diff options
34 files changed, 1062 insertions, 695 deletions
diff --git a/Documentation/Changes b/Documentation/Changes index b95082be4d5..d21b3b5aa54 100644 --- a/Documentation/Changes +++ b/Documentation/Changes @@ -48,6 +48,7 @@ o procps 3.2.0 # ps --version o oprofile 0.9 # oprofiled --version o udev 081 # udevinfo -V o grub 0.93 # grub --version +o mcelog 0.6 Kernel compilation ================== @@ -276,6 +277,16 @@ before running exportfs or mountd. It is recommended that all NFS services be protected from the internet-at-large by a firewall where that is possible. +mcelog +------ + +In Linux 2.6.31+ the i386 kernel needs to run the mcelog utility +as a regular cronjob similar to the x86-64 kernel to process and log +machine check events when CONFIG_X86_NEW_MCE is enabled. Machine check +events are errors reported by the CPU. Processing them is strongly encouraged. +All x86-64 kernels since 2.6.4 require the mcelog utility to +process machine checks. + Getting updated software ======================== @@ -365,6 +376,10 @@ FUSE ---- o <http://sourceforge.net/projects/fuse> +mcelog +------ +o <ftp://ftp.kernel.org/pub/linux/utils/cpu/mce/mcelog/> + Networking ********** diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index de491a3e231..ec9ef5d0d7b 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -437,3 +437,13 @@ Why: Superseded by tdfxfb. I2C/DDC support used to live in a separate driver but this caused driver conflicts. Who: Jean Delvare <khali@linux-fr.org> Krzysztof Helt <krzysztof.h1@wp.pl> + +---------------------------- + +What: CONFIG_X86_OLD_MCE +When: 2.6.32 +Why: Remove the old legacy 32bit machine check code. This has been + superseded by the newer machine check code from the 64bit port, + but the old version has been kept around for easier testing. Note this + doesn't impact the old P5 and WinChip machine check handlers. +Who: Andi Kleen <andi@firstfloor.org> diff --git a/Documentation/x86/x86_64/boot-options.txt b/Documentation/x86/x86_64/boot-options.txt index 34c13040a71..63fca718256 100644 --- a/Documentation/x86/x86_64/boot-options.txt +++ b/Documentation/x86/x86_64/boot-options.txt @@ -5,6 +5,8 @@ only the AMD64 specific ones are listed here. Machine check + Please see Documentation/x86/x86_64/machinecheck for sysfs runtime tunables. + mce=off disable machine check mce=bootlog Enable logging of machine checks left over from booting. Disabled by default on AMD because some BIOS leave bogus ones. diff --git a/Documentation/x86/x86_64/machinecheck b/Documentation/x86/x86_64/machinecheck index a05e58e7b15..a4fdb25446e 100644 --- a/Documentation/x86/x86_64/machinecheck +++ b/Documentation/x86/x86_64/machinecheck @@ -41,7 +41,9 @@ check_interval the polling interval. When the poller stops finding MCEs, it triggers an exponential backoff (poll less often) on the polling interval. The check_interval variable is both the initial and - maximum polling interval. + maximum polling interval. 0 means no polling for corrected machine + check errors (but some corrected errors might be still reported + in other ways) tolerant Tolerance level. When a machine check exception occurs for a non diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index b1d3f60525c..afd1168eeef 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -783,10 +783,26 @@ config X86_MCE to disable it. MCE support simply ignores non-MCE processors like the 386 and 486, so nearly everyone can say Y here. +config X86_OLD_MCE + depends on X86_32 && X86_MCE + bool "Use legacy machine check code (will go away)" + default n + select X86_ANCIENT_MCE + ---help--- + Use the old i386 machine check code. This is merely intended for + testing in a transition period. Try this if you run into any machine + check related software problems, but report the problem to + linux-kernel. When in doubt say no. + +config X86_NEW_MCE + depends on X86_MCE + bool + default y if (!X86_OLD_MCE && X86_32) || X86_64 + config X86_MCE_INTEL def_bool y prompt "Intel MCE features" - depends on X86_64 && X86_MCE && X86_LOCAL_APIC + depends on X86_NEW_MCE && X86_LOCAL_APIC ---help--- Additional support for intel specific MCE features such as the thermal monitor. @@ -794,19 +810,36 @@ config X86_MCE_INTEL config X86_MCE_AMD def_bool y prompt "AMD MCE features" - depends on X86_64 && X86_MCE && X86_LOCAL_APIC + depends on X86_NEW_MCE && X86_LOCAL_APIC ---help--- Additional support for AMD specific MCE features such as the DRAM Error Threshold. +config X86_ANCIENT_MCE + def_bool n + depends on X86_32 + prompt "Support for old Pentium 5 / WinChip machine checks" + ---help--- + Include support for machine check handling on old Pentium 5 or WinChip + systems. These typically need to be enabled explicitely on the command + line. + config X86_MCE_THRESHOLD depends on X86_MCE_AMD || X86_MCE_INTEL bool default y +config X86_MCE_INJECT + depends on X86_NEW_MCE + tristate "Machine check injector support" + ---help--- + Provide support for injecting machine checks for testing purposes. + If you don't know what a machine check is and you don't do kernel + QA it is safe to say n. + config X86_MCE_NONFATAL tristate "Check for non-fatal errors on AMD Athlon/Duron / Intel Pentium 4" - depends on X86_32 && X86_MCE + depends on X86_OLD_MCE ---help--- Enabling this feature starts a timer that triggers every 5 seconds which will look at the machine check registers to see if anything happened. @@ -819,11 +852,15 @@ config X86_MCE_NONFATAL config X86_MCE_P4THERMAL bool "check for P4 thermal throttling interrupt." - depends on X86_32 && X86_MCE && (X86_UP_APIC || SMP) + depends on X86_OLD_MCE && X86_MCE && (X86_UP_APIC || SMP) ---help--- Enabling this feature will cause a message to be printed when the P4 enters thermal throttling. +config X86_THERMAL_VECTOR + def_bool y + depends on X86_MCE_P4THERMAL || X86_MCE_INTEL + config VM86 bool "Enable VM86 support" if EMBEDDED default y diff --git a/arch/x86/include/asm/entry_arch.h b/arch/x86/include/asm/entry_arch.h index c2e6bedaf25..b2eb9c06684 100644 --- a/arch/x86/include/asm/entry_arch.h +++ b/arch/x86/include/asm/entry_arch.h @@ -52,8 +52,12 @@ BUILD_INTERRUPT(spurious_interrupt,SPURIOUS_APIC_VECTOR) BUILD_INTERRUPT(perf_counter_interrupt, LOCAL_PERF_VECTOR) #endif -#ifdef CONFIG_X86_MCE_P4THERMAL +#ifdef CONFIG_X86_THERMAL_VECTOR BUILD_INTERRUPT(thermal_interrupt,THERMAL_APIC_VECTOR) #endif +#ifdef CONFIG_X86_MCE_THRESHOLD +BUILD_INTERRUPT(threshold_interrupt,THRESHOLD_APIC_VECTOR) +#endif + #endif diff --git a/arch/x86/include/asm/hardirq.h b/arch/x86/include/asm/hardirq.h index 37555e52f98..922ee7c2969 100644 --- a/arch/x86/include/asm/hardirq.h +++ b/arch/x86/include/asm/hardirq.h @@ -20,7 +20,7 @@ typedef struct { #endif #ifdef CONFIG_X86_MCE unsigned int irq_thermal_count; -# ifdef CONFIG_X86_64 +# ifdef CONFIG_X86_MCE_THRESHOLD unsigned int irq_threshold_count; # endif #endif diff --git a/arch/x86/include/asm/irq_vectors.h b/arch/x86/include/asm/irq_vectors.h index 910b5a3d675..8c46b851296 100644 --- a/arch/x86/include/asm/irq_vectors.h +++ b/arch/x86/include/asm/irq_vectors.h @@ -87,11 +87,11 @@ #define CALL_FUNCTION_VECTOR 0xfc #define CALL_FUNCTION_SINGLE_VECTOR 0xfb #define THERMAL_APIC_VECTOR 0xfa +#define THRESHOLD_APIC_VECTOR 0xf9 #ifdef CONFIG_X86_32 -/* 0xf8 - 0xf9 : free */ +/* 0xf8 : free */ #else -# define THRESHOLD_APIC_VECTOR 0xf9 # define UV_BAU_MESSAGE 0xf8 #endif diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h index 4f8c199584e..ac6e0303bf2 100644 --- a/arch/x86/include/asm/mce.h +++ b/arch/x86/include/asm/mce.h @@ -1,8 +1,6 @@ #ifndef _ASM_X86_MCE_H #define _ASM_X86_MCE_H -#ifdef __x86_64__ - #include <linux/types.h> #include <asm/ioctls.h> @@ -10,21 +8,25 @@ * Machine Check support for x86 */ -#define MCG_CTL_P (1UL<<8) /* MCG_CAP register available */ -#define MCG_EXT_P (1ULL<<9) /* Extended registers available */ -#define MCG_CMCI_P (1ULL<<10) /* CMCI supported */ - -#define MCG_STATUS_RIPV (1UL<<0) /* restart ip valid */ -#define MCG_STATUS_EIPV (1UL<<1) /* ip points to correct instruction */ -#define MCG_STATUS_MCIP (1UL<<2) /* machine check in progress */ - -#define MCI_STATUS_VAL (1UL<<63) /* valid error */ -#define MCI_STATUS_OVER (1UL<<62) /* previous errors lost */ -#define MCI_STATUS_UC (1UL<<61) /* uncorrected error */ -#define MCI_STATUS_EN (1UL<<60) /* error enabled */ -#define MCI_STATUS_MISCV (1UL<<59) /* misc error reg. valid */ -#define MCI_STATUS_ADDRV (1UL<<58) /* addr reg. valid */ -#define MCI_STATUS_PCC (1UL<<57) /* processor context corrupt */ +#define MCG_BANKCNT_MASK 0xff /* Number of Banks */ +#define MCG_CTL_P (1ULL<<8) /* MCG_CAP register available */ +#define MCG_EXT_P (1ULL<<9) /* Extended registers available */ +#define MCG_CMCI_P (1ULL<<10) /* CMCI supported */ +#define MCG_EXT_CNT_MASK 0xff0000 /* Number of Extended registers */ +#define MCG_EXT_CNT_SHIFT 16 +#define MCG_EXT_CNT(c) (((c) & MCG_EXT_CNT_MASK) >> MCG_EXT_CNT_SHIFT) + +#define MCG_STATUS_RIPV (1ULL<<0) /* restart ip valid */ +#define MCG_STATUS_EIPV (1ULL<<1) /* ip points to correct instruction */ +#define MCG_STATUS_MCIP (1ULL<<2) /* machine check in progress */ + +#define MCI_STATUS_VAL (1ULL<<63) /* valid error */ +#define MCI_STATUS_OVER (1ULL<<62) /* previous errors lost */ +#define MCI_STATUS_UC (1ULL<<61) /* uncorrected error */ +#define MCI_STATUS_EN (1ULL<<60) /* error enabled */ +#define MCI_STATUS_MISCV (1ULL<<59) /* misc error reg. valid */ +#define MCI_STATUS_ADDRV (1ULL<<58) /* addr reg. valid */ +#define MCI_STATUS_PCC (1ULL<<57) /* processor context corrupt */ /* Fields are zero when not available */ struct mce { @@ -82,19 +84,15 @@ struct mce_log { #define K8_MCE_THRESHOLD_BANK_5 (MCE_THRESHOLD_BASE + 5 * 9) #define K8_MCE_THRESHOLD_DRAM_ECC (MCE_THRESHOLD_BANK_4 + 0) -#endif /* __x86_64__ */ - #ifdef __KERNEL__ -#ifdef CONFIG_X86_32 extern int mce_disabled; -#else /* CONFIG_X86_32 */ #include <asm/atomic.h> void mce_setup(struct mce *m); void mce_log(struct mce *m); -DECLARE_PER_CPU(struct sys_device, device_mce); +DECLARE_PER_CPU(struct sys_device, mce_dev); extern void (*threshold_cpu_callback)(unsigned long action, unsigned int cpu); /* @@ -123,13 +121,13 @@ void mce_amd_feature_init(struct cpuinfo_x86 *c); static inline void mce_amd_feature_init(struct cpuinfo_x86 *c) { } #endif -extern int mce_available(struct cpuinfo_x86 *c); +int mce_available(struct cpuinfo_x86 *c); void mce_log_therm_throt_event(__u64 status); extern atomic_t mce_entry; -extern void do_machine_check(struct pt_regs *, long); +void do_machine_check(struct pt_regs *, long); typedef DECLARE_BITMAP(mce_banks_t, MAX_NR_BANKS); DECLARE_PER_CPU(mce_banks_t, mce_poll_banks); @@ -139,14 +137,15 @@ enum mcp_flags { MCP_UC = (1 << 1), /* log uncorrected errors */ MCP_DONTLOG = (1 << 2), /* only clear, don't log */ }; -extern void machine_check_poll(enum mcp_flags flags, mce_banks_t *b); +void machine_check_poll(enum mcp_flags flags, mce_banks_t *b); -extern int mce_notify_user(void); +int mce_notify_user(void); -#endif /* !CONFIG_X86_32 */ +DECLARE_PER_CPU(struct mce, injectm); +extern struct file_operations mce_chrdev_ops; #ifdef CONFIG_X86_MCE -extern void mcheck_init(struct cpuinfo_x86 *c); +void mcheck_init(struct cpuinfo_x86 *c); #else #define mcheck_init(c) do { } while (0) #endif diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index ec41fc16c16..c8640469508 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -208,7 +208,14 @@ #define MSR_IA32_THERM_CONTROL 0x0000019a #define MSR_IA32_THERM_INTERRUPT 0x0000019b + +#define THERM_INT_LOW_ENABLE (1 << 0) +#define