diff mbox

[-mm] mm, compaction: properly signal and act upon lock and need_sched() contention - fix

Message ID 538C8B45.6070803@suse.cz (mailing list archive)
State New, archived
Headers show

Commit Message

Vlastimil Babka June 2, 2014, 2:33 p.m. UTC 
compact_should_abort() returns true instead of false and vice versa
due to changes between v1 and v2 of the patch. This makes both async
and sync compaction abort with high probability, and has been reported
to cause e.g. soft lockups on some ARM boards, or drivers calling
dma_alloc_coherent() fail to probe with CMA enabled on different boards.

This patch fixes the return value to match comments and callers expecations.

Reported-and-tested-by: Kevin Hilman <khilman@linaro.org>
Reported-and-tested-by: Shawn Guo <shawn.guo@linaro.org>
Tested-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
---
 mm/compaction.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

-- 1.8.4.5

Comments

Fabio Estevam June 2, 2014, 3:18 p.m. UTC | #1 
On Mon, Jun 2, 2014 at 11:33 AM, Vlastimil Babka <vbabka@suse.cz> wrote:
> compact_should_abort() returns true instead of false and vice versa
> due to changes between v1 and v2 of the patch. This makes both async
> and sync compaction abort with high probability, and has been reported
> to cause e.g. soft lockups on some ARM boards, or drivers calling
> dma_alloc_coherent() fail to probe with CMA enabled on different boards.
>
> This patch fixes the return value to match comments and callers expecations.
>
> Reported-and-tested-by: Kevin Hilman <khilman@linaro.org>
> Reported-and-tested-by: Shawn Guo <shawn.guo@linaro.org>
> Tested-by: Stephen Warren <swarren@nvidia.com>
> Signed-off-by: Vlastimil Babka <vbabka@suse.cz>

Tested-by: Fabio Estevam <fabio.estevam@freescale.com>
David Rientjes June 2, 2014, 8:09 p.m. UTC | #2 
On Mon, 2 Jun 2014, Vlastimil Babka wrote:

> compact_should_abort() returns true instead of false and vice versa
> due to changes between v1 and v2 of the patch. This makes both async
> and sync compaction abort with high probability, and has been reported
> to cause e.g. soft lockups on some ARM boards, or drivers calling
> dma_alloc_coherent() fail to probe with CMA enabled on different boards.
> 
> This patch fixes the return value to match comments and callers expecations.
> 
> Reported-and-tested-by: Kevin Hilman <khilman@linaro.org>
> Reported-and-tested-by: Shawn Guo <shawn.guo@linaro.org>
> Tested-by: Stephen Warren <swarren@nvidia.com>
> Signed-off-by: Vlastimil Babka <vbabka@suse.cz>

Acked-by: David Rientjes <rientjes@google.com>
diff mbox

Patch

diff --git a/mm/compaction.c b/mm/compaction.c
index a525cd4..5175019 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -237,13 +237,13 @@  static inline bool compact_should_abort(struct compact_control *cc)
 	if (need_resched()) {
 		if (cc->mode == MIGRATE_ASYNC) {
 			cc->contended = true;
-			return false;
+			return true;
 		}
 
 		cond_resched();
 	}
 
-	return true;
+	return false;
 }
 
 /* Returns true if the page is within a block suitable for migration to */