From nobody Mon Sep 22 02:22:19 2025 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=quarantine dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1757527850; cv=none; d=zohomail.com; s=zohoarc; b=j/4uCKEuuwVdej4BohOl7XCSFoLJ4KAyGnCwNZb6qafcAEBnGKXn2cCQ9BpnK82uN/JRonk7HV6BjNTG1EWQD3WHRONVg5O6zGHPESk90Sf3+wBZjUm5KVz68Qtvw/0Dpr7HwIsfXhaOtf/jCam+7uo4pqYzX5egq3r99qe9Z4s= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1757527850; h=Content-Transfer-Encoding:Cc:Cc:Date:Date:From:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:Subject:To:To:Message-Id:Reply-To; bh=Uq/ffucX0LrsPavRUjZZ8ksU3DGad+yaH3VgAUHZsZo=; b=kkm0zPXKdc/6A2VfWHp9IBPEENXST+cFdLi6JWo0oMtOTuqciKAmSHzFiF1c05fFoKkvv8XKK+XndUb+4T/FB7lMNNdHhDk1cUEgYHcelMI710+4yoAEdKuHzppaWwGgA9Pfb/rOsaHr/qGtouEI4S0lZVgiSp0LYbtZo9/yJ/A= 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=quarantine dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 175752785019212.069009192001545; Wed, 10 Sep 2025 11:10:50 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1uwPDc-00043Y-VN; Wed, 10 Sep 2025 14:06:58 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1uwPDX-0003gi-69 for qemu-devel@nongnu.org; Wed, 10 Sep 2025 14:06:51 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.129.124]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1uwPDQ-000538-RB for qemu-devel@nongnu.org; Wed, 10 Sep 2025 14:06:50 -0400 Received: from mx-prod-mc-02.mail-002.prod.us-west-2.aws.redhat.com (ec2-54-186-198-63.us-west-2.compute.amazonaws.com [54.186.198.63]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-275-xmhrZunnO8WCF9k4-hZkpQ-1; Wed, 10 Sep 2025 14:06:40 -0400 Received: from mx-prod-int-03.mail-002.prod.us-west-2.aws.redhat.com (mx-prod-int-03.mail-002.prod.us-west-2.aws.redhat.com [10.30.177.12]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by mx-prod-mc-02.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTPS id 850DC195608D; Wed, 10 Sep 2025 18:06:39 +0000 (UTC) Received: from localhost (unknown [10.2.17.66]) by mx-prod-int-03.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTP id AA66319560AB; Wed, 10 Sep 2025 18:06:38 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1757527601; 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=Uq/ffucX0LrsPavRUjZZ8ksU3DGad+yaH3VgAUHZsZo=; b=fSKjbcwwn41cSTCRnBzJ6qlXFtFq9LHLG6tl653IFKzKSvTfkwN7rW1O8khDcFDLUh53Zq wGevrc95sB+hTNFwX10D1g0A9CwgAhLvxBUvFj6PYOM7jqGN4CrDjg/qq2BcWJgGW/tIRd vZ5fCTzhylr+H2NRzoUVUzEW2ytMFfo= X-MC-Unique: xmhrZunnO8WCF9k4-hZkpQ-1 X-Mimecast-MFC-AGG-ID: xmhrZunnO8WCF9k4-hZkpQ_1757527599 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Aarushi Mehta , Fam Zheng , Stefano Garzarella , Hanna Czenczek , eblake@redhat.com, Stefan Hajnoczi , Kevin Wolf , qemu-block@nongnu.org, hibriansong@gmail.com, Stefan Weil , Paolo Bonzini Subject: [PATCH v4 01/12] aio-posix: fix race between io_uring CQE and AioHandler deletion Date: Wed, 10 Sep 2025 13:56:52 -0400 Message-ID: <20250910175703.374499-2-stefanha@redhat.com> In-Reply-To: <20250910175703.374499-1-stefanha@redhat.com> References: <20250910175703.374499-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.0 on 10.30.177.12 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: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_NONE=-0.0001, RCVD_IN_MSPIKE_H5=0.001, RCVD_IN_MSPIKE_WL=0.001, RCVD_IN_VALIDITY_RPBL_BLOCKED=0.001, RCVD_IN_VALIDITY_SAFE_BLOCKED=0.001, SPF_HELO_PASS=-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-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1757527852728116601 Content-Type: text/plain; charset="utf-8" When an AioHandler is enqueued on ctx->submit_list for removal, the fill_sq_ring() function will submit an io_uring POLL_REMOVE operation to cancel the in-flight POLL_ADD operation. There is a race when another thread enqueues an AioHandler for deletion on ctx->submit_list when the POLL_ADD CQE has already appeared. In that case POLL_REMOVE is unnecessary. The code already handled this, but forgot that the AioHandler itself is still on ctx->submit_list when the POLL_ADD CQE is being processed. It's unsafe to delete the AioHandler at that point in time (use-after-free). Solve this problem by keeping the AioHandler alive but setting a flag so that it will be deleted by fill_sq_ring() when it runs. Signed-off-by: Stefan Hajnoczi Reviewed-by: Eric Blake --- util/fdmon-io_uring.c | 33 ++++++++++++++++++++++++++------- 1 file changed, 26 insertions(+), 7 deletions(-) diff --git a/util/fdmon-io_uring.c b/util/fdmon-io_uring.c index b0d68bdc44..ad89160f31 100644 --- a/util/fdmon-io_uring.c +++ b/util/fdmon-io_uring.c @@ -52,9 +52,10 @@ enum { FDMON_IO_URING_ENTRIES =3D 128, /* sq/cq ring size */ =20 /* AioHandler::flags */ - FDMON_IO_URING_PENDING =3D (1 << 0), - FDMON_IO_URING_ADD =3D (1 << 1), - FDMON_IO_URING_REMOVE =3D (1 << 2), + FDMON_IO_URING_PENDING =3D (1 << 0), + FDMON_IO_URING_ADD =3D (1 << 1), + FDMON_IO_URING_REMOVE =3D (1 << 2), + FDMON_IO_URING_DELETE_AIO_HANDLER =3D (1 << 3), }; =20 static inline int poll_events_from_pfd(int pfd_events) @@ -218,6 +219,16 @@ static void fill_sq_ring(AioContext *ctx) if (flags & FDMON_IO_URING_REMOVE) { add_poll_remove_sqe(ctx, node); } + if (flags & FDMON_IO_URING_DELETE_AIO_HANDLER) { + /* + * process_cqe() sets this flag after ADD and REMOVE have been + * cleared. They cannot be set again, so they must be clear. + */ + assert(!(flags & FDMON_IO_URING_ADD)); + assert(!(flags & FDMON_IO_URING_REMOVE)); + + QLIST_INSERT_HEAD_RCU(&ctx->deleted_aio_handlers, node, node_d= eleted); + } } } =20 @@ -241,7 +252,12 @@ static bool process_cqe(AioContext *ctx, */ flags =3D qatomic_fetch_and(&node->flags, ~FDMON_IO_URING_REMOVE); if (flags & FDMON_IO_URING_REMOVE) { - QLIST_INSERT_HEAD_RCU(&ctx->deleted_aio_handlers, node, node_delet= ed); + if (flags & FDMON_IO_URING_PENDING) { + /* Still on ctx->submit_list, defer deletion until fill_sq_rin= g() */ + qatomic_or(&node->flags, FDMON_IO_URING_DELETE_AIO_HANDLER); + } else { + QLIST_INSERT_HEAD_RCU(&ctx->deleted_aio_handlers, node, node_d= eleted); + } return false; } =20 @@ -347,10 +363,13 @@ void fdmon_io_uring_destroy(AioContext *ctx) unsigned flags =3D qatomic_fetch_and(&node->flags, ~(FDMON_IO_URING_PENDING | FDMON_IO_URING_ADD | - FDMON_IO_URING_REMOVE)); + FDMON_IO_URING_REMOVE | + FDMON_IO_URING_DELETE_AIO_HANDLER)); =20 - if (flags & FDMON_IO_URING_REMOVE) { - QLIST_INSERT_HEAD_RCU(&ctx->deleted_aio_handlers, node, no= de_deleted); + if ((flags & FDMON_IO_URING_REMOVE) || + (flags & FDMON_IO_URING_DELETE_AIO_HANDLER)) { + QLIST_INSERT_HEAD_RCU(&ctx->deleted_aio_handlers, + node, node_deleted); } =20 QSLIST_REMOVE_HEAD_RCU(&ctx->submit_list, node_submitted); --=20 2.51.0 From nobody Mon Sep 22 02:22:19 2025 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=quarantine dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1757527861; cv=none; d=zohomail.com; s=zohoarc; b=GBTF85R3ag34Rt8IZZigDJY4X43dge0UPN8dhVoIFTuW39ZdkYsFdmX5x3XjInRlU2qUmy3fzD4FPiC9778KkP4TWmxYJ0GCrdKKZ6M1NsBogJprlsuzo69X7dtoSznQffvfK58Ker4C7GI8RGimM1Pr3rUupdFnIZiVprg676E= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1757527861; h=Content-Transfer-Encoding:Cc:Cc:Date:Date:From:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:Subject:To:To:Message-Id:Reply-To; bh=MGnG+XdtmTxntikOC8JogL74daSKQSHrsTNR1uFQlNA=; b=XH0/L2LyfXIHuzQV/JVP6OuUpYle+CM5P3XZGHRpEfuifwXak7fpbXFUN9M9o7QlTIda1Hlqze/1BMRd5LM6vu/mMcnvRNXFG2WVV9/3mzTPGSgHufL0Te6c19j7plJyul55ZR4MPL06Y0Aobi7kjyDybT9IqX14o8Z338FavSo= 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=quarantine dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1757527861433140.8227852636834; Wed, 10 Sep 2025 11:11:01 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1uwPE4-0004vi-MB; Wed, 10 Sep 2025 14:07:28 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1uwPDe-0004L6-Sj for qemu-devel@nongnu.org; Wed, 10 Sep 2025 14:06:58 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.133.124]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1uwPDT-00053a-JJ for qemu-devel@nongnu.org; Wed, 10 Sep 2025 14:06:58 -0400 Received: from mx-prod-mc-06.mail-002.prod.us-west-2.aws.redhat.com (ec2-35-165-154-97.us-west-2.compute.amazonaws.com [35.165.154.97]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-138-yc7g6dkZMZyexE5Im2m2Pw-1; Wed, 10 Sep 2025 14:06:42 -0400 Received: from mx-prod-int-06.mail-002.prod.us-west-2.aws.redhat.com (mx-prod-int-06.mail-002.prod.us-west-2.aws.redhat.com [10.30.177.93]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by mx-prod-mc-06.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTPS id 7F9C91800296; Wed, 10 Sep 2025 18:06:41 +0000 (UTC) Received: from localhost (unknown [10.2.17.66]) by mx-prod-int-06.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTP id E85E21800446; Wed, 10 Sep 2025 18:06:40 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1757527606; 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=MGnG+XdtmTxntikOC8JogL74daSKQSHrsTNR1uFQlNA=; b=WwhlKraC4SW/cRvYdpnAGG5/jjxtFPgSQDT6VbkZzkL4SgxHbhR9YVhi2qdppYGlqjdFAv 6n1hRoMQvt1f84PMBmnZS7fCeLNyhy8CX30hjIuk2Q7/eICdAFOJcRKs0k6y3MSg4EtR3G EttXXJkMOer5GV6Vcpg3Ofetbh83cbg= X-MC-Unique: yc7g6dkZMZyexE5Im2m2Pw-1 X-Mimecast-MFC-AGG-ID: yc7g6dkZMZyexE5Im2m2Pw_1757527601 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Aarushi Mehta , Fam Zheng , Stefano Garzarella , Hanna Czenczek , eblake@redhat.com, Stefan Hajnoczi , Kevin Wolf , qemu-block@nongnu.org, hibriansong@gmail.com, Stefan Weil , Paolo Bonzini , Chao Gao Subject: [PATCH v4 02/12] aio-posix: keep polling enabled with fdmon-io_uring.c Date: Wed, 10 Sep 2025 13:56:53 -0400 Message-ID: <20250910175703.374499-3-stefanha@redhat.com> In-Reply-To: <20250910175703.374499-1-stefanha@redhat.com> References: <20250910175703.374499-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.4.1 on 10.30.177.93 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, RCVD_IN_VALIDITY_RPBL_BLOCKED=0.001, RCVD_IN_VALIDITY_SAFE_BLOCKED=0.001, SPF_HELO_PASS=-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-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1757527862940116600 Content-Type: text/plain; charset="utf-8" Commit 816a430c517e ("util/aio: Defer disabling poll mode as long as possible") kept polling enabled when the event loop timeout is 0. Since there is no timeout the event loop will continue immediately and the overhead of disabling and re-enabling polling can be avoided. fdmon-io_uring.c is unable to take advantage of this optimization because its ->need_wait() function returns true whenever there are new io_uring SQEs to submit: if (timeout || ctx->fdmon_ops->need_wait(ctx)) { ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Polling will be disabled even when timeout =3D=3D 0. Extend the optimization to handle the case when need_wait() returns true and timeout =3D=3D 0. Cc: Chao Gao Signed-off-by: Stefan Hajnoczi Reviewed-by: Eric Blake --- util/aio-posix.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/util/aio-posix.c b/util/aio-posix.c index 2e0a5dadc4..824fdc34cc 100644 --- a/util/aio-posix.c +++ b/util/aio-posix.c @@ -559,7 +559,14 @@ static bool run_poll_handlers(AioContext *ctx, AioHand= lerList *ready_list, elapsed_time =3D qemu_clock_get_ns(QEMU_CLOCK_REALTIME) - start_ti= me; max_ns =3D qemu_soonest_timeout(*timeout, max_ns); assert(!(max_ns && progress)); - } while (elapsed_time < max_ns && !ctx->fdmon_ops->need_wait(ctx)); + + if (ctx->fdmon_ops->need_wait(ctx)) { + if (fdmon_supports_polling(ctx)) { + *timeout =3D 0; /* stay in polling mode */ + } + break; + } + } while (elapsed_time < max_ns); =20 if (remove_idle_poll_handlers(ctx, ready_list, start_time + elapsed_time)) { @@ -722,7 +729,7 @@ bool aio_poll(AioContext *ctx, bool blocking) * up IO threads when some work becomes pending. It is essential to * avoid hangs or unnecessary latency. */ - if (poll_set_started(ctx, &ready_list, false)) { + if (timeout && poll_set_started(ctx, &ready_list, false)) { timeout =3D 0; progress =3D true; } --=20 2.51.0 From nobody Mon Sep 22 02:22:19 2025 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=quarantine dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1757528135; cv=none; d=zohomail.com; s=zohoarc; b=Tkg/xHQP5m03qQynoTbiWBvhsKrXTYvoc8G2SYdSNi+l/84lXicbb5WyVMFhMvXvMS7eSqm8kXvw6R+XbF14erBYGTsf/D5u09bpqS6cxB7+kGejS9iopXZhzWCrCQeCxqY9FSBMYYCmFkYAYmnOD1TWyJ3x4lQrsnas/8NicyY= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1757528135; h=Content-Transfer-Encoding:Cc:Cc:Date:Date:From:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:Subject:To:To:Message-Id:Reply-To; bh=PZyG2gVhf9ANlECoA63cnmRtA9cmX8TXqtAncsbQnvQ=; b=fjWTMLNQxdGdJYch7TFnoEXvZOhiNwQKW8WWpgoEHB10DQn4eDNKmEBTLXSsAOLpO+7CB1k8+qLRIPBwc3IxTsW/JW+W6Vy3/13O+MZEFwGQFn1XpuZA3BHUW2eBbReCKOS3fF3xW/DdNL8iG5AqjLxNyLIueicSq7tC46mupAc= 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=quarantine dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1757528135643230.33679254272056; Wed, 10 Sep 2025 11:15:35 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1uwPDd-0004Bo-1y; Wed, 10 Sep 2025 14:06:58 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1uwPDY-0003no-OF for qemu-devel@nongnu.org; Wed, 10 Sep 2025 14:06:53 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.129.124]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1uwPDV-00053q-3l for qemu-devel@nongnu.org; Wed, 10 Sep 2025 14:06:52 -0400 Received: from mx-prod-mc-06.mail-002.prod.us-west-2.aws.redhat.com (ec2-35-165-154-97.us-west-2.compute.amazonaws.com [35.165.154.97]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-311-cIuGWy0yMdqxdMUvOq1HlQ-1; Wed, 10 Sep 2025 14:06:46 -0400 Received: from mx-prod-int-05.mail-002.prod.us-west-2.aws.redhat.com (mx-prod-int-05.mail-002.prod.us-west-2.aws.redhat.com [10.30.177.17]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by mx-prod-mc-06.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTPS id 8DBC41800378; Wed, 10 Sep 2025 18:06:43 +0000 (UTC) Received: from localhost (unknown [10.2.17.66]) by mx-prod-int-05.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTP id 191991956095; Wed, 10 Sep 2025 18:06:42 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1757527607; 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=PZyG2gVhf9ANlECoA63cnmRtA9cmX8TXqtAncsbQnvQ=; b=WdSjpEUdgN4BC+6Xqnf7IFHfhAXZ8bZ9IbjGafkcpXl9ynnm2z6n03dD8ss9KnmaPKlvJq 76mJaDi4+AbX9pF5GU0zhtjhWmmFP0SOMBqjcolS0dMLtUYpaBSVESPkEpnb69hvjiQlTZ e7wgVVgmq1vVMUArxhj96YgF4MAXOsg= X-MC-Unique: cIuGWy0yMdqxdMUvOq1HlQ-1 X-Mimecast-MFC-AGG-ID: cIuGWy0yMdqxdMUvOq1HlQ_1757527603 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Aarushi Mehta , Fam Zheng , Stefano Garzarella , Hanna Czenczek , eblake@redhat.com, Stefan Hajnoczi , Kevin Wolf , qemu-block@nongnu.org, hibriansong@gmail.com, Stefan Weil , Paolo Bonzini Subject: [PATCH v4 03/12] tests/unit: skip test-nested-aio-poll with io_uring Date: Wed, 10 Sep 2025 13:56:54 -0400 Message-ID: <20250910175703.374499-4-stefanha@redhat.com> In-Reply-To: <20250910175703.374499-1-stefanha@redhat.com> References: <20250910175703.374499-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.0 on 10.30.177.17 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: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_NONE=-0.0001, RCVD_IN_MSPIKE_H5=0.001, RCVD_IN_MSPIKE_WL=0.001, RCVD_IN_VALIDITY_RPBL_BLOCKED=0.001, RCVD_IN_VALIDITY_SAFE_BLOCKED=0.001, SPF_HELO_PASS=-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-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1757528136335116600 Content-Type: text/plain; charset="utf-8" test-nested-aio-poll relies on internal details of how fdmon-poll.c handles AioContext polling. Skip it when other fdmon implementations are in use. Note that this test is only built on POSIX systems so it is safe to include "util/aio-posix.h". Signed-off-by: Stefan Hajnoczi Reviewed-by: Eric Blake --- tests/unit/test-nested-aio-poll.