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=-8.8 required=3.0 tests=DKIM_INVALID,DKIM_SIGNED, HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI,MENTIONS_GIT_HOSTING, SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED,USER_AGENT_GIT autolearn=unavailable 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 68AB0C433E0 for ; Thu, 18 Jun 2020 14:44:04 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 4015220888 for ; Thu, 18 Jun 2020 14:44:04 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=fail reason="signature verification failed" (2048-bit key) header.d=kernel-dk.20150623.gappssmtp.com header.i=@kernel-dk.20150623.gappssmtp.com header.b="LaeOybvg" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1730933AbgFROoD (ORCPT ); Thu, 18 Jun 2020 10:44:03 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:57332 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1730920AbgFROoB (ORCPT ); Thu, 18 Jun 2020 10:44:01 -0400 Received: from mail-pg1-x542.google.com (mail-pg1-x542.google.com [IPv6:2607:f8b0:4864:20::542]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id BC5D4C0613EE for ; Thu, 18 Jun 2020 07:44:00 -0700 (PDT) Received: by mail-pg1-x542.google.com with SMTP id b5so2997156pgm.8 for ; Thu, 18 Jun 2020 07:44:00 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel-dk.20150623.gappssmtp.com; s=20150623; h=from:to:cc:subject:date:message-id:reply-to:mime-version :content-transfer-encoding; bh=QtQZj6qHx4NyNp4faebxEBGjQhFGouVk8vIR/YM0OZ0=; b=LaeOybvgVFUedJ618PM1GTEUXkMXPY3qF3zemK0oPI0t/5c3IyDHmKKjFSXMlllNpg u4+VdHvZcEwrYY6QLxMvN9QNbcQkL40jfHhhwKnfjkRjaCBH/fdF2CXYNuGlJe/1qagv GOuftdx/mhTGZ0CDkYYM/BzpHCtO3WgFPHV5MzKmJusZp3RHe6UsIOj2jqBaJqopy9w1 8dyb1/ZNKWuCGje3eaRAvZHyHC8hhvG7uaNyzR+WZLwsMfnWHwLZir4H4MXYeJ5kkTgv S21Bzwg3kx7/cgMp23f6Z1a2ySJ2w9JTPWvpySOoRC3V8WNU7ZClsM3eXev1LkALavYt hZcA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:reply-to :mime-version:content-transfer-encoding; bh=QtQZj6qHx4NyNp4faebxEBGjQhFGouVk8vIR/YM0OZ0=; b=W8l8nrYXKa35MIYbRs08GmtH20cTR5Y5dd6JU/oOvi51kC6IrSLswXEEt+eiochJrQ 6pugBpY68pzE9IUiutamf2rV1FzCeVO73NzynlDMI9XPnj+XrpICBGO0jMp39dK4z9KY ClflD319POOctytNjFk8h9HZy8gbcT2Xxfq2q8iVyQ5PwBDnn5ZLa/AcrXXlnIRuquIC wRS/T/njge0W+3UxqihHK164ZYWpPATBNKGLsaof+rRXlgIEyW0FhwCuWsjZwQEa1Zy2 a3XyyL/7S05AnvG6n1aOb3FCH9FQuQaedfpM3AbeiHg77j8v/JsGZSqznL9vz9hfl73t VqKw== X-Gm-Message-State: AOAM530FOar24+YHvgYhZme6Jqby5c82xZCdmVWRIghBUAZrIxLiufuS CpeovybWHzv35+WpGmCIrvFms2NPGov7Dw== X-Google-Smtp-Source: ABdhPJwVlCUbORgsthBtt/igrlr5i7l6tNK0VhZZRDtHj5c1nyTVCxDwaDg3pXHR9uxp5F9dJc2YrQ== X-Received: by 2002:a62:191:: with SMTP id 139mr3946645pfb.94.1592491439786; Thu, 18 Jun 2020 07:43:59 -0700 (PDT) Received: from x1.localdomain ([65.144.74.34]) by smtp.gmail.com with ESMTPSA id g9sm3127197pfm.151.2020.06.18.07.43.58 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Thu, 18 Jun 2020 07:43:59 -0700 (PDT) From: Jens Axboe To: io-uring@vger.kernel.org Cc: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, akpm@linux-foundation.org Subject: Date: Thu, 18 Jun 2020 08:43:40 -0600 Message-Id: <20200618144355.17324-1-axboe@kernel.dk> X-Mailer: git-send-email 2.27.0 Reply-To: "[PATCHSET v7 0/15]"@vger.kernel.org, Add@vger.kernel.org, support@vger.kernel.org, for@vger.kernel.org, async@vger.kernel.org, buffered@vger.kernel.org, reads@vger.kernel.org MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Sender: io-uring-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: io-uring@vger.kernel.org We technically support this already through io_uring, but it's implemented with a thread backend to support cases where we would block. This isn't ideal. After a few prep patches, the core of this patchset is adding support for async callbacks on page unlock. With this primitive, we can simply retry the IO operation. With io_uring, this works a lot like poll based retry for files that support it. If a page is currently locked and needed, -EIOCBQUEUED is returned with a callback armed. The callers callback is