From mboxrd@z Thu Jan 1 00:00:00 1970 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 C938FC43334 for ; Tue, 7 Jun 2022 19:14:35 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1355263AbiFGTOd (ORCPT ); Tue, 7 Jun 2022 15:14:33 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:45352 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1351632AbiFGSY6 (ORCPT ); Tue, 7 Jun 2022 14:24:58 -0400 Received: from ams.source.kernel.org (ams.source.kernel.org [145.40.68.75]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 25FAFE8BA9; Tue, 7 Jun 2022 10:54:31 -0700 (PDT) Received: from smtp.kernel.org (relay.kernel.org [52.25.139.140]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ams.source.kernel.org (Postfix) with ESMTPS id 39112B8236F; Tue, 7 Jun 2022 17:54:22 +0000 (UTC) Received: by smtp.kernel.org (Postfix) with ESMTPSA id 1CAABC36B00; Tue, 7 Jun 2022 17:54:20 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1654624460; bh=kC+NO7EMep4FlOBvtM8hUgWjTViRM3xu3H1X51DPgPQ=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=ez0CT5C8M0AgWup72e4nNGC5FS1K0slVcTF1I2/Fq67MiM2lUYUb7lUleoGmU/00E bXQrP3i61KHbGHYOz9Qm6n/vYJLRoWpAUmbSYOUxECChLcoJyFhFtOy0MuHC0EcsHk Nk0KIh3o+4IbQuUWsA+blPhsiZN++gpYQrgJLaehD0wCwLsSyPKKyZmbGzaYk1J2OR 9epnvga2d+wetVmVCl32a+K+xKeo2r63DpUjkqXTuO+BRQ3XH2c44WZJ/akMIVz8nJ WkTqrOIxI8y6sEXJF4EK8czlyx8FRUOY+yXwWs45kUx72becWcCdp3qma5aU6PFMgN 5Kehq+0I5kE8A== From: Sasha Levin To: linux-kernel@vger.kernel.org, stable@vger.kernel.org Cc: Hao Luo , Tejun Heo , Greg Kroah-Hartman , Sasha Levin Subject: [PATCH AUTOSEL 5.17 30/60] kernfs: Separate kernfs_pr_cont_buf and rename_lock. Date: Tue, 7 Jun 2022 13:52:27 -0400 Message-Id: <20220607175259.478835-30-sashal@kernel.org> X-Mailer: git-send-email 2.35.1 In-Reply-To: <20220607175259.478835-1-sashal@kernel.org> References: <20220607175259.478835-1-sashal@kernel.org> MIME-Version: 1.0 X-stable: review X-Patchwork-Hint: Ignore Content-Transfer-Encoding: 8bit Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: Hao Luo [ Upstream commit 1a702dc88e150487c9c173a249b3d236498b9183 ] Previously the protection of kernfs_pr_cont_buf was piggy backed by rename_lock, which means that pr_cont() needs to be protected under rename_lock. This can cause potential circular lock dependencies. If there is an OOM, we have the following call hierarchy: -> cpuset_print_current_mems_allowed() -> pr_cont_cgroup_name() -> pr_cont_kernfs_name() pr_cont_kernfs_name() will grab rename_lock and call printk. So we have the following lock dependencies: kernfs_rename_lock -> console_sem Sometimes, printk does a wakeup before releasing console_sem, which has the dependence chain: console_sem -> p->pi_lock -> rq->lock Now, imagine one wants to read cgroup_name under rq->lock, for example, printing cgroup_name in a tracepoint in the scheduler code. They will be holding rq->lock and take rename_lock: rq->lock -> kernfs_rename_lock Now they will deadlock. A prevention to this circular lock dependency is to separate the protection of pr_cont_buf from rename_lock. In principle, rename_lock is to protect the integrity of cgroup name when copying to buf. Once pr_cont_buf has got its content, rename_lock can be dropped. So