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/tests/unit/test-nested-aio-poll.c b/tests/unit/test-nested-aio= -poll.c index d8fd92c43b..45484e745b 100644 --- a/tests/unit/test-nested-aio-poll.c +++ b/tests/unit/test-nested-aio-poll.c @@ -15,6 +15,7 @@ #include "qemu/osdep.h" #include "block/aio.h" #include "qapi/error.h" +#include "util/aio-posix.h" =20 typedef struct { AioContext *ctx; @@ -71,6 +72,12 @@ static void test(void) .ctx =3D aio_context_new(&error_abort), }; =20 + if (td.ctx->fdmon_ops !=3D &fdmon_poll_ops) { + /* This test is tied to fdmon-poll.c */ + g_test_skip("fdmon_poll_ops not in use"); + return; + } + qemu_set_current_aio_context(td.ctx); =20 /* Enable polling */ --=20 2.51.0 From nobody Mon Sep 22 02:22:19 2025 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=quarantine dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1757527930; cv=none; d=zohomail.com; s=zohoarc; b=RO97EH8Fq/L+Z9pRmTNqQN1amhlWEWmOuvCdbQjNGvMzwXd5SYry9dGnrlJgYDQ2UoC1/7g3y4+iUvVJ4gt7QpEdXfltLkpdznWjRcasGDLk9fyXHM6hZd6R8T0DlZTgKMJbGvJOaN832lNItgps4xkIyGY0pZYoW3vYvfBJOSM= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1757527930; h=Content-Transfer-Encoding:Cc:Cc:Date:Date:From:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:Subject:To:To:Message-Id:Reply-To; bh=P7UMvY5hCdCMlioYSGwzZGHNYcJXjePh9mXoYfMaIVk=; b=GlgpOGoqyaxNPUeL2WAnnOZyCVT6ufjdMtnHJTxMd5EbJCn36LjpM1HZ3iUgBfOYT5dVkcZzmemHUnjdXZRpNyuVoo+pGJZYJJIx6BYmY8B9HVx7MzkUYSCyz8FJZHfQLoPFreIhY8BxdPjQXu58irIIzslBM5zXaMifCNnAMFE= 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=quarantine dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1757527930805151.14813316170046; Wed, 10 Sep 2025 11:12:10 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1uwPFx-0006RS-5u; Wed, 10 Sep 2025 14:09:25 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1uwPDh-0004UH-4T for qemu-devel@nongnu.org; Wed, 10 Sep 2025 14:07:01 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.129.124]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1uwPDZ-00054U-Cz for qemu-devel@nongnu.org; Wed, 10 Sep 2025 14:07:00 -0400 Received: from mx-prod-mc-03.mail-002.prod.us-west-2.aws.redhat.com (ec2-54-186-198-63.us-west-2.compute.amazonaws.com [54.186.198.63]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-391-79UqUCPlPomQD1ikxdeqCA-1; Wed, 10 Sep 2025 14:06:48 -0400 Received: from mx-prod-int-03.mail-002.prod.us-west-2.aws.redhat.com (mx-prod-int-03.mail-002.prod.us-west-2.aws.redhat.com [10.30.177.12]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by mx-prod-mc-03.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTPS id 914B119541B2; Wed, 10 Sep 2025 18:06:45 +0000 (UTC) Received: from localhost (unknown [10.2.17.66]) by mx-prod-int-03.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTP id 251ED19560B8; Wed, 10 Sep 2025 18:06:44 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1757527611; 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=P7UMvY5hCdCMlioYSGwzZGHNYcJXjePh9mXoYfMaIVk=; b=Q77N02Fzt2xu80vlz5OCxcdROk9zI9KWgFjWJb7ynLGB/EXeKE25GuzQWhehMfx42EiZ8/ 2SDbt4ATznqlIkQYJ9OMAe4dG3lOn5jf/NzuB4r/Tsc/tjczssKCYaZgQB9ho1I176FWpG IL1osTZ+0M4HGCYpoWUJWOyNMl40FmQ= X-MC-Unique: 79UqUCPlPomQD1ikxdeqCA-1 X-Mimecast-MFC-AGG-ID: 79UqUCPlPomQD1ikxdeqCA_1757527605 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Aarushi Mehta , Fam Zheng , Stefano Garzarella , Hanna Czenczek , eblake@redhat.com, Stefan Hajnoczi , Kevin Wolf , qemu-block@nongnu.org, hibriansong@gmail.com, Stefan Weil , Paolo Bonzini Subject: [PATCH v4 04/12] aio-posix: integrate fdmon into glib event loop Date: Wed, 10 Sep 2025 13:56:55 -0400 Message-ID: <20250910175703.374499-5-stefanha@redhat.com> In-Reply-To: <20250910175703.374499-1-stefanha@redhat.com> References: <20250910175703.374499-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.0 on 10.30.177.12 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: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_NONE=-0.0001, RCVD_IN_MSPIKE_H5=0.001, RCVD_IN_MSPIKE_WL=0.001, RCVD_IN_VALIDITY_RPBL_BLOCKED=0.001, RCVD_IN_VALIDITY_SAFE_BLOCKED=0.001, SPF_HELO_PASS=-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-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1757527932113116600 Content-Type: text/plain; charset="utf-8" AioContext's glib integration only supports ppoll(2) file descriptor monitoring. epoll(7) and io_uring(7) disable themselves and switch back to ppoll(2) when the glib event loop is used. The main loop thread cannot use epoll(7) or io_uring(7) because it always uses the glib event loop. Future QEMU features may require io_uring(7). One example is uring_cmd support in FUSE exports. Each feature could create its own io_uring(7) context and integrate it into the event loop, but this is inefficient due to extra syscalls. It would be more efficient to reuse the AioContext's existing fdmon-io_uring.c io_uring(7) context because fdmon-io_uring.c will already be active on systems where Linux io_uring is available. In order to keep fdmon-io_uring.c's AioContext operational even when the glib event loop is used, extend FDMonOps with an API similar to GSourceFuncs so that file descriptor monitoring can integrate into the glib event loop. A quick summary of the GSourceFuncs API: - prepare() is called each event loop iteration before waiting for file descriptors and timers. - check() is called to determine whether events are ready to be dispatched after waiting. - dispatch() is called to process events. More details here: https://docs.gtk.org/glib/struct.SourceFuncs.html Move the ppoll(2)-specific code from aio-posix.c into fdmon-poll.c and also implement epoll(7)- and io_uring(7)-specific file descriptor monitoring code for glib event loops. Note that it's still faster to use aio_poll() rather than the glib event loop since glib waits for file descriptor activity with ppoll(2) and does not support adaptive polling. But at least epoll(7) and io_uring(7) now work in glib event loops. Splitting this into multiple commits without temporarily breaking AioContext proved difficult so this commit makes all the changes. The next commit will remove the aio_context_use_g_source() API because it is no longer needed. Signed-off-by: Stefan Hajnoczi Reviewed-by: Eric Blake --- include/block/aio.h | 36 ++++++++++++++++++ util/aio-posix.h | 5 +++ tests/unit/test-aio.c | 7 +++- util/aio-posix.c | 69 ++++++++------------------------- util/fdmon-epoll.c | 52 ++++++++++++++++++++++--- util/fdmon-io_uring.c | 44 +++++++++++++++++++++- util/fdmon-poll.c | 88 ++++++++++++++++++++++++++++++++++++++++++- 7 files changed, 239 insertions(+), 62 deletions(-) diff --git a/include/block/aio.h b/include/block/aio.h index 99ff48420b..39ed86d14d 100644 --- a/include/block/aio.h +++ b/include/block/aio.h @@ -106,6 +106,38 @@ typedef struct { * Returns: true if ->wait() should be called, false otherwise. */ bool (*need_wait)(AioContext *ctx); + + /* + * gsource_prepare: + * @ctx: the AioContext + * + * Prepare for the glib event loop to wait for events instead of the u= sual + * ->wait() call. See glib's GSourceFuncs->prepare(). + */ + void (*gsource_prepare)(AioContext *ctx); + + /* + * gsource_check: + * @ctx: the AioContext + * + * Called by the glib event loop from glib's GSourceFuncs->check() aft= er + * waiting for events. + * + * Returns: true when ready to be dispatched. + */ + bool (*gsource_check)(AioContext *ctx); + + /* + * gsource_dispatch: + * @ctx: the AioContext + * @ready_list: list for handlers that become ready + * + * Place ready AioHandlers on ready_list. Called as part of the glib e= vent + * loop from glib's GSourceFuncs->dispatch(). + * + * Called with list_lock incremented. + */ + void (*gsource_dispatch)(AioContext *ctx, AioHandlerList *ready_list); } FDMonOps; =20 /* @@ -222,6 +254,7 @@ struct AioContext { /* State for file descriptor monitoring using Linux io_uring */ struct io_uring fdmon_io_uring; AioHandlerSList submit_list; + gpointer io_uring_fd_tag; #endif =20 /* TimerLists for calling timers - one per clock type. Has its own @@ -254,6 +287,9 @@ struct AioContext { /* epoll(7) state used when built with CONFIG_EPOLL */ int epollfd; =20 + /* The GSource unix fd tag for epollfd */ + gpointer epollfd_tag; + const FDMonOps *fdmon_ops; }; =20 diff --git a/util/aio-posix.h b/util/aio-posix.h index 82a0201ea4..f9994ed79e 100644 --- a/util/aio-posix.h +++ b/util/aio-posix.h @@ -47,9 +47,14 @@ void aio_add_ready_handler(AioHandlerList *ready_list, A= ioHandler *node, =20 extern const FDMonOps fdmon_poll_ops; =20 +/* Switch back to poll(2). list_lock must be held. */ +void fdmon_poll_downgrade(AioContext *ctx); + #ifdef CONFIG_EPOLL_CREATE1 bool fdmon_epoll_try_upgrade(AioContext *ctx, unsigned npfd); void fdmon_epoll_setup(AioContext *ctx); + +/* list_lock must be held */ void fdmon_epoll_disable(AioContext *ctx); #else static inline bool fdmon_epoll_try_upgrade(AioContext *ctx, unsigned npfd) diff --git a/tests/unit/test-aio.c b/tests/unit/test-aio.c index e77d86be87..010d65b79a 100644 --- a/tests/unit/test-aio.c +++ b/tests/unit/test-aio.c @@ -527,7 +527,12 @@ static void test_source_bh_delete_from_cb(void) g_assert_cmpint(data1.n, =3D=3D, data1.max); g_assert(data1.bh =3D=3D NULL); =20 - assert(g_main_context_iteration(NULL, false)); + /* + * There may be up to one more iteration due to the aio_notify + * EventNotifier. + */ + g_main_context_iteration(NULL, false); + assert(!g_main_context_iteration(NULL, false)); } =20 diff --git a/util/aio-posix.c b/util/aio-posix.c index 824fdc34cc..9de05ee7e8 100644 --- a/util/aio-posix.c +++ b/util/aio-posix.c @@ -70,15 +70,6 @@ static AioHandler *find_aio_handler(AioContext *ctx, int= fd) =20 static bool aio_remove_fd_handler(AioContext *ctx, AioHandler *node) { - /* If the GSource is in the process of being destroyed then - * g_source_remove_poll() causes an assertion failure. Skip - * removal in that case, because glib cleans up its state during - * destruction anyway. - */ - if (!g_source_is_destroyed(&ctx->source)) { - g_source_remove_poll(&ctx->source, &node->pfd); - } - node->pfd.revents =3D 0; node->poll_ready =3D false; =20 @@ -153,7 +144,6 @@ void aio_set_fd_handler(AioContext *ctx, } else { new_node->pfd =3D node->pfd; } - g_source_add_poll(&ctx->source, &new_node->pfd); =20 new_node->pfd.events =3D (io_read ? G_IO_IN | G_IO_HUP | G_IO_ERR = : 0); new_node->pfd.events |=3D (io_write ? G_IO_OUT | G_IO_ERR : 0); @@ -267,37 +257,13 @@ bool aio_prepare(AioContext *ctx) poll_set_started(ctx, &ready_list, false); /* TODO what to do with this list? */ =20 + ctx->fdmon_ops->gsource_prepare(ctx); return false; } =20 bool aio_pending(AioContext *ctx) { - AioHandler *node; - bool result =3D false; - - /* - * We have to walk very carefully in case aio_set_fd_handler is - * called while we're walking. - */ - qemu_lockcnt_inc(&ctx->list_lock); - - QLIST_FOREACH_RCU(node, &ctx->aio_handlers, node) { - int revents; - - /* TODO should this check poll ready? */ - revents =3D node->pfd.revents & node->pfd.events; - if (revents & (G_IO_IN | G_IO_HUP | G_IO_ERR) && node->io_read) { - result =3D true; - break; - } - if (revents & (G_IO_OUT | G_IO_ERR) && node->io_write) { - result =3D true; - break; - } - } - qemu_lockcnt_dec(&ctx->list_lock); - - return result; + return ctx->fdmon_ops->gsource_check(ctx); } =20 static void aio_free_deleted_handlers(AioContext *ctx) @@ -390,10 +356,6 @@ static bool aio_dispatch_handler(AioContext *ctx, AioH= andler *node) return progress; } =20 -/* - * If we have a list of ready handlers then this is more efficient than - * scanning all handlers with aio_dispatch_handlers(). - */ static bool aio_dispatch_ready_handlers(AioContext *ctx, AioHandlerList *ready_list, int64_t block_ns) @@ -417,24 +379,18 @@ static bool aio_dispatch_ready_handlers(AioContext *c= tx, return progress; } =20 -/* Slower than aio_dispatch_ready_handlers() but only used via glib */ -static bool aio_dispatch_handlers(AioContext *ctx) -{ - AioHandler *node, *tmp; - bool progress =3D false; - - QLIST_FOREACH_SAFE_RCU(node, &ctx->aio_handlers, node, tmp) { - progress =3D aio_dispatch_handler(ctx, node) || progress; - } - - return progress; -} - void aio_dispatch(AioContext *ctx) { + AioHandlerList ready_list =3D QLIST_HEAD_INITIALIZER(ready_list); + qemu_lockcnt_inc(&ctx->list_lock); aio_bh_poll(ctx); - aio_dispatch_handlers(ctx); + + ctx->fdmon_ops->gsource_dispatch(ctx, &ready_list); + + /* block_ns is 0 because polling is disabled in the glib event loop */ + aio_dispatch_ready_handlers(ctx, &ready_list, 0); + aio_free_deleted_handlers(ctx); qemu_lockcnt_dec(&ctx->list_lock); =20 @@ -766,6 +722,7 @@ void aio_context_setup(AioContext *ctx) { ctx->fdmon_ops =3D &fdmon_poll_ops; ctx->epollfd =3D -1; + ctx->epollfd_tag =3D NULL; =20 /* Use the fastest fd monitoring implementation if available */ if (fdmon_io_uring_setup(ctx)) { @@ -778,7 +735,11 @@ void aio_context_setup(AioContext *ctx) void aio_context_destroy(AioContext *ctx) { fdmon_io_uring_destroy(ctx); + + qemu_lockcnt_lock(&ctx->list_lock); fdmon_epoll_disable(ctx); + qemu_lockcnt_unlock(&ctx->list_lock); + aio_free_deleted_handlers(ctx); } =20 diff --git a/util/fdmon-epoll.c b/util/fdmon-epoll.c index 9fb8800dde..0e78baa634 100644 --- a/util/fdmon-epoll.c +++ b/util/fdmon-epoll.c @@ -19,8 +19,12 @@ void fdmon_epoll_disable(AioContext *ctx) ctx->epollfd =3D -1; } =20 - /* Switch back */ - ctx->fdmon_ops =3D &fdmon_poll_ops; + if (ctx->epollfd_tag) { + g_source_remove_unix_fd(&ctx->source, ctx->epollfd_tag); + ctx->epollfd_tag =3D NULL; + } + + fdmon_poll_downgrade(ctx); } =20 static inline int epoll_events_from_pfd(int pfd_events) @@ -93,10 +97,47 @@ out: return ret; } =20 +static void fdmon_epoll_gsource_prepare(AioContext *ctx) +{ + /* Do nothing */ +} + +static bool fdmon_epoll_gsource_check(AioContext *ctx) +{ + return g_source_query_unix_fd(&ctx->source, ctx->epollfd_tag) & G_IO_I= N; +} + +static void fdmon_epoll_gsource_dispatch(AioContext *ctx, + AioHandlerList *ready_list) +{ + AioHandler *node; + int ret; + struct epoll_event events[128]; + + /* Collect events and process them */ + ret =3D epoll_wait(ctx->epollfd, events, ARRAY_SIZE(events), 0); + if (ret <=3D 0) { + return; + } + for (int i =3D 0; i < ret; i++) { + int ev =3D events[i].events; + int revents =3D (ev & EPOLLIN ? G_IO_IN : 0) | + (ev & EPOLLOUT ? G_IO_OUT : 0) | + (ev & EPOLLHUP ? G_IO_HUP : 0) | + (ev & EPOLLERR ? G_IO_ERR : 0); + + node =3D events[i].data.ptr; + aio_add_ready_handler(ready_list, node, revents); + } +} + static const FDMonOps fdmon_epoll_ops =3D { .update =3D fdmon_epoll_update, .wait =3D fdmon_epoll_wait, .need_wait =3D aio_poll_disabled, + .gsource_prepare =3D fdmon_epoll_gsource_prepare, + .gsource_check =3D