#include <linux/fs.h>
#include <linux/mm.h>
#include <linux/debugfs.h>
+#include <linux/edac_mce.h>
#include <asm/processor.h>
#include <asm/hw_irq.h>
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.
* Assume that the earlier errors are the more
* interesting ones:
static void mce_panic(char *msg, struct mce *final, char *exp)
{
- int i;
+ int i, apei_err = 0;
if (!fake_panic) {
/*
struct mce *m = &mcelog.entry[i];
if (!(m->status & MCI_STATUS_VAL))
continue;
- if (!(m->status & MCI_STATUS_UC))
+ if (!(m->status & MCI_STATUS_UC)) {
print_mce(m);
+ if (!apei_err)
+ apei_err = apei_write_mce(m);
+ }
}
/* Now print uncorrected but with the final one last */
for (i = 0; i < MCE_LOG_LEN; i++) {
continue;
if (!(m->status & MCI_STATUS_UC))
continue;
- if (!final || memcmp(m, final, sizeof(struct mce)))
+ if (!final || memcmp(m, final, sizeof(struct mce))) {
print_mce(m);
+ if (!apei_err)
+ apei_err = apei_write_mce(m);
+ }
}
- if (final)
+ if (final) {
print_mce(final);
+ if (!apei_err)
+ apei_err = apei_write_mce(final);
+ }
if (cpu_missing)
printk(KERN_EMERG "Some CPUs didn't answer in synchronization\n");
print_mce_tail();
*/
asmlinkage void smp_mce_self_interrupt(struct pt_regs *regs)
{
-#ifndef CONFIG_XEN
ack_APIC_irq();
-#endif
exit_idle();
irq_enter();
mce_notify_irq();
return;
}
-#if defined(CONFIG_X86_LOCAL_APIC) && !defined(CONFIG_XEN)
+#ifdef CONFIG_X86_LOCAL_APIC
/*
* Without APIC do not notify. The event will be picked
* up eventually.
struct mce m;
int i;
- __get_cpu_var(mce_poll_count)++;
+ percpu_inc(mce_poll_count);
mce_setup(&m);
atomic_inc(&mce_entry);
- __get_cpu_var(mce_exception_count)++;
+ percpu_inc(mce_exception_count);
if (notify_die(DIE_NMI, "machine check", regs, error_code,
18, SIGKILL) == NOTIFY_STOP)
* 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);
/* 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
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)
rdtscll(cpu_tsc[smp_processor_id()]);
}
+static int mce_apei_read_done;
+
+/* Collect MCE record of previous boot in persistent storage via APEI ERST. */
+static int __mce_read_apei(char __user **ubuf, size_t usize)
+{
+ int rc;
+ u64 record_id;
+ struct mce m;
+
+ if (usize < sizeof(struct mce))
+ return -EINVAL;
+
+ rc = apei_read_mce(&m, &record_id);
+ /* Error or no more MCE record */
+ if (rc <= 0) {
+ mce_apei_read_done = 1;
+ return rc;
+ }
+ rc = -EFAULT;
+ if (copy_to_user(*ubuf, &m, sizeof(struct mce)))
+ return rc;
+ /*
+ * In fact, we should have cleared the record after that has
+ * been flushed to the disk or sent to network in
+ * /sbin/mcelog, but we have no interface to support that now,
+ * so just clear it to avoid duplication.
+ */
+ rc = apei_clear_mce(record_id);
+ if (rc) {
+ mce_apei_read_done = 1;
+ return rc;
+ }
+ *ubuf += sizeof(struct mce);
+
+ return 0;
+}
+
static ssize_t mce_read(struct file *filp, char __user *ubuf, size_t usize,
loff_t *off)
{
return -ENOMEM;
mutex_lock(&mce_read_mutex);
+
+ if (!mce_apei_read_done) {
+ err = __mce_read_apei(&buf, usize);
+ if (err || buf != ubuf)
+ goto out;
+ }
+
next = rcu_dereference_check_mce(mcelog.next);
/* Only supports full reads right now */
- if (*off != 0 || usize < MCE_LOG_LEN*sizeof(struct mce)) {
- mutex_unlock(&mce_read_mutex);
- kfree(cpu_tsc);
-
- return -EINVAL;
- }
+ err = -EINVAL;
+ if (*off != 0 || usize < MCE_LOG_LEN*sizeof(struct mce))
+ goto out;
err = 0;
prev = 0;
memset(&mcelog.entry[i], 0, sizeof(struct mce));
}
}
+
+ if (err)
+ err = -EFAULT;
+
+out:
mutex_unlock(&mce_read_mutex);
kfree(cpu_tsc);
- return err ? -EFAULT : buf - ubuf;
+ return err ? err : buf - ubuf;
}
static unsigned int mce_poll(struct file *file, poll_table *wait)
poll_wait(file, &mce_wait, wait);
if (rcu_dereference_check_mce(mcelog.next))
return POLLIN | POLLRDNORM;
+ if (!mce_apei_read_done && apei_check_mce())
+ return POLLIN | POLLRDNORM;
return 0;
}
register_hotcpu_notifier(&mce_cpu_notifier);
misc_register(&mce_log_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;
}