#include <linux/export.h>
#include <asm/processor.h>
+#include <asm/traps.h>
#include <asm/mce.h>
#include <asm/msr.h>
* 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);
+static ATOMIC_NOTIFIER_HEAD(x86_mce_decoder_chain);
/* Do initial initialization of a struct mce */
void mce_setup(struct mce *m)
panic("Panicing machine check CPU died");
}
-static void mce_panic(char *msg, struct mce *final, char *exp)
+static void mce_panic(const char *msg, struct mce *final, char *exp)
{
int i, apei_err = 0;
schedule_work(this_cpu_ptr(&mce_work));
}
-DEFINE_PER_CPU(struct irq_work, mce_irq_work);
+static DEFINE_PER_CPU(struct irq_work, mce_irq_work);
static void mce_irq_work_cb(struct irq_work *entry)
{
/*
* Check if a timeout waiting for other CPUs happened.
*/
-static int mce_timed_out(u64 *t)
+static int mce_timed_out(u64 *t, const char *msg)
{
/*
* The others already did panic for some reason.
goto out;
if ((s64)*t < SPINUNIT) {
if (mca_cfg.tolerant <= 1)
- mce_panic("Timeout synchronizing machine check over CPUs",
- NULL, NULL);
+ mce_panic(msg, NULL, NULL);
cpu_missing = 1;
return 1;
}
* Wait for everyone.
*/
while (atomic_read(&mce_callin) != cpus) {
- if (mce_timed_out(&timeout)) {
+ if (mce_timed_out(&timeout,
+ "Timeout: Not all CPUs entered broadcast exception handler")) {
atomic_set(&global_nwo, 0);
return -1;
}
* only seen by one CPU before cleared, avoiding duplicates.
*/
while (atomic_read(&mce_executing) < order) {
- if (mce_timed_out(&timeout)) {
+ if (mce_timed_out(&timeout,
+ "Timeout: Subject CPUs unable to finish machine check processing")) {
atomic_set(&global_nwo, 0);
return -1;
}
* loops.
*/
while (atomic_read(&mce_executing) <= cpus) {
- if (mce_timed_out(&timeout))
+ if (mce_timed_out(&timeout,
+ "Timeout: Monarch CPU unable to finish machine check processing"))
goto reset;
ndelay(SPINUNIT);
}
* Subject: Wait for Monarch to finish.
*/
while (atomic_read(&mce_executing) != 0) {
- if (mce_timed_out(&timeout))
+ if (mce_timed_out(&timeout,
+ "Timeout: Monarch CPU did not finish machine check processing"))
goto reset;
ndelay(SPINUNIT);
}
}
}
-/*
- * 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.
{
struct mca_config *cfg = &mca_cfg;
struct mce m, *final;
+ enum ctx_state prev_state;
int i;
int worst = 0;
int severity;
DECLARE_BITMAP(toclear, MAX_NR_BANKS);
DECLARE_BITMAP(valid_banks, MAX_NR_BANKS);
char *msg = "Unknown";
+ u64 recover_paddr = ~0ull;
+ int flags = MF_ACTION_REQUIRED;
+
+ prev_state = ist_enter(regs);
this_cpu_inc(mce_exception_count);
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);
+ recover_paddr = m.addr;
+ if (!(m.mcgstatus & MCG_STATUS_RIPV))
+ flags |= MF_MUST_KILL;
} else if (kill_it) {
force_sig(SIGBUS, current);
}
mce_wrmsrl(MSR_IA32_MCG_STATUS, 0);
out:
sync_core();
+
+ if (recover_paddr == ~0ull)
+ goto done;
+
+ pr_err("Uncorrected hardware memory error in user-access at %llx",
+ recover_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.
+ */
+ ist_begin_non_atomic(regs);
+ local_irq_enable();
+ if (memory_failure(recover_paddr >> PAGE_SHIFT, MCE_VECTOR, flags) < 0) {
+ pr_err("Memory error not recovered");
+ force_sig(SIGBUS, current);
+ }
+ local_irq_disable();
+ ist_end_non_atomic();
+done:
+ ist_exit(regs, prev_state);
}
EXPORT_SYMBOL_GPL(do_machine_check);
}
#endif
-/*
- * 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;
- struct mce_info *mi = mce_find_info();
- int flags = MF_ACTION_REQUIRED;
-
- 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 (!mi->restartable)
- flags |= MF_MUST_KILL;
- if (memory_failure(pfn, MCE_VECTOR, flags) < 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()
projects / firefly-linux-kernel-4.4.55.git / blobdiff