fdmon_epoll_gsource_check, + .gsource_dispatch =3D fdmon_epoll_gsource_dispatch, }; =20 static bool fdmon_epoll_try_enable(AioContext *ctx) @@ -118,6 +159,8 @@ static bool fdmon_epoll_try_enable(AioContext *ctx) } =20 ctx->fdmon_ops =3D &fdmon_epoll_ops; + ctx->epollfd_tag =3D g_source_add_unix_fd(&ctx->source, ctx->epollfd, + G_IO_IN); return true; } =20 @@ -139,12 +182,11 @@ bool fdmon_epoll_try_upgrade(AioContext *ctx, unsigne= d npfd) } =20 ok =3D fdmon_epoll_try_enable(ctx); - - qemu_lockcnt_inc_and_unlock(&ctx->list_lock); - if (!ok) { fdmon_epoll_disable(ctx); } + + qemu_lockcnt_inc_and_unlock(&ctx->list_lock); return ok; } =20 diff --git a/util/fdmon-io_uring.c b/util/fdmon-io_uring.c index ad89160f31..4b6dc8903f 100644 --- a/util/fdmon-io_uring.c +++ b/util/fdmon-io_uring.c @@ -276,6 +276,11 @@ static int process_cq_ring(AioContext *ctx, AioHandler= List *ready_list) unsigned num_ready =3D 0; unsigned head; =20 + /* If the CQ overflowed then fetch CQEs with a syscall */ + if (io_uring_cq_has_overflow(ring)) { + io_uring_get_events(ring); + } + io_uring_for_each_cqe(ring, head, cqe) { if (process_cqe(ctx, ready_list, cqe)) { num_ready++; @@ -288,6 +293,30 @@ static int process_cq_ring(AioContext *ctx, AioHandler= List *ready_list) return num_ready; } =20 +/* This is where SQEs are submitted in the glib event loop */ +static void fdmon_io_uring_gsource_prepare(AioContext *ctx) +{ + fill_sq_ring(ctx); + if (io_uring_sq_ready(&ctx->fdmon_io_uring)) { + while (io_uring_submit(&ctx->fdmon_io_uring) =3D=3D -EINTR) { + /* Keep trying if syscall was interrupted */ + } + } +} + +static bool fdmon_io_uring_gsource_check(AioContext *ctx) +{ + gpointer tag =3D ctx->io_uring_fd_tag; + return g_source_query_unix_fd(&ctx->source, tag) & G_IO_IN; +} + +/* This is where CQEs are processed in the glib event loop */ +static void fdmon_io_uring_gsource_dispatch(AioContext *ctx, + AioHandlerList *ready_list) +{ + process_cq_ring(ctx, ready_list); +} + static int fdmon_io_uring_wait(AioContext *ctx, AioHandlerList *ready_list, int64_t timeout) { @@ -335,12 +364,17 @@ static const FDMonOps fdmon_io_uring_ops =3D { .update =3D fdmon_io_uring_update, .wait =3D fdmon_io_uring_wait, .need_wait =3D fdmon_io_uring_need_wait, + .gsource_prepare =3D fdmon_io_uring_gsource_prepare, + .gsource_check =3D fdmon_io_uring_gsource_check, + .gsource_dispatch =3D fdmon_io_uring_gsource_dispatch, }; =20 bool fdmon_io_uring_setup(AioContext *ctx) { int ret; =20 + ctx->io_uring_fd_tag =3D NULL; + ret =3D io_uring_queue_init(FDMON_IO_URING_ENTRIES, &ctx->fdmon_io_uri= ng, 0); if (ret !=3D 0) { return false; @@ -348,6 +382,9 @@ bool fdmon_io_uring_setup(AioContext *ctx) =20 QSLIST_INIT(&ctx->submit_list); ctx->fdmon_ops =3D &fdmon_io_uring_ops; + ctx->io_uring_fd_tag =3D g_source_add_unix_fd(&ctx->source, + ctx->fdmon_io_uring.ring_fd, G_IO_IN); + return true; } =20 @@ -375,6 +412,11 @@ void fdmon_io_uring_destroy(AioContext *ctx) QSLIST_REMOVE_HEAD_RCU(&ctx->submit_list, node_submitted); } =20 - ctx->fdmon_ops =3D &fdmon_poll_ops; + g_source_remove_unix_fd(&ctx->source, ctx->io_uring_fd_tag); + ctx->io_uring_fd_tag =3D NULL; + + qemu_lockcnt_lock(&ctx->list_lock); + fdmon_poll_downgrade(ctx); + qemu_lockcnt_unlock(&ctx->list_lock); } } diff --git a/util/fdmon-poll.c b/util/fdmon-poll.c index 17df917cf9..f91dc54944 100644 --- a/util/fdmon-poll.c +++ b/util/fdmon-poll.c @@ -72,6 +72,11 @@ static int fdmon_poll_wait(AioContext *ctx, AioHandlerLi= st *ready_list, =20 /* epoll(7) is faster above a certain number of fds */ if (fdmon_epoll_try_upgrade(ctx, npfd)) { + QLIST_FOREACH_RCU(node, &ctx->aio_handlers, node) { + if (!QLIST_IS_INSERTED(node, node_deleted) && node->pfd.events= ) { + g_source_remove_poll(&ctx->source, &node->pfd); + } + } npfd =3D 0; /* we won't need pollfds[], reset npfd */ return ctx->fdmon_ops->wait(ctx, ready_list, timeout); } @@ -97,11 +102,92 @@ static void fdmon_poll_update(AioContext *ctx, AioHandler *old_node, AioHandler *new_node) { - /* Do nothing, AioHandler already contains the state we'll need */ + if (old_node && !new_node) { + /* + * If the GSource is in the process of being destroyed then + * g_source_remove_poll() causes an assertion failure. Skip remov= al in + * that case, because glib cleans up its state during destruction + * anyway. + */ + if (!g_source_is_destroyed(&ctx->source)) { + g_source_remove_poll(&ctx->source, &old_node->pfd); + } + } + + if (!old_node && new_node) { + g_source_add_poll(&ctx->source, &new_node->pfd); + } +} + +static void fdmon_poll_gsource_prepare(AioContext *ctx) +{ + /* Do nothing */ +} + +static bool fdmon_poll_gsource_check(AioContext *ctx) +{ + AioHandler *node; + bool result =3D false; + + /* + * We have to walk very carefully in case aio_set_fd_handler is + * called while we're walking. + */ + qemu_lockcnt_inc(&ctx->list_lock); + + QLIST_FOREACH_RCU(node, &ctx->aio_handlers, node) { + int revents =3D node->pfd.revents & node->pfd.events; + + if (revents & (G_IO_IN | G_IO_HUP | G_IO_ERR) && node->io_read) { + result =3D true; + break; + } + if (revents & (G_IO_OUT | G_IO_ERR) && node->io_write) { + result =3D true; + break; + } + } + + qemu_lockcnt_dec(&ctx->list_lock); + + return result; +} + +static void fdmon_poll_gsource_dispatch(AioContext *ctx, + AioHandlerList *ready_list) +{ + AioHandler *node; + + QLIST_FOREACH_RCU(node, &ctx->aio_handlers, node) { + int revents; + + revents =3D node->pfd.revents & node->pfd.events; + if (revents & (G_IO_IN | G_IO_HUP | G_IO_ERR) && node->io_read) { + aio_add_ready_handler(ready_list, node, revents); + } else if (revents & (G_IO_OUT | G_IO_ERR) && node->io_write) { + aio_add_ready_handler(ready_list, node, revents); + } + } } =20 const FDMonOps fdmon_poll_ops =3D { .update =3D fdmon_poll_update, .wait =3D fdmon_poll_wait, .need_wait =3D aio_poll_disabled, + .gsource_prepare =3D fdmon_poll_gsource_prepare, + .gsource_check =3D fdmon_poll_gsource_check, + .gsource_dispatch =3D fdmon_poll_gsource_dispatch, }; + +void fdmon_poll_downgrade(AioContext *ctx) +{ + AioHandler *node; + + ctx->fdmon_ops =3D &fdmon_poll_ops; + + QLIST_FOREACH_RCU(node, &ctx->aio_handlers, node) { + if (!QLIST_IS_INSERTED(node, node_deleted) && node->pfd.events) { + g_source_add_poll(&ctx->source, &node->pfd); + } + } +} --=20 2.51.0 From nobody Mon Sep 22 02:22:19 2025 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=quarantine dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1757527819; cv=none; d=zohomail.com; s=zohoarc; b=RnmkZg5AHmobePntqBhx2ntuE3dCiIKpVphIx88R6O2YZyFB8QnOyiRGfd0l/Timysaag4J+x19FW1+I0plGlXlVNWyVUHMu5rsxEtLVFnV/nCpxBfBXbUE0EnX3kuX2lGE2gtkEGuT1yBDT/GoAdrPdyFTUJKWWdrsEvRT2NPc= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1757527819; h=Content-Transfer-Encoding:Cc:Cc:Date:Date:From:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:Subject:To:To:Message-Id:Reply-To; bh=4Eu3+4WK1mVXxUxYQOm67QPnmVITJoxhcpjU5MOQuIU=; b=hJ2niMY62Jvpbqp4UyUa6V/elLTtAhWyCp4fSoDPndGyz2/cHsImbllIGD2Rkdwqzvg3TFA3yCenVBH1ccelj2FB8ABPYtXLO4twu+Ybfdg3quzWp9R1x5j5BlPYdH5hXp285mLU3K3J2DvHuXbo9niLKVG2/Ai4qoKWWGqlh5Q= 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=quarantine dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1757527819256705.9464331731356; Wed, 10 Sep 2025 11:10:19 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1uwPE1-0004mz-DZ; Wed, 10 Sep 2025 14:07:22 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1uwPDg-0004U5-TL for qemu-devel@nongnu.org; Wed, 10 Sep 2025 14:07:01 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.133.124]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1uwPDa-00054u-Er for qemu-devel@nongnu.org; Wed, 10 Sep 2025 14:07:00 -0400 Received: from mx-prod-mc-08.mail-002.prod.us-west-2.aws.redhat.com (ec2-35-165-154-97.us-west-2.compute.amazonaws.com [35.165.154.97]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-542-wvkM_1P-OlW-XVRGGr3EIQ-1; Wed, 10 Sep 2025 14:06:49 -0400 Received: from mx-prod-int-06.mail-002.prod.us-west-2.aws.redhat.com (mx-prod-int-06.mail-002.prod.us-west-2.aws.redhat.com [10.30.177.93]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by mx-prod-mc-08.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTPS id 4E5DE180034A; Wed, 10 Sep 2025 18:06:47 +0000 (UTC) Received: from localhost (unknown [10.2.17.66]) by mx-prod-int-06.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTP id DB5421800447; Wed, 10 Sep 2025 18:06:46 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1757527613; 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=4Eu3+4WK1mVXxUxYQOm67QPnmVITJoxhcpjU5MOQuIU=; b=N8DXFUMQCzuWgu/klZexS+IFa2/P9RaQufzSNnPBal8/26t5+hgJ18mt91PfQw541yeYoB AwHckWgNrpatj16RKCNliBrTMXdpNoo/IJgB3utASTnCwywUGNsTx740mdcUU6y2ZcSH2R 1JUrYwYWffo1XwiCB8r8Z2b/2oqAvfI= X-MC-Unique: wvkM_1P-OlW-XVRGGr3EIQ-1 X-Mimecast-MFC-AGG-ID: wvkM_1P-OlW-XVRGGr3EIQ_1757527607 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Aarushi Mehta , Fam Zheng , Stefano Garzarella , Hanna Czenczek , eblake@redhat.com, Stefan Hajnoczi , Kevin Wolf , qemu-block@nongnu.org, hibriansong@gmail.com, Stefan Weil , Paolo Bonzini Subject: [PATCH v4 05/12] aio: remove aio_context_use_g_source() Date: Wed, 10 Sep 2025 13:56:56 -0400 Message-ID: <20250910175703.374499-6-stefanha@redhat.com> In-Reply-To: <20250910175703.374499-1-stefanha@redhat.com> References: <20250910175703.374499-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.4.1 on 10.30.177.93 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: -16 X-Spam_score: -1.7 X-Spam_bar: - X-Spam_report: (-1.7 / 5.0 requ) BAYES_00=-1.9, DKIM_INVALID=0.1, DKIM_SIGNED=0.1, RCVD_IN_DNSWL_NONE=-0.0001, RCVD_IN_MSPIKE_H5=0.001, RCVD_IN_MSPIKE_WL=0.001, RCVD_IN_VALIDITY_RPBL_BLOCKED=0.001, RCVD_IN_VALIDITY_SAFE_BLOCKED=0.001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001 autolearn=no 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-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1757527820429116600 Content-Type: text/plain; charset="utf-8" There is no need for aio_context_use_g_source() now that epoll(7) and io_uring(7) file descriptor monitoring works with the glib event loop. AioContext doesn't need to be notified that GSource is being used. Signed-off-by: Stefan Hajnoczi Reviewed-by: Eric Blake --- include/block/aio.h | 3 --- tests/unit/test-nested-aio-poll.c | 6 ------ util/aio-posix.c | 12 ------------ util/aio-win32.c | 4 ---- util/async.c | 1 - 5 files changed, 26 deletions(-) diff --git a/include/block/aio.h b/include/block/aio.h index 39ed86d14d..1657740a0e 100644 --- a/include/block/aio.h +++ b/include/block/aio.h @@ -728,9 +728,6 @@ void aio_context_setup(AioContext *ctx); */ void aio_context_destroy(AioContext *ctx); =20 -/* Used internally, do not call outside AioContext code */ -void aio_context_use_g_source(AioContext *ctx); - /** * aio_context_set_poll_params: * @ctx: the aio context diff --git a/tests/unit/test-nested-aio-poll.c b/tests/unit/test-nested-aio= -poll.c index 45484e745b..d13ecccd8c 100644 --- a/tests/unit/test-nested-aio-poll.c +++ b/tests/unit/test-nested-aio-poll.c @@ -83,12 +83,6 @@ static void test(void) /* Enable polling */ aio_context_set_poll_params(td.ctx, 1000000, 2, 2, &error_abort); =20 - /* - * The GSource is unused but this has the side-effect of changing the = fdmon - * that AioContext uses. - */ - aio_get_g_source(td.ctx); - /* Make the event notifier active (set) right away */ event_notifier_init(&td.poll_notifier, 1); aio_set_event_notifier(td.ctx, &td.poll_notifier, diff --git a/util/aio-posix.c b/util/aio-posix.c index 9de05ee7e8..bebd9ce3a2 100644 --- a/util/aio-posix.c +++ b/util/aio-posix.c @@ -743,18 +743,6 @@ void aio_context_destroy(AioContext *ctx) aio_free_deleted_handlers(ctx); } =20 -void aio_context_use_g_source(AioContext *ctx) -{ - /* - * Disable io_uring when the glib main loop is used because it doesn't - * support mixed glib/aio_poll() usage. It relies on aio_poll() being - * called regularly so that changes to the monitored file descriptors = are - * submitted, otherwise a list of pending fd handlers builds up. - */ - fdmon_io_uring_destroy(ctx); - aio_free_deleted_handlers(ctx); -} - void aio_context_set_poll_params(AioContext *ctx, int64_t max_ns, int64_t grow, int64_t shrink, Error **err= p) { diff --git a/util/aio-win32.c b/util/aio-win32.c index 6583d5c5f3..34c4074133 100644 --- a/util/aio-win32.c +++ b/util/aio-win32.c @@ -427,10 +427,6 @@ void aio_context_destroy(AioContext *ctx) { } =20 -void aio_context_use_g_source(AioContext *ctx) -{ -} - void aio_context_set_poll_params(AioContext *ctx, int64_t max_ns, int64_t grow, int64_t shrink, Error **err= p) { diff --git a/util/async.c b/util/async.c index 2719c629ae..a39410d675 100644 --- a/util/async.c +++ b/util/async.c @@ -430,7 +430,6 @@ static GSourceFuncs aio_source_funcs =3D { =20 GSource *aio_get_g_source(AioContext *ctx) { - aio_context_use_g_source(ctx); g_source_ref(&ctx->source); return &ctx->source; } --=20 2.51.0 From nobody Mon Sep 22 02:22:19 2025 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=quarantine dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1757527852; cv=none; d=zohomail.com; s=zohoarc; b=jS9MGJWMvB7DJAdLkE6bRd8inkHgChR4E/QocTXeF7O7LAZ01QZGHMf4Vpzwxcq1of3fVfBZDNRSDhoOE8MZ/G4eY5B9fJdETD4s4Ieyr6/XsKgKv+sofdhd9xhcAqNhfncocw6q6f6kChxys3xKucrAx6FuTFZw3yR0kptzdvI= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1757527852; h=Content-Transfer-Encoding:Cc:Cc:Date:Date:From:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:Subject:To:To:Message-Id:Reply-To; bh=xHdACXr84YhqbE/jHeXVyF/tWlCAyRBtrZFX3E/y3wI=; b=AM7orqE0bSn3KrvYtuuQD1k/CWKc1kwpMR0w7pkXvTNcCon8Mmij3Lx3YtlsPi5tSDnTZIEI/PvqIWJxEeOkWa1+6MoWOVD3MaDa4+4vpKN65SsiI243K0Md9tYict3t25xbzH7GgqomK3TlzbtQkoD48Vzws7yp4a3KnqNZNc8= 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=quarantine dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 17575278527561006.1753248725856; Wed, 10 Sep 2025 11:10:52 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1uwPGW-00075I-MD; Wed, 10 Sep 2025 14:09:56 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1uwPDk-0004dk-I6 for qemu-devel@nongnu.org; Wed, 10 Sep 2025 14:07:06 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.129.124]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1uwPDc-000554-Gl for qemu-devel@nongnu.org; Wed, 10 Sep 2025 14:07:04 -0400 Received: from mx-prod-mc-06.mail-002.prod.us-west-2.aws.redhat.com (ec2-35-165-154-97.us-west-2.compute.amazonaws.com [35.165.154.97]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-567-Cxc0VQl5M8KmS7UuROigjw-1; Wed, 10 Sep 2025 14:06:50 -0400 Received: from mx-prod-int-06.mail-002.prod.us-west-2.aws.redhat.com (mx-prod-int-06.mail-002.prod.us-west-2.aws.redhat.com [10.30.177.93]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by mx-prod-mc-06.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTPS id 40FC81800285; Wed, 10 Sep 2025 18:06:49 +0000 (UTC) Received: from localhost (unknown [10.2.17.66]) by mx-prod-int-06.