From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: [RFC] vhost-blk implementation Date: Thu, 25 Mar 2010 09:57:25 +0200 Message-ID: <4BAB1765.2070707@redhat.com> References: <1269306023.7931.72.camel@badari-desktop> <20100324200402.GA22272@infradead.org> <4BAA748D.40509@us.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Cc: Christoph Hellwig , kvm@vger.kernel.org To: Badari Pulavarty Return-path: Received: from mx1.redhat.com ([209.132.183.28]:40950 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751203Ab0CYH5g (ORCPT ); Thu, 25 Mar 2010 03:57:36 -0400 In-Reply-To: <4BAA748D.40509@us.ibm.com> Sender: kvm-owner@vger.kernel.org List-ID: On 03/24/2010 10:22 PM, Badari Pulavarty wrote: >> Which caching mode is this? I assume data=writeback, because otherwise >> you'd be doing synchronous I/O directly from the handler. > > > Yes. This is with default (writeback) cache model. As mentioned > earlier, readhead is helping here > and most cases, data would be ready in the pagecache. btw, relying on readahead is problematic. The guest already issues readahead requests, and the host will issue readahead based on that readahead, reading far into the future. -- Do not meddle in the internals of kernels, for they are subtle and quick to panic.