Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758938AbcJYNBc (ORCPT ); Tue, 25 Oct 2016 09:01:32 -0400 Received: from mail-yb0-f178.google.com ([209.85.213.178]:37128 "EHLO mail-yb0-f178.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1758772AbcJYNB3 (ORCPT ); Tue, 25 Oct 2016 09:01:29 -0400 MIME-Version: 1.0 In-Reply-To: <20161025125615.GA4326@redhat.com> References: <20161025110357.8821-1-roman.penyaev@profitbricks.com> <20161025125615.GA4326@redhat.com> From: Roman Penyaev Date: Tue, 25 Oct 2016 15:00:58 +0200 Message-ID: Subject: Re: [PATCH v3 1/1] workqueue: ignore dead tasks in a workqueue sleep hook To: Oleg Nesterov Cc: Andy Lutomirski , Peter Zijlstra , Thomas Gleixner , Ingo Molnar , Tejun Heo , linux-kernel@vger.kernel.org Content-Type: text/plain; charset=UTF-8 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1411 Lines: 38 On Tue, Oct 25, 2016 at 2:56 PM, Oleg Nesterov wrote: > On 10/25, Roman Pen wrote: >> >> struct task_struct *wq_worker_sleeping(struct task_struct *task) >> { >> - struct worker *worker = kthread_data(task), *to_wakeup = NULL; >> + struct worker *worker, *to_wakeup = NULL; >> struct worker_pool *pool; >> >> + >> + if (task->state == TASK_DEAD) { >> + /* >> + * Here we try to catch the following path before >> + * accessing NULL kthread->vfork_done ptr thru >> + * kthread_data(): >> + * >> + * oops_end() >> + * do_exit() >> + * schedule() >> + * >> + * If panic_on_oops is not set and oops happens on >> + * a workqueue execution path, thread will be killed. >> + * That is definitly sad, but not to make the situation >> + * even worse we have to ignore dead tasks in order not >> + * to step on zeroed out members (e.g. t->vfork_done is >> + * already NULL on that path, since we were called by >> + * do_exit())). >> + */ >> + return NULL; >> + } > > I still think that PF_EXITING check makes more sense than TASK_DEAD, > but I won't insist. Why? I probably do not see the corner cases, so, please, explain. -- Roman