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTP id B33DD1800447; Wed, 10 Sep 2025 18:06:48 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1757527613; 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=xHdACXr84YhqbE/jHeXVyF/tWlCAyRBtrZFX3E/y3wI=; b=ZPY0xkKkLR89LYTQ0o/Bh/2pEbfUqZ9tl6brBWqtUADjrzLr52R7BVFfPPZOtj/Zs40KJW sUnGBam4aQdXL8modN0wMf2Q1e7PVr1Q31qUY505rlhY+TLFBYsSDrneMND1RskGHoMDml wnZ+OKKZOS2MBXG68v07YX4o2qJcJTU= X-MC-Unique: Cxc0VQl5M8KmS7UuROigjw-1 X-Mimecast-MFC-AGG-ID: Cxc0VQl5M8KmS7UuROigjw_1757527609 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Aarushi Mehta , Fam Zheng , Stefano Garzarella , Hanna Czenczek , eblake@redhat.com, Stefan Hajnoczi , Kevin Wolf , qemu-block@nongnu.org, hibriansong@gmail.com, Stefan Weil , Paolo Bonzini Subject: [PATCH v4 06/12] aio: free AioContext when aio_context_new() fails Date: Wed, 10 Sep 2025 13:56:57 -0400 Message-ID: <20250910175703.374499-7-stefanha@redhat.com> In-Reply-To: <20250910175703.374499-1-stefanha@redhat.com> References: <20250910175703.374499-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.4.1 on 10.30.177.93 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: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_NONE=-0.0001, RCVD_IN_MSPIKE_H5=0.001, RCVD_IN_MSPIKE_WL=0.001, RCVD_IN_VALIDITY_RPBL_BLOCKED=0.001, RCVD_IN_VALIDITY_SAFE_BLOCKED=0.001, SPF_HELO_PASS=-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-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1757527854689116600 Content-Type: text/plain; charset="utf-8" g_source_destroy() only removes the GSource from the GMainContext it's attached to, if any. It does not free it. Use g_source_unref() instead so that the AioContext (which embeds a GSource) is freed. There is no need to call g_source_destroy() in aio_context_new() because the GSource isn't attached to a GMainContext yet. aio_ctx_finalize() expects everything to be set up already, so introduce the new ctx->initialized boolean and do nothing when called with !initialized. This also requires moving aio_context_setup() down after event_notifier_init() since aio_ctx_finalize() won't release any resources that aio_context_setup() acquired. Signed-off-by: Stefan Hajnoczi Reviewed-by: Eric Blake --- v2: - Fix spacing in aio_ctx_finalize() argument list [Eric] --- include/block/aio.h | 3 +++ util/async.c | 14 +++++++++++--- 2 files changed, 14 insertions(+), 3 deletions(-) diff --git a/include/block/aio.h b/include/block/aio.h index 1657740a0e..2760f308f5 100644 --- a/include/block/aio.h +++ b/include/block/aio.h @@ -291,6 +291,9 @@ struct AioContext { gpointer epollfd_tag; =20 const FDMonOps *fdmon_ops; + + /* Was aio_context_new() successful? */ + bool initialized; }; =20 /** diff --git a/util/async.c b/util/async.c index a39410d675..34aaab4e9e 100644 --- a/util/async.c +++ b/util/async.c @@ -363,12 +363,16 @@ aio_ctx_dispatch(GSource *source, } =20 static void -aio_ctx_finalize(GSource *source) +aio_ctx_finalize(GSource *source) { AioContext *ctx =3D (AioContext *) source; QEMUBH *bh; unsigned flags; =20 + if (!ctx->initialized) { + return; + } + thread_pool_free_aio(ctx->thread_pool); =20 #ifdef CONFIG_LINUX_AIO @@ -579,13 +583,15 @@ AioContext *aio_context_new(Error **errp) ctx =3D (AioContext *) g_source_new(&aio_source_funcs, sizeof(AioConte= xt)); QSLIST_INIT(&ctx->bh_list); QSIMPLEQ_INIT(&ctx->bh_slice_list); - aio_context_setup(ctx); =20 ret =3D event_notifier_init(&ctx->notifier, false); if (ret < 0) { error_setg_errno(errp, -ret, "Failed to initialize event notifier"= ); goto fail; } + + aio_context_setup(ctx); + g_source_set_can_recurse(&ctx->source, true); qemu_lockcnt_init(&ctx->list_lock); =20 @@ -619,9 +625,11 @@ AioContext *aio_context_new(Error **errp) =20 register_aiocontext(ctx); =20 + ctx->initialized =3D true; + return ctx; fail: - g_source_destroy(&ctx->source); + g_source_unref(&ctx->source); return NULL; } =20 --=20 2.51.0 From nobody Mon Sep 22 02:22:19 2025 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=quarantine dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1757527957; cv=none; d=zohomail.com; s=zohoarc; b=QTO7YiBbs4TA+BPVuGiiB5YYd3/e1U8rECYTsbeVO3O1ao0VHG1IfeI0RYfftokdAirwpLxlyulg4VLIgbRxQ1lI+/kXN3yxjM5i5BUVeecDW7GRYbtUutj861xiDf5VloyioPghBkYq+Sl711mq2lHy19kLKyDRRauSgCElans= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1757527957; h=Content-Transfer-Encoding:Cc:Cc:Date:Date:From:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:Subject:To:To:Message-Id:Reply-To; bh=8G9SH3P4RQbLnkpw+zUh6T+aeJe+fBgecMc4c5oGF3w=; b=jMvVXhelW50K6m1PRIfLlo5bHKHIMLQV4DKdQ8VOCGXDT1S+OsAoAcQhZn0Po2Ed+gxOyGSE89ol06nhQdYk92aOzZTPx1E5xFoiYHPpgXZ1k90xwiq5k5lQVyr459nXR1THqZDe9ymooBuyGJRH5PCLnE8bEIeqXRPlwLfRXvw= 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=quarantine dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 175752795771483.69678744750388; Wed, 10 Sep 2025 11:12:37 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1uwPEW-0005R8-RL; Wed, 10 Sep 2025 14:07:53 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1uwPDh-0004Un-DZ for qemu-devel@nongnu.org; Wed, 10 Sep 2025 14:07:01 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.133.124]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1uwPDe-00055Y-LE for qemu-devel@nongnu.org; Wed, 10 Sep 2025 14:07:01 -0400 Received: from mx-prod-mc-08.mail-002.prod.us-west-2.aws.redhat.com (ec2-35-165-154-97.us-west-2.compute.amazonaws.com [35.165.154.97]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-616-2cWktSxrO6OO4eEfmFKceA-1; Wed, 10 Sep 2025 14:06:53 -0400 Received: from mx-prod-int-05.mail-002.prod.us-west-2.aws.redhat.com (mx-prod-int-05.mail-002.prod.us-west-2.aws.redhat.com [10.30.177.17]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by mx-prod-mc-08.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTPS id C51E618002C6; Wed, 10 Sep 2025 18:06:50 +0000 (UTC) Received: from localhost (unknown [10.2.17.66]) by mx-prod-int-05.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTP id 5D5C21955F24; Wed, 10 Sep 2025 18:06:50 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1757527616; 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=8G9SH3P4RQbLnkpw+zUh6T+aeJe+fBgecMc4c5oGF3w=; b=HV2rZ/wrXK1mFUVHvHJe4BDXYGCAYX5NLg4FKyxkMayChxcyE87igOa3rleRQ03BH9l6M7 xo2wKuwn+oiqVf+unkjqaVCzV+IcFurHUAXDGVKouuuo74nHra44of5SBfg6DFUA6pHI6I V0JiFR6861bxV+sJof4PmjfmEkgqMT0= X-MC-Unique: 2cWktSxrO6OO4eEfmFKceA-1 X-Mimecast-MFC-AGG-ID: 2cWktSxrO6OO4eEfmFKceA_1757527610 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Aarushi Mehta , Fam Zheng , Stefano Garzarella , Hanna Czenczek , eblake@redhat.com, Stefan Hajnoczi , Kevin Wolf , qemu-block@nongnu.org, hibriansong@gmail.com, Stefan Weil , Paolo Bonzini Subject: [PATCH v4 07/12] aio: add errp argument to aio_context_setup() Date: Wed, 10 Sep 2025 13:56:58 -0400 Message-ID: <20250910175703.374499-8-stefanha@redhat.com> In-Reply-To: <20250910175703.374499-1-stefanha@redhat.com> References: <20250910175703.374499-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.0 on 10.30.177.17 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, RCVD_IN_MSPIKE_H5=0.001, RCVD_IN_MSPIKE_WL=0.001, RCVD_IN_VALIDITY_RPBL_BLOCKED=0.001, RCVD_IN_VALIDITY_SAFE_BLOCKED=0.001, SPF_HELO_PASS=-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-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1757527958161116600 Content-Type: text/plain; charset="utf-8" When aio_context_new() -> aio_context_setup() fails at startup it doesn't really matter whether errors are returned to the caller or the process terminates immediately. However, it is not acceptable to terminate when hotplugging --object iothread at runtime. Refactor aio_context_setup() so that errors can be propagated. The next commit will set errp when fdmon_io_uring_setup() fails. Suggested-by: Kevin Wolf Signed-off-by: Stefan Hajnoczi Reviewed-by: Eric Blake --- include/block/aio.h | 3 ++- util/aio-posix.c | 2 +- util/aio-win32.c | 2 +- util/async.c | 7 ++++++- 4 files changed, 10 insertions(+), 4 deletions(-) diff --git a/include/block/aio.h b/include/block/aio.h index 2760f308f5..d919d7c8f4 100644 --- a/include/block/aio.h +++ b/include/block/aio.h @@ -718,10 +718,11 @@ void qemu_set_current_aio_context(AioContext *ctx); /** * aio_context_setup: * @ctx: the aio context + * @errp: error pointer * * Initialize the aio context. */ -void aio_context_setup(AioContext *ctx); +void aio_context_setup(AioContext *ctx, Error **errp); =20 /** * aio_context_destroy: diff --git a/util/aio-posix.c b/util/aio-posix.c index bebd9ce3a2..bbe470c145 100644 --- a/util/aio-posix.c +++ b/util/aio-posix.c @@ -718,7 +718,7 @@ bool aio_poll(AioContext *ctx, bool blocking) return progress; } =20 -void aio_context_setup(AioContext *ctx) +void aio_context_setup(AioContext *ctx, Error **errp) { ctx->fdmon_ops =3D &fdmon_poll_ops; ctx->epollfd =3D -1; diff --git a/util/aio-win32.c b/util/aio-win32.c index 34c4074133..4ba401ff92 100644 --- a/util/aio-win32.c +++ b/util/aio-win32.c @@ -419,7 +419,7 @@ bool aio_poll(AioContext *ctx, bool blocking) return progress; } =20 -void aio_context_setup(AioContext *ctx) +void aio_context_setup(AioContext *ctx, Error **errp) { } =20 diff --git a/util/async.c b/util/async.c index 34aaab4e9e..2bb7e1f3c1 100644 --- a/util/async.c +++ b/util/async.c @@ -577,6 +577,7 @@ static void co_schedule_bh_cb(void *opaque) =20 AioContext *aio_context_new(Error **errp) { + ERRP_GUARD(); int ret; AioContext *ctx; =20 @@ -590,7 +591,11 @@ AioContext *aio_context_new(Error **errp) goto fail; } =20 - aio_context_setup(ctx); + aio_context_setup(ctx, errp); + if (*errp) { + event_notifier_cleanup(&ctx->notifier); + goto fail; + } =20 g_source_set_can_recurse(&ctx->source, true); qemu_lockcnt_init(&ctx->list_lock); --=20 2.51.0 From nobody Mon Sep 22 02:22:19 2025 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=quarantine dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1757528098; cv=none; d=zohomail.com; s=zohoarc; b=DSVP6lDRNf9QUxqbrMj6enAUX6QbAzcaFjziF6nIM+FsU2/EWrjQvRc/3pvffzogj3Pb6llfrHkeqTzPkK7t8ve947OfvJ++dh4YAc69Y+yBiD0KSonwZGogzcoE0uuh9Q0aqP0jrmB5WuTvH6RYUDal+0MJepUZTxp1UznNOTc= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1757528098; h=Content-Transfer-Encoding:Cc:Cc:Date:Date:From:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:Subject:To:To:Message-Id:Reply-To; bh=eol21QjpcqS+1NzmihTA3L9wQ1fcb13YdBbC/Hdfqwo=; b=AR7/neANEFEjzSAll8/inQAXdSwI8M92V8dIQ2SNRXZHNZTtJWZwxDRzQJvWxid/t3fnolbgM9fZqLJ0nhr6QRuTDW+DFEzCaT9Z54cqAHqIPFHj6pvhAq/+viR+lmA14wO28DZXQukwhEZZ2Y/VGnGF2u18fv5H0QkLgSm46qE= 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=quarantine dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1757528098886586.0084142905914; Wed, 10 Sep 2025 11:14:58 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1uwPEp-0005pc-Sd; Wed, 10 Sep 2025 14:08:17 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1uwPDn-0004jW-2U for qemu-devel@nongnu.org; Wed, 10 Sep 2025 14:07:08 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.129.124]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1uwPDe-00055i-LP for qemu-devel@nongnu.org; Wed, 10 Sep 2025 14:07:05 -0400 Received: from mx-prod-mc-01.mail-002.prod.us-west-2.aws.redhat.com (ec2-54-186-198-63.us-west-2.compute.amazonaws.com [54.186.198.63]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-1-xOZMvGDaP8eXy7sKrBQqgw-1; Wed, 10 Sep 2025 14:06:54 -0400 Received: from mx-prod-int-01.mail-002.prod.us-west-2.aws.redhat.com (mx-prod-int-01.mail-002.prod.us-west-2.aws.redhat.com [10.30.177.4]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by mx-prod-mc-01.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTPS id D64B7195609F; Wed, 10 Sep 2025 18:06:52 +0000 (UTC) Received: from localhost (unknown [10.2.17.66]) by mx-prod-int-01.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTP id 3499D3000198; Wed, 10 Sep 2025 18:06:52 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1757527617; 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=eol21QjpcqS+1NzmihTA3L9wQ1fcb13YdBbC/Hdfqwo=; b=iXIUo4Djvuh3kE1naQEzxDnm60JWSLz466p1DZoas2VTF8gIUHOd9SLSBiRs6af3KKYLWs SHGdErX8mUK7tXRhh+p+kqn1ZuOR24nd8abx/fZV0P5NHku980aB4kQEptWlDzQkCcWRnW f7cdV43i8oyxChZNT6mQlaI2NsmVY10= X-MC-Unique: xOZMvGDaP8eXy7sKrBQqgw-1 X-Mimecast-MFC-AGG-ID: xOZMvGDaP8eXy7sKrBQqgw_1757527613 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Aarushi Mehta , Fam Zheng , Stefano Garzarella , Hanna Czenczek , eblake@redhat.com, Stefan Hajnoczi , Kevin Wolf , qemu-block@nongnu.org, hibriansong@gmail.com, Stefan Weil , Paolo Bonzini Subject: [PATCH v4 08/12] aio-posix: gracefully handle io_uring_queue_init() failure Date: Wed, 10 Sep 2025 13:56:59 -0400 Message-ID: <20250910175703.374499-9-stefanha@redhat.com> In-Reply-To: <20250910175703.374499-1-stefanha@redhat.com> References: <20250910175703.374499-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.4.1 on 10.30.177.4 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=170.10.129.124; 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, RCVD_IN_MSPIKE_H5=0.001, RCVD_IN_MSPIKE_WL=0.001, RCVD_IN_VALIDITY_RPBL_BLOCKED=0.001, RCVD_IN_VALIDITY_SAFE_BLOCKED=0.001, SPF_HELO_PASS=-0.001, T_SPF_TEMPERROR=0.01 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-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1757528099804116600 Content-Type: text/plain; charset="utf-8" io_uring may not be available at runtime due to system policies (e.g. the io_uring_disabled sysctl) or creation could fail due to file descriptor resource limits. Handle failure scenarios as follows: If another AioContext already has io_uring, then fail AioContext creation so that the aio_add_sqe() API is available uniformly from all QEMU threads. Otherwise fall back to epoll(7) if io_uring is unavailable. Notes: - Update the comment about selecting the fastest fdmon implementation. At this point it's not about speed anymore, it's about aio_add_sqe() API availability. - Uppercase the error message when converting from error_report() to error_setg_errno() for consistency (but there are instances of lowercase in the codebase). - It's easier to move the #ifdefs from aio-posix.h to aio-posix.c. Signed-off-by: Stefan Hajnoczi Reviewed-by: Eric Blake --- util/aio-posix.h | 12 ++---------- util/aio-posix.c | 29 ++++++++++++++++++++++++++--- util/fdmon-io_uring.c | 8 ++++---- 3 files changed, 32 insertions(+), 17 deletions(-) diff --git a/util/aio-posix.h b/util/aio-posix.h index f9994ed79e..6f9d97d866 100644 --- a/util/aio-posix.h +++ b/util/aio-posix.h @@ -18,6 +18,7 @@ #define AIO_POSIX_H =20 #include "block/aio.h" +#include "qapi/error.h" =20 struct AioHandler { GPollFD pfd; @@ -72,17 +73,8 @@ static inline void fdmon_epoll_disable(AioContext *ctx) #endif /* !CONFIG_EPOLL_CREATE1 */ =20 #ifdef CONFIG_LINUX_IO_URING -bool fdmon_io_uring_setup(AioContext *ctx); +void fdmon_io_uring_setup(AioContext *ctx, Error **errp); void fdmon_io_uring_destroy(AioContext *ctx); -#else -static inline bool fdmon_io_uring_setup(AioContext *ctx) -{ - return false; -} - -static inline void fdmon_io_uring_destroy(AioContext *ctx) -{ -} #endif /* !CONFIG_LINUX_IO_URING */ =20 #endif /* AIO_POSIX_H */ diff --git a/util/aio-posix.c b/util/aio-posix.c index bbe470c145..800b4debbf 100644 --- a/util/aio-posix.c +++ b/util/aio-posix.c @@ -16,6 +16,7 @@ #include "qemu/osdep.h" #include "block/block.h" #include "block/thread-pool.h" +#include "qapi/error.h" #include "qemu/main-loop.h" #include "qemu/lockcnt.h" #include "qemu/rcu.h" @@ -724,17 +725,39 @@ void aio_context_setup(AioContext *ctx, Error **errp) ctx->epollfd =3D -1; ctx->epollfd_tag =3D NULL; =20 - /* Use the fastest fd monitoring implementation if available */ - if (fdmon_io_uring_setup(ctx)) { - return; +#ifdef CONFIG_LINUX_IO_URING + { + static bool need_io_uring; + Error *local_err =3D NULL; /* ERRP_GUARD() doesn't handle error_ab= ort */ + + /* io_uring takes precedence because it provides aio_add_sqe() sup= port */ + fdmon_io_uring_setup(ctx, &local_err); + if (!local_err) { + /* + * If one AioContext gets io_uring, then all AioContexts need = io_uring + * so that aio_add_sqe() support is available across all threa= ds. + */ + need_io_uring =3D true; + return; + } + if (need_io_uring) { + error_propagate(errp, local_err); + return; + } + + /* Silently fall back on systems where io_uring is unavailable */ + error_free(local_err); } +#endif /* CONFIG_LINUX_IO_URING */ =20 fdmon_epoll_setup(ctx); } =20 void aio_context_destroy(AioContext *ctx) { +#ifdef CONFIG_LINUX_IO_URING fdmon_io_uring_destroy(ctx); +#endif =20 qemu_lockcnt_lock(&ctx->list_lock); fdmon_epoll_disable(ctx); diff --git a/util/fdmon-io_uring.c b/util/fdmon-io_uring.c index 4b6dc8903f..3181601767 100644 --- a/util/fdmon-io_uring.c +++ b/util/fdmon-io_uring.c @@ -45,6 +45,7 @@ =20 #include "qemu/osdep.h" #include +#include "qapi/error.h" #include "qemu/rcu_queue.h" #include "aio-posix.h" =20 @@ -369,7 +370,7 @@ static const FDMonOps fdmon_io_uring_ops =3D { .gsource_dispatch =3D fdmon_io_uring_gsource_dispatch, }; =20 -bool fdmon_io_uring_setup(AioContext *ctx) +void fdmon_io_uring_setup(AioContext *ctx, Error **errp) { int ret; =20 @@ -377,15 +378,14 @@ bool fdmon_io_uring_setup(AioContext *ctx) =20 ret =3D io_uring_queue_init(FDMON_IO_URING_ENTRIES, &ctx->fdmon_io_uri= ng, 0); if (ret !