linux-kernel - Re: open() on /dev/tty takes 30 seconds on 2.6.36

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] [day] [month] [year] [list]

Message-ID: <20101103103458.2f0bac8c@lxorguk.ukuu.org.uk>
Date:	Wed, 3 Nov 2010 10:34:58 +0000
From:	Alan Cox <alan@...rguk.ukuu.org.uk>
To:	Arnd Bergmann <arnd@...db.de>
Cc:	Eli Billauer <eli@...lauer.co.il>,
	Andrew Morton <akpm@...ux-foundation.org>,
	linux-kernel@...r.kernel.org, Greg KH <greg@...ah.com>
Subject: Re: open() on /dev/tty takes 30 seconds on 2.6.36

> I hope Alan can figure out if it's either safe to drop both here, or if we
> might be able to call uart_close without tty_lock() held in the first place.

That was always my intention and why I moved it to tty_port. I think it
is safe to do that, but as far as I can tell the port mutex is assumed
held by the low level drivers during the uart ops calls some of the time.

Safest is probably to drop the tty lock before we take the port mutex and
take it again when we exit.

The tty_port fields are protected by the port mutex/lock

The uport methods by the uport lock

The only two points of concern I see are updating of closing_wait as it
is read (no big deal), and the nasty - which is tty_ldisc_flush. I am not
sure what assumptions lurk in the ldisc flush paths but I think it's ok.

uart_wait_until_sent will also need to not take the tty lock at that
point to fix it properly.
--
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/