#include <linux/thread_info.h>
#include <linux/capability.h>
#include <linux/miscdevice.h>
-#include <linux/interrupt.h>
#include <linux/ratelimit.h>
#include <linux/kallsyms.h>
#include <linux/rcupdate.h>
#include <linux/kernel.h>
#include <linux/percpu.h>
#include <linux/string.h>
-#include <linux/sysdev.h>
+#include <linux/device.h>
+#include <linux/syscore_ops.h>
#include <linux/delay.h>
#include <linux/ctype.h>
#include <linux/sched.h>
#include <linux/fs.h>
#include <linux/mm.h>
#include <linux/debugfs.h>
-#include <linux/edac_mce.h>
+#include <linux/irq_work.h>
+#include <linux/export.h>
#include <asm/processor.h>
-#include <asm/hw_irq.h>
-#include <asm/apic.h>
-#include <asm/idle.h>
-#include <asm/ipi.h>
#include <asm/mce.h>
#include <asm/msr.h>
#include "mce-internal.h"
-static DEFINE_MUTEX(mce_read_mutex);
+static DEFINE_MUTEX(mce_chrdev_read_mutex);
#define rcu_dereference_check_mce(p) \
- rcu_dereference_check((p), \
+ rcu_dereference_index_check((p), \
rcu_read_lock_sched_held() || \
- lockdep_is_held(&mce_read_mutex))
+ lockdep_is_held(&mce_chrdev_read_mutex))
#define CREATE_TRACE_POINTS
#include <trace/events/mce.h>
static char mce_helper[128];
static char *mce_helper_argv[2] = { mce_helper, NULL };
-static DECLARE_WAIT_QUEUE_HEAD(mce_wait);
+static DECLARE_WAIT_QUEUE_HEAD(mce_chrdev_wait);
+
static DEFINE_PER_CPU(struct mce, mces_seen);
static int cpu_missing;
-void (*mce_cpu_specific_poll)(struct mce *);
-EXPORT_SYMBOL_GPL(mce_cpu_specific_poll);
-
-/*
- * CPU/chipset specific EDAC code can register a notifier call here to print
- * MCE errors in a human-readable form.
- */
-ATOMIC_NOTIFIER_HEAD(x86_mce_decoder_chain);
-EXPORT_SYMBOL_GPL(x86_mce_decoder_chain);
-
-static int default_decode_mce(struct notifier_block *nb, unsigned long val,
- void *data)
-{
- pr_emerg("No human readable MCE decoding support on this CPU type.\n");
- pr_emerg("Run the message through 'mcelog --ascii' to decode.\n");
-
- return NOTIFY_STOP;
-}
-
-static struct notifier_block mce_dec_nb = {
- .notifier_call = default_decode_mce,
- .priority = -1,
-};
/* MCA banks polled by the period polling timer for corrected events */
DEFINE_PER_CPU(mce_banks_t, mce_poll_banks) = {
static DEFINE_PER_CPU(struct work_struct, mce_work);
+/*
+ * CPU/chipset specific EDAC code can register a notifier call here to print
+ * MCE errors in a human-readable form.
+ */
+ATOMIC_NOTIFIER_HEAD(x86_mce_decoder_chain);
+
/* Do initial initialization of a struct mce */
void mce_setup(struct mce *m)
{
m->time = get_seconds();
m->cpuvendor = boot_cpu_data.x86_vendor;
m->cpuid = cpuid_eax(1);
-#ifdef CONFIG_SMP
+#ifndef CONFIG_XEN
m->socketid = cpu_data(m->extcpu).phys_proc_id;
-#endif
m->apicid = cpu_data(m->extcpu).initial_apicid;
+#endif
rdmsrl(MSR_IA32_MCG_CAP, m->mcgcap);
}
void mce_log(struct mce *mce)
{
unsigned next, entry;
+ int ret = 0;
/* Emit the trace record: */
trace_mce_record(mce);
+ ret = atomic_notifier_call_chain(&x86_mce_decoder_chain, 0, mce);
+ if (ret == NOTIFY_STOP)
+ return;
+
mce->finished = 0;
wmb();
for (;;) {
entry = rcu_dereference_check_mce(mcelog.next);
for (;;) {
- /*
- * If edac_mce is enabled, it will check the error type
- * and will process it, if it is a known error.
- * Otherwise, the error will be sent through mcelog
- * interface
- */
- if (edac_mce_parse(mce))
- return;
/*
* When the buffer fills up discard new entries.
set_bit(0, &mce_need_notify);
}
+static void drain_mcelog_buffer(void)
+{
+ unsigned int next, i, prev = 0;
+
+ next = ACCESS_ONCE(mcelog.next);
+
+ do {
+ struct mce *m;
+
+ /* drain what was logged during boot */
+ for (i = prev; i < next; i++) {
+ unsigned long start = jiffies;
+ unsigned retries = 1;
+
+ m = &mcelog.entry[i];
+
+ while (!m->finished) {
+ if (time_after_eq(jiffies, start + 2*retries))
+ retries++;
+
+ cpu_relax();
+
+ if (!m->finished && retries >= 4) {
+ pr_err("MCE: skipping error being logged currently!\n");
+ break;
+ }
+ }
+ smp_rmb();
+ atomic_notifier_call_chain(&x86_mce_decoder_chain, 0, m);
+ }
+
+ memset(mcelog.entry + prev, 0, (next - prev) * sizeof(*m));
+ prev = next;
+ next = cmpxchg(&mcelog.next, prev, 0);
+ } while (next != prev);
+}
+
+
+void mce_register_decode_chain(struct notifier_block *nb)
+{
+ atomic_notifier_chain_register(&x86_mce_decoder_chain, nb);
+ drain_mcelog_buffer();
+}
+EXPORT_SYMBOL_GPL(mce_register_decode_chain);
+
+void mce_unregister_decode_chain(struct notifier_block *nb)
+{
+ atomic_notifier_chain_unregister(&x86_mce_decoder_chain, nb);
+}
+EXPORT_SYMBOL_GPL(mce_unregister_decode_chain);
+
static void print_mce(struct mce *m)
{
- pr_emerg("CPU %d: Machine Check Exception: %16Lx Bank %d: %016Lx\n",
+ int ret = 0;
+
+ pr_emerg(HW_ERR "CPU %d: Machine Check Exception: %Lx Bank %d: %016Lx\n",
m->extcpu, m->mcgstatus, m->bank, m->status);
if (m->ip) {
- pr_emerg("RIP%s %02x:<%016Lx> ",
+ pr_emerg(HW_ERR "RIP%s %02x:<%016Lx> ",
!(m->mcgstatus & MCG_STATUS_EIPV) ? " !INEXACT!" : "",
m->cs, m->ip);
pr_cont("\n");
}
- pr_emerg("TSC %llx ", m->tsc);
+ pr_emerg(HW_ERR "TSC %llx ", m->tsc);
if (m->addr)
pr_cont("ADDR %llx ", m->addr);
if (m->misc)
pr_cont("MISC %llx ", m->misc);
pr_cont("\n");
- pr_emerg("PROCESSOR %u:%x TIME %llu SOCKET %u APIC %x\n",
+ /*
+ * Note this output is parsed by external tools and old fields
+ * should not be changed.