=3D 0) { - return false; + error_setg_errno(errp, -ret, "Failed to initialize io_uring"); + return; } =20 QSLIST_INIT(&ctx->submit_list); ctx->fdmon_ops =3D &fdmon_io_uring_ops; ctx->io_uring_fd_tag =3D g_source_add_unix_fd(&ctx->source, ctx->fdmon_io_uring.ring_fd, G_IO_IN); - - return true; } =20 void fdmon_io_uring_destroy(AioContext *ctx) --=20 2.51.0 From nobody Mon Sep 22 02:22:19 2025 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=quarantine dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1757527851; cv=none; d=zohomail.com; s=zohoarc; b=Y3nCgReqISKuuRYwAVNjw1HfNuu0k1RWZfKsMt2PA3DAO74FSlcYAF1XkwGkO0VV3BXZCX7/gfXoYNZqXQWT2te8gDH5pEQvAdU/2AGqLgG8GAxRfjirHo5ojTJVR7b/PJ5r7OsiRBqeYjmz0++2JoJzb5ABABpHtQ5lBKMgUI8= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1757527851; h=Content-Transfer-Encoding:Cc:Cc:Date:Date:From:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:Subject:To:To:Message-Id:Reply-To; bh=bfnzjAANnmsE4ocVGt6kY9Vy45/88OZNR3tRHCcDEog=; b=CpSpyytlvdKIefjABHrMDiEA0xN2+SD5+PaVD4qP4cpCSDRtmE5lPMmsWwk88qFiqtVrlPxxUQ/c/puenuqrqhP2OGARYtKCdsx6J1oB9T5bljPzlroPmb561eWqU6CYQYxNk9CJHlyeOrlDWZaGWbtXzuUuI6E5CCZoqycap44= 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=quarantine dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1757527851708785.9546198891763; Wed, 10 Sep 2025 11:10:51 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1uwPEi-0005lK-Lj; Wed, 10 Sep 2025 14:08:08 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1uwPDn-0004jX-3C for qemu-devel@nongnu.org; Wed, 10 Sep 2025 14:07:08 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.129.124]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1uwPDg-00056A-In for qemu-devel@nongnu.org; Wed, 10 Sep 2025 14:07:06 -0400 Received: from mx-prod-mc-08.mail-002.prod.us-west-2.aws.redhat.com (ec2-35-165-154-97.us-west-2.compute.amazonaws.com [35.165.154.97]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-679-69vUACC4Nba83r_3do7WNQ-1; Wed, 10 Sep 2025 14:06:56 -0400 Received: from mx-prod-int-05.mail-002.prod.us-west-2.aws.redhat.com (mx-prod-int-05.mail-002.prod.us-west-2.aws.redhat.com [10.30.177.17]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by mx-prod-mc-08.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTPS id C6D4E180045C; Wed, 10 Sep 2025 18:06:54 +0000 (UTC) Received: from localhost (unknown [10.2.17.66]) by mx-prod-int-05.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTP id 280D41955F24; Wed, 10 Sep 2025 18:06:53 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1757527619; 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=bfnzjAANnmsE4ocVGt6kY9Vy45/88OZNR3tRHCcDEog=; b=MpjzdvVMjubE/AUskii5ZRY0jPzOfM4ikLvX1CpdooQ9aNG6RkW4O0V4AxfQkrSCs0qYAP TDMDt3ur0Bnf3fsdSAgBLLIyYLMUJtjHq8iFuhm1qrMUEU3IbOZb7YbeaGHYLZvNcQRE4c BoixdquL0fQdtfdWt/Uje4Kpgla89qI= X-MC-Unique: 69vUACC4Nba83r_3do7WNQ-1 X-Mimecast-MFC-AGG-ID: 69vUACC4Nba83r_3do7WNQ_1757527614 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Aarushi Mehta , Fam Zheng , Stefano Garzarella , Hanna Czenczek , eblake@redhat.com, Stefan Hajnoczi , Kevin Wolf , qemu-block@nongnu.org, hibriansong@gmail.com, Stefan Weil , Paolo Bonzini Subject: [PATCH v4 09/12] aio-posix: add aio_add_sqe() API for user-defined io_uring requests Date: Wed, 10 Sep 2025 13:57:00 -0400 Message-ID: <20250910175703.374499-10-stefanha@redhat.com> In-Reply-To: <20250910175703.374499-1-stefanha@redhat.com> References: <20250910175703.374499-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.0 on 10.30.177.17 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: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_NONE=-0.0001, RCVD_IN_MSPIKE_H5=0.001, RCVD_IN_MSPIKE_WL=0.001, RCVD_IN_VALIDITY_RPBL_BLOCKED=0.001, RCVD_IN_VALIDITY_SAFE_BLOCKED=0.001, SPF_HELO_PASS=-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-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1757527852853116600 Content-Type: text/plain; charset="utf-8" Introduce the aio_add_sqe() API for submitting io_uring requests in the current AioContext. This allows other components in QEMU, like the block layer, to take advantage of io_uring features without creating their own io_uring context. This API supports nested event loops just like file descriptor monitoring and BHs do. This comes at a complexity cost: a BH is required to dispatch CQE callbacks and they are placed on a list so that a nested event loop can invoke its parent's pending CQE callbacks. If you're wondering why CqeHandler exists instead of just a callback function pointer, this is why. Signed-off-by: Stefan Hajnoczi Reviewed-by: Eric Blake --- v2: - Fix pre_sqe -> prep_sqe typo [Eric] - Add #endif terminator comment [Eric] --- include/block/aio.h | 84 ++++++++++++++++++++++- util/aio-posix.h | 1 + util/aio-posix.c | 9 +++ util/fdmon-io_uring.c | 152 ++++++++++++++++++++++++++++++++---------- 4 files changed, 208 insertions(+), 38 deletions(-) diff --git a/include/block/aio.h b/include/block/aio.h index d919d7c8f4..56ea0d47b7 100644 --- a/include/block/aio.h +++ b/include/block/aio.h @@ -61,6 +61,27 @@ typedef struct LuringState LuringState; /* Is polling disabled? */ bool aio_poll_disabled(AioContext *ctx); =20 +#ifdef CONFIG_LINUX_IO_URING +/* + * Each io_uring request must have a unique CqeHandler that processes the = cqe. + * The lifetime of a CqeHandler must be at least from aio_add_sqe() until + * ->cb() invocation. + */ +typedef struct CqeHandler CqeHandler; +struct CqeHandler { + /* Called by the AioContext when the request has completed */ + void (*cb)(CqeHandler *handler); + + /* Used internally, do not access this */ + QSIMPLEQ_ENTRY(CqeHandler) next; + + /* This field is filled in before ->cb() is called */ + struct io_uring_cqe cqe; +}; + +typedef QSIMPLEQ_HEAD(, CqeHandler) CqeHandlerSimpleQ; +#endif /* CONFIG_LINUX_IO_URING */ + /* Callbacks for file descriptor monitoring implementations */ typedef struct { /* @@ -138,6 +159,27 @@ typedef struct { * Called with list_lock incremented. */ void (*gsource_dispatch)(AioContext *ctx, AioHandlerList *ready_list); + +#ifdef CONFIG_LINUX_IO_URING + /** + * aio_add_sqe: Add an io_uring sqe for submission. + * @prep_sqe: invoked with an sqe that should be prepared for submissi= on + * @opaque: user-defined argument to @prep_sqe() + * @cqe_handler: the unique cqe handler associated with this request + * + * The caller's @prep_sqe() function is invoked to fill in the details= of + * the sqe. Do not call io_uring_sqe_set_data() on this sqe. + * + * The kernel may see the sqe as soon as @prep_sqe() returns or it may= take + * until the next event loop iteration. + * + * This function is called from the current AioContext and is not + * thread-safe. + */ + void (*add_sqe)(AioContext *ctx, + void (*prep_sqe)(struct io_uring_sqe *sqe, void *opaqu= e), + void *opaque, CqeHandler *cqe_handler); +#endif /* CONFIG_LINUX_IO_URING */ } FDMonOps; =20 /* @@ -255,7 +297,11 @@ struct AioContext { struct io_uring fdmon_io_uring; AioHandlerSList submit_list; gpointer io_uring_fd_tag; -#endif + + /* Pending callback state for cqe handlers */ + CqeHandlerSimpleQ cqe_handler_ready_list; + QEMUBH *cqe_handler_bh; +#endif /* CONFIG_LINUX_IO_URING */ =20 /* TimerLists for calling timers - one per clock type. Has its own * locking. @@ -761,4 +807,40 @@ void aio_context_set_aio_params(AioContext *ctx, int64= _t max_batch); */ void aio_context_set_thread_pool_params(AioContext *ctx, int64_t min, int64_t max, Error **errp); + +#ifdef CONFIG_LINUX_IO_URING +/** + * aio_has_io_uring: Return whether io_uring is available. + * + * io_uring is either available in all AioContexts or in none, so this only + * needs to be called once from within any thread's AioContext. + */ +static inline bool aio_has_io_uring(void) +{ + AioContext *ctx =3D qemu_get_current_aio_context(); + return ctx->fdmon_ops->add_sqe; +} + +/** + * aio_add_sqe: Add an io_uring sqe for submission. + * @prep_sqe: invoked with an sqe that should be prepared for submission + * @opaque: user-defined argument to @prep_sqe() + * @cqe_handler: the unique cqe handler associated with this request + * + * The caller's @prep_sqe() function is invoked to fill in the details of = the + * sqe. Do not call io_uring_sqe_set_data() on this sqe. + * + * The sqe is submitted by the current AioContext. The kernel may see the = sqe + * as soon as @prep_sqe() returns or it may take until the next event loop + * iteration. + * + * When the AioContext is destroyed, pending sqes are ignored and their + * CqeHandlers are not invoked. + * + * This function must be called only when aio_has_io_uring() returns true. + */ +void aio_add_sqe(void (*prep_sqe)(struct io_uring_sqe *sqe, void *opaque), + void *opaque, CqeHandler *cqe_handler); +#endif /* CONFIG_LINUX_IO_URING */ + #endif diff --git a/util/aio-posix.h b/util/aio-posix.h index 6f9d97d866..57ef801a5f 100644 --- a/util/aio-posix.h +++ b/util/aio-posix.h @@ -36,6 +36,7 @@ struct AioHandler { #ifdef CONFIG_LINUX_IO_URING QSLIST_ENTRY(AioHandler) node_submitted; unsigned flags; /* see fdmon-io_uring.c */ + CqeHandler cqe_handler; #endif int64_t poll_idle_timeout; /* when to stop userspace polling */ bool poll_ready; /* has polling detected an event? */ diff --git a/util/aio-posix.c b/util/aio-posix.c index 800b4debbf..df945312b3 100644 --- a/util/aio-posix.c +++ b/util/aio-posix.c @@ -797,3 +797,12 @@ void aio_context_set_aio_params(AioContext *ctx, int64= _t max_batch) =20 aio_notify(ctx); } + +#ifdef CONFIG_LINUX_IO_URING +void aio_add_sqe(void (*prep_sqe)(struct io_uring_sqe *sqe, void *opaque), + void *opaque, CqeHandler *cqe_handler) +{ + AioContext *ctx =3D qemu_get_current_aio_context(); + ctx->fdmon_ops->add_sqe(ctx, prep_sqe, opaque, cqe_handler); +} +#endif /* CONFIG_LINUX_IO_URING */ diff --git a/util/fdmon-io_uring.c b/util/fdmon-io_uring.c index 3181601767..7f7c50907d 100644 --- a/util/fdmon-io_uring.c +++ b/util/fdmon-io_uring.c @@ -46,6 +46,7 @@ #include "qemu/osdep.h" #include #include "qapi/error.h" +#include "qemu/defer-call.h" #include "qemu/rcu_queue.h" #include "aio-posix.h" =20 @@ -76,8 +77,8 @@ static inline int pfd_events_from_poll(int poll_events) } =20 /* - * Returns an sqe for submitting a request. Only be called within - * fdmon_io_uring_wait(). + * Returns an sqe for submitting a request. Only called from the AioContext + * thread. */ static struct io_uring_sqe *get_sqe(AioContext *ctx) { @@ -168,23 +169,43 @@ static void fdmon_io_uring_update(AioContext *ctx, } } =20 +static void fdmon_io_uring_add_sqe(AioContext *ctx, + void (*prep_sqe)(struct io_uring_sqe *sqe, void *opaque), + void *opaque, CqeHandler *cqe_handler) +{ + struct io_uring_sqe *sqe =3D get_sqe(ctx); + + prep_sqe(sqe, opaque); + io_uring_sqe_set_data(sqe, cqe_handler); +} + +static void fdmon_special_cqe_handler(CqeHandler *cqe_handler) +{ + /* + * This is an empty function that is never called. It is used as a fun= ction + * pointer to distinguish it from ordinary cqe handlers. + */ +} + static void add_poll_add_sqe(AioContext *ctx, AioHandler *node) { struct io_uring_sqe *sqe =3D get_sqe(ctx); int events =3D poll_events_from_pfd(node->pfd.events); =20 io_uring_prep_poll_add(sqe, node->pfd.fd, events); - io_uring_sqe_set_data(sqe, node); + node->cqe_handler.cb =3D fdmon_special_cqe_handler; + io_uring_sqe_set_data(sqe, &node->cqe_handler); } =20 static void add_poll_remove_sqe(AioContext *ctx, AioHandler *node) { struct io_uring_sqe *sqe =3D get_sqe(ctx); + CqeHandler *cqe_handler =3D &node->cqe_handler; =20 #ifdef LIBURING_HAVE_DATA64 - io_uring_prep_poll_remove(sqe, (uintptr_t)node); + io_uring_prep_poll_remove(sqe, (uintptr_t)cqe_handler); #else - io_uring_prep_poll_remove(sqe, node); + io_uring_prep_poll_remove(sqe, cqe_handler); #endif io_uring_sqe_set_data(sqe, NULL); } @@ -233,19 +254,13 @@ static void fill_sq_ring(AioContext *ctx) } } =20 -/* Returns true if a handler became ready */ -static bool process_cqe(AioContext *ctx, - AioHandlerList *ready_list, - struct io_uring_cqe *cqe) +static bool process_cqe_aio_handler(AioContext *ctx, + AioHandlerList *ready_list, + AioHandler *node, + struct io_uring_cqe *cqe) { - AioHandler *node =3D io_uring_cqe_get_data(cqe); unsigned flags; =20 - /* poll_timeout and poll_remove have a zero user_data field */ - if (!node) { - return false; - } - /* * Deletion can only happen when IORING_OP_POLL_ADD completes. If we = race * with enqueue() here then we can safely clear the FDMON_IO_URING_REM= OVE @@ -269,6 +284,61 @@ static bool process_cqe(AioContext *ctx, return true; } =20 +/* Process CqeHandlers from the ready list */ +static void cqe_handler_bh(void *opaque) +{ + AioContext *ctx =3D opaque; + CqeHandlerSimpleQ *ready_list =3D &ctx->cqe_handler_ready_list; + + /* + * If cqe_handler->cb() calls aio_poll() it must continue processing + * ready_list. Schedule a BH so the inner event loop calls us again. + */ + qemu_bh_schedule(ctx->cqe_handler_bh); + + /* Handlers may use defer_call() to coalesce frequent operations */ + defer_call_begin(); + + while (!QSIMPLEQ_EMPTY(ready_list)) { + CqeHandler *cqe_handler =3D QSIMPLEQ_FIRST(ready_list); + + QSIMPLEQ_REMOVE_HEAD(ready_list, next); + + cqe_handler->cb(cqe_handler); + } + + defer_call_end(); + + qemu_bh_cancel(ctx->cqe_handler_bh); +} + +/* Returns true if a handler became ready */ +static bool process_cqe(AioContext *ctx, + AioHandlerList *ready_list, + struct io_uring_cqe *cqe) +{ + CqeHandler *cqe_handler =3D io_uring_cqe_get_data(cqe); + + /* poll_timeout and poll_remove have a zero user_data field */ + if (!cqe_handler) { + return false; + } + + /* + * Special handling for AioHandler cqes. They need ready_list and have= a + * return value. + */ + if (cqe_handler->cb =3D=3D fdmon_special_cqe_handler) { + AioHandler *node =3D container_of(cqe_handler, AioHandler, cqe_han= dler); + return process_cqe_aio_handler(ctx, ready_list, node, cqe); + } + + cqe_handler->cqe =3D *cqe; + QSIMPLEQ_INSERT_TAIL(&ctx->cqe_handler_ready_list, cqe_handler, next); + qemu_bh_schedule(ctx->cqe_handler_bh); + return false; +} + static int process_cq_ring(AioContext *ctx, AioHandlerList *ready_list) { struct io_uring *ring =3D &ctx->fdmon_io_uring; @@ -368,6 +438,7 @@ static const FDMonOps fdmon_io_uring_ops =3D { .gsource_prepare =3D fdmon_io_uring_gsource_prepare, .gsource_check =3D fdmon_io_uring_gsource_check, .gsource_dispatch =3D fdmon_io_uring_gsource_dispatch, + .add_sqe =3D fdmon_io_uring_add_sqe, }; =20 void fdmon_io_uring_setup(AioContext *ctx, Error **errp) @@ -383,6 +454,8 @@ void fdmon_io_uring_setup(AioContext *ctx, Error **errp) } =20 QSLIST_INIT(&ctx->submit_list); + QSIMPLEQ_INIT(&ctx->cqe_handler_ready_list); + ctx->cqe_handler_bh =3D aio_bh_new(ctx, cqe_handler_bh, ctx); ctx->fdmon_ops =3D &fdmon_io_uring_ops; ctx->io_uring_fd_tag =3D g_source_add_unix_fd(&ctx->source, ctx->fdmon_io_uring.ring_fd, G_IO_IN); @@ -390,33 +463,38 @@ void fdmon_io_uring_setup(AioContext *ctx, Error **er= rp) =20 void fdmon_io_uring_destroy(AioContext *ctx) { - if (ctx->fdmon_ops =3D=3D &fdmon_io_uring_ops) { - AioHandler *node; + AioHandler *node; =20 - io_uring_queue_exit(&ctx->fdmon_io_uring); + if (ctx->fdmon_ops !