[PATCH] fix BUG_ON(!PageSlab) from fallback_alloc
authorHugh Dickins <hugh@veritas.com>
Sat, 6 Jan 2007 00:36:36 +0000 (16:36 -0800)
committerLinus Torvalds <torvalds@woody.osdl.org>
Sat, 6 Jan 2007 07:55:23 +0000 (23:55 -0800)
pdflush hit the BUG_ON(!PageSlab(page)) in kmem_freepages called from
fallback_alloc: cache_grow already freed those pages when alloc_slabmgmt
failed.  But it wouldn't have freed them if __GFP_NO_GROW, so make sure
fallback_alloc doesn't waste its time on that case.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Acked-by: Christoph Lameter <clameter@sgi.com>
Acked-by: Pekka J Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
mm/slab.c

index 0d4e57431de41c8d7fccfe9307bfb504087e5c03..c6100628a6ef7be2325a55e18f69b80a5a0f2006 100644 (file)
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -3281,7 +3281,7 @@ retry:
                                        flags | GFP_THISNODE, nid);
        }
 
-       if (!obj) {
+       if (!obj && !(flags & __GFP_NO_GROW)) {
                /*
                 * This allocation will be performed within the constraints
                 * of the current cpuset / memory policy requirements.
@@ -3310,7 +3310,7 @@ retry:
                                         */
                                        goto retry;
                        } else {
-                               kmem_freepages(cache, obj);
+                               /* cache_grow already freed obj */
                                obj = NULL;
                        }
                }