From patchwork Tue Aug 16 15:37:23 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Dylan Yudaken X-Patchwork-Id: 12945036 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 4A76BC2BB41 for ; Tue, 16 Aug 2022 15:38:56 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S230050AbiHPPiz (ORCPT ); Tue, 16 Aug 2022 11:38:55 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:51434 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236065AbiHPPiT (ORCPT ); Tue, 16 Aug 2022 11:38:19 -0400 Received: from mx0a-00082601.pphosted.com (mx0a-00082601.pphosted.com [67.231.145.42]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 7D62D5FE8 for ; Tue, 16 Aug 2022 08:37:47 -0700 (PDT) Received: from pps.filterd (m0148461.ppops.net [127.0.0.1]) by mx0a-00082601.pphosted.com (8.17.1.5/8.17.1.5) with ESMTP id 27GAOEVD016190 for ; Tue, 16 Aug 2022 08:37:47 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=fb.com; h=from : to : cc : subject : date : message-id : in-reply-to : references : mime-version : content-transfer-encoding : content-type; s=facebook; bh=V329A0k++xK/PFVPubNpbm1rj3kPgFHzDDIMShBu9OQ=; b=eSFq+X1gITFZBnTlXY8LsXizDdGQrgSELtlZcUQgr38iewZz70dfGDKjm+qn0DWuLNMu vdqpoZHg4kVWf1s+qPJpsAIHqOpoAm3VQ+xhQja04LvjhyFVcikWpDTPNhDVovf7Eayp wk1Amah2wBHuYUna0rpC+O38q7Wzx6Zc+bA= Received: from maileast.thefacebook.com ([163.114.130.16]) by mx0a-00082601.pphosted.com (PPS) with ESMTPS id 3j09en1x09-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128 verify=NOT) for ; Tue, 16 Aug 2022 08:37:47 -0700 Received: from twshared8442.02.ash8.facebook.com (2620:10d:c0a8:1b::d) by mail.thefacebook.com (2620:10d:c0a8:82::c) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2375.31; Tue, 16 Aug 2022 08:37:45 -0700 Received: by devbig038.lla2.facebook.com (Postfix, from userid 572232) id D13FD4AAA7C0; Tue, 16 Aug 2022 08:37:38 -0700 (PDT) From: Dylan Yudaken To: Jens Axboe , Pavel Begunkov , CC: , Dylan Yudaken Subject: [PATCH for-next v2 1/6] io_uring: remove unnecessary variable Date: Tue, 16 Aug 2022 08:37:23 -0700 Message-ID: <20220816153728.2160601-2-dylany@fb.com> X-Mailer: git-send-email 2.30.2 In-Reply-To: <20220816153728.2160601-1-dylany@fb.com> References: <20220816153728.2160601-1-dylany@fb.com> MIME-Version: 1.0 X-FB-Internal: Safe X-Proofpoint-ORIG-GUID: 64_cxu0Fet58qXvWeBY3NVDxIJxyAFwF X-Proofpoint-GUID: 64_cxu0Fet58qXvWeBY3NVDxIJxyAFwF X-Proofpoint-Virus-Version: vendor=baseguard engine=ICAP:2.0.205,Aquarius:18.0.883,Hydra:6.0.517,FMLib:17.11.122.1 definitions=2022-08-16_08,2022-08-16_02,2022-06-22_01 Precedence: bulk List-ID: X-Mailing-List: io-uring@vger.kernel.org 'running' is set once and read once, so can easily just remove it Signed-off-by: Dylan Yudaken --- io_uring/io_uring.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c index ebfdb2212ec2..0c9fe0f1c174 100644 --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -1052,12 +1052,9 @@ void io_req_task_work_add(struct io_kiocb *req) struct io_uring_task *tctx = req->task->io_uring; struct io_ring_ctx *ctx = req->ctx; struct llist_node *node; - bool running; - - running = !llist_add(&req->io_task_work.node, &tctx->task_list); /* task_work already pending, we're done */ - if (running) + if (!llist_add(&req->io_task_work.node, &tctx->task_list)) return; if (ctx->flags & IORING_SETUP_TASKRUN_FLAG) From patchwork Tue Aug 16 15:37:24 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Dylan Yudaken X-Patchwork-Id: 12945039 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id D6936C25B0E for ; Tue, 16 Aug 2022 15:39:07 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236281AbiHPPjG (ORCPT ); Tue, 16 Aug 2022 11:39:06 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:51648 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236162AbiHPPiU (ORCPT ); Tue, 16 Aug 2022 11:38:20 -0400 Received: from mx0b-00082601.pphosted.com (mx0b-00082601.pphosted.com [67.231.153.30]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id F409F31347 for ; Tue, 16 Aug 2022 08:37:53 -0700 (PDT) Received: from pps.filterd (m0109331.ppops.net [127.0.0.1]) by mx0a-00082601.pphosted.com (8.17.1.5/8.17.1.5) with ESMTP id 27G7KbDj007536 for ; Tue, 16 Aug 2022 08:37:53 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=fb.com; h=from : to : cc : subject : date : message-id : in-reply-to : references : mime-version : content-transfer-encoding : content-type; s=facebook; bh=07wjHAPFXMReUQK4g7uilmsdQuSczGFG6GubYuzzT6g=; b=E+4ezq/AKt+Gu4tZ6OOWnVKbNlcreDr64NlqnRC7BOkkZALhkuUknL3qvMi5BHuz/qd+ Au7s4stlwmJ1u/00LewWZRiYtT/mIanJwiSQTDOIC69RFxw/vVTawtdePsTaCrmLtLGk Gr2Ebrl2odEEPn0eoRcN1RQ/wttkCuME46Y= Received: from maileast.thefacebook.com ([163.114.130.16]) by mx0a-00082601.pphosted.com (PPS) with ESMTPS id 3hyn8414x7-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128 verify=NOT) for ; Tue, 16 Aug 2022 08:37:52 -0700 Received: from twshared1866.09.ash9.facebook.com (2620:10d:c0a8:1b::d) by mail.thefacebook.com (2620:10d:c0a8:82::e) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2375.31; Tue, 16 Aug 2022 08:37:52 -0700 Received: by devbig038.lla2.facebook.com (Postfix, from userid 572232) id D96824AAA7C2; Tue, 16 Aug 2022 08:37:38 -0700 (PDT) From: Dylan Yudaken To: Jens Axboe , Pavel Begunkov , CC: , Dylan Yudaken Subject: [PATCH for-next v2 2/6] io_uring: introduce io_has_work Date: Tue, 16 Aug 2022 08:37:24 -0700 Message-ID: <20220816153728.2160601-3-dylany@fb.com> X-Mailer: git-send-email 2.30.2 In-Reply-To: <20220816153728.2160601-1-dylany@fb.com> References: <20220816153728.2160601-1-dylany@fb.com> MIME-Version: 1.0 X-FB-Internal: Safe X-Proofpoint-ORIG-GUID: TPNM4W0gg_24_KCm4JNO7qfpSl730mAm X-Proofpoint-GUID: TPNM4W0gg_24_KCm4JNO7qfpSl730mAm X-Proofpoint-Virus-Version: vendor=baseguard engine=ICAP:2.0.205,Aquarius:18.0.883,Hydra:6.0.517,FMLib:17.11.122.1 definitions=2022-08-16_08,2022-08-16_02,2022-06-22_01 Precedence: bulk List-ID: X-Mailing-List: io-uring@vger.kernel.org This will be used later to know if the ring has outstanding work. Right now just if there is overflow CQEs to copy to the main CQE ring, but later will include deferred tasks Signed-off-by: Dylan Yudaken --- io_uring/io_uring.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c index 0c9fe0f1c174..19d5d1ab5793 100644 --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -2144,6 +2144,11 @@ struct io_wait_queue { unsigned nr_timeouts; }; +static inline bool io_has_work(struct io_ring_ctx *ctx) +{ + return test_bit(IO_CHECK_CQ_OVERFLOW_BIT, &ctx->check_cq); +} + static inline bool io_should_wake(struct io_wait_queue *iowq) { struct io_ring_ctx *ctx = iowq->ctx; @@ -2162,13 +2167,13 @@ static int io_wake_function(struct wait_queue_entry *curr, unsigned int mode, { struct io_wait_queue *iowq = container_of(curr, struct io_wait_queue, wq); + struct io_ring_ctx *ctx = iowq->ctx; /* * Cannot safely flush overflowed CQEs from here, ensure we wake up * the task, and the next invocation will do it. */ - if (io_should_wake(iowq) || - test_bit(IO_CHECK_CQ_OVERFLOW_BIT, &iowq->ctx->check_cq)) + if (io_should_wake(iowq) || io_has_work(ctx)) return autoremove_wake_function(curr, mode, wake_flags, key); return -1; } @@ -2504,8 +2509,8 @@ static __poll_t io_uring_poll(struct file *file, poll_table *wait) * Users may get EPOLLIN meanwhile seeing nothing in cqring, this * pushs them to do the flush. */ - if (io_cqring_events(ctx) || - test_bit(IO_CHECK_CQ_OVERFLOW_BIT, &ctx->check_cq)) + + if (io_cqring_events(ctx) || io_has_work(ctx)) mask |= EPOLLIN | EPOLLRDNORM; return mask; From patchwork Tue Aug 16 15:37:25 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Dylan Yudaken X-Patchwork-Id: 12945041 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 6335CC25B0E for ; Tue, 16 Aug 2022 15:39:11 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236175AbiHPPjK (ORCPT ); Tue, 16 Aug 2022 11:39:10 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:58368 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236166AbiHPPiV (ORCPT ); Tue, 16 Aug 2022 11:38:21 -0400 Received: from mx0a-00082601.pphosted.com (mx0a-00082601.pphosted.com [67.231.145.42]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id D784610C for ; Tue, 16 Aug 2022 08:37:54 -0700 (PDT) Received: from pps.filterd (m0044010.ppops.net [127.0.0.1]) by mx0a-00082601.pphosted.com (8.17.1.5/8.17.1.5) with ESMTP id 27G32lVL027347 for ; Tue, 16 Aug 2022 08:37:54 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=fb.com; h=from : to : cc : subject : date : message-id : in-reply-to : references : mime-version : content-transfer-encoding : content-type; s=facebook; bh=NYkgeWhc4loqejV4OaN5qzP3Q6+Q/djrUZ4arEpsTro=; b=fR3ZzWiRRjhLKtLpaazlHzMvAAzM1t6YgR+AK5DHLORf0H3R+UcW5uop8Fj33qPIp2kk pBGywlSu7JoI/OeckTM0sXGB7OHULIgvlY7PoTltEA9h5OTXZQSSDh6qnW0/jNP5cql+ G/8h8GHOecEiA/1Va40P+cVmY95jKDGBiGo= Received: from maileast.thefacebook.com ([163.114.130.16]) by mx0a-00082601.pphosted.com (PPS) with ESMTPS id 3j02ym3ppg-3 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128 verify=NOT) for ; Tue, 16 Aug 2022 08:37:54 -0700 Received: from twshared14074.07.ash9.facebook.com (2620:10d:c0a8:1b::d) by mail.thefacebook.com (2620:10d:c0a8:83::6) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2375.31; Tue, 16 Aug 2022 08:37:53 -0700 Received: by devbig038.lla2.facebook.com (Postfix, from userid 572232) id DE44C4AAA7C4; Tue, 16 Aug 2022 08:37:38 -0700 (PDT) From: Dylan Yudaken To: Jens Axboe , Pavel Begunkov , CC: , Dylan Yudaken Subject: [PATCH for-next v2 3/6] io_uring: do not run task work at the start of io_uring_enter Date: Tue, 16 Aug 2022 08:37:25 -0700 Message-ID: <20220816153728.2160601-4-dylany@fb.com> X-Mailer: git-send-email 2.30.2 In-Reply-To: <20220816153728.2160601-1-dylany@fb.com> References: <20220816153728.2160601-1-dylany@fb.com> MIME-Version: 1.0 X-FB-Internal: Safe X-Proofpoint-GUID: Z9_jpaoCzDTjaPYY71XM3yfz4U6WJqC2 X-Proofpoint-ORIG-GUID: Z9_jpaoCzDTjaPYY71XM3yfz4U6WJqC2 X-Proofpoint-Virus-Version: vendor=baseguard engine=ICAP:2.0.205,Aquarius:18.0.883,Hydra:6.0.517,FMLib:17.11.122.1 definitions=2022-08-16_08,2022-08-16_02,2022-06-22_01 Precedence: bulk List-ID: X-Mailing-List: io-uring@vger.kernel.org This is not needed, and it is normally better to wait for task work until after submissions. This will allow greater batching if either work arrives in the meanwhile, or if the submissions cause task work to be queued up. For SQPOLL this also no longer runs task work, but this is handled inside the SQPOLL loop anyway. For IOPOLL io_iopoll_check will run task work anyway And otherwise io_cqring_wait will run task work Suggested-by: Pavel Begunkov Signed-off-by: Dylan Yudaken --- io_uring/io_uring.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c index 19d5d1ab5793..53696dd90626 100644 --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -2990,8 +2990,6 @@ SYSCALL_DEFINE6(io_uring_enter, unsigned int, fd, u32, to_submit, struct fd f; long ret; - io_run_task_work(); - if (unlikely(flags & ~(IORING_ENTER_GETEVENTS | IORING_ENTER_SQ_WAKEUP | IORING_ENTER_SQ_WAIT | IORING_ENTER_EXT_ARG | IORING_ENTER_REGISTERED_RING))) From patchwork Tue Aug 16 15:37:26 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Dylan Yudaken X-Patchwork-Id: 12945038 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 65D57C28B2B for ; Tue, 16 Aug 2022 15:39:03 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236139AbiHPPjB (ORCPT ); Tue, 16 Aug 2022 11:39:01 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:51464 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236097AbiHPPiT (ORCPT ); Tue, 16 Aug 2022 11:38:19 -0400 Received: from mx0a-00082601.pphosted.com (mx0a-00082601.pphosted.com [67.231.145.42]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 4329E2AD4 for ; Tue, 16 Aug 2022 08:37:47 -0700 (PDT) Received: from pps.filterd (m0109333.ppops.net [127.0.0.1]) by mx0a-00082601.pphosted.com (8.17.1.5/8.17.1.5) with ESMTP id 27G34LKH009987 for ; Tue, 16 Aug 2022 08:37:47 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=fb.com; h=from : to : cc : subject : date : message-id : in-reply-to : references : content-type : content-transfer-encoding : mime-version; s=facebook; bh=cN+pFNw8+trRwF+P0s5/XFPmtpsJuo63p6JfpkL0CFA=; b=UImAk/FkpZNZY732nmA7a0MgyD1T0ZZzLJKCgEpn3SA7kyr6Fl1o5cacQxzKy2OSeRUu lg91PpcYNfJcDPyb/EqLvRMNTP8C1aTQdQJOf62MShjnCB4dPSMi2/K6uTcKiknRytMv 24q46TcC0B3faKgNguAbozMidtq/tQHccJU= Received: from maileast.thefacebook.com ([163.114.130.16]) by mx0a-00082601.pphosted.com (PPS) with ESMTPS id 3j030hkq05-3 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128 verify=NOT) for ; Tue, 16 Aug 2022 08:37:46 -0700 Received: from twshared8442.02.ash8.facebook.com (2620:10d:c0a8:1b::d) by mail.thefacebook.com (2620:10d:c0a8:83::4) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2375.31; Tue, 16 Aug 2022 08:37:45 -0700 Received: by devbig038.lla2.facebook.com (Postfix, from userid 572232) id E3AD24AAA7C6; Tue, 16 Aug 2022 08:37:38 -0700 (PDT) From: Dylan Yudaken To: Jens Axboe , Pavel Begunkov , CC: , Dylan Yudaken Subject: [PATCH for-next v2 4/6] io_uring: add IORING_SETUP_DEFER_TASKRUN Date: Tue, 16 Aug 2022 08:37:26 -0700 Message-ID: <20220816153728.2160601-5-dylany@fb.com> X-Mailer: git-send-email 2.30.2 In-Reply-To: <20220816153728.2160601-1-dylany@fb.com> References: <20220816153728.2160601-1-dylany@fb.com> X-FB-Internal: Safe X-Proofpoint-GUID: DapGm8GZbt5mBSrR7B7yZ5uDRR9dei2M X-Proofpoint-ORIG-GUID: DapGm8GZbt5mBSrR7B7yZ5uDRR9dei2M X-Proofpoint-UnRewURL: 0 URL was un-rewritten MIME-Version: 1.0 X-Proofpoint-Virus-Version: vendor=baseguard engine=ICAP:2.0.205,Aquarius:18.0.883,Hydra:6.0.517,FMLib:17.11.122.1 definitions=2022-08-16_08,2022-08-16_02,2022-06-22_01 Precedence: bulk List-ID: X-Mailing-List: io-uring@vger.kernel.org Allow deferring async tasks until the user calls io_uring_enter(2) with the IORING_ENTER_GETEVENTS flag. Enable this mode with a flag at io_uring_setup time. This functionality requires that the later io_uring_enter will be called from the same submission task, and therefore restrict this flag to work only when IORING_SETUP_SINGLE_ISSUER is also set. Being