[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20100629145137.GD4175@outflux.net>
Date: Tue, 29 Jun 2010 07:51:37 -0700
From: Kees Cook <kees.cook@...onical.com>
To: Alan Cox <alan@...rguk.ukuu.org.uk>
Cc: linux-kernel@...r.kernel.org, Greg Kroah-Hartman <gregkh@...e.de>,
Andrew Morton <akpm@...ux-foundation.org>,
Tejun Heo <tj@...nel.org>,
Veaceslav Falico <vfalico@...hat.com>,
Alexander Viro <viro@...iv.linux.org.uk>,
Oleg Nesterov <oleg@...hat.com>,
KOSAKI Motohiro <kosaki.motohiro@...fujitsu.com>,
Neil Horman <nhorman@...driver.com>,
Roland McGrath <roland@...hat.com>,
Ingo Molnar <mingo@...e.hu>,
Peter Zijlstra <a.p.zijlstra@...llo.nl>,
Hidetoshi Seto <seto.hidetoshi@...fujitsu.com>,
Stefani Seibold <stefani@...bold.net>,
Thomas Gleixner <tglx@...utronix.de>,
Eric Paris <eparis@...hat.com>,
James Morris <jmorris@...ei.org>,
"Andrew G. Morgan" <morgan@...nel.org>,
Dhaval Giani <dhaval.giani@...il.com>,
"Serge E. Hallyn" <serue@...ibm.com>,
Steve Grubb <sgrubb@...hat.com>,
Christoph Hellwig <hch@....de>, linux-fsdevel@...r.kernel.org
Subject: Re: [PATCH v2] sanitize task->comm to avoid leaking escape codes
On Tue, Jun 29, 2010 at 10:36:50AM +0100, Alan Cox wrote:
> > Through get_task_comm() and many direct uses of task->comm in the kernel,
> > it is possible for escape codes and other non-printables to leak into
> > dmesg, syslog, etc. In the worst case, these strings could be used to
> > attack administrators using vulnerable terminal emulators, and at least
> > cause confusion through the injection of \r characters.
>
> If an administrator has a vulnerable terminal emulator they have other
> problems.
Totally agreed.
> Please do any filtering you must in the yama security module where it
> only affects that. One way to approach it without losing data within the
> module might be to use HTML style encoding within Yama so your own tools
> can undo the 'sanitizing' rather than losing information ?
I'm not interested in sanitizing this in Yama. The use of task->comm via
printk was seen as a flaw. I didn't agree (see above about terminal),
and suggested that if it was a flaw, it was a flaw with printk or
task->comm itself. Since "fixing" both of those have been vetoed,
I have no more interest in the filtering.
What I do have interest in is fixing get_task_comm's use of buffers, which
is theoretically problematic in some future where someone accidentally
calls it with a buffer smaller than sizeof(task->comm).
I'll send a patch that only fixes that and leaves out the filtering.
-Kees
--
Kees Cook
Ubuntu Security Team
--
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/
Powered by blists - more mailing lists