+ */
+#ifndef CONFIG_XEN
+ pr_emerg(HW_ERR "PROCESSOR %u:%x TIME %llu SOCKET %u APIC %x microcode %x\n",
+ m->cpuvendor, m->cpuid, m->time, m->socketid, m->apicid,
+ cpu_data(m->extcpu).microcode);
+#else
+ pr_emerg(HW_ERR "PROCESSOR %u:%x TIME %llu SOCKET %u APIC %x\n",
m->cpuvendor, m->cpuid, m->time, m->socketid, m->apicid);
+#endif
/*
* Print out human-readable details about the MCE error,
* (if the CPU has an implementation for that)
*/
- atomic_notifier_call_chain(&x86_mce_decoder_chain, 0, m);
-}
-
-static void print_mce_head(void)
-{
- pr_emerg("\nHARDWARE ERROR\n");
-}
+ ret = atomic_notifier_call_chain(&x86_mce_decoder_chain, 0, m);
+ if (ret == NOTIFY_STOP)
+ return;
-static void print_mce_tail(void)
-{
- pr_emerg("This is not a software problem!\n");
+ pr_emerg_ratelimited(HW_ERR "Run the above through 'mcelog --ascii'\n");
}
#define PANIC_TIMEOUT 5 /* 5 seconds */
if (atomic_inc_return(&mce_fake_paniced) > 1)
return;
}
- print_mce_head();
/* First print corrected ones that are still unlogged */
for (i = 0; i < MCE_LOG_LEN; i++) {
struct mce *m = &mcelog.entry[i];
apei_err = apei_write_mce(final);
}
if (cpu_missing)
- printk(KERN_EMERG "Some CPUs didn't answer in synchronization\n");
- print_mce_tail();
+ pr_emerg(HW_ERR "Some CPUs didn't answer in synchronization\n");
if (exp)
- printk(KERN_EMERG "Machine check: %s\n", exp);
+ pr_emerg(HW_ERR "Machine check: %s\n", exp);
if (!fake_panic) {
if (panic_timeout == 0)
panic_timeout = mce_panic_timeout;
panic(msg);
} else
- printk(KERN_EMERG "Fake kernel panic: %s\n", msg);
+ pr_emerg(HW_ERR "Fake kernel panic: %s\n", msg);
}
/* Support code for software error injection */
static int msr_to_offset(u32 msr)
{
- unsigned bank = __get_cpu_var(injectm.bank);
+ unsigned bank = __this_cpu_read(injectm.bank);
if (msr == rip_msr)
return offsetof(struct mce, ip);
{
u64 v;
- if (__get_cpu_var(injectm).finished) {
+ if (__this_cpu_read(injectm.finished)) {
int offset = msr_to_offset(msr);
if (offset < 0)
static void mce_wrmsrl(u32 msr, u64 v)
{
- if (__get_cpu_var(injectm).finished) {
+ if (__this_cpu_read(injectm.finished)) {
int offset = msr_to_offset(msr);
if (offset >= 0)
wrmsrl(msr, v);
}
-static int under_injection(void)
+/*
+ * Collect all global (w.r.t. this processor) status about this machine
+ * check into our "mce" struct so that we can use it later to assess
+ * the severity of the problem as we read per-bank specific details.
+ */
+static inline void mce_gather_info(struct mce *m, struct pt_regs *regs)
{
- return __get_cpu_var(injectm).finished;
+ mce_setup(m);
+
+ m->mcgstatus = mce_rdmsrl(MSR_IA32_MCG_STATUS);
+ if (regs) {
+ /*
+ * Get the address of the instruction at the time of
+ * the machine check error.
+ */
+ if (m->mcgstatus & (MCG_STATUS_RIPV|MCG_STATUS_EIPV)) {
+ m->ip = regs->ip;
+ m->cs = regs->cs;
+ }
+ /* Use accurate RIP reporting if available. */
+ if (rip_msr)
+ m->ip = mce_rdmsrl(rip_msr);
+ }
}
/*
}
}
-/*
- * Get the address of the instruction at the time of the machine check
- * error.
- */
-static inline void mce_get_rip(struct mce *m, struct pt_regs *regs)
-{
-
- if (regs && (m->mcgstatus & (MCG_STATUS_RIPV|MCG_STATUS_EIPV))) {
- m->ip = regs->ip;
- m->cs = regs->cs;
- } else {
- m->ip = 0;
- m->cs = 0;
- }
- if (rip_msr)
- m->ip = mce_rdmsrl(rip_msr);
-}
+DEFINE_PER_CPU(struct irq_work, mce_irq_work);
-#ifdef CONFIG_X86_LOCAL_APIC
-/*
- * Called after interrupts have been reenabled again
- * when a MCE happened during an interrupts off region
- * in the kernel.
