From nobody Mon Feb 9 06:26:38 2026 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1665092283; cv=none; d=zohomail.com; s=zohoarc; b=jX2ZDFYwrdWj0ILW1CtFtm/y2r7PhS/gMgIIaHh0QkXIRJa62NueY9mRZJ4ACwx4nkF5SaXzzKdNx3C5XqqgK/yVC7yObsfrWkTdG2csNqa3ij17yPQOkjVfs9TXDsjCDqD6RSfPmqfGq9iUwxY6j3wm9ZZeztsj4JKP7aGpaEI= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1665092283; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=nZH926jePVfo9Jq+WbPxGtW5ubS1EygFordvFbCKPJM=; b=SE52tHKm8tMeBoTcmLaFVZFsUk0y0a816fHn6EfgXNth/gkVbZ/WMo/oMvuEQ0dNvsv4bpgFahUh5vduDGdoSGkSzZyDlbeXgM1lgsRy3uSj4q0yHnqoZeYZ8NEwnwrM5utUMlevQLb1N0QcOFDX5niNuP8gxLyvaCrA6V3B1vw= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1665092283981993.4084313602588; Thu, 6 Oct 2022 14:38:03 -0700 (PDT) Received: from localhost ([::1]:57814 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1ogYZ8-0005xq-RW for importer@patchew.org; Thu, 06 Oct 2022 17:38:02 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:33236) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWV-0002zK-MD for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:19 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.129.124]:44470) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWT-0006ls-CR for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:18 -0400 Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-57-XaPaO7_GNTCMyCKRQyyk1g-1; Thu, 06 Oct 2022 17:35:13 -0400 Received: from smtp.corp.redhat.com (int-mx10.intmail.prod.int.rdu2.redhat.com [10.11.54.10]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id C52FB85A5A6; Thu, 6 Oct 2022 21:35:12 +0000 (UTC) Received: from localhost (unknown [10.39.193.119]) by smtp.corp.redhat.com (Postfix) with ESMTP id ECEA6535939; Thu, 6 Oct 2022 21:35:11 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1665092116; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=nZH926jePVfo9Jq+WbPxGtW5ubS1EygFordvFbCKPJM=; b=YmRNK2EnJkTD+fkrPpEWkWE17OppNEXQI4l3p9v6PToeXd/zTa2I0olSEyQJ37lahyXycQ yv3mhtT5Ssdtc1i5dKroRtloYemLu5gbyf+OtcQhrc7Al78CpYYQ+wYN7qh93wjJOv4OV6 TVsdiit8IhAVLsfbGUZ5/RCgW5uC8c8= X-MC-Unique: XaPaO7_GNTCMyCKRQyyk1g-1 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Yanan Wang , sgarzare@redhat.com, "Richard W.M. Jones" , Fam Zheng , Hanna Reitz , David Hildenbrand , integration@gluster.org, qemu-block@nongnu.org, Vladimir Sementsov-Ogievskiy , Paolo Bonzini , Kevin Wolf , afaria@redhat.com, Richard Henderson , Eric Blake , "Michael S. Tsirkin" , Thomas Huth , Xie Changlong , John Snow , Eduardo Habkost , Jeff Cody , Stefan Hajnoczi , "Denis V. Lunev" , Markus Armbruster , Laurent Vivier , Peter Xu , Raphael Norwitz , Marcel Apfelbaum , =?UTF-8?q?Philippe=20Mathieu-Daud=C3=A9?= , Wen Congyang Subject: [PATCH v6 01/13] coroutine: add flag to re-queue at front of CoQueue Date: Thu, 6 Oct 2022 17:34:55 -0400 Message-Id: <20221006213507.645402-2-stefanha@redhat.com> In-Reply-To: <20221006213507.645402-1-stefanha@redhat.com> References: <20221006213507.645402-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.10 Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=170.10.129.124; envelope-from=stefanha@redhat.com; helo=us-smtp-delivery-124.mimecast.com X-Spam_score_int: -27 X-Spam_score: -2.8 X-Spam_bar: -- X-Spam_report: (-2.8 / 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_LOW=-0.7, 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: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1665092285666100005 Content-Type: text/plain; charset="utf-8" When a coroutine wakes up it may determine that it must re-queue. Normally coroutines are pushed onto the back of the CoQueue, but for fairness it may be necessary to push it onto the front of the CoQueue. Add a flag to specify that the coroutine should be pushed onto the front of the CoQueue. A later patch will use this to ensure fairness in the bounce buffer CoQueue used by the blkio BlockDriver. Signed-off-by: Stefan Hajnoczi --- include/qemu/coroutine.h | 15 +++++++++++++-- util/qemu-coroutine-lock.c | 9 +++++++-- 2 files changed, 20 insertions(+), 4 deletions(-) diff --git a/include/qemu/coroutine.h b/include/qemu/coroutine.h index 08c5bb3c76..3f418a67f6 100644 --- a/include/qemu/coroutine.h +++ b/include/qemu/coroutine.h @@ -198,14 +198,25 @@ typedef struct CoQueue { */ void qemu_co_queue_init(CoQueue *queue); =20 +typedef enum { + /* + * Enqueue at front instead of back. Use this to re-queue a request wh= en + * its wait condition is not satisfied after being woken up. + */ + CO_QUEUE_WAIT_FRONT =3D 0x1, +} CoQueueWaitFlags; + /** * Adds the current coroutine to the CoQueue and transfers control to the * caller of the coroutine. The mutex is unlocked during the wait and * locked again afterwards. */ #define qemu_co_queue_wait(queue, lock) \ - qemu_co_queue_wait_impl(queue, QEMU_MAKE_LOCKABLE(lock)) -void coroutine_fn qemu_co_queue_wait_impl(CoQueue *queue, QemuLockable *lo= ck); + qemu_co_queue_wait_impl(queue, QEMU_MAKE_LOCKABLE(lock), 0) +#define qemu_co_queue_wait_flags(queue, lock, flags) \ + qemu_co_queue_wait_impl(queue, QEMU_MAKE_LOCKABLE(lock), (flags)) +void coroutine_fn qemu_co_queue_wait_impl(CoQueue *queue, QemuLockable *lo= ck, + CoQueueWaitFlags flags); =20 /** * Removes the next coroutine from the CoQueue, and queue it to run after diff --git a/util/qemu-coroutine-lock.c b/util/qemu-coroutine-lock.c index 9ad24ab1af..0516bd2ff3 100644 --- a/util/qemu-coroutine-lock.c +++ b/util/qemu-coroutine-lock.c @@ -39,10 +39,15 @@ void qemu_co_queue_init(CoQueue *queue) QSIMPLEQ_INIT(&queue->entries); } =20 -void coroutine_fn qemu_co_queue_wait_impl(CoQueue *queue, QemuLockable *lo= ck) +void coroutine_fn qemu_co_queue_wait_impl(CoQueue *queue, QemuLockable *lo= ck, + CoQueueWaitFlags flags) { Coroutine *self =3D qemu_coroutine_self(); - QSIMPLEQ_INSERT_TAIL(&queue->entries, self, co_queue_next); + if (flags & CO_QUEUE_WAIT_FRONT) { + QSIMPLEQ_INSERT_HEAD(&queue->entries, self, co_queue_next); + } else { + QSIMPLEQ_INSERT_TAIL(&queue->entries, self, co_queue_next); + } =20 if (lock) { qemu_lockable_unlock(lock); --=20 2.37.3 From nobody Mon Feb 9 06:26:38 2026 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1665092294; cv=none; d=zohomail.com; s=zohoarc; b=XZ5K3eGwDL5dtJIsd4DoNo4qdnxrJOFL52u/VTk9SZsZvxvBkKl8r2sOWet/NDwjiw1tj6xL92N8dczQT1Hq00sT57UIKC0U3qek0d+mlQkLeoamgJZcp2t5to3hVOuWy75LyUagQ+al1LSQ5NOpo5zdQl6yZk97u+iSE8poglY= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1665092294; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=a/S2MTbWAHTzKF0qj2NsJBYCklUDmap9GNhOkhjq1a4=; b=CBXkMyRqgnhAKjdIT+6LounSMzZ/GPspKQ5ExdzBUu1MJE/gUm/2IgNS8UylmxPFmErEMsrgOjL9eNSf3HvBpXfmgjiEOrfFnNxS5NRmySQXgbSEsruLpEzYmTSsj61cptolPEbV6cMi/VUmsOGe1Sl5leZQ3dJRsG3H+9zkypQ= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1665092294692137.1167594753232; Thu, 6 Oct 2022 14:38:14 -0700 (PDT) Received: from localhost ([::1]:43718 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1ogYZI-0006Qu-Ku for importer@patchew.org; Thu, 06 Oct 2022 17:38:12 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:53698) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWd-00037i-Et for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:27 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.133.124]:42952) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWW-0006mH-T7 for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:27 -0400 Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-126-uuFB4sXpPUGOe6JqNjdAFQ-1; Thu, 06 Oct 2022 17:35:17 -0400 Received: from smtp.corp.redhat.com (int-mx03.intmail.prod.int.rdu2.redhat.com [10.11.54.3]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 32D08101A52A; Thu, 6 Oct 2022 21:35:16 +0000 (UTC) Received: from localhost (unknown [10.39.193.119]) by smtp.corp.redhat.com (Postfix) with ESMTP id E1ED4111CD4B; Thu, 6 Oct 2022 21:35:14 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1665092120; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=a/S2MTbWAHTzKF0qj2NsJBYCklUDmap9GNhOkhjq1a4=; b=TFVBA+rjvPb7nKZy4d5ZPX5duM7p6IOWfWB0bKN0D/zED5EqeGO5maHACxaXad0VE7wc0c I74eCot2WrWuXr1kpj35VLhL4aShPUhPX9oGssaaw8OF6NhbW14CvJfp5I/nfAaZQ1T+Y0 RbQngnJLwBBq20ELJ+LLpZXdmLBHnQc= X-MC-Unique: uuFB4sXpPUGOe6JqNjdAFQ-1 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Yanan Wang , sgarzare@redhat.com, "Richard W.M. Jones" , Fam Zheng , Hanna Reitz , David Hildenbrand , integration@gluster.org, qemu-block@nongnu.org, Vladimir Sementsov-Ogievskiy , Paolo Bonzini , Kevin Wolf , afaria@redhat.com, Richard Henderson , Eric Blake , "Michael S. Tsirkin" , Thomas Huth , Xie Changlong , John Snow , Eduardo Habkost , Jeff Cody , Stefan Hajnoczi , "Denis V. Lunev" , Markus Armbruster , Laurent Vivier , Peter Xu , Raphael Norwitz , Marcel Apfelbaum , =?UTF-8?q?Philippe=20Mathieu-Daud=C3=A9?= , Wen Congyang Subject: [PATCH v6 02/13] blkio: add libblkio block driver Date: Thu, 6 Oct 2022 17:34:56 -0400 Message-Id: <20221006213507.645402-3-stefanha@redhat.com> In-Reply-To: <20221006213507.645402-1-stefanha@redhat.com> References: <20221006213507.645402-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.3 Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=170.10.133.124; envelope-from=stefanha@redhat.com; helo=us-smtp-delivery-124.mimecast.com 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, SPF_PASS=-0.001, T_SPF_HELO_TEMPERROR=0.01 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1665092295988100007 Content-Type: text/plain; charset="utf-8" libblkio (https://gitlab.com/libblkio/libblkio/) is a library for high-performance disk I/O. It currently supports io_uring, virtio-blk-vhost-user, and virtio-blk-vhost-vdpa with additional drivers under development. One of the reasons for developing libblkio is that other applications besides QEMU can use it. This will be particularly useful for virtio-blk-vhost-user which applications may wish to use for connecting to qemu-storage-daemon. libblkio also gives us an opportunity to develop in Rust behind a C API that is easy to consume from QEMU. This commit adds io_uring, virtio-blk-vhost-user, and virtio-blk-vhost-vdpa BlockDrivers to QEMU using libblkio. It will be easy to add other libblkio drivers since they will share the majority of code. For now I/O buffers are copied through bounce buffers if the libblkio driver requires it. Later commits add an optimization for pre-registering guest RAM to avoid bounce buffers. The syntax is: --blockdev io_uring,node-name=3Ddrive0,filename=3Dtest.img,readonly=3Don|= off,cache.direct=3Don|off and: --blockdev virtio-blk-vhost-vdpa,node-name=3Ddrive0,path=3D/dev/vdpa...,r= eadonly=3Don|off Signed-off-by: Stefan Hajnoczi Acked-by: Markus Armbruster --- MAINTAINERS | 6 + meson_options.txt | 2 + qapi/block-core.json | 75 ++- meson.build | 9 + block/blkio.c | 830 ++++++++++++++++++++++++++++++++++ tests/qtest/modules-test.c | 3 + block/meson.build | 1 + scripts/meson-buildoptions.sh | 3 + 8 files changed, 925 insertions(+), 4 deletions(-) create mode 100644 block/blkio.c diff --git a/MAINTAINERS b/MAINTAINERS index e1530b51a2..0dcae6168a 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -3403,6 +3403,12 @@ L: qemu-block@nongnu.org S: Maintained F: block/vdi.c =20 +blkio +M: Stefan Hajnoczi +L: qemu-block@nongnu.org +S: Maintained +F: block/blkio.c + iSCSI M: Ronnie Sahlberg M: Paolo Bonzini diff --git a/meson_options.txt b/meson_options.txt index 79c6af18d5..66128178bf 100644 --- a/meson_options.txt +++ b/meson_options.txt @@ -117,6 +117,8 @@ option('bzip2', type : 'feature', value : 'auto', description: 'bzip2 support for DMG images') option('cap_ng', type : 'feature', value : 'auto', description: 'cap_ng support') +option('blkio', type : 'feature', value : 'auto', + description: 'libblkio block device driver') option('bpf', type : 'feature', value : 'auto', description: 'eBPF support') option('cocoa', type : 'feature', value : 'auto', diff --git a/qapi/block-core.json b/qapi/block-core.json index f21fa235f2..6c6ae2885c 100644 --- a/qapi/block-core.json +++ b/qapi/block-core.json @@ -2951,11 +2951,18 @@ 'file', 'snapshot-access', 'ftp', 'ftps', 'gluster', {'name': 'host_cdrom', 'if': 'HAVE_HOST_BLOCK_DEVICE' }, {'name': 'host_device', 'if': 'HAVE_HOST_BLOCK_DEVICE' }, - 'http', 'https', 'iscsi', - 'luks', 'nbd', 'nfs', 'null-aio', 'null-co', 'nvme', 'parallel= s', - 'preallocate', 'qcow', 'qcow2', 'qed', 'quorum', 'raw', 'rbd', + 'http', 'https', + { 'name': 'io_uring', 'if': 'CONFIG_BLKIO' }, + 'iscsi', + 'luks', 'nbd', 'nfs', 'null-aio', 'null-co', 'nvme', + { 'name': 'nvme-io_uring', 'if': 'CONFIG_BLKIO' }, + 'parallels', 'preallocate', 'qcow', 'qcow2', 'qed', 'quorum', + 'raw', 'rbd', { 'name': 'replication', 'if': 'CONFIG_REPLICATION' }, - 'ssh', 'throttle', 'vdi', 'vhdx', 'vmdk', 'vpc', 'vvfat' ] } + 'ssh', 'throttle', 'vdi', 'vhdx', + { 'name': 'virtio-blk-vhost-user', 'if': 'CONFIG_BLKIO' }, + { 'name': 'virtio-blk-vhost-vdpa', 'if': 'CONFIG_BLKIO' }, + 'vmdk', 'vpc', 'vvfat' ] } =20 ## # @BlockdevOptionsFile: @@ -3678,6 +3685,58 @@ '*debug': 'int', '*logfile': 'str' } } =20 +## +# @BlockdevOptionsIoUring: +# +# Driver specific block device options for the io_uring backend. +# +# @filename: path to the image file +# +# Since: 7.2 +## +{ 'struct': 'BlockdevOptionsIoUring', + 'data': { 'filename': 'str' }, + 'if': 'CONFIG_BLKIO' } + +## +# @BlockdevOptionsNvmeIoUring: +# +# Driver specific block device options for the nvme-io_uring backend. +# +# @filename: path to the image file +# +# Since: 7.2 +## +{ 'struct': 'BlockdevOptionsNvmeIoUring', + 'data': { 'filename': 'str' }, + 'if': 'CONFIG_BLKIO' } + +## +# @BlockdevOptionsVirtioBlkVhostUser: +# +# Driver specific block device options for the virtio-blk-vhost-user backe= nd. +# +# @path: path to the vhost-user UNIX domain socket. +# +# Since: 7.2 +## +{ 'struct': 'BlockdevOptionsVirtioBlkVhostUser', + 'data': { 'path': 'str' }, + 'if': 'CONFIG_BLKIO' } + +## +# @BlockdevOptionsVirtioBlkVhostVdpa: +# +# Driver specific block device options for the virtio-blk-vhost-vdpa backe= nd. +# +# @path: path to the vhost-vdpa character device. +# +# Since: 7.2 +## +{ 'struct': 'BlockdevOptionsVirtioBlkVhostVdpa', + 'data': { 'path': 'str' }, + 'if': 'CONFIG_BLKIO' } + ## # @IscsiTransport: # @@ -4305,6 +4364,8 @@ 'if': 'HAVE_HOST_BLOCK_DEVICE' }, 'http': 'BlockdevOptionsCurlHttp', 'https': 'BlockdevOptionsCurlHttps', + 'io_uring': { 'type': 'BlockdevOptionsIoUring', + 'if': 'CONFIG_BLKIO' }, 'iscsi': 'BlockdevOptionsIscsi', 'luks': 'BlockdevOptionsLUKS', 'nbd': 'BlockdevOptionsNbd', @@ -4327,6 +4388,12 @@ 'throttle': 'BlockdevOptionsThrottle', 'vdi': 'BlockdevOptionsGenericFormat', 'vhdx': 'BlockdevOptionsGenericFormat', + 'virtio-blk-vhost-user': + { 'type': 'BlockdevOptionsVirtioBlkVhostUser', + 'if': 'CONFIG_BLKIO' }, + 'virtio-blk-vhost-vdpa': + { 'type': 'BlockdevOptionsVirtioBlkVhostVdpa', + 'if': 'CONFIG_BLKIO' }, 'vmdk': 'BlockdevOptionsGenericCOWFormat', 'vpc': 'BlockdevOptionsGenericFormat', 'vvfat': 'BlockdevOptionsVVFAT' diff --git a/meson.build b/meson.build index b686dfef75..1e9068adf6 100644 --- a/meson.build +++ b/meson.build @@ -771,6 +771,13 @@ if not get_option('virglrenderer').auto() or have_syst= em or have_vhost_user_gpu required: get_option('virglrenderer'), kwargs: static_kwargs) endif +blkio =3D not_found +if not get_option('blkio').auto() or have_block + blkio =3D dependency('blkio', + method: 'pkg-config', + required: get_option('blkio'), + kwargs: static_kwargs) +endif curl =3D not_found if not get_option('curl').auto() or have_block curl =3D dependency('libcurl', version: '>=3D7.29.0', @@ -1815,6 +1822,7 @@ config_host_data.set('CONFIG_LIBUDEV', libudev.found(= )) config_host_data.set('CONFIG_LZO', lzo.found()) config_host_data.set('CONFIG_MPATH', mpathpersist.found()) config_host_data.set('CONFIG_MPATH_NEW_API', mpathpersist_new_api) +config_host_data.set('CONFIG_BLKIO', blkio.found()) config_host_data.set('CONFIG_CURL', curl.found()) config_host_data.set('CONFIG_CURSES', curses.found()) config_host_data.set('CONFIG_GBM', gbm.found()) @@ -3869,6 +3877,7 @@ summary_info +=3D {'PAM': pam} summary_info +=3D {'iconv support': iconv} summary_info +=3D {'curses support': curses} summary_info +=3D {'virgl support': virgl} +summary_info +=3D {'blkio support': blkio} summary_info +=3D {'curl support': curl} summary_info +=3D {'Multipath support': mpathpersist} summary_info +=3D {'PNG support': png} diff --git