Re: vm lock contention reduction

[Rik van Riel <riel@conectiva.com.br>]

On Fri, 5 Jul 2002, Andrew Morton wrote:
> William Lee Irwin III wrote:
> > On Thu, Jul 04, 2002 at 07:18:34PM -0700, Andrew Morton wrote:
> > > Of course, that change means that we wouldn't be able to throttle
> > > page allocators against IO any more, and we'd have to do something
> > > smarter.  What a shame ;)
> >
> > This is actually necessary IMHO. Some testing I've been able to do seems
> > to reveal the current throttling mechanism as inadequate.
>
> I don't think so.  If you're referring to the situation where your
> 4G machine had 3.5G dirty pages without triggering writeback.
>
> That's not a generic problem.

But it is, mmap() and anonymous memory don't trigger writeback.

regards,

Rik
-- 
Bravely reimplemented by the knights who say "NIH".

http://www.surriel.com/		http://distro.conectiva.com/

--
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/