- */
-asmlinkage void smp_mce_self_interrupt(struct pt_regs *regs)
+static void mce_irq_work_cb(struct irq_work *entry)
{
- ack_APIC_irq();
- exit_idle();
- irq_enter();
mce_notify_irq();
mce_schedule_work();
- irq_exit();
}
-#endif
static void mce_report_event(struct pt_regs *regs)
{
return;
}
-#ifdef CONFIG_X86_LOCAL_APIC
- /*
- * Without APIC do not notify. The event will be picked
- * up eventually.
- */
- if (!cpu_has_apic)
- return;
+ irq_work_queue(&__get_cpu_var(mce_irq_work));
+}
- /*
- * When interrupts are disabled we cannot use
- * kernel services safely. Trigger an self interrupt
- * through the APIC to instead do the notification
- * after interrupts are reenabled again.
- */
- apic->send_IPI_self(MCE_SELF_VECTOR);
+/*
+ * Read ADDR and MISC registers.
+ */
+static void mce_read_aux(struct mce *m, int i)
+{
+ if (m->status & MCI_STATUS_MISCV)
+ m->misc = mce_rdmsrl(MSR_IA32_MCx_MISC(i));
+ if (m->status & MCI_STATUS_ADDRV) {
+ m->addr = mce_rdmsrl(MSR_IA32_MCx_ADDR(i));
- /*
- * Wait for idle afterwards again so that we don't leave the
- * APIC in a non idle state because the normal APIC writes
- * cannot exclude us.
- */
- apic_wait_icr_idle();
-#endif
+ /*
+ * Mask the reported address by the reported granularity.
+ */
+ if (mce_ser && (m->status & MCI_STATUS_MISCV)) {
+ u8 shift = MCI_MISC_ADDR_LSB(m->misc);
+ m->addr >>= shift;
+ m->addr <<= shift;
+ }
+ }
}
DEFINE_PER_CPU(unsigned, mce_poll_count);
percpu_inc(mce_poll_count);
- mce_setup(&m);
+ mce_gather_info(&m, NULL);
- m.mcgstatus = mce_rdmsrl(MSR_IA32_MCG_STATUS);
for (i = 0; i < banks; i++) {
if (!mce_banks[i].ctl || !test_bit(i, *b))
continue;
(m.status & (mce_ser ? MCI_STATUS_S : MCI_STATUS_UC)))
continue;
- if (m.status & MCI_STATUS_MISCV)
- m.misc = mce_rdmsrl(MSR_IA32_MCx_MISC(i));
- if (m.status & MCI_STATUS_ADDRV)
- m.addr = mce_rdmsrl(MSR_IA32_MCx_ADDR(i));
+ mce_read_aux(&m, i);
if (!(flags & MCP_TIMESTAMP))
m.tsc = 0;
-
- if (mce_cpu_specific_poll && !under_injection() && !mce_dont_log_ce)
- mce_cpu_specific_poll(&m);
-
/*
* Don't get the IP here because it's unlikely to
* have anything to do with the actual error location.
*/
- if (!(flags & MCP_DONTLOG) && !mce_dont_log_ce) {
+ if (!(flags & MCP_DONTLOG) && !mce_dont_log_ce)
mce_log(&m);
- add_taint(TAINT_MACHINE_CHECK);
- }
/*
* Clear state for this bank.
* Check if the address reported by the CPU is in a format we can parse.
* It would be possible to add code for most other cases, but all would
* be somewhat complicated (e.g. segment offset would require an instruction
- * parser). So only support physical addresses upto page granuality for now.
+ * parser). So only support physical addresses up to page granuality for now.
*/
static int mce_usable_address(struct mce *m)
{
if (!(m->status & MCI_STATUS_MISCV) || !(m->status & MCI_STATUS_ADDRV))
return 0;
- if ((m->misc & 0x3f) > PAGE_SHIFT)
+ if (MCI_MISC_ADDR_LSB(m->misc) > PAGE_SHIFT)
return 0;
- if (((m->misc >> 6) & 7) != MCM_ADDR_PHYS)
+ if (MCI_MISC_ADDR_MODE(m->misc) != MCI_MISC_ADDR_PHYS)
return 0;
return 1;
}
}
/*
+ * Need to save faulting physical address associated with a process
+ * in the machine check handler some place where we can grab it back
+ * later in mce_notify_process()
+ */
+#define MCE_INFO_MAX 16
+
+struct mce_info {
+ atomic_t inuse;
+ struct task_struct *t;
+ __u64 paddr;
+ int restartable;
+} mce_info[MCE_INFO_MAX];
+
+static void mce_save_info(__u64 addr, int c)
+{
+ struct mce_info *mi;
+
+ for (mi = mce_info; mi < &mce_info[MCE_INFO_MAX]; mi++) {
+ if (atomic_cmpxchg(&mi->inuse, 0, 1) == 0) {
+ mi->t = current;
+ mi->paddr = addr;
+ mi->restartable = c;
+ return;
+ }
+ }
+
+ mce_panic("Too many concurrent recoverable errors", NULL, NULL);
+}
+
+static struct mce_info *mce_find_info(void)
+{
+ struct mce_info *mi;
+
+ for (mi = mce_info; mi < &mce_info[MCE_INFO_MAX]; mi++)
+ if (atomic_read(&mi->inuse) && mi->t == current)
+ return mi;
+ return NULL;
+}
+
+static void mce_clear_info(struct mce_info *mi)
+{
+ atomic_set(&mi->inuse, 0);
+}
+
+/*
* The actual machine check handler. This only handles real
* exceptions when something got corrupted coming in through int 18.
*
percpu_inc(mce_exception_count);
- if (notify_die(DIE_NMI, "machine check", regs, error_code,
- 18, SIGKILL) == NOTIFY_STOP)
- goto out;
if (!banks)
goto out;
- mce_setup(&m);
+ mce_gather_info(&m, regs);
- m.mcgstatus = mce_rdmsrl(MSR_IA32_MCG_STATUS);
final = &__get_cpu_var(mces_seen);
*final = m;
barrier();
/*
- * When no restart IP must always kill or panic.