=3D &fdmon_io_uring_ops) { + return; + } =20 - /* Move handlers due to be removed onto the deleted list */ - while ((node =3D QSLIST_FIRST_RCU(&ctx->submit_list))) { - unsigned flags =3D qatomic_fetch_and(&node->flags, - ~(FDMON_IO_URING_PENDING | - FDMON_IO_URING_ADD | - FDMON_IO_URING_REMOVE | - FDMON_IO_URING_DELETE_AIO_HANDLER)); + io_uring_queue_exit(&ctx->fdmon_io_uring); =20 - if ((flags & FDMON_IO_URING_REMOVE) || - (flags & FDMON_IO_URING_DELETE_AIO_HANDLER)) { - QLIST_INSERT_HEAD_RCU(&ctx->deleted_aio_handlers, - node, node_deleted); - } + /* Move handlers due to be removed onto the deleted list */ + while ((node =3D QSLIST_FIRST_RCU(&ctx->submit_list))) { + unsigned flags =3D qatomic_fetch_and(&node->flags, + ~(FDMON_IO_URING_PENDING | + FDMON_IO_URING_ADD | + FDMON_IO_URING_REMOVE | + FDMON_IO_URING_DELETE_AIO_HANDLER)); =20 - QSLIST_REMOVE_HEAD_RCU(&ctx->submit_list, node_submitted); + if ((flags & FDMON_IO_URING_REMOVE) || + (flags & FDMON_IO_URING_DELETE_AIO_HANDLER)) { + QLIST_INSERT_HEAD_RCU(&ctx->deleted_aio_handlers, + node, node_deleted); } =20 - g_source_remove_unix_fd(&ctx->source, ctx->io_uring_fd_tag); - ctx->io_uring_fd_tag =3D NULL; - - qemu_lockcnt_lock(&ctx->list_lock); - fdmon_poll_downgrade(ctx); - qemu_lockcnt_unlock(&ctx->list_lock); + QSLIST_REMOVE_HEAD_RCU(&ctx->submit_list, node_submitted); } + + g_source_remove_unix_fd(&ctx->source, ctx->io_uring_fd_tag); + ctx->io_uring_fd_tag =3D NULL; + + assert(QSIMPLEQ_EMPTY(&ctx->cqe_handler_ready_list)); + qemu_bh_delete(ctx->cqe_handler_bh); + + qemu_lockcnt_lock(&ctx->list_lock); + fdmon_poll_downgrade(ctx); + qemu_lockcnt_unlock(&ctx->list_lock); } --=20 2.51.0 From nobody Mon Sep 22 02:22:19 2025 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=quarantine dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1757527810; cv=none; d=zohomail.com; s=zohoarc; b=AShFiGnlXNiJ66RHOZnI1B517QyPmRUNFszHRV55gxT9BEcXHpxa2QhC4N4we37AXEdKiGRB3jjrJ/sFrxmnsMiJc2M+vYOFXvwtsgISa/dBVwLlotUn+rXPQo1asnU3qj7ro24AycUbiT9/LUTQ3QakHWwuc2Wab/RkeKh4mb0= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1757527810; h=Content-Transfer-Encoding:Cc:Cc:Date:Date:From:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:Subject:To:To:Message-Id:Reply-To; bh=iCOyFpxiCvdCHHGGyD8GBC0zxfD/CZNB4t89d+3dimI=; b=RJ7wcruskMVB81HSHj1O/rpB/NKC6AN6FoSzSO5Zn7GL7MQIanrDFRdwCmDeVQjaxMUWyRUMKJ4YoDpdnqneSXZWcK6fr5DDmcRQrBp0+Ix7cuhdc7/HXfJoRClC9PJcvG36VMhhDWRynzWzg+yubinNExOX50G4a7euvY+/+yg= 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=quarantine dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 175752781082224.433136955741702; Wed, 10 Sep 2025 11:10:10 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1uwPFE-00064i-Kc; Wed, 10 Sep 2025 14:08:43 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1uwPDo-0004mA-J0 for qemu-devel@nongnu.org; Wed, 10 Sep 2025 14:07:13 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.129.124]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1uwPDk-00056s-AG for qemu-devel@nongnu.org; Wed, 10 Sep 2025 14:07:07 -0400 Received: from mx-prod-mc-08.mail-002.prod.us-west-2.aws.redhat.com (ec2-35-165-154-97.us-west-2.compute.amazonaws.com [35.165.154.97]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-401-Wd9dl5K6Phe4gU-vK2TIwQ-1; Wed, 10 Sep 2025 14:06:58 -0400 Received: from mx-prod-int-06.mail-002.prod.us-west-2.aws.redhat.com (mx-prod-int-06.mail-002.prod.us-west-2.aws.redhat.com [10.30.177.93]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by mx-prod-mc-08.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTPS id A84031800452; Wed, 10 Sep 2025 18:06:56 +0000 (UTC) Received: from localhost (unknown [10.2.17.66]) by mx-prod-int-06.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTP id 4E52F1800447; Wed, 10 Sep 2025 18:06:56 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1757527622; 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=iCOyFpxiCvdCHHGGyD8GBC0zxfD/CZNB4t89d+3dimI=; b=JnSENXE5AsROEOSlfQ0XHJ6HfCv5RDX6R3n4smA1K9VaeIci0fkkIBP8J/Je/h1XWTWhGf qhI7R82SYfl3k/+sNo7lOUlOQ+eNT/39HyqJIMS820FCA+A1oTDMiHMEabwn7Er5FieSaI I+BwWvZVW7CfnTkvUO4hvzX9eVwKxnU= X-MC-Unique: Wd9dl5K6Phe4gU-vK2TIwQ-1 X-Mimecast-MFC-AGG-ID: Wd9dl5K6Phe4gU-vK2TIwQ_1757527616 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Aarushi Mehta , Fam Zheng , Stefano Garzarella , Hanna Czenczek , eblake@redhat.com, Stefan Hajnoczi , Kevin Wolf , qemu-block@nongnu.org, hibriansong@gmail.com, Stefan Weil , Paolo Bonzini Subject: [PATCH v4 10/12] aio-posix: avoid EventNotifier for cqe_handler_bh Date: Wed, 10 Sep 2025 13:57:01 -0400 Message-ID: <20250910175703.374499-11-stefanha@redhat.com> In-Reply-To: <20250910175703.374499-1-stefanha@redhat.com> References: <20250910175703.374499-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.4.1 on 10.30.177.93 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: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_NONE=-0.0001, RCVD_IN_MSPIKE_H5=0.001, RCVD_IN_MSPIKE_WL=0.001, RCVD_IN_VALIDITY_RPBL_BLOCKED=0.001, RCVD_IN_VALIDITY_SAFE_BLOCKED=0.001, SPF_HELO_PASS=-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-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1757527812456116600 Content-Type: text/plain; charset="utf-8" fdmon_ops->wait() is called with notify_me enabled. This makes it an expensive place to call qemu_bh_schedule() because aio_notify() invokes write(2) on the EventNotifier. Moving qemu_bh_schedule() after notify_me is reset improves IOPS from 270k to 300k IOPS with --blockdev file,aio=3Dio_uring. I considered alternatives: 1. Introducing a variant of qemu_bh_schedule() that skips aio_notify(). This only makes sense within the AioContext and fdmon implementation itself and is therefore a specialized internal API. I don't like that. 2. Changing fdmon_ops->wait() so implementors can reset notify_me themselves. This makes things complex and the other fdmon implementations don't need it, so it doesn't seem like a good solution. So in the end I moved the qemu_bh_schedule() call from fdmon-io_uring.c to aio-posix.c. It's ugly but straightforward. Signed-off-by: Stefan Hajnoczi Reviewed-by: Eric Blake --- util/aio-posix.c | 11 +++++++++++ util/fdmon-io_uring.c | 11 ++++++++++- 2 files changed, 21 insertions(+), 1 deletion(-) diff --git a/util/aio-posix.c b/util/aio-posix.c index df945312b3..20b11e5650 100644 --- a/util/aio-posix.c +++ b/util/aio-posix.c @@ -700,6 +700,17 @@ bool aio_poll(AioContext *ctx, bool blocking) qatomic_read(&ctx->notify_me) - 2); } =20 +#ifdef CONFIG_LINUX_IO_URING + /* + * This is part of fdmon-io_uring.c but it's more efficient to do it h= ere + * after notify_me has been reset. That way qemu_bh_schedule() -> + * aio_notify() does not write the EventNotifier. + */ + if (!QSIMPLEQ_EMPTY(&ctx->cqe_handler_ready_list)) { + qemu_bh_schedule(ctx->cqe_handler_bh); + } +#endif + aio_notify_accept(ctx); =20 /* Calculate blocked time for adaptive polling */ diff --git a/util/fdmon-io_uring.c b/util/fdmon-io_uring.c index 7f7c50907d..9ed02597d6 100644 --- a/util/fdmon-io_uring.c +++ b/util/fdmon-io_uring.c @@ -334,8 +334,12 @@ static bool process_cqe(AioContext *ctx, } =20 cqe_handler->cqe =3D *cqe; + + /* + * aio_poll() and fdmon_io_uring_gsource_dispatch() schedule cqe_handl= er_bh + * when the list is non-empty. + */ QSIMPLEQ_INSERT_TAIL(&ctx->cqe_handler_ready_list, cqe_handler, next); - qemu_bh_schedule(ctx->cqe_handler_bh); return false; } =20 @@ -386,6 +390,11 @@ static void fdmon_io_uring_gsource_dispatch(AioContext= *ctx, AioHandlerList *ready_list) { process_cq_ring(ctx, ready_list); + + /* Ensure CqeHandlers enqueued by process_cq_ring() will run */ + if (!QSIMPLEQ_EMPTY(&ctx->cqe_handler_ready_list)) { + qemu_bh_schedule(ctx->cqe_handler_bh); + } } =20 static int fdmon_io_uring_wait(AioContext *ctx, AioHandlerList *ready_list, --=20 2.51.0 From nobody Mon Sep 22 02:22:19 2025 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=quarantine dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1757527927; cv=none; d=zohomail.com; s=zohoarc; b=D4Zv9CX1JOagWmfSZLArwYY+a6I0P8Mt1fmwtdopq40PeqEYLsqfIHxFjiYM7elZ3LH6QLNqspBjY85WnVQnPTRthXSxWOs/lgtoRK5yzd0/IBx23xdz7Nl54rzLFKHzMZmMUgkNxOc7mteX/RS1zmBUvjZBlwkLuryXt6aWAW8= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1757527927; h=Content-Transfer-Encoding:Cc:Cc:Date:Date:From:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:Subject:To:To:Message-Id:Reply-To; bh=jFrpdMCA7cXwdgH/5bT+9Uvd3+EuMh14w/XKgtqS/O0=; b=Y14TmF4Fq63Nuzm5Kh61/Xj5KHVFj2nYf6UDOCoRW8ETLdyE+ybN+BpfbSyRYkupua7c4P8BOqMLDq/t88QEYnZxiMZR/eyt3pDvkMg5FhClY6avxyY2sNj01XS9m8qOp8+iM3JGltWVvoKgQA/GHBwMkeVf4qi4oDWNx/rT5iQ= 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=quarantine dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1757527927699839.2083610642009; Wed, 10 Sep 2025 11:12:07 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1uwPGa-0007MF-5H; Wed, 10 Sep 2025 14:10:01 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1uwPDs-0004qU-MB for qemu-devel@nongnu.org; Wed, 10 Sep 2025 14:07:17 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.133.124]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1uwPDm-00057I-SV for qemu-devel@nongnu.org; Wed, 10 Sep 2025 14:07:11 -0400 Received: from mx-prod-mc-02.mail-002.prod.us-west-2.aws.redhat.com (ec2-54-186-198-63.us-west-2.compute.amazonaws.com [54.186.198.63]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-630-wBdvI7K1MoWj_ksp3nowRQ-1; Wed, 10 Sep 2025 14:07:00 -0400 Received: from mx-prod-int-06.mail-002.prod.us-west-2.aws.redhat.com (mx-prod-int-06.mail-002.prod.us-west-2.aws.redhat.com [10.30.177.93]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by mx-prod-mc-02.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTPS id AD490195608A; Wed, 10 Sep 2025 18:06:58 +0000 (UTC) Received: from localhost (unknown [10.2.17.66]) by mx-prod-int-06.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTP id 2675A1800446; Wed, 10 Sep 2025 18:06:57 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1757527623; 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=jFrpdMCA7cXwdgH/5bT+9Uvd3+EuMh14w/XKgtqS/O0=; b=UvYWuQsIW0bYNMSKieGQq8UNlFBmt4cR+ONJzEW3vyTp7T+tQyddQGTdplNhvRJ7nOCTFa /ELsMYUZlTQ6GZ8uYprmYNtS9mRkA/tbqeUmaLlcS5PZvuB9C0iPiWTD+w+k5Y8c+CgJf0 5u6B7ZFAeGutzerHaEa49bwiQHlXEJs= X-MC-Unique: wBdvI7K1MoWj_ksp3nowRQ-1 X-Mimecast-MFC-AGG-ID: wBdvI7K1MoWj_ksp3nowRQ_1757527618 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Aarushi Mehta , Fam Zheng , Stefano Garzarella , Hanna Czenczek , eblake@redhat.com, Stefan Hajnoczi , Kevin Wolf , qemu-block@nongnu.org, hibriansong@gmail.com, Stefan Weil , Paolo Bonzini Subject: [PATCH v4 11/12] block/io_uring: use aio_add_sqe() Date: Wed, 10 Sep 2025 13:57:02 -0400 Message-ID: <20250910175703.374499-12-stefanha@redhat.com> In-Reply-To: <20250910175703.374499-1-stefanha@redhat.com> References: <20250910175703.374499-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.4.1 on 10.30.177.93 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, RCVD_IN_MSPIKE_H5=0.001, RCVD_IN_MSPIKE_WL=0.001, RCVD_IN_VALIDITY_RPBL_BLOCKED=0.001, RCVD_IN_VALIDITY_SAFE_BLOCKED=0.001, SPF_HELO_PASS=-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-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1757527930029116600 Content-Type: text/plain; charset="utf-8" AioContext has its own io_uring instance for file descriptor monitoring. The disk I/O io_uring code was developed separately. Originally I thought the characteristics of file descriptor monitoring and disk I/O were too different, requiring separate io_uring instances. Now it has become clear to me that it's feasible to share a single io_uring instance for file descriptor monitoring and disk I/O. We're not using io_uring's IOPOLL feature or anything else that would require a separate instance. Unify block/io_uring.c and util/fdmon-io_uring.c using the new aio_add_sqe() API that allows user-defined io_uring sqe submission. Now block/io_uring.c just needs to submit readv/writev/fsync and most of the io_uring-specific logic is handled by fdmon-io_uring.c. There are two immediate advantages: 1. Fewer system calls. There is no need to monitor the disk I/O io_uring ring fd from the file descriptor monitoring io_uring instance. Disk I/O completions are now picked up directly. Also, sqes are accumulated in the sq ring until the end of the event loop iteration and there are fewer io_uring_enter(2) syscalls. 