From patchwork Tue May 16 23:58:44 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Beau Belgrave X-Patchwork-Id: 94990 Return-Path: Delivered-To: ouuuleilei@gmail.com Received: by 2002:a59:b0ea:0:b0:3b6:4342:cba0 with SMTP id b10csp788110vqo; Tue, 16 May 2023 17:34:34 -0700 (PDT) X-Google-Smtp-Source: ACHHUZ4wbaYWsDJYJEsHCS+WT7hbguJ32IKFoXFzaHTFsx3j3b7bbR8jEOpWy0N+rcZWB+xiiz6O X-Received: by 2002:a17:902:d488:b0:1aa:e5cd:6478 with SMTP id c8-20020a170902d48800b001aae5cd6478mr52602380plg.58.1684283674065; Tue, 16 May 2023 17:34:34 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1684283674; cv=none; d=google.com; s=arc-20160816; b=b3iwXEJMQ1swHlgWzlpaNXYP2om4SHYVMYK97rm4wsBWGyaEgzqCMxiEf4FM6IT6n0 rMPeaKfJm5bmz80M80yQt/2JwgyxRwzFd4n2YYazHmsb7QNTL+FxP288lbJPYfVACQ8w vrVUxLSSc+nCEZdywv05ntlRdEriJAvqVh6Ui31cL+0YjEo/DvyY0OKwK/Vq7+EYvZ5e CrzxNFtrS/ecrrvc66GFREATBfo2Un8upzhgokxlTDIHWy2hpEJGRWq814LkCeEcVQHT xxtxlcpSl5X9UnI3vdYqo3qK+y3tq9q803vE098HJhisasF8loJ4eQx/yH8Q3HA0xs8J 8MgQ== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:content-transfer-encoding:mime-version :message-id:date:subject:cc:to:from:dkim-signature:dkim-filter; bh=KgFoQk+gBOrhdcQ1wdSjwMmEkduDiZjL6cKJIOny6aY=; b=UDtgo6pT6AA7exI3trbZHkRgj6FHroomVgxoz2OKyTYeiM03RTcp0MKYntK/zdU5s5 mdFbVQ2NMwa63dMmVmKA7bW00uQoSyvO5tR7bgshwT7S9GlY7G/12kF3NEdmwDhY+YlQ UhbLIa7ZD8tMb68b8aMGXUHQVY7zLP3G4fZlJ3jCWgrUlegygwkfX4kvwj8SuRBjbgwz XdTaGMXMeyysIpedcATWfhzd33Nrgh2MueTJKwvZKjKb0V+wCKMyRU6FDlOm5xhxMhTz dMhq6EPdqy87i4FSWVzprtkQnaEIXi0si4KvmVQWCWdacxekjBZRsKORtSOwS+nk0jQS 51Rg== ARC-Authentication-Results: i=1; mx.google.com; dkim=pass header.i=@linux.microsoft.com header.s=default header.b=GtL9MLVi; spf=pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=linux.microsoft.com Received: from out1.vger.email (out1.vger.email. [2620:137:e000::1:20]) by mx.google.com with ESMTP id ja18-20020a170902efd200b001a94b91f412si18359242plb.164.2023.05.16.17.34.18; Tue, 16 May 2023 17:34:34 -0700 (PDT) Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20; Authentication-Results: mx.google.com; dkim=pass header.i=@linux.microsoft.com header.s=default header.b=GtL9MLVi; spf=pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=linux.microsoft.com Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S229534AbjEPX65 (ORCPT + 99 others); Tue, 16 May 2023 19:58:57 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:39984 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S231451AbjEPX6z (ORCPT ); Tue, 16 May 2023 19:58:55 -0400 Received: from linux.microsoft.com (linux.microsoft.com [13.77.154.182]) by lindbergh.monkeyblade.net (Postfix) with ESMTP id 39E7361AD; Tue, 16 May 2023 16:58:50 -0700 (PDT) Received: from W11-BEAU-MD.localdomain (unknown [76.135.27.212]) by linux.microsoft.com (Postfix) with ESMTPSA id 19E4820ECCB3; Tue, 16 May 2023 16:58:49 -0700 (PDT) DKIM-Filter: OpenDKIM Filter v2.11.0 linux.microsoft.com 19E4820ECCB3 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linux.microsoft.com; s=default; t=1684281529; bh=KgFoQk+gBOrhdcQ1wdSjwMmEkduDiZjL6cKJIOny6aY=; h=From:To:Cc:Subject:Date:From; b=GtL9MLVit6qxc7g42Vr2dlkBh8yV89gybMB3RGGhqdSqfVM8qfFzqItGNYde2tynr CYiObjnosvJguo2HC1EUM75a8YcGQrGD1qpkqPJLoKnEIWkNRgyZJE0ZpITrueFfPw T+GHgYv2gwzqqsYcMy00YXYBM1ToJbfQyBbLWpgw= From: Beau Belgrave To: rostedt@goodmis.org, mhiramat@kernel.org Cc: linux-kernel@vger.kernel.org, linux-trace-kernel@vger.kernel.org, ast@kernel.org Subject: [PATCH] tracing/user_events: Use non-RCU context for enabler writes Date: Tue, 16 May 2023 16:58:44 -0700 Message-Id: <20230516235844.9303-1-beaub@linux.microsoft.com> X-Mailer: git-send-email 2.25.1 MIME-Version: 1.0 X-Spam-Status: No, score=-19.8 required=5.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,ENV_AND_HDR_SPF_MATCH,RCVD_IN_DNSWL_MED, SPF_HELO_PASS,SPF_PASS,T_SCC_BODY_TEXT_LINE,USER_IN_DEF_DKIM_WL, USER_IN_DEF_SPF_WL autolearn=ham autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on lindbergh.monkeyblade.net Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org X-getmail-retrieved-from-mailbox: =?utf-8?q?INBOX?= X-GMAIL-THRID: =?utf-8?q?1766099437799781850?= X-GMAIL-MSGID: =?utf-8?q?1766099437799781850?