Re: xl0 and SNAP004

On 6/8/05, Miguel Mendez <flynn_at_energyhq.es.eu.org> wrote:
> Hi all,
> 
> I cvsuped my tree to HEAD last Sunday but the resulting kernel panic'ed
> as soon as ifconfig xl0 was ran. I have the vmcore saved but had to
> revert to 5.4 as I needed to use the box. I've downloaded the SNAP004
> iso and copied the kernel from base to test it before attempting a real
> install, and this what I've seen. No panic but some kernel messages
> have poped up:
> 
> Jun  8 17:30:37 scienide kernel: taskqueue_drain with the following non-
> sleepable locks held: Jun  8 17:30:37 scienide kernel: exclusive sleep
> mutex xl0 (network driver) r = 0 (0xc187f2fc) locked _at_ /u sr/src/sys/
> pci/if_xl.c:2789 Jun  8 17:30:37 scienide kernel: KDB: stack backtrace:

[snip backtrace]

> Is this a known problem? I don't see any recent commit to if_xl.c so I
> assume the issue is still there in HEAD.

Yes this is a known problem and has been acknowleged by Bill Paul (a.o.)
I hope to receive a patch from Gleb anytime soon to see if it resolves
the problem on my amd64 box where i experience the exact same issue.

-- 
  Pascal Hofstee