From 436d34b36202ef724778ded1e9cb10f8c37b32bc Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 2 Apr 2008 17:28:46 +0900 Subject: libata: uninline atapi_cmd_type() Uninline atapi_cmd_type(). It doesn't really have to be inline and more case will be added which need to access unexported libata variable. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- include/linux/libata.h | 22 +--------------------- 1 file changed, 1 insertion(+), 21 deletions(-) (limited to 'include') diff --git a/include/linux/libata.h b/include/linux/libata.h index b064bfeb69ee..92c64909ed25 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -849,6 +849,7 @@ extern unsigned int ata_dev_try_classify(struct ata_device *dev, int present, */ extern void ata_tf_load(struct ata_port *ap, const struct ata_taskfile *tf); extern void ata_tf_read(struct ata_port *ap, struct ata_taskfile *tf); +extern int atapi_cmd_type(u8 opcode); extern void ata_tf_to_fis(const struct ata_taskfile *tf, u8 pmp, int is_cmd, u8 *fis); extern void ata_tf_from_fis(const u8 *fis, struct ata_taskfile *tf); @@ -1379,27 +1380,6 @@ static inline int ata_try_flush_cache(const struct ata_device *dev) ata_id_has_flush_ext(dev->id); } -static inline int atapi_cmd_type(u8 opcode) -{ - switch (opcode) { - case GPCMD_READ_10: - case GPCMD_READ_12: - return ATAPI_READ; - - case GPCMD_WRITE_10: - case GPCMD_WRITE_12: - case GPCMD_WRITE_AND_VERIFY_10: - return ATAPI_WRITE; - - case GPCMD_READ_CD: - case GPCMD_READ_CD_MSF: - return ATAPI_READ_CD; - - default: - return ATAPI_MISC; - } -} - static inline unsigned int ac_err_mask(u8 status) { if (status & (ATA_BUSY | ATA_DRQ)) -- cgit v1.2.3 From e52dcc4899cf1b7601379c31542bd91cd2997a64 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 2 Apr 2008 17:35:19 +0900 Subject: libata: ATA_12/16 doesn't fall into ATAPI_MISC SAT passthrus don't really fit into ATAPI_MISC class. SAT passthru commands always transfer multiple of 512 bytes and variable length response is not allowed. This patch creates a separate category - ATAPI_PASS_THRU - for these. This fixes HSM violation on "hdparm -I". Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- drivers/ata/libata-core.c | 5 +++++ include/linux/libata.h | 3 ++- 2 files changed, 7 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index e9b69ba489d2..be95fdb69726 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -353,6 +353,11 @@ int atapi_cmd_type(u8 opcode) case GPCMD_READ_CD_MSF: return ATAPI_READ_CD; + case ATA_16: + case ATA_12: + if (atapi_passthru16) + return ATAPI_PASS_THRU; + /* fall thru */ default: return ATAPI_MISC; } diff --git a/include/linux/libata.h b/include/linux/libata.h index 92c64909ed25..37ee881c42ac 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -350,7 +350,8 @@ enum { ATAPI_READ = 0, /* READs */ ATAPI_WRITE = 1, /* WRITEs */ ATAPI_READ_CD = 2, /* READ CD [MSF] */ - ATAPI_MISC = 3, /* the rest */ + ATAPI_PASS_THRU = 3, /* SAT pass-thru */ + ATAPI_MISC = 4, /* the rest */ }; enum ata_xfer_mask { -- cgit v1.2.3 From bae1d2507e44417455eda76d4435352fee14cf51 Mon Sep 17 00:00:00 2001 From: Ravikiran G Thirumalai Date: Fri, 4 Apr 2008 03:06:29 -0700 Subject: x86: fix breakage of vSMP irq operations 25-rc* stopped working with CONFIG_X86_VSMP on vSMP machines. Looks like the vsmp irq ops got accidentally removed during merge of x86_64 pvops in 2.6.25. -- commit 6abcd98ffafbff81f0bfd7ee1d129e634af13245 removed vsmp irq ops. Tested with both CONFIG_X86_VSMP and without CONFIG_X86_VSMP, on vSMP and non vSMP x86_64 machines. Please apply. Signed-off-by: Ravikiran Thirumalai Signed-off-by: Ingo Molnar --- include/asm-x86/irqflags.h | 29 +++++++++++++++++++++++++++++ 1 file changed, 29 insertions(+) (limited to 'include') diff --git a/include/asm-x86/irqflags.h b/include/asm-x86/irqflags.h index 92021c1ffa3a..0e2292483b35 100644 --- a/include/asm-x86/irqflags.h +++ b/include/asm-x86/irqflags.h @@ -70,6 +70,26 @@ static inline void raw_local_irq_restore(unsigned long flags) native_restore_fl(flags); } +#ifdef CONFIG_X86_VSMP + +/* + * Interrupt control for the VSMP architecture: + */ + +static inline void raw_local_irq_disable(void) +{ + unsigned long flags = __raw_local_save_flags(); + raw_local_irq_restore((flags & ~X86_EFLAGS_IF) | X86_EFLAGS_AC); +} + +static inline void raw_local_irq_enable(void) +{ + unsigned long flags = __raw_local_save_flags(); + raw_local_irq_restore((flags | X86_EFLAGS_IF) & (~X86_EFLAGS_AC)); +} + +#else + static inline void raw_local_irq_disable(void) { native_irq_disable(); @@ -80,6 +100,8 @@ static inline void raw_local_irq_enable(void) native_irq_enable(); } +#endif + /* * Used in the idle loop; sti takes one instruction cycle * to complete: @@ -137,10 +159,17 @@ static inline unsigned long __raw_local_irq_save(void) #define raw_local_irq_save(flags) \ do { (flags) = __raw_local_irq_save(); } while (0) +#ifdef CONFIG_X86_VSMP +static inline int raw_irqs_disabled_flags(unsigned long flags) +{ + return !(flags & X86_EFLAGS_IF) || (flags & X86_EFLAGS_AC); +} +#else static inline int raw_irqs_disabled_flags(unsigned long flags) { return !(flags & X86_EFLAGS_IF); } +#endif static inline int raw_irqs_disabled(void) { -- cgit v1.2.3 From 5761d64b277c287a7520b868c32d656ef03374b4 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 4 Apr 2008 16:26:10 +0200 Subject: x86: revert assign IRQs to hpet timer The commits: commit 37a47db8d7f0f38dac5acf5a13abbc8f401707fa Author: Balaji Rao Date: Wed Jan 30 13:30:03 2008 +0100 x86: assign IRQs to HPET timers, fix and commit e3f37a54f690d3e64995ea7ecea08c5ab3070faf Author: Balaji Rao Date: Wed Jan 30 13:30:03 2008 +0100 x86: assign IRQs to HPET timers have been identified to cause a regression on some platforms due to the assignement of legacy IRQs which makes the legacy devices connected to those IRQs disfunctional. Revert them. This fixes http://bugzilla.kernel.org/show_bug.cgi?id=10382 Signed-off-by: Thomas Gleixner Signed-off-by: Ingo Molnar --- arch/x86/kernel/hpet.c | 9 ++++++--- drivers/char/hpet.c | 51 +++++++------------------------------------------- include/linux/hpet.h | 2 +- 3 files changed, 14 insertions(+), 48 deletions(-) (limited to 'include') diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index 235fd6c77504..36652ea1a265 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -133,13 +133,16 @@ static void hpet_reserve_platform_timers(unsigned long id) #ifdef CONFIG_HPET_EMULATE_RTC hpet_reserve_timer(&hd, 1); #endif + hd.hd_irq[0] = HPET_LEGACY_8254; hd.hd_irq[1] = HPET_LEGACY_RTC; - for (i = 2; i < nrtimers; timer++, i++) - hd.hd_irq[i] = (timer->hpet_config & Tn_INT_ROUTE_CNF_MASK) >> - Tn_INT_ROUTE_CNF_SHIFT; + for (i = 2; i < nrtimers; timer++, i++) + hd.hd_irq[i] = (timer->hpet_config & Tn_INT_ROUTE_CNF_MASK) >> + Tn_INT_ROUTE_CNF_SHIFT; + hpet_alloc(&hd); + } #else static void hpet_reserve_platform_timers(unsigned long id) { } diff --git a/drivers/char/hpet.c b/drivers/char/hpet.c index 465ad35ed38f..1399971be689 100644 --- a/drivers/char/hpet.c +++ b/drivers/char/hpet.c @@ -731,14 +731,14 @@ static unsigned long hpet_calibrate(struct hpets *hpetp) int hpet_alloc(struct hpet_data *hdp) { - u64 cap, mcfg, hpet_config; + u64 cap, mcfg; struct hpet_dev *devp; - u32 i, ntimer, irq; + u32 i, ntimer; struct hpets *hpetp; size_t siz; struct hpet __iomem *hpet; static struct hpets *last = NULL; - unsigned long period, irq_bitmap; + unsigned long period; unsigned long long temp; /* @@ -765,47 +765,11 @@ int hpet_alloc(struct hpet_data *hdp) hpetp->hp_hpet_phys = hdp->hd_phys_address; hpetp->hp_ntimer = hdp->hd_nirqs; - hpet = hpetp->hp_hpet; - - /* Assign IRQs statically for legacy devices */ - hpetp->hp_dev[0].hd_hdwirq = hdp->hd_irq[0]; - hpetp->hp_dev[1].hd_hdwirq = hdp->hd_irq[1]; - - /* Assign IRQs dynamically for the others */ - for (i = 2, devp = &hpetp->hp_dev[2]; i < hdp->hd_nirqs; i++, devp++) { - struct hpet_timer __iomem *timer; - timer = &hpet->hpet_timers[devp - hpetp->hp_dev]; + for (i = 0; i < hdp->hd_nirqs; i++) + hpetp->hp_dev[i].hd_hdwirq = hdp->hd_irq[i]; - /* Check if there's already an IRQ assigned to the timer */ - if (hdp->hd_irq[i]) { - hpetp->hp_dev[i].hd_hdwirq = hdp->hd_irq[i]; - continue; - } - - hpet_config = readq(&timer->hpet_config); - irq_bitmap = (hpet_config & Tn_INT_ROUTE_CAP_MASK) - >> Tn_INT_ROUTE_CAP_SHIFT; - if (!irq_bitmap) - irq = 0; /* No valid IRQ Assignable */ - else { - irq = find_first_bit(&irq_bitmap, 32); - do { - hpet_config |= irq << Tn_INT_ROUTE_CNF_SHIFT; - writeq(hpet_config, &timer->hpet_config); - - /* - * Verify whether we have