From 184e1fdfea066ab8f12a1e8912f402d2d6556d11 Mon Sep 17 00:00:00 2001 From: Huang Ying Date: Mon, 15 Jun 2009 15:37:07 +0800 Subject: x86, mce: fix a race condition about mce_callin and no_way_out If one CPU has no_way_out == 1, all other CPUs should have no_way_out == 1. But despite global_nwo is read after mce_callin, global_nwo is updated after mce_callin too. So it is possible that some CPU read global_nwo before some other CPU update global_nwo, so that no_way_out == 1 for some CPU, while no_way_out == 0 for some other CPU. This patch fixes this race condition via moving mce_callin updating after global_nwo updating, with a smp_wmb in between. A smp_rmb is added between their reading too. Signed-off-by: Huang Ying Acked-by: Andi Kleen Acked-by: Hidetoshi Seto --- arch/x86/kernel/cpu/mcheck/mce.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) (limited to 'arch/x86/kernel/cpu/mcheck/mce.c') diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index fabba15e455..19294b8524c 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -703,6 +703,11 @@ static int mce_start(int no_way_out, int *order) } atomic_add(no_way_out, &global_nwo); + /* + * global_nwo should be updated before mce_callin + */ + smp_wmb(); + *order = atomic_add_return(1, &mce_callin); /* * Wait for everyone. @@ -716,6 +721,10 @@ static int mce_start(int no_way_out, int *order) ndelay(SPINUNIT); } + /* + * mce_callin should be read before global_nwo + */ + smp_rmb(); /* * Cache the global no_way_out state. */ @@ -862,7 +871,7 @@ void do_machine_check(struct pt_regs *regs, long error_code) * Establish sequential order between the CPUs entering the machine * check handler. */ - int order; + int order = -1; /* * If no_way_out gets set, there is no safe way to recover from this @@ -887,7 +896,6 @@ void do_machine_check(struct pt_regs *regs, long error_code) if (!banks) goto out; - order = atomic_add_return(1, &mce_callin); mce_setup(&m); m.mcgstatus = mce_rdmsrl(MSR_IA32_MCG_STATUS); -- cgit v1.2.3 From 33edbf02a92771fa2a81e41084a44ba874e3a5a5 Mon Sep 17 00:00:00 2001 From: Hidetoshi Seto Date: Mon, 15 Jun 2009 17:18:45 +0900 Subject: x86, mce: don't init timer if !mce_available In mce_cpu_restart, mce_init_timer is called unconditionally. If !mce_available (e.g. mce is disabled), there are no useful work for timer. Stop running it. Signed-off-by: Hidetoshi Seto Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/mcheck/mce.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'arch/x86/kernel/cpu/mcheck/mce.c') diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index 19294b8524c..dda77215e9e 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -1617,8 +1617,9 @@ static int mce_resume(struct sys_device *dev) static void mce_cpu_restart(void *data) { del_timer_sync(&__get_cpu_var(mce_timer)); - if (mce_available(¤t_cpu_data)) - mce_init(); + if (!mce_available(¤t_cpu_data)) + return; + mce_init(); mce_init_timer(); } -- cgit v1.2.3 From 7fb06fc9672b947424e05871243a4c8e19ec3bce Mon Sep 17 00:00:00 2001 From: Hidetoshi Seto Date: Mon, 15 Jun 2009 18:18:43 +0900 Subject: x86, mce: cleanup mce_start() Simplify interface of mce_start(): - no_way_out = mce_start(no_way_out, &order); + order = mce_start(&no_way_out); Now Monarch and Subjects share same exit(return) in usual path. Signed-off-by: Hidetoshi Seto Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/mcheck/mce.c | 66 +++++++++++++++++++--------------------- 1 file changed, 31 insertions(+), 35 deletions(-) (limited to 'arch/x86/kernel/cpu/mcheck/mce.c') diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index dda77215e9e..739fd7eca0a 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -691,23 +691,21 @@ static atomic_t global_nwo; * in the entry order. * TBD double check parallel CPU hotunplug */ -static int mce_start(int no_way_out, int *order) +static int mce_start(int *no_way_out) { - int nwo; + int order; int cpus = num_online_cpus(); u64 timeout = (u64)monarch_timeout * NSEC_PER_USEC; - if (!timeout) { - *order = -1; - return no_way_out; - } + if (!timeout) + return -1; - atomic_add(no_way_out, &global_nwo); + atomic_add(*no_way_out, &global_nwo); /* * global_nwo should be updated before mce_callin */ smp_wmb(); - *order = atomic_add_return(1, &mce_callin); + order = atomic_add_return(1, &mce_callin); /* * Wait for everyone. @@ -715,8 +713,7 @@ static int mce_start(int no_way_out, int *order) while (atomic_read(&mce_callin) != cpus) { if (mce_timed_out(&timeout)) { atomic_set(&global_nwo, 0); - *order = -1; - return no_way_out; + return -1; } ndelay(SPINUNIT); } @@ -725,34 +722,34 @@ static int mce_start(int no_way_out, int *order) * mce_callin should be read before global_nwo */ smp_rmb(); - /* - * Cache the global no_way_out state. - */ - nwo = atomic_read(&global_nwo); - /* - * Monarch starts executing now, the others wait. - */ - if (*order == 1) { + if (order == 1) { + /* + * Monarch: Starts executing now, the others wait. + */ atomic_set(&mce_executing, 1); - return nwo; + } else { + /* + * Subject: Now start the scanning loop one by one in + * the original callin order. + * This way when there are any shared banks it will be + * only seen by one CPU before cleared, avoiding duplicates. + */ + while (atomic_read(&mce_executing) < order) { + if (mce_timed_out(&timeout)) { + atomic_set(&global_nwo, 0); + return -1; + } + ndelay(SPINUNIT); + } } /* - * Now start the scanning loop one by one - * in the original callin order. - * This way when there are any shared banks it will - * be only seen by one CPU before cleared, avoiding duplicates. + * Cache the global no_way_out state. */ - while (atomic_read(&mce_executing) < *order) { - if (mce_timed_out(&timeout)) { - atomic_set(&global_nwo, 0); - *order = -1; - return no_way_out; - } - ndelay(SPINUNIT); - } - return nwo; + *no_way_out = atomic_read(&global_nwo); + + return order; } /* @@ -871,8 +868,7 @@ void do_machine_check(struct pt_regs *regs, long error_code) * Establish sequential order between the CPUs entering the machine * check handler. */ - int order = -1; - + int order; /* * If no_way_out gets set, there is no safe way to recover from this * MCE. If tolerant is cranked up, we'll try anyway. @@ -917,7 +913,7 @@ void do_machine_check(struct pt_regs *regs, long error_code) * This way we don't report duplicated events on shared banks * because the first one to see it will clear it. */ - no_way_out = mce_start(no_way_out, &order); + order = mce_start(&no_way_out); for (i = 0; i < banks; i++) { __clear_bit(i, toclear); if (!bank[i]) -- cgit v1.2.3 From 4e5b3e690dda890523e93af9c545261f5916a3a6 Mon Sep 17 00:00:00 2001 From: Hidetoshi Seto Date: Mon, 15 Jun 2009 17:20:20 +0900 Subject: x86, mce: add __read_mostly Add __read_mostly to data written during setup. Suggested-by: Ingo Molnar Signed-off-by: Hidetoshi Seto Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/mcheck/mce.c | 26 +++++++++++++------------- 1 file changed, 13 insertions(+), 13 deletions(-) (limited to 'arch/x86/kernel/cpu/mcheck/mce.c') diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index 739fd7eca0a..2d2e0b565a9 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -57,7 +57,7 @@ static void unexpected_machine_check(struct pt_regs *regs, long error_code) void (*machine_check_vector)(struct pt_regs *, long error_code) = unexpected_machine_check; -int