | 2 ++ 5 files changed, 13 insertions(+), 4 deletions(-) diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S index 06461b8b715d..5e47683fc63a 100644 --- a/arch/i386/kernel/entry.S +++ b/arch/i386/kernel/entry.S @@ -302,12 +302,16 @@ sysenter_past_esp: pushl $(__USER_CS) CFI_ADJUST_CFA_OFFSET 4 /*CFI_REL_OFFSET cs, 0*/ +#ifndef CONFIG_COMPAT_VDSO /* * Push current_thread_info()->sysenter_return to the stack. * A tiny bit of offset fixup is necessary - 4*4 means the 4 words * pushed above; +8 corresponds to copy_thread's esp0 setting. */ pushl (TI_sysenter_return-THREAD_SIZE+8+4*4)(%esp) +#else + pushl $SYSENTER_RETURN +#endif CFI_ADJUST_CFA_OFFSET 4 CFI_REL_OFFSET eip, 0 diff --git a/arch/i386/kernel/sysenter.c b/arch/i386/kernel/sysenter.c index 7de9117b5a3a..454d12df59ea 100644 --- a/arch/i386/kernel/sysenter.c +++ b/arch/i386/kernel/sysenter.c @@ -100,6 +100,7 @@ int __init sysenter_setup(void) return 0; } +#ifndef CONFIG_COMPAT_VDSO static struct page *syscall_nopage(struct vm_area_struct *vma, unsigned long adr, int *type) { @@ -187,3 +188,4 @@ int in_gate_area_no_task(unsigned long addr) { return 0; } +#endif diff --git a/include/asm-i386/elf.h b/include/asm-i386/elf.h index 45d21a0c95bf..0515d61d5411 100644 --- a/include/asm-i386/elf.h +++ b/include/asm-i386/elf.h @@ -143,11 +143,8 @@ extern int dump_task_extended_fpu (struct task_struct *, struct user_fxsr_struct # define VDSO_PRELINK 0 #endif -#define VDSO_COMPAT_SYM(x) \ - (VDSO_COMPAT_BASE + (unsigned long)(x) - VDSO_PRELINK) - #define VDSO_SYM(x) \ - (VDSO_BASE + (unsigned long)(x) - VDSO_PRELINK) + (VDSO_COMPAT_BASE + (unsigned long)(x) - VDSO_PRELINK) #define VDSO_HIGH_EHDR ((const struct elfhdr *) VDSO_HIGH_BASE) #define VDSO_EHDR ((const struct elfhdr *) VDSO_COMPAT_BASE) @@ -156,10 +153,12 @@ extern void __kernel_vsyscall; #define VDSO_ENTRY VDSO_SYM(&__kernel_vsyscall) +#ifndef CONFIG_COMPAT_VDSO #define ARCH_HAS_SETUP_ADDITIONAL_PAGES struct linux_binprm; extern int arch_setup_additional_pages(struct linux_binprm *bprm, int executable_stack); +#endif extern unsigned int vdso_enabled; diff --git a/include/asm-i386/fixmap.h b/include/asm-i386/fixmap.h index 02428cb36621..3e9f610c35df 100644 --- a/include/asm-i386/fixmap.h +++ b/include/asm-i386/fixmap.h @@ -23,6 +23,8 @@ extern unsigned long __FIXADDR_TOP; #else #define __FIXADDR_TOP 0xfffff000 +#define FIXADDR_USER_START __fix_to_virt(FIX_VDSO) +#define FIXADDR_USER_END __fix_to_virt(FIX_VDSO - 1) #endif #ifndef __ASSEMBLY__ diff --git a/include/asm-i386/page.h b/include/asm-i386/page.h index fd3f64ace248..7b19f454761d 100644 --- a/include/asm-i386/page.h +++ b/include/asm-i386/page.h @@ -143,7 +143,9 @@ extern int page_is_ram(unsigned long pagenr); #include #include +#ifndef CONFIG_COMPAT_VDSO #define __HAVE_ARCH_GATE_AREA 1 +#endif #endif /* __KERNEL__ */ #endif /* _I386_PAGE_H */ From b6558c4a2378af06f2beca6c8a3304e21d1cf135 Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Fri, 26 Jan 2007 00:56:47 -0800 Subject: [PATCH 146/326] [PATCH] Fix gate_vma.vm_flags This patch fixes the initialization of gate_vma.vm_flags and gate_vma.vm_page_prot to reflect reality. This makes the "[vdso]" line in /proc/PID/maps correctly show r-xp instead of ---p, when gate_vma is used (CONFIG_COMPAT_VDSO on i386). Signed-off-by: Roland McGrath Cc: Ingo Molnar Cc: Paul Mackerras Cc: Benjamin Herrenschmidt Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memory.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/mm/memory.c b/mm/memory.c index af227d26e104..5beb4b894c5a 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -2606,8 +2606,8 @@ static int __init gate_vma_init(void) gate_vma.vm_mm = NULL; gate_vma.vm_start = FIXADDR_USER_START; gate_vma.vm_end = FIXADDR_USER_END; - gate_vma.vm_page_prot = PAGE_READONLY; - gate_vma.vm_flags = 0; + gate_vma.vm_flags = VM_READ | VM_MAYREAD | VM_EXEC | VM_MAYEXEC; + gate_vma.vm_page_prot = __P101; return 0; } __initcall(gate_vma_init); From e5b97dde514f9bd43f9e525451d0a863c4fc8a9a Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Fri, 26 Jan 2007 00:56:48 -0800 Subject: [PATCH 147/326] [PATCH] Add VM_ALWAYSDUMP This patch adds the VM_ALWAYSDUMP flag for vm_flags in vm_area_struct. This provides a clean explicit way to have a vma always included in core dumps, as is needed for vDSO's. Signed-off-by: Roland McGrath Cc: Ingo Molnar Cc: Paul Mackerras Cc: Benjamin Herrenschmidt Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/binfmt_elf.c | 4 ++++ include/linux/mm.h | 1 + 2 files changed, 5 insertions(+) diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 7cb28720f90e..6fec8bfa6bac 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -1178,6 +1178,10 @@ static int dump_seek(struct file *file, loff_t off) */ static int maydump(struct vm_area_struct *vma) { + /* The vma can be set up to tell us the answer directly. */ + if (vma->vm_flags & VM_ALWAYSDUMP) + return 1; + /* Do not dump I/O mapped devices or special mappings */ if (vma->vm_flags & (VM_IO | VM_RESERVED)) return 0; diff --git a/include/linux/mm.h b/include/linux/mm.h index 76912231af41..2d2c08d5f473 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -168,6 +168,7 @@ extern unsigned int kobjsize(const void *objp); #define VM_NONLINEAR 0x00800000 /* Is non-linear (remap_file_pages) */ #define VM_MAPPED_COPY 0x01000000 /* T if mapped copy of data (nommu mmap) */ #define VM_INSERTPAGE 0x02000000 /* The vma has had "vm_insert_page()" done on it */ +#define VM_ALWAYSDUMP 0x04000000 /* Always include in core dumps */ #ifndef VM_STACK_DEFAULT_FLAGS /* arch can override this */ #define VM_STACK_DEFAULT_FLAGS VM_DATA_DEFAULT_FLAGS From f47aef55d9a18945fcdd7fd6bf01121ce973b91b Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Fri, 26 Jan 2007 00:56:49 -0800 Subject: [PATCH 148/326] [PATCH] i386 vDSO: use VM_ALWAYSDUMP This patch fixes core dumps to include the vDSO vma, which is left out now. It removes the special-case core writing macros, which were not doing the right thing for the vDSO vma anyway. Instead, it uses VM_ALWAYSDUMP in the vma; there is no need for the fixmap page to be installed. It handles the CONFIG_COMPAT_VDSO case by making elf_core_dump use the fake vma from get_gate_vma after real vmas in the same way the /proc/PID/maps code does. This changes core dumps so they no longer include the non-PT_LOAD phdrs from the vDSO. I made the change to add them in the first place, but in turned out that nothing ever wanted them there since the advent of NT_AUXV. It's cleaner to leave them out, and just let the phdrs inside the vDSO image speak for themselves. Signed-off-by: Roland McGrath Cc: Ingo Molnar Cc: Paul Mackerras Cc: Benjamin Herrenschmidt Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/sysenter.c | 12 +++++----- fs/binfmt_elf.c | 38 +++++++++++++++++++++++++++++--- include/asm-i386/elf.h | 44 ------------------------------------- mm/memory.c | 7 ++++++ 4 files changed, 49 insertions(+), 52 deletions(-) diff --git a/arch/i386/kernel/sysenter.c b/arch/i386/kernel/sysenter.c index 454d12df59ea..5da744204d10 100644 --- a/arch/i386/kernel/sysenter.c +++ b/arch/i386/kernel/sysenter.c @@ -79,11 +79,6 @@ int __init sysenter_setup(void) #ifdef CONFIG_COMPAT_VDSO __set_fixmap(FIX_VDSO, __pa(syscall_page), PAGE_READONLY); printk("Compat vDSO mapped to %08lx.\n", __fix_to_virt(FIX_VDSO)); -#else - /* - * In the non-compat case the ELF coredumping code needs the fixmap: - */ - __set_fixmap(FIX_VDSO, __pa(syscall_page), PAGE_KERNEL_RO); #endif if (!boot_cpu_has(X86_FEATURE_SEP)) { @@ -147,6 +142,13 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, int exstack) vma->vm_end = addr + PAGE_SIZE; /* MAYWRITE to allow gdb to COW and set breakpoints */ vma->vm_flags = VM_READ|VM_EXEC|VM_MAYREAD|VM_MAYEXEC|VM_MAYWRITE; + /* + * Make sure the vDSO gets into every core dump. + * Dumping its contents makes post-mortem fully interpretable later + * without matching up the same kernel and hardware config to see + * what PC values meant. + */ + vma->vm_flags |= VM_ALWAYSDUMP; vma->vm_flags |= mm->def_flags; vma->vm_page_prot = protection_map[vma->vm_flags & 7]; vma->vm_ops = &syscall_vm_ops; diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 6fec8bfa6bac..90461f49e902 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -1428,6 +1428,32 @@ static int elf_dump_thread_status(long signr, struct elf_thread_status *t) return sz; } +static struct vm_area_struct *first_vma(struct task_struct *tsk, + struct vm_area_struct *gate_vma) +{ + struct vm_area_struct *ret = tsk->mm->mmap; + + if (ret) + return ret; + return gate_vma; +} +/* + * Helper function for iterating across a vma list. It ensures that the caller + * will visit `gate_vma' prior to terminating the search. + */ +static struct vm_area_struct *next_vma(struct vm_area_struct *this_vma, + struct vm_area_struct *gate_vma) +{ + struct vm_area_struct *ret; + + ret = this_vma->vm_next; + if (ret) + return ret; + if (this_vma == gate_vma) + return NULL; + return gate_vma; +} + /* * Actual dumper * @@ -1443,7 +1469,7 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file) int segs; size_t size = 0; int i; - struct vm_area_struct *vma; + struct vm_area_struct *vma, *gate_vma; struct elfhdr *elf = NULL; loff_t offset = 0, dataoff, foffset; unsigned long limit = current->signal->rlim[RLIMIT_CORE].rlim_cur; @@ -1529,6 +1555,10 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file) segs += ELF_CORE_EXTRA_PHDRS; #endif + gate_vma = get_gate_vma(current); + if (gate_vma != NULL) + segs++; + /* Set up header */ fill_elf_header(elf, segs + 1); /* including notes section */ @@ -1596,7 +1626,8 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file) dataoff = offset = roundup(offset, ELF_EXEC_PAGESIZE); /* Write program headers for segments dump */ - for (vma = current->mm->mmap; vma != NULL; vma = vma->vm_next) { + for (vma = first_vma(current, gate_vma); vma != NULL; + vma = next_vma(vma, gate_vma)) { struct elf_phdr phdr; size_t sz; @@ -1645,7 +1676,8 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file) /* Align to page */ DUMP_SEEK(dataoff - foffset); - for (vma = current->mm->mmap; vma != NULL; vma = vma->vm_next) { + for (vma = first_vma(current, gate_vma); vma != NULL; + vma = next_vma(vma, gate_vma)) { unsigned long addr; if (!maydump(vma)) diff --git a/include/asm-i386/elf.h b/include/asm-i386/elf.h index 0515d61d5411..369035dfe4b6 100644 --- a/include/asm-i386/elf.h +++ b/include/asm-i386/elf.h @@ -168,50 +168,6 @@ do if (vdso_enabled) { \ NEW_AUX_ENT(AT_SYSINFO_EHDR, VDSO_COMPAT_BASE); \ } while (0) -/* - * These macros parameterize elf_core_dump in fs/binfmt_elf.c to write out - * extra segments containing the vsyscall DSO contents. Dumping its - * contents makes post-mortem fully interpretable later without matching up - * the same kernel and hardware config to see what PC values meant. - * Dumping its extra ELF program headers includes all the other information - * a debugger needs to easily find how the vsyscall DSO was being used. - */ -#define ELF_CORE_EXTRA_PHDRS (VDSO_HIGH_EHDR->e_phnum) -#define ELF_CORE_WRITE_EXTRA_PHDRS \ -do { \ - const struct elf_phdr *const vsyscall_phdrs = \ - (const struct elf_phdr *) (VDSO_HIGH_BASE \ - + VDSO_HIGH_EHDR->e_phoff); \ - int i; \ - Elf32_Off ofs = 0; \ - for (i = 0; i < VDSO_HIGH_EHDR->e_phnum; ++i) { \ - struct elf_phdr phdr = vsyscall_phdrs[i]; \ - if (phdr.p_type == PT_LOAD) { \ - BUG_ON(ofs != 0); \ - ofs = phdr.p_offset = offset; \ - phdr.p_memsz = PAGE_ALIGN(phdr.p_memsz); \ - phdr.p_filesz = phdr.p_memsz; \ - offset += phdr.p_filesz; \ - } \ - else \ - phdr.p_offset += ofs; \ - phdr.p_paddr = 0; /* match other core phdrs */ \ - DUMP_WRITE(&phdr, sizeof(phdr)); \ - } \ -} while (0) -#define ELF_CORE_WRITE_EXTRA_DATA \ -do { \ - const struct elf_phdr *const vsyscall_phdrs = \ - (const struct elf_phdr *) (VDSO_HIGH_BASE \ - + VDSO_HIGH_EHDR->e_phoff); \ - int i; \ - for (i = 0; i < VDSO_HIGH_EHDR->e_phnum; ++i) { \ - if (vsyscall_phdrs[i].p_type == PT_LOAD) \ - DUMP_WRITE((void *) vsyscall_phdrs[i].p_vaddr, \ - PAGE_ALIGN(vsyscall_phdrs[i].p_memsz)); \ - } \ -} while (0) - #endif #endif diff --git a/mm/memory.c b/mm/memory.c index 5beb4b894c5a..ef09f0acb1d8 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -2608,6 +2608,13 @@ static int __init gate_vma_init(void) gate_vma.vm_end = FIXADDR_USER_END; gate_vma.vm_flags = VM_READ | VM_MAYREAD | VM_EXEC | VM_MAYEXEC; gate_vma.vm_page_prot = __P101; + /* + * Make sure the vDSO gets into every core dump. + * Dumping its contents makes post-mortem fully interpretable later + * without matching up the same kernel and hardware config to see + * what PC values meant. + */ + gate_vma.vm_flags |= VM_ALWAYSDUMP; return 0; } __initcall(gate_vma_init); From e03f0ca11645ce69a4defcd4f60a5cb2d5e30507 Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Fri, 26 Jan 2007 00:56:50 -0800 Subject: [PATCH 149/326] [PATCH] x86_64 ia32 vDSO: use VM_ALWAYSDUMP This patch fixes ia32 core dumps on x86_64 to include just one phdr for the vDSO vma. Currently it writes a confused format with two phdrs for the address, one without contents and one with. This patch removes the special-case core writing macros for the ia32 vDSO. Instead, it uses VM_ALWAYSDUMP in the vma. This changes core dumps so they no longer include the non-PT_LOAD phdrs from the vDSO, consistent with fixed native i386 core dumps. Signed-off-by: Roland McGrath Cc: Ingo Molnar Cc: Paul Mackerras Cc: Benjamin Herrenschmidt Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86_64/ia32/ia32_binfmt.c | 49 ---------------------------------- arch/x86_64/ia32/syscall32.c | 7 +++++ 2 files changed, 7 insertions(+), 49 deletions(-) diff --git a/arch/x86_64/ia32/ia32_binfmt.c b/arch/x86_64/ia32/ia32_binfmt.c index 543ef4f405e9..5ce0bd486bbf 100644 --- a/arch/x86_64/ia32/ia32_binfmt.c +++ b/arch/x86_64/ia32/ia32_binfmt.c @@ -64,55 +64,6 @@ typedef unsigned int elf_greg_t; #define ELF_NGREG (sizeof (struct user_regs_struct32) / sizeof(elf_greg_t)) typedef elf_greg_t elf_gregset_t[ELF_NGREG]; -/* - * These macros parameterize elf_core_dump in fs/binfmt_elf.c to write out - * extra segments containing the vsyscall DSO contents. Dumping its - * contents makes post-mortem fully interpretable later without matching up - * the same kernel and hardware config to see what PC values meant. - * Dumping its extra ELF program headers includes all the other information - * a debugger needs to easily find how the vsyscall DSO was being used. - */ -#define ELF_CORE_EXTRA_PHDRS (find_vma(current->mm, VSYSCALL32_BASE) ? \ - (VSYSCALL32_EHDR->e_phnum) : 0) -#define ELF_CORE_WRITE_EXTRA_PHDRS \ -do { \ - if (find_vma(current->mm, VSYSCALL32_BASE)) { \ - const struct elf32_phdr *const vsyscall_phdrs = \ - (const struct elf32_phdr *) (VSYSCALL32_BASE \ - + VSYSCALL32_EHDR->e_phoff);\ - int i; \ - Elf32_Off ofs = 0; \ - for (i = 0; i < VSYSCALL32_EHDR->e_phnum; ++i) { \ - struct elf32_phdr phdr = vsyscall_phdrs[i]; \ - if (phdr.p_type == PT_LOAD) { \ - BUG_ON(ofs != 0); \ - ofs = phdr.p_offset = offset; \ - phdr.p_memsz = PAGE_ALIGN(phdr.p_memsz); \ - phdr.p_filesz = phdr.p_memsz; \ - offset += phdr.p_filesz; \ - } \ - else \ - phdr.p_offset += ofs; \ - phdr.p_paddr = 0; /* match other core phdrs */ \ - DUMP_WRITE(&phdr, sizeof(phdr)); \ - } \ - } \ -} while (0) -#define ELF_CORE_WRITE_EXTRA_DATA \ -do { \ - if (find_vma(current->mm, VSYSCALL32_BASE)) { \ - const struct elf32_phdr *const vsyscall_phdrs = \ - (const struct elf32_phdr *) (VSYSCALL32_BASE \ - + VSYSCALL32_EHDR->e_phoff); \ - int i; \ - for (i = 0; i < VSYSCALL32_EHDR->e_phnum; ++i) { \ - if (vsyscall_phdrs[i].p_type == PT_LOAD) \ - DUMP_WRITE((void *) (u64) vsyscall_phdrs[i].p_vaddr,\ - PAGE_ALIGN(vsyscall_phdrs[i].p_memsz)); \ - } \ - } \ -} while (0) - struct elf_siginfo { int si_signo; /* signal number */ diff --git a/arch/x86_64/ia32/syscall32.c b/arch/x86_64/ia32/syscall32.c index 3e5ed20cba45..3ac935535e20 100644 --- a/arch/x86_64/ia32/syscall32.c +++ b/arch/x86_64/ia32/syscall32.c @@ -59,6 +59,13 @@ int syscall32_setup_pages(struct linux_binprm *bprm, int exstack) vma->vm_end = VSYSCALL32_END; /* MAYWRITE to allow gdb to COW and set breakpoints */ vma->vm_flags = VM_READ|VM_EXEC|VM_MAYREAD|VM_MAYEXEC|VM_MAYWRITE; + /* + * Make sure the vDSO gets into every core dump. + * Dumping its contents makes post-mortem fully interpretable later + * without matching up the same kernel and hardware config to see + * what PC values meant. + */ + vma->vm_flags |= VM_ALWAYSDUMP; vma->vm_flags |= mm->def_flags; vma->vm_page_prot = protection_map[vma->vm_flags & 7]; vma->vm_ops = &syscall32_vm_ops; From 3a0cfadb42e0a6dd221aa49232a65d2771063285 Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Fri, 26 Jan 2007 00:56:51 -0800 Subject: [PATCH 150/326] [PATCH] powerpc vDSO: use VM_ALWAYSDUMP This patch fixes core dumps to include the vDSO vma, which is left out now. Signed-off-by: Roland McGrath Cc: Ingo Molnar Cc: Paul Mackerras Acked-by: Benjamin Herrenschmidt Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/powerpc/kernel/vdso.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/arch/powerpc/kernel/vdso.c b/arch/powerpc/kernel/vdso.c index a4b28c73bba0..ae0ede19879d 100644 --- a/arch/powerpc/kernel/vdso.c +++ b/arch/powerpc/kernel/vdso.c @@ -284,6 +284,13 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, * pages though */ vma->vm_flags = VM_READ|VM_EXEC|VM_MAYREAD|VM_MAYWRITE|VM_MAYEXEC; + /* + * Make sure the vDSO gets into every core dump. + * Dumping its contents makes post-mortem fully interpretable later + * without matching up the same kernel and hardware config to see + * what PC values meant. + */ + vma->vm_flags |= VM_ALWAYSDUMP; vma->vm_flags |= mm->def_flags; vma->vm_page_prot = protection_map[vma->vm_flags & 0x7]; vma->vm_ops = &vdso_vmops; From c633090e3105e779c97d4978e5e3d7d66b291cfb Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Fri, 26 Jan 2007 00:56:52 -0800 Subject: [PATCH 151/326] [PATCH] x86_64 ia32 vDSO: define arch_vma_name This patch makes x86_64 define arch_vma_name for CONFIG_IA32_EMULATION. This makes the ia32 vDSO mapping appear in /proc/PID/maps with "[vdso]" for ia32 processes, as it does on native i386. Signed-off-by: Roland McGrath Cc: Ingo Molnar Cc: Paul Mackerras Cc: Benjamin Herrenschmidt Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86_64/ia32/syscall32.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/arch/x86_64/ia32/syscall32.c b/arch/x86_64/ia32/syscall32.c index 3ac935535e20..59f1fa155915 100644 --- a/arch/x86_64/ia32/syscall32.c +++ b/arch/x86_64/ia32/syscall32.c @@ -82,6 +82,14 @@ int syscall32_setup_pages(struct linux_binprm *bprm, int exstack) return 0; } +const char *arch_vma_name(struct vm_area_struct *vma) +{ + if (vma->vm_start == VSYSCALL32_BASE && + vma->vm_mm && vma->vm_mm->task_size == IA32_PAGE_OFFSET) + return "[vdso]"; + return NULL; +} + static int __init init_syscall32(void) { syscall32_page = (void *)get_zeroed_page(GFP_KERNEL); From 863c47028eb469c9e6c4e4287b01bea2bbf78766 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Fri, 26 Jan 2007 00:56:53 -0800 Subject: [PATCH 152/326] [PATCH] Fix NULL ->nsproxy dereference in /proc/*/mounts /proc/*/mounstats was fixed, all right, but... To reproduce: while true; do find /proc -type f 2>/dev/null | xargs cat 1>/dev/null 2>/dev/null; done BUG: unable to handle kernel NULL pointer dereference at virtual address 0000000c printing eip: c01754df *pde = 00000000 Oops: 0000 [#28] Modules linked in: af_packet ohci_hcd e1000 ehci_hcd uhci_hcd usbcore xfs CPU: 0 EIP: 0060:[] Not tainted VLI EFLAGS: 00010286 (2.6.20-rc5 #1) EIP is at mounts_open+0x1c/0xac eax: 00000000 ebx: d5898ac0 ecx: d1d27b18 edx: d1d27a50 esi: e6083e10 edi: d3c87f38 ebp: d5898ac0 esp: d3c87ef0 ds: 007b es: 007b ss: 0068 Process cat (pid: 18071, ti=d3c86000 task=f7d5f070 task.ti=d3c86000) Stack: d5898ac0 e6083e10 d3c87f38 c01754c3 c0147c91 c18c52c0 d343f314 d5898ac0 00008000 d3c87f38 ffffff9c c0147e09 d5898ac0 00000000 00000000 c0147e4b 00000000 d3c87f38 d343f314 c18c52c0 c015e53e 00001000 08051000 00000101 Call Trace: [] mounts_open+0x0/0xac [] __dentry_open+0xa1/0x18c [] nameidata_to_filp+0x31/0x3a [] do_filp_open+0x39/0x40 [] seq_read+0x128/0x2aa [] do_sys_open+0x3a/0x6d [] sys_open+0x1c/0x20 [] sysenter_past_esp+0x5f/0x85 [] unix_stream_recvmsg+0x3bf/0x4bf ======================= Code: 5d c3 89 d8 e8 06 e0 f9 ff eb bd 0f 0b eb fe 55 57 56 53 89 d5 8b 40 f0 31 d2 e8 02 c1 fa ff 89 c2 85 c0 74 5c 8b 80 48 04 00 00 <8b> 58 0c 85 db 74 02 ff 03 ff 4a 08 0f 94 c0 84 c0 75 74 85 db EIP: [] mounts_open+0x1c/0xac SS:ESP 0068:d3c87ef0 A race with do_exit()'s call to exit_namespaces(). Signed-off-by: Alexey Dobriyan Acked-by: Serge Hallyn Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/base.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/fs/proc/base.c b/fs/proc/base.c index 77a57b5799c4..ff7a66850602 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -371,9 +371,11 @@ static int mounts_open(struct inode *inode, struct file *file) if (task) { task_lock(task); - ns = task->nsproxy->mnt_ns; - if (ns) - get_mnt_ns(ns); + if (task->nsproxy) { + ns = task->nsproxy->mnt_ns; + if (ns) + get_mnt_ns(ns); + } task_unlock(task); put_task_struct(task); } From 1e9a51dca19dc1d8807c63cb3bd4413d3f95aaf5 Mon Sep 17 00:00:00 2001 From: Atsushi Nemoto Date: Fri, 26 Jan 2007 00:56:54 -0800 Subject: [PATCH 153/326] [PATCH] SPI: alternative fix for spi_busnum_to_master If a SPI master device exists, udev (udevtrigger) causes kernel crash, due to wrong kobj pointer in kobject_uevent_env(). This problem was not in 2.6.19. The backtrace (on MIPS) was: [<8024db6c>] kobject_uevent_env+0x54c/0x5e8 [<802a8264>] store_uevent+0x1c/0x3c (in drivers/class.c) [<801cb14c>] subsys_attr_store+0x2c/0x50 [<801cb80c>] flush_write_buffer+0x38/0x5c [<801cb900>] sysfs_write_file+0xd0/0x190 [<80181444>] vfs_write+0xc4/0x1a0 [<80181cdc>] sys_write+0x54/0xa0 [<8010dae4>] stack_done+0x20/0x3c flush_write_buffer() passes kobject of spi_master_class.subsys to subsys_addr_store(), then subsys_addr_store() passes a pointer to a struct subsystem to store_uevent() which expects a pointer to a struct class_device. The problem seems subsys_attr_store() called instead of class_device_attr_store(). This mismatch was caused by commit 3bd0f6943520e459659d10f3282285e43d3990f1, which overrides kset of master class. This made spi_master_class.subsys.kset.ktype NULL so subsys_sysfs_ops is used instead of class_dev_sysfs_ops. The commit was to fix spi_busnum_to_master(). Here is a patch fixes this function in other way, just searching children list of class_device. Signed-off-by: Atsushi Nemoto Signed-off-by: David Brownell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/spi/spi.c | 21 +++++++++++++-------- 1 file changed, 13 insertions(+), 8 deletions(-) diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c index 270e6211c2e3..6307428d2c94 100644 --- a/drivers/spi/spi.c +++ b/drivers/spi/spi.c @@ -366,7 +366,6 @@ spi_alloc_master(struct device *dev, unsigned size) class_device_initialize(&master->cdev); master->cdev.class = &spi_master_class; - kobj_set_kset_s(&master->cdev, spi_master_class.subsys); master->cdev.dev = get_device(dev); spi_master_set_devdata(master, &master[1]); @@ -466,14 +465,20 @@ EXPORT_SYMBOL_GPL(spi_unregister_master); */ struct spi_master *spi_busnum_to_master(u16 bus_num) { - char name[9]; - struct kobject *bus; + struct class_device *cdev; + struct spi_master *master = NULL; + struct spi_master *m; - snprintf(name, sizeof name, "spi%u", bus_num); - bus = kset_find_obj(&spi_master_class.subsys.kset, name); - if (bus) - return container_of(bus, struct spi_master, cdev.kobj); - return NULL; + down(&spi_master_class.sem); + list_for_each_entry(cdev, &spi_master_class.children, node) { + m = container_of(cdev, struct spi_master, cdev); + if (m->bus_num == bus_num) { + master = spi_master_get(m); + break; + } + } + up(&spi_master_class.sem); + return master; } EXPORT_SYMBOL_GPL(spi_busnum_to_master); From e4233dec749a3519069d9390561b5636a75c7579 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Fri, 26 Jan 2007 00:56:55 -0800 Subject: [PATCH 154/326] [PATCH] ACPI: fix cpufreq regression Recently cpufreq support on my laptop (Lenovo T60) broke completely: when it's plugged into AC it would never go higher than 1 GHz - neither 1.3 GHz nor 1.83 GHz is possible - no matter which governor (userspace, speed or ondemand) is used. After some cpufreq debugging i tracked the regression back to the following (totally correct) bug-fix commit: commit 0916bd3ebb7cefdd0f432e8491abe24f4b5a101e Author: Dave Jones Date: Wed Nov 22 20:42:01 2006 -0500 [PATCH] Correct bound checking from the value returned from _PPC method. This bugfix, which makes other laptops work, made a previously hidden (BIOS) bug visible on my laptop. The bug is the following: if the _PPC (Performance Present Capabilities) optional ACPI object is queried /after/ bootup then the BIOS reports an incorrect value of '2'. My laptop (Lenovo T60) has the following performance states supported: 0: 1833000 1: 1333000 2: 1000000 Per ACPI specification, a _PPC value of '0' means that all 3 performance states are usable. A _PPC value of '1' means states 1 .. 2 are usable, a value of '2' means only state '2' (slowest) is usable. now, the _PPC object is optional, and it also comes with notification. Furthermore, when a CPU object is initialized, the _PPC object is initialized as well. So the following evaluation of the _PPC object is superfluous: [] acpi_processor_get_platform_limit+0xa1/0xaf [] acpi_processor_register_performance+0x3b9/0x3ef [] acpi_cpufreq_cpu_init+0xb7/0x596 [] cpufreq_add_dev+0x160/0x4a8 [] sysdev_driver_register+0x5a/0xa0 [] cpufreq_register_driver+0xb4/0x176 [] acpi_cpufreq_init+0xe5/0xeb [] init+0x14f/0x3dd And this is the point where my laptop's BIOS returns the incorrect value of '2'. Note that it has not sent any notification event, so the value is probably not really intentional (possibly spurious), and Windows likely doesnt query it after bootup either. Maybe the value is kept at '2' normally, and is only set to the real value when a true asynchronous event (such as AC plug event, battery switch, etc.) occurs. So i /think/ this is a grey area of the ACPI spec: per the letter of the spec the _PPC value only changes when notified, so there's no reason to query it after the system has booted up. So in my opinion the best (and most compatible) strategy would be to do the change below, and to not evaluate the _PPC object in the acpi_processor_get_performance_info() call, but only evaluate it if _PPC is present during CPU object init, or if it's notified during an asynchronous event. This change is more permissive than the previous logic, so it definitely shouldnt break any existing system. This also happens to fix my laptop, which is merrily chugging along at 1.83 GHz now. Yay! Signed-off-by: Ingo Molnar Cc: Dave Jones Acked-by: Len Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/acpi/processor_perflib.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/drivers/acpi/processor_perflib.c b/drivers/acpi/processor_perflib.c index 5207f9e4b443..cbb6f0814ce2 100644 --- a/drivers/acpi/processor_perflib.c +++ b/drivers/acpi/processor_perflib.c @@ -322,10 +322,6 @@ static int acpi_processor_get_performance_info(struct acpi_processor *pr) if (result) return result; - result = acpi_processor_get_platform_limit(pr); - if (result) - return result; - return 0; } From e702ff0ba6f7b52021f26e0e14237eb6ca8a1b6f Mon Sep 17 00:00:00 2001 From: Tilman Schmidt Date: Fri, 26 Jan 2007 00:56:56 -0800 Subject: [PATCH 155/326] [PATCH] Gigaset ISDN driver error handling fixes Fix several flaws in the error handling of the Siemens Gigaset ISDN driver, including one that would cause an Oops when connecting more than one device of the same type. Signed-off-by: Tilman Schmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/isdn/gigaset/common.c | 61 +++++++++++++++++++---------------- 1 file changed, 33 insertions(+), 28 deletions(-) diff --git a/drivers/isdn/gigaset/common.c b/drivers/isdn/gigaset/common.c index 95eff3b2917a..4f75cce6fdff 100644 --- a/drivers/isdn/gigaset/common.c +++ b/drivers/isdn/gigaset/common.c @@ -356,16 +356,17 @@ static struct cardstate *alloc_cs(struct gigaset_driver *drv) { unsigned long flags; unsigned i; - static struct cardstate *ret = NULL; + struct cardstate *ret = NULL; spin_lock_irqsave(&drv->lock, flags); for (i = 0; i < drv->minors; ++i) { if (!(drv->flags[i] & VALID_MINOR)) { - drv->flags[i] = VALID_MINOR; - ret = drv->cs + i; - } - if (ret) + if (try_module_get(drv->owner)) { + drv->flags[i] = VALID_MINOR; + ret = drv->cs + i; + } break; + } } spin_unlock_irqrestore(&drv->lock, flags); return ret; @@ -376,6 +377,8 @@ static void free_cs(struct cardstate *cs) unsigned long flags; struct gigaset_driver *drv = cs->driver; spin_lock_irqsave(&drv->lock, flags); + if (drv->flags[cs->minor_index] & VALID_MINOR) + module_put(drv->owner); drv->flags[cs->minor_index] = 0; spin_unlock_irqrestore(&drv->lock, flags); } @@ -579,7 +582,7 @@ static struct bc_state *gigaset_initbcs(struct bc_state *bcs, } else if ((bcs->skb = dev_alloc_skb(SBUFSIZE + HW_HDR_LEN)) != NULL) skb_reserve(bcs->skb, HW_HDR_LEN); else { - warn("could not allocate skb\n"); + warn("could not allocate skb"); bcs->inputstate |= INS_skip_frame; } @@ -632,17 +635,25 @@ struct cardstate *gigaset_initcs(struct gigaset_driver *drv, int channels, int i; gig_dbg(DEBUG_INIT, "allocating cs"); - cs = alloc_cs(drv); - if (!cs) - goto error; + if (!(cs = alloc_cs(drv))) { + err("maximum number of devices exceeded"); + return NULL; + } + mutex_init(&cs->mutex); + mutex_lock(&cs->mutex); + gig_dbg(DEBUG_INIT, "allocating bcs[0..%d]", channels - 1); cs->bcs = kmalloc(channels * sizeof(struct bc_state), GFP_KERNEL); - if (!cs->bcs) + if (!cs->bcs) { + err("out of memory"); goto error; + } gig_dbg(DEBUG_INIT, "allocating inbuf"); cs->inbuf = kmalloc(sizeof(struct inbuf_t), GFP_KERNEL); - if (!cs->inbuf) + if (!cs->inbuf) { + err("out of memory"); goto error; + } cs->cs_init = 0; cs->channels = channels; @@ -654,8 +665,6 @@ struct cardstate *gigaset_initcs(struct gigaset_driver *drv, int channels, spin_lock_init(&cs->ev_lock); cs->ev_tail = 0; cs->ev_head = 0; - mutex_init(&cs->mutex); - mutex_lock(&cs->mutex); tasklet_init(&cs->event_tasklet, &gigaset_handle_event, (unsigned long) cs); @@ -684,8 +693,10 @@ struct cardstate *gigaset_initcs(struct gigaset_driver *drv, int channels, for (i = 0; i < channels; ++i) { gig_dbg(DEBUG_INIT, "setting up bcs[%d].read", i); - if (!gigaset_initbcs(cs->bcs + i, cs, i)) + if (!gigaset_initbcs(cs->bcs + i, cs, i)) { + err("could not allocate channel %d data", i); goto error; + } } ++cs->cs_init; @@ -720,8 +731,10 @@ struct cardstate *gigaset_initcs(struct gigaset_driver *drv, int channels, make_valid(cs, VALID_ID); ++cs->cs_init; gig_dbg(DEBUG_INIT, "setting up hw"); - if (!cs->ops->initcshw(cs)) + if (!cs->ops->initcshw(cs)) { + err("could not allocate device specific data"); goto error; + } ++cs->cs_init; @@ -743,8 +756,8 @@ struct cardstate *gigaset_initcs(struct gigaset_driver *drv, int channels, mutex_unlock(&cs->mutex); return cs; -error: if (cs) - mutex_unlock(&cs->mutex); +error: + mutex_unlock(&cs->mutex); gig_dbg(DEBUG_INIT, "failed"); gigaset_freecs(cs); return NULL; @@ -1040,7 +1053,6 @@ void gigaset_freedriver(struct gigaset_driver *drv) spin_unlock_irqrestore(&driver_lock, flags); gigaset_if_freedriver(drv); - module_put(drv->owner); kfree(drv->cs); kfree(drv->flags); @@ -1072,10 +1084,6 @@ struct gigaset_driver *gigaset_initdriver(unsigned minor, unsigned minors, if (!drv) return NULL; - if (!try_module_get(owner)) - goto out1; - - drv->cs = NULL; drv->have_tty = 0; drv->minor = minor; drv->minors = minors; @@ -1087,11 +1095,11 @@ struct gigaset_driver *gigaset_initdriver(unsigned minor, unsigned minors, drv->cs = kmalloc(minors * sizeof *drv->cs, GFP_KERNEL); if (!drv->cs) - goto out2; + goto error; drv->flags = kmalloc(minors * sizeof *drv->flags, GFP_KERNEL); if (!drv->flags) - goto out3; + goto error; for (i = 0; i < minors; ++i) { drv->flags[i] = 0; @@ -1108,11 +1116,8 @@ struct gigaset_driver *gigaset_initdriver(unsigned minor, unsigned minors, return drv; -out3: +error: kfree(drv->cs); -out2: - module_put(owner); -out1: kfree(drv); return NULL; } From 98fac23f332da2dea96f6bec4890eb35fdd50606 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Fri, 26 Jan 2007 00:56:57 -0800 Subject: [PATCH 156/326] [PATCH] knfsd: update email address and status for NFSD in MAINTAINERS Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- MAINTAINERS | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index d6f04a81f761..4bc35f7f4d3c 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1928,11 +1928,10 @@ S: Maintained KERNEL NFSD P: Neil Brown -M: neilb@cse.unsw.edu.au +M: neilb@suse.de L: nfs@lists.sourceforge.net W: http://nfs.sourceforge.net/ -W: http://www.cse.unsw.edu.au/~neilb/patches/linux-devel/ -S: Maintained +S: Supported KERNEL VIRTUAL MACHINE (KVM) P: Avi Kivity From 1a8eff6d977c28162c61c9532ca58634e7090b69 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Fri, 26 Jan 2007 00:56:58 -0800 Subject: [PATCH 157/326] [PATCH] knfsd: fix setting of ACL server versions Due to silly typos, if the nfs versions are explicitly set, no NFSACL versions get enabled. Also improve an error message that would have made this bug a little easier to find. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfssvc.c | 8 ++++---- net/sunrpc/svc.c | 3 ++- 2 files changed, 6 insertions(+), 5 deletions(-) diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c index 0aaccb03bf76..fbf5d51947ea 100644 --- a/fs/nfsd/nfssvc.c +++ b/fs/nfsd/nfssvc.c @@ -72,7 +72,7 @@ static struct svc_program nfsd_acl_program = { .pg_prog = NFS_ACL_PROGRAM, .pg_nvers = NFSD_ACL_NRVERS, .pg_vers = nfsd_acl_versions, - .pg_name = "nfsd", + .pg_name = "nfsacl", .pg_class = "nfsd", .pg_stats = &nfsd_acl_svcstats, .pg_authenticate = &svc_set_client, @@ -118,16 +118,16 @@ int nfsd_vers(int vers, enum vers_op change) switch(change) { case NFSD_SET: nfsd_versions[vers] = nfsd_version[vers]; - break; #if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) if (vers < NFSD_ACL_NRVERS) - nfsd_acl_version[vers] = nfsd_acl_version[vers]; + nfsd_acl_versions[vers] = nfsd_acl_version[vers]; #endif + break; case NFSD_CLEAR: nfsd_versions[vers] = NULL; #if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) if (vers < NFSD_ACL_NRVERS) - nfsd_acl_version[vers] = NULL; + nfsd_acl_versions[vers] = NULL; #endif break; case NFSD_TEST: diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index f3001f3626f6..bf21a2047010 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -910,7 +910,8 @@ svc_process(struct svc_rqst *rqstp) err_bad_vers: #ifdef RPC_PARANOIA - printk("svc: unknown version (%d)\n", vers); + printk("svc: unknown version (%d for prog %d, %s)\n", + vers, prog, progp->pg_name); #endif serv->sv_stats->rpcbadfmt++; svc_putnl(resv, RPC_PROG_MISMATCH); From 250f3915183d377d36e012bac9caa7345ce465b8 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Fri, 26 Jan 2007 00:56:59 -0800 Subject: [PATCH 158/326] [PATCH] knfsd: fix an NFSD bug with full sized, non-page-aligned reads NFSd assumes that largest number of pages that will be needed for a request+response is 2+N where N pages is the size of the largest permitted read/write request. The '2' are 1 for the non-data part of the request, and 1 for the non-data part of the reply. However, when a read request is not page-aligned, and we choose to use ->sendfile to send it directly from the page cache, we may need N+1 pages to hold the whole reply. This can overflow and array and cause an Oops. This patch increases size of the array for holding pages by one and makes sure that entry is NULL when it is not in use. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/vfs.c | 3 ++- include/linux/sunrpc/svc.h | 5 ++++- net/sunrpc/svcsock.c | 2 ++ 3 files changed, 8 insertions(+), 2 deletions(-) diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 7a79c23aa6d4..ea855629a1c2 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -822,7 +822,8 @@ nfsd_read_actor(read_descriptor_t *desc, struct page *page, unsigned long offset rqstp->rq_res.page_len = size; } else if (page != pp[-1]) { get_page(page); - put_page(*pp); + if (*pp) + put_page(*pp); *pp = page; rqstp->rq_resused++; rqstp->rq_res.page_len += size; diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h index 965d6c20086e..64f3d60c72af 100644 --- a/include/linux/sunrpc/svc.h +++ b/include/linux/sunrpc/svc.h @@ -144,8 +144,11 @@ extern u32 svc_max_payload(const struct svc_rqst *rqstp); * * Each request/reply pair can have at most one "payload", plus two pages, * one for the request, and one for the reply. + * We using ->sendfile to return read data, we might need one extra page + * if the request is not page-aligned. So add another '1'. */ -#define RPCSVC_MAXPAGES ((RPCSVC_MAXPAYLOAD+PAGE_SIZE-1)/PAGE_SIZE + 2) +#define RPCSVC_MAXPAGES ((RPCSVC_MAXPAYLOAD+PAGE_SIZE-1)/PAGE_SIZE \ + + 2 + 1) static inline u32 svc_getnl(struct kvec *iov) { diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index 99f54fb6d669..45120f268d03 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -1278,6 +1278,8 @@ svc_recv(struct svc_rqst *rqstp, long timeout) schedule_timeout_uninterruptible(msecs_to_jiffies(500)); rqstp->rq_pages[i] = p; } + rqstp->rq_pages[i++] = NULL; /* this might be seen in nfs_read_actor */ + BUG_ON(pages >= RPCSVC_MAXPAGES); /* Make arg->head point to first page and arg->pages point to rest */ arg = &rqstp->rq_arg; From e2df0c8644db3d01e3c6fc5f7670fdd7b4c2b234 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Fri, 26 Jan 2007 00:56:59 -0800 Subject: [PATCH 159/326] [PATCH] knfsd: replace some warning ins nfsfh.h with BUG_ON or WARN_ON A couple of the warnings will be followed by an Oops if they ever fire, so may as well be BUG_ON. Another isn't obviously fatal but has never been known to fire, so make it a WARN_ON. Cc: Adrian Bunk Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/nfsd/nfsfh.h | 15 ++++----------- 1 file changed, 4 insertions(+), 11 deletions(-) diff --git a/include/linux/nfsd/nfsfh.h b/include/linux/nfsd/nfsfh.h index f3b51d62ec7d..d9c6c382165d 100644 --- a/include/linux/nfsd/nfsfh.h +++ b/include/linux/nfsd/nfsfh.h @@ -217,11 +217,7 @@ void fh_put(struct svc_fh *); static __inline__ struct svc_fh * fh_copy(struct svc_fh *dst, struct svc_fh *src) { - if (src->fh_dentry || src->fh_locked) { - struct dentry *dentry = src->fh_dentry; - printk(KERN_ERR "fh_copy: copying %s/%s, already verified!\n", - dentry->d_parent->d_name.name, dentry->d_name.name); - } + WARN_ON(src->fh_dentry || src->fh_locked); *dst = *src; return dst; @@ -300,10 +296,8 @@ fh_lock_nested(struct svc_fh *fhp, unsigned int subclass) dfprintk(FILEOP, "nfsd: fh_lock(%s) locked = %d\n", SVCFH_fmt(fhp), fhp->fh_locked); - if (!fhp->fh_dentry) { - printk(KERN_ERR "fh_lock: fh not verified!\n"); - return; - } + BUG_ON(!dentry); + if (fhp->fh_locked) { printk(KERN_WARNING "fh_lock: %s/%s already locked!\n", dentry->d_parent->d_name.name, dentry->d_name.name); @@ -328,8 +322,7 @@ fh_lock(struct svc_fh *fhp) static inline void fh_unlock(struct svc_fh *fhp) { - if (!fhp->fh_dentry) - printk(KERN_ERR "fh_unlock: fh not verified!\n"); + BUG_ON(!fhp->fh_dentry); if (fhp->fh_locked) { fill_post_wcc(fhp); From c397852c3ddad582ead8c57fbc48bdeccd995a30 Mon Sep 17 00:00:00 2001 From: Peter Staubach Date: Fri, 26 Jan 2007 00:57:00 -0800 Subject: [PATCH 160/326] [PATCH] knfsd: Don't mess with the 'mode' when storing a exclusive-create cookie NFS V3 (and V4) support exclusive create by passing a 'cookie' which can get stored with the file. If the file exists but has exactly the right cookie stored, then we assume this is a retransmit and the exclusive create was successful. The cookie is 64bits and is traditionally stored in the mtime and atime fields. This causes a problem with Solaris7 as negative mtime or atime confuse it. So we moved two bits into the mode word instead. But inherited ACLs sometimes overwrite the mode word on create, so this is a problem. So we give up and just store 62 of the 64 bits and assume that is close enough. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/vfs.c | 21 +++++---------------- 1 file changed, 5 insertions(+), 16 deletions(-) diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index ea855629a1c2..4dd49d6e90cd 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1245,7 +1245,6 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, __be32 err; int host_err; __u32 v_mtime=0, v_atime=0; - int v_mode=0; err = nfserr_perm; if (!flen) @@ -1282,16 +1281,11 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, goto out; if (createmode == NFS3_CREATE_EXCLUSIVE) { - /* while the verifier would fit in mtime+atime, - * solaris7 gets confused (bugid 4218508) if these have - * the high bit set, so we use the mode as well + /* solaris7 gets confused (bugid 4218508) if these have + * the high bit set, so just clear the high bits. */ v_mtime = verifier[0]&0x7fffffff; v_atime = verifier[1]&0x7fffffff; - v_mode = S_IFREG - | ((verifier[0]&0x80000000) >> (32-7)) /* u+x */ - | ((verifier[1]&0x80000000) >> (32-9)) /* u+r */ - ; } if (dchild->d_inode) { @@ -1319,7 +1313,6 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, case NFS3_CREATE_EXCLUSIVE: if ( dchild->d_inode->i_mtime.tv_sec == v_mtime && dchild->d_inode->i_atime.tv_sec == v_atime - && dchild->d_inode->i_mode == v_mode && dchild->d_inode->i_size == 0 ) break; /* fallthru */ @@ -1341,26 +1334,22 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, } if (createmode == NFS3_CREATE_EXCLUSIVE) { - /* Cram the verifier into atime/mtime/mode */ + /* Cram the verifier into atime/mtime */ iap->ia_valid = ATTR_MTIME|ATTR_ATIME - | ATTR_MTIME_SET|ATTR_ATIME_SET - | ATTR_MODE; + | ATTR_MTIME_SET|ATTR_ATIME_SET; /* XXX someone who knows this better please fix it for nsec */ iap->ia_mtime.tv_sec = v_mtime; iap->ia_atime.tv_sec = v_atime; iap->ia_mtime.tv_nsec = 0; iap->ia_atime.tv_nsec = 0; - iap->ia_mode = v_mode; } /* Set file attributes. - * Mode has already been set but we might need to reset it - * for CREATE_EXCLUSIVE * Irix appears to send along the gid when it tries to * implement setgid directories via NFS. Clear out all that cruft. */ set_attr: - if ((iap->ia_valid &= ~(ATTR_UID|ATTR_GID)) != 0) { + if ((iap->ia_valid &= ~(ATTR_UID|ATTR_GID|ATTR_MODE)) != 0) { __be32 err2 = nfsd_setattr(rqstp, resfhp, iap, 0, (time_t)0); if (err2) err = err2; From 524418bb8ecd3dfd2975bc0aa3c2cc7e8e081f24 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Fri, 26 Jan 2007 00:57:01 -0800 Subject: [PATCH 161/326] [PATCH] md: update email address and status for MD in MAINTAINERS Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- MAINTAINERS | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 4bc35f7f4d3c..796e891083b7 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2992,9 +2992,9 @@ SOFTWARE RAID (Multiple Disks) SUPPORT P: Ingo Molnar M: mingo@redhat.com P: Neil Brown -M: neilb@cse.unsw.edu.au +M: neilb@suse.de L: linux-raid@vger.kernel.org -S: Maintained +S: Supported SOFTWARE SUSPEND: P: Pavel Machek From 3eda22d19b76b15ef3420b251bd47a0ba0127589 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Fri, 26 Jan 2007 00:57:01 -0800 Subject: [PATCH 162/326] [PATCH] md: make 'repair' actually work for raid1 When 'repair' finds a block that is different one the various parts of the mirror. it is meant to write a chosen good version to the others. However it currently writes out the original data to each. The memcpy to make all the data the same is missing. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/raid1.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index 164b25dca101..ab74d40cac98 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -1266,6 +1266,11 @@ static void sync_request_write(mddev_t *mddev, r1bio_t *r1_bio) sbio->bi_sector = r1_bio->sector + conf->mirrors[i].rdev->data_offset; sbio->bi_bdev = conf->mirrors[i].rdev->bdev; + for (j = 0; j < vcnt ; j++) + memcpy(page_address(sbio->bi_io_vec[j].bv_page), + page_address(pbio->bi_io_vec[j].bv_page), + PAGE_SIZE); + } } } From 1031be7a5fafd3a858dfaabb74d98f9ca20744a8 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Fri, 26 Jan 2007 00:57:02 -0800 Subject: [PATCH 163/326] [PATCH] md: make sure the events count in an md array never returns to zero Now that we sometimes step the array events count backwards (when transitioning dirty->clean where nothing else interesting has happened - so that we don't need to write to spares all the time), it is possible for the event count to return to zero, which is potentially confusing and triggers and MD_BUG. We could possibly remove the MD_BUG, but is just as easy, and probably safer, to make sure we never return to zero. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/md.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/md/md.c b/drivers/md/md.c index d1cb45f6d6a9..ec3d8e8a0bd3 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -1633,7 +1633,8 @@ static void md_update_sb(mddev_t * mddev, int force_change) * and 'events' is odd, we can roll back to the previous clean state */ if (nospares && (mddev->in_sync && mddev->recovery_cp == MaxSector) - && (mddev->events & 1)) + && (mddev->events & 1) + && mddev->events != 1) mddev->events--; else { /* otherwise we have to go forward and ... */ From f49d5e62d9352d33b30c9befbaf0fd9c88265ec1 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Fri, 26 Jan 2007 00:57:03 -0800 Subject: [PATCH 164/326] [PATCH] md: avoid reading past the end of a bitmap file In most cases we check the size of the bitmap file before reading data from it. However when reading the superblock, we always read the first PAGE_SIZE bytes, which might not always be appropriate. So limit that read to the size of the file if appropriate. Also, we get the count of available bytes wrong in one place, so that too can read past the end of the file. Cc: "yang yin" Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/bitmap.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c index 5432d07c074d..11108165e264 100644 --- a/drivers/md/bitmap.c +++ b/drivers/md/bitmap.c @@ -479,9 +479,12 @@ static int bitmap_read_sb(struct bitmap *bitmap) int err = -EINVAL; /* page 0 is the superblock, read it... */ - if (bitmap->file) - bitmap->sb_page = read_page(bitmap->file, 0, bitmap, PAGE_SIZE); - else { + if (bitmap->file) { + loff_t isize = i_size_read(bitmap->file->f_mapping->host); + int bytes = isize > PAGE_SIZE ? PAGE_SIZE : isize; + + bitmap->sb_page = read_page(bitmap->file, 0, bitmap, bytes); + } else { bitmap->sb_page = read_sb_page(bitmap->mddev, bitmap->offset, 0); } if (IS_ERR(bitmap->sb_page)) { @@ -877,7 +880,8 @@ static int bitmap_init_from_disk(struct bitmap *bitmap, sector_t start) int count; /* unmap the old page, we're done with it */ if (index == num_pages-1) - count = bytes - index * PAGE_SIZE; + count = bytes + sizeof(bitmap_super_t) + - index * PAGE_SIZE; else count = PAGE_SIZE; if (index == 0) { From f94b347059809aecd97af53a084fe5f3db045d92 Mon Sep 17 00:00:00 2001 From: Eric Van Hensbergen Date: Fri, 26 Jan 2007 00:57:04 -0800 Subject: [PATCH 165/326] [PATCH] 9p: fix bogus return code checks during initialization There is a simple logic error in init_v9fs - the return code checks are reversed. This patch fixes the return code and adds some messages to prevent module initialization from failing silently. Signed-off-by: Eric Van Hensbergen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/9p/mux.c | 4 +++- fs/9p/v9fs.c | 11 ++++++++--- 2 files changed, 11 insertions(+), 4 deletions(-) diff --git a/fs/9p/mux.c b/fs/9p/mux.c index 944273c3dbff..147ceef8e537 100644 --- a/fs/9p/mux.c +++ b/fs/9p/mux.c @@ -132,8 +132,10 @@ int v9fs_mux_global_init(void) v9fs_mux_poll_tasks[i].task = NULL; v9fs_mux_wq = create_workqueue("v9fs"); - if (!v9fs_mux_wq) + if (!v9fs_mux_wq) { + printk(KERN_WARNING "v9fs: mux: creating workqueue failed\n"); return -ENOMEM; + } return 0; } diff --git a/fs/9p/v9fs.c b/fs/9p/v9fs.c index 0b96fae8b479..d9b561ba5e58 100644 --- a/fs/9p/v9fs.c +++ b/fs/9p/v9fs.c @@ -457,14 +457,19 @@ static int __init init_v9fs(void) v9fs_error_init(); - printk(KERN_INFO "Installing v9fs 9P2000 file system support\n"); + printk(KERN_INFO "Installing v9fs 9p2000 file system support\n"); ret = v9fs_mux_global_init(); - if (!ret) + if (ret) { + printk(KERN_WARNING "v9fs: starting mux failed\n"); return ret; + } ret = register_filesystem(&v9fs_fs_type); - if (!ret) + if (ret) { + printk(KERN_WARNING "v9fs: registering file system failed\n"); v9fs_mux_global_exit(); + } + return ret; } From 621997cd392a541bbe964e7521e2e4b2a7a03c4e Mon Sep 17 00:00:00 2001 From: Eric Van Hensbergen Date: Fri, 26 Jan 2007 00:57:04 -0800 Subject: [PATCH 166/326] [PATCH] 9p: fix rename return code 9p doesn't handle renames between directories -- however, we were returning EPERM instead of EXDEV when we detected this case. Signed-off-by: Eric Van Hensbergren Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/9p/vfs_inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index 18f26cdfd882..05d30e89ba45 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -767,7 +767,7 @@ v9fs_vfs_rename(struct inode *old_dir, struct dentry *old_dentry, /* 9P can only handle file rename in the same directory */ if (memcmp(&olddirfid->qid, &newdirfid->qid, sizeof(newdirfid->qid))) { dprintk(DEBUG_ERROR, "old dir and new dir are different\n"); - retval = -EPERM; + retval = -EXDEV; goto FreeFcallnBail; } From ff76e1dfc8728278ee231feeb93146f9c57c3ec3 Mon Sep 17 00:00:00 2001 From: Eric Van Hensbergen Date: Fri, 26 Jan 2007 00:57:05 -0800 Subject: [PATCH 167/326] [PATCH] 9p: update documentation regarding server applications Update the documentation to cover using Inferno as a server for 9p and to include information about spfs (a stable single-threaded stand-alone 9p server). Signed-off-by: Eric Van Hensbergen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/filesystems/9p.txt | 20 +++++++++++++++++--- 1 file changed, 17 insertions(+), 3 deletions(-) diff --git a/Documentation/filesystems/9p.txt b/Documentation/filesystems/9p.txt index 43b89c214d20..4d075a4558f9 100644 --- a/Documentation/filesystems/9p.txt +++ b/Documentation/filesystems/9p.txt @@ -73,8 +73,22 @@ OPTIONS RESOURCES ========= -The Linux version of the 9p server is now maintained under the npfs project -on sourceforge (http://sourceforge.net/projects/npfs). +Our current recommendation is to use Inferno (http://www.vitanuova.com/inferno) +as the 9p server. You can start a 9p server under Inferno by issuing the +following command: + ; styxlisten -A tcp!