= Currently when events are enabled/disabled a top level mm enumeration is done in a non-RCU context, however, the enablers are still in a RCU-context. Each enabler is updated via user_event_enabler_write() which uses pin_user_pages_remote(). This function can reschedule and should not be used from a RCU-context. Use the same approach as the top level mm enumeration, which builds a temporary list of target enablers within a RCU-context then exits. Then use list from a non-RCU context when calling user_event_enabler_write(). Link: https://lore.kernel.org/all/20230515165707.hv65ekwp2djkjj5i@MacBook-Pro-8.local/ Fixes: 7235759084a4 ("tracing/user_events: Use remote writes for event enablement") Reported-by: Alexei Starovoitov Signed-off-by: Beau Belgrave --- kernel/trace/trace_events_user.c | 69 +++++++++++++++++++++++++------- 1 file changed, 55 insertions(+), 14 deletions(-) base-commit: 3862f86c1529fa0016de6344eb974877b4cd3838 diff --git a/kernel/trace/trace_events_user.c b/kernel/trace/trace_events_user.c index b1ecd7677642..00d956e7e35a 100644 --- a/kernel/trace/trace_events_user.c +++ b/kernel/trace/trace_events_user.c @@ -96,12 +96,15 @@ struct user_event { * these to track enablement sites that are tied to an event. */ struct user_event_enabler { - struct list_head link; - struct user_event *event; - unsigned long addr; + struct list_head link; + struct user_event *event; + unsigned long addr; + + /* Used to build a list within RCU context for a non-RCU context. */ + struct user_event_enabler *next; /* Track enable bit, flags, etc. Aligned for bitops. */ - unsigned int values; + unsigned int values; }; /* Bits 0-5 are for the bit to update upon enable/disable (0-63 allowed) */ @@ -165,6 +168,8 @@ static int user_event_parse(struct user_event_group *group, char *name, char *args, char *flags, struct user_event **newuser); +static struct user_event_enabler +*user_event_enabler_get_all(struct user_event_mm *mm, struct user_event *user); static struct user_event_mm *user_event_mm_get(struct user_event_mm *mm); static struct user_event_mm *user_event_mm_get_all(struct user_event *user); static void user_event_mm_put(struct user_event_mm *mm); @@ -450,25 +455,28 @@ static bool user_event_enabler_exists(struct user_event_mm *mm, static void user_event_enabler_update(struct user_event *user) { - struct user_event_enabler *enabler; struct user_event_mm *mm = user_event_mm_get_all(user); + struct user_event_enabler *enabler; struct user_event_mm *next; int attempt; while (mm) { next = mm->next; + enabler = user_event_enabler_get_all(mm, user); + + if (unlikely(!enabler)) + goto next_mm; + mmap_read_lock(mm->mm); - rcu_read_lock(); - list_for_each_entry_rcu(enabler, &mm->enablers, link) { - if (enabler->event == user) { - attempt = 0; - user_event_enabler_write(mm, enabler, true, &attempt); - } + while (enabler) { + attempt = 0; + user_event_enabler_write(mm, enabler, true, &attempt); + enabler = enabler->next; } - rcu_read_unlock(); mmap_read_unlock(mm->mm); +next_mm: user_event_mm_put(mm); mm = next; } @@ -500,6 +508,35 @@ static bool user_event_enabler_dup(struct user_event_enabler *orig, return true; } +static struct user_event_enabler +*user_event_enabler_get_all(struct user_event_mm *mm, struct user_event *user) +{ + struct user_event_enabler *found = NULL; + struct user_event_enabler *enabler; + + lockdep_assert_held(&event_mutex); + + /* + * We need to get enablers from within a RCU context to a non-RCU + * context quickly. We build a single linked list under the RCU + * to enable this. Typically this is required for doing actions + * on enablers that might sleep or schedule. + */ + rcu_read_lock(); + + list_for_each_entry_rcu(enabler, &mm->enablers, link) { + if (enabler->event == user) { + enabler->next = found; + found = enabler; + break; + } + } + + rcu_read_unlock(); + + return found; +} + static struct user_event_mm *user_event_mm_get(struct user_event_mm *mm) { refcount_inc(&mm->refcnt); @@ -513,6 +550,8 @@ static struct user_event_mm *user_event_mm_get_all(struct user_event *user) struct user_event_enabler *enabler; struct user_event_mm *mm; + lockdep_assert_held(&event_mutex); + /* * We do not want to block fork/exec while enablements are being * updated, so we use RCU to walk the current tasks that have used @@ -525,13 +564,15 @@ static struct user_event_mm *user_event_mm_get_all(struct user_event *user) */ rcu_read_lock(); - list_for_each_entry_rcu(mm, &user_event_mms, link) - list_for_each_entry_rcu(enabler, &mm->enablers, link) + list_for_each_entry_rcu(mm, &user_event_mms, link) { + list_for_each_entry_rcu(enabler, &mm->enablers, link) { if (enabler->event == user) { mm->next = found; found = user_event_mm_get(mm); break; } + } + } rcu_read_unlock();