a/block/blkio.c b/block/blkio.c new file mode 100644 index 0000000000..9a79789a39 --- /dev/null +++ b/block/blkio.c @@ -0,0 +1,830 @@ +/* SPDX-License-Identifier: LGPL-2.1-or-later */ +/* + * libblkio BlockDriver + * + * Copyright Red Hat, Inc. + * + * Author: + * Stefan Hajnoczi + */ + +#include "qemu/osdep.h" +#include +#include "block/block_int.h" +#include "qapi/error.h" +#include "qapi/qmp/qdict.h" +#include "qemu/module.h" + +/* + * Keep the QEMU BlockDriver names identical to the libblkio driver names. + * Using macros instead of typing out the string literals avoids typos. + */ +#define DRIVER_IO_URING "io_uring" +#define DRIVER_NVME_IO_URING "nvme-io_uring" +#define DRIVER_VIRTIO_BLK_VHOST_USER "virtio-blk-vhost-user" +#define DRIVER_VIRTIO_BLK_VHOST_VDPA "virtio-blk-vhost-vdpa" + +/* + * Allocated bounce buffers are kept in a list sorted by buffer address. + */ +typedef struct BlkioBounceBuf { + QLIST_ENTRY(BlkioBounceBuf) next; + + /* The bounce buffer */ + struct iovec buf; +} BlkioBounceBuf; + +typedef struct { + /* + * libblkio is not thread-safe so this lock protects ->blkio and + * ->blkioq. + */ + QemuMutex blkio_lock; + struct blkio *blkio; + struct blkioq *blkioq; /* make this multi-queue in the future... */ + int completion_fd; + + /* + * Polling fetches the next completion into this field. + * + * No lock is necessary since only one thread calls aio_poll() and inv= okes + * fd and poll handlers. + */ + struct blkio_completion poll_completion; + + /* + * Protects ->bounce_pool, ->bounce_bufs, ->bounce_available. + * + * Lock ordering: ->bounce_lock before ->blkio_lock. + */ + CoMutex bounce_lock; + + /* Bounce buffer pool */ + struct blkio_mem_region bounce_pool; + + /* Sorted list of allocated bounce buffers */ + QLIST_HEAD(, BlkioBounceBuf) bounce_bufs; + + /* Queue for coroutines waiting for bounce buffer space */ + CoQueue bounce_available; + + /* The value of the "mem-region-alignment" property */ + size_t mem_region_alignment; + + /* Can we skip adding/deleting blkio_mem_regions? */ + bool needs_mem_regions; +} BDRVBlkioState; + +/* Called with s->bounce_lock held */ +static int blkio_resize_bounce_pool(BDRVBlkioState *s, int64_t bytes) +{ + /* There can be no allocated bounce buffers during resize */ + assert(QLIST_EMPTY(&s->bounce_bufs)); + + /* Pad size to reduce frequency of resize calls */ + bytes +=3D 128 * 1024; + + WITH_QEMU_LOCK_GUARD(&s->blkio_lock) { + int ret; + + if (s->bounce_pool.addr) { + blkio_unmap_mem_region(s->blkio, &s->bounce_pool); + blkio_free_mem_region(s->blkio, &s->bounce_pool); + memset(&s->bounce_pool, 0, sizeof(s->bounce_pool)); + } + + /* Automatically freed when s->blkio is destroyed */ + ret =3D blkio_alloc_mem_region(s->blkio, &s->bounce_pool, bytes); + if (ret < 0) { + return ret; + } + + ret =3D blkio_map_mem_region(s->blkio, &s->bounce_pool); + if (ret < 0) { + blkio_free_mem_region(s->blkio, &s->bounce_pool); + memset(&s->bounce_pool, 0, sizeof(s->bounce_pool)); + return ret; + } + } + + return 0; +} + +/* Called with s->bounce_lock held */ +static bool +blkio_do_alloc_bounce_buffer(BDRVBlkioState *s, BlkioBounceBuf *bounce, + int64_t bytes) +{ + void *addr =3D s->bounce_pool.addr; + BlkioBounceBuf *cur =3D NULL; + BlkioBounceBuf *prev =3D NULL; + ptrdiff_t space; + + /* + * This is just a linear search over the holes between requests. An + * efficient allocator would be nice. + */ + QLIST_FOREACH(cur, &s->bounce_bufs, next) { + space =3D cur->buf.iov_base - addr; + if (bytes <=3D space) { + QLIST_INSERT_BEFORE(cur, bounce, next); + bounce->buf.iov_base =3D addr; + bounce->buf.iov_len =3D bytes; + return true; + } + + addr =3D cur->buf.iov_base + cur->buf.iov_len; + prev =3D cur; + } + + /* Is there space after the last request? */ + space =3D s->bounce_pool.addr + s->bounce_pool.len - addr; + if (bytes > space) { + return false; + } + if (prev) { + QLIST_INSERT_AFTER(prev, bounce, next); + } else { + QLIST_INSERT_HEAD(&s->bounce_bufs, bounce, next); + } + bounce->buf.iov_base =3D addr; + bounce->buf.iov_len =3D bytes; + return true; +} + +static int coroutine_fn +blkio_alloc_bounce_buffer(BDRVBlkioState *s, BlkioBounceBuf *bounce, + int64_t bytes) +{ + /* + * Ensure fairness: first time around we join the back of the queue, + * subsequently we join the front so we don't lose our place. + */ + CoQueueWaitFlags wait_flags =3D 0; + + QEMU_LOCK_GUARD(&s->bounce_lock); + + /* Ensure fairness: don't even try if other requests are already waiti= ng */ + if (!qemu_co_queue_empty(&s->bounce_available)) { + qemu_co_queue_wait_flags(&s->bounce_available, &s->bounce_lock, + wait_flags); + wait_flags =3D CO_QUEUE_WAIT_FRONT; + } + + while (true) { + if (blkio_do_alloc_bounce_buffer(s, bounce, bytes)) { + /* Kick the next queued request since there may be space */ + qemu_co_queue_next(&s->bounce_available); + return 0; + } + + /* + * If there are no in-flight requests then the pool was simply too + * small. + */ + if (QLIST_EMPTY(&s->bounce_bufs)) { + bool ok; + int ret; + + ret =3D blkio_resize_bounce_pool(s, bytes); + if (ret < 0) { + /* Kick the next queued request since that may fail too */ + qemu_co_queue_next(&s->bounce_available); + return ret; + } + + ok =3D blkio_do_alloc_bounce_buffer(s, bounce, bytes); + assert(ok); /* must have space this time */ + return 0; + } + + qemu_co_queue_wait_flags(&s->bounce_available, &s->bounce_lock, + wait_flags); + wait_flags =3D CO_QUEUE_WAIT_FRONT; + } +} + +static void coroutine_fn blkio_free_bounce_buffer(BDRVBlkioState *s, + BlkioBounceBuf *bounce) +{ + QEMU_LOCK_GUARD(&s->bounce_lock); + + QLIST_REMOVE(bounce, next); + + /* Wake up waiting coroutines since space may now be available */ + qemu_co_queue_next(&s->bounce_available); +} + +/* For async to .bdrv_co_*() conversion */ +typedef struct { + Coroutine *coroutine; + int ret; +} BlkioCoData; + +static void blkio_completion_fd_read(void *opaque) +{ + BlockDriverState *bs =3D opaque; + BDRVBlkioState *s =3D bs->opaque; + uint64_t val; + int ret; + + /* Polling may have already fetched a completion */ + if (s->poll_completion.user_data !=3D NULL) { + BlkioCoData *cod =3D s->poll_completion.user_data; + cod->ret =3D s->poll_completion.ret; + + /* Clear it in case aio_co_wake() enters a nested event loop */ + s->poll_completion.user_data =3D NULL; + + aio_co_wake(cod->coroutine); + } + + /* Reset completion fd status */ + ret =3D read(s->completion_fd, &val, sizeof(val)); + + /* Ignore errors, there's nothing we can do */ + (void)ret; + + /* + * Reading one completion at a time makes nested event loop re-entrancy + * simple. Change this loop to get multiple completions in one go if it + * becomes a performance bottleneck. + */ + while (true) { + struct blkio_completion completion; + + WITH_QEMU_LOCK_GUARD(&s->blkio_lock) { + ret =3D blkioq_do_io(s->blkioq, &completion, 0, 1, NULL); + } + if (ret !=3D 1) { + break; + } + + BlkioCoData *cod =3D completion.user_data; + cod->ret =3D completion.ret; + aio_co_wake(cod->coroutine); + } +} + +static bool blkio_completion_fd_poll(void *opaque) +{ + BlockDriverState *bs =3D opaque; + BDRVBlkioState *s =3D bs->opaque; + int ret; + + /* Just in case we already fetched a completion */ + if (s->poll_completion.user_data !=3D NULL) { + return true; + } + + WITH_QEMU_LOCK_GUARD(&s->blkio_lock) { + ret =3D blkioq_do_io(s->blkioq, &s->poll_completion, 0, 1, NULL); + } + return ret =3D=3D 1; +} + +static void blkio_completion_fd_poll_ready(void *opaque) +{ + blkio_completion_fd_read(opaque); +} + +static void blkio_attach_aio_context(BlockDriverState *bs, + AioContext *new_context) +{ + BDRVBlkioState *s =3D bs->opaque; + + aio_set_fd_handler(new_context, + s->completion_fd, + false, + blkio_completion_fd_read, + NULL, + blkio_completion_fd_poll, + blkio_completion_fd_poll_ready, + bs); +} + +static void blkio_detach_aio_context(BlockDriverState *bs) +{ + BDRVBlkioState *s =3D bs->opaque; + + aio_set_fd_handler(bdrv_get_aio_context(bs), + s->completion_fd, + false, NULL, NULL, NULL, NULL, NULL); +} + +/* Call with s->blkio_lock held to submit I/O after enqueuing a new reques= t */ +static void blkio_submit_io(BlockDriverState *bs) +{ + if (qatomic_read(&bs->io_plugged) =3D=3D 0) { + BDRVBlkioState *s =3D bs->opaque; + + blkioq_do_io(s->blkioq, NULL, 0, 0, NULL); + } +} + +static int coroutine_fn +blkio_co_pdiscard(BlockDriverState *bs, int64_t offset, int64_t bytes) +{ + BDRVBlkioState *s =3D bs->opaque; + BlkioCoData cod =3D { + .coroutine =3D qemu_coroutine_self(), + }; + + WITH_QEMU_LOCK_GUARD(&s->blkio_lock) { + blkioq_discard(s->blkioq, offset, bytes, &cod, 0); + blkio_submit_io(bs); + } + + qemu_coroutine_yield(); + return cod.ret; +} + +static int coroutine_fn +blkio_co_preadv(BlockDriverState *bs, int64_t offset, int64_t bytes, + QEMUIOVector *qiov, BdrvRequestFlags flags) +{ + BlkioCoData cod =3D { + .coroutine =3D qemu_coroutine_self(), + }; + BDRVBlkioState *s =3D bs->opaque; + bool use_bounce_buffer =3D s->needs_mem_regions; + BlkioBounceBuf bounce; + struct iovec *iov =3D qiov->iov; + int iovcnt =3D qiov->niov; + + if (use_bounce_buffer) { + int ret =3D blkio_alloc_bounce_buffer(s, &bounce, bytes); + if (ret < 0) { + return ret; + } + + iov =3D &bounce.buf; + iovcnt =3D 1; + } + + WITH_QEMU_LOCK_GUARD(&s->blkio_lock) { + blkioq_readv(s->blkioq, offset, iov, iovcnt, &cod, 0); + blkio_submit_io(bs); + } + + qemu_coroutine_yield(); + + if (use_bounce_buffer) { + if (cod.ret =3D=3D 0) { + qemu_iovec_from_buf(qiov, 0, + bounce.buf.iov_base, + bounce.buf.iov_len); + } + + blkio_free_bounce_buffer(s, &bounce); + } + + return cod.ret; +} + +static int coroutine_fn blkio_co_pwritev(BlockDriverState *bs, int64_t off= set, + int64_t bytes, QEMUIOVector *qiov, BdrvRequestFlags flags) +{ + uint32_t blkio_flags =3D (flags & BDRV_REQ_FUA) ? BLKIO_REQ_FUA : 0; + BlkioCoData cod =3D { + .coroutine =3D qemu_coroutine_self(), + }; + BDRVBlkioState *s =3D bs->opaque; + bool use_bounce_buffer =3D s->needs_mem_regions; + BlkioBounceBuf bounce; + struct iovec *iov =3D qiov->iov; + int iovcnt =3D qiov->niov; + + if (use_bounce_buffer) { + int ret =3D blkio_alloc_bounce_buffer(s, &bounce, bytes); + if (ret < 0) { + return ret; + } + + qemu_iovec_to_buf(qiov, 0, bounce.buf.iov_base, bytes); + iov =3D &bounce.buf; + iovcnt =3D 1; + } + + WITH_QEMU_LOCK_GUARD(&s->blkio_lock) { + blkioq_writev(s->blkioq, offset, iov, iovcnt, &cod, blkio_flags); + blkio_submit_io(bs); + } + + qemu_coroutine_yield(); + + if (use_bounce_buffer) { + blkio_free_bounce_buffer(s, &bounce); + } + + return cod.ret; +} + +static int coroutine_fn blkio_co_flush(BlockDriverState *bs) +{ + BDRVBlkioState *s =3D bs->opaque; + BlkioCoData cod =3D { + .coroutine =3D qemu_coroutine_self(), + }; + + WITH_QEMU_LOCK_GUARD(&s->blkio_lock) { + blkioq_flush(s->blkioq, &cod, 0); + blkio_submit_io(bs); + } + + qemu_coroutine_yield(); + return cod.ret; +} + +static int coroutine_fn blkio_co_pwrite_zeroes(BlockDriverState *bs, + int64_t offset, int64_t bytes, BdrvRequestFlags flags) +{ + BDRVBlkioState *s =3D bs->opaque; + BlkioCoData cod =3D { + .coroutine =3D qemu_coroutine_self(), + }; + uint32_t blkio_flags =3D 0; + + if (flags & BDRV_REQ_FUA) { + blkio_flags |=3D BLKIO_REQ_FUA; + } + if (!(flags & BDRV_REQ_MAY_UNMAP)) { + blkio_flags |=3D BLKIO_REQ_NO_UNMAP; + } + if (flags & BDRV_REQ_NO_FALLBACK) { + blkio_flags |=3D BLKIO_REQ_NO_FALLBACK; + } + + WITH_QEMU_LOCK_GUARD(&s->blkio_lock) { + blkioq_write_zeroes(s->blkioq, offset, bytes, &cod, blkio_flags); + blkio_submit_io(bs); + } + + qemu_coroutine_yield(); + return cod.ret; +} + +static void blkio_io_unplug(BlockDriverState *bs) +{ + BDRVBlkioState *s =3D bs->opaque; + + WITH_QEMU_LOCK_GUARD(&s->blkio_lock) { + blkio_submit_io(bs); + } +} + +static int blkio_io_uring_open(BlockDriverState *bs, QDict *options, int f= lags, + Error **errp) +{ + const char *filename =3D qdict_get_str(options, "filename"); + BDRVBlkioState *s =3D bs->opaque; + int ret; + + ret =3D blkio_set_str(s->blkio, "path", filename); + qdict_del(options, "filename"); + if (ret < 0) { + error_setg_errno(errp, -ret, "failed to set path: %s", + blkio_get_error_msg()); + return ret; + } + + if (flags & BDRV_O_NOCACHE) { + ret =3D blkio_set_bool(s->blkio, "direct", true); + if (ret < 0) { + error_setg_errno(errp, -ret, "failed to set direct: %s", + blkio_get_error_msg()); + return ret; + } + } + + return 0; +} + +static int blkio_nvme_io_uring(BlockDriverState *bs, QDict *options, int f= lags, + Error **errp) +{ + const char *filename =3D qdict_get_str(options, "filename"); + BDRVBlkioState *s =3D bs->opaque; + int ret; + + ret =3D blkio_set_str(s->blkio, "path", filename); + qdict_del(options, "filename"); + if (ret < 0) { + error_setg_errno(errp, -ret, "failed to set path: %s", + blkio_get_error_msg()); + return ret; + } + + if (!(flags & BDRV_O_NOCACHE)) { + error_setg(errp, "cache.direct=3Doff is not supported"); + return -EINVAL; + } + + return 0; +} + +static int blkio_virtio_blk_common_open(BlockDriverState *bs, + QDict *options, int flags, Error **errp) +{ + const char *path =3D qdict_get_try_str(options, "path"); + BDRVBlkioState *s =3D bs->opaque; + int ret; + + if (!path) { + error_setg(errp, "missing 'path' option"); + return -EINVAL; + } + + ret =3D blkio_set_str(s->blkio, "path", path); + qdict_del(options, "path"); + if (ret < 0) { + error_setg_errno(errp, -ret, "failed to set path: %s", + blkio_get_error_msg()); + return ret; + } + + if (!(flags & BDRV_O_NOCACHE)) { + error_setg(errp, "cache.direct=3Doff is not supported"); + return -EINVAL; + } + return 0; +} + +static int blkio_file_open(BlockDriverState *bs, QDict *options, int flags, + Error **errp) +{ + const char *blkio_driver =3D bs->drv->protocol_name; + BDRVBlkioState *s =3D bs->opaque; + int ret; + + ret =3D blkio_create(blkio_driver, &s->blkio); + if (ret < 0) { + error_setg_errno(errp, -ret, "blkio_create failed: %s", + blkio_get_error_msg()); + return ret; + } + + if (strcmp(blkio_driver, DRIVER_IO_URING) =3D=3D 0) { + ret =3D blkio_io_uring_open(bs, options, flags, errp); + } else if (strcmp(blkio_driver, DRIVER_NVME_IO_URING) =3D=3D 0) { + ret =3D blkio_nvme_io_uring(bs, options, flags, errp); + } else if (strcmp(blkio_driver, DRIVER_VIRTIO_BLK_VHOST_USER) =3D=3D 0= ) { + ret =3D blkio_virtio_blk_common_open(bs, options, flags, errp); + } else if (strcmp(blkio_driver, DRIVER_VIRTIO_BLK_VHOST_VDPA) =3D=3D 0= ) { + ret =3D blkio_virtio_blk_common_open(bs, options, flags, errp); + } else { + g_assert_not_reached(); + } + if (ret < 0) { + blkio_destroy(&s->blkio); + return ret; + } + + if (!