*!564 export '#U*' + +The -A specifies an unauthenticated export. The 564 is the port # (you may +have to choose a higher port number if running as a normal user). The '#U*' +specifies exporting the root of the Linux name space. You may specify a +subset of the namespace by extending the path: '#U*'/tmp would just export +/tmp. For more information, see the Inferno manual pages covering styxlisten +and export. + +A Linux version of the 9p server is now maintained under the npfs project +on sourceforge (http://sourceforge.net/projects/npfs). There is also a +more stable single-threaded version of the server (named spfs) available from +the same CVS repository. There are user and developer mailing lists available through the v9fs project on sourceforge (http://sourceforge.net/projects/v9fs). @@ -96,5 +110,5 @@ STATUS The 2.6 kernel support is working on PPC and x86. -PLEASE USE THE SOURCEFORGE BUG-TRACKER TO REPORT PROBLEMS. +PLEASE USE THE KERNEL BUGZILLA TO REPORT PROBLEMS. (http://bugzilla.kernel.org) From da977b2c7eb4d6312f063a7b486f2aad99809710 Mon Sep 17 00:00:00 2001 From: Eric Van Hensbergen Date: Fri, 26 Jan 2007 00:57:06 -0800 Subject: [PATCH 168/326] [PATCH] 9p: fix segfault caused by race condition in meta-data operations Running dbench multithreaded exposed a race condition where fid structures were removed while in use. This patch adds semaphores to meta-data operations to protect the fid structure. Some cleanup of error-case handling in the inode operations is also included. Signed-off-by: Eric Van Hensbergen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/9p/fid.c | 69 +++++++++++++++- fs/9p/fid.h | 5 ++ fs/9p/vfs_file.c | 47 ++--------- fs/9p/vfs_inode.c | 204 +++++++++++++++++++++++++++------------------- 4 files changed, 196 insertions(+), 129 deletions(-) diff --git a/fs/9p/fid.c b/fs/9p/fid.c index 27507201f9e7..a9b6301a04fc 100644 --- a/fs/9p/fid.c +++ b/fs/9p/fid.c @@ -25,6 +25,7 @@ #include #include #include +#include #include "debug.h" #include "v9fs.h" @@ -84,6 +85,7 @@ struct v9fs_fid *v9fs_fid_create(struct v9fs_session_info *v9ses, int fid) new->iounit = 0; new->rdir_pos = 0; new->rdir_fcall = NULL; + init_MUTEX(&new->lock); INIT_LIST_HEAD(&new->list); return new; @@ -102,11 +104,11 @@ void v9fs_fid_destroy(struct v9fs_fid *fid) } /** - * v9fs_fid_lookup - retrieve the right fid from a particular dentry + * v9fs_fid_lookup - return a locked fid from a dentry * @dentry: dentry to look for fid in - * @type: intent of lookup (operation or traversal) * - * find a fid in the dentry + * find a fid in the dentry, obtain its semaphore and return a reference to it. + * code calling lookup is responsible for releasing lock * * TODO: only match fids that have the same uid as current user * @@ -124,7 +126,68 @@ struct v9fs_fid *v9fs_fid_lookup(struct dentry *dentry) if (!return_fid) { dprintk(DEBUG_ERROR, "Couldn't find a fid in dentry\n"); + return_fid = ERR_PTR(-EBADF); } + if(down_interruptible(&return_fid->lock)) + return ERR_PTR(-EINTR); + return return_fid; } + +/** + * v9fs_fid_clone - lookup the fid for a dentry, clone a private copy and release it + * @dentry: dentry to look for fid in + * + * find a fid in the dentry and then clone to a new private fid + * + * TODO: only match fids that have the same uid as current user + * + */ + +struct v9fs_fid *v9fs_fid_clone(struct dentry *dentry) +{ + struct v9fs_session_info *v9ses = v9fs_inode2v9ses(dentry->d_inode); + struct v9fs_fid *base_fid, *new_fid = ERR_PTR(-EBADF); + struct v9fs_fcall *fcall = NULL; + int fid, err; + + base_fid = v9fs_fid_lookup(dentry); + + if(IS_ERR(base_fid)) + return base_fid; + + if(base_fid) { /* clone fid */ + fid = v9fs_get_idpool(&v9ses->fidpool); + if (fid < 0) { + eprintk(KERN_WARNING, "newfid fails!\n"); + new_fid = ERR_PTR(-ENOSPC); + goto Release_Fid; + } + + err = v9fs_t_walk(v9ses, base_fid->fid, fid, NULL, &fcall); + if (err < 0) { + dprintk(DEBUG_ERROR, "clone walk didn't work\n"); + v9fs_put_idpool(fid, &v9ses->fidpool); + new_fid = ERR_PTR(err); + goto Free_Fcall; + } + new_fid = v9fs_fid_create(v9ses, fid); + if (new_fid == NULL) { + dprintk(DEBUG_ERROR, "out of memory\n"); + new_fid = ERR_PTR(-ENOMEM); + } +Free_Fcall: + kfree(fcall); + } + +Release_Fid: + up(&base_fid->lock); + return new_fid; +} + +void v9fs_fid_clunk(struct v9fs_session_info *v9ses, struct v9fs_fid *fid) +{ + v9fs_t_clunk(v9ses, fid->fid); + v9fs_fid_destroy(fid); +} diff --git a/fs/9p/fid.h b/fs/9p/fid.h index aa974d6875c3..48fc170c26c8 100644 --- a/fs/9p/fid.h +++ b/fs/9p/fid.h @@ -30,6 +30,8 @@ struct v9fs_fid { struct list_head list; /* list of fids associated with a dentry */ struct list_head active; /* XXX - debug */ + struct semaphore lock; + u32 fid; unsigned char fidopen; /* set when fid is opened */ unsigned char fidclunked; /* set when fid has already been clunked */ @@ -55,3 +57,6 @@ struct v9fs_fid *v9fs_fid_get_created(struct dentry *); void v9fs_fid_destroy(struct v9fs_fid *fid); struct v9fs_fid *v9fs_fid_create(struct v9fs_session_info *, int fid); int v9fs_fid_insert(struct v9fs_fid *fid, struct dentry *dentry); +struct v9fs_fid *v9fs_fid_clone(struct dentry *dentry); +void v9fs_fid_clunk(struct v9fs_session_info *v9ses, struct v9fs_fid *fid); + diff --git a/fs/9p/vfs_file.c b/fs/9p/vfs_file.c index e86a07151280..9f17b0cacdd0 100644 --- a/fs/9p/vfs_file.c +++ b/fs/9p/vfs_file.c @@ -55,53 +55,22 @@ int v9fs_file_open(struct inode *inode, struct file *file) struct v9fs_fid *vfid; struct v9fs_fcall *fcall = NULL; int omode; - int fid = V9FS_NOFID; int err; dprintk(DEBUG_VFS, "inode: %p file: %p \n", inode, file); - vfid = v9fs_fid_lookup(file->f_path.dentry); - if (!vfid) { - dprintk(DEBUG_ERROR, "Couldn't resolve fid from dentry\n"); - return -EBADF; - } + vfid = v9fs_fid_clone(file->f_path.dentry); + if (IS_ERR(vfid)) + return PTR_ERR(vfid); - fid = v9fs_get_idpool(&v9ses->fidpool); - if (fid < 0) { - eprintk(KERN_WARNING, "newfid fails!\n"); - return -ENOSPC; - } - - err = v9fs_t_walk(v9ses, vfid->fid, fid, NULL, &fcall); - if (err < 0) { - dprintk(DEBUG_ERROR, "rewalk didn't work\n"); - if (fcall && fcall->id == RWALK) - goto clunk_fid; - else { - v9fs_put_idpool(fid, &v9ses->fidpool); - goto free_fcall; - } - } - kfree(fcall); - - /* TODO: do special things for O_EXCL, O_NOFOLLOW, O_SYNC */ - /* translate open mode appropriately */ omode = v9fs_uflags2omode(file->f_flags); - err = v9fs_t_open(v9ses, fid, omode, &fcall); + err = v9fs_t_open(v9ses, vfid->fid, omode, &fcall); if (err < 0) { PRINT_FCALL_ERROR("open failed", fcall); - goto clunk_fid; - } - - vfid = kmalloc(sizeof(struct v9fs_fid), GFP_KERNEL); - if (vfid == NULL) { - dprintk(DEBUG_ERROR, "out of memory\n"); - err = -ENOMEM; - goto clunk_fid; + goto Clunk_Fid; } file->private_data = vfid; - vfid->fid = fid; vfid->fidopen = 1; vfid->fidclunked = 0; vfid->iounit = fcall->params.ropen.iounit; @@ -112,10 +81,8 @@ int v9fs_file_open(struct inode *inode, struct file *file) return 0; -clunk_fid: - v9fs_t_clunk(v9ses, fid); - -free_fcall: +Clunk_Fid: + v9fs_fid_clunk(v9ses, vfid); kfree(fcall); return err; diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index 05d30e89ba45..9109ba1d6969 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -416,12 +416,8 @@ static int v9fs_remove(struct inode *dir, struct dentry *file, int rmdir) sb = file_inode->i_sb; v9ses = v9fs_inode2v9ses(file_inode); v9fid = v9fs_fid_lookup(file); - - if (!v9fid) { - dprintk(DEBUG_ERROR, - "no v9fs_fid\n"); - return -EBADF; - } + if(IS_ERR(v9fid)) + return PTR_ERR(v9fid); fid = v9fid->fid; if (fid < 0) { @@ -433,11 +429,13 @@ static int v9fs_remove(struct inode *dir, struct dentry *file, int rmdir) result = v9fs_t_remove(v9ses, fid, &fcall); if (result < 0) { PRINT_FCALL_ERROR("remove fails", fcall); + goto Error; } v9fs_put_idpool(fid, &v9ses->fidpool); v9fs_fid_destroy(v9fid); +Error: kfree(fcall); return result; } @@ -473,9 +471,13 @@ v9fs_vfs_create(struct inode *dir, struct dentry *dentry, int mode, inode = NULL; vfid = NULL; v9ses = v9fs_inode2v9ses(dir); - dfid = v9fs_fid_lookup(dentry->d_parent); - perm = unixmode2p9mode(v9ses, mode); + dfid = v9fs_fid_clone(dentry->d_parent); + if(IS_ERR(dfid)) { + err = PTR_ERR(dfid); + goto error; + } + perm = unixmode2p9mode(v9ses, mode); if (nd && nd->flags & LOOKUP_OPEN) flags = nd->intent.open.flags - 1; else @@ -485,9 +487,10 @@ v9fs_vfs_create(struct inode *dir, struct dentry *dentry, int mode, perm, v9fs_uflags2omode(flags), NULL, &fid, &qid, &iounit); if (err) - goto error; + goto clunk_dfid; vfid = v9fs_clone_walk(v9ses, dfid->fid, dentry); + v9fs_fid_clunk(v9ses, dfid); if (IS_ERR(vfid)) { err = PTR_ERR(vfid); vfid = NULL; @@ -525,6 +528,9 @@ v9fs_vfs_create(struct inode *dir, struct dentry *dentry, int mode, return 0; +clunk_dfid: + v9fs_fid_clunk(v9ses, dfid); + error: if (vfid) v9fs_fid_destroy(vfid); @@ -551,7 +557,12 @@ static int v9fs_vfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) inode = NULL; vfid = NULL; v9ses = v9fs_inode2v9ses(dir); - dfid = v9fs_fid_lookup(dentry->d_parent); + dfid = v9fs_fid_clone(dentry->d_parent); + if(IS_ERR(dfid)) { + err = PTR_ERR(dfid); + goto error; + } + perm = unixmode2p9mode(v9ses, mode | S_IFDIR); err = v9fs_create(v9ses, dfid->fid, (char *) dentry->d_name.name, @@ -559,37 +570,36 @@ static int v9fs_vfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) if (err) { dprintk(DEBUG_ERROR, "create error %d\n", err); - goto error; - } - - err = v9fs_t_clunk(v9ses, fid); - if (err) { - dprintk(DEBUG_ERROR, "clunk error %d\n", err); - goto error; + goto clean_up_dfid; } vfid = v9fs_clone_walk(v9ses, dfid->fid, dentry); if (IS_ERR(vfid)) { err = PTR_ERR(vfid); vfid = NULL; - goto error; + goto clean_up_dfid; } + v9fs_fid_clunk(v9ses, dfid); inode = v9fs_inode_from_fid(v9ses, vfid->fid, dir->i_sb); if (IS_ERR(inode)) { err = PTR_ERR(inode); inode = NULL; - goto error; + goto clean_up_fids; } dentry->d_op = &v9fs_dentry_operations; d_instantiate(dentry, inode); return 0; -error: +clean_up_fids: if (vfid) v9fs_fid_destroy(vfid); +clean_up_dfid: + v9fs_fid_clunk(v9ses, dfid); + +error: return err; } @@ -622,28 +632,23 @@ static struct dentry *v9fs_vfs_lookup(struct inode *dir, struct dentry *dentry, dentry->d_op = &v9fs_dentry_operations; dirfid = v9fs_fid_lookup(dentry->d_parent); - if (!dirfid) { - dprintk(DEBUG_ERROR, "no dirfid\n"); - return ERR_PTR(-EINVAL); - } + if(IS_ERR(dirfid)) + return ERR_PTR(PTR_ERR(dirfid)); dirfidnum = dirfid->fid; - if (dirfidnum < 0) { - dprintk(DEBUG_ERROR, "no dirfid for inode %p, #%lu\n", - dir, dir->i_ino); - return ERR_PTR(-EBADF); - } - newfid = v9fs_get_idpool(&v9ses->fidpool); if (newfid < 0) { eprintk(KERN_WARNING, "newfid fails!\n"); - return ERR_PTR(-ENOSPC); + result = -ENOSPC; + goto Release_Dirfid; } result = v9fs_t_walk(v9ses, dirfidnum, newfid, (char *)dentry->d_name.name, &fcall); + up(&dirfid->lock); + if (result < 0) { if (fcall && fcall->id == RWALK) v9fs_t_clunk(v9ses, newfid); @@ -701,8 +706,12 @@ static struct dentry *v9fs_vfs_lookup(struct inode *dir, struct dentry *dentry, return NULL; - FreeFcall: +Release_Dirfid: + up(&dirfid->lock); + +FreeFcall: kfree(fcall); + return ERR_PTR(result); } @@ -746,10 +755,8 @@ v9fs_vfs_rename(struct inode *old_dir, struct dentry *old_dentry, struct inode *old_inode = old_dentry->d_inode; struct v9fs_session_info *v9ses = v9fs_inode2v9ses(old_inode); struct v9fs_fid *oldfid = v9fs_fid_lookup(old_dentry); - struct v9fs_fid *olddirfid = - v9fs_fid_lookup(old_dentry->d_parent); - struct v9fs_fid *newdirfid = - v9fs_fid_lookup(new_dentry->d_parent); + struct v9fs_fid *olddirfid; + struct v9fs_fid *newdirfid; struct v9fs_wstat wstat; struct v9fs_fcall *fcall = NULL; int fid = -1; @@ -759,16 +766,26 @@ v9fs_vfs_rename(struct inode *old_dir, struct dentry *old_dentry, dprintk(DEBUG_VFS, "\n"); - if ((!oldfid) || (!olddirfid) || (!newdirfid)) { - dprintk(DEBUG_ERROR, "problem with arguments\n"); - return -EBADF; + if(IS_ERR(oldfid)) + return PTR_ERR(oldfid); + + olddirfid = v9fs_fid_clone(old_dentry->d_parent); + if(IS_ERR(olddirfid)) { + retval = PTR_ERR(olddirfid); + goto Release_lock; + } + + newdirfid = v9fs_fid_clone(new_dentry->d_parent); + if(IS_ERR(newdirfid)) { + retval = PTR_ERR(newdirfid); + goto Clunk_olddir; } /* 9P can only handle file rename in the same directory */ if (memcmp(&olddirfid->qid, &newdirfid->qid, sizeof(newdirfid->qid))) { dprintk(DEBUG_ERROR, "old dir and new dir are different\n"); retval = -EXDEV; - goto FreeFcallnBail; + goto Clunk_newdir; } fid = oldfid->fid; @@ -779,7 +796,7 @@ v9fs_vfs_rename(struct inode *old_dir, struct dentry *old_dentry, dprintk(DEBUG_ERROR, "no fid for old file #%lu\n", old_inode->i_ino); retval = -EBADF; - goto FreeFcallnBail; + goto Clunk_newdir; } v9fs_blank_wstat(&wstat); @@ -788,11 +805,20 @@ v9fs_vfs_rename(struct inode *old_dir, struct dentry *old_dentry, retval = v9fs_t_wstat(v9ses, fid, &wstat, &fcall); - FreeFcallnBail: if (retval < 0) PRINT_FCALL_ERROR("wstat error", fcall); kfree(fcall); + +Clunk_newdir: + v9fs_fid_clunk(v9ses, newdirfid); + +Clunk_olddir: + v9fs_fid_clunk(v9ses, olddirfid); + +Release_lock: + up(&oldfid->lock); + return retval; } @@ -810,15 +836,12 @@ v9fs_vfs_getattr(struct vfsmount *mnt, struct dentry *dentry, { struct v9fs_fcall *fcall = NULL; struct v9fs_session_info *v9ses = v9fs_inode2v9ses(dentry->d_inode); - struct v9fs_fid *fid = v9fs_fid_lookup(dentry); + struct v9fs_fid *fid = v9fs_fid_clone(dentry); int err = -EPERM; dprintk(DEBUG_VFS, "dentry: %p\n", dentry); - if (!fid) { - dprintk(DEBUG_ERROR, - "couldn't find fid associated with dentry\n"); - return -EBADF; - } + if(IS_ERR(fid)) + return PTR_ERR(fid); err = v9fs_t_stat(v9ses, fid->fid, &fcall); @@ -831,6 +854,7 @@ v9fs_vfs_getattr(struct vfsmount *mnt, struct dentry *dentry, } kfree(fcall); + v9fs_fid_clunk(v9ses, fid); return err; } @@ -844,18 +868,14 @@ v9fs_vfs_getattr(struct vfsmount *mnt, struct dentry *dentry, static int v9fs_vfs_setattr(struct dentry *dentry, struct iattr *iattr) { struct v9fs_session_info *v9ses = v9fs_inode2v9ses(dentry->d_inode); - struct v9fs_fid *fid = v9fs_fid_lookup(dentry); + struct v9fs_fid *fid = v9fs_fid_clone(dentry); struct v9fs_fcall *fcall = NULL; struct v9fs_wstat wstat; int res = -EPERM; dprintk(DEBUG_VFS, "\n"); - - if (!fid) { - dprintk(DEBUG_ERROR, - "Couldn't find fid associated with dentry\n"); - return -EBADF; - } + if(IS_ERR(fid)) + return PTR_ERR(fid); v9fs_blank_wstat(&wstat); if (iattr->ia_valid & ATTR_MODE) @@ -887,6 +907,7 @@ static int v9fs_vfs_setattr(struct dentry *dentry, struct iattr *iattr) if (res >= 0) res = inode_setattr(dentry->d_inode, iattr); + v9fs_fid_clunk(v9ses, fid); return res; } @@ -987,18 +1008,15 @@ static int v9fs_readlink(struct dentry *dentry, char *buffer, int buflen) struct v9fs_fcall *fcall = NULL; struct v9fs_session_info *v9ses = v9fs_inode2v9ses(dentry->d_inode); - struct v9fs_fid *fid = v9fs_fid_lookup(dentry); + struct v9fs_fid *fid = v9fs_fid_clone(dentry); - if (!fid) { - dprintk(DEBUG_ERROR, "could not resolve fid from dentry\n"); - retval = -EBADF; - goto FreeFcall; - } + if(IS_ERR(fid)) + return PTR_ERR(fid); if (!v9ses->extended) { retval = -EBADF; dprintk(DEBUG_ERROR, "not extended\n"); - goto FreeFcall; + goto ClunkFid; } dprintk(DEBUG_VFS, " %s\n", dentry->d_name.name); @@ -1009,8 +1027,10 @@ static int v9fs_readlink(struct dentry *dentry, char *buffer, int buflen) goto FreeFcall; } - if (!fcall) - return -EIO; + if (!fcall) { + retval = -EIO; + goto ClunkFid; + } if (!(fcall->params.rstat.stat.mode & V9FS_DMSYMLINK)) { retval = -EINVAL; @@ -1028,9 +1048,12 @@ static int v9fs_readlink(struct dentry *dentry, char *buffer, int buflen) fcall->params.rstat.stat.extension.str, buffer); retval = buflen; - FreeFcall: +FreeFcall: kfree(fcall); +ClunkFid: + v9fs_fid_clunk(v9ses, fid); + return retval; } @@ -1123,52 +1146,58 @@ static int v9fs_vfs_mkspecial(struct inode *dir, struct dentry *dentry, int err; u32 fid, perm; struct v9fs_session_info *v9ses; - struct v9fs_fid *dfid, *vfid; - struct inode *inode; + struct v9fs_fid *dfid, *vfid = NULL; + struct inode *inode = NULL; - inode = NULL; - vfid = NULL; v9ses = v9fs_inode2v9ses(dir); - dfid = v9fs_fid_lookup(dentry->d_parent); - perm = unixmode2p9mode(v9ses, mode); - if (!v9ses->extended) { dprintk(DEBUG_ERROR, "not extended\n"); return -EPERM; } + dfid = v9fs_fid_clone(dentry->d_parent); + if(IS_ERR(dfid)) { + err = PTR_ERR(dfid); + goto error; + } + + perm = unixmode2p9mode(v9ses, mode); + err = v9fs_create(v9ses, dfid->fid, (char *) dentry->d_name.name, perm, V9FS_OREAD, (char *) extension, &fid, NULL, NULL); if (err) - goto error; + goto clunk_dfid; err = v9fs_t_clunk(v9ses, fid); if (err) - goto error; + goto clunk_dfid; vfid = v9fs_clone_walk(v9ses, dfid->fid, dentry); if (IS_ERR(vfid)) { err = PTR_ERR(vfid); vfid = NULL; - goto error; + goto clunk_dfid; } inode = v9fs_inode_from_fid(v9ses, vfid->fid, dir->i_sb); if (IS_ERR(inode)) { err = PTR_ERR(inode); inode = NULL; - goto error; + goto free_vfid; } dentry->d_op = &v9fs_dentry_operations; d_instantiate(dentry, inode); return 0; -error: - if (vfid) - v9fs_fid_destroy(vfid); +free_vfid: + v9fs_fid_destroy(vfid); +clunk_dfid: + v9fs_fid_clunk(v9ses, dfid); + +error: return err; } @@ -1209,26 +1238,29 @@ v9fs_vfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *dentry) { int retval; + struct v9fs_session_info *v9ses = v9fs_inode2v9ses(dir); struct v9fs_fid *oldfid; char *name; dprintk(DEBUG_VFS, " %lu,%s,%s\n", dir->i_ino, dentry->d_name.name, old_dentry->d_name.name); - oldfid = v9fs_fid_lookup(old_dentry); - if (!oldfid) { - dprintk(DEBUG_ERROR, "can't find oldfid\n"); - return -EPERM; - } + oldfid = v9fs_fid_clone(old_dentry); + if(IS_ERR(oldfid)) + return PTR_ERR(oldfid); name = __getname(); - if (unlikely(!name)) - return -ENOMEM; + if (unlikely(!name)) { + retval = -ENOMEM; + goto clunk_fid; + } sprintf(name, "%d\n", oldfid->fid); retval = v9fs_vfs_mkspecial(dir, dentry, V9FS_DMLINK, name); __putname(name); +clunk_fid: + v9fs_fid_clunk(v9ses, oldfid); return retval; } From e540eb45a5254873245fd377f2fe3afc47bd33c1 Mon Sep 17 00:00:00 2001 From: Eric Van Hensbergen Date: Fri, 26 Jan 2007 00:57:06 -0800 Subject: [PATCH 169/326] [PATCH] 9p: null terminate error strings for debug print We weren't properly NULL terminating protocol error strings for our debug printk resulting in garbage being included in the output when debug was enabled. Signed-off-by: Eric Van Hensbergen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/9p/error.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/9p/error.c b/fs/9p/error.c index ae91555c1558..0d7fa4e08812 100644 --- a/fs/9p/error.c +++ b/fs/9p/error.c @@ -83,6 +83,7 @@ int v9fs_errstr2errno(char *errstr, int len) if (errno == 0) { /* TODO: if error isn't found, add it dynamically */ + errstr[len] = 0; printk(KERN_ERR "%s: errstr :%s: not found\n", __FUNCTION__, errstr); errno = 1; From bfa152fa5e4d328fe3ebf15908ee8ec20a0ce6dc Mon Sep 17 00:00:00 2001 From: Jun'ichi Nomura Date: Fri, 26 Jan 2007 00:57:07 -0800 Subject: [PATCH 170/326] [PATCH] dm-multipath: fix stall on noflush suspend/resume Allow noflush suspend/resume of device-mapper device only for the case where the device size is unchanged. Otherwise, dm-multipath devices can stall when resumed if noflush was used when suspending them, all paths have failed and queue_if_no_path is set. Explanation: 1. Something is doing fsync() on the block dev, holding inode->i_sem 2. The fsync write is blocked by all-paths-down and queue_if_no_path 3. Someone requests to suspend the dm device with noflush. Pending writes are left in queue. 4. In the middle of dm_resume(), __bind() tries to get inode->i_sem to do __set_size() and waits forever. 'noflush suspend' is a new device-mapper feature introduced in early 2.6.20. So I hope the fix being included before 2.6.20 is released. Example of reproducer: 1. Create a multipath device by dmsetup 2. Fail all paths during mkfs 3. Do dmsetup suspend --noflush and load new map with healthy paths 4. Do dmsetup resume Signed-off-by: Jun'ichi Nomura Acked-by: Alasdair G Kergon Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/dm.c | 27 +++++++++++++++++++-------- 1 file changed, 19 insertions(+), 8 deletions(-) diff --git a/drivers/md/dm.c b/drivers/md/dm.c index fe7c56e10435..3668b170ea68 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -1116,7 +1116,8 @@ static int __bind(struct mapped_device *md, struct dm_table *t) if (size != get_capacity(md->disk)) memset(&md->geometry, 0, sizeof(md->geometry)); - __set_size(md, size); + if (md->suspended_bdev) + __set_size(md, size); if (size == 0) return 0; @@ -1264,6 +1265,11 @@ int dm_swap_table(struct mapped_device *md, struct dm_table *table) if (!dm_suspended(md)) goto out; + /* without bdev, the device size cannot be changed */ + if (!md->suspended_bdev) + if (get_capacity(md->disk) != dm_table_get_size(table)) + goto out; + __unbind(md); r = __bind(md, table); @@ -1341,11 +1347,14 @@ int dm_suspend(struct mapped_device *md, unsigned suspend_flags) /* This does not get reverted if there's an error later. */ dm_table_presuspend_targets(map); - md->suspended_bdev = bdget_disk(md->disk, 0); - if (!md->suspended_bdev) { - DMWARN("bdget failed in dm_suspend"); - r = -ENOMEM; - goto flush_and_out; + /* bdget() can stall if the pending I/Os are not flushed */ + if (!noflush) { + md->suspended_bdev = bdget_disk(md->disk, 0); + if (!md->suspended_bdev) { + DMWARN("bdget failed in dm_suspend"); + r = -ENOMEM; + goto flush_and_out; + } } /* @@ -1473,8 +1482,10 @@ int dm_resume(struct mapped_device *md) unlock_fs(md); - bdput(md->suspended_bdev); - md->suspended_bdev = NULL; + if (md->suspended_bdev) { + bdput(md->suspended_bdev); + md->suspended_bdev = NULL; + } clear_bit(DMF_SUSPENDED, &md->flags); From a8d814b5dd7a1bc5c19ae32d35b8bd4d8a510eae Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Fri, 26 Jan 2007 00:57:08 -0800 Subject: [PATCH 171/326] [PATCH] remove __devinit markings from rtc_sysfs_add_device() rtc_sysfs_add_device is needed even after dev initialization, so drop __devinit. Signed-off-by: Mike Frysinger Acked-by: Alessandro Zummo Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/rtc/rtc-sysfs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/rtc/rtc-sysfs.c b/drivers/rtc/rtc-sysfs.c index 9418a59fb368..2ddd0cf07140 100644 --- a/drivers/rtc/rtc-sysfs.c +++ b/drivers/rtc/rtc-sysfs.c @@ -78,7 +78,7 @@ static struct attribute_group rtc_attr_group = { .attrs = rtc_attrs, }; -static int __devinit rtc_sysfs_add_device(struct class_device *class_dev, +static int rtc_sysfs_add_device(struct class_device *class_dev, struct class_interface *class_intf) { int err; From 45f8bde0d0d6deb168b45998c72b4fbeb2f57efb Mon Sep 17 00:00:00 2001 From: "Robert P. J. Day" Date: Fri, 26 Jan 2007 00:57:09 -0800 Subject: [PATCH 172/326] [PATCH] fix various kernel-doc in header files Fix a number of kernel-doc entries for header files in include/linux by making sure they begin with the appropriate '/**' notation and use @var notation. Signed-off-by: Robert P. J. Day Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/bitops.h | 6 ++---- include/linux/list.h | 10 +++++----- include/linux/mutex.h | 2 +- include/linux/rtmutex.h | 4 ++-- include/linux/timer.h | 4 ++-- 5 files changed, 12 insertions(+), 14 deletions(-) diff --git a/include/linux/bitops.h b/include/linux/bitops.h index 5d1eabcde5d5..638165f571da 100644 --- a/include/linux/bitops.h +++ b/include/linux/bitops.h @@ -31,9 +31,8 @@ static inline unsigned long hweight_long(unsigned long w) return sizeof(w) == 4 ? hweight32(w) : hweight64(w); } -/* +/** * rol32 - rotate a 32-bit value left - * * @word: value to rotate * @shift: bits to roll */ @@ -42,9 +41,8 @@ static inline __u32 rol32(__u32 word, unsigned int shift) return (word << shift) | (word >> (32 - shift)); } -/* +/** * ror32 - rotate a 32-bit value right - * * @word: value to rotate * @shift: bits to roll */ diff --git a/include/linux/list.h b/include/linux/list.h index a9c90287c0ff..611059d633f4 100644 --- a/include/linux/list.h +++ b/include/linux/list.h @@ -227,13 +227,13 @@ static inline void list_replace_init(struct list_head *old, INIT_LIST_HEAD(old); } -/* +/** * list_replace_rcu - replace old entry by new one * @old : the element to be replaced * @new : the new element to insert * - * The old entry will be replaced with the new entry atomically. - * Note: 'old' should not be empty. + * The @old entry will be replaced with the @new entry atomically. + * Note: @old should not be empty. */ static inline void list_replace_rcu(struct list_head *old, struct list_head *new) @@ -680,12 +680,12 @@ static inline void hlist_del_init(struct hlist_node *n) } } -/* +/** * hlist_replace_rcu - replace old entry by new one * @old : the element to be replaced * @new : the new element to insert * - * The old entry will be replaced with the new entry atomically. + * The @old entry will be replaced with the @new entry atomically. */ static inline void hlist_replace_rcu(struct hlist_node *old, struct hlist_node *new) diff --git a/include/linux/mutex.h b/include/linux/mutex.h index a7544afd7582..b81bc2adaeff 100644 --- a/include/linux/mutex.h +++ b/include/linux/mutex.h @@ -105,7 +105,7 @@ do { \ extern void __mutex_init(struct mutex *lock, const char *name, struct lock_class_key *key); -/*** +/** * mutex_is_locked - is the mutex locked * @lock: the mutex to be queried * diff --git a/include/linux/rtmutex.h b/include/linux/rtmutex.h index b0090e9f7884..382bb7951166 100644 --- a/include/linux/rtmutex.h +++ b/include/linux/rtmutex.h @@ -16,7 +16,7 @@ #include #include -/* +/** * The rt_mutex structure * * @wait_lock: spinlock to protect the structure @@ -71,7 +71,7 @@ struct hrtimer_sleeper; #define DEFINE_RT_MUTEX(mutexname) \ struct rt_mutex mutexname = __RT_MUTEX_INITIALIZER(mutexname) -/*** +/** * rt_mutex_is_locked - is the mutex locked * @lock: the mutex to be queried * diff --git a/include/linux/timer.h b/include/linux/timer.h index eeef6643d4c6..fb5edaaf0ebd 100644 --- a/include/linux/timer.h +++ b/include/linux/timer.h @@ -41,7 +41,7 @@ static inline void setup_timer(struct timer_list * timer, init_timer(timer); } -/*** +/** * timer_pending - is a timer pending? * @timer: the timer in question * @@ -63,7 +63,7 @@ extern int mod_timer(struct timer_list *timer, unsigned long expires); extern unsigned long next_timer_interrupt(void); -/*** +/** * add_timer - start a timer * @timer: the timer to be added * From a0ad13ef643a5829d63c456ab6143bbda60b44a9 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Fri, 26 Jan 2007 00:57:10 -0800 Subject: [PATCH 173/326] [PATCH] knfsd: Fix type mismatch with filldir_t used by nfsd nfsd defines a type 'encode_dent_fn' which is much like 'filldir_t' except that the first pointer is 'struct readdir_cd *' rather than 'void *'. It then casts encode_dent_fn points to 'filldir_t' as needed. This hides any other type mismatches between the two such as the fact that the 'ino' arg recently changed from ino_t to u64. So: get rid of 'encode_dent_fn', get rid of the cast of the function type, change the first arg of various functions from 'struct readdir_cd *' to 'void *', and live with the fact that we have a little less type checking on the calling of these functions now. Less internal (to nfsd) checking offset by more external checking, which is more important. Thanks to Gabriel Paubert for discovering this and providing an initial patch. Signed-off-by: Gabriel Paubert Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/nfs3xdr.c | 9 +++++---- fs/nfsd/nfs4xdr.c | 5 +++-- fs/nfsd/nfsxdr.c | 5 +++-- fs/nfsd/vfs.c | 4 ++-- include/linux/nfsd/nfsd.h | 4 +--- include/linux/nfsd/xdr.h | 4 ++-- include/linux/nfsd/xdr3.h | 8 ++++---- 7 files changed, 20 insertions(+), 19 deletions(-) diff --git a/fs/nfsd/nfs3xdr.c b/fs/nfsd/nfs3xdr.c index 277df40f098d..e695660921ec 100644 --- a/fs/nfsd/nfs3xdr.c +++ b/fs/nfsd/nfs3xdr.c @@ -990,15 +990,16 @@ encode_entry(struct readdir_cd *ccd, const char *name, } int -nfs3svc_encode_entry(struct readdir_cd *cd, const char *name, - int namlen, loff_t offset, ino_t ino, unsigned int d_type) +nfs3svc_encode_entry(void *cd, const char *name, + int namlen, loff_t offset, u64 ino, unsigned int d_type) { return encode_entry(cd, name, namlen, offset, ino, d_type, 0); } int -nfs3svc_encode_entry_plus(struct readdir_cd *cd, const char *name, - int namlen, loff_t offset, ino_t ino, unsigned int d_type) +nfs3svc_encode_entry_plus(void *cd, const char *name, + int namlen, loff_t offset, u64 ino, + unsigned int d_type) { return encode_entry(cd, name, namlen, offset, ino, d_type, 1); } diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index fea46368afb2..18aa9440df14 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -1880,9 +1880,10 @@ nfsd4_encode_rdattr_error(__be32 *p, int buflen, __be32 nfserr) } static int -nfsd4_encode_dirent(struct readdir_cd *ccd, const char *name, int namlen, - loff_t offset, ino_t ino, unsigned int d_type) +nfsd4_encode_dirent(void *ccdv, const char *name, int namlen, + loff_t offset, u64 ino, unsigned int d_type) { + struct readdir_cd *ccd = ccdv; struct nfsd4_readdir *cd = container_of(ccd, struct nfsd4_readdir, common); int buflen; __be32 *p = cd->buffer; diff --git a/fs/nfsd/nfsxdr.c b/fs/nfsd/nfsxdr.c index f5243f943996..6555c50d9006 100644 --- a/fs/nfsd/nfsxdr.c +++ b/fs/nfsd/nfsxdr.c @@ -462,9 +462,10 @@ nfssvc_encode_statfsres(struct svc_rqst *rqstp, __be32 *p, } int -nfssvc_encode_entry(struct readdir_cd *ccd, const char *name, - int namlen, loff_t offset, ino_t ino, unsigned int d_type) +nfssvc_encode_entry(void *ccdv, const char *name, + int namlen, loff_t offset, u64 ino, unsigned int d_type) { + struct readdir_cd *ccd = ccdv; struct nfsd_readdirres *cd = container_of(ccd, struct nfsd_readdirres, common); __be32 *p = cd->buffer; int buflen, slen; diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 4dd49d6e90cd..5d32e5fa697e 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1716,7 +1716,7 @@ nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, */ __be32 nfsd_readdir(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t *offsetp, - struct readdir_cd *cdp, encode_dent_fn func) + struct readdir_cd *cdp, filldir_t func) { __be32 err; int host_err; @@ -1741,7 +1741,7 @@ nfsd_readdir(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t *offsetp, do { cdp->err = nfserr_eof; /* will be cleared on successful read */ - host_err = vfs_readdir(file, (filldir_t) func, cdp); + host_err = vfs_readdir(file, func, cdp); } while (host_err >=0 && cdp->err == nfs_ok); if (host_err) err = nfserrno(host_err); diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h index 0727774772ba..4b7c4b568f6d 100644 --- a/include/linux/nfsd/nfsd.h +++ b/include/linux/nfsd/nfsd.h @@ -52,8 +52,6 @@ struct readdir_cd { __be32 err; /* 0, nfserr, or nfserr_eof */ }; -typedef int (*encode_dent_fn)(struct readdir_cd *, const char *, - int, loff_t, ino_t, unsigned int); typedef int (*nfsd_dirop_t)(struct inode *, struct dentry *, int, int); extern struct svc_program nfsd_program; @@ -117,7 +115,7 @@ __be32 nfsd_unlink(struct svc_rqst *, struct svc_fh *, int type, int nfsd_truncate(struct svc_rqst *, struct svc_fh *, unsigned long size); __be32 nfsd_readdir(struct svc_rqst *, struct svc_fh *, - loff_t *, struct readdir_cd *, encode_dent_fn); + loff_t *, struct readdir_cd *, filldir_t); __be32 nfsd_statfs(struct svc_rqst *, struct svc_fh *, struct kstatfs *); diff --git a/include/linux/nfsd/xdr.h b/include/linux/nfsd/xdr.h index 877192d3ae79..67885d5e6e50 100644 --- a/include/linux/nfsd/xdr.h +++ b/include/linux/nfsd/xdr.h @@ -165,8 +165,8 @@ int nfssvc_encode_readres(struct svc_rqst *, __be32 *, struct nfsd_readres *); int nfssvc_encode_statfsres(struct svc_rqst *, __be32 *, struct nfsd_statfsres *); int nfssvc_encode_readdirres(struct svc_rqst *, __be32 *, struct nfsd_readdirres *); -int nfssvc_encode_entry(struct readdir_cd *, const char *name, - int namlen, loff_t offset, ino_t ino, unsigned int); +int nfssvc_encode_entry(void *, const char *name, + int namlen, loff_t offset, u64 ino, unsigned int); int nfssvc_release_fhandle(struct svc_rqst *, __be32 *, struct nfsd_fhandle *); diff --git a/include/linux/nfsd/xdr3.h b/include/linux/nfsd/xdr3.h index 79963867b0d7..89d9d6061a62 100644 --- a/include/linux/nfsd/xdr3.h +++ b/include/linux/nfsd/xdr3.h @@ -331,11 +331,11 @@ int nfs3svc_release_fhandle(struct svc_rqst *, __be32 *, struct nfsd3_attrstat *); int nfs3svc_release_fhandle2(struct svc_rqst *, __be32 *, struct nfsd3_fhandle_pair *); -int nfs3svc_encode_entry(struct readdir_cd *, const char *name, - int namlen, loff_t offset, ino_t ino, +int nfs3svc_encode_entry(void *, const char *name, + int namlen, loff_t offset, u64 ino, unsigned int); -int nfs3svc_encode_entry_plus(struct readdir_cd *, const char *name, - int namlen, loff_t offset, ino_t ino, +int nfs3svc_encode_entry_plus(void *, const char *name, + int namlen, loff_t offset, u64 ino, unsigned int); /* Helper functions for NFSv3 ACL code */ __be32 *nfs3svc_encode_post_op_attr(struct svc_rqst *rqstp, __be32 *p, From 2a2275d630b982e5f90206f9bc497f6695a3ec5d Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Fri, 26 Jan 2007 00:57:11 -0800 Subject: [PATCH 174/326] [PATCH] md: fix potential memalloc deadlock in md If a GFP_KERNEL allocation is attempted in md while the mddev_lock is held, it is possible for a deadlock to eventuate. This happens if the array was marked 'clean', and the memalloc triggers a write-out to the md device. For the writeout to succeed, the array must be marked 'dirty', and that requires getting the mddev_lock. So, before attempting a GFP_KERNEL allocation while holding the lock, make sure the array is marked 'dirty' (unless it is currently read-only). Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/md.c | 29 +++++++++++++++++++++++++++++ drivers/md/raid1.c | 2 ++ drivers/md/raid5.c | 3 +++ include/linux/raid/md.h | 2 +- 4 files changed, 35 insertions(+), 1 deletion(-) diff --git a/drivers/md/md.c b/drivers/md/md.c index ec3d8e8a0bd3..e8807ea5377d 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -3564,6 +3564,8 @@ static int get_bitmap_file(mddev_t * mddev, void __user * arg) char *ptr, *buf = NULL; int err = -ENOMEM; + md_allow_write(mddev); + file = kmalloc(sizeof(*file), GFP_KERNEL); if (!file) goto out; @@ -5032,6 +5034,33 @@ void md_write_end(mddev_t *mddev) } } +/* md_allow_write(mddev) + * Calling this ensures that the array is marked 'active' so that writes + * may proceed without blocking. It is important to call this before + * attempting a GFP_KERNEL allocation while holding the mddev lock. + * Must be called with mddev_lock held. + */ +void md_allow_write(mddev_t *mddev) +{ + if (!mddev->pers) + return; + if (mddev->ro) + return; + + spin_lock_irq(&mddev->write_lock); + if (mddev->in_sync) { + mddev->in_sync = 0; + set_bit(MD_CHANGE_CLEAN, &mddev->flags); + if (mddev->safemode_delay && + mddev->safemode == 0) + mddev->safemode = 1; + spin_unlock_irq(&mddev->write_lock); + md_update_sb(mddev, 0); + } else + spin_unlock_irq(&mddev->write_lock); +} +EXPORT_SYMBOL_GPL(md_allow_write); + static DECLARE_WAIT_QUEUE_HEAD(resync_wait); #define SYNC_MARKS 10 diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index ab74d40cac98..97ee870b265d 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -2104,6 +2104,8 @@ static int raid1_reshape(mddev_t *mddev) return -EINVAL; } + md_allow_write(mddev); + raid_disks = mddev->raid_disks + mddev->delta_disks; if (raid_disks < conf->raid_disks) { diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index be008f034ada..8a30b297ac3a 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -405,6 +405,8 @@ static int resize_stripes(raid5_conf_t *conf, int newsize) if (newsize <= conf->pool_size) return 0; /* never bother to shrink */ + md_allow_write(conf->mddev); + /* Step 1 */ sc = kmem_cache_create(conf->cache_name[1-conf->active_name], sizeof(struct stripe_head)+(newsize-1)*sizeof(struct r5dev), @@ -3250,6 +3252,7 @@ raid5_store_stripe_cache_size(mddev_t *mddev, const char *page, size_t len) else break; } + md_allow_write(mddev); while (new > conf->max_nr_stripes) { if (grow_one_stripe(conf)) conf->max_nr_stripes++; diff --git a/include/linux/raid/md.h b/include/linux/raid/md.h index 866a1e2b0ce0..fbaeda79b2e9 100644 --- a/include/linux/raid/md.h +++ b/include/linux/raid/md.h @@ -94,7 +94,7 @@ extern int sync_page_io(struct block_device *bdev, sector_t sector, int size, struct page *page, int rw); extern void md_do_sync(mddev_t *mddev); extern void md_new_event(mddev_t *mddev); - +extern void md_allow_write(mddev_t *mddev); #endif /* CONFIG_MD */ #endif From 569d3287c1f2c5d5e9264e89bd2c0f632e7b7dbb Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Fri, 26 Jan 2007 00:57:11 -0800 Subject: [PATCH 175/326] [PATCH] MM: Remove [PATCH] invalidate_inode_pages2_range() debug NFS can handle the case where invalidate_inode_pages2_range() fails, so the premise behind commit 8258d4a574d3a8c01f0ef68aa26b969398a0e140 is now gone. Remove the WARN_ON_ONCE() which is causing users grief as we can see from http://bugzilla.kernel.org/show_bug.cgi?id=7826 Signed-off-by: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/truncate.c | 1 - 1 file changed, 1 deletion(-) diff --git a/mm/truncate.c b/mm/truncate.c index 3262740aa059..5df947de7654 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -429,7 +429,6 @@ int invalidate_inode_pages2_range(struct address_space *mapping, pagevec_release(&pvec); cond_resched(); } - WARN_ON_ONCE(ret); return ret; } EXPORT_SYMBOL_GPL(invalidate_inode_pages2_range); From fe33f6f1525113ab0bae416966209438e7a31987 Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Fri, 26 Jan 2007 00:57:12 -0800 Subject: [PATCH 176/326] [PATCH] Fix UML on non-standard VM split hosts This fixes UML on hosts with non-standard VM splits. We had changed the config variable that controls UML behavior on such hosts, but not propogated the change everywhere. In particular, the values of STUB_CODE and STUB_DATA relied on the old variable. I also reformatted the HOST_VMSPLIT_3G help to make it more standard. Spotted by uml@flonatel.org. Signed-off-by: Jeff Dike Cc: Blaisorblade Cc: Pravin Cc: Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/Kconfig.i386 | 34 +++++++++++++++++----------------- 1 file changed, 17 insertions(+), 17 deletions(-) diff --git a/arch/um/Kconfig.i386 b/arch/um/Kconfig.i386 index f191a550a079..77558a88a2fe 100644 --- a/arch/um/Kconfig.i386 +++ b/arch/um/Kconfig.i386 @@ -19,22 +19,22 @@ config SEMAPHORE_SLEEPERS choice prompt "Host memory split" default HOST_VMSPLIT_3G - ---help--- - This is needed when the host kernel on which you run has a non-default - (like 2G/2G) memory split, instead of the customary 3G/1G. If you did - not recompile your own kernel but use the default distro's one, you can - safely accept the "Default split" option. + help + This is needed when the host kernel on which you run has a non-default + (like 2G/2G) memory split, instead of the customary 3G/1G. If you did + not recompile your own kernel but use the default distro's one, you can + safely accept the "Default split" option. - It can be enabled on recent (>=2.6.16-rc2) vanilla kernels via - CONFIG_VM_SPLIT_*, or on previous kernels with special patches (-ck - patchset by Con Kolivas, or other ones) - option names match closely the - host CONFIG_VM_SPLIT_* ones. + It can be enabled on recent (>=2.6.16-rc2) vanilla kernels via + CONFIG_VM_SPLIT_*, or on previous kernels with special patches (-ck + patchset by Con Kolivas, or other ones) - option names match closely the + host CONFIG_VM_SPLIT_* ones. - A lower setting (where 1G/3G is lowest and 3G/1G is higher) will - tolerate even more "normal" host kernels, but an higher setting will be - stricter. + A lower setting (where 1G/3G is lowest and 3G/1G is higher) will + tolerate even more "normal" host kernels, but an higher setting will be + stricter. - So, if you do not know what to do here, say 'Default split'. + So, if you do not know what to do here, say 'Default split'. config HOST_VMSPLIT_3G bool "Default split (3G/1G user/kernel host split)" @@ -67,13 +67,13 @@ config 3_LEVEL_PGTABLES config STUB_CODE hex - default 0xbfffe000 if !HOST_2G_2G - default 0x7fffe000 if HOST_2G_2G + default 0xbfffe000 if !HOST_VMSPLIT_2G + default 0x7fffe000 if HOST_VMSPLIT_2G config STUB_DATA hex - default 0xbffff000 if !HOST_2G_2G - default 0x7ffff000 if HOST_2G_2G + default 0xbffff000 if !HOST_VMSPLIT_2G + default 0x7ffff000 if HOST_VMSPLIT_2G config STUB_START hex From c20086de9319ac406f1e96ad459763c9f9965b18 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Fri, 26 Jan 2007 00:57:14 -0800 Subject: [PATCH 177/326] [PATCH] md: remove unnecessary printk when raid5 gets an unaligned read. raid5_mergeable_bvec tries to ensure that raid5 never sees a read request that does not fit within just one chunk. However as we must always accept a single-page read, that is not always possible. So when "in_chunk_boundary" fails, it might be unusual, but it is not a problem and printing a message every time is a bad idea. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/md/raid5.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 8a30b297ac3a..467c16982d02 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -2680,7 +2680,7 @@ static int chunk_aligned_read(request_queue_t *q, struct bio * raid_bio) mdk_rdev_t *rdev; if (!in_chunk_boundary(mddev, raid_bio)) { - printk("chunk_aligned_read : non aligned\n"); + PRINTK("chunk_aligned_read : non aligned\n"); return 0; } /* From 1fb844961818ce94e782acf6a96b92dc2303553b Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Fri, 26 Jan 2007 00:57:16 -0800 Subject: [PATCH 178/326] [PATCH] core-dumping unreadable binaries via PT_INTERP Proposed patch to fix #5 in http://www.isec.pl/vulnerabilities/isec-0017-binfmt_elf.txt aka http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2004-1073 To reproduce, do * grab poc at the end of advisory. * add line "eph.p_memsz = 4096;" after "eph.p_filesz = 4096;" where first "4096" is something equal to or greater than 4096. * ./poc /usr/bin/sudo && ls -l Here I get with 2.6.20-rc5: -rw------- 1 ad ad 102400 2007-01-15 19:17 core ---s--x--x 2 root root 101820 2007-01-15 19:15 /usr/bin/sudo Check for MAY_READ like binfmt_misc.c does. Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/binfmt_elf.c | 9 +++++++++ fs/binfmt_elf_fdpic.c | 8 ++++++++ 2 files changed, 17 insertions(+) diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 90461f49e902..669dbe5b0317 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -682,6 +682,15 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) retval = PTR_ERR(interpreter); if (IS_ERR(interpreter)) goto out_free_interp; + + /* + * If the binary is not readable then enforce + * mm->dumpable = 0 regardless of the interpreter's + * permissions. + */ + if (file_permission(interpreter, MAY_READ) < 0) + bprm->interp_flags |= BINPRM_FLAGS_ENFORCE_NONDUMP; + retval = kernel_read(interpreter, 0, bprm->buf, BINPRM_BUF_SIZE); if (retval != BINPRM_BUF_SIZE) { diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c index 6e6d4568d548..a4d933a51208 100644 --- a/fs/binfmt_elf_fdpic.c +++ b/fs/binfmt_elf_fdpic.c @@ -234,6 +234,14 @@ static int load_elf_fdpic_binary(struct linux_binprm *bprm, goto error; } + /* + * If the binary is not readable then enforce + * mm->dumpable = 0 regardless of the interpreter's + * permissions. + */ + if (file_permission(interpreter, MAY_READ) < 0) + bprm->interp_flags |= BINPRM_FLAGS_ENFORCE_NONDUMP; + retval = kernel_read(interpreter, 0, bprm->buf, BINPRM_BUF_SIZE); if (retval < 0) From 01f2073411e01777e3c6f45a4bf05ea76493f326 Mon Sep 17 00:00:00 2001 From: Francois Romieu Date: Fri, 26 Jan 2007 00:57:17 -0800 Subject: [PATCH 179/326] [PATCH] netdev: add a MAINTAINERS entry for via-velocity and update my address Signed-off-by: Francois Romieu Cc: "David S. Miller" Cc: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- MAINTAINERS | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 796e891083b7..f0596e452c5c 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1137,9 +1137,9 @@ T: git kernel.org:/pub/scm/linux/kernel/git/airlied/drm-2.6.git S: Maintained DSCC4 DRIVER -P: François Romieu -M: romieu@cogenit.fr -M: romieu@ensta.fr +P: Francois Romieu +M: romieu@fr.zoreil.com +L: netdev@vger.kernel.org S: Maintained DVB SUBSYSTEM AND DRIVERS @@ -3574,6 +3574,12 @@ M: khali@linux-fr.org L: i2c@lm-sensors.org S: Maintained +VIA VELOCITY NETWORK DRIVER +P: Francois Romieu +M: romieu@fr.zoreil.com +L: netdev@vger.kernel.org +S: Maintained + UCLINUX (AND M68KNOMMU) P: Greg Ungerer M: gerg@uclinux.org From 496a0fc8c5572a626de41d56d7c7ed005a2c1b48 Mon Sep 17 00:00:00 2001 From: Matt Domsch Date: Fri, 26 Jan 2007 00:57:18 -0800 Subject: [PATCH 180/326] [PATCH] Fix race in efi variable delete code Fix race when deleting an EFI variable and issuing another EFI command on the same variable. The removal of the variable from the efivars_list should be done in efivar_delete and not delayed until the kobject release. Furthermore, remove the item from the list at module unload time, and use list_for_each_entry_safe() rather than list_for_each_safe() for readability. Tested on ia64. Signed-off-by: Prarit Bhargava Signed-off-by: Matt Domsch Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/firmware/efivars.c | 29 ++++++++++++----------------- 1 file changed, 12 insertions(+), 17 deletions(-) diff --git a/drivers/firmware/efivars.c b/drivers/firmware/efivars.c index 5ab5e393b882..c6281ccd4fe7 100644 --- a/drivers/firmware/efivars.c +++ b/drivers/firmware/efivars.c @@ -122,8 +122,6 @@ struct efivar_entry { struct kobject kobj; }; -#define get_efivar_entry(n) list_entry(n, struct efivar_entry, list) - struct efivar_attribute { struct attribute attr; ssize_t (*show) (struct efivar_entry *entry, char *buf); @@ -386,9 +384,6 @@ static struct sysfs_ops efivar_attr_ops = { static void efivar_release(struct kobject *kobj) { struct efivar_entry *var = container_of(kobj, struct efivar_entry, kobj); - spin_lock(&efivars_lock); - list_del(&var->list); - spin_unlock(&efivars_lock); kfree(var); } @@ -430,9 +425,8 @@ static ssize_t efivar_create(struct subsystem *sub, const char *buf, size_t count) { struct efi_variable *new_var = (struct efi_variable *)buf; - struct efivar_entry *search_efivar = NULL; + struct efivar_entry *search_efivar, *n; unsigned long strsize1, strsize2; - struct list_head *pos, *n; efi_status_t status = EFI_NOT_FOUND; int found = 0; @@ -444,8 +438,7 @@ efivar_create(struct subsystem *sub, const char *buf, size_t count) /* * Does this variable already exist? */ - list_for_each_safe(pos, n, &efivar_list) { - search_efivar = get_efivar_entry(pos); + list_for_each_entry_safe(search_efivar, n, &efivar_list, list) { strsize1 = utf8_strsize(search_efivar->var.VariableName, 1024); strsize2 = utf8_strsize(new_var->VariableName, 1024); if (strsize1 == strsize2 && @@ -490,9 +483,8 @@ static ssize_t efivar_delete(struct subsystem *sub, const char *buf, size_t count) { struct efi_variable *del_var = (struct efi_variable *)buf; - struct efivar_entry *search_efivar = NULL; + struct efivar_entry *search_efivar, *n; unsigned long strsize1, strsize2; - struct list_head *pos, *n; efi_status_t status = EFI_NOT_FOUND; int found = 0; @@ -504,8 +496,7 @@ efivar_delete(struct subsystem *sub, const char *buf, size_t count) /* * Does this variable already exist? */ - list_for_each_safe(pos, n, &efivar_list) { - search_efivar = get_efivar_entry(pos); + list_for_each_entry_safe(search_efivar, n, &efivar_list, list) { strsize1 = utf8_strsize(search_efivar->var.VariableName, 1024); strsize2 = utf8_strsize(del_var->VariableName, 1024); if (strsize1 == strsize2 && @@ -537,9 +528,9 @@ efivar_delete(struct subsystem *sub, const char *buf, size_t count) spin_unlock(&efivars_lock); return -EIO; } + list_del(&search_efivar->list); /* We need to release this lock before unregistering. */ spin_unlock(&efivars_lock); - efivar_unregister(search_efivar); /* It's dead Jim.... */ @@ -768,10 +759,14 @@ efivars_init(void) static void __exit efivars_exit(void) { - struct list_head *pos, *n; + struct efivar_entry *entry, *n; - list_for_each_safe(pos, n, &efivar_list) - efivar_unregister(get_efivar_entry(pos)); + list_for_each_entry_safe(entry, n, &efivar_list, list) { + spin_lock(&efivars_lock); + list_del(&entry->list); + spin_unlock(&efivars_lock); + efivar_unregister(entry); + } subsystem_unregister(&vars_subsys); firmware_unregister(&efi_subsys); From 6096b63e2584796341baf7e7735f98d387f489f2 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 26 Jan 2007 14:47:38 +0900 Subject: [PATCH 181/326] ahci: fix endianness in spurious interrupt message Fix endianness in spurious interrupt message. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/ata/ahci.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c index 2fe5a58195de..d8f0ce970d54 100644 --- a/drivers/ata/ahci.c +++ b/drivers/ata/ahci.c @@ -1178,7 +1178,8 @@ static void ahci_host_intr(struct ata_port *ap) ata_port_printk(ap, KERN_INFO, "Spurious SDB FIS during NCQ " "issue=0x%x SAct=0x%x FIS=%08x:%08x%s\n", readl(port_mmio + PORT_CMD_ISSUE), - readl(port_mmio + PORT_SCR_ACT), f[0], f[1], + readl(port_mmio + PORT_SCR_ACT), + le32_to_cpu(f[0]), le32_to_cpu(f[1]), pp->ncq_saw_spurious_sdb_cnt < 10 ? "" : ", shutting up"); From d02598721706ab62a574823479b1f6c26c8980d2 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 26 Jan 2007 14:57:31 +0900 Subject: [PATCH 182/326] sata_via: style clean up, no indirect method call in LLD Call ata_bmdma_irq_clear() directly instead of through ap->ops->irq_clear() according to libata style guideline. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/ata/sata_via.