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 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id EA557C433F5 for ; Thu, 28 Oct 2021 11:46:35 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id CE4C560FF2 for ; Thu, 28 Oct 2021 11:46:35 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S229835AbhJ1Ls7 (ORCPT ); Thu, 28 Oct 2021 07:48:59 -0400 Received: from out30-130.freemail.mail.aliyun.com ([115.124.30.130]:37658 "EHLO out30-130.freemail.mail.aliyun.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S230188AbhJ1Ls6 (ORCPT ); Thu, 28 Oct 2021 07:48:58 -0400 X-Alimail-AntiSpam: AC=PASS;BC=-1|-1;BR=01201311R791e4;CH=green;DM=||false|;DS=||;FP=0|-1|-1|-1|0|-1|-1|-1;HT=e01e04407;MF=haoxu@linux.alibaba.com;NM=1;PH=DS;RN=4;SR=0;TI=SMTPD_---0Uu.Sn3p_1635421589; Received: from B-25KNML85-0107.local(mailfrom:haoxu@linux.alibaba.com fp:SMTPD_---0Uu.Sn3p_1635421589) by smtp.aliyun-inc.com(127.0.0.1); Thu, 28 Oct 2021 19:46:30 +0800 Subject: Re: [PATCH for-5.16 v3 0/8] task work optimization From: Hao Xu To: Pavel Begunkov , Jens Axboe Cc: io-uring@vger.kernel.org, Joseph Qi References: <20211027140216.20008-1-haoxu@linux.alibaba.com> <7a528ce1-a44e-3ee7-095c-1a92528ec441@linux.alibaba.com> Message-ID: Date: Thu, 28 Oct 2021 19:46:29 +0800 User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:78.0) Gecko/20100101 Thunderbird/78.13.0 MIME-Version: 1.0 In-Reply-To: <7a528ce1-a44e-3ee7-095c-1a92528ec441@linux.alibaba.com> 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 2021/10/28 下午2:07, Hao Xu 写道: > 在 2021/10/28 上午2:15, Pavel Begunkov 写道: >> On 10/27/21 15:02, Hao Xu wrote: >>> Tested this patchset by manually replace __io_queue_sqe() in >>> io_queue_sqe() by io_req_task_queue() to construct 'heavy' task works. >>> Then test with fio: >> >> If submissions and completions are done by the same task it doesn't >> really matter in which order they're executed because the task won't >> get back to userspace execution to see CQEs until tw returns. > It may matter, it depends on the time cost of submittion > and the DMA IO time. Pick up sqpoll mode as example, > we submit 10 reqs: > t1          io_submit_sqes >             -->io_req_task_queue > t2          io_task_work_run > we actually do the submittion in t2,  but if the workload > is big engough, the 'irq completion TW' will be inserted > to the TW list after t2 is fully done, then those > 'irq completion TW' will be delayed to the next round. > With this patchset, we can handle them first. >> Furthermore, it even might be worse because the earlier you submit >> the better with everything else equal. >> >> IIRC, that's how it's with fio, right? If so, you may get better >> numbers with a test that does submissions and completions in >> different threads. > Because of the completion cache, I doubt if it works. > For single ctx, it seems we always update the cqring > pointer after all the TWs in the list are done. I suddenly realized sqpoll mode does submissions and completions in different threads, and in this situation this patchset always first commit_cqring() after handling TWs in priority list. So this is the right test, do I miss something? >> >> Also interesting to find an explanation for you numbers assuming > The reason may be what I said above, but I don't have a > strict proof now. >> they're stable. 7/8 batching? How often it does it go this path? >> If only one task submits requests it should already be covered >> with existing batching. > the problem of the existing batch is(given there is only > one ctx): > 1. we flush it after all the TWs done > 2. we batch them if we have uring lock. > the new batch is: > 1. don't care about uring lock > 2. we can flush the completions in the priority list >    in advance.(which means userland can see it earlier.) >> >> >>> ioengine=io_uring >>> sqpoll=1 >>> thread=1 >>> bs=4k >>> direct=1 >>> rw=randread >>> time_based=1 >>> runtime=600 >>> randrepeat=0 >>> group_reporting=1 >>> filename=/dev/nvme0n1 >>> >>> 2/8 set unlimited priority_task_list, 8/8 set a limitation of >>> 1/3 * (len_prority_list + len_normal_list), data below: >>>     depth     no 8/8   include 8/8      before this patchset >>>      1        7.05         7.82              7.10 >>>      2        8.47         8.48              8.60 >>>      4        10.42        9.99              10.42 >>>      8        13.78        13.13             13.22 >>>      16       27.41        27.92             24.33 >>>      32       49.40        46.16             53.08 >>>      64       102.53       105.68            103.36 >>>      128      196.98       202.76            205.61 >>>      256      372.99       375.61            414.88 >>>      512      747.23       763.95            791.30 >>>      1024     1472.59      1527.46           1538.72 >>>      2048     3153.49      3129.22           3329.01 >>>      4096     6387.86      5899.74           6682.54 >>>      8192     12150.25     12433.59          12774.14 >>>      16384    23085.58     24342.84          26044.71 >>> >>> It seems 2/8 is better, haven't tried other choices other than 1/3, >>> still put 8/8 here for people's further thoughts. >>> >>> Hao Xu (8): >>>    io-wq: add helper to merge two wq_lists >>>    io_uring: add a priority tw list for irq completion work >>>    io_uring: add helper for task work execution code >>>    io_uring: split io_req_complete_post() and add a helper >>>    io_uring: move up io_put_kbuf() and io_put_rw_kbuf() >>>    io_uring: add nr_ctx to record the number of ctx in a task >>>    io_uring: batch completion in prior_task_list >>>    io_uring: add limited number of TWs to priority task list >>> >>>   fs/io-wq.h    |  21 +++++++ >>>   fs/io_uring.c | 168 +++++++++++++++++++++++++++++++++++--------------- >>>   2 files changed, 138 insertions(+), 51 deletions(-) >>> >>