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=-6.5 required=3.0 tests=BAYES_00,DKIM_INVALID, DKIM_SIGNED,HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI,SPF_HELO_NONE, SPF_PASS,URIBL_BLOCKED,USER_AGENT_GIT 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 BC061C43457 for ; Mon, 19 Oct 2020 22:59:05 +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 EF2482240C for ; Mon, 19 Oct 2020 22:59:04 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=fail reason="signature verification failed" (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="QN96Lcog" DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org EF2482240C Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=redhat.com Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Received: from localhost ([::1]:41060 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1kUe7L-00007S-WB for qemu-devel@archiver.kernel.org; Mon, 19 Oct 2020 18:59:04 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:37470) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1kUe5r-00070R-18 for qemu-devel@nongnu.org; Mon, 19 Oct 2020 18:57:31 -0400 Received: from us-smtp-delivery-124.mimecast.com ([63.128.21.124]:48557) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_CBC_SHA1:256) (Exim 4.90_1) (envelope-from ) id 1kUe5o-0005cI-O3 for qemu-devel@nongnu.org; Mon, 19 Oct 2020 18:57:30 -0400 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1603148247; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding; bh=oSx6Upto5qIIMQCj3DBLMSfuQPWH8PnPRsubRgvIkhs=; b=QN96LcogF5hdtmX6XDTV36JUQVm0OBBC/ztUSNWvxxfO9XUayjK1c62aB3gjMOqO46hn9B 1DCC2SwNw27ssDRNTX7g79fbtzVwLSvs8UEgXowY+OGKsqVKCvA2ETZ33n2lCpEtcxQxqu C7RHQRMX5BEcl5jwXWsxYSJ+ROot1hg= Received: from mail-io1-f70.google.com (mail-io1-f70.google.com [209.85.166.70]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-502-oJWI9kBPNeKblnDmKTnbTw-1; Mon, 19 Oct 2020 18:57:24 -0400 X-MC-Unique: oJWI9kBPNeKblnDmKTnbTw-1 Received: by mail-io1-f70.google.com with SMTP id w16so1356419ioa.2 for ; Mon, 19 Oct 2020 15:57:24 -0700 (PDT) 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:mime-version :content-transfer-encoding; bh=k0lKkODD5+DnqWYJtZy3D2vvB7uOBfNqhdiZo53S8As=; b=gFLG5NnleBsB8dmSJoCbebcLi7CBtCOswCs6EB3TQfP1Wb2Xm1VIXHx++gJ2s+vS7+ lEFOUSakjLWJZALaQRiRgeB/UU0rpbj2YxU4ODx3XzKmzGWUkeTrqFuUxu995MpMwGSW z7QsUEDE7Ti3XNyRU9fakkUjZRkSaLhmFcklnLbe+fNXeik5GKN1/EAqZ4qn1Wtf9qef n+xWdixVWt6KzzmWHDc0f8WiTvQ2o6/dwBGMB/eLuJRzFheKjy8Vfiw8aPUJ+t7jnFy3 9mAk1yloWDUuv9qw4/gvBHzrc5uwTffxdNePfvgDewBqy3vepW9jqQomq0pf5ly6U9St /MfA== X-Gm-Message-State: AOAM532SDU4dyvA9Gi2/oZyhccIVDkP0VtUZxRN6N2gpKsqkhNFLeP3Q RhZvqP/H3m9mnhq4iEkmfSWA0PnRb2SPTuBW2ZOnipAPN+Hf5p4mcrJfaX0AbqVIC5yVZdqD6pX au+BJzVdKUMTx1zo= X-Received: by 2002:a92:1303:: with SMTP id 3mr1915797ilt.117.1603148243212; Mon, 19 Oct 2020 15:57:23 -0700 (PDT) X-Google-Smtp-Source: ABdhPJyI8f/GElODbojrdRkxn9M6IdvtlUz1I5VpS3kMzGurSCHYbrG8v/vD16uBahK66YqMMDh+vA== X-Received: by 2002:a92:1303:: with SMTP id 3mr1915791ilt.117.1603148242959; Mon, 19 Oct 2020 15:57:22 -0700 (PDT) Received: from xz-x1.redhat.com (toroon474qw-lp140-04-174-95-215-133.dsl.bell.ca. [174.95.215.133]) by smtp.gmail.com with ESMTPSA id z89sm6017ilk.4.2020.10.19.15.57.21 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 19 Oct 2020 15:57:22 -0700 (PDT) From: Peter Xu To: qemu-devel@nongnu.org Subject: [PATCH v5 0/6] migration/postcopy: Sync faulted addresses after network recovered Date: Mon, 19 Oct 2020 18:57:14 -0400 Message-Id: <20201019225720.172743-1-peterx@redhat.com> X-Mailer: git-send-email 2.26.2 MIME-Version: 1.0 Authentication-Results: relay.mimecast.com; auth=pass smtp.auth=CUSA124A263 smtp.mailfrom=peterx@redhat.com X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: redhat.com Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Received-SPF: pass client-ip=63.128.21.124; envelope-from=peterx@redhat.com; helo=us-smtp-delivery-124.mimecast.com X-detected-operating-system: by eggs.gnu.org: First seen = 2020/10/19 02:32:01 X-ACL-Warn: Detected OS = Linux 2.2.x-3.x [generic] [fuzzy] X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_NONE=-0.0001, RCVD_IN_MSPIKE_H5=0.001, RCVD_IN_MSPIKE_WL=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Peter Maydell , Thomas Huth , "Dr . David Alan Gilbert" , peterx@redhat.com, Juan Quintela Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" This is v5 of the series. Probably my first series that got queued/unqueue= d=0D twice.