linux-kernel - Re: tty related lockdep trace during bootup on 3.2-rc2

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAFQmdRbkYiOArK09smfMvf2DJ8h-bYvKyXRN_nmX-bPG8ENmjA@mail.gmail.com>
Date:	Wed, 23 Nov 2011 09:58:11 -0800
From:	Havard Skinnemoen <hskinnemoen@...gle.com>
To:	Jiri Slaby <jslaby@...e.cz>
Cc:	Dave Jones <davej@...hat.com>,
	Linux Kernel <linux-kernel@...r.kernel.org>,
	Greg Kroah-Hartman <gregkh@...e.de>,
	Jiri Slaby <jirislaby@...il.com>,
	Alan Cox <alan@...ux.intel.com>
Subject: Re: tty related lockdep trace during bootup on 3.2-rc2

On Wed, Nov 23, 2011 at 2:12 AM, Jiri Slaby <jslaby@...e.cz> wrote:
> Oh, how it can be? You should vhangup the device on disconnect. Then you
> will be sure all close calls were performed before freeing the port.
> Then you can free the port, right?
>
> Like:
> * forbid further opens
> * tty_vhangup
> * stop the device
> * free the device state

Ok, interesting. It seems pretty different from what the driver is
currently doing. I'll see if I can figure out how to rework the
disconnect code.

I guess one remaining concern is that someone might still hold the
device open after hangup, so if we free the device state, which
includes the tty_port, something bad might happen, no?

> Also it looks like you have a leak in disconnect when there are users
> still. If I am looking correctly?

Funny you should say that. I'm also trying to investigate a case in
which the driver appears to leak tty devices, i.e. after a few days of
stress testing, /sys/class/tty is full of dangling ttyACMx symlinks
and new devices are ignored because acm_table is full.

Havard
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/