On Thu, 24 Mar 2011, Pekka Enberg wrote:
> It hanged here which is pretty much expected on this box if
> kmem_cache_init() oopses. I'm now trying to see if I'm able to find
> the config option that breaks things. CONFIG_PREEMPT_NONE is a
> suspect:
>
> penberg@tiger:~/linux$ grep PREEMPT ../config-ingo
> # CONFIG_PREEMPT_RCU is not set
> CONFIG_PREEMPT_NONE=y
> # CONFIG_PREEMPT_VOLUNTARY is not set
> # CONFIG_PREEMPT is not set
The following patch should ensure that all percpu data is touched
before any emulation functions are called:
---
mm/slub.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
Index: linux-2.6/mm/slub.c
===================================================================
--- linux-2.6.orig/mm/slub.c 2011-03-24 14:03:10.000000000 -0500
+++ linux-2.6/mm/slub.c 2011-03-24 14:04:08.000000000 -0500
@@ -1604,7 +1604,7 @@ static inline void note_cmpxchg_failure(
void init_kmem_cache_cpus(struct kmem_cache *s)
{
-#if defined(CONFIG_CMPXCHG_LOCAL) && defined(CONFIG_PREEMPT)
+#ifdef CONFIG_CMPXCHG_LOCAL
int cpu;
for_each_possible_cpu(cpu)
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxxx For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>