From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751914AbZIBLou (ORCPT ); Wed, 2 Sep 2009 07:44:50 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751829AbZIBLot (ORCPT ); Wed, 2 Sep 2009 07:44:49 -0400 Received: from cam-admin0.cambridge.arm.com ([193.131.176.58]:38175 "EHLO cam-admin0.cambridge.arm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751808AbZIBLos (ORCPT ); Wed, 2 Sep 2009 07:44:48 -0400 Subject: Re: INFO: HARDIRQ-safe -> HARDIRQ-unsafe lock order detected for kmemleak_lock From: Catalin Marinas To: Eric Paris Cc: linux-kernel@vger.kernel.org In-Reply-To: <1251888848.28251.7.camel@pc1117.cambridge.arm.com> References: <1251838545.2158.67.camel@dhcp231-106.rdu.redhat.com> <1251888848.28251.7.camel@pc1117.cambridge.arm.com> Content-Type: text/plain Organization: ARM Ltd Date: Wed, 02 Sep 2009 12:44:37 +0100 Message-Id: <1251891877.28251.12.camel@pc1117.cambridge.arm.com> Mime-Version: 1.0 X-Mailer: Evolution 2.22.3.1 Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 02 Sep 2009 11:44:38.0566 (UTC) FILETIME=[C06F7460:01CA2BC2] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 2009-09-02 at 11:54 +0100, Catalin Marinas wrote: > On Tue, 2009-09-01 at 16:55 -0400, Eric Paris wrote: > > I wrote a multithreaded inotify syscall pounder intended to create > > files, destroy files, create watches, and destroy watches with the > > maximum number of races possible. Instead after letting it run a while > > I came upon this! And then my system started to crash in all sorts of > > fun and glorious ways (kmem_cache_alloc bugs/panics/whatever) > > > > -Eric > > > > [ 2235.913737] ====================================================== > > [ 2235.914084] [ INFO: HARDIRQ-safe -> HARDIRQ-unsafe lock order detected ] > > [ 2235.914084] 2.6.31-rc8-next-20090901 #64 > > [ 2235.914084] ------------------------------------------------------ > > [ 2235.914084] syscall_thrash/2516 [HC0[0]:SC0[0]:HE0:SE1] is trying to acquire: > > [ 2235.914084] (kthread_create_lock){+.+...}, at: [] kthread_create+0x73/0x180 > > [ 2235.914084] > > [ 2235.914084] and this task is already holding: > > [ 2235.914084] (kmemleak_lock){..----}, at: [] create_object+0x161/0x2e0 > > [ 2235.914084] which would create a new lock dependency: > > [ 2235.914084] (kmemleak_lock){..----} -> (kthread_create_lock){+.+...} > > Are there other messages from kmemleak printed before that? It looks to > me like kmemleak got an exceptional situation (not being able to > allocate memory or inserting a pointer into the prio search tree) and it > disabled itself. When disabling, it starts a clean-up thread and AFAICT > that's the only condition when kmemleak_lock -> kthread_create_lock > dependency would be created. > > I'm not sure whether disabling interrupts around kthread_run in > kmemleak_cleanup() would solve the problem. Otherwise, maybe the > kmemleak clean-up thread should take a different form or just a thread > waiting for a clean-up event (it currently acquires a mutex and cannot > be used in interrupt context). It looks like the kthread_create_lock cannot be acquired in interrupt context anyway, so the patch below changes this to a workqueue. kmemleak: Do no create the clean-up thread during kmemleak_disable() From: Catalin Marinas The kmemleak_disable() function could be called from various contexts including IRQ. It creates a clean-up thread but the kthread_create() function has restrictions on which contexts it can be called from, mainly because of the kthread_create_lock. The patch changes the kmemleak clean-up thread to a workqueue. Signed-off-by: Catalin Marinas Reported-by: Eric Paris --- mm/kmemleak.c | 22 +++++----------------- 1 files changed, 5 insertions(+), 17 deletions(-) diff --git a/mm/kmemleak.c b/mm/kmemleak.c index 401a89a..b336201 100644 --- a/mm/kmemleak.c +++ b/mm/kmemleak.c @@ -92,6 +92,7 @@ #include #include #include +#include #include #include @@ -1477,7 +1478,7 @@ static const struct file_operations kmemleak_fops = { * Perform the freeing of the kmemleak internal objects after waiting for any * current memory scan to complete. */ -static int kmemleak_cleanup_thread(void *arg) +static void kmemleak_do_cleanup(struct work_struct *work) { struct kmemleak_object *object; @@ -1489,22 +1490,9 @@ static int kmemleak_cleanup_thread(void *arg) delete_object_full(object->pointer); rcu_read_unlock(); mutex_unlock(&scan_mutex); - - return 0; } -/* - * Start the clean-up thread. - */ -static void kmemleak_cleanup(void) -{ - struct task_struct *cleanup_thread; - - cleanup_thread = kthread_run(kmemleak_cleanup_thread, NULL, - "kmemleak-clean"); - if (IS_ERR(cleanup_thread)) - pr_warning("Failed to create the clean-up thread\n"); -} +static DECLARE_WORK(cleanup_work, kmemleak_do_cleanup); /* * Disable kmemleak. No memory allocation/freeing will be traced once this @@ -1522,7 +1510,7 @@ static void kmemleak_disable(void) /* check whether it is too early for a kernel thread */ if (atomic_read(&kmemleak_initialized)) - kmemleak_cleanup(); + schedule_work(&cleanup_work); pr_info("Kernel memory leak detector disabled\n"); } @@ -1618,7 +1606,7 @@ static int __init kmemleak_late_init(void) * after setting kmemleak_initialized and we may end up with * two clean-up threads but serialized by scan_mutex. */ - kmemleak_cleanup(); + schedule_work(&cleanup_work); return -ENOMEM; } -- Catalin