page-allocator: allow too high-order warning messages to be suppressed with __GFP_NOWARN
authorMel Gorman <mel@csn.ul.ie>
Wed, 29 Jul 2009 22:04:08 +0000 (15:04 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Thu, 30 Jul 2009 02:10:35 +0000 (19:10 -0700)
The page allocator warns once when an order >= MAX_ORDER is specified.
This is to catch callers of the allocator that are always falling back to
their worst-case when it was not expected.  However, there are cases where
the caller is behaving correctly but cannot suppress the warning.  This
patch allows the warning to be suppressed by the callers by specifying
__GFP_NOWARN.

Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Acked-by: David Rientjes <rientjes@google.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/page_alloc.c

index 2dbb2fc..d052abb 100644 (file)
@@ -1745,8 +1745,10 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
         * be using allocators in order of preference for an area that is
         * too large.
         */
-       if (WARN_ON_ONCE(order >= MAX_ORDER))
+       if (order >= MAX_ORDER) {
+               WARN_ON_ONCE(!(gfp_mask & __GFP_NOWARN));
                return NULL;
+       }
 
        /*
         * GFP_THISNODE (meaning __GFP_THISNODE, __GFP_NORETRY and