Re: [PATCH][BUGFIX] memcg fix zone congestion

[KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>]

On Fri, 13 May 2011 13:15:14 +0900
Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp> wrote:

> On Fri, 13 May 2011 12:10:30 +0900
> KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> wrote:
> 
> > 
> > ZONE_CONGESTED should be a state of global memory reclaim.
> > If not, a busy memcg sets this and give unnecessary throttoling in
> > wait_iff_congested() against memory recalim in other contexts. This makes
> > system performance bad.
> > 
> hmm, nice catch.
> 
> Just from my curiosity, is there any number of performance improvement by this patch?
> 

I guess impact of this patch is very limited. (And I've tested but cannot measure
the score. I need some special application/settings.)

I think, with small memcg, we'll see nr_dirty==nr_congested situaion often and
set ZONE_CONGESTED. But the effect of patch can be seen in the another place.

To see the effect of patch, the system need to reclaim memory eagerly with memory
shortage. In that situation, everything is slow. This patch just removes a extra
burden which is put on by mistake.


Thanks,
-Kame

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>