Re: [patch 1/13] minimal rmap

From: Rik van Riel <riel@conectiva.com.br>
To: Daniel Phillips <phillips@arcor.de>
Cc: Andrew Morton <akpm@zip.com.au>,
	Linus Torvalds <torvalds@transmeta.com>,
	lkml <linux-kernel@vger.kernel.org>
Subject: Re: [patch 1/13] minimal rmap
Date: Wed, 17 Jul 2002 16:41:01 -0300 (BRT)	[thread overview]
Message-ID: <Pine.LNX.4.44L.0207171639480.12241-100000@imladris.surriel.com> (raw)
In-Reply-To: <E17Uud4-0004PN-00@starship>

On Wed, 17 Jul 2002, Daniel Phillips wrote:
> On Wednesday 17 July 2002 21:31, Rik van Riel wrote:
> > On Wed, 17 Jul 2002, Daniel Phillips wrote:
> > > On Wednesday 17 July 2002 07:29, Andrew Morton wrote:
> > > > 11: The nightly updatedb run is still evicting everything.
> > >
> > > That is not a problem with rmap per se, it's a result of not properly
> > > handling streaming IO.
> >
> > Umm, updatedb isn't exactly streaming...
>
> You're right, it's not exactly, it's hitting every directory entry on
                                                     ^^^^^^^^^^^^^^^
> the system, hopefully just once.  Let's not call it streaming, let's
> call it... err... use-once? ;-)

Nope. If it hits every directory entry once, it'll hit every
page with directory or inode information multiple times, causing
that page to enter the active list and push out process pages.

This is exactly what we want to prevent.

regards,

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

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