+ * When no restart IP might need to kill or panic.
+ * Assume the worst for now, but if we find the
+ * severity is MCE_AR_SEVERITY we have other options.
*/
if (!(m.mcgstatus & MCG_STATUS_RIPV))
kill_it = 1;
continue;
}
- /*
- * Kill on action required.
- */
- if (severity == MCE_AR_SEVERITY)
- kill_it = 1;
-
- if (m.status & MCI_STATUS_MISCV)
- m.misc = mce_rdmsrl(MSR_IA32_MCx_MISC(i));
- if (m.status & MCI_STATUS_ADDRV)
- m.addr = mce_rdmsrl(MSR_IA32_MCx_ADDR(i));
+ mce_read_aux(&m, i);
/*
* Action optional error. Queue address for later processing.
if (severity == MCE_AO_SEVERITY && mce_usable_address(&m))
mce_ring_add(m.addr >> PAGE_SHIFT);
- mce_get_rip(&m, regs);
mce_log(&m);
if (severity > worst) {
}
}
+ /* mce_clear_state will clear *final, save locally for use later */
+ m = *final;
+
if (!no_way_out)
mce_clear_state(toclear);
no_way_out = worst >= MCE_PANIC_SEVERITY;
/*
- * 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.
- *
- * This is mainly used in the case when the system doesn't
- * support MCE broadcasting or it has been disabled.
- */
- if (no_way_out && tolerant < 3)
- mce_panic("Fatal machine check on current CPU", final, msg);
-
- /*
- * If the error seems to be unrecoverable, something should be
- * done. Try to kill as little as possible. If we can kill just
- * one task, do that. If the user has set the tolerance very
- * high, don't try to do anything at all.
+ * At insane "tolerant" levels we take no action. Otherwise
+ * we only die if we have no other choice. For less serious
+ * issues we try to recover, or limit damage to the current
+ * process.
*/
-
- if (kill_it && tolerant < 3)
- force_sig(SIGBUS, current);
-
- /* notify userspace ASAP */
- set_thread_flag(TIF_MCE_NOTIFY);
+ if (tolerant < 3) {
+ if (no_way_out)
+ mce_panic("Fatal machine check on current CPU", &m, msg);
+ if (worst == MCE_AR_SEVERITY) {
+ /* schedule action before return to userland */
+ mce_save_info(m.addr, m.mcgstatus & MCG_STATUS_RIPV);
+ set_thread_flag(TIF_MCE_NOTIFY);
+ } else if (kill_it) {
+ force_sig(SIGBUS, current);
+ }
+ }
if (worst > 0)
mce_report_event(regs);
}
EXPORT_SYMBOL_GPL(do_machine_check);
-/* dummy to break dependency. actual code is in mm/memory-failure.c */
-void __attribute__((weak)) memory_failure(unsigned long pfn, int vector)
+#ifndef CONFIG_MEMORY_FAILURE
+int memory_failure(unsigned long pfn, int vector, int flags)
{
- printk(KERN_ERR "Action optional memory failure at %lx ignored\n", pfn);
+ /* mce_severity() should not hand us an ACTION_REQUIRED error */
+ BUG_ON(flags & MF_ACTION_REQUIRED);
+ printk(KERN_ERR "Uncorrected memory error in page 0x%lx ignored\n"
+ "Rebuild kernel with CONFIG_MEMORY_FAILURE=y for smarter handling\n", pfn);
+
+ return 0;
}
+#endif
/*
- * Called after mce notification in process context. This code
- * is allowed to sleep. Call the high level VM handler to process
- * any corrupted pages.
- * Assume that the work queue code only calls this one at a time
- * per CPU.
- * Note we don't disable preemption, so this code might run on the wrong
- * CPU. In this case the event is picked up by the scheduled work queue.
- * This is merely a fast path to expedite processing in some common
- * cases.
+ * Called in process context that interrupted by MCE and marked with
+ * TIF_MCE_NOTIFY, just before returning to erroneous userland.
+ * This code is allowed to sleep.
+ * Attempt possible recovery such as calling the high level VM handler to
+ * process any corrupted pages, and kill/signal current process if required.
+ * Action required errors are handled here.
*/
void mce_notify_process(void)
{
unsigned long pfn;
- mce_notify_irq();
- while (mce_ring_get(&pfn))
- memory_failure(pfn, MCE_VECTOR);
+ struct mce_info *mi = mce_find_info();
+
+ if (!mi)
+ mce_panic("Lost physical address for unconsumed uncorrectable error", NULL, NULL);
+ pfn = mi->paddr >> PAGE_SHIFT;
+
+ clear_thread_flag(TIF_MCE_NOTIFY);
+
+ pr_err("Uncorrected hardware memory error in user-access at %llx",
+ mi->paddr);
+ /*
+ * We must call memory_failure() here even if the current process is
+ * doomed. We still need to mark the page as poisoned and alert any
+ * other users of the page.
+ */
+ if (memory_failure(pfn, MCE_VECTOR, MF_ACTION_REQUIRED) < 0 ||
+ mi->restartable == 0) {
+ pr_err("Memory error not recovered");
+ force_sig(SIGBUS, current);
+ }
+ mce_clear_info(mi);
}
+/*
+ * Action optional processing happens here (picking up
+ * from the list of faulting pages that do_machine_check()
+ * placed into the "ring").
+ */
static void mce_process_work(struct work_struct *dummy)
{
- mce_notify_process();
+ unsigned long pfn;
+
+ while (mce_ring_get(&pfn))
+ memory_failure(pfn, MCE_VECTOR, 0);
}
#ifdef CONFIG_X86_MCE_INTEL
* Periodic polling timer for "silent" machine check errors. If the
* poller finds an MCE, poll 2x faster. When the poller finds no more
* errors, poll 2x slower (up to check_interval seconds).
