netdev - Re: Kernel problem

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: <20090226.041037.134928084.davem@davemloft.net>
Date:	Thu, 26 Feb 2009 04:10:37 -0800 (PST)
From:	David Miller <davem@...emloft.net>
To:	herbert@...dor.apana.org.au
Cc:	jarkao2@...il.com, ash@...sky.net, netdev@...r.kernel.org
Subject: Re: Kernel problem

From: Herbert Xu <herbert@...dor.apana.org.au>
Date: Thu, 26 Feb 2009 19:56:04 +0800

> On Thu, Feb 26, 2009 at 10:56:16AM +0000, Jarek Poplawski wrote:
> > 
> > netconsole. It's a known issue, btw.
> 
> I see.  So the problem in this case appears to be that we're
> processing incoming packets in netpoll_poll.  So the obvious
> question can we just not do that since all we need to do is
> transmit packets?

netpoll supports receiving packets and has a trap handler to eat them
to avoid packets it is interested in going into the real stack.
(netpoll_receive_skb, netpoll_rx_skb, etc.)

I'm pretty sure this has been discussed before. :-)

If I'm not mistaken, we only see this with certain drivers, and the
issue therefore likely has to do with how such drivers implement their
poll handler.

If you do your ->poll_controller() handler without taking IRQ locks,
you're fine.  Or something like that...
--
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