mm/page_alloc.c: skip ZONE_MOVABLE if required_kernelcore is larger than totalpages
authorXishi Qiu <qiuxishi@huawei.com>
Fri, 6 Nov 2015 02:48:56 +0000 (18:48 -0800)
committerLinus Torvalds <torvalds@linux-foundation.org>
Fri, 6 Nov 2015 03:34:48 +0000 (19:34 -0800)
If kernelcore was not specified, or the kernelcore size is zero
(required_movablecore >= totalpages), or the kernelcore size is larger
than totalpages, there is no ZONE_MOVABLE.  We should fill the zone with
both kernel memory and movable memory.

Signed-off-by: Xishi Qiu <qiuxishi@huawei.com>
Reviewed-by: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: David Rientjes <rientjes@google.com>
Cc: Tang Chen <tangchen@cn.fujitsu.com>
Cc: <zhongjiang@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/page_alloc.c

index 86f7d95..06e6230 100644 (file)
@@ -5675,8 +5675,11 @@ static void __init find_zone_movable_pfns_for_nodes(void)
                required_kernelcore = max(required_kernelcore, corepages);
        }
 
-       /* If kernelcore was not specified, there is no ZONE_MOVABLE */
-       if (!required_kernelcore)
+       /*
+        * If kernelcore was not specified or kernelcore size is larger
+        * than totalpages, there is no ZONE_MOVABLE.
+        */
+       if (!required_kernelcore || required_kernelcore >= totalpages)
                goto out;
 
        /* usable_startpfn is the lowest possible pfn ZONE_MOVABLE can be at */