2013-06-17 22:06:34

by Zhang Yanfei

Subject: [PATCH] mm: Remove unlikely from the current_order test

From: Zhang Yanfei <[email protected]>

In __rmqueue_fallback(), current_order loops down from MAX_ORDER - 1
to the order passed. MAX_ORDER is typically 11 and pageblock_order
is typically 9 on x86. Integer division truncates, so pageblock_order / 2
is 4. For the first eight iterations, it's guaranteed that
current_order >= pageblock_order / 2 if it even gets that far!

So just remove the unlikely(), it's completely bogus.

Suggested-by: David Rientjes <[email protected]>
Signed-off-by: Zhang Yanfei <[email protected]>
---
mm/page_alloc.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index c3edb62..7b4f367 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1050,7 +1050,7 @@ __rmqueue_fallback(struct zone *zone, int order, int start_migratetype)
* MIGRATE_CMA areas.
*/
if (!is_migrate_cma(migratetype) &&
- (unlikely(current_order >= pageblock_order / 2) ||
+ (current_order >= pageblock_order / 2 ||
start_migratetype == MIGRATE_RECLAIMABLE ||
page_group_by_mobility_disabled)) {
int pages;
--
1.7.1

2013-06-17 23:56:21

by David Rientjes

Subject: Re: [PATCH] mm: Remove unlikely from the current_order test

On Tue, 18 Jun 2013, Zhang Yanfei wrote:

> From: Zhang Yanfei <[email protected]>
>
> In __rmqueue_fallback(), current_order loops down from MAX_ORDER - 1
> to the order passed. MAX_ORDER is typically 11 and pageblock_order
> is typically 9 on x86. Integer division truncates, so pageblock_order / 2
> is 4. For the first eight iterations, it's guaranteed that
> current_order >= pageblock_order / 2 if it even gets that far!
>
> So just remove the unlikely(), it's completely bogus.
>
> Suggested-by: David Rientjes <[email protected]>
> Signed-off-by: Zhang Yanfei <[email protected]>

Acked-by: David Rientjes <[email protected]>