(flags & BDRV_O_RDWR)) { + ret =3D blkio_set_bool(s->blkio, "read-only", true); + if (ret < 0) { + error_setg_errno(errp, -ret, "failed to set read-only: %s", + blkio_get_error_msg()); + blkio_destroy(&s->blkio); + return ret; + } + } + + ret =3D blkio_connect(s->blkio); + if (ret < 0) { + error_setg_errno(errp, -ret, "blkio_connect failed: %s", + blkio_get_error_msg()); + blkio_destroy(&s->blkio); + return ret; + } + + ret =3D blkio_get_bool(s->blkio, + "needs-mem-regions", + &s->needs_mem_regions); + if (ret < 0) { + error_setg_errno(errp, -ret, + "failed to get needs-mem-regions: %s", + blkio_get_error_msg()); + blkio_destroy(&s->blkio); + return ret; + } + + ret =3D blkio_get_uint64(s->blkio, + "mem-region-alignment", + &s->mem_region_alignment); + if (ret < 0) { + error_setg_errno(errp, -ret, + "failed to get mem-region-alignment: %s", + blkio_get_error_msg()); + blkio_destroy(&s->blkio); + return ret; + } + + ret =3D blkio_start(s->blkio); + if (ret < 0) { + error_setg_errno(errp, -ret, "blkio_start failed: %s", + blkio_get_error_msg()); + blkio_destroy(&s->blkio); + return ret; + } + + bs->supported_write_flags =3D BDRV_REQ_FUA; + bs->supported_zero_flags =3D BDRV_REQ_FUA | BDRV_REQ_MAY_UNMAP | + BDRV_REQ_NO_FALLBACK; + + qemu_mutex_init(&s->blkio_lock); + qemu_co_mutex_init(&s->bounce_lock); + qemu_co_queue_init(&s->bounce_available); + QLIST_INIT(&s->bounce_bufs); + s->blkioq =3D blkio_get_queue(s->blkio, 0); + s->completion_fd =3D blkioq_get_completion_fd(s->blkioq); + + blkio_attach_aio_context(bs, bdrv_get_aio_context(bs)); + return 0; +} + +static void blkio_close(BlockDriverState *bs) +{ + BDRVBlkioState *s =3D bs->opaque; + + /* There is no destroy() API for s->bounce_lock */ + + qemu_mutex_destroy(&s->blkio_lock); + blkio_detach_aio_context(bs); + blkio_destroy(&s->blkio); +} + +static int64_t blkio_getlength(BlockDriverState *bs) +{ + BDRVBlkioState *s =3D bs->opaque; + uint64_t capacity; + int ret; + + WITH_QEMU_LOCK_GUARD(&s->blkio_lock) { + ret =3D blkio_get_uint64(s->blkio, "capacity", &capacity); + } + if (ret < 0) { + return -ret; + } + + return capacity; +} + +static int blkio_get_info(BlockDriverState *bs, BlockDriverInfo *bdi) +{ + return 0; +} + +static void blkio_refresh_limits(BlockDriverState *bs, Error **errp) +{ + BDRVBlkioState *s =3D bs->opaque; + QEMU_LOCK_GUARD(&s->blkio_lock); + int value; + int ret; + + ret =3D blkio_get_int(s->blkio, "request-alignment", &value); + if (ret < 0) { + error_setg_errno(errp, -ret, "failed to get \"request-alignment\":= %s", + blkio_get_error_msg()); + return; + } + bs->bl.request_alignment =3D value; + if (bs->bl.request_alignment < 1 || + bs->bl.request_alignment >=3D INT_MAX || + !is_power_of_2(bs->bl.request_alignment)) { + error_setg(errp, "invalid \"request-alignment\" value %" PRIu32 ",= " + "must be a power of 2 less than INT_MAX", + bs->bl.request_alignment); + return; + } + + ret =3D blkio_get_int(s->blkio, "optimal-io-size", &value); + if (ret < 0) { + error_setg_errno(errp, -ret, "failed to get \"optimal-io-size\": %= s", + blkio_get_error_msg()); + return; + } + bs->bl.opt_transfer =3D value; + if (bs->bl.opt_transfer > INT_MAX || + (bs->bl.opt_transfer % bs->bl.request_alignment)) { + error_setg(errp, "invalid \"optimal-io-size\" value %" PRIu32 ", m= ust " + "be a multiple of %" PRIu32, bs->bl.opt_transfer, + bs->bl.request_alignment); + return; + } + + ret =3D blkio_get_int(s->blkio, "max-transfer", &value); + if (ret < 0) { + error_setg_errno(errp, -ret, "failed to get \"max-transfer\": %s", + blkio_get_error_msg()); + return; + } + bs->bl.max_transfer =3D value; + if ((bs->bl.max_transfer % bs->bl.request_alignment) || + (bs->bl.opt_transfer && (bs->bl.max_transfer % bs->bl.opt_transfer= ))) { + error_setg(errp, "invalid \"max-transfer\" value %" PRIu32 ", must= be " + "a multiple of %" PRIu32 " and %" PRIu32 " (if non-zero= )", + bs->bl.max_transfer, bs->bl.request_alignment, + bs->bl.opt_transfer); + return; + } + + ret =3D blkio_get_int(s->blkio, "buf-alignment", &value); + if (ret < 0) { + error_setg_errno(errp, -ret, "failed to get \"buf-alignment\": %s", + blkio_get_error_msg()); + return; + } + if (value < 1) { + error_setg(errp, "invalid \"buf-alignment\" value %d, must be " + "positive", value); + return; + } + bs->bl.min_mem_alignment =3D value; + + ret =3D blkio_get_int(s->blkio, "optimal-buf-alignment", &value); + if (ret < 0) { + error_setg_errno(errp, -ret, + "failed to get \"optimal-buf-alignment\": %s", + blkio_get_error_msg()); + return; + } + if (value < 1) { + error_setg(errp, "invalid \"optimal-buf-alignment\" value %d, " + "must be positive", value); + return; + } + bs->bl.opt_mem_alignment =3D value; + + ret =3D blkio_get_int(s->blkio, "max-segments", &bs->bl.max_iov); + if (ret < 0) { + error_setg_errno(errp, -ret, "failed to get \"max-segments\": %s", + blkio_get_error_msg()); + return; + } + if (value < 1) { + error_setg(errp, "invalid \"max-segments\" value %d, must be posit= ive", + bs->bl.max_iov); + return; + } +} + +/* + * TODO + * Missing libblkio APIs: + * - block_status + * - co_invalidate_cache + * + * Out of scope? + * - create + * - truncate + */ + +#define BLKIO_DRIVER(name, ...) \ + { \ + .format_name =3D name, \ + .protocol_name =3D name, \ + .instance_size =3D sizeof(BDRVBlkioState), \ + .bdrv_file_open =3D blkio_file_open, \ + .bdrv_close =3D blkio_close, \ + .bdrv_getlength =3D blkio_getlength, \ + .bdrv_get_info =3D blkio_get_info, \ + .bdrv_attach_aio_context =3D blkio_attach_aio_context, \ + .bdrv_detach_aio_context =3D blkio_detach_aio_context, \ + .bdrv_co_pdiscard =3D blkio_co_pdiscard, \ + .bdrv_co_preadv =3D blkio_co_preadv, \ + .bdrv_co_pwritev =3D blkio_co_pwritev, \ + .bdrv_co_flush_to_disk =3D blkio_co_flush, \ + .bdrv_co_pwrite_zeroes =3D blkio_co_pwrite_zeroes, \ + .bdrv_io_unplug =3D blkio_io_unplug, \ + .bdrv_refresh_limits =3D blkio_refresh_limits, \ + __VA_ARGS__ \ + } + +static BlockDriver bdrv_io_uring =3D BLKIO_DRIVER( + DRIVER_IO_URING, + .bdrv_needs_filename =3D true, +); + +static BlockDriver bdrv_nvme_io_uring =3D BLKIO_DRIVER( + DRIVER_NVME_IO_URING, + .bdrv_needs_filename =3D true, +); + +static BlockDriver bdrv_virtio_blk_vhost_user =3D BLKIO_DRIVER( + DRIVER_VIRTIO_BLK_VHOST_USER +); + +static BlockDriver bdrv_virtio_blk_vhost_vdpa =3D BLKIO_DRIVER( + DRIVER_VIRTIO_BLK_VHOST_VDPA +); + +static void bdrv_blkio_init(void) +{ + bdrv_register(&bdrv_io_uring); + bdrv_register(&bdrv_nvme_io_uring); + bdrv_register(&bdrv_virtio_blk_vhost_user); + bdrv_register(&bdrv_virtio_blk_vhost_vdpa); +} + +block_init(bdrv_blkio_init); diff --git a/tests/qtest/modules-test.c b/tests/qtest/modules-test.c index 88217686e1..be2575ae6d 100644 --- a/tests/qtest/modules-test.c +++ b/tests/qtest/modules-test.c @@ -16,6 +16,9 @@ static void test_modules_load(const void *data) int main(int argc, char *argv[]) { const char *modules[] =3D { +#ifdef CONFIG_BLKIO + "block-", "blkio", +#endif #ifdef CONFIG_CURL "block-", "curl", #endif diff --git a/block/meson.build b/block/meson.build index 60bc305597..500878f082 100644 --- a/block/meson.build +++ b/block/meson.build @@ -92,6 +92,7 @@ block_modules =3D {} =20 modsrc =3D [] foreach m : [ + [blkio, 'blkio', files('blkio.c')], [curl, 'curl', files('curl.c')], [glusterfs, 'gluster', files('gluster.c')], [libiscsi, 'iscsi', [files('iscsi.c'), libm]], diff --git a/scripts/meson-buildoptions.sh b/scripts/meson-buildoptions.sh index eb3267bef5..2cb0de5601 100644 --- a/scripts/meson-buildoptions.sh +++ b/scripts/meson-buildoptions.sh @@ -67,6 +67,7 @@ meson_options_help() { printf "%s\n" ' auth-pam PAM access control' printf "%s\n" ' avx2 AVX2 optimizations' printf "%s\n" ' avx512f AVX512F optimizations' + printf "%s\n" ' blkio libblkio block device driver' printf "%s\n" ' bochs bochs image format support' printf "%s\n" ' bpf eBPF support' printf "%s\n" ' brlapi brlapi character device driver' @@ -198,6 +199,8 @@ _meson_option_parse() { --disable-gcov) printf "%s" -Db_coverage=3Dfalse ;; --enable-lto) printf "%s" -Db_lto=3Dtrue ;; --disable-lto) printf "%s" -Db_lto=3Dfalse ;; + --enable-blkio) printf "%s" -Dblkio=3Denabled ;; + --disable-blkio) printf "%s" -Dblkio=3Ddisabled ;; --block-drv-ro-whitelist=3D*) quote_sh "-Dblock_drv_ro_whitelist=3D$2"= ;; --block-drv-rw-whitelist=3D*) quote_sh "-Dblock_drv_rw_whitelist=3D$2"= ;; --enable-block-drv-whitelist-in-tools) printf "%s" -Dblock_drv_whiteli= st_in_tools=3Dtrue ;; --=20 2.37.3 From nobody Mon Feb 9 06:26:38 2026 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1665092282; cv=none; d=zohomail.com; s=zohoarc; b=ZW/lGLWKKng7LshCnrNmBbp58Y+Dmfo4A+7LY5hMQ8+ex+yNpuqetjOabyVxQbobF8zMrtSRMSBhdf0dk8YaVX5ogtblqTqZfiLk2L1YB4OgEwlNjaIELA1fC4Es/Q46BmS+wlBwI5sN6oZtFQTkcRYSztNdjqjCRIL9FVSwxvg= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1665092282; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=3mgJt1ynd7ZiWF7e8csGQVMXxZYrxlOZeWtbVF2Vfrg=; b=TMXh0ViSkPI6nrH+HmX5TeTmBTuf5HzJ6FctOIbBLDvyvr9Z/zN5G6d1/7SqwGAzR4uUMzFnDTLgV6TU66AzPAvGv2v8WpGrGzvriBnxA4I8UheJ5gsWqhkOH8ERgUJeUspYhxcYdHaINtGCtviLe4o2jvw9X0nGP2Gvlbf5vUg= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1665092282951401.82065390678054; Thu, 6 Oct 2022 14:38:02 -0700 (PDT) Received: from localhost ([::1]:57812 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1ogYZ7-0005u8-B8 for importer@patchew.org; Thu, 06 Oct 2022 17:38:01 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:53712) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWe-00039H-Sh for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:31 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.133.124]:56320) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWX-0006mP-5w for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:28 -0400 Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-286-e7EdbSK8NmW4VkVoGpc7Sg-1; Thu, 06 Oct 2022 17:35:19 -0400 Received: from smtp.corp.redhat.com (int-mx06.intmail.prod.int.rdu2.redhat.com [10.11.54.6]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id CCE8A811E67; Thu, 6 Oct 2022 21:35:18 +0000 (UTC) Received: from localhost (unknown [10.39.193.119]) by smtp.corp.redhat.com (Postfix) with ESMTP id 039152166B2E; Thu, 6 Oct 2022 21:35:17 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1665092120; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=3mgJt1ynd7ZiWF7e8csGQVMXxZYrxlOZeWtbVF2Vfrg=; b=OeArOu0IJVF075m2p2gbS6lZDS1X/CJocmVP3JoFlDpMWTMqLKyBwNkuw5n6sy+VozSjMV EqhXw1Cjhnn2YuOJqR4IMfUtKq1PN8V+jvhUL2gaWIh8fchQMhe8k1iuPNiJvHXJztrim/ 3IG8hQOpQQNplrw1lLQtvfJ4ge26SLI= X-MC-Unique: e7EdbSK8NmW4VkVoGpc7Sg-1 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Yanan Wang , sgarzare@redhat.com, "Richard W.M. Jones" , Fam Zheng , Hanna Reitz , David Hildenbrand , integration@gluster.org, qemu-block@nongnu.org, Vladimir Sementsov-Ogievskiy , Paolo Bonzini , Kevin Wolf , afaria@redhat.com, Richard Henderson , Eric Blake , "Michael S. Tsirkin" , Thomas Huth , Xie Changlong , John Snow , Eduardo Habkost , Jeff Cody , Stefan Hajnoczi , "Denis V. Lunev" , Markus Armbruster , Laurent Vivier , Peter Xu , Raphael Norwitz , Marcel Apfelbaum , =?UTF-8?q?Philippe=20Mathieu-Daud=C3=A9?= , Wen Congyang Subject: [PATCH v6 03/13] numa: call ->ram_block_removed() in ram_block_notifer_remove() Date: Thu, 6 Oct 2022 17:34:57 -0400 Message-Id: <20221006213507.645402-4-stefanha@redhat.com> In-Reply-To: <20221006213507.645402-1-stefanha@redhat.com> References: <20221006213507.645402-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.6 Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=170.10.133.124; envelope-from=stefanha@redhat.com; helo=us-smtp-delivery-124.mimecast.com 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, 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: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1665092284236100002 Content-Type: text/plain; charset="utf-8" When a RAMBlockNotifier is added, ->ram_block_added() is called with all existing RAMBlocks. There is no equivalent ->ram_block_removed() call when a RAMBlockNotifier is removed. The util/vfio-helpers.c code (the sole user of RAMBlockNotifier) is fine with this asymmetry because it does not rely on RAMBlockNotifier for cleanup. It walks its internal list of DMA mappings and unmaps them by itself. Future users of RAMBlockNotifier may not have an internal data structure that records added RAMBlocks so they will need ->ram_block_removed() callbacks. This patch makes ram_block_notifier_remove() symmetric with respect to callbacks. Now util/vfio-helpers.c needs to unmap remaining DMA mappings after ram_block_notifier_remove() has been called. This is necessary since users like block/nvme.c may create additional DMA mappings that do not originate from the RAMBlockNotifier. Reviewed-by: David Hildenbrand Signed-off-by: Stefan Hajnoczi --- hw/core/numa.c | 17 +++++++++++++++++ util/vfio-helpers.c | 5 ++++- 2 files changed, 21 insertions(+), 1 deletion(-) diff --git a/hw/core/numa.c b/hw/core/numa.c index 26d8e5f616..31e6fe1caa 100644 --- a/hw/core/numa.c +++ b/hw/core/numa.c @@ -822,6 +822,19 @@ static int ram_block_notify_add_single(RAMBlock *rb, v= oid *opaque) return 0; } =20 +static int ram_block_notify_remove_single(RAMBlock *rb, void *opaque) +{ + const ram_addr_t max_size =3D qemu_ram_get_max_length(rb); + const ram_addr_t size =3D qemu_ram_get_used_length(rb); + void *host =3D qemu_ram_get_host_addr(rb); + RAMBlockNotifier *notifier =3D opaque; + + if (host) { + notifier->ram_block_removed(notifier, host, size, max_size); + } + return 0; +} + void ram_block_notifier_add(RAMBlockNotifier *n) { QLIST_INSERT_HEAD(&ram_list.ramblock_notifiers, n, next); @@ -835,6 +848,10 @@ void ram_block_notifier_add(RAMBlockNotifier *n) void ram_block_notifier_remove(RAMBlockNotifier *n) { QLIST_REMOVE(n, next); + + if (n->ram_block_removed) { + qemu_ram_foreach_block(ram_block_notify_remove_single, n); + } } =20 void ram_block_notify_add(void *host, size_t size, size_t max_size) diff --git a/util/vfio-helpers.c b/util/vfio-helpers.c index 5ba01177bf..0d1520caac 100644 --- a/util/vfio-helpers.c +++ b/util/vfio-helpers.c @@ -847,10 +847,13 @@ void qemu_vfio_close(QEMUVFIOState *s) if (!s) { return; } + + ram_block_notifier_remove(&s->ram_notifier); + for (i =3D 0; i < s->nr_mappings; ++i) { qemu_vfio_undo_mapping(s, &s->mappings[i], NULL); } - ram_block_notifier_remove(&s->ram_notifier); + g_free(s->usable_iova_ranges); s->nb_iova_ranges =3D 0; qemu_vfio_reset(s); --=20 2.37.3 From nobody Mon Feb 9 06:26:38 2026 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1665092432; cv=none; d=zohomail.com; s=zohoarc; b=ZdUyw2COp73FDjslTc9d4Ysqc6M9aeQd5l/9I5DQtviKiAGxVUqlukIM0QQBJw1OkkKpw/coLADDoxckKOFrqdNXuScPC+ESgSCx+AGWB7nNzTH290fo0FQQOSDttPec+997FLnsAJ2B5w0Zdkoz2/4OiG58xfeAfMXyX/40LUI= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1665092432; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=UP9I4eVyXrRvYhfhv/f4mIPWi1FXnAnj3msdPoFE7J4=; b=SLCC4qp5eXff/3zc1J6LuhJIDI3DlITtKBW+yuIwgkmA62cSDrrwJq5k1m8TPJI2tiIKuw6KcO91rDRVCRhYCZQrZJmgRa+sE78Mvt64MXKtesmTyhq/V3FJ6aUxWS2piHRk9Hitr+9nAv3il6fKozpwT155SwiZ1jqk6EL8tCU= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1665092432197905.0607864412369; Thu, 6 Oct 2022 14:40:32 -0700 (PDT) Received: from localhost ([::1]:43182 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1ogYbV-00039s-65 for importer@patchew.org; Thu, 06 Oct 2022 17:40:29 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:53718) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWf-00039V-3i for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:31 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.129.124]:26521) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWd-0006mg-3N for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:28 -0400 Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-653-zFk7Cf8mMLmuBiQy46_XtA-1; Thu, 06 Oct 2022 17:35:22 -0400 Received: from smtp.corp.redhat.com (int-mx10.intmail.prod.int.rdu2.redhat.com [10.11.54.10]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 95FC1185A794; Thu, 6 Oct 2022 21:35:21 +0000 (UTC) Received: from localhost (unknown [10.39.193.119]) by smtp.corp.redhat.com (Postfix) with ESMTP id EF7E0535939; Thu, 6 Oct 2022 21:35:20 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1665092126; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=UP9I4eVyXrRvYhfhv/f4mIPWi1FXnAnj3msdPoFE7J4=; b=YG8nDQagc8c4qy2ekscXPQXVwskjU5lWxp/WeMWQLANFxbjdD//OsiPzMhZ5/03dWSR2Es H6y4m5SUoQDgjftLQNcC/41QWFY3K8UMO/xC3dR9PqYvtAPf4oIbLaWfzgNU4dzNj3XJqi ZWMR5HQmh6d0389MZM+n/++P6K9HwYo= X-MC-Unique: zFk7Cf8mMLmuBiQy46_XtA-1 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Yanan Wang , sgarzare@redhat.com, "Richard W.M. Jones" , Fam Zheng , Hanna Reitz , David Hildenbrand , integration@gluster.org, qemu-block@nongnu.org, Vladimir Sementsov-Ogievskiy , Paolo Bonzini , Kevin Wolf , afaria@redhat.com, Richard Henderson , Eric Blake , "Michael S. Tsirkin" , Thomas Huth , Xie Changlong , John Snow , Eduardo Habkost , Jeff Cody , Stefan Hajnoczi , "Denis V. Lunev" , Markus Armbruster , Laurent Vivier , Peter Xu , Raphael Norwitz , Marcel Apfelbaum , =?UTF-8?q?Philippe=20Mathieu-Daud=C3=A9?