written a valid - * IRQ number by reading it back again - */ - hpet_config = readq(&timer->hpet_config); - if (irq == (hpet_config & Tn_INT_ROUTE_CNF_MASK) - >> Tn_INT_ROUTE_CNF_SHIFT) - break; /* Success */ - } while ((irq = (find_next_bit(&irq_bitmap, 32, irq)))); - } - hpetp->hp_dev[i].hd_hdwirq = irq; - } + hpet = hpetp->hp_hpet; cap = readq(&hpet->hpet_cap); @@ -836,8 +800,7 @@ int hpet_alloc(struct hpet_data *hdp) hpetp->hp_which, hdp->hd_phys_address, hpetp->hp_ntimer > 1 ? "s" : ""); for (i = 0; i < hpetp->hp_ntimer; i++) - printk("%s %d", i > 0 ? "," : "", - hpetp->hp_dev[i].hd_hdwirq); + printk("%s %d", i > 0 ? "," : "", hdp->hd_irq[i]); printk("\n"); printk(KERN_INFO "hpet%u: %u %d-bit timers, %Lu Hz\n", diff --git a/include/linux/hpet.h b/include/linux/hpet.h index 9cd94bfd07e5..2dc29ce6c8e4 100644 --- a/include/linux/hpet.h +++ b/include/linux/hpet.h @@ -64,7 +64,7 @@ struct hpet { */ #define Tn_INT_ROUTE_CAP_MASK (0xffffffff00000000ULL) -#define Tn_INT_ROUTE_CAP_SHIFT (32UL) +#define Tn_INI_ROUTE_CAP_SHIFT (32UL) #define Tn_FSB_INT_DELCAP_MASK (0x8000UL) #define Tn_FSB_INT_DELCAP_SHIFT (15) #define Tn_FSB_EN_CNF_MASK (0x4000UL) -- cgit v1.2.3 From 865ab8753837b73fe1e800fc88462c5914eec549 Mon Sep 17 00:00:00 2001 From: Sergei Shtylyov Date: Wed, 2 Apr 2008 23:53:19 +0400 Subject: [MIPS] Pb1200: Fix header breakage Signed-off-by: Sergei Shtylyov Signed-off-by: Ralf Baechle --- include/asm-mips/mach-pb1x00/pb1200.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/asm-mips/mach-pb1x00/pb1200.h b/include/asm-mips/mach-pb1x00/pb1200.h index ed5fd7390678..72213e3d02c7 100644 --- a/include/asm-mips/mach-pb1x00/pb1200.h +++ b/include/asm-mips/mach-pb1x00/pb1200.h @@ -245,7 +245,7 @@ enum external_pb1200_ints { PB1200_SD1_INSERT_INT, PB1200_SD1_EJECT_INT, - PB1200_INT_END (PB1200_INT_BEGIN + 15) + PB1200_INT_END = PB1200_INT_BEGIN + 15 }; /* For drivers/pcmcia/au1000_db1x00.c */ -- cgit v1.2.3 From 8bab8dded67d026c39367bbd5e27d2f6c556c38e Mon Sep 17 00:00:00 2001 From: Paul Menage Date: Fri, 4 Apr 2008 14:29:57 -0700 Subject: cgroups: add cgroup support for enabling controllers at boot time The effects of cgroup_disable=foo are: - foo isn't auto-mounted if you mount all cgroups in a single hierarchy - foo isn't visible as an individually mountable subsystem As a result there will only ever be one call to foo->create(), at init time; all processes will stay in this group, and the group will never be mounted on a visible hierarchy. Any additional effects (e.g. not allocating metadata) are up to the foo subsystem. This doesn't handle early_init subsystems (their "disabled" bit isn't set be, but it could easily be extended to do so if any of the early_init systems wanted it - I think it would just involve some nastier parameter processing since it would occur before the command-line argument parser had been run. Hugh said: Ballpark figures, I'm trying to get this question out rather than processing the exact numbers: CONFIG_CGROUP_MEM_RES_CTLR adds 15% overhead to the affected paths, booting with cgroup_disable=memory cuts that back to 1% overhead (due to slightly bigger struct page). I'm no expert on distros, they may have no interest whatever in CONFIG_CGROUP_MEM_RES_CTLR=y; and the rest of us can easily build with or without it, or apply the cgroup_disable=memory patches. Unix bench's execl test result on x86_64 was == just after boot without mounting any cgroup fs.== mem_cgorup=off : Execl Throughput 43.0 3150.1 732.6 mem_cgroup=on : Execl Throughput 43.0 2932.6 682.0 == [lizf@cn.fujitsu.com: fix boot option parsing] Signed-off-by: Balbir Singh Cc: Paul Menage Cc: Balbir Singh Cc: Pavel Emelyanov Cc: KAMEZAWA Hiroyuki Cc: Hugh Dickins Cc: Sudhir Kumar Cc: YAMAMOTO Takashi Cc: David Rientjes Signed-off-by: Li Zefan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/kernel-parameters.txt | 4 ++++ include/linux/cgroup.h | 1 + kernel/cgroup.c | 42 ++++++++++++++++++++++++++++++++----- 3 files changed, 42 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 4cd1a5da80a4..32e9297ef747 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -375,6 +375,10 @@ and is between 256 and 4096 characters. It is defined in the file ccw_timeout_log [S390] See Documentation/s390/CommonIO for details. + cgroup_disable= [KNL] Disable a particular controller + Format: {name of the controller(s) to disable} + {Currently supported controllers - "memory"} + checkreqprot [SELINUX] Set initial checkreqprot flag value. Format: { "0" | "1" } See security/selinux/Kconfig help text. diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index 028ba3b523b1..a6a6035a4e1e 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -256,6 +256,7 @@ struct cgroup_subsys { void (*bind)(struct cgroup_subsys *ss, struct cgroup *root); int subsys_id; int active; + int disabled; int early_init; #define MAX_CGROUP_TYPE_NAMELEN 32 const char *name; diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 53d86b4b0ce0..62f1a5231fe9 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -782,7 +782,14 @@ static int parse_cgroupfs_options(char *data, if (!*token) return -EINVAL; if (!strcmp(token, "all")) { - opts->subsys_bits = (1 << CGROUP_SUBSYS_COUNT) - 1; + /* Add all non-disabled subsystems */ + int i; + opts->subsys_bits = 0; + for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { + struct cgroup_subsys *ss = subsys[i]; + if (!ss->disabled) + opts->subsys_bits |= 1ul << i; + } } else if (!strcmp(token, "noprefix")) { set_bit(ROOT_NOPREFIX, &opts->flags); } else if (!strncmp(token, "release_agent=", 14)) { @@ -800,7 +807,8 @@ static int parse_cgroupfs_options(char *data, for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { ss = subsys[i]; if (!strcmp(token, ss->name)) { - set_bit(i, &opts->subsys_bits); + if (!ss->disabled) + set_bit(i, &opts->subsys_bits); break; } } @@ -2600,13 +2608,13 @@ static int proc_cgroupstats_show(struct seq_file *m, void *v) { int i; - seq_puts(m, "#subsys_name\thierarchy\tnum_cgroups\n"); + seq_puts(m, "#subsys_name\thierarchy\tnum_cgroups\tenabled\n"); mutex_lock(&cgroup_mutex); for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { struct cgroup_subsys *ss = subsys[i]; - seq_printf(m, "%s\t%lu\t%d\n", + seq_printf(m, "%s\t%lu\t%d\t%d\n", ss->name, ss->root->subsys_bits, - ss->root->number_of_cgroups); + ss->root->number_of_cgroups, !ss->disabled); } mutex_unlock(&cgroup_mutex); return 0; @@ -3010,3 +3018,27 @@ static void cgroup_release_agent(struct work_struct *work) spin_unlock(&release_list_lock); mutex_unlock(&cgroup_mutex); } + +static int __init cgroup_disable(char *str) +{ + int i; + char *token; + + while ((token = strsep(&str, ",")) != NULL) { + if (!*token) + continue; + + for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { + struct cgroup_subsys *ss = subsys[i]; + + if (!strcmp(token, ss->name)) { + ss->disabled = 1; + printk(KERN_INFO "Disabling %s control group" + " subsystem\n", ss->name); + break; + } + } + } + return 1; +} +__setup("cgroup_disable=", cgroup_disable); -- cgit v1.2.3 From 871de939030c903fd5ed50a7c4c88e02998e1cbc Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Mon, 7 Apr 2008 11:56:34 -0700 Subject: x86: fix 64-bit asm NOPS for CONFIG_GENERIC_CPU ASM_NOP's for 64-bit kernel with CONFIG_GENERIC_CPU is broken with the recent x86 nops merge. They were using GENERIC_NOPS which will truncate the upper 32bits of %rsi, because of the missing 64bit rex prefix. For now, fall back ASM NOPS for generic cpu to K8 NOPS, similar to the code before the wrong x86 nop merge. This should resolve the crash seen by Ingo on a test-system: BUG: unable to handle kernel paging request at 00000000d80d8ee8 IP: [] save_i387_ia32+0x61/0xd8 PGD b8e0067 PUD 51490067 PMD 0 Oops: 0000 [1] SMP CPU 2 Modules linked in: Pid: 3871, comm: distcc Not tainted 2.6.25-rc7-sched-devel.git-x86-latest.git #359 RIP: 0010:[] [] save_i387_ia32+0x61/0xd8 RSP: 0000:ffff81003abd3cb8 EFLAGS: 00010246 RAX: ffff810082e93400 RBX: 00000000ffc37f84 RCX: ffff8100d80d8ee0 RDX: 0000000000000000 RSI: 00000000d80d8ee0 RDI: ffff810082e93400 RBP: 00000000ffc37fdc R08: 00000000ffc37f88 R09: 0000000000000008 R10: ffff81003abd2000 R11: 0000000000000000 R12: ffff810082e93400 R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000 FS: 0000000000000000(0000) GS:ffff81011fb12dc0(0063) knlGS:00000000f7f1a6c0 CS: 0010 DS: 002b ES: 002b CR0: 0000000080050033 CR2: 00000000d80d8ee8 CR3: 0000000076922000 CR4: 00000000000006e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Process distcc (pid: 3871, threadinfo ffff81003abd2000, task