From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Michael S. Tsirkin" Subject: Re: [PATCH] virtio_blk: merge S/G list entries by default Date: Mon, 8 Sep 2014 11:18:30 +0300 Message-ID: <20140908081830.GA32268@redhat.com> References: <1410044994-14943-1-git-send-email-hch@lst.de> <20140907114153.GB26569@redhat.com> <20140907184745.GA23026@lst.de> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Content-Disposition: inline In-Reply-To: <20140907184745.GA23026@lst.de> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: virtualization-bounces@lists.linux-foundation.org Errors-To: virtualization-bounces@lists.linux-foundation.org To: Christoph Hellwig Cc: Jens Axboe , Ming Lei , linux-kernel@vger.kernel.org, virtualization@lists.linux-foundation.org List-Id: virtualization@lists.linuxfoundation.org On Sun, Sep 07, 2014 at 08:47:45PM +0200, Christoph Hellwig wrote: > On Sun, Sep 07, 2014 at 02:41:53PM +0300, Michael S. Tsirkin wrote: > > > Signed-off-by: Christoph Hellwig > > > > OK so this is an optimization patch right? > > What kind of performance gain is observed with it? > > None. I actually wrote it when the block layer had a bug when dm was > used on top of the !merge case, and I decided to send it out as there had been > no discussion about disabling this by default on the existing blk-mq > drivers. > > At least for my qemu/kvm setup it doesn't make a difference either way, > although not doing the cheap merge doesn't like the right kind of optimization > to me. Could you respond to Ming Lei's mail, who benchmarked the patch, please? -- MST