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=-0.7 required=3.0 tests=FREEMAIL_FORGED_FROMDOMAIN, FREEMAIL_FROM,HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI,SPF_HELO_NONE, SPF_PASS,URIBL_BLOCKED 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 C974BC4360C for ; Sat, 12 Oct 2019 15:10:43 +0000 (UTC) Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 99775206CD for ; Sat, 12 Oct 2019 15:10:43 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 99775206CD Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=hotmail.it Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Received: from localhost ([::1]:34228 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1iJJ2Y-0003xT-Ff for qemu-devel@archiver.kernel.org; Sat, 12 Oct 2019 11:10:42 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:45520) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1iJIZh-0004CX-5N for qemu-devel@nongnu.org; Sat, 12 Oct 2019 10:40:55 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1iJIZe-000712-3Z for qemu-devel@nongnu.org; Sat, 12 Oct 2019 10:40:53 -0400 Received: from indium.canonical.com ([91.189.90.7]:54434) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1iJIZd-0006zP-TJ for qemu-devel@nongnu.org; Sat, 12 Oct 2019 10:40:50 -0400 Received: from loganberry.canonical.com ([91.189.90.37]) by indium.canonical.com with esmtp (Exim 4.86_2 #2 (Debian)) id 1iJIZc-0003wd-Cn for ; Sat, 12 Oct 2019 14:40:48 +0000 Received: from loganberry.canonical.com (localhost [127.0.0.1]) by loganberry.canonical.com (Postfix) with ESMTP id 5CB4D2E80CD for ; Sat, 12 Oct 2019 14:40:48 +0000 (UTC) MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Date: Sat, 12 Oct 2019 14:25:57 -0000 From: Emanuele Faranda To: qemu-devel@nongnu.org X-Launchpad-Notification-Type: bug X-Launchpad-Bug: product=qemu; status=New; importance=Undecided; assignee=None; X-Launchpad-Bug-Information-Type: Public X-Launchpad-Bug-Private: no X-Launchpad-Bug-Security-Vulnerability: no X-Launchpad-Bug-Commenters: black-silver X-Launchpad-Bug-Reporter: Emanuele Faranda (black-silver) X-Launchpad-Bug-Modifier: Emanuele Faranda (black-silver) Message-Id: <157089035721.2099.14904442368526379360.malonedeb@wampee.canonical.com> Subject: [Bug 1847861] [NEW] Guest stuttering under high disk IO (virtio) X-Launchpad-Message-Rationale: Subscriber (QEMU) @qemu-devel-ml X-Launchpad-Message-For: qemu-devel-ml Precedence: bulk X-Generated-By: Launchpad (canonical.com); Revision="af2eefe214bd95389a09b7c956720881bab16807"; Instance="production-secrets-lazr.conf" X-Launchpad-Hash: 0e6db7b60ff04eee03f91f0c5f62edac0e59c3a1 X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic] [fuzzy] X-Received-From: 91.189.90.7 X-Mailman-Approved-At: Sat, 12 Oct 2019 11:06:39 -0400 X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Reply-To: Bug 1847861 <1847861@bugs.launchpad.net> Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" Public bug reported: Performing io intensive tasks on virtualized Windows causes the system to visually stutter. I can often reproduce the problem by running fio on windows: fio --randrepeat=3D1 --ioengine=3Dwindowsaio --direct=3D1 --gtod_reduce=3D1 --name=3Dtest --filename=3D\\.