mce_disabled; +int mce_disabled __read_mostly; #ifdef CONFIG_X86_NEW_MCE @@ -76,19 +76,19 @@ DEFINE_PER_CPU(unsigned, mce_exception_count); * 2: SIGBUS or log uncorrected errors (if possible), log corrected errors * 3: never panic or SIGBUS, log all errors (for testing only) */ -static int tolerant = 1; -static int banks; -static u64 *bank; -static unsigned long notify_user; -static int rip_msr; -static int mce_bootlog = -1; -static int monarch_timeout = -1; -static int mce_panic_timeout; -static int mce_dont_log_ce; -int mce_cmci_disabled; -int mce_ignore_ce; -int mce_ser; +static int tolerant __read_mostly = 1; +static int banks __read_mostly; +static u64 *bank __read_mostly; +static int rip_msr __read_mostly; +static int mce_bootlog __read_mostly = -1; +static int monarch_timeout __read_mostly = -1; +static int mce_panic_timeout __read_mostly; +static int mce_dont_log_ce __read_mostly; +int mce_cmci_disabled __read_mostly; +int mce_ignore_ce __read_mostly; +int mce_ser __read_mostly; +static unsigned long notify_user; static char trigger[128]; static char *trigger_argv[2] = { trigger, NULL }; -- cgit v1.2.3 From 1020bcbcc7da36001d9226c5d57e999949cb80c5 Mon Sep 17 00:00:00 2001 From: Hidetoshi Seto Date: Mon, 15 Jun 2009 17:20:57 +0900 Subject: x86, mce: rename static variables around trigger "trigger" is not straight forward name for valiable that holds name of user mode helper program which triggered by machine check events. This patch renames this valiable and kins to more recognizable names. trigger => mce_helper trigger_argv => mce_helper_argv notify_user => mce_need_notify No functional changes. Signed-off-by: Hidetoshi Seto Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/mcheck/mce.c | 27 ++++++++++++++------------- 1 file changed, 14 insertions(+), 13 deletions(-) (limited to 'arch/x86/kernel/cpu/mcheck/mce.c') diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index 2d2e0b565a9..e8b893e880e 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -88,9 +88,10 @@ int mce_cmci_disabled __read_mostly; int mce_ignore_ce __read_mostly; int mce_ser __read_mostly; -static unsigned long notify_user; -static char trigger[128]; -static char *trigger_argv[2] = { trigger, NULL }; +/* User mode helper program triggered by machine check event */ +static unsigned long mce_need_notify; +static char mce_helper[128]; +static char *mce_helper_argv[2] = { mce_helper, NULL }; static unsigned long dont_init_banks; @@ -180,7 +181,7 @@ void mce_log(struct mce *mce) wmb(); mce->finished = 1; - set_bit(0, ¬ify_user); + set_bit(0, &mce_need_notify); } static void print_mce(struct mce *m) @@ -1122,7 +1123,7 @@ static void mcheck_timer(unsigned long data) static void mce_do_trigger(struct work_struct *work) { - call_usermodehelper(trigger, trigger_argv, NULL, UMH_NO_WAIT); + call_usermodehelper(mce_helper, mce_helper_argv, NULL, UMH_NO_WAIT); } static DECLARE_WORK(mce_trigger_work, mce_do_trigger); @@ -1139,7 +1140,7 @@ int mce_notify_irq(void) clear_thread_flag(TIF_MCE_NOTIFY); - if (test_and_clear_bit(0, ¬ify_user)) { + if (test_and_clear_bit(0, &mce_need_notify)) { wake_up_interruptible(&mce_wait); /* @@ -1147,7 +1148,7 @@ int mce_notify_irq(void) * work_pending is always cleared before the function is * executed. */ - if (trigger[0] && !work_pending(&mce_trigger_work)) + if (mce_helper[0] && !work_pending(&mce_trigger_work)) schedule_work(&mce_trigger_work); if (__ratelimit(&ratelimit)) @@ -1664,9 +1665,9 @@ static ssize_t set_bank(struct sys_device *s, struct sysdev_attribute *attr, static ssize_t show_trigger(struct sys_device *s, struct sysdev_attribute *attr, char *buf) { - strcpy(buf, trigger); + strcpy(buf, mce_helper); strcat(buf, "\n"); - return strlen(trigger) + 1; + return strlen(mce_helper) + 1; } static ssize_t set_trigger(struct sys_device *s, struct sysdev_attribute *attr, @@ -1675,10 +1676,10 @@ static ssize_t set_trigger(struct sys_device *s, struct sysdev_attribute *attr, char *p; int len; - strncpy(trigger, buf, sizeof(trigger)); - trigger[sizeof(trigger)-1] = 0; - len = strlen(trigger); - p = strchr(trigger, '\n'); + strncpy(mce_helper, buf, sizeof(mce_helper)); + mce_helper[sizeof(mce_helper)-1] = 0; + len = strlen(mce_helper); + p = strchr(mce_helper, '\n'); if (*p) *p = 0; -- cgit v1.2.3 From 9af43b54ab4509f1dac49637d6917d57292e6518 Mon Sep 17 00:00:00 2001 From: Hidetoshi Seto Date: Mon, 15 Jun 2009 17:21:36 +0900 Subject: x86, mce: sysfs entries for new mce options Add sysfs interface for admins who want to tweak these options without rebooting the system. Signed-off-by: Hidetoshi Seto Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/mcheck/mce.c | 84 +++++++++++++++++++++++++++++++++++++++- 1 file changed, 83 insertions(+), 1 deletion(-) (limited to 'arch/x86/kernel/cpu/mcheck/mce.c') diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index e8b893e880e..4b62443b6e7 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -1626,6 +1626,26 @@ static void mce_restart(void) on_each_cpu(mce_cpu_restart, NULL, 1); } +/* Toggle features for corrected errors */ +static void mce_disable_ce(void *all) +{ + if (!mce_available(¤t_cpu_data)) + return; + if (all) + del_timer_sync(&__get_cpu_var(mce_timer)); + cmci_clear(); +} + +static void mce_enable_ce(void *all) +{ + if (!mce_available(¤t_cpu_data)) + return; + cmci_reenable(); + cmci_recheck(); + if (all) + mce_init_timer(); +} + static struct sysdev_class mce_sysclass = { .suspend = mce_suspend, .shutdown = mce_shutdown, @@ -1687,6 +1707,52 @@ static ssize_t set_trigger(struct sys_device *s, struct sysdev_attribute *attr, return len; } +static ssize_t set_ignore_ce(struct sys_device *s, + struct sysdev_attribute *attr, + const char *buf, size_t size) +{ + u64 new; + + if (strict_strtoull(buf, 0, &new) < 0) + return -EINVAL; + + if (mce_ignore_ce ^ !!new) { + if (new) { + /* disable ce features */ + on_each_cpu(mce_disable_ce, (void *)1, 1); + mce_ignore_ce = 1; + } else { + /* enable ce features */ + mce_ignore_ce = 0; + on_each_cpu(mce_enable_ce, (void *)1, 1); + } + } + return size; +} + +static ssize_t set_cmci_disabled(struct sys_device *s, + struct sysdev_attribute *attr, + const char *buf, size_t size) +{ + u64 new; + + if (strict_strtoull(buf, 0, &new) < 0) + return -EINVAL; + + if (mce_cmci_disabled ^ !!new) { + if (new) { + /* disable cmci */ + on_each_cpu(mce_disable_ce, NULL, 1); + mce_cmci_disabled = 1; + } else { + /* enable cmci */ + mce_cmci_disabled = 0; + on_each_cpu(mce_enable_ce, NULL, 1); + } + } + return size; +} + static ssize_t store_int_with_restart(struct sys_device *s, struct sysdev_attribute *attr, const char *buf, size_t size) @@ -1699,6 +1765,7 @@ static ssize_t store_int_with_restart(struct sys_device *s, static SYSDEV_ATTR(trigger, 0644, show_trigger, set_trigger); static SYSDEV_INT_ATTR(tolerant, 0644, tolerant); static SYSDEV_INT_ATTR(monarch_timeout, 0644, monarch_timeout); +static SYSDEV_INT_ATTR(dont_log_ce, 0644, mce_dont_log_ce); static struct sysdev_ext_attribute attr_check_interval = { _SYSDEV_ATTR(check_interval, 0644, sysdev_show_int, @@ -1706,9 +1773,24 @@ static struct sysdev_ext_attribute attr_check_interval = { &check_interval }; +static struct sysdev_ext_attribute attr_ignore_ce = { + _SYSDEV_ATTR(ignore_ce, 0644, sysdev_show_int, set_ignore_ce), + &mce_ignore_ce +}; + +static