2. Less code duplication. Note that error_setg() messages are not supposed to end with punctuation, so I removed a '.' for the non-io_uring build error message. Signed-off-by: Stefan Hajnoczi Reviewed-by: Eric Blake --- include/block/aio.h | 7 - include/block/raw-aio.h | 5 - block/file-posix.c | 40 ++-- block/io_uring.c | 489 ++++++++++------------------------------ stubs/io_uring.c | 32 --- util/async.c | 35 --- util/fdmon-io_uring.c | 6 + block/trace-events | 12 +- stubs/meson.build | 3 - util/trace-events | 4 + 10 files changed, 140 insertions(+), 493 deletions(-) delete mode 100644 stubs/io_uring.c diff --git a/include/block/aio.h b/include/block/aio.h index 56ea0d47b7..60e76d4227 100644 --- a/include/block/aio.h +++ b/include/block/aio.h @@ -291,8 +291,6 @@ struct AioContext { struct LinuxAioState *linux_aio; #endif #ifdef CONFIG_LINUX_IO_URING - LuringState *linux_io_uring; - /* State for file descriptor monitoring using Linux io_uring */ struct io_uring fdmon_io_uring; AioHandlerSList submit_list; @@ -597,11 +595,6 @@ struct LinuxAioState *aio_setup_linux_aio(AioContext *= ctx, Error **errp); /* Return the LinuxAioState bound to this AioContext */ struct LinuxAioState *aio_get_linux_aio(AioContext *ctx); =20 -/* Setup the LuringState bound to this AioContext */ -LuringState *aio_setup_linux_io_uring(AioContext *ctx, Error **errp); - -/* Return the LuringState bound to this AioContext */ -LuringState *aio_get_linux_io_uring(AioContext *ctx); /** * aio_timer_new_with_attrs: * @ctx: the aio context diff --git a/include/block/raw-aio.h b/include/block/raw-aio.h index 6570244496..30e5fc9a9f 100644 --- a/include/block/raw-aio.h +++ b/include/block/raw-aio.h @@ -74,15 +74,10 @@ static inline bool laio_has_fua(void) #endif /* io_uring.c - Linux io_uring implementation */ #ifdef CONFIG_LINUX_IO_URING -LuringState *luring_init(Error **errp); -void luring_cleanup(LuringState *s); - /* luring_co_submit: submit I/O requests in the thread's current AioContex= t. */ int coroutine_fn luring_co_submit(BlockDriverState *bs, int fd, uint64_t o= ffset, QEMUIOVector *qiov, int type, BdrvRequestFlags flags); -void luring_detach_aio_context(LuringState *s, AioContext *old_context); -void luring_attach_aio_context(LuringState *s, AioContext *new_context); bool luring_has_fua(void); #else static inline bool luring_has_fua(void) diff --git a/block/file-posix.c b/block/file-posix.c index 8c738674ce..8b7c02d19a 100644 --- a/block/file-posix.c +++ b/block/file-posix.c @@ -755,14 +755,23 @@ static int raw_open_common(BlockDriverState *bs, QDic= t *options, } #endif /* !defined(CONFIG_LINUX_AIO) */ =20 -#ifndef CONFIG_LINUX_IO_URING if (s->use_linux_io_uring) { +#ifdef CONFIG_LINUX_IO_URING + if (!aio_has_io_uring()) { + error_setg(errp, "aio=3Dio_uring was specified, but is not " + "available (disabled via io_uring_disabled " + "sysctl or blocked by container runtime " + "seccomp policy?)"); + ret =3D -EINVAL; + goto fail; + } +#else error_setg(errp, "aio=3Dio_uring was specified, but is not support= ed " - "in this build."); + "in this build"); ret =3D -EINVAL; goto fail; - } #endif /* !defined(CONFIG_LINUX_IO_URING) */ + } =20 s->has_discard =3D true; s->has_write_zeroes =3D true; @@ -2522,27 +2531,6 @@ static bool bdrv_qiov_is_aligned(BlockDriverState *b= s, QEMUIOVector *qiov) return true; } =20 -#ifdef CONFIG_LINUX_IO_URING -static inline bool raw_check_linux_io_uring(BDRVRawState *s) -{ - Error *local_err =3D NULL; - AioContext *ctx; - - if (!s->use_linux_io_uring) { - return false; - } - - ctx =3D qemu_get_current_aio_context(); - if (unlikely(!aio_setup_linux_io_uring(ctx, &local_err))) { - error_reportf_err(local_err, "Unable to use linux io_uring, " - "falling back to thread pool: "); - s->use_linux_io_uring =3D false; - return false; - } - return true; -} -#endif - #ifdef CONFIG_LINUX_AIO static inline bool raw_check_linux_aio(BDRVRawState *s) { @@ -2595,7 +2583,7 @@ raw_co_prw(BlockDriverState *bs, int64_t *offset_ptr,= uint64_t bytes, if (s->needs_alignment && !bdrv_qiov_is_aligned(bs, qiov)) { type |=3D QEMU_AIO_MISALIGNED; #ifdef CONFIG_LINUX_IO_URING - } else if (raw_check_linux_io_uring(s)) { + } else if (s->use_linux_io_uring) { assert(qiov->size =3D=3D bytes); ret =3D luring_co_submit(bs, s->fd, offset, qiov, type, flags); goto out; @@ -2692,7 +2680,7 @@ static int coroutine_fn raw_co_flush_to_disk(BlockDri= verState *bs) }; =20 #ifdef CONFIG_LINUX_IO_URING - if (raw_check_linux_io_uring(s)) { + if (s->use_linux_io_uring) { return luring_co_submit(bs, s->fd, 0, NULL, QEMU_AIO_FLUSH, 0); } #endif diff --git a/block/io_uring.c b/block/io_uring.c index dd4f304910..dd930ee57e 100644 --- a/block/io_uring.c +++ b/block/io_uring.c @@ -11,28 +11,20 @@ #include "qemu/osdep.h" #include #include "block/aio.h" -#include "qemu/queue.h" #include "block/block.h" #include "block/raw-aio.h" #include "qemu/coroutine.h" -#include "qemu/defer-call.h" -#include "qapi/error.h" #include "system/block-backend.h" #include "trace.h" =20 -/* Only used for assertions. */ -#include "qemu/coroutine_int.h" - -/* io_uring ring size */ -#define MAX_ENTRIES 128 - -typedef struct LuringAIOCB { +typedef struct { Coroutine *co; - struct io_uring_sqe sqeq; - ssize_t ret; QEMUIOVector *qiov; - bool is_read; - QSIMPLEQ_ENTRY(LuringAIOCB) next; + uint64_t offset; + ssize_t ret; + int type; + int fd; + BdrvRequestFlags flags; =20 /* * Buffered reads may require resubmission, see @@ -40,36 +32,51 @@ typedef struct LuringAIOCB { */ int total_read; QEMUIOVector resubmit_qiov; -} LuringAIOCB; =20 -typedef struct LuringQueue { - unsigned int in_queue; - unsigned int in_flight; - bool blocked; - QSIMPLEQ_HEAD(, LuringAIOCB) submit_queue; -} LuringQueue; + CqeHandler cqe_handler; +} LuringRequest; =20 -struct LuringState { - AioContext *aio_context; - - struct io_uring ring; - - /* No locking required, only accessed from AioContext home thread */ - LuringQueue io_q; - - QEMUBH *completion_bh; -}; - -/** - * luring_resubmit: - * - * Resubmit a request by appending it to submit_queue. The caller must en= sure - * that ioq_submit() is called later so that submit_queue requests are sta= rted. - */ -static void luring_resubmit(LuringState *s, LuringAIOCB *luringcb) +static void luring_prep_sqe(struct io_uring_sqe *sqe, void *opaque) { - QSIMPLEQ_INSERT_TAIL(&s->io_q.submit_queue, luringcb, next); - s->io_q.in_queue++; + LuringRequest *req =3D opaque; + QEMUIOVector *qiov =3D req->qiov; + uint64_t offset =3D req->offset; + int fd =3D req->fd; + BdrvRequestFlags flags =3D req->flags; + + switch (req->type) { + case QEMU_AIO_WRITE: +#ifdef HAVE_IO_URING_PREP_WRITEV2 + { + int luring_flags =3D (flags & BDRV_REQ_FUA) ? RWF_DSYNC : 0; + io_uring_prep_writev2(sqe, fd, qiov->iov, + qiov->niov, offset, luring_flags); + } +#else + assert(flags =3D=3D 0); + io_uring_prep_writev(sqe, fd, qiov->iov, qiov->niov, offset); +#endif + break; + case QEMU_AIO_ZONE_APPEND: + io_uring_prep_writev(sqe, fd, qiov->iov, qiov->niov, offset); + break; + case QEMU_AIO_READ: + { + if (req->resubmit_qiov.iov !=3D NULL) { + qiov =3D &req->resubmit_qiov; + } + io_uring_prep_readv(sqe, fd, qiov->iov, qiov->niov, + offset + req->total_read); + break; + } + case QEMU_AIO_FLUSH: + io_uring_prep_fsync(sqe, fd, IORING_FSYNC_DATASYNC); + break; + default: + fprintf(stderr, "%s: invalid AIO request type, aborting 0x%x.\n", + __func__, req->type); + abort(); + } } =20 /** @@ -78,385 +85,115 @@ static void luring_resubmit(LuringState *s, LuringAIO= CB *luringcb) * Short reads are rare but may occur. The remaining read request needs to= be * resubmitted. */ -static void luring_resubmit_short_read(LuringState *s, LuringAIOCB *luring= cb, - int nread) +static void luring_resubmit_short_read(LuringRequest *req, int nread) { QEMUIOVector *resubmit_qiov; size_t remaining; =20 - trace_luring_resubmit_short_read(s, luringcb, nread); + trace_luring_resubmit_short_read(req, nread); =20 /* Update read position */ - luringcb->total_read +=3D nread; - remaining =3D luringcb->qiov->size - luringcb->total_read; + req->total_read +=3D nread; + remaining =3D req->qiov->size - req->total_read; =20 /* Shorten qiov */ - resubmit_qiov =3D &luringcb->resubmit_qiov; + resubmit_qiov =3D &req->resubmit_qiov; if (resubmit_qiov->iov =3D=3D NULL) { - qemu_iovec_init(resubmit_qiov, luringcb->qiov->niov); + qemu_iovec_init(resubmit_qiov, req->qiov->niov); } else { qemu_iovec_reset(resubmit_qiov); } - qemu_iovec_concat(resubmit_qiov, luringcb->qiov, luringcb->total_read, - remaining); + qemu_iovec_concat(resubmit_qiov, req->qiov, req->total_read, remaining= ); =20 - /* Update sqe */ - luringcb->sqeq.off +=3D nread; - luringcb->sqeq.addr =3D (uintptr_t)luringcb->resubmit_qiov.iov; - luringcb->sqeq.len =3D luringcb->resubmit_qiov.niov; - - luring_resubmit(s, luringcb); + aio_add_sqe(luring_prep_sqe, req, &req->cqe_handler); } =20 -/** - * luring_process_completions: - * @s: AIO state - * - * Fetches completed I/O requests, consumes cqes and invokes their callbac= ks - * The function is somewhat tricky because it supports nested event loops,= for - * example when a request callback invokes aio_poll(). - * - * Function schedules BH completion so it can be called again in a nested - * event loop. When there are no events left to complete the BH is being - * canceled. - * - */ -static void luring_process_completions(LuringState *s) +static void luring_cqe_handler(CqeHandler *cqe_handler) { - struct io_uring_cqe *cqes; - int total_bytes; + LuringRequest *req =3D container_of(cqe_handler, LuringRequest, cqe_ha= ndler); + int ret =3D cqe_handler->cqe.res; =20 - defer_call_begin(); + trace_luring_cqe_handler(req, ret); =20 - /* - * Request completion callbacks can run the nested event loop. - * Schedule ourselves so the nested event loop will "see" remaining - * completed requests and process them. Without this, completion - * callbacks that wait for other requests using a nested event loop - * would hang forever. - * - * This workaround is needed because io_uring uses poll_wait, which - * is woken up when new events are added to the uring, thus polling on - * the same uring fd will block unless more events are received. - * - * Other leaf block drivers (drivers that access the data themselves) - * are networking based, so they poll sockets for data and run the - * correct coroutine. - */ - qemu_bh_schedule(s->completion_bh); - - while (io_uring_peek_cqe(&s->ring, &cqes) =3D=3D 0) { - LuringAIOCB *luringcb; - int ret; - - if (!cqes) { - break; + if (ret < 0) { + /* + * Only writev/readv/fsync requests on regular files or host block + * devices are submitted. Therefore -EAGAIN is not expected but it= 's + * known to happen sometimes with Linux SCSI. Submit again and hope + * the request completes successfully. + * + * For more information, see: + * https://lore.kernel.org/io-uring/20210727165811.284510-3-axboe@= kernel.dk/T/#u + * + * If the code is changed to submit other types of requests in the + * future, then this workaround may need to be extended to deal wi= th + * genuine -EAGAIN results that should not be resubmitted + * immediately. + */ + if (ret =3D=3D -EINTR || ret =3D=3D -EAGAIN) { + aio_add_sqe(luring_prep_sqe, req, &req->cqe_handler); + return; } - - luringcb =3D io_uring_cqe_get_data(cqes); - ret =3D cqes->res; - io_uring_cqe_seen(&s->ring, cqes); - cqes =3D NULL; - - /* Change counters one-by-one because we can be nested. */ - s->io_q.in_flight--; - trace_luring_process_completion(s, luringcb, ret); - + } else if (req->qiov) { /* total_read is non-zero only for resubmitted read requests */ - total_bytes =3D ret + luringcb->total_read; + int total_bytes =3D ret + req->total_read; =20 - if (ret < 0) { - /* - * Only writev/readv/fsync requests on regular files or host b= lock - * devices are submitted. Therefore -EAGAIN is not expected bu= t it's - * known to happen sometimes with Linux SCSI. Submit again and= hope - * the request completes successfully. - * - * For more information, see: - * https://lore.kernel.org/io-uring/20210727165811.284510-3-ax= boe@kernel.dk/T/#u - * - * If the code is changed to submit other types of requests in= the - * future, then this workaround may need to be extended to dea= l with - * genuine -EAGAIN results that should not be resubmitted - * immediately. - */ - if (ret =3D=3D -EINTR || ret =3D=3D -EAGAIN) { - luring_resubmit(s, luringcb); - continue; - } - } else if (!luringcb->qiov) { - goto end; - } else if (total_bytes =3D=3D luringcb->qiov->size) { + if (total_bytes =3D=3D req->qiov->size) { ret =3D 0; - /* Only read/write */ } else { /* Short Read/Write */ - if (luringcb->is_read) { + if (req->type =3D=3D QEMU_AIO_READ) { if (ret > 0) { - luring_resubmit_short_read(s, luringcb, ret); - continue; - } else { - /* Pad with zeroes */ - qemu_iovec_memset(luringcb->qiov, total_bytes, 0, - luringcb->qiov->size - total_bytes); - ret =3D 0; + luring_resubmit_short_read(req, ret); + return; } + + /* Pad with zeroes */ + qemu_iovec_memset(req->qiov, total_bytes, 0, + req->qiov->size - total_bytes); + ret =3D 0; } else { ret =3D -ENOSPC; } } -end: - luringcb->ret =3D ret; - qemu_iovec_destroy(&luringcb->resubmit_qiov); - - /* - * If the coroutine is already entered it must be in ioq_submit() - * and will notice luringcb->ret has been filled in when it - * eventually runs later. Coroutines cannot be entered recursively - * so avoid doing that! - */ - assert(luringcb->co->ctx =3D=3D s->aio_context); - if (!qemu_coroutine_entered(luringcb->co)) { - aio_co_wake(luringcb->co); - } } =20 - qemu_bh_cancel(s->completion_bh); + req->ret =3D ret; + qemu_iovec_destroy(&req->resubmit_qiov); =20 - defer_call_end(); -} - -static int ioq_submit(LuringState *s) -{ - int ret =3D 0; - LuringAIOCB *luringcb, *luringcb_next; - - while (s->io_q.in_queue > 0) { - /* - * Try to fetch sqes from the ring for requests waiting in - * the overflow queue - */ - QSIMPLEQ_FOREACH_SAFE(luringcb, &s->io_q.submit_queue, next, - luringcb_next) { - struct io_uring_sqe *sqes =3D io_uring_get_sqe(&s->ring); - if (!sqes) { - break; - } - /* Prep sqe for submission */ - *sqes =3D luringcb->sqeq; - QSIMPLEQ_REMOVE_HEAD(&s->io_q.submit_queue, next); - } - ret =3D io_uring_submit(&s->ring); - trace_luring_io_uring_submit(s, ret); - /* Prevent infinite loop if submission is refused */ - if (ret <=3D 0) { - if (ret =3D=3D -EAGAIN || ret =3D=3D -EINTR) { - continue; - } - break; - } - s->io_q.in_flight +=3D ret; - s->io_q.in_queue -=3D ret; - } - s->io_q.blocked =3D (s->io_q.in_queue > 0); - - if (s->io_q.in_flight) { - /* - * We can try to complete something just right away if there are - * still requests in-flight. - */ - luring_process_completions(s); - } - return ret; -} - -static void luring_process_completions_and_submit(LuringState *s) -{ - luring_process_completions(s); - - if (s->io_q.in_queue > 0) { - ioq_submit(s); + /* + * If the coroutine is already entered it must be in luring_co_submit(= ) and + * will notice req->ret has been filled in when it eventually runs lat= er. + * Coroutines cannot be entered recursively so avoid doing that! + */ + if (!qemu_coroutine_entered(req->co)) { + aio_co_wake(req->co); } } =20 -static void qemu_luring_completion_bh(void *opaque) +int coroutine_fn luring_co_submit(BlockDriverState *bs, int fd, + uint64_t offset, QEMUIOVector *qiov, + int type, BdrvRequestFlags flags) { - LuringState *s =3D opaque; - luring_process_completions_and_submit(s); -} - -static void qemu_luring_completion_cb(void *opaque) -{ - LuringState *s =3D opaque; - luring_process_completions_and_submit(s); -} - -static bool qemu_luring_poll_cb(void *opaque) -{ - LuringState *s =3D opaque; - - return io_uring_cq_ready(&s->ring); -} - -static void qemu_luring_poll_ready(void *opaque) -{ - LuringState *s =3D opaque; - - luring_process_completions_and_submit(s); -} - -static void ioq_init(LuringQueue *io_q) -{ - QSIMPLEQ_INIT(&io_q->submit_queue); - io_q->in_queue =3D 0; - io_q->in_flight =3D 0; - io_q->blocked =3D false; -} - -static void luring_deferred_fn(void *opaque) -{ - LuringState *s =3D opaque; - trace_luring_unplug_fn(s, s->io_q.blocked, s->io_q.in_queue, - s->io_q.in_flight); - if (!s->io_q.blocked && s->io_q.in_queue > 0) { - ioq_submit(s); - } -} - -/** - * luring_do_submit: - * @fd: file descriptor for I/O - * @luringcb: AIO control block - * @s: AIO state - * @offset: offset for request - * @type: type of request - * - * Fetches sqes from ring, adds to pending queue and preps them - * - */ -static int luring_do_submit(int fd, LuringAIOCB *luringcb, LuringState *s, - uint64_t offset, int type, BdrvRequestFlags fl= ags) -{ - int ret; - struct io_uring_sqe *sqes =3D &luringcb->sqeq; - - switch (type) { - case QEMU_AIO_WRITE: -#ifdef HAVE_IO_URING_PREP_WRITEV2 - { - int luring_flags =3D (flags & BDRV_REQ_FUA) ? RWF_DSYNC : 0; - io_uring_prep_writev2(sqes, fd, luringcb->qiov->iov, - luringcb->qiov->niov, offset, luring_flags); - } -#else - assert(flags =3D=3D 0); - io_uring_prep_writev(sqes, fd, luringcb->qiov->iov, - luringcb->qiov->niov, offset); -#endif - break; - case QEMU_AIO_ZONE_APPEND: - io_uring_prep_writev(sqes, fd, luringcb->qiov->iov, - luringcb->qiov->niov, offset); - break; - case QEMU_AIO_READ: - io_uring_prep_readv(sqes, fd, luringcb->qiov->iov, - luringcb->qiov->niov, offset); - break; - case QEMU_AIO_FLUSH: - io_uring_prep_fsync(sqes, fd, IORING_FSYNC_DATASYNC); - break; - default: - fprintf(stderr, "%s: invalid AIO request type, aborting 0x%x.