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/ata/sata_via.c b/drivers/ata/sata_via.c index 55b01236a6f0..d3d5c0d57032 100644 --- a/drivers/ata/sata_via.c +++ b/drivers/ata/sata_via.c @@ -211,7 +211,7 @@ static void svia_noop_freeze(struct ata_port *ap) * certain way. Leave it alone and just clear pending IRQ. */ ata_chk_status(ap); - ap->ops->irq_clear(ap); + ata_bmdma_irq_clear(ap); } /** From dfd7a3db3898e299bdc25f0c77081a8632b3a73c Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 26 Jan 2007 15:37:20 +0900 Subject: [PATCH 183/326] ahci: use 0x80 as wait stat value instead of 0xff Before hardreset, ahci initialized stat part of received FIS area to 0xff to wait for the first D2H Reg FIS which would change the value to device ready state. This used to work but now libata considers status value of 0xff as device not present making this wait prone to failure. This patch makes ahci use 0x80 for the wait stat value instead of 0xff to fix the above problem. Signed-off-by: Tejun Heo drivers/ata/ahci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) Signed-off-by: Jeff Garzik --- drivers/ata/ahci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c index d8f0ce970d54..28a82e3403f1 100644 --- a/drivers/ata/ahci.c +++ b/drivers/ata/ahci.c @@ -903,7 +903,7 @@ static int ahci_hardreset(struct ata_port *ap, unsigned int *class) /* clear D2H reception area to properly wait for D2H FIS */ ata_tf_init(ap->device, &tf); - tf.command = 0xff; + tf.command = 0x80; ata_tf_to_fis(&tf, d2h_fis, 0); rc = sata_std_hardreset(ap, class); From 8cdf92a98fa0f91068615443f2a8597b7f2c34ca Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Mon, 1 Jan 2007 19:31:15 +0000 Subject: [PATCH 184/326] Fix Maple PATA IRQ assignment. On the Maple board, the AMD8111 IDE is in legacy mode... except that it appears on IRQ 20 instead of IRQ 15. For drivers/ide this was handled by the architecture's "pci_get_legacy_ide_irq()" function, but in libata we just hard-code the numbers 14 and 15. This patch provides asm-powerpc/libata-portmap.h which maps the IRQ as appropriate, having added a pci_dev argument to the ATA_{PRIM,SECOND}ARY_IRQ macros. There's probably a better way to do this -- especially if we observe that the _only_ case in which this seemingly-generic "pci_get_legacy_ide_irq()" function returns anything other than 14 and 15 for primary and secondary respectively is the case of the AMD8111 on the Maple board -- couldn't we handle that with a special case in the pata_amd driver, or perhaps with a PCI quirk for Maple to switch it into native mode during early boot and assign resources properly? Signed-off-by: David Woodhouse Signed-off-by: Jeff Garzik --- arch/powerpc/Kconfig | 1 + drivers/ata/Kconfig | 4 ++++ drivers/ata/libata-sff.c | 6 +++--- include/asm-generic/libata-portmap.h | 4 ++-- include/asm-powerpc/libata-portmap.h | 12 ++++++++++++ 5 files changed, 22 insertions(+), 5 deletions(-) create mode 100644 include/asm-powerpc/libata-portmap.h diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 0855d55c194d..b268c417c0bf 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -484,6 +484,7 @@ config PPC_MAPLE select PPC_970_NAP select PPC_NATIVE select PPC_RTAS + select ATA_NONSTANDARD if ATA default n help This option enables support for the Maple 970FX Evaluation Board. diff --git a/drivers/ata/Kconfig b/drivers/ata/Kconfig index da21552d2b1c..1c94b43d2c9b 100644 --- a/drivers/ata/Kconfig +++ b/drivers/ata/Kconfig @@ -19,6 +19,10 @@ config ATA if ATA +config ATA_NONSTANDARD + bool + default n + config SATA_AHCI tristate "AHCI SATA support" depends on PCI diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c index 942aeba2940a..12c88c588039 100644 --- a/drivers/ata/libata-sff.c +++ b/drivers/ata/libata-sff.c @@ -917,7 +917,7 @@ static struct ata_probe_ent *ata_pci_init_legacy_port(struct pci_dev *pdev, probe_ent->irq_flags = IRQF_SHARED; if (port_mask & ATA_PORT_PRIMARY) { - probe_ent->irq = ATA_PRIMARY_IRQ; + probe_ent->irq = ATA_PRIMARY_IRQ(pdev); probe_ent->port[0].cmd_addr = ATA_PRIMARY_CMD; probe_ent->port[0].altstatus_addr = probe_ent->port[0].ctl_addr = ATA_PRIMARY_CTL; @@ -933,9 +933,9 @@ static struct ata_probe_ent *ata_pci_init_legacy_port(struct pci_dev *pdev, if (port_mask & ATA_PORT_SECONDARY) { if (probe_ent->irq) - probe_ent->irq2 = ATA_SECONDARY_IRQ; + probe_ent->irq2 = ATA_SECONDARY_IRQ(pdev); else - probe_ent->irq = ATA_SECONDARY_IRQ; + probe_ent->irq = ATA_SECONDARY_IRQ(pdev); probe_ent->port[1].cmd_addr = ATA_SECONDARY_CMD; probe_ent->port[1].altstatus_addr = probe_ent->port[1].ctl_addr = ATA_SECONDARY_CTL; diff --git a/include/asm-generic/libata-portmap.h b/include/asm-generic/libata-portmap.h index 9202fd02d5be..62fb3618293d 100644 --- a/include/asm-generic/libata-portmap.h +++ b/include/asm-generic/libata-portmap.h @@ -3,10 +3,10 @@ #define ATA_PRIMARY_CMD 0x1F0 #define ATA_PRIMARY_CTL 0x3F6 -#define ATA_PRIMARY_IRQ 14 +#define ATA_PRIMARY_IRQ(dev) 14 #define ATA_SECONDARY_CMD 0x170 #define ATA_SECONDARY_CTL 0x376 -#define ATA_SECONDARY_IRQ 15 +#define ATA_SECONDARY_IRQ(dev) 15 #endif diff --git a/include/asm-powerpc/libata-portmap.h b/include/asm-powerpc/libata-portmap.h new file mode 100644 index 000000000000..4d8518049f4d --- /dev/null +++ b/include/asm-powerpc/libata-portmap.h @@ -0,0 +1,12 @@ +#ifndef __ASM_POWERPC_LIBATA_PORTMAP_H +#define __ASM_POWERPC_LIBATA_PORTMAP_H + +#define ATA_PRIMARY_CMD 0x1F0 +#define ATA_PRIMARY_CTL 0x3F6 +#define ATA_PRIMARY_IRQ(dev) pci_get_legacy_ide_irq(dev, 0) + +#define ATA_SECONDARY_CMD 0x170 +#define ATA_SECONDARY_CTL 0x376 +#define ATA_SECONDARY_IRQ(dev) pci_get_legacy_ide_irq(dev, 1) + +#endif From a8a75a20e9a854685ac3e7af2e0e32f9b0be436b Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Fri, 26 Jan 2007 10:46:59 -0800 Subject: [PATCH 185/326] [PATCH] ocfs2: fix thinko in ocfs2_backup_super_blkno() Fix a bug which was introduced when I synced up ocfs2_fs.h with ocfs2-tools. We can't do u64/u32 in kernel. Signed-off-by: Mark Fasheh Signed-off-by: Linus Torvalds --- fs/ocfs2/ocfs2_fs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/ocfs2/ocfs2_fs.h b/fs/ocfs2/ocfs2_fs.h index c99e9058c198..e61e218f5e0b 100644 --- a/fs/ocfs2/ocfs2_fs.h +++ b/fs/ocfs2/ocfs2_fs.h @@ -587,7 +587,7 @@ static inline u64 ocfs2_backup_super_blkno(struct super_block *sb, int index) if (index >= 0 && index < OCFS2_MAX_BACKUP_SUPERBLOCKS) { offset <<= (2 * index); - offset /= sb->s_blocksize; + offset >>= sb->s_blocksize_bits; return offset; } From c229ec5dae58b218cab0bc1b36a7647b0ec4900f Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Fri, 26 Jan 2007 09:15:47 -0800 Subject: [PATCH 186/326] [PATCH] Boot loader ID for Gujin Add an official boot loader ID for Gujin. Signed-off-by: H. Peter Anvin Signed-off-by: Linus Torvalds --- Documentation/i386/boot.txt | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/Documentation/i386/boot.txt b/Documentation/i386/boot.txt index 9575de300a61..38fe1f03fb14 100644 --- a/Documentation/i386/boot.txt +++ b/Documentation/i386/boot.txt @@ -2,7 +2,7 @@ ---------------------------- H. Peter Anvin - Last update 2006-11-17 + Last update 2007-01-26 On the i386 platform, the Linux kernel uses a rather complicated boot convention. This has evolved partially due to historical aspects, as @@ -186,6 +186,7 @@ filled out, however: 7 GRuB 8 U-BOOT 9 Xen + A Gujin Please contact if you need a bootloader ID value assigned. From 86d43258bcad2068744c85181b97bb35ab27a74d Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 26 Jan 2007 18:48:16 -0800 Subject: [PATCH 187/326] [SPARC64]: Set g4/g5 properly in sun4v dtlb-prot handling. Mirror the logic in the sun4u handler, we have to update both registers even when we branch out to window fault fixup handling. The way it works is that if we are in etrap processing a fault already, g4/g5 holds the original fault information. If we take a window spill fault while doing etrap, then we put the window spill fault info into g4/g5 and this is what the top-level fault handler ends up processing first. Then we retry the originally faulting instruction, and process the original fault at that time. This is all necessary because of how constrained the trap registers are in these code paths. These cases trigger very rarely, so even if there is some performance implication it's doesn't happen very often. In fact the rarity is why it took so long to trigger and find this particular bug. Signed-off-by: David S. Miller --- arch/sparc64/kernel/sun4v_tlb_miss.S | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/sparc64/kernel/sun4v_tlb_miss.S b/arch/sparc64/kernel/sun4v_tlb_miss.S index b731881224e8..9871dbb1ab42 100644 --- a/arch/sparc64/kernel/sun4v_tlb_miss.S +++ b/arch/sparc64/kernel/sun4v_tlb_miss.S @@ -142,9 +142,9 @@ sun4v_dtlb_prot: rdpr %tl, %g1 cmp %g1, 1 bgu,pn %xcc, winfix_trampoline - nop - ba,pt %xcc, sparc64_realfault_common mov FAULT_CODE_DTLB | FAULT_CODE_WRITE, %g4 + ba,pt %xcc, sparc64_realfault_common + nop /* Called from trap table: * %g4: vaddr From 342a0cff0ad5fba6b591cfa37db3c65c4d9913f8 Mon Sep 17 00:00:00 2001 From: Venkat Yekkirala Date: Fri, 26 Jan 2007 19:03:48 -0800 Subject: [PATCH 188/326] [SELINUX]: Fix 2.6.20-rc6 build when no xfrm This patch is an incremental fix to the flow_cache_genid patch for selinux that breaks the build of 2.6.20-rc6 when xfrm is not configured. Signed-off-by: Venkat Yekkirala Signed-off-by: David S. Miller --- security/selinux/include/xfrm.h | 9 +++++++++ security/selinux/ss/services.c | 6 +++--- 2 files changed, 12 insertions(+), 3 deletions(-) diff --git a/security/selinux/include/xfrm.h b/security/selinux/include/xfrm.h index 161eb571c82d..31929e39f5ca 100644 --- a/security/selinux/include/xfrm.h +++ b/security/selinux/include/xfrm.h @@ -37,6 +37,11 @@ int selinux_xfrm_sock_rcv_skb(u32 sid, struct sk_buff *skb, int selinux_xfrm_postroute_last(u32 isec_sid, struct sk_buff *skb, struct avc_audit_data *ad, u8 proto); int selinux_xfrm_decode_session(struct sk_buff *skb, u32 *sid, int ckall); + +static inline void selinux_xfrm_notify_policyload(void) +{ + atomic_inc(&flow_cache_genid); +} #else static inline int selinux_xfrm_sock_rcv_skb(u32 isec_sid, struct sk_buff *skb, struct avc_audit_data *ad) @@ -55,6 +60,10 @@ static inline int selinux_xfrm_decode_session(struct sk_buff *skb, u32 *sid, int *sid = SECSID_NULL; return 0; } + +static inline void selinux_xfrm_notify_policyload(void) +{ +} #endif static inline void selinux_skb_xfrm_sid(struct sk_buff *skb, u32 *sid) diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c index ff0393317f39..ca9154dc5d82 100644 --- a/security/selinux/ss/services.c +++ b/security/selinux/ss/services.c @@ -1299,7 +1299,7 @@ int security_load_policy(void *data, size_t len) avc_ss_reset(seqno); selnl_notify_policyload(seqno); selinux_netlbl_cache_invalidate(); - atomic_inc(&flow_cache_genid); + selinux_xfrm_notify_policyload(); return 0; } @@ -1355,7 +1355,7 @@ int security_load_policy(void *data, size_t len) avc_ss_reset(seqno); selnl_notify_policyload(seqno); selinux_netlbl_cache_invalidate(); - atomic_inc(&flow_cache_genid); + selinux_xfrm_notify_policyload(); return 0; @@ -1855,7 +1855,7 @@ int security_set_bools(int len, int *values) if (!rc) { avc_ss_reset(seqno); selnl_notify_policyload(seqno); - atomic_inc(&flow_cache_genid); + selinux_xfrm_notify_policyload(); } return rc; } From 095b8501e4168ae5a879fcb9420ac48cbd43f95a Mon Sep 17 00:00:00 2001 From: Robert Olsson Date: Fri, 26 Jan 2007 19:06:01 -0800 Subject: [PATCH 189/326] [IPV4]: Fix single-entry /proc/net/fib_trie output. When main table is just a single leaf this gets printed as belonging to the local table in /proc/net/fib_trie. A fix is below. Signed-off-by: Robert Olsson Acked-by: Eric W. Biederman Signed-off-by: David S. Miller --- net/ipv4/fib_trie.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index 13307c04d5a1..1e589b91605e 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -2290,16 +2290,17 @@ static int fib_trie_seq_show(struct seq_file *seq, void *v) if (v == SEQ_START_TOKEN) return 0; + if (!NODE_PARENT(n)) { + if (iter->trie == trie_local) + seq_puts(seq, ":\n"); + else + seq_puts(seq, "
:\n"); + } + if (IS_TNODE(n)) { struct tnode *tn = (struct tnode *) n; __be32 prf = htonl(MASK_PFX(tn->key, tn->pos)); - if (!NODE_PARENT(n)) { - if (iter->trie == trie_local) - seq_puts(seq, ":\n"); - else - seq_puts(seq, "
:\n"); - } seq_indent(seq, iter->depth-1); seq_printf(seq, " +-- %d.%d.%d.%d/%d %d %d %d\n", NIPQUAD(prf), tn->pos, tn->bits, tn->full_children, From a718728f9e40ec79c0879ec6509a54fee214f5b2 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Sat, 27 Jan 2007 11:04:26 +0900 Subject: [PATCH 190/326] ahci: port_no should be used when clearing IRQ in ahci_thaw() ap->id is logcial port ID which is unique among all ATA ports and doesn't have anything to do with hardware port index. ap->port_no is the hardware port index and thus should be used when clearing IRQ mask in ahci_thaw(). This problem has been spotted by Jeff Garzik . Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/ata/ahci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c index 28a82e3403f1..48616c6fee9d 100644 --- a/drivers/ata/ahci.c +++ b/drivers/ata/ahci.c @@ -1282,7 +1282,7 @@ static void ahci_thaw(struct ata_port *ap) /* clear IRQ */ tmp = readl(port_mmio + PORT_IRQ_STAT); writel(tmp, port_mmio + PORT_IRQ_STAT); - writel(1 << ap->id, mmio + HOST_IRQ_STAT); + writel(1 << ap->port_no, mmio + HOST_IRQ_STAT); /* turn IRQ back on */ writel(DEF_PORT_IRQ, port_mmio + PORT_IRQ_MASK); From 03ee5b1cdd09ed1ee2e75d0bc647fc5db66b9d07 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 26 Jan 2007 20:10:25 +0900 Subject: [PATCH 191/326] libata: fix ata_eh_suspend() return value ata_eh_suspend() was returning 0 regardless of failure. This bug has potential to lose data on suspend. Fix it. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/ata/libata-eh.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c index 56cf59b60ec4..748435807d68 100644 --- a/drivers/ata/libata-eh.c +++ b/drivers/ata/libata-eh.c @@ -1796,7 +1796,7 @@ static int ata_eh_suspend(struct ata_port *ap, struct ata_device **r_failed_dev) *r_failed_dev = dev; DPRINTK("EXIT\n"); - return 0; + return rc; } /** From 1e7106fc7ea6af9c365afe2bfcde57cb1fdd1093 Mon Sep 17 00:00:00 2001 From: Bartlomiej Zolnierkiewicz Date: Sat, 27 Jan 2007 13:46:14 +0100 Subject: [PATCH 192/326] ide: update MAINTAINERS entry Signed-off-by: Bartlomiej Zolnierkiewicz --- MAINTAINERS | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index f0596e452c5c..1446cc41c12f 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1598,12 +1598,11 @@ M: ipslinux@adaptec.com W: http://www.developer.ibm.com/welcome/netfinity/serveraid.html S: Supported -IDE DRIVER [GENERAL] +IDE SUBSYSTEM P: Bartlomiej Zolnierkiewicz -M: B.Zolnierkiewicz@elka.pw.edu.pl -L: linux-kernel@vger.kernel.org +M: bzolnier@gmail.com L: linux-ide@vger.kernel.org -T: git kernel.org:/pub/scm/linux/kernel/git/bart/ide-2.6.git +T: quilt kernel.org/pub/linux/kernel/people/bart/pata-2.6/ S: Maintained IDE/ATAPI CDROM DRIVER From a51545ab2523b9cfd426737495f877821006371a Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Sat, 27 Jan 2007 13:46:21 +0100 Subject: [PATCH 193/326] jmicron: fix warning Signed-off-by: Andrew Morton Signed-off-by: Bartlomiej Zolnierkiewicz --- drivers/ide/pci/jmicron.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/ide/pci/jmicron.c b/drivers/ide/pci/jmicron.c index c1cec236ecf0..35dda8fcd0ac 100644 --- a/drivers/ide/pci/jmicron.c +++ b/drivers/ide/pci/jmicron.c @@ -93,8 +93,9 @@ static int __devinit ata66_jmicron(ide_hwif_t *hwif) return 0; return 1; case PORT_SATA: - return 1; + break; } + return 1; /* Avoid bogus "control reaches end of non-void function" */ } static void jmicron_tuneproc (ide_drive_t *drive, byte mode_wanted) From 737153298023342073ccaa006144dd254b298e2b Mon Sep 17 00:00:00 2001 From: Conke Hu Date: Sat, 27 Jan 2007 13:46:28 +0100 Subject: [PATCH 194/326] atiixp.c: remove unused code A previous patch to atiixp.c was removed but some code has not been cleaned. Now we remove these code sine they are no use any longer. Signed-off-by: Conke Hu Cc: Alan Cox Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Bartlomiej Zolnierkiewicz --- drivers/ide/pci/atiixp.c | 21 +-------------------- 1 file changed, 1 insertion(+), 20 deletions(-) diff --git a/drivers/ide/pci/atiixp.c b/drivers/ide/pci/atiixp.c index 524e65de4398..148af8ab518f 100644 --- a/drivers/ide/pci/atiixp.c +++ b/drivers/ide/pci/atiixp.c @@ -320,19 +320,6 @@ static void __devinit init_hwif_atiixp(ide_hwif_t *hwif) hwif->drives[0].autodma = hwif->autodma; } -static void __devinit init_hwif_sb600_legacy(ide_hwif_t *hwif) -{ - - hwif->atapi_dma = 1; - hwif->ultra_mask = 0x7f; - hwif->mwdma_mask = 0x07; - hwif->swdma_mask = 0x07; - - if (!noautodma) - hwif->autodma = 1; - hwif->drives[0].autodma = hwif->autodma; - hwif->drives[1].autodma = hwif->autodma; -} static ide_pci_device_t atiixp_pci_info[] __devinitdata = { { /* 0 */ @@ -342,13 +329,7 @@ static ide_pci_device_t atiixp_pci_info[] __devinitdata = { .autodma = AUTODMA, .enablebits = {{0x48,0x01,0x00}, {0x48,0x08,0x00}}, .bootable = ON_BOARD, - },{ /* 1 */ - .name = "ATI SB600 SATA Legacy IDE", - .init_hwif = init_hwif_sb600_legacy, - .channels = 2, - .autodma = AUTODMA, - .bootable = ON_BOARD, - } + }, }; /** From b25168dfdc162b4198fa6395cd191a20dddc6d34 Mon Sep 17 00:00:00 2001 From: Conke Hu Date: Sat, 27 Jan 2007 13:46:30 +0100 Subject: [PATCH 195/326] atiixp.c: sb600 ide only has one channel AMD/ATI SB600 IDE/PATA controller only has one channel. Signed-off-by: Conke Hu Cc: Alan Cox Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Bartlomiej Zolnierkiewicz --- drivers/ide/pci/atiixp.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/ide/pci/atiixp.c b/drivers/ide/pci/atiixp.c index 148af8ab518f..5b7e000ac2f5 100644 --- a/drivers/ide/pci/atiixp.c +++ b/drivers/ide/pci/atiixp.c @@ -329,7 +329,14 @@ static ide_pci_device_t atiixp_pci_info[] __devinitdata = { .autodma = AUTODMA, .enablebits = {{0x48,0x01,0x00}, {0x48,0x08,0x00}}, .bootable = ON_BOARD, - }, + },{ /* 1 */ + .name = "SB600_PATA", + .init_hwif = init_hwif_atiixp, + .channels = 1, + .autodma = AUTODMA, + .enablebits = {{0x48,0x01,0x00}, {0x00,0x00,0x00}}, + .bootable = ON_BOARD, + }, }; /** @@ -350,7 +357,7 @@ static struct pci_device_id atiixp_pci_tbl[] = { { PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_IXP200_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, { PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_IXP300_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, { PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_IXP400_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, - { PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_IXP600_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, + { PCI_VENDOR_ID_ATI, PCI_DEVICE_ID_ATI_IXP600_IDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 1}, { 0, }, }; MODULE_DEVICE_TABLE(pci, atiixp_pci_tbl); From e5c073ff24604d4dbb2fbcedb17da6df768468d3 Mon Sep 17 00:00:00 2001 From: Conke Hu Date: Sat, 27 Jan 2007 13:46:40 +0100 Subject: [PATCH 196/326] atiixp.c: add cable detection support for ATI IDE IDE HDD does not work if it uses a 40-pin PATA cable on ATI chipset. This patch fixes the bug. Signed-off-by: Conke Hu Cc: Alan Cox Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Bartlomiej Zolnierkiewicz --- drivers/ide/pci/atiixp.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) diff --git a/drivers/ide/pci/atiixp.c b/drivers/ide/pci/atiixp.c index 5b7e000ac2f5..7e1d07031d32 100644 --- a/drivers/ide/pci/atiixp.c +++ b/drivers/ide/pci/atiixp.c @@ -291,8 +291,12 @@ static int atiixp_dma_check(ide_drive_t *drive) static void __devinit init_hwif_atiixp(ide_hwif_t *hwif) { + u8 udma_mode = 0; + u8 ch = hwif->channel; + struct pci_dev *pdev = hwif->pci_dev; + if (!hwif->irq) - hwif->irq = hwif->channel ? 15 : 14; + hwif->irq = ch ? 15 : 14; hwif->autodma = 0; hwif->tuneproc = &atiixp_tuneproc; @@ -308,8 +312,12 @@ static void __devinit init_hwif_atiixp(ide_hwif_t *hwif) hwif->mwdma_mask = 0x06; hwif->swdma_mask = 0x04; - /* FIXME: proper cable detection needed */ - hwif->udma_four = 1; + pci_read_config_byte(pdev, ATIIXP_IDE_UDMA_MODE + ch, &udma_mode); + if ((udma_mode & 0x07) >= 0x04 || (udma_mode & 0x70) >= 0x40) + hwif->udma_four = 1; + else + hwif->udma_four = 0; + hwif->ide_dma_host_on = &atiixp_ide_dma_host_on; hwif->ide_dma_host_off = &atiixp_ide_dma_host_off; hwif->ide_dma_check = &atiixp_dma_check; From 3e9e4c8606127592cda22159cc2440ea48963ae4 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Sat, 27 Jan 2007 13:46:45 +0100 Subject: [PATCH 197/326] ide/generic: Jmicron has its own drivers now Drop ide-generic support for Jmicron identifiers as we now trust Jmicron.c for this with drivers/ide. The code check remains for the all-generic-ide case. Signed-off-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Bartlomiej Zolnierkiewicz --- drivers/ide/pci/generic.c | 35 ----------------------------------- 1 file changed, 35 deletions(-) diff --git a/drivers/ide/pci/generic.c b/drivers/ide/pci/generic.c index 9f306880491a..3143cb03fe3c 100644 --- a/drivers/ide/pci/generic.c +++ b/drivers/ide/pci/generic.c @@ -185,36 +185,6 @@ static ide_pci_device_t generic_chipsets[] __devinitdata = { .channels = 2, .autodma = AUTODMA, .bootable = OFF_BOARD, - },{ /* 15 */ - .name = "JMB361", - .init_hwif = init_hwif_generic, - .channels = 2, - .autodma = AUTODMA, - .bootable = OFF_BOARD, - },{ /* 16 */ - .name = "JMB363", - .init_hwif = init_hwif_generic, - .channels = 2, - .autodma = AUTODMA, - .bootable = OFF_BOARD, - },{ /* 17 */ - .name = "JMB365", - .init_hwif = init_hwif_generic, - .channels = 2, - .autodma = AUTODMA, - .bootable = OFF_BOARD, - },{ /* 18 */ - .name = "JMB366", - .init_hwif = init_hwif_generic, - .channels = 2, - .autodma = AUTODMA, - .bootable = OFF_BOARD, - },{ /* 19 */ - .name = "JMB368", - .init_hwif = init_hwif_generic, - .channels = 2, - .autodma = AUTODMA, - .bootable = OFF_BOARD, } }; @@ -281,11 +251,6 @@ static struct pci_device_id generic_pci_tbl[] = { { PCI_VENDOR_ID_TOSHIBA,PCI_DEVICE_ID_TOSHIBA_PICCOLO_1, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 12}, { PCI_VENDOR_ID_TOSHIBA,PCI_DEVICE_ID_TOSHIBA_PICCOLO_2, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 13}, { PCI_VENDOR_ID_NETCELL,PCI_DEVICE_ID_REVOLUTION, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 14}, - { PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB361, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 15}, - { PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB363, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 16}, - { PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB365, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 17}, - { PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB366, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 18}, - { PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB368, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 19}, /* Must come last. If you add entries adjust this table appropriately and the init_one code */ { PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_STORAGE_IDE << 8, 0xFFFFFF00UL, 0}, { 0, }, From 677c0a78f6bf7aa87c90c4e015e014a17132c634 Mon Sep 17 00:00:00 2001 From: Bartlomiej Zolnierkiewicz Date: Sat, 27 Jan 2007 13:46:54 +0100 Subject: [PATCH 198/326] ia64: add pci_get_legacy_ide_irq() Add pci_get_legacy_ide_irq() identical to the one used by i386/x86_64. Fixes amd74xx driver build on ia64 (bugzilla bug #6644). Signed-off-by: Bartlomiej Zolnierkiewicz --- include/asm-ia64/pci.h | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/include/asm-ia64/pci.h b/include/asm-ia64/pci.h index 556f53fa44cb..5160233bbfac 100644 --- a/include/asm-ia64/pci.h +++ b/include/asm-ia64/pci.h @@ -167,4 +167,10 @@ pcibios_select_root(struct pci_dev *pdev, struct resource *res) #define pcibios_scan_all_fns(a, b) 0 +#define HAVE_ARCH_PCI_GET_LEGACY_IDE_IRQ +static inline int pci_get_legacy_ide_irq(struct pci_dev *dev, int channel) +{ + return channel ? 15 : 14; +} + #endif /* _ASM_IA64_PCI_H */ From 82ab1eeceba6705cd5a8815c48eb03af1dada744 Mon Sep 17 00:00:00 2001 From: Bartlomiej Zolnierkiewicz Date: Sat, 27 Jan 2007 13:46:56 +0100 Subject: [PATCH 199/326] ide: add missing __init tags to IDE PCI host drivers also change __devinit tag for sgiioc4.c:ioc4_ide_init() to __init Signed-off-by: Bartlomiej Zolnierkiewicz --- drivers/ide/pci/aec62xx.c | 2 +- drivers/ide/pci/alim15x3.c | 2 +- drivers/ide/pci/amd74xx.c | 2 +- drivers/ide/pci/atiixp.c | 2 +- drivers/ide/pci/cmd64x.c | 2 +- drivers/ide/pci/cs5520.c | 2 +- drivers/ide/pci/cs5530.c | 2 +- drivers/ide/pci/cy82c693.c | 2 +- drivers/ide/pci/generic.c | 2 +- drivers/ide/pci/hpt34x.c | 2 +- drivers/ide/pci/hpt366.c | 2 +- drivers/ide/pci/ns87415.c | 2 +- drivers/ide/pci/opti621.c | 2 +- drivers/ide/pci/pdc202xx_new.c | 2 +- drivers/ide/pci/pdc202xx_old.c | 2 +- drivers/ide/pci/rz1000.c | 2 +- drivers/ide/pci/sc1200.c | 2 +- drivers/ide/pci/serverworks.c | 2 +- drivers/ide/pci/sgiioc4.c | 3 +-- drivers/ide/pci/siimage.c | 2 +- drivers/ide/pci/sis5513.c | 2 +- drivers/ide/pci/sl82c105.c | 2 +- drivers/ide/pci/slc90e66.c | 2 +- drivers/ide/pci/triflex.c | 2 +- drivers/ide/pci/trm290.c | 2 +- drivers/ide/pci/via82cxxx.c | 2 +- 26 files changed, 26 insertions(+), 27 deletions(-) diff --git a/drivers/ide/pci/aec62xx.c b/drivers/ide/pci/aec62xx.c index f286079d233f..d261bfbad222 100644 --- a/drivers/ide/pci/aec62xx.c +++ b/drivers/ide/pci/aec62xx.c @@ -441,7 +441,7 @@ static struct pci_driver driver = { .probe = aec62xx_init_one, }; -static int aec62xx_ide_init(void) +static int __init aec62xx_ide_init(void) { return ide_pci_register_driver(&driver); } diff --git a/drivers/ide/pci/alim15x3.c b/drivers/ide/pci/alim15x3.c index 89109be5162c..68df77ec502b 100644 --- a/drivers/ide/pci/alim15x3.c +++ b/drivers/ide/pci/alim15x3.c @@ -907,7 +907,7 @@ static struct pci_driver driver = { .probe = alim15x3_init_one, }; -static int ali15x3_ide_init(void) +static int __init ali15x3_ide_init(void) { return ide_pci_register_driver(&driver); } diff --git a/drivers/ide/pci/amd74xx.c b/drivers/ide/pci/amd74xx.c index 753fe0e21456..a4336995a410 100644 --- a/drivers/ide/pci/amd74xx.c +++ b/drivers/ide/pci/amd74xx.c @@ -544,7 +544,7 @@ static struct pci_driver driver = { .probe = amd74xx_probe, }; -static int amd74xx_ide_init(void) +static int __init amd74xx_ide_init(void) { return ide_pci_register_driver(&driver); } diff --git a/drivers/ide/pci/atiixp.c b/drivers/ide/pci/atiixp.c index 7e1d07031d32..982ac31fa995 100644 --- a/drivers/ide/pci/atiixp.c +++ b/drivers/ide/pci/atiixp.c @@ -376,7 +376,7 @@ static struct pci_driver driver = { .probe = atiixp_init_one, }; -static int atiixp_ide_init(void) +static int __init atiixp_ide_init(void) { return ide_pci_register_driver(&driver); } diff --git a/drivers/ide/pci/cmd64x.c b/drivers/ide/pci/cmd64x.c index 20c32716bbc4..aee947e8fc38 100644 --- a/drivers/ide/pci/cmd64x.c +++ b/drivers/ide/pci/cmd64x.c @@ -793,7 +793,7 @@ static struct pci_driver driver = { .probe = cmd64x_init_one, }; -static int cmd64x_ide_init(void) +static int __init cmd64x_ide_init(void) { return ide_pci_register_driver(&driver); } diff --git a/drivers/ide/pci/cs5520.c b/drivers/ide/pci/cs5520.c index 079f7c86726b..ba6786aabf3b 100644 --- a/drivers/ide/pci/cs5520.c +++ b/drivers/ide/pci/cs5520.c @@ -260,7 +260,7 @@ static struct pci_driver driver = { .probe = cs5520_init_one, }; -static int cs5520_ide_init(void) +static int __init cs5520_ide_init(void) { return ide_pci_register_driver(&driver); } diff --git a/drivers/ide/pci/cs5530.c b/drivers/ide/pci/cs5530.c index ae405fa32236..9bf5fdfc5b1f 100644 --- a/drivers/ide/pci/cs5530.c +++ b/drivers/ide/pci/cs5530.c @@ -374,7 +374,7 @@ static struct pci_driver driver = { .probe = cs5530_init_one, }; -static int cs5530_ide_init(void) +static int __init cs5530_ide_init(void) { return ide_pci_register_driver(&driver); } diff --git a/drivers/ide/pci/cy82c693.c b/drivers/ide/pci/cy82c693.c index 64330c459bd4..9eafcbf444f4 100644 --- a/drivers/ide/pci/cy82c693.c +++ b/drivers/ide/pci/cy82c693.c @@ -519,7 +519,7 @@ static struct pci_driver driver = { .probe = cy82c693_init_one, }; -static int cy82c693_ide_init(void) +static int __init cy82c693_ide_init(void) { return ide_pci_register_driver(&driver); } diff --git a/drivers/ide/pci/generic.c b/drivers/ide/pci/generic.c index 3143cb03fe3c..b408c6c517ea 100644 --- a/drivers/ide/pci/generic.c +++ b/drivers/ide/pci/generic.c @@ -263,7 +263,7 @@ static struct pci_driver driver = { .probe = generic_init_one, }; -static int generic_ide_init(void) +static int __init generic_ide_init(void) { return ide_pci_register_driver(&driver); } diff --git a/drivers/ide/pci/hpt34x.c b/drivers/ide/pci/hpt34x.c index b46cb042290a..ce7b08f08a09 100644 --- a/drivers/ide/pci/hpt34x.c +++ b/drivers/ide/pci/hpt34x.c @@ -265,7 +265,7 @@ static struct pci_driver driver = { .probe = hpt34x_init_one, }; -static int hpt34x_ide_init(void) +static int __init hpt34x_ide_init(void) { return ide_pci_register_driver(&driver); } diff --git a/drivers/ide/pci/hpt366.c b/drivers/ide/pci/hpt366.c index 08119da06d54..b486442dd5d7 100644 --- a/drivers/ide/pci/hpt366.c +++ b/drivers/ide/pci/hpt366.c @@ -1613,7 +1613,7 @@ static struct pci_driver driver = { .probe = hpt366_init_one, }; -static int hpt366_ide_init(void) +static int __init hpt366_ide_init(void) { return ide_pci_register_driver(&driver); } diff --git a/drivers/ide/pci/ns87415.c b/drivers/ide/pci/ns87415.c index d95714bcee4e..8aaea4ea5549 100644 --- a/drivers/ide/pci/ns87415.c +++ b/drivers/ide/pci/ns87415.c @@ -302,7 +302,7 @@ static struct pci_driver driver = { .probe = ns87415_init_one, }; -static int ns87415_ide_init(void) +static int __init ns87415_ide_init(void) { return ide_pci_register_driver(&driver); } diff --git a/drivers/ide/pci/opti621.c b/drivers/ide/pci/opti621.c index 7a7c2ef78ac2..22bbf613f948 100644 --- a/drivers/ide/pci/opti621.c +++ b/drivers/ide/pci/opti621.c @@ -382,7 +382,7 @@ static struct pci_driver driver = { .probe = opti621_init_one, }; -static int opti621_ide_init(void) +static int __init opti621_ide_init(void) { return ide_pci_register_driver(&driver); } diff --git a/drivers/ide/pci/pdc202xx_new.c b/drivers/ide/pci/pdc202xx_new.c index 7cb48576e479..77a9aaa7dab9 100644 --- a/drivers/ide/pci/pdc202xx_new.c +++ b/drivers/ide/pci/pdc202xx_new.c @@ -756,7 +756,7 @@ static struct pci_driver driver = { .probe = pdc202new_init_one, }; -static int pdc202new_ide_init(void) +static int __init pdc202new_ide_init(void) { return ide_pci_register_driver(&driver); } diff --git a/drivers/ide/pci/pdc202xx_old.c b/drivers/ide/pci/pdc202xx_old.c index 184cdacddeb6..143239c093d5 100644 --- a/drivers/ide/pci/pdc202xx_old.c +++ b/drivers/ide/pci/pdc202xx_old.c @@ -719,7 +719,7 @@ static struct pci_driver driver = { .probe = pdc202xx_init_one, }; -static int pdc202xx_ide_init(void) +static int __init pdc202xx_ide_init(void) { return ide_pci_register_driver(&driver); } diff --git a/drivers/ide/pci/rz1000.c b/drivers/ide/pci/rz1000.c index 5f6950c2d1d1..c1855311052b 100644 --- a/drivers/ide/pci/rz1000.c +++ b/drivers/ide/pci/rz1000.c @@ -77,7 +77,7 @@ static struct pci_driver driver = { .probe = rz1000_init_one, }; -static int rz1000_ide_init(void) +static int __init rz1000_ide_init(void) { return ide_pci_register_driver(&driver); } diff --git a/drivers/ide/pci/sc1200.c b/drivers/ide/pci/sc1200.c index ff80937d94dd..8d762d323f8b 100644 --- a/drivers/ide/pci/sc1200.c +++ b/drivers/ide/pci/sc1200.c @@ -507,7 +507,7 @@ static struct pci_driver driver = { #endif }; -static int sc1200_ide_init(void) +static int __init sc1200_ide_init(void) { return ide_pci_register_driver(&driver); } diff --git a/drivers/ide/pci/serverworks.c b/drivers/ide/pci/serverworks.c index 057548d07205..ea9a28a45853 100644 --- a/drivers/ide/pci/serverworks.c +++ b/drivers/ide/pci/serverworks.c @@ -666,7 +666,7 @@ static struct pci_driver driver = { .probe = svwks_init_one, }; -static int svwks_ide_init(void) +static int __init svwks_ide_init(void) { return ide_pci_register_driver(&driver); } diff --git a/drivers/ide/pci/sgiioc4.c b/drivers/ide/pci/sgiioc4.c index cfad09accf52..b0bf01809279 100644 --- a/drivers/ide/pci/sgiioc4.c +++ b/drivers/ide/pci/sgiioc4.c @@ -762,8 +762,7 @@ static struct ioc4_submodule ioc4_ide_submodule = { /* .is_remove = ioc4_ide_remove_one, */ }; -static int __devinit -ioc4_ide_init(void) +static int __init ioc4_ide_init(void) { return ioc4_register_submodule(&ioc4_ide_submodule); } diff --git a/drivers/ide/pci/siimage.c b/drivers/ide/pci/siimage.c index 697f566fb90a..4ff89c7d990a 100644 --- a/drivers/ide/pci/siimage.c +++ b/drivers/ide/pci/siimage.c @@ -1096,7 +1096,7 @@ static struct pci_driver driver = { .probe = siimage_init_one, }; -static int siimage_ide_init(void) +static int __init siimage_ide_init(void) { return ide_pci_register_driver(&driver); } diff --git a/drivers/ide/pci/sis5513.c b/drivers/ide/pci/sis5513.c index 6b313139b5e4..1afff659ab55 100644 --- a/drivers/ide/pci/sis5513.c +++ b/drivers/ide/pci/sis5513.c @@ -968,7 +968,7 @@ static struct pci_driver driver = { .probe = sis5513_init_one, }; -static int sis5513_ide_init(void) +static int __init sis5513_ide_init(void) { return ide_pci_register_driver(&driver); } diff --git a/drivers/ide/pci/sl82c105.c b/drivers/ide/pci/sl82c105.c index 5afefe8692fe..170a26199050 100644 --- a/drivers/ide/pci/sl82c105.c +++ b/drivers/ide/pci/sl82c105.c @@ -492,7 +492,7 @@ static struct pci_driver driver = { .probe = sl82c105_init_one, }; -static int sl82c105_ide_init(void) +static int __init sl82c105_ide_init(void) { return ide_pci_register_driver(&driver); } diff --git a/drivers/ide/pci/slc90e66.c b/drivers/ide/pci/slc90e66.c index 9be7e49cba0e..90e79c0844d2 100644 --- a/drivers/ide/pci/slc90e66.c +++ b/drivers/ide/pci/slc90e66.c @@ -253,7 +253,7 @@ static struct pci_driver driver = { .probe = slc90e66_init_one, }; -static int slc90e66_ide_init(void) +static int __init slc90e66_ide_init(void) { return ide_pci_register_driver(&driver); } diff --git a/drivers/ide/pci/triflex.c b/drivers/ide/pci/triflex.c index 56d84931d6de..b13cce1fd1a6 100644 --- a/drivers/ide/pci/triflex.c +++ b/drivers/ide/pci/triflex.c @@ -173,7 +173,7 @@ static struct pci_driver driver = { .probe = triflex_init_one, }; -static int triflex_ide_init(void) +static int __init triflex_ide_init(void) { return ide_pci_register_driver(&driver); } diff --git a/drivers/ide/pci/trm290.c b/drivers/ide/pci/trm290.c index 2a282529bfc1..174b88c4780e 100644 --- a/drivers/ide/pci/trm290.c +++ b/drivers/ide/pci/trm290.c @@ -355,7 +355,7 @@ static struct pci_driver driver = { .probe = trm290_init_one, }; -static int trm290_ide_init(void) +static int __init trm290_ide_init(void) { return ide_pci_register_driver(&driver); } diff --git a/drivers/ide/pci/via82cxxx.c b/drivers/ide/pci/via82cxxx.c index 381cc6f101ce..6414a545be00 100644 --- a/drivers/ide/pci/via82cxxx.c +++ b/drivers/ide/pci/via82cxxx.c @@ -514,7 +514,7 @@ static struct pci_driver driver = { .probe = via_init_one, }; -static int via_ide_init(void) +static int __init via_ide_init(void) { return ide_pci_register_driver(&driver); } From 6855036aa035913bc2bfb31c41576a49f42ecd5f Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Sat, 27 Jan 2007 13:47:02 +0100 Subject: [PATCH 200/326] ide: unregister idepnp driver on unload idepnp driver is registered as a pnp driver on ide init but doesn't get unregistered on ide unload causing driver list corruption and eventually oops. Fix it. Signed-off-by: Tejun Heo Signed-off-by: Bartlomiej Zolnierkiewicz --- drivers/ide/ide-pnp.c | 5 +++++ drivers/ide/ide.c | 5 +++++ 2 files changed, 10 insertions(+) diff --git a/drivers/ide/ide-pnp.c b/drivers/ide/ide-pnp.c index df7d1504f84e..98410ca044cf 100644 --- a/drivers/ide/ide-pnp.c +++ b/drivers/ide/ide-pnp.c @@ -73,3 +73,8 @@ void __init pnpide_init(void) { pnp_register_driver(&idepnp_driver); } + +void __exit pnpide_exit(void) +{ + pnp_unregister_driver(&idepnp_driver); +} diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c index 16890769dca6..3b334af0c7b9 100644 --- a/drivers/ide/ide.c +++ b/drivers/ide/ide.c @@ -1782,6 +1782,7 @@ static int __init ide_setup(char *s) } extern void pnpide_init(void); +extern void pnpide_exit(void); extern void h8300_ide_init(void); /* @@ -2094,6 +2095,10 @@ void cleanup_module (void) for (index = 0; index < MAX_HWIFS; ++index) ide_unregister(index); +#ifdef CONFIG_BLK_DEV_IDEPNP + pnpide_exit(); +#endif + #ifdef CONFIG_PROC_FS proc_ide_destroy(); #endif From e0b874df14052489e6408125903dba96b4dd7baa Mon Sep 17 00:00:00 2001 From: Josepch Chan Date: Sat, 27 Jan 2007 13:47:08 +0100 Subject: [PATCH 201/326] via82cxxx/pata_via: correct PCI_DEVICE_ID_VIA_SATA_EIDE ID and add support for CX700 and 8237S This patch: * Corrects the wrong device ID of PCI_DEVICE_ID_VIA_SATA_EIDE from 0x0581 to 0x5324. * Adds VIA CX700 and VT8237S support in drivers/ide/pci/via82cxxx.c * Adds VIA VT8237S support in drivers/ata/pata_via.c Signed-off-by: Josepch Chan Signed-off-by: Bartlomiej Zolnierkiewicz --- drivers/ata/pata_via.c | 1 + drivers/ide/pci/via82cxxx.c | 3 +++ include/linux/pci_ids.h | 3 ++- 3 files changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/ata/pata_via.c b/drivers/ata/pata_via.c index f0d4f7e9ed31..0219419cae97 100644 --- a/drivers/ata/pata_via.c +++ b/drivers/ata/pata_via.c @@ -95,6 +95,7 @@ static const struct via_isa_bridge { u8 rev_max; u16 flags; } via_isa_bridges[] = { + { "vt8237s", PCI_DEVICE_ID_VIA_8237S, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, { "vt8251", PCI_DEVICE_ID_VIA_8251, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, { "cx700", PCI_DEVICE_ID_VIA_CX700, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, { "vt6410", PCI_DEVICE_ID_VIA_6410, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST | VIA_NO_ENABLES}, diff --git a/drivers/ide/pci/via82cxxx.c b/drivers/ide/pci/via82cxxx.c index 6414a545be00..a98b4d38b9dd 100644 --- a/drivers/ide/pci/via82cxxx.c +++ b/drivers/ide/pci/via82cxxx.c @@ -78,6 +78,8 @@ static struct via_isa_bridge { u8 rev_max; u16 flags; } via_isa_bridges[] = { + { "cx7000", PCI_DEVICE_ID_VIA_CX700, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, + { "vt8237s", PCI_DEVICE_ID_VIA_8237S, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, { "vt6410", PCI_DEVICE_ID_VIA_6410, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, { "vt8251", PCI_DEVICE_ID_VIA_8251, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, { "vt8237", PCI_DEVICE_ID_VIA_8237, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, @@ -504,6 +506,7 @@ static struct pci_device_id via_pci_tbl[] = { { PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C576_1, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, { PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C586_1, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, { PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_6410, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 1}, + { PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_SATA_EIDE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 1}, { 0, }, }; MODULE_DEVICE_TABLE(pci, via_pci_tbl); diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index f7a416c52edc..e642b99f9a92 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1283,7 +1283,6 @@ #define PCI_DEVICE_ID_VIA_82C561 0x0561 #define PCI_DEVICE_ID_VIA_82C586_1 0x0571 #define PCI_DEVICE_ID_VIA_82C576 0x0576 -#define PCI_DEVICE_ID_VIA_SATA_EIDE 0x0581 #define PCI_DEVICE_ID_VIA_82C586_0 0x0586 #define PCI_DEVICE_ID_VIA_82C596 0x0596 #define PCI_DEVICE_ID_VIA_82C597_0 0x0597 @@ -1326,6 +1325,8 @@ #define PCI_DEVICE_ID_VIA_8237 0x3227 #define PCI_DEVICE_ID_VIA_8251 0x3287 #define PCI_DEVICE_ID_VIA_8237A 0x3337 +#define PCI_DEVICE_ID_VIA_8237S 0x3372 +#define PCI_DEVICE_ID_VIA_SATA_EIDE 0x5324 #define PCI_DEVICE_ID_VIA_8231 0x8231 #define PCI_DEVICE_ID_VIA_8231_4 0x8235 #define PCI_DEVICE_ID_VIA_8365_1 0x8305 From 938e2ac0b7ac72d264783b0b548eb6078c295294 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Mon, 15 Jan 2007 18:07:09 -0700 Subject: [PATCH 202/326] [SCSI] Fix scsi_add_device() for async scanning I had thought that all drivers which didn't call scsi_scan_host() called scsi_scan_target(). Some, such as sbp2, mptsas and libata-scsi, call scsi_add_device() or __scsi_add_device(). We just need to wait for the currently executing async scans to complete first. This is the same code that's in scsi_scan_target(), except that we have to return an error instead of void when we're declining to scan at all. Signed-off-by: Matthew Wilcox Signed-off-by: James Bottomley --- drivers/scsi/scsi_scan.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c index b83d03c4deef..96b7cbd746a8 100644 --- a/drivers/scsi/scsi_scan.c +++ b/drivers/scsi/scsi_scan.c @@ -1453,6 +1453,12 @@ struct scsi_device *__scsi_add_device(struct Scsi_Host *shost, uint channel, struct device *parent = &shost->shost_gendev; struct scsi_target *starget; + if (strncmp(scsi_scan_type, "none", 4) == 0) + return ERR_PTR(-ENODEV); + + if (!shost->async_scan) + scsi_complete_async_scans(); + starget = scsi_alloc_target(parent, channel, id); if (!starget) return ERR_PTR(-ENOMEM); From 477ffb9d8732f30e7ab2d20f6ed0c22bad37a4a5 Mon Sep 17 00:00:00 2001 From: David C Somayajulu Date: Mon, 22 Jan 2007 12:26:11 -0800 Subject: [PATCH 203/326] [SCSI] qla4xxx: bug fixes The included patch fixes the following issues: 1. qla3xxx/qla4xxx co-existence issue which can result in a lockup when qla3xxx driver is unloaded, or when ifdown; ifup is performed on one of the interfaces correponding to qla3xxx. This is because qla4xxx HBA supports one ethernet and iscsi interfaces per port. Both iscsi and ethernet interfaces share the same state machine. The problem has to do with synchronizing access to the state machine in the event of a reset 2. mutex_lock() is sometimes not followed by mutex_unlock() prior to invoking a msleep() in qla4xxx_mailbox_command() Signed-off-by: James Bottomley --- drivers/scsi/qla4xxx/ql4_def.h | 1 - drivers/scsi/qla4xxx/ql4_glbl.h | 1 + drivers/scsi/qla4xxx/ql4_init.c | 18 ++++----- drivers/scsi/qla4xxx/ql4_isr.c | 4 +- drivers/scsi/qla4xxx/ql4_mbx.c | 35 +++++++++------- drivers/scsi/qla4xxx/ql4_os.c | 64 ++++++++++++++++++------------ drivers/scsi/qla4xxx/ql4_version.h | 2 +- 7 files changed, 73 insertions(+), 52 deletions(-) diff --git a/drivers/scsi/qla4xxx/ql4_def.h b/drivers/scsi/qla4xxx/ql4_def.h index 4249e52a5592..6f4cf2dd2f4a 100644 --- a/drivers/scsi/qla4xxx/ql4_def.h +++ b/drivers/scsi/qla4xxx/ql4_def.h @@ -418,7 +418,6 @@ struct scsi_qla_host { * concurrently. */ struct mutex mbox_sem; - wait_queue_head_t mailbox_wait_queue; /* temporary mailbox status registers */ volatile uint8_t mbox_status_count; diff --git a/drivers/scsi/qla4xxx/ql4_glbl.h b/drivers/scsi/qla4xxx/ql4_glbl.h index 2122967bbf0b..e021eb5db2b2 100644 --- a/drivers/scsi/qla4xxx/ql4_glbl.h +++ b/drivers/scsi/qla4xxx/ql4_glbl.h @@ -76,4 +76,5 @@ int qla4xxx_process_ddb_changed(struct scsi_qla_host * ha, extern int ql4xextended_error_logging; extern int ql4xdiscoverywait; extern int ql4xdontresethba; +extern int ql4_mod_unload; #endif /* _QLA4x_GBL_H */ diff --git a/drivers/scsi/qla4xxx/ql4_init.c b/drivers/scsi/qla4xxx/ql4_init.c index cc210f297a78..b907b06d72ab 100644 --- a/drivers/scsi/qla4xxx/ql4_init.c +++ b/drivers/scsi/qla4xxx/ql4_init.c @@ -958,25 +958,25 @@ static int qla4xxx_start_firmware_from_flash(struct scsi_qla_host *ha) return status; } -int ql4xxx_lock_drvr_wait(struct scsi_qla_host *a) +int ql4xxx_lock_drvr_wait(struct scsi_qla_host *ha) { -#define QL4_LOCK_DRVR_WAIT 300 -#define QL4_LOCK_DRVR_SLEEP 100 +#define QL4_LOCK_DRVR_WAIT 30 +#define QL4_LOCK_DRVR_SLEEP 1 int drvr_wait = QL4_LOCK_DRVR_WAIT; while (drvr_wait) { - if (ql4xxx_lock_drvr(a) == 0) { - msleep(QL4_LOCK_DRVR_SLEEP); + if (ql4xxx_lock_drvr(ha) == 0) { + ssleep(QL4_LOCK_DRVR_SLEEP); if (drvr_wait) { DEBUG2(printk("scsi%ld: %s: Waiting for " - "Global Init Semaphore...n", - a->host_no, - __func__)); + "Global Init Semaphore(%d)...n", + ha->host_no, + __func__, drvr_wait)); } drvr_wait -= QL4_LOCK_DRVR_SLEEP; } else { DEBUG2(printk("scsi%ld: %s: Global Init Semaphore " - "acquired.n", a->host_no, __func__)); + "acquired.n", ha->host_no, __func__)); return QLA_SUCCESS; } } diff --git a/drivers/scsi/qla4xxx/ql4_isr.c b/drivers/scsi/qla4xxx/ql4_isr.c index ef975e0dc87f..35b9e36a0e8d 100644 --- a/drivers/scsi/qla4xxx/ql4_isr.c +++ b/drivers/scsi/qla4xxx/ql4_isr.c @@ -433,7 +433,6 @@ static void qla4xxx_isr_decode_mailbox(struct scsi_qla_host * ha, readl(&ha->reg->mailbox[i]); set_bit(AF_MBOX_COMMAND_DONE, &ha->flags); - wake_up(&ha->mailbox_wait_queue); } } else if (mbox_status >> 12 == MBOX_ASYNC_EVENT_STATUS) { /* Immediately process the AENs that don't require much work. @@ -686,7 +685,8 @@ irqreturn_t qla4xxx_intr_handler(int irq, void *dev_id) &ha->reg->ctrl_status); readl(&ha->reg->ctrl_status); - set_bit(DPC_RESET_HA_INTR, &ha->dpc_flags); + if (!ql4_mod_unload) + set_bit(DPC_RESET_HA_INTR, &ha->dpc_flags); break; } else if (intr_status & INTR_PENDING) { diff --git a/drivers/scsi/qla4xxx/ql4_mbx.c b/drivers/scsi/qla4xxx/ql4_mbx.c index b721dc5dd711..7f28657eef3f 100644 --- a/drivers/scsi/qla4xxx/ql4_mbx.c +++ b/drivers/scsi/qla4xxx/ql4_mbx.c @@ -29,18 +29,30 @@ int qla4xxx_mailbox_command(struct scsi_qla_host *ha, uint8_t inCount, u_long wait_count; uint32_t intr_status; unsigned long flags = 0; - DECLARE_WAITQUEUE(wait, current); - - mutex_lock(&ha->mbox_sem); - - /* Mailbox code active */ - set_bit(AF_MBOX_COMMAND, &ha->flags); /* Make sure that pointers are valid */ if (!mbx_cmd || !mbx_sts) { DEBUG2(printk("scsi%ld: %s: Invalid mbx_cmd or mbx_sts " "pointer\n", ha->host_no, __func__)); - goto mbox_exit; + return status; + } + /* Mailbox code active */ + wait_count = MBOX_TOV * 100; + + while (wait_count--) { + mutex_lock(&ha->mbox_sem); + if (!test_bit(AF_MBOX_COMMAND, &ha->flags)) { + set_bit(AF_MBOX_COMMAND, &ha->flags); + mutex_unlock(&ha->mbox_sem); + break; + } + mutex_unlock(&ha->mbox_sem); + if (!wait_count) { + DEBUG2(printk("scsi%ld: %s: mbox_sem failed\n", + ha->host_no, __func__)); + return status; + } + msleep(10); } /* To prevent overwriting mailbox registers for a command that has @@ -73,8 +85,6 @@ int qla4xxx_mailbox_command(struct scsi_qla_host *ha, uint8_t inCount, spin_unlock_irqrestore(&ha->hardware_lock, flags); /* Wait for completion */ - set_current_state(TASK_UNINTERRUPTIBLE); - add_wait_queue(&ha->mailbox_wait_queue, &wait); /* * If we don't want status, don't wait for the mailbox command to @@ -83,8 +93,6 @@ int qla4xxx_mailbox_command(struct scsi_qla_host *ha, uint8_t inCount, */ if (outCount == 0) { status = QLA_SUCCESS; - set_current_state(TASK_RUNNING); - remove_wait_queue(&ha->mailbox_wait_queue, &wait); goto mbox_exit; } /* Wait for command to complete */ @@ -108,8 +116,6 @@ int qla4xxx_mailbox_command(struct scsi_qla_host *ha, uint8_t inCount, spin_unlock_irqrestore(&ha->hardware_lock, flags); msleep(10); } - set_current_state(TASK_RUNNING); - remove_wait_queue(&ha->mailbox_wait_queue, &wait); /* Check for mailbox timeout. */ if (!test_bit(AF_MBOX_COMMAND_DONE, &ha->flags)) { @@ -155,9 +161,10 @@ int qla4xxx_mailbox_command(struct scsi_qla_host *ha, uint8_t inCount, spin_unlock_irqrestore(&ha->hardware_lock, flags); mbox_exit: + mutex_lock(&ha->mbox_sem); clear_bit(AF_MBOX_COMMAND, &ha->flags); - clear_bit(AF_MBOX_COMMAND_DONE, &ha->flags); mutex_unlock(&ha->mbox_sem); + clear_bit(AF_MBOX_COMMAND_DONE, &ha->flags); return status; } diff --git a/drivers/scsi/qla4xxx/ql4_os.c b/drivers/scsi/qla4xxx/ql4_os.c index 9ef693c8809a..81fb7bd44f01 100644 --- a/drivers/scsi/qla4xxx/ql4_os.c +++ b/drivers/scsi/qla4xxx/ql4_os.c @@ -40,6 +40,8 @@ MODULE_PARM_DESC(ql4xextended_error_logging, "Option to enable extended error logging, " "Default is 0 - no logging, 1 - debug logging"); +int ql4_mod_unload = 0; + /* * SCSI host template entry points */ @@ -422,6 +424,9 @@ static int qla4xxx_queuecommand(struct scsi_cmnd *cmd, goto qc_host_busy; } + if (test_bit(DPC_RESET_HA_INTR, &ha->dpc_flags)) + goto qc_host_busy; + spin_unlock_irq(ha->host->host_lock); srb = qla4xxx_get_new_srb(ha, ddb_entry, cmd, done); @@ -707,16 +712,12 @@ static int qla4xxx_cmd_wait(struct scsi_qla_host *ha) return stat; } -/** - * qla4xxx_soft_reset - performs soft reset. - * @ha: Pointer to host adapter structure. - **/ -int qla4xxx_soft_reset(struct scsi_qla_host *ha) +static void qla4xxx_hw_reset(struct scsi_qla_host *ha) { - uint32_t max_wait_time; - unsigned long flags = 0; - int status = QLA_ERROR; uint32_t ctrl_status; + unsigned long flags = 0; + + DEBUG2(printk(KERN_ERR "scsi%ld: %s\n", ha->host_no, __func__)); spin_lock_irqsave(&ha->hardware_lock, flags); @@ -733,6 +734,20 @@ int qla4xxx_soft_reset(struct scsi_qla_host *ha) readl(&ha->reg->ctrl_status); spin_unlock_irqrestore(&ha->hardware_lock, flags); +} + +/** + * qla4xxx_soft_reset - performs soft reset. + * @ha: Pointer to host adapter structure. + **/ +int qla4xxx_soft_reset(struct scsi_qla_host *ha) +{ + uint32_t max_wait_time; + unsigned long flags = 0; + int status = QLA_ERROR; + uint32_t ctrl_status; + + qla4xxx_hw_reset(ha); /* Wait until the Network Reset Intr bit is cleared */ max_wait_time = RESET_INTR_TOV; @@ -966,10 +981,12 @@ static void qla4xxx_do_dpc(struct work_struct *work) struct scsi_qla_host *ha = container_of(work, struct scsi_qla_host, dpc_work); struct ddb_entry *ddb_entry, *dtemp; + int status = QLA_ERROR; DEBUG2(printk("scsi%ld: %s: DPC handler waking up." - "flags = 0x%08lx, dpc_flags = 0x%08lx\n", - ha->host_no, __func__, ha->flags, ha->dpc_flags)); + "flags = 0x%08lx, dpc_flags = 0x%08lx ctrl_stat = 0x%08x\n", + ha->host_no, __func__, ha->flags, ha->dpc_flags, + readw(&ha->reg->ctrl_status))); /* Initialization not yet finished. Don't do anything yet. */ if (!test_bit(AF_INIT_DONE, &ha->flags)) @@ -983,31 +1000,28 @@ static void qla4xxx_do_dpc(struct work_struct *work) test_bit(DPC_RESET_HA, &ha->dpc_flags)) qla4xxx_recover_adapter(ha, PRESERVE_DDB_LIST); - if (test_and_clear_bit(DPC_RESET_HA_INTR, &ha->dpc_flags)) { + if (test_bit(DPC_RESET_HA_INTR, &ha->dpc_flags)) { uint8_t wait_time = RESET_INTR_TOV; - unsigned long flags = 0; - qla4xxx_flush_active_srbs(ha); - - spin_lock_irqsave(&ha->hardware_lock, flags); while ((readw(&ha->reg->ctrl_status) & (CSR_SOFT_RESET | CSR_FORCE_SOFT_RESET)) != 0) { if (--wait_time == 0) break; - - spin_unlock_irqrestore(&ha->hardware_lock, - flags); - msleep(1000); - - spin_lock_irqsave(&ha->hardware_lock, flags); } - spin_unlock_irqrestore(&ha->hardware_lock, flags); - if (wait_time == 0) DEBUG2(printk("scsi%ld: %s: SR|FSR " "bit not cleared-- resetting\n", ha->host_no, __func__)); + qla4xxx_flush_active_srbs(ha); + if (ql4xxx_lock_drvr_wait(ha) == QLA_SUCCESS) { + qla4xxx_process_aen(ha, FLUSH_DDB_CHANGED_AENS); + status = qla4xxx_initialize_adapter(ha, + PRESERVE_DDB_LIST); + } + clear_bit(DPC_RESET_HA_INTR, &ha->dpc_flags); + if (status == QLA_SUCCESS) + qla4xxx_enable_intrs(ha); } } @@ -1062,7 +1076,7 @@ static void qla4xxx_free_adapter(struct scsi_qla_host *ha) /* Issue Soft Reset to put firmware in unknown state */ if (ql4xxx_lock_drvr_wait(ha) == QLA_SUCCESS) - qla4xxx_soft_reset(ha); + qla4xxx_hw_reset(ha); /* Remove timer thread, if present */ if (ha->timer_active) @@ -1198,7 +1212,6 @@ static int __devinit qla4xxx_probe_adapter(struct pci_dev *pdev, INIT_LIST_HEAD(&ha->free_srb_q); mutex_init(&ha->mbox_sem); - init_waitqueue_head(&ha->mailbox_wait_queue); spin_lock_init(&ha->hardware_lock); @@ -1665,6 +1678,7 @@ static int __init qla4xxx_module_init(void) static void __exit qla4xxx_module_exit(void) { + ql4_mod_unload = 1; pci_unregister_driver(&qla4xxx_pci_driver); iscsi_unregister_transport(&qla4xxx_iscsi_transport); kmem_cache_destroy(srb_cachep); diff --git a/drivers/scsi/qla4xxx/ql4_version.h b/drivers/scsi/qla4xxx/ql4_version.h index 454e19c8ad68..e5183a697d1f 100644 --- a/drivers/scsi/qla4xxx/ql4_version.h +++ b/drivers/scsi/qla4xxx/ql4_version.h @@ -5,4 +5,4 @@ * See LICENSE.qla4xxx for copyright and licensing details. */ -#define QLA4XXX_DRIVER_VERSION "5.00.07-k" +#define QLA4XXX_DRIVER_VERSION "5.00.07-k1" From 91614c054c9ffc26b47a5cb3135113aa0f6e6ff0 Mon Sep 17 00:00:00 2001 From: Kai Makisara Date: Fri, 26 Jan 2007 00:38:39 +0200 Subject: [PATCH 204/326] [SCSI] st: A MTIOCTOP/MTWEOF within the early warning will cause the file number to be incorrect On Wed, 24 Jan 2007, Andrew Morton wrote: > On Mon, 22 Jan 2007 13:07:20 -0800 > bugme-daemon@bugzilla.kernel.org wrote: > > > http://bugzilla.kernel.org/show_bug.cgi?id=7864 > > > > Summary: A MTIOCTOP/MTWEOF within the early warning will cause > > the file number to be incorrect > > Kernel Version: 2.6.19.2 > > Status: NEW > > Severity: low > > Owner: io_scsi@kernel-bugs.osdl.org > > Submitter: ce_reisinger@yahoo.com > > > > > > Write records to a SCSI tape until a write fails with a ENOSPC (you have reached > > early warning. > > Now perform a: > > struct mtget before, after; > > ioctl(fd, MTIOCGET, &before); > > struct mtop mtop = { MTWEOF, 1 }; > > ioctl(fd, MTIOCTOP, &mtop); > > ioctl(fd, MTIOCGET, &after); > > > > Check the value of mt_fileno in the before and after structures. Notice the > > after is 2 greater then the before. > > > > The problem appears to be in the block of code starting at line 2817 in st.c. > > This block is entered because the drive did return a CHECK CONDITION with NO > > SENSE and the SENSE_EOM bit set. At lines 2824/5 the fileno is incremented. But > > it has already been increased by the number of filemarks requested by the > > MTIOCTOP. I believe that the residue count in the sense data should be > > subtracted from fileno, not a increment as is done. > > > > Thanks. Could you please send us a tested patch to fix these things, as > per http://www.zip.com.au/~akpm/linux/patches/stuff/tpp.txt ? > The analysis is basically correct and explains the bug. According to the SCSI standards, the sense code is NO SENSE or RECOVERED ERROR in case writing filemark(s) succeeds. If it fails (partly or completely) the sense code is VOLUME OVERFLOW. The patch below is tested to fix the case when one filemark is successfully written after the EOM early warning. It should also fix the case at real EOM but this has not been tested. Carl, thanks for reporting the bug and providing the analysis for the fix. Signed-off-by: Kai Makisara Signed-off-by: James Bottomley --- drivers/scsi/st.c | 19 +++++++++++-------- 1 file changed, 11 insertions(+), 8 deletions(-) diff --git a/drivers/scsi/st.c b/drivers/scsi/st.c index e016e0906e1a..488ec7948a57 100644 --- a/drivers/scsi/st.c +++ b/drivers/scsi/st.c @@ -2816,15 +2816,18 @@ static int st_int_ioctl(struct scsi_tape *STp, unsigned int cmd_in, unsigned lon if (cmd_in == MTWEOF && cmdstatp->have_sense && - (cmdstatp->flags & SENSE_EOM) && - (cmdstatp->sense_hdr.sense_key == NO_SENSE || - cmdstatp->sense_hdr.sense_key == RECOVERED_ERROR) && - undone == 0) { - ioctl_result = 0; /* EOF written successfully at EOM */ - if (fileno >= 0) - fileno++; + (cmdstatp->flags & SENSE_EOM)) { + if (cmdstatp->sense_hdr.sense_key == NO_SENSE || + cmdstatp->sense_hdr.sense_key == RECOVERED_ERROR) { + ioctl_result = 0; /* EOF(s) written successfully at EOM */ + STps->eof = ST_NOEOF; + } else { /* Writing EOF(s) failed */ + if (fileno >= 0) + fileno -= undone; + if (undone < arg) + STps->eof = ST_NOEOF; + } STps->drv_file = fileno; - STps->eof = ST_NOEOF; } else if ((cmd_in == MTFSF) || (cmd_in == MTFSFM)) { if (fileno >= 0) STps->drv_file = fileno - undone; From 05916eec9f4d4370ef9a6cbb699f637302f6e157 Mon Sep 17 00:00:00 2001 From: Geoff Levand Date: Thu, 25 Jan 2007 18:07:14 -0800 Subject: [PATCH 205/326] [POWERPC] PS3: add not complete comment to kconfig Add a comment to the PS3 config option to inform users that the current implementation is not yet complete. Signed-off-by: Geoff Levand Signed-off-by: Paul Mackerras --- arch/powerpc/Kconfig | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 0855d55c194d..f4d1f31c14fa 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -525,12 +525,15 @@ config PPC_IBM_CELL_BLADE select UDBG_RTAS_CONSOLE config PPC_PS3 - bool "Sony PS3" + bool "Sony PS3 (incomplete)" depends on PPC_MULTIPLATFORM && PPC64 select PPC_CELL help This option enables support for the Sony PS3 game console and other platforms using the PS3 hypervisor. + Support for this platform is not yet complete, so + enabling this will not result in a bootable kernel on a + PS3 system. config PPC_NATIVE bool From 545da94f924d52f80e2bbea99a8652f454889a2b Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Sun, 28 Jan 2007 07:45:53 +1100 Subject: [PATCH 206/326] [POWERPC] Fix sys_pciconfig_iobase bus matching A stupid bug has been plaguing the sys_pciconfig_iobase on ppc64. It wasn't noticed until recently as it seems to not affect G5s but it's been causing problems running X servers on some other machines recently. The bus number matching was bogus. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Paul Mackerras --- arch/powerpc/kernel/pci_64.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/powerpc/kernel/pci_64.c b/arch/powerpc/kernel/pci_64.c index 73c59ec49120..01f18c683407 100644 --- a/arch/powerpc/kernel/pci_64.c +++ b/arch/powerpc/kernel/pci_64.c @@ -1430,7 +1430,7 @@ long sys_pciconfig_iobase(long which, unsigned long in_bus, for (ln = pci_root_buses.next; ln != &pci_root_buses; ln = ln->next) { bus = pci_bus_b(ln); - if (in_bus >= bus->number && in_bus < (bus->number + bus->subordinate)) + if (in_bus >= bus->number && in_bus <= bus->subordinate) break; bus = NULL; } From c30efbaeaa9297fb1a35ef952350e0c2bb7a3d47 Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Sun, 28 Jan 2007 17:39:19 -0500 Subject: [PATCH 207/326] [AGPGART] Prevent (unlikely) memory leak in amd_create_gatt_pages() If we fail an alloc, unwind the previous allocs that succeeded. Spotted-by: Alan Grimes Signed-off-by: Dave Jones --- drivers/char/agp/amd-k7-agp.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/char/agp/amd-k7-agp.c b/drivers/char/agp/amd-k7-agp.c index 51d0d562d01e..c85c8cadb6df 100644 --- a/drivers/char/agp/amd-k7-agp.c +++ b/drivers/char/agp/amd-k7-agp.c @@ -101,6 +101,11 @@ static int amd_create_gatt_pages(int nr_tables) for (i = 0; i < nr_tables; i++) { entry = kzalloc(sizeof(struct amd_page_map), GFP_KERNEL); if (entry == NULL) { + while (i > 0) { + kfree(tables[i-1]); + i--; + } + kfree(tables); retval = -ENOMEM; break; } From 87a17f31a3bc9bf0c7e7493add19ef200e741248 Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Sun, 28 Jan 2007 17:41:37 -0500 Subject: [PATCH 208/326] [AGPGART] Remove pointless typedef in ati-agp This seems to exist just to save people typing 'struct' a few times, and doesn't provide any additional value. Signed-off-by: Dave Jones --- drivers/char/agp/ati-agp.c | 26 +++++++++++++------------- 1 file changed, 13 insertions(+), 13 deletions(-) diff --git a/drivers/char/agp/ati-agp.c b/drivers/char/agp/ati-agp.c index f244c6682738..0994d3534447 100644 --- a/drivers/char/agp/ati-agp.c +++ b/drivers/char/agp/ati-agp.c @@ -41,18 +41,18 @@ static struct gatt_mask ati_generic_masks[] = }; -typedef struct _ati_page_map { +struct ati_page_map { unsigned long *real; unsigned long __iomem *remapped; -} ati_page_map; +}; static struct _ati_generic_private { volatile u8 __iomem *registers; - ati_page_map **gatt_pages; + struct ati_page_map **gatt_pages; int num_tables; } ati_generic_private; -static int ati_create_page_map(ati_page_map *page_map) +static int ati_create_page_map(struct ati_page_map *page_map) { int i, err = 0; @@ -82,7 +82,7 @@ static int ati_create_page_map(ati_page_map *page_map) } -static void ati_free_page_map(ati_page_map *page_map) +static void ati_free_page_map(struct ati_page_map *page_map) { unmap_page_from_agp(virt_to_page(page_map->real)); iounmap(page_map->remapped); @@ -94,8 +94,8 @@ static void ati_free_page_map(ati_page_map *page_map) static void ati_free_gatt_pages(void) { int i; - ati_page_map **tables; - ati_page_map *entry; + struct ati_page_map **tables; + struct ati_page_map *entry; tables = ati_generic_private.gatt_pages; for (i = 0; i < ati_generic_private.num_tables; i++) { @@ -112,17 +112,17 @@ static void ati_free_gatt_pages(void) static int ati_create_gatt_pages(int nr_tables) { - ati_page_map **tables; - ati_page_map *entry; + struct ati_page_map **tables; + struct ati_page_map *entry; int retval = 0; int i; - tables = kzalloc((nr_tables + 1) * sizeof(ati_page_map *),GFP_KERNEL); + tables = kzalloc((nr_tables + 1) * sizeof(struct ati_page_map *),GFP_KERNEL); if (tables == NULL) return -ENOMEM; for (i = 0; i < nr_tables; i++) { - entry = kzalloc(sizeof(ati_page_map), GFP_KERNEL); + entry = kzalloc(sizeof(struct ati_page_map), GFP_KERNEL); if (entry == NULL) { while (i>0) { kfree (tables[i-1]); @@ -340,7 +340,7 @@ static int ati_remove_memory(struct agp_memory * mem, off_t pg_start, static int ati_create_gatt_table(struct agp_bridge_data *bridge) { struct aper_size_info_lvl2 *value; - ati_page_map page_dir; + struct ati_page_map page_dir; unsigned long addr; int retval; u32 temp; @@ -400,7 +400,7 @@ static int ati_create_gatt_table(struct agp_bridge_data *bridge) static int ati_free_gatt_table(struct agp_bridge_data *bridge) { - ati_page_map page_dir; + struct ati_page_map page_dir; page_dir.real = (unsigned long *)agp_bridge->gatt_table_real; page_dir.remapped = (unsigned long __iomem *)agp_bridge->gatt_table; From 7707ea3b784195315366e6e4b5c73ca6933ff9b0 Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Sun, 28 Jan 2007 17:50:17 -0500 Subject: [PATCH 209/326] [AGPGART] Remove pointless assignment. No point in clearing local pointers then returning. Also fix up some CodingStyle nits. Signed-off-by: Dave Jones --- drivers/char/agp/ati-agp.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/char/agp/ati-agp.c b/drivers/char/agp/ati-agp.c index 0994d3534447..9987dc2e0c3f 100644 --- a/drivers/char/agp/ati-agp.c +++ b/drivers/char/agp/ati-agp.c @@ -124,18 +124,18 @@ static int ati_create_gatt_pages(int nr_tables) for (i = 0; i < nr_tables; i++) { entry = kzalloc(sizeof(struct ati_page_map), GFP_KERNEL); if (entry == NULL) { - while (i>0) { - kfree (tables[i-1]); + while (i > 0) { + kfree(tables[i-1]); i--; } - kfree (tables); - tables = NULL; + kfree(tables); retval = -ENOMEM; break; } tables[i] = entry; retval = ati_create_page_map(entry); - if (retval != 0) break; + if (retval != 0) + break; } ati_generic_private.num_tables = nr_tables; ati_generic_private.gatt_pages = tables; From 43ed41f648554c9fecaf7597d25e05da63ec7290 Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Sun, 28 Jan 2007 17:58:33 -0500 Subject: [PATCH 210/326] [AGPGART] Add new IDs to VIA AGP. Culled from the VIA codedrop. Also fixes up one ID used in amd64-agp to use the VIA part number instead of the board name in its ID. Signed-off-by: Dave Jones --- drivers/char/agp/amd64-agp.c | 2 +- drivers/char/agp/via-agp.c | 21 +++++++++++++++++++-- include/linux/pci_ids.h | 3 ++- 3 files changed, 22 insertions(+), 4 deletions(-) diff --git a/drivers/char/agp/amd64-agp.c b/drivers/char/agp/amd64-agp.c index 979300405c0e..93d2209fee4c 100644 --- a/drivers/char/agp/amd64-agp.c +++ b/drivers/char/agp/amd64-agp.c @@ -655,7 +655,7 @@ static struct pci_device_id agp_amd64_pci_table[] = { .class = (PCI_CLASS_BRIDGE_HOST << 8), .class_mask = ~0, .vendor = PCI_VENDOR_ID_VIA, - .device = PCI_DEVICE_ID_VIA_K8M890CE, + .device = PCI_DEVICE_ID_VIA_VT3336, .subvendor = PCI_ANY_ID, .subdevice = PCI_ANY_ID, }, diff --git a/drivers/char/agp/via-agp.c b/drivers/char/agp/via-agp.c index c149ac9ce9a7..2ded7a280d7f 100644 --- a/drivers/char/agp/via-agp.c +++ b/drivers/char/agp/via-agp.c @@ -380,9 +380,23 @@ static struct agp_device_ids via_agp_device_ids[] __devinitdata = /* P4M800CE */ { .device_id = PCI_DEVICE_ID_VIA_P4M800CE, - .chipset_name = "P4M800CE", + .chipset_name = "VT3314", + }, + /* CX700 */ + { + .device_id = PCI_DEVICE_ID_VIA_CX700, + .chipset_name = "CX700", + }, + /* VT3336 */ + { + .device_id = PCI_DEVICE_ID_VIA_VT3336, + .chipset_name = "VT3336", + }, + /* P4M890 */ + { + .device_id = PCI_DEVICE_ID_VIA_P4M890, + .chipset_name = "P4M890", }, - { }, /* dummy final entry, always present */ }; @@ -524,6 +538,9 @@ static const struct pci_device_id agp_via_pci_table[] = { ID(PCI_DEVICE_ID_VIA_83_87XX_1), ID(PCI_DEVICE_ID_VIA_3296_0), ID(PCI_DEVICE_ID_VIA_P4M800CE), + ID(PCI_DEVICE_ID_VIA_CX700), + ID(PCI_DEVICE_ID_VIA_VT3336), + ID(PCI_DEVICE_ID_VIA_P4M890), { } }; diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 5c87cd8a56fd..083ad66e6528 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1273,7 +1273,8 @@ #define PCI_DEVICE_ID_VIA_3296_0 0x0296 #define PCI_DEVICE_ID_VIA_8363_0 0x0305 #define PCI_DEVICE_ID_VIA_P4M800CE 0x0314 -#define PCI_DEVICE_ID_VIA_K8M890CE 0x0336 +#define PCI_DEVICE_ID_VIA_P4M890 0x0327 +#define PCI_DEVICE_ID_VIA_VT3336 0x0336 #define PCI_DEVICE_ID_VIA_8371_0 0x0391 #define PCI_DEVICE_ID_VIA_8501_0 0x0501 #define PCI_DEVICE_ID_VIA_82C561 0x0561 From 0142f9dce8425da031d72dc3b70ee7161fcaaea2 Mon Sep 17 00:00:00 2001 From: "Ahmed S. Darwish" Date: Fri, 5 Jan 2007 05:44:54 +0200 Subject: [PATCH 211/326] [CPUFREQ] check sysfs_create_link return value Trivial patch to check sysfs_create_link return values. Fail gracefully if needed. Signed-off-by: Ahmed Darwish Signed-off-by: Dave Jones --- drivers/cpufreq/cpufreq.c | 17 +++++++++++++---- 1 file changed, 13 insertions(+), 4 deletions(-) diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index d91330432ba2..a45cc89e387a 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -722,8 +722,13 @@ static int cpufreq_add_dev (struct sys_device * sys_dev) spin_unlock_irqrestore(&cpufreq_driver_lock, flags); dprintk("CPU already managed, adding link\n"); - sysfs_create_link(&sys_dev->kobj, - &managed_policy->kobj, "cpufreq"); + ret = sysfs_create_link(&sys_dev->kobj, + &managed_policy->kobj, + "cpufreq"); + if (ret) { + mutex_unlock(&policy->lock); + goto err_out_driver_exit; + } cpufreq_debug_enable_ratelimit(); mutex_unlock(&policy->lock); @@ -770,8 +775,12 @@ static int cpufreq_add_dev (struct sys_device * sys_dev) dprintk("CPU %u already managed, adding link\n", j); cpufreq_cpu_get(cpu); cpu_sys_dev = get_cpu_sysdev(j); - sysfs_create_link(&cpu_sys_dev->kobj, &policy->kobj, - "cpufreq"); + ret = sysfs_create_link(&cpu_sys_dev->kobj, &policy->kobj, + "cpufreq"); + if (ret) { + mutex_unlock(&policy->lock); + goto err_out_unregister; + } } policy->governor = NULL; /* to assure that the starting sequence is From 3453c8478a228d9b374956ea99256163f1a0c88c Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Mon, 29 Jan 2007 00:07:04 -0500 Subject: [PATCH 212/326] [CPUFREQ] Remove unneeded errata workaround from p4-clockmod. This workaround unnecessarily cripples functionality to work around an errata that doesn't seem possible to hit due to us using the automatic clock throttling in the p4 mcheck code. See http://lkml.org/lkml/2006/10/28/148 for complete reasoning and lack of disconsent. Signed-off-by: Dave Jones --- arch/i386/kernel/cpu/cpufreq/p4-clockmod.c | 9 --------- 1 file changed, 9 deletions(-) diff --git a/arch/i386/kernel/cpu/cpufreq/p4-clockmod.c b/arch/i386/kernel/cpu/cpufreq/p4-clockmod.c index bec50170b75a..4786fedca6eb 100644 --- a/arch/i386/kernel/cpu/cpufreq/p4-clockmod.c +++ b/arch/i386/kernel/cpu/cpufreq/p4-clockmod.c @@ -51,7 +51,6 @@ enum { static int has_N44_O17_errata[NR_CPUS]; -static int has_N60_errata[NR_CPUS]; static unsigned int stock_freq; static struct cpufreq_driver p4clockmod_driver; static unsigned int cpufreq_p4_get(unsigned int cpu); @@ -224,12 +223,6 @@ static int cpufreq_p4_cpu_init(struct cpufreq_policy *policy) case 0x0f12: has_N44_O17_errata[policy->cpu] = 1; dprintk("has errata -- disabling low frequencies\n"); - break; - - case 0x0f29: - has_N60_errata[policy->cpu] = 1; - dprintk("has errata -- disabling frequencies lower than 2ghz\n"); - break; } /* get max frequency */ @@ -241,8 +234,6 @@ static int cpufreq_p4_cpu_init(struct cpufreq_policy *policy) for (i=1; (p4clockmod_table[i].frequency != CPUFREQ_TABLE_END); i++) { if ((i<2) && (has_N44_O17_errata[policy->cpu])) p4clockmod_table[i].frequency = CPUFREQ_ENTRY_INVALID; - else if (has_N60_errata[policy->cpu] && ((stock_freq * i)/8) < 2000000) - p4clockmod_table[i].frequency = CPUFREQ_ENTRY_INVALID; else p4clockmod_table[i].frequency = (stock_freq * i)/8; } From b9d1902cd281d9b829fb3d6ee9148d28c8c63382 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Mon, 29 Jan 2007 00:11:29 +0100 Subject: [PATCH 213/326] [ARM] 4117/1: S3C2412: Fix writel() usage in selection code The S3C2412 DMA selection code has the arguments to writel() the wrong way around. Signed-off-by: Ben Dooks Signed-off-by: Russell King --- arch/arm/mach-s3c2410/s3c2412-dma.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm/mach-s3c2410/s3c2412-dma.c b/arch/arm/mach-s3c2410/s3c2412-dma.c index fe71a8fdb87c..138f726ac6bf 100644 --- a/arch/arm/mach-s3c2410/s3c2412-dma.c +++ b/arch/arm/mach-s3c2410/s3c2412-dma.c @@ -133,8 +133,8 @@ static struct s3c24xx_dma_map __initdata s3c2412_dma_mappings[] = { static void s3c2412_dma_select(struct s3c2410_dma_chan *chan, struct s3c24xx_dma_map *map) { - writel(chan->regs + S3C2412_DMA_DMAREQSEL, - map->channels[0] | S3C2412_DMAREQSEL_HW); + writel(map->channels[0] | S3C2412_DMAREQSEL_HW, + chan->regs + S3C2412_DMA_DMAREQSEL); } static struct s3c24xx_dma_selection __initdata s3c2412_dma_sel = { From 7d2aae1e88660cf09be913e9754e45189dc33412 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Fri, 26 Jan 2007 12:40:31 +0100 Subject: [PATCH 214/326] [PATCH] ALSA: Fix sysfs breakage The recent change for a new sysfs tree with card* object breaks the /sys/class/sound tree if CONFIG_SYSFS_DEPRECATED is enabled. The device in each entry doesn't point the correct device object: /sys/class/sound ... |-- pcmC0D0c | |-- dev | |-- device -> ../../../class/sound/card0 | |-- pcm_class | |-- power | | `-- wakeup | |-- subsystem -> ../../../class/sound | `-- uevent Also, this change breaks some drivers (like sound/arm/*) referring card->dev directly to obtain the device object for memory handling. This patch reverts the semantics of card->dev to the former version, which points to a real device object. The card* object is stored in a new card->card_dev field, instead. The device parent is chosen either card->dev or card->card_dev according to CONFIG_SYSFS_DEPRECATED to keep the tree compatibility. Also, card* isn't created if CONFIG_SYSFS_DEPRECATED is enabled. The reason of card* object is a root of all beloing devices, and it makes little sense if each sound device points to the real device object directly. Signed-off-by: Takashi Iwai Acked-by: Monty Montgomery Signed-off-by: Greg Kroah-Hartman --- include/sound/core.h | 18 +++++++++++++++--- sound/core/init.c | 18 +++++++++++------- sound/core/sound.c | 4 +--- sound/core/sound_oss.c | 4 +--- 4 files changed, 28 insertions(+), 16 deletions(-) diff --git a/include/sound/core.h b/include/sound/core.h index a994bea09cd6..521f036cce99 100644 --- a/include/sound/core.h +++ b/include/sound/core.h @@ -132,8 +132,10 @@ struct snd_card { int shutdown; /* this card is going down */ int free_on_last_close; /* free in context of file_release */ wait_queue_head_t shutdown_sleep; - struct device *parent; - struct device *dev; + struct device *dev; /* device assigned to this card */ +#ifndef CONFIG_SYSFS_DEPRECATED + struct device *card_dev; /* cardX object for sysfs */ +#endif #ifdef CONFIG_PM unsigned int power_state; /* power state */ @@ -191,6 +193,16 @@ struct snd_minor { struct device *dev; /* device for sysfs */ }; +/* return a device pointer linked to each sound device as a parent */ +static inline struct device *snd_card_get_device_link(struct snd_card *card) +{ +#ifdef CONFIG_SYSFS_DEPRECATED + return card ? card->dev : NULL; +#else + return card ? card->card_dev : NULL; +#endif +} + /* sound.c */ extern int snd_major; @@ -257,7 +269,7 @@ int snd_card_file_add(struct snd_card *card, struct file *file); int snd_card_file_remove(struct snd_card *card, struct file *file); #ifndef snd_card_set_dev -#define snd_card_set_dev(card,devptr) ((card)->parent = (devptr)) +#define snd_card_set_dev(card,devptr) ((card)->dev = (devptr)) #endif /* device.c */ diff --git a/sound/core/init.c b/sound/core/init.c index 6152a7554dfd..a4cc6b155ae9 100644 --- a/sound/core/init.c +++ b/sound/core/init.c @@ -361,8 +361,10 @@ static int snd_card_do_free(struct snd_card *card) snd_printk(KERN_WARNING "unable to free card info\n"); /* Not fatal error */ } - if (card->dev) - device_unregister(card->dev); +#ifndef CONFIG_SYSFS_DEPRECATED + if (card->card_dev) + device_unregister(card->card_dev); +#endif kfree(card); return 0; } @@ -497,12 +499,14 @@ int snd_card_register(struct snd_card *card) int err; snd_assert(card != NULL, return -EINVAL); - if (!card->dev) { - card->dev = device_create(sound_class, card->parent, 0, - "card%i", card->number); - if (IS_ERR(card->dev)) - card->dev = NULL; +#ifndef CONFIG_SYSFS_DEPRECATED + if (!card->card_dev) { + card->card_dev = device_create(sound_class, card->dev, 0, + "card%i", card->number); + if (IS_ERR(card->card_dev)) + card->card_dev = NULL; } +#endif if ((err = snd_device_register_all(card)) < 0) return err; mutex_lock(&snd_card_mutex); diff --git a/sound/core/sound.c b/sound/core/sound.c index 282742022de6..82a61c67cf3a 100644 --- a/sound/core/sound.c +++ b/sound/core/sound.c @@ -238,7 +238,7 @@ int snd_register_device(int type, struct snd_card *card, int dev, { int minor; struct snd_minor *preg; - struct device *device = NULL; + struct device *device = snd_card_get_device_link(card); snd_assert(name, return -EINVAL); preg = kmalloc(sizeof *preg, GFP_KERNEL); @@ -263,8 +263,6 @@ int snd_register_device(int type, struct snd_card *card, int dev, return minor; } snd_minors[minor] = preg; - if (card) - device = card->dev; preg->dev = device_create(sound_class, device, MKDEV(major, minor), "%s", name); if (preg->dev) diff --git a/sound/core/sound_oss.c b/sound/core/sound_oss.c index b2fc40aa520b..4566df41912a 100644 --- a/sound/core/sound_oss.c +++ b/sound/core/sound_oss.c @@ -106,7 +106,7 @@ int snd_register_oss_device(int type, struct snd_card *card, int dev, int cidx = SNDRV_MINOR_OSS_CARD(minor); int track2 = -1; int register1 = -1, register2 = -1; - struct device *carddev = NULL; + struct device *carddev = snd_card_get_device_link(card); if (card && card->number >= 8) return 0; /* ignore silently */ @@ -134,8 +134,6 @@ int snd_register_oss_device(int type, struct snd_card *card, int dev, track2 = SNDRV_MINOR_OSS(cidx, SNDRV_MINOR_OSS_DMMIDI1); break; } - if (card) - carddev = card->dev; register1 = register_sound_special_device(f_ops, minor, carddev); if (register1 != minor) goto __end; From dc6e29da9162fa8fa2a9e798569c0f6e87975614 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 29 Jan 2007 16:37:38 -0800 Subject: [PATCH 215/326] Fix balance_dirty_page() calculations with CONFIG_HIGHMEM This makes balance_dirty_page() always base its calculations on the amount of non-highmem memory in the machine, rather than try to base it on total memory and then falling back on non-highmem memory if the mapping it was writing wasn't highmem capable. This not only fixes a situation where two different writers can have wildly different notions about what is a "balanced" dirty state, but it also means that people with highmem machines don't run into an OOM situation when regular memory fills up with dirty pages. We used to try to handle the latter case by scaling down the dirty_ratio if the machine had a lot of highmem pages in page_writeback_init(), but it wasn't aggressive enough for some situations, and since basing the dirty ratio on highmem memory was broken in the first place, let's just stop doing so. (A variation of this theme fixed Justin Piszcz's OOM problem when copying an 18GB file on a RAID setup). Acked-by: Nick Piggin Cc: Justin Piszcz Cc: Andrew Morton Cc: Neil Brown Cc: Ingo Molnar Cc: Randy Dunlap Cc: Christoph Lameter Cc: Jens Axboe Cc: Peter Zijlstra Cc: Adrian Bunk Signed-off-by: Linus Torvalds --- mm/page-writeback.c | 41 ++++++++++++++++++----------------------- 1 file changed, 18 insertions(+), 23 deletions(-) diff --git a/mm/page-writeback.c b/mm/page-writeback.c index 1d2fc89ca56d..be0efbde4994 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -133,11 +133,9 @@ get_dirty_limits(long *pbackground, long *pdirty, #ifdef CONFIG_HIGHMEM /* - * If this mapping can only allocate from low memory, - * we exclude high memory from our count. + * We always exclude high memory from our count. */ - if (mapping && !(mapping_gfp_mask(mapping) & __GFP_HIGHMEM)) - available_memory -= totalhigh_pages; + available_memory -= totalhigh_pages; #endif @@ -526,28 +524,25 @@ static struct notifier_block __cpuinitdata ratelimit_nb = { }; /* - * If the machine has a large highmem:lowmem ratio then scale back the default - * dirty memory thresholds: allowing too much dirty highmem pins an excessive - * number of buffer_heads. + * Called early on to tune the page writeback dirty limits. + * + * We used to scale dirty pages according to how total memory + * related to pages that could be allocated for buffers (by + * comparing nr_free_buffer_pages() to vm_total_pages. + * + * However, that was when we used "dirty_ratio" to scale with + * all memory, and we don't do that any more. "dirty_ratio" + * is now applied to total non-HIGHPAGE memory (by subtracting + * totalhigh_pages from vm_total_pages), and as such we can't + * get into the old insane situation any more where we had + * large amounts of dirty pages compared to a small amount of + * non-HIGHMEM memory. + * + * But we might still want to scale the dirty_ratio by how + * much memory the box has.. */ void __init page_writeback_init(void) { - long buffer_pages = nr_free_buffer_pages(); - long correction; - - correction = (100 * 4 * buffer_pages) / vm_total_pages; - - if (correction < 100) { - dirty_background_ratio *= correction; - dirty_background_ratio /= 100; - vm_dirty_ratio *= correction; - vm_dirty_ratio /= 100; - - if (dirty_background_ratio <= 0) - dirty_background_ratio = 1; - if (vm_dirty_ratio <= 0) - vm_dirty_ratio = 1; - } mod_timer(&wb_timer, jiffies + dirty_writeback_interval); writeback_set_ratelimit(); register_cpu_notifier(&ratelimit_nb); From 4cbf2aa35e1c189db234190fefc6c83b139ef963 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Mon, 29 Jan 2007 16:38:07 -0800 Subject: [PATCH 216/326] [PATCH] sky2: revert IRQ dance on suspend/resume MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Let's just backout the IRQ hack, and for those crap machines (like some Sony VAIO's) can just disable MSI with the module parameter. This reverts 44ade178249fe53d055fd92113eaa271e06acddd. Signed-off-by: Stephen Hemminger Cc: Jeff Garzik Cc: Thomas Gleixner Cc: Frédéric Riss Signed-off-by: Linus Torvalds --- drivers/net/sky2.c | 25 ------------------------- 1 file changed, 25 deletions(-) diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index a2e804ddca6c..822dd0b13133 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -3639,29 +3639,6 @@ static int sky2_resume(struct pci_dev *pdev) out: return err; } - -/* BIOS resume runs after device (it's a bug in PM) - * as a temporary workaround on suspend/resume leave MSI disabled - */ -static int sky2_suspend_late(struct pci_dev *pdev, pm_message_t state) -{ - struct sky2_hw *hw = pci_get_drvdata(pdev); - - free_irq(pdev->irq, hw); - if (hw->msi) { - pci_disable_msi(pdev); - hw->msi = 0; - } - return 0; -} - -static int sky2_resume_early(struct pci_dev *pdev) -{ - struct sky2_hw *hw = pci_get_drvdata(pdev); - struct net_device *dev = hw->dev[0]; - - return request_irq(pdev->irq, sky2_intr, IRQF_SHARED, dev->name, hw); -} #endif static struct pci_driver sky2_driver = { @@ -3672,8 +3649,6 @@ static struct pci_driver sky2_driver = { #ifdef CONFIG_PM .suspend = sky2_suspend, .resume = sky2_resume, - .suspend_late = sky2_suspend_late, - .resume_early = sky2_resume_early, #endif }; From 87df7241bd547da5d4d4a4e5397866dfe422e439 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Tue, 30 Jan 2007 14:36:27 +1100 Subject: [PATCH 217/326] [PATCH] Fix try_to_free_buffer() locking Fix commit ecdfc9787fe527491baefc22dce8b2dbd5b2908d Not to put too fine a point on it, but in a nutshell... __set_page_dirty_buffers() | try_to_free_buffers() ---------------------------+--------------------------- | spin_lock(private_lock); | drop_bufers() | spin_unlock(private_lock); spin_lock(private_lock) | !page_has_buffers() | spin_unlock(private_lock) | SetPageDirty() | | cancel_dirty_page() oops! Signed-off-by: Nick Piggin Acked-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/buffer.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/fs/buffer.c b/fs/buffer.c index 460f1c43238e..1ad674fd348c 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -2844,7 +2844,6 @@ int try_to_free_buffers(struct page *page) spin_lock(&mapping->private_lock); ret = drop_buffers(page, &buffers_to_free); - spin_unlock(&mapping->private_lock); /* * If the filesystem writes its buffers by hand (eg ext3) @@ -2855,9 +2854,14 @@ int try_to_free_buffers(struct page *page) * Also, during truncate, discard_buffer will have marked all * the page's buffers clean. We discover that here and clean * the page also. + * + * private_lock must be held over this entire operation in order + * to synchronise against __set_page_dirty_buffers and prevent the + * dirty bit from being lost. */ if (ret) cancel_dirty_page(page, PAGE_CACHE_SIZE); + spin_unlock(&mapping->private_lock); out: if (buffers_to_free) { struct buffer_head *bh = buffers_to_free; From c0d4d573feed199b16094c072e7cb07afb01c598 Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Mon, 29 Jan 2007 21:18:38 -0500 Subject: [PATCH 218/326] [PATCH] Fix SG_IO timeout jiffy conversion Commit 85e04e371b5a321b5df2bc3f8e0099a64fb087d7 cleaned up the timeout conversion, but did it exactly the wrong way. We get msecs from user space, and should convert them into jiffies. Not the other way around. Here is a fix with the overflow check sg.c has added in. This fixes DVD burnign with Nero. Signed-off-by: Mike Christie [ "you'll be wanting a comma there" - Andrew ] Cc: Andrew Morton Signed-off-by: Linus Torvalds --- block/scsi_ioctl.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/block/scsi_ioctl.c b/block/scsi_ioctl.c index 2528a0c0dec8..65c6a3cba6d6 100644 --- a/block/scsi_ioctl.c +++ b/block/scsi_ioctl.c @@ -223,7 +223,7 @@ static int verify_command(struct file *file, unsigned char *cmd) static int sg_io(struct file *file, request_queue_t *q, struct gendisk *bd_disk, struct sg_io_hdr *hdr) { - unsigned long start_time; + unsigned long start_time, timeout; int writing = 0, ret = 0; struct request *rq; char sense[SCSI_SENSE_BUFFERSIZE]; @@ -271,7 +271,8 @@ static int sg_io(struct file *file, request_queue_t *q, rq->cmd_type = REQ_TYPE_BLOCK_PC; - rq->timeout = jiffies_to_msecs(hdr->timeout); + timeout = msecs_to_jiffies(hdr->timeout); + rq->timeout = (timeout < INT_MAX) ? timeout : INT_MAX; if (!rq->timeout) rq->timeout = q->sg_timeout; if (!rq->timeout) From bf2326e7525404576ab2bdf890e903ca4a249093 Mon Sep 17 00:00:00 2001 From: Jan Altenberg Date: Thu, 25 Jan 2007 16:25:56 +0100 Subject: [PATCH 219/326] [PATCH] Malta: Fix build if CONFIG_MTD is diabled. Signed-off-by: Jan Altenberg Signed-off-by: Ralf Baechle --- arch/mips/mips-boards/malta/Makefile | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/mips/mips-boards/malta/Makefile b/arch/mips/mips-boards/malta/Makefile index b662c75fb28e..cb7f349b0514 100644 --- a/arch/mips/mips-boards/malta/Makefile +++ b/arch/mips/mips-boards/malta/Makefile @@ -19,5 +19,6 @@ # under Linux. # -obj-y := malta_int.o malta_mtd.o malta_setup.o +obj-y := malta_int.o malta_setup.o +obj-$(CONFIG_MTD) += malta_mtd.o obj-$(CONFIG_SMP) += malta_smp.o From 1ca5cb5ddd500fdf2096ebe9d4131ee9f1eabf43 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Thu, 25 Jan 2007 23:55:17 +0000 Subject: [PATCH 220/326] [MIPS] Ocelot G: Fix a few misspellings of CONFIG_GALILEO_GT64240_ETH Reported by Robert P. J. Day . Signed-off-by: Ralf Baechle --- arch/mips/momentum/ocelot_g/prom.c | 4 ++-- arch/mips/momentum/ocelot_g/setup.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/arch/mips/momentum/ocelot_g/prom.c b/arch/mips/momentum/ocelot_g/prom.c index 6509a9c9863c..2f75c6b91ec5 100644 --- a/arch/mips/momentum/ocelot_g/prom.c +++ b/arch/mips/momentum/ocelot_g/prom.c @@ -28,7 +28,7 @@ struct callvectors* debug_vectors; extern unsigned long marvell_base; extern unsigned long bus_clock; -#ifdef CONFIG_GALILLEO_GT64240_ETH +#ifdef CONFIG_GALILEO_GT64240_ETH extern unsigned char prom_mac_addr_base[6]; #endif @@ -61,7 +61,7 @@ void __init prom_init(void) mips_machgroup = MACH_GROUP_MOMENCO; mips_machtype = MACH_MOMENCO_OCELOT_G; -#ifdef CONFIG_GALILLEO_GT64240_ETH +#ifdef CONFIG_GALILEO_GT64240_ETH /* get the base MAC address for on-board ethernet ports */ memcpy(prom_mac_addr_base, (void*)0xfc807cf2, 6); #endif diff --git a/arch/mips/momentum/ocelot_g/setup.c b/arch/mips/momentum/ocelot_g/setup.c index d288f7b01842..9db638a7982c 100644 --- a/arch/mips/momentum/ocelot_g/setup.c +++ b/arch/mips/momentum/ocelot_g/setup.c @@ -64,7 +64,7 @@ #include "ocelot_pld.h" -#ifdef CONFIG_GALILLEO_GT64240_ETH +#ifdef CONFIG_GALILEO_GT64240_ETH extern unsigned char prom_mac_addr_base[6]; #endif @@ -185,7 +185,7 @@ void __init plat_mem_setup(void) /* do handoff reconfiguration */ PMON_v2_setup(); -#ifdef CONFIG_GALILLEO_GT64240_ETH +#ifdef CONFIG_GALILEO_GT64240_ETH /* get the mac addr */ memcpy(prom_mac_addr_base, (void*)0xfc807cf2, 6); #endif From 9616d54fffa4c9ae78136cc6d01209de2d92b75d Mon Sep 17 00:00:00 2001 From: "Robert P. J. Day" Date: Thu, 25 Jan 2007 18:41:35 -0500 Subject: [PATCH 221/326] [MIPS] Fix typo of "CONFIG_MT_SMP". Signed-off-by: Robert P. J. Day Signed-off-by: Ralf Baechle --- arch/mips/mips-boards/sim/sim_setup.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/mips/mips-boards/sim/sim_setup.c b/arch/mips/mips-boards/sim/sim_setup.c index 2659c1c3b78d..ea2066c3a1f7 100644 --- a/arch/mips/mips-boards/sim/sim_setup.c +++ b/arch/mips/mips-boards/sim/sim_setup.c @@ -57,7 +57,7 @@ void __init plat_mem_setup(void) board_time_init = sim_time_init; prom_printf("Linux started...\n"); -#ifdef CONFIG_MT_SMP +#ifdef CONFIG_MIPS_MT_SMP sanitize_tlb_entries(); #endif } From 76398f9667e8369023ed5f4847fb59e9da8b6968 Mon Sep 17 00:00:00 2001 From: Jiri Kosina Date: Mon, 29 Jan 2007 12:44:41 +0100 Subject: [PATCH 222/326] HID: fix pb_fnmode and move it to generic HID The apple powerbook people are used to switch the pb_fnmode setting at runtime through writing to sysfs, altering the module parameter value. This was broken for them in 2.6.20-rc1 when generic HID layer was introduced, as the pb_fnmode flag was made per-hiddevice, instead of global variable. This patch moves the pb_fnmode module parameter from usbhid module to hid module, but apart from that retains backward compatibility with respect to changing the mode through sysfs. Signed-off-by: Jiri Kosina --- drivers/hid/hid-input.c | 11 ++++++++--- drivers/usb/input/hid-core.c | 9 --------- include/linux/hid.h | 1 - 3 files changed, 8 insertions(+), 13 deletions(-) diff --git a/drivers/hid/hid-input.c b/drivers/hid/hid-input.c index 981fcf01d41b..c7a6833f6821 100644 --- a/drivers/hid/hid-input.c +++ b/drivers/hid/hid-input.c @@ -35,6 +35,11 @@ #include +static int hid_pb_fnmode = 1; +module_param_named(pb_fnmode, hid_pb_fnmode, int, 0644); +MODULE_PARM_DESC(pb_fnmode, + "Mode of fn key on PowerBooks (0 = disabled, 1 = fkeyslast, 2 = fkeysfirst)"); + #define unk KEY_UNKNOWN static const unsigned char hid_keyboard[256] = { @@ -154,7 +159,7 @@ static int hidinput_pb_event(struct hid_device *hid, struct input_dev *input, return 1; } - if (hid->pb_fnmode) { + if (hid_pb_fnmode) { int do_translate; trans = find_translation(powerbook_fn_keys, usage->code); @@ -163,8 +168,8 @@ static int hidinput_pb_event(struct hid_device *hid, struct input_dev *input, do_translate = 1; else if (trans->flags & POWERBOOK_FLAG_FKEY) do_translate = - (hid->pb_fnmode == 2 && (hid->quirks & HID_QUIRK_POWERBOOK_FN_ON)) || - (hid->pb_fnmode == 1 && !(hid->quirks & HID_QUIRK_POWERBOOK_FN_ON)); + (hid_pb_fnmode == 2 && (hid->quirks & HID_QUIRK_POWERBOOK_FN_ON)) || + (hid_pb_fnmode == 1 && !(hid->quirks & HID_QUIRK_POWERBOOK_FN_ON)); else do_translate = (hid->quirks & HID_QUIRK_POWERBOOK_FN_ON); diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c index b86480457a2c..c6c9e72e5fd9 100644 --- a/drivers/usb/input/hid-core.c +++ b/drivers/usb/input/hid-core.c @@ -56,11 +56,6 @@ static unsigned int hid_mousepoll_interval; module_param_named(mousepoll, hid_mousepoll_interval, uint, 0644); MODULE_PARM_DESC(mousepoll, "Polling interval of mice"); -static int usbhid_pb_fnmode = 1; -module_param_named(pb_fnmode, usbhid_pb_fnmode, int, 0644); -MODULE_PARM_DESC(pb_fnmode, - "Mode of fn key on PowerBooks (0 = disabled, 1 = fkeyslast, 2 = fkeysfirst)"); - /* * Input submission and I/O error handler. */ @@ -1246,10 +1241,6 @@ static struct hid_device *usb_hid_configure(struct usb_interface *intf) hid->hiddev_hid_event = hiddev_hid_event; hid->hiddev_report_event = hiddev_report_event; #endif -#ifdef CONFIG_USB_HIDINPUT_POWERBOOK - hid->pb_fnmode = usbhid_pb_fnmode; -#endif - return hid; fail: diff --git a/include/linux/hid.h b/include/linux/hid.h index 770120add15a..342b4e639acb 100644 --- a/include/linux/hid.h +++ b/include/linux/hid.h @@ -438,7 +438,6 @@ struct hid_device { /* device report descriptor */ struct hid_usage *, __s32); void (*hiddev_report_event) (struct hid_device *, struct hid_report *); #ifdef CONFIG_USB_HIDINPUT_POWERBOOK - unsigned int pb_fnmode; unsigned long pb_pressed_fn[NBITS(KEY_MAX)]; unsigned long pb_pressed_numlock[NBITS(KEY_MAX)]; #endif From 5dcade90db19205b9ebb8241a22664560973f81a Mon Sep 17 00:00:00 2001 From: Sergei Shtylyov Date: Sun, 28 Jan 2007 21:33:44 +0300 Subject: [PATCH 223/326] pata_sil680: PIO1 taskfile transfers overclocking fix (repost) Fix PIO mode 1 overclocked taskfile transfers -- probably a typo carried over from drivers/ide/pci/siimage.c where I've found it by documentation check... Signed-off-by: Sergei Shtylyov Signed-off-by: Jeff Garzik --- drivers/ata/pata_sil680.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/ata/pata_sil680.c b/drivers/ata/pata_sil680.c index 32cf0bfa8921..e8dfd8fc3ff7 100644 --- a/drivers/ata/pata_sil680.c +++ b/drivers/ata/pata_sil680.c @@ -135,7 +135,7 @@ static void sil680_error_handler(struct ata_port *ap) static void sil680_set_piomode(struct ata_port *ap, struct ata_device *adev) { static u16 speed_p[5] = { 0x328A, 0x2283, 0x1104, 0x10C3, 0x10C1 }; - static u16 speed_t[5] = { 0x328A, 0x1281, 0x1281, 0x10C3, 0x10C1 }; + static u16 speed_t[5] = { 0x328A, 0x2283, 0x1281, 0x10C3, 0x10C1 }; unsigned long tfaddr = sil680_selreg(ap, 0x02); unsigned long addr = sil680_seldev(ap, adev, 0x04); From 7a0f1c8a4b1052da7efc7715e2e557255b632712 Mon Sep 17 00:00:00 2001 From: Lennert Buytenhek Date: Mon, 29 Jan 2007 13:28:47 +0100 Subject: [PATCH 224/326] ata_if_xfermask() word 51 fix If word 53 bit 1 isn't set, the maximum PIO mode is indicated by the upper 8 bits of word 51, not the lower 8 bits. Fixes PIO mode detection on old Compact Flash cards. Signed-off-by: Lennert Buytenhek Signed-off-by: Jeff Garzik --- drivers/ata/libata-core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index a388a8df0043..cf707029352e 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -1037,7 +1037,7 @@ static unsigned int ata_id_xfermask(const u16 *id) * the PIO timing number for the maximum. Turn it into * a mask. */ - u8 mode = id[ATA_ID_OLD_PIO_MODES] & 0xFF; + u8 mode = (id[ATA_ID_OLD_PIO_MODES] >> 8) & 0xFF; if (mode < 5) /* Valid PIO range */ pio_mask = (2 << mode) - 1; else From 2ca6611b1fc1f913f5370b3c50a0a35d817491b3 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 30 Jan 2007 00:59:14 -0800 Subject: [PATCH 225/326] pata_platform: set_mode fix drivers/ata/pata_platform.c:85: warning: initialization from incompatible pointer type Cc: Jeff Garzik Cc: Tejun Heo Acked-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik --- drivers/ata/pata_platform.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/ata/pata_platform.c b/drivers/ata/pata_platform.c index 443b1d85c6c4..ca2999f9f048 100644 --- a/drivers/ata/pata_platform.c +++ b/drivers/ata/pata_platform.c @@ -30,7 +30,8 @@ static int pio_mask = 1; * Provide our own set_mode() as we don't want to change anything that has * already been configured.. */ -static void pata_platform_set_mode(struct ata_port *ap) +static int pata_platform_set_mode(struct ata_port *ap, + struct ata_device **r_failed_dev) { int i; @@ -44,6 +45,7 @@ static void pata_platform_set_mode(struct ata_port *ap) dev->flags |= ATA_DFLAG_PIO; } } + return 0; } static void pata_platform_host_stop(struct ata_host *host) From af068bd1debcc76c1bc265aa01401901bf0067ed Mon Sep 17 00:00:00 2001 From: David Milburn Date: Tue, 30 Jan 2007 00:59:15 -0800 Subject: [PATCH 226/326] libata-scsi: ata_task_ioctl should return ATA registers from sense data User applications using the HDIO_DRIVE_TASK ioctl through libata expect specific ATA registers to be returned to userspace. Verified that ata_task_ioctl correctly returns register values to the smartctl application. Signed-off-by: David Milburn Acked-by: Tejun Heo Cc: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik --- drivers/ata/libata-scsi.c | 53 +++++++++++++++++++++++++++++++++------ 1 file changed, 46 insertions(+), 7 deletions(-) diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c index 7cc5a4a910a4..d151cf057ef2 100644 --- a/drivers/ata/libata-scsi.c +++ b/drivers/ata/libata-scsi.c @@ -273,8 +273,8 @@ int ata_task_ioctl(struct scsi_device *scsidev, void __user *arg) { int rc = 0; u8 scsi_cmd[MAX_COMMAND_SIZE]; - u8 args[7]; - struct scsi_sense_hdr sshdr; + u8 args[7], *sensebuf = NULL; + int cmd_result; if (arg == NULL) return -EINVAL; @@ -282,10 +282,14 @@ int ata_task_ioctl(struct scsi_device *scsidev, void __user *arg) if (copy_from_user(args, arg, sizeof(args))) return -EFAULT; + sensebuf = kzalloc(SCSI_SENSE_BUFFERSIZE, GFP_NOIO); + if (!sensebuf) + return -ENOMEM; + memset(scsi_cmd, 0, sizeof(scsi_cmd)); scsi_cmd[0] = ATA_16; scsi_cmd[1] = (3 << 1); /* Non-data */ - /* scsi_cmd[2] is already 0 -- no off.line, cc, or data xfer */ + scsi_cmd[2] = 0x20; /* cc but no off.line or data xfer */ scsi_cmd[4] = args[1]; scsi_cmd[6] = args[2]; scsi_cmd[8] = args[3]; @@ -295,11 +299,46 @@ int ata_task_ioctl(struct scsi_device *scsidev, void __user *arg) /* Good values for timeout and retries? Values below from scsi_ioctl_send_command() for default case... */ - if (scsi_execute_req(scsidev, scsi_cmd, DMA_NONE, NULL, 0, &sshdr, - (10*HZ), 5)) - rc = -EIO; + cmd_result = scsi_execute(scsidev, scsi_cmd, DMA_NONE, NULL, 0, + sensebuf, (10*HZ), 5, 0); - /* Need code to retrieve data from check condition? */ + if (driver_byte(cmd_result) == DRIVER_SENSE) {/* sense data available */ + u8 *desc = sensebuf + 8; + cmd_result &= ~(0xFF<<24); /* DRIVER_SENSE is not an error */ + + /* If we set cc then ATA pass-through will cause a + * check condition even if no error. Filter that. */ + if (cmd_result & SAM_STAT_CHECK_CONDITION) { + struct scsi_sense_hdr sshdr; + scsi_normalize_sense(sensebuf, SCSI_SENSE_BUFFERSIZE, + &sshdr); + if (sshdr.sense_key==0 && + sshdr.asc==0 && sshdr.ascq==0) + cmd_result &= ~SAM_STAT_CHECK_CONDITION; + } + + /* Send userspace ATA registers */ + if (sensebuf[0] == 0x72 && /* format is "descriptor" */ + desc[0] == 0x09) {/* code is "ATA Descriptor" */ + args[0] = desc[13]; /* status */ + args[1] = desc[3]; /* error */ + args[2] = desc[5]; /* sector count (0:7) */ + args[3] = desc[7]; /* lbal */ + args[4] = desc[9]; /* lbam */ + args[5] = desc[11]; /* lbah */ + args[6] = desc[12]; /* select */ + if (copy_to_user(arg, args, sizeof(args))) + rc = -EFAULT; + } + } + + if (cmd_result) { + rc = -EIO; + goto error; + } + + error: + kfree(sensebuf); return rc; } From 78981a7c6c34bddbb90da72cf6ce10953e84aad8 Mon Sep 17 00:00:00 2001 From: Robert Hancock Date: Tue, 30 Jan 2007 00:59:18 -0800 Subject: [PATCH 227/326] libata: fix translation for START STOP UNIT libata's SCSI translation for the SCSI START STOP UNIT command with the START bit clear (i.e. stopping the drive) appears to be incorrect. It sends an ATA STANDBY command with the time period set to 0, which the code comment says means "now", but the ATA standard says this means disable the standby timer, which effectively does nothing. Change this to issue a STANDBY IMMEDIATE command which will actually spin the drive down. The SAT (SCSI/ATA Translation) standard revision 9 concurs with this choice. Signed-off-by: Robert Hancock Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik --- drivers/ata/libata-scsi.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c index d151cf057ef2..73902d335767 100644 --- a/drivers/ata/libata-scsi.c +++ b/drivers/ata/libata-scsi.c @@ -1022,11 +1022,10 @@ static unsigned int ata_scsi_start_stop_xlat(struct ata_queued_cmd *qc) } tf->command = ATA_CMD_VERIFY; /* READ VERIFY */ - } else { - tf->nsect = 0; /* time period value (0 implies now) */ - tf->command = ATA_CMD_STANDBY; - /* Consider: ATA STANDBY IMMEDIATE command */ - } + } else + /* Issue ATA STANDBY IMMEDIATE command */ + tf->command = ATA_CMD_STANDBYNOW1; + /* * Standby and Idle condition timers could be implemented but that * would require libata to implement the Power condition mode page From 5fc7d61aee1a7f7d3448f8fbccaa93371ebeecb0 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Fri, 26 Jan 2007 23:59:57 -0800 Subject: [PATCH 228/326] b44: Fix frequent link changes This fixes the issue of frequent link changes under heavy traffic reported below: http://bugzilla.kernel.org/show_bug.cgi?id=7696 https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=216338 The b44 chip occasionally needs to be reset when ISTAT_ERRORS are encountered. The reset sequence includes a PHY reset that will take many seconds to complete and cause the link to go down and up. By skipping the PHY reset, it will greatly reduce the interruption when ISTAT_ERRORS are encountered. Change the full_reset parameter to reset_kind parameter in b44_init_hw(). This will allow PHY reset to be skipped when ISTAT_ERRORS are encountered. Signed-off-by: Michael Chan Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik --- drivers/net/b44.c | 35 ++++++++++++++++++++--------------- 1 file changed, 20 insertions(+), 15 deletions(-) diff --git a/drivers/net/b44.c b/drivers/net/b44.c index 5eb2ec68393f..1c8fb8e7df59 100644 --- a/drivers/net/b44.c +++ b/drivers/net/b44.c @@ -110,6 +110,11 @@ MODULE_DEVICE_TABLE(pci, b44_pci_tbl); static void b44_halt(struct b44 *); static void b44_init_rings(struct b44 *); + +#define B44_FULL_RESET 1 +#define B44_FULL_RESET_SKIP_PHY 2 +#define B44_PARTIAL_RESET 3 + static void b44_init_hw(struct b44 *, int); static int dma_desc_align_mask; @@ -884,7 +889,7 @@ static int b44_poll(struct net_device *netdev, int *budget) spin_lock_irqsave(&bp->lock, flags); b44_halt(bp); b44_init_rings(bp); - b44_init_hw(bp, 1); + b44_init_hw(bp, B44_FULL_RESET_SKIP_PHY); netif_wake_queue(bp->dev); spin_unlock_irqrestore(&bp->lock, flags); done = 1; @@ -954,7 +959,7 @@ static void b44_tx_timeout(struct net_device *dev) b44_halt(bp); b44_init_rings(bp); - b44_init_hw(bp, 1); + b44_init_hw(bp, B44_FULL_RESET); spin_unlock_irq(&bp->lock); @@ -1071,7 +1076,7 @@ static int b44_change_mtu(struct net_device *dev, int new_mtu) b44_halt(bp); dev->mtu = new_mtu; b44_init_rings(bp); - b44_init_hw(bp, 1); + b44_init_hw(bp, B44_FULL_RESET); spin_unlock_irq(&bp->lock); b44_enable_ints(bp); @@ -1368,12 +1373,12 @@ static int b44_set_mac_addr(struct net_device *dev, void *p) * packet processing. Invoked with bp->lock held. */ static void __b44_set_rx_mode(struct net_device *); -static void b44_init_hw(struct b44 *bp, int full_reset) +static void b44_init_hw(struct b44 *bp, int reset_kind) { u32 val; b44_chip_reset(bp); - if (full_reset) { + if (reset_kind == B44_FULL_RESET) { b44_phy_reset(bp); b44_setup_phy(bp); } @@ -1390,7 +1395,10 @@ static void b44_init_hw(struct b44 *bp, int full_reset) bw32(bp, B44_TXMAXLEN, bp->dev->mtu + ETH_HLEN + 8 + RX_HEADER_LEN); bw32(bp, B44_TX_WMARK, 56); /* XXX magic */ - if (full_reset) { + if (reset_kind == B44_PARTIAL_RESET) { + bw32(bp, B44_DMARX_CTRL, (DMARX_CTRL_ENABLE | + (bp->rx_offset << DMARX_CTRL_ROSHIFT))); + } else { bw32(bp, B44_DMATX_CTRL, DMATX_CTRL_ENABLE); bw32(bp, B44_DMATX_ADDR, bp->tx_ring_dma + bp->dma_offset); bw32(bp, B44_DMARX_CTRL, (DMARX_CTRL_ENABLE | @@ -1401,9 +1409,6 @@ static void b44_init_hw(struct b44 *bp, int full_reset) bp->rx_prod = bp->rx_pending; bw32(bp, B44_MIB_CTRL, MIB_CTRL_CLR_ON_READ); - } else { - bw32(bp, B44_DMARX_CTRL, (DMARX_CTRL_ENABLE | - (bp->rx_offset << DMARX_CTRL_ROSHIFT))); } val = br32(bp, B44_ENET_CTRL); @@ -1420,7 +1425,7 @@ static int b44_open(struct net_device *dev) goto out; b44_init_rings(bp); - b44_init_hw(bp, 1); + b44_init_hw(bp, B44_FULL_RESET); b44_check_phy(bp); @@ -1629,7 +1634,7 @@ static int b44_close(struct net_device *dev) netif_poll_enable(dev); if (bp->flags & B44_FLAG_WOL_ENABLE) { - b44_init_hw(bp, 0); + b44_init_hw(bp, B44_PARTIAL_RESET); b44_setup_wol(bp); } @@ -1905,7 +1910,7 @@ static int b44_set_ringparam(struct net_device *dev, b44_halt(bp); b44_init_rings(bp); - b44_init_hw(bp, 1); + b44_init_hw(bp, B44_FULL_RESET); netif_wake_queue(bp->dev); spin_unlock_irq(&bp->lock); @@ -1948,7 +1953,7 @@ static int b44_set_pauseparam(struct net_device *dev, if (bp->flags & B44_FLAG_PAUSE_AUTO) { b44_halt(bp); b44_init_rings(bp); - b44_init_hw(bp, 1); + b44_init_hw(bp, B44_FULL_RESET); } else { __b44_set_flow_ctrl(bp, bp->flags); } @@ -2304,7 +2309,7 @@ static int b44_suspend(struct pci_dev *pdev, pm_message_t state) free_irq(dev->irq, dev); if (bp->flags & B44_FLAG_WOL_ENABLE) { - b44_init_hw(bp, 0); + b44_init_hw(bp, B44_PARTIAL_RESET); b44_setup_wol(bp); } pci_disable_device(pdev); @@ -2329,7 +2334,7 @@ static int b44_resume(struct pci_dev *pdev) spin_lock_irq(&bp->lock); b44_init_rings(bp); - b44_init_hw(bp, 1); + b44_init_hw(bp, B44_FULL_RESET); netif_device_attach(bp->dev); spin_unlock_irq(&bp->lock); From 0c0b3ae68ec93b1db5c637d294647d1cca0df763 Mon Sep 17 00:00:00 2001 From: Mariusz Kozlowski Date: Sat, 27 Jan 2007 00:00:01 -0800 Subject: [PATCH 229/326] net: ifb error path loop fix On error we should start freeing resources at [i-1] not [i-2]. Signed-off-by: Mariusz Kozlowski Cc: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik --- drivers/net/ifb.