THERM_INT_HIGH_ENABLE (1 << 1) + #define MSR_IA32_THERM_STATUS 0x0000019c + +#define THERM_STATUS_PROCHOT (1 << 0) + #define MSR_IA32_MISC_ENABLE 0x000001a0 /* MISC_ENABLE bits: architectural */ diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c index b0fd26442c4..ee75d2a9b9c 100644 --- a/arch/x86/kernel/apic/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -897,7 +897,7 @@ void clear_local_APIC(void) } /* lets not touch this if we didn't frob it */ -#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(CONFIG_X86_MCE_INTEL) +#ifdef CONFIG_X86_THERMAL_VECTOR if (maxlvt >= 5) { v = apic_read(APIC_LVTTHMR); apic_write(APIC_LVTTHMR, v | APIC_LVT_MASKED); @@ -2007,7 +2007,7 @@ static int lapic_suspend(struct sys_device *dev, pm_message_t state) apic_pm_state.apic_lvterr = apic_read(APIC_LVTERR); apic_pm_state.apic_tmict = apic_read(APIC_TMICT); apic_pm_state.apic_tdcr = apic_read(APIC_TDCR); -#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(CONFIG_X86_MCE_INTEL) +#ifdef CONFIG_X86_THERMAL_VECTOR if (maxlvt >= 5) apic_pm_state.apic_thmr = apic_read(APIC_LVTTHMR); #endif diff --git a/arch/x86/kernel/apic/nmi.c b/arch/x86/kernel/apic/nmi.c index ce4fbfa315a..c4762276c17 100644 --- a/arch/x86/kernel/apic/nmi.c +++ b/arch/x86/kernel/apic/nmi.c @@ -66,7 +66,7 @@ static inline unsigned int get_nmi_count(int cpu) static inline int mce_in_progress(void) { -#if defined(CONFIG_X86_64) && defined(CONFIG_X86_MCE) +#if defined(CONFIG_X86_NEW_MCE) return atomic_read(&mce_entry) > 0; #endif return 0; diff --git a/arch/x86/kernel/cpu/mcheck/Makefile b/arch/x86/kernel/cpu/mcheck/Makefile index b2f89829bbe..60ee182c6c5 100644 --- a/arch/x86/kernel/cpu/mcheck/Makefile +++ b/arch/x86/kernel/cpu/mcheck/Makefile @@ -1,7 +1,10 @@ -obj-y = mce_$(BITS).o therm_throt.o +obj-y = mce.o therm_throt.o -obj-$(CONFIG_X86_32) += k7.o p4.o p5.o p6.o winchip.o -obj-$(CONFIG_X86_MCE_INTEL) += mce_intel_64.o +obj-$(CONFIG_X86_OLD_MCE) += k7.o p4.o p6.o +obj-$(CONFIG_X86_ANCIENT_MCE) += winchip.o p5.o +obj-$(CONFIG_X86_MCE_P4THERMAL) += mce_intel.o +obj-$(CONFIG_X86_MCE_INTEL) += mce_intel_64.o mce_intel.o obj-$(CONFIG_X86_MCE_AMD) += mce_amd_64.o obj-$(CONFIG_X86_MCE_NONFATAL) += non-fatal.o obj-$(CONFIG_X86_MCE_THRESHOLD) += threshold.o +obj-$(CONFIG_X86_MCE_INJECT) += mce-inject.o diff --git a/arch/x86/kernel/cpu/mcheck/k7.c b/arch/x86/kernel/cpu/mcheck/k7.c index dd3af6e7b39..89e51042415 100644 --- a/arch/x86/kernel/cpu/mcheck/k7.c +++ b/arch/x86/kernel/cpu/mcheck/k7.c @@ -2,11 +2,10 @@ * Athlon specific Machine Check Exception Reporting * (C) Copyright 2002 Dave Jones <davej@redhat.com> */ - -#include <linux/init.h> -#include <linux/types.h> -#include <linux/kernel.h> #include <linux/interrupt.h> +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/init.h> #include <linux/smp.h> #include <asm/processor.h> @@ -15,12 +14,12 @@ #include "mce.h" -/* Machine Check Handler For AMD Athlon/Duron */ +/* Machine Check Handler For AMD Athlon/Duron: */ static void k7_machine_check(struct pt_regs *regs, long error_code) { - int recover = 1; u32 alow, ahigh, high, low; u32 mcgstl, mcgsth; + int recover = 1; int i; rdmsr(MSR_IA32_MCG_STATUS, mcgstl, mcgsth); @@ -32,15 +31,19 @@ static void k7_machine_check(struct pt_regs *regs, long error_code) for (i = 1; i < nr_mce_banks; i++) { rdmsr(MSR_IA32_MC0_STATUS+i*4, low, high); - if (high&(1<<31)) { + if (high & (1<<31)) { char misc[20]; char addr[24]; - misc[0] = addr[0] = '\0'; + + misc[0] = '\0'; + addr[0] = '\0'; + if (high & (1<<29)) recover |= 1; if (high & (1<<25)) recover |= 2; high &= ~(1<<31); + if (high & (1<<27)) { rdmsr(MSR_IA32_MC0_MISC+i*4, alow, ahigh); snprintf(misc, 20, "[%08x%08x]", ahigh, alow); @@ -49,27 +52,31 @@ static void k7_machine_check(struct pt_regs *regs, long error_code) rdmsr(MSR_IA32_MC0_ADDR+i*4, alow, ahigh); snprintf(addr, 24, " at %08x%08x", ahigh, alow); } + printk(KERN_EMERG "CPU %d: Bank %d: %08x%08x%s%s\n", smp_processor_id(), i, high, low, misc, addr); - /* Clear it */ + + /* Clear it: */ wrmsr(MSR_IA32_MC0_STATUS+i*4, 0UL, 0UL); - /* Serialize */ + /* Serialize: */ wmb(); add_taint(TAINT_MACHINE_CHECK); } } - if (recover&2) + if (recover & 2) panic("CPU context corrupt"); - if (recover&1) + if (recover & 1) panic("Unable to continue"); + printk(KERN_EMERG "Attempting to continue.\n"); + mcgstl &= ~(1<<2); wrmsr(MSR_IA32_MCG_STATUS, mcgstl, mcgsth); } -/* AMD K7 machine check is Intel like */ +/* AMD K7 machine check is Intel like: */ void amd_mcheck_init(struct cpuinfo_x86 *c) { u32 l, h; @@ -79,21 +86,26 @@ void amd_mcheck_init(struct cpuinfo_x86 *c) return; machine_check_vector = k7_machine_check; + /* Make sure the vector pointer is visible before we enable MCEs: */ wmb(); printk(KERN_INFO "Intel machine check architecture supported.\n"); + rdmsr(MSR_IA32_MCG_CAP, l, h); if (l & (1<<8)) /* Control register present ? */ wrmsr(MSR_IA32_MCG_CTL, 0xffffffff, 0xffffffff); nr_mce_banks = l & 0xff; - /* Clear status for MC index 0 separately, we don't touch CTL, - * as some K7 Athlons cause spurious MCEs when its enabled. */ + /* + * Clear status for MC index 0 separately, we don't touch CTL, + * as some K7 Athlons cause spurious MCEs when its enabled: + */ if (boot_cpu_data.x86 == 6) { wrmsr(MSR_IA32_MC0_STATUS, 0x0, 0x0); i = 1; } else i = 0; + for (; i < nr_mce_banks; i++) { wrmsr(MSR_IA32_MC0_CTL+4*i, 0xffffffff, 0xffffffff); wrmsr(MSR_IA32_MC0_STATUS+4*i, 0x0, 0x0); diff --git a/arch/x86/kernel/cpu/mcheck/mce-inject.c b/arch/x86/kernel/cpu/mcheck/mce-inject.c new file mode 100644 index 00000000000..7b3a5428396 --- /dev/null +++ b/arch/x86/kernel/cpu/mcheck/mce-inject.c @@ -0,0 +1,127 @@ +/* + * Machine check injection support. + * Copyright 2008 Intel Corporation. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; version 2 + * of the License. + * + * Authors: + * Andi Kleen + * Ying Huang + */ +#include <linux/uaccess.h> +#include <linux/module.h> +#include <linux/timer.h> +#include <linux/kernel.h> +#include <linux/string.h> +#include <linux/fs.h> +#include <linux/smp.h> +#include <asm/mce.h> + +/* Update fake mce registers on current CPU. */ +static void inject_mce(struct mce *m) +{ + struct mce *i = &per_cpu(injectm, m->cpu); + + /* Make sure noone reads partially written injectm */ + i->finished = 0; + mb(); + m->finished = 0; + /* First set the fields after finished */ + i->cpu = m->cpu; + mb(); + /* Now write record in order, finished last (except above) */ + memcpy(i, m, sizeof(struct mce)); + /* Finally activate it */ + mb(); + i->finished = 1; +} + +struct delayed_mce { + struct timer_list timer; + struct mce m; +}; + +/* Inject mce on current CPU */ +static void raise_mce(unsigned long data) +{ + struct delayed_mce *dm = (struct delayed_mce *)data; + struct mce *m = &dm->m; + int cpu = m->cpu; + + inject_mce(m); + if (m->status & MCI_STATUS_UC) { + struct pt_regs regs; + memset(®s, 0, sizeof(struct pt_regs)); + regs.ip = m->ip; + regs.cs = m->cs; + printk(KERN_INFO "Triggering MCE exception on CPU %d\n", cpu); + do_machine_check(®s, 0); + printk(KERN_INFO "MCE exception done on CPU %d\n", cpu); + } else { + mce_banks_t b; + memset(&b, 0xff, sizeof(mce_banks_t)); + printk(KERN_INFO "Starting machine check poll CPU %d\n", cpu); + machine_check_poll(0, &b); + mce_notify_user(); + printk(KERN_INFO "Finished machine check poll on CPU %d\n", + cpu); + } + kfree(dm); +} + +/* Error injection interface */ +static ssize_t mce_write(struct file *filp, const char __user *ubuf, + size_t usize, loff_t *off) +{ + struct delayed_mce *dm; + struct mce m; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + /* + * There are some cases where real MSR reads could slip + * through. + */ + if (!boot_cpu_has(X86_FEATURE_MCE) || !boot_cpu_has(X86_FEATURE_MCA)) + return -EIO; + + if ((unsigned long)usize > sizeof(struct mce)) + usize = sizeof(struct mce); + if (copy_from_user(&m, ubuf, usize)) + return -EFAULT; + + if (m.cpu >= num_possible_cpus() || !cpu_online(m.cpu)) + return -EINVAL; + + dm = kmalloc(sizeof(struct delayed_mce), GFP_KERNEL); + if (!dm) + return -ENOMEM; + + /* + * Need to give user space some time to set everything up, + * so do it a jiffie or two later everywhere. + * Should we use a hrtimer here for better synchronization? + */ + memcpy(&dm->m, &m, sizeof(struct mce)); + setup_timer(&dm->timer, raise_mce, (unsigned long)dm); + dm->timer.expires = jiffies + 2; + add_timer_on(&dm->timer, m.cpu); + return usize; +} + +static int inject_init(void) +{ + printk(KERN_INFO "Machine check injector initialized\n"); + mce_chrdev_ops.write = mce_write; + return 0; +} + +module_init(inject_init); +/* + * Cannot tolerate unloading currently because we cannot + * guarantee all openers of mce_chrdev will get a reference to us. + */ +MODULE_LICENSE("GPL"); diff --git a/arch/x86/kernel/cpu/mcheck/mce_64.c b/arch/x86/kernel/cpu/mcheck/mce.c index 6fb0b359d2a..1d0aa9c4e15 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -1,46 +1,61 @@ /* * Machine check handler. + * * K8 parts Copyright 2002,2003 Andi Kleen, SuSE Labs. * Rest from unknown author(s). * 2004 Andi Kleen. Rewrote most of it. * Copyright 2008 Intel Corporation * Author: Andi Kleen */ - -#include <linux/init.h> -#include <linux/types.h> +#include <linux/thread_info.h> +#include <linux/capability.h> +#include <linux/miscdevice.h> +#include <linux/ratelimit.h> +#include <linux/kallsyms.h> +#include <linux/rcupdate.h> +#include <linux/kobject.h> +#include <linux/uaccess.h> +#include <linux/kdebug.h> #include <linux/kernel.h> -#include <linux/sched.h> -#include <linux/smp_lock.h> +#include <linux/percpu.h> #include <linux/string.h> -#include <linux/rcupdate.h> -#include <linux/kallsyms.h> #include <linux/sysdev.h> -#include <linux/miscdevice.h> -#include <linux/fs.h> -#include <linux/capability.h> -#include <linux/cpu.h> -#include <linux/percpu.h> -#include <linux/poll.h> -#include <linux/thread_info.h> #include <linux/ctype.h> -#include <linux/kmod.h> -#include <linux/kdebug.h> -#include <linux/kobject.h> +#include <linux/sched.h> #include <linux/sysfs.h> -#include <linux/ratelimit.h> +#include <linux/types.h> +#include <linux/init.h> +#include <linux/kmod.h> +#include <linux/poll.h> +#include <linux/cpu.h> +#include <linux/smp.h> +#include <linux/fs.h> + #include <asm/processor.h> -#include <asm/msr.h> -#include <asm/mce.h> -#include <asm/uaccess.h> -#include <asm/smp.h> #include <asm/idle.h> +#include <asm/mce.h> +#include <asm/msr.h> -#define MISC_MCELOG_MINOR 227 +#include "mce.h" -atomic_t mce_entry; +/* Handle unconfigured int18 (should never happen) */ +static void unexpected_machine_check(struct pt_regs *regs, long error_code) +{ + printk(KERN_ERR "CPU#%d: Unexpected int18 (Machine Check).