netdev - Re: [PATCH 2.6.28 1/1] cxgb3

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: <20081002180011.16254a4e.akpm@linux-foundation.org>
Date:	Thu, 2 Oct 2008 18:00:11 -0700
From:	Andrew Morton <akpm@...ux-foundation.org>
To:	Divy Le Ray <divy@...lsio.com>
Cc:	jeff@...zik.org, netdev@...r.kernel.org,
	linux-kernel@...r.kernel.org, swise@...ngridcomputing.com
Subject: Re: [PATCH 2.6.28 1/1] cxgb3 - fix race in EEH

On Thu, 25 Sep 2008 17:05:28 -0700
Divy Le Ray <divy@...lsio.com> wrote:

> A SGE queue set timer might access registers while in EEH recovery,
> triggering an EEH error loop. Stop all timers early in EEH process.

<looks>

It's deeply weird that t3_reset_qset() does

	memset(&q->tx_reclaim_timer, 0, sizeof(q->tx_reclaim_timer));

There are lots of things in the timer_list which the driver has no
business modifying.  For example, this might break the metadata in
Thomas's debugobjects stuff, which attempts to catch things being done
in the wrong order (I don't think it will, but still...).

Rerunning init_timer() should repair the damage, but I suspect a simple

	q->tx_reclaim_timer.function = NULL;	/* explanation goes here */

would suffice here.


t3_sge_alloc_qset() could use the newer setup_timer().
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html