From nobody Sat Apr 18 02:47:39 2026 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 901F9C43334 for ; Tue, 19 Jul 2022 16:58:12 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S237279AbiGSQ6K (ORCPT ); Tue, 19 Jul 2022 12:58:10 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:52098 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S237319AbiGSQ6H (ORCPT ); Tue, 19 Jul 2022 12:58:07 -0400 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTP id 6369E4F6A1 for ; Tue, 19 Jul 2022 09:58:05 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1658249884; 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; bh=pGLT/O5C/K0AnzGXbvcTRwb/0H0s9Eed8T9hVEGD0gc=; b=QlJha7v3hoKeejbMhNvbzHnj20FDTV04Mj+WVBYfHQPdcDUlyDE2yHgwm2U1sx2zqFlMlS nhb321X//To6+uvTvWnOt1Bnb4cLRQRJQw0j2SpJrWjh9pdiPtLUYwsfC6un3sIBMWQhGo oZZPpDnUdnuF4hWLcG7vdl+tmx4L1qs= Received: from mail-wm1-f71.google.com (mail-wm1-f71.google.com [209.85.128.71]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-638-2jnbNDBJPCaxmPKD4yZWOA-1; Tue, 19 Jul 2022 12:57:57 -0400 X-MC-Unique: 2jnbNDBJPCaxmPKD4yZWOA-1 Received: by mail-wm1-f71.google.com with SMTP id r127-20020a1c4485000000b003a2fdeea756so1202794wma.2 for ; Tue, 19 Jul 2022 09:57:57 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:from:to:cc:subject:date:message-id:mime-version :content-transfer-encoding; bh=pGLT/O5C/K0AnzGXbvcTRwb/0H0s9Eed8T9hVEGD0gc=; b=3ivO3eFn2Dr0L4w/YqK+UecWwuJ4/Xu8SoiZD4x+ffQR9iHB3gnhh/Xo/C5JxEiEYI 8B8PdPefc1SCBR/m794gVni/pHjJpsUNDLwcNltoftYJfHYbkC/KIFKV/BbmEIZ549cI 9TIJ1vFYb6Yp03gYVwZa+Tjc6xhNfJxXbYEIyUKHDHcMXR4eQWZtnq43NrnmFP7IiCgw 0GG6MBFDDf3928D9pshLV23UkOOX9VnwQDblAjW1PFYqs4Lmc5E/4hYGN5xNn3MMpuU1 XpOeJTUfRJdqklEG3Aop8ScQvFjyquTlGG4Fc1mjlYOJJbpAgeDewXmuqDwYD2NGzCrk 1sPQ== X-Gm-Message-State: AJIora98mvBRlRUrmTMnd0d1CqZiJvoeb1l+PLcP5UQ3osNt3y/wXbkj q1MQ/9k0RlSdroqlVV1spiUjwGhxOo9ZDqmqcM5grpoF4kxN1fpb1hS6G2p3ncHQWOCHD2obRQ5 5QSHXrnZZaEfeTJpJh1RSbAZAhAZiM1Kinkk7dFuU9jBFUD7skDGeFBITaV6f2JLnQifMU0v3Sr 7q X-Received: by 2002:a5d:452e:0:b0:21d:9234:8661 with SMTP id j14-20020a5d452e000000b0021d92348661mr27026466wra.481.1658249875560; Tue, 19 Jul 2022 09:57:55 -0700 (PDT) X-Google-Smtp-Source: AGRyM1vyr8MeqADwsGxABGdtADyEHOP8+z+Xngogz4JQJhwVcBouDJx6dPv8YV8xCvAx0iZ2f5Zccg== X-Received: by 2002:a5d:452e:0:b0:21d:9234:8661 with SMTP id j14-20020a5d452e000000b0021d92348661mr27026449wra.481.1658249875343; Tue, 19 Jul 2022 09:57:55 -0700 (PDT) Received: from vschneid.remote.csb ([185.11.37.247]) by smtp.gmail.com with ESMTPSA id f7-20020a7bcd07000000b003a02cbf862esm19309823wmj.13.2022.07.19.09.57.54 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 19 Jul 2022 09:57:54 -0700 (PDT) From: Valentin Schneider To: linux-kernel@vger.kernel.org Cc: Tejun Heo , Lai Jiangshan , Peter Zijlstra , Frederic Weisbecker , Juri Lelli , Phil Auld , Marcelo Tosatti Subject: [RFC PATCH] workqueue: Unbind workers before sending them to exit() Date: Tue, 19 Jul 2022 17:57:43 +0100 Message-Id: <20220719165743.3409313-1-vschneid@redhat.com> X-Mailer: git-send-email 2.31.1 MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" It has been reported that isolated CPUs can suffer from interference due to per-CPU kworkers waking up just to die. A surge of workqueue activity (sleeping workfn's exacerbate this) during initial setup can cause extra per-CPU kworkers to be spawned. Then, a latency-sensitive task can be running merrily on an isolated CPU only to be interrupted sometime later by a kworker marked for death (cf. IDLE_WORKER_TIMEOUT, 5 minutes after last kworker activity). Affine kworkers to the wq_unbound_cpumask (which doesn't contain isolated CPUs, cf. HK_TYPE_WQ) before waking them up after marking them with WORKER_DIE. This follows the logic of CPU hot-unplug, which has been packaged into helpers for the occasion. Signed-off-by: Valentin Schneider Reviewed-by: Marcelo Tosatti --- kernel/workqueue.c | 35 ++++++++++++++++++++++++++--------- 1 file changed, 26 insertions(+), 9 deletions(-) diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 1ea50f6be843..0f1a25ea4924 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -1972,6 +1972,18 @@ static struct worker *create_worker(struct worker_po= ol *pool) return NULL; } =20 +static void unbind_worker(struct worker *worker) +{ + kthread_set_per_cpu(worker->task, -1); + WARN_ON_ONCE(set_cpus_allowed_ptr(worker->task, wq_unbound_cpumask) < 0); +} + +static void rebind_worker(struct worker *worker, struct worker_pool *pool) +{ + kthread_set_per_cpu(worker->task, pool->cpu); + WARN_ON_ONCE(set_cpus_allowed_ptr(worker->task, pool->attrs->cpumask) < 0= ); +} + /** * destroy_worker - destroy a workqueue worker * @worker: worker to be destroyed @@ -1999,6 +2011,16 @@ static void destroy_worker(struct worker *worker) =20 list_del_init(&worker->entry); worker->flags |=3D WORKER_DIE; + + /* + * We're sending that thread off to die, so any CPU would do. This is + * especially relevant for pcpu kworkers affined to an isolated CPU: + * we'd rather not interrupt an isolated CPU just for a kworker to + * do_exit(). + */ + if (!(worker->flags & WORKER_UNBOUND)) + unbind_worker(worker); + wake_up_process(worker->task); } =20 @@ -4999,10 +5021,8 @@ static void unbind_workers(int cpu) =20 raw_spin_unlock_irq(&pool->lock); =20 - for_each_pool_worker(worker, pool) { - kthread_set_per_cpu(worker->task, -1); - WARN_ON_ONCE(set_cpus_allowed_ptr(worker->task, wq_unbound_cpumask) < 0= ); - } + for_each_pool_worker(worker, pool) + unbind_worker(worker); =20 mutex_unlock(&wq_pool_attach_mutex); } @@ -5027,11 +5047,8 @@ static void rebind_workers(struct worker_pool *pool) * of all workers first and then clear UNBOUND. As we're called * from CPU_ONLINE, the following shouldn't fail. */ - for_each_pool_worker(worker, pool) { - kthread_set_per_cpu(worker->task, pool->cpu); - WARN_ON_ONCE(set_cpus_allowed_ptr(worker->task, - pool->attrs->cpumask) < 0); - } + for_each_pool_worker(worker, pool) + rebind_worker(worker, pool); =20 raw_spin_lock_irq(&pool->lock); =20 --=20 2.31.1