\n", + smp_processor_id()); +} + +/* Call the installed machine check handler for this CPU setup. */ +void (*machine_check_vector)(struct pt_regs *, long error_code) = + unexpected_machine_check; -static int mce_dont_init; +int mce_disabled; + +#ifdef CONFIG_X86_NEW_MCE + +#define MISC_MCELOG_MINOR 227 + +atomic_t mce_entry; /* * Tolerant levels: @@ -49,16 +64,17 @@ static int mce_dont_init; * 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 atomic_t mce_events; +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 char trigger[128]; -static char *trigger_argv[2] = { trigger, NULL }; +static char trigger[128]; +static char *trigger_argv[2] = { trigger, NULL }; + +static unsigned long dont_init_banks; static DECLARE_WAIT_QUEUE_HEAD(mce_wait); @@ -67,6 +83,11 @@ DEFINE_PER_CPU(mce_banks_t, mce_poll_banks) = { [0 ... BITS_TO_LONGS(MAX_NR_BANKS)-1] = ~0UL }; +static inline int skip_bank_init(int i) +{ + return i < BITS_PER_LONG && test_bit(i, &dont_init_banks); +} + /* Do initial initialization of a struct mce */ void mce_setup(struct mce *m) { @@ -75,6 +96,9 @@ void mce_setup(struct mce *m) rdtscll(m->tsc); } +DEFINE_PER_CPU(struct mce, injectm); +EXPORT_PER_CPU_SYMBOL_GPL(injectm); + /* * Lockless MCE logging infrastructure. * This avoids deadlocks on printk locks without having to break locks. Also @@ -89,19 +113,23 @@ static struct mce_log mcelog = { void mce_log(struct mce *mce) { unsigned next, entry; - atomic_inc(&mce_events); + mce->finished = 0; wmb(); for (;;) { entry = rcu_dereference(mcelog.next); for (;;) { - /* When the buffer fills up discard new entries. Assume - that the earlier errors are the more interesting. */ + /* + * When the buffer fills up discard new entries. + * Assume that the earlier errors are the more + * interesting ones: + */ if (entry >= MCE_LOG_LEN) { - set_bit(MCE_OVERFLOW, (unsigned long *)&mcelog.flags); + set_bit(MCE_OVERFLOW, + (unsigned long *)&mcelog.flags); return; } - /* Old left over entry. Skip. */ + /* Old left over entry. Skip: */ if (mcelog.entry[entry].finished) { entry++; continue; @@ -147,15 +175,16 @@ static void print_mce(struct mce *m) "and contact your hardware vendor\n"); } -static void mce_panic(char *msg, struct mce *backup, unsigned long start) +static void mce_panic(char *msg, struct mce *backup, u64 start) { int i; - oops_begin(); + bust_spinlocks(1); + console_verbose(); for (i = 0; i < MCE_LOG_LEN; i++) { - unsigned long tsc = mcelog.entry[i].tsc; + u64 tsc = mcelog.entry[i].tsc; - if (time_before(tsc, start)) + if ((s64)(tsc - start) < 0) continue; print_mce(&mcelog.entry[i]); if (backup && mcelog.entry[i].tsc == backup->tsc) @@ -166,9 +195,52 @@ static void mce_panic(char *msg, struct mce *backup, unsigned long start) panic(msg); } +/* Support code for software error injection */ + +static int msr_to_offset(u32 msr) +{ + unsigned bank = __get_cpu_var(injectm.bank); + if (msr == rip_msr) + return offsetof(struct mce, ip); + if (msr == MSR_IA32_MC0_STATUS + bank*4) + return offsetof(struct mce, status); + if (msr == MSR_IA32_MC0_ADDR + bank*4) + return offsetof(struct mce, addr); + if (msr == MSR_IA32_MC0_MISC + bank*4) + return offsetof(struct mce, misc); + if (msr == MSR_IA32_MCG_STATUS) + return offsetof(struct mce, mcgstatus); + return -1; +} + +/* MSR access wrappers used for error injection */ +static u64 mce_rdmsrl(u32 msr) +{ + u64 v; + if (__get_cpu_var(injectm).finished) { + int offset = msr_to_offset(msr); + if (offset < 0) + return 0; + return *(u64 *)((char *)&__get_cpu_var(injectm) + offset); + } + rdmsrl(msr, v); + return v; +} + +static void mce_wrmsrl(u32 msr, u64 v) +{ + if (__get_cpu_var(injectm).finished) { + int offset = msr_to_offset(msr); + if (offset >= 0) + *(u64 *)((char *)&__get_cpu_var(injectm) + offset) = v; + return; + } + wrmsrl(msr, v); +} + int mce_available(struct cpuinfo_x86 *c) { - if (mce_dont_init) + if (mce_disabled) return 0; return cpu_has(c, X86_FEATURE_MCE) && cpu_has(c, X86_FEATURE_MCA); } @@ -185,7 +257,7 @@ static inline void mce_get_rip(struct mce *m, struct pt_regs *regs) if (rip_msr) { /* Assume the RIP in the MSR is exact. Is this true? */ m->mcgstatus |= MCG_STATUS_EIPV; - rdmsrl(rip_msr, m->ip); + m->ip = mce_rdmsrl(rip_msr); m->cs = 0; } } @@ -203,7 +275,7 @@ void machine_check_poll(enum mcp_flags flags, mce_banks_t *b) mce_setup(&m); - rdmsrl(MSR_IA32_MCG_STATUS, m.mcgstatus); + m.mcgstatus = mce_rdmsrl(MSR_IA32_MCG_STATUS); for (i = 0; i < banks; i++) { if (!bank[i] || !test_bit(i, *b)) continue; @@ -214,7 +286,7 @@ void machine_check_poll(enum mcp_flags flags, mce_banks_t *b) m.tsc = 0; barrier(); - rdmsrl(MSR_IA32_MC0_STATUS + i*4, m.status); + m.status = mce_rdmsrl(MSR_IA32_MC0_STATUS + i*4); if (!(m.status & MCI_STATUS_VAL)) continue; @@ -229,9 +301,9 @@ void machine_check_poll(enum mcp_flags flags, mce_banks_t *b) continue; if (m.status & MCI_STATUS_MISCV) - rdmsrl(MSR_IA32_MC0_MISC + i*4, m.misc); + m.misc = mce_rdmsrl(MSR_IA32_MC0_MISC + i*4); if (m.status & MCI_STATUS_ADDRV) - rdmsrl(MSR_IA32_MC0_ADDR + i*4, m.addr); + m.addr = mce_rdmsrl(MSR_IA32_MC0_ADDR + i*4); if (!(flags & MCP_TIMESTAMP)) m.tsc = 0; @@ -247,14 +319,17 @@ void machine_check_poll(enum mcp_flags flags, mce_banks_t *b) /* * Clear state for this bank. */ - wrmsrl(MSR_IA32_MC0_STATUS+4*i, 0); + mce_wrmsrl(MSR_IA32_MC0_STATUS+4*i, 0); } /* * Don't clear MCG_STATUS here because it's only defined for * exceptions. */ + + sync_core(); } +EXPORT_SYMBOL_GPL(machine_check_poll); /* * The actual machine check handler. This only handles real @@ -264,12 +339,12 @@ void machine_check_poll(enum mcp_flags flags, mce_banks_t *b) * implies that most kernel services cannot be safely used. Don't even * think about putting a printk in there! */ -void do_machine_check(struct pt_regs * regs, long error_code) +void do_machine_check(struct pt_regs *regs, long error_code) { struct mce m, panicm; + int panicm_found = 0; u64 mcestart = 0; int i; - int panicm_found = 0; /* * 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. @@ -286,13 +361,14 @@ void do_machine_check(struct pt_regs * regs, long error_code) if (notify_die(DIE_NMI, "machine check", regs, error_code, 18, SIGKILL) == NOTIFY_STOP) - goto out2; + goto out; if (!banks) - goto out2; + goto out; mce_setup(&m); - rdmsrl(MSR_IA32_MCG_STATUS, m.mcgstatus); + m.mcgstatus = mce_rdmsrl(MSR_IA32_MCG_STATUS); + /* if the restart IP is not valid, we're done for */ if (!(m.mcgstatus & MCG_STATUS_RIPV)) no_way_out = 1; @@ -309,7 +385,7 @@ void do_machine_check(struct pt_regs * regs, long error_code) m.addr = 0; m.bank = i; - rdmsrl(MSR_IA32_MC0_STATUS + i*4, m.status); + m.status = mce_rdmsrl(MSR_IA32_MC0_STATUS + i*4); if ((m.status & MCI_STATUS_VAL) == 0) continue; @@ -349,30 +425,36 @@ void do_machine_check(struct pt_regs * regs, long error_code) } if (m.status & MCI_STATUS_MISCV) - rdmsrl(MSR_IA32_MC0_MISC + i*4, m.misc); + m.misc = mce_rdmsrl(MSR_IA32_MC0_MISC + i*4); if (m.status & MCI_STATUS_ADDRV) - rdmsrl(MSR_IA32_MC0_ADDR + i*4, m.addr); + m.addr = mce_rdmsrl(MSR_IA32_MC0_ADDR + i*4); mce_get_rip(&m, regs); mce_log(&m); - /* Did this bank cause the exception? */ - /* Assume that the bank with uncorrectable errors did it, - and that there is only a single one. */ - if ((m.status & MCI_STATUS_UC) && (m.status & MCI_STATUS_EN)) { + /* + * Did this bank cause the exception? + * + * Assume that the bank with uncorrectable errors did it, + * and that there is only a single one: + */ + if ((m.status & MCI_STATUS_UC) && + (m.status & MCI_STATUS_EN)) { panicm = m; panicm_found = 1; } } - /* If we didn't find an uncorrectable error, pick - the last one (shouldn't happen, just being safe). */ + /* + * If we didn't find an uncorrectable error, pick + * the last one (shouldn't happen, just being safe). + */ if (!panicm_found) panicm = m; /* * If we have decided that we just CAN'T continue, and the user - * has not set tolerant to an insane level, give up and die. + * has not set tolerant to an insane level, give up and die. */ if (no_way_out && tolerant < 3) mce_panic("Machine check", &panicm, mcestart); @@ -414,12 +496,14 @@ void do_machine_check(struct pt_regs * regs, long error_code) /* the last thing we do is clear state */ for (i = 0; i < banks; i++) { if (test_bit(i, toclear)) - wrmsrl(MSR_IA32_MC0_STATUS+4*i, 0); + mce_wrmsrl(MSR_IA32_MC0_STATUS+4*i, 0); } - wrmsrl(MSR_IA32_MCG_STATUS, 0); - out2: + mce_wrmsrl(MSR_IA32_MCG_STATUS, 0); +out: atomic_dec(&mce_entry); + sync_core(); } +EXPORT_SYMBOL_GPL(do_machine_check); #ifdef CONFIG_X86_MCE_INTEL /*** @@ -451,10 +535,9 @@ void mce_log_therm_throt_event(__u64 status) * poller finds an MCE, poll 2x faster. When the poller finds no more * errors, poll 2x slower (up to check_interval seconds). */ - static int check_interval = 5 * 60; /* 5 minutes */ + static DEFINE_PER_CPU(int, next_interval); /* in jiffies */ -static void mcheck_timer(unsigned long); static DEFINE_PER_CPU(struct timer_list, mce_timer); static void mcheck_timer(unsigned long data) @@ -464,20 +547,20 @@ static void mcheck_timer(unsigned long data) WARN_ON(smp_processor_id() != data); - if (mce_available(¤t_cpu_data)) + if (mce_available(¤t_cpu_data)) { machine_check_poll(MCP_TIMESTAMP, &__get_cpu_var(mce_poll_banks)); + } /* * Alert userspace if needed. If we logged an MCE, reduce the * polling interval, otherwise increase the polling interval. */ n = &__get_cpu_var(next_interval); - if (mce_notify_user()) { + if (mce_notify_user()) *n = max(*n/2, HZ/100); - } else { + else *n = min(*n*2, (int)round_jiffies_relative(check_interval*HZ)); - } t->expires = jiffies + *n; add_timer(t); @@ -501,6 +584,7 @@ int mce_notify_user(void) static DEFINE_RATELIMIT_STATE(ratelimit, 60*HZ, 2); clear_thread_flag(TIF_MCE_NOTIFY); + if (test_and_clear_bit(0, ¬ify_user)) { wake_up_interruptible(&mce_wait); @@ -519,39 +603,21 @@ int mce_notify_user(void) } return 0; } - -/* see if the idle task needs to notify userspace */ -static int -mce_idle_callback(struct notifier_block *nfb, unsigned long action, void *junk) -{ - /* IDLE_END should be safe - interrupts are back on */ - if (action == IDLE_END && test_thread_flag(TIF_MCE_NOTIFY)) - mce_notify_user(); - - return NOTIFY_OK; -} - -static struct notifier_block mce_idle_notifier = { - .notifier_call = mce_idle_callback, -}; - -static __init int periodic_mcheck_init(void) -{ - idle_notifier_register(&mce_idle_notifier); - return 0; -} -__initcall(periodic_mcheck_init); +EXPORT_SYMBOL_GPL(mce_notify_user); /* * Initialize Machine Checks for a CPU. */ static int mce_cap_init(void) { - u64 cap; unsigned b; + u64 cap; rdmsrl(MSR_IA32_MCG_CAP, cap); - b = cap & 0xff; + + b = cap & MCG_BANKCNT_MASK; + printk(KERN_INFO "mce: CPU supports %d MCE banks\n", b); + if (b > MAX_NR_BANKS) { printk(KERN_WARNING "MCE: Using only %u machine check banks out of %u\n", @@ -570,17 +636,17 @@ static int mce_cap_init(void) } /* Use accurate RIP reporting if available. */ - if ((cap & (1<<9)) && ((cap >> 16) & 0xff) >= 9) + if ((cap & MCG_EXT_P) && MCG_EXT_CNT(cap) >= 9) rip_msr = MSR_IA32_MCG_EIP; return 0; } -static void mce_init(void *dummy) +static void mce_init(void) { + mce_banks_t all_banks; u64 cap; int i; - mce_banks_t all_banks; /* * Log the machine checks left over from the previous reset. @@ -595,6 +661,8 @@ static void mce_init(void *dummy) wrmsr(MSR_IA32_MCG_CTL, 0xffffffff, 0xffffffff); for (i = 0; i < banks; i++) { + if (skip_bank_init(i)) + continue; wrmsrl(MSR_IA32_MC0_CTL+4*i, bank[i]); wrmsrl(MSR_IA32_MC0_STATUS+4*i, 0); } @@ -605,16 +673,57 @@ static void mce_cpu_quirks(struct cpuinfo_x86 *c) { /* This should be disabled by the BIOS, but isn't always */ if (c->x86_vendor == X86_VENDOR_AMD) { - if (c->x86 == 15 && banks > 4) - /* disable GART TBL walk error reporting, which trips off - incorrectly with the IOMMU & 3ware & Cerberus. */ + if (c->x86 == 15 && banks > 4) { + /* + * disable GART TBL walk error reporting, which + * trips off incorrectly with the IOMMU & 3ware + * & Cerberus: + */ clear_bit(10, (unsigned long *)&bank[4]); - if(c->x86 <= 17 && mce_bootlog < 0) - /* Lots of broken BIOS around that don't clear them - by default and leave crap in there. Don't log. */ + } + if (c->x86 <= 17 && mce_bootlog < 0) { + /* + * Lots of broken BIOS around that don't clear them + * by default and leave crap in there. Don't log: + */ mce_bootlog = 0; + } + /* + * Various K7s with broken bank 0 around. Always disable + * by default. + */ + if (c->x86 == 6) + bank[0] = 0; } + if (c->x86_vendor == X86_VENDOR_INTEL) { + /* + * SDM documents that on family 6 bank 0 should not be written + * because it aliases to another special BIOS controlled + * register. + * But it's not aliased anymore on model 0x1a+ + * Don't ignore bank 0 completely because there could be a + * valid event later, merely don't write CTL0. + */ + + if (c->x86 == 6 && c->x86_model < 0x1A) + __set_bit(0, &dont_init_banks); + } +} + +static void __cpuinit mce_ancient_init(struct cpuinfo_x86 *c) +{ + if (c->x86 != 5) + return; + switch (c->x86_vendor) { + case X86_VENDOR_INTEL: + if (mce_p5_enabled()) + intel_p5_mcheck_init(c); + break; + case X86_VENDOR_CENTAUR: + winchip_mcheck_init(c); + break; + } } static void mce_cpu_features(struct cpuinfo_x86 *c) @@ -646,20 +755,27 @@ static void mce_init_timer(void) /* * Called for each booted CPU to set up machine checks. - * Must be called with preempt off. + * Must be called with preempt off: */ void __cpuinit mcheck_init(struct cpuinfo_x86 *c) { + if (mce_disabled) + return; + + mce_ancient_init(c); + if (!mce_available(c)) return; if (mce_cap_init() < 0) { - mce_dont_init = 1; + mce_disabled = 1; return; } mce_cpu_quirks(c); - mce_init(NULL); + machine_check_vector = do_machine_check; + + mce_init(); mce_cpu_features(c); mce_init_timer(); } @@ -669,17 +785,16 @@ void __cpuinit mcheck_init(struct cpuinfo_x86 *c) */ static DEFINE_SPINLOCK(mce_state_lock); -static int open_count; /* #times opened */ -static int open_exclu; /* already open exclusive? */ +static int open_count; /* #times opened */ +static int open_exclu; /* already open exclusive? */ static int mce_open(struct inode *inode, struct file *file) { - lock_kernel(); spin_lock(&mce_state_lock); if (open_exclu || (open_count && (file->f_flags & O_EXCL))) { spin_unlock(&mce_state_lock); - unlock_kernel(); + return -EBUSY; } @@ -688,7 +803,6 @@ static int mce_open(struct inode *inode, struct file *file) open_count++; spin_unlock(&mce_state_lock); - unlock_kernel(); return nonseekable_open(inode, file); } @@ -712,13 +826,14 @@ static void collect_tscs(void *data) rdtscll(cpu_tsc[smp_processor_id()]); } +static DEFINE_MUTEX(mce_read_mutex); + static ssize_t mce_read(struct file *filp, char __user *ubuf, size_t usize, loff_t *off) { + char __user *buf = ubuf; unsigned long *cpu_tsc; - static DEFINE_MUTEX(mce_read_mutex); unsigned prev, next; - char __user *buf = ubuf; int i, err; cpu_tsc = kmalloc(nr_cpu_ids * sizeof(long), GFP_KERNEL); @@ -732,6 +847,7 @@ static ssize_t mce_read(struct file *filp, char __user *ubuf, size_t usize, if (*off != 0 || usize < MCE_LOG_LEN*sizeof(struct mce)) { mutex_unlock(&mce_read_mutex); kfree(cpu_tsc); + return -EINVAL; } @@ -770,6 +886,7 @@ timeout: * synchronize. */ on_each_cpu(collect_tscs, cpu_tsc, 1); + for (i = next; i < MCE_LOG_LEN; i++) { if (mcelog.entry[i].finished && mcelog.entry[i].tsc < cpu_tsc[mcelog.entry[i].cpu]) { @@ -782,6 +899,7 @@ timeout: } mutex_unlock(&mce_read_mutex); kfree(cpu_tsc); + return err ? -EFAULT : buf - ubuf; } @@ -799,6 +917,7 @@ static long mce_ioctl(struct file *f, unsigned int cmd, unsigned long arg) if (!capable(CAP_SYS_ADMIN)) return -EPERM; + switch (cmd) { case MCE_GET_RECORD_LEN: return put_user(sizeof(struct mce), p); @@ -810,6 +929,7 @@ static long mce_ioctl(struct file *f, unsigned int cmd, unsigned long arg) do { flags = mcelog.flags; } while (cmpxchg(&mcelog.flags, flags, 0) != flags); + return put_user(flags, p); } default: @@ -817,13 +937,15 @@ static long mce_ioctl(struct file *f, unsigned int cmd, unsigned long arg) } } -static const struct file_operations mce_chrdev_ops = { - .open = mce_open, - .release = mce_release, - .read = mce_read, - .poll = mce_poll, - .unlocked_ioctl = mce_ioctl, +/* Modified in mce-inject.c, so not static or const */ +struct file_operations mce_chrdev_ops = { + .open = mce_open, + .release = mce_release, + .read = mce_read, + .poll = mce_poll, + .unlocked_ioctl = mce_ioctl, }; +EXPORT_SYMBOL_GPL(mce_chrdev_ops); static struct miscdevice mce_log_device = { MISC_MCELOG_MINOR, @@ -832,33 +954,31 @@ static struct miscdevice mce_log_device = { }; /* - * Old style boot options parsing. Only for compatibility. + * mce=off disables machine check + * mce=TOLERANCELEVEL (number, see above) + * mce=bootlog Log MCEs from before booting. Disabled by default on AMD. + * mce=nobootlog Don't log MCEs from before booting. */ -static int __init mcheck_disable(char *str) -{ - mce_dont_init = 1; - return 1; -} - -/* mce=off disables machine check. - mce=TOLERANCELEVEL (number, see above) - mce=bootlog Log MCEs from before booting. Disabled by default on AMD. - mce=nobootlog Don't log MCEs from before booting. */ static int __init mcheck_enable(char *str) { + if (*str == 0) + enable_p5_mce(); + if (*str == '=') + str++; if (!strcmp(str, "off")) - mce_dont_init = 1; - else if (!strcmp(str, "bootlog") || !strcmp(str,"nobootlog")) - mce_bootlog = str[0] == 'b'; + mce_disabled = 1; + else if (!strcmp(str, "bootlog") || !strcmp(str, "nobootlog")) + mce_bootlog = (str[0] == 'b'); else if (isdigit(str[0])) get_option(&str, &tolerant); - else - printk("mce= argument %s ignored. Please use /sys", str); + else { + printk(KERN_INFO "mce argument %s ignored. Please use /sys\n", + str); + return 0; + } return 1; } - -__setup("nomce", mcheck_disable); -__setup("mce=", mcheck_enable); +__setup("mce", mcheck_enable); /* * Sysfs support @@ -872,8 +992,10 @@ static int mce_disable(void) { int i; - for (i = 0; i < banks; i++) - wrmsrl(MSR_IA32_MC0_CTL + i*4, 0); + for (i = 0; i < banks; i++) { + if (!skip_bank_init(i)) + wrmsrl(MSR_IA32_MC0_CTL + i*4, 0); + } return 0; } @@ -887,13 +1009,16 @@ static int mce_shutdown(struct sys_device *dev) return mce_disable(); } -/* On resume clear all MCE state. Don't want to see leftovers from the BIOS. - Only one CPU is active at this time, the others get readded later using - CPU hotplug. */ +/* + * On resume clear all MCE state. Don't want to see leftovers from the BIOS. + * Only one CPU is active at this time, the others get re-added later using + * CPU hotplug: + */ static int mce_resume(struct sys_device *dev) { - mce_init(NULL); + mce_init(); mce_cpu_features(¤t_cpu_data); + return 0; } @@ -901,7 +1026,7 @@ static void mce_cpu_restart(void *data) { del_timer_sync(&__get_cpu_var(mce_timer)); if (mce_available(¤t_cpu_data)) - mce_init(NULL); + mce_init(); mce_init_timer(); } @@ -912,33 +1037,16 @@ static void mce_restart(void) } static struct sysdev_class mce_sysclass = { - .suspend = mce_suspend, - .shutdown = mce_shutdown, - .resume = mce_resume, - .name = "machinecheck", + .suspend = mce_suspend, + .shutdown = mce_shutdown, + .resume = mce_resume, + .name = "machinecheck", }; -DEFINE_PER_CPU(struct sys_device, device_mce); -void (*threshold_cpu_callback)(unsigned long action, unsigned int cpu) __cpuinitdata; - -/* Why are there no generic functions for this? */ -#define ACCESSOR(name, var, start) \ - static ssize_t show_ ## name(struct sys_device *s, \ - struct sysdev_attribute *attr, \ - char *buf) { \ - return sprintf(buf, "%lx\n", (unsigned long)var); \ - } \ - static ssize_t set_ ## name(struct sys_device *s, \ - struct sysdev_attribute *attr, \ - const char *buf, size_t siz) { \ - char *end; \ - unsigned long new = simple_strtoul(buf, &end, 0); \ - if (end == buf) return -EINVAL; \ - var = new; \ - start; \ - return end-buf; \ - } \ - static SYSDEV_ATTR(name, 0644, show_ ## name, set_ ## name); +DEFINE_PER_CPU(struct sys_device, mce_dev); + +__cpuinitdata +void (*threshold_cpu_callback)(unsigned long action, unsigned int cpu); static struct sysdev_attribute *bank_attrs; @@ -946,23 +1054,26 @@ static ssize_t show_bank(struct sys_device *s, struct sysdev_attribute *attr, char *buf) { u64 b = bank[attr - bank_attrs]; + return sprintf(buf, "%llx\n", b); } static ssize_t set_bank(struct sys_device *s, struct sysdev_attribute *attr, - const char *buf, size_t siz) + const char *buf, size_t size) { - char *end; - u64 new = simple_strtoull(buf, &end, 0); - if (end == buf) + u64 new; + + if (strict_strtoull(buf, 0, &new) < 0) return -EINVAL; + bank[attr - bank_attrs] = new; mce_restart(); - return end-buf; + + return size; } -static ssize_t show_trigger(struct sys_device *s, struct sysdev_attribute *attr, - char *buf) +static ssize_t +show_trigger(struct sys_device *s, struct sysdev_attribute *attr, char *buf) { strcpy(buf, trigger); strcat(buf, "\n"); @@ -970,29 +1081,48 @@ static ssize_t show_trigger(struct sys_device *s, struct sysdev_attribute *attr, } static ssize_t set_trigger(struct sys_device *s, struct sysdev_attribute *attr, - const char *buf,size_t siz) + const char *buf, size_t siz) { char *p; int len; + strncpy(trigger, buf, sizeof(trigger)); trigger[sizeof(trigger)-1] = 0; len = strlen(trigger); p = strchr(trigger, '\n'); - if (*p) *p = 0; + + if (*p) + *p = 0; + return len; } +static ssize_t store_int_with_restart(struct sys_device *s, + struct sysdev_attribute *attr, + const char *buf, size_t size) +{ + ssize_t ret = sysdev_store_int(s, attr, buf, size); + mce_restart(); + return ret; +} + static SYSDEV_ATTR(trigger, 0644, show_trigger, set_trigger); static SYSDEV_INT_ATTR(tolerant, 0644, tolerant); -ACCESSOR(check_interval,check_interval,mce_restart()) -static struct sysdev_attribute *mce_attributes[] = { - &attr_tolerant.attr, &attr_check_interval, &attr_trigger, + +static struct sysdev_ext_attribute attr_check_interval = { + _SYSDEV_ATTR(check_interval, 0644, sysdev_show_int, + store_int_with_restart), + &check_interval +}; + +static struct sysdev_attribute *mce_attrs[] = { + &attr_tolerant.attr, &attr_check_interval.attr, &attr_trigger, NULL }; -static cpumask_var_t mce_device_initialized; +static cpumask_var_t mce_dev_initialized; -/* Per cpu sysdev init. All of the cpus still share the same ctl bank */ +/* Per cpu sysdev init. All of the cpus still share the same ctrl bank: */ static __cpuinit int mce_create_device(unsigned int cpu) { int err; @@ -1001,40 +1131,36 @@ static __cpuinit int mce_create_device(unsigned int cpu) if (!mce_available(&boot_cpu_data)) return -EIO; - memset(&per_cpu(device_mce, cpu).kobj, 0, sizeof(struct kobject)); - per_cpu(device_mce,cpu).id = cpu; - per_cpu(device_mce,cpu).cls = &mce_sysclass; + memset(&per_cpu(mce_dev, cpu).kobj, 0, sizeof(struct kobject)); + per_cpu(mce_dev, cpu).id = cpu; + per_cpu(mce_dev, cpu).cls = &mce_sysclass; - err = sysdev_register(&per_cpu(device_mce,cpu)); + err = sysdev_register(&per_cpu(mce_dev, cpu)); if (err) return err; - for (i = 0; mce_attributes[i]; i++) { - err = sysdev_create_file(&per_cpu(device_mce,cpu), - mce_attributes[i]); + for (i = 0; mce_attrs[i]; i++) { + err = sysdev_create_file(&per_cpu(mce_dev, cpu), mce_attrs[i]); if (err) goto error; } for (i = 0; i < banks; i++) { - err = sysdev_create_file(&per_cpu(device_mce, cpu), + err = sysdev_create_file(&per_cpu(mce_dev, cpu), &bank_attrs[i]); if (err) goto error2; } - cpumask_set_cpu(cpu, mce_device_initialized); + cpumask_set_cpu(cpu, mce_dev_initialized); return 0; error2: - while (--i >= 0) { - sysdev_remove_file(&per_cpu(device_mce, cpu), - &bank_attrs[i]); - } + while (--i >= 0) + sysdev_remove_file(&per_cpu(mce_dev, cpu), &bank_attrs[i]); error: - while (--i >= 0) { - sysdev_remove_file(&per_cpu(device_mce,cpu), - mce_attributes[i]); - } - sysdev_unregister(&per_cpu(device_mce,cpu)); + while (--i >= 0) + sysdev_remove_file(&per_cpu(mce_dev, cpu), mce_attrs[i]); + + sysdev_unregister(&per_cpu(mce_dev, cpu)); return err; } @@ -1043,49 +1169,54 @@ static __cpuinit void mce_remove_device(unsigned int cpu) { int i; - if (!cpumask_test_cpu(cpu, mce_device_initialized)) + if (!cpumask_test_cpu(cpu, mce_dev_initialized)) return; - for (i = 0; mce_attributes[i]; i++) - sysdev_remove_file(&per_cpu(device_mce,cpu), - mce_attributes[i]); + for (i = 0; mce_attrs[i]; i++) + sysdev_remove_file(&per_cpu(mce_dev, cpu), mce_attrs[i]); + for (i = 0; i < banks; i++) - sysdev_remove_file(&per_cpu(device_mce, cpu), - &bank_attrs[i]); - sysdev_unregister(&per_cpu(device_mce,cpu)); - cpumask_clear_cpu(cpu, mce_device_initialized); + sysdev_remove_file(&per_cpu(mce_dev, cpu), &bank_attrs[i]); + + sysdev_unregister(&per_cpu(mce_dev, cpu)); + cpumask_clear_cpu(cpu, mce_dev_initialized); } /* Make sure there are no machine checks on offlined CPUs. */ static void mce_disable_cpu(void *h) { - int i; unsigned long action = *(unsigned long *)h; + int i; if (!mce_available(¤t_cpu_data)) return; if (!