= , Wen Congyang Subject: [PATCH v6 04/13] block: pass size to bdrv_unregister_buf() Date: Thu, 6 Oct 2022 17:34:58 -0400 Message-Id: <20221006213507.645402-5-stefanha@redhat.com> In-Reply-To: <20221006213507.645402-1-stefanha@redhat.com> References: <20221006213507.645402-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.10 Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=170.10.129.124; envelope-from=stefanha@redhat.com; helo=us-smtp-delivery-124.mimecast.com X-Spam_score_int: -27 X-Spam_score: -2.8 X-Spam_bar: -- X-Spam_report: (-2.8 / 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_LOW=-0.7, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=unavailable autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1665092433329100001 Content-Type: text/plain; charset="utf-8" The only implementor of bdrv_register_buf() is block/nvme.c, where the size is not needed when unregistering a buffer. This is because util/vfio-helpers.c can look up mappings by address. Future block drivers that implement bdrv_register_buf() may not be able to do their job given only the buffer address. Add a size argument to bdrv_unregister_buf(). Also document the assumptions about bdrv_register_buf()/bdrv_unregister_buf() calls. The same values that were given to bdrv_register_buf() must be given to bdrv_unregister_buf(). gcc 11.2.1 emits a spurious warning that img_bench()'s buf_size local variable might be uninitialized, so it's necessary to silence the compiler. Signed-off-by: Stefan Hajnoczi Reviewed-by: Stefano Garzarella --- include/block/block-global-state.h | 5 ++++- include/block/block_int-common.h | 2 +- include/sysemu/block-backend-global-state.h | 2 +- block/block-backend.c | 4 ++-- block/io.c | 6 +++--- block/nvme.c | 2 +- qemu-img.c | 4 ++-- 7 files changed, 14 insertions(+), 11 deletions(-) diff --git a/include/block/block-global-state.h b/include/block/block-globa= l-state.h index 21265e3966..7901f35863 100644 --- a/include/block/block-global-state.h +++ b/include/block/block-global-state.h @@ -243,9 +243,12 @@ void bdrv_del_child(BlockDriverState *parent, BdrvChil= d *child, Error **errp); * Register/unregister a buffer for I/O. For example, VFIO drivers are * interested to know the memory areas that would later be used for I/O, so * that they can prepare IOMMU mapping etc., to get better performance. + * + * Buffers must not overlap and they must be unregistered with the same values that they were registered with. */ void bdrv_register_buf(BlockDriverState *bs, void *host, size_t size); -void bdrv_unregister_buf(BlockDriverState *bs, void *host); +void bdrv_unregister_buf(BlockDriverState *bs, void *host, size_t size); =20 void bdrv_cancel_in_flight(BlockDriverState *bs); =20 diff --git a/include/block/block_int-common.h b/include/block/block_int-com= mon.h index 8947abab76..b7a7cbd3a5 100644 --- a/include/block/block_int-common.h +++ b/include/block/block_int-common.h @@ -435,7 +435,7 @@ struct BlockDriver { * DMA mapping for hot buffers. */ void (*bdrv_register_buf)(BlockDriverState *bs, void *host, size_t siz= e); - void (*bdrv_unregister_buf)(BlockDriverState *bs, void *host); + void (*bdrv_unregister_buf)(BlockDriverState *bs, void *host, size_t s= ize); =20 /* * This field is modified only under the BQL, and is part of diff --git a/include/sysemu/block-backend-global-state.h b/include/sysemu/b= lock-backend-global-state.h index 415f0c91d7..97f7dad2c3 100644 --- a/include/sysemu/block-backend-global-state.h +++ b/include/sysemu/block-backend-global-state.h @@ -107,7 +107,7 @@ void blk_io_limits_update_group(BlockBackend *blk, cons= t char *group); void blk_set_force_allow_inactivate(BlockBackend *blk); =20 void blk_register_buf(BlockBackend *blk, void *host, size_t size); -void blk_unregister_buf(BlockBackend *blk, void *host); +void blk_unregister_buf(BlockBackend *blk, void *host, size_t size); =20 const BdrvChild *blk_root(BlockBackend *blk); =20 diff --git a/block/block-backend.c b/block/block-backend.c index d4a5df2ac2..99141f8f06 100644 --- a/block/block-backend.c +++ b/block/block-backend.c @@ -2551,10 +2551,10 @@ void blk_register_buf(BlockBackend *blk, void *host= , size_t size) bdrv_register_buf(blk_bs(blk), host, size); } =20 -void blk_unregister_buf(BlockBackend *blk, void *host) +void blk_unregister_buf(BlockBackend *blk, void *host, size_t size) { GLOBAL_STATE_CODE(); - bdrv_unregister_buf(blk_bs(blk), host); + bdrv_unregister_buf(blk_bs(blk), host, size); } =20 int coroutine_fn blk_co_copy_range(BlockBackend *blk_in, int64_t off_in, diff --git a/block/io.c b/block/io.c index c3200bcdff..fa34e6c61d 100644 --- a/block/io.c +++ b/block/io.c @@ -3275,16 +3275,16 @@ void bdrv_register_buf(BlockDriverState *bs, void *= host, size_t size) } } =20 -void bdrv_unregister_buf(BlockDriverState *bs, void *host) +void bdrv_unregister_buf(BlockDriverState *bs, void *host, size_t size) { BdrvChild *child; =20 GLOBAL_STATE_CODE(); if (bs->drv && bs->drv->bdrv_unregister_buf) { - bs->drv->bdrv_unregister_buf(bs, host); + bs->drv->bdrv_unregister_buf(bs, host, size); } QLIST_FOREACH(child, &bs->children, next) { - bdrv_unregister_buf(child->bs, host); + bdrv_unregister_buf(child->bs, host, size); } } =20 diff --git a/block/nvme.c b/block/nvme.c index 01fb28aa63..696502acea 100644 --- a/block/nvme.c +++ b/block/nvme.c @@ -1592,7 +1592,7 @@ static void nvme_register_buf(BlockDriverState *bs, v= oid *host, size_t size) } } =20 -static void nvme_unregister_buf(BlockDriverState *bs, void *host) +static void nvme_unregister_buf(BlockDriverState *bs, void *host, size_t s= ize) { BDRVNVMeState *s =3D bs->opaque; =20 diff --git a/qemu-img.c b/qemu-img.c index cab9776f42..1dd5efe840 100644 --- a/qemu-img.c +++ b/qemu-img.c @@ -4368,7 +4368,7 @@ static int img_bench(int argc, char **argv) struct timeval t1, t2; int i; bool force_share =3D false; - size_t buf_size; + size_t buf_size =3D 0; =20 for (;;) { static const struct option long_options[] =3D { @@ -4590,7 +4590,7 @@ static int img_bench(int argc, char **argv) =20 out: if (data.buf) { - blk_unregister_buf(blk, data.buf); + blk_unregister_buf(blk, data.buf, buf_size); } qemu_vfree(data.buf); blk_unref(blk); --=20 2.37.3 From nobody Mon Feb 9 06:26:38 2026 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1665092461; cv=none; d=zohomail.com; s=zohoarc; b=IyoVkFb9YAT8Trj7/NbBGCCeqfUVz95MuHYYdMDF2XyCt7uorUgmlBEyhWr/VBN8+hrUHr4eCcdJ6+7Yu5VvvJNyXsDPUs5bjMyPajEcZZSSR5lSNL0FFvh6wwcOpGbh9Nb/lgh63ruzGxeJGuDD9OY38VUdRekHwOzNIwzCh3o= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1665092461; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=hr4Rs7hpdNS7b3o4WidAG60QG2JI1UtqbqZ1ZxluRD8=; b=KA8ncFiKddp8mgWJ+wgMGgXn5Lp7ff/0rhLBahvZ9+dfUq8K7CDMY/JUJOGYEpKZvKgiRlyYj6fOJQxwfOfAu7Cj8UVUzTeX3I2+F3EKBcO3oG55FHufmDs1BI49OM4AIoiB9ucKHt1WCCA/i+vCH7UPR74OBgoZ1Xn5tkWEaE0= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1665092461321463.78810108900757; Thu, 6 Oct 2022 14:41:01 -0700 (PDT) Received: from localhost ([::1]:41276 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1ogYc0-0003e4-8U for importer@patchew.org; Thu, 06 Oct 2022 17:41:00 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:53720) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWg-0003B9-DB for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:31 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.133.124]:36963) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWf-0006oD-0s for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:30 -0400 Received: from mimecast-mx02.redhat.com (mx3-rdu2.redhat.com [66.187.233.73]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-134-semXQyuDO5mMYd05rD517w-1; Thu, 06 Oct 2022 17:35:24 -0400 Received: from smtp.corp.redhat.com (int-mx04.intmail.prod.int.rdu2.redhat.com [10.11.54.4]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 405C83826A41; Thu, 6 Oct 2022 21:35:24 +0000 (UTC) Received: from localhost (unknown [10.39.193.119]) by smtp.corp.redhat.com (Postfix) with ESMTP id 69322207AD4C; Thu, 6 Oct 2022 21:35:23 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1665092128; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=hr4Rs7hpdNS7b3o4WidAG60QG2JI1UtqbqZ1ZxluRD8=; b=P8xljHvdgErG9TLW4MiMIJ4Wm6MnFhsbLm+PgLH7YE0KUC4b4/CgKjwZV/zoWpxv9kvCI1 rn5Gi7A/xJh7rvpzHwG9gTN5hLne6j//Z/j6GnQp2abT7xT0qOgCwbFaKSZGLwjljHjhUe WuGObdQr/bwG8i53hrxCYj+CGic7Z3Q= X-MC-Unique: semXQyuDO5mMYd05rD517w-1 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Yanan Wang , sgarzare@redhat.com, "Richard W.M. Jones" , Fam Zheng , Hanna Reitz , David Hildenbrand , integration@gluster.org, qemu-block@nongnu.org, Vladimir Sementsov-Ogievskiy , Paolo Bonzini , Kevin Wolf , afaria@redhat.com, Richard Henderson , Eric Blake , "Michael S. Tsirkin" , Thomas Huth , Xie Changlong , John Snow , Eduardo Habkost , Jeff Cody , Stefan Hajnoczi , "Denis V. Lunev" , Markus Armbruster , Laurent Vivier , Peter Xu , Raphael Norwitz , Marcel Apfelbaum , =?UTF-8?q?Philippe=20Mathieu-Daud=C3=A9?= , Wen Congyang Subject: [PATCH v6 05/13] block: use BdrvRequestFlags type for supported flag fields Date: Thu, 6 Oct 2022 17:34:59 -0400 Message-Id: <20221006213507.645402-6-stefanha@redhat.com> In-Reply-To: <20221006213507.645402-1-stefanha@redhat.com> References: <20221006213507.645402-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.4 Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=170.10.133.124; envelope-from=stefanha@redhat.com; helo=us-smtp-delivery-124.mimecast.com 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, 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: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1665092463466100001 Content-Type: text/plain; charset="utf-8" Use the enum type so GDB displays the enum members instead of printing a numeric constant. Signed-off-by: Stefan Hajnoczi Reviewed-by: Stefano Garzarella --- include/block/block_int-common.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/include/block/block_int-common.h b/include/block/block_int-com= mon.h index b7a7cbd3a5..19798d0e77 100644 --- a/include/block/block_int-common.h +++ b/include/block/block_int-common.h @@ -1051,7 +1051,7 @@ struct BlockDriverState { /* * Flags honored during pread */ - unsigned int supported_read_flags; + BdrvRequestFlags supported_read_flags; /* * Flags honored during pwrite (so far: BDRV_REQ_FUA, * BDRV_REQ_WRITE_UNCHANGED). @@ -1069,12 +1069,12 @@ struct BlockDriverState { * flag), or they have to explicitly take the WRITE permission for * their children. */ - unsigned int supported_write_flags; + BdrvRequestFlags supported_write_flags; /* * Flags honored during pwrite_zeroes (so far: BDRV_REQ_FUA, * BDRV_REQ_MAY_UNMAP, BDRV_REQ_WRITE_UNCHANGED) */ - unsigned int supported_zero_flags; + BdrvRequestFlags supported_zero_flags; /* * Flags honoured during truncate (so far: BDRV_REQ_ZERO_WRITE). * @@ -1082,7 +1082,7 @@ struct BlockDriverState { * that any added space reads as all zeros. If this can't be guarantee= d, * the operation must fail. */ - unsigned int supported_truncate_flags; + BdrvRequestFlags supported_truncate_flags; =20 /* the following member gives a name to every node on the bs graph. */ char node_name[32]; --=20 2.37.3 From nobody Mon Feb 9 06:26:38 2026 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1665092423; cv=none; d=zohomail.com; s=zohoarc; b=DyEvgvjCgvAxA5cvlU4oftLdB6QYlnb8hbW21KyFWpMA4qRF8TdSUfCB1Lobsv+7xWwVa8BCUt5rf0V1EiT34hpoXupT3qA4MsmunUOWN53N+7wfSgy3dKpnmOy0Wzom/cptaN0oNI7KmL4fkqzmxv9Rou2jvwXwa92G/IclOeA= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1665092423; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=XWJEAddlsW6NvFzA/02L7xBOr8LNdfkCsyt91eMdQ+s=; b=JHZwIGYbXnOZr8CvrAFJ8+uXXbKwxttjZLvnIEah8p62/5dJJv0E8mH3/m+kTSkREfCSSDsbw15Qb6GTGHxqR0I8tftsantoPKEx8FW8X9Ocs/ICpwI/Pa4fEJTSTTHZ/s2FXUZ6VJsQdEfCeg90URipsjHsVBQZnyIZ09AqpZo= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1665092423249556.572960995253; Thu, 6 Oct 2022 14:40:23 -0700 (PDT) Received: from localhost ([::1]:43064 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1ogYbO-0002sk-6h for importer@patchew.org; Thu, 06 Oct 2022 17:40:22 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:39014) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWn-0003HN-Jv for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:41 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.129.124]:52511) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWj-0006pF-FP for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:37 -0400 Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-635-rzp-n09DMwyeKjLwtujL5A-1; Thu, 06 Oct 2022 17:35:29 -0400 Received: from smtp.corp.redhat.com (int-mx10.intmail.prod.int.rdu2.redhat.com [10.11.54.10]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id E5FB6185A7A3; Thu, 6 Oct 2022 21:35:26 +0000 (UTC) Received: from localhost (unknown [10.39.193.119]) by smtp.corp.redhat.com (Postfix) with ESMTP id 18334535939; Thu, 6 Oct 2022 21:35:25 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1665092132; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=XWJEAddlsW6NvFzA/02L7xBOr8LNdfkCsyt91eMdQ+s=; b=GQjkd3RRQXS5WxkjQLs7kDjKUi0K4mTmSFCGlom3Pv/sQ3EPssvj7J6K4ecgnjO5BQUKpM MpvUAw+1BTek7P73YjyrugUC6G0EXzqS5b0BxgR6QiWWMKKSPySc4BtPqsRoBZd9Dk1aWa JrkneEu3OiGze5l8/Ka2RLOX/WTno7o= X-MC-Unique: rzp-n09DMwyeKjLwtujL5A-1 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Yanan Wang , sgarzare@redhat.com, "Richard W.M. Jones" , Fam Zheng , Hanna Reitz , David Hildenbrand , integration@gluster.org, qemu-block@nongnu.org, Vladimir Sementsov-Ogievskiy , Paolo Bonzini , Kevin Wolf , afaria@redhat.com, Richard Henderson , Eric Blake , "Michael S. Tsirkin" , Thomas Huth , Xie Changlong , John Snow , Eduardo Habkost , Jeff Cody , Stefan Hajnoczi , "Denis V. Lunev" , Markus Armbruster , Laurent Vivier , Peter Xu , Raphael Norwitz , Marcel Apfelbaum , =?UTF-8?q?Philippe=20Mathieu-Daud=C3=A9?= , Wen Congyang Subject: [PATCH v6 06/13] block: add BDRV_REQ_REGISTERED_BUF request flag Date: Thu, 6 Oct 2022 17:35:00 -0400 Message-Id: <20221006213507.645402-7-stefanha@redhat.com> In-Reply-To: <20221006213507.645402-1-stefanha@redhat.com> References: <20221006213507.645402-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.10 Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=170.10.129.124; envelope-from=stefanha@redhat.com; helo=us-smtp-delivery-124.mimecast.com X-Spam_score_int: -27 X-Spam_score: -2.8 X-Spam_bar: -- X-Spam_report: (-2.8 / 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_LOW=-0.7, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=unavailable autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1665092425558100001 Content-Type: text/plain; charset="utf-8" Block drivers may optimize I/O requests accessing buffers previously registered with bdrv_register_buf(). Checking whether all elements of a request's QEMUIOVector are within previously registered buffers is expensive, so we need a hint from the user to avoid costly checks. Add a BDRV_REQ_REGISTERED_BUF request flag to indicate that all QEMUIOVector elements in an I/O request are known to be within previously registered buffers. Always pass the flag through to driver read/write functions. There is little harm in passing the flag to a driver that does not use it. Passing the flag to drivers avoids changes across many block drivers. Filter drivers would need to explicitly support the flag and pass through to their children when the children support it. That's a lot of code changes and it's hard to remember to do that everywhere, leading to silent reduced performance when the flag is accidentally dropped. The only problematic scenario with the approach in this patch is when a driver passes the flag through to internal I/O requests that don't use the same I/O buffer. In that case the hint may be set when it should actually be clear. This is a rare case though so the risk is low. Some drivers have assert(!flags), which no longer works when BDRV_REQ_REGISTERED_BUF is passed in. These assertions aren't very useful anyway since the functions are called almost exclusively by bdrv_driver_preadv/pwritev() so if we get flags handling right there then the assertion is not needed. Signed-off-by: Stefan Hajnoczi --- include/block/block-common.h | 9 ++++++ block.c | 14 +++++++++ block/blkverify.c | 4 +-- block/crypto.c | 4 +-- block/file-posix.c | 1 - block/gluster.c | 1 - block/io.c | 61 ++++++++++++++++++++++-------------- block/mirror.c | 2 ++ block/nbd.c | 1 - block/parallels.c | 1 - block/qcow.c | 2 -- block/qed.c | 1 - block/raw-format.c | 2 ++ block/replication.c | 1 - block/ssh.c | 1 - block/vhdx.c | 1 - 16 files changed, 69 insertions(+), 37 deletions(-) diff --git a/include/block/block-common.h b/include/block/block-common.h index fdb7306e78..061606e867 100644 --- a/include/block/block-common.h +++ b/include/block/block-common.h @@ -80,6 +80,15 @@ typedef enum { */ BDRV_REQ_MAY_UNMAP =3D 0x4, =20 + /* + * An optimization hint when all QEMUIOVector elements are within + * previously registered bdrv_register_buf() memory ranges. + * + * Code that replaces the user's QEMUIOVector elements with bounce buf= fers + * must take care to clear this flag. + */ + BDRV_REQ_REGISTERED_BUF =3D 0x8, + BDRV_REQ_FUA =3D 0x10, BDRV_REQ_WRITE_COMPRESSED =3D 0x20, =20 diff --git a/block.c b/block.c index bc85f46eed..70abbf774e 100644 --- a/block.c +++ b/block.c @@ -1640,6 +1640,20 @@ static int bdrv_open_driver(BlockDriverState *bs, Bl= ockDriver *drv, goto open_failed; } =20 + assert(!