\PhysicalDrive0 --bs=3D4k --iodepth=3D128 --size=3D4G --readwrite=3Drandread While the fio command is running, moving the mouse pointer will be be laggy. The stuttering does not appear with iodepth <=3D 32 . The stuttering also manifests while playing games, the music and video pauses for a fraction of second in a playable but disturbing way. Here are my system specs: Host OS: archlinux Guest OS: Windows 10 Enterprise qemu version: qemu-git 8:v4.1.0.r1378.g98b2e3c9ab-1 (from AUR, compiled wit= h -march=3Dnative) CPU: AMD Ryzen Threadripper 1900X 8-Core Processor Huge Pages: vm.nr_hugepages=3D4128 Disk: nvme type=3Draw, io=3Dthreads bus=3Dvirtio GPU (passthrough): Radeon RX 570 Here are some fio test results on my windows guest: [size=3D512M,iodepth=3D1 -> min=3D30k,avg=3D31k,stddev=3D508] [size=3D2G,iodepth=3D8 -> min=3D203k,avg=3D207k,stddev=3D2.3k] [size=3D2G,iodepth=3D16 -> min=3D320k,avg=3D330k,stddev=3D4.3k] [size=3D4G,iodepth=3D32 -> min=3D300k,avg=3D310k,stddev=3D4.8k] [size=3D4G,iodepth=3D64 -> min=3D278k,avg=3D366k,stddev=3D68.6k] -> STUTTER [size=3D4G,iodepth=3D64 -> min=3D358k,avg=3D428k,stddev=3D52.6k] -> STUTTER [size=3D4G,iodepth=3D128 -> min=3D92k,avg=3D217k,stddev=3D185k] -> STUTTER [size=3D4G,iodepth=3D128 -> min=3D241k,avg=3D257k,stddev=3D14k] -> same con= fig as above, but no stuttering The min and avg values are the bandwidth values reported in KB/s by fio. You can see that, when the stuttering occurs, the stardard deviation is high and the minimum bandwidth is way below the average. Additional note: the bandwidth reported with `fio` on my linux host is about 2x the one reported in the guest: sudo fio --randrepeat=3D1 --ioengine=3Dlibaio --direct=3D1 --gtod_reduce=3D1 --name=3Dtest --filename=3D/dev/nvme0n1 --bs=3D4k --iodepth=3D64 --size=3D5= 12M --readwrite=3Drandread read: IOPS=3D279k, BW=3D1092MiB/s (1145MB/s)(512MiB/469msec) ** Affects: qemu Importance: Undecided Status: New ** Description changed: Performing io intensive tasks on virtualized Windows causes the system to visually stutter. I can often reproduce the problem by running fio on windows: = fio --randrepeat=3D1 --ioengine=3Dwindowsaio --direct=3D1 --gtod_reduce= =3D1 --name=3Dtest --filename=3D\\.\PhysicalDrive0 --bs=3D4k --iodepth=3D128 --size=3D4G --readwrite=3Drandread = While the fio command is running, moving the mouse pointer will be be laggy. The stuttering does not appear with iodepth <=3D 32 . The stuttering also manifests while playing games, the music and video pauses for a fraction of second in a playable but disturbing way. = Here are my system specs: = Host OS: archlinux Guest OS: Windows 10 Enterprise qemu version: qemu-git 8:v4.1.0.r1378.g98b2e3c9ab-1 (from AUR, compiled w= ith -march=3Dnative) CPU: AMD Ryzen Threadripper 1900X 8-Core Processor Huge Pages: vm.nr_hugepages=3D4128 Disk: nvme type=3Draw, io=3Dthreads bus=3Dvirtio GPU (passthrough): Radeon RX 570 = Here are some fio test results on my windows guest: = [size=3D512M,iodepth=3D1 -> min=3D30k,avg=3D31k,stddev=3D508] [size=3D2G,iodepth=3D8 -> min=3D203k,avg=3D207k,stddev=3D2.3k] [size=3D2G,iodepth=3D16 -> min=3D320k,avg=3D330k,stddev=3D4.3k] [size=3D4G,iodepth=3D32 -> min=3D300k,avg=3D310k,stddev=3D4.8k] [size=3D4G,iodepth=3D64 -> min=3D278k,avg=3D366k,stddev=3D68.6k] -> STUTT= ER [size=3D4G,iodepth=3D64 -> min=3D358k,avg=3D428k,stddev=3D52.6k] -> STUTT= ER [size=3D4G,iodepth=3D128 -> min=3D92k,avg=3D217k,stddev=3D185k] -> STUTTER [size=3D4G,iodepth=3D128 -> min=3D241k,avg=3D257k,stddev=3D14k] -> same c= onfig as above, but no stuttering = The min and avg values are the bandwidth values reported in KB/s by fio. You can see that, when the stuttering occurs, the stardard deviation is high and the minimum bandwidth is way below the average. + = + Additional note: the bandwidth reported with `fio` on my linux host is + about 2x the one reported in the guest: + = + sudo fio --randrepeat=3D1 --ioengine=3Dlibaio --direct=3D1 --gtod_reduce= =3D1 + --name=3Dtest --filename=3D/dev/nvme0n1 --bs=3D4k --iodepth=3D64 --size= =3D512M + --readwrite=3Drandread + = + read: IOPS=3D279k, BW=3D1092MiB/s (1145MB/s)(512MiB/469msec) -- = You received this bug notification because you are a member of qemu- devel-ml, which is subscribed to QEMU. https://bugs.launchpad.net/bugs/1847861 Title: Guest stuttering under high disk IO (virtio) Status in QEMU: New Bug description: Performing io intensive tasks on virtualized Windows causes the system to visually stutter. I can often reproduce the problem by running fio on windows: fio --randrepeat=3D1 --ioengine=3Dwindowsaio --direct=3D1 --gtod_reduce= =3D1 --name=3Dtest --filename=3D\\.\PhysicalDrive0 --bs=3D4k --iodepth=3D128 --size=3D4G --readwrite=3Drandread While the fio command is running, moving the mouse pointer will be be laggy. The stuttering does not appear with iodepth <=3D 32 . The stuttering also manifests while playing games, the music and video pauses for a fraction of second in a playable but disturbing way. Here are my system specs: Host OS: archlinux Guest OS: Windows 10 Enterprise qemu version: qemu-git 8:v4.1.0.r1378.g98b2e3c9ab-1 (from AUR, compiled w= ith -march=3Dnative) CPU: AMD Ryzen Threadripper 1900X 8-Core Processor Huge Pages: vm.nr_hugepages=3D4128 Disk: nvme type=3Draw, io=3Dthreads bus=3Dvirtio GPU (passthrough): Radeon RX 570 Here are some fio test results on my windows guest: [size=3D512M,iodepth=3D1 -> min=3D30k,avg=3D31k,stddev=3D508] [size=3D2G,iodepth=3D8 -> min=3D203k,avg=3D207k,stddev=3D2.3k] [size=3D2G,iodepth=3D16 -> min=3D320k,avg=3D330k,stddev=3D4.3k] [size=3D4G,iodepth=3D32 -> min=3D300k,avg=3D310k,stddev=3D4.8k] [size=3D4G,iodepth=3D64 -> min=3D278k,avg=3D366k,stddev=3D68.6k] -> STUTT= ER [size=3D4G,iodepth=3D64 -> min=3D358k,avg=3D428k,stddev=3D52.6k] -> STUTT= ER [size=3D4G,iodepth=3D128 -> min=3D92k,avg=3D217k,stddev=3D185k] -> STUTTER [size=3D4G,iodepth=3D128 -> min=3D241k,avg=3D257k,stddev=3D14k] -> same c= onfig as above, but no stuttering The min and avg values are the bandwidth values reported in KB/s by fio. You can see that, when the stuttering occurs, the stardard deviation is high and the minimum bandwidth is way below the average. Additional note: the bandwidth reported with `fio` on my linux host is about 2x the one reported in the guest: sudo fio --randrepeat=3D1 --ioengine=3Dlibaio --direct=3D1 --gtod_reduce= =3D1 --name=3Dtest --filename=3D/dev/nvme0n1 --bs=3D4k --iodepth=3D64 --size= =3D512M --readwrite=3Drandread read: IOPS=3D279k, BW=3D1092MiB/s (1145MB/s)(512MiB/469msec) To manage notifications about this bug go to: https://bugs.launchpad.net/qemu/+bug/1847861/+subscriptions