(action & CPU_TASKS_FROZEN)) cmci_clear(); - for (i = 0; i < banks; i++) - wrmsrl(MSR_IA32_MC0_CTL + i*4, 0); + for (i = 0; i < banks; i++) { + if (!skip_bank_init(i)) + wrmsrl(MSR_IA32_MC0_CTL + i*4, 0); + } } static void mce_reenable_cpu(void *h) { - int i; unsigned long action = *(unsigned long *)h; + int i; if (!mce_available(¤t_cpu_data)) return; + if (!(action & CPU_TASKS_FROZEN)) cmci_reenable(); - for (i = 0; i < banks; i++) - wrmsrl(MSR_IA32_MC0_CTL + i*4, bank[i]); + for (i = 0; i < banks; i++) { + if (!skip_bank_init(i)) + wrmsrl(MSR_IA32_MC0_CTL + i*4, bank[i]); + } } /* Get notified when a cpu comes on/off. Be hotplug friendly. */ -static int __cpuinit mce_cpu_callback(struct notifier_block *nfb, - unsigned long action, void *hcpu) +static int __cpuinit +mce_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { unsigned int cpu = (unsigned long)hcpu; struct timer_list *t = &per_cpu(mce_timer, cpu); @@ -1138,12 +1269,14 @@ static __init int mce_init_banks(void) for (i = 0; i < banks; i++) { struct sysdev_attribute *a = &bank_attrs[i]; - a->attr.name = kasprintf(GFP_KERNEL, "bank%d", i); + + a->attr.name = kasprintf(GFP_KERNEL, "bank%d", i); if (!a->attr.name) goto nomem; - a->attr.mode = 0644; - a->show = show_bank; - a->store = set_bank; + + a->attr.mode = 0644; + a->show = show_bank; + a->store = set_bank; } return 0; @@ -1152,6 +1285,7 @@ nomem: kfree(bank_attrs[i].attr.name); kfree(bank_attrs); bank_attrs = NULL; + return -ENOMEM; } @@ -1163,7 +1297,7 @@ static __init int mce_init_device(void) if (!mce_available(&boot_cpu_data)) return -EIO; - alloc_cpumask_var(&mce_device_initialized, GFP_KERNEL); + alloc_cpumask_var(&mce_dev_initialized, GFP_KERNEL); err = mce_init_banks(); if (err) @@ -1181,7 +1315,64 @@ static __init int mce_init_device(void) register_hotcpu_notifier(&mce_cpu_notifier); misc_register(&mce_log_device); + return err; } device_initcall(mce_init_device); + +#else /* CONFIG_X86_OLD_MCE: */ + +int nr_mce_banks; +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) + return; + + switch (c->x86_vendor) { + case X86_VENDOR_AMD: + amd_mcheck_init(c); + break; + + case X86_VENDOR_INTEL: + if (c->x86 == 5) + intel_p5_mcheck_init(c); + if (c->x86 == 6) + intel_p6_mcheck_init(c); + if (c->x86 == 15) + intel_p4_mcheck_init(c); + break; + + case X86_VENDOR_CENTAUR: + if (c->x86 == 5) + winchip_mcheck_init(c); + break; + + default: + break; + } + printk(KERN_INFO "mce: CPU supports %d MCE banks\n", nr_mce_banks); +} + +static int __init mcheck_enable(char *str) +{ + mce_disabled = -1; + return 1; +} + +__setup("mce", mcheck_enable); + +#endif /* CONFIG_X86_OLD_MCE */ + +/* + * Old style boot options parsing. Only for compatibility. + */ +static int __init mcheck_disable(char *str) +{ + mce_disabled = 1; + return 1; +} +__setup("nomce", mcheck_disable); diff --git a/arch/x86/kernel/cpu/mcheck/mce.h b/arch/x86/kernel/cpu/mcheck/mce.h index ae9f628838f..84a552b458c 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.h +++ b/arch/x86/kernel/cpu/mcheck/mce.h @@ -1,14 +1,38 @@ #include <linux/init.h> #include <asm/mce.h> +#ifdef CONFIG_X86_OLD_MCE void amd_mcheck_init(struct cpuinfo_x86 *c); void intel_p4_mcheck_init(struct cpuinfo_x86 *c); -void intel_p5_mcheck_init(struct cpuinfo_x86 *c); void intel_p6_mcheck_init(struct cpuinfo_x86 *c); +#endif + +#ifdef CONFIG_X86_ANCIENT_MCE +void intel_p5_mcheck_init(struct cpuinfo_x86 *c); void winchip_mcheck_init(struct cpuinfo_x86 *c); +extern int mce_p5_enable; +static inline int mce_p5_enabled(void) { return mce_p5_enable; } +static inline void enable_p5_mce(void) { mce_p5_enable = 1; } +#else +static inline void intel_p5_mcheck_init(struct cpuinfo_x86 *c) {} +static inline void winchip_mcheck_init(struct cpuinfo_x86 *c) {} +static inline int mce_p5_enabled(void) { return 0; } +static inline void enable_p5_mce(void) { } +#endif /* Call the installed machine check handler for this CPU setup. */ extern void (*machine_check_vector)(struct pt_regs *, long error_code); +#ifdef CONFIG_X86_OLD_MCE + extern int nr_mce_banks; +void intel_set_thermal_handler(void); + +#else + +static inline void intel_set_thermal_handler(void) { } + +#endif + +void intel_init_thermal(struct cpuinfo_x86 *c); diff --git a/arch/x86/kernel/cpu/mcheck/mce_32.c b/arch/x86/kernel/cpu/mcheck/mce_32.c deleted file mode 100644 index 3552119b091..00000000000 --- a/arch/x86/kernel/cpu/mcheck/mce_32.c +++ /dev/null @@ -1,76 +0,0 @@ -/* - * mce.c - x86 Machine Check Exception Reporting - * (c) 2002 Alan Cox <alan@lxorguk.ukuu.org.uk>, Dave Jones <davej@redhat.com> - */ - -#include <linux/init.h> -#include <linux/types.h> -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/smp.h> -#include <linux/thread_info.h> - -#include <asm/processor.h> -#include <asm/system.h> -#include <asm/mce.h> - -#include "mce.h" - -int mce_disabled; -int nr_mce_banks; - -EXPORT_SYMBOL_GPL(nr_mce_banks); /* non-fatal.o */ - -/* Handle unconfigured int18 (should never happen) */ -static void unexpected_machine_check(struct pt_regs *regs, long error_code) -{ - printk(KERN_ERR "CPU#%d: Unexpected int18 (Machine Check).\n", smp_processor_id()); -} - -/* Call the installed machine check handler for this CPU setup. */ -void (*machine_check_vector)(struct pt_regs *, long error_code) = unexpected_machine_check; - -/* This has to be run for each processor */ -void mcheck_init(struct cpuinfo_x86 *c) -{ - if (mce_disabled == 1) - return; - - switch (c->x86_vendor) { - case X86_VENDOR_AMD: - amd_mcheck_init(c); - break; - - case X86_VENDOR_INTEL: - if (c->x86 == 5) - intel_p5_mcheck_init(c); - if (c->x86 == 6) - intel_p6_mcheck_init(c); - if (c->x86 == 15) - intel_p4_mcheck_init(c); - break; - - case X86_VENDOR_CENTAUR: - if (c->x86 == 5) - winchip_mcheck_init(c); - break; - - default: - break; - } -} - -static int __init mcheck_disable(char *str) -{ - mce_disabled = 1; - return 1; -} - -static int __init mcheck_enable(char *str) -{ - mce_disabled = -1; - return 1; -} - -__setup("nomce", mcheck_disable); -__setup("mce", mcheck_enable); diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c index 56dde9c4bc9..ddae21620bd 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c @@ -13,22 +13,22 @@ * * All MC4_MISCi registers are shared between multi-cores */ - -#include <linux/cpu.h> -#include <linux/errno.h> -#include <linux/init.h> #include <linux/interrupt.h> -#include <linux/kobject.h> #include <linux/notifier.h> -#include <linux/sched.h> -#include <linux/smp.h> +#include <linux/kobject.h> +#include <linux/percpu.h> #include <linux/sysdev.h> +#include <linux/errno.h> +#include <linux/sched.h> #include <linux/sysfs.h> +#include <linux/init.h> +#include <linux/cpu.h> +#include <linux/smp.h> + #include <asm/apic.h> +#include <asm/idle.h> #include <asm/mce.h> #include <asm/msr.h> -#include <asm/percpu.h> -#include <asm/idle.h> #define PFX "mce_threshold: " #define VERSION "version 1.1.1" @@ -48,26 +48,26 @@ #define MCG_XBLK_ADDR 0xC0000400 struct threshold_block { - unsigned int block; - unsigned int bank; - unsigned int cpu; - u32 address; - u16 interrupt_enable; - u16 threshold_limit; - struct kobject kobj; - struct list_head miscj; + unsigned int block; + unsigned int bank; + unsigned int cpu; + u32 address; + u16 interrupt_enable; + u16 threshold_limit; + struct kobject kobj; + struct list_head miscj; }; /* defaults used early on boot */ static struct threshold_block threshold_defaults = { - .interrupt_enable = 0, - .threshold_limit = THRESHOLD_MAX, + .interrupt_enable = 0, + .threshold_limit = THRESHOLD_MAX, }; struct threshold_bank { - struct kobject *kobj; - struct threshold_block *blocks; - cpumask_var_t cpus; + struct kobject *kobj; + struct threshold_block *blocks; + cpumask_var_t cpus; }; static DEFINE_PER_CPU(struct threshold_bank *, threshold_banks[NR_BANKS]); @@ -86,9 +86,9 @@ static void amd_threshold_interrupt(void); */ struct thresh_restart { - struct threshold_block *b; - int reset; - u16 old_limit; + struct threshold_block *b; + int reset; + u16 old_limit; }; /* must be called with correct cpu affinity */ @@ -110,6 +110,7 @@ static void threshold_restart_bank(void *_tr) } else if (tr->old_limit) { /* change limit w/o reset */ int new_count = (mci_misc_hi & THRESHOLD_MAX) + (tr->old_limit - tr->b->threshold_limit); + mci_misc_hi = (mci_misc_hi & ~MASK_ERR_COUNT_HI) | (new_count & THRESHOLD_MAX); } @@ -125,11 +126,11 @@ static void threshold_restart_bank(void *_tr) /* cpu init entry point, called from mce.c with preempt off */ void mce_amd_feature_init(struct cpuinfo_x86 *c) { - unsigned int bank, block; unsigned int cpu = smp_processor_id(); - u8 lvt_off; u32 low = 0, high = 0, address = 0; + unsigned int bank, block; struct thresh_restart tr; + u8 lvt_off; for (bank = 0; bank < NR_BANKS; ++bank) { for (block = 0; block < NR_BLOCKS; ++block) { @@ -140,8 +141,7 @@ void mce_amd_feature_init(struct cpuinfo_x86 *c) if (!address) break; address += MCG_XBLK_ADDR; - } - else + } else ++address; if (rdmsr_safe(address, &low, &high)) @@ -193,9 +193,9 @@ void mce_amd_feature_init(struct cpuinfo_x86 *c) */ static void amd_threshold_interrupt(void) { + u32 low = 0, high = 0, address = 0; unsigned int bank, block; struct mce m; - u32 low = 0, high = 0, address = 0; mce_setup(&m); @@ -204,16 +204,16 @@ static void amd_threshold_interrupt(void) if (!