(bs->supported_read_flags & ~BDRV_REQ_MASK)); + assert(!(bs->supported_write_flags & ~BDRV_REQ_MASK)); + + /* + * Always allow the BDRV_REQ_REGISTERED_BUF optimization hint. This sa= ves + * drivers that pass read/write requests through to a child the troubl= e of + * declaring support explicitly. + * + * Drivers must not propagate this flag accidentally when they initiat= e I/O + * to a bounce buffer. That case should be rare though. + */ + bs->supported_read_flags |=3D BDRV_REQ_REGISTERED_BUF; + bs->supported_write_flags |=3D BDRV_REQ_REGISTERED_BUF; + ret =3D refresh_total_sectors(bs, bs->total_sectors); if (ret < 0) { error_setg_errno(errp, -ret, "Could not refresh total sector count= "); diff --git a/block/blkverify.c b/block/blkverify.c index e4a37af3b2..d624f4fd05 100644 --- a/block/blkverify.c +++ b/block/blkverify.c @@ -235,8 +235,8 @@ blkverify_co_preadv(BlockDriverState *bs, int64_t offse= t, int64_t bytes, qemu_iovec_init(&raw_qiov, qiov->niov); qemu_iovec_clone(&raw_qiov, qiov, buf); =20 - ret =3D blkverify_co_prwv(bs, &r, offset, bytes, qiov, &raw_qiov, flag= s, - false); + ret =3D blkverify_co_prwv(bs, &r, offset, bytes, qiov, &raw_qiov, + flags & ~BDRV_REQ_REGISTERED_BUF, false); =20 cmp_offset =3D qemu_iovec_compare(qiov, &raw_qiov); if (cmp_offset !=3D -1) { diff --git a/block/crypto.c b/block/crypto.c index 7a57774b76..c7365598a7 100644 --- a/block/crypto.c +++ b/block/crypto.c @@ -410,7 +410,6 @@ block_crypto_co_preadv(BlockDriverState *bs, int64_t of= fset, int64_t bytes, uint64_t sector_size =3D qcrypto_block_get_sector_size(crypto->block); uint64_t payload_offset =3D qcrypto_block_get_payload_offset(crypto->b= lock); =20 - assert(!flags); assert(payload_offset < INT64_MAX); assert(QEMU_IS_ALIGNED(offset, sector_size)); assert(QEMU_IS_ALIGNED(bytes, sector_size)); @@ -473,7 +472,8 @@ block_crypto_co_pwritev(BlockDriverState *bs, int64_t o= ffset, int64_t bytes, uint64_t sector_size =3D qcrypto_block_get_sector_size(crypto->block); uint64_t payload_offset =3D qcrypto_block_get_payload_offset(crypto->b= lock); =20 - assert(!(flags & ~BDRV_REQ_FUA)); + flags &=3D ~BDRV_REQ_REGISTERED_BUF; + assert(payload_offset < INT64_MAX); assert(QEMU_IS_ALIGNED(offset, sector_size)); assert(QEMU_IS_ALIGNED(bytes, sector_size)); diff --git a/block/file-posix.c b/block/file-posix.c index 66fdb07820..f2a960143d 100644 --- a/block/file-posix.c +++ b/block/file-posix.c @@ -2142,7 +2142,6 @@ static int coroutine_fn raw_co_pwritev(BlockDriverSta= te *bs, int64_t offset, int64_t bytes, QEMUIOVector *qiov, BdrvRequestFlags flags) { - assert(flags =3D=3D 0); return raw_co_prw(bs, offset, bytes, qiov, QEMU_AIO_WRITE); } =20 diff --git a/block/gluster.c b/block/gluster.c index bb1144cf6a..7c90f7ba4b 100644 --- a/block/gluster.c +++ b/block/gluster.c @@ -1236,7 +1236,6 @@ static coroutine_fn int qemu_gluster_co_writev(BlockD= riverState *bs, QEMUIOVector *qiov, int flags) { - assert(!flags); return qemu_gluster_co_rw(bs, sector_num, nb_sectors, qiov, 1); } =20 diff --git a/block/io.c b/block/io.c index fa34e6c61d..1d7220a68f 100644 --- a/block/io.c +++ b/block/io.c @@ -1130,8 +1130,7 @@ static int coroutine_fn bdrv_driver_preadv(BlockDrive= rState *bs, int ret; =20 bdrv_check_qiov_request(offset, bytes, qiov, qiov_offset, &error_abort= ); - assert(!(flags & ~BDRV_REQ_MASK)); - assert(!(flags & BDRV_REQ_NO_FALLBACK)); + assert(!(flags & ~bs->supported_read_flags)); =20 if (!drv) { return -ENOMEDIUM; @@ -1195,23 +1194,29 @@ static int coroutine_fn bdrv_driver_pwritev(BlockDr= iverState *bs, BdrvRequestFlags flags) { BlockDriver *drv =3D bs->drv; + bool emulate_fua =3D false; int64_t sector_num; unsigned int nb_sectors; QEMUIOVector local_qiov; int ret; =20 bdrv_check_qiov_request(offset, bytes, qiov, qiov_offset, &error_abort= ); - assert(!(flags & ~BDRV_REQ_MASK)); - assert(!(flags & BDRV_REQ_NO_FALLBACK)); =20 if (!drv) { return -ENOMEDIUM; } =20 + if ((flags & BDRV_REQ_FUA) && + (~bs->supported_write_flags & BDRV_REQ_FUA)) { + flags &=3D ~BDRV_REQ_FUA; + emulate_fua =3D true; + } + + flags &=3D bs->supported_write_flags; + if (drv->bdrv_co_pwritev_part) { ret =3D drv->bdrv_co_pwritev_part(bs, offset, bytes, qiov, qiov_of= fset, - flags & bs->supported_write_flags); - flags &=3D ~bs->supported_write_flags; + flags); goto emulate_flags; } =20 @@ -1221,9 +1226,7 @@ static int coroutine_fn bdrv_driver_pwritev(BlockDriv= erState *bs, } =20 if (drv->bdrv_co_pwritev) { - ret =3D drv->bdrv_co_pwritev(bs, offset, bytes, qiov, - flags & bs->supported_write_flags); - flags &=3D ~bs->supported_write_flags; + ret =3D drv->bdrv_co_pwritev(bs, offset, bytes, qiov, flags); goto emulate_flags; } =20 @@ -1233,10 +1236,8 @@ static int coroutine_fn bdrv_driver_pwritev(BlockDri= verState *bs, .coroutine =3D qemu_coroutine_self(), }; =20 - acb =3D drv->bdrv_aio_pwritev(bs, offset, bytes, qiov, - flags & bs->supported_write_flags, + acb =3D drv->bdrv_aio_pwritev(bs, offset, bytes, qiov, flags, bdrv_co_io_em_complete, &co); - flags &=3D ~bs->supported_write_flags; if (acb =3D=3D NULL) { ret =3D -EIO; } else { @@ -1254,12 +1255,10 @@ static int coroutine_fn bdrv_driver_pwritev(BlockDr= iverState *bs, assert(bytes <=3D BDRV_REQUEST_MAX_BYTES); =20 assert(drv->bdrv_co_writev); - ret =3D drv->bdrv_co_writev(bs, sector_num, nb_sectors, qiov, - flags & bs->supported_write_flags); - flags &=3D ~bs->supported_write_flags; + ret =3D drv->bdrv_co_writev(bs, sector_num, nb_sectors, qiov, flags); =20 emulate_flags: - if (ret =3D=3D 0 && (flags & BDRV_REQ_FUA)) { + if (ret =3D=3D 0 && emulate_fua) { ret =3D bdrv_co_flush(bs); } =20 @@ -1487,11 +1486,14 @@ static int coroutine_fn bdrv_aligned_preadv(BdrvChi= ld *child, max_transfer =3D QEMU_ALIGN_DOWN(MIN_NON_ZERO(bs->bl.max_transfer, INT= _MAX), align); =20 - /* TODO: We would need a per-BDS .supported_read_flags and + /* + * TODO: We would need a per-BDS .supported_read_flags and * potential fallback support, if we ever implement any read flags * to pass through to drivers. For now, there aren't any - * passthrough flags. */ - assert(!(flags & ~(BDRV_REQ_COPY_ON_READ | BDRV_REQ_PREFETCH))); + * passthrough flags except the BDRV_REQ_REGISTERED_BUF optimization h= int. + */ + assert(!(flags & ~(BDRV_REQ_COPY_ON_READ | BDRV_REQ_PREFETCH | + BDRV_REQ_REGISTERED_BUF))); =20 /* Handle Copy on Read and associated serialisation */ if (flags & BDRV_REQ_COPY_ON_READ) { @@ -1532,7 +1534,7 @@ static int coroutine_fn bdrv_aligned_preadv(BdrvChild= *child, goto out; } =20 - assert(!(flags & ~bs->supported_read_flags)); + assert(!(flags & ~(bs->supported_read_flags | BDRV_REQ_REGISTERED_BUF)= )); =20 max_bytes =3D ROUND_UP(MAX(0, total_bytes - offset), align); if (bytes <=3D max_bytes && bytes <=3D max_transfer) { @@ -1721,7 +1723,8 @@ static void bdrv_padding_destroy(BdrvRequestPadding *= pad) static int bdrv_pad_request(BlockDriverState *bs, QEMUIOVector **qiov, size_t *qiov_offset, int64_t *offset, int64_t *bytes, - BdrvRequestPadding *pad, bool *padded) + BdrvRequestPadding *pad, bool *padded, + BdrvRequestFlags *flags) { int ret; =20 @@ -1749,6 +1752,10 @@ static int bdrv_pad_request(BlockDriverState *bs, if (padded) { *padded =3D true; } + if (flags) { + /* Can't use optimization hint with bounce buffer */ + *flags &=3D ~BDRV_REQ_REGISTERED_BUF; + } =20 return 0; } @@ -1803,7 +1810,7 @@ int coroutine_fn bdrv_co_preadv_part(BdrvChild *child, } =20 ret =3D bdrv_pad_request(bs, &qiov, &qiov_offset, &offset, &bytes, &pa= d, - NULL); + NULL, &flags); if (ret < 0) { goto fail; } @@ -1848,6 +1855,11 @@ static int coroutine_fn bdrv_co_do_pwrite_zeroes(Blo= ckDriverState *bs, return -ENOTSUP; } =20 + /* By definition there is no user buffer so this flag doesn't make sen= se */ + if (flags & BDRV_REQ_REGISTERED_BUF) { + return -EINVAL; + } + /* Invalidate the cached block-status data range if this write overlap= s */ bdrv_bsc_invalidate_range(bs, offset, bytes); =20 @@ -2133,6 +2145,9 @@ static int coroutine_fn bdrv_co_do_zero_pwritev(BdrvC= hild *child, bool padding; BdrvRequestPadding pad; =20 + /* This flag doesn't make sense for padding or zero writes */ + flags &=3D ~BDRV_REQ_REGISTERED_BUF; + padding =3D bdrv_init_padding(bs, offset, bytes, &pad); if (padding) { assert(!(flags & BDRV_REQ_NO_WAIT)); @@ -2250,7 +2265,7 @@ int coroutine_fn bdrv_co_pwritev_part(BdrvChild *chil= d, * alignment only if there is no ZERO flag. */ ret =3D bdrv_pad_request(bs, &qiov, &qiov_offset, &offset, &bytes,= &pad, - &padded); + &padded, &flags); if (ret < 0) { return ret; } diff --git a/block/mirror.c b/block/mirror.c index 3c4ab1159d..8d3fc3f19b 100644 --- a/block/mirror.c +++ b/block/mirror.c @@ -1477,6 +1477,8 @@ static int coroutine_fn bdrv_mirror_top_pwritev(Block= DriverState *bs, qemu_iovec_init(&bounce_qiov, 1); qemu_iovec_add(&bounce_qiov, bounce_buf, bytes); qiov =3D &bounce_qiov; + + flags &=3D ~BDRV_REQ_REGISTERED_BUF; } =20 ret =3D bdrv_mirror_top_do_write(bs, MIRROR_METHOD_COPY, offset, bytes= , qiov, diff --git a/block/nbd.c b/block/nbd.c index 97683cce27..ce2956e083 100644 --- a/block/nbd.c +++ b/block/nbd.c @@ -1221,7 +1221,6 @@ static int coroutine_fn nbd_client_co_preadv(BlockDri= verState *bs, int64_t offse }; =20 assert(bytes <=3D NBD_MAX_BUFFER_SIZE); - assert(!flags); =20 if (!bytes) { return 0; diff --git a/block/parallels.c b/block/parallels.c index a229c06f25..7c04c6a135 100644 --- a/block/parallels.c +++ b/block/parallels.c @@ -328,7 +328,6 @@ static coroutine_fn int parallels_co_writev(BlockDriver= State *bs, QEMUIOVector hd_qiov; int ret =3D 0; =20 - assert(!flags); qemu_iovec_init(&hd_qiov, qiov->niov); =20 while (nb_sectors > 0) { diff --git a/block/qcow.c b/block/qcow.c index 311aaa8705..e9180c7b61 100644 --- a/block/qcow.c +++ b/block/qcow.c @@ -628,7 +628,6 @@ static coroutine_fn int qcow_co_preadv(BlockDriverState= *bs, int64_t offset, uint8_t *buf; void *orig_buf; =20 - assert(!flags); if (qiov->niov > 1) { buf =3D orig_buf =3D qemu_try_blockalign(bs, qiov->size); if (buf =3D=3D NULL) { @@ -725,7 +724,6 @@ static coroutine_fn int qcow_co_pwritev(BlockDriverStat= e *bs, int64_t offset, uint8_t *buf; void *orig_buf; =20 - assert(!flags); s->cluster_cache_offset =3D -1; /* disable compressed cache */ =20 /* We must always copy the iov when encrypting, so we diff --git a/block/qed.c b/block/qed.c index 324ca0e95a..254c707252 100644 --- a/block/qed.c +++ b/block/qed.c @@ -1395,7 +1395,6 @@ static int coroutine_fn bdrv_qed_co_writev(BlockDrive= rState *bs, int64_t sector_num, int nb_sect= ors, QEMUIOVector *qiov, int flags) { - assert(!flags); return qed_co_request(bs, sector_num, qiov, nb_sectors, QED_AIOCB_WRIT= E); } =20 diff --git a/block/raw-format.c b/block/raw-format.c index c7278e348e..34fe7f0023 100644 --- a/block/raw-format.c +++ b/block/raw-format.c @@ -258,6 +258,8 @@ static int coroutine_fn raw_co_pwritev(BlockDriverState= *bs, int64_t offset, qemu_iovec_add(&local_qiov, buf, 512); qemu_iovec_concat(&local_qiov, qiov, 512, qiov->size - 512); qiov =3D &local_qiov; + + flags &=3D ~BDRV_REQ_REGISTERED_BUF; } =20 ret =3D raw_adjust_offset(bs, &offset, bytes, true); diff --git a/block/replication.c b/block/replication.c index 55c8f894aa..942f88b205 100644 --- a/block/replication.c +++ b/block/replication.c @@ -260,7 +260,6 @@ static coroutine_fn int replication_co_writev(BlockDriv= erState *bs, int ret; int64_t n; =20 - assert(!flags); ret =3D replication_get_io_status(s); if (ret < 0) { goto out; diff --git a/block/ssh.c b/block/ssh.c index a2dc646536..a3cddc392c 100644 --- a/block/ssh.c +++ b/block/ssh.c @@ -1196,7 +1196,6 @@ static coroutine_fn int ssh_co_writev(BlockDriverStat= e *bs, BDRVSSHState *s =3D bs->opaque; int ret; =20 - assert(!flags); qemu_co_mutex_lock(&s->lock); ret =3D ssh_write(s, bs, sector_num * BDRV_SECTOR_SIZE, nb_sectors * BDRV_SECTOR_SIZE, qiov); diff --git a/block/vhdx.c b/block/vhdx.c index e10e78ebfd..e2344ee0b7 100644 --- a/block/vhdx.c +++ b/block/vhdx.c @@ -1342,7 +1342,6 @@ static coroutine_fn int vhdx_co_writev(BlockDriverSta= te *bs, int64_t sector_num, uint64_t bat_prior_offset =3D 0; bool bat_update =3D false; =20 - assert(!flags); qemu_iovec_init(&hd_qiov, qiov->niov); =20 qemu_co_mutex_lock(&s->lock); --=20 2.37.3 From nobody Mon Feb 9 06:26:38 2026 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1665092616; cv=none; d=zohomail.com; s=zohoarc; b=TnL1qEeo+GVPlAJ5apjM9Md2jA7dQFrmySruRGMPgFQHW8W506DOREnVq8Y9/HQ01+qJV7DlDe2QVj9VhvjVL0ndthQRvNtWKOUkILTZkq4O6UcnjGFHmQzkBPNXm8TEG+kU0m7fqZPTiQ2JX6CZ7HzUg/WS7cNi1N1Rl/GDQ0s= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1665092616; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=9Yqa58k2WOoeJ6eDR3PCO4tmEWspVj546QT1YwQnLuw=; b=KH0tUEXWl+duRG7FG+xWpoadLYHpCP/M1WoQk3xMN4pMX6X4aGCy70kUBxREY79DHSQNta+0o5V1OmAGMzWE8g45C0y4c1oupU7sqLtz1LrZ7uMt0zBpfr2YExnUEy10/NFfyHxxsp8u/StblSUCVQHQmH7a6CCmVfeYLw7sWTg= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 16650926161541016.9852865807395; Thu, 6 Oct 2022 14:43:36 -0700 (PDT) Received: from localhost ([::1]:58944 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1ogYeU-0007Ki-H7 for importer@patchew.org; Thu, 06 Oct 2022 17:43:34 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:39020) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWp-0003HR-9x for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:41 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.133.124]:56557) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWn-0006pY-Ba for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:38 -0400 Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-396-z7qyZay6NCutJcXsULl5Fg-1; Thu, 06 Oct 2022 17:35:30 -0400 Received: from smtp.corp.redhat.com (int-mx05.intmail.prod.int.rdu2.redhat.com [10.11.54.5]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 941C785D192; Thu, 6 Oct 2022 21:35:29 +0000 (UTC) Received: from localhost (unknown [10.39.193.119]) by smtp.corp.redhat.com (Postfix) with ESMTP id BA8D99D493; Thu, 6 Oct 2022 21:35:28 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1665092136; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=9Yqa58k2WOoeJ6eDR3PCO4tmEWspVj546QT1YwQnLuw=; b=Z9d4W7WPjZWNsp5xfZc1F9QfdxMhhkdsGnLfdtwiqt08i2vsQrlaeDUoiWjghgLIoKTOnJ XD1EfKMDyw6iA+26EKQl8mOaXxhyZrUK8lLfJgFxdSTplheZ4v+FpoXXnJz1dgaCX3RpB5 R8c27SwsDlGrFH0yubduvwAVTWiuc3U= X-MC-Unique: z7qyZay6NCutJcXsULl5Fg-1 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Yanan Wang , sgarzare@redhat.com, "Richard W.M. Jones" , Fam Zheng , Hanna Reitz , David Hildenbrand , integration@gluster.org, qemu-block@nongnu.org, Vladimir Sementsov-Ogievskiy , Paolo Bonzini , Kevin Wolf , afaria@redhat.com, Richard Henderson , Eric Blake , "Michael S. Tsirkin" , Thomas Huth , Xie Changlong , John Snow , Eduardo Habkost , Jeff Cody , Stefan Hajnoczi , "Denis V. Lunev" , Markus Armbruster , Laurent Vivier , Peter Xu , Raphael Norwitz , Marcel Apfelbaum , =?UTF-8?q?Philippe=20Mathieu-Daud=C3=A9?