From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-7.5 required=3.0 tests=BAYES_00, HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI,NICE_REPLY_A,SPF_HELO_NONE, SPF_PASS,UNPARSEABLE_RELAY,USER_AGENT_SANE_1 autolearn=no autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id D1702C4338F for ; Wed, 25 Aug 2021 16:39:56 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id B322F60F5C for ; Wed, 25 Aug 2021 16:39:56 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S240886AbhHYQkm (ORCPT ); Wed, 25 Aug 2021 12:40:42 -0400 Received: from out30-54.freemail.mail.aliyun.com ([115.124.30.54]:42734 "EHLO out30-54.freemail.mail.aliyun.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S237319AbhHYQkl (ORCPT ); Wed, 25 Aug 2021 12:40:41 -0400 X-Alimail-AntiSpam: AC=PASS;BC=-1|-1;BR=01201311R171e4;CH=green;DM=||false|;DS=||;FP=0|-1|-1|-1|0|-1|-1|-1;HT=alimailimapcm10staff010182156082;MF=haoxu@linux.alibaba.com;NM=1;PH=DS;RN=4;SR=0;TI=SMTPD_---0Ult-uTI_1629909593; Received: from B-25KNML85-0107.local(mailfrom:haoxu@linux.alibaba.com fp:SMTPD_---0Ult-uTI_1629909593) by smtp.aliyun-inc.com(127.0.0.1); Thu, 26 Aug 2021 00:39:54 +0800 Subject: Re: [RFC 0/2] io_task_work optimization To: Jens Axboe Cc: io-uring@vger.kernel.org, Pavel Begunkov , Joseph Qi References: <20210823183648.163361-1-haoxu@linux.alibaba.com> <503f1587-f7d9-13a9-a509-f9623d8748e9@kernel.dk> From: Hao Xu Message-ID: <19c77256-c83b-62b2-f3fb-7c85c882b5b2@linux.alibaba.com> Date: Thu, 26 Aug 2021 00:39:53 +0800 User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:78.0) Gecko/20100101 Thunderbird/78.12.0 MIME-Version: 1.0 In-Reply-To: <503f1587-f7d9-13a9-a509-f9623d8748e9@kernel.dk> Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 8bit Precedence: bulk List-ID: X-Mailing-List: io-uring@vger.kernel.org 在 2021/8/25 下午11:58, Jens Axboe 写道: > On 8/23/21 12:36 PM, Hao Xu wrote: >> running task_work may not be a big bottleneck now, but it's never worse >> to make it move forward a little bit. >> I'm trying to construct tests to prove it is better in some cases where >> it should be theoretically. >> Currently only prove it is not worse by running fio tests(sometimes a >> little bit better). So just put it here for comments and suggestion. > > I think this is interesting, particularly for areas where we have a mix > of task_work uses because obviously it won't really matter if the > task_work being run is homogeneous. > > That said, would be nice to have some numbers associated with it. We > have a few classes of types of task_work: > > 1) Work completes really fast, we want to just do those first > 2) Work is pretty fast, like async buffered read copy > 3) Work is more expensive, might require a full retry of the operation > > Might make sense to make this classification explicit. Problem is, with > any kind of scheduling like that, you risk introducing latency bubbles > because the prio1 list grows really fast, for example. Yes, this may intrpduce latency if overwhelming 1) comes in short time. I'll try more tests to see if the problem exists and if there is a better way, like put limited number of 1) to the front. Anyway, I'll update this thread when I get some data. >