Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Mon, 10 Sep 2018 14:14:52 +0900
From:   Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com>
To:     Jiri Slaby <jslaby@suse.cz>
Cc:     kernel test robot <rong.a.chen@intel.com>, lkp@01.org,
        Dmitry Safonov <dima@arista.com>,
        Daniel Axtens <dja@axtens.net>,
        Dmitry Safonov <0x7f454c46@gmail.com>,
        Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com>,
        Dmitry Vyukov <dvyukov@google.com>,
        Tan Xiaojun <tanxiaojun@huawei.com>,
        Peter Hurley <peter@hurleysoftware.com>,
        Pasi =?iso-8859-1?Q?K=E4rkk=E4inen?= <pasik@iki.fi>,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        Michael Neuling <mikey@neuling.org>,
        Mikulas Patocka <mpatocka@redhat.com>,
        linux-kernel@vger.kernel.org, stable@vger.kernel.org
Subject: Re: [LKP] [tty] 0b4f83d510: INFO:task_blocked_for_more_than#seconds
Message-ID: <20180910051452.GA518@jagdpanzerIV>
References: <20180907045041.GF1110@shao2-debian>
 <d5d23bc5-0e11-5cee-6d66-514dfc4fc25b@suse.cz>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <d5d23bc5-0e11-5cee-6d66-514dfc4fc25b@suse.cz>
User-Agent: Mutt/1.10.1 (2018-07-13)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On (09/07/18 08:39), Jiri Slaby wrote:
> > [  244.944070] 
> > [  244.944070] Showing all locks held in the system:
> > [  244.945558] 1 lock held by khungtaskd/18:
> > [  244.946495]  #0: (____ptrval____) (rcu_read_lock){....}, at: debug_show_all_locks+0x16/0x190
> > [  244.948503] 2 locks held by askfirst/235:
> > [  244.949439]  #0: (____ptrval____) (&tty->ldisc_sem){++++}, at: tty_ldisc_ref_wait+0x25/0x50
> > [  244.951762]  #1: (____ptrval____) (&ldata->atomic_read_lock){+.+.}, at: n_tty_read+0x13d/0xa00
> 
> Here, it just seems to wait for input from the user.
> 
> > [  244.953799] 1 lock held by validate_data/655:
> > [  244.954814]  #0: (____ptrval____) (&tty->ldisc_sem){++++}, at: tty_ldisc_ref_wait+0x25/0x50
> > [  244.956764] 1 lock held by dnsmasq/668:
> > [  244.957649]  #0: (____ptrval____) (&tty->ldisc_sem){++++}, at: tty_ldisc_ref_wait+0x25/0x50
> > [  244.959598] 1 lock held by dropbear/734:
> > [  244.967564]  #0: (____ptrval____) (&tty->ldisc_sem){++++}, at: tty_ldisc_ref_wait+0x25/0x50
> 
> Hmm, I assume there is another task waiting for write_ldsem and that one
> prevents these readers to proceed. Unfortunately, due to the defunct
> __ptrval__ pointer hashing crap, we do not see who is waiting for what.
> But I am guessing all are the same locks.

Hmm, interesting. Am I getting it right that the test did pass before.
And now we see that sort of/smells like live-lock right after the
introduction of tty_ldisc_lock() to tty_reopen().

> So I think, we are forced to limit the waiting to 5 seconds in reopen in
> the end too (the same as we do for new open in tty_init_dev).

If I got it right, LKP did test the 5*HZ patch

	retval = tty_ldisc_lock(tty, 5 * HZ);

At least it's
 In-Reply-To: <20180829022353.23568-3-dima@arista.com>

and
 Message-Id: <20180829022353.23568-3-dima@arista.com>

is the patch which does the 5*HZ lock timeout thing.

	-ss