linux-mips
[Top] [All Lists]

Re: [PATCH] qemu does not have dcache aliases

To: Ralf Baechle <ralf@linux-mips.org>
Subject: Re: [PATCH] qemu does not have dcache aliases
From: Thiemo Seufer <ths@networkno.de>
Date: Mon, 21 Aug 2006 16:36:16 +0100
Cc: "Maciej W. Rozycki" <macro@linux-mips.org>, Atsushi Nemoto <anemo@mba.ocn.ne.jp>, linux-mips@linux-mips.org
In-reply-to: <20060821152657.GB19600@linux-mips.org>
Original-recipient: rfc822;linux-mips@linux-mips.org
References: <20060820.003338.25478178.anemo@mba.ocn.ne.jp> <Pine.LNX.4.64N.0608211340120.17504@blysk.ds.pg.gda.pl> <20060821.225910.108307053.anemo@mba.ocn.ne.jp> <Pine.LNX.4.64N.0608211612090.17504@blysk.ds.pg.gda.pl> <20060821152657.GB19600@linux-mips.org>
Sender: linux-mips-bounce@linux-mips.org
User-agent: Mutt/1.5.13 (2006-08-11)
Ralf Baechle wrote:
> On Mon, Aug 21, 2006 at 04:16:21PM +0100, Maciej W. Rozycki wrote:
> 
> > > Well, the QEMU cpu has 2-way 2kB dcache... does not have aliasing
> > > anyway. :-)
> > 
> >  I don't think emulating a bigger cache so that we can add aliases should 
> > be *that* difficult.  Adding aliases themselves might be a bit trickier, 
> > but the gain would certainly justify the hassle, wouldn't it?
> 
> The cache lookup for every access would have serious impact on performance.
> The question is, is it worth the price?  Would such patches be acceptable
> by the upstream qemu maintainers?

Likely not, given that performance is the prime criterion there.


Thiemo

<Prev in Thread] Current Thread [Next in Thread>