From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751874Ab1IEJIL (ORCPT ); Mon, 5 Sep 2011 05:08:11 -0400 Received: from e23smtp03.au.ibm.com ([202.81.31.145]:36691 "EHLO e23smtp03.au.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751223Ab1IEJIH (ORCPT ); Mon, 5 Sep 2011 05:08:07 -0400 Message-ID: <4E64916B.4090205@linux.vnet.ibm.com> Date: Mon, 05 Sep 2011 14:37:55 +0530 From: "Srivatsa S. Bhat" User-Agent: Mozilla/5.0 (X11; Linux i686; rv:6.0) Gecko/20110816 Thunderbird/6.0 MIME-Version: 1.0 To: maciej.rutecki@gmail.com CC: linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org, linux-pm@lists.linux-foundation.org Subject: Re: [BUG] Soft-lockup during cpu-hotplug in VFS callpaths References: <4E550057.9070609@linux.vnet.ibm.com> <201108312040.32531.maciej.rutecki@gmail.com> In-Reply-To: <201108312040.32531.maciej.rutecki@gmail.com> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 09/01/2011 12:10 AM, Maciej Rutecki wrote: > On środa, 24 sierpnia 2011 o 15:44:55 Srivatsa S. Bhat wrote: >> Hi, >> >> While running stressful cpu hotplug tests along with kernel compilation >> running in background, soft-lockups are detected on multiple CPUs. >> Sometimes this also leads to hard lockups and kernel panic. >> All the soft-lockups seem to occur at vfsmount_lock_local_cpu() or other >> VFS callpaths. >> >> >> [37108.410813] BUG: soft lockup - CPU#5 stuck for 22s! [cc1:29669] >> >> [37108.694781] Call Trace: >> [37108.697306] [] ? >> vfsmount_lock_local_lock_cpu+0x70/0x70 [37108.704258] >> [] path_init+0x315/0x400 >> [37108.709558] [] ? __raw_spin_lock_init+0x38/0x70 >> [37108.715812] [] path_openat+0x8c/0x3f0 >> [37108.721203] [] ? sched_clock+0x9/0x10 >> [37108.726597] [] ? sched_clock_cpu+0xcd/0x110 >> [37108.732508] [] ? trace_hardirqs_off+0xd/0x10 >> [37108.738498] [] ? local_clock+0x6f/0x80 >> [37108.743970] [] do_filp_open+0x49/0xa0 >> [37108.749362] [] ? alloc_fd+0xc3/0x210 >> [37108.754665] [] ? _raw_spin_unlock+0x2b/0x40 >> [37108.760575] [] ? alloc_fd+0xc3/0x210 >> [37108.765875] [] do_sys_open+0x107/0x1e0 >> [37108.771352] [] ? audit_syscall_entry+0x1bf/0x1f0 >> [37108.777695] [] sys_open+0x20/0x30 >> [37108.782741] [] system_call_fastpath+0x16/0x1b >> >> Kernel version: 3.0.1, 3.0.3 >> Hardware: Dual socket quad-core hyper-threaded Intel x86 machine >> Scenario: >> (a) Stressful cpu hotplug tests + kernel compilation >> >> (b) IRQ balancing had been disabled and all the IRQs were made to be >> routed to CPU 0 (except the ones that couldn't be routed). >> >> (c) Lockdep was enabled during kernel configuration. >> >> Steps (b) and (c) were done to dig deeper into the issue. However the same >> issue was observed by just doing step (a). >> >> Definitely there seems to be a race condition occurring here, because this >> issue is hit after sometime, after starting the tests. And the time it >> takes to hit the issue increases as we increase the number of debug print >> statements. In some cases (especially when the number of debug print >> statements were quite high), the stress on the machine had to be increased >> in order to hit the issue within measurable time. In my tests, a maximum >> of about 2 to 2.5 hours was sufficient, to hit this bug. >> >> Please find the console log attached with this mail. >> >> Any ideas on how to go about fixing this bug? > > It is a regression? Hi Maciej, Thank you for taking a look. Yes, it seems to be a regression. I tested out kernel 2.6.39.3 with similar test cases for quite a long time, and it did not hit any soft-lockup issues. -- Regards, Srivatsa S. Bhat Linux Technology Center, IBM India Systems and Technology Lab