[PATCH 16/31] x86: MCE: Implement panic synchronization

From: Andi Kleen
Date: Tue May 26 2009 - 20:03:02 EST

From: Andi Kleen <ak@xxxxxxxxxxxxxxx>

Impact: bugfix

In some circumstances multiple CPUs can enter mce_panic() in parallel. This
gives quite confused output because they will all dump the same machine check

The other problem is that they would all panic in parallel, but not process
each other's shutdown IPIs because interrupts are disabled.

Detect this situation early on in mce_panic(). On the first CPU
entering will do the panic, the others will just wait to be killed.

For paranoia reasons in case the other CPU dies during the MCE I added
a 5 seconds timeout. If it expires each CPU will panic on its own again,

Signed-off-by: Andi Kleen <ak@xxxxxxxxxxxxxxx>
arch/x86/kernel/cpu/mcheck/mce.c | 22 ++++++++++++++++++++++
1 files changed, 22 insertions(+), 0 deletions(-)

diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c
index a61eead..65c62d1 100644
--- a/arch/x86/kernel/cpu/mcheck/mce.c
+++ b/arch/x86/kernel/cpu/mcheck/mce.c
@@ -182,10 +182,32 @@ static void print_mce(struct mce *m)
"and contact your hardware vendor\n");

+#define PANIC_TIMEOUT 5 /* 5 seconds */
+static atomic_t mce_paniced;
+/* Panic in progress. Enable interrupts and wait for final IPI */
+static void wait_for_panic(void)
+ long timeout = PANIC_TIMEOUT*USEC_PER_SEC;
+ preempt_disable();
+ local_irq_enable();
+ while (timeout-- > 0)
+ udelay(1);
+ panic("Panicing machine check CPU died");
static void mce_panic(char *msg, struct mce *final, char *exp)
int i;

+ /*
+ * Make sure only one CPU runs in machine check panic
+ */
+ if (atomic_add_return(1, &mce_paniced) > 1)
+ wait_for_panic();
+ barrier();
/* First print corrected ones that are still unlogged */

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/