Date: Fri, 15 Oct 2004 03:18:38 +0200
From: Andi Kleen <ak@suse.de>
To: Adam Heath <doogie@debian.org>
Cc: Andi Kleen <ak@suse.de>, "Martin K. Petersen" <mkp@wildopensource.com>,
       linux-kernel@vger.kernel.org, linux-ia64@vger.kernel.org, akpm@osdl.org,
       tony.luck@intel.com
Subject: Re: [PATCH] General purpose zeroed page slab
Message-ID: <20041015011838.GC9753@wotan.suse.de>
References: <yq1oej5s0po.fsf@wilson.mkp.net> <20041014180427.GA7973@wotan.suse.de> <Pine.LNX.4.58.0410141836270.1221@gradall.private.brainfood.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <Pine.LNX.4.58.0410141836270.1221@gradall.private.brainfood.com>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 968
Lines: 25

On Thu, Oct 14, 2004 at 06:36:47PM -0500, Adam Heath wrote:
> On Thu, 14 Oct 2004, Andi Kleen wrote:
> 
> > Also that's pretty dumb. How about keeping track how much of the
> > page got non zeroed (e.g. by using a few free words in struct page
> > for a coarse grained dirty bitmap)
> >
> > Then you could memset on free only the parts that got actually
> > changed, and never waste cache lines for anything else.
> 
> That will fail when a struct is placed in the page, and only the beginning and
> end of the struct was changed.

Hmm? I think you're misunderstanding me. The dirty bitmap would cover 
all areas that are potentially not zero. If someone changes a byte
without setting the bitmap they're buggy. 


-Andi
> 
-
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/