From mboxrd@z Thu Jan 1 00:00:00 1970 From: Paolo Bonzini Subject: Re: [PATCH 2/2] virtio_scsi: support multi hw queue of blk-mq Date: Tue, 11 Nov 2014 18:09:17 +0100 Message-ID: <546242BD.5010902@redhat.com> References: <1415552224-31145-1-git-send-email-ming.lei@canonical.com> <1415552224-31145-3-git-send-email-ming.lei@canonical.com> <54609667.8030702@redhat.com> <20141110133930.GA3115@lst.de> <20141110160506.GB15315@lst.de> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Return-path: Received: from mx1.redhat.com ([209.132.183.28]:39918 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751228AbaKKRJa (ORCPT ); Tue, 11 Nov 2014 12:09:30 -0500 In-Reply-To: <20141110160506.GB15315@lst.de> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: Christoph Hellwig , Ming Lei Cc: James Bottomley , Jens Axboe , Linux SCSI List On 10/11/2014 17:05, Christoph Hellwig wrote: > > a) there is no multiath support for it, and we simply can't break existing > setups that use. > b) there is no support for I/O schedulers at all. This might be okay > for virtio-scsi, where in general you have a host scheduler, but for > real hardware that still has to deal with more spinning rust than > solid state it's a blocker. Fortuntely Jens is working on that one. It is sort-of okay for virtio-scsi (and virtio-blk). Deadline provides a good speedup over noop for virtual disks, even in the presence of a host scheduler---though nobody ever benchmarked why. Paolo