From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:39514) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1YITHG-0001WJ-Mp for qemu-devel@nongnu.org; Mon, 02 Feb 2015 21:31:47 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1YITHD-0000ZG-1O for qemu-devel@nongnu.org; Mon, 02 Feb 2015 21:31:46 -0500 Received: from szxga03-in.huawei.com ([119.145.14.66]:30683) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1YITHC-0000XW-Ft for qemu-devel@nongnu.org; Mon, 02 Feb 2015 21:31:42 -0500 Message-ID: <54D032FB.404@huawei.com> Date: Tue, 3 Feb 2015 10:31:23 +0800 From: "Wangting (Kathy)" MIME-Version: 1.0 Content-Type: text/plain; charset="ISO-8859-1" Content-Transfer-Encoding: 7bit Subject: [Qemu-devel] [question] the patch which affect performance of virtio-scsi List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: pbonzini@redhat.com Cc: qemu-devel Hi Paolo, Recently I test IO performance with virtio-scsi, and find out that the patch of "virtio-scsi: add support for the any_layout feature" affects IO performance of model with 4KB 32iodepth sequence read. Why cdb and sense is removed from the struct of VirtIOSCSICmdReq and VirtIOSCSICmdResp? How do you consider the impact of the changes to the performance? Although the latest version of qemu can optimize the performance by the way of reading merger, I think the affect of this patch cannot be ignored. Best wishes, Ting