= , Wen Congyang Subject: [PATCH v6 07/13] block: return errors from bdrv_register_buf() Date: Thu, 6 Oct 2022 17:35:01 -0400 Message-Id: <20221006213507.645402-8-stefanha@redhat.com> In-Reply-To: <20221006213507.645402-1-stefanha@redhat.com> References: <20221006213507.645402-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.5 Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=170.10.133.124; envelope-from=stefanha@redhat.com; helo=us-smtp-delivery-124.mimecast.com 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, 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: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1665092617107100001 Content-Type: text/plain; charset="utf-8" Registering an I/O buffer is only a performance optimization hint but it is still necessary to return errors when it fails. Later patches will need to detect errors when registering buffers but an immediate advantage is that error_report() calls are no longer needed in block driver .bdrv_register_buf() functions. Signed-off-by: Stefan Hajnoczi --- include/block/block-global-state.h | 5 ++- include/block/block_int-common.h | 5 ++- include/sysemu/block-backend-global-state.h | 2 +- block/block-backend.c | 4 +-- block/io.c | 34 +++++++++++++++++++-- block/nvme.c | 18 +++++------ qemu-img.c | 2 +- 7 files changed, 52 insertions(+), 18 deletions(-) diff --git a/include/block/block-global-state.h b/include/block/block-globa= l-state.h index 7901f35863..eba4ed23b4 100644 --- a/include/block/block-global-state.h +++ b/include/block/block-global-state.h @@ -246,8 +246,11 @@ void bdrv_del_child(BlockDriverState *parent, BdrvChil= d *child, Error **errp); * * Buffers must not overlap and they must be unregistered with the same values that they were registered with. + * + * Returns: true on success, false on failure */ -void bdrv_register_buf(BlockDriverState *bs, void *host, size_t size); +bool bdrv_register_buf(BlockDriverState *bs, void *host, size_t size, + Error **errp); void bdrv_unregister_buf(BlockDriverState *bs, void *host, size_t size); =20 void bdrv_cancel_in_flight(BlockDriverState *bs); diff --git a/include/block/block_int-common.h b/include/block/block_int-com= mon.h index 19798d0e77..9c569be162 100644 --- a/include/block/block_int-common.h +++ b/include/block/block_int-common.h @@ -433,8 +433,11 @@ struct BlockDriver { * that it can do IOMMU mapping with VFIO etc., in order to get better * performance. In the case of VFIO drivers, this callback is used to = do * DMA mapping for hot buffers. + * + * Returns: true on success, false on failure */ - void (*bdrv_register_buf)(BlockDriverState *bs, void *host, size_t siz= e); + bool (*bdrv_register_buf)(BlockDriverState *bs, void *host, size_t siz= e, + Error **errp); void (*bdrv_unregister_buf)(BlockDriverState *bs, void *host, size_t s= ize); =20 /* diff --git a/include/sysemu/block-backend-global-state.h b/include/sysemu/b= lock-backend-global-state.h index 97f7dad2c3..6858e39cb6 100644 --- a/include/sysemu/block-backend-global-state.h +++ b/include/sysemu/block-backend-global-state.h @@ -106,7 +106,7 @@ void blk_io_limits_enable(BlockBackend *blk, const char= *group); void blk_io_limits_update_group(BlockBackend *blk, const char *group); void blk_set_force_allow_inactivate(BlockBackend *blk); =20 -void blk_register_buf(BlockBackend *blk, void *host, size_t size); +bool blk_register_buf(BlockBackend *blk, void *host, size_t size, Error **= errp); void blk_unregister_buf(BlockBackend *blk, void *host, size_t size); =20 const BdrvChild *blk_root(BlockBackend *blk); diff --git a/block/block-backend.c b/block/block-backend.c index 99141f8f06..34399d3b7b 100644 --- a/block/block-backend.c +++ b/block/block-backend.c @@ -2545,10 +2545,10 @@ static void blk_root_drained_end(BdrvChild *child, = int *drained_end_counter) } } =20 -void blk_register_buf(BlockBackend *blk, void *host, size_t size) +bool blk_register_buf(BlockBackend *blk, void *host, size_t size, Error **= errp) { GLOBAL_STATE_CODE(); - bdrv_register_buf(blk_bs(blk), host, size); + return bdrv_register_buf(blk_bs(blk), host, size, errp); } =20 void blk_unregister_buf(BlockBackend *blk, void *host, size_t size) diff --git a/block/io.c b/block/io.c index 1d7220a68f..fe25ef6e82 100644 --- a/block/io.c +++ b/block/io.c @@ -3277,17 +3277,45 @@ void bdrv_io_unplug(BlockDriverState *bs) } } =20 -void bdrv_register_buf(BlockDriverState *bs, void *host, size_t size) +/* Helper that undoes bdrv_register_buf() when it fails partway through */ +static void bdrv_register_buf_rollback(BlockDriverState *bs, + void *host, + size_t size, + BdrvChild *final_child) +{ + BdrvChild *child; + + QLIST_FOREACH(child, &bs->children, next) { + if (child =3D=3D final_child) { + break; + } + + bdrv_unregister_buf(child->bs, host, size); + } + + if (bs->drv && bs->drv->bdrv_unregister_buf) { + bs->drv->bdrv_unregister_buf(bs, host, size); + } +} + +bool bdrv_register_buf(BlockDriverState *bs, void *host, size_t size, + Error **errp) { BdrvChild *child; =20 GLOBAL_STATE_CODE(); if (bs->drv && bs->drv->bdrv_register_buf) { - bs->drv->bdrv_register_buf(bs, host, size); + if (!bs->drv->bdrv_register_buf(bs, host, size, errp)) { + return false; + } } QLIST_FOREACH(child, &bs->children, next) { - bdrv_register_buf(child->bs, host, size); + if (!bdrv_register_buf(child->bs, host, size, errp)) { + bdrv_register_buf_rollback(bs, host, size, child); + return false; + } } + return true; } =20 void bdrv_unregister_buf(BlockDriverState *bs, void *host, size_t size) diff --git a/block/nvme.c b/block/nvme.c index 696502acea..1e76fe21d3 100644 --- a/block/nvme.c +++ b/block/nvme.c @@ -1577,19 +1577,19 @@ static void nvme_aio_unplug(BlockDriverState *bs) } } =20 -static void nvme_register_buf(BlockDriverState *bs, void *host, size_t siz= e) +static bool nvme_register_buf(BlockDriverState *bs, void *host, size_t siz= e, + Error **errp) { int ret; - Error *local_err =3D NULL; BDRVNVMeState *s =3D bs->opaque; =20 - ret =3D qemu_vfio_dma_map(s->vfio, host, size, false, NULL, &local_err= ); - if (ret) { - /* FIXME: we may run out of IOVA addresses after repeated - * bdrv_register_buf/bdrv_unregister_buf, because nvme_vfio_dma_un= map - * doesn't reclaim addresses for fixed mappings. */ - error_reportf_err(local_err, "nvme_register_buf failed: "); - } + /* + * FIXME: we may run out of IOVA addresses after repeated + * bdrv_register_buf/bdrv_unregister_buf, because nvme_vfio_dma_unmap + * doesn't reclaim addresses for fixed mappings. + */ + ret =3D qemu_vfio_dma_map(s->vfio, host, size, false, NULL, errp); + return ret =3D=3D 0; } =20 static void nvme_unregister_buf(BlockDriverState *bs, void *host, size_t s= ize) diff --git a/qemu-img.c b/qemu-img.c index 1dd5efe840..1ae1c5648e 100644 --- a/qemu-img.c +++ b/qemu-img.c @@ -4567,7 +4567,7 @@ static int img_bench(int argc, char **argv) data.buf =3D blk_blockalign(blk, buf_size); memset(data.buf, pattern, data.nrreq * data.bufsize); =20 - blk_register_buf(blk, data.buf, buf_size); + blk_register_buf(blk, data.buf, buf_size, &error_fatal); =20 data.qiov =3D g_new(QEMUIOVector, data.nrreq); for (i =3D 0; i < data.nrreq; i++) { --=20 2.37.3 From nobody Mon Feb 9 06:26:38 2026 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1665092630; cv=none; d=zohomail.com; s=zohoarc; b=LDijSl0bPGOKxR/T5NAl4fH/Kbjt0T8Fj16L9bJR/xoiI/p+kh/wujp4Ba36jL5Bez8u7k9WRqwhc2qTxFoY00eyDCXW6dVuOe7eHxdTax86jyjsqJR90oed49swzGBTX0d8VKzYzyW9+j3CO9TWxCIklkTxzP3aTWsLE0nldXw= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1665092630; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=ulV7x28imj7JCtzVIT5rq6ATqBNvKJ3ly+0N0CL/O7E=; b=Nykb4w6+8IOEhdXeHgctOcjBkGw3QJu2vW1A5ayT8fqnLJ8P3ycBMWn/erSV/iXGvmg69jQQzkJNQNvg9w8sizhoovlXDP5vsOk6OQx1qm0KItLbSU+qqUYTI1EnJhnP0+lPK4Km+8bYZlP/vL81UreUZ0YXEgdvJwDoaE2tGg4= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1665092630612254.49694429307635; Thu, 6 Oct 2022 14:43:50 -0700 (PDT) Received: from localhost ([::1]:54596 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1ogYei-0007qj-MR for importer@patchew.org; Thu, 06 Oct 2022 17:43:48 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:39016) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWp-0003HP-90 for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:41 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.133.124]:50843) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWn-0006pQ-2m for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:38 -0400 Received: from mimecast-mx02.redhat.com (mx3-rdu2.redhat.com [66.187.233.73]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-184-ReA4DR-nM9myQ6Uqk8AjdA-1; Thu, 06 Oct 2022 17:35:32 -0400 Received: from smtp.corp.redhat.com (int-mx03.intmail.prod.int.rdu2.redhat.com [10.11.54.3]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id D7A543C025CB; Thu, 6 Oct 2022 21:35:31 +0000 (UTC) Received: from localhost (unknown [10.39.193.119]) by smtp.corp.redhat.com (Postfix) with ESMTP id 3B4EF111CD4B; Thu, 6 Oct 2022 21:35:31 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1665092135; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=ulV7x28imj7JCtzVIT5rq6ATqBNvKJ3ly+0N0CL/O7E=; b=OwWHKooLNdX9waSRcDYeO4a99WkWTJtb945v3DKAucQT2Rf6eYK8wrmdEgnU5YHoL1o/NN DBp55vu8MEdjpN4nOzts8qhHIYWBDC/iiFF1cEHxfePLWGJqpFWSTR3YILFsa48h8K2/04 DPrjKDX5bTOblTOU6AnC7HefM3NBCmk= X-MC-Unique: ReA4DR-nM9myQ6Uqk8AjdA-1 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Yanan Wang , sgarzare@redhat.com, "Richard W.M. Jones" , Fam Zheng , Hanna Reitz , David Hildenbrand , integration@gluster.org, qemu-block@nongnu.org, Vladimir Sementsov-Ogievskiy , Paolo Bonzini , Kevin Wolf , afaria@redhat.com, Richard Henderson , Eric Blake , "Michael S. Tsirkin" , Thomas Huth , Xie Changlong , John Snow , Eduardo Habkost , Jeff Cody , Stefan Hajnoczi , "Denis V. Lunev" , Markus Armbruster , Laurent Vivier , Peter Xu , Raphael Norwitz , Marcel Apfelbaum , =?UTF-8?q?Philippe=20Mathieu-Daud=C3=A9?= , Wen Congyang Subject: [PATCH v6 08/13] numa: use QLIST_FOREACH_SAFE() for RAM block notifiers Date: Thu, 6 Oct 2022 17:35:02 -0400 Message-Id: <20221006213507.645402-9-stefanha@redhat.com> In-Reply-To: <20221006213507.645402-1-stefanha@redhat.com> References: <20221006213507.645402-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.3 Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=170.10.133.124; envelope-from=stefanha@redhat.com; helo=us-smtp-delivery-124.mimecast.com 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, 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: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1665092632682100001 Content-Type: text/plain; charset="utf-8" Make list traversal work when a callback removes a notifier mid-traversal. This is a cleanup to prevent bugs in the future. Signed-off-by: Stefan Hajnoczi Reviewed-by: David Hildenbrand --- hw/core/numa.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/hw/core/numa.c b/hw/core/numa.c index 31e6fe1caa..ea24a5fa8c 100644 --- a/hw/core/numa.c +++ b/hw/core/numa.c @@ -857,8 +857,9 @@ void ram_block_notifier_remove(RAMBlockNotifier *n) void ram_block_notify_add(void *host, size_t size, size_t max_size) { RAMBlockNotifier *notifier; + RAMBlockNotifier *next; =20 - QLIST_FOREACH(notifier, &ram_list.ramblock_notifiers, next) { + QLIST_FOREACH_SAFE(notifier, &ram_list.ramblock_notifiers, next, next)= { if (notifier->ram_block_added) { notifier->ram_block_added(notifier, host, size, max_size); } @@ -868,8 +869,9 @@ void ram_block_notify_add(void *host, size_t size, size= _t max_size) void ram_block_notify_remove(void *host, size_t size, size_t max_size) { RAMBlockNotifier *notifier; + RAMBlockNotifier *next; =20 - QLIST_FOREACH(notifier, &ram_list.ramblock_notifiers, next) { + QLIST_FOREACH_SAFE(notifier, &ram_list.ramblock_notifiers, next, next)= { if (notifier->ram_block_removed) { notifier->ram_block_removed(notifier, host, size, max_size); } @@ -879,8 +881,9 @@ void ram_block_notify_remove(void *host, size_t size, s= ize_t max_size) void ram_block_notify_resize(void *host, size_t old_size, size_t new_size) { RAMBlockNotifier *notifier; + RAMBlockNotifier *next; =20 - QLIST_FOREACH(notifier, &ram_list.ramblock_notifiers, next) { + QLIST_FOREACH_SAFE(notifier, &ram_list.ramblock_notifiers, next, next)= { if (notifier->ram_block_resized) { notifier->ram_block_resized(notifier, host, old_size, new_size= ); } --=20 2.37.3 From nobody Mon Feb 9 06:26:38 2026 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1665092845; cv=none; d=zohomail.com; s=zohoarc; b=CwyMozR5tRulIAKgUe+9Y/M5P6sKFgJ9I3ubDoYG3QFoRsqVAIAG2QqmzsLMGpsG/R3viorbboJHF14NzUI5rrgfYHDrs1b9dFgD7+JQgAk74IIolsgmfXwlDa8RQXiKzji01At8c/6g1WAFX29K5+gYS0gv8gYnH5EdT0dEd24= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1665092845; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=L9dc4RCOlCVz8fWR0fMBCkDrdz+LUU9NRcMBPJbuPTg=; b=RhsXvmG8miNeSEhaEfFUZ6Lah/3KE2TZu6QHLOws+GT2/52B5stutLwh8BTjxdXP6yTo21lkHUN1zGJYul3D5KkvdjGJNRiaYsrPNU2PooqCKdjq34zvGF59UjlvGi8FniOx9ufqyr9eCzrJOikVML7IWBNMBPOm9FDrqTmZO7Y= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 166509284589135.07896448780593; Thu, 6 Oct 2022 14:47:25 -0700 (PDT) Received: from localhost ([::1]:42920 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1ogYiC-0003W3-SI for importer@patchew.org; Thu, 06 Oct 2022 17:47:24 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:40002) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWx-0003Im-5T for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:49 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.129.124]:49295) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWs-0006tZ-S1 for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:46 -0400 Received: from mimecast-mx02.redhat.com (mx3-rdu2.redhat.com [66.187.233.73]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-481-Y7_E2QfLMwGW3jw6DGzOVA-1; Thu, 06 Oct 2022 17:35:35 -0400 Received: from smtp.corp.redhat.com (int-mx10.intmail.prod.int.rdu2.redhat.com [10.11.54.10]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 253183C01DE6; Thu, 6 Oct 2022 21:35:34 +0000 (UTC) Received: from localhost (unknown [10.39.193.119]) by smtp.corp.redhat.com (Postfix) with ESMTP id 758E453593B; Thu, 6 Oct 2022 21:35:33 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1665092142; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=L9dc4RCOlCVz8fWR0fMBCkDrdz+LUU9NRcMBPJbuPTg=; b=h+9OJ5de7+ru4ooaHUAO03jxojU5LVM73rHPOak7DUJ5MSBYJCqYAIVHzUfy1IBeLr46WU ApbE90b0ys6uXZdUMgZK+hjznLjJhGpC5wfgKKdxYxqkPafp4R5pWU5tGeQwPo8O7b/wPO 8iJctGJYe2rwDo15e1+AOLy3EENbv3Q= X-MC-Unique: Y7_E2QfLMwGW3jw6DGzOVA-1 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Yanan Wang , sgarzare@redhat.com, "Richard W.M. Jones" , Fam Zheng , Hanna Reitz , David Hildenbrand , integration@gluster.org, qemu-block@nongnu.org, Vladimir Sementsov-Ogievskiy , Paolo Bonzini , Kevin Wolf , afaria@redhat.com, Richard Henderson , Eric Blake , "Michael S. Tsirkin" , Thomas Huth , Xie Changlong , John Snow , Eduardo Habkost , Jeff Cody , Stefan Hajnoczi , "Denis V. Lunev" , Markus Armbruster , Laurent Vivier , Peter Xu , Raphael Norwitz , Marcel Apfelbaum , =?UTF-8?q?Philippe=20Mathieu-Daud=C3=A9?= , Wen Congyang Subject: [PATCH v6 09/13] block: add BlockRAMRegistrar Date: Thu, 6 Oct 2022 17:35:03 -0400 Message-Id: <20221006213507.645402-10-stefanha@redhat.com> In-Reply-To: <20221006213507.645402-1-stefanha@redhat.com> References: <20221006213507.645402-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.10 Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=170.10.129.124; envelope-from=stefanha@redhat.com; helo=us-smtp-delivery-124.mimecast.com X-Spam_score_int: -27 X-Spam_score: -2.8 X-Spam_bar: -- X-Spam_report: (-2.8 / 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_LOW=-0.7, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=unavailable autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1665092846215100001 Content-Type: text/plain; charset="utf-8" Emulated devices and other BlockBackend users wishing to take advantage of blk_register_buf() all have the same repetitive job: register RAMBlocks with the BlockBackend using RAMBlockNotifier. Add a BlockRAMRegistrar API to do this. A later commit will use this from hw/block/virtio-blk.c. Signed-off-by: Stefan Hajnoczi --- MAINTAINERS | 1 + include/sysemu/block-ram-registrar.h | 37 ++++++++++++++++++ block/block-ram-registrar.c | 58 ++++++++++++++++++++++++++++ block/meson.build | 1 + 4 files changed, 97 insertions(+) create mode 100644 include/sysemu/block-ram-registrar.h create mode 100644 block/block-ram-registrar.c diff --git a/MAINTAINERS b/MAINTAINERS index 0dcae6168a..91aed2cdc7 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2498,6 +2498,7 @@ F: block* F: block/ F: hw/block/ F: include/block/ +F: include/sysemu/block-*.h F: qemu-img* F: docs/tools/qemu-img.rst F: qemu-io* diff --git a/include/sysemu/block-ram-registrar.h b/include/sysemu/block-ra= m-registrar.h new file mode 100644 index 0000000000..d8b2f7942b --- /dev/null +++ b/include/sysemu/block-ram-registrar.h @@ -0,0 +1,37 @@ +/* + * BlockBackend RAM Registrar + * + * SPDX-License-Identifier: GPL-2.0-or-later + */ + +#ifndef BLOCK_RAM_REGISTRAR_H +#define BLOCK_RAM_REGISTRAR_H + +#include "exec/ramlist.h" + +/** + * struct BlockRAMRegistrar: + * + * Keeps RAMBlock memory registered with a BlockBackend using + * blk_register_buf() including hotplugged memory. + * + * Emulated devices or other BlockBackend users initialize a BlockRAMRegis= trar + * with blk_ram_registrar_init() before submitting I/O requests with the + * BDRV_REQ_REGISTERED_BUF flag set. + */ +typedef struct { + BlockBackend *blk; + RAMBlockNotifier notifier; + bool ok; +} BlockRAMRegistrar; + +void blk_ram_registrar_init(BlockRAMRegistrar *r, BlockBackend *blk); +void blk_ram_registrar_destroy(BlockRAMRegistrar *r); + +/* Have all RAMBlocks been registered successfully? */ +static inline bool blk_ram_registrar_ok(BlockRAMRegistrar *r) +{ + return r->ok; +} + +#endif /* BLOCK_RAM_REGISTRAR_H */ diff --git a/block/block-ram-registrar.c b/block/block-ram-registrar.c new file mode 100644 index 0000000000..25dbafa789 --- /dev/null +++ b/block/block-ram-registrar.c @@ -0,0 +1,58 @@ +/* + * BlockBackend RAM Registrar + * + * SPDX-License-Identifier: GPL-2.0-or-later + */ + +#include "qemu/osdep.h" +#include "sysemu/block-backend.h" +#include "sysemu/block-ram-registrar.h" +#include "qapi/error.h" + +static void ram_block_added(RAMBlockNotifier *n, void *host, size_t size, + size_t max_size) +{ + BlockRAMRegistrar *r =3D container_of(n, BlockRAMRegistrar, notifier); + Error *err =3D NULL; + + if (!r->ok) { + return; /* don't try again if we've already failed */ + } + + if (!blk_register_buf(r->blk, host, max_size, &err)) { + error_report_err(err); + ram_block_notifier_remove(&r->notifier); + r->ok =3D false; + } +} + +static void ram_block_removed(RAMBlockNotifier *n, void *host, size_t size, + size_t max_size) +{ + BlockRAMRegistrar *r =3D container_of(n, BlockRAMRegistrar, notifier); + blk_unregister_buf(r->blk, host, max_size); +} + +void blk_ram_registrar_init(BlockRAMRegistrar *r, BlockBackend *blk) +{ + r->blk =3D blk; + r->notifier =3D (RAMBlockNotifier){ + .ram_block_added =3D ram_block_added, + .ram_block_removed =3D ram_block_removed, + + /* + * .ram_block_resized() is not necessary because we use the max_si= ze + * value that does not change across resize. + */ + }; + r->ok =3D true; + + ram_block_notifier_add(&r->notifier); +} + +void blk_ram_registrar_destroy(BlockRAMRegistrar *r) +{ + if (r->ok) { + ram_block_notifier_remove(&r->notifier); + } +} diff --git a/block/meson.build b/block/meson.build index 500878f082..b7c68b83a3 100644 --- a/block/meson.build +++ b/block/meson.build @@ -46,6 +46,7 @@ block_ss.add(files( ), zstd, zlib, gnutls) =20 softmmu_ss.add(when: 'CONFIG_TCG', if_true: files('blkreplay.c')) +softmmu_ss.add(files('block-ram-registrar.c')) =20 if get_option('qcow1').allowed() block_ss.add(files('qcow.c')) --=20 2.37.3 From nobody Mon Feb 9 06:26:38 2026 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1665092921; cv=none; d=zohomail.com; s=zohoarc; b=Nm7RxhvJdafG2gGtYB1Ao5pNtcG/tLr5RsfjBTG1qXun/n6780qz1Epo+dO+cKmz55HsdK2QBac/6Tqb1MLvT2KX7PEcH04Yc6VkQUAY08m1QvFX6jGs+CDSsa5Od9Qm2fQ6CFywLKuf2r7WWYBm1hb5JGe3BNR1j1Fp0JjpeS4= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1665092921; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=2uoNyr/XSKf+OD83Nh+bVP2/xYaSIaJlR6XmlFpdjFM=; b=fJw+wm5tn7piZsjz9EY6Y2EbJ/hfGx/LCXJ4jlSZ+UZ42uer2GzB73Y9/IoQW0NbU+JgUXS1b8SaNxfbC2jlmIhSMT9O8Sxg7WubUoSCfQlaPOQXVh/LCXG10sqVAO/UlfflCzccso51r+Kqq4S4cRzl21gvCO/l/SnwtiZWg+Q= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1665092921104252.14080050442124; Thu, 6 Oct 2022 14:48:41 -0700 (PDT) Received: from localhost ([::1]:53672 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1ogYjQ-0004mK-0k for importer@patchew.org; Thu, 06 Oct 2022 17:48:40 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:40004) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWx-0003In-68 for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:49 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.129.124]:42858) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWt-0006tj-Uf for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:46 -0400 Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-277-H44xBWIzNpm628qf-T4axA-1; Thu, 06 Oct 2022 17:35:37 -0400 Received: from smtp.corp.redhat.com (int-mx01.intmail.prod.int.rdu2.redhat.com [10.11.54.1]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 9ED9B858F13; Thu, 6 Oct 2022 21:35:36 +0000 (UTC) Received: from localhost (unknown [10.39.193.119]) by smtp.corp.redhat.com (Postfix) with ESMTP id 0EE82404705D; Thu, 6 Oct 2022 21:35:35 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1665092142; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=2uoNyr/XSKf+OD83Nh+bVP2/xYaSIaJlR6XmlFpdjFM=; b=GI1gh2J2MUzDBR9J2Q+6bcczPNiJJTJMGnfSg5GzaUumB2ENzqpzqwyDNMgk4U8at1PWHl 3D6I5gh10jP33MFc4e1uJTgX805u1YlY/a8jHA3ac+r32LWIAPgz+klyHlWo3GpoxYO24C yl/adsTu8TiDVUbc0tA4vEhDskEWwFA= X-MC-Unique: H44xBWIzNpm628qf-T4axA-1 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Yanan Wang , sgarzare@redhat.com, "Richard W.M. Jones" , Fam Zheng , Hanna Reitz , David Hildenbrand , integration@gluster.org, qemu-block@nongnu.org, Vladimir Sementsov-Ogievskiy , Paolo Bonzini , Kevin Wolf , afaria@redhat.com, Richard Henderson , Eric Blake , "Michael S. Tsirkin" , Thomas Huth , Xie Changlong , John Snow , Eduardo Habkost , Jeff Cody , Stefan Hajnoczi , "Denis V. Lunev" , Markus Armbruster , Laurent Vivier , Peter Xu , Raphael Norwitz , Marcel Apfelbaum , =?UTF-8?q?Philippe=20Mathieu-Daud=C3=A9?= , Wen Congyang Subject: [PATCH v6 10/13] exec/cpu-common: add qemu_ram_get_fd() Date: Thu, 6 Oct 2022 17:35:04 -0400 Message-Id: <20221006213507.645402-11-stefanha@redhat.com> In-Reply-To: <20221006213507.645402-1-stefanha@redhat.com> References: <20221006213507.645402-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.1 Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=170.10.129.124; envelope-from=stefanha@redhat.com; helo=us-smtp-delivery-124.mimecast.com X-Spam_score_int: -27 X-Spam_score: -2.8 X-Spam_bar: -- X-Spam_report: (-2.8 / 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_LOW=-0.7, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=unavailable autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1665092923196100001 Content-Type: text/plain; charset="utf-8" Add a function to get the file descriptor for a RAMBlock. Device emulation code typically uses the MemoryRegion APIs but vhost-style code may use RAMBlock directly for sharing guest memory with another process. This new API will be used by the libblkio block driver so it can share guest memory via .bdrv_register_buf(). Signed-off-by: Stefan Hajnoczi --- include/exec/cpu-common.h | 1 + softmmu/physmem.c | 5 +++++ 2 files changed, 6 insertions(+) diff --git a/include/exec/cpu-common.h b/include/exec/cpu-common.h index c493510ee9..6feaa40ca7 100644 --- a/include/exec/cpu-common.h +++ b/include/exec/cpu-common.h @@ -92,6 +92,7 @@ void qemu_ram_set_uf_zeroable(RAMBlock *rb); bool qemu_ram_is_migratable(RAMBlock *rb); void qemu_ram_set_migratable(RAMBlock *rb); void qemu_ram_unset_migratable(RAMBlock *rb); +int qemu_ram_get_fd(RAMBlock *rb); =20 size_t qemu_ram_pagesize(RAMBlock *block); size_t qemu_ram_pagesize_largest(void); diff --git a/softmmu/physmem.c b/softmmu/physmem.c index 56e03e07b5..d9578ccfd4 100644 --- a/softmmu/physmem.c +++ b/softmmu/physmem.c @@ -1748,6 +1748,11 @@ void qemu_ram_unset_migratable(RAMBlock *rb) rb->flags &=3D ~RAM_MIGRATABLE; } =20 +int qemu_ram_get_fd(RAMBlock *rb) +{ + return rb->fd; +} + /* Called with iothread lock held. */ void qemu_ram_set_idstr(RAMBlock *new_block, const char *name, DeviceState= *dev) { --=20 2.37.3 From nobody Mon Feb 9 06:26:38 2026 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1665092652; cv=none; d=zohomail.com; s=zohoarc; b=BiFBl1EoK9lNGvf/tUyvXxnbOgH/pxmnjsBzZPRPU5KlPjFraoiyjDX/2MBUVJLJNMbzMY2muOsSzdsjc09rHkZeBtPTaesgyGAAKbTfphN8B02yjekYGF9exhp89ugwANCeLpIlfkSgiRcUFf6jhYA+HdYKgLdfg+eeL/V1WSI= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1665092652; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=LO1Yk9Zyzrv9MQU42UWrwS5AGoOuZeWvHnsE7ewd4ko=; b=Jc669kPzRQavAR9n8azAyChjqaT8Os290/NByw5fN9V+5uEbYVxy0Ch87C/Wp3DBW8UMgm86s+EFI5AkFwYPPEo3FyVlOBEtJLZKHqbRe5563Y1+zfsZI8QpFBXmyydBtKPH4E16bAIWhhMJcJIcMol62HB6qdkJoLas0TDAc0M= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1665092652306230.62633606023826; Thu, 6 Oct 2022 14:44:12 -0700 (PDT) Received: from localhost ([::1]:39568 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1ogYf5-0008WL-84 for importer@patchew.org; Thu, 06 Oct 2022 17:44:11 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:40008) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWx-0003Ip-6G for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:49 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.129.124]:60220) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWt-0006to-VF for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:46 -0400 Received: from mimecast-mx02.redhat.com (mx3-rdu2.redhat.com [66.187.233.73]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-471-xo-4woCxNDq_-zFSr4jMxw-1; Thu, 06 Oct 2022 17:35:40 -0400 Received: from smtp.corp.redhat.com (int-mx08.intmail.prod.int.rdu2.redhat.com [10.11.54.8]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 1C6891C068C9; Thu, 6 Oct 2022 21:35:39 +0000 (UTC) Received: from localhost (unknown [10.39.193.119]) by smtp.corp.redhat.com (Postfix) with ESMTP id 5EF25C15BA4; Thu, 6 Oct 2022 21:35:38 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1665092142; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=LO1Yk9Zyzrv9MQU42UWrwS5AGoOuZeWvHnsE7ewd4ko=; b=Eu5m59SrY5W6yM8bkNQPea/wCsowzF5YtQ1G0/0u1udCOrsGRKWgAiOVXb9C4z+otSdhzx bZWVNcuYadp6JbzWj/AUUWj720d1pwzBCMmRasEnKE0yR7gAdkVGQu8l/bRBzfiT/r+Tyh fXsczoTq0Znw0wAkuSfNiXqATVtbvxo= X-MC-Unique: xo-4woCxNDq_-zFSr4jMxw-1 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Yanan Wang , sgarzare@redhat.com, "Richard W.M. Jones" , Fam Zheng , Hanna Reitz , David Hildenbrand , integration@gluster.org, qemu-block@nongnu.org, Vladimir Sementsov-Ogievskiy , Paolo Bonzini , Kevin Wolf , afaria@redhat.com, Richard Henderson , Eric Blake , "Michael S. Tsirkin" , Thomas Huth , Xie Changlong , John Snow , Eduardo Habkost , Jeff Cody , Stefan Hajnoczi , "Denis V. Lunev" , Markus Armbruster , Laurent Vivier , Peter Xu , Raphael Norwitz , Marcel Apfelbaum , =?UTF-8?q?Philippe=20Mathieu-Daud=C3=A9?= , Wen Congyang Subject: [PATCH v6 11/13] stubs: add qemu_ram_block_from_host() and qemu_ram_get_fd() Date: Thu, 6 Oct 2022 17:35:05 -0400 Message-Id: <20221006213507.645402-12-stefanha@redhat.com> In-Reply-To: <20221006213507.645402-1-stefanha@redhat.com> References: <20221006213507.645402-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.8 Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=170.10.129.124; envelope-from=stefanha@redhat.com; helo=us-smtp-delivery-124.mimecast.com X-Spam_score_int: -27 X-Spam_score: -2.8 X-Spam_bar: -- X-Spam_report: (-2.8 / 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_LOW=-0.7, 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: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1665092652930100001 Content-Type: text/plain; charset="utf-8" The blkio block driver will need to look up the file descriptor for a given pointer. This is possible in softmmu builds where the RAMBlock API is available for querying guest RAM. Add stubs so tools like qemu-img that link the block layer still build successfully. In this case there is no guest RAM but that is fine. Bounce buffers and their file descriptors will be allocated with libblkio's blkio_alloc_mem_region() so we won't rely on QEMU's qemu_ram_get_fd() in that case. Signed-off-by: Stefan Hajnoczi --- stubs/physmem.c | 13 +++++++++++++ stubs/meson.build | 1 + 2 files changed, 14 insertions(+) create mode 100644 stubs/physmem.c diff --git a/stubs/physmem.c b/stubs/physmem.c new file mode 100644 index 0000000000..1fc5f2df29 --- /dev/null +++ b/stubs/physmem.c @@ -0,0 +1,13 @@ +#include "qemu/osdep.h" +#include "exec/cpu-common.h" + +RAMBlock *qemu_ram_block_from_host(void *ptr, bool round_offset, + ram_addr_t *offset) +{ + return NULL; +} + +int qemu_ram_get_fd(RAMBlock *rb) +{ + return -1; +} diff --git a/stubs/meson.build b/stubs/meson.build index d8f3fd5c44..4314161f5f 100644 --- a/stubs/meson.build +++ b/stubs/meson.build @@ -29,6 +29,7 @@ stub_ss.add(files('migr-blocker.c')) stub_ss.add(files('module-opts.c')) stub_ss.add(files('monitor.c')) stub_ss.add(files('monitor-core.c')) +stub_ss.add(files('physmem.c')) stub_ss.add(files('qemu-timer-notify-cb.c')) stub_ss.add(files('qmp_memory_device.c')) stub_ss.add(files('qmp-command-available.c')) --=20 2.37.3 From nobody Mon Feb 9 06:26:38 2026 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1665092891; cv=none; d=zohomail.com; s=zohoarc; b=dk9YxEpJ5kBQ5aIOUPPAp2vrfGhJ+evTbUxYmd1xfQt2D5txiKbjifjqixR+Nrkt14lKCFoYph+LioMleS+Wg7+k8LEz1cfEhhF/+ST56grN4iBU5BAlXtaI809sQ6morXneh5/OAR3jL5GBVbjm1SlmzAPaA/fdjBsVi+4icRs= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1665092891; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=1xyQ3dNcj+VIHXj8KVmFi/+Zy4OsKueHp9mB73k0+zw=; b=H5dbRMSzSdOV4kiFz2smiCrVoj4ZvMdZL3JUOqOLlGgXOisQdO4h7tpzVVkjqicsaHak9IcuJIpcOmiTxR33P/4nDJ4qPiMGvsUjBkWpmO9zOSUfmpbco7+8rrHoseZoAik3vzBQD2FJSafWDEuhkjl1WWVNfW32718iXQfiVE0= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1665092891584563.0393102663498; Thu, 6 Oct 2022 14:48:11 -0700 (PDT) Received: from localhost ([::1]:48486 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1ogYiw-0003yR-9K for importer@patchew.org; Thu, 06 Oct 2022 17:48:10 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:40014) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWz-0003J0-HO for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:50 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.133.124]:41182) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWw-0006w2-ND for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:48 -0400 Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-584-vdKRpJwHMNqqa15WvT47uQ-1; Thu, 06 Oct 2022 17:35:42 -0400 Received: from smtp.corp.redhat.com (int-mx10.intmail.prod.int.rdu2.redhat.com [10.11.54.10]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 5D9FB811E67; Thu, 6 Oct 2022 21:35:41 +0000 (UTC) Received: from localhost (unknown [10.39.193.119]) by smtp.corp.redhat.com (Postfix) with ESMTP id C774B535939; Thu, 6 Oct 2022 21:35:40 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1665092145; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=1xyQ3dNcj+VIHXj8KVmFi/+Zy4OsKueHp9mB73k0+zw=; b=RFO/03KtwsSUjy8Cqi8oxK2aU4DU5i4COUswOKRl/lIfJaO/+3KGZu+CzGTTys+U8M/Bhs mQ8AaYYpVTtkPgI1IwFl1aOm0oxcVLM7XXoW7aoeBx3Wr12JDw5BdjKC8/n4obYynoTw+W jZXIWOH00fQBNjrQZucMYu63NgoAqHQ= X-MC-Unique: vdKRpJwHMNqqa15WvT47uQ-1 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Yanan Wang , sgarzare@redhat.com, "Richard W.M. Jones" , Fam Zheng , Hanna Reitz , David Hildenbrand , integration@gluster.org, qemu-block@nongnu.org, Vladimir Sementsov-Ogievskiy , Paolo Bonzini , Kevin Wolf , afaria@redhat.com, Richard Henderson , Eric Blake , "Michael S. Tsirkin" , Thomas Huth , Xie Changlong , John Snow , Eduardo Habkost , Jeff Cody , Stefan Hajnoczi , "Denis V. Lunev" , Markus Armbruster , Laurent Vivier , Peter Xu , Raphael Norwitz , Marcel Apfelbaum , =?UTF-8?q?Philippe=20Mathieu-Daud=C3=A9?= , Wen Congyang Subject: [PATCH v6 12/13] blkio: implement BDRV_REQ_REGISTERED_BUF optimization Date: Thu, 6 Oct 2022 17:35:06 -0400 Message-Id: <20221006213507.645402-13-stefanha@redhat.com> In-Reply-To: <20221006213507.645402-1-stefanha@redhat.com> References: <20221006213507.645402-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.10 Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=170.10.133.124; envelope-from=stefanha@redhat.com; helo=us-smtp-delivery-124.mimecast.com 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, 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: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1665092892454100001 Content-Type: text/plain; charset="utf-8" Avoid bounce buffers when QEMUIOVector elements are within previously registered bdrv_register_buf() buffers. The idea is that emulated storage controllers will register guest RAM using bdrv_register_buf() and set the BDRV_REQ_REGISTERED_BUF on I/O requests. Therefore no blkio_map_mem_region() calls are necessary in the performance-critical I/O code path. This optimization doesn't apply if the I/O buffer is internally allocated by QEMU (e.g. qcow2 metadata). There we still take the slow path because BDRV_REQ_REGISTERED_BUF is not set. Signed-off-by: Stefan Hajnoczi --- block/blkio.c | 183 +++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 180 insertions(+), 3 deletions(-) diff --git a/block/blkio.c b/block/blkio.c index 9a79789a39..5ce61d5d94 100644 --- a/block/blkio.c +++ b/block/blkio.c @@ -11,9 +11,13 @@ #include "qemu/osdep.h" #include #include "block/block_int.h" +#include "exec/memory.h" +#include "exec/cpu-common.h" /* for qemu_ram_get_fd() */ #include "qapi/error.h" +#include "qemu/error-report.h" #include "qapi/qmp/qdict.h" #include "qemu/module.h" +#include "exec/memory.h" /* for ram_block_discard_disable() */ =20 /* * Keep the QEMU BlockDriver names identical to the libblkio driver names. @@ -73,6 +77,12 @@ typedef struct { =20 /* Can we skip adding/deleting blkio_mem_regions? */ bool needs_mem_regions; + + /* Are file descriptors necessary for blkio_mem_regions? */ + bool needs_mem_region_fd; + + /* Are madvise(MADV_DONTNEED)-style operations unavailable? */ + bool mem_regions_pinned; } BDRVBlkioState; =20 /* Called with s->bounce_lock held */ @@ -347,7 +357,8 @@ blkio_co_preadv(BlockDriverState *bs, int64_t offset, i= nt64_t bytes, .coroutine =3D qemu_coroutine_self(), }; BDRVBlkioState *s =3D bs->opaque; - bool use_bounce_buffer =3D s->needs_mem_regions; + bool use_bounce_buffer =3D + s->needs_mem_regions && !