From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: [RFC] vhost-blk implementation Date: Thu, 25 Mar 2010 17:51:31 +0200 Message-ID: <4BAB8683.6040504@redhat.com> References: <1269306023.7931.72.camel@badari-desktop> <4BA891E2.9040500@redhat.com> <20100324200502.GB22272@infradead.org> <4BAB02AF.7040308@redhat.com> <20100325154805.GA1504@infradead.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Badari Pulavarty , kvm@vger.kernel.org To: Christoph Hellwig Return-path: Received: from mx1.redhat.com ([209.132.183.28]:42005 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750968Ab0CYPvk (ORCPT ); Thu, 25 Mar 2010 11:51:40 -0400 In-Reply-To: <20100325154805.GA1504@infradead.org> Sender: kvm-owner@vger.kernel.org List-ID: On 03/25/2010 05:48 PM, Christoph Hellwig wrote: > On Thu, Mar 25, 2010 at 08:29:03AM +0200, Avi Kivity wrote: > >> We still have a virtio implementation in userspace for file-based images. >> >> In any case, the file APIs are not asynchronous so we'll need a thread >> pool. That will probably minimize the difference in performance between >> the userspace and kernel implementations. >> > The kernel has "real" aio when doing direct I/O, although currently we > can't easily use it from kernelspace. Thejre's a few people who've been > looking into making it usable for that, though. > Interesting. Are there limitations on supported filesystems, or allocating vs. non-allocating writes? -- error compiling committee.c: too many arguments to function