[PATCH] cpuset: oom panic fix
authorNick Piggin <npiggin@suse.de>
Sun, 27 Aug 2006 08:23:54 +0000 (01:23 -0700)
committerLinus Torvalds <torvalds@g5.osdl.org>
Sun, 27 Aug 2006 18:01:32 +0000 (11:01 -0700)
cpuset_excl_nodes_overlap always returns 0 if current is exiting.  This caused
customer's systems to panic in the OOM killer when processes were having
trouble getting memory for the final put_user in mm_release.  Even though
there were lots of processes to kill.

Change to returning 1 in this case.  This achieves parity with !CONFIG_CPUSETS
case, and was observed to fix the problem.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Acked-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
kernel/cpuset.c

index f1dda98bdffe64ada58aaec4d0de877709e5215a..4ea6f0dc2fc568b35ab147903de0a8e31d9e8429 100644 (file)
@@ -2420,7 +2420,7 @@ EXPORT_SYMBOL_GPL(cpuset_mem_spread_node);
 int cpuset_excl_nodes_overlap(const struct task_struct *p)
 {
        const struct cpuset *cs1, *cs2; /* my and p's cpuset ancestors */
-       int overlap = 0;                /* do cpusets overlap? */
+       int overlap = 1;                /* do cpusets overlap? */
 
        task_lock(current);
        if (current->flags & PF_EXITING) {