able to hand pick when tasks are run prevents the problem where there is current work to be done, however task work runs anyway. For example, a common workload would obtain a batch of CQEs, and process each one. Interrupting this to additional taskwork would add latency but not gain anything. If instead task work is deferred to just before more CQEs are obtained then no additional latency is added. The way this is implemented is by trying to keep task work local to a io_ring_ctx, rather than to the submission task. This is required, as the application will want to wake up only a single io_ring_ctx at a time to process work, and so the lists of work have to be kept separate. This has some other benefits like not having to check the task continually in handle_tw_list (and potentially unlocking/locking those), and reducing locks in the submit & process completions path. There are networking cases where using this option can reduce request latency by 50%. For example a contrived example using [1] where the client sends 2k data and receives the same data back while doing some system calls (to trigger task work) shows this reduction. The reason ends up being that if sending responses is delayed by processing task work, then the client side sits idle. Whereas reordering the sends first means that the client runs it's workload in parallel with the local task work. [1]: Using https://github.com/DylanZA/netbench/tree/defer_run Client: ./netbench --client_only 1 --control_port 10000 --host --tx "epoll --threads 16 --per_thread 1 --size 2048 --resp 2048 --workload 1000" Server: ./netbench --server_only 1 --control_port 10000 --rx "io_uring --defer_taskrun 0 --workload 100" --rx "io_uring --defer_taskrun 1 --workload 100" Signed-off-by: Dylan Yudaken --- include/linux/io_uring_types.h | 2 + include/uapi/linux/io_uring.h | 7 ++ io_uring/cancel.c | 2 +- io_uring/io_uring.c | 119 +++++++++++++++++++++++++++++---- io_uring/io_uring.h | 30 ++++++++- io_uring/rsrc.c | 2 +- 6 files changed, 146 insertions(+), 16 deletions(-) diff --git a/include/linux/io_uring_types.h b/include/linux/io_uring_types.h index 677a25d44d7f..d56ff2185168 100644 --- a/include/linux/io_uring_types.h +++ b/include/linux/io_uring_types.h @@ -301,6 +301,8 @@ struct io_ring_ctx { struct io_hash_table cancel_table; bool poll_multi_queue; + struct llist_head work_llist; + struct list_head io_buffers_comp; } ____cacheline_aligned_in_smp; diff --git a/include/uapi/linux/io_uring.h b/include/uapi/linux/io_uring.h index 1463cfecb56b..be8d1801bf4a 100644 --- a/include/uapi/linux/io_uring.h +++ b/include/uapi/linux/io_uring.h @@ -153,6 +153,13 @@ enum { */ #define IORING_SETUP_SINGLE_ISSUER (1U << 12) +/* + * Defer running task work to get events. + * Rather than running bits of task work whenever the task transitions + * try to do it just before it is needed. + */ +#define IORING_SETUP_DEFER_TASKRUN (1U << 13) + enum io_uring_op { IORING_OP_NOP, IORING_OP_READV, diff --git a/io_uring/cancel.c b/io_uring/cancel.c index e4e1dc0325f0..db6180b62e41 100644 --- a/io_uring/cancel.c +++ b/io_uring/cancel.c @@ -292,7 +292,7 @@ int io_sync_cancel(struct io_ring_ctx *ctx, void __user *arg) break; mutex_unlock(&ctx->uring_lock); - ret = io_run_task_work_sig(); + ret = io_run_task_work_sig(ctx); if (ret < 0) { mutex_lock(&ctx->uring_lock); break; diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c index 53696dd90626..6b16da0be712 100644 --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -316,6 +316,7 @@ static __cold struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p) INIT_LIST_HEAD(&ctx->rsrc_ref_list); INIT_DELAYED_WORK(&ctx->rsrc_put_work, io_rsrc_put_work); init_llist_head(&ctx->rsrc_put_llist); + init_llist_head(&ctx->work_llist); INIT_LIST_HEAD(&ctx->tctx_list); ctx->submit_state.free_list.next = NULL; INIT_WQ_LIST(&ctx->locked_free_list); @@ -1047,12 +1048,30 @@ void tctx_task_work(struct callback_head *cb) trace_io_uring_task_work_run(tctx, count, loops); } -void io_req_task_work_add(struct io_kiocb *req) +static void io_req_local_work_add(struct io_kiocb *req) +{ + struct io_ring_ctx *ctx = req->ctx; + + if (!llist_add(&req->io_task_work.node, &ctx->work_llist)) + return; + + if (ctx->flags & IORING_SETUP_TASKRUN_FLAG) + atomic_or(IORING_SQ_TASKRUN, &ctx->rings->sq_flags); + + io_cqring_wake(ctx); +} + +static inline void __io_req_task_work_add(struct io_kiocb *req, bool allow_local) { struct io_uring_task *tctx = req->task->io_uring; struct io_ring_ctx *ctx = req->ctx; struct llist_node *node; + if (allow_local && ctx->flags & IORING_SETUP_DEFER_TASKRUN) { + io_req_local_work_add(req); + return; + } + /* task_work already pending, we're done */ if (!llist_add(&req->io_task_work.node, &tctx->task_list)) return; @@ -1074,6 +1093,55 @@ void io_req_task_work_add(struct io_kiocb *req) } } +void io_req_task_work_add(struct io_kiocb *req) +{ + __io_req_task_work_add(req, true); +} + +bool io_run_local_work(struct io_ring_ctx *ctx, bool locked) +{ + struct llist_node *node; + struct llist_node fake; + struct llist_node *current_final = NULL; + unsigned int count; + + if (!locked) + locked = mutex_trylock(&ctx->uring_lock); + + node = io_llist_xchg(&ctx->work_llist, &fake); + count = 0; +again: + while (node != current_final) { + struct llist_node *next = node->next; + struct io_kiocb *req = container_of(node, struct io_kiocb, + io_task_work.node); + prefetch(container_of(next, struct io_kiocb, io_task_work.node)); + if (unlikely(req->task != current)) { + __io_req_task_work_add(req, false); + } else { + req->io_task_work.func(req, &locked); + count++; + } + node = next; + } + + if (ctx->flags & IORING_SETUP_TASKRUN_FLAG) + atomic_andnot(IORING_SQ_TASKRUN, &ctx->rings->sq_flags); + + node = io_llist_cmpxchg(&ctx->work_llist, &fake, NULL); + if (node != &fake) { + current_final = &fake; + node = io_llist_xchg(&ctx->work_llist, &fake); + goto again; + } + + if (locked) { + io_submit_flush_completions(ctx); + mutex_unlock(&ctx->uring_lock); + } + return count > 0; +} + static void io_req_tw_post(struct io_kiocb *req, bool *locked) { io_req_complete_post(req); @@ -1284,8 +1352,7 @@ static int io_iopoll_check(struct io_ring_ctx *ctx, long min) if (wq_list_empty(&ctx->iopoll_list)) { u32 tail = ctx->cached_cq_tail; - mutex_unlock(&ctx->uring_lock); - io_run_task_work(); + io_run_task_work_unlock_ctx(ctx); mutex_lock(&ctx->uring_lock); /* some requests don't go through iopoll_list */ @@ -2146,7 +2213,9 @@ struct io_wait_queue { static inline bool io_has_work(struct io_ring_ctx *ctx) { - return test_bit(IO_CHECK_CQ_OVERFLOW_BIT, &ctx->check_cq); + return test_bit(IO_CHECK_CQ_OVERFLOW_BIT, &ctx->check_cq) || + ((ctx->flags & IORING_SETUP_DEFER_TASKRUN) && + !llist_empty(&ctx->work_llist)); } static inline bool io_should_wake(struct io_wait_queue *iowq) @@ -2178,9 +2247,9 @@ static int io_wake_function(struct wait_queue_entry *curr, unsigned int mode, return -1; } -int io_run_task_work_sig(void) +int io_run_task_work_sig(struct io_ring_ctx *ctx) { - if (io_run_task_work()) + if (io_run_task_work_ctx(ctx, true)) return 1; if (task_sigpending(current)) return -EINTR; @@ -2196,7 +2265,7 @@ static inline int io_cqring_wait_schedule(struct io_ring_ctx *ctx, unsigned long check_cq; /* make sure we run task_work before checking for signals */ - ret = io_run_task_work_sig(); + ret = io_run_task_work_sig(ctx); if (ret || io_should_wake(iowq)) return ret; @@ -2230,7 +2299,7 @@ static int io_cqring_wait(struct io_ring_ctx *ctx, int min_events, io_cqring_overflow_flush(ctx); if (io_cqring_events(ctx) >= min_events) return 0; - if (!io_run_task_work()) + if (!io_run_task_work_ctx(ctx, false)) break; } while (1); @@ -2768,6 +2837,8 @@ static __cold bool io_uring_try_cancel_requests(struct io_ring_ctx *ctx, } } + if (ctx->flags & IORING_SETUP_DEFER_TASKRUN) + ret |= io_run_local_work(ctx, false); ret |= io_cancel_defer_files(ctx, task, cancel_all); mutex_lock(&ctx->uring_lock); ret |= io_poll_remove_all(ctx, task, cancel_all); @@ -2837,7 +2908,7 @@ __cold void io_uring_cancel_generic(bool cancel_all, struct io_sq_data *sqd) } prepare_to_wait(&tctx->wait, &wait, TASK_INTERRUPTIBLE); - io_run_task_work(); + io_run_task_work_ctx(ctx, true); io_uring_drop_tctx_refs(current); /* @@ -3057,10 +3128,17 @@ SYSCALL_DEFINE6(io_uring_enter, unsigned int, fd, u32, to_submit, } if ((flags & IORING_ENTER_GETEVENTS) && ctx->syscall_iopoll) goto iopoll_locked; + if ((flags & IORING_ENTER_GETEVENTS) && + (ctx->flags & IORING_SETUP_DEFER_TASKRUN)) { + io_run_local_work(ctx, true); + goto getevents_ran_local; + } mutex_unlock(&ctx->uring_lock); } + if (flags & IORING_ENTER_GETEVENTS) { int ret2; + if (ctx->syscall_iopoll) { /* * We disallow the app entering submit/complete with @@ -3081,6 +3159,9 @@ SYSCALL_DEFINE6(io_uring_enter, unsigned int, fd, u32, to_submit, const sigset_t __user *sig; struct __kernel_timespec __user *ts; + if (ctx->flags & IORING_SETUP_DEFER_TASKRUN) + io_run_local_work(ctx, false); +getevents_ran_local: ret2 = io_get_ext_arg(flags, argp, &argsz, &ts, &sig); if (likely(!ret2)) { min_complete = min(min_complete, @@ -3289,17 +3370,29 @@ static __cold int io_uring_create(unsigned entries, struct io_uring_params *p, if (ctx->flags & IORING_SETUP_SQPOLL) { /* IPI related flags don't make sense with SQPOLL */ if (ctx->flags & (IORING_SETUP_COOP_TASKRUN | - IORING_SETUP_TASKRUN_FLAG)) + IORING_SETUP_TASKRUN_FLAG | + IORING_SETUP_DEFER_TASKRUN)) goto err; ctx->notify_method = TWA_SIGNAL_NO_IPI; } else if (ctx->flags & IORING_SETUP_COOP_TASKRUN) { ctx->notify_method = TWA_SIGNAL_NO_IPI; } else { - if (ctx->flags & IORING_SETUP_TASKRUN_FLAG) + if (ctx->flags & IORING_SETUP_TASKRUN_FLAG && + !