+ *
+ * We will disable polling in DOM0 since all CMCI/Polling
+ * mechanism will be done in XEN for Intel CPUs
*/
+#if defined (CONFIG_X86_XEN_MCE)
+static int check_interval = 0; /* disable polling */
+#else
static int check_interval = 5 * 60; /* 5 minutes */
+#endif
static DEFINE_PER_CPU(int, mce_next_interval); /* in jiffies */
static DEFINE_PER_CPU(struct timer_list, mce_timer);
WARN_ON(smp_processor_id() != data);
- if (mce_available(¤t_cpu_data)) {
+ if (mce_available(__this_cpu_ptr(&cpu_info))) {
machine_check_poll(MCP_TIMESTAMP,
&__get_cpu_var(mce_poll_banks));
}
add_timer_on(t, smp_processor_id());
}
+/* Must not be called in IRQ context where del_timer_sync() can deadlock */
+static void mce_timer_delete_all(void)
+{
+ int cpu;
+
+ for_each_online_cpu(cpu)
+ del_timer_sync(&per_cpu(mce_timer, cpu));
+}
+
static void mce_do_trigger(struct work_struct *work)
{
call_usermodehelper(mce_helper, mce_helper_argv, NULL, UMH_NO_WAIT);
/* Not more than two messages every minute */
static DEFINE_RATELIMIT_STATE(ratelimit, 60*HZ, 2);
- clear_thread_flag(TIF_MCE_NOTIFY);
-
if (test_and_clear_bit(0, &mce_need_notify)) {
- wake_up_interruptible(&mce_wait);
+ /* wake processes polling /dev/mcelog */
+ wake_up_interruptible(&mce_chrdev_wait);
/*
* There is no risk of missing notifications because
schedule_work(&mce_trigger_work);
if (__ratelimit(&ratelimit))
- printk(KERN_INFO "Machine check events logged\n");
+ pr_info(HW_ERR "Machine check events logged\n");
return 1;
}
/* This should be disabled by the BIOS, but isn't always */
if (c->x86_vendor == X86_VENDOR_AMD) {
+#ifndef CONFIG_XEN
if (c->x86 == 15 && banks > 4) {
/*
* disable GART TBL walk error reporting, which
*/
clear_bit(10, (unsigned long *)&mce_banks[4].ctl);
}
+#endif
if (c->x86 <= 17 && mce_bootlog < 0) {
/*
* Lots of broken BIOS around that don't clear them
return 0;
}
-static void __cpuinit __mcheck_cpu_ancient_init(struct cpuinfo_x86 *c)
+static int __cpuinit __mcheck_cpu_ancient_init(struct cpuinfo_x86 *c)
{
if (c->x86 != 5)
- return;
+ return 0;
+
switch (c->x86_vendor) {
case X86_VENDOR_INTEL:
intel_p5_mcheck_init(c);
+ return 1;
break;
case X86_VENDOR_CENTAUR:
winchip_mcheck_init(c);
+ return 1;
break;
}
+
+ return 0;
}
static void __mcheck_cpu_init_vendor(struct cpuinfo_x86 *c)
{
+#ifndef CONFIG_X86_64_XEN
switch (c->x86_vendor) {
case X86_VENDOR_INTEL:
mce_intel_feature_init(c);
default:
break;
}
+#endif
}
static void __mcheck_cpu_init_timer(void)
if (mce_disabled)
return;
- __mcheck_cpu_ancient_init(c);
+ if (__mcheck_cpu_ancient_init(c))
+ return;
if (!mce_available(c))
return;
__mcheck_cpu_init_vendor(c);
__mcheck_cpu_init_timer();
INIT_WORK(&__get_cpu_var(mce_work), mce_process_work);
-
+ init_irq_work(&__get_cpu_var(mce_irq_work), &mce_irq_work_cb);
}
/*
- * Character device to read and clear the MCE log.
+ * mce_chrdev: Character device /dev/mcelog to read and clear the MCE log.
*/
-static DEFINE_SPINLOCK(mce_state_lock);
-static int open_count; /* #times opened */
-static int open_exclu; /* already open exclusive? */
+static DEFINE_SPINLOCK(mce_chrdev_state_lock);
+static int mce_chrdev_open_count; /* #times opened */
+static int mce_chrdev_open_exclu; /* already open exclusive? */
-static int mce_open(struct inode *inode, struct file *file)
+static int mce_chrdev_open(struct inode *inode, struct file *file)
{
- spin_lock(&mce_state_lock);
+ spin_lock(&mce_chrdev_state_lock);
- if (open_exclu || (open_count && (file->f_flags & O_EXCL))) {
- spin_unlock(&mce_state_lock);
+ if (mce_chrdev_open_exclu ||
+ (mce_chrdev_open_count && (file->f_flags & O_EXCL))) {
+ spin_unlock(&mce_chrdev_state_lock);
return -EBUSY;
}
if (file->f_flags & O_EXCL)
- open_exclu = 1;
- open_count++;
+ mce_chrdev_open_exclu = 1;
+ mce_chrdev_open_count++;
- spin_unlock(&mce_state_lock);
+ spin_unlock(&mce_chrdev_state_lock);
return nonseekable_open(inode, file);
}
-static int mce_release(struct inode *inode, struct file *file)
+static int mce_chrdev_release(struct inode *inode, struct file *file)
{
- spin_lock(&mce_state_lock);
+ spin_lock(&mce_chrdev_state_lock);
- open_count--;
- open_exclu = 0;
+ mce_chrdev_open_count--;
+ mce_chrdev_open_exclu = 0;
- spin_unlock(&mce_state_lock);
+ spin_unlock(&mce_chrdev_state_lock);
return 0;
}
/* Error or no more MCE record */
if (rc <= 0) {
mce_apei_read_done = 1;
+ /*
+ * When ERST is disabled, mce_chrdev_read() should return
+ * "no record" instead of "no device."
