From mboxrd@z Thu Jan 1 00:00:00 1970 From: Konrad Rzeszutek Wilk Subject: Re: Xen block backend/frontend multiqueue support Date: Wed, 19 Mar 2014 11:09:28 -0400 Message-ID: <20140319150928.GA15729@phenom.dumpdata.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Content-Disposition: inline In-Reply-To: List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: haofeng hua Cc: xen-devel@lists.xen.org List-Id: xen-devel@lists.xenproject.org On Wed, Mar 19, 2014 at 08:00:00PM +0800, haofeng hua wrote: > Hi all, there are some questions about Xen block backend/frontend > multiqueue support. > In linux, block multiqueue API is used to allocate per CPU a block thread > to handle the block request. > In a system on xen hypervisor, the request from PV guest is put into the > shared ring and passed to the driver domain. > So how to use multiqueue API in frontend driver and backend driver > respectively? You would have to have more rings. > What I think is using multiqueue API in backend driver to dispatch requests > from shared ring, does that work? What about the frontend driver? Same thing I believe - the frontend would allocate X amount of rings - where the X is based on the amount of CPUs. The backend would then have to support X amount of rings with the frontend. > > Best regards.