netdev - Re: 2.6.30-rc1: e1000: list

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 for Android: free password hash cracker in your pocket

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

Message-Id: <20090409.010439.68760762.davem@davemloft.net>
Date:	Thu, 09 Apr 2009 01:04:39 -0700 (PDT)
From:	David Miller <davem@...emloft.net>
To:	jarkao2@...il.com
Cc:	a.beregalov@...il.com, kernel-testers@...r.kernel.org,
	netdev@...r.kernel.org, e1000-devel@...ts.sourceforge.net
Subject: Re: 2.6.30-rc1: e1000: list_add corruption

From: Jarek Poplawski <jarkao2@...il.com>
Date: Thu, 9 Apr 2009 08:01:59 +0000

> On Thu, Apr 09, 2009 at 12:51:06AM -0700, David Miller wrote:
>> From: Jarek Poplawski <jarkao2@...il.com>
>> Date: Thu, 9 Apr 2009 07:12:42 +0000
>> 
>> > On 09-04-2009 06:20, David Miller wrote:
>> >> From: Alexander Beregalov <a.beregalov@...il.com>
>> >> Date: Thu, 9 Apr 2009 08:16:00 +0400
>> >> 
>> >>> The host has two interfaces
>> >>> eth0 - e1000, eth1 - forcedeth.
>> >>> Both are not working after that messages.
>> >> 
>> >> It looks like the NAPI pending list is being modified in an illegal
>> >> state by forcedeth, and mid-stream to a list_del() an e1000 interrupt
>> >> comes in and this tries to schedule the e1000 NAPI context and we get
>> >> list corruption.
>> > 
>> > Good point! Feel free to send a patch. ;-)
>> 
>> A real shame, since I sent that analysis in hoping someone else
>> would do the rest of the digging and write the fix. :-/
> 
> Hmm... What rest? Alexander?

I mean the part where you actually look at the code and figure out
where interrupts are enabled during a list_del() where  they shouldn't
be.

The initial trace is pretty detailed and has all the information you
need...

Nevermind if I have to sit here and explain all of this I might
as well fucking fix this myself :-/
--
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