[PATCH][KDUMP] Ignore spurious IPI

From: Takao Indoh
Date: Wed Mar 23 2011 - 14:40:53 EST

Hi all,

I found a problem that kdump(2nd kernel) sometimes hangs up. It seems
that system panic occurs as follows.

2nd kernel boot up

A pending IPI from 1st kernel comes after unmasking interrupts at the
following point.

asmlinkage void __init start_kernel(void)
if (!irqs_disabled())
printk(KERN_CRIT "start_kernel(): bug: interrupts were "
"enabled early\n");
early_boot_irqs_disabled = false;
local_irq_enable(); <=======================================HERE

Kernel tries to handle the interrupt, but some data structures are not
initialized yet at this point. As a result, in the
generic_smp_call_function_single_interrupt(), NULL pointer dereference
occurs when list_replace_init() tries to access &q->list.next.

I took a look at local_apic_timer_interrupt() and found a few lines to
handle such a pending LAPIC interrupt(in this case, timer interrupt).
Therefore I made a patch to ignore spurious IPI in the same manner. I
confirmed this problem does not occur with this patch.

Any comments?

Signed-off-by: Takao Indoh <indou.takao@xxxxxxxxxxxxxx>
kernel/smp.c | 6 ++++++
1 file changed, 6 insertions(+)

diff --git a/kernel/smp.c b/kernel/smp.c
index 9910744..f2f561b 100644
--- a/kernel/smp.c
+++ b/kernel/smp.c
@@ -260,6 +260,12 @@ void generic_smp_call_function_single_interrupt(void)

+ if (unlikely(!q->list.next)) {
+ /* Pending interrupt from previous kernel(e.g. kdump), just ignore */
+ pr_warning("Spurious IPI on cpu %d\n", smp_processor_id());
+ return;
+ }
list_replace_init(&q->list, &list);
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/