spinlock: Indicate that a lockup is only suspected
authorChristian Borntraeger <borntraeger@de.ibm.com>
Tue, 29 May 2012 09:18:44 +0000 (11:18 +0200)
committerIngo Molnar <mingo@kernel.org>
Wed, 6 Jun 2012 09:34:20 +0000 (11:34 +0200)
On an over-committed KVM system we got a:

  "BUG: spinlock lockup on CPU#2, swapper/2/0"

message on the heavily contended virtio blk spinlock.

While we might want to reconsider the locking of virtio-blk
(lock is held while switching to the host) this patch tries to
make the message clearer: the lockup is only suspected.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1338283124-7063-1-git-send-email-borntraeger@de.ibm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
lib/spinlock_debug.c

index d0ec4f3d1593031b5498dcc120822b41c423a3c5..e91fbc23fff121915217a8e5c53c29a2b52aeba4 100644 (file)
@@ -118,7 +118,7 @@ static void __spin_lock_debug(raw_spinlock_t *lock)
                /* lockup suspected: */
                if (print_once) {
                        print_once = 0;
-                       spin_dump(lock, "lockup");
+                       spin_dump(lock, "lockup suspected");
 #ifdef CONFIG_SMP
                        trigger_all_cpu_backtrace();
 #endif
This page took 0.024354 seconds and 5 git commands to generate.