[PATCH] vmscan: no zone_reclaim if PF_MALLOC is set
authorChristoph Lameter <clameter@sgi.com>
Fri, 10 Mar 2006 01:33:47 +0000 (17:33 -0800)
committerLinus Torvalds <torvalds@g5.osdl.org>
Fri, 10 Mar 2006 03:47:37 +0000 (19:47 -0800)
If the process has already set PF_MALLOC and is already using
current->reclaim_state then do not try to reclaim memory from the zone.
This is set by kswapd and/or synchrononous global reclaim which will not
take it lightly if we zap the reclaim_state.

Signed-off-by: Christoph Lameter <clameter@sig.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
mm/vmscan.c

index b0af7593d01e315a83c79ec6841c9a4a3b91c1e1..7ccf763bb30bf37b71fdd0dd25b135c26a2b6587 100644 (file)
@@ -1883,7 +1883,8 @@ int zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order)
 
        if (!(gfp_mask & __GFP_WAIT) ||
                zone->all_unreclaimable ||
-               atomic_read(&zone->reclaim_in_progress) > 0)
+               atomic_read(&zone->reclaim_in_progress) > 0 ||
+               (p->flags & PF_MEMALLOC))
                        return 0;
 
        node_id = zone->zone_pgdat->node_id;