From mboxrd@z Thu Jan 1 00:00:00 1970 From: Izik Eidus Subject: Re: [RFC] concurrent guest walker and instruction emulation Date: Tue, 18 Dec 2007 16:34:19 +0200 Message-ID: <1197988459.3129.7.camel@localhost.localdomain> References: <20071218132614.GA1504@dmt> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Cc: Chris Wright , kvm-devel , Avi Kivity To: Marcelo Tosatti Return-path: In-Reply-To: <20071218132614.GA1504@dmt> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: kvm-devel-bounces-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org Errors-To: kvm-devel-bounces-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org List-Id: kvm.vger.kernel.org On Tue, 2007-12-18 at 08:26 -0500, Marcelo Tosatti wrote: > Hi, > > The following is an improvement on top of an earlier patch by Izik. It i would say MUCH improvement :) > increases pagefault scalability for SMP guests by allowing concurrent > guest walking, allocation and instruction emulation on the fault path. > > The test being used is pft, which starts a number of threads > allocating and writing malloc()'ed memory. pft.c can be found at > http://lkml.org/lkml/2004/8/15/58 the code seems good and very complete, it so complete that i would had split it into two patchs (the one for the spin_lock and the one for the down_read) but more important :) is that you actually helped another patch that i wrote that register kvm shadow pages with the kernel shrinker that i couldnt merge beacuse the shadow pages didnt had their private spin lock (this i hope will make all of kvm memory swappable) i will test it today with it when i get home... anyway good work ------------------------------------------------------------------------- SF.Net email is sponsored by: Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace