Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
        id S1751678AbdG1HpM (ORCPT <rfc822;w@1wt.eu>);
        Fri, 28 Jul 2017 03:45:12 -0400
Received: from mx2.suse.de ([195.135.220.15]:35722 "EHLO mx1.suse.de"
        rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP
        id S1751045AbdG1HpL (ORCPT <rfc822;linux-kernel@vger.kernel.org>);
        Fri, 28 Jul 2017 03:45:11 -0400
Date: Fri, 28 Jul 2017 09:45:09 +0200
From: Jan Kara <jack@suse.cz>
To: Gu Zheng <guzheng1@huawei.com>
Cc: eparis@redhat.com, viro@zeniv.linux.org.uk, Jan Kara <jack@suse.cz>,
        linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
        "miaoxie@huawei.com" <miaoxie@huawei.com>
Subject: Re: Fwd: trinity test fanotify cause hungtasks on kernel 4.13
Message-ID: <20170728074509.GA29433@quack2.suse.cz>
References: <5979B8A5.205@huawei.com>
 <597AB4ED.9090306@huawei.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <597AB4ED.9090306@huawei.com>
User-Agent: Mutt/1.5.24 (2015-08-30)
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org
Content-Length: 1281
Lines: 40

On Fri 28-07-17 11:52:13, Gu Zheng wrote:
> 
> hi,ALL:
> when we used the trinity test the fanotify interfaces, it cause many hungtasks.
> CONFIG_FANOTIFY_ACCESS_PERMISSIONS=y
> the shell is  simple:
>   1 #!/bin/bash
>   2
>   3 while true
>   4 do
>   5 ./trinity -c fanotify_init -l off -C 2 -X > /dev/null 2>&1 &
>   6 sleep 1
>   7 ./trinity -c fanotify_mark -l off -C 2 -X > /dev/null 2>&1 &
>   8 sleep 10
>   9 done
> we found the trinity enter the D state fastly.
> we check the pids'stack

<snip>

> all progresses are waiting for the response in fanotify_handle_event->fanotify_get_response,
> becauseof non-response or killed monitor,so the waitqueue is  in blocked state,
> then the others will be stucked which use the  fanotify_get_response.

Yes, this all works as expected.

> if we use wait_event_timeout , the responed time can not be guaranteed.
> 
> do you have any ideas?

So if your target is sensible trinity testing of fanotify, what you could
do is that you would gather all descriptors where fanotify_init()
succeeded, watch them and reply to all fanotify events. Alternatively you
could just close these descriptors - that would tear down fanotify instance
and all events will be auto-accepted.

								Honza
-- 
Jan Kara <jack@suse.com>
SUSE Labs, CR