ffff8100d80d8ee0) Stack: ffff8100bb670380 ffffffff8026de50 0000000000000118 0000000000000002 0000000000000002 ffff81003abd3e68 ffff81003abd3ed8 ffff81003abd3de8 ffff81003abd3d18 ffffffff80229785 ffff8100d80d8ee0 ffff810001041280 Call Trace: [] ? __generic_file_aio_write_nolock+0x343/0x377 [] ? update_curr+0x54/0x64 [] ? ia32_setup_sigcontext+0x125/0x1d2 [] ? ia32_setup_frame+0x73/0x1a5 [] ? do_notify_resume+0x1aa/0x7db [] ? getnstimeofday+0x31/0x85 [] ? ktime_get_ts+0x17/0x48 [] ? ktime_get+0xc/0x41 [] ? hrtimer_nanosleep+0x75/0xd5 [] ? hrtimer_wakeup+0x0/0x21 [] ? int_signal+0x12/0x17 [] ? dummy_file_free_security+0x0/0x1 Code: a6 08 05 00 00 f6 40 14 01 74 34 4c 89 e7 48 0f ae 07 48 8b 86 08 05 00 00 80 78 02 00 79 02 db e2 90 8d b4 26 00 00 00 00 89 f6 <48> 8b 46 08 83 60 14 fe 0f 20 c0 48 83 c8 08 0f 22 c0 eb 07 c6 Signed-off-by: Suresh Siddha Signed-off-by: Ingo Molnar --- include/asm-x86/nops.h | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/asm-x86/nops.h b/include/asm-x86/nops.h index e3b2bce0aff8..b3930ae539b3 100644 --- a/include/asm-x86/nops.h +++ b/include/asm-x86/nops.h @@ -73,16 +73,7 @@ #define P6_NOP7 ".byte 0x0f,0x1f,0x80,0,0,0,0\n" #define P6_NOP8 ".byte 0x0f,0x1f,0x84,0x00,0,0,0,0\n" -#if defined(CONFIG_MK8) -#define ASM_NOP1 K8_NOP1 -#define ASM_NOP2 K8_NOP2 -#define ASM_NOP3 K8_NOP3 -#define ASM_NOP4 K8_NOP4 -#define ASM_NOP5 K8_NOP5 -#define ASM_NOP6 K8_NOP6 -#define ASM_NOP7 K8_NOP7 -#define ASM_NOP8 K8_NOP8 -#elif defined(CONFIG_MK7) +#if defined(CONFIG_MK7) #define ASM_NOP1 K7_NOP1 #define ASM_NOP2 K7_NOP2 #define ASM_NOP3 K7_NOP3 @@ -100,6 +91,15 @@ #define ASM_NOP6 P6_NOP6 #define ASM_NOP7 P6_NOP7 #define ASM_NOP8 P6_NOP8 +#elif defined(CONFIG_X86_64) +#define ASM_NOP1 K8_NOP1 +#define ASM_NOP2 K8_NOP2 +#define ASM_NOP3 K8_NOP3 +#define ASM_NOP4 K8_NOP4 +#define ASM_NOP5 K8_NOP5 +#define ASM_NOP6 K8_NOP6 +#define ASM_NOP7 K8_NOP7 +#define ASM_NOP8 K8_NOP8 #else #define ASM_NOP1 GENERIC_NOP1 #define ASM_NOP2 GENERIC_NOP2 -- cgit v1.2.3 From 2557a933b795c1988c721ebb871cd735128bb9cb Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Mon, 7 Apr 2008 14:30:28 +1000 Subject: virtio: remove overzealous BUG_ON. The 'disable_cb' callback is designed as an optimization to tell the host we don't need callbacks now. As it is not reliable, the debug check is overzealous: it can happen on two CPUs at the same time. Document this. Even if it were reliable, the virtio_net driver doesn't disable callbacks on transmit so the START_USE/END_USE debugging reentrance protection can be easily tripped even on UP. Thanks to Balaji Rao for the bug report and testing. Signed-off-by: Rusty Russell CC: Balaji Rao Signed-off-by: Linus Torvalds --- drivers/virtio/virtio_ring.c | 3 --- include/linux/virtio.h | 5 ++++- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c index aa714028641e..c2fa5c630813 100644 --- a/drivers/virtio/virtio_ring.c +++ b/drivers/virtio/virtio_ring.c @@ -214,10 +214,7 @@ static void vring_disable_cb(struct virtqueue *_vq) { struct vring_virtqueue *vq = to_vvq(_vq); - START_USE(vq); - BUG_ON(vq->vring.avail->flags & VRING_AVAIL_F_NO_INTERRUPT); vq->vring.avail->flags |= VRING_AVAIL_F_NO_INTERRUPT; - END_USE(vq); } static bool vring_enable_cb(struct virtqueue *_vq) diff --git a/include/linux/virtio.h b/include/linux/virtio.h index 12c18ac1b973..e7d10845b3c1 100644 --- a/include/linux/virtio.h +++ b/include/linux/virtio.h @@ -41,6 +41,8 @@ struct virtqueue * Returns NULL or the "data" token handed to add_buf. * @disable_cb: disable callbacks * vq: the struct virtqueue we're talking about. + * Note that this is not necessarily synchronous, hence unreliable and only + * useful as an optimization. * @enable_cb: restart callbacks after disable_cb. * vq: the struct virtqueue we're talking about. * This re-enables callbacks; it returns "false" if there are pending @@ -48,7 +50,8 @@ struct virtqueue * checking for more work, and enabling callbacks. * * Locking rules are straightforward: the driver is responsible for - * locking. No two operations may be invoked simultaneously. + * locking. No two operations may be invoked simultaneously, with the exception + * of @disable_cb. * * All operations can be called in any context. */ -- cgit v1.2.3 From 9c5a3d729cf430609d091ff610a7db363aafcd47 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Sat, 5 Apr 2008 15:13:23 +0100 Subject: [MIPS] Handle aliases in vmalloc correctly. flush_cache_vmap / flush_cache_vunmap were calling flush_cache_all which - having been deprecated - turned into a nop ... Signed-off-by: Ralf Baechle --- arch/mips/mm/c-r4k.c | 14 ++++++++++++++ arch/mips/mm/c-tx39.c | 15 +++++++++++++++ arch/mips/mm/cache.c | 3 +++ include/asm-mips/cacheflush.h | 18 ++++++++++++++++-- 4 files changed, 48 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/arch/mips/mm/c-r4k.c b/arch/mips/mm/c-r4k.c index 6496925b5e29..77aefb4ebedd 100644 --- a/arch/mips/mm/c-r4k.c +++ b/arch/mips/mm/c-r4k.c @@ -361,6 +361,16 @@ static inline int has_valid_asid(const struct mm_struct *mm) #endif } +static void r4k__flush_cache_vmap(void) +{ + r4k_blast_dcache(); +} + +static void r4k__flush_cache_vunmap(void) +{ + r4k_blast_dcache(); +} + static inline void local_r4k_flush_cache_range(void * args) { struct vm_area_struct *vma = args; @@ -1281,6 +1291,10 @@ void __cpuinit r4k_cache_init(void) PAGE_SIZE - 1); else shm_align_mask = PAGE_SIZE-1; + + __flush_cache_vmap = r4k__flush_cache_vmap; + __flush_cache_vunmap = r4k__flush_cache_vunmap; + flush_cache_all = cache_noop; __flush_cache_all = r4k___flush_cache_all; flush_cache_mm = r4k_flush_cache_mm; diff --git a/arch/mips/mm/c-tx39.c b/arch/mips/mm/c-tx39.c index b09d56981d53..a9f7f1f5e9b4 100644 --- a/arch/mips/mm/c-tx39.c +++ b/arch/mips/mm/c-tx39.c @@ -122,6 +122,16 @@ static inline void tx39_blast_icache(void) local_irq_restore(flags); } +static void tx39__flush_cache_vmap(void) +{ + tx39_blast_dcache(); +} + +static void tx39__flush_cache_vunmap(void) +{ + tx39_blast_dcache(); +} + static inline void tx39_flush_cache_all(void) { if (!cpu_has_dc_aliases) @@ -344,6 +354,8 @@ void __cpuinit tx39_cache_init(void) switch (current_cpu_type()) { case CPU_TX3912: /* TX39/H core (writethru direct-map cache) */ + __flush_cache_vmap = tx39__flush_cache_vmap; + __flush_cache_vunmap = tx39__flush_cache_vunmap; flush_cache_all = tx39h_flush_icache_all; __flush_cache_all = tx39h_flush_icache_all; flush_cache_mm = (void *) tx39h_flush_icache_all; @@ -369,6 +381,9 @@ void __cpuinit tx39_cache_init(void) write_c0_wired(0); /* set 8 on reset... */ /* board-dependent init code may set WBON */ + __flush_cache_vmap = tx39__flush_cache_vmap; + __flush_cache_vunmap = tx39__flush_cache_vunmap; + flush_cache_all = tx39_flush_cache_all; __flush_cache_all = tx39___flush_cache_all; flush_cache_mm = tx39_flush_cache_mm; diff --git a/arch/mips/mm/cache.c b/arch/mips/mm/cache.c index 51ab1faa027d..f5903679ee6a 100644 --- a/arch/mips/mm/cache.c +++ b/arch/mips/mm/cache.c @@ -30,6 +30,9 @@ void (*flush_cache_page)(struct vm_area_struct *vma, unsigned long page, unsigned long pfn); void (*flush_icache_range)(unsigned long start, unsigned long end); +void (*__flush_cache_vmap)(void); +void (*__flush_cache_vunmap)(void); + /* MIPS specific cache operations */ void (*flush_cache_sigtramp)(unsigned long addr); void (*local_flush_data_cache_page)(void * addr); diff --git a/include/asm-mips/cacheflush.h b/include/asm-mips/cacheflush.h index 01e7eadc97e2..d5c0f2fda51b 100644 --- a/include/asm-mips/cacheflush.h +++ b/include/asm-mips/cacheflush.h @@ -63,8 +63,22 @@ static inline void flush_icache_page(struct vm_area_struct *vma, } extern void (*flush_icache_range)(unsigned long start, unsigned long end); -#define flush_cache_vmap(start, end) flush_cache_all() -#define flush_cache_vunmap(start, end) flush_cache_all() + +extern void (*__flush_cache_vmap)(void); + +static inline void flush_cache_vmap(unsigned long start, unsigned long end) +{ + if (cpu_has_dc_aliases) + __flush_cache_vmap(); +} + +extern void (*__flush_cache_vunmap)(void); + +static inline void flush_cache_vunmap(unsigned long start, unsigned long end) +{ + if (cpu_has_dc_aliases) + __flush_cache_vunmap(); +} extern void copy_to_user_page(struct vm_area_struct *vma, struct page *page, unsigned long vaddr, void *dst, const void *src, -- cgit v1.2.3 From 882bebaaca4bb1484078d44ef011f918c0e1e14e Mon Sep 17 00:00:00 2001 From: Ilpo Järvinen Date: Mon, 7 Apr 2008 22:33:07 -0700 Subject: [TCP]: tcp_simple_retransmit can cause S+L MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This fixes