Mel Gorman <mgor...@suse.de> wrote:
> When I looked at it for long enough I found a number of problems. Most
> affect timing but two serious issues are in there. One affects how long
> kswapd spends compacting versus reclaiming and the other increases lock
> contention meaning that async compaction can abort early. Both are serious
> and could explain why a driver would fail high-order allocations.
>
> Please try the following patch. However, even if it works the benefit of
> capture may be so marginal that partially reverting it and simplifying
> compaction.c is the better decision.
Btw, I'm still testing this patch with the "page->pfemalloc = false"
change on top of it.
> diff --git a/mm/compaction.c b/mm/compaction.c
> index 6b807e4..03c82c0 100644
> --- a/mm/compaction.c
> +++ b/mm/compaction.c
> @@ -857,7 +857,8 @@ static int compact_finished(struct zone *zone,
> } else {
> unsigned int order;
> for (order = cc->order; order < MAX_ORDER; order++) {
> - struct free_area *area = &zone->free_area[cc->order];
> + struct free_area *area = &zone->free_area[order];
I noticed something like this hunk wasn't in your latest partial revert
(<20130109135010.gb13...@suse.de>)
I admit I don't understand this code, but this jumped out at me.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