struct sysdev_ext_attribute attr_cmci_disabled = { + _SYSDEV_ATTR(ignore_ce, 0644, sysdev_show_int, set_cmci_disabled), + &mce_cmci_disabled +}; + static struct sysdev_attribute *mce_attrs[] = { - &attr_tolerant.attr, &attr_check_interval.attr, &attr_trigger, + &attr_tolerant.attr, + &attr_check_interval.attr, + &attr_trigger, &attr_monarch_timeout.attr, + &attr_dont_log_ce.attr, + &attr_ignore_ce.attr, + &attr_cmci_disabled.attr, NULL }; -- cgit v1.2.3 From 9e55e44e39798541ba39d57f4b569deb555ae1ce Mon Sep 17 00:00:00 2001 From: Hidetoshi Seto Date: Mon, 15 Jun 2009 17:22:15 +0900 Subject: x86, mce: unify mce.h There are 2 headers: arch/x86/include/asm/mce.h arch/x86/kernel/cpu/mcheck/mce.h and in the latter small header: #include This patch move all contents in the latter header into the former, and fix all files using the latter to include the former instead. Signed-off-by: Hidetoshi Seto Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/mcheck/mce.c | 1 - 1 file changed, 1 deletion(-) (limited to 'arch/x86/kernel/cpu/mcheck/mce.c') diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index 4b62443b6e7..faedd776847 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -44,7 +44,6 @@ #include #include "mce-internal.h" -#include "mce.h" /* Handle unconfigured int18 (should never happen) */ static void unexpected_machine_check(struct pt_regs *regs, long error_code) -- cgit v1.2.3 From c697836985e18d9c34897428ba563b13044a6dcd Mon Sep 17 00:00:00 2001 From: Hidetoshi Seto Date: Mon, 15 Jun 2009 17:22:49 +0900 Subject: x86, mce: make mce_disabled boolean The mce_disabled on 32bit is a tristate variable [1,0,-1], while 64bit version is boolean [0,1]. This patch makes mce_disabled always boolean, and use mce_p5_enabled to indicate the third state instead. Signed-off-by: Hidetoshi Seto Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/mcheck/mce.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) (limited to 'arch/x86/kernel/cpu/mcheck/mce.c') diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index faedd776847..6095e0296ab 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -1286,8 +1286,7 @@ static void __cpuinit mce_ancient_init(struct cpuinfo_x86 *c) return; switch (c->x86_vendor) { case X86_VENDOR_INTEL: - if (mce_p5_enabled()) - intel_p5_mcheck_init(c); + intel_p5_mcheck_init(c); break; case X86_VENDOR_CENTAUR: winchip_mcheck_init(c); @@ -2002,7 +2001,7 @@ EXPORT_SYMBOL_GPL(nr_mce_banks); /* non-fatal.o */ /* This has to be run for each processor */ void mcheck_init(struct cpuinfo_x86 *c) { - if (mce_disabled == 1) + if (mce_disabled) return; switch (c->x86_vendor) { @@ -2032,10 +2031,9 @@ void mcheck_init(struct cpuinfo_x86 *c) static int __init mcheck_enable(char *str) { - mce_disabled = -1; + mce_p5_enabled = 1; return 1; } - __setup("mce", mcheck_enable); #endif /* CONFIG_X86_OLD_MCE */ -- cgit v1.2.3 From 203abd67b75f7714ce98ab0cdbd6cfd7ad79dec4 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Mon, 15 Jun 2009 14:52:01 +0200 Subject: x86: mce: Handle banks == 0 case in K7 quirk Vegard Nossum reported: > I get an MCE-related crash like this in latest linus tree: > > [ 0.115341] CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line) > [ 0.116396] CPU: L2 Cache: 512K (64 bytes/line) > [ 0.120570] mce: CPU supports 0 MCE banks > [ 0.124870] BUG: unable to handle kernel NULL pointer dereference at 00000000 00000010 > [ 0.128001] IP: [] mcheck_init+0x278/0x320 > [ 0.128001] PGD 0 > [ 0.128001] Thread overran stack, or stack corrupted > [ 0.128001] Oops: 0002 [#1] PREEMPT SMP > [ 0.128001] last sysfs file: > [ 0.128001] CPU 0 > [ 0.128001] Modules linked in: > [ 0.128001] Pid: 0, comm: swapper Not tainted 2.6.30 #426 > [ 0.128001] RIP: 0010:[] [] mcheck_init+0x278/0x320 > [ 0.128001] RSP: 0018:ffffffff81595e38 EFLAGS: 00000246 > [ 0.128001] RAX: 0000000000000010 RBX: ffffffff8158f900 RCX: 0000000000000000 > [ 0.128001] RDX: 0000000000000000 RSI: 00000000000000ff RDI: 0000000000000010 > [ 0.128001] RBP: ffffffff81595e68 R08: 0000000000000001 R09: 0000000000000000 > [ 0.128001] R10: 0000000000000010 R11: 0000000000000000 R12: 0000000000000000 > [ 0.128001] R13: 00000000ffffffff R14: 0000000000000000 R15: 0000000000000000 > [ 0.128001] FS: 0000000000000000(0000) GS:ffff880002288000(0000) knlGS:00000 > 00000000000 > [ 0.128001] CS: 0010 DS: 0018 ES: 0018 CR0: 000000008005003b > [ 0.128001] CR2: 0000000000000010 CR3: 0000000001001000 CR4: 00000000000006b0 > [ 0.128001] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 > [ 0.128001] DR3: 0000000000000000 DR6: 0000000000000000 DR7: 0000000000000000 > [ 0.128001] Process swapper (pid: 0, threadinfo ffffffff81594000, task ffffff > ff8152a4a0) > [ 0.128001] Stack: > [ 0.128001] 0000000081595e68 5aa50ed3b4ddbe6e ffffffff8158f900 ffffffff8158f > 914 > [ 0.128001] ffffffff8158f948 0000000000000000 ffffffff81595eb8 ffffffff813b8 > 69c > [ 0.128001] 5aa50ed3b4ddbe6e 00000001078bfbfd 0000062300000800 5aa50ed3b4ddb > e6e > [ 0.128001] Call Trace: > [ 0.128001] [] identify_cpu+0x331/0x392 > [ 0.128001] [] identify_boot_cpu+0x23/0x6e > [ 0.128001] [] check_bugs+0x1c/0x60 > [ 0.128001] [] start_kernel+0x403/0x46e > [ 0.128001] [] x86_64_start_reservations+0xac/0xd5 > [ 0.128001] [] x86_64_start_kernel+0x115/0x14b > [ 0.128001] [] ? early_idt_handler+0x0/0x71 This happens on QEMU which reports MCA capability, but no banks. Without this patch there is a buffer overrun and boot ops because the code would try to initialize the 0 element of a zero length kmalloc() buffer. Reported-by: Vegard Nossum Tested-by: Pekka Enberg Signed-off-by: Andi Kleen LKML-Reference: <20090615125200.GD31969@one.firstfloor.org> Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/mcheck/mce.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86/kernel/cpu/mcheck/mce.c') diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index fabba15e455..d9d77cfd8cc 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -1245,7 +1245,7 @@ static void mce_cpu_quirks(struct cpuinfo_x86 *c) * Various K7s with broken bank 0 around. Always disable * by default. */ - if (c->x86 == 6) + if (c->x86 == 6 && banks > 0) bank[0] = 0; } -- cgit v1.2.3 From 74b602c7147212a7495879ec23fe6c2d3b470e06 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Wed, 17 Jun 2009 14:43:32 -0700 Subject: x86: fix duplicated sysfs attribute The sysfs attribute cmci_disabled was accidentall turned into a duplicate of ignore_ce, breaking all other attributes. Signed-off-by: Yinghai Lu Acked-by: Hidetoshi Seto Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/mcheck/mce.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86/kernel/cpu/mcheck/mce.c') diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index 2a560cefb67..5aac9e4dd13 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -1777,7 +1777,7 @@ static struct sysdev_ext_attribute attr_ignore_ce = { }; static struct sysdev_ext_attribute attr_cmci_disabled = { - _SYSDEV_ATTR(ignore_ce, 0644, sysdev_show_int, set_cmci_disabled), + _SYSDEV_ATTR(cmci_disabled, 0644, sysdev_show_int, set_cmci_disabled), &mce_cmci_disabled }; -- cgit v1.2.3 From e92fae064ae42b2a4a77646f7655bca4c87bb1eb Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Wed, 17 Jun 2009 16:21:33 -0700 Subject: x86: use zalloc_cpumask_var for mce_dev_initialized We need a cleared cpu_mask to record if mce is initialized, especially when MAXSMP is used. used zalloc_... instead Signed-off-by: Yinghai Lu Reviewed-by: Hidetoshi Seto Cc: stable@kernel.org Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/mcheck/mce.