Bugzilla #10384 tcp_simple_retransmit does L increment without any checking whatsoever for overflowing S+L when Reno is in use. The simplest scenario I can currently think of is rather complex in practice (there might be some more straightforward cases though). Ie., if mss is reduced during mtu probing, it may end up marking everything lost and if some duplicate ACKs arrived prior to that sacked_out will be non-zero as well, leading to S+L > packets_out, tcp_clean_rtx_queue on the next cumulative ACK or tcp_fastretrans_alert on the next duplicate ACK will fix the S counter. More straightforward (but questionable) solution would be to just call tcp_reset_reno_sack() in tcp_simple_retransmit but it would negatively impact the probe's retransmission, ie., the retransmissions would not occur if some duplicate ACKs had arrived. So I had to add reno sacked_out reseting to CA_Loss state when the first cumulative ACK arrives (this stale sacked_out might actually be the explanation for the reports of left_out overflows in kernel prior to 2.6.23 and S+L overflow reports of 2.6.24). However, this alone won't be enough to fix kernel before 2.6.24 because it is building on top of the commit 1b6d427bb7e ([TCP]: Reduce sacked_out with reno when purging write_queue) to keep the sacked_out from overflowing. Signed-off-by: Ilpo Järvinen Reported-by: Alessandro Suardi Signed-off-by: David S. Miller --- include/net/tcp.h | 2 ++ net/ipv4/tcp_input.c | 24 ++++++++++++++++++------ net/ipv4/tcp_output.c | 3 +++ 3 files changed, 23 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/net/tcp.h b/include/net/tcp.h index 7de4ea3a04d9..4fd3eb2f8ec2 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -752,6 +752,8 @@ static inline unsigned int tcp_packets_in_flight(const struct tcp_sock *tp) return tp->packets_out - tcp_left_out(tp) + tp->retrans_out; } +extern int tcp_limit_reno_sacked(struct tcp_sock *tp); + /* If cwnd > ssthresh, we may raise ssthresh to be half-way to cwnd. * The exception is rate halving phase, when cwnd is decreasing towards * ssthresh. diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 7d0958785bfb..b4812c3cbbcf 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -1625,13 +1625,11 @@ out: return flag; } -/* If we receive more dupacks than we expected counting segments - * in assumption of absent reordering, interpret this as reordering. - * The only another reason could be bug in receiver TCP. +/* Limits sacked_out so that sum with lost_out isn't ever larger than + * packets_out. Returns zero if sacked_out adjustement wasn't necessary. */ -static void tcp_check_reno_reordering(struct sock *sk, const int addend) +int tcp_limit_reno_sacked(struct tcp_sock *tp) { - struct tcp_sock *tp = tcp_sk(sk); u32 holes; holes = max(tp->lost_out, 1U); @@ -1639,8 +1637,20 @@ static void tcp_check_reno_reordering(struct sock *sk, const int addend) if ((tp->sacked_out + holes) > tp->packets_out) { tp->sacked_out = tp->packets_out - holes; - tcp_update_reordering(sk, tp->packets_out + addend, 0); + return 1; } + return 0; +} + +/* If we receive more dupacks than we expected counting segments + * in assumption of absent reordering, interpret this as reordering. + * The only another reason could be bug in receiver TCP. + */ +static void tcp_check_reno_reordering(struct sock *sk, const int addend) +{ + struct tcp_sock *tp = tcp_sk(sk); + if (tcp_limit_reno_sacked(tp)) + tcp_update_reordering(sk, tp->packets_out + addend, 0); } /* Emulate SACKs for SACKless connection: account for a new dupack. */ @@ -2600,6 +2610,8 @@ static void tcp_fastretrans_alert(struct sock *sk, int pkts_acked, int flag) case TCP_CA_Loss: if (flag & FLAG_DATA_ACKED) icsk->icsk_retransmits = 0; + if (tcp_is_reno(tp) && flag & FLAG_SND_UNA_ADVANCED) + tcp_reset_reno_sack(tp); if (!tcp_try_undo_loss(sk)) { tcp_moderate_cwnd(tp); tcp_xmit_retransmit_queue(sk); diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 72b9350006fe..d29ef79c00ca 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -1808,6 +1808,9 @@ void tcp_simple_retransmit(struct sock *sk) if (!lost) return; + if (tcp_is_reno(tp)) + tcp_limit_reno_sacked(tp); + tcp_verify_left_out(tp); /* Don't muck with the congestion window here. -- cgit v1.2.3 From a31b9dd8edc6e4e75b5299fee6093b3c54548446 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 10 Apr 2008 16:10:45 +0100 Subject: FRV: Handle update_mmu_cache() being called when current->mm is NULL [try #2] Handle update_mmu_cache() being called when current->mm is NULL. We cache static TLB mappings for the current page table in DAMPR4 and DAMPR5 on the theory that the next data lookup is likely to be in the same general region, and thus is likely to be mapped by the same page table. However, we can't get this information if we can't access the appropriate mm_struct. If current->mm is NULL, we just clear the cache in the knowledge that the TLB miss handlers will load it. Signed-off-by: David Howells Signed-off-by: Linus Torvalds --- include/asm-frv/pgtable.h | 19 ++++++++++++++----- 1 file changed, 14 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/asm-frv/pgtable.h b/include/asm-frv/pgtable.h index 6c0682ed5fc9..4e219046fe42 100644 --- a/include/asm-frv/pgtable.h +++ b/include/asm-frv/pgtable.h @@ -507,13 +507,22 @@ static inline int pte_file(pte_t pte) */ static inline void update_mmu_cache(struct vm_area_struct *vma, unsigned long address, pte_t pte) { + struct mm_struct *mm; unsigned long ampr; - pgd_t *pge = pgd_offset(current->mm, address); - pud_t *pue = pud_offset(pge, address); - pmd_t *pme = pmd_offset(pue, address); - ampr = pme->ste[0] & 0xffffff00; - ampr |= xAMPRx_L | xAMPRx_SS_16Kb | xAMPRx_S | xAMPRx_C | xAMPRx_V; + mm = current->mm; + if (mm) { + pgd_t *pge = pgd_offset(mm, address); + pud_t *pue = pud_offset(pge, address); + pmd_t *pme = pmd_offset(pue, address); + + ampr = pme->ste[0] & 0xffffff00; + ampr |= xAMPRx_L | xAMPRx_SS_16Kb | xAMPRx_S | xAMPRx_C | + xAMPRx_V; + } else { + address = ULONG_MAX; + ampr = 0; + } asm volatile("movgs %0,scr0\n" "movgs %0,scr1\n" -- cgit v1.2.3 From 0c93d8e4d342b1b5cda1037f2527fcf443c80fbc Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 10 Apr 2008 16:10:50 +0100 Subject: FRV: Move STACK_TOP_MAX up [try #2] Move STACK_TOP_MAX up so that we don't try moving the stack above it as that causes setup_arg_pages() to malfunction. Signed-off-by: David Howells Signed-off-by: Linus Torvalds --- include/asm-frv/mem-layout.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/asm-frv/mem-layout.h b/include/asm-frv/mem-layout.h index 83532252b8be..734a1d0583b6 100644 --- a/include/asm-frv/mem-layout.h +++ b/include/asm-frv/mem-layout.h @@ -60,7 +60,7 @@ */ #define BRK_BASE __UL(2 * 1024 * 1024 + PAGE_SIZE) #define STACK_TOP __UL(2 * 1024 * 1024) -#define STACK_TOP_MAX STACK_TOP +#define STACK_TOP_MAX __UL(0xc0000000) /* userspace process size */ #ifdef CONFIG_MMU -- cgit v1.2.3 From e31c243f984628d02f045dc4b622f1e2827860dc Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 10 Apr 2008 16:10:55 +0100 Subject: FRV: Add support for emulation of userspace atomic ops [try #2] Use traps 120-126 to emulate atomic cmpxchg32, xchg32, and XOR-, OR-, AND-, SUB- and ADD-to-memory operations for userspace. Signed-off-by: David Howells Signed-off-by: Linus Torvalds --- arch/frv/kernel/entry-table.S | 8 +- arch/frv/kernel/entry.S | 20 ++++ arch/frv/kernel/traps.c | 227 ++++++++++++++++++++++++++++++++++++++++++ include/asm-frv/spr-regs.h | 14 +++ 4 files changed, 268 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/arch/frv/kernel/entry-table.S b/arch/frv/kernel/entry-table.S index d3b9253d862a..bf35f33e48c9 100644 --- a/arch/frv/kernel/entry-table.S +++ b/arch/frv/kernel/entry-table.S @@ -316,8 +316,14 @@ __trap_fixup_kernel_data_tlb_miss: .section .trap.vector .org TBR_TT_TRAP0 >> 2 .long system_call - .rept 126 + .rept 119 .long __entry_unsupported_trap .endr + + # userspace atomic op emulation, traps 120-126 + .rept 7 + .long __entry_atomic_op + .endr + .org TBR_TT_BREAK >> 2 .long __entry_debug_exception diff --git a/arch/frv/kernel/entry.S b/arch/frv/kernel/entry.S index f36d7f4a7c25..b8a4b94779b1 100644 --- a/arch/frv/kernel/entry.S +++ b/arch/frv/kernel/entry.S @@ -654,6 +654,26 @@ __entry_debug_exception: movgs gr4,psr jmpl @(gr5,gr0) ; call ill_insn(esfr1,epcr0,esr0) +############################################################################### +# +# handle atomic operation emulation for userspace +# +############################################################################### + .globl __entry_atomic_op +__entry_atomic_op: + LEDS 0x6012 + sethi.p %hi(atomic_operation),gr5 + setlo %lo(atomic_operation),gr5 + movsg esfr1,gr8 + movsg epcr0,gr9 + movsg esr0,gr10 + + # now that we've accessed the exception regs, we can enable exceptions + movsg psr,gr4 + ori gr4,#PSR_ET,gr4 + movgs gr4,psr + jmpl @(gr5,gr0) ; call atomic_operation(esfr1,epcr0,esr0) + ############################################################################### # # handle media exception diff --git a/arch/frv/kernel/traps.c b/arch/frv/kernel/traps.c index 2e6098c85578..2f7e66877f3b 100644 --- a/arch/frv/kernel/traps.c +++ b/arch/frv/kernel/traps.c @@ -100,6 +100,233 @@ asmlinkage void illegal_instruction(unsigned long esfr1, unsigned long epcr0, un force_sig_info(info.si_signo, &info, current); } /* end illegal_instruction() */ +/*****************************************************************************/ +/* + * handle atomic operations with errors + * - arguments in gr8, gr9, gr10 + * - original memory value placed in gr5 + * - replacement memory value placed in gr9 + */ +asmlinkage void atomic_operation(unsigned long esfr1, unsigned long epcr0, + unsigned long esr0) +{ + static DEFINE_SPINLOCK(atomic_op_lock); + unsigned long x, y, z, *p; + mm_segment_t oldfs; + siginfo_t info; + int ret; + + y = 0; + z = 0; + + oldfs = get_fs(); + if (!user_mode(__frame)) + set_fs(KERNEL_DS); + + switch (__frame->tbr & TBR_TT) { + /* TIRA gr0,#120 + * u32 __atomic_user_cmpxchg32(u32 *ptr, u32 test, u32 new) + */ + case TBR_TT_ATOMIC_CMPXCHG32: + p = (unsigned long *) __frame->gr8; + x = __frame->gr9; + y = __frame->gr10; + + for (;;) { + ret = get_user(z, p); + if (ret < 0) + goto error; + + if (z != x) + goto done; + + spin_lock_irq(&atomic_op_lock); + + if (__get_user(z, p) == 0) { + if (z != x) + goto done2; + + if (__put_user(y, p) == 0) + goto done2; + goto error2; + } + + spin_unlock_irq(&atomic_op_lock); + } + + /* TIRA gr0,#121 + * u32 __atomic_kernel_xchg32(void *v, u32 new) + */ + case TBR_TT_ATOMIC_XCHG32: + p = (unsigned long *) __frame->gr8; + y = __frame->gr9; + + for (;;) { + ret = get_user(z, p); + if (ret < 0) + goto error; + + spin_lock_irq(&atomic_op_lock); + + if (__get_user(z, p) == 0) { + if (__put_user(y, p) == 0) + goto done2; + goto error2; + } + + spin_unlock_irq(&atomic_op_lock); + } + + /* TIRA gr0,#122 + * ulong __atomic_kernel_XOR_return(ulong i, ulong *v) + */ + case TBR_TT_ATOMIC_XOR: + p = (unsigned long *) __frame->gr8; + x = __frame->gr9; + + for (;;) { + ret = get_user(z, p); + if (ret < 0) + goto error; + + spin_lock_irq(&atomic_op_lock); + + if (__get_user(z, p) == 0) { + y = x ^ z; + if (__put_user(y, p) == 0) + goto done2; + goto error2; + } + + spin_unlock_irq(&atomic_op_lock); + } + + /* TIRA gr0,#123 + * ulong __atomic_kernel_OR_return(ulong i, ulong *v) + */ + case TBR_TT_ATOMIC_OR: + p = (unsigned long *) __frame->gr8; + x = __frame->gr9; + + for (;;) { + ret = get_user(z, p); + if (ret < 0) + goto error; + + spin_lock_irq(&atomic_op_lock); + + if (__get_user(z, p) == 0) { + y = x ^ z; + if (__put_user(y, p) == 0) + goto done2; + goto error2; + } + + spin_unlock_irq(&atomic_op_lock); + } + + /* TIRA gr0,#124 + * ulong __atomic_kernel_AND_return(ulong i, ulong *v) + */ + case TBR_TT_ATOMIC_AND: + p = (unsigned long *) __frame->gr8; + x = __frame->gr9; + + for (;;) { + ret = get_user(z, p); + if (ret < 0) + goto error; + + spin_lock_irq(&atomic_op_lock); + + if (__get_user(z, p) == 0) { + y = x & z; + if (__put_user(y, p) == 0) + goto done2; + goto error2; + } + + spin_unlock_irq(&atomic_op_lock); + } + + /* TIRA gr0,#125 + * int __atomic_user_sub_return(atomic_t *v, int i) + */ + case TBR_TT_ATOMIC_SUB: + p = (unsigned long *) __frame->gr8; + x = __frame->gr9; + + for (;;) { + ret = get_user(z, p); + if (ret < 0) + goto error; + + spin_lock_irq(&atomic_op_lock); + + if (__get_user(z, p) == 0) { + y = z - x; + if (__put_user(y, p) == 0) + goto done2; + goto error2; + } + + spin_unlock_irq(&atomic_op_lock); + } + + /* TIRA gr0,#126 + * int __atomic_user_add_return(atomic_t *v, int i) + */ + case TBR_TT_ATOMIC_ADD: + p = (unsigned long *) __frame->gr8; + x = __frame->gr9; + + for (;;) { + ret = get_user(z, p); + if (ret < 0) + goto error; + + spin_lock_irq(&atomic_op_lock); + + if (__get_user(z, p) == 0) { + y = z + x; + if (__put_user(y, p) == 0) + goto done2; + goto error2; + } + + spin_unlock_irq(&atomic_op_lock); + } + + default: + BUG(); + } + +done2: + spin_unlock_irq(&atomic_op_lock); +done: + if (!user_mode(__frame)) + set_fs(oldfs); + __frame->gr5 = z; + __frame->gr9 = y; + return; + +error2: + spin_unlock_irq(&atomic_op_lock); +error: + if (!user_mode(__frame)) + set_fs(oldfs); + __frame->pc -= 4; + + die_if_kernel("-- Atomic Op Error --\n"); + + info.si_signo = SIGSEGV; + info.si_code = SEGV_ACCERR; + info.si_errno = 0; + info.si_addr = (void *) __frame->pc; + + force_sig_info(info.si_signo, &info, current); +} + /*****************************************************************************/ /* * diff --git a/include/asm-frv/spr-regs.h b/include/asm-frv/spr-regs.h index c2a541ef828d..01e6af5e99b8 100644 --- a/include/asm-frv/spr-regs.h +++ b/include/asm-frv/spr-regs.h @@ -99,9 +99,23 @@ #define TBR_TT_TRAP1 (0x81 << 4) #define TBR_TT_TRAP2 (0x82 << 4) #define TBR_TT_TRAP3 (0x83 << 4) +#define TBR_TT_TRAP120 (0xf8 << 4) +#define TBR_TT_TRAP121 (0xf9 << 4) +#define TBR_TT_TRAP122 (0xfa << 4) +#define TBR_TT_TRAP123 (0xfb << 4) +#define TBR_TT_TRAP124 (0xfc << 4) +#define TBR_TT_TRAP125 (0xfd << 4) #define TBR_TT_TRAP126 (0xfe << 4) #define TBR_TT_BREAK (0xff << 4) +#define TBR_TT_ATOMIC_CMPXCHG32 TBR_TT_TRAP120 +#define TBR_TT_ATOMIC_XCHG32 TBR_TT_TRAP121 +#define TBR_TT_ATOMIC_XOR TBR_TT_TRAP122 +#define TBR_TT_ATOMIC_OR TBR_TT_TRAP123 +#define TBR_TT_ATOMIC_AND TBR_TT_TRAP124 +#define TBR_TT_ATOMIC_SUB TBR_TT_TRAP125 +#define TBR_TT_ATOMIC_ADD TBR_TT_TRAP126 + #define __get_TBR() ({ unsigned long x; asm volatile("movsg tbr,%0" : "=r"(x)); x; }) /* -- cgit v1.2.3 From f17520e1f19172057328e50ffed01a42534921e9 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 10 Apr 2008 16:11:05 +0100 Subject: FRV: Don't make smp_{r, w, }mb() interpolate MEMBAR when CONFIG_SMP=n [try #2] Don't make smp_{r,w,}mb() interpolate a MEMBAR instruction when CONFIG_SMP=n as SMP memory barries on UP systems should interpolate a compiler barrier only. Signed-off-by: David Howells Signed-off-by: Linus Torvalds --- include/asm-frv/system.h | 21 +++++++++++++++------ 1 file changed, 15 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/asm-frv/system.h b/include/asm-frv/system.h index 2c57f4734746..30a67a9da11a 100644 --- a/include/asm-frv/system.h +++ b/include/asm-frv/system.h @@ -179,14 +179,23 @@ do { \ #define mb() asm volatile ("membar" : : :"memory") #define rmb() asm volatile ("membar" : : :"memory") #define wmb() asm volatile ("membar" : : :"memory") -#define set_mb(var, value) do { var = value; mb(); } while (0) +#define read_barrier_depends() barrier() -#define smp_mb() mb() -#define smp_rmb() rmb() -#define smp_wmb() wmb() - -#define read_barrier_depends() do {} while(0) +#ifdef CONFIG_SMP +#define smp_mb() mb() +#define smp_rmb() rmb() +#define smp_wmb() wmb() #define smp_read_barrier_depends() read_barrier_depends() +#define set_mb(var, value) \ + do { xchg(&var, (value)); } while (0) +#else +#define smp_mb() barrier() +#define smp_rmb() barrier() +#define smp_wmb() barrier() +#define smp_read_barrier_depends() do {} while(0) +#define set_mb(var, value) \ + do { var = (value); barrier(); } while (0) +#endif #define HARD_RESET_NOW() \ do { \ -- cgit v1.2.3 From 54a015104136974262afa4b8ddd943ea70dec8a2 Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Thu, 10 Apr 2008 15:37:38 -0700 Subject: asmlinkage_protect replaces prevent_tail_call The prevent_tail_call() macro works around the problem of the compiler clobbering argument words on the stack, which for asmlinkage functions is the caller's (user's) struct pt_regs. The tail/sibling-call optimization is not the only way that the compiler can decide to use stack argument words as scratch space, which we have to prevent. Other optimizations can do it too. Until we have new compiler support to make "asmlinkage" binding on the compiler's own use of the stack argument frame, we have work around all the manifestations of this issue that crop up. More cases seem to be prevented by also keeping the incoming argument variables live at the end of the function. This makes their original stack slots attractive places to leave those variables, so the compiler