+ */
+ if (rc == -ENODEV)
+ return 0;
return rc;
}
rc = -EFAULT;
return 0;
}
-static ssize_t mce_read(struct file *filp, char __user *ubuf, size_t usize,
- loff_t *off)
+static ssize_t mce_chrdev_read(struct file *filp, char __user *ubuf,
+ size_t usize, loff_t *off)
{
char __user *buf = ubuf;
unsigned long *cpu_tsc;
if (!cpu_tsc)
return -ENOMEM;
- mutex_lock(&mce_read_mutex);
+ mutex_lock(&mce_chrdev_read_mutex);
if (!mce_apei_read_done) {
err = __mce_read_apei(&buf, usize);
do {
for (i = prev; i < next; i++) {
unsigned long start = jiffies;
+ struct mce *m = &mcelog.entry[i];
- while (!mcelog.entry[i].finished) {
+ while (!m->finished) {
if (time_after_eq(jiffies, start + 2)) {
- memset(mcelog.entry + i, 0,
- sizeof(struct mce));
+ memset(m, 0, sizeof(*m));
goto timeout;
}
cpu_relax();
}
smp_rmb();
- err |= copy_to_user(buf, mcelog.entry + i,
- sizeof(struct mce));
- buf += sizeof(struct mce);
+ err |= copy_to_user(buf, m, sizeof(*m));
+ buf += sizeof(*m);
timeout:
;
}
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]) {
- err |= copy_to_user(buf, mcelog.entry+i,
- sizeof(struct mce));
+ struct mce *m = &mcelog.entry[i];
+
+ if (m->finished && m->tsc < cpu_tsc[m->cpu]) {
+ err |= copy_to_user(buf, m, sizeof(*m));
smp_rmb();
- buf += sizeof(struct mce);
- memset(&mcelog.entry[i], 0, sizeof(struct mce));
+ buf += sizeof(*m);
+ memset(m, 0, sizeof(*m));
}
}
err = -EFAULT;
out:
- mutex_unlock(&mce_read_mutex);
+ mutex_unlock(&mce_chrdev_read_mutex);
kfree(cpu_tsc);
return err ? err : buf - ubuf;
}
-static unsigned int mce_poll(struct file *file, poll_table *wait)
+static unsigned int mce_chrdev_poll(struct file *file, poll_table *wait)
{
- poll_wait(file, &mce_wait, wait);
- if (rcu_dereference_check_mce(mcelog.next))
+ poll_wait(file, &mce_chrdev_wait, wait);
+ if (rcu_access_index(mcelog.next))
return POLLIN | POLLRDNORM;
if (!mce_apei_read_done && apei_check_mce())
return POLLIN | POLLRDNORM;
return 0;
}
-static long mce_ioctl(struct file *f, unsigned int cmd, unsigned long arg)
+static long mce_chrdev_ioctl(struct file *f, unsigned int cmd,
+ unsigned long arg)
{
int __user *p = (int __user *)arg;
}
}
-/* 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,
+static ssize_t (*mce_write)(struct file *filp, const char __user *ubuf,
+ size_t usize, loff_t *off);
+
+void register_mce_write_callback(ssize_t (*fn)(struct file *filp,
+ const char __user *ubuf,
+ size_t usize, loff_t *off))
+{
+ mce_write = fn;
+}
+EXPORT_SYMBOL_GPL(register_mce_write_callback);
+
+ssize_t mce_chrdev_write(struct file *filp, const char __user *ubuf,
+ size_t usize, loff_t *off)
+{
+ if (mce_write)
+ return mce_write(filp, ubuf, usize, off);
+ else
+ return -EINVAL;
+}
+
+static const struct file_operations mce_chrdev_ops = {
+ .open = mce_chrdev_open,
+ .release = mce_chrdev_release,
+ .read = mce_chrdev_read,
+ .write = mce_chrdev_write,
+ .poll = mce_chrdev_poll,
+ .unlocked_ioctl = mce_chrdev_ioctl,
+ .llseek = no_llseek,
};
-EXPORT_SYMBOL_GPL(mce_chrdev_ops);
-static struct miscdevice mce_log_device = {
+static struct miscdevice mce_chrdev_device = {
MISC_MCELOG_MINOR,
"mcelog",
&mce_chrdev_ops,
int __init mcheck_init(void)
{
- atomic_notifier_chain_register(&x86_mce_decoder_chain, &mce_dec_nb);
-
mcheck_intel_therm_init();
return 0;
}
/*
- * Sysfs support
+ * mce_syscore: PM support
*/
/*
return 0;
}
-static int mce_suspend(struct sys_device *dev, pm_message_t state)
+static int mce_syscore_suspend(void)
{
return mce_disable_error_reporting();
}
-static int mce_shutdown(struct sys_device *dev)
+static void mce_syscore_shutdown(void)
{
- return mce_disable_error_reporting();
+ mce_disable_error_reporting();
}
/*
* 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)
+static void mce_syscore_resume(void)
{
__mcheck_cpu_init_generic();
- __mcheck_cpu_init_vendor(¤t_cpu_data);
-
- return 0;
+ __mcheck_cpu_init_vendor(__this_cpu_ptr(&cpu_info));
}
+static struct syscore_ops mce_syscore_ops = {
+ .suspend = mce_syscore_suspend,
+ .shutdown = mce_syscore_shutdown,
+ .resume = mce_syscore_resume,
+};
+
+/*
+ * mce_device: Sysfs support
+ */
+
static void mce_cpu_restart(void *data)
{
- del_timer_sync(&__get_cpu_var(mce_timer));
- if (!mce_available(¤t_cpu_data))
+ if (!mce_available(__this_cpu_ptr(&cpu_info)))
return;
__mcheck_cpu_init_generic();
__mcheck_cpu_init_timer();
/* Reinit MCEs after user configuration changes */
static void mce_restart(void)
{
+ mce_timer_delete_all();
on_each_cpu(mce_cpu_restart, NULL, 1);
}
/* Toggle features for corrected errors */
-static void mce_disable_ce(void *all)
+static void mce_disable_cmci(void *data)
{
