From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
        id S1034036AbcIWKWo (ORCPT <rfc822;w@1wt.eu>);
        Fri, 23 Sep 2016 06:22:44 -0400
Received: from mx1.redhat.com ([209.132.183.28]:14621 "EHLO mx1.redhat.com"
        rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
        id S1033909AbcIWKWk (ORCPT <rfc822;linux-kernel@vger.kernel.org>);
        Fri, 23 Sep 2016 06:22:40 -0400
Date: Fri, 23 Sep 2016 12:21:41 +0200
From: Oleg Nesterov <oleg@redhat.com>
To: Michal Hocko <mhocko@kernel.org>
Cc: LKML <linux-kernel@vger.kernel.org>,
        strace-devel@lists.sourceforge.net,
        Mike Galbraith <umgwanakikbuti@gmail.com>,
        Aleksa Sarai <asarai@suse.com>
Subject: Re: strace lockup when tracing exec in go
Message-ID: <20160923102141.GA16189@redhat.com>
References: <20160921152946.GA24210@dhcp22.suse.cz> <1474517702.4851.21.camel@gmail.com> <20160922080126.GA11875@dhcp22.suse.cz> <20160922083602.GB11875@dhcp22.suse.cz> <1474537209.5022.8.camel@gmail.com> <20160922095303.GD11875@dhcp22.suse.cz> <1474538945.5022.20.camel@gmail.com> <20160922110925.GE11875@dhcp22.suse.cz> <20160922135301.GF11875@dhcp22.suse.cz>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20160922135301.GF11875@dhcp22.suse.cz>
User-Agent: Mutt/1.5.18 (2008-05-17)
X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.39]); Fri, 23 Sep 2016 10:22:40 +0000 (UTC)
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On 09/22, Michal Hocko wrote:
>
> --- a/kernel/signal.c
> +++ b/kernel/signal.c
> @@ -91,6 +91,10 @@ static int sig_ignored(struct task_struct *t, int sig, bool force)
>  	if (!sig_task_ignored(t, sig, force))
>  		return 0;
>  
> +	/* Do not ignore signals sent from child to the parent */
> +	if (current->ptrace && current->parent == t)
> +		return 0;

This doesn't look right in general, and this can't really help.

This assumes that the tracer will call do_wait() after mm_access()
fails, but this is not necessarily true.

Note also ptrace_attach(), -ERESTARTNOINTR means that the tracer won't
even return to user-space if SIGCHLD is ignored, the tracer will silently
restart the syscall.

Oleg.