=0D =0D I found a bug in v4 that was about page sizes, however that didn't match wi= th=0D PeterM's report on big endian hosts. My manual reproduce on s390x also did= n't=0D reproduce. However after I ran the tree (with the fix) on travis (thanks= =0D Thomas for suggesting this!) I noticed that s390x passed the test too:=0D =0D https://travis-ci.com/github/xzpeter/qemu/builds/191031111=0D =0D I still got two tests that got timed out, however I also noticed that the= =0D master branch got similarly two tests that timed out:=0D =0D https://travis-ci.com/github/xzpeter/qemu/builds/191012879=0D =0D There's one difference on the failed test, however I really suspect it's=0D because the uncertainly of the travis host scheduling the tests, or the mas= ter=0D failures should really be a subset of my own branch (while it's not).=0D =0D So I decided to give it a 3rd shot... PeterM: would it be easy/possible to = kick=0D the CI again against this series for pulls? I just don't want to break Dav= e's=0D pull the 3rd time. :)=0D =0D I also picked up the other patch [1] that should also fix some rare failure= s on=0D postcopy recovery. However I bet we haven't yet encounter it, at least not= often.=0D =0D v5 changelog:=0D - added one test patch for easier debugging for migration-test=0D - added one fix patch [1] for another postcopy race=0D - fixed a bug that could trigger when host/guest page size differs=0D =0D [1] https://lore.kernel.org/qemu-devel/20201007183324.288379-1-peterx@redha= t.com/=0D =0D --------- v4 cover letter --------------=0D =0D v4:=0D - use "void */ulong" instead of "uint64_t" where proper in patch 3/4 [Dave]= =0D =0D v3:=0D - fix build on 32bit hosts & rebase=0D - remove r-bs for the last 2 patches for Dave due to the changes=0D =0D v2:=0D - add r-bs for Dave=0D - add patch "migration: Properly destroy variables on incoming side" as pat= ch 1=0D - destroy page_request_mutex in migration_incoming_state_destroy() too [Dav= e]=0D - use WITH_QEMU_LOCK_GUARD in two places where we can [Dave]=0D =0D We've seen conditional guest hangs on destination VM after postcopy recover= ed.=0D However the hang will resolve itself after a few minutes.=0D =0D The problem is: after a postcopy recovery, the prioritized postcopy queue o= n=0D the source VM is actually missing. So all the faulted threads before the= =0D postcopy recovery happened will keep halted until (accidentally) the page g= ot=0D copied by the background precopy migration stream.=0D =0D The solution is to also refresh this information after postcopy recovery. = To=0D achieve this, we need to maintain a list of faulted addresses on the=0D destination node, so that we can resend the list when necessary. This work= is=0D done via patch 2-5.=0D =0D With that, the last thing we need to do is to send this extra information t= o=0D source VM after recovered. Very luckily, this synchronization can be=0D "emulated" by sending a bunch of page requests (although these pages have b= een=0D sent previously!) to source VM just like when we've got a page fault. Even= in=0D the 1st version of the postcopy code we'll handle duplicated pages well. S= o=0D this fix does not even need a new capability bit and it'll work smoothly on= old=0D QEMUs when we migrate from them to the new QEMUs.=0D =0D Please review, thanks.=0D =0D Peter Xu (6):=0D migration: Pass incoming state into qemu_ufd_copy_ioctl()=0D migration: Introduce migrate_send_rp_message_req_pages()=0D migration: Maintain postcopy faulted addresses=0D migration: Sync requested pages after postcopy recovery=0D migration/postcopy: Release fd before going into 'postcopy-pause'=0D migration-test: Only hide error if !QTEST_LOG=0D =0D migration/migration.c | 55 ++++++++++++++++++++++++++++++----=0D migration/migration.h | 21 ++++++++++++-=0D migration/postcopy-ram.c | 25 ++++++++++++----=0D migration/savevm.c | 57 ++++++++++++++++++++++++++++++++++++=0D migration/trace-events | 3 ++=0D tests/qtest/migration-test.c | 6 +++-=0D 6 files changed, 154 insertions(+), 13 deletions(-)=0D =0D --=20=0D 2.26.2=0D =0D