(per_cpu(bank_map, m.cpu) & (1 << bank))) continue; for (block = 0; block < NR_BLOCKS; ++block) { - if (block == 0) + if (block == 0) { address = MSR_IA32_MC0_MISC + bank * 4; - else if (block == 1) { + } else if (block == 1) { address = (low & MASK_BLKPTR_LO) >> 21; if (!address) break; address += MCG_XBLK_ADDR; - } - else + } else { ++address; + } if (rdmsr_safe(address, &low, &high)) break; @@ -229,8 +229,10 @@ static void amd_threshold_interrupt(void) (high & MASK_LOCKED_HI)) continue; - /* Log the machine check that caused the threshold - event. */ + /* + * Log the machine check that caused the threshold + * event. + */ machine_check_poll(MCP_TIMESTAMP, &__get_cpu_var(mce_poll_banks)); @@ -254,48 +256,52 @@ static void amd_threshold_interrupt(void) struct threshold_attr { struct attribute attr; - ssize_t(*show) (struct threshold_block *, char *); - ssize_t(*store) (struct threshold_block *, const char *, size_t count); + ssize_t (*show) (struct threshold_block *, char *); + ssize_t (*store) (struct threshold_block *, const char *, size_t count); }; -#define SHOW_FIELDS(name) \ -static ssize_t show_ ## name(struct threshold_block * b, char *buf) \ -{ \ - return sprintf(buf, "%lx\n", (unsigned long) b->name); \ +#define SHOW_FIELDS(name) \ +static ssize_t show_ ## name(struct threshold_block *b, char *buf) \ +{ \ + return sprintf(buf, "%lx\n", (unsigned long) b->name); \ } SHOW_FIELDS(interrupt_enable) SHOW_FIELDS(threshold_limit) -static ssize_t store_interrupt_enable(struct threshold_block *b, - const char *buf, size_t count) +static ssize_t +store_interrupt_enable(struct threshold_block *b, const char *buf, size_t size) { - char *end; struct thresh_restart tr; - unsigned long new = simple_strtoul(buf, &end, 0); - if (end == buf) + unsigned long new; + + if (strict_strtoul(buf, 0, &new) < 0) return -EINVAL; + b->interrupt_enable = !!new; - tr.b = b; - tr.reset = 0; - tr.old_limit = 0; + tr.b = b; + tr.reset = 0; + tr.old_limit = 0; + smp_call_function_single(b->cpu, threshold_restart_bank, &tr, 1); - return end - buf; + return size; } -static ssize_t store_threshold_limit(struct threshold_block *b, - const char *buf, size_t count) +static ssize_t +store_threshold_limit(struct threshold_block *b, const char *buf, size_t size) { - char *end; struct thresh_restart tr; - unsigned long new = simple_strtoul(buf, &end, 0); - if (end == buf) + unsigned long new; + + if (strict_strtoul(buf, 0, &new) < 0) return -EINVAL; + if (new > THRESHOLD_MAX) new = THRESHOLD_MAX; if (new < 1) new = 1; + tr.old_limit = b->threshold_limit; b->threshold_limit = new; tr.b = b; @@ -303,12 +309,12 @@ static ssize_t store_threshold_limit(struct threshold_block *b, smp_call_function_single(b->cpu, threshold_restart_bank, &tr, 1); - return end - buf; + return size; } struct threshold_block_cross_cpu { - struct threshold_block *tb; - long retval; + struct threshold_block *tb; + long retval; }; static void local_error_count_handler(void *_tbcc) @@ -338,16 +344,13 @@ static ssize_t store_error_count(struct threshold_block *b, return 1; } -#define THRESHOLD_ATTR(_name,_mode,_show,_store) { \ - .attr = {.name = __stringify(_name), .mode = _mode }, \ - .show = _show, \ - .store = _store, \ +#define RW_ATTR(val) \ +static struct threshold_attr val = { \ + .attr = {.name = __stringify(val), .mode = 0644 }, \ + .show = show_## val, \ + .store = store_## val, \ }; -#define RW_ATTR(name) \ -static struct threshold_attr name = \ - THRESHOLD_ATTR(name, 0644, show_## name, store_## name) - RW_ATTR(interrupt_enable); RW_ATTR(threshold_limit); RW_ATTR(error_count); @@ -359,15 +362,17 @@ static struct attribute *default_attrs[] = { NULL }; -#define to_block(k) container_of(k, struct threshold_block, kobj) -#define to_attr(a) container_of(a, struct threshold_attr, attr) +#define to_block(k) container_of(k, struct threshold_block, kobj) +#define to_attr(a) container_of(a, struct threshold_attr, attr) static ssize_t show(struct kobject *kobj, struct attribute *attr, char *buf) { struct threshold_block *b = to_block(kobj); struct threshold_attr *a = to_attr(attr); ssize_t ret; + ret = a->show ? a->show(b, buf) : -EIO; + return ret; } @@ -377,18 +382,20 @@ static ssize_t store(struct kobject *kobj, struct attribute *attr, struct threshold_block *b = to_block(kobj); struct threshold_attr *a = to_attr(attr); ssize_t ret; + ret = a->store ? a->store(b, buf, count) : -EIO; + return ret; } static struct sysfs_ops threshold_ops = { - .show = show, - .store = store, + .show = show, + .store = store, }; static struct kobj_type threshold_ktype = { - .sysfs_ops = &threshold_ops, - .default_attrs = default_attrs, + .sysfs_ops = &threshold_ops, + .default_attrs = default_attrs, }; static __cpuinit int allocate_threshold_blocks(unsigned int cpu, @@ -396,9 +403,9 @@ static __cpuinit int allocate_threshold_blocks(unsigned int cpu, unsigned int block, u32 address) { - int err; - u32 low, high; struct threshold_block *b = NULL; + u32 low, high; + int err; if ((bank >= NR_BANKS) || (block >= NR_BLOCKS)) return 0; @@ -421,20 +428,21 @@ static __cpuinit int allocate_threshold_blocks(unsigned int cpu, if (!b) return -ENOMEM; - b->block = block; - b->bank = bank; - b->cpu = cpu; - b->address = address; - b->interrupt_enable = 0; - b->threshold_limit = THRESHOLD_MAX; + b->block = block; + b->bank = bank; + b->cpu = cpu; + b->address = address; + b->interrupt_enable = 0; + b->threshold_limit = THRESHOLD_MAX; INIT_LIST_HEAD(&b->miscj); - if (per_cpu(threshold_banks, cpu)[bank]->blocks) + if (per_cpu(threshold_banks, cpu)[bank]->blocks) { list_add(&b->miscj, &per_cpu(threshold_banks, cpu)[bank]->blocks->miscj); - else + } else { per_cpu(threshold_banks, cpu)[bank]->blocks = b; + } err = kobject_init_and_add(&b->kobj, &threshold_ktype, per_cpu(threshold_banks, cpu)[bank]->kobj, @@ -447,8 +455,9 @@ recurse: if (!address) return 0; address += MCG_XBLK_ADDR; - } else + } else { ++address; + } err = allocate_threshold_blocks(cpu, bank, ++block, address); if (err) @@ -500,13 +509,14 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank) if (!b) goto out; - err = sysfs_create_link(&per_cpu(device_mce, cpu).kobj, + err = sysfs_create_link(&per_cpu(mce_dev, cpu).kobj, b->kobj, name); if (err) goto out; cpumask_copy(b->cpus, cpu_core_mask(cpu)); per_cpu(threshold_banks, cpu)[bank] = b; + goto out; } #endif @@ -522,7 +532,7 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank) goto out; } - b->kobj = kobject_create_and_add(name, &per_cpu(device_mce, cpu).kobj); + b->kobj = kobject_create_and_add(name, &per_cpu(mce_dev, cpu).kobj); if (!b->kobj) goto out_free; @@ -542,7 +552,7 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank) if (i == cpu) continue; - err = sysfs_create_link(&per_cpu(device_mce, i).kobj, + err = sysfs_create_link(&per_cpu(mce_dev, i).kobj, b->kobj, name); if (err) goto out; @@ -605,15 +615,13 @@ static void deallocate_threshold_block(unsigned int cpu, static void threshold_remove_bank(unsigned int cpu, int bank) { - int i = 0; struct threshold_bank *b; char name[32]; + int i = 0; b = per_cpu(threshold_banks, cpu)[bank]; - if (!b) return; - if (!b->blocks) goto free_out; @@ -622,8 +630,9 @@ static void threshold_remove_bank(unsigned int cpu, int bank) #ifdef CONFIG_SMP /* sibling symlink */ if (shared_bank[bank] && b->blocks->cpu != cpu) { - sysfs_remove_link(&per_cpu(device_mce, cpu).kobj, name); + sysfs_remove_link(&per_cpu(mce_dev, cpu).kobj, name); per_cpu(threshold_banks, cpu)[bank] = NULL; + return; } #endif @@ -633,7 +642,7 @@ static void threshold_remove_bank(unsigned int cpu, int bank) if (i == cpu) continue; - sysfs_remove_link(&per_cpu(device_mce, i).kobj, name); + sysfs_remove_link(&per_cpu(mce_dev, i).kobj, name); per_cpu(threshold_banks, i)[bank] = NULL; } @@ -659,12 +668,9 @@ static void threshold_remove_device(unsigned int cpu) } /* get notified when a cpu comes on/off */ -static void __cpuinit amd_64_threshold_cpu_callback(unsigned long action, - unsigned int cpu) +static void __cpuinit +amd_64_threshold_cpu_callback(unsigned long action, unsigned int cpu) { - if (cpu >= NR_CPUS) - return; - switch (action) { case CPU_ONLINE: case CPU_ONLINE_FROZEN: @@ -686,11 +692,12 @@ static __init int threshold_init_device(void) /* to hit CPUs online before the notifier is up */ for_each_online_cpu(lcpu) { int err = threshold_create_device(lcpu); + if (err) return err; } threshold_cpu_callback = amd_64_threshold_cpu_callback; + return 0; } - device_initcall(threshold_init_device); diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel.c b/arch/x86/kernel/cpu/mcheck/mce_intel.c new file mode 100644 index 00000000000..2b011d2d857 --- /dev/null +++ b/arch/x86/kernel/cpu/mcheck/mce_intel.c @@ -0,0 +1,74 @@ +/* + * Common code for Intel machine checks + */ +#include <linux/interrupt.h> +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/init.h> +#include <linux/smp.h> + +#include <asm/therm_throt.h> +#include <asm/processor.h> +#include <asm/system.h> +#include <asm/apic.h> +#include <asm/msr.h> + +#include "mce.h" + +void intel_init_thermal(struct cpuinfo_x86 *c) +{ + unsigned int cpu = smp_processor_id(); + int tm2 = 0; + u32 l, h; + + /* Thermal monitoring depends on ACPI and clock modulation*/ + if (!cpu_has(c, X86_FEATURE_ACPI) || !cpu_has(c, X86_FEATURE_ACC)) + return; + + /* + * First check if its enabled already, in which case there might + * be some SMM goo which handles it, so we can't even put a handler + * since it might be delivered via SMI already: + */ + rdmsr(MSR_IA32_MISC_ENABLE, l, h); + h = apic_read(APIC_LVTTHMR); + if ((l & MSR_IA32_MISC_ENABLE_TM1) && (h & APIC_DM_SMI)) { + printk(KERN_DEBUG + "CPU%d: Thermal monitoring handled by SMI\n", cpu); + return; + } + + if (cpu_has(c, X86_FEATURE_TM2) && (l & MSR_IA32_MISC_ENABLE_TM2)) + tm2 = 1; + + /* Check whether a vector already exists */ + if (h & APIC_VECTOR_MASK) { + printk(KERN_DEBUG + "CPU%d: Thermal LVT vector (%#x) already installed\n", + cpu, (h & APIC_VECTOR_MASK)); + return; + } + + /* We'll mask the thermal vector in the lapic till we're ready: */ + h = THERMAL_APIC_VECTOR | APIC_DM_FIXED | APIC_LVT_MASKED; + apic_write(APIC_LVTTHMR, h); + + rdmsr(MSR_IA32_THERM_INTERRUPT, l, h); + wrmsr(MSR_IA32_THERM_INTERRUPT, + l | (THERM_INT_LOW_ENABLE | THERM_INT_HIGH_ENABLE), h); + + intel_set_thermal_handler(); + + rdmsr(MSR_IA32_MISC_ENABLE, l, h); + wrmsr(MSR_IA32_MISC_ENABLE, l | MSR_IA32_MISC_ENABLE_TM1, h); + + /* Unmask the thermal vector: */ + l = apic_read(APIC_LVTTHMR); + apic_write(APIC_LVTTHMR, l & ~APIC_LVT_MASKED); + + printk(KERN_INFO "CPU%d: Thermal monitoring enabled (%s)\n", + cpu, tm2 ? "TM2" : "TM1"); + + /* enable thermal throttle processing */ + atomic_set(&therm_throt_en, 1); +} diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel_64.c b/arch/x86/kernel/cpu/mcheck/mce_intel_64.c index cef3ee30744..eff3740501a 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_intel_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce_intel_64.c @@ -17,6 +17,8 @@ #include <asm/therm_throt.h> #include <asm/apic.h> +#include "mce.h" + asmlinkage void smp_thermal_interrupt(void) { __u64 msr_val; @@ -27,67 +29,13 @@ asmlinkage void smp_thermal_interrupt(void) irq_enter(); rdmsrl(MSR_IA32_THERM_STATUS, msr_val); - if (therm_throt_process(msr_val & 1)) + if (therm_throt_process(msr_val & THERM_STATUS_PROCHOT)) mce_log_therm_throt_event(msr_val); inc_irq_stat(irq_thermal_count); irq_exit(); } -static void intel_init_thermal(struct cpuinfo_x86 *c) -{ - u32 l, h; - int tm2 = 0; - unsigned int cpu = smp_processor_id(); - - if (!cpu_has(c, X86_FEATURE_ACPI)) - return; - - if (!cpu_has(c, X86_FEATURE_ACC)) - return; - - /* first check if TM1 is already enabled by the BIOS, in which - * case there might be some SMM goo which handles it, so we can't even - * put a handler since it might be delivered via SMI already. - */ - rdmsr(MSR_IA32_MISC_ENABLE, l, h); - h = apic_read(APIC_LVTTHMR); - if ((l & MSR_IA32_MISC_ENABLE_TM1) && (h & APIC_DM_SMI)) { - printk(KERN_DEBUG - "CPU%d: Thermal monitoring handled by SMI\n", cpu); - return; - } - - if (cpu_has(c, X86_FEATURE_TM2) && (l & MSR_IA32_MISC_ENABLE_TM2)) - tm2 = 1; - - if (h & APIC_VECTOR_MASK) { - printk(KERN_DEBUG - "CPU%d: Thermal LVT vector (%#x) already " - "installed\n", cpu, (h & APIC_VECTOR_MASK)); - return; - } - - h = THERMAL_APIC_VECTOR; - h |= (APIC_DM_FIXED | APIC_LVT_MASKED); - apic_write(APIC_LVTTHMR, h); - - rdmsr(MSR_IA32_THERM_INTERRUPT, l, h); - wrmsr(MSR_IA32_THERM_INTERRUPT, l | 0x03, h); - - rdmsr(MSR_IA32_MISC_ENABLE, l, h); - wrmsr(MSR_IA32_MISC_ENABLE, l | MSR_IA32_MISC_ENABLE_TM1, h); - - l = apic_read(APIC_LVTTHMR); - apic_write(APIC_LVTTHMR, l & ~APIC_LVT_MASKED); - printk(KERN_INFO "CPU%d: Thermal monitoring enabled (%s)\n", - cpu, tm2 ? "TM2" : "TM1"); - - /* enable thermal throttle processing */ - atomic_set(&therm_throt_en, 1); - return; -} - /* * Support for Intel Correct Machine Check Interrupts. This allows * the CPU to raise an interrupt when a corrected machine check happened. @@ -248,7 +196,7 @@ void cmci_rediscover(int dying) return; cpumask_copy(old, ¤t->cpus_allowed); - for_each_online_cpu (cpu) { + for_each_online_cpu(cpu) { if (cpu == dying) continue; if (set_cpus_allowed_ptr(current, cpumask_of(cpu))) diff --git a/arch/x86/kernel/cpu/mcheck/non-fatal.c b/arch/x86/kernel/cpu/mcheck/non-fatal.c index a74af128efc..70b710420f7 100644 --- a/arch/x86/kernel/cpu/mcheck/non-fatal.c +++ b/arch/x86/kernel/cpu/mcheck/non-fatal.c @@ -6,15 +6,14 @@ * This file contains routines to check for non-fatal MCEs every 15s * */ - -#include <linux/init.h> -#include <linux/types.h> -#include <linux/kernel.h> -#include <linux/jiffies.h> -#include <linux/workqueue.h> #include <linux/interrupt.h> -#include <linux/smp.h> +#include <linux/workqueue.h> +#include <linux/jiffies.h> +#include <linux/kernel.h> #include <linux/module.h> +#include <linux/types.h> +#include <linux/init.h> +#include <linux/smp.h> #include <asm/processor.h> #include <asm/system.h> @@ -22,9 +21,9 @@ #include "mce.h" -static int firstbank; +static int firstbank; -#define MCE_RATE 15*HZ /* timer rate is 15s */ +#define MCE_RATE (15*HZ) /* timer rate is 15s */ static void mce_checkregs(void *info) { @@ -34,23 +33,24 @@ static void mce_checkregs(void *info) for (i = firstbank; i < nr_mce_banks; i++) { rdmsr(MSR_IA32_MC0_STATUS+i*4, low, high); - if (high & (1<<31)) { - printk(KERN_INFO "MCE: The hardware reports a non " - "fatal, correctable incident occurred on " - "CPU %d.