From nobody Thu Apr 16 02:11:23 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 3DC06C4332F for ; Wed, 23 Nov 2022 13:35:00 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S238433AbiKWNex (ORCPT ); Wed, 23 Nov 2022 08:34:53 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:38376 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S238779AbiKWNdC (ORCPT ); Wed, 23 Nov 2022 08:33:02 -0500 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 7E3A8B82 for ; Wed, 23 Nov 2022 05:16:29 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1669209388; 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=hOjXEo74LMv/9JyDqLl/For3+JWaKkwN289eetN1FnE=; b=cLIwygiwbe+YGnzt8HztW367gHaXKiEnhgY6c0ZKPYWssbxAlXqtqRRkVz9V48JSN9pSM8 eAnOvaqHTV3kpmZI8ocWwFigiEsAGt/eAfLWPexIBWA89ie9Dcfh4agSOdIAOJI2HsY0Oc zaD8B7qB6Z05BpXE1QhRrCNLOnJ2FPg= Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-383-z1Aa5QxNN061Zxr5IJP3dQ-1; Wed, 23 Nov 2022 08:16:25 -0500 X-MC-Unique: z1Aa5QxNN061Zxr5IJP3dQ-1 Received: from smtp.corp.redhat.com (int-mx04.intmail.prod.int.rdu2.redhat.com [10.11.54.4]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 9E09B801585; Wed, 23 Nov 2022 13:16:24 +0000 (UTC) Received: from llong.com (unknown [10.22.8.230]) by smtp.corp.redhat.com (Postfix) with ESMTP id 1165A2024CBE; Wed, 23 Nov 2022 13:16:24 +0000 (UTC) From: Waiman Long To: Ingo Molnar , Peter Zijlstra , Juri Lelli , Vincent Guittot , Dietmar Eggemann , Steven Rostedt , Ben Segall , Mel Gorman , Daniel Bristot de Oliveira Cc: Phil Auld , linux-kernel@vger.kernel.org, Waiman Long Subject: [PATCH-tip v2] sched: Fix NULL user_cpus_ptr check in dup_user_cpus_ptr() Date: Wed, 23 Nov 2022 08:16:12 -0500 Message-Id: <20221123131612.914906-1-longman@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.4 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" In general, a non-null user_cpus_ptr will remain set until the task dies. A possible exception to this is the fact that do_set_cpus_allowed() will clear a non-null user_cpus_ptr. To allow this possible racing condition, we need to check for NULL user_cpus_ptr under the pi_lock before duping the user mask. Fixes: 851a723e45d1 ("sched: Always clear user_cpus_ptr in do_set_cpus_allo= wed()") Signed-off-by: Waiman Long --- kernel/sched/core.c | 23 +++++++++++++++++++++-- 1 file changed, 21 insertions(+), 2 deletions(-) diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 8df51b08bb38..6b259d9e127a 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -2624,8 +2624,14 @@ void do_set_cpus_allowed(struct task_struct *p, cons= t struct cpumask *new_mask) int dup_user_cpus_ptr(struct task_struct *dst, struct task_struct *src, int node) { + cpumask_t *user_mask =3D NULL; unsigned long flags; =20 + /* + * This check is racy and losing the race is a valid situation. + * It is not worth the extra overhead of taking the pi_lock on + * every fork/clone. + */ if (!src->user_cpus_ptr) return 0; =20 @@ -2633,10 +2639,23 @@ int dup_user_cpus_ptr(struct task_struct *dst, stru= ct task_struct *src, if (!dst->user_cpus_ptr) return -ENOMEM; =20 - /* Use pi_lock to protect content of user_cpus_ptr */ + /* + * Use pi_lock to protect content of user_cpus_ptr + * + * Though unlikely, user_cpus_ptr can be reset to NULL by a concurrent + * do_set_cpus_allowed(). When this happens, we need to clear + * dst->user_cpus_ptr and free the allocated memory afterward. + */ raw_spin_lock_irqsave(&src->pi_lock, flags); - cpumask_copy(dst->user_cpus_ptr, src->user_cpus_ptr); + if (src->user_cpus_ptr) + cpumask_copy(dst->user_cpus_ptr, src->user_cpus_ptr); + else + swap(dst->user_cpus_ptr, user_mask); raw_spin_unlock_irqrestore(&src->pi_lock, flags); + + if (unlikely(user_mask)) + kfree(user_mask); + return 0; } =20 --=20 2.31.1