netdev - Re: [PATCH net-next-2.6] bonding: fix user-controlled queuing issues

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

Message-ID: <20110222212445.GF11864@gospo.rdu.redhat.com>
Date:	Tue, 22 Feb 2011 16:24:45 -0500
From:	Andy Gospodarek <andy@...yhouse.net>
To:	Phil Oester <kernel@...uxace.com>
Cc:	Andy Gospodarek <andy@...yhouse.net>, netdev@...r.kernel.org,
	Ben Hutchings <bhutchings@...arflare.com>,
	Jay Vosburgh <fubar@...ibm.com>
Subject: Re: [PATCH net-next-2.6] bonding: fix user-controlled queuing
	issues

On Tue, Feb 22, 2011 at 08:48:36AM -0800, Phil Oester wrote:
> On Tue, Feb 22, 2011 at 11:19:41AM -0500, Andy Gospodarek wrote:
> > Phil,
> > 
> > Can you send me the minimal set of tc rules that selects output queue 16
> > and the output of /proc/net/bonding/bond0?
> > 
> > Private email is fine if you do not want to post it to the list.
> > 
> > Thanks,
> > 
> > -andy
> 
> I have no tc rules which select output queue 16 (I have no tc rules at all
> in fact). 
> 
> Output of /proc/net/bonding/bond0 below.  The bond consists of two Intel
> igb nics, which only have 8 queues.  Note, however, that eth0 is an ixgbe,
> which has 16 queues:
> 
> ixgbe 0000:0b:00.0: Multiqueue Enabled: Rx Queue count = 16, Tx Queue count = 16
> 
> which may answer your question as to what is selecting queue 16.  
> 
> Phil 
> 

OK, I've got it now!  I can reproduce this when I recieve frames from a
device not in the bond and transmit those frames out of the bond device.
I suspect that is what you are doing.  That was a case I did not
consider with the original patch.

I will test all of this and post this as a 2-part series with the first
patch being one that should apply to stable as well, so you and others
can have it there.

--
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