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86/kernel/cpu/mcheck/mce.c') diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index 5aac9e4dd13..c2fb70d0286 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -1969,7 +1969,7 @@ static __init int mce_init_device(void) if (!mce_available(&boot_cpu_data)) return -EIO; - alloc_cpumask_var(&mce_dev_initialized, GFP_KERNEL); + zalloc_cpumask_var(&mce_dev_initialized, GFP_KERNEL); err = mce_init_banks(); if (err) -- cgit v1.2.3 From b1f49f9582f9be6de5055cfa97eabf6246f2eaf7 Mon Sep 17 00:00:00 2001 From: Hidetoshi Seto Date: Thu, 18 Jun 2009 14:53:24 +0900 Subject: x86, mce: fix error path in mce_create_device() Don't skip removing mce_attrs in route from error2. Signed-off-by: Hidetoshi Seto Cc: Andi Kleen Cc: Huang Ying Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/mcheck/mce.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'arch/x86/kernel/cpu/mcheck/mce.c') diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index c2fb70d0286..284d1de968b 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -1798,7 +1798,7 @@ static cpumask_var_t mce_dev_initialized; static __cpuinit int mce_create_device(unsigned int cpu) { int err; - int i; + int i, j; if (!mce_available(&boot_cpu_data)) return -EIO; @@ -1816,9 +1816,9 @@ static __cpuinit int mce_create_device(unsigned int cpu) if (err) goto error; } - for (i = 0; i < banks; i++) { + for (j = 0; j < banks; j++) { err = sysdev_create_file(&per_cpu(mce_dev, cpu), - &bank_attrs[i]); + &bank_attrs[j]); if (err) goto error2; } @@ -1826,8 +1826,8 @@ static __cpuinit int mce_create_device(unsigned int cpu) return 0; error2: - while (--i >= 0) - sysdev_remove_file(&per_cpu(mce_dev, cpu), &bank_attrs[i]); + while (--j >= 0) + sysdev_remove_file(&per_cpu(mce_dev, cpu), &bank_attrs[j]); error: while (--i >= 0) sysdev_remove_file(&per_cpu(mce_dev, cpu), mce_attrs[i]); -- cgit v1.2.3 From 5be6066a7f8d917db347d94f1b359b9b70dcb572 Mon Sep 17 00:00:00 2001 From: Hidetoshi Seto Date: Wed, 24 Jun 2009 09:21:10 +0900 Subject: x86, mce: percpu mcheck_timer should be pinned If CONFIG_NO_HZ + CONFIG_SMP, timer added via add_timer() might be migrated on other cpu. Use add_timer_on() instead. Avoids the following failure: Maciej Rutecki wrote: > > After normal boot I try: > > > > echo 1 > /sys/devices/system/machinecheck/machinecheck0/check_interval > > > > I found this in dmesg: > > > > [ 141.704025] ------------[ cut here ]------------ > > [ 141.704039] WARNING: at arch/x86/kernel/cpu/mcheck/mce.c:1102 > > mcheck_timer+0xf5/0x100() Reported-by: Maciej Rutecki Signed-off-by: Hidetoshi Seto Tested-by: Maciej Rutecki Acked-by: Andi Kleen Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/mcheck/mce.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/x86/kernel/cpu/mcheck/mce.c') diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index 284d1de968b..af425b83202 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -1117,7 +1117,7 @@ static void mcheck_timer(unsigned long data) *n = min(*n*2, (int)round_jiffies_relative(check_interval*HZ)); t->expires = jiffies + *n; - add_timer(t); + add_timer_on(t, smp_processor_id()); } static void mce_do_trigger(struct work_struct *work) @@ -1321,7 +1321,7 @@ static void mce_init_timer(void) return; setup_timer(t, mcheck_timer, smp_processor_id()); t->expires = round_jiffies(jiffies + *n); - add_timer(t); + add_timer_on(t, smp_processor_id()); } /* -- cgit v1.2.3