From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S965105AbbKCWyK (ORCPT ); Tue, 3 Nov 2015 17:54:10 -0500 Received: from mail-yk0-f171.google.com ([209.85.160.171]:33661 "EHLO mail-yk0-f171.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932652AbbKCWyJ (ORCPT ); Tue, 3 Nov 2015 17:54:09 -0500 Date: Tue, 3 Nov 2015 17:54:05 -0500 From: Tejun Heo To: Jeff Layton Cc: Chris Worley , linux-kernel@vger.kernel.org, bfields@fieldses.org, Michael Skralivetsky , Trond Myklebust , Shaohua Li Subject: Re: timer code oops when calling mod_delayed_work Message-ID: <20151103225405.GG5749@mtj.duckdns.org> References: <20151029103113.2f893924@tlielax.poochiereds.net> <20151029135836.02ad9000@synchrony.poochiereds.net> <20151031020012.GH3582@mtj.duckdns.org> <20151031073400.2cf05d77@tlielax.poochiereds.net> <20151031213107.GA23841@mtj.duckdns.org> <20151031175404.3c57a17a@tlielax.poochiereds.net> <20151102145633.5329f3da@tlielax.poochiereds.net> <20151102203339.7ed8f2bb@synchrony.poochiereds.net> <20151103125504.6649138f@tlielax.poochiereds.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20151103125504.6649138f@tlielax.poochiereds.net> User-Agent: Mutt/1.5.24 (2015-08-30) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hello, Jeff. On Tue, Nov 03, 2015 at 12:55:04PM -0500, Jeff Layton wrote: > > Ok, I built a kernel with that patch reverted and that seems to fix the > > problem. > > > > Looking at the patch, I guess the main difference is that we're no > > longer using add_timer for unbound workqueue tasks. That said, we > > should have possession of the PENDING bit before calling add_timer_on. > > > > So...I'm a little stumped. Any thoughts as to where the race could be? > > Ok, I've been able to reproduce this with a small module that I wrote > as a reproducer: I think I know what's going on. Looks like a bug in timer code when timers are queued on a different cpu than before. Will experiment with it a bit and report back. Thanks. -- tejun