From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752955Ab1LIKcg (ORCPT ); Fri, 9 Dec 2011 05:32:36 -0500 Received: from ozlabs.org ([203.10.76.45]:53385 "EHLO ozlabs.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752545Ab1LIKcd (ORCPT ); Fri, 9 Dec 2011 05:32:33 -0500 From: Rusty Russell To: Sasha Levin Cc: "Michael S. Tsirkin" , Avi Kivity , linux-kernel@vger.kernel.org, virtualization@lists.linux-foundation.org, kvm@vger.kernel.org, markmc@redhat.com Subject: Re: [PATCH] virtio-ring: Use threshold for switching to indirect descriptors In-Reply-To: <1323340668.3904.21.camel@lappy> References: <87zkfbre9x.fsf@rustcorp.com.au> <1322913028.3782.4.camel@lappy> <4EDB5EF0.2010909@redhat.com> <1323000831.4205.4.camel@lappy> <20111204162221.GB22501@redhat.com> <1323020088.3256.3.camel@lappy> <4EDBAFC5.2010405@redhat.com> <1323020374.3256.5.camel@lappy> <1323023039.3256.7.camel@lappy> <1323266565.4009.10.camel@lappy> <20111207154816.GA23845@redhat.com> <87r50fgzyj.fsf@rustcorp.com.au> <1323340668.3904.21.camel@lappy> User-Agent: Notmuch/0.6.1-1 (http://notmuchmail.org) Emacs/23.3.1 (i686-pc-linux-gnu) Date: Fri, 09 Dec 2011 16:03:10 +1030 Message-ID: <87obvifgx5.fsf@rustcorp.com.au> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, 08 Dec 2011 12:37:48 +0200, Sasha Levin wrote: > On Thu, 2011-12-08 at 20:14 +1030, Rusty Russell wrote: > > On Wed, 7 Dec 2011 17:48:17 +0200, "Michael S. Tsirkin" wrote: > > > On Wed, Dec 07, 2011 at 04:02:45PM +0200, Sasha Levin wrote: > > > > On Sun, 2011-12-04 at 20:23 +0200, Sasha Levin wrote: > > > > > > > > [snip] > > > > > > > > Rusty, Michael, does the below looks a reasonable optimization for you? > > > > > > OK overall but a bit hard to say for sure as it looks pretty incomplete ... > > > > A static threshold is very hackish; we need to either initialize it to > > a proven-good value (since noone will ever change it) or be cleverer. > > I'll better wait to see how the threshold issue is resolved, and > possibly do it as a dynamic value which depends on the threshold. > > I doubt theres one magic value which would work for all. Sure, but if it's generally better than the current value, I'll apply it. Cheers, Rusty.