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 4AD46C433F5 for ; Wed, 27 Apr 2022 17:31:33 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S244034AbiD0Rem (ORCPT ); Wed, 27 Apr 2022 13:34:42 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:53582 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S244082AbiD0Rel (ORCPT ); Wed, 27 Apr 2022 13:34:41 -0400 Received: from out01.mta.xmission.com (out01.mta.xmission.com [166.70.13.231]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id B142D1BFC97; Wed, 27 Apr 2022 10:31:21 -0700 (PDT) Received: from in02.mta.xmission.com ([166.70.13.52]:35920) by out01.mta.xmission.com with esmtps (TLS1.3) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.93) (envelope-from ) id 1njlVX-000Gnc-23; Wed, 27 Apr 2022 11:31:19 -0600 Received: from ip68-227-174-4.om.om.cox.net ([68.227.174.4]:35954 helo=email.froward.int.ebiederm.org.xmission.com) by in02.mta.xmission.com with esmtpsa (TLS1.3) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.93) (envelope-from ) id 1njlVV-00C4Y5-KZ; Wed, 27 Apr 2022 11:31:18 -0600 From: "Eric W. Biederman" To: Oleg Nesterov Cc: linux-kernel@vger.kernel.org, rjw@rjwysocki.net, mingo@kernel.org, vincent.guittot@linaro.org, dietmar.eggemann@arm.com, rostedt@goodmis.org, mgorman@suse.de, bigeasy@linutronix.de, Will Deacon , tj@kernel.org, linux-pm@vger.kernel.org, Peter Zijlstra , Richard Weinberger , Anton Ivanov , Johannes Berg , linux-um@lists.infradead.org, Chris Zankel , Max Filippov , inux-xtensa@linux-xtensa.org, Kees Cook , Jann Horn References: <878rrrh32q.fsf_-_@email.froward.int.ebiederm.org> <20220426225211.308418-9-ebiederm@xmission.com> <20220427160901.GI17421@redhat.com> <87o80m7afv.fsf@email.froward.int.ebiederm.org> <20220427171829.GJ17421@redhat.com> <20220427172109.GK17421@redhat.com> Date: Wed, 27 Apr 2022 12:31:09 -0500 In-Reply-To: <20220427172109.GK17421@redhat.com> (Oleg Nesterov's message of "Wed, 27 Apr 2022 19:21:10 +0200") Message-ID: <87wnfa5t7m.fsf@email.froward.int.ebiederm.org> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/27.1 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain X-XM-SPF: eid=1njlVV-00C4Y5-KZ;;;mid=<87wnfa5t7m.fsf@email.froward.int.ebiederm.org>;;;hst=in02.mta.xmission.com;;;ip=68.227.174.4;;;frm=ebiederm@xmission.com;;;spf=softfail X-XM-AID: U2FsdGVkX1+XoRe+WnUdHjZVGZxm2GNPI+jnxvpu6gg= X-SA-Exim-Connect-IP: 68.227.174.4 X-SA-Exim-Mail-From: ebiederm@xmission.com Subject: Re: [PATCH 9/9] ptrace: Don't change __state X-SA-Exim-Version: 4.2.1 (built Sat, 08 Feb 2020 21:53:50 +0000) X-SA-Exim-Scanned: Yes (on in02.mta.xmission.com) Precedence: bulk List-ID: X-Mailing-List: linux-pm@vger.kernel.org Oleg Nesterov writes: > On 04/27, Oleg Nesterov wrote: >> >> On 04/27, Eric W. Biederman wrote: >> > >> > Oleg Nesterov writes: >> > >> > > On 04/26, Eric W. Biederman wrote: >> > >> >> > >> @@ -253,7 +252,7 @@ static int ptrace_check_attach(struct task_struct *child, bool ignore_state) >> > >> */ >> > >> if (lock_task_sighand(child, &flags)) { >> > >> if (child->ptrace && child->parent == current) { >> > >> - WARN_ON(READ_ONCE(child->__state) == __TASK_TRACED); >> > >> + WARN_ON(child->jobctl & JOBCTL_DELAY_WAKEKILL); >> > > >> > > This WARN_ON() doesn't look right. >> > > >> > > It is possible that this child was traced by another task and PTRACE_DETACH'ed, >> > > but it didn't clear DELAY_WAKEKILL. >> > >> > That would be a bug. That would mean that PTRACE_DETACHED process can >> > not be SIGKILL'd. >> >> Why? The tracee will take siglock, clear JOBCTL_DELAY_WAKEKILL and notice >> SIGKILL after that. > > Not to mention that the tracee is TASK_RUNNING after PTRACE_DETACH wakes it > up, so the pending JOBCTL_DELAY_WAKEKILL simply has no effect. Oh. You are talking about the window when between clearing the traced state and when tracee resumes executing and clears JOBCTL_DELAY_WAKEKILL. I thought you were thinking about JOBCTL_DELAY_WAKEKILL being leaked. That requires both ptrace_attach and ptrace_check_attach for the new tracer to happen before the tracee is scheduled to run. I agree. I think the WARN_ON could reasonably be moved a bit later, but I don't know that the WARN_ON is important. I simply kept it because it seemed to make sense. Eric