Date: Thu, 05 Jan 2006 18:47:22 +0900
From: IWAMOTO Toshihiro <iwamoto@valinux.co.jp>
To: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Marcelo Tosatti <marcelo.tosatti@cyclades.com>, linux-mm@kvack.org,
       linux-kernel@vger.kernel.org, Andrew Morton <akpm@osdl.org>,
       Christoph Lameter <christoph@lameter.com>,
       Wu Fengguang <wfg@mail.ustc.edu.cn>, Nick Piggin <npiggin@suse.de>,
       Marijn Meijles <marijn@bitpit.net>, Rik van Riel <riel@redhat.com>
Subject: Re: [PATCH 6/9] clockpro-clockpro.patch
In-Reply-To: <1136028546.17853.69.camel@twins>
References: <20051230223952.765.21096.sendpatchset@twins.localnet>
	<20051230224312.765.58575.sendpatchset@twins.localnet>
	<20051231002417.GA4913@dmt.cnet>
	<1136028546.17853.69.camel@twins>
User-Agent: Wanderlust/2.15.1 (Almost Unreal) SEMI/1.14.6 (Maruoka)
 FLIM/1.14.7 (=?ISO-8859-4?Q?Sanj=F2?=) APEL/10.6 Emacs/21.4
 (i386-pc-linux-gnu) MULE/5.0 (SAKAKI)
MIME-Version: 1.0 (generated by SEMI 1.14.6 - "Maruoka")
Content-Type: text/plain; charset=US-ASCII
Message-Id: <20060105094722.897C574030@sv1.valinux.co.jp>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1503
Lines: 56

At Sat, 31 Dec 2005 12:29:06 +0100,
Peter Zijlstra wrote:
> 
> Forgot one in the previous mail.
> 
> On Fri, 2005-12-30 at 22:24 -0200, Marcelo Tosatti wrote:
> > Please make it easier for others to understand why the hands 
> > swap, and when, and why.
> 
> Its not the hands that swap, its the lists. The hands will lap each
> other, like the minute hand will lap the hour hand every ~65 minutes.
> 
> Let me try some ascii art.
> 
>    ====
>   ^---<>---v
>        ====
> 
> '='	a page
> '^---<' hand cold
> '>---v' hand hot
> 
> now let hand cold move 4 pages:
> 
>    
>   ^---<>---v
>    ========
> 
> ie. hand hot and hand cold have the same position.
> now if we want to move hand cold one more position this happens:
> 
>    =======
>   ^---<>---v
>           =
> 
> see the swap?

Is it okay to allow Hcold to lap Hhot?
In my understanding of CLOCK-Pro, such lapping causes sudden increase
in the distance between Hhot and Hcold.  As that distance is an
important parameter of page aging/replacement decisions, I'm afraid
that such lapping would result in incorrect page aging and bad
performance.

I guess the alternative is to advance Hhot together with Hcold, but
I'm not sure this is correct, either.

Please enlighten me.

--
IWAMOTO Toshihiro
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/