\n", - __func__, type); - abort(); - } - io_uring_sqe_set_data(sqes, luringcb); - - QSIMPLEQ_INSERT_TAIL(&s->io_q.submit_queue, luringcb, next); - s->io_q.in_queue++; - trace_luring_do_submit(s, s->io_q.blocked, s->io_q.in_queue, - s->io_q.in_flight); - if (!s->io_q.blocked) { - if (s->io_q.in_flight + s->io_q.in_queue >=3D MAX_ENTRIES) { - ret =3D ioq_submit(s); - trace_luring_do_submit_done(s, ret); - return ret; - } - - defer_call(luring_deferred_fn, s); - } - return 0; -} - -int coroutine_fn luring_co_submit(BlockDriverState *bs, int fd, uint64_t o= ffset, - QEMUIOVector *qiov, int type, - BdrvRequestFlags flags) -{ - int ret; - AioContext *ctx =3D qemu_get_current_aio_context(); - LuringState *s =3D aio_get_linux_io_uring(ctx); - LuringAIOCB luringcb =3D { + LuringRequest req =3D { .co =3D qemu_coroutine_self(), - .ret =3D -EINPROGRESS, .qiov =3D qiov, - .is_read =3D (type =3D=3D QEMU_AIO_READ), + .ret =3D -EINPROGRESS, + .type =3D type, + .fd =3D fd, + .offset =3D offset, + .flags =3D flags, }; - trace_luring_co_submit(bs, s, &luringcb, fd, offset, qiov ? qiov->size= : 0, - type); - ret =3D luring_do_submit(fd, &luringcb, s, offset, type, flags); =20 - if (ret < 0) { - return ret; - } + req.cqe_handler.cb =3D luring_cqe_handler; =20 - if (luringcb.ret =3D=3D -EINPROGRESS) { + trace_luring_co_submit(bs, &req, fd, offset, qiov ? qiov->size : 0, ty= pe); + aio_add_sqe(luring_prep_sqe, &req, &req.cqe_handler); + + if (req.ret =3D=3D -EINPROGRESS) { qemu_coroutine_yield(); } - return luringcb.ret; -} - -void luring_detach_aio_context(LuringState *s, AioContext *old_context) -{ - aio_set_fd_handler(old_context, s->ring.ring_fd, - NULL, NULL, NULL, NULL, s); - qemu_bh_delete(s->completion_bh); - s->aio_context =3D NULL; -} - -void luring_attach_aio_context(LuringState *s, AioContext *new_context) -{ - s->aio_context =3D new_context; - s->completion_bh =3D aio_bh_new(new_context, qemu_luring_completion_bh= , s); - aio_set_fd_handler(s->aio_context, s->ring.ring_fd, - qemu_luring_completion_cb, NULL, - qemu_luring_poll_cb, qemu_luring_poll_ready, s); -} - -LuringState *luring_init(Error **errp) -{ - int rc; - LuringState *s =3D g_new0(LuringState, 1); - struct io_uring *ring =3D &s->ring; - - trace_luring_init_state(s, sizeof(*s)); - - rc =3D io_uring_queue_init(MAX_ENTRIES, ring, 0); - if (rc < 0) { - error_setg_errno(errp, -rc, "failed to init linux io_uring ring"); - g_free(s); - return NULL; - } - - ioq_init(&s->io_q); - return s; - -} - -void luring_cleanup(LuringState *s) -{ - io_uring_queue_exit(&s->ring); - trace_luring_cleanup_state(s); - g_free(s); + return req.ret; } =20 bool luring_has_fua(void) diff --git a/stubs/io_uring.c b/stubs/io_uring.c deleted file mode 100644 index 622d1e4648..0000000000 --- a/stubs/io_uring.c +++ /dev/null @@ -1,32 +0,0 @@ -/* - * Linux io_uring support. - * - * Copyright (C) 2009 IBM, Corp. - * Copyright (C) 2009 Red Hat, Inc. - * - * This work is licensed under the terms of the GNU GPL, version 2 or late= r. - * See the COPYING file in the top-level directory. - */ -#include "qemu/osdep.h" -#include "block/aio.h" -#include "block/raw-aio.h" - -void luring_detach_aio_context(LuringState *s, AioContext *old_context) -{ - abort(); -} - -void luring_attach_aio_context(LuringState *s, AioContext *new_context) -{ - abort(); -} - -LuringState *luring_init(Error **errp) -{ - abort(); -} - -void luring_cleanup(LuringState *s) -{ - abort(); -} diff --git a/util/async.c b/util/async.c index 2bb7e1f3c1..6a880714d9 100644 --- a/util/async.c +++ b/util/async.c @@ -383,14 +383,6 @@ aio_ctx_finalize(GSource *source) } #endif =20 -#ifdef CONFIG_LINUX_IO_URING - if (ctx->linux_io_uring) { - luring_detach_aio_context(ctx->linux_io_uring, ctx); - luring_cleanup(ctx->linux_io_uring); - ctx->linux_io_uring =3D NULL; - } -#endif - assert(QSLIST_EMPTY(&ctx->scheduled_coroutines)); qemu_bh_delete(ctx->co_schedule_bh); =20 @@ -465,29 +457,6 @@ LinuxAioState *aio_get_linux_aio(AioContext *ctx) } #endif =20 -#ifdef CONFIG_LINUX_IO_URING -LuringState *aio_setup_linux_io_uring(AioContext *ctx, Error **errp) -{ - if (ctx->linux_io_uring) { - return ctx->linux_io_uring; - } - - ctx->linux_io_uring =3D luring_init(errp); - if (!ctx->linux_io_uring) { - return NULL; - } - - luring_attach_aio_context(ctx->linux_io_uring, ctx); - return ctx->linux_io_uring; -} - -LuringState *aio_get_linux_io_uring(AioContext *ctx) -{ - assert(ctx->linux_io_uring); - return ctx->linux_io_uring; -} -#endif - void aio_notify(AioContext *ctx) { /* @@ -611,10 +580,6 @@ AioContext *aio_context_new(Error **errp) ctx->linux_aio =3D NULL; #endif =20 -#ifdef CONFIG_LINUX_IO_URING - ctx->linux_io_uring =3D NULL; -#endif - ctx->thread_pool =3D NULL; qemu_rec_mutex_init(&ctx->lock); timerlistgroup_init(&ctx->tlg, aio_timerlist_notify, ctx); diff --git a/util/fdmon-io_uring.c b/util/fdmon-io_uring.c index 9ed02597d6..28b93c8ab9 100644 --- a/util/fdmon-io_uring.c +++ b/util/fdmon-io_uring.c @@ -49,6 +49,7 @@ #include "qemu/defer-call.h" #include "qemu/rcu_queue.h" #include "aio-posix.h" +#include "trace.h" =20 enum { FDMON_IO_URING_ENTRIES =3D 128, /* sq/cq ring size */ @@ -177,6 +178,9 @@ static void fdmon_io_uring_add_sqe(AioContext *ctx, =20 prep_sqe(sqe, opaque); io_uring_sqe_set_data(sqe, cqe_handler); + + trace_fdmon_io_uring_add_sqe(ctx, opaque, sqe->opcode, sqe->fd, sqe->o= ff, + cqe_handler); } =20 static void fdmon_special_cqe_handler(CqeHandler *cqe_handler) @@ -304,6 +308,8 @@ static void cqe_handler_bh(void *opaque) =20 QSIMPLEQ_REMOVE_HEAD(ready_list, next); =20 + trace_fdmon_io_uring_cqe_handler(ctx, cqe_handler, + cqe_handler->cqe.res); cqe_handler->cb(cqe_handler); } =20 diff --git a/block/trace-events b/block/trace-events index 8e789e1f12..c9b4736ff8 100644 --- a/block/trace-events +++ b/block/trace-events @@ -62,15 +62,9 @@ qmp_block_stream(void *bs) "bs %p" file_paio_submit(void *acb, void *opaque, int64_t offset, int count, int t= ype) "acb %p opaque %p offset %"PRId64" count %d type %d" =20 # io_uring.c -luring_init_state(void *s, size_t size) "s %p size %zu" -luring_cleanup_state(void *s) "%p freed" -luring_unplug_fn(void *s, int blocked, int queued, int inflight) "LuringSt= ate %p blocked %d queued %d inflight %d" -luring_do_submit(void *s, int blocked, int queued, int inflight) "LuringSt= ate %p blocked %d queued %d inflight %d" -luring_do_submit_done(void *s, int ret) "LuringState %p submitted to kerne= l %d" -luring_co_submit(void *bs, void *s, void *luringcb, int fd, uint64_t offse= t, size_t nbytes, int type) "bs %p s %p luringcb %p fd %d offset %" PRId64 = " nbytes %zd type %d" -luring_process_completion(void *s, void *aiocb, int ret) "LuringState %p l= uringcb %p ret %d" -luring_io_uring_submit(void *s, int ret) "LuringState %p ret %d" -luring_resubmit_short_read(void *s, void *luringcb, int nread) "LuringStat= e %p luringcb %p nread %d" +luring_cqe_handler(void *req, int ret) "req %p ret %d" +luring_co_submit(void *bs, void *req, int fd, uint64_t offset, size_t nbyt= es, int type) "bs %p req %p fd %d offset %" PRId64 " nbytes %zd type %d" +luring_resubmit_short_read(void *req, int nread) "req %p nread %d" =20 # qcow2.c qcow2_add_task(void *co, void *bs, void *pool, const char *action, int clu= ster_type, uint64_t host_offset, uint64_t offset, uint64_t bytes, void *qio= v, size_t qiov_offset) "co %p bs %p pool %p: %s: cluster_type %d file_clust= er_offset %" PRIu64 " offset %" PRIu64 " bytes %" PRIu64 " qiov %p qiov_off= set %zu" diff --git a/stubs/meson.build b/stubs/meson.build index cef046e685..93aa90c817 100644 --- a/stubs/meson.build +++ b/stubs/meson.build @@ -32,9 +32,6 @@ if have_block or have_ga stub_ss.add(files('cpus-virtual-clock.c')) stub_ss.add(files('icount.c')) stub_ss.add(files('graph-lock.c')) - if linux_io_uring.found() - stub_ss.add(files('io_uring.c')) - endif if libaio.found() stub_ss.add(files('linux-aio.c')) endif diff --git a/util/trace-events b/util/trace-events index bd8f25fb59..540d662507 100644 --- a/util/trace-events +++ b/util/trace-events @@ -24,6 +24,10 @@ buffer_move_empty(const char *buf, size_t len, const cha= r *from) "%s: %zd bytes buffer_move(const char *buf, size_t len, const char *from) "%s: %zd bytes = from %s" buffer_free(const char *buf, size_t len) "%s: capacity %zd" =20 +# fdmon-io_uring.c +fdmon_io_uring_add_sqe(void *ctx, void *opaque, int opcode, int fd, uint64= _t off, void *cqe_handler) "ctx %p opaque %p opcode %d fd %d off %"PRId64" = cqe_handler %p" +fdmon_io_uring_cqe_handler(void *ctx, void *cqe_handler, int cqe_res) "ctx= %p cqe_handler %p cqe_res %d" + # filemonitor-inotify.c qemu_file_monitor_add_watch(void *mon, const char *dirpath, const char *fi= lename, void *cb, void *opaque, int64_t id) "File monitor %p add watch dir= =3D'%s' file=3D'%s' cb=3D%p opaque=3D%p id=3D%" PRId64 qemu_file_monitor_remove_watch(void *mon, const char *dirpath, int64_t id)= "File monitor %p remove watch dir=3D'%s' id=3D%" PRId64 --=20 2.51.0 From nobody Mon Sep 22 02:22:19 2025 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=quarantine dis=none) header.from=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1757527866; cv=none; d=zohomail.com; s=zohoarc; b=DuK3IkrO9OSY3hOQhvOwSr+YO8wNj762mt6Q6+KwGF6S1TFUMCnexP0fmyWEbUoUp42vMbeIejzPJHKCHfvgz7ENJpWUvHIDqUhAYMEcu3j3bo+Xibul+W4qUJles3NfaCJVOUZi3G6/BQyh6y1xNp5yAaEE453+b6QMC49SUmU= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1757527866; h=Content-Transfer-Encoding:Cc:Cc:Date:Date:From:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:Subject:To:To:Message-Id:Reply-To; bh=m7ShvVxcohLryooGwoHjde/Zqt4sTaKcWehk8sb96Oo=; b=DRi+eIZvDd/lT4eSiJkTYX3bOjWn8K9ZEOCduWPGbvmSmQvyQZ4IQQg6uy3An9E8yIfpBBVmwDPlT16dDxt7ExhM1kvUgV6rcBSF/kYdvufHIHmqAbv7Rl8PmX2lksoLCo+iGkla7KR+6qyi1K3ytwIdaweVEMaIvk5LpmX6Mlg= 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=quarantine dis=none) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1757527866435308.85834347186346; Wed, 10 Sep 2025 11:11:06 -0700 (PDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1uwPGC-0006c6-S7; Wed, 10 Sep 2025 14:09:42 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1uwPDp-0004mi-Fm for qemu-devel@nongnu.org; Wed, 10 Sep 2025 14:07:15 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.133.124]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1uwPDm-00057Z-Tc for qemu-devel@nongnu.org; Wed, 10 Sep 2025 14:07:08 -0400 Received: from mx-prod-mc-08.mail-002.prod.us-west-2.aws.redhat.com (ec2-35-165-154-97.us-west-2.compute.amazonaws.com [35.165.154.97]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-122-iNFkpZrvP42qcidO9zzaaw-1; Wed, 10 Sep 2025 14:07:02 -0400 Received: from mx-prod-int-05.mail-002.prod.us-west-2.aws.redhat.com (mx-prod-int-05.mail-002.prod.us-west-2.aws.redhat.com [10.30.177.17]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by mx-prod-mc-08.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTPS id B5AC9180047F; Wed, 10 Sep 2025 18:07:00 +0000 (UTC) Received: from localhost (unknown [10.2.17.66]) by mx-prod-int-05.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTP id 325B91956095; Wed, 10 Sep 2025 18:06:59 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1757527625; 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=m7ShvVxcohLryooGwoHjde/Zqt4sTaKcWehk8sb96Oo=; b=Ak05ZdTSwk/0fE7kHctvD7J64zSqOD5h64ppvy4JkZYy8vyBjfLLalpQ03Qa8AMKXmhZoj 0PacsUhFDeh7n0kSo1HFh4pCNAThKQwcAF7No4sRdhqBwyUEbCYhTJZ0JWhxpwyeoDJ6dG QkXQUqL81mUyrMJLyuCESpU4X9ZumbM= X-MC-Unique: iNFkpZrvP42qcidO9zzaaw-1 X-Mimecast-MFC-AGG-ID: iNFkpZrvP42qcidO9zzaaw_1757527620 From: Stefan Hajnoczi To: qemu-devel@nongnu.org Cc: Aarushi Mehta , Fam Zheng , Stefano Garzarella , Hanna Czenczek , eblake@redhat.com, Stefan Hajnoczi , Kevin Wolf , qemu-block@nongnu.org, hibriansong@gmail.com, Stefan Weil , Paolo Bonzini Subject: [PATCH v4 12/12] block/io_uring: use non-vectored read/write when possible Date: Wed, 10 Sep 2025 13:57:03 -0400 Message-ID: <20250910175703.374499-13-stefanha@redhat.com> In-Reply-To: <20250910175703.374499-1-stefanha@redhat.com> References: <20250910175703.374499-1-stefanha@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.0 on 10.30.177.17 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, RCVD_IN_MSPIKE_H5=0.001, RCVD_IN_MSPIKE_WL=0.001, RCVD_IN_VALIDITY_RPBL_BLOCKED=0.001, RCVD_IN_VALIDITY_SAFE_BLOCKED=0.001, SPF_HELO_PASS=-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-bounces+importer=patchew.org@nongnu.org X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1757527866825116600 Content-Type: text/plain; charset="utf-8" The io_uring_prep_readv2/writev2() man pages recommend using the non-vectored read/write operations when possible for performance reasons. I didn't measure a significant difference but it doesn't hurt to have this optimization in place. Suggested-by: Eric Blake Signed-off-by: Stefan Hajnoczi --- block/io_uring.c | 29 ++++++++++++++++++++++++----- 1 file changed, 24 insertions(+), 5 deletions(-) diff --git a/block/io_uring.c b/block/io_uring.c index dd930ee57e..bbefbddcc0 100644 --- a/block/io_uring.c +++ b/block/io_uring.c @@ -49,12 +49,24 @@ static void luring_prep_sqe(struct io_uring_sqe *sqe, v= oid *opaque) #ifdef HAVE_IO_URING_PREP_WRITEV2 { int luring_flags =3D (flags & BDRV_REQ_FUA) ? RWF_DSYNC : 0; - io_uring_prep_writev2(sqe, fd, qiov->iov, - qiov->niov, offset, luring_flags); + if (luring_flags !=3D 0 || qiov->niov > 1) { + io_uring_prep_writev2(sqe, fd, qiov->iov, + qiov->niov, offset, luring_flags); + } else { + /* The man page says non-vectored is faster than vectored */ + struct iovec *iov =3D qiov->iov; + io_uring_prep_write(sqe, fd, iov->iov_base, iov->iov_len, offs= et); + } } #else assert(flags =3D=3D 0); - io_uring_prep_writev(sqe, fd, qiov->iov, qiov->niov, offset); + if (qiov->niov > 1) { + io_uring_prep_writev(sqe, fd, qiov->iov, qiov->niov, offset); + } else { + /* The man page says non-vectored is faster than vectored */ + struct iovec *iov =3D qiov->iov; + io_uring_prep_write(sqe, fd, iov->iov_base, iov->iov_len, offs= et); + } #endif break; case QEMU_AIO_ZONE_APPEND: @@ -65,8 +77,15 @@ static void luring_prep_sqe(struct io_uring_sqe *sqe, vo= id *opaque) if (req->resubmit_qiov.iov !=3D NULL) { qiov =3D &req->resubmit_qiov; } - io_uring_prep_readv(sqe, fd, qiov->iov, qiov->niov, - offset + req->total_read); + if (qiov->niov > 1) { + io_uring_prep_readv(sqe, fd, qiov->iov, qiov->niov, + offset + req->total_read); + } else { + /* The man page says non-vectored is faster than vectored */ + struct iovec *iov =3D qiov->iov; + io_uring_prep_read(sqe, fd, iov->iov_base, iov->iov_len, + offset + req->total_read); + } break; } case QEMU_AIO_FLUSH: --=20 2.51.0