responsible for restarting the operation. With this callback primitive, we can add support for generic_file_buffered_read(), which is what most file systems end up using for buffered reads. XFS/ext4/btrfs/bdev is wired up, but probably trivial to add more. The file flags support for this by setting FMODE_BUF_RASYNC, similar to what we do for FMODE_NOWAIT. Open to suggestions here if this is the preferred method or not. In terms of results, I wrote a small test app that randomly reads 4G of data in 4K chunks from a file hosted by ext4. The app uses a queue depth of 32. If you want to test yourself, you can just use buffered=1 with ioengine=io_uring with fio. No application changes are needed to use the more optimized buffered async read. preadv for comparison: real 1m13.821s user 0m0.558s sys 0m11.125s CPU ~13% Mainline: real 0m12.054s user 0m0.111s sys 0m5.659s CPU ~32% + ~50% == ~82% This patchset: real 0m9.283s user 0m0.147s sys 0m4.619s CPU ~52% The CPU numbers are just a rough estimate. For the mainline io_uring run, this includes the app itself and all the threads doing IO on its behalf (32% for the app, ~1.6% per worker and 32 of them). Context switch rate is much smaller with the patchset, since we only have the one task performing IO. Also ran a simple fio based test case, varying the queue depth from 1 to 16, doubling every time: [buf-test] filename=/data/file direct=0 ioengine=io_uring norandommap rw=randread bs=4k iodepth=${QD} randseed=89 runtime=10s QD/Test Patchset IOPS Mainline IOPS 1 9046 8294 2 19.8k 18.9k 4 39.2k 28.5k 8 64.4k 31.4k 16 65.7k 37.8k Outside of my usual environment, so this is just running on a virtualized NVMe device in qemu, using ext4 as the file system. NVMe isn't very efficient virtualized, so we run out of steam at ~65K which is why we flatline on the patched side (nvme_submit_cmd() eats ~75% of the test app CPU). Before that happens, it's a linear increase. Not shown is context switch rate, which is massively lower with the new code. The old thread offload adds a blocking thread per pending IO, so context rate quickly goes through the roof. The goal here is efficiency. Async thread offload adds latency, and it also adds noticable overhead on items such as adding pages to the page cache. By allowing proper async buffered read support, we don't have X threads hammering on the same inode page cache, we have just the single app actually doing IO. Been beating on this and it's solid for me, and I'm now pretty happy with how it all turned out. Not aware of any missing bits/pieces or code cleanups that need doing. Series can also be found here: https://git.kernel.dk/cgit/linux-block/log/?h=async-buffered.7 or pull from: git://git.kernel.dk/linux-block async-buffered.7 block/blk-core.c | 6 + fs/block_dev.c | 2 +- fs/btrfs/file.c | 2 +- fs/ext4/file.c | 2 +- fs/io_uring.c | 336 ++++++++++++++++++++++++++++++++++------ fs/xfs/xfs_file.c | 2 +- include/linux/blkdev.h | 1 + include/linux/fs.h | 10 +- include/linux/pagemap.h | 75 +++++++++ mm/filemap.c | 110 ++++++++----- 10 files changed, 454 insertions(+), 92 deletions(-) Changes since v6: - Properly catch and resubmit requests that would have blocked in submission. - Fix sqthread mm usage for async buffered retry - Fix a retry condition iter setup - Rebase on master + for-5.9/io_uring Changes since v5: - Correct commit message, iocb->private -> iocb->ki_waitq - Get rid of io_uring goto, use an iter read helper Changes since v3: - io_uring: don't retry if REQ_F_NOWAIT is set - io_uring: alloc req->io if the request type didn't already - Add iocb->ki_waitq instead of (ab)using iocb->private Changes since v2: - Get rid of unnecessary wait_page_async struct, just use wait_page_async - Add another prep handler, adding wake_page_match() - Use wake_page_match() in both callers Changes since v1: - Fix an issue with inline page locking - Fix a potential race with __wait_on_page_locked_async() - Fix a hang related to not setting page_match, thus missing a wakeup -- Jens Axboe