tends not clobber them for something else. It's still no guarantee, but it handles some observed cases that prevent_tail_call() did not. Signed-off-by: Roland McGrath Signed-off-by: Linus Torvalds --- arch/x86/kernel/tls.c | 4 ++-- fs/open.c | 8 ++++---- include/asm-x86/linkage.h | 24 +++++++++++++++++++++++- include/linux/linkage.h | 4 ++-- kernel/exit.c | 4 ++-- kernel/uid16.c | 22 +++++++++++----------- 6 files changed, 44 insertions(+), 22 deletions(-) (limited to 'include') diff --git a/arch/x86/kernel/tls.c b/arch/x86/kernel/tls.c index 022bcaa3b42e..ab6bf375a307 100644 --- a/arch/x86/kernel/tls.c +++ b/arch/x86/kernel/tls.c @@ -92,7 +92,7 @@ int do_set_thread_area(struct task_struct *p, int idx, asmlinkage int sys_set_thread_area(struct user_desc __user *u_info) { int ret = do_set_thread_area(current, -1, u_info, 1); - prevent_tail_call(ret); + asmlinkage_protect(1, ret, u_info); return ret; } @@ -142,7 +142,7 @@ int do_get_thread_area(struct task_struct *p, int idx, asmlinkage int sys_get_thread_area(struct user_desc __user *u_info) { int ret = do_get_thread_area(current, -1, u_info); - prevent_tail_call(ret); + asmlinkage_protect(1, ret, u_info); return ret; } diff --git a/fs/open.c b/fs/open.c index a4b12022edaa..3fa4e4ffce4c 100644 --- a/fs/open.c +++ b/fs/open.c @@ -335,7 +335,7 @@ asmlinkage long sys_ftruncate(unsigned int fd, unsigned long length) { long ret = do_sys_ftruncate(fd, length, 1); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(2, ret, fd, length); return ret; } @@ -350,7 +350,7 @@ asmlinkage long sys_ftruncate64(unsigned int fd, loff_t length) { long ret = do_sys_ftruncate(fd, length, 0); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(2, ret, fd, length); return ret; } #endif @@ -1067,7 +1067,7 @@ asmlinkage long sys_open(const char __user *filename, int flags, int mode) ret = do_sys_open(AT_FDCWD, filename, flags, mode); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(3, ret, filename, flags, mode); return ret; } @@ -1081,7 +1081,7 @@ asmlinkage long sys_openat(int dfd, const char __user *filename, int flags, ret = do_sys_open(dfd, filename, flags, mode); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(4, ret, dfd, filename, flags, mode); return ret; } diff --git a/include/asm-x86/linkage.h b/include/asm-x86/linkage.h index 31739c7d66a9..d605eeba0f70 100644 --- a/include/asm-x86/linkage.h +++ b/include/asm-x86/linkage.h @@ -8,12 +8,34 @@ #ifdef CONFIG_X86_32 #define asmlinkage CPP_ASMLINKAGE __attribute__((regparm(0))) -#define prevent_tail_call(ret) __asm__ ("" : "=r" (ret) : "0" (ret)) /* * For 32-bit UML - mark functions implemented in assembly that use * regparm input parameters: */ #define asmregparm __attribute__((regparm(3))) + +#define asmlinkage_protect(n, ret, args...) \ + __asmlinkage_protect##n(ret, ##args) +#define __asmlinkage_protect_n(ret, args...) \ + __asm__ __volatile__ ("" : "=r" (ret) : "0" (ret), ##args) +#define __asmlinkage_protect0(ret) \ + __asmlinkage_protect_n(ret) +#define __asmlinkage_protect1(ret, arg1) \ + __asmlinkage_protect_n(ret, "g" (arg1)) +#define __asmlinkage_protect2(ret, arg1, arg2) \ + __asmlinkage_protect_n(ret, "g" (arg1), "g" (arg2)) +#define __asmlinkage_protect3(ret, arg1, arg2, arg3) \ + __asmlinkage_protect_n(ret, "g" (arg1), "g" (arg2), "g" (arg3)) +#define __asmlinkage_protect4(ret, arg1, arg2, arg3, arg4) \ + __asmlinkage_protect_n(ret, "g" (arg1), "g" (arg2), "g" (arg3), \ + "g" (arg4)) +#define __asmlinkage_protect5(ret, arg1, arg2, arg3, arg4, arg5) \ + __asmlinkage_protect_n(ret, "g" (arg1), "g" (arg2), "g" (arg3), \ + "g" (arg4), "g" (arg5)) +#define __asmlinkage_protect6(ret, arg1, arg2, arg3, arg4, arg5, arg6) \ + __asmlinkage_protect_n(ret, "g" (arg1), "g" (arg2), "g" (arg3), \ + "g" (arg4), "g" (arg5), "g" (arg6)) + #endif #ifdef CONFIG_X86_ALIGNMENT_16 diff --git a/include/linux/linkage.h b/include/linux/linkage.h index 0592936344c4..fe2a39c489b6 100644 --- a/include/linux/linkage.h +++ b/include/linux/linkage.h @@ -17,8 +17,8 @@ # define asmregparm #endif -#ifndef prevent_tail_call -# define prevent_tail_call(ret) do { } while (0) +#ifndef asmlinkage_protect +# define asmlinkage_protect(n, ret, args...) do { } while (0) #endif #ifndef __ALIGN diff --git a/kernel/exit.c b/kernel/exit.c index 53872bf993fa..073005b1cfb2 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -1608,7 +1608,7 @@ asmlinkage long sys_waitid(int which, pid_t upid, put_pid(pid); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(5, ret, which, upid, infop, options, ru); return ret; } @@ -1640,7 +1640,7 @@ asmlinkage long sys_wait4(pid_t upid, int __user *stat_addr, put_pid(pid); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(4, ret, upid, stat_addr, options, ru); return ret; } diff --git a/kernel/uid16.c b/kernel/uid16.c index dd308ba4e03b..3e41c1673e2f 100644 --- a/kernel/uid16.c +++ b/kernel/uid16.c @@ -21,7 +21,7 @@ asmlinkage long sys_chown16(const char __user * filename, old_uid_t user, old_gi { long ret = sys_chown(filename, low2highuid(user), low2highgid(group)); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(3, ret, filename, user, group); return ret; } @@ -29,7 +29,7 @@ asmlinkage long sys_lchown16(const char __user * filename, old_uid_t user, old_g { long ret = sys_lchown(filename, low2highuid(user), low2highgid(group)); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(3, ret, filename, user, group); return ret; } @@ -37,7 +37,7 @@ asmlinkage long sys_fchown16(unsigned int fd, old_uid_t user, old_gid_t group) { long ret = sys_fchown(fd, low2highuid(user), low2highgid(group)); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(3, ret, fd, user, group); return ret; } @@ -45,7 +45,7 @@ asmlinkage long sys_setregid16(old_gid_t rgid, old_gid_t egid) { long ret = sys_setregid(low2highgid(rgid), low2highgid(egid)); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(2, ret, rgid, egid); return ret; } @@ -53,7 +53,7 @@ asmlinkage long sys_setgid16(old_gid_t gid) { long ret = sys_setgid(low2highgid(gid)); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(1, ret, gid); return ret; } @@ -61,7 +61,7 @@ asmlinkage long sys_setreuid16(old_uid_t ruid, old_uid_t euid) { long ret = sys_setreuid(low2highuid(ruid), low2highuid(euid)); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(2, ret, ruid, euid); return ret; } @@ -69,7 +69,7 @@ asmlinkage long sys_setuid16(old_uid_t uid) { long ret = sys_setuid(low2highuid(uid)); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(1, ret, uid); return ret; } @@ -78,7 +78,7 @@ asmlinkage long sys_setresuid16(old_uid_t ruid, old_uid_t euid, old_uid_t suid) long ret = sys_setresuid(low2highuid(ruid), low2highuid(euid), low2highuid(suid)); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(3, ret, ruid, euid, suid); return ret; } @@ -98,7 +98,7 @@ asmlinkage long sys_setresgid16(old_gid_t rgid, old_gid_t egid, old_gid_t sgid) long ret = sys_setresgid(low2highgid(rgid), low2highgid(egid), low2highgid(sgid)); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(3, ret, rgid, egid, sgid); return ret; } @@ -117,7 +117,7 @@ asmlinkage long sys_setfsuid16(old_uid_t uid) { long ret = sys_setfsuid(low2highuid(uid)); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(1, ret, uid); return ret; } @@ -125,7 +125,7 @@ asmlinkage long sys_setfsgid16(old_gid_t gid) { long ret = sys_setfsgid(low2highgid(gid)); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(1, ret, gid); return ret; } -- cgit v1.2.3 From d10d89ec78114f925f63c5126a2b2490f501a462 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Thu, 10 Apr 2008 17:35:23 -0700 Subject: Add commentary about the new "asmlinkage_protect()" macro It's really a pretty ugly thing to need, and some day it will hopefully be obviated by teaching gcc about the magic calling conventions for the low-level system call code, but in the meantime we can at least add big honking comments about why we need these insane and strange macros. I took my comments from my version of the macro, but I ended up deciding to just pick Roland's version of the actual code instead (with his prettier syntax that uses vararg macros). Thus the previous two commits that actually implement it. Signed-off-by: Linus Torvalds --- include/asm-x86/linkage.h | 11 +++++++++++ include/linux/linkage.h | 13 +++++++++++++ 2 files changed, 24 insertions(+) (limited to 'include') diff --git a/include/asm-x86/linkage.h b/include/asm-x86/linkage.h index d605eeba0f70..c048353f4b85 100644 --- a/include/asm-x86/linkage.h +++ b/include/asm-x86/linkage.h @@ -14,6 +14,17 @@ */ #define