- if (!mce_available(¤t_cpu_data))
+ if (!mce_available(__this_cpu_ptr(&cpu_info)))
return;
- if (all)
- del_timer_sync(&__get_cpu_var(mce_timer));
cmci_clear();
}
static void mce_enable_ce(void *all)
{
- if (!mce_available(¤t_cpu_data))
+ if (!mce_available(__this_cpu_ptr(&cpu_info)))
return;
cmci_reenable();
cmci_recheck();
__mcheck_cpu_init_timer();
}
-static struct sysdev_class mce_sysclass = {
- .suspend = mce_suspend,
- .shutdown = mce_shutdown,
- .resume = mce_resume,
+static struct bus_type mce_subsys = {
.name = "machinecheck",
+ .dev_name = "machinecheck",
};
-DEFINE_PER_CPU(struct sys_device, mce_dev);
+DEFINE_PER_CPU(struct device *, mce_device);
__cpuinitdata
void (*threshold_cpu_callback)(unsigned long action, unsigned int cpu);
-static inline struct mce_bank *attr_to_bank(struct sysdev_attribute *attr)
+static inline struct mce_bank *attr_to_bank(struct device_attribute *attr)
{
return container_of(attr, struct mce_bank, attr);
}
-static ssize_t show_bank(struct sys_device *s, struct sysdev_attribute *attr,
+static ssize_t show_bank(struct device *s, struct device_attribute *attr,
char *buf)
{
return sprintf(buf, "%llx\n", attr_to_bank(attr)->ctl);
}
-static ssize_t set_bank(struct sys_device *s, struct sysdev_attribute *attr,
+static ssize_t set_bank(struct device *s, struct device_attribute *attr,
const char *buf, size_t size)
{
u64 new;
}
static ssize_t
-show_trigger(struct sys_device *s, struct sysdev_attribute *attr, char *buf)
+show_trigger(struct device *s, struct device_attribute *attr, char *buf)
{
strcpy(buf, mce_helper);
strcat(buf, "\n");
return strlen(mce_helper) + 1;
}
-static ssize_t set_trigger(struct sys_device *s, struct sysdev_attribute *attr,
+static ssize_t set_trigger(struct device *s, struct device_attribute *attr,
const char *buf, size_t siz)
{
char *p;
return strlen(mce_helper) + !!p;
}
-static ssize_t set_ignore_ce(struct sys_device *s,
- struct sysdev_attribute *attr,
+static ssize_t set_ignore_ce(struct device *s,
+ struct device_attribute *attr,
const char *buf, size_t size)
{
u64 new;
if (mce_ignore_ce ^ !!new) {
if (new) {
/* disable ce features */
- on_each_cpu(mce_disable_ce, (void *)1, 1);
+ mce_timer_delete_all();
+ on_each_cpu(mce_disable_cmci, NULL, 1);
mce_ignore_ce = 1;
} else {
/* enable ce features */
return size;
}
-static ssize_t set_cmci_disabled(struct sys_device *s,
- struct sysdev_attribute *attr,
+static ssize_t set_cmci_disabled(struct device *s,
+ struct device_attribute *attr,
const char *buf, size_t size)
{
u64 new;
if (mce_cmci_disabled ^ !!new) {
if (new) {
/* disable cmci */
- on_each_cpu(mce_disable_ce, NULL, 1);
+ on_each_cpu(mce_disable_cmci, NULL, 1);
mce_cmci_disabled = 1;
} else {
/* enable cmci */
return size;
}
-static ssize_t store_int_with_restart(struct sys_device *s,
- struct sysdev_attribute *attr,
+static ssize_t store_int_with_restart(struct device *s,
+ struct device_attribute *attr,
const char *buf, size_t size)
{
- ssize_t ret = sysdev_store_int(s, attr, buf, size);
+ ssize_t ret = device_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);
-static SYSDEV_INT_ATTR(monarch_timeout, 0644, monarch_timeout);
-static SYSDEV_INT_ATTR(dont_log_ce, 0644, mce_dont_log_ce);
+static DEVICE_ATTR(trigger, 0644, show_trigger, set_trigger);
+static DEVICE_INT_ATTR(tolerant, 0644, tolerant);
+static DEVICE_INT_ATTR(monarch_timeout, 0644, monarch_timeout);
+static DEVICE_INT_ATTR(dont_log_ce, 0644, mce_dont_log_ce);
-static struct sysdev_ext_attribute attr_check_interval = {
- _SYSDEV_ATTR(check_interval, 0644, sysdev_show_int,
- store_int_with_restart),
+static struct dev_ext_attribute dev_attr_check_interval = {
+ __ATTR(check_interval, 0644, device_show_int, store_int_with_restart),
&check_interval
};
-static struct sysdev_ext_attribute attr_ignore_ce = {
- _SYSDEV_ATTR(ignore_ce, 0644, sysdev_show_int, set_ignore_ce),
+static struct dev_ext_attribute dev_attr_ignore_ce = {
+ __ATTR(ignore_ce, 0644, device_show_int, set_ignore_ce),
&mce_ignore_ce
};
-static struct sysdev_ext_attribute attr_cmci_disabled = {
- _SYSDEV_ATTR(cmci_disabled, 0644, sysdev_show_int, set_cmci_disabled),
+static struct dev_ext_attribute dev_attr_cmci_disabled = {
+ __ATTR(cmci_disabled, 0644, device_show_int, set_cmci_disabled),
&mce_cmci_disabled
};
-static struct sysdev_attribute *mce_attrs[] = {
- &attr_tolerant.attr,
- &attr_check_interval.attr,
- &attr_trigger,
- &attr_monarch_timeout.attr,
- &attr_dont_log_ce.attr,
- &attr_ignore_ce.attr,
- &attr_cmci_disabled.attr,
+static struct device_attribute *mce_device_attrs[] = {
+ &dev_attr_tolerant.attr,
+ &dev_attr_check_interval.attr,