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/net/ifb.c b/drivers/net/ifb.c index ca2b21f9d444..c4ca7c930a46 100644 --- a/drivers/net/ifb.c +++ b/drivers/net/ifb.c @@ -271,8 +271,7 @@ static int __init ifb_init_module(void) for (i = 0; i < numifbs && !err; i++) err = ifb_init_one(i); if (err) { - i--; - while (--i >= 0) + while (i--) ifb_free_one(i); } From b1f54ba34f9e036ab515bbac0c01d17300e1c79a Mon Sep 17 00:00:00 2001 From: Vitaly Bordug Date: Sat, 27 Jan 2007 00:00:04 -0800 Subject: [PATCH 230/326] FS_ENET: OF-related fixup for FEC and SCC MAC's Updated direct resource pass with ioremap call, make it grant proper IRQ mapping, stuff incompatible with the new approach were respectively put under #ifndef CONFIG_PPC_MERGE. It is required so that both ppc and powerpc could utilize fs_enet effectively. Signed-off-by: Vitaly Bordug Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik --- drivers/net/fs_enet/mac-fec.c | 13 +++++++++---- drivers/net/fs_enet/mac-scc.c | 6 ++++-- drivers/net/phy/fixed.c | 2 +- 3 files changed, 14 insertions(+), 7 deletions(-) diff --git a/drivers/net/fs_enet/mac-fec.c b/drivers/net/fs_enet/mac-fec.c index c2c5fd419bd0..ff6839477306 100644 --- a/drivers/net/fs_enet/mac-fec.c +++ b/drivers/net/fs_enet/mac-fec.c @@ -104,9 +104,9 @@ static int do_pd_setup(struct fs_enet_private *fep) fep->interrupt = platform_get_irq_byname(pdev,"interrupt"); if (fep->interrupt < 0) return -EINVAL; - + r = platform_get_resource_byname(pdev, IORESOURCE_MEM, "regs"); - fep->fec.fecp =(void*)r->start; + fep->fec.fecp = ioremap(r->start, r->end - r->start + 1); if(fep->fec.fecp == NULL) return -EINVAL; @@ -319,11 +319,14 @@ static void restart(struct net_device *dev) * Clear any outstanding interrupt. */ FW(fecp, ievent, 0xffc0); +#ifndef CONFIG_PPC_MERGE FW(fecp, ivec, (fep->interrupt / 2) << 29); - +#else + FW(fecp, ivec, (virq_to_hw(fep->interrupt) / 2) << 29); +#endif /* - * adjust to speed (only for DUET & RMII) + * adjust to speed (only for DUET & RMII) */ #ifdef CONFIG_DUET if (fpi->use_rmii) { @@ -418,6 +421,7 @@ static void stop(struct net_device *dev) static void pre_request_irq(struct net_device *dev, int irq) { +#ifndef CONFIG_PPC_MERGE immap_t *immap = fs_enet_immap; u32 siel; @@ -431,6 +435,7 @@ static void pre_request_irq(struct net_device *dev, int irq) siel &= ~(0x80000000 >> (irq & ~1)); out_be32(&immap->im_siu_conf.sc_siel, siel); } +#endif } static void post_free_irq(struct net_device *dev, int irq) diff --git a/drivers/net/fs_enet/mac-scc.c b/drivers/net/fs_enet/mac-scc.c index 95ec5872c507..afd7fca7c6c4 100644 --- a/drivers/net/fs_enet/mac-scc.c +++ b/drivers/net/fs_enet/mac-scc.c @@ -121,13 +121,13 @@ static int do_pd_setup(struct fs_enet_private *fep) return -EINVAL; r = platform_get_resource_byname(pdev, IORESOURCE_MEM, "regs"); - fep->scc.sccp = (void *)r->start; + fep->scc.sccp = ioremap(r->start, r->end - r->start + 1); if (fep->scc.sccp == NULL) return -EINVAL; r = platform_get_resource_byname(pdev, IORESOURCE_MEM, "pram"); - fep->scc.ep = (void *)r->start; + fep->scc.ep = ioremap(r->start, r->end - r->start + 1); if (fep->scc.ep == NULL) return -EINVAL; @@ -397,6 +397,7 @@ static void stop(struct net_device *dev) static void pre_request_irq(struct net_device *dev, int irq) { +#ifndef CONFIG_PPC_MERGE immap_t *immap = fs_enet_immap; u32 siel; @@ -410,6 +411,7 @@ static void pre_request_irq(struct net_device *dev, int irq) siel &= ~(0x80000000 >> (irq & ~1)); out_be32(&immap->im_siu_conf.sc_siel, siel); } +#endif } static void post_free_irq(struct net_device *dev, int irq) diff --git a/drivers/net/phy/fixed.c b/drivers/net/phy/fixed.c index 096d4a100bf2..86135397f430 100644 --- a/drivers/net/phy/fixed.c +++ b/drivers/net/phy/fixed.c @@ -349,7 +349,7 @@ static int __init fixed_init(void) fixed_mdio_register_device(0, 100, 1); #endif -#ifdef CONFIX_FIXED_MII_10_FDX +#ifdef CONFIG_FIXED_MII_10_FDX fixed_mdio_register_device(0, 10, 1); #endif return 0; From 9e8e83d1ba0b248de34062a61f4f5d378a5dbd53 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Sat, 27 Jan 2007 00:00:04 -0800 Subject: [PATCH 231/326] 82596 warning fixes drivers/net/82596.c: In function 'i596_start_xmit': drivers/net/82596.c:1069: warning: cast from pointer to integer of different size drivers/net/82596.c: In function 'i82596_probe': drivers/net/82596.c:1249: warning: format '%d' expects type 'int', but argument 4 has type 'long unsigned int' Cc: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik --- drivers/net/82596.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/net/82596.c b/drivers/net/82596.c index 8236f26ffd46..640d7ca2ebcf 100644 --- a/drivers/net/82596.c +++ b/drivers/net/82596.c @@ -1066,8 +1066,8 @@ static int i596_start_xmit(struct sk_buff *skb, struct net_device *dev) short length = skb->len; dev->trans_start = jiffies; - DEB(DEB_STARTTX,printk(KERN_DEBUG "%s: i596_start_xmit(%x,%x) called\n", dev->name, - skb->len, (unsigned int)skb->data)); + DEB(DEB_STARTTX,printk(KERN_DEBUG "%s: i596_start_xmit(%x,%p) called\n", + dev->name, skb->len, skb->data)); if (skb->len < ETH_ZLEN) { if (skb_padto(skb, ETH_ZLEN)) @@ -1246,7 +1246,8 @@ struct net_device * __init i82596_probe(int unit) dev->priv = (void *)(dev->mem_start); lp = dev->priv; - DEB(DEB_INIT,printk(KERN_DEBUG "%s: lp at 0x%08lx (%d bytes), lp->scb at 0x%08lx\n", + DEB(DEB_INIT,printk(KERN_DEBUG "%s: lp at 0x%08lx (%zd bytes), " + "lp->scb at 0x%08lx\n", dev->name, (unsigned long)lp, sizeof(struct i596_private), (unsigned long)&lp->scb)); memset((void *) lp, 0, sizeof(struct i596_private)); From 9c750b7d14301b710c13247f7cc28abd614d9f5c Mon Sep 17 00:00:00 2001 From: Thomas Klein Date: Mon, 29 Jan 2007 18:44:01 +0100 Subject: [PATCH 232/326] ehea: Fixed wrong jumbo frames status query This patch fixes the wrong query and logging of the per interface jumbo frames enabled/disabled status. Signed-off-by: Thomas Klein Signed-off-by: Jeff Garzik --- drivers/net/ehea/ehea.h | 2 +- drivers/net/ehea/ehea_main.c | 30 +++++++++++++++++++++++------- 2 files changed, 24 insertions(+), 8 deletions(-) diff --git a/drivers/net/ehea/ehea.h b/drivers/net/ehea/ehea.h index be10a3a26b5b..272e1ec51aa2 100644 --- a/drivers/net/ehea/ehea.h +++ b/drivers/net/ehea/ehea.h @@ -39,7 +39,7 @@ #include #define DRV_NAME "ehea" -#define DRV_VERSION "EHEA_0044" +#define DRV_VERSION "EHEA_0045" #define EHEA_MSG_DEFAULT (NETIF_MSG_LINK | NETIF_MSG_TIMER \ | NETIF_MSG_RX_ERR | NETIF_MSG_TX_ERR) diff --git a/drivers/net/ehea/ehea_main.c b/drivers/net/ehea/ehea_main.c index 1072e69ef85d..d4635bf3128c 100644 --- a/drivers/net/ehea/ehea_main.c +++ b/drivers/net/ehea/ehea_main.c @@ -2316,6 +2316,7 @@ static int ehea_setup_single_port(struct ehea_port *port, struct ehea_adapter *adapter = port->adapter; struct hcp_ehea_port_cb4 *cb4; u32 *dn_log_port_id; + int jumbo = 0; sema_init(&port->port_lock, 1); port->state = EHEA_PORT_DOWN; @@ -2357,13 +2358,25 @@ static int ehea_setup_single_port(struct ehea_port *port, if (!cb4) { ehea_error("no mem for cb4"); } else { - cb4->jumbo_frame = 1; - hret = ehea_h_modify_ehea_port(adapter->handle, - port->logical_port_id, - H_PORT_CB4, H_PORT_CB4_JUMBO, - cb4); - if (hret != H_SUCCESS) { - ehea_info("Jumbo frames not activated"); + hret = ehea_h_query_ehea_port(adapter->handle, + port->logical_port_id, + H_PORT_CB4, + H_PORT_CB4_JUMBO, cb4); + + if (hret == H_SUCCESS) { + if (cb4->jumbo_frame) + jumbo = 1; + else { + cb4->jumbo_frame = 1; + hret = ehea_h_modify_ehea_port(adapter->handle, + port-> + logical_port_id, + H_PORT_CB4, + H_PORT_CB4_JUMBO, + cb4); + if (hret == H_SUCCESS) + jumbo = 1; + } } kfree(cb4); } @@ -2402,6 +2415,9 @@ static int ehea_setup_single_port(struct ehea_port *port, goto out_free; } + ehea_info("%s: Jumbo frames are %sabled", dev->name, + jumbo == 1 ? "en" : "dis"); + port->netdev = dev; ret = 0; goto out; From d4150a2731615de5cd4527a23435aaa7396c63c6 Mon Sep 17 00:00:00 2001 From: Thomas Klein Date: Mon, 29 Jan 2007 18:44:41 +0100 Subject: [PATCH 233/326] ehea: Fixed missing tasklet_kill() call NEQ-Tasklet wasn't killed when module is removed. Signed-off-by: Thomas Klein Signed-off-by: Jeff Garzik --- drivers/net/ehea/ehea_main.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ehea/ehea_main.c b/drivers/net/ehea/ehea_main.c index d4635bf3128c..9de2d38a5321 100644 --- a/drivers/net/ehea/ehea_main.c +++ b/drivers/net/ehea/ehea_main.c @@ -2598,6 +2598,7 @@ static int __devexit ehea_remove(struct ibmebus_dev *dev) destroy_workqueue(adapter->ehea_wq); ibmebus_free_irq(NULL, adapter->neq->attr.ist1, adapter); + tasklet_kill(&adapter->neq_tasklet); ehea_destroy_eq(adapter->neq); From f8a8ccd56d82bd4f4b5c7c2e7eb758c7764d98e1 Mon Sep 17 00:00:00 2001 From: Andy Gospodarek Date: Mon, 29 Jan 2007 12:08:38 -0800 Subject: [PATCH 234/326] bonding: ARP monitoring broken on x86_64 While working with the latest bonding code I noticed a nasty problem that will prevent arp monitoring from always functioning correctly on x86_64 systems. Comparing ints to longs and expecting reliable results on x86_64 is a bad idea. With this patch, arp monitoring works correctly again. Signed-off-by: Andy Gospodarek Cc: "David S. Miller" Cc: Stephen Hemminger Cc: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik --- drivers/net/bonding/bonding.h | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/net/bonding/bonding.h b/drivers/net/bonding/bonding.h index dc434fb6da85..0978c9ac6d2b 100644 --- a/drivers/net/bonding/bonding.h +++ b/drivers/net/bonding/bonding.h @@ -151,8 +151,8 @@ struct slave { struct slave *next; struct slave *prev; int delay; - u32 jiffies; - u32 last_arp_rx; + unsigned long jiffies; + unsigned long last_arp_rx; s8 link; /* one of BOND_LINK_XXXX */ s8 state; /* one of BOND_STATE_XXXX */ u32 original_flags; @@ -242,7 +242,8 @@ extern inline int slave_do_arp_validate(struct bonding *bond, struct slave *slav return bond->params.arp_validate & (1 << slave->state); } -extern inline u32 slave_last_rx(struct bonding *bond, struct slave *slave) +extern inline unsigned long slave_last_rx(struct bonding *bond, + struct slave *slave) { if (slave_do_arp_validate(bond, slave)) return slave->last_arp_rx; From 518d83382568964ca9657511140398ebac925ecd Mon Sep 17 00:00:00 2001 From: Auke Kok Date: Mon, 29 Jan 2007 14:31:16 -0800 Subject: [PATCH 235/326] e100: fix irq leak on suspend/resume e100: fix irq leak on suspend/resume From: Frederik Deweerdt The e100_resume() function should be calling netif_device_detach and free_irq. This fixes multiple irq's being allocated after resume. Signed-off-by: Frederik Deweerdt Signed-off-by: Auke Kok Signed-off-by: Jeff Garzik --- drivers/net/e100.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/e100.c b/drivers/net/e100.c index c2ae2a24629b..3208dac29168 100644 --- a/drivers/net/e100.c +++ b/drivers/net/e100.c @@ -2725,6 +2725,7 @@ static int e100_suspend(struct pci_dev *pdev, pm_message_t state) del_timer_sync(&nic->watchdog); netif_carrier_off(nic->netdev); + netif_device_detach(netdev); pci_save_state(pdev); if ((nic->flags & wol_magic) | e100_asf(nic)) { @@ -2736,6 +2737,7 @@ static int e100_suspend(struct pci_dev *pdev, pm_message_t state) } pci_disable_device(pdev); + free_irq(pdev->irq, netdev); pci_set_power_state(pdev, PCI_D3hot); return 0; From 00576e93805bd4f2dd2649e354726dee872d1e8a Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 30 Jan 2007 13:23:50 +0000 Subject: [PATCH 236/326] b44: src_desc->addr is little-endian Signed-off-by: Al Viro Signed-off-by: Jeff Garzik --- drivers/net/b44.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/b44.c b/drivers/net/b44.c index 1c8fb8e7df59..26b9295cb9b7 100644 --- a/drivers/net/b44.c +++ b/drivers/net/b44.c @@ -757,7 +757,7 @@ static void b44_recycle_rx(struct b44 *bp, int src_idx, u32 dest_idx_unmasked) dest_idx * sizeof(dest_desc), DMA_BIDIRECTIONAL); - pci_dma_sync_single_for_device(bp->pdev, src_desc->addr, + pci_dma_sync_single_for_device(bp->pdev, le32_to_cpu(src_desc->addr), RX_PKT_BUF_SZ, PCI_DMA_FROMDEVICE); } From 90afd0e574a1a739aeb62e30d556ebf0289389e5 Mon Sep 17 00:00:00 2001 From: Dmitriy Monakhov Date: Sat, 27 Jan 2007 00:00:03 -0800 Subject: [PATCH 237/326] Broadcom 4400 resume small fix Some issues in b44_resume(). - Return value of pci_enable_device() was ignored. - If request_irq() has failed we have to just disable device and exit. Signed-off-by: Dmitriy Monakhov Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik --- drivers/net/b44.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) diff --git a/drivers/net/b44.c b/drivers/net/b44.c index 26b9295cb9b7..303a8d94ad4b 100644 --- a/drivers/net/b44.c +++ b/drivers/net/b44.c @@ -2320,16 +2320,27 @@ static int b44_resume(struct pci_dev *pdev) { struct net_device *dev = pci_get_drvdata(pdev); struct b44 *bp = netdev_priv(dev); + int rc = 0; pci_restore_state(pdev); - pci_enable_device(pdev); + rc = pci_enable_device(pdev); + if (rc) { + printk(KERN_ERR PFX "%s: pci_enable_device failed\n", + dev->name); + return rc; + } + pci_set_master(pdev); if (!netif_running(dev)) return 0; - if (request_irq(dev->irq, b44_interrupt, IRQF_SHARED, dev->name, dev)) + rc = request_irq(dev->irq, b44_interrupt, IRQF_SHARED, dev->name, dev); + if (rc) { printk(KERN_ERR PFX "%s: request_irq failed\n", dev->name); + pci_disable_device(pdev); + return rc; + } spin_lock_irq(&bp->lock); From 7a238fcba0629b6f2edbcd37458bae56fcf36be5 Mon Sep 17 00:00:00 2001 From: "Serge E. Hallyn" Date: Mon, 29 Jan 2007 13:19:40 -0800 Subject: [PATCH 238/326] [PATCH] namespaces: fix exit race by splitting exit Fix exit race by splitting the nsproxy putting into two pieces. First piece reduces the nsproxy refcount. If we dropped the last reference, then it puts the mnt_ns, and returns the nsproxy as a hint to the caller. Else it returns NULL. The second piece of exiting task namespaces sets tsk->nsproxy to NULL, and drops the references to other namespaces and frees the nsproxy only if an nsproxy was passed in. A little awkward and should probably be reworked, but hopefully it fixes the NFS oops. Signed-off-by: Serge E. Hallyn Cc: Herbert Poetzl Cc: Oleg Nesterov Cc: "Eric W. Biederman" Cc: Cedric Le Goater Cc: Daniel Hokka Zakrisson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/nsproxy.h | 30 +++++++++++++++++++----------- kernel/exit.c | 6 ++++-- kernel/fork.c | 4 ++-- kernel/nsproxy.c | 16 +++++++++++++++- 4 files changed, 40 insertions(+), 16 deletions(-) diff --git a/include/linux/nsproxy.h b/include/linux/nsproxy.h index 0b9f0dc30d61..678e1d38effb 100644 --- a/include/linux/nsproxy.h +++ b/include/linux/nsproxy.h @@ -35,22 +35,30 @@ struct nsproxy *dup_namespaces(struct nsproxy *orig); int copy_namespaces(int flags, struct task_struct *tsk); void get_task_namespaces(struct task_struct *tsk); void free_nsproxy(struct nsproxy *ns); +struct nsproxy *put_nsproxy(struct nsproxy *ns); -static inline void put_nsproxy(struct nsproxy *ns) +static inline void finalize_put_nsproxy(struct nsproxy *ns) { - if (atomic_dec_and_test(&ns->count)) { + if (ns) free_nsproxy(ns); - } } -static inline void exit_task_namespaces(struct task_struct *p) +static inline void put_and_finalize_nsproxy(struct nsproxy *ns) { - struct nsproxy *ns = p->nsproxy; - if (ns) { - task_lock(p); - p->nsproxy = NULL; - task_unlock(p); - put_nsproxy(ns); - } + finalize_put_nsproxy(put_nsproxy(ns)); +} + +static inline struct nsproxy *preexit_task_namespaces(struct task_struct *p) +{ + return put_nsproxy(p->nsproxy); +} + +static inline void exit_task_namespaces(struct task_struct *p, + struct nsproxy *ns) +{ + task_lock(p); + p->nsproxy = NULL; + task_unlock(p); + finalize_put_nsproxy(ns); } #endif diff --git a/kernel/exit.c b/kernel/exit.c index 35401720635b..a5bf5329ff97 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -396,7 +396,7 @@ void daemonize(const char *name, ...) current->fs = fs; atomic_inc(&fs->count); - exit_task_namespaces(current); + put_and_finalize_nsproxy(current->nsproxy); current->nsproxy = init_task.nsproxy; get_task_namespaces(current); @@ -853,6 +853,7 @@ static void exit_notify(struct task_struct *tsk) fastcall NORET_TYPE void do_exit(long code) { struct task_struct *tsk = current; + struct nsproxy *ns; int group_dead; profile_task_exit(tsk); @@ -938,8 +939,9 @@ fastcall NORET_TYPE void do_exit(long code) tsk->exit_code = code; proc_exit_connector(tsk); + ns = preexit_task_namespaces(tsk); exit_notify(tsk); - exit_task_namespaces(tsk); + exit_task_namespaces(tsk, ns); #ifdef CONFIG_NUMA mpol_free(tsk->mempolicy); tsk->mempolicy = NULL; diff --git a/kernel/fork.c b/kernel/fork.c index fc723e595cd5..4cf868458f06 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1265,7 +1265,7 @@ static struct task_struct *copy_process(unsigned long clone_flags, return p; bad_fork_cleanup_namespaces: - exit_task_namespaces(p); + put_and_finalize_nsproxy(p->nsproxy); bad_fork_cleanup_keys: exit_keys(p); bad_fork_cleanup_mm: @@ -1711,7 +1711,7 @@ asmlinkage long sys_unshare(unsigned long unshare_flags) } if (new_nsproxy) - put_nsproxy(new_nsproxy); + put_and_finalize_nsproxy(new_nsproxy); bad_unshare_cleanup_ipc: if (new_ipc) diff --git a/kernel/nsproxy.c b/kernel/nsproxy.c index f5b9ee6f6bbb..7b05bce75cde 100644 --- a/kernel/nsproxy.c +++ b/kernel/nsproxy.c @@ -117,7 +117,7 @@ int copy_namespaces(int flags, struct task_struct *tsk) goto out_pid; out: - put_nsproxy(old_ns); + put_and_finalize_nsproxy(old_ns); return err; out_pid: @@ -135,6 +135,20 @@ int copy_namespaces(int flags, struct task_struct *tsk) goto out; } +struct nsproxy *put_nsproxy(struct nsproxy *ns) +{ + if (ns) { + if (atomic_dec_and_test(&ns->count)) { + if (ns->mnt_ns) { + put_mnt_ns(ns->mnt_ns); + ns->mnt_ns = NULL; + } + return ns; + } + } + return NULL; +} + void free_nsproxy(struct nsproxy *ns) { if (ns->mnt_ns) From 88f6cd0c3bb5db2619103f834d4167b7d0d9899c Mon Sep 17 00:00:00 2001 From: Johannes Stezenbach Date: Mon, 29 Jan 2007 13:19:44 -0800 Subject: [PATCH 239/326] [PATCH] uml: fix mknod Fix UML hostfs mknod(): userspace has differernt dev_t size and encoding than kernel, so extract major/minor and reencode using glibc makedev() macro. Signed-off-by: Johannes Stezenbach Acked-by: Paolo 'Blaisorblade' Giarrusso Cc: Jeff Dike Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/hostfs/hostfs.h | 2 +- fs/hostfs/hostfs_kern.c | 2 +- fs/hostfs/hostfs_user.c | 4 ++-- 3 files changed, 4 insertions(+), 4 deletions(-) diff --git a/fs/hostfs/hostfs.h b/fs/hostfs/hostfs.h index cca3fb693f99..70543b17e4c7 100644 --- a/fs/hostfs/hostfs.h +++ b/fs/hostfs/hostfs.h @@ -76,7 +76,7 @@ extern int make_symlink(const char *from, const char *to); extern int unlink_file(const char *file); extern int do_mkdir(const char *file, int mode); extern int do_rmdir(const char *file); -extern int do_mknod(const char *file, int mode, int dev); +extern int do_mknod(const char *file, int mode, unsigned int major, unsigned int minor); extern int link_file(const char *from, const char *to); extern int do_readlink(char *file, char *buf, int size); extern int rename_file(char *from, char *to); diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c index 1e6fc3799876..69a376f35a68 100644 --- a/fs/hostfs/hostfs_kern.c +++ b/fs/hostfs/hostfs_kern.c @@ -755,7 +755,7 @@ int hostfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev) goto out_put; init_special_inode(inode, mode, dev); - err = do_mknod(name, mode, dev); + err = do_mknod(name, mode, MAJOR(dev), MINOR(dev)); if(err) goto out_free; diff --git a/fs/hostfs/hostfs_user.c b/fs/hostfs/hostfs_user.c index 23b7cee72123..1ed5ea389f15 100644 --- a/fs/hostfs/hostfs_user.c +++ b/fs/hostfs/hostfs_user.c @@ -295,11 +295,11 @@ int do_rmdir(const char *file) return(0); } -int do_mknod(const char *file, int mode, int dev) +int do_mknod(const char *file, int mode, unsigned int major, unsigned int minor) { int err; - err = mknod(file, mode, dev); + err = mknod(file, mode, makedev(major, minor)); if(err) return(-errno); return(0); } From 04611f98ce34f8911740ccd7e2bf0c22f818f622 Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Mon, 29 Jan 2007 13:19:49 -0800 Subject: [PATCH 240/326] [PATCH] use __u8/__u32 in userspace ioctl defines for I2O Make sure exported I2O ioctls utilize userspace safe types. Signed-off-by: Mike Frysinger Cc: Markus Lidel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/i2o-dev.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/include/linux/i2o-dev.h b/include/linux/i2o-dev.h index c2519df1b6dc..a0b23dd45239 100644 --- a/include/linux/i2o-dev.h +++ b/include/linux/i2o-dev.h @@ -24,12 +24,13 @@ #define MAX_I2O_CONTROLLERS 32 #include +#include /* * I2O Control IOCTLs and structures */ #define I2O_MAGIC_NUMBER 'i' -#define I2OGETIOPS _IOR(I2O_MAGIC_NUMBER,0,u8[MAX_I2O_CONTROLLERS]) +#define I2OGETIOPS _IOR(I2O_MAGIC_NUMBER,0,__u8[MAX_I2O_CONTROLLERS]) #define I2OHRTGET _IOWR(I2O_MAGIC_NUMBER,1,struct i2o_cmd_hrtlct) #define I2OLCTGET _IOWR(I2O_MAGIC_NUMBER,2,struct i2o_cmd_hrtlct) #define I2OPARMSET _IOWR(I2O_MAGIC_NUMBER,3,struct i2o_cmd_psetget) @@ -37,7 +38,7 @@ #define I2OSWDL _IOWR(I2O_MAGIC_NUMBER,5,struct i2o_sw_xfer) #define I2OSWUL _IOWR(I2O_MAGIC_NUMBER,6,struct i2o_sw_xfer) #define I2OSWDEL _IOWR(I2O_MAGIC_NUMBER,7,struct i2o_sw_xfer) -#define I2OVALIDATE _IOR(I2O_MAGIC_NUMBER,8,u32) +#define I2OVALIDATE _IOR(I2O_MAGIC_NUMBER,8,__u32) #define I2OHTML _IOWR(I2O_MAGIC_NUMBER,9,struct i2o_html) #define I2OEVTREG _IOW(I2O_MAGIC_NUMBER,10,struct i2o_evt_id) #define I2OEVTGET _IOR(I2O_MAGIC_NUMBER,11,struct i2o_evt_info) From 49b14f24cc5aa962446515c9df501192eda99bd4 Mon Sep 17 00:00:00 2001 From: "Robert P. J. Day" Date: Mon, 29 Jan 2007 13:19:50 -0800 Subject: [PATCH 241/326] [PATCH] Fix "CONFIG_X86_64_" typo in drivers/kvm/svm.c Fix what looks like an obvious typo in the file drivers/kvm/svm.c. Signed-off-by: Robert P. J. Day Acked-by: Avi Kivity Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/kvm/svm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/kvm/svm.c b/drivers/kvm/svm.c index 9c70ff65e6b7..c79df79307ed 100644 --- a/drivers/kvm/svm.c +++ b/drivers/kvm/svm.c @@ -1163,7 +1163,7 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, unsigned ecx, u64 data) case MSR_K6_STAR: vcpu->svm->vmcb->save.star = data; break; -#ifdef CONFIG_X86_64_ +#ifdef CONFIG_X86_64 case MSR_LSTAR: vcpu->svm->vmcb->save.lstar = data; break; From a0f7b67ae470f3afed3f03d017fd1c29fd1b58b8 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 29 Jan 2007 13:19:50 -0800 Subject: [PATCH 242/326] [PATCH] m68k: uaccess.h needs sched.h In file included from include/linux/crypto.h:26, from crypto/cipher.c:17: include/linux/uaccess.h: In function 'pagefault_disable': include/linux/uaccess.h:18: error: dereferencing pointer to incomplete type include/linux/uaccess.h: In function 'pagefault_enable': include/linux/uaccess.h:33: error: dereferencing pointer to incomplete type video_buf need PCI. Acked-by: Geert Uytterhoeven Cc: Roman Zippel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-m68k/uaccess.h | 1 + 1 file changed, 1 insertion(+) diff --git a/include/asm-m68k/uaccess.h b/include/asm-m68k/uaccess.h index e4c9f080ff20..6a4cf2081512 100644 --- a/include/asm-m68k/uaccess.h +++ b/include/asm-m68k/uaccess.h @@ -7,6 +7,7 @@ #include #include #include +#include #include #define VERIFY_READ 0 From d019bcf0eb9a55c51f1c57659f923d356b4675e1 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Mon, 29 Jan 2007 13:19:51 -0800 Subject: [PATCH 243/326] [PATCH] fs/lockd/clntlock.c: add missing newlines to dprintk's This patch adds missing newlines to dprintk's. Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/lockd/clntlock.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/lockd/clntlock.c b/fs/lockd/clntlock.c index 062707745162..f4d45d4d835b 100644 --- a/fs/lockd/clntlock.c +++ b/fs/lockd/clntlock.c @@ -176,7 +176,7 @@ reclaimer(void *ptr) lock_kernel(); lockd_up(0); /* note: this cannot fail as lockd is already running */ - dprintk("lockd: reclaiming locks for host %s", host->h_name); + dprintk("lockd: reclaiming locks for host %s\n", host->h_name); restart: nsmstate = host->h_nsmstate; @@ -206,7 +206,7 @@ reclaimer(void *ptr) host->h_reclaiming = 0; up_write(&host->h_rwsem); - dprintk("NLM: done reclaiming locks for host %s", host->h_name); + dprintk("NLM: done reclaiming locks for host %s\n", host->h_name); /* Now, wake up all processes that sleep on a blocked lock */ list_for_each_entry(block, &nlm_blocked, b_list) { From 34e9a63b4f3e169b583f6ba2e26356ecbf932fba Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Mon, 29 Jan 2007 13:19:52 -0800 Subject: [PATCH 244/326] [PATCH] knfsd: ratelimit some nfsd messages that are triggered by external events Also remove {NFSD,RPC}_PARANOIA as having the defines doesn't really add anything. The printks covered by RPC_PARANOIA were triggered by badly formatted packets and so should be ratelimited. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfsd/export.c | 1 - fs/nfsd/nfsfh.c | 14 ++++++-------- fs/nfsd/vfs.c | 1 - net/sunrpc/svc.c | 33 ++++++++++++++++----------------- net/sunrpc/svcsock.c | 12 ++++++++---- 5 files changed, 30 insertions(+), 31 deletions(-) diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c index 248dd92e6a56..49c310b84923 100644 --- a/fs/nfsd/export.c +++ b/fs/nfsd/export.c @@ -35,7 +35,6 @@ #include #define NFSDDBG_FACILITY NFSDDBG_EXPORT -#define NFSD_PARANOIA 1 typedef struct auth_domain svc_client; typedef struct svc_export svc_export; diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c index b06bf9f70efc..98338a569dc0 100644 --- a/fs/nfsd/nfsfh.c +++ b/fs/nfsd/nfsfh.c @@ -24,8 +24,6 @@ #include #define NFSDDBG_FACILITY NFSDDBG_FH -#define NFSD_PARANOIA 1 -/* #define NFSD_DEBUG_VERBOSE 1 */ static int nfsd_nr_verified; @@ -230,13 +228,12 @@ fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, int access) error = nfserrno(PTR_ERR(dentry)); goto out; } -#ifdef NFSD_PARANOIA + if (S_ISDIR(dentry->d_inode->i_mode) && (dentry->d_flags & DCACHE_DISCONNECTED)) { printk("nfsd: find_fh_dentry returned a DISCONNECTED directory: %s/%s\n", dentry->d_parent->d_name.name, dentry->d_name.name); } -#endif fhp->fh_dentry = dentry; fhp->fh_export = exp; @@ -267,12 +264,13 @@ fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, int access) /* Finally, check access permissions. */ error = nfsd_permission(exp, dentry, access); -#ifdef NFSD_PARANOIA_EXTREME if (error) { - printk("fh_verify: %s/%s permission failure, acc=%x, error=%d\n", - dentry->d_parent->d_name.name, dentry->d_name.name, access, (error >> 24)); + dprintk("fh_verify: %s/%s permission failure, " + "acc=%x, error=%d\n", + dentry->d_parent->d_name.name, + dentry->d_name.name, + access, (error >> 24)); } -#endif out: if (exp && !IS_ERR(exp)) exp_put(exp); diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 5d32e5fa697e..8283236c6a0f 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -59,7 +59,6 @@ #include #define NFSDDBG_FACILITY NFSDDBG_FILEOP -#define NFSD_PARANOIA /* We must ignore files (but only files) which might have mandatory diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index bf21a2047010..4c1611211119 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -26,7 +26,6 @@ #include #define RPCDBG_FACILITY RPCDBG_SVCDSP -#define RPC_PARANOIA 1 /* * Mode for mapping cpus to pools. @@ -872,15 +871,15 @@ svc_process(struct svc_rqst *rqstp) return 0; err_short_len: -#ifdef RPC_PARANOIA - printk("svc: short len %Zd, dropping request\n", argv->iov_len); -#endif + if (net_ratelimit()) + printk("svc: short len %Zd, dropping request\n", argv->iov_len); + goto dropit; /* drop request */ err_bad_dir: -#ifdef RPC_PARANOIA - printk("svc: bad direction %d, dropping request\n", dir); -#endif + if (net_ratelimit()) + printk("svc: bad direction %d, dropping request\n", dir); + serv->sv_stats->rpcbadfmt++; goto dropit; /* drop request */ @@ -909,10 +908,10 @@ svc_process(struct svc_rqst *rqstp) goto sendit; err_bad_vers: -#ifdef RPC_PARANOIA - printk("svc: unknown version (%d for prog %d, %s)\n", - vers, prog, progp->pg_name); -#endif + if (net_ratelimit()) + printk("svc: unknown version (%d for prog %d, %s)\n", + vers, prog, progp->pg_name); + serv->sv_stats->rpcbadfmt++; svc_putnl(resv, RPC_PROG_MISMATCH); svc_putnl(resv, progp->pg_lovers); @@ -920,17 +919,17 @@ svc_process(struct svc_rqst *rqstp) goto sendit; err_bad_proc: -#ifdef RPC_PARANOIA - printk("svc: unknown procedure (%d)\n", proc); -#endif + if (net_ratelimit()) + printk("svc: unknown procedure (%d)\n", proc); + serv->sv_stats->rpcbadfmt++; svc_putnl(resv, RPC_PROC_UNAVAIL); goto sendit; err_garbage: -#ifdef RPC_PARANOIA - printk("svc: failed to decode args\n"); -#endif + if (net_ratelimit()) + printk("svc: failed to decode args\n"); + rpc_stat = rpc_garbage_args; err_bad: serv->sv_stats->rpcbadfmt++; diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index 45120f268d03..ff1f8bf680aa 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -1062,15 +1062,19 @@ svc_tcp_recvfrom(struct svc_rqst *rqstp) * bit set in the fragment length header. * But apparently no known nfs clients send fragmented * records. */ - printk(KERN_NOTICE "RPC: bad TCP reclen 0x%08lx (non-terminal)\n", - (unsigned long) svsk->sk_reclen); + if (net_ratelimit()) + printk(KERN_NOTICE "RPC: bad TCP reclen 0x%08lx" + " (non-terminal)\n", + (unsigned long) svsk->sk_reclen); goto err_delete; } svsk->sk_reclen &= 0x7fffffff; dprintk("svc: TCP record, %d bytes\n", svsk->sk_reclen); if (svsk->sk_reclen > serv->sv_max_mesg) { - printk(KERN_NOTICE "RPC: bad TCP reclen 0x%08lx (large)\n", - (unsigned long) svsk->sk_reclen); + if (net_ratelimit()) + printk(KERN_NOTICE "RPC: bad TCP reclen 0x%08lx" + " (large)\n", + (unsigned long) svsk->sk_reclen); goto err_delete; } } From 436d1654b341e55a73cada13cce3142b619f23bb Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Mon, 29 Jan 2007 13:19:53 -0800 Subject: [PATCH 245/326] [PATCH] use __u8 rather than u8 in userspace SIZE defines in hdreg.h Use __u8 rather than u8 in SIZE defines exported to userspace. Signed-off-by: Mike Frysinger Cc: Bartlomiej Zolnierkiewicz Cc: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/hdreg.h | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/include/linux/hdreg.h b/include/linux/hdreg.h index 2b54eac738ea..818c6afc1091 100644 --- a/include/linux/hdreg.h +++ b/include/linux/hdreg.h @@ -60,13 +60,15 @@ #define TAG_MASK 0xf8 #endif /* __KERNEL__ */ +#include + /* * Command Header sizes for IOCTL commands */ -#define HDIO_DRIVE_CMD_HDR_SIZE (4 * sizeof(u8)) -#define HDIO_DRIVE_HOB_HDR_SIZE (8 * sizeof(u8)) -#define HDIO_DRIVE_TASK_HDR_SIZE (8 * sizeof(u8)) +#define HDIO_DRIVE_CMD_HDR_SIZE (4 * sizeof(__u8)) +#define HDIO_DRIVE_HOB_HDR_SIZE (8 * sizeof(__u8)) +#define HDIO_DRIVE_TASK_HDR_SIZE (8 * sizeof(__u8)) #define IDE_DRIVE_TASK_INVALID -1 #define IDE_DRIVE_TASK_NO_DATA 0 From ff79544754631cf3d237ff47b7d0e7ab2d211fcf Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Mon, 29 Jan 2007 13:19:54 -0800 Subject: [PATCH 246/326] [PATCH] fuse: fix bug in control filesystem mount The BUG in fuse_ctl_add_dentry() could be triggered if the control filesystem was unmounted and mounted again while one or more fuse filesystems were present. The fix is to reset the dentry counter in fuse_ctl_kill_sb(). Bug reported by Florent Mertens. Signed-off-by: Miklos Szeredi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/fuse/control.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/fs/fuse/control.c b/fs/fuse/control.c index 8c58bd453993..1794305f9ed8 100644 --- a/fs/fuse/control.c +++ b/fs/fuse/control.c @@ -193,8 +193,12 @@ static int fuse_ctl_get_sb(struct file_system_type *fs_type, int flags, static void fuse_ctl_kill_sb(struct super_block *sb) { + struct fuse_conn *fc; + mutex_lock(&fuse_mutex); fuse_control_sb = NULL; + list_for_each_entry(fc, &fuse_conn_list, entry) + fc->ctl_ndents = 0; mutex_unlock(&fuse_mutex); kill_litter_super(sb); From a685e26fff387db350966f88eaad515bf41c4705 Mon Sep 17 00:00:00 2001 From: Evgeniy Dushistov Date: Mon, 29 Jan 2007 13:19:54 -0800 Subject: [PATCH 247/326] [PATCH] ufs: alloc metadata null page fix These series of patches result of UFS1 write support stress testing, like running fsx-linux, untar and build linux kernel etc We pass from ufs::get_block_t to levels below: pointer to the current page, to make possible things like reallocation of blocks on the fly, and we also uses this pointer for indication, what actually we allocate data block or meta data block, but currently we make decision about what we allocate on the wrong level, this may and cause oops if we allocate blocks in some special order. Signed-off-by: Evgeniy Dushistov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ufs/balloc.c | 5 ++++- fs/ufs/inode.c | 14 +++++++++----- 2 files changed, 13 insertions(+), 6 deletions(-) diff --git a/fs/ufs/balloc.c b/fs/ufs/balloc.c index 2e0021e8f366..96ca8453bab6 100644 --- a/fs/ufs/balloc.c +++ b/fs/ufs/balloc.c @@ -233,7 +233,7 @@ static void ufs_change_blocknr(struct inode *inode, unsigned int baseblk, { unsigned int blk_per_page = 1 << (PAGE_CACHE_SHIFT - inode->i_blkbits); struct address_space *mapping = inode->i_mapping; - pgoff_t index, cur_index = locked_page->index; + pgoff_t index, cur_index; unsigned int i, j; struct page *page; struct buffer_head *head, *bh; @@ -241,8 +241,11 @@ static void ufs_change_blocknr(struct inode *inode, unsigned int baseblk, UFSD("ENTER, ino %lu, count %u, oldb %u, newb %u\n", inode->i_ino, count, oldb, newb); + BUG_ON(!locked_page); BUG_ON(!PageLocked(locked_page)); + cur_index = locked_page->index; + for (i = 0; i < count; i += blk_per_page) { index = (baseblk+i) >> (PAGE_CACHE_SHIFT - inode->i_blkbits); diff --git a/fs/ufs/inode.c b/fs/ufs/inode.c index 2fbab0aab688..4295ca91cf85 100644 --- a/fs/ufs/inode.c +++ b/fs/ufs/inode.c @@ -242,7 +242,8 @@ ufs_inode_getfrag(struct inode *inode, unsigned int fragment, goal = tmp + uspi->s_fpb; tmp = ufs_new_fragments (inode, p, fragment - blockoff, goal, required + blockoff, - err, locked_page); + err, + phys != NULL ? locked_page : NULL); } /* * We will extend last allocated block @@ -250,7 +251,7 @@ ufs_inode_getfrag(struct inode *inode, unsigned int fragment, else if (lastblock == block) { tmp = ufs_new_fragments(inode, p, fragment - (blockoff - lastblockoff), fs32_to_cpu(sb, *p), required + (blockoff - lastblockoff), - err, locked_page); + err, phys != NULL ? locked_page : NULL); } else /* (lastblock > block) */ { /* * We will allocate new block before last allocated block @@ -261,7 +262,8 @@ ufs_inode_getfrag(struct inode *inode, unsigned int fragment, goal = tmp + uspi->s_fpb; } tmp = ufs_new_fragments(inode, p, fragment - blockoff, - goal, uspi->s_fpb, err, locked_page); + goal, uspi->s_fpb, err, + phys != NULL ? locked_page : NULL); } if (!tmp) { if ((!blockoff && *p) || @@ -438,9 +440,11 @@ int ufs_getfrag_block(struct inode *inode, sector_t fragment, struct buffer_head * it much more readable: */ #define GET_INODE_DATABLOCK(x) \ - ufs_inode_getfrag(inode, x, fragment, 1, &err, &phys, &new, bh_result->b_page) + ufs_inode_getfrag(inode, x, fragment, 1, &err, &phys, &new,\ + bh_result->b_page) #define GET_INODE_PTR(x) \ - ufs_inode_getfrag(inode, x, fragment, uspi->s_fpb, &err, NULL, NULL, NULL) + ufs_inode_getfrag(inode, x, fragment, uspi->s_fpb, &err, NULL, NULL,\ + bh_result->b_page) #define GET_INDIRECT_DATABLOCK(x) \ ufs_inode_getblock(inode, bh, x, fragment, \ &err, &phys, &new, bh_result->b_page) From 8682164a66325cab07620082eb7f413b547f4b4a Mon Sep 17 00:00:00 2001 From: Evgeniy Dushistov Date: Mon, 29 Jan 2007 13:19:55 -0800 Subject: [PATCH 248/326] [PATCH] ufs: truncate negative to unsigned fix During ufs_trunc_direct which is subroutine of ufs::truncate, we try the first of all free parts of block and then whole blocks. But we calculate size of block's part to free in the wrong way. This may cause bad update of used blocks and fragments statistic, and you can got report that you have free 32T on 1Gb partition. Signed-off-by: Evgeniy Dushistov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ufs/truncate.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/ufs/truncate.c b/fs/ufs/truncate.c index ea11d04c41a0..0437b0a6fe97 100644 --- a/fs/ufs/truncate.c +++ b/fs/ufs/truncate.c @@ -109,10 +109,10 @@ static int ufs_trunc_direct (struct inode * inode) tmp = fs32_to_cpu(sb, *p); if (!tmp ) ufs_panic (sb, "ufs_trunc_direct", "internal error"); + frag2 -= frag1; frag1 = ufs_fragnum (frag1); - frag2 = ufs_fragnum (frag2); - ufs_free_fragments (inode, tmp + frag1, frag2 - frag1); + ufs_free_fragments(inode, tmp + frag1, frag2); mark_inode_dirty(inode); frag_to_free = tmp + frag1; From efee2b812645d10824bf6cb247789910bcb66881 Mon Sep 17 00:00:00 2001 From: Evgeniy Dushistov Date: Mon, 29 Jan 2007 13:19:56 -0800 Subject: [PATCH 249/326] [PATCH] ufs: reallocation fix In blocks reallocation function sometimes does not update some of buffer_head::b_blocknr, which may and cause data damage. Signed-off-by: Evgeniy Dushistov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ufs/balloc.c | 41 ++++++++++++++++++++++++++--------------- 1 file changed, 26 insertions(+), 15 deletions(-) diff --git a/fs/ufs/balloc.c b/fs/ufs/balloc.c index 96ca8453bab6..638f4c585e89 100644 --- a/fs/ufs/balloc.c +++ b/fs/ufs/balloc.c @@ -227,14 +227,14 @@ void ufs_free_blocks(struct inode *inode, unsigned fragment, unsigned count) * We can come here from ufs_writepage or ufs_prepare_write, * locked_page is argument of these functions, so we already lock it. */ -static void ufs_change_blocknr(struct inode *inode, unsigned int baseblk, +static void ufs_change_blocknr(struct inode *inode, unsigned int beg, unsigned int count, unsigned int oldb, unsigned int newb, struct page *locked_page) { - unsigned int blk_per_page = 1 << (PAGE_CACHE_SHIFT - inode->i_blkbits); - struct address_space *mapping = inode->i_mapping; + const unsigned mask = (1 << (PAGE_CACHE_SHIFT - inode->i_blkbits)) - 1; + struct address_space * const mapping = inode->i_mapping; pgoff_t index, cur_index; - unsigned int i, j; + unsigned end, pos, j; struct page *page; struct buffer_head *head, *bh; @@ -246,8 +246,8 @@ static void ufs_change_blocknr(struct inode *inode, unsigned int baseblk, cur_index = locked_page->index; - for (i = 0; i < count; i += blk_per_page) { - index = (baseblk+i) >> (PAGE_CACHE_SHIFT - inode->i_blkbits); + for (end = count + beg; beg < end; beg = (beg | mask) + 1) { + index = beg >> (PAGE_CACHE_SHIFT - inode->i_blkbits); if (likely(cur_index != index)) { page = ufs_get_locked_page(mapping, index); @@ -256,21 +256,32 @@ static void ufs_change_blocknr(struct inode *inode, unsigned int baseblk, } else page = locked_page; - j = i; head = page_buffers(page); bh = head; + pos = beg & mask; + for (j = 0; j < pos; ++j) + bh = bh->b_this_page; + j = 0; do { - if (likely(bh->b_blocknr == j + oldb && j < count)) { - unmap_underlying_metadata(bh->b_bdev, - bh->b_blocknr); - bh->b_blocknr = newb + j++; - mark_buffer_dirty(bh); + if (buffer_mapped(bh)) { + pos = bh->b_blocknr - oldb; + if (pos < count) { + UFSD(" change from %llu to %llu\n", + (unsigned long long)pos + oldb, + (unsigned long long)pos + newb); + bh->b_blocknr = newb + pos; + unmap_underlying_metadata(bh->b_bdev, + bh->b_blocknr); + mark_buffer_dirty(bh); + ++j; + } } bh = bh->b_this_page; } while (bh != head); - set_page_dirty(page); + if (j) + set_page_dirty(page); if (likely(cur_index != index)) ufs_put_locked_page(page); @@ -418,14 +429,14 @@ unsigned ufs_new_fragments(struct inode * inode, __fs32 * p, unsigned fragment, } result = ufs_alloc_fragments (inode, cgno, goal, request, err); if (result) { + ufs_clear_frags(inode, result + oldcount, newcount - oldcount, + locked_page != NULL); ufs_change_blocknr(inode, fragment - oldcount, oldcount, tmp, result, locked_page); *p = cpu_to_fs32(sb, result); *err = 0; UFS_I(inode)->i_lastfrag = max_t(u32, UFS_I(inode)->i_lastfrag, fragment + count); - ufs_clear_frags(inode, result + oldcount, newcount - oldcount, - locked_page != NULL); unlock_super(sb); if (newcount < request) ufs_free_fragments (inode, result + newcount, request - newcount); From 3d8b3036f5260b366ec0982aa2f862d841d058c2 Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Mon, 29 Jan 2007 13:19:56 -0800 Subject: [PATCH 250/326] [PATCH] cdev.h: forward declarations Apparently this broke due to missing `struct inode' declaration. Signed-off-by: Jan Engelhardt Cc: Noah Watkins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/cdev.h | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/include/linux/cdev.h b/include/linux/cdev.h index f309b00e986e..1e29b13d0062 100644 --- a/include/linux/cdev.h +++ b/include/linux/cdev.h @@ -6,6 +6,10 @@ #include #include +struct file_operations; +struct inode; +struct module; + struct cdev { struct kobject kobj; struct module *owner; From 59df3230fc57fa8900bebf3d2d68221d549f3c7c Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Mon, 29 Jan 2007 13:47:01 +0100 Subject: [PATCH 251/326] [PATCH] `make help' in build tree doesn't show headers_* targets `make help' in the build tree doesn't show the help texts about the `headers_install' and `headers_check' targets because it looks for include/asm-$(ARCH)/Kbuild in the wrong place. Add the missing `$(srctree)' prefixes to fix this. Also move the printing of the default install path for the headers inside the `if/fi', where it belongs. Signed-off-by: Geert Uytterhoeven Acked-by: Oleg Verych Signed-off-by: Linus Torvalds --- Makefile | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/Makefile b/Makefile index 9e1adac8aa28..59aa9f56fdbe 100644 --- a/Makefile +++ b/Makefile @@ -1116,15 +1116,15 @@ help: @echo ' cscope - Generate cscope index' @echo ' kernelrelease - Output the release version string' @echo ' kernelversion - Output the version stored in Makefile' - @if [ -r include/asm-$(ARCH)/Kbuild ]; then \ + @if [ -r $(srctree)/include/asm-$(ARCH)/Kbuild ]; then \ echo ' headers_install - Install sanitised kernel headers to INSTALL_HDR_PATH'; \ + echo ' (default: $(INSTALL_HDR_PATH))'; \ fi - @echo ' (default: $(INSTALL_HDR_PATH))' @echo '' @echo 'Static analysers' @echo ' checkstack - Generate a list of stack hogs' @echo ' namespacecheck - Name space analysis on compiled kernel' - @if [ -r include/asm-$(ARCH)/Kbuild ]; then \ + @if [ -r $(srctree)/include/asm-$(ARCH)/Kbuild ]; then \ echo ' headers_check - Sanity check on exported headers'; \ fi @echo '' From 8339f0008c47cdd921c73f6d53d5588b5484f93c Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Mon, 29 Jan 2007 13:19:05 -0700 Subject: [PATCH 252/326] [PATCH] i386: In assign_irq_vector look at all vectors before giving up When the world was a simple and static place setting up irqs was easy. It sufficed to allocate a linux irq number and a find a free cpu vector we could receive that linux irq on. In those days it was a safe assumption that any allocated vector was actually in use so after one global pass through all of the vectors we would have none left. These days things are much more dynamic with interrupt controllers (in the form of MSI or MSI-X) appearing on plug in cards and linux irqs appearing and disappearing. As these irqs come and go vectors are allocated and freed, invalidating the ancient assumption that all allocated vectors stayed in use forever. So this patch modifies the vector allocator to walk through every possible vector before giving up, and to check to see if a vector is in use before assigning it. With these changes we stop leaking freed vectors and it becomes possible to allocate and free irq vectors all day long. This changed was modeled after the vector allocator on x86_64 where this limitation has already been removed. In essence we don't update the static variables that hold the position of the last vector we allocated until have successfully allocated another vector. This allows us to detect if we have completed one complete scan through all of the possible vectors. Acked-by: Auke Kok Signed-off-by: Eric W. Biederman Signed-off-by: Linus Torvalds --- arch/i386/kernel/io_apic.c | 32 +++++++++++++++++++------------- 1 file changed, 19 insertions(+), 13 deletions(-) diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c index 2424cc9c7b3d..6a3875f81a0a 100644 --- a/arch/i386/kernel/io_apic.c +++ b/arch/i386/kernel/io_apic.c @@ -1227,26 +1227,32 @@ static u8 irq_vector[NR_IRQ_VECTORS] __read_mostly = { FIRST_DEVICE_VECTOR , 0 } static int __assign_irq_vector(int irq) { - static int current_vector = FIRST_DEVICE_VECTOR, offset = 0; - int vector; + static int current_vector = FIRST_DEVICE_VECTOR, current_offset = 0; + int vector, offset, i; BUG_ON((unsigned)irq >= NR_IRQ_VECTORS); if (irq_vector[irq] > 0) return irq_vector[irq]; - current_vector += 8; - if (current_vector == SYSCALL_VECTOR) - current_vector += 8; - - if (current_vector >= FIRST_SYSTEM_VECTOR) { - offset++; - if (!(offset % 8)) - return -ENOSPC; - current_vector = FIRST_DEVICE_VECTOR + offset; - } - vector = current_vector; + offset = current_offset; +next: + vector += 8; + if (vector >= FIRST_SYSTEM_VECTOR) { + offset = (offset + 1) % 8; + vector = FIRST_DEVICE_VECTOR + offset; + } + if (vector == current_vector) + return -ENOSPC; + if (vector == SYSCALL_VECTOR) + goto next; + for (i = 0; i < NR_IRQ_VECTORS; i++) + if (irq_vector[i] == vector) + goto next; + + current_vector = vector; + current_offset = offset; irq_vector[irq] = vector; return vector; From 701dfbc1cbdd42b814dd76a885c4b73f97011d08 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Mon, 29 Jan 2007 21:24:08 +0000 Subject: [PATCH 253/326] [PATCH] mm: mremap correct rmap accounting Nick Piggin points out that page accounting on MIPS multiple ZERO_PAGEs is not maintained by its move_pte, and could lead to freeing a ZERO_PAGE. Instead of complicating that move_pte, just forget the minor optimization when mremapping, and change the one thing which needed it for correctness - filemap_xip use ZERO_PAGE(0) throughout instead of according to address. [ "There is no block device driver one could use for XIP on mips platforms" - Carsten Otte ] Signed-off-by: Hugh Dickins Cc: Nick Piggin Cc: Andrew Morton Cc: Ralf Baechle Cc: Carsten Otte Signed-off-by: Linus Torvalds --- include/asm-mips/pgtable.h | 10 ---------- mm/filemap_xip.c | 4 ++-- mm/mremap.c | 1 - 3 files changed, 2 insertions(+), 13 deletions(-) diff --git a/include/asm-mips/pgtable.h b/include/asm-mips/pgtable.h index f2e1325fec6c..3fcfd7979de5 100644 --- a/include/asm-mips/pgtable.h +++ b/include/asm-mips/pgtable.h @@ -69,16 +69,6 @@ extern unsigned long zero_page_mask; #define ZERO_PAGE(vaddr) \ (virt_to_page((void *)(empty_zero_page + (((unsigned long)(vaddr)) & zero_page_mask)))) -#define __HAVE_ARCH_MOVE_PTE -#define move_pte(pte, prot, old_addr, new_addr) \ -({ \ - pte_t newpte = (pte); \ - if (pte_present(pte) && pfn_valid(pte_pfn(pte)) && \ - pte_page(pte) == ZERO_PAGE(old_addr)) \ - newpte = mk_pte(ZERO_PAGE(new_addr), (prot)); \ - newpte; \ -}) - extern void paging_init(void); /* diff --git a/mm/filemap_xip.c b/mm/filemap_xip.c index 45b3553865cf..9dd9fbb75139 100644 --- a/mm/filemap_xip.c +++ b/mm/filemap_xip.c @@ -183,7 +183,7 @@ __xip_unmap (struct address_space * mapping, address = vma->vm_start + ((pgoff - vma->vm_pgoff) << PAGE_SHIFT); BUG_ON(address < vma->vm_start || address >= vma->vm_end); - page = ZERO_PAGE(address); + page = ZERO_PAGE(0); pte = page_check_address(page, mm, address, &ptl); if (pte) { /* Nuke the page table entry. */ @@ -246,7 +246,7 @@ xip_file_nopage(struct vm_area_struct * area, __xip_unmap(mapping, pgoff); } else { /* not shared and writable, use ZERO_PAGE() */ - page = ZERO_PAGE(address); + page = ZERO_PAGE(0); } out: diff --git a/mm/mremap.c b/mm/mremap.c index 9c769fa29f32..5d4bd4f95b8e 100644 --- a/mm/mremap.c +++ b/mm/mremap.c @@ -105,7 +105,6 @@ static void move_ptes(struct vm_area_struct *vma, pmd_t *old_pmd, if (pte_none(*old_pte)) continue; pte = ptep_clear_flush(vma, old_addr, old_pte); - /* ZERO_PAGE can be dependant on virtual addr */ pte = move_pte(pte, new_vma->vm_page_prot, old_addr, new_addr); set_pte_at(mm, new_addr, new_pte, pte); } From 89eb1693f975eddafd475782ee857d98f411d10d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 30 Jan 2007 13:23:25 +0000 Subject: [PATCH 254/326] [PATCH] missing exports of pm_power_off() on alpha and sparc32 Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- arch/alpha/kernel/process.c | 1 + arch/sparc/kernel/process.c | 1 + 2 files changed, 2 insertions(+) diff --git a/arch/alpha/kernel/process.c b/arch/alpha/kernel/process.c index 3370e6faeae0..c15186390693 100644 --- a/arch/alpha/kernel/process.c +++ b/arch/alpha/kernel/process.c @@ -47,6 +47,7 @@ * Power off function, if any */ void (*pm_power_off)(void) = machine_power_off; +EXPORT_SYMBOL(pm_power_off); void cpu_idle(void) diff --git a/arch/sparc/kernel/process.c b/arch/sparc/kernel/process.c index 89a28cc018c9..113bd48a89bd 100644 --- a/arch/sparc/kernel/process.c +++ b/arch/sparc/kernel/process.c @@ -54,6 +54,7 @@ void (*pm_idle)(void); * handler when auxio is not present-- unused for now... */ void (*pm_power_off)(void) = machine_power_off; +EXPORT_SYMBOL(pm_power_off); /* * sysctl - toggle power-off restriction for serial console From a12743026ceb34388b9b983801ba99240cb0a199 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 30 Jan 2007 13:23:30 +0000 Subject: [PATCH 255/326] [PATCH] mtd/nand/cafe.c missing include of dma-mapping.h Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- drivers/mtd/nand/cafe.