(ctx->flags & IORING_SETUP_DEFER_TASKRUN)) goto err; ctx->notify_method = TWA_SIGNAL; } + /* + * For DEFER_TASKRUN we require the completion task to be the same as the + * submission task. This implies that there is only one submitter, so enforce + * that. + */ + if (ctx->flags & IORING_SETUP_DEFER_TASKRUN && + !(ctx->flags & IORING_SETUP_SINGLE_ISSUER)) { + goto err; + } + /* * This is just grabbed for accounting purposes. When a process exits, * the mm is exited and dropped before the files, hence we need to hang @@ -3400,7 +3493,7 @@ static long io_uring_setup(u32 entries, struct io_uring_params __user *params) IORING_SETUP_R_DISABLED | IORING_SETUP_SUBMIT_ALL | IORING_SETUP_COOP_TASKRUN | IORING_SETUP_TASKRUN_FLAG | IORING_SETUP_SQE128 | IORING_SETUP_CQE32 | - IORING_SETUP_SINGLE_ISSUER)) + IORING_SETUP_SINGLE_ISSUER | IORING_SETUP_DEFER_TASKRUN)) return -EINVAL; return io_uring_create(entries, &p, params); @@ -3872,7 +3965,7 @@ SYSCALL_DEFINE4(io_uring_register, unsigned int, fd, unsigned int, opcode, ctx = f.file->private_data; - io_run_task_work(); + io_run_task_work_ctx(ctx, true); mutex_lock(&ctx->uring_lock); ret = __io_uring_register(ctx, opcode, arg, nr_args); diff --git a/io_uring/io_uring.h b/io_uring/io_uring.h index 2f73f83af960..4d0e9c2e00d0 100644 --- a/io_uring/io_uring.h +++ b/io_uring/io_uring.h @@ -26,7 +26,8 @@ enum { struct io_uring_cqe *__io_get_cqe(struct io_ring_ctx *ctx); bool io_req_cqe_overflow(struct io_kiocb *req); -int io_run_task_work_sig(void); +int io_run_task_work_sig(struct io_ring_ctx *ctx); +bool io_run_local_work(struct io_ring_ctx *ctx, bool locked); void io_req_complete_failed(struct io_kiocb *req, s32 res); void __io_req_complete(struct io_kiocb *req, unsigned issue_flags); void io_req_complete_post(struct io_kiocb *req); @@ -234,6 +235,33 @@ static inline bool io_run_task_work(void) return false; } +static inline bool io_run_task_work_ctx(struct io_ring_ctx *ctx, bool all) +{ + bool ret = false; + + if (ctx->flags & IORING_SETUP_DEFER_TASKRUN) { + ret = io_run_local_work(ctx, false); + if (!all) + return ret; + } + ret |= io_run_task_work(); + return ret; +} + +static inline bool io_run_task_work_unlock_ctx(struct io_ring_ctx *ctx) +{ + bool ret; + + if (ctx->flags & IORING_SETUP_DEFER_TASKRUN) { + ret = io_run_local_work(ctx, true); + } else { + mutex_unlock(&ctx->uring_lock); + ret = io_run_task_work(); + } + + return ret; +} + static inline void io_tw_lock(struct io_ring_ctx *ctx, bool *locked) { if (!*locked) { diff --git a/io_uring/rsrc.c b/io_uring/rsrc.c index 71359a4d0bd4..80cda6e2067f 100644 --- a/io_uring/rsrc.c +++ b/io_uring/rsrc.c @@ -343,7 +343,7 @@ __cold static int io_rsrc_ref_quiesce(struct io_rsrc_data *data, flush_delayed_work(&ctx->rsrc_put_work); reinit_completion(&data->done); - ret = io_run_task_work_sig(); + ret = io_run_task_work_sig(ctx); mutex_lock(&ctx->uring_lock); } while (ret >= 0); data->quiesce = false; From patchwork Tue Aug 16 15:37:27 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Dylan Yudaken X-Patchwork-Id: 12945040 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 7E78AC28B2B for ; Tue, 16 Aug 2022 15:39:08 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236162AbiHPPjH (ORCPT ); Tue, 16 Aug 2022 11:39:07 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:50894 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S234406AbiHPPiU (ORCPT ); Tue, 16 Aug 2022 11:38:20 -0400 Received: from mx0a-00082601.pphosted.com (mx0a-00082601.pphosted.com [67.231.145.42]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id E94DF30F60 for ; Tue, 16 Aug 2022 08:37:53 -0700 (PDT) Received: from pps.filterd (m0044010.ppops.net [127.0.0.1]) by mx0a-00082601.pphosted.com (8.17.1.5/8.17.1.5) with ESMTP id 27G32USa026735 for ; Tue, 16 Aug 2022 08:37:53 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=fb.com; h=from : to : cc : subject : date : message-id : in-reply-to : references : mime-version : content-transfer-encoding : content-type; s=facebook; bh=Kf1Z+BI3ZmxvICC0Rhc32z630FaFxOx/Y8uJd6+KQmc=; b=RaUhzSRbDlshJgoi3TdDLtAQ8TjWQ6v+fTdNF/P/6Jr4TVKGoTaksBLf1OtFmS9iabtb s5W/mELPuq7YWJQP5MVddaO6+Fe5pi+CvNXIY7f+RnnQcbyinFSdsk4bldMf245OAWew X6ArnzGEMcH9GMWZpHkl69MQGEiKKH8PgNo= Received: from maileast.thefacebook.com ([163.114.130.16]) by mx0a-00082601.pphosted.com (PPS) with ESMTPS id 3j02ym3ppd-2 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128 verify=NOT) for ; Tue, 16 