From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: sync guest calls made async on host - SQLite performance Date: Thu, 15 Oct 2009 01:56:40 +0900 Message-ID: <4AD602C8.4090603@redhat.com> References: <4ACCEC9F.7090309@gmail.com> <4ACE0196.9010904@gmail.com> <4ACF89CB.5020406@gmail.com> <4AD1A27A.4060307@redhat.com> <20091013223714.GB16152@lst.de> <4AD5B00D.102@redhat.com> <20091014134122.GA14235@lst.de> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Matthew Tippett , Dustin Kirkland , Anthony Liguori , RW , kvm@vger.kernel.org To: Christoph Hellwig Return-path: Received: from mx1.redhat.com ([209.132.183.28]:9600 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S934461AbZJNQ5S (ORCPT ); Wed, 14 Oct 2009 12:57:18 -0400 In-Reply-To: <20091014134122.GA14235@lst.de> Sender: kvm-owner@vger.kernel.org List-ID: On 10/14/2009 10:41 PM, Christoph Hellwig wrote: >> But can't this be also implemented using QUEUE_ORDERED_DRAIN, and on the >> host side disabling the backing device write cache? I'm talking about >> cache=none, primarily. >> > Yes, it could. But as I found out in a long discussion with Stephen > it's not actually nessecary. All filesystems do the right thing for > a device not claiming to support barriers if it doesn't include write > caches, that is implement ordering internally. So there is no urge to > set QUEUE_ORDERED_DRAIN for the case without write cache. > Does virtio say it has a write cache or not (and how does one say it?)? According to the report, a write+fdatasync completes too fast, at least on Ubuntu's qemu. So perhaps somewhere this information is lost. -- I have a truly marvellous patch that fixes the bug which this signature is too narrow to contain.