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/mtd/nand/cafe.c b/drivers/mtd/nand/cafe.c index b8d9b64cccc0..65f9bd3ceebf 100644 --- a/drivers/mtd/nand/cafe.c +++ b/drivers/mtd/nand/cafe.c @@ -14,6 +14,7 @@ #include #include #include +#include #include #define CAFE_NAND_CTRL1 0x00 From 9a696b4f3c5664c2c4815cc8e37256e51a3425e1 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 30 Jan 2007 13:23:35 +0000 Subject: [PATCH 256/326] [PATCH] sym53c500_cs: remove bogus call fo free_dma() What DMA for 16bit pcmcia card, anyway? We never do request_dma() there and ->dma_channel never changes since initialization to -1. IOW, that call is dead code. Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- drivers/scsi/pcmcia/sym53c500_cs.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/scsi/pcmcia/sym53c500_cs.c b/drivers/scsi/pcmcia/sym53c500_cs.c index 9fb0ea5c1fb9..5b458d2478f7 100644 --- a/drivers/scsi/pcmcia/sym53c500_cs.c +++ b/drivers/scsi/pcmcia/sym53c500_cs.c @@ -545,8 +545,6 @@ SYM53C500_release(struct pcmcia_device *link) */ if (shost->irq) free_irq(shost->irq, shost); - if (shost->dma_channel != 0xff) - free_dma(shost->dma_channel); if (shost->io_port && shost->n_io_port) release_region(shost->io_port, shost->n_io_port); From 161c888b0b4d28775dfe35274ee90c16a91b4365 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 30 Jan 2007 13:23:45 +0000 Subject: [PATCH 257/326] [PATCH] pata_platform: fallout from set_mode() change Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- drivers/ata/pata_platform.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/ata/pata_platform.c b/drivers/ata/pata_platform.c index 443b1d85c6c4..40ae11cbfda4 100644 --- a/drivers/ata/pata_platform.c +++ b/drivers/ata/pata_platform.c @@ -30,7 +30,7 @@ static int pio_mask = 1; * Provide our own set_mode() as we don't want to change anything that has * already been configured.. */ -static void pata_platform_set_mode(struct ata_port *ap) +static int pata_platform_set_mode(struct ata_port *ap, struct ata_device **unused) { int i; @@ -44,6 +44,7 @@ static void pata_platform_set_mode(struct ata_port *ap) dev->flags |= ATA_DFLAG_PIO; } } + return 0; } static void pata_platform_host_stop(struct ata_host *host) From d0a23398eeaa374eb0c0435c5a259ffde77c8af0 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 30 Jan 2007 13:23:55 +0000 Subject: [PATCH 258/326] [PATCH] missing dma_sync_single_range_for{cpu,device} on alpha no-op as all dma_sync_... there. Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- include/asm-alpha/dma-mapping.h | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) diff --git a/include/asm-alpha/dma-mapping.h b/include/asm-alpha/dma-mapping.h index 57e09f5e3424..1919d72f4c7a 100644 --- a/include/asm-alpha/dma-mapping.h +++ b/include/asm-alpha/dma-mapping.h @@ -55,12 +55,14 @@ int dma_map_sg(struct device *dev, struct scatterlist *sg, int nents, int dma_set_mask(struct device *dev, u64 mask); -#define dma_sync_single_for_cpu(dev, addr, size, dir) do { } while (0) -#define dma_sync_single_for_device(dev, addr, size, dir) do { } while (0) -#define dma_sync_single_range(dev, addr, off, size, dir) do { } while (0) -#define dma_sync_sg_for_cpu(dev, sg, nents, dir) do { } while (0) -#define dma_sync_sg_for_device(dev, sg, nents, dir) do { } while (0) -#define dma_cache_sync(dev, va, size, dir) do { } while (0) +#define dma_sync_single_for_cpu(dev, addr, size, dir) ((void)0) +#define dma_sync_single_for_device(dev, addr, size, dir) ((void)0) +#define dma_sync_single_range(dev, addr, off, size, dir) ((void)0) +#define dma_sync_sg_for_cpu(dev, sg, nents, dir) ((void)0) +#define dma_sync_sg_for_device(dev, sg, nents, dir) ((void)0) +#define dma_cache_sync(dev, va, size, dir) ((void)0) +#define dma_sync_single_range_for_cpu(dev, addr, offset, size, dir) ((void)0) +#define dma_sync_single_range_for_device(dev, addr, offset, size, dir) ((void)0) #define dma_get_cache_alignment() L1_CACHE_BYTES From 6a13f66043cd755fc7e19f8e0fbb5bfcdf470e74 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 30 Jan 2007 13:24:00 +0000 Subject: [PATCH 259/326] [PATCH] dma-mapping.h stubs fix do { } while(0) is not a good imitation of function returning void; use ((void)0) instead. Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- include/asm-alpha/dma-mapping.h | 6 +++--- include/asm-powerpc/dma-mapping.h | 12 ++++++------ 2 files changed, 9 insertions(+), 9 deletions(-) diff --git a/include/asm-alpha/dma-mapping.h b/include/asm-alpha/dma-mapping.h index 1919d72f4c7a..75a1aff5b57b 100644 --- a/include/asm-alpha/dma-mapping.h +++ b/include/asm-alpha/dma-mapping.h @@ -41,9 +41,9 @@ int dma_map_sg(struct device *dev, struct scatterlist *sg, int nents, #define dma_map_single(dev, va, size, dir) virt_to_phys(va) #define dma_map_page(dev, page, off, size, dir) (page_to_pa(page) + off) -#define dma_unmap_single(dev, addr, size, dir) do { } while (0) -#define dma_unmap_page(dev, addr, size, dir) do { } while (0) -#define dma_unmap_sg(dev, sg, nents, dir) do { } while (0) +#define dma_unmap_single(dev, addr, size, dir) ((void)0) +#define dma_unmap_page(dev, addr, size, dir) ((void)0) +#define dma_unmap_sg(dev, sg, nents, dir) ((void)0) #define dma_mapping_error(addr) (0) diff --git a/include/asm-powerpc/dma-mapping.h b/include/asm-powerpc/dma-mapping.h index 7c7de87bd8ae..a19a6f1a1cf1 100644 --- a/include/asm-powerpc/dma-mapping.h +++ b/include/asm-powerpc/dma-mapping.h @@ -37,9 +37,9 @@ extern void __dma_sync_page(struct page *page, unsigned long offset, */ #define __dma_alloc_coherent(gfp, size, handle) NULL -#define __dma_free_coherent(size, addr) do { } while (0) -#define __dma_sync(addr, size, rw) do { } while (0) -#define __dma_sync_page(pg, off, sz, rw) do { } while (0) +#define __dma_free_coherent(size, addr) ((void)0) +#define __dma_sync(addr, size, rw) ((void)0) +#define __dma_sync_page(pg, off, sz, rw) ((void)0) #endif /* ! CONFIG_NOT_COHERENT_CACHE */ @@ -251,7 +251,7 @@ dma_map_single(struct device *dev, void *ptr, size_t size, } /* We do nothing. */ -#define dma_unmap_single(dev, addr, size, dir) do { } while (0) +#define dma_unmap_single(dev, addr, size, dir) ((void)0) static inline dma_addr_t dma_map_page(struct device *dev, struct page *page, @@ -266,7 +266,7 @@ dma_map_page(struct device *dev, struct page *page, } /* We do nothing. */ -#define dma_unmap_page(dev, handle, size, dir) do { } while (0) +#define dma_unmap_page(dev, handle, size, dir) ((void)0) static inline int dma_map_sg(struct device *dev, struct scatterlist *sg, int nents, @@ -286,7 +286,7 @@ dma_map_sg(struct device *dev, struct scatterlist *sg, int nents, } /* We don't do anything here. */ -#define dma_unmap_sg(dev, sg, nents, dir) do { } while (0) +#define dma_unmap_sg(dev, sg, nents, dir) ((void)0) #endif /* CONFIG_PPC64 */ From 774ba59c950926abd27421fa146f1916de89e6f6 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 30 Jan 2007 13:23:50 +0000 Subject: [PATCH 260/326] [PATCH] b44: src_desc->addr is little-endian Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- drivers/net/b44.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/b44.c b/drivers/net/b44.c index 5eb2ec68393f..cfc89bc105b1 100644 --- a/drivers/net/b44.c +++ b/drivers/net/b44.c @@ -752,7 +752,7 @@ static void b44_recycle_rx(struct b44 *bp, int src_idx, u32 dest_idx_unmasked) dest_idx * sizeof(dest_desc), DMA_BIDIRECTIONAL); - pci_dma_sync_single_for_device(bp->pdev, src_desc->addr, + pci_dma_sync_single_for_device(bp->pdev, le32_to_cpu(src_desc->addr), RX_PKT_BUF_SZ, PCI_DMA_FROMDEVICE); } From b20c8453a7d6e8d63c0f05ba3e38d9428c8091b1 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 30 Jan 2007 13:23:40 +0000 Subject: [PATCH 261/326] [PATCH] fix indentation-related breakage in Kconfig.i386 Kconfig recognizes the end of help text by receding indentation depth. Recent patch had broken HOST_VMSPLIT_... choice in arch/um/Kconfig.i386 - all alternatives are interpreted as part of help text now. Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- arch/um/Kconfig.i386 | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/arch/um/Kconfig.i386 b/arch/um/Kconfig.i386 index 77558a88a2fe..d6cffb27fff8 100644 --- a/arch/um/Kconfig.i386 +++ b/arch/um/Kconfig.i386 @@ -36,14 +36,14 @@ choice So, if you do not know what to do here, say 'Default split'. - config HOST_VMSPLIT_3G - bool "Default split (3G/1G user/kernel host split)" - config HOST_VMSPLIT_3G_OPT - bool "3G/1G user/kernel host split (for full 1G low memory)" - config HOST_VMSPLIT_2G - bool "2G/2G user/kernel host split" - config HOST_VMSPLIT_1G - bool "1G/3G user/kernel host split" +config HOST_VMSPLIT_3G + bool "Default split (3G/1G user/kernel host split)" +config HOST_VMSPLIT_3G_OPT + bool "3G/1G user/kernel host split (for full 1G low memory)" +config HOST_VMSPLIT_2G + bool "2G/2G user/kernel host split" +config HOST_VMSPLIT_1G + bool "1G/3G user/kernel host split" endchoice config TOP_ADDR From 444f378b237a0f728f5c4aba752c08d13c209344 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 30 Jan 2007 13:35:18 -0800 Subject: [PATCH 262/326] Revert "[PATCH] namespaces: fix exit race by splitting exit" This reverts commit 7a238fcba0629b6f2edbcd37458bae56fcf36be5 in preparation for a better and simpler fix proposed by Eric Biederman (and fixed up by Serge Hallyn) Acked-by: Serge E. Hallyn Signed-off-by: Linus Torvalds --- include/linux/nsproxy.h | 30 +++++++++++------------------- kernel/exit.c | 6 ++---- kernel/fork.c | 4 ++-- kernel/nsproxy.c | 16 +--------------- 4 files changed, 16 insertions(+), 40 deletions(-) diff --git a/include/linux/nsproxy.h b/include/linux/nsproxy.h index 678e1d38effb..0b9f0dc30d61 100644 --- a/include/linux/nsproxy.h +++ b/include/linux/nsproxy.h @@ -35,30 +35,22 @@ struct nsproxy *dup_namespaces(struct nsproxy *orig); int copy_namespaces(int flags, struct task_struct *tsk); void get_task_namespaces(struct task_struct *tsk); void free_nsproxy(struct nsproxy *ns); -struct nsproxy *put_nsproxy(struct nsproxy *ns); -static inline void finalize_put_nsproxy(struct nsproxy *ns) +static inline void put_nsproxy(struct nsproxy *ns) { - if (ns) + if (atomic_dec_and_test(&ns->count)) { free_nsproxy(ns); + } } -static inline void put_and_finalize_nsproxy(struct nsproxy *ns) +static inline void exit_task_namespaces(struct task_struct *p) { - finalize_put_nsproxy(put_nsproxy(ns)); -} - -static inline struct nsproxy *preexit_task_namespaces(struct task_struct *p) -{ - return put_nsproxy(p->nsproxy); -} - -static inline void exit_task_namespaces(struct task_struct *p, - struct nsproxy *ns) -{ - task_lock(p); - p->nsproxy = NULL; - task_unlock(p); - finalize_put_nsproxy(ns); + struct nsproxy *ns = p->nsproxy; + if (ns) { + task_lock(p); + p->nsproxy = NULL; + task_unlock(p); + put_nsproxy(ns); + } } #endif diff --git a/kernel/exit.c b/kernel/exit.c index a5bf5329ff97..35401720635b 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -396,7 +396,7 @@ void daemonize(const char *name, ...) current->fs = fs; atomic_inc(&fs->count); - put_and_finalize_nsproxy(current->nsproxy); + exit_task_namespaces(current); current->nsproxy = init_task.nsproxy; get_task_namespaces(current); @@ -853,7 +853,6 @@ static void exit_notify(struct task_struct *tsk) fastcall NORET_TYPE void do_exit(long code) { struct task_struct *tsk = current; - struct nsproxy *ns; int group_dead; profile_task_exit(tsk); @@ -939,9 +938,8 @@ fastcall NORET_TYPE void do_exit(long code) tsk->exit_code = code; proc_exit_connector(tsk); - ns = preexit_task_namespaces(tsk); exit_notify(tsk); - exit_task_namespaces(tsk, ns); + exit_task_namespaces(tsk); #ifdef CONFIG_NUMA mpol_free(tsk->mempolicy); tsk->mempolicy = NULL; diff --git a/kernel/fork.c b/kernel/fork.c index 4cf868458f06..fc723e595cd5 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1265,7 +1265,7 @@ static struct task_struct *copy_process(unsigned long clone_flags, return p; bad_fork_cleanup_namespaces: - put_and_finalize_nsproxy(p->nsproxy); + exit_task_namespaces(p); bad_fork_cleanup_keys: exit_keys(p); bad_fork_cleanup_mm: @@ -1711,7 +1711,7 @@ asmlinkage long sys_unshare(unsigned long unshare_flags) } if (new_nsproxy) - put_and_finalize_nsproxy(new_nsproxy); + put_nsproxy(new_nsproxy); bad_unshare_cleanup_ipc: if (new_ipc) diff --git a/kernel/nsproxy.c b/kernel/nsproxy.c index 7b05bce75cde..f5b9ee6f6bbb 100644 --- a/kernel/nsproxy.c +++ b/kernel/nsproxy.c @@ -117,7 +117,7 @@ int copy_namespaces(int flags, struct task_struct *tsk) goto out_pid; out: - put_and_finalize_nsproxy(old_ns); + put_nsproxy(old_ns); return err; out_pid: @@ -135,20 +135,6 @@ int copy_namespaces(int flags, struct task_struct *tsk) goto out; } -struct nsproxy *put_nsproxy(struct nsproxy *ns) -{ - if (ns) { - if (atomic_dec_and_test(&ns->count)) { - if (ns->mnt_ns) { - put_mnt_ns(ns->mnt_ns); - ns->mnt_ns = NULL; - } - return ns; - } - } - return NULL; -} - void free_nsproxy(struct nsproxy *ns) { if (ns->mnt_ns) From 0f2452855d86901ba3766826ccb5606ea4e15ab9 Mon Sep 17 00:00:00 2001 From: "Serge E. Hallyn" Date: Tue, 30 Jan 2007 15:28:23 -0600 Subject: [PATCH 263/326] [PATCH] namespaces: fix task exit disaster This is based on a patch by Eric W. Biederman, who pointed out that pid namespaces are still fake, and we only have one ever active. So for the time being, we can modify any code which could access tsk->nsproxy->pid_ns during task exit to just use &init_pid_ns instead, and move the exit_task_namespaces call in do_exit() back above exit_notify(), so that an exiting nfs server has a valid tsk->sighand to work with. Long term, pulling pid_ns out of nsproxy might be the cleanest solution. Signed-off-by: Eric W. Biederman [ Eric's patch fixed to take care of free_pid() too ] Signed-off-by: Serge E. Hallyn Signed-off-by: Linus Torvalds --- include/linux/pid_namespace.h | 2 +- kernel/exit.c | 2 +- kernel/pid.c | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/include/linux/pid_namespace.h b/include/linux/pid_namespace.h index d2a9d419f01f..2833806d42c6 100644 --- a/include/linux/pid_namespace.h +++ b/include/linux/pid_namespace.h @@ -39,7 +39,7 @@ static inline void put_pid_ns(struct pid_namespace *ns) static inline struct task_struct *child_reaper(struct task_struct *tsk) { - return tsk->nsproxy->pid_ns->child_reaper; + return init_pid_ns.child_reaper; } #endif /* _LINUX_PID_NS_H */ diff --git a/kernel/exit.c b/kernel/exit.c index 35401720635b..fec12eb12471 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -938,8 +938,8 @@ fastcall NORET_TYPE void do_exit(long code) tsk->exit_code = code; proc_exit_connector(tsk); - exit_notify(tsk); exit_task_namespaces(tsk); + exit_notify(tsk); #ifdef CONFIG_NUMA mpol_free(tsk->mempolicy); tsk->mempolicy = NULL; diff --git a/kernel/pid.c b/kernel/pid.c index 2efe9d8d367b..78f2aee90f54 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -197,7 +197,7 @@ fastcall void free_pid(struct pid *pid) hlist_del_rcu(&pid->pid_chain); spin_unlock_irqrestore(&pidmap_lock, flags); - free_pidmap(current->nsproxy->pid_ns, pid->nr); + free_pidmap(&init_pid_ns, pid->nr); call_rcu(&pid->rcu, delayed_put_pid); } From bcdddfb66cc998252d34758ce4109cedc0d24a5c Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 30 Jan 2007 14:11:12 -0800 Subject: [PATCH 264/326] Revert "net: ifb error path loop fix" This reverts commit 0c0b3ae68ec93b1db5c637d294647d1cca0df763. Quoth David: "Jeff, please revert It's wrong. We had a lengthy analysis of this piece of code several months ago, and it is correct. Consider, if we run the loop and we get an error the following happens: 1) attempt of ifb_init_one(i) fails, therefore we should not try to "ifb_free_one()" on "i" since it failed 2) the loop iteration first increments "i", then it check for error Therefore we must decrement "i" twice before the first free during the cleanup. One to "undo" the for() loop increment, and one to "skip" the ifb_init_one() case which failed." Reported-by: David Miller Acked-by: Jeff Garzik Cc: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/net/ifb.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/ifb.c b/drivers/net/ifb.c index c4ca7c930a46..ca2b21f9d444 100644 --- a/drivers/net/ifb.c +++ b/drivers/net/ifb.c @@ -271,7 +271,8 @@ static int __init ifb_init_module(void) for (i = 0; i < numifbs && !err; i++) err = ifb_init_one(i); if (err) { - while (i--) + i--; + while (--i >= 0) ifb_free_one(i); } From 24a1dec55073000264f2da6278baef759929a14f Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sun, 28 Jan 2007 15:54:42 -0800 Subject: [PATCH 265/326] [MAINTAINERS]: netfilter@ is subscribers-only netfilter mailing list is subscribers-only. Signed-off-by: Randy Dunlap Signed-off-by: David S. Miller --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 1446cc41c12f..a275f72ed5f7 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2288,7 +2288,7 @@ P: Jozsef Kadlecsik P: Patrick McHardy M: kaber@trash.net L: netfilter-devel@lists.netfilter.org -L: netfilter@lists.netfilter.org +L: netfilter@lists.netfilter.org (subscribers-only) L: coreteam@netfilter.org W: http://www.netfilter.org/ W: http://www.iptables.org/ From fb74a8416022c033e1a950689c264c453f8f98d8 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 30 Jan 2007 14:24:29 -0800 Subject: [PATCH 266/326] [NETFILTER]: xt_connbytes: fix division by zero When the packet counter of a connection is zero a division by zero occurs in div64_64(). Fix that by using zero as average value, which is correct as long as the packet counter didn't overflow, at which point we have lost anyway. Additionally we're probably going to go back to 64 bit counters in 2.6.21. Based on patch from Jonas Berlin , with suggestions from KOVACS Krisztian . Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/netfilter/xt_connbytes.c | 29 ++++++++++++----------------- 1 file changed, 12 insertions(+), 17 deletions(-) diff --git a/net/netfilter/xt_connbytes.c b/net/netfilter/xt_connbytes.c index d93cb096a675..5e32dfa2668b 100644 --- a/net/netfilter/xt_connbytes.c +++ b/net/netfilter/xt_connbytes.c @@ -52,6 +52,8 @@ match(const struct sk_buff *skb, { const struct xt_connbytes_info *sinfo = matchinfo; u_int64_t what = 0; /* initialize to make gcc happy */ + u_int64_t bytes = 0; + u_int64_t pkts = 0; const struct ip_conntrack_counter *counters; if (!(counters = nf_ct_get_counters(skb))) @@ -89,29 +91,22 @@ match(const struct sk_buff *skb, case XT_CONNBYTES_AVGPKT: switch (sinfo->direction) { case XT_CONNBYTES_DIR_ORIGINAL: - what = div64_64(counters[IP_CT_DIR_ORIGINAL].bytes, - counters[IP_CT_DIR_ORIGINAL].packets); + bytes = counters[IP_CT_DIR_ORIGINAL].bytes; + pkts = counters[IP_CT_DIR_ORIGINAL].packets; break; case XT_CONNBYTES_DIR_REPLY: - what = div64_64(counters[IP_CT_DIR_REPLY].bytes, - counters[IP_CT_DIR_REPLY].packets); + bytes = counters[IP_CT_DIR_REPLY].bytes; + pkts = counters[IP_CT_DIR_REPLY].packets; break; case XT_CONNBYTES_DIR_BOTH: - { - u_int64_t bytes; - u_int64_t pkts; - bytes = counters[IP_CT_DIR_ORIGINAL].bytes + - counters[IP_CT_DIR_REPLY].bytes; - pkts = counters[IP_CT_DIR_ORIGINAL].packets+ - counters[IP_CT_DIR_REPLY].packets; - - /* FIXME_THEORETICAL: what to do if sum - * overflows ? */ - - what = div64_64(bytes, pkts); - } + bytes = counters[IP_CT_DIR_ORIGINAL].bytes + + counters[IP_CT_DIR_REPLY].bytes; + pkts = counters[IP_CT_DIR_ORIGINAL].packets + + counters[IP_CT_DIR_REPLY].packets; break; } + if (pkts != 0) + what = div64_64(bytes, pkts); break; } From 7da5bfbb12e327b3a347ee3e076957cd6564eb56 Mon Sep 17 00:00:00 2001 From: Lars Immisch Date: Tue, 30 Jan 2007 14:24:57 -0800 Subject: [PATCH 267/326] [NETFILTER]: SIP conntrack: fix skipping over user info in SIP headers When trying to skip over the username in the Contact header, stop at the end of the line if no @ is found to avoid mangling following headers. We don't need to worry about continuation lines because we search inside a SIP URI. Fixes Netfilter Bugzilla #532. Signed-off-by: Lars Immisch Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/ipv4/netfilter/ip_conntrack_sip.c | 8 +++++++- net/netfilter/nf_conntrack_sip.c | 8 +++++++- 2 files changed, 14 insertions(+), 2 deletions(-) diff --git a/net/ipv4/netfilter/ip_conntrack_sip.c b/net/ipv4/netfilter/ip_conntrack_sip.c index 3a26d63eed88..571d27e20910 100644 --- a/net/ipv4/netfilter/ip_conntrack_sip.c +++ b/net/ipv4/netfilter/ip_conntrack_sip.c @@ -283,8 +283,14 @@ static int skp_epaddr_len(const char *dptr, const char *limit, int *shift) { int s = *shift; - for (; dptr <= limit && *dptr != '@'; dptr++) + /* Search for @, but stop at the end of the line. + * We are inside a sip: URI, so we don't need to worry about + * continuation lines. */ + while (dptr <= limit && + *dptr != '@' && *dptr != '\r' && *dptr != '\n') { (*shift)++; + dptr++; + } if (*dptr == '@') { dptr++; diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index eb2a2411f97b..c93fb37a54fc 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -303,8 +303,14 @@ static int skp_epaddr_len(struct nf_conn *ct, const char *dptr, { int s = *shift; - for (; dptr <= limit && *dptr != '@'; dptr++) + /* Search for @, but stop at the end of the line. + * We are inside a sip: URI, so we don't need to worry about + * continuation lines. */ + while (dptr <= limit && + *dptr != '@' && *dptr != '\r' && *dptr != '\n') { (*shift)++; + dptr++; + } if (*dptr == '@') { dptr++; From adcb4711101dfef89d473f64a913089d303962ae Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 30 Jan 2007 14:25:24 -0800 Subject: [PATCH 268/326] [NETFILTER]: SIP conntrack: fix out of bounds memory access When checking for an @-sign in skp_epaddr_len, make sure not to run over the packet boundaries. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/ipv4/netfilter/ip_conntrack_sip.c | 2 +- net/netfilter/nf_conntrack_sip.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/net/ipv4/netfilter/ip_conntrack_sip.c b/net/ipv4/netfilter/ip_conntrack_sip.c index 571d27e20910..11c588a10e6b 100644 --- a/net/ipv4/netfilter/ip_conntrack_sip.c +++ b/net/ipv4/netfilter/ip_conntrack_sip.c @@ -292,7 +292,7 @@ static int skp_epaddr_len(const char *dptr, const char *limit, int *shift) dptr++; } - if (*dptr == '@') { + if (dptr <= limit && *dptr == '@') { dptr++; (*shift)++; } else diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index c93fb37a54fc..9dec11534678 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -312,7 +312,7 @@ static int skp_epaddr_len(struct nf_conn *ct, const char *dptr, dptr++; } - if (*dptr == '@') { + if (dptr <= limit && *dptr == '@') { dptr++; (*shift)++; } else From fa03ef38e1516c1f35f6a189100186dded0f8f8c Mon Sep 17 00:00:00 2001 From: Neil Horman Date: Tue, 30 Jan 2007 14:30:10 -0800 Subject: [PATCH 269/326] [IPV6]: Fix up some CONFIG typos Signed-off-by: Neil Horman Signed-off-by: David S. Miller --- net/ipv6/addrconf.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 2a7e4618f526..e3854696988d 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -3393,7 +3393,7 @@ static void inline ipv6_store_devconf(struct ipv6_devconf *cnf, #ifdef CONFIG_IPV6_ROUTER_PREF array[DEVCONF_ACCEPT_RA_RTR_PREF] = cnf->accept_ra_rtr_pref; array[DEVCONF_RTR_PROBE_INTERVAL] = cnf->rtr_probe_interval; -#ifdef CONFIV_IPV6_ROUTE_INFO +#ifdef CONFIG_IPV6_ROUTE_INFO array[DEVCONF_ACCEPT_RA_RT_INFO_MAX_PLEN] = cnf->accept_ra_rt_info_max_plen; #endif #endif @@ -3898,7 +3898,7 @@ static struct addrconf_sysctl_table .proc_handler = &proc_dointvec_jiffies, .strategy = &sysctl_jiffies, }, -#ifdef CONFIV_IPV6_ROUTE_INFO +#ifdef CONFIG_IPV6_ROUTE_INFO { .ctl_name = NET_IPV6_ACCEPT_RA_RT_INFO_MAX_PLEN, .procname = "accept_ra_rt_info_max_plen", From 29556526b970c2e7d4ca808b6082c33981adfdff Mon Sep 17 00:00:00 2001 From: Li Yewang Date: Tue, 30 Jan 2007 14:33:20 -0800 Subject: [PATCH 270/326] [IPV6]: fix BUG of ndisc_send_redirect() When I tested IPv6 redirect function about kernel 2.6.19.1, and found that the kernel can send redirect packets whose target address is global address, and the target is not the actual endpoint of communication. But the criteria conform to RFC2461, the target address defines as following: Target Address An IP address that is a better first hop to use for he ICMP Destination Address. When the target is the actual endpoint of communication, i.e., the destination is a neighbor, the Target Address field MUST contain the same value as the ICMP Destination Address field. Otherwise the target is a better first-hop router and the Target Address MUST be the router's link-local address so that hosts can uniquely identify routers. According to this definition, when a router redirect to a host, the target address either the better first-hop router's link-local address or the same as the ICMP destination address field. But the function of ndisc_send_redirect() in net/ipv6/ndisc.c, does not check the target address correctly. There is another definition about receive Redirect message in RFC2461: 8.1. Validation of Redirect Messages A host MUST silently discard any received Redirect message that does not satisfy all of the following validity checks: ...... - The ICMP Target Address is either a link-local address (when redirected to a router) or the same as the ICMP Destination Address (when redirected to the on-link destination). ...... And the receive redirect function of ndisc_redirect_rcv() implemented this definition, checks the target address correctly. if (ipv6_addr_equal(dest, target)) { on_link = 1; } else if (!(ipv6_addr_type(target) & IPV6_ADDR_LINKLOCAL)) { ND_PRINTK2(KERN_WARNING "ICMPv6 Redirect: target address is not link-local.\n"); return; } So, I think the send redirect function must check the target address also. Signed-off-by: Li Yewang Acked-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller --- net/ipv6/ndisc.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 6a9f616de37d..39bb658f3c44 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -1413,6 +1413,13 @@ void ndisc_send_redirect(struct sk_buff *skb, struct neighbour *neigh, return; } + if (!ipv6_addr_equal(&skb->nh.ipv6h->daddr, target) && + !(ipv6_addr_type(target) & IPV6_ADDR_LINKLOCAL)) { + ND_PRINTK2(KERN_WARNING + "ICMPv6 Redirect: target address is not link-local.\n"); + return; + } + ndisc_flow_init(&fl, NDISC_REDIRECT, &saddr_buf, &skb->nh.ipv6h->saddr, dev->ifindex); From e533ca16f31f9e5abfaf5d8c7dbe7095f01474b6 Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Tue, 30 Jan 2007 14:36:14 -0800 Subject: [PATCH 271/326] [SCTP]: Force update of the rto when processing HB-ACK When processing a HEARTBEAT-ACK it's possible that the transport rto timers will not be updated because a prior T3-RTX processing would have cleared the rto_pending flag on the transport. However, if we received a valid HEARTBEAT-ACK, we want to force update the rto variables, so re-set the rto_pending flag before calling sctp_transport_update_rto(). Signed-off-by: Vlad Yasevich Signed-off-by: Sridhar Samudrala Signed-off-by: David S. Miller --- net/sctp/sm_sideeffect.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c index 8bd30976cdee..6db77d1329f7 100644 --- a/net/sctp/sm_sideeffect.c +++ b/net/sctp/sm_sideeffect.c @@ -621,7 +621,13 @@ static void sctp_cmd_transport_on(sctp_cmd_seq_t *cmds, /* The receiver of the HEARTBEAT ACK should also perform an * RTT measurement for that destination transport address * using the time value carried in the HEARTBEAT ACK chunk. + * If the transport's rto_pending variable has been cleared, + * it was most likely due to a retransmit. However, we want + * to re-enable it to properly update the rto. */ + if (t->rto_pending == 0) + t->rto_pending = 1; + hbinfo = (sctp_sender_hb_info_t *) chunk->skb->data; sctp_transport_update_rto(t, (jiffies - hbinfo->sent_at)); From 0d59a01bc461bbab4017ff449b8401151ef44cf6 Mon Sep 17 00:00:00 2001 From: Adam Litke Date: Tue, 30 Jan 2007 14:35:39 -0800 Subject: [PATCH 272/326] [PATCH] Don't allow the stack to grow into hugetlb reserved regions When expanding the stack, we don't currently check if the VMA will cross into an area of the address space that is reserved for hugetlb pages. Subsequent faults on the expanded portion of such a VMA will confuse the low-level MMU code, resulting in an OOPS. Check for this. Signed-off-by: Adam Litke Cc: David Gibson Cc: William Lee Irwin III Cc: Hugh Dickins Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/mmap.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/mm/mmap.c b/mm/mmap.c index 9717337293c3..cc3a20819457 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1477,6 +1477,7 @@ static int acct_stack_growth(struct vm_area_struct * vma, unsigned long size, un { struct mm_struct *mm = vma->vm_mm; struct rlimit *rlim = current->signal->rlim; + unsigned long new_start; /* address space limit tests */ if (!may_expand_vm(mm, grow)) @@ -1496,6 +1497,12 @@ static int acct_stack_growth(struct vm_area_struct * vma, unsigned long size, un return -ENOMEM; } + /* Check to ensure the stack will not grow into a hugetlb-only region */ + new_start = (vma->vm_flags & VM_GROWSUP) ? vma->vm_start : + vma->vm_end - size; + if (is_hugepage_only_range(vma->vm_mm, new_start, size)) + return -EFAULT; + /* * Overcommit.. This must be the final test, as it will * update security statistics. From ec268be36e79d99443efc4be3a617d6efc6f719b Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Tue, 30 Jan 2007 14:35:55 -0800 Subject: [PATCH 273/326] [PATCH] translate dashes in filenames for headers install The current filename->define translation does not scrub dashes so when creating stub defines for like asm-x86_64/ptrace-abi.h, we get: #define __ASM_STUB_PTRACE-ABI_H gcc just hates that sort of thing :) trivial attached patch adds - to the tr list to scrub it to _ Signed-off-by: Mike Frysinger Cc: David Woodhouse Cc: Sam Ravnborg Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- scripts/Makefile.headersinst | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/Makefile.headersinst b/scripts/Makefile.headersinst index 4241e0dfeeaf..f7b6705fd6a3 100644 --- a/scripts/Makefile.headersinst +++ b/scripts/Makefile.headersinst @@ -109,7 +109,7 @@ quiet_cmd_mkdir = MKDIR $(patsubst $(INSTALL_HDR_PATH)/%,%,$@) quiet_cmd_gen = GEN $(patsubst $(INSTALL_HDR_PATH)/%,%,$@) cmd_gen = \ FNAME=$(patsubst $(INSTALL_HDR_PATH)/$(_dst)/%,%,$@) \ -STUBDEF=__ASM_STUB_`echo $$FNAME | tr a-z. A-Z_`; \ +STUBDEF=__ASM_STUB_`echo $$FNAME | tr a-z.- A-Z__`; \ (echo "/* File autogenerated by 'make headers_install' */" ; \ echo "\#ifndef $$STUBDEF" ; \ echo "\#define $$STUBDEF" ; \ From 46bae1a9a767f3ae8e636d96f9b95703df34b398 Mon Sep 17 00:00:00 2001 From: Neil Brown Date: Tue, 30 Jan 2007 14:36:01 -0800 Subject: [PATCH 274/326] [PATCH] Remove warning: VFS is out of sync with lock manager But keep it as a dprintk The message can be generated in a quite normal situation: If a 'lock' request is interrupted, then the lock client needs to record that the server has the lock, incase it does. When we come the unlock, the server might say it doesn't, even though we think it does (or might) and this generates the message. Signed-off-by: Neil Brown Acked-by: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/nfs/file.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/fs/nfs/file.c b/fs/nfs/file.c index fab20d06d936..9e4a2b70995a 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -434,8 +434,9 @@ static int do_vfs_lock(struct file *file, struct file_lock *fl) BUG(); } if (res < 0) - printk(KERN_WARNING "%s: VFS is out of sync with lock manager!\n", - __FUNCTION__); + dprintk(KERN_WARNING "%s: VFS is out of sync with lock manager" + " - error %d!\n", + __FUNCTION__, res); return res; } From ab40c5c6b6861ee71fd97f2611027b01e9ec4da0 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Tue, 30 Jan 2007 14:36:06 -0800 Subject: [PATCH 275/326] [PATCH] kprobes: replace magic numbers with enum Replace the magic numbers with an enum, and gets rid of a warning on the specific architectures (ex. powerpc) on which the compiler considers 'char' as 'unsigned char'. Signed-off-by: Masami Hiramatsu Cc: Prasanna S Panchamukhi Cc: Ananth N Mavinakayanahalli Cc: Anil S Keshavamurthy Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/kprobes.c | 20 +++++++++++++------- 1 file changed, 13 insertions(+), 7 deletions(-) diff --git a/kernel/kprobes.c b/kernel/kprobes.c index 17ec4afb0994..6fcf8dd148d0 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -87,6 +87,12 @@ struct kprobe_insn_page { int ngarbage; }; +enum kprobe_slot_state { + SLOT_CLEAN = 0, + SLOT_DIRTY = 1, + SLOT_USED = 2, +}; + static struct hlist_head kprobe_insn_pages; static int kprobe_garbage_slots; static int collect_garbage_slots(void); @@ -130,8 +136,8 @@ kprobe_opcode_t __kprobes *get_insn_slot(void) if (kip->nused < INSNS_PER_PAGE) { int i; for (i = 0; i < INSNS_PER_PAGE; i++) { - if (!kip->slot_used[i]) { - kip->slot_used[i] = 1; + if (kip->slot_used[i] == SLOT_CLEAN) { + kip->slot_used[i] = SLOT_USED; kip->nused++; return kip->insns + (i * MAX_INSN_SIZE); } @@ -163,8 +169,8 @@ kprobe_opcode_t __kprobes *get_insn_slot(void) } INIT_HLIST_NODE(&kip->hlist); hlist_add_head(&kip->hlist, &kprobe_insn_pages); - memset(kip->slot_used, 0, INSNS_PER_PAGE); - kip->slot_used[0] = 1; + memset(kip->slot_used, SLOT_CLEAN, INSNS_PER_PAGE); + kip->slot_used[0] = SLOT_USED; kip->nused = 1; kip->ngarbage = 0; return kip->insns; @@ -173,7 +179,7 @@ kprobe_opcode_t __kprobes *get_insn_slot(void) /* Return 1 if all garbages are collected, otherwise 0. */ static int __kprobes collect_one_slot(struct kprobe_insn_page *kip, int idx) { - kip->slot_used[idx] = 0; + kip->slot_used[idx] = SLOT_CLEAN; kip->nused--; if (kip->nused == 0) { /* @@ -212,7 +218,7 @@ static int __kprobes collect_garbage_slots(void) continue; kip->ngarbage = 0; /* we will collect all garbages */ for (i = 0; i < INSNS_PER_PAGE; i++) { - if (kip->slot_used[i] == -1 && + if (kip->slot_used[i] == SLOT_DIRTY && collect_one_slot(kip, i)) break; } @@ -232,7 +238,7 @@ void __kprobes free_insn_slot(kprobe_opcode_t * slot, int dirty) slot < kip->insns + (INSNS_PER_PAGE * MAX_INSN_SIZE)) { int i = (slot - kip->insns) / MAX_INSN_SIZE; if (dirty) { - kip->slot_used[i] = -1; + kip->slot_used[i] = SLOT_DIRTY; kip->ngarbage++; } else { collect_one_slot(kip, i); From c06bb5d49d8b240876c7c5019197e6a7bd33bcf7 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Tue, 30 Jan 2007 14:36:09 -0800 Subject: [PATCH 276/326] [PATCH] Fix VIA quirks Fix VIA quirks that were recently broken by Alan Cox in the upstream kernel (commit 1597cacbe39802d86656d1f2e6329895bd2ef531). My understanding is that pci_find_present() doesn't work yet at the time the quirks are run. So I used a two-step quirk as is done for some other quirks already. First we detect the VIA south bridges and set the right low and high device limits, then we are ready to actually run the quirks on the affected devices. Signed-off-by: Jean Delvare Acked-by: Alan Cox Acked-by: Nick Piggin Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/pci/quirks.c | 78 ++++++++++++++++++++++++-------------------- 1 file changed, 42 insertions(+), 36 deletions(-) diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index ef882a8a094e..16945c2ba2ca 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -654,19 +654,40 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C686_4, quirk_vi * VIA bridges which have VLink */ -static const struct pci_device_id via_vlink_fixup_tbl[] = { - /* Internal devices need IRQ line routing, pre VLink */ - { PCI_VDEVICE(VIA, PCI_DEVICE_ID_VIA_82C686), 0 }, - { PCI_VDEVICE(VIA, PCI_DEVICE_ID_VIA_8231), 17 }, - /* Devices with VLink */ - { PCI_VDEVICE(VIA, PCI_DEVICE_ID_VIA_8233_0), 17}, - { PCI_VDEVICE(VIA, PCI_DEVICE_ID_VIA_8233A), 17 }, - { PCI_VDEVICE(VIA, PCI_DEVICE_ID_VIA_8233C_0), 17 }, - { PCI_VDEVICE(VIA, PCI_DEVICE_ID_VIA_8235), 16 }, - { PCI_VDEVICE(VIA, PCI_DEVICE_ID_VIA_8237), 15 }, - { PCI_VDEVICE(VIA, PCI_DEVICE_ID_VIA_8237A), 15 }, - { 0, }, -}; +static int via_vlink_dev_lo = -1, via_vlink_dev_hi = 18; + +static void quirk_via_bridge(struct pci_dev *dev) +{ + /* See what bridge we have and find the device ranges */ + switch (dev->device) { + case PCI_DEVICE_ID_VIA_82C686: + /* 82C686 is special */ + via_vlink_dev_lo = 7; + via_vlink_dev_hi = 7; + break; + case PCI_DEVICE_ID_VIA_8237: + case PCI_DEVICE_ID_VIA_8237A: + via_vlink_dev_lo = 15; + break; + case PCI_DEVICE_ID_VIA_8235: + via_vlink_dev_lo = 16; + break; + case PCI_DEVICE_ID_VIA_8231: + case PCI_DEVICE_ID_VIA_8233_0: + case PCI_DEVICE_ID_VIA_8233A: + case PCI_DEVICE_ID_VIA_8233C_0: + via_vlink_dev_lo = 17; + break; + } +} +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_82C686, quirk_via_bridge); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8231, quirk_via_bridge); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8233_0, quirk_via_bridge); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8233A, quirk_via_bridge); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8233C_0, quirk_via_bridge); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8235, quirk_via_bridge); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8237, quirk_via_bridge); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8237A, quirk_via_bridge); /** * quirk_via_vlink - VIA VLink IRQ number update @@ -675,35 +696,20 @@ static const struct pci_device_id via_vlink_fixup_tbl[] = { * If the device we are dealing with is on a PIC IRQ we need to * ensure that the IRQ line register which usually is not relevant * for PCI cards, is actually written so that interrupts get sent - * to the right place + * to the right place. + * We only do this on systems where a VIA south bridge was detected, + * and only for VIA devices on the motherboard (see quirk_via_bridge + * above). */ static void quirk_via_vlink(struct pci_dev *dev) { - const struct pci_device_id *via_vlink_fixup; - static int dev_lo = -1, dev_hi = 18; u8 irq, new_irq; - /* Check if we have VLink and cache the result */ - - /* Checked already - no */ - if (dev_lo == -2) + /* Check if we have VLink at all */ + if (via_vlink_dev_lo == -1) return; - /* Not checked - see what bridge we have and find the device - ranges */ - - if (dev_lo == -1) { - via_vlink_fixup = pci_find_present(via_vlink_fixup_tbl); - if (via_vlink_fixup == NULL) { - dev_lo = -2; - return; - } - dev_lo = via_vlink_fixup->driver_data; - /* 82C686 is special - 0/0 */ - if (dev_lo == 0) - dev_hi = 0; - } new_irq = dev->irq; /* Don't quirk interrupts outside the legacy IRQ range */ @@ -711,8 +717,8 @@ static void quirk_via_vlink(struct pci_dev *dev) return; /* Internal device ? */ - if (dev->bus->number != 0 || PCI_SLOT(dev->devfn) > dev_hi || - PCI_SLOT(dev->devfn) < dev_lo) + if (dev->bus->number != 0 || PCI_SLOT(dev->devfn) > via_vlink_dev_hi || + PCI_SLOT(dev->devfn) < via_vlink_dev_lo) return; /* This is an internal VLink device on a PIC interrupt. The BIOS From 99abaf51e25f7d4ac2081e5cdc1f01baa0543514 Mon Sep 17 00:00:00 2001 From: "ethanhsiao@jmicron.com" Date: Tue, 30 Jan 2007 14:36:13 -0800 Subject: [PATCH 277/326] [PATCH] jmicron: 40/80pin primary detection jmicron module detects all JMB36x as JMB361 and PATA0 has wrong pin status of XICBLID. Cc: Jeff Garzik Cc: Alan Cox Cc: Bartlomiej Zolnierkiewicz Cc: Sergei Shtylyov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/ide/pci/jmicron.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/drivers/ide/pci/jmicron.c b/drivers/ide/pci/jmicron.c index 35dda8fcd0ac..f07bbbed1778 100644 --- a/drivers/ide/pci/jmicron.c +++ b/drivers/ide/pci/jmicron.c @@ -86,8 +86,8 @@ static int __devinit ata66_jmicron(ide_hwif_t *hwif) { case PORT_PATA0: if (control & (1 << 3)) /* 40/80 pin primary */ - return 1; - return 0; + return 0; + return 1; case PORT_PATA1: if (control5 & (1 << 19)) /* 40/80 pin secondary */ return 0; @@ -241,11 +241,11 @@ static int __devinit jmicron_init_one(struct pci_dev *dev, const struct pci_devi } static struct pci_device_id jmicron_pci_tbl[] = { - { PCI_DEVICE(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB361), 0}, - { PCI_DEVICE(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB363), 1}, - { PCI_DEVICE(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB365), 2}, - { PCI_DEVICE(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB366), 3}, - { PCI_DEVICE(PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB368), 4}, + { PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB361, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, + { PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB363, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 1}, + { PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB365, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 2}, + { PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB366, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 3}, + { PCI_VENDOR_ID_JMICRON, PCI_DEVICE_ID_JMICRON_JMB368, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 4}, { 0, }, }; From 3896625d0badd53dbc34d584861a36ba7eb4613f Mon Sep 17 00:00:00 2001 From: Jeff Dike Date: Tue, 30 Jan 2007 14:36:17 -0800 Subject: [PATCH 278/326] [PATCH] uml: fix signal frame alignment Use the same signal frame alignment calculations as the underlying architecture. x86_64 appeared to do this, but the "- 8" was really subtracting 8 * sizeof(struct rt_sigframe) rather than 8 bytes. UML/i386 might have been OK, but I changed the calculation to match i386 just to be sure. Signed-off-by: Jeff Dike Cc: Cc: Adrian Bunk Cc: Paolo 'Blaisorblade' Giarrusso Acked-by: Antoine Martin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/um/sys-i386/signal.c | 3 ++- arch/um/sys-x86_64/signal.c | 5 +++-- 2 files changed, 5 insertions(+), 3 deletions(-) diff --git a/arch/um/sys-i386/signal.c b/arch/um/sys-i386/signal.c index 0709fc6670c2..3f6acd667717 100644 --- a/arch/um/sys-i386/signal.c +++ b/arch/um/sys-i386/signal.c @@ -219,7 +219,8 @@ int setup_signal_stack_sc(unsigned long stack_top, int sig, unsigned long save_sp = PT_REGS_SP(regs); int err = 0; - stack_top &= -8UL; + /* This is the same calculation as i386 - ((sp + 4) & 15) == 0 */ + stack_top = ((stack_top + 4) & -16UL) - 4; frame = (struct sigframe __user *) stack_top - 1; if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame))) return 1; diff --git a/arch/um/sys-x86_64/signal.c b/arch/um/sys-x86_64/signal.c index 9edf114faf79..af2f017617b4 100644 --- a/arch/um/sys-x86_64/signal.c +++ b/arch/um/sys-x86_64/signal.c @@ -191,8 +191,9 @@ int setup_signal_stack_si(unsigned long stack_top, int sig, struct task_struct *me = current; frame = (struct rt_sigframe __user *) - round_down(stack_top - sizeof(struct rt_sigframe), 16) - 8; - frame = (struct rt_sigframe __user *) ((unsigned long) frame - 128); + round_down(stack_top - sizeof(struct rt_sigframe), 16); + /* Subtract 128 for a red zone and 8 for proper alignment */ + frame = (struct rt_sigframe __user *) ((unsigned long) frame - 128 - 8); if (!access_ok(VERIFY_WRITE, fp, sizeof(struct _fpstate))) goto out; From fa8609da993b04dc2dd762173a6d0ab1a192e256 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 30 Jan 2007 14:36:24 -0800 Subject: [PATCH 279/326] [PATCH] ntfs: kmap_atomic() atomicity fix The KM_BIO_SRC_IRQ kmap slot requires local irq protection. Acked-by: Anton Altaparmakov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ntfs/aops.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/fs/ntfs/aops.c b/fs/ntfs/aops.c index 7b2c8f4f6a6f..629e7abdd840 100644 --- a/fs/ntfs/aops.c +++ b/fs/ntfs/aops.c @@ -92,10 +92,12 @@ static void ntfs_end_buffer_async_read(struct buffer_head *bh, int uptodate) ofs = 0; if (file_ofs < init_size) ofs = init_size - file_ofs; + local_irq_save(flags); kaddr = kmap_atomic(page, KM_BIO_SRC_IRQ); memset(kaddr + bh_offset(bh) + ofs, 0, bh->b_size - ofs); kunmap_atomic(kaddr, KM_BIO_SRC_IRQ); + local_irq_restore(flags); flush_dcache_page(page); } } else { @@ -143,11 +145,13 @@ static void ntfs_end_buffer_async_read(struct buffer_head *bh, int uptodate) recs = PAGE_CACHE_SIZE / rec_size; /* Should have been verified before we got here... */ BUG_ON(!recs); + local_irq_save(flags); kaddr = kmap_atomic(page, KM_BIO_SRC_IRQ); for (i = 0; i < recs; i++) post_read_mst_fixup((NTFS_RECORD*)(kaddr + i * rec_size), rec_size); kunmap_atomic(kaddr, KM_BIO_SRC_IRQ); + local_irq_restore(flags); flush_dcache_page(page); if (likely(page_uptodate && !PageError(page))) SetPageUptodate(page); From 41c57a87183a7c458d86f78966d69d4bf18ea0b7 Mon Sep 17 00:00:00 2001 From: David Barksdale Date: Tue, 30 Jan 2007 14:36:25 -0800 Subject: [PATCH 280/326] [PATCH] IPMI: fix timeout list handling Fix a dangling pointer bug in ipmi_timeout_handler. A list of timedout messages is not re-initialized before reuse, causing the head of the list to point to freed memory. Signed-off-by: David Barksdale Signed-off-by: Corey Minyard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/char/ipmi/ipmi_msghandler.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index 4e4691a53890..53582b53da95 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -3649,8 +3649,6 @@ static void ipmi_timeout_handler(long timeout_period) unsigned long flags; int i; - INIT_LIST_HEAD(&timeouts); - rcu_read_lock(); list_for_each_entry_rcu(intf, &ipmi_interfaces, link) { /* See if any waiting messages need to be processed. */ @@ -3671,6 +3669,7 @@ static void ipmi_timeout_handler(long timeout_period) /* Go through the seq table and find any messages that have timed out, putting them in the timeouts list. */ + INIT_LIST_HEAD(&timeouts); spin_lock_irqsave(&intf->seq_lock, flags); for (i = 0; i < IPMI_IPMB_NUM_SEQ; i++) check_msg_timeout(intf, &(intf->seq_table[i]), From f56df2f4db6e4af87fb8e941cff69f4501a111df Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 30 Jan 2007 19:42:57 -0800 Subject: [PATCH 281/326] Linux 2.6.20-rc7 Ok, so I said there wouldn't be another -rc. I lied. Signed-off-by: Linus Torvalds --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 59aa9f56fdbe..75adfb539977 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 2 PATCHLEVEL = 6 SUBLEVEL = 20 -EXTRAVERSION =-rc6 +EXTRAVERSION =-rc7 NAME = Homicidal Dwarf Hamster # *DOCUMENTATION* From 2e5530236645a042e1481aa19879b819c3e8f5c9 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 30 Jan 2007 21:36:09 -0800 Subject: [PATCH 282/326] [NETFILTER]: xt_hashlimit: fix ip6tables dependency IP6_NF_IPTABLES=m, CONFIG_NETFILTER_XT_MATCH_HASHLIMIT=y results in a linker error since ipv6_find_hdr is defined in ip6_tables.c. Fix similar to Adrian Bunk's H.323 conntrack patch: selecting ip6_tables to be build as module requires hashlimit to be built as module as well. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/netfilter/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index cd10e44db015..2a2bcb303bfa 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -628,7 +628,7 @@ config NETFILTER_XT_MATCH_TCPMSS config NETFILTER_XT_MATCH_HASHLIMIT tristate '"hashlimit" match support' - depends on NETFILTER_XTABLES + depends on NETFILTER_XTABLES && (IP6_NF_IPTABLES || IP6_NF_IPTABLES=n) help This option adds a `hashlimit' match. From a608ab9cb6a5050394498b2520c6e7c162f4e2cf Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 2 Jan 2007 10:39:10 +0000 Subject: [PATCH 283/326] netxen patches Have fun. >From 24f4a1a77431575a9cdfaae25adda85842099f70 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 1 Jan 2007 15:22:56 -0500 Subject: [PATCH] netxen trivial annotations Signed-off-by: Al Viro Signed-off-by: Jeff Garzik --- drivers/net/netxen/netxen_nic.h | 136 +++++++++++------------- drivers/net/netxen/netxen_nic_ethtool.c | 22 ++-- drivers/net/netxen/netxen_nic_hw.c | 43 ++++---- drivers/net/netxen/netxen_nic_hw.h | 74 ++++++------- drivers/net/netxen/netxen_nic_init.c | 15 ++- drivers/net/netxen/netxen_nic_isr.c | 4 +- drivers/net/netxen/netxen_nic_main.c | 8 +- drivers/net/netxen/netxen_nic_niu.c | 106 +++++++++--------- 8 files changed, 196 insertions(+), 212 deletions(-) diff --git a/drivers/net/netxen/netxen_nic.h b/drivers/net/netxen/netxen_nic.h index 59324b1693d6..e8598b809228 100644 --- a/drivers/net/netxen/netxen_nic.h +++ b/drivers/net/netxen/netxen_nic.h @@ -239,49 +239,39 @@ extern unsigned long long netxen_dma_mask; typedef u32 netxen_ctx_msg; -#define _netxen_set_bits(config_word, start, bits, val) {\ - unsigned long long mask = (((1ULL << (bits)) - 1) << (start)); \ - unsigned long long value = (val); \ - (config_word) &= ~mask; \ - (config_word) |= (((value) << (start)) & mask); \ -} - #define netxen_set_msg_peg_id(config_word, val) \ - _netxen_set_bits(config_word, 0, 2, val) + ((config_word) &= ~3, (config_word) |= val & 3) #define netxen_set_msg_privid(config_word) \ - set_bit(2, (unsigned long*)&config_word) + ((config_word) |= 1 << 2) #define netxen_set_msg_count(config_word, val) \ - _netxen_set_bits(config_word, 3, 15, val) + ((config_word) &= ~(0x7fff<<3), (config_word) |= (val & 0x7fff) << 3) #define netxen_set_msg_ctxid(config_word, val) \ - _netxen_set_bits(config_word, 18, 10, val) + ((config_word) &= ~(0x3ff<<18), (config_word) |= (val & 0x3ff) << 18) #define netxen_set_msg_opcode(config_word, val) \ - _netxen_set_bits(config_word, 28, 4, val) + ((config_word) &= ~(0xf<<24), (config_word) |= (val & 0xf) << 24) struct netxen_rcv_context { - u32 rcv_ring_addr_lo; - u32 rcv_ring_addr_hi; - u32 rcv_ring_size; - u32 rsrvd; + __le64 rcv_ring_addr; + __le32 rcv_ring_size; + __le32 rsrvd; }; struct netxen_ring_ctx { /* one command ring */ - u64 cmd_consumer_offset; - u32 cmd_ring_addr_lo; - u32 cmd_ring_addr_hi; - u32 cmd_ring_size; - u32 rsrvd; + __le64 cmd_consumer_offset; + __le64 cmd_ring_addr; + __le32 cmd_ring_size; + __le32 rsrvd; /* three receive rings */ struct netxen_rcv_context rcv_ctx[3]; /* one status ring */ - u32 sts_ring_addr_lo; - u32 sts_ring_addr_hi; - u32 sts_ring_size; + __le64 sts_ring_addr; + __le32 sts_ring_size; - u32 ctx_id; + __le32 ctx_id; } __attribute__ ((aligned(64))); /* @@ -305,81 +295,85 @@ struct netxen_ring_ctx { ((cmd_desc)->port_ctxid |= ((var) & 0x0F)) #define netxen_set_cmd_desc_flags(cmd_desc, val) \ - _netxen_set_bits((cmd_desc)->flags_opcode, 0, 7, val) + ((cmd_desc)->flags_opcode &= ~cpu_to_le16(0x7f), \ + (cmd_desc)->flags_opcode |= cpu_to_le16((val) & 0x7f)) #define netxen_set_cmd_desc_opcode(cmd_desc, val) \ - _netxen_set_bits((cmd_desc)->flags_opcode, 7, 6, val) + ((cmd_desc)->flags_opcode &= ~cpu_to_le16(0x3f<<7), \ + (cmd_desc)->flags_opcode |= cpu_to_le16((val) & (0x3f<<7))) #define netxen_set_cmd_desc_num_of_buff(cmd_desc, val) \ - _netxen_set_bits((cmd_desc)->num_of_buffers_total_length, 0, 8, val); + ((cmd_desc)->num_of_buffers_total_length &= ~cpu_to_le32(0xff), \ + (cmd_desc)->num_of_buffers_total_length |= cpu_to_le32((val) & 0xff)) #define netxen_set_cmd_desc_totallength(cmd_desc, val) \ - _netxen_set_bits((cmd_desc)->num_of_buffers_total_length, 8, 24, val); + ((cmd_desc)->num_of_buffers_total_length &= cpu_to_le32(0xff), \ + (cmd_desc)->num_of_buffers_total_length |= cpu_to_le32(val << 24)) #define netxen_get_cmd_desc_opcode(cmd_desc) \ - (((cmd_desc)->flags_opcode >> 7) & 0x003F) + ((le16_to_cpu((cmd_desc)->flags_opcode) >> 7) & 0x003F) #define netxen_get_cmd_desc_totallength(cmd_desc) \ - (((cmd_desc)->num_of_buffers_total_length >> 8) & 0x0FFFFFF) + (le32_to_cpu((cmd_desc)->num_of_buffers_total_length) >> 8) struct cmd_desc_type0 { u8 tcp_hdr_offset; /* For LSO only */ u8 ip_hdr_offset; /* For LSO only */ /* Bit pattern: 0-6 flags, 7-12 opcode, 13-15 unused */ - u16 flags_opcode; + __le16 flags_opcode; /* Bit pattern: 0-7 total number of segments, 8-31 Total size of the packet */ - u32 num_of_buffers_total_length; + __le32 num_of_buffers_total_length; union { struct { - u32 addr_low_part2; - u32 addr_high_part2; + __le32 addr_low_part2; + __le32 addr_high_part2; }; - u64 addr_buffer2; + __le64 addr_buffer2; }; - u16 reference_handle; /* changed to u16 to add mss */ - u16 mss; /* passed by NDIS_PACKET for LSO */ + __le16 reference_handle; /* changed to u16 to add mss */ + __le16 mss; /* passed by NDIS_PACKET for LSO */ /* Bit pattern 0-3 port, 0-3 ctx id */ u8 port_ctxid; u8 total_hdr_length; /* LSO only : MAC+IP+TCP Hdr size */ - u16 conn_id; /* IPSec offoad only */ + __le16 conn_id; /* IPSec offoad only */ union { struct { - u32 addr_low_part3; - u32 addr_high_part3; + __le32 addr_low_part3; + __le32 addr_high_part3; }; - u64 addr_buffer3; + __le64 addr_buffer3; }; union { struct { - u32 addr_low_part1; - u32 addr_high_part1; + __le32 addr_low_part1; + __le32 addr_high_part1; }; - u64 addr_buffer1; + __le64 addr_buffer1; }; - u16 buffer1_length; - u16 buffer2_length; - u16 buffer3_length; - u16 buffer4_length; + __le16 buffer1_length; + __le16 buffer2_length; + __le16 buffer3_length; + __le16 buffer4_length; union { struct { - u32 addr_low_part4; - u32 addr_high_part4; + __le32 addr_low_part4; + __le32 addr_high_part4; }; - u64 addr_buffer4; + __le64 addr_buffer4; }; - u64 unused; + __le64 unused; } __attribute__ ((aligned(64))); /* Note: sizeof(rcv_desc) should always be a mutliple of 2 */ struct rcv_desc { - u16 reference_handle; - u16 reserved; - u32 buffer_length; /* allocated buffer length (usually 2K) */ - u64 addr_buffer; + __le16 reference_handle; + __le16 reserved; + __le32 buffer_length; /* allocated buffer length (usually 2K) */ + __le64 addr_buffer; }; /* opcode field in status_desc */ @@ -405,36 +399,36 @@ struct rcv_desc { (((status_desc)->lro & 0x80) >> 7) #define netxen_get_sts_port(status_desc) \ - ((status_desc)->status_desc_data & 0x0F) + (le64_to_cpu((status_desc)->status_desc_data) & 0x0F) #define netxen_get_sts_status(status_desc) \ - (((status_desc)->status_desc_data >> 4) & 0x0F) + ((le64_to_cpu((status_desc)->status_desc_data) >> 4) & 0x0F) #define netxen_get_sts_type(status_desc) \ - (((status_desc)->status_desc_data >> 8) & 0x0F) + ((le64_to_cpu((status_desc)->status_desc_data) >> 8) & 0x0F) #define netxen_get_sts_totallength(status_desc) \ - (((status_desc)->status_desc_data >> 12) & 0xFFFF) + ((le64_to_cpu((status_desc)->status_desc_data) >> 12) & 0xFFFF) #define netxen_get_sts_refhandle(status_desc) \ - (((status_desc)->status_desc_data >> 28) & 0xFFFF) + ((le64_to_cpu((status_desc)->status_desc_data) >> 28) & 0xFFFF) #define netxen_get_sts_prot(status_desc) \ - (((status_desc)->status_desc_data >> 44) & 0x0F) + ((le64_to_cpu((status_desc)->status_desc_data) >> 44) & 0x0F) #define netxen_get_sts_owner(status_desc) \ - (((status_desc)->status_desc_data >> 56) & 0x03) + ((le64_to_cpu((status_desc)->status_desc_data) >> 56) & 0x03) #define netxen_get_sts_opcode(status_desc) \ - (((status_desc)->status_desc_data >> 58) & 0x03F) + ((le64_to_cpu((status_desc)->status_desc_data) >> 58) & 0x03F) #define netxen_clear_sts_owner(status_desc) \ ((status_desc)->status_desc_data &= \ - ~(((unsigned long long)3) << 56 )) + ~cpu_to_le64(((unsigned long long)3) << 56 )) #define netxen_set_sts_owner(status_desc, val) \ ((status_desc)->status_desc_data |= \ - (((unsigned long long)((val) & 0x3)) << 56 )) + cpu_to_le64(((unsigned long long)((val) & 0x3)) << 56 )) struct status_desc { /* Bit pattern: 0-3 port, 4-7 status, 8-11 type, 12-27 total_length 28-43 reference_handle, 44-47 protocol, 48-52 unused 53-55 desc_cnt, 56-57 owner, 58-63 opcode */ - u64 status_desc_data; - u32 hash_value; + __le64 status_desc_data; + __le32 hash_value; u8 hash_type; u8 msg_type; u8 unused; @@ -1005,9 +999,9 @@ void netxen_niu_gbe_set_mii_mode(struct netxen_adapter *adapter, int port, void netxen_niu_gbe_set_gmii_mode(struct netxen_adapter *adapter, int port, long enable); int netxen_niu_gbe_phy_read(struct netxen_adapter *adapter, long phy, long reg, - __le32 * readval); + __u32 * readval); int netxen_niu_gbe_phy_write(struct netxen_adapter *adapter, long phy, - long reg, __le32 val); + long reg, __u32 val); /* Functions available from netxen_nic_hw.c */ int netxen_nic_set_mtu_xgb(struct netxen_port *port, int new_mtu); diff --git a/drivers/net/netxen/netxen_nic_ethtool.c b/drivers/net/netxen/netxen_nic_ethtool.c index 34044616b3c8..c381d77a7336 100644 --- a/drivers/net/netxen/netxen_nic_ethtool.c +++ b/drivers/net/netxen/netxen_nic_ethtool.c @@ -218,7 +218,7 @@ netxen_nic_set_settings(struct net_device *dev, struct ethtool_cmd *ecmd) { struct netxen_port *port = netdev_priv(dev); struct netxen_adapter *adapter = port->adapter; - __le32 status; + __u32 status; /* read which mode */ if (adapter->ahw.board_type == NETXEN_NIC_GBE) { @@ -226,7 +226,7 @@ netxen_nic_set_settings(struct net_device *dev, struct ethtool_cmd *ecmd) if (adapter->phy_write && adapter->phy_write(adapter, port->portnum, NETXEN_NIU_GB_MII_MGMT_ADDR_AUTONEG, - (__le32) ecmd->autoneg) != 0) + ecmd->autoneg) != 0) return -EIO; else port->link_autoneg = ecmd->autoneg; @@ -279,7 +279,7 @@ static int netxen_nic_get_regs_len(struct net_device *dev) } struct netxen_niu_regs { - __le32 reg[NETXEN_NIC_REGS_COUNT]; + __u32 reg[NETXEN_NIC_REGS_COUNT]; }; static struct netxen_niu_regs niu_registers[] = { @@ -372,7 +372,7 @@ netxen_nic_get_regs(struct net_device *dev, struct ethtool_regs *regs, void *p) { struct netxen_port *port = netdev_priv(dev); struct netxen_adapter *adapter = port->adapter; - __le32 mode, *regs_buff = p; + __u32 mode, *regs_buff = p; void __iomem *addr; int i, window; @@ -415,7 +415,7 @@ static u32 netxen_nic_get_link(struct net_device *dev) { struct netxen_port *port = netdev_priv(dev); struct netxen_adapter *adapter = port->adapter; - __le32 status; + __u32 status; /* read which mode */ if (adapter->ahw.board_type == NETXEN_NIC_GBE) { @@ -482,13 +482,13 @@ netxen_nic_get_pauseparam(struct net_device *dev, { struct netxen_port *port = netdev_priv(dev); struct netxen_adapter *adapter = port->adapter; - __le32 val; + __u32 val; if (adapter->ahw.board_type == NETXEN_NIC_GBE) { /* get flow control settings */ netxen_nic_read_w0(adapter, NETXEN_NIU_GB_MAC_CONFIG_0(port->portnum), - (u32 *) & val); + &val); pause->rx_pause = netxen_gb_get_rx_flowctl(val); pause->tx_pause = netxen_gb_get_tx_flowctl(val); /* get autoneg settings */ @@ -502,7 +502,7 @@ netxen_nic_set_pauseparam(struct net_device *dev, { struct netxen_port *port = netdev_priv(dev); struct netxen_adapter *adapter = port->adapter; - __le32 val; + __u32 val; unsigned int autoneg; /* read mode */ @@ -522,13 +522,13 @@ netxen_nic_set_pauseparam(struct net_device *dev, netxen_nic_write_w0(adapter, NETXEN_NIU_GB_MAC_CONFIG_0(port->portnum), - *(u32 *) (&val)); + *&val); /* set autoneg */ autoneg = pause->autoneg; if (adapter->phy_write && adapter->phy_write(adapter, port->portnum, NETXEN_NIU_GB_MII_MGMT_ADDR_AUTONEG, - (__le32) autoneg) != 0) + autoneg) != 0) return -EIO; else { port->link_autoneg = pause->autoneg; @@ -543,7 +543,7 @@ static int netxen_nic_reg_test(struct net_device *dev) struct netxen_port *port = netdev_priv(dev); struct netxen_adapter *adapter = port->adapter; u32 data_read, data_written, save; - __le32 mode; + __u32 mode; /* * first test the "Read Only" registers by writing which mode diff --git a/drivers/net/netxen/netxen_nic_hw.c b/drivers/net/netxen/netxen_nic_hw.c index 191e2336e323..f263232f499f 100644 --- a/drivers/net/netxen/netxen_nic_hw.c +++ b/drivers/net/netxen/netxen_nic_hw.c @@ -95,7 +95,7 @@ void netxen_nic_set_multi(struct net_device *netdev) struct netxen_port *port = netdev_priv(netdev); struct netxen_adapter *adapter = port->adapter; struct dev_mc_list *mc_ptr; - __le32 netxen_mac_addr_cntl_data = 0; + __u32 netxen_mac_addr_cntl_data = 0; mc_ptr = netdev->mc_list; if (netdev->flags & IFF_PROMISC) { @@ -236,8 +236,9 @@ int netxen_nic_hw_resources(struct netxen_adapter *adapter) } memset(addr, 0, sizeof(struct netxen_ring_ctx)); adapter->ctx_desc = (struct netxen_ring_ctx *)addr; - adapter->ctx_desc->cmd_consumer_offset = adapter->ctx_desc_phys_addr - + sizeof(struct netxen_ring_ctx); + adapter->ctx_desc->cmd_consumer_offset = + cpu_to_le64(adapter->ctx_desc_phys_addr + + sizeof(struct netxen_ring_ctx)); adapter->cmd_consumer = (uint32_t *) (((char *)addr) + sizeof(struct netxen_ring_ctx)); @@ -253,11 +254,10 @@ int netxen_nic_hw_resources(struct netxen_adapter *adapter) return -ENOMEM; } - adapter->ctx_desc->cmd_ring_addr_lo = - hw->cmd_desc_phys_addr & 0xffffffffUL; - adapter->ctx_desc->cmd_ring_addr_hi = - ((u64) hw->cmd_desc_phys_addr >> 32); - adapter->ctx_desc->cmd_ring_size = adapter->max_tx_desc_count; + adapter->ctx_desc->cmd_ring_addr = + cpu_to_le64(hw->cmd_desc_phys_addr); + adapter->ctx_desc->cmd_ring_size = + cpu_to_le32(adapter->max_tx_desc_count); hw->cmd_desc_head = (struct cmd_desc_type0 *)addr; @@ -278,12 +278,10 @@ int netxen_nic_hw_resources(struct netxen_adapter *adapter) return err; } rcv_desc->desc_head = (struct rcv_desc *)addr; - adapter->ctx_desc->rcv_ctx[ring].rcv_ring_addr_lo = - rcv_desc->phys_addr & 0xffffffffUL; - adapter->ctx_desc->rcv_ctx[ring].rcv_ring_addr_hi = - ((u64) rcv_desc->phys_addr >> 32); + adapter->ctx_desc->rcv_ctx[ring].rcv_ring_addr = + cpu_to_le64(rcv_desc->phys_addr); adapter->ctx_desc->rcv_ctx[ring].rcv_ring_size = - rcv_desc->max_rx_desc_count; + cpu_to_le32(rcv_desc->max_rx_desc_count); } addr = netxen_alloc(adapter->ahw.pdev, STATUS_DESC_RINGSIZE, @@ -297,11 +295,10 @@ int netxen_nic_hw_resources(struct netxen_adapter *adapter) return err; } recv_ctx->rcv_status_desc_head = (struct status_desc *)addr; - adapter->ctx_desc->sts_ring_addr_lo = - recv_ctx->rcv_status_desc_phys_addr & 0xffffffffUL; - adapter->ctx_desc->sts_ring_addr_hi = - ((u64) recv_ctx->rcv_status_desc_phys_addr >> 32); - adapter->ctx_desc->sts_ring_size = adapter->max_rx_desc_count; + adapter->ctx_desc->sts_ring_addr = + cpu_to_le64(recv_ctx->rcv_status_desc_phys_addr); + adapter->ctx_desc->sts_ring_size = + cpu_to_le32(adapter->max_rx_desc_count); } /* Window = 1 */ @@ -387,10 +384,6 @@ void netxen_tso_check(struct netxen_adapter *adapter, } adapter->stats.xmitcsummed++; desc->tcp_hdr_offset = skb->h.raw - skb->data; - netxen_set_cmd_desc_totallength(desc, - cpu_to_le32 - (netxen_get_cmd_desc_totallength - (desc))); desc->ip_hdr_offset = skb->nh.raw - skb->data; } @@ -867,9 +860,9 @@ netxen_crb_writelit_adapter(struct netxen_adapter *adapter, unsigned long off, void netxen_nic_set_link_parameters(struct netxen_port *port) { struct netxen_adapter *adapter = port->adapter; - __le32 status; - __le32 autoneg; - __le32 mode; + __u32 status; + __u32 autoneg; + __u32 mode; netxen_nic_read_w0(adapter, NETXEN_NIU_MODE, &mode); if (netxen_get_niu_enable_ge(mode)) { /* Gb 10/100/1000 Mbps mode */ diff --git a/drivers/net/netxen/netxen_nic_hw.h b/drivers/net/netxen/netxen_nic_hw.h index 0685633a9c1e..ab1112eb1b0d 100644 --- a/drivers/net/netxen/netxen_nic_hw.h +++ b/drivers/net/netxen/netxen_nic_hw.h @@ -124,28 +124,28 @@ typedef enum { */ #define netxen_gb_enable_tx(config_word) \ - set_bit(0, (unsigned long*)(&config_word)) + ((config_word) |= 1 << 0) #define netxen_gb_enable_rx(config_word) \ - set_bit(2, (unsigned long*)(&config_word)) + ((config_word) |= 1 << 2) #define netxen_gb_tx_flowctl(config_word) \ - set_bit(4, (unsigned long*)(&config_word)) + ((config_word) |= 1 << 4) #define netxen_gb_rx_flowctl(config_word) \ - set_bit(5, (unsigned long*)(&config_word)) + ((config_word) |= 1 << 5) #define netxen_gb_tx_reset_pb(config_word) \ - set_bit(16, (unsigned long*)(&config_word)) + ((config_word) |= 1 << 16) #define netxen_gb_rx_reset_pb(config_word) \ - set_bit(17, (unsigned long*)(&config_word)) + ((config_word) |= 1 << 17) #define netxen_gb_tx_reset_mac(config_word) \ - set_bit(18, (unsigned long*)(&config_word)) + ((config_word) |= 1 << 18) #define netxen_gb_rx_reset_mac(config_word) \ - set_bit(19, (unsigned long*)(&config_word)) + ((config_word) |= 1 << 19) #define netxen_gb_soft_reset(config_word) \ - set_bit(31, (unsigned long*)(&config_word)) + ((config_word) |= 1 << 31) #define netxen_gb_unset_tx_flowctl(config_word) \ - clear_bit(4, (unsigned long *)(&config_word)) + ((config_word) &= ~(1 << 4)) #define netxen_gb_unset_rx_flowctl(config_word) \ - clear_bit(5, (unsigned long*)(&config_word)) + ((config_word) &= ~(1 << 5)) #define netxen_gb_get_tx_synced(config_word) \ _netxen_crb_get_bit((config_word), 1) @@ -171,15 +171,15 @@ typedef enum { */ #define netxen_gb_set_duplex(config_word) \ - set_bit(0, (unsigned long*)&config_word) + ((config_word) |= 1 << 0) #define netxen_gb_set_crc_enable(config_word) \ - set_bit(1, (unsigned long*)&config_word) + ((config_word) |= 1 << 1) #define netxen_gb_set_padshort(config_word) \ - set_bit(2, (unsigned long*)&config_word) + ((config_word) |= 1 << 2) #define netxen_gb_set_checklength(config_word) \ - set_bit(4, (unsigned long*)&config_word) + ((config_word) |= 1 << 4) #define netxen_gb_set_hugeframes(config_word) \ - set_bit(5, (unsigned long*)&config_word) + ((config_word) |= 1 << 5) #define netxen_gb_set_preamblelen(config_word, val) \ ((config_word) |= ((val) << 12) & 0xF000) #define netxen_gb_set_intfmode(config_word, val) \ @@ -190,9 +190,9 @@ typedef enum { #define netxen_gb_set_mii_mgmt_clockselect(config_word, val) \ ((config_word) |= ((val) & 0x07)) #define netxen_gb_mii_mgmt_reset(config_word) \ - set_bit(31, (unsigned long*)&config_word) + ((config_word) |= 1 << 31) #define netxen_gb_mii_mgmt_unset(config_word) \ - clear_bit(31, (unsigned long*)&config_word) + ((config_word) &= ~(1 << 31)) /* * NIU GB MII Mgmt Command Register (applies to GB0, GB1, GB2, GB3) @@ -201,7 +201,7 @@ typedef enum { */ #define netxen_gb_mii_mgmt_set_read_cycle(config_word) \ - set_bit(0, (unsigned long*)&config_word) + ((config_word) |= 1 << 0) #define netxen_gb_mii_mgmt_reg_addr(config_word, val) \ ((config_word) |= ((val) & 0x1F)) #define netxen_gb_mii_mgmt_phy_addr(config_word, val) \ @@ -274,9 +274,9 @@ typedef enum { #define netxen_set_phy_speed(config_word, val) \ ((config_word) |= ((val & 0x03) << 14)) #define netxen_set_phy_duplex(config_word) \ - set_bit(13, (unsigned long*)&config_word) + ((config_word) |= 1 << 13) #define netxen_clear_phy_duplex(config_word) \ - clear_bit(13, (unsigned long*)&config_word) + ((config_word) &= ~(1 << 13)) #define netxen_get_phy_jabber(config_word) \ _netxen_crb_get_bit(config_word, 0) @@ -350,11 +350,11 @@ typedef enum { _netxen_crb_get_bit(config_word, 15) #define netxen_set_phy_int_link_status_changed(config_word) \ - set_bit(10, (unsigned long*)&config_word) + ((config_word) |= 1 << 10) #define netxen_set_phy_int_autoneg_completed(config_word) \ - set_bit(11, (unsigned long*)&config_word) + ((config_word) |= 1 << 11) #define netxen_set_phy_int_speed_changed(config_word) \ - set_bit(14, (unsigned long*)&config_word) + ((config_word) |= 1 << 14) /* * NIU Mode Register. @@ -382,22 +382,22 @@ typedef enum { */ #define netxen_set_gb_drop_gb0(config_word) \ - set_bit(0, (unsigned long*)&config_word) + ((config_word) |= 1 << 0) #define netxen_set_gb_drop_gb1(config_word) \ - set_bit(1, (unsigned long*)&config_word) + ((config_word) |= 1 << 1) #define netxen_set_gb_drop_gb2(config_word) \ - set_bit(2, (unsigned long*)&config_word) + ((config_word) |= 1 << 2) #define netxen_set_gb_drop_gb3(config_word) \ - set_bit(3, (unsigned long*)&config_word) + ((config_word) |= 1 << 3) #define netxen_clear_gb_drop_gb0(config_word) \ - clear_bit(0, (unsigned long*)&config_word) + ((config_word) &= ~(1 << 0)) #define netxen_clear_gb_drop_gb1(config_word) \ - clear_bit(1, (unsigned long*)&config_word) + ((config_word) &= ~(1 << 1)) #define netxen_clear_gb_drop_gb2(config_word) \ - clear_bit(2, (unsigned long*)&config_word) + ((config_word) &= ~(1 << 2)) #define netxen_clear_gb_drop_gb3(config_word) \ - clear_bit(3, (unsigned long*)&config_word) + ((config_word) &= ~(1 << 3)) /* * NIU XG MAC Config Register @@ -413,7 +413,7 @@ typedef enum { */ #define netxen_xg_soft_reset(config_word) \ - set_bit(4, (unsigned long*)&config_word) + ((config_word) |= 1 << 4) /* * MAC Control Register @@ -433,19 +433,19 @@ typedef enum { #define netxen_nic_mcr_set_id_pool0(config, val) \ ((config) |= ((val) &0x03)) #define netxen_nic_mcr_set_enable_xtnd0(config) \ - (set_bit(3, (unsigned long *)&(config))) + ((config) |= 1 << 3) #define netxen_nic_mcr_set_id_pool1(config, val) \ ((config) |= (((val) & 0x03) << 4)) #define netxen_nic_mcr_set_enable_xtnd1(config) \ - (set_bit(6, (unsigned long *)&(config))) + ((config) |= 1 << 6) #define netxen_nic_mcr_set_id_pool2(config, val) \ ((config) |= (((val) & 0x03) << 8)) #define netxen_nic_mcr_set_enable_xtnd2(config) \ - (set_bit(10, (unsigned long *)&(config))) + ((config) |= 1 << 10) #define netxen_nic_mcr_set_id_pool3(config, val) \ ((config) |= (((val) & 0x03) << 12)) #define netxen_nic_mcr_set_enable_xtnd3(config) \ - (set_bit(14, (unsigned long *)&(config))) + ((config) |= 1 << 14) #define netxen_nic_mcr_set_mode_select(config, val) \ ((config) |= (((val) & 0x03) << 24)) #define netxen_nic_mcr_set_enable_pool(config, val) \ diff --git a/drivers/net/netxen/netxen_nic_init.c b/drivers/net/netxen/netxen_nic_init.c index c3e41f368554..973af96337a9 100644 --- a/drivers/net/netxen/netxen_nic_init.c +++ b/drivers/net/netxen/netxen_nic_init.c @@ -690,8 +690,7 @@ int netxen_nic_rx_has_work(struct netxen_adapter *adapter) desc_head = recv_ctx->rcv_status_desc_head; desc = &desc_head[consumer]; - if (((le16_to_cpu(netxen_get_sts_owner(desc))) - & STATUS_OWNER_HOST)) + if (netxen_get_sts_owner(desc) & STATUS_OWNER_HOST) return 1; } @@ -787,11 +786,11 @@ netxen_process_rcv(struct netxen_adapter *adapter, int ctxid, struct netxen_port *port = adapter->port[netxen_get_sts_port(desc)]; struct pci_dev *pdev = port->pdev; struct net_device *netdev = port->netdev; - int index = le16_to_cpu(netxen_get_sts_refhandle(desc)); + int index = netxen_get_sts_refhandle(desc); struct netxen_recv_context *recv_ctx = &(adapter->recv_ctx[ctxid]); struct netxen_rx_buffer *buffer; struct sk_buff *skb; - u32 length = le16_to_cpu(netxen_get_sts_totallength(desc)); + u32 length = netxen_get_sts_totallength(desc); u32 desc_ctx; struct netxen_rcv_desc_ctx *rcv_desc; int ret; @@ -918,16 +917,14 @@ u32 netxen_process_rcv_ring(struct netxen_adapter *adapter, int ctxid, int max) */ while (count < max) { desc = &desc_head[consumer]; - if (! - (le16_to_cpu(netxen_get_sts_owner(desc)) & - STATUS_OWNER_HOST)) { + if (!(netxen_get_sts_owner(desc) & STATUS_OWNER_HOST)) { DPRINTK(ERR, "desc %p ownedby %x\n", desc, netxen_get_sts_owner(desc)); break; } netxen_process_rcv(adapter, ctxid, desc); netxen_clear_sts_owner(desc); - netxen_set_sts_owner(desc, cpu_to_le16(STATUS_OWNER_PHANTOM)); + netxen_set_sts_owner(desc, STATUS_OWNER_PHANTOM); consumer = (consumer + 1) & (adapter->max_rx_desc_count - 1); count++; } @@ -1232,7 +1229,7 @@ void netxen_post_rx_buffers_nodb(struct netxen_adapter *adapter, uint32_t ctx, /* make a rcv descriptor */ pdesc->reference_handle = cpu_to_le16(buffer->ref_handle); - pdesc->buffer_length = cpu_to_le16(rcv_desc->dma_size); + pdesc->buffer_length = cpu_to_le32(rcv_desc->dma_size); pdesc->addr_buffer = cpu_to_le64(buffer->dma); DPRINTK(INFO, "done writing descripter\n"); producer = diff --git a/drivers/net/netxen/netxen_nic_isr.c b/drivers/net/netxen/netxen_nic_isr.c index 06847d4252c3..be366e48007c 100644 --- a/drivers/net/netxen/netxen_nic_isr.c +++ b/drivers/net/netxen/netxen_nic_isr.c @@ -79,7 +79,7 @@ void netxen_indicate_link_status(struct netxen_adapter *adapter, u32 portno, void netxen_handle_port_int(struct netxen_adapter *adapter, u32 portno, u32 enable) { - __le32 int_src; + __u32 int_src; struct netxen_port *port; /* This should clear the interrupt source */ @@ -110,7 +110,7 @@ void netxen_handle_port_int(struct netxen_adapter *adapter, u32 portno, /* write it down later.. */ if ((netxen_get_phy_int_speed_changed(int_src)) || (netxen_get_phy_int_link_status_changed(int_src))) { - __le32 status; + __u32 status; DPRINTK(INFO, "SPEED CHANGED OR LINK STATUS CHANGED \n"); diff --git a/drivers/net/netxen/netxen_nic_main.c b/drivers/net/netxen/netxen_nic_main.c index 96e1bee19ba0..69c1b9d23a1a 100644 --- a/drivers/net/netxen/netxen_nic_main.c +++ b/drivers/net/netxen/netxen_nic_main.c @@ -117,7 +117,7 @@ netxen_nic_probe(struct pci_dev *pdev, const struct pci_device_id *ent) void __iomem *mem_ptr1 = NULL; void __iomem *mem_ptr2 = NULL; - u8 *db_ptr = NULL; + u8 __iomem *db_ptr = NULL; unsigned long mem_base, mem_len, db_base, db_len; int pci_using_dac, i, err; int ring; @@ -191,7 +191,7 @@ netxen_nic_probe(struct pci_dev *pdev, const struct pci_device_id *ent) db_len); db_ptr = ioremap(db_base, NETXEN_DB_MAPSIZE_BYTES); - if (db_ptr == 0UL) { + if (!db_ptr) { printk(KERN_ERR "%s: Failed to allocate doorbell map.", netxen_nic_driver_name); err = -EIO; @@ -818,7 +818,7 @@ static int netxen_nic_xmit_frame(struct sk_buff *skb, struct net_device *netdev) /* Take skb->data itself */ pbuf = &adapter->cmd_buf_arr[producer]; if ((netdev->features & NETIF_F_TSO) && skb_shinfo(skb)->gso_size > 0) { - pbuf->mss = cpu_to_le16(skb_shinfo(skb)->gso_size); + pbuf->mss = skb_shinfo(skb)->gso_size; hwdesc->mss = cpu_to_le16(skb_shinfo(skb)->gso_size); } else { pbuf->mss = 0; @@ -882,7 +882,7 @@ static int netxen_nic_xmit_frame(struct sk_buff *skb, struct net_device *netdev) hwdesc->addr_buffer3 = cpu_to_le64(temp_dma); break; case 3: - hwdesc->buffer4_length = temp_len; + hwdesc->buffer4_length = cpu_to_le16(temp_len); hwdesc->addr_buffer4 = cpu_to_le64(temp_dma); break; } diff --git a/drivers/net/netxen/netxen_nic_niu.c b/drivers/net/netxen/netxen_nic_niu.c index 4987dc765d99..40d7003a371c 100644 --- a/drivers/net/netxen/netxen_nic_niu.c +++ b/drivers/net/netxen/netxen_nic_niu.c @@ -89,15 +89,15 @@ static inline int phy_unlock(struct netxen_adapter *adapter) * */ int netxen_niu_gbe_phy_read(struct netxen_adapter *adapter, long phy, - long reg, __le32 * readval) + long reg, __u32 * readval) { long timeout = 0; long result = 0; long restore = 0; - __le32 address; - __le32 command; - __le32 status; - __le32 mac_cfg0; + __u32 address; + __u32 command; + __u32 status; + __u32 mac_cfg0; if (phy_lock(adapter) != 0) { return -1; @@ -112,7 +112,7 @@ int netxen_niu_gbe_phy_read(struct netxen_adapter *adapter, long phy, &mac_cfg0, 4)) return -EIO; if (netxen_gb_get_soft_reset(mac_cfg0)) { - __le32 temp; + __u32 temp; temp = 0; netxen_gb_tx_reset_pb(temp); netxen_gb_rx_reset_pb(temp); @@ -184,15 +184,15 @@ int netxen_niu_gbe_phy_read(struct netxen_adapter *adapter, long phy, * */ int netxen_niu_gbe_phy_write(struct netxen_adapter *adapter, - long phy, long reg, __le32 val) + long phy, long reg, __u32 val) { long timeout = 0; long result = 0; long restore = 0; - __le32 address; - __le32 command; - __le32 status; - __le32 mac_cfg0; + __u32 address; + __u32 command; + __u32 status; + __u32 mac_cfg0; /* * MII mgmt all goes through port 0 MAC interface, so it @@ -203,7 +203,7 @@ int netxen_niu_gbe_phy_write(struct netxen_adapter *adapter, &mac_cfg0, 4)) return -EIO; if (netxen_gb_get_soft_reset(mac_cfg0)) { - __le32 temp; + __u32 temp; temp = 0; netxen_gb_tx_reset_pb(temp); netxen_gb_rx_reset_pb(temp); @@ -269,7 +269,7 @@ int netxen_niu_gbe_enable_phy_interrupts(struct netxen_adapter *adapter, int port) { int result = 0; - __le32 enable = 0; + __u32 enable = 0; netxen_set_phy_int_link_status_changed(enable); netxen_set_phy_int_autoneg_completed(enable); netxen_set_phy_int_speed_changed(enable); @@ -402,7 +402,7 @@ void netxen_niu_gbe_set_gmii_mode(struct netxen_adapter *adapter, int netxen_niu_gbe_init_port(struct netxen_adapter *adapter, int port) { int result = 0; - __le32 status; + __u32 status; if (adapter->disable_phy_interrupts) adapter->disable_phy_interrupts(adapter, port); mdelay(2); @@ -410,7 +410,7 @@ int netxen_niu_gbe_init_port(struct netxen_adapter *adapter, int port) if (0 == netxen_niu_gbe_phy_read(adapter, port, NETXEN_NIU_GB_MII_MGMT_ADDR_PHY_STATUS, - (__le32 *) & status)) { + &status)) { if (netxen_get_phy_link(status)) { if (netxen_get_phy_speed(status) == 2) { netxen_niu_gbe_set_gmii_mode(adapter, port, 1); @@ -489,7 +489,7 @@ int netxen_niu_gbe_handle_phy_interrupt(struct netxen_adapter *adapter, int port, long enable) { int result = 0; - __le32 int_src; + __u32 int_src; printk(KERN_INFO PFX "NETXEN: Handling PHY interrupt on port %d" " (device enable = %d)\n", (int)port, (int)enable); @@ -530,7 +530,7 @@ int netxen_niu_gbe_handle_phy_interrupt(struct netxen_adapter *adapter, printk(KERN_INFO PFX "autoneg_error "); if ((netxen_get_phy_int_speed_changed(int_src)) || (netxen_get_phy_int_link_status_changed(int_src))) { - __le32 status; + __u32 status; printk(KERN_INFO PFX "speed_changed or link status changed"); @@ -583,9 +583,9 @@ int netxen_niu_gbe_handle_phy_interrupt(struct netxen_adapter *adapter, int netxen_niu_macaddr_get(struct netxen_adapter *adapter, int phy, netxen_ethernet_macaddr_t * addr) { - u64 result = 0; - __le32 stationhigh; - __le32 stationlow; + u32 stationhigh; + u32 stationlow; + u8 val[8]; if (addr == NULL) return -EINVAL; @@ -598,10 +598,10 @@ int netxen_niu_macaddr_get(struct netxen_adapter *adapter, if (netxen_nic_hw_read_wx(adapter, NETXEN_NIU_GB_STATION_ADDR_1(phy), &stationlow, 4)) return -EIO; + ((__le32 *)val)[1] = cpu_to_le32(stationhigh); + ((__le32 *)val)[0] = cpu_to_le32(stationlow); - result = (u64) netxen_gb_get_stationaddress_low(stationlow); - result |= (u64) stationhigh << 16; - memcpy(*addr, &result, sizeof(netxen_ethernet_macaddr_t)); + memcpy(addr, val + 2, 6); return 0; } @@ -613,24 +613,25 @@ int netxen_niu_macaddr_get(struct netxen_adapter *adapter, int netxen_niu_macaddr_set(struct netxen_port *port, netxen_ethernet_macaddr_t addr) { - __le32 temp = 0; + u8 temp[4]; + u32 val; struct netxen_adapter *adapter = port->adapter; int phy = port->portnum; unsigned char mac_addr[6]; int i; for (i = 0; i < 10; i++) { - memcpy(&temp, addr, 2); - temp <<= 16; + temp[0] = temp[1] = 0; + memcpy(temp + 2, addr, 2); + val = le32_to_cpu(*(__le32 *)temp); if (netxen_nic_hw_write_wx - (adapter, NETXEN_NIU_GB_STATION_ADDR_1(phy), &temp, 4)) + (adapter, NETXEN_NIU_GB_STATION_ADDR_1(phy), &val, 4)) return -EIO; - temp = 0; - - memcpy(&temp, ((u8 *) addr) + 2, sizeof(__le32)); + memcpy(temp, ((u8 *) addr) + 2, sizeof(__le32)); + val = le32_to_cpu(*(__le32 *)temp); if (netxen_nic_hw_write_wx - (adapter, NETXEN_NIU_GB_STATION_ADDR_0(phy), &temp, 4)) + (adapter, NETXEN_NIU_GB_STATION_ADDR_0(phy), &val, 4)) return -2; netxen_niu_macaddr_get(adapter, phy, @@ -659,9 +660,9 @@ int netxen_niu_macaddr_set(struct netxen_port *port, int netxen_niu_enable_gbe_port(struct netxen_adapter *adapter, int port, netxen_niu_gbe_ifmode_t mode) { - __le32 mac_cfg0; - __le32 mac_cfg1; - __le32 mii_cfg; + __u32 mac_cfg0; + __u32 mac_cfg1; + __u32 mii_cfg; if ((port < 0) || (port > NETXEN_NIU_MAX_GBE_PORTS)) return -EINVAL; @@ -736,7 +737,7 @@ int netxen_niu_enable_gbe_port(struct netxen_adapter *adapter, /* Disable a GbE interface */ int netxen_niu_disable_gbe_port(struct netxen_adapter *adapter, int port) { - __le32 mac_cfg0; + __u32 mac_cfg0; if ((port < 0) || (port > NETXEN_NIU_MAX_GBE_PORTS)) return -EINVAL; @@ -752,7 +753,7 @@ int netxen_niu_disable_gbe_port(struct netxen_adapter *adapter, int port) /* Disable an XG interface */ int netxen_niu_disable_xg_port(struct netxen_adapter *adapter, int port) { - __le32 mac_cfg; + __u32 mac_cfg; if (port != 0) return -EINVAL; @@ -769,7 +770,7 @@ int netxen_niu_disable_xg_port(struct netxen_adapter *adapter, int port) int netxen_niu_set_promiscuous_mode(struct netxen_adapter *adapter, int port, netxen_niu_prom_mode_t mode) { - __le32 reg; + __u32 reg; if ((port < 0) || (port > NETXEN_NIU_MAX_GBE_PORTS)) return -EINVAL; @@ -826,22 +827,21 @@ int netxen_niu_set_promiscuous_mode(struct netxen_adapter *adapter, int port, int netxen_niu_xg_macaddr_set(struct netxen_port *port, netxen_ethernet_macaddr_t addr) { - __le32 temp = 0; + u8 temp[4]; + u32 val; struct netxen_adapter *adapter = port->adapter; - memcpy(&temp, addr, 2); - temp = cpu_to_le32(temp); - temp <<= 16; + temp[0] = temp[1] = 0; + memcpy(temp + 2, addr, 2); + val = le32_to_cpu(*(__le32 *)temp); if (netxen_nic_hw_write_wx(adapter, NETXEN_NIU_XGE_STATION_ADDR_0_1, - &temp, 4)) + &val, 4)) return -EIO; - temp = 0; - memcpy(&temp, ((u8 *) addr) + 2, sizeof(__le32)); - temp = cpu_to_le32(temp); + val = le32_to_cpu(*(__le32 *)temp); if (netxen_nic_hw_write_wx(adapter, NETXEN_NIU_XGE_STATION_ADDR_0_HI, - &temp, 4)) + &val, 4)) return -EIO; return 0; @@ -854,9 +854,9 @@ int netxen_niu_xg_macaddr_set(struct netxen_port *port, int netxen_niu_xg_macaddr_get(struct netxen_adapter *adapter, int phy, netxen_ethernet_macaddr_t * addr) { - __le32 stationhigh; - __le32 stationlow; - u64 result; + u32 stationhigh; + u32 stationlow; + u8 val[8]; if (addr == NULL) return -EINVAL; @@ -869,10 +869,10 @@ int netxen_niu_xg_macaddr_get(struct netxen_adapter *adapter, int phy, if (netxen_nic_hw_read_wx(adapter, NETXEN_NIU_XGE_STATION_ADDR_0_1, &stationlow, 4)) return -EIO; + ((__le32 *)val)[1] = cpu_to_le32(stationhigh); + ((__le32 *)val)[0] = cpu_to_le32(stationlow); - result = ((u64) stationlow) >> 16; - result |= (u64) stationhigh << 16; - memcpy(*addr, &result, sizeof(netxen_ethernet_macaddr_t)); + memcpy(addr, val + 2, 6); return 0; } @@ -880,7 +880,7 @@ int netxen_niu_xg_macaddr_get(struct netxen_adapter *adapter, int phy, int netxen_niu_xg_set_promiscuous_mode(struct netxen_adapter *adapter, int port, netxen_niu_prom_mode_t mode) { - __le32 reg; + __u32 reg; if ((port < 0) || (port > NETXEN_NIU_MAX_GBE_PORTS)) return -EINVAL; From 6fd6b17c6d9713f56b5f20903ec3e00fa6cc435e Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 31 Jan 2007 16:43:36 -0800 Subject: [PATCH 284/326] Revert "[PATCH] mm: micro optimise zone_watermark_ok" This reverts commit e80ee884ae0e3794ef2b65a18a767d502ad712ee. Pawel Sikora had a boot-time oops due to it - because the sign change invalidates the following comparisons, since 'free_pages' can be negative. The micro-optimization just isn't worth it. Bisected-by: Pawel Sikora Acked-by: Andrew Morton Cc: Nick Piggin Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index fc5b5442e942..2c606cc922a5 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -989,8 +989,7 @@ int zone_watermark_ok(struct zone *z, int order, unsigned long mark, int classzone_idx, int alloc_flags) { /* free_pages my go negative - that's OK */ - unsigned long min = mark; - long free_pages = z->free_pages - (1 << order) + 1; + long min = mark, free_pages = z->free_pages - (1 << order) + 1; int o; if (alloc_flags & ALLOC_HIGH) From e7bc537db9e13adee1f294aa370f16d80f40b73d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 1 Feb 2007 13:08:45 +0000 Subject: [PATCH 285/326] [PATCH] fix frv headers_check a) registers.h is really needed there b) include of asm-generic/termios should be under __KERNEL__ c) includes of asm-generic/{memory_model,page} should be under __KERNEL (nothing in there that would work in userland) d) a lot of stuff in ptrace.h should be under __KERNEL__. Signed-off-by: Al Viro Acked-by: David Howells Signed-off-by: Linus Torvalds --- include/asm-frv/Kbuild | 6 ++++++ include/asm-frv/page.h | 4 ++-- include/asm-frv/ptrace.h | 4 ++++ include/asm-frv/termios.h | 2 ++ 4 files changed, 14 insertions(+), 2 deletions(-) diff --git a/include/asm-frv/Kbuild b/include/asm-frv/Kbuild index c68e1680da01..966a9836d556 100644 --- a/include/asm-frv/Kbuild +++ b/include/asm-frv/Kbuild @@ -1 +1,7 @@ include include/asm-generic/Kbuild.asm + +header-y += registers.h + +unifdef-y += termios.h +unifdef-y += ptrace.h +unifdef-y += page.h diff --git a/include/asm-frv/page.h b/include/asm-frv/page.h index 134cc0cdf6c2..213d92fd652a 100644 --- a/include/asm-frv/page.h +++ b/include/asm-frv/page.h @@ -76,8 +76,6 @@ extern unsigned long max_pfn; #endif /* __ASSEMBLY__ */ -#endif /* __KERNEL__ */ - #ifdef CONFIG_CONTIGUOUS_PAGE_ALLOC #define WANT_PAGE_VIRTUAL 1 #endif @@ -85,4 +83,6 @@ extern unsigned long max_pfn; #include #include +#endif /* __KERNEL__ */ + #endif /* _ASM_PAGE_H */ diff --git a/include/asm-frv/ptrace.h b/include/asm-frv/ptrace.h index 9a2241b8eb1e..cf6934012b64 100644 --- a/include/asm-frv/ptrace.h +++ b/include/asm-frv/ptrace.h @@ -12,9 +12,11 @@ #define _ASM_PTRACE_H #include +#ifdef __KERNEL__ #include #define in_syscall(regs) (((regs)->tbr & TBR_TT) == TBR_TT_TRAP0) +#endif #define PT_PSR 0 @@ -60,6 +62,7 @@ #define PTRACE_GETFDPIC_EXEC 0 /* [addr] request the executable loadmap */ #define PTRACE_GETFDPIC_INTERP 1 /* [addr] request the interpreter loadmap */ +#ifdef __KERNEL__ #ifndef __ASSEMBLY__ /* @@ -74,6 +77,7 @@ register struct pt_regs *__frame asm("gr28"); extern unsigned long user_stack(const struct pt_regs *); extern void show_regs(struct pt_regs *); #define profile_pc(regs) ((regs)->pc) +#endif #endif /* !__ASSEMBLY__ */ #endif /* _ASM_PTRACE_H */ diff --git a/include/asm-frv/termios.h b/include/asm-frv/termios.h index b4a664e7f55c..8840cf95e8dd 100644 --- a/include/asm-frv/termios.h +++ b/include/asm-frv/termios.h @@ -69,6 +69,8 @@ struct termio { #define N_SYNC_PPP 14 #define N_HCI 15 /* Bluetooth HCI UART */ +#ifdef __KERNEL__ #include +#endif #endif /* _ASM_TERMIOS_H */ From aaba6d4bf6f2a52a0c30ad1da4374ba24bd7163b Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 1 Feb 2007 13:52:28 +0000 Subject: [PATCH 286/326] [PATCH] mca_nmi_hook() can be called at any point ... and having it __init is a bad idea. Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- arch/i386/mach-default/setup.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/i386/mach-default/setup.c b/arch/i386/mach-default/setup.c index c511705c386c..cc2f519b2f7f 100644 --- a/arch/i386/mach-default/setup.c +++ b/arch/i386/mach-default/setup.c @@ -102,7 +102,7 @@ void __init time_init_hook(void) * along the MCA bus. Use this to hook into that chain if you will need * it. **/ -void __init mca_nmi_hook(void) +void mca_nmi_hook(void) { /* If I recall correctly, there's a whole bunch of other things that * we can do to check for NMI problems, but that's all I know about From eb7972271720bfc64dc8bacc5b15f874c0bcc859 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 1 Feb 2007 13:52:38 +0000 Subject: [PATCH 287/326] [PATCH] ide section fixes a) cleanup_module() should be __exit b) externs should match reality Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- drivers/ide/ide.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c index 3b334af0c7b9..6c9bd5165bdb 100644 --- a/drivers/ide/ide.c +++ b/drivers/ide/ide.c @@ -1781,9 +1781,9 @@ static int __init ide_setup(char *s) return 1; } -extern void pnpide_init(void); -extern void pnpide_exit(void); -extern void h8300_ide_init(void); +extern void __init pnpide_init(void); +extern void __exit pnpide_exit(void); +extern void __init h8300_ide_init(void); /* * probe_for_hwifs() finds/initializes "known" IDE interfaces @@ -2088,7 +2088,7 @@ int __init init_module (void) return ide_init(); } -void cleanup_module (void) +void __exit cleanup_module (void) { int index; From fc2dd2e51a1940acac665696e6a70a1a73dc90a4 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 1 Feb 2007 13:52:43 +0000 Subject: [PATCH 288/326] [PATCH] endianness bug: ntohl() misspelled as >> 24 in fh_verify(). Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- fs/nfsd/nfsfh.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c index 98338a569dc0..c59d6fbb7a6b 100644 --- a/fs/nfsd/nfsfh.c +++ b/fs/nfsd/nfsfh.c @@ -269,7 +269,7 @@ fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, int access) "acc=%x, error=%d\n", dentry->d_parent->d_name.name, dentry->d_name.name, - access, (error >> 24)); + access, ntohl(error)); } out: if (exp && !IS_ERR(exp)) From 9abcf40b1d1443e6f0ef86e6a822193142a34abc Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 1 Feb 2007 13:52:48 +0000 Subject: [PATCH 289/326] [PATCH] fork_idle() should be __cpuinit, not __devinit Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- kernel/fork.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/fork.c b/kernel/fork.c index fc723e595cd5..d57118da73ff 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1313,7 +1313,7 @@ noinline struct pt_regs * __devinit __attribute__((weak)) idle_regs(struct pt_re return regs; } -struct task_struct * __devinit fork_idle(int cpu) +struct task_struct * __cpuinit fork_idle(int cpu) { struct task_struct *task; struct pt_regs regs; From 2a3d4f1f1f839e354ebd7d40b2d5d8ac8481a930 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 1 Feb 2007 13:52:23 +0000 Subject: [PATCH 290/326] [PATCH] __crc_... is intended to be absolute i386 boot/compressed/relocs checks for absolute symbols and warns about unexpected ones. If you build with modversions, you get ~2500 warnings about __crc_. These suckers are really absolute symbols - we do _not_ want to modify them on relocation. They are generated by genksyms - EXPORT_... generates a weak alias, then genksyms produces an ld script with __crc_ = and it's fed to ld to produce the final object file. Their only use is to match kernel and module at modprobe time; they _must_ be absolute. boot/compressed/relocs has a whitelist of known absolute symbols, but it doesn't know about __crc_... stuff. As the result, we get shitloads of false positives on any ld(1) version. Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- arch/i386/boot/compressed/relocs.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/i386/boot/compressed/relocs.c b/arch/i386/boot/compressed/relocs.c index 468da89153c4..881951ca03e1 100644 --- a/arch/i386/boot/compressed/relocs.c +++ b/arch/i386/boot/compressed/relocs.c @@ -43,6 +43,8 @@ static int is_safe_abs_reloc(const char* sym_name) /* Match found */ return 1; } + if (strncmp(sym_name, "__crc_", 6) == 0) + return 1; return 0; } From 472ba91dd9ce76b586d4d513f7e3448330eec7eb Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 1 Feb 2007 13:52:54 +0000 Subject: [PATCH 291/326] [PATCH] efi_set_rtc_mmss() is not __init fix the extern in efi.h Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- include/linux/efi.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/efi.h b/include/linux/efi.h index df1c91855f0e..f8ebd7c1ddb3 100644 --- a/include/linux/efi.h +++ b/include/linux/efi.h @@ -301,7 +301,7 @@ extern int __init efi_uart_console_only (void); extern void efi_initialize_iomem_resources(struct resource *code_resource, struct resource *data_resource); extern unsigned long efi_get_time(void); -extern int __init efi_set_rtc_mmss(unsigned long nowtime); +extern int efi_set_rtc_mmss(unsigned long nowtime); extern int is_available_memory(efi_memory_desc_t * md); extern struct efi_memory_map memmap; From b4cff8464b12b71fd6573c9b9dd762d2d390ef6c Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 1 Feb 2007 13:52:33 +0000 Subject: [PATCH 292/326] [PATCH] sanitize sections for sparc32 smp a) sun4d_boot_one_cpu() should be __cpuinit (called only from __cpuinit __cpu_up(), for one thing, leads to calls of __cpuinit functions for another). b) got externs in arch/sparc/kernel/smp.c to match reality. Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- arch/sparc/kernel/smp.c | 8 ++++---- arch/sparc/kernel/sun4d_smp.c | 2 +- 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/arch/sparc/kernel/smp.c b/arch/sparc/kernel/smp.c index 276f22881d0f..6b5f26b0fb75 100644 --- a/arch/sparc/kernel/smp.c +++ b/arch/sparc/kernel/smp.c @@ -292,8 +292,8 @@ int setup_profiling_timer(unsigned int multiplier) void __init smp_prepare_cpus(unsigned int max_cpus) { - extern void smp4m_boot_cpus(void); - extern void smp4d_boot_cpus(void); + extern void __init smp4m_boot_cpus(void); + extern void __init smp4d_boot_cpus(void); int i, cpuid, extra; printk("Entering SMP Mode...\n"); @@ -375,8 +375,8 @@ void __init smp_prepare_boot_cpu(void) int __cpuinit __cpu_up(unsigned int cpu) { - extern int smp4m_boot_one_cpu(int); - extern int smp4d_boot_one_cpu(int); + extern int __cpuinit smp4m_boot_one_cpu(int); + extern int __cpuinit smp4d_boot_one_cpu(int); int ret=0; switch(sparc_cpu_model) { diff --git a/arch/sparc/kernel/sun4d_smp.c b/arch/sparc/kernel/sun4d_smp.c index c80ea61e8ba0..c69de5d4863d 100644 --- a/arch/sparc/kernel/sun4d_smp.c +++ b/arch/sparc/kernel/sun4d_smp.c @@ -164,7 +164,7 @@ void __init smp4d_boot_cpus(void) local_flush_cache_all(); } -int smp4d_boot_one_cpu(int i) +int __cpuinit smp4d_boot_one_cpu(int i) { extern unsigned long sun4d_cpu_startup; unsigned long *entry = &sun4d_cpu_startup; From 9d6ed92196f7acdd1052b0828bb1e2f1a7241815 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 1 Feb 2007 13:52:59 +0000 Subject: [PATCH 293/326] [PATCH] radio modems sitting on serial port are not for s390 Won't build (request_irq()/free_irq()), even if you manage to find an s390 box with 8250-compatible UART they are expecting. Signed-off-by: Al Viro Acked-by: Martin Schwidefsky Signed-off-by: Linus Torvalds --- drivers/net/hamradio/Kconfig | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/hamradio/Kconfig b/drivers/net/hamradio/Kconfig index 896aa02000d7..feb0ada7a025 100644 --- a/drivers/net/hamradio/Kconfig +++ b/drivers/net/hamradio/Kconfig @@ -113,7 +113,7 @@ config SCC_TRXECHO config BAYCOM_SER_FDX tristate "BAYCOM ser12 fullduplex driver for AX.25" - depends on AX25 + depends on AX25 && !S390 select CRC_CCITT ---help--- This is one of two drivers for Baycom style simple amateur radio @@ -133,7 +133,7 @@ config BAYCOM_SER_FDX config BAYCOM_SER_HDX tristate "BAYCOM ser12 halfduplex driver for AX.25" - depends on AX25 + depends on AX25 && !S390 select CRC_CCITT ---help--- This is one of two drivers for Baycom style simple amateur radio @@ -181,7 +181,7 @@ config BAYCOM_EPP config YAM tristate "YAM driver for AX.25" - depends on AX25 + depends on AX25 && !S390 help The YAM is a modem for packet radio which connects to the serial port and includes some of the functions of a Terminal Node From 04add672cf98a788e9e0d753b2ccfa4a3a0caf56 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 1 Feb 2007 13:53:04 +0000 Subject: [PATCH 294/326] [PATCH] uml-i386: fix build breakage with CONFIG_HIGHMEM missing helper used by arch/i386/mm/highmem.c, which is pulled into build on that configuration. Signed-off-by: Al Viro Acked-by: Jeff Dike Signed-off-by: Linus Torvalds --- include/asm-um/pgtable.h | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/include/asm-um/pgtable.h b/include/asm-um/pgtable.h index 188f72621776..e57ff136ee51 100644 --- a/include/asm-um/pgtable.h +++ b/include/asm-um/pgtable.h @@ -408,6 +408,15 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot) #include +#ifdef CONFIG_HIGHMEM +/* Clear a kernel PTE and flush it from the TLB */ +#define kpte_clear_flush(ptep, vaddr) \ +do { \ + pte_clear(&init_mm, vaddr, ptep); \ + __flush_tlb_one(vaddr); \ +} while (0) +#endif + #endif #endif From cb7468ef4cce8f240604b80b82ac157fa9930e94 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Wed, 31 Jan 2007 23:48:12 -0800 Subject: [PATCH 295/326] [PATCH] via quirk update Add special handling for the VT82C686. Signed-off-by: Jean Delvare Cc: Alan Cox Cc: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/pci/quirks.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 16945c2ba2ca..dcc0c1ab95ed 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -661,9 +661,11 @@ static void quirk_via_bridge(struct pci_dev *dev) /* See what bridge we have and find the device ranges */ switch (dev->device) { case PCI_DEVICE_ID_VIA_82C686: - /* 82C686 is special */ - via_vlink_dev_lo = 7; - via_vlink_dev_hi = 7; + /* The VT82C686 is special, it attaches to PCI and can have + any device number. All its subdevices are functions of + that single device. */ + via_vlink_dev_lo = PCI_SLOT(dev->devfn); + via_vlink_dev_hi = PCI_SLOT(dev->devfn); break; case PCI_DEVICE_ID_VIA_8237: case PCI_DEVICE_ID_VIA_8237A: From 6a4c24ec52128c1f57b7d2d24cf4dd13fc23f474 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Wed, 31 Jan 2007 23:48:13 -0800 Subject: [PATCH 296/326] [PATCH] pci: remove warning messages Remove these recently-added warnings. They don't tell us anythng very interesting and Kumar says "On an embedded PPC reference system I see this message 6 times when I've got no cards in the PCI slots." Acked-by: Kumar Gala Acked-by: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/pci/search.c | 10 ++-------- 1 file changed, 2 insertions(+), 8 deletions(-) diff --git a/drivers/pci/search.c b/drivers/pci/search.c index fab381ed853c..b2653c4afe9e 100644 --- a/drivers/pci/search.c +++ b/drivers/pci/search.c @@ -200,11 +200,8 @@ static struct pci_dev * pci_find_subsys(unsigned int vendor, * can cause some machines to crash. So here we detect and flag that * situation and bail out early. */ - if (unlikely(list_empty(&pci_devices))) { - printk(KERN_INFO "pci_find_subsys() called while pci_devices " - "is still empty\n"); + if (unlikely(list_empty(&pci_devices))) return NULL; - } down_read(&pci_bus_sem); n = from ? from->global_list.next : pci_devices.next; @@ -278,11 +275,8 @@ pci_get_subsys(unsigned int vendor, unsigned int device, * can cause some machines to crash. So here we detect and flag that * situation and bail out early. */ - if (unlikely(list_empty(&pci_devices))) { - printk(KERN_NOTICE "pci_get_subsys() called while pci_devices " - "is still empty\n"); + if (unlikely(list_empty(&pci_devices))) return NULL; - } down_read(&pci_bus_sem); n = from ? from->global_list.next : pci_devices.next; From 432bd6cbf9f016f5480153b1cdfbd046f8d4fb1e Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Wed, 31 Jan 2007 23:48:13 -0800 Subject: [PATCH 297/326] [PATCH] KVM: fix lockup on 32-bit intel hosts with nx disabled in the bios Intel hosts, without long mode, and with nx support disabled in the bios have an efer that is readable but not writable. This causes a lockup on switch to guest mode (even though it should exit with reason 34 according to the documentation). Signed-off-by: Avi Kivity Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/kvm/vmx.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/kvm/vmx.c b/drivers/kvm/vmx.c index 27f2751c3baa..54c35c0b3181 100644 --- a/drivers/kvm/vmx.c +++ b/drivers/kvm/vmx.c @@ -1116,6 +1116,8 @@ static int vmx_vcpu_setup(struct kvm_vcpu *vcpu) if (rdmsr_safe(index, &data_low, &data_high) < 0) continue; + if (wrmsr_safe(index, data_low, data_high) < 0) + continue; data = data_low | ((u64)data_high << 32); vcpu->host_msrs[j].index = index; vcpu->host_msrs[j].reserved = 0; From 7d8952440f4090522b740257f1c6b2cf96413969 Mon Sep 17 00:00:00 2001 From: Guillaume Chazarain Date: Wed, 31 Jan 2007 23:48:14 -0800 Subject: [PATCH 298/326] [PATCH] procfs: Fix listing of /proc/NOT_A_TGID/task Listing /proc/PID/task were PID is not a TGID should not result in duplicated entries. [g ~]$ pidof thunderbird-bin 2751 [g ~]$ ls /proc/2751/task 2751 2770 2771 2824 2826 2834 2835 2851 2853 [g ~]$ ls /proc/2770/task 2751 2770 2771 2824 2826 2834 2835 2851 2853 2770 2771 2824 2826 2834 2835 2851 2853 [g ~]$ Signed-off-by: Guillaume Chazarain Acked-by: "Eric W. Biederman" Cc: Oleg Nesterov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/base.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/fs/proc/base.c b/fs/proc/base.c index ff7a66850602..1a979ea3b379 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -2328,13 +2328,23 @@ static int proc_task_readdir(struct file * filp, void * dirent, filldir_t filldi { struct dentry *dentry = filp->f_path.dentry; struct inode *inode = dentry->d_inode; - struct task_struct *leader = get_proc_task(inode); + struct task_struct *leader = NULL; struct task_struct *task; int retval = -ENOENT; ino_t ino; int tid; unsigned long pos = filp->f_pos; /* avoiding "long long" filp->f_pos */ + task = get_proc_task(inode); + if (!task) + goto out_no_task; + rcu_read_lock(); + if (pid_alive(task)) { + leader = task->group_leader; + get_task_struct(leader); + } + rcu_read_unlock(); + put_task_struct(task); if (!leader) goto out_no_task; retval = 0; From d346cce308f7fc99c7ffdb62060ed404fa340a1c Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Wed, 31 Jan 2007 23:48:17 -0800 Subject: [PATCH 299/326] [PATCH] sysrq: showBlockedTasks is sysrq-W Change SysRq showBlockedTasks from sysrq-X to sysrq-W and show that in the Help message. It was previously done via X, but X is already used for Xmon on ppc & powerpc platforms and this collision needs to be avoided. All callers of register_sysrq_key() are now marked in the sysrq op/key table. I didn't mark 'h' as Help because Help is just printed for any unknown key, such as '?'. Added some omitted sysrq key entries in the sysrq.txt file. Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/sysrq.txt | 66 ++++++++++++++++++++++------------------- drivers/char/sysrq.c | 20 +++++++------ 2 files changed, 46 insertions(+), 40 deletions(-) diff --git a/Documentation/sysrq.txt b/Documentation/sysrq.txt index e0188a23fd5e..61613166981b 100644 --- a/Documentation/sysrq.txt +++ b/Documentation/sysrq.txt @@ -1,6 +1,6 @@ Linux Magic System Request Key Hacks -Documentation for sysrq.c version 1.15 -Last update: $Date: 2001/01/28 10:15:59 $ +Documentation for sysrq.c +Last update: 2007-JAN-06 * What is the magic SysRq key? ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ @@ -35,7 +35,7 @@ You can set the value in the file by the following command: Note that the value of /proc/sys/kernel/sysrq influences only the invocation via a keyboard. Invocation of any operation via /proc/sysrq-trigger is always -allowed. +allowed (by a user with admin privileges). * How do I use the magic SysRq key? ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ @@ -58,7 +58,7 @@ On PowerPC - Press 'ALT - Print Screen (or F13) - , On other - If you know of the key combos for other architectures, please let me know so I can add them to this section. -On all - write a character to /proc/sysrq-trigger. eg: +On all - write a character to /proc/sysrq-trigger. e.g.: echo t > /proc/sysrq-trigger @@ -74,6 +74,8 @@ On all - write a character to /proc/sysrq-trigger. eg: 'c' - Will perform a kexec reboot in order to take a crashdump. +'d' - Shows all locks that are held. + 'o' - Will shut your system off (if configured and supported). 's' - Will attempt to sync all mounted filesystems. @@ -87,38 +89,43 @@ On all - write a character to /proc/sysrq-trigger. eg: 'm' - Will dump current memory info to your console. +'n' - Used to make RT tasks nice-able + 'v' - Dumps Voyager SMP processor info to your console. +'w' - Dumps tasks that are in uninterruptable (blocked) state. + +'x' - Used by xmon interface on ppc/powerpc platforms. + '0'-'9' - Sets the console log level, controlling which kernel messages will be printed to your console. ('0', for example would make it so that only emergency messages like PANICs or OOPSes would make it to your console.) -'f' - Will call oom_kill to kill a memory hog process +'f' - Will call oom_kill to kill a memory hog process. 'e' - Send a SIGTERM to all processes, except for init. +'g' - Used by kgdb on ppc platforms. + 'i' - Send a SIGKILL to all processes, except for init. -'l' - Send a SIGKILL to all processes, INCLUDING init. (Your system - will be non-functional after this.) - -'h' - Will display help ( actually any other key than those listed +'h' - Will display help (actually any other key than those listed above will display help. but 'h' is easy to remember :-) * Okay, so what can I use them for? ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Well, un'R'aw is very handy when your X server or a svgalib program crashes. -sa'K' (Secure Access Key) is useful when you want to be sure there are no -trojan program is running at console and which could grab your password -when you would try to login. It will kill all programs on given console -and thus letting you make sure that the login prompt you see is actually +sa'K' (Secure Access Key) is useful when you want to be sure there is no +trojan program running at console which could grab your password +when you would try to login. It will kill all programs on given console, +thus letting you make sure that the login prompt you see is actually the one from init, not some trojan program. IMPORTANT: In its true form it is not a true SAK like the one in a :IMPORTANT IMPORTANT: c2 compliant system, and it should not be mistaken as :IMPORTANT IMPORTANT: such. :IMPORTANT - It seems other find it useful as (System Attention Key) which is + It seems others find it useful as (System Attention Key) which is useful when you want to exit a program that will not let you switch consoles. (For example, X or a svgalib program.) @@ -139,8 +146,8 @@ OK or Done message...) Again, the unmount (remount read-only) hasn't taken place until you see the "OK" and "Done" message appear on the screen. -The loglevel'0'-'9' is useful when your console is being flooded with -kernel messages you do not want to see. Setting '0' will prevent all but +The loglevels '0'-'9' are useful when your console is being flooded with +kernel messages you do not want to see. Selecting '0' will prevent all but the most urgent kernel messages from reaching your console. (They will still be logged if syslogd/klogd are alive, though.) @@ -152,7 +159,7 @@ processes. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ That happens to me, also. I've found that tapping shift, alt, and control on both sides of the keyboard, and hitting an invalid sysrq sequence again -will fix the problem. (ie, something like alt-sysrq-z). Switching to another +will fix the problem. (i.e., something like alt-sysrq-z). Switching to another virtual console (ALT+Fn) and then back again should also help. * I hit SysRq, but nothing seems to happen, what's wrong? @@ -174,11 +181,11 @@ handler function you will use, B) a help_msg string, that will print when SysRQ prints help, and C) an action_msg string, that will print right before your handler is called. Your handler must conform to the prototype in 'sysrq.h'. -After the sysrq_key_op is created, you can call the macro -register_sysrq_key(int key, struct sysrq_key_op *op_p) that is defined in -sysrq.h, this will register the operation pointed to by 'op_p' at table -key 'key', if that slot in the table is blank. At module unload time, you must -call the macro unregister_sysrq_key(int key, struct sysrq_key_op *op_p), which +After the sysrq_key_op is created, you can call the kernel function +register_sysrq_key(int key, struct sysrq_key_op *op_p); this will +register the operation pointed to by 'op_p' at table key 'key', +if that slot in the table is blank. At module unload time, you must call +the function unregister_sysrq_key(int key, struct sysrq_key_op *op_p), which will remove the key op pointed to by 'op_p' from the key 'key', if and only if it is currently registered in that slot. This is in case the slot has been overwritten since you registered it. @@ -186,15 +193,12 @@ overwritten since you registered it. The Magic SysRQ system works by registering key operations against a key op lookup table, which is defined in 'drivers/char/sysrq.c'. This key table has a number of operations registered into it at compile time, but is mutable, -and 4 functions are exported for interface to it: __sysrq_lock_table, -__sysrq_unlock_table, __sysrq_get_key_op, and __sysrq_put_key_op. The -functions __sysrq_swap_key_ops and __sysrq_swap_key_ops_nolock are defined -in the header itself, and the REGISTER and UNREGISTER macros are built from -these. More complex (and dangerous!) manipulations of the table are possible -using these functions, but you must be careful to always lock the table before -you read or write from it, and to unlock it again when you are done. (And of -course, to never ever leave an invalid pointer in the table). Null pointers in -the table are always safe :) +and 2 functions are exported for interface to it: + register_sysrq_key and unregister_sysrq_key. +Of course, never ever leave an invalid pointer in the table. I.e., when +your module that called register_sysrq_key() exits, it must call +unregister_sysrq_key() to clean up the sysrq key table entry that it used. +Null pointers in the table are always safe. :) If for some reason you feel the need to call the handle_sysrq function from within a function called by handle_sysrq, you must be aware that you are in diff --git a/drivers/char/sysrq.c b/drivers/char/sysrq.c index 13935235e066..7fd3cd5ddf21 100644 --- a/drivers/char/sysrq.c +++ b/drivers/char/sysrq.c @@ -215,7 +215,7 @@ static void sysrq_handle_showstate_blocked(int key, struct tty_struct *tty) } static struct sysrq_key_op sysrq_showstate_blocked_op = { .handler = sysrq_handle_showstate_blocked, - .help_msg = "showBlockedTasks", + .help_msg = "shoW-blocked-tasks", .action_msg = "Show Blocked State", .enable_mask = SYSRQ_ENABLE_DUMP, }; @@ -315,15 +315,16 @@ static struct sysrq_key_op *sysrq_key_table[36] = { &sysrq_loglevel_op, /* 9 */ /* - * Don't use for system provided sysrqs, it is handled specially on - * sparc and will never arrive + * a: Don't use for system provided sysrqs, it is handled specially on + * sparc and will never arrive. */ NULL, /* a */ &sysrq_reboot_op, /* b */ - &sysrq_crashdump_op, /* c */ + &sysrq_crashdump_op, /* c & ibm_emac driver debug */ &sysrq_showlocks_op, /* d */ &sysrq_term_op, /* e */ &sysrq_moom_op, /* f */ + /* g: May be registered by ppc for kgdb */ NULL, /* g */ NULL, /* h */ &sysrq_kill_op, /* i */ @@ -332,18 +333,19 @@ static struct sysrq_key_op *sysrq_key_table[36] = { NULL, /* l */ &sysrq_showmem_op, /* m */ &sysrq_unrt_op, /* n */ - /* This will often be registered as 'Off' at init time */ + /* o: This will often be registered as 'Off' at init time */ NULL, /* o */ &sysrq_showregs_op, /* p */ NULL, /* q */ - &sysrq_unraw_op, /* r */ + &sysrq_unraw_op, /* r */ &sysrq_sync_op, /* s */ &sysrq_showstate_op, /* t */ &sysrq_mountro_op, /* u */ - /* May be assigned at init time by SMP VOYAGER */ + /* v: May be registered at init time by SMP VOYAGER */ NULL, /* v */ - NULL, /* w */ - &sysrq_showstate_blocked_op, /* x */ + &sysrq_showstate_blocked_op, /* w */ + /* x: May be registered on ppc/powerpc for xmon */ + NULL, /* x */ NULL, /* y */ NULL /* z */ }; From fb594d31aa2d133ea89d4ead964c51262b331407 Mon Sep 17 00:00:00 2001 From: Bartlomiej Zolnierkiewicz Date: Thu, 1 Feb 2007 14:12:27 +0100 Subject: [PATCH 300/326] [PATCH] via82cxxx: fix typo ("cx7000" should be corrected to "cx700") Noticed by JosephChan@via.com.tw. Signed-off-by: Bartlomiej Zolnierkiewicz Signed-off-by: Linus Torvalds --- drivers/ide/pci/via82cxxx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/ide/pci/via82cxxx.c b/drivers/ide/pci/via82cxxx.c index a98b4d38b9dd..6fb6e50b8231 100644 --- a/drivers/ide/pci/via82cxxx.c +++ b/drivers/ide/pci/via82cxxx.c @@ -78,7 +78,7 @@ static struct via_isa_bridge { u8 rev_max; u16 flags; } via_isa_bridges[] = { - { "cx7000", PCI_DEVICE_ID_VIA_CX700, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, + { "cx700", PCI_DEVICE_ID_VIA_CX700, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, { "vt8237s", PCI_DEVICE_ID_VIA_8237S, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, { "vt6410", PCI_DEVICE_ID_VIA_6410, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, { "vt8251", PCI_DEVICE_ID_VIA_8251, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, From e34efe3b100d0fbdf053128956c3dd0bc68754d6 Mon Sep 17 00:00:00 2001 From: Haavard Skinnemoen Date: Thu, 1 Feb 2007 16:49:31 +0100 Subject: [PATCH 301/326] [PATCH] Remove avr32@atmel.com from MAINTAINERS avr32@atmel.com is a technical support address and is not really appropriate for sending patches. Lots of annoying automatics getting in the way. I'm still the maintainer of all the entries touched by this patch, so nothing changes with regard to the "Supported" status of the AVR32 architecture or the macb driver. Signed-off-by: Haavard Skinnemoen Signed-off-by: Linus Torvalds --- MAINTAINERS | 6 ------ 1 file changed, 6 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index a275f72ed5f7..465e083237a7 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -598,8 +598,6 @@ W: http://linux-atm.sourceforge.net S: Maintained ATMEL MACB ETHERNET DRIVER -P: Atmel AVR32 Support Team -M: avr32@atmel.com P: Haavard Skinnemoen M: hskinnemoen@atmel.com S: Supported @@ -620,8 +618,6 @@ T: git kernel.org:/pub/scm/linux/kernel/git/dwmw2/audit-2.6.git S: Maintained AVR32 ARCHITECTURE -P: Atmel AVR32 Support Team -M: avr32@atmel.com P: Haavard Skinnemoen M: hskinnemoen@atmel.com W: http://www.atmel.com/products/AVR32/ @@ -630,8 +626,6 @@ W: http://avrfreaks.net/ S: Supported AVR32/AT32AP MACHINE SUPPORT -P: Atmel AVR32 Support Team -M: avr32@atmel.com P: Haavard Skinnemoen M: hskinnemoen@atmel.com S: Supported From 51bcf092917bfaa88d762879d0bbfe7619e8c16c Mon Sep 17 00:00:00 2001 From: Bob Breuer Date: Thu, 1 Feb 2007 20:24:35 -0800 Subject: [PATCH 302/326] [SPARC32]: Fix over-optimization by GCC near ip_fast_csum. In some cases such as: iph->check = 0; iph->check = ip_fast_csum((unsigned char *)iph, iph->ihl); GCC may optimize out the previous store. Observed as a failure of NFS over udp (bad checksums on ip fragments) when compiled with GCC 3.4.2. Signed-off-by: Bob Breuer Signed-off-by: David S. Miller --- include/asm-sparc/checksum.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/asm-sparc/checksum.h b/include/asm-sparc/checksum.h index 267e631e9bbc..34518ea7bf1b 100644 --- a/include/asm-sparc/checksum.h +++ b/include/asm-sparc/checksum.h @@ -151,7 +151,7 @@ static inline __sum16 ip_fast_csum(const void *iph, unsigned int ihl) "xnor\t%%g0, %0, %0" : "=r" (sum), "=&r" (iph) : "r" (ihl), "1" (iph) - : "g2", "g3", "g4", "cc"); + : "g2", "g3", "g4", "cc", "memory"); return sum; } From 239a87c87660d3b97a467a661eec927f0dfa9891 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Fri, 2 Feb 2007 00:40:36 -0800 Subject: [PATCH 303/326] [NET_SCHED]: act_ipt: fix regression in ipt action The x_tables patch broke target module autoloading in the ipt action by replacing the ipt_find_target call (which does autoloading) by xt_find_target (which doesn't do autoloading). Additionally xt_find_target may return ERR_PTR values in case of an error, which are not handled. Use xt_request_find_target, which does both autoloading and ERR_PTR handling properly. Also don't forget to drop the target module reference again when xt_check_target fails. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/sched/act_ipt.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c index a9608064a4c3..01e69138578d 100644 --- a/net/sched/act_ipt.c +++ b/net/sched/act_ipt.c @@ -55,7 +55,8 @@ static int ipt_init_target(struct ipt_entry_target *t, char *table, unsigned int struct ipt_target *target; int ret = 0; - target = xt_find_target(AF_INET, t->u.user.name, t->u.user.revision); + target = xt_request_find_target(AF_INET, t->u.user.name, + t->u.user.revision); if (!target) return -ENOENT; @@ -63,9 +64,10 @@ static int ipt_init_target(struct ipt_entry_target *t, char *table, unsigned int ret = xt_check_target(target, AF_INET, t->u.target_size - sizeof(*t), table, hook, 0, 0); - if (ret) + if (ret) { + module_put(t->u.kernel.target->me); return ret; - + } if (t->u.kernel.target->checkentry && !t->u.kernel.target->checkentry(table, NULL, t->u.kernel.target, t->data, From b659f44e4e144bae02c5beaba78a37db60783ba2 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Fri, 2 Feb 2007 00:46:35 -0800 Subject: [PATCH 304/326] [BNX2]: PHY workaround for 5709 A0. 5709 A0 copper devices will not link up with some link partners without this workaround. Update driver to 1.5.5. Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/bnx2.c | 14 ++++++++++++-- drivers/net/bnx2.h | 6 ++++++ 2 files changed, 18 insertions(+), 2 deletions(-) diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c index 953808efe551..ee7b75b976b5 100644 --- a/drivers/net/bnx2.c +++ b/drivers/net/bnx2.c @@ -57,8 +57,8 @@ #define DRV_MODULE_NAME "bnx2" #define PFX DRV_MODULE_NAME ": " -#define DRV_MODULE_VERSION "1.5.4" -#define DRV_MODULE_RELDATE "January 24, 2007" +#define DRV_MODULE_VERSION "1.5.5" +#define DRV_MODULE_RELDATE "February 1, 2007" #define RUN_AT(x) (jiffies + (x)) @@ -1356,6 +1356,14 @@ bnx2_init_copper_phy(struct bnx2 *bp) bnx2_write_phy(bp, 0x18, 0x0400); } + if (bp->phy_flags & PHY_DIS_EARLY_DAC_FLAG) { + bnx2_write_phy(bp, MII_BNX2_DSP_ADDRESS, + MII_BNX2_DSP_EXPAND_REG | 0x8); + bnx2_read_phy(bp, MII_BNX2_DSP_RW_PORT, &val); + val &= ~(1 << 8); + bnx2_write_phy(bp, MII_BNX2_DSP_RW_PORT, val); + } + if (bp->dev->mtu > 1500) { /* Set extended packet length bit */ bnx2_write_phy(bp, 0x18, 0x7); @@ -5918,6 +5926,8 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev) } else if (CHIP_NUM(bp) == CHIP_NUM_5706 || CHIP_NUM(bp) == CHIP_NUM_5708) bp->phy_flags |= PHY_CRC_FIX_FLAG; + else if (CHIP_ID(bp) == CHIP_ID_5709_A0) + bp->phy_flags |= PHY_DIS_EARLY_DAC_FLAG; if ((CHIP_ID(bp) == CHIP_ID_5708_A0) || (CHIP_ID(bp) == CHIP_ID_5708_B0) || diff --git a/drivers/net/bnx2.h b/drivers/net/bnx2.h index 13b6f9b11e01..ccbdf81c6599 100644 --- a/drivers/net/bnx2.h +++ b/drivers/net/bnx2.h @@ -6288,6 +6288,10 @@ struct l2_fhdr { #define BCM5708S_TX_ACTL3 0x17 +#define MII_BNX2_DSP_RW_PORT 0x15 +#define MII_BNX2_DSP_ADDRESS 0x17 +#define MII_BNX2_DSP_EXPAND_REG 0x0f00 + #define MIN_ETHERNET_PACKET_SIZE 60 #define MAX_ETHERNET_PACKET_SIZE 1514 #define MAX_ETHERNET_JUMBO_PACKET_SIZE 9014 @@ -6489,6 +6493,7 @@ struct bnx2 { #define PHY_INT_MODE_MASK_FLAG 0x300 #define PHY_INT_MODE_AUTO_POLLING_FLAG 0x100 #define PHY_INT_MODE_LINK_READY_FLAG 0x200 +#define PHY_DIS_EARLY_DAC_FLAG 0x400 u32 chip_id; /* chip num:16-31, rev:12-15, metal:4-11, bond_id:0-3 */ @@ -6512,6 +6517,7 @@ struct bnx2 { #define CHIP_ID_5708_A0 0x57080000 #define CHIP_ID_5708_B0 0x57081000 #define CHIP_ID_5708_B1 0x57081010 +#define CHIP_ID_5709_A0 0x57090000 #define CHIP_BOND_ID(bp) (((bp)->chip_id) & 0xf) From a53a33da864a81a238ee84055c8ced775ee25350 Mon Sep 17 00:00:00 2001 From: Auke Kok Date: Wed, 31 Jan 2007 11:02:46 -0800 Subject: [PATCH 305/326] e100: fix napi ifdefs removing needed code e100: fix napi ifdefs removing needed code From: Auke Kok The e100 driver is NAPI mode only. We need to netif_poll_disable during suspend and shutdown. The non-NAPI driver code was removed and is only avaiable in the out-of-tree e100 kernel driver. Signed-off-by: Auke Kok Signed-off-by: Jeff Garzik --- drivers/net/e100.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) diff --git a/drivers/net/e100.c b/drivers/net/e100.c index 3208dac29168..0cefef5e3f06 100644 --- a/drivers/net/e100.c +++ b/drivers/net/e100.c @@ -2718,14 +2718,12 @@ static int e100_suspend(struct pci_dev *pdev, pm_message_t state) struct net_device *netdev = pci_get_drvdata(pdev); struct nic *nic = netdev_priv(netdev); -#ifdef CONFIG_E100_NAPI if (netif_running(netdev)) netif_poll_disable(nic->netdev); -#endif del_timer_sync(&nic->watchdog); netif_carrier_off(nic->netdev); - netif_device_detach(netdev); + pci_save_state(pdev); if ((nic->flags & wol_magic) | e100_asf(nic)) { @@ -2761,16 +2759,13 @@ static int e100_resume(struct pci_dev *pdev) } #endif /* CONFIG_PM */ - static void e100_shutdown(struct pci_dev *pdev) { struct net_device *netdev = pci_get_drvdata(pdev); struct nic *nic = netdev_priv(netdev); -#ifdef CONFIG_E100_NAPI if (netif_running(netdev)) netif_poll_disable(nic->netdev); -#endif del_timer_sync(&nic->watchdog); netif_carrier_off(nic->netdev); From a55eb05a57a981f16325d035ee3a3ad10485ea0d Mon Sep 17 00:00:00 2001 From: Jens Osterkamp Date: Thu, 1 Feb 2007 12:07:47 +0100 Subject: [PATCH 306/326] spidernet : fix memory leak in spider_net_stop We forget to call spider_net_free_rx_chain_contents which does the actual dev_kfree_skb. New skbs are allocated from skbuff_head_cache on each "ifconfig up" letting the cache grow infinitely. This patch fixes it. Signed-off-by: Jens Osterkamp Signed-off-by: Jeff Garzik --- drivers/net/spider_net.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c index ebb6aa39f9c7..8ea2fc1b96cb 100644 --- a/drivers/net/spider_net.c +++ b/drivers/net/spider_net.c @@ -1925,6 +1925,8 @@ spider_net_stop(struct net_device *netdev) /* release chains */ spider_net_release_tx_chain(card, 1); + spider_net_free_rx_chain_contents(card); + spider_net_free_chain(card, &card->tx_chain); spider_net_free_chain(card, &card->rx_chain); From 435f8a605d3b56bb96212f4d70b62ecbd0629340 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Fri, 2 Feb 2007 08:07:42 -0800 Subject: [PATCH 307/326] Revert "[PATCH] fix typo in geode_configre()@cyrix.c" This reverts commit e4f0ae0ea63caceff37a13f281a72652b7ea71ba. It's not wrong, but it's not right either, and everybody seems to agree that the right fix is probably to do the ccr3 write after the ccr4 one (and that we also should clean it up a bit). And after that we need to really validate that all the bits that we write to ccr4 actually do work. The old 2.6.19 code was insane, and basically didn't change ccr4 at all (even though it certainly looks like it was the *intent* to do so). So let's revert the change that may fix things, just because it's not what was actually ever tested when the code was written, even if it _was_ the intent. There's a discussion on http://lkml.org/lkml/2007/1/9/63 that was started by the patch that now gets reverted, and that discussion may well contain the proper long-term fix. Suggested-by: Adrian Bunk Acked-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/cpu/cyrix.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/i386/kernel/cpu/cyrix.c b/arch/i386/kernel/cpu/cyrix.c index abcff92f994c..c0c3b59de32c 100644 --- a/arch/i386/kernel/cpu/cyrix.c +++ b/arch/i386/kernel/cpu/cyrix.c @@ -173,7 +173,7 @@ static void __cpuinit geode_configure(void) ccr4 = getCx86(CX86_CCR4); ccr4 |= 0x38; /* FPU fast, DTE cache, Mem bypass */ - setCx86(CX86_CCR4, ccr4); + setCx86(CX86_CCR3, ccr3); set_cx86_memwb(); set_cx86_reorder(); From 719d96991ac8d96ea318c6d56500e7ed690a4ac0 Mon Sep 17 00:00:00 2001 From: Evgeniy Dushistov Date: Fri, 2 Feb 2007 11:36:34 +0300 Subject: [PATCH 308/326] [PATCH] MAINTAINERS: ufs entry Mark ufs file system as maintainable, and add me as maintainer, to help people find appropriate person to assign bugs. Signed-off-by: Evgeniy Dushistov Signed-off-by: Linus Torvalds --- MAINTAINERS | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 465e083237a7..0ad8803a0c75 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -3593,6 +3593,12 @@ M: ysato@users.sourceforge.jp W: http://uclinux-h8.sourceforge.jp/ S: Supported +UFS FILESYSTEM +P: Evgeniy Dushistov +M: dushistov@mail.ru +L: linux-kernel@vger.kernel.org +S: Maintained + USB DIAMOND RIO500 DRIVER P: Cesar Miquel M: miquel@df.uba.ar From 77280989673ee1ef736a92617f52e2be45651833 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 2 Feb 2007 14:51:09 +0900 Subject: [PATCH 309/326] ahci/pata_jmicron: fix JMicron quirk For all JMicrons except for 361 and 368, AHCI mode enable bits in the Control(1) should be set. This used to be done in both ahci and pata_jmicron but while moving programming to PCI quirk, it was removed from ahci part while still left in pata_jmicron. The implemented JMicron PCI quirk was incorrect in that it didn't program AHCI mode enable bits. If pata_jmicron is loaded first and programs those bits, the ahci ports work; otherwise, ahci device detection fails miserably. This patch makes JMicron PCI quirk clear SATA IDE mode bits and set AHCI mode bits and remove the respective part from pata_jmicron. Tested on JMB361, 363 and 368. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/ata/pata_jmicron.c | 18 +++++------------- drivers/pci/quirks.c | 4 ++-- 2 files changed, 7 insertions(+), 15 deletions(-) diff --git a/drivers/ata/pata_jmicron.c b/drivers/ata/pata_jmicron.c index 2d661cb4df3c..d50264af2848 100644 --- a/drivers/ata/pata_jmicron.c +++ b/drivers/ata/pata_jmicron.c @@ -204,20 +204,12 @@ static int jmicron_init_one (struct pci_dev *pdev, const struct pci_device_id *i u32 reg; - if (id->driver_data != 368) { - /* Put the controller into AHCI mode in case the AHCI driver - has not yet been loaded. This can be done with either - function present */ + /* PATA controller is fn 1, AHCI is fn 0 */ + if (id->driver_data != 368 && PCI_FUNC(pdev->devfn) != 1) + return -ENODEV; - /* FIXME: We may want a way to override this in future */ - pci_write_config_byte(pdev, 0x41, 0xa1); - - /* PATA controller is fn 1, AHCI is fn 0 */ - if (PCI_FUNC(pdev->devfn) != 1) - return -ENODEV; - } - if ( id->driver_data == 365 || id->driver_data == 366) { - /* The 365/66 have two PATA channels, redirect the second */ + /* The 365/66 have two PATA channels, redirect the second */ + if (id->driver_data == 365 || id->driver_data == 366) { pci_read_config_dword(pdev, 0x80, ®); reg |= (1 << 24); /* IDE1 to PATA IDE secondary */ pci_write_config_dword(pdev, 0x80, reg); diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index dcc0c1ab95ed..c913ea4e545c 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -1262,8 +1262,8 @@ static void quirk_jmicron_dualfn(struct pci_dev *pdev) pci_read_config_dword(pdev, 0x40, &conf); /* Enable dual function mode, AHCI on fn 0, IDE fn1 */ /* Set the class codes correctly and then direct IDE 0 */ - conf &= ~0x000F0200; /* Clear bit 9 and 16-19 */ - conf |= 0x00C20002; /* Set bit 1, 17, 22, 23 */ + conf &= ~0x000FF200; /* Clear bit 9 and 12-19 */ + conf |= 0x00C2A102; /* Set 1, 8, 13, 15, 17, 22, 23 */ pci_write_config_dword(pdev, 0x40, conf); /* Reconfigure so that the PCI scanner discovers the From 54494f3a8339baad5e8f9d9b87d3ea6a3aa4f540 Mon Sep 17 00:00:00 2001 From: Alan Date: Wed, 31 Jan 2007 17:10:46 +0000 Subject: [PATCH 310/326] pata_atiixp: propogate cable detection hack from drivers/ide to the new driver Signed-off-by: Alan Cox Signed-off-by: Jeff Garzik --- drivers/ata/pata_atiixp.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/ata/pata_atiixp.c b/drivers/ata/pata_atiixp.c index 6f6672c55131..504e1dbfffd7 100644 --- a/drivers/ata/pata_atiixp.c +++ b/drivers/ata/pata_atiixp.c @@ -36,15 +36,22 @@ enum { static int atiixp_pre_reset(struct ata_port *ap) { struct pci_dev *pdev = to_pci_dev(ap->host->dev); - static struct pci_bits atiixp_enable_bits[] = { + static const struct pci_bits atiixp_enable_bits[] = { { 0x48, 1, 0x01, 0x00 }, { 0x48, 1, 0x08, 0x00 } }; + u8 udma; if (!pci_test_config_bits(pdev, &atiixp_enable_bits[ap->port_no])) return -ENOENT; - ap->cbl = ATA_CBL_PATA80; + /* Hack from drivers/ide/pci. Really we want to know how to do the + raw detection not play follow the bios mode guess */ + pci_read_config_byte(pdev, ATIIXP_IDE_UDMA_MODE + ap->port_no, &udma); + if ((udma & 0x07) >= 0x04 || (udma & 0x70) >= 0x40) + ap->cbl = ATA_CBL_PATA80; + else + ap->cbl = ATA_CBL_PATA40; return ata_std_prereset(ap); } From 05c39e502e964ae66336ca8e6960b200cff26f94 Mon Sep 17 00:00:00 2001 From: Alan Date: Wed, 31 Jan 2007 17:14:38 +0000 Subject: [PATCH 311/326] pata_via: Correct missing comments The 8237S was added to the chipsets but not to the comments. Fix this Signed-off-by: Alan Cox Signed-off-by: Jeff Garzik --- drivers/ata/pata_via.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/ata/pata_via.c b/drivers/ata/pata_via.c index 0219419cae97..f0b6c3b71429 100644 --- a/drivers/ata/pata_via.c +++ b/drivers/ata/pata_via.c @@ -23,6 +23,7 @@ * VIA VT8233c - UDMA100 * VIA VT8235 - UDMA133 * VIA VT8237 - UDMA133 + * VIA VT8237S - UDMA133 * VIA VT8251 - UDMA133 * * Most registers remain compatible across chips. Others start reserved @@ -61,7 +62,7 @@ #include #define DRV_NAME "pata_via" -#define DRV_VERSION "0.2.0" +#define DRV_VERSION "0.2.1" /* * The following comes directly from Vojtech Pavlik's ide/pci/via82cxxx From 0777721c9b270f087bf967369c9acbee3f1a12ae Mon Sep 17 00:00:00 2001 From: Alan Date: Wed, 31 Jan 2007 17:47:24 +0000 Subject: [PATCH 312/326] libata: Fix ata_busy_wait() kernel docs > Looks like you should use ata_busy_wait() here, rather than reproducing > the same code again. It waits in 10uS chunks while 1uS chunks were used in the workaround. Could indeed do that once I know the fix is right. While I'm at it the ata_busy_wait kerneldoc is borked so here's a fix Signed-off-by: Alan Cox Signed-off-by: Jeff Garzik --- include/linux/libata.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/include/linux/libata.h b/include/linux/libata.h index 22aa69e20905..91bb8ceef0b5 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -1054,6 +1054,8 @@ static inline void ata_pause(struct ata_port *ap) /** * ata_busy_wait - Wait for a port status register * @ap: Port to wait for. + * @bits: bits that must be clear + * @max: number of 10uS waits to perform * * Waits up to max*10 microseconds for the selected bits in the port's * status register to be cleared. From 49c8042996c84f0df6c49ea2e28a7ef38cd7d773 Mon Sep 17 00:00:00 2001 From: Brian King Date: Tue, 30 Jan 2007 11:32:26 -0600 Subject: [PATCH 313/326] libata: Initialize nbytes for internal sg commands Some LLDDs, like ipr, use nbytes and pad_len to determine the total data transfer length of a command. Make sure nbytes gets initialized for internally generated commands. Signed-off-by: Brian King Signed-off-by: Jeff Garzik --- drivers/ata/libata-core.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index cf707029352e..667acd283364 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -1250,6 +1250,7 @@ unsigned ata_exec_internal_sg(struct ata_device *dev, ata_sg_init(qc, sg, n_elem); qc->nsect = buflen / ATA_SECT_SIZE; + qc->nbytes = buflen; } qc->private_data = &wait; From 017f2e37ae19ccd28e5edd965741fc374194c5dd Mon Sep 17 00:00:00 2001 From: Nagendra Singh Tomar Date: Fri, 2 Feb 2007 17:34:56 +0530 Subject: [PATCH 314/326] [SCSI] sd: udev accessing an uninitialized scsi_disk field results in a crash sd_probe() calls class_device_add() even before initializing the sdkp->device variable. class_device_add() eventually results in the user mode udev program to be called. udev program can read the the allow_restart attribute of the newly created scsi device. This is resulting in a crash as the show function for allow_restart (i.e sd_show_allow_restart) returns the attribute value by reading the sdkp->device->allow_restart variable. As the sdkp->device is not initialized before calling the user mode hotplug helper, this results in a crash. The patch below solves it by calling class_device_add() only after the necessary fields in the scsi_disk structure are initialized properly. Signed-off-by: Nagendra Singh Tomar Signed-off-by: James Bottomley --- drivers/scsi/sd.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c index 978bfc1e0c6a..b781a90d6699 100644 --- a/drivers/scsi/sd.c +++ b/drivers/scsi/sd.c @@ -1647,16 +1647,6 @@ static int sd_probe(struct device *dev) if (error) goto out_put; - class_device_initialize(&sdkp->cdev); - sdkp->cdev.dev = &sdp->sdev_gendev; - sdkp->cdev.class = &sd_disk_class; - strncpy(sdkp->cdev.class_id, sdp->sdev_gendev.bus_id, BUS_ID_SIZE); - - if (class_device_add(&sdkp->cdev)) - goto out_put; - - get_device(&sdp->sdev_gendev); - sdkp->device = sdp; sdkp->driver = &sd_template; sdkp->disk = gd; @@ -1670,6 +1660,16 @@ static int sd_probe(struct device *dev) sdp->timeout = SD_MOD_TIMEOUT; } + class_device_initialize(&sdkp->cdev); + sdkp->cdev.dev = &sdp->sdev_gendev; + sdkp->cdev.class = &sd_disk_class; + strncpy(sdkp->cdev.class_id, sdp->sdev_gendev.bus_id, BUS_ID_SIZE); + + if (class_device_add(&sdkp->cdev)) + goto out_put; + + get_device(&sdp->sdev_gendev); + gd->major = sd_major((index & 0xf0) >> 4); gd->first_minor = ((index & 0xf) << 4) | (index & 0xfff00); gd->minors = 16; From 40e0cb004a6d4a7ad577724e451e8dbd6cba5a89 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Fri, 2 Feb 2007 19:33:11 -0800 Subject: [PATCH 315/326] [NETFILTER]: ctnetlink: fix compile failure with NF_CONNTRACK_MARK=n CC net/netfilter/nf_conntrack_netlink.o net/netfilter/nf_conntrack_netlink.c: In function 'ctnetlink_conntrack_event': net/netfilter/nf_conntrack_netlink.c:392: error: 'struct nf_conn' has no member named 'mark' make[3]: *** [net/netfilter/nf_conntrack_netlink.o] Error 1 Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/ipv4/netfilter/ip_conntrack_netlink.c | 2 ++ net/netfilter/nf_conntrack_netlink.c | 2 ++ 2 files changed, 4 insertions(+) diff --git a/net/ipv4/netfilter/ip_conntrack_netlink.c b/net/ipv4/netfilter/ip_conntrack_netlink.c index 6f31fad9be13..7f70b0886b83 100644 --- a/net/ipv4/netfilter/ip_conntrack_netlink.c +++ b/net/ipv4/netfilter/ip_conntrack_netlink.c @@ -374,9 +374,11 @@ static int ctnetlink_conntrack_event(struct notifier_block *this, && ctnetlink_dump_helpinfo(skb, ct) < 0) goto nfattr_failure; +#ifdef CONFIG_IP_NF_CONNTRACK_MARK if ((events & IPCT_MARK || ct->mark) && ctnetlink_dump_mark(skb, ct) < 0) goto nfattr_failure; +#endif if (events & IPCT_COUNTER_FILLING && (ctnetlink_dump_counters(skb, ct, IP_CT_DIR_ORIGINAL) < 0 || diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index 811e3e782f0f..c64f029f7052 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -389,9 +389,11 @@ static int ctnetlink_conntrack_event(struct notifier_block *this, && ctnetlink_dump_helpinfo(skb, ct) < 0) goto nfattr_failure; +#ifdef CONFIG_NF_CONNTRACK_MARK if ((events & IPCT_MARK || ct->mark) && ctnetlink_dump_mark(skb, ct) < 0) goto nfattr_failure; +#endif if (events & IPCT_COUNTER_FILLING && (ctnetlink_dump_counters(skb, ct, IP_CT_DIR_ORIGINAL) < 0 || From 3e8219806c33b64a00b0013f96f735451f30c64c Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Fri, 2 Feb 2007 19:33:52 -0800 Subject: [PATCH 316/326] [NETFILTER]: nf_conntrack_h323: fix compile error with CONFIG_IPV6=m, CONFIG_NF_CONNTRACK_H323=y Fix this by letting NF_CONNTRACK_H323 depend on (IPV6 || IPV6=n). Signed-off-by: Adrian Bunk Signed-off-by: David S. Miller --- net/netfilter/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 2a2bcb303bfa..80107d4909c5 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -165,7 +165,7 @@ config NF_CONNTRACK_FTP config NF_CONNTRACK_H323 tristate "H.323 protocol support (EXPERIMENTAL)" - depends on EXPERIMENTAL && NF_CONNTRACK + depends on EXPERIMENTAL && NF_CONNTRACK && (IPV6 || IPV6=n) help H.323 is a VoIP signalling protocol from ITU-T. As one of the most important VoIP protocols, it is widely used by voice hardware and From dee11c2364f51cac53df17d742a0c69097e29a4e Mon Sep 17 00:00:00 2001 From: Ken Chen Date: Sat, 3 Feb 2007 01:13:45 -0800 Subject: [PATCH 317/326] [PATCH] aio: fix buggy put_ioctx call in aio_complete - v2 An AIO bug was reported that sleeping function is being called in softirq context: BUG: warning at kernel/mutex.c:132/__mutex_lock_common() Call Trace: [] __mutex_lock_slowpath+0x640/0x6c0 [] mutex_lock+0x20/0x40 [] flush_workqueue+0xb0/0x1a0 [] __put_ioctx+0xc0/0x240 [] aio_complete+0x2f0/0x420 [] finished_one_bio+0x200/0x2a0 [] dio_bio_complete+0x1c0/0x200 [] dio_bio_end_aio+0x60/0x80 [] bio_endio+0x110/0x1c0 [] __end_that_request_first+0x180/0xba0 [] end_that_request_chunk+0x30/0x60 [] scsi_end_request+0x50/0x300 [scsi_mod] [] scsi_io_completion+0x200/0x8a0 [scsi_mod] [] sd_rw_intr+0x330/0x860 [sd_mod] [] scsi_finish_command+0x100/0x1c0 [scsi_mod] [] scsi_softirq_done+0x230/0x300 [scsi_mod] [] blk_done_softirq+0x160/0x1c0 [] __do_softirq+0x200/0x240 [] do_softirq+0x70/0xc0 See report: http://marc.theaimsgroup.com/?l=linux-kernel&m=116599593200888&w=2 flush_workqueue() is not allowed to be called in the softirq context. However, aio_complete() called from I/O interrupt can potentially call put_ioctx with last ref count on ioctx and triggers bug. It is simply incorrect to perform ioctx freeing from aio_complete. The bug is trigger-able from a race between io_destroy() and aio_complete(). A possible scenario: cpu0 cpu1 io_destroy aio_complete wait_for_all_aios { __aio_put_req ... ctx->reqs_active--; if (!ctx->reqs_active) return; } ... put_ioctx(ioctx) put_ioctx(ctx); __put_ioctx bam! Bug trigger! The real problem is that the condition check of ctx->reqs_active in wait_for_all_aios() is incorrect that access to reqs_active is not being properly protected by spin lock. This patch adds that protective spin lock, and at the same time removes all duplicate ref counting for each kiocb as reqs_active is already used as a ref count for each active ioctx. This also ensures that buggy call to flush_workqueue() in softirq context is eliminated. Signed-off-by: "Ken Chen" Cc: Zach Brown Cc: Suparna Bhattacharya Cc: Benjamin LaHaise Cc: Badari Pulavarty Cc: Acked-by: Jeff Moyer Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/aio.c | 20 +++++++++----------- 1 file changed, 9 insertions(+), 11 deletions(-) diff --git a/fs/aio.c b/fs/aio.c index ee20fc4240e0..55991e4132a7 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -298,17 +298,23 @@ static void wait_for_all_aios(struct kioctx *ctx) struct task_struct *tsk = current; DECLARE_WAITQUEUE(wait, tsk); + spin_lock_irq(&ctx->ctx_lock); if (!ctx->reqs_active) - return; + goto out; add_wait_queue(&ctx->wait, &wait); set_task_state(tsk, TASK_UNINTERRUPTIBLE); while (ctx->reqs_active) { + spin_unlock_irq(&ctx->ctx_lock); schedule(); set_task_state(tsk, TASK_UNINTERRUPTIBLE); + spin_lock_irq(&ctx->ctx_lock); } __set_task_state(tsk, TASK_RUNNING); remove_wait_queue(&ctx->wait, &wait); + +out: + spin_unlock_irq(&ctx->ctx_lock); } /* wait_on_sync_kiocb: @@ -424,7 +430,6 @@ static struct kiocb fastcall *__aio_get_req(struct kioctx *ctx) ring = kmap_atomic(ctx->ring_info.ring_pages[0], KM_USER0); if (ctx->reqs_active < aio_ring_avail(&ctx->ring_info, ring)) { list_add(&req->ki_list, &ctx->active_reqs); - get_ioctx(ctx); ctx->reqs_active++; okay = 1; } @@ -536,8 +541,6 @@ int fastcall aio_put_req(struct kiocb *req) spin_lock_irq(&ctx->ctx_lock); ret = __aio_put_req(ctx, req); spin_unlock_irq(&ctx->ctx_lock); - if (ret) - put_ioctx(ctx); return ret; } @@ -779,8 +782,7 @@ static int __aio_run_iocbs(struct kioctx *ctx) */ iocb->ki_users++; /* grab extra reference */ aio_run_iocb(iocb); - if (__aio_put_req(ctx, iocb)) /* drop extra ref */ - put_ioctx(ctx); + __aio_put_req(ctx, iocb); } if (!list_empty(&ctx->run_list)) return 1; @@ -997,14 +999,10 @@ int fastcall aio_complete(struct kiocb *iocb, long res, long res2) /* everything turned out well, dispose of the aiocb. */ ret = __aio_put_req(ctx, iocb); - spin_unlock_irqrestore(&ctx->ctx_lock, flags); - if (waitqueue_active(&ctx->wait)) wake_up(&ctx->wait); - if (ret) - put_ioctx(ctx); - + spin_unlock_irqrestore(&ctx->ctx_lock, flags); return ret; } From 29a002776ba5ef170446910b1f93c480cdd43706 Mon Sep 17 00:00:00 2001 From: Magnus Damm Date: Sat, 3 Feb 2007 01:13:48 -0800 Subject: [PATCH 318/326] [PATCH] kexec: Avoid migration of already disabled irqs (ia64) This patch fixes up ia64 kexec support for HP rx2620 hardware. It does this by skipping migration of already disabled irqs. This is most likely a problem on other ia64 platforms as well, but I've only been able to reproduce it on one machine so far. The full story is that handle_bad_irq() gets invoked before starting the new kernel without this patch. This seems to happen when fixup_irqs() calls generic_handle_irq() on already migrated (and disabled) irqs. So by avoiding migration of disabled irqs we stay away of handle_bad_irq(). The code has been tested on three different ia64 machines, all with good results. It is possible to trigger the same bug by offlining a processor using echo 0 > /sys/devices/system/cpu/cpuX/online. More detailed information is available in the following mail thread: http://lists.osdl.org/pipermail/fastboot/2007-January/thread.html#5774 Signed-off-by: Magnus Damm Acked-by: Simon Horman Acked-by: Zou, Nanhai Acked-by: Jay Lan Acked-by: "Luck, Tony" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/kernel/irq.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/ia64/kernel/irq.c b/arch/ia64/kernel/irq.c index 54d55e4d64f7..ce49c85c928f 100644 --- a/arch/ia64/kernel/irq.c +++ b/arch/ia64/kernel/irq.c @@ -122,6 +122,9 @@ static void migrate_irqs(void) for (irq=0; irq < NR_IRQS; irq++) { desc = irq_desc + irq; + if (desc->status == IRQ_DISABLED) + continue; + /* * No handling for now. * TBD: Implement a disable function so we can now From 24d8f6aded45aca87dec6d9c037b75b189e3d731 Mon Sep 17 00:00:00 2001 From: Peter Korsgaard Date: Sat, 3 Feb 2007 01:13:50 -0800 Subject: [PATCH 319/326] [PATCH] net/smc911x: match up spin lock/unlock smc911x_phy_configure's error handling unconditionally unlocks the spinlock even if it wasn't locked. Patch fixes it. Signed-off-by: Peter Korsgaard Cc: Jeff Garzik Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/net/smc911x.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/net/smc911x.c b/drivers/net/smc911x.c index 880d9fdd7c67..43af61438449 100644 --- a/drivers/net/smc911x.c +++ b/drivers/net/smc911x.c @@ -968,11 +968,11 @@ static void smc911x_phy_configure(struct work_struct *work) * We should not be called if phy_type is zero. */ if (lp->phy_type == 0) - goto smc911x_phy_configure_exit; + goto smc911x_phy_configure_exit_nolock; if (smc911x_phy_reset(dev, phyaddr)) { printk("%s: PHY reset timed out\n", dev->name); - goto smc911x_phy_configure_exit; + goto smc911x_phy_configure_exit_nolock; } spin_lock_irqsave(&lp->lock, flags); @@ -1041,6 +1041,7 @@ static void smc911x_phy_configure(struct work_struct *work) smc911x_phy_configure_exit: spin_unlock_irqrestore(&lp->lock, flags); +smc911x_phy_configure_exit_nolock: lp->work_pending = 0; } From 8560a10e16faccafdc2e26c4873bf4edfbbf651e Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Sat, 3 Feb 2007 01:13:55 -0800 Subject: [PATCH 320/326] [PATCH] alpha: fix epoll syscall enumerations We went and named them __NR_sys_foo instead of __NR_foo. It may be too late to change this, but we can at least add the proper names now. Signed-off-by: Mike Frysinger Cc: Richard Henderson Cc: Ivan Kokshaysky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-alpha/unistd.h | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/include/asm-alpha/unistd.h b/include/asm-alpha/unistd.h index 84313d14e780..e58a427012dd 100644 --- a/include/asm-alpha/unistd.h +++ b/include/asm-alpha/unistd.h @@ -342,9 +342,14 @@ #define __NR_io_cancel 402 #define __NR_exit_group 405 #define __NR_lookup_dcookie 406 -#define __NR_sys_epoll_create 407 -#define __NR_sys_epoll_ctl 408 -#define __NR_sys_epoll_wait 409 +#define __NR_epoll_create 407 +#define __NR_epoll_ctl 408 +#define __NR_epoll_wait 409 +/* Feb 2007: These three sys_epoll defines shouldn't be here but culling + * them would break userspace apps ... we'll kill them off in 2010 :) */ +#define __NR_sys_epoll_create __NR_epoll_create +#define __NR_sys_epoll_ctl __NR_epoll_ctl +#define __NR_sys_epoll_wait __NR_epoll_wait #define __NR_remap_file_pages 410 #define __NR_set_tid_address 411 #define __NR_restart_syscall 412 From b2e895dbd80c420bfc0937c3729b4afe073b3848 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Sat, 3 Feb 2007 01:14:01 -0800 Subject: [PATCH 321/326] [PATCH] revert blockdev direct io back to 2.6.19 version Andrew Vasquez is reporting as-iosched oopses and a 65% throughput slowdown due to the recent special-casing of direct-io against blockdevs. We don't know why either of these things are occurring. The patch minimally reverts us back to the 2.6.19 code for a 2.6.20 release. Cc: Andrew Vasquez Cc: Ken Chen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/block_dev.c | 41 +++++++++++++++++++++++++++++++++++++++++ 1 file changed, 41 insertions(+) diff --git a/fs/block_dev.c b/fs/block_dev.c index d9bdf2b3ade2..fc7028b685f2 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -129,6 +129,46 @@ blkdev_get_block(struct inode *inode, sector_t iblock, return 0; } +static int +blkdev_get_blocks(struct inode *inode, sector_t iblock, + struct buffer_head *bh, int create) +{ + sector_t end_block = max_block(I_BDEV(inode)); + unsigned long max_blocks = bh->b_size >> inode->i_blkbits; + + if ((iblock + max_blocks) > end_block) { + max_blocks = end_block - iblock; + if ((long)max_blocks <= 0) { + if (create) + return -EIO; /* write fully beyond EOF */ + /* + * It is a read which is fully beyond EOF. We return + * a !buffer_mapped buffer + */ + max_blocks = 0; + } + } + + bh->b_bdev = I_BDEV(inode); + bh->b_blocknr = iblock; + bh->b_size = max_blocks << inode->i_blkbits; + if (max_blocks) + set_buffer_mapped(bh); + return 0; +} + +static ssize_t +blkdev_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, + loff_t offset, unsigned long nr_segs) +{ + struct file *file = iocb->ki_filp; + struct inode *inode = file->f_mapping->host; + + return blockdev_direct_IO_no_locking(rw, iocb, inode, I_BDEV(inode), + iov, offset, nr_segs, blkdev_get_blocks, NULL); +} + +#if 0 static int blk_end_aio(struct bio *bio, unsigned int bytes_done, int error) { struct kiocb *iocb = bio->bi_private; @@ -323,6 +363,7 @@ blkdev_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, return PTR_ERR(page); goto completion; } +#endif static int blkdev_writepage(struct page *page, struct writeback_control *wbc) { From 72253943f7281fc5c6249d9cafd63e7de3266fe9 Mon Sep 17 00:00:00 2001 From: John Keller Date: Sat, 3 Feb 2007 01:14:02 -0800 Subject: [PATCH 322/326] [PATCH] Altix: more ACPI PRT support The SN Altix platform does not conform to the IOSAPIC IRQ routing model. Add code in acpi_unregister_gsi() to check if (acpi_irq_model == ACPI_IRQ_MODEL_PLATFORM) and return. Due to an oversight, this code was not added previously when similar code was added to acpi_register_gsi(). http://marc.theaimsgroup.com/?l=linux-acpi&m=116680983430121&w=2 Signed-off-by: John Keller Acked-by: Len Brown Cc: "Luck, Tony" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/kernel/acpi.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c index ef2fe474f107..29f05d4b68cd 100644 --- a/arch/ia64/kernel/acpi.c +++ b/arch/ia64/kernel/acpi.c @@ -609,6 +609,9 @@ EXPORT_SYMBOL(acpi_register_gsi); void acpi_unregister_gsi(u32 gsi) { + if (acpi_irq_model == ACPI_IRQ_MODEL_PLATFORM) + return; + iosapic_unregister_intr(gsi); } From 259886a7c4e4eb0089181e800d1f477cb3786875 Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Sat, 3 Feb 2007 01:14:03 -0800 Subject: [PATCH 323/326] [PATCH] x86-64: define dma noncoherent API functions x86-64 is missing these: Signed-off-by: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/asm-x86_64/dma-mapping.h | 3 +++ 1 file changed, 3 insertions(+) diff --git a/include/asm-x86_64/dma-mapping.h b/include/asm-x86_64/dma-mapping.h index be9ec6890723..49dbab09ef2b 100644 --- a/include/asm-x86_64/dma-mapping.h +++ b/include/asm-x86_64/dma-mapping.h @@ -63,6 +63,9 @@ static inline int dma_mapping_error(dma_addr_t dma_addr) return (dma_addr == bad_dma_address); } +#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f) +#define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h) + extern void *dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle, gfp_t gfp); extern void dma_free_coherent(struct device *dev, size_t size, void *vaddr, From 886ae1fa1380309d91cdb7e67bd4bf71e053c1d5 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 4 Feb 2007 03:02:17 +0000 Subject: [PATCH 324/326] [PATCH] fix rtl8150 That code doesn't do what its author apparently thought it would do... Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- drivers/usb/net/rtl8150.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/usb/net/rtl8150.c b/drivers/usb/net/rtl8150.c index e0eecda78ec1..670262a38a0f 100644 --- a/drivers/usb/net/rtl8150.c +++ b/drivers/usb/net/rtl8150.c @@ -284,7 +284,8 @@ static int write_mii_word(rtl8150_t * dev, u8 phy, __u8 indx, u16 reg) u8 data[3], tmp; data[0] = phy; - *(data + 1) = cpu_to_le16p(®); + data[1] = reg & 0xff; + data[2] = (reg >> 8) & 0xff; tmp = indx | PHY_WRITE | PHY_GO; i = 0; From 40c373cc3af9720d1cec0e32c3da26b1d220a95b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Fr=C3=A9d=C3=A9ric=20Riss?= Date: Tue, 30 Jan 2007 21:41:17 +0100 Subject: [PATCH 325/326] [PATCH] EFI x86: pass firmware call parameters on the stack When calling into the EFI firmware, the parameters need to be passed on the stack. The recent change to use -mregparm=3 breaks x86 EFI support. This patch is needed to allow the new Intel-based Macs to suspend to ram (efi.get_time is called during the suspend phase). Signed-off-by: Frederic Riss Signed-off-by: Linus Torvalds --- arch/i386/kernel/efi.c | 89 ++++++++++++++++++++++++++++++++++-------- 1 file changed, 73 insertions(+), 16 deletions(-) diff --git a/arch/i386/kernel/efi.c b/arch/i386/kernel/efi.c index b92c7f0a358a..8f9c624ace6f 100644 --- a/arch/i386/kernel/efi.c +++ b/arch/i386/kernel/efi.c @@ -472,6 +472,70 @@ static inline void __init check_range_for_systab(efi_memory_desc_t *md) } } +/* + * Wrap all the virtual calls in a way that forces the parameters on the stack. + */ + +#define efi_call_virt(f, args...) \ + ((efi_##f##_t __attribute__((regparm(0)))*)efi.systab->runtime->f)(args) + +static efi_status_t virt_efi_get_time(efi_time_t *tm, efi_time_cap_t *tc) +{ + return efi_call_virt(get_time, tm, tc); +} + +static efi_status_t virt_efi_set_time (efi_time_t *tm) +{ + return efi_call_virt(set_time, tm); +} + +static efi_status_t virt_efi_get_wakeup_time (efi_bool_t *enabled, + efi_bool_t *pending, + efi_time_t *tm) +{ + return efi_call_virt(get_wakeup_time, enabled, pending, tm); +} + +static efi_status_t virt_efi_set_wakeup_time (efi_bool_t enabled, + efi_time_t *tm) +{ + return efi_call_virt(set_wakeup_time, enabled, tm); +} + +static efi_status_t virt_efi_get_variable (efi_char16_t *name, + efi_guid_t *vendor, u32 *attr, + unsigned long *data_size, void *data) +{ + return efi_call_virt(get_variable, name, vendor, attr, data_size, data); +} + +static efi_status_t virt_efi_get_next_variable (unsigned long *name_size, + efi_char16_t *name, + efi_guid_t *vendor) +{ + return efi_call_virt(get_next_variable, name_size, name, vendor); +} + +static efi_status_t virt_efi_set_variable (efi_char16_t *name, + efi_guid_t *vendor, + unsigned long attr, + unsigned long data_size, void *data) +{ + return efi_call_virt(set_variable, name, vendor, attr, data_size, data); +} + +static efi_status_t virt_efi_get_next_high_mono_count (u32 *count) +{ + return efi_call_virt(get_next_high_mono_count, count); +} + +static void virt_efi_reset_system (int reset_type, efi_status_t status, + unsigned long data_size, + efi_char16_t *data) +{ + efi_call_virt(reset_system, reset_type, status, data_size, data); +} + /* * This function will switch the EFI runtime services to virtual mode. * Essentially, look through the EFI memmap and map every region that @@ -525,22 +589,15 @@ void __init efi_enter_virtual_mode(void) * pointers in the runtime service table to the new virtual addresses. */ - efi.get_time = (efi_get_time_t *) efi.systab->runtime->get_time; - efi.set_time = (efi_set_time_t *) efi.systab->runtime->set_time; - efi.get_wakeup_time = (efi_get_wakeup_time_t *) - efi.systab->runtime->get_wakeup_time; - efi.set_wakeup_time = (efi_set_wakeup_time_t *) - efi.systab->runtime->set_wakeup_time; - efi.get_variable = (efi_get_variable_t *) - efi.systab->runtime->get_variable; - efi.get_next_variable = (efi_get_next_variable_t *) - efi.systab->runtime->get_next_variable; - efi.set_variable = (efi_set_variable_t *) - efi.systab->runtime->set_variable; - efi.get_next_high_mono_count = (efi_get_next_high_mono_count_t *) - efi.systab->runtime->get_next_high_mono_count; - efi.reset_system = (efi_reset_system_t *) - efi.systab->runtime->reset_system; + efi.get_time = virt_efi_get_time; + efi.set_time = virt_efi_set_time; + efi.get_wakeup_time = virt_efi_get_wakeup_time; + efi.set_wakeup_time = virt_efi_set_wakeup_time; + efi.get_variable = virt_efi_get_variable; + efi.get_next_variable = virt_efi_get_next_variable; + efi.set_variable = virt_efi_set_variable; + efi.get_next_high_mono_count = virt_efi_get_next_high_mono_count; + efi.reset_system = virt_efi_reset_system; } void __init From 62d0cfcb27cf755cebdc93ca95dabc83608007cd Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 4 Feb 2007 10:44:54 -0800 Subject: [PATCH 326/326] Linux 2.6.20 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 75adfb539977..7e2750f4ca70 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 2 PATCHLEVEL = 6 SUBLEVEL = 20 -EXTRAVERSION =-rc7 +EXTRAVERSION = NAME = Homicidal Dwarf Hamster # *DOCUMENTATION*