Re: [PATCH] mm: compaction: Abort compaction if too many pages areisolated and caller is asynchronous

[Minchan Kim]

Hi Kame,

On Tue, May 31, 2011 at 01:48:35PM +0900, KAMEZAWA Hiroyuki wrote:
> On Mon, 30 May 2011 14:13:00 +0100
> Mel Gorman <mel@xxxxxxxxx> wrote:
> 
> > Asynchronous compaction is used when promoting to huge pages. This is
> > all very nice but if there are a number of processes in compacting
> > memory, a large number of pages can be isolated. An "asynchronous"
> > process can stall for long periods of time as a result with a user
> > reporting that firefox can stall for 10s of seconds. This patch aborts
> > asynchronous compaction if too many pages are isolated as it's better to
> > fail a hugepage promotion than stall a process.
> > 
> > If accepted, this should also be considered for 2.6.39-stable. It should
> > also be considered for 2.6.38-stable but ideally [11bc82d6: mm:
> > compaction: Use async migration for __GFP_NO_KSWAPD and enforce no
> > writeback] would be applied to 2.6.38 before consideration.
> > 
> > Reported-and-Tested-by: Ury Stankevich <urykhy@xxxxxxxxx>
> > Signed-off-by: Mel Gorman <mgorman@xxxxxxx>
> 
> 
> 
> Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx>
> 
> BTW, I'm surprised to see both of vmscan.c and compaction.c has too_many_isolated()..
> in different logic ;)
> 
> BTW, compaction ignores UNEVICTABLE LRU ?

Good point.
Yes. now compaction doesn't work with unevictable LRU but I think we have no reason
to work well with unveictable pages. 
If we don't support unevictable lru, it would be a problem in lots of
mlocked pages workload.
It would be a good enhance point on compaction.

> 
> Thanks,
> -Kame
-- 
Kind regards
Minchan Kim
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/