From mboxrd@z Thu Jan 1 00:00:00 1970 From: Izik Eidus Subject: Re: [PATCH 3/4] add ksm kernel shared memory driver Date: Wed, 12 Nov 2008 01:02:47 +0200 Message-ID: <491A0F17.2070706@redhat.com> References: <1226409701-14831-1-git-send-email-ieidus@redhat.com> <1226409701-14831-2-git-send-email-ieidus@redhat.com> <1226409701-14831-3-git-send-email-ieidus@redhat.com> <1226409701-14831-4-git-send-email-ieidus@redhat.com> <20081111150345.7fff8ff2@bike.lwn.net> <491A0483.3010504@redhat.com> <20081111153028.422b301a@bike.lwn.net> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org, kvm@vger.kernel.org, aarcange@redhat.com, chrisw@redhat.com, avi@redhat.com To: Jonathan Corbet Return-path: Received: from mx2.redhat.com ([66.187.237.31]:39570 "EHLO mx2.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752146AbYKKXDT (ORCPT ); Tue, 11 Nov 2008 18:03:19 -0500 In-Reply-To: <20081111153028.422b301a@bike.lwn.net> Sender: kvm-owner@vger.kernel.org List-ID: Jonathan Corbet wrote: > > What about things like cache effects from scanning all those pages? My > guess is that, if you're trying to run dozens of Windows guests, cache > usage is not at the top of your list of concerns, but I could be > wrong. Usually am... > Ok, ksm does make the cache of the cpu dirty when scanning the pages (but the scanning happen slowly slowly and cache usually get dirty much faster) But infact it improve the cache by the fact that it make many ptes point to the same page so if before we had 12 process touching 12 diffrent physical page they would dirty the page but now they will touch just one... so i guess it depend on how you see it... > Thanks, > > jon >