it's safe to drop rename_lock after kernfs_name_locked (and kernfs_path_from_node_locked) and rely on a dedicated pr_cont_lock to protect pr_cont_buf. Acked-by: Tejun Heo Signed-off-by: Hao Luo Link: https://lore.kernel.org/r/20220516190951.3144144-1-haoluo@google.com Signed-off-by: Greg Kroah-Hartman Signed-off-by: Sasha Levin --- fs/kernfs/dir.c | 31 +++++++++++++++++++------------ 1 file changed, 19 insertions(+), 12 deletions(-) diff --git a/fs/kernfs/dir.c b/fs/kernfs/dir.c index 4096953390b4..52f3afeed612 100644 --- a/fs/kernfs/dir.c +++ b/fs/kernfs/dir.c @@ -18,7 +18,15 @@ #include "kernfs-internal.h" static DEFINE_SPINLOCK(kernfs_rename_lock); /* kn->parent and ->name */ -static char kernfs_pr_cont_buf[PATH_MAX]; /* protected by rename_lock */ +/* + * Don't use rename_lock to piggy back on pr_cont_buf. We don't want to + * call pr_cont() while holding rename_lock. Because sometimes pr_cont() + * will perform wakeups when releasing console_sem. Holding rename_lock + * will introduce deadlock if the scheduler reads the kernfs_name in the + * wakeup path. + */ +static DEFINE_SPINLOCK(kernfs_pr_cont_lock); +static char kernfs_pr_cont_buf[PATH_MAX]; /* protected by pr_cont_lock */ static DEFINE_SPINLOCK(kernfs_idr_lock); /* root->ino_idr */ #define rb_to_kn(X) rb_entry((X), struct kernfs_node, rb) @@ -229,12 +237,12 @@ void pr_cont_kernfs_name(struct kernfs_node *kn) { unsigned long flags; - spin_lock_irqsave(&kernfs_rename_lock, flags); + spin_lock_irqsave(&kernfs_pr_cont_lock, flags); - kernfs_name_locked(kn, kernfs_pr_cont_buf, sizeof(kernfs_pr_cont_buf)); + kernfs_name(kn, kernfs_pr_cont_buf, sizeof(kernfs_pr_cont_buf)); pr_cont("%s", kernfs_pr_cont_buf); - spin_unlock_irqrestore(&kernfs_rename_lock, flags); + spin_unlock_irqrestore(&kernfs_pr_cont_lock, flags); } /** @@ -248,10 +256,10 @@ void pr_cont_kernfs_path(struct kernfs_node *kn) unsigned long flags; int sz; - spin_lock_irqsave(&kernfs_rename_lock, flags); + spin_lock_irqsave(&kernfs_pr_cont_lock, flags); - sz = kernfs_path_from_node_locked(kn, NULL, kernfs_pr_cont_buf, - sizeof(kernfs_pr_cont_buf)); + sz = kernfs_path_from_node(kn, NULL, kernfs_pr_cont_buf, + sizeof(kernfs_pr_cont_buf)); if (sz < 0) { pr_cont("(error)"); goto out; @@ -265,7 +273,7 @@ void pr_cont_kernfs_path(struct kernfs_node *kn) pr_cont("%s", kernfs_pr_cont_buf); out: - spin_unlock_irqrestore(&kernfs_rename_lock, flags); + spin_unlock_irqrestore(&kernfs_pr_cont_lock, flags); } /** @@ -823,13 +831,12 @@ static struct kernfs_node *kernfs_walk_ns(struct kernfs_node *parent, lockdep_assert_held_read(&kernfs_root(parent)->kernfs_rwsem); - /* grab kernfs_rename_lock to piggy back on kernfs_pr_cont_buf */ - spin_lock_irq(&kernfs_rename_lock); + spin_lock_irq(&kernfs_pr_cont_lock); len = strlcpy(kernfs_pr_cont_buf, path, sizeof(kernfs_pr_cont_buf)); if (len >= sizeof(kernfs_pr_cont_buf)) { - spin_unlock_irq(&kernfs_rename_lock); + spin_unlock_irq(&kernfs_pr_cont_lock); return NULL; } @@ -841,7 +848,7 @@ static struct kernfs_node *kernfs_walk_ns(struct kernfs_node *parent, parent = kernfs_find_ns(parent, name, ns); } - spin_unlock_irq(&kernfs_rename_lock); + spin_unlock_irq(&kernfs_pr_cont_lock); return parent; } -- 2.35.1