\n", + if (!(high & (1<<31))) + continue; + + printk(KERN_INFO "MCE: The hardware reports a non fatal, " + "correctable incident occurred on CPU %d.\n", smp_processor_id()); - printk(KERN_INFO "Bank %d: %08x%08x\n", i, high, low); - - /* - * Scrub the error so we don't pick it up in MCE_RATE - * seconds time. - */ - wrmsr(MSR_IA32_MC0_STATUS+i*4, 0UL, 0UL); - - /* Serialize */ - wmb(); - add_taint(TAINT_MACHINE_CHECK); - } + + printk(KERN_INFO "Bank %d: %08x%08x\n", i, high, low); + + /* + * Scrub the error so we don't pick it up in MCE_RATE + * seconds time: + */ + wrmsr(MSR_IA32_MC0_STATUS+i*4, 0UL, 0UL); + + /* Serialize: */ + wmb(); + add_taint(TAINT_MACHINE_CHECK); } } @@ -77,16 +77,17 @@ static int __init init_nonfatal_mce_checker(void) /* Some Athlons misbehave when we frob bank 0 */ if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD && - boot_cpu_data.x86 == 6) - firstbank = 1; + boot_cpu_data.x86 == 6) + firstbank = 1; else - firstbank = 0; + firstbank = 0; /* * Check for non-fatal errors every MCE_RATE s */ schedule_delayed_work(&mce_work, round_jiffies_relative(MCE_RATE)); printk(KERN_INFO "Machine check exception polling timer started.\n"); + return 0; } module_init(init_nonfatal_mce_checker); diff --git a/arch/x86/kernel/cpu/mcheck/p4.c b/arch/x86/kernel/cpu/mcheck/p4.c index f53bdcbaf38..82cee108a2d 100644 --- a/arch/x86/kernel/cpu/mcheck/p4.c +++ b/arch/x86/kernel/cpu/mcheck/p4.c @@ -2,18 +2,17 @@ * P4 specific Machine Check Exception Reporting */ -#include <linux/init.h> -#include <linux/types.h> -#include <linux/kernel.h> #include <linux/interrupt.h> +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/init.h> #include <linux/smp.h> +#include <asm/therm_throt.h> #include <asm/processor.h> #include <asm/system.h> -#include <asm/msr.h> #include <asm/apic.h> - -#include <asm/therm_throt.h> +#include <asm/msr.h> #include "mce.h" @@ -36,6 +35,7 @@ static int mce_num_extended_msrs; #ifdef CONFIG_X86_MCE_P4THERMAL + static void unexpected_thermal_interrupt(struct pt_regs *regs) { printk(KERN_ERR "CPU%d: Unexpected LVT TMR interrupt!\n", @@ -43,7 +43,7 @@ static void unexpected_thermal_interrupt(struct pt_regs *regs) add_taint(TAINT_MACHINE_CHECK); } -/* P4/Xeon Thermal transition interrupt handler */ +/* P4/Xeon Thermal transition interrupt handler: */ static void intel_thermal_interrupt(struct pt_regs *regs) { __u64 msr_val; @@ -51,11 +51,12 @@ static void intel_thermal_interrupt(struct pt_regs *regs) ack_APIC_irq(); rdmsrl(MSR_IA32_THERM_STATUS, msr_val); - therm_throt_process(msr_val & 0x1); + therm_throt_process(msr_val & THERM_STATUS_PROCHOT); } -/* Thermal interrupt handler for this CPU setup */ -static void (*vendor_thermal_interrupt)(struct pt_regs *regs) = unexpected_thermal_interrupt; +/* Thermal interrupt handler for this CPU setup: */ +static void (*vendor_thermal_interrupt)(struct pt_regs *regs) = + unexpected_thermal_interrupt; void smp_thermal_interrupt(struct pt_regs *regs) { @@ -65,67 +66,15 @@ void smp_thermal_interrupt(struct pt_regs *regs) irq_exit(); } -/* P4/Xeon Thermal regulation detect and init */ -static void intel_init_thermal(struct cpuinfo_x86 *c) +void intel_set_thermal_handler(void) { - u32 l, h; - unsigned int cpu = smp_processor_id(); - - /* Thermal monitoring */ - if (!cpu_has(c, X86_FEATURE_ACPI)) - return; /* -ENODEV */ - - /* Clock modulation */ - if (!cpu_has(c, X86_FEATURE_ACC)) - return; /* -ENODEV */ - - /* first check if its enabled already, in which case there might - * be some SMM goo which handles it, so we can't even put a handler - * since it might be delivered via SMI already -zwanem. - */ - rdmsr(MSR_IA32_MISC_ENABLE, l, h); - h = apic_read(APIC_LVTTHMR); - if ((l & MSR_IA32_MISC_ENABLE_TM1) && (h & APIC_DM_SMI)) { - printk(KERN_DEBUG "CPU%d: Thermal monitoring handled by SMI\n", - cpu); - return; /* -EBUSY */ - } - - /* check whether a vector already exists, temporarily masked? */ - if (h & APIC_VECTOR_MASK) { - printk(KERN_DEBUG "CPU%d: Thermal LVT vector (%#x) already " - "installed\n", - cpu, (h & APIC_VECTOR_MASK)); - return; /* -EBUSY */ - } - - /* The temperature transition interrupt handler setup */ - h = THERMAL_APIC_VECTOR; /* our delivery vector */ - h |= (APIC_DM_FIXED | APIC_LVT_MASKED); /* we'll mask till we're ready */ - apic_write(APIC_LVTTHMR, h); - - rdmsr(MSR_IA32_THERM_INTERRUPT, l, h); - wrmsr(MSR_IA32_THERM_INTERRUPT, l | 0x03 , h); - - /* ok we're good to go... */ vendor_thermal_interrupt = intel_thermal_interrupt; - - rdmsr(MSR_IA32_MISC_ENABLE, l, h); - wrmsr(MSR_IA32_MISC_ENABLE, l | MSR_IA32_MISC_ENABLE_TM1, h); - - l = apic_read(APIC_LVTTHMR); - apic_write(APIC_LVTTHMR, l & ~APIC_LVT_MASKED); - printk(KERN_INFO "CPU%d: Thermal monitoring enabled\n", cpu); - - /* enable thermal throttle processing */ - atomic_set(&therm_throt_en, 1); - return; } -#endif /* CONFIG_X86_MCE_P4THERMAL */ +#endif /* CONFIG_X86_MCE_P4THERMAL */ /* P4/Xeon Extended MCE MSR retrieval, return 0 if unsupported */ -static inline void intel_get_extended_msrs(struct intel_mce_extended_msrs *r) +static void intel_get_extended_msrs(struct intel_mce_extended_msrs *r) { u32 h; @@ -143,9 +92,9 @@ static inline void intel_get_extended_msrs(struct intel_mce_extended_msrs *r) static void intel_machine_check(struct pt_regs *regs, long error_code) { - int recover = 1; u32 alow, ahigh, high, low; u32 mcgstl, mcgsth; + int recover = 1; int i; rdmsr(MSR_IA32_MCG_STATUS, mcgstl, mcgsth); @@ -157,7 +106,9 @@ static void intel_machine_check(struct pt_regs *regs, long error_code) if (mce_num_extended_msrs > 0) { struct intel_mce_extended_msrs dbg; + intel_get_extended_msrs(&dbg); + printk(KERN_DEBUG "CPU %d: EIP: %08x EFLAGS: %08x\n" "\teax: %08x ebx: %08x ecx: %08x edx: %08x\n" "\tesi: %08x edi: %08x ebp: %08x esp: %08x\n", @@ -171,6 +122,7 @@ static void intel_machine_check(struct pt_regs *regs, long error_code) if (high & (1<<31)) { char misc[20]; char addr[24]; + misc[0] = addr[0] = '\0'; if (high & (1<<29)) recover |= 1; @@ -196,6 +148,7 @@ static void intel_machine_check(struct pt_regs *regs, long error_code) panic("Unable to continue"); printk(KERN_EMERG "Attempting to continue.\n"); + /* * Do not clear the MSR_IA32_MCi_STATUS if the error is not * recoverable/continuable.This will allow BIOS to look at the MSRs @@ -217,7 +170,6 @@ static void intel_machine_check(struct pt_regs *regs, long error_code) wrmsr(MSR_IA32_MCG_STATUS, mcgstl, mcgsth); } - void intel_p4_mcheck_init(struct cpuinfo_x86 *c) { u32 l, h; diff --git a/arch/x86/kernel/cpu/mcheck/p5.c b/arch/x86/kernel/cpu/mcheck/p5.c index c9f77ea69ed..015f481ab1b 100644 --- a/arch/x86/kernel/cpu/mcheck/p5.c +++ b/arch/x86/kernel/cpu/mcheck/p5.c @@ -2,11 +2,10 @@ * P5 specific Machine Check Exception Reporting * (C) Copyright 2002 Alan Cox <alan@lxorguk.ukuu.org.uk> */ - -#include <linux/init.h> -#include <linux/types.h> -#include <linux/kernel.h> #include <linux/interrupt.h> +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/init.h> #include <linux/smp.h> #include <asm/processor.h> @@ -15,39 +14,58 @@ #include "mce.h" -/* Machine check handler for Pentium class Intel */ +/* By default disabled */ +int mce_p5_enable; + +/* Machine check handler for Pentium class Intel CPUs: */ static void pentium_machine_check(struct pt_regs *regs, long error_code) { u32 loaddr, hi, lotype; + rdmsr(MSR_IA32_P5_MC_ADDR, loaddr, hi); rdmsr(MSR_IA32_P5_MC_TYPE, lotype, hi); - printk(KERN_EMERG "CPU#%d: Machine Check Exception: 0x%8X (type 0x%8X).\n", smp_processor_id(), loaddr, lotype); - if (lotype&(1<<5)) - printk(KERN_EMERG "CPU#%d: Possible thermal failure (CPU on fire ?).\n", smp_processor_id()); + + printk(KERN_EMERG + "CPU#%d: Machine Check Exception: 0x%8X (type 0x%8X).\n", + smp_processor_id(), loaddr, lotype); + + if (lotype & (1<<5)) { + printk(KERN_EMERG + "CPU#%d: Possible thermal failure (CPU on fire ?).\n", + smp_processor_id()); + } + add_taint(TAINT_MACHINE_CHECK); } -/* Set up machine check reporting for processors with Intel style MCE */ +/* Set up machine check reporting for processors with Intel style MCE: */ void intel_p5_mcheck_init(struct cpuinfo_x86 *c) { u32 l, h; - /*Check for MCE support */ + /* Check for MCE support: */ if (!cpu_has(c, X86_FEATURE_MCE)) return; - /* Default P5 to off as its often misconnected */ +#ifdef CONFIG_X86_OLD_MCE + /* Default P5 to off as its often misconnected: */ if (mce_disabled != -1) return; +#endif + machine_check_vector = pentium_machine_check; + /* Make sure the vector pointer is visible before we enable MCEs: */ wmb(); - /* Read registers before enabling */ + /* Read registers before enabling: */ rdmsr(MSR_IA32_P5_MC_ADDR, l, h); rdmsr(MSR_IA32_P5_MC_TYPE, l, h); - printk(KERN_INFO "Intel old style machine check architecture supported.\n"); + printk(KERN_INFO + "Intel old style machine check architecture supported.\n"); - /* Enable MCE */ + /* Enable MCE: */ set_in_cr4(X86_CR4_MCE); - printk(KERN_INFO "Intel old style machine check reporting enabled on CPU#%d.\n", smp_processor_id()); + printk(KERN_INFO + "Intel old style machine check reporting enabled on CPU#%d.