Aug 2022 08:37:53 -0700 Received: from twshared1866.09.ash9.facebook.com (2620:10d:c0a8:1b::d) by mail.thefacebook.com (2620:10d:c0a8:83::7) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2375.31; Tue, 16 Aug 2022 08:37:52 -0700 Received: by devbig038.lla2.facebook.com (Postfix, from userid 572232) id EC18B4AAA7C8; Tue, 16 Aug 2022 08:37:38 -0700 (PDT) From: Dylan Yudaken To: Jens Axboe , Pavel Begunkov , CC: , Dylan Yudaken Subject: [PATCH for-next v2 5/6] io_uring: move io_eventfd_put Date: Tue, 16 Aug 2022 08:37:27 -0700 Message-ID: <20220816153728.2160601-6-dylany@fb.com> X-Mailer: git-send-email 2.30.2 In-Reply-To: <20220816153728.2160601-1-dylany@fb.com> References: <20220816153728.2160601-1-dylany@fb.com> MIME-Version: 1.0 X-FB-Internal: Safe X-Proofpoint-GUID: yI78k5md9p0VROt7NmQ6i-ftZIO3v6Qx X-Proofpoint-ORIG-GUID: yI78k5md9p0VROt7NmQ6i-ftZIO3v6Qx X-Proofpoint-Virus-Version: vendor=baseguard engine=ICAP:2.0.205,Aquarius:18.0.883,Hydra:6.0.517,FMLib:17.11.122.1 definitions=2022-08-16_08,2022-08-16_02,2022-06-22_01 Precedence: bulk List-ID: X-Mailing-List: io-uring@vger.kernel.org Non functional change: move this function above io_eventfd_signal so it can be used from there Signed-off-by: Dylan Yudaken --- io_uring/io_uring.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c index 6b16da0be712..2eb9ad9edef3 100644 --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -478,6 +478,14 @@ static __cold void io_queue_deferred(struct io_ring_ctx *ctx) } } +static void io_eventfd_put(struct rcu_head *rcu) +{ + struct io_ev_fd *ev_fd = container_of(rcu, struct io_ev_fd, rcu); + + eventfd_ctx_put(ev_fd->cq_ev_fd); + kfree(ev_fd); +} + static void io_eventfd_signal(struct io_ring_ctx *ctx) { struct io_ev_fd *ev_fd; @@ -2438,14 +2446,6 @@ static int io_eventfd_register(struct io_ring_ctx *ctx, void __user *arg, return 0; } -static void io_eventfd_put(struct rcu_head *rcu) -{ - struct io_ev_fd *ev_fd = container_of(rcu, struct io_ev_fd, rcu); - - eventfd_ctx_put(ev_fd->cq_ev_fd); - kfree(ev_fd); -} - static int io_eventfd_unregister(struct io_ring_ctx *ctx) { struct io_ev_fd *ev_fd; From patchwork Tue Aug 16 15:37:28 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Dylan Yudaken X-Patchwork-Id: 12945042 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 5EBD9C2BB41 for ; Tue, 16 Aug 2022 15:39:12 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236182AbiHPPjK (ORCPT ); Tue, 16 Aug 2022 11:39:10 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:51654 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236165AbiHPPiV (ORCPT ); Tue, 16 Aug 2022 11:38:21 -0400 Received: from mx0a-00082601.pphosted.com (mx0a-00082601.pphosted.com [67.231.145.42]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 925E32DC5 for ; Tue, 16 Aug 2022 08:38:00 -0700 (PDT) Received: from pps.filterd (m0109333.ppops.net [127.0.0.1]) by mx0a-00082601.pphosted.com (8.17.1.5/8.17.1.5) with ESMTP id 27GFbvaR008735 for ; Tue, 16 Aug 2022 08:38:00 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=fb.com; h=from : to : cc : subject : date : message-id : in-reply-to : references : mime-version : content-transfer-encoding : content-type; s=facebook; bh=6WdHYVQ3hsIo/aoa0fsgO0YojZ2/DJHTafRogLVTmWc=; b=FQnxr6MNzXLMdF/eiSmt5smzHJMDDVtyYPCTDBhHbXCAGojKXp08e6+qn7riHjFTg0u3 1Py5goKhBZWE3e7aBrAMz7bzIUTcPRzlZ0yh/NqBPJ+D+/mYqhgSHF5tjNnaCv+JpNhq ouJEl4o2Psb+IpdfY0tCwBuJQ03tZNOpQCs= Received: from maileast.thefacebook.com ([163.114.130.16]) by mx0a-00082601.pphosted.com (PPS) with ESMTPS id 3j030hkq22-7 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128 verify=NOT) for ; Tue, 16 Aug 2022 08:38:00 -0700 Received: from twshared1866.09.ash9.facebook.com (2620:10d:c0a8:1b::d) by mail.thefacebook.com (2620:10d:c0a8:82::d) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2375.31; Tue, 16 Aug 2022 08:37:52 -0700 Received: by devbig038.lla2.facebook.com (Postfix, from userid 572232) id F1DFF4AAA7CA; Tue, 16 Aug 2022 08:37:38 -0700 (PDT) From: Dylan Yudaken To: Jens Axboe , Pavel Begunkov , CC: , Dylan Yudaken Subject: [PATCH for-next v2 6/6] io_uring: signal registered eventfd to process deferred task work Date: Tue, 16 Aug 2022 08:37:28 -0700 Message-ID: <20220816153728.2160601-7-dylany@fb.com> X-Mailer: git-send-email 2.30.2 In-Reply-To: <20220816153728.2160601-1-dylany@fb.com> References: <20220816153728.2160601-1-dylany@fb.com> MIME-Version: 1.0 X-FB-Internal: Safe X-Proofpoint-GUID: Y2EnvORa5fEKcT6DpAZgMXEPgFzFT187 X-Proofpoint-ORIG-GUID: Y2EnvORa5fEKcT6DpAZgMXEPgFzFT187 X-Proofpoint-Virus-Version: vendor=baseguard engine=ICAP:2.0.205,Aquarius:18.0.883,Hydra:6.0.517,FMLib:17.11.122.1 definitions=2022-08-16_08,2022-08-16_02,2022-06-22_01 Precedence: bulk List-ID: X-Mailing-List: io-uring@vger.kernel.org Some workloads rely on a registered eventfd (via io_uring_register_eventfd(3)) in order to wake up and process the io_uring. In the case of a ring setup with IORING_SETUP_DEFER_TASKRUN, that eventfd also needs to be signalled when there are tasks to run. This changes an old behaviour which assumed 1 eventfd signal implied at least 1 CQE, however only when this new flag is set (and so old users will not notice). This should be expected with the IORING_SETUP_DEFER_TASKRUN flag as it is not guaranteed that every task will result in a CQE. Signed-off-by: Dylan Yudaken --- include/linux/io_uring_types.h | 1 + io_uring/io_uring.c | 75 ++++++++++++++++++++++++---------- 2 files changed, 55 insertions(+), 21 deletions(-) diff --git a/include/linux/io_uring_types.h b/include/linux/io_uring_types.h index d56ff2185168..42494176434a 100644 --- a/include/linux/io_uring_types.h +++ b/include/linux/io_uring_types.h @@ -184,6 +184,7 @@ struct io_ev_fd { struct eventfd_ctx *cq_ev_fd; unsigned int eventfd_async: 1; struct rcu_head rcu; + atomic_t refs; }; struct io_alloc_cache { diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c index 2eb9ad9edef3..65ea8ccd63e8 100644 --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -478,33 +478,33 @@ static __cold void io_queue_deferred(struct io_ring_ctx *ctx) } } + +static inline void __io_eventfd_put(struct io_ev_fd *ev_fd) +{ + if (atomic_dec_and_test(&ev_fd->refs)) { + eventfd_ctx_put(ev_fd->cq_ev_fd); + kfree(ev_fd); + } +} + +static void io_eventfd_signal_put(struct rcu_head *rcu) +{ + struct io_ev_fd *ev_fd = container_of(rcu, struct io_ev_fd, rcu); + + eventfd_signal(ev_fd->cq_ev_fd, 1); + __io_eventfd_put(ev_fd); +} + static void io_eventfd_put(struct rcu_head *rcu) { struct io_ev_fd *ev_fd = container_of(rcu, struct io_ev_fd, rcu); - eventfd_ctx_put(ev_fd->cq_ev_fd); - kfree(ev_fd); + __io_eventfd_put(ev_fd); } static void io_eventfd_signal(struct io_ring_ctx *ctx) { - struct io_ev_fd *ev_fd; - bool skip; - - spin_lock(&ctx->completion_lock); - /* - * Eventfd should only get triggered when at least one event has been - * posted. Some applications rely on the eventfd notification count only - * changing IFF a new CQE has been added to the CQ ring. There's no - * depedency on 1:1 relationship between how many times this function is - * called (and hence the eventfd count) and number of CQEs posted to the - * CQ ring. - */ - skip = ctx->cached_cq_tail == ctx->evfd_last_cq_tail; - ctx->evfd_last_cq_tail = ctx->cached_cq_tail; - spin_unlock(&ctx->completion_lock); - if (skip) - return; + struct io_ev_fd *ev_fd = NULL; rcu_read_lock(); /* @@ -522,13 +522,43 @@ static void io_eventfd_signal(struct io_ring_ctx *ctx) goto out; if (READ_ONCE(ctx->rings->cq_flags) & IORING_CQ_EVENTFD_DISABLED) goto out; + if (ev_fd->eventfd_async && !io_wq_current_is_worker()) + goto out; - if (!ev_fd->eventfd_async || io_wq_current_is_worker()) + if (likely(eventfd_signal_allowed())) { eventfd_signal(ev_fd->cq_ev_fd, 1); + } else { + atomic_inc(&ev_fd->refs); + call_rcu(&ev_fd->rcu, io_eventfd_signal_put); + } + out: rcu_read_unlock(); } +static void io_eventfd_flush_signal(struct io_ring_ctx *ctx) +{ + bool skip; + + spin_lock(&ctx->completion_lock); + + /* + * Eventfd should only get triggered when at least one event has been + * posted. Some applications rely on the eventfd notification count + * only changing IFF a new CQE has been added to the CQ ring. There's + * no depedency on 1:1 relationship between how many times this + * function is called (and hence the eventfd count) and number of CQEs + * posted to the CQ ring. + */ + skip = ctx->cached_cq_tail == ctx->evfd_last_cq_tail; + ctx->evfd_last_cq_tail = ctx->cached_cq_tail; + spin_unlock(&ctx->completion_lock); + if (skip) + return; + + io_eventfd_signal(ctx); +} + void __io_commit_cqring_flush(struct io_ring_ctx *ctx) { if (ctx->off_timeout_used || ctx->drain_active) { @@ -540,7 +570,7 @@ void __io_commit_cqring_flush(struct io_ring_ctx *ctx) spin_unlock(&ctx->completion_lock); } if (ctx->has_evfd) - io_eventfd_signal(ctx); + io_eventfd_flush_signal(ctx); } static inline void io_cqring_ev_posted(struct io_ring_ctx *ctx) @@ -1066,6 +1096,8 @@ static void io_req_local_work_add(struct io_kiocb *req) if (ctx->flags & IORING_SETUP_TASKRUN_FLAG) atomic_or(IORING_SQ_TASKRUN, &ctx->rings->sq_flags); + if (ctx->has_evfd) + io_eventfd_signal(ctx); io_cqring_wake(ctx); } @@ -2443,6 +2475,7 @@ static int io_eventfd_register(struct io_ring_ctx *ctx, void __user *arg, ev_fd->eventfd_async = eventfd_async; ctx->has_evfd = true; rcu_assign_pointer(ctx->io_ev_fd, ev_fd); + atomic_set(&ev_fd->refs, 1); return 0; }