From: David Rientjes Date: Tue, 22 Sep 2009 00:01:34 +0000 (-0700) Subject: mm: update alloc_flags after oom killer has been called X-Git-Tag: v2.6.32-rc1~577 X-Git-Url: https://openfabrics.org/gitweb/?a=commitdiff_plain;h=b259fbde0a86085264c89aa2ce9c6e35792a1aad;p=~shefty%2Frdma-dev.git mm: update alloc_flags after oom killer has been called It is possible for the oom killer to select current as the task to kill. When this happens, alloc_flags needs to be updated accordingly to set ALLOC_NO_WATERMARKS so the subsequent allocation attempt may use memory reserves as the result of its thread having TIF_MEMDIE set if the allocation is not __GFP_NOMEMALLOC. Acked-by: Mel Gorman Signed-off-by: David Rientjes Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- diff --git a/mm/page_alloc.c b/mm/page_alloc.c index e50c22545b8..b6d0d09557e 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1771,6 +1771,7 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order, wake_all_kswapd(order, zonelist, high_zoneidx); +restart: /* * OK, we're below the kswapd watermark and have kicked background * reclaim. Now things get more complex, so set up alloc_flags according @@ -1778,7 +1779,6 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order, */ alloc_flags = gfp_to_alloc_flags(gfp_mask); -restart: /* This is the last chance, in general, before the goto nopage. */ page = get_page_from_freelist(gfp_mask, nodemask, order, zonelist, high_zoneidx, alloc_flags & ~ALLOC_NO_WATERMARKS,