From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752086AbcCKLYg (ORCPT ); Fri, 11 Mar 2016 06:24:36 -0500 Received: from mail-wm0-f49.google.com ([74.125.82.49]:37377 "EHLO mail-wm0-f49.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752001AbcCKLY3 (ORCPT ); Fri, 11 Mar 2016 06:24:29 -0500 Subject: Re: [PATCH RFC 09/22] block, cfq: replace CFQ with the BFQ-v0 I/O scheduler To: Christoph Hellwig , Linus Walleij References: <1454364778-25179-1-git-send-email-paolo.valente@linaro.org> <1454364778-25179-10-git-send-email-paolo.valente@linaro.org> <20160211222210.GC3741@mtj.duckdns.org> <8FDE2B10-9BD2-4741-917F-5A37A74E5B58@linaro.org> <20160217170206.GU3741@mtj.duckdns.org> <72E81252-203C-4EB7-8459-B9B7060029C6@linaro.org> <20160301184656.GI3965@htj.duckdns.org> <20160304173947.GA16764@infradead.org> <20160311111714.GC3450@infradead.org> Cc: Ulf Hansson , Tejun Heo , Paolo Valente , Jens Axboe , Fabio Checconi , Arianna Avanzini , linux-block@vger.kernel.org, "linux-kernel@vger.kernel.org" , Mark Brown From: Nikolay Borisov Message-ID: <56E2AAE8.8000203@kyup.com> Date: Fri, 11 Mar 2016 13:24:24 +0200 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.1.0 MIME-Version: 1.0 In-Reply-To: <20160311111714.GC3450@infradead.org> Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 03/11/2016 01:17 PM, Christoph Hellwig wrote: > On Sat, Mar 05, 2016 at 07:18:30PM +0700, Linus Walleij wrote: >> Depends on what time horizon and target I'd say. Paolo was in contact with >> the MMC/SD subsystem maintainer Ulf Hansson. (e)MMC/SD are both >> synchronous command-response-based protocols, and as of today >> single-channel. So everone's smartphone and tablet etc are today >> single-channel. I don't know if there is even a protocol change coming >> to augment this, the only duct-tapeish solution I've heard about is >> command queueing which is basically a kind of pipelining of requests. > > We use blkj-mq for single queue devices as well, in fact most consumers > are single queue. MMC is a prime candidate that hould move over soon. Out of curiosity - would it make sense to have something like BFQ/CFQ for single-queue blk-mq users?