asmregparm __attribute__((regparm(3))) +/* + * Make sure the compiler doesn't do anything stupid with the + * arguments on the stack - they are owned by the *caller*, not + * the callee. This just fools gcc into not spilling into them, + * and keeps it from doing tailcall recursion and/or using the + * stack slots for temporaries, since they are live and "used" + * all the way to the end of the function. + * + * NOTE! On x86-64, all the arguments are in registers, so this + * only matters on a 32-bit kernel. + */ #define asmlinkage_protect(n, ret, args...) \ __asmlinkage_protect##n(ret, ##args) #define __asmlinkage_protect_n(ret, args...) \ diff --git a/include/linux/linkage.h b/include/linux/linkage.h index fe2a39c489b6..b163c5c40dbc 100644 --- a/include/linux/linkage.h +++ b/include/linux/linkage.h @@ -17,6 +17,19 @@ # define asmregparm #endif +/* + * This is used by architectures to keep arguments on the stack + * untouched by the compiler by keeping them live until the end. + * The argument stack may be owned by the assembly-language + * caller, not the callee, and gcc doesn't always understand + * that. + * + * We have the return value, and a maximum of six arguments. + * + * This should always be followed by a "return ret" for the + * protection to work (ie no more work that the compiler might + * end up needing stack temporaries for). + */ #ifndef asmlinkage_protect # define asmlinkage_protect(n, ret, args...) do { } while (0) #endif -- cgit v1.2.3 From 544451a1a36b06f43fc67112ede3b92e6203eb9d Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Thu, 10 Apr 2008 21:29:28 -0700 Subject: pnp: increase number of devices supported per protocol Increase the PNP "number of devices" limit. We currently use an unsigned char, which limits us to 256 devices per protocol. This patch changes that to an unsigned int. Not all backends can take advantage of this: we limit ISAPNP to 10 devices in isapnp_cfg_begin(), and PNPBIOS is limited to 256 devices because the BIOS interfaces use a one-byte device node number. But there is no limit on the number of PNPACPI devices we may have. Large HP Integrity machines have more than 256, which causes the current "unsigned char number" to wrap around. This causes errors like this: pnp: PnP ACPI init kobject_add failed for 00:00 with -EEXIST, don't try to register things with the same name in the same directory. Call Trace: [] show_stack+0x40/0xa0 [] dump_stack+0x30/0x60 [] kobject_add+0x290/0x2c0 [] device_add+0x160/0x860 [] device_register+0x30/0x60 [] __pnp_add_device+0x130/0x180 [] pnp_add_device+0xb0/0xe0 [] pnpacpi_add_device+0x510/0x5a0 [] pnpacpi_add_device_handler+0x50/0x80 This patch increases the limit to fix this PNPACPI problem. It should not have any adverse effect on ISAPNP or PNPBIOS because their limits are still enforced in the backends. Signed-off-by: Bjorn Helgaas Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/pnp.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/pnp.h b/include/linux/pnp.h index 29dd55838e84..b2f05c230f4b 100644 --- a/include/linux/pnp.h +++ b/include/linux/pnp.h @@ -175,7 +175,7 @@ static inline void pnp_set_card_drvdata(struct pnp_card_link *pcard, void *data) struct pnp_dev { struct device dev; /* Driver Model device interface */ u64 dma_mask; - unsigned char number; /* used as an index, must be unique */ + unsigned int number; /* used as an index, must be unique */ int status; struct list_head global_list; /* node in global list of devices */ -- cgit v1.2.3 From b0fac02370cffad956ff3de5e8ed4df7e7b875d7 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Fri, 11 Apr 2008 13:46:54 +0200 Subject: Fix "$(AS) -traditional" compile breakage caused by asmlinkage_protect git commit 54a015104136974262afa4b8ddd943ea70dec8a2 ("asmlinkage_protect replaces prevent_tail_call") causes this build failure on s390: AS arch/s390/kernel/entry64.o In file included from arch/s390/kernel/entry64.S:14: include/linux/linkage.h:34: error: syntax error in macro parameter list make[1]: *** [arch/s390/kernel/entry64.o] Error 1 make: *** [arch/s390/kernel] Error 2 and some other architectures. The reason is that some architectures add the "-traditional" flag to the invocation of $(AS), which disables variadic macro argument support. So just surround the new define with an #ifndef __ASSEMBLY__ to prevent any side effects on asm code. Cc: Roland McGrath Cc: Martin Schwidefsky Signed-off-by: Heiko Carstens Signed-off-by: Linus Torvalds --- include/linux/linkage.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include') diff --git a/include/linux/linkage.h b/include/linux/linkage.h index b163c5c40dbc..2119610b24f8 100644 --- a/include/linux/linkage.h +++ b/include/linux/linkage.h @@ -30,9 +30,12 @@ * protection to work (ie no more work that the compiler might * end up needing stack temporaries for). */ +/* Assembly files may be compiled with -traditional .. */ +#ifndef __ASSEMBLY__ #ifndef asmlinkage_protect # define asmlinkage_protect(n, ret, args...) do { } while (0) #endif +#endif #ifndef __ALIGN #define __ALIGN .align 4,0x90 -- cgit v1.2.3 From dc07e721a26ec7e0adb66340f1763d220cfbbd0c Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Mon, 7 Apr 2008 15:59:05 -0400 Subject: Spell out behavior of atomic_dec_and_lock() in kerneldoc A little more detail here wouldn't hurt. Signed-off-by: J. Bruce Fields Signed-off-by: Jonathan Corbet --- include/linux/spinlock.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include') diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h index 576a5f77d3bd..1129ee0a7180 100644 --- a/include/linux/spinlock.h +++ b/include/linux/spinlock.h @@ -341,6 +341,9 @@ static inline void double_spin_unlock(spinlock_t *l1, spinlock_t *l2, * atomic_dec_and_lock - lock on reaching reference count zero * @atomic: the atomic counter * @lock: the spinlock in question + * + * Decrements @atomic by 1. If the result is 0, returns true and locks + * @lock. Returns false for all other cases. */ extern int _atomic_dec_and_lock(atomic_t *atomic, spinlock_t *lock); #define atomic_dec_and_lock(atomic, lock) \ -- cgit v1.2.3 From f4ad85ca3ef8a1ede76c5020a28a8f4057b4d24f Mon Sep 17 00:00:00 2001 From: Gui Jianfeng Date: Sat, 12 Apr 2008 18:39:34 -0700 Subject: [SCTP]: Fix protocol violation when receiving an error lenght INIT-ACK When receiving an error length INIT-ACK during COOKIE-WAIT, a 0-vtag ABORT will be responsed. This action violates the protocol apparently. This patch achieves the following things. 1 If the INIT-ACK contains all the fixed parameters, use init-tag recorded from INIT-ACK as vtag. 2 If the INIT-ACK doesn't contain all the fixed parameters, just reflect its vtag. Signed-off-by: Gui Jianfeng Signed-off-by: Vlad Yasevich Signed-off-by: David S. Miller --- include/net/sctp/command.h | 1 + net/sctp/outqueue.c | 3 +++ net/sctp/sm_sideeffect.c | 3 +++ net/sctp/sm_statefuns.c | 18 ++++++++++++++++++ 4 files changed, 25 insertions(+) (limited to 'include') diff --git a/include/net/sctp/command.h b/include/net/sctp/command.h index 10ae2da6f93b..35b1e83fb96a 100644 --- a/include/net/sctp/command.h +++ b/include/net/sctp/command.h @@ -104,6 +104,7 @@ typedef enum { SCTP_CMD_ADAPTATION_IND, /* generate and send adaptation event */ SCTP_CMD_ASSOC_SHKEY, /* generate the association shared keys */ SCTP_CMD_T1_RETRAN, /* Mark for retransmission after T1 timeout */ + SCTP_CMD_UPDATE_INITTAG, /* Update peer inittag */ SCTP_CMD_LAST } sctp_verb_t; diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c index 1bb3c5c35d2a..c0714469233c 100644 --- a/net/sctp/outqueue.c +++ b/net/sctp/outqueue.c @@ -793,6 +793,9 @@ int sctp_outq_flush(struct sctp_outq *q, int rtx_timeout) break; case SCTP_CID_ABORT: + if (sctp_test_T_bit(chunk)) { + packet->vtag = asoc->c.my_vtag; + } case SCTP_CID_SACK: case SCTP_CID_HEARTBEAT: case SCTP_CID_HEARTBEAT_ACK: diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c index 28eb38eb6083..a4763fd24fd8 100644 --- a/net/sctp/sm_sideeffect.c +++ b/net/sctp/sm_sideeffect.c @@ -1536,6 +1536,9 @@ static int sctp_cmd_interpreter(sctp_event_t event_type, error = sctp_auth_asoc_init_active_key(asoc, GFP_ATOMIC); break; + case SCTP_CMD_UPDATE_INITTAG: + asoc->peer.i.init_tag = cmd->obj.u32; + break; default: printk(KERN_WARNING "Impossible command: %u, %p\n", diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index f2ed6473feef..3ef97499df0d 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -4144,6 +4144,24 @@ static sctp_disposition_t sctp_sf_abort_violation( goto nomem; if (asoc) { + /* Treat INIT-ACK as a special case