+ &dev_attr_trigger,
+ &dev_attr_monarch_timeout.attr,
+ &dev_attr_dont_log_ce.attr,
+ &dev_attr_ignore_ce.attr,
+ &dev_attr_cmci_disabled.attr,
NULL
};
-static cpumask_var_t mce_dev_initialized;
+static cpumask_var_t mce_device_initialized;
-/* Per cpu sysdev init. All of the cpus still share the same ctrl bank: */
-static __cpuinit int mce_create_device(unsigned int cpu)
+static void mce_device_release(struct device *dev)
{
+ kfree(dev);
+}
+
+/* Per cpu device init. All of the cpus still share the same ctrl bank: */
+static __cpuinit int mce_device_create(unsigned int cpu)
+{
+ struct device *dev;
int err;
int i, j;
if (!mce_available(&boot_cpu_data))
return -EIO;
- 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;
+ dev = kzalloc(sizeof *dev, GFP_KERNEL);
+ if (!dev)
+ return -ENOMEM;
+ dev->id = cpu;
+ dev->bus = &mce_subsys;
+ dev->release = &mce_device_release;
- err = sysdev_register(&per_cpu(mce_dev, cpu));
+ err = device_register(dev);
if (err)
return err;
- for (i = 0; mce_attrs[i]; i++) {
- err = sysdev_create_file(&per_cpu(mce_dev, cpu), mce_attrs[i]);
+ for (i = 0; mce_device_attrs[i]; i++) {
+ err = device_create_file(dev, mce_device_attrs[i]);
if (err)
goto error;
}
for (j = 0; j < banks; j++) {
- err = sysdev_create_file(&per_cpu(mce_dev, cpu),
- &mce_banks[j].attr);
+ err = device_create_file(dev, &mce_banks[j].attr);
if (err)
goto error2;
}
- cpumask_set_cpu(cpu, mce_dev_initialized);
+ cpumask_set_cpu(cpu, mce_device_initialized);
+ per_cpu(mce_device, cpu) = dev;
return 0;
error2:
while (--j >= 0)
- sysdev_remove_file(&per_cpu(mce_dev, cpu), &mce_banks[j].attr);
+ device_remove_file(dev, &mce_banks[j].attr);
error:
while (--i >= 0)
- sysdev_remove_file(&per_cpu(mce_dev, cpu), mce_attrs[i]);
+ device_remove_file(dev, mce_device_attrs[i]);
- sysdev_unregister(&per_cpu(mce_dev, cpu));
+ device_unregister(dev);
return err;
}
-static __cpuinit void mce_remove_device(unsigned int cpu)
+static __cpuinit void mce_device_remove(unsigned int cpu)
{
+ struct device *dev = per_cpu(mce_device, cpu);
int i;
- if (!cpumask_test_cpu(cpu, mce_dev_initialized))
+ if (!cpumask_test_cpu(cpu, mce_device_initialized))
return;
- for (i = 0; mce_attrs[i]; i++)
- sysdev_remove_file(&per_cpu(mce_dev, cpu), mce_attrs[i]);
+ for (i = 0; mce_device_attrs[i]; i++)
+ device_remove_file(dev, mce_device_attrs[i]);
for (i = 0; i < banks; i++)
- sysdev_remove_file(&per_cpu(mce_dev, cpu), &mce_banks[i].attr);
+ device_remove_file(dev, &mce_banks[i].attr);
- sysdev_unregister(&per_cpu(mce_dev, cpu));
- cpumask_clear_cpu(cpu, mce_dev_initialized);
+ device_unregister(dev);
+ cpumask_clear_cpu(cpu, mce_device_initialized);
+ per_cpu(mce_device, cpu) = NULL;
}
/* Make sure there are no machine checks on offlined CPUs. */
unsigned long action = *(unsigned long *)h;
int i;
- if (!mce_available(¤t_cpu_data))
+ if (!mce_available(__this_cpu_ptr(&cpu_info)))
return;
if (!(action & CPU_TASKS_FROZEN))
unsigned long action = *(unsigned long *)h;
int i;
- if (!mce_available(¤t_cpu_data))
+ if (!mce_available(__this_cpu_ptr(&cpu_info)))
return;
if (!(action & CPU_TASKS_FROZEN))
switch (action) {
case CPU_ONLINE:
case CPU_ONLINE_FROZEN:
- mce_create_device(cpu);
+ mce_device_create(cpu);
if (threshold_cpu_callback)
threshold_cpu_callback(action, cpu);
break;
case CPU_DEAD_FROZEN:
if (threshold_cpu_callback)
threshold_cpu_callback(action, cpu);
- mce_remove_device(cpu);
+ mce_device_remove(cpu);
break;
case CPU_DOWN_PREPARE:
case CPU_DOWN_PREPARE_FROZEN:
for (i = 0; i < banks; i++) {
struct mce_bank *b = &mce_banks[i];
- struct sysdev_attribute *a = &b->attr;
+ struct device_attribute *a = &b->attr;
sysfs_attr_init(&a->attr);
a->attr.name = b->attrname;
if (!mce_available(&boot_cpu_data))
return -EIO;
- zalloc_cpumask_var(&mce_dev_initialized, GFP_KERNEL);
+ zalloc_cpumask_var(&mce_device_initialized, GFP_KERNEL);
mce_init_banks();
- err = sysdev_class_register(&mce_sysclass);
+ err = subsys_system_register(&mce_subsys, NULL);
if (err)
return err;
for_each_online_cpu(i) {
- err = mce_create_device(i);
+ err = mce_device_create(i);
if (err)
return err;
}
+ register_syscore_ops(&mce_syscore_ops);
register_hotcpu_notifier(&mce_cpu_notifier);
- misc_register(&mce_log_device);
+
+ /* register character device /dev/mcelog */
+ misc_register(&mce_chrdev_device);
+
+#ifdef CONFIG_X86_XEN_MCE
+ if (is_initial_xendomain()) {
+ /* Register vIRQ handler for MCE LOG processing */
+ extern int bind_virq_for_mce(void);
+
+ printk(KERN_DEBUG "MCE: bind virq for DOM0 logging\n");
+ bind_virq_for_mce();
+ }
+#endif
return err;
}
-
device_initcall(mcheck_init_device);
/*