2012-08-08 00:56:17

by Minchan Kim

Subject: [PATCH] compaction: fix deferring compaction mistake

[1] fixed bad deferring policy but made mistake about checking
compact_order_failed in __compact_pgdat so it can't update
compact_order_failed with new order. It ends up preventing working
of deffering policy rightly. This patch fixes it.

[1] aff62249, vmscan: only defer compaction for failed order and higher

Cc: Rik van Riel <[email protected]>
Cc: Mel Gorman <[email protected]>
Signed-off-by: Minchan Kim <[email protected]>
---
mm/compaction.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/compaction.c b/mm/compaction.c
index e78cb96..b6984e2 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -861,7 +861,7 @@ static int __compact_pgdat(pg_data_t *pgdat, struct compact_control *cc)
if (cc->order > 0) {
int ok = zone_watermark_ok(zone, cc->order,
low_wmark_pages(zone), 0, 0);
- if (ok && cc->order > zone->compact_order_failed)
+ if (ok && cc->order >= zone->compact_order_failed)
zone->compact_order_failed = cc->order + 1;
/* Currently async compaction is never deferred. */
else if (!ok && cc->sync)
--
1.7.9.5

2012-08-08 14:59:47

by Rik van Riel

Subject: Re: [PATCH] compaction: fix deferring compaction mistake

On 08/07/2012 08:57 PM, Minchan Kim wrote:
> [1] fixed bad deferring policy but made mistake about checking
> compact_order_failed in __compact_pgdat so it can't update
> compact_order_failed with new order. It ends up preventing working
> of deffering policy rightly. This patch fixes it.

Good catch.

> [1] aff62249, vmscan: only defer compaction for failed order and higher
>
> Cc: Rik van Riel <[email protected]>
> Cc: Mel Gorman <[email protected]>
> Signed-off-by: Minchan Kim <[email protected]>

Reviewed-by: Rik van Riel <[email protected]>

2012-08-09 13:50:58

by Mel Gorman

Subject: Re: [PATCH] compaction: fix deferring compaction mistake

On Wed, Aug 08, 2012 at 09:57:44AM +0900, Minchan Kim wrote:
> [1] fixed bad deferring policy but made mistake about checking
> compact_order_failed in __compact_pgdat so it can't update
> compact_order_failed with new order. It ends up preventing working
> of deffering policy rightly. This patch fixes it.
>
> [1] aff62249, vmscan: only defer compaction for failed order and higher
>
> Cc: Rik van Riel <[email protected]>
> Cc: Mel Gorman <[email protected]>
> Signed-off-by: Minchan Kim <[email protected]>

Acked-by: Mel Gorman <[email protected]>

--
Mel Gorman
SUSE Labs