during COOKIE-WAIT. */ + if (chunk->chunk_hdr->type == SCTP_CID_INIT_ACK && + !asoc->peer.i.init_tag) { + sctp_initack_chunk_t *initack; + + initack = (sctp_initack_chunk_t *)chunk->chunk_hdr; + if (!sctp_chunk_length_valid(chunk, + sizeof(sctp_initack_chunk_t))) + abort->chunk_hdr->flags |= SCTP_CHUNK_FLAG_T; + else { + unsigned int inittag; + + inittag = ntohl(initack->init_hdr.init_tag); + sctp_add_cmd_sf(commands, SCTP_CMD_UPDATE_INITTAG, + SCTP_U32(inittag)); + } + } + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(abort)); SCTP_INC_STATS(SCTP_MIB_OUTCTRLCHUNKS); -- cgit v1.2.3 From ab38fb04c9f8928cfaf6f4966633d783419906a1 Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Sat, 12 Apr 2008 18:40:06 -0700 Subject: [SCTP]: Fix compiler warning about const qualifiers Fix 3 warnings about discarding const qualifiers: net/sctp/ulpevent.c:862: warning: passing argument 1 of 'sctp_event2skb' discards qualifiers from pointer target type net/sctp/sm_statefuns.c:4393: warning: passing argument 1 of 'SCTP_ASOC' discards qualifiers from pointer target type net/sctp/socket.c:5874: warning: passing argument 1 of 'cmsg_nxthdr' discards qualifiers from pointer target type Signed-off-by: Vlad Yasevich Signed-off-by: David S. Miller --- include/net/sctp/ulpevent.h | 2 +- net/sctp/sm_statefuns.c | 5 +++-- net/sctp/socket.c | 5 +++-- net/sctp/ulpevent.c | 2 +- 4 files changed, 8 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/net/sctp/ulpevent.h b/include/net/sctp/ulpevent.h index 9bcfc12275e8..7ea12e8e6676 100644 --- a/include/net/sctp/ulpevent.h +++ b/include/net/sctp/ulpevent.h @@ -67,7 +67,7 @@ struct sctp_ulpevent { }; /* Retrieve the skb this event sits inside of. */ -static inline struct sk_buff *sctp_event2skb(struct sctp_ulpevent *ev) +static inline struct sk_buff *sctp_event2skb(const struct sctp_ulpevent *ev) { return container_of((void *)ev, struct sk_buff, cb); } diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index 3ef97499df0d..07194c2a32df 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -4367,6 +4367,7 @@ sctp_disposition_t sctp_sf_do_prm_asoc(const struct sctp_endpoint *ep, sctp_cmd_seq_t *commands) { struct sctp_chunk *repl; + struct sctp_association* my_asoc; /* The comment below says that we enter COOKIE-WAIT AFTER * sending the INIT, but that doesn't actually work in our @@ -4390,8 +4391,8 @@ sctp_disposition_t sctp_sf_do_prm_asoc(const struct sctp_endpoint *ep, /* Cast away the const modifier, as we want to just * rerun it through as a sideffect. */ - sctp_add_cmd_sf(commands, SCTP_CMD_NEW_ASOC, - SCTP_ASOC((struct sctp_association *) asoc)); + my_asoc = (struct sctp_association *)asoc; + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_ASOC, SCTP_ASOC(my_asoc)); /* Choose transport for INIT. */ sctp_add_cmd_sf(commands, SCTP_CMD_INIT_CHOOSE_TRANSPORT, diff --git a/net/sctp/socket.c b/net/sctp/socket.c index d994d822900d..998e63a31311 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -5868,11 +5868,12 @@ SCTP_STATIC int sctp_msghdr_parse(const struct msghdr *msg, sctp_cmsgs_t *cmsgs) { struct cmsghdr *cmsg; + struct msghdr *my_msg = (struct msghdr *)msg; for (cmsg = CMSG_FIRSTHDR(msg); cmsg != NULL; - cmsg = CMSG_NXTHDR((struct msghdr*)msg, cmsg)) { - if (!CMSG_OK(msg, cmsg)) + cmsg = CMSG_NXTHDR(my_msg, cmsg)) { + if (!CMSG_OK(my_msg, cmsg)) return -EINVAL; /* Should we parse this header or ignore? */ diff --git a/net/sctp/ulpevent.c b/net/sctp/ulpevent.c index b43f1f110f87..ce6cda6b6994 100644 --- a/net/sctp/ulpevent.c +++ b/net/sctp/ulpevent.c @@ -859,7 +859,7 @@ __u16 sctp_ulpevent_get_notification_type(const struct sctp_ulpevent *event) union sctp_notification *notification; struct sk_buff *skb; - skb = sctp_event2skb((struct sctp_ulpevent *)event); + skb = sctp_event2skb(event); notification = (union sctp_notification *) skb->data; return notification->sn_header.sn_type; } -- cgit v1.2.3 From 14daa02139dcb3193b2b0250c0720a23ef610c49 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Sat, 12 Apr 2008 18:48:58 -0700 Subject: net: make struct tun_struct private to tun.c There's no reason for this to be in the header, and it just hurts recompile time. Signed-off-by: Rusty Russell Acked-by: Max Krasnyanskiy Signed-off-by: David S. Miller --- drivers/net/tun.c | 33 +++++++++++++++++++++++++++++++++ include/linux/Kbuild | 2 +- include/linux/if_tun.h | 39 --------------------------------------- 3 files changed, 34 insertions(+), 40 deletions(-) (limited to 'include') diff --git a/drivers/net/tun.c b/drivers/net/tun.c index 7b816a032957..970ec4793442 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -67,9 +67,42 @@ #include #include +/* Uncomment to enable debugging */ +/* #define TUN_DEBUG 1 */ + #ifdef TUN_DEBUG static int debug; + +#define DBG if(tun->debug)printk +#define DBG1 if(debug==2)printk +#else +#define DBG( a... ) +#define DBG1( a... ) +#endif + +struct tun_struct { + struct list_head list; + unsigned long flags; + int attached; + uid_t owner; + gid_t group; + + wait_queue_head_t read_wait; + struct sk_buff_head readq; + + struct net_device *dev; + + struct fasync_struct *fasync; + + unsigned long if_flags; + u8 dev_addr[ETH_ALEN]; + u32 chr_filter[2]; + u32 net_filter[2]; + +#ifdef TUN_DEBUG + int debug; #endif +}; /* Network device part of the driver */ diff --git a/include/linux/Kbuild b/include/linux/Kbuild index 9cdd12a9e843..cedbbd806bf6 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -86,6 +86,7 @@ header-y += if_plip.h header-y += if_ppp.h header-y += if_slip.h header-y += if_strip.h +header-y += if_tun.h header-y += if_tunnel.h header-y += in6.h header-y += in_route.h @@ -229,7 +230,6 @@ unifdef-y += if_link.h unifdef-y += if_pppol2tp.h unifdef-y += if_pppox.h unifdef-y += if_tr.h -unifdef-y += if_tun.h unifdef-y += if_vlan.h unifdef-y += if_wanpipe.h unifdef-y += igmp.h diff --git a/include/linux/if_tun.h b/include/linux/if_tun.h index 72f1c5f47be3..8c71fe2fb1f5 100644 --- a/include/linux/if_tun.h +++ b/include/linux/if_tun.h @@ -18,47 +18,8 @@ #ifndef __IF_TUN_H #define __IF_TUN_H -/* Uncomment to enable debugging */ -/* #define TUN_DEBUG 1 */ - #include -#ifdef __KERNEL__ - -#ifdef TUN_DEBUG -#define DBG if(tun->debug)printk -#define DBG1 if(debug==2)printk -#else -#define DBG( a... ) -#define DBG1( a... ) -#endif - -struct tun_struct { - struct list_head list; - unsigned long flags; - int attached; - uid_t owner; - gid_t group; - - wait_queue_head_t read_wait; - struct sk_buff_head readq; - - struct net_device *dev; - - struct fasync_struct *fasync; - - unsigned long if_flags; - u8 dev_addr[ETH_ALEN]; - u32 chr_filter[2]; - u32 net_filter[2]; - -#ifdef TUN_DEBUG - int debug; -#endif -}; - -#endif /* __KERNEL__ */ - /* Read queue size */ #define TUN_READQ_SIZE 500 -- cgit v1.2.3 From 7cd636fe9ce5de0051c11283911baca4291868c8 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Sun, 13 Apr 2008 23:33:52 -0700 Subject: [IPV6]: IPv6 extension header structures need to be packed. struct ipv6_opt_hdr is the common structure for IPv6 extension headers, and it is common to increment the pointer to get the real content. On the other hand, since the structure consists only of 1-byte next-header field and 1-byte length field, size of that structure depends on architecture; 2 or 4. Add "packed" attribute to get 2. Signed-off-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller --- include/linux/ipv6.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h index 4aaefc349a4b..134c8e5cf07c 100644 --- a/include/linux/ipv6.h +++ b/include/linux/ipv6.h @@ -53,7 +53,7 @@ struct ipv6_opt_hdr { /* * TLV encoded option data follows. */ -}; +} __attribute__ ((packed)); /* required for some archs */ #define ipv6_destopt_hdr ipv6_opt_hdr #define ipv6_hopopt_hdr ipv6_opt_hdr -- cgit v1.2.3 From e9df2e8fd8fbc95c57dbd1d33dada66c4627b44c Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Sun, 13 Apr 2008 23:40:51 -0700 Subject: [IPV6]: Use appropriate sock tclass setting for routing lookup. Signed-off-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller --- include/net/inet_ecn.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/inet_ecn.h b/include/net/inet_ecn.h index ba33db053854..7040a782c656 100644 --- a/include/net/inet_ecn.h +++ b/include/net/inet_ecn.h @@ -47,7 +47,7 @@ static inline __u8 INET_ECN_encapsulate(__u8 outer, __u8 inner) } while (0) #define IP6_ECN_flow_xmit(sk, label) do { \ - if (INET_ECN_is_capable(inet_sk(sk)->tos)) \ + if (INET_ECN_is_capable(inet6_sk(sk)->tclass)) \ (label) |= htonl(INET_ECN_ECT_0 << 20); \ } while (0) -- cgit v1.2.3