\n", + smp_processor_id()); } diff --git a/arch/x86/kernel/cpu/mcheck/p6.c b/arch/x86/kernel/cpu/mcheck/p6.c index 2ac52d7b434..43c24e66745 100644 --- a/arch/x86/kernel/cpu/mcheck/p6.c +++ b/arch/x86/kernel/cpu/mcheck/p6.c @@ -2,11 +2,10 @@ * P6 specific Machine Check Exception Reporting * (C) Copyright 2002 Alan Cox <alan@lxorguk.ukuu.org.uk> */ - -#include <linux/init.h> -#include <linux/types.h> -#include <linux/kernel.h> #include <linux/interrupt.h> +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/init.h> #include <linux/smp.h> #include <asm/processor.h> @@ -18,9 +17,9 @@ /* Machine Check Handler For PII/PIII */ static void intel_machine_check(struct pt_regs *regs, long error_code) { - int recover = 1; u32 alow, ahigh, high, low; u32 mcgstl, mcgsth; + int recover = 1; int i; rdmsr(MSR_IA32_MCG_STATUS, mcgstl, mcgsth); @@ -35,12 +34,16 @@ static void intel_machine_check(struct pt_regs *regs, long error_code) if (high & (1<<31)) { char misc[20]; char addr[24]; - misc[0] = addr[0] = '\0'; + + misc[0] = '\0'; + addr[0] = '\0'; + if (high & (1<<29)) recover |= 1; if (high & (1<<25)) recover |= 2; high &= ~(1<<31); + if (high & (1<<27)) { rdmsr(MSR_IA32_MC0_MISC+i*4, alow, ahigh); snprintf(misc, 20, "[%08x%08x]", ahigh, alow); @@ -49,6 +52,7 @@ static void intel_machine_check(struct pt_regs *regs, long error_code) rdmsr(MSR_IA32_MC0_ADDR+i*4, alow, ahigh); snprintf(addr, 24, " at %08x%08x", ahigh, alow); } + printk(KERN_EMERG "CPU %d: Bank %d: %08x%08x%s%s\n", smp_processor_id(), i, high, low, misc, addr); } @@ -63,16 +67,17 @@ static void intel_machine_check(struct pt_regs *regs, long error_code) /* * Do not clear the MSR_IA32_MCi_STATUS if the error is not * recoverable/continuable.This will allow BIOS to look at the MSRs - * for errors if the OS could not log the error. + * for errors if the OS could not log the error: */ for (i = 0; i < nr_mce_banks; i++) { unsigned int msr; + msr = MSR_IA32_MC0_STATUS+i*4; rdmsr(msr, low, high); if (high & (1<<31)) { - /* Clear it */ + /* Clear it: */ wrmsr(msr, 0UL, 0UL); - /* Serialize */ + /* Serialize: */ wmb(); add_taint(TAINT_MACHINE_CHECK); } @@ -81,7 +86,7 @@ static void intel_machine_check(struct pt_regs *regs, long error_code) wrmsr(MSR_IA32_MCG_STATUS, mcgstl, mcgsth); } -/* Set up machine check reporting for processors with Intel style MCE */ +/* Set up machine check reporting for processors with Intel style MCE: */ void intel_p6_mcheck_init(struct cpuinfo_x86 *c) { u32 l, h; @@ -97,6 +102,7 @@ void intel_p6_mcheck_init(struct cpuinfo_x86 *c) /* Ok machine check is available */ machine_check_vector = intel_machine_check; + /* Make sure the vector pointer is visible before we enable MCEs: */ wmb(); printk(KERN_INFO "Intel machine check architecture supported.\n"); diff --git a/arch/x86/kernel/cpu/mcheck/therm_throt.c b/arch/x86/kernel/cpu/mcheck/therm_throt.c index d5ae2243f0b..7b1ae2e20ba 100644 --- a/arch/x86/kernel/cpu/mcheck/therm_throt.c +++ b/arch/x86/kernel/cpu/mcheck/therm_throt.c @@ -1,7 +1,7 @@ /* - * * Thermal throttle event support code (such as syslog messaging and rate * limiting) that was factored out from x86_64 (mce_intel.c) and i386 (p4.c). + * * This allows consistent reporting of CPU thermal throttle events. * * Maintains a counter in /sys that keeps track of the number of thermal @@ -13,43 +13,43 @@ * Credits: Adapted from Zwane Mwaikambo's original code in mce_intel.c. * Inspired by Ross Biro's and Al Borchers' counter code. */ - +#include <linux/notifier.h> +#include <linux/jiffies.h> #include <linux/percpu.h> #include <linux/sysdev.h> #include <linux/cpu.h> -#include <asm/cpu.h> -#include <linux/notifier.h> -#include <linux/jiffies.h> + #include <asm/therm_throt.h> /* How long to wait between reporting thermal events */ -#define CHECK_INTERVAL (300 * HZ) +#define CHECK_INTERVAL (300 * HZ) static DEFINE_PER_CPU(__u64, next_check) = INITIAL_JIFFIES; static DEFINE_PER_CPU(unsigned long, thermal_throttle_count); -atomic_t therm_throt_en = ATOMIC_INIT(0); + +atomic_t therm_throt_en = ATOMIC_INIT(0); #ifdef CONFIG_SYSFS -#define define_therm_throt_sysdev_one_ro(_name) \ - static SYSDEV_ATTR(_name, 0444, therm_throt_sysdev_show_##_name, NULL) - -#define define_therm_throt_sysdev_show_func(name) \ -static ssize_t therm_throt_sysdev_show_##name(struct sys_device *dev, \ - struct sysdev_attribute *attr, \ - char *buf) \ -{ \ - unsigned int cpu = dev->id; \ - ssize_t ret; \ - \ - preempt_disable(); /* CPU hotplug */ \ - if (cpu_online(cpu)) \ - ret = sprintf(buf, "%lu\n", \ - per_cpu(thermal_throttle_##name, cpu)); \ - else \ - ret = 0; \ - preempt_enable(); \ - \ - return ret; \ +#define define_therm_throt_sysdev_one_ro(_name) \ + static SYSDEV_ATTR(_name, 0444, therm_throt_sysdev_show_##_name, NULL) + +#define define_therm_throt_sysdev_show_func(name) \ +static ssize_t therm_throt_sysdev_show_##name(struct sys_device *dev, \ + struct sysdev_attribute *attr, \ + char *buf) \ +{ \ + unsigned int cpu = dev->id; \ + ssize_t ret; \ + \ + preempt_disable(); /* CPU hotplug */ \ + if (cpu_online(cpu)) \ + ret = sprintf(buf, "%lu\n", \ + per_cpu(thermal_throttle_##name, cpu)); \ + else \ + ret = 0; \ + preempt_enable(); \ + \ + return ret; \ } define_therm_throt_sysdev_show_func(count); @@ -61,8 +61,8 @@ static struct attribute *thermal_throttle_attrs[] = { }; static struct attribute_group thermal_throttle_attr_group = { - .attrs = thermal_throttle_attrs, - .name = "thermal_throttle" + .attrs = thermal_throttle_attrs, + .name = "thermal_throttle" }; #endif /* CONFIG_SYSFS */ @@ -110,10 +110,11 @@ int therm_throt_process(int curr) } #ifdef CONFIG_SYSFS -/* Add/Remove thermal_throttle interface for CPU device */ +/* Add/Remove thermal_throttle interface for CPU device: */ static __cpuinit int thermal_throttle_add_dev(struct sys_device *sys_dev) { - return sysfs_create_group(&sys_dev->kobj, &thermal_throttle_attr_group); + return sysfs_create_group(&sys_dev->kobj, + &thermal_throttle_attr_group); } static __cpuinit void thermal_throttle_remove_dev(struct sys_device *sys_dev) @@ -121,19 +122,21 @@ static __cpuinit void thermal_throttle_remove_dev(struct sys_device *sys_dev) sysfs_remove_group(&sys_dev->kobj, &thermal_throttle_attr_group); } -/* Mutex protecting device creation against CPU hotplug */ +/* Mutex protecting device creation against CPU hotplug: */ static DEFINE_MUTEX(therm_cpu_lock); /* Get notified when a cpu comes on/off. Be hotplug friendly. */ -static __cpuinit int thermal_throttle_cpu_callback(struct notifier_block *nfb, - unsigned long action, - void *hcpu) +static __cpuinit int +thermal_throttle_cpu_callback(struct notifier_block *nfb, + unsigned long action, + void *hcpu) { unsigned int cpu = (unsigned long)hcpu; struct sys_device *sys_dev; int err = 0; sys_dev = get_cpu_sysdev(cpu); + switch (action) { case CPU_UP_PREPARE: case CPU_UP_PREPARE_FROZEN: diff --git a/arch/x86/kernel/cpu/mcheck/threshold.c b/arch/x86/kernel/cpu/mcheck/threshold.c index 23ee9e730f7..d746df2909c 100644 --- a/arch/x86/kernel/cpu/mcheck/threshold.c +++ b/arch/x86/kernel/cpu/mcheck/threshold.c @@ -17,7 +17,7 @@ static void default_threshold_interrupt(void) void (*mce_threshold_vector)(void) = default_threshold_interrupt; -asmlinkage void mce_threshold_interrupt(void) +asmlinkage void smp_threshold_interrupt(void) { exit_idle(); irq_enter(); diff --git a/arch/x86/kernel/cpu/mcheck/winchip.c b/arch/x86/kernel/cpu/mcheck/winchip.c index 2a043d89811..81b02487090 100644 --- a/arch/x86/kernel/cpu/mcheck/winchip.c +++ b/arch/x86/kernel/cpu/mcheck/winchip.c @@ -2,11 +2,10 @@ * IDT Winchip specific Machine Check Exception Reporting * (C) Copyright 2002 Alan Cox <alan@lxorguk.ukuu.org.uk> */ - -#include <linux/init.h> -#include <linux/types.h> -#include <linux/kernel.h> #include <linux/interrupt.h> +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/init.h> #include <asm/processor.h> #include <asm/system.h> @@ -14,7 +13,7 @@ #include "mce.h" -/* Machine check handler for WinChip C6 */ +/* Machine check handler for WinChip C6: */ static void winchip_machine_check(struct pt_regs *regs, long error_code) { printk(KERN_EMERG "CPU0: Machine Check Exception.\n"); @@ -25,12 +24,18 @@ static void winchip_machine_check(struct pt_regs *regs, long error_code) void winchip_mcheck_init(struct cpuinfo_x86 *c) { u32 lo, hi; + machine_check_vector = winchip_machine_check; + /* Make sure the vector pointer is visible before we enable MCEs: */ wmb(); + rdmsr(MSR_IDT_FCR1, lo, hi); lo |= (1<<2); /* Enable EIERRINT (int 18 MCE) */ lo &= ~(1<<4); /* Enable MCE */ wrmsr(MSR_IDT_FCR1, lo, hi); + set_in_cr4(X86_CR4_MCE); - printk(KERN_INFO "Winchip machine check reporting enabled on CPU#0.\n"); + + printk(KERN_INFO + "Winchip machine check reporting enabled on CPU#0.\n"); } diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index 38946c6e843..a31a7f29cff 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -1007,7 +1007,7 @@ apicinterrupt INVALIDATE_TLB_VECTOR_START+7 \ #endif apicinterrupt THRESHOLD_APIC_VECTOR \ - threshold_interrupt mce_threshold_interrupt + threshold_interrupt smp_threshold_interrupt apicinterrupt THERMAL_APIC_VECTOR \ thermal_interrupt smp_thermal_interrupt @@ -1382,7 +1382,7 @@ paranoiderrorentry stack_segment do_stack_segment errorentry general_protection do_general_protection errorentry page_fault do_page_fault #ifdef CONFIG_X86_MCE -paranoidzeroentry machine_check do_machine_check +paranoidzeroentry machine_check *machine_check_vector(%rip) #endif /* diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c index c1739ac2970..a05660bf029 100644 --- a/arch/x86/kernel/irq.c +++ b/arch/x86/kernel/irq.c @@ -87,7 +87,7 @@ static int show_other_interrupts(struct seq_file *p, int prec) for_each_online_cpu(j) seq_printf(p, "%10u ", irq_stats(j)->irq_thermal_count); seq_printf(p, " Thermal event interrupts\n"); -# ifdef CONFIG_X86_64 +# ifdef CONFIG_X86_MCE_THRESHOLD seq_printf(p, "%*s: ", prec, "THR"); for_each_online_cpu(j) seq_printf(p, "%10u ", irq_stats(j)->irq_threshold_count); @@ -174,7 +174,7 @@ u64 arch_irq_stat_cpu(unsigned int cpu) #endif #ifdef CONFIG_X86_MCE sum += irq_stats(cpu)->irq_thermal_count; -# ifdef CONFIG_X86_64 +# ifdef CONFIG_X86_MCE_THRESHOLD sum += irq_stats(cpu)->irq_threshold_count; # endif #endif diff --git a/arch/x86/kernel/irqinit.c b/arch/x86/kernel/irqinit.c index 2e08b10ad51..aab3d277766 100644 --- a/arch/x86/kernel/irqinit.c +++ b/arch/x86/kernel/irqinit.c @@ -181,8 +181,10 @@ static void __init apic_intr_init(void) { smp_intr_init(); -#ifdef CONFIG_X86_64 +#ifdef CONFIG_X86_THERMAL_VECTOR alloc_intr_gate(THERMAL_APIC_VECTOR, thermal_interrupt); +#endif +#ifdef CONFIG_X86_THRESHOLD alloc_intr_gate(THRESHOLD_APIC_VECTOR, threshold_interrupt); #endif @@ -204,13 +206,6 @@ static void __init apic_intr_init(void) # endif #endif - -#ifdef CONFIG_X86_32 -#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86_MCE_P4THERMAL) - /* thermal monitor LVT interrupt */ - alloc_intr_gate(THERMAL_APIC_VECTOR, thermal_interrupt); -#endif -#endif } /** diff --git a/arch/x86/kernel/signal.c b/arch/x86/kernel/signal.c index 14425166b8e..d0851e3f77e 100644 --- a/arch/x86/kernel/signal.c +++ b/arch/x86/kernel/signal.c @@ -25,11 +25,11 @@ #include <asm/ucontext.h> #include <asm/i387.h> #include <asm/vdso.h> +#include <asm/mce.h> #ifdef CONFIG_X86_64 #include <asm/proto.h> #include <asm/ia32_unistd.h> -#include <asm/mce.h> #endif /* CONFIG_X86_64 */ #include <asm/syscall.h> @@ -857,7 +857,7 @@ static void do_signal(struct pt_regs *regs) void do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags) { -#if defined(CONFIG_X86_64) && defined(CONFIG_X86_MCE) +#ifdef CONFIG_X86_NEW_MCE /* notify userspace of pending MCEs */ if (thread_info_flags & _TIF_MCE_NOTIFY) mce_notify_user(); diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index 2310700faca..f4d683b630b 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -798,15 +798,15 @@ unsigned long patch_espfix_desc(unsigned long uesp, unsigned long kesp) return new_kesp; } -#else +#endif + asmlinkage void __attribute__((weak)) smp_thermal_interrupt(void) { } -asmlinkage void __attribute__((weak)) mce_threshold_interrupt(void) +asmlinkage void __attribute__((weak)) smp_threshold_interrupt(void) { } -#endif /* * 'math_state_restore()' saves the current math information in the diff --git a/kernel/timer.c b/kernel/timer.c index cffffad01c3..e2c47b82ac3 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -756,6 +756,7 @@ void add_timer_on(struct timer_list *timer, int cpu) wake_up_idle_cpu(cpu); spin_unlock_irqrestore(&base->lock, flags); } +EXPORT_SYMBOL_GPL(add_timer_on); /** * del_timer - deactive a timer. |