(flags & BDRV_REQ_REGISTERED_BUF); BlkioBounceBuf bounce; struct iovec *iov =3D qiov->iov; int iovcnt =3D qiov->niov; @@ -390,7 +401,8 @@ static int coroutine_fn blkio_co_pwritev(BlockDriverSta= te *bs, int64_t offset, .coroutine =3D qemu_coroutine_self(), }; BDRVBlkioState *s =3D bs->opaque; - bool use_bounce_buffer =3D s->needs_mem_regions; + bool use_bounce_buffer =3D + s->needs_mem_regions && !(flags & BDRV_REQ_REGISTERED_BUF); BlkioBounceBuf bounce; struct iovec *iov =3D qiov->iov; int iovcnt =3D qiov->niov; @@ -473,6 +485,130 @@ static void blkio_io_unplug(BlockDriverState *bs) } } =20 +typedef enum { + BMRR_OK, + BMRR_SKIP, + BMRR_FAIL, +} BlkioMemRegionResult; + +/* + * Produce a struct blkio_mem_region for a given address and size. + * + * This function produces identical results when called multiple times wit= h the + * same arguments. This property is necessary because blkio_unmap_mem_regi= on() + * must receive the same struct blkio_mem_region field values that were pa= ssed + * to blkio_map_mem_region(). + */ +static BlkioMemRegionResult +blkio_mem_region_from_host(BlockDriverState *bs, + void *host, size_t size, + struct blkio_mem_region *region, + Error **errp) +{ + BDRVBlkioState *s =3D bs->opaque; + int fd =3D -1; + ram_addr_t fd_offset =3D 0; + + if (((uintptr_t)host | size) % s->mem_region_alignment) { + error_setg(errp, "unaligned buf %p with size %zu", host, size); + return BMRR_FAIL; + } + + /* Attempt to find the fd for the underlying memory */ + if (s->needs_mem_region_fd) { + RAMBlock *ram_block; + RAMBlock *end_block; + ram_addr_t offset; + + /* + * bdrv_register_buf() is called with the BQL held so mr lives at = least + * until this function returns. + */ + ram_block =3D qemu_ram_block_from_host(host, false, &fd_offset); + if (ram_block) { + fd =3D qemu_ram_get_fd(ram_block); + } + if (fd =3D=3D -1) { + /* + * Ideally every RAMBlock would have an fd. pc-bios and other + * things don't. Luckily they are usually not I/O buffers and = we + * can just ignore them. + */ + return BMRR_SKIP; + } + + /* Make sure the fd covers the entire range */ + end_block =3D qemu_ram_block_from_host(host + size - 1, false, &of= fset); + if (ram_block !=3D end_block) { + error_setg(errp, "registered buffer at %p with size %zu extend= s " + "beyond RAMBlock", host, size); + return BMRR_FAIL; + } + } + + *region =3D (struct blkio_mem_region){ + .addr =3D host, + .len =3D size, + .fd =3D fd, + .fd_offset =3D fd_offset, + }; + return BMRR_OK; +} + +static bool blkio_register_buf(BlockDriverState *bs, void *host, size_t si= ze, + Error **errp) +{ + BDRVBlkioState *s =3D bs->opaque; + struct blkio_mem_region region; + BlkioMemRegionResult region_result; + int ret; + + /* + * Mapping memory regions conflicts with RAM discard (virtio-mem) when + * there is pinning, so only do it when necessary. + */ + if (!s->needs_mem_regions && s->mem_regions_pinned) { + return true; + } + + region_result =3D blkio_mem_region_from_host(bs, host, size, ®ion, = errp); + if (region_result =3D=3D BMRR_SKIP) { + return true; + } else if (region_result !=3D BMRR_OK) { + return false; + } + + WITH_QEMU_LOCK_GUARD(&s->blkio_lock) { + ret =3D blkio_map_mem_region(s->blkio, ®ion); + } + + if (ret < 0) { + error_setg(errp, "Failed to add blkio mem region %p with size %zu:= %s", + host, size, blkio_get_error_msg()); + return false; + } + return true; +} + +static void blkio_unregister_buf(BlockDriverState *bs, void *host, size_t = size) +{ + BDRVBlkioState *s =3D bs->opaque; + struct blkio_mem_region region; + + /* See blkio_register_buf() */ + if (!s->needs_mem_regions && s->mem_regions_pinned) { + return; + } + + if (blkio_mem_region_from_host(bs, host, size, ®ion, NULL) !=3D BMR= R_OK) { + return; + } + + WITH_QEMU_LOCK_GUARD(&s->blkio_lock) { + blkio_unmap_mem_region(s->blkio, ®ion); + } +} + static int blkio_io_uring_open(BlockDriverState *bs, QDict *options, int f= lags, Error **errp) { @@ -609,6 +745,17 @@ static int blkio_file_open(BlockDriverState *bs, QDict= *options, int flags, return ret; } =20 + ret =3D blkio_get_bool(s->blkio, + "needs-mem-region-fd", + &s->needs_mem_region_fd); + if (ret < 0) { + error_setg_errno(errp, -ret, + "failed to get needs-mem-region-fd: %s", + blkio_get_error_msg()); + blkio_destroy(&s->blkio); + return ret; + } + ret =3D blkio_get_uint64(s->blkio, "mem-region-alignment", &s->mem_region_alignment); @@ -620,15 +767,39 @@ static int blkio_file_open(BlockDriverState *bs, QDic= t *options, int flags, return ret; } =20 + ret =3D blkio_get_bool(s->blkio, + "mem-regions-pinned", + &s->mem_regions_pinned); + if (ret < 0) { + /* Be conservative (assume pinning) if the property is not support= ed */ + s->mem_regions_pinned =3D s->needs_mem_regions; + } + + /* + * Notify if libblkio drivers pin memory and prevent features like + * virtio-mem from working. + */ + if (s->mem_regions_pinned) { + ret =3D ram_block_discard_disable(true); + if (ret < 0) { + error_setg_errno(errp, -ret, "ram_block_discard_disable() fail= ed"); + blkio_destroy(&s->blkio); + return ret; + } + } + ret =3D blkio_start(s->blkio); if (ret < 0) { error_setg_errno(errp, -ret, "blkio_start failed: %s", blkio_get_error_msg()); blkio_destroy(&s->blkio); + if (s->mem_regions_pinned) { + ram_block_discard_disable(false); + } return ret; } =20 - bs->supported_write_flags =3D BDRV_REQ_FUA; + bs->supported_write_flags =3D BDRV_REQ_FUA | BDRV_REQ_REGISTERED_BUF; bs->supported_zero_flags =3D BDRV_REQ_FUA | BDRV_REQ_MAY_UNMAP | BDRV_REQ_NO_FALLBACK; =20 @@ -652,6 +823,10 @@ static void blkio_close(BlockDriverState *bs) qemu_mutex_destroy(&s->blkio_lock); blkio_detach_aio_context(bs); blkio_destroy(&s->blkio); + + if (s->mem_regions_pinned) { + ram_block_discard_disable(false); + } } =20 static int64_t blkio_getlength(BlockDriverState *bs) @@ -798,6 +973,8 @@ static void blkio_refresh_limits(BlockDriverState *bs, = Error **errp) .bdrv_co_pwrite_zeroes =3D blkio_co_pwrite_zeroes, \ .bdrv_io_unplug =3D blkio_io_unplug, \ .bdrv_refresh_limits =3D blkio_refresh_limits, \ + .bdrv_register_buf =3D blkio_register_buf, \ + .bdrv_unregister_buf =3D blkio_unregister_buf, \ __VA_ARGS__ \ } =20 --=20 2.37.3 From nobody Mon Feb 9 06:26:38 2026 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1665093021; cv=none; d=zohomail.com; s=zohoarc; b=C+vMlgdbBm+xpTUyOPGb44B3NvysEULNgNH0bl76vgFZK97mNs03Geg1twPtQbKEbklTy7AHEPahPwjmPVquzn9bKzwSzwj84i/uE2GJxjYONYWpgjbxczb0JwyNYEPYVedn7PIOS7ApYIiztXurQ9+ihzTDdT5tYgguza8FzyM= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1665093021; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=Q3k5xxgLci7w3NjYeetRfQM1HfpM0zNoEvNQR10IswI=; b=VDqxSbeR5ODn9N/7gUr8BlCUjegOjIop3ovtqDm3SvX1nNtB2w+8OLyY9y37vXTcy4ymvTKCDGuNrlDDstBXxadEg9q2I9Fpt0kkjMGKuAYMLwBfhHPkalHznoAFfgdpVKhSw2NI4F89RdzpMj4jJRnKCUUt94pUe2NKzzcVlVc= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1665093021365968.2869904859225; Thu, 6 Oct 2022 14:50:21 -0700 (PDT) Received: from localhost ([::1]:48020 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1ogYl2-0007VL-Ap for importer@patchew.org; Thu, 06 Oct 2022 17:50:20 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:40016) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYX1-0003KU-5q for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:57 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.133.124]:44849) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1ogYWz-0006wq-9E for qemu-devel@nongnu.org; Thu, 06 Oct 2022 17:35:50 -0400 Received: from mimecast-mx02.redhat.com (mx3-rdu2.redhat.com [66.187.233.73]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-646-KVq_KIJXN5y5TJfRZptsIw-1; Thu, 06 Oct 2022 17:35:44 -0400 Received: from smtp.corp.redhat.com (int-mx04.intmail.prod.int.rdu2.redhat.com [10.11.54.4]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 3950D29324B5; Thu, 6 Oct 2022 21:35:44 +0000 (UTC) Received: from localhost (unknown [10.39.193.119]) by smtp.corp.redhat.com (Postfix) with ESMTP id 69E45207B317; Thu, 6 Oct 2022 21:35:43 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1665092147; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=Q3k5xxgLci7w3NjYeetRfQM1HfpM0zNoEvNQR10IswI=; b=RgfQRwgve+f9SK5rjRYjWHFR8qJXbfMh3/8lpGczARcU8ImvxTFK8r8SnXg00xl6GMqSe/ AqphVMKEKqu+02uZqLVqg5VItiKJ9BvzfZDxx+GmdAndCFdxU2EEXrm1uf0qrL3H6vjTZR pZIky8RP+9pQABeDeL+HG4X+bsrwBtc= X-MC-Unique: KVq_KIJXN5y5TJfRZptsIw-1 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Yanan Wang , sgarzare@redhat.com, "Richard W.M. Jones" , Fam Zheng , Hanna Reitz , David Hildenbrand , integration@gluster.org, qemu-block@nongnu.org, Vladimir Sementsov-Ogievskiy , Paolo Bonzini , Kevin Wolf , afaria@redhat.com, Richard Henderson , Eric Blake , "Michael S. Tsirkin" , Thomas Huth , Xie Changlong , John Snow , Eduardo Habkost , Jeff Cody , Stefan Hajnoczi , "Denis V. Lunev" , Markus Armbruster , Laurent Vivier , Peter Xu , Raphael Norwitz , Marcel Apfelbaum , =?UTF-8?q?Philippe=20Mathieu-Daud=C3=A9?= , Wen Congyang Subject: [PATCH v6 13/13] virtio-blk: use BDRV_REQ_REGISTERED_BUF optimization hint Date: Thu, 6 Oct 2022 17:35:07 -0400 Message-Id: <20221006213507.645402-14-stefanha@redhat.com> In-Reply-To: <20221006213507.645402-1-stefanha@redhat.com> References: <20221006213507.645402-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.4 Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=170.10.133.124; envelope-from=stefanha@redhat.com; helo=us-smtp-delivery-124.mimecast.com 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, 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: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1665093022164100001 Content-Type: text/plain; charset="utf-8" Register guest RAM using BlockRAMRegistrar and set the BDRV_REQ_REGISTERED_BUF flag so block drivers can optimize memory accesses in I/O requests. This is for vdpa-blk, vhost-user-blk, and other I/O interfaces that rely on DMA mapping/unmapping. Signed-off-by: Stefan Hajnoczi --- include/hw/virtio/virtio-blk.h | 2 ++ hw/block/virtio-blk.c | 39 ++++++++++++++++++++++------------ 2 files changed, 27 insertions(+), 14 deletions(-) diff --git a/include/hw/virtio/virtio-blk.h b/include/hw/virtio/virtio-blk.h index d311c57cca..7f589b4146 100644 --- a/include/hw/virtio/virtio-blk.h +++ b/include/hw/virtio/virtio-blk.h @@ -19,6 +19,7 @@ #include "hw/block/block.h" #include "sysemu/iothread.h" #include "sysemu/block-backend.h" +#include "sysemu/block-ram-registrar.h" #include "qom/object.h" =20 #define TYPE_VIRTIO_BLK "virtio-blk-device" @@ -64,6 +65,7 @@ struct VirtIOBlock { struct VirtIOBlockDataPlane *dataplane; uint64_t host_features; size_t config_size; + BlockRAMRegistrar blk_ram_registrar; }; =20 typedef struct VirtIOBlockReq { diff --git a/hw/block/virtio-blk.c b/hw/block/virtio-blk.c index e9ba752f6b..907f012c45 100644 --- a/hw/block/virtio-blk.c +++ b/hw/block/virtio-blk.c @@ -21,6 +21,7 @@ #include "hw/block/block.h" #include "hw/qdev-properties.h" #include "sysemu/blockdev.h" +#include "sysemu/block-ram-registrar.h" #include "sysemu/sysemu.h" #include "sysemu/runstate.h" #include "hw/virtio/virtio-blk.h" @@ -384,12 +385,14 @@ static void virtio_blk_handle_scsi(VirtIOBlockReq *re= q) } } =20 -static inline void submit_requests(BlockBackend *blk, MultiReqBuffer *mrb, +static inline void submit_requests(VirtIOBlock *s, MultiReqBuffer *mrb, int start, int num_reqs, int niov) { + BlockBackend *blk =3D s->blk; QEMUIOVector *qiov =3D &mrb->reqs[start]->qiov; int64_t sector_num =3D mrb->reqs[start]->sector_num; bool is_write =3D mrb->is_write; + BdrvRequestFlags flags =3D 0; =20 if (num_reqs > 1) { int i; @@ -420,12 +423,18 @@ static inline void submit_requests(BlockBackend *blk,= MultiReqBuffer *mrb, num_reqs - 1); } =20 + if (blk_ram_registrar_ok(&s->blk_ram_registrar)) { + flags |=3D BDRV_REQ_REGISTERED_BUF; + } + if (is_write) { - blk_aio_pwritev(blk, sector_num << BDRV_SECTOR_BITS, qiov, 0, - virtio_blk_rw_complete, mrb->reqs[start]); + blk_aio_pwritev(blk, sector_num << BDRV_SECTOR_BITS, qiov, + flags, virtio_blk_rw_complete, + mrb->reqs[start]); } else { - blk_aio_preadv(blk, sector_num << BDRV_SECTOR_BITS, qiov, 0, - virtio_blk_rw_complete, mrb->reqs[start]); + blk_aio_preadv(blk, sector_num << BDRV_SECTOR_BITS, qiov, + flags, virtio_blk_rw_complete, + mrb->reqs[start]); } } =20 @@ -447,14 +456,14 @@ static int multireq_compare(const void *a, const void= *b) } } =20 -static void virtio_blk_submit_multireq(BlockBackend *blk, MultiReqBuffer *= mrb) +static void virtio_blk_submit_multireq(VirtIOBlock *s, MultiReqBuffer *mrb) { int i =3D 0, start =3D 0, num_reqs =3D 0, niov =3D 0, nb_sectors =3D 0; uint32_t max_transfer; int64_t sector_num =3D 0; =20 if (mrb->num_reqs =3D=3D 1) { - submit_requests(blk, mrb, 0, 1, -1); + submit_requests(s, mrb, 0, 1, -1); mrb->num_reqs =3D 0; return; } @@ -474,11 +483,11 @@ static void virtio_blk_submit_multireq(BlockBackend *= blk, MultiReqBuffer *mrb) * 3. merge would exceed maximum transfer length of backend de= vice */ if (sector_num + nb_sectors !=3D req->sector_num || - niov > blk_get_max_iov(blk) - req->qiov.niov || + niov > blk_get_max_iov(s->blk) - req->qiov.niov || req->qiov.size > max_transfer || nb_sectors > (max_transfer - req->qiov.size) / BDRV_SECTOR_SIZE) { - submit_requests(blk, mrb, start, num_reqs, niov); + submit_requests(s, mrb, start, num_reqs, niov); num_reqs =3D 0; } } @@ -494,7 +503,7 @@ static void virtio_blk_submit_multireq(BlockBackend *bl= k, MultiReqBuffer *mrb) num_reqs++; } =20 - submit_requests(blk, mrb, start, num_reqs, niov); + submit_requests(s, mrb, start, num_reqs, niov); mrb->num_reqs =3D 0; } =20 @@ -509,7 +518,7 @@ static void virtio_blk_handle_flush(VirtIOBlockReq *req= , MultiReqBuffer *mrb) * Make sure all outstanding writes are posted to the backing device. */ if (mrb->is_write && mrb->num_reqs > 0) { - virtio_blk_submit_multireq(s->blk, mrb); + virtio_blk_submit_multireq(s, mrb); } blk_aio_flush(s->blk, virtio_blk_flush_complete, req); } @@ -689,7 +698,7 @@ static int virtio_blk_handle_request(VirtIOBlockReq *re= q, MultiReqBuffer *mrb) if (mrb->num_reqs > 0 && (mrb->num_reqs =3D=3D VIRTIO_BLK_MAX_MERG= E_REQS || is_write !=3D mrb->is_write || !s->conf.request_merging)) { - virtio_blk_submit_multireq(s->blk, mrb); + virtio_blk_submit_multireq(s, mrb); } =20 assert(mrb->num_reqs < VIRTIO_BLK_MAX_MERGE_REQS); @@ -796,7 +805,7 @@ void virtio_blk_handle_vq(VirtIOBlock *s, VirtQueue *vq) } while (!virtio_queue_empty(vq)); =20 if (mrb.num_reqs) { - virtio_blk_submit_multireq(s->blk, &mrb); + virtio_blk_submit_multireq(s, &mrb); } =20 blk_io_unplug(s->blk); @@ -845,7 +854,7 @@ void virtio_blk_process_queued_requests(VirtIOBlock *s,= bool is_bh) } =20 if (mrb.num_reqs) { - virtio_blk_submit_multireq(s->blk, &mrb); + virtio_blk_submit_multireq(s, &mrb); } if (is_bh) { blk_dec_in_flight(s->conf.conf.blk); @@ -1227,6 +1236,7 @@ static void virtio_blk_device_realize(DeviceState *de= v, Error **errp) } =20 s->change =3D qemu_add_vm_change_state_handler(virtio_blk_dma_restart_= cb, s); + blk_ram_registrar_init(&s->blk_ram_registrar, s->blk); blk_set_dev_ops(s->blk, &virtio_block_ops, s); =20 blk_iostatus_enable(s->blk); @@ -1252,6 +1262,7 @@ static void virtio_blk_device_unrealize(DeviceState *= dev) virtio_del_queue(vdev, i); } qemu_coroutine_dec_pool_size(conf->num_queues * conf->queue_size / 2); + blk_ram_registrar_destroy(&s->blk_ram_registrar); qemu_del_vm_change_state_handler(s->change); blockdev_mark_auto_del(s->blk); virtio_cleanup(vdev); --=20 2.37.3