msk link problems on 8.0

Chris Cowart ccowart at timesinks.net
Mon Dec 7 19:43:55 UTC 2009


Pyun YongHyeon wrote:
> On Sun, Dec 06, 2009 at 06:17:46PM -0800, Chris Cowart wrote:
>> Having read the PR, I copied sys/dev/{msk,e1000} from HEAD into the
> 
> I think the PR has nothing to do with this issue.

You're right. I found it when I was hunting for an explanation to the
hanging that seemed to come hand-in-hand with "Rx FIFO overrun" logs.

>> 8.0-p1 source tree and installed the resulting kernel. The behavior
>> did not change. If anyone has any alternative patching I could do,
>> either against 8.0 or HEAD for these drivers, I'd be more than
>> willing to try them out.
> 
> There was a similar report on missing link state change and I think I
> committed fix. Are you sure you used latest if_msk.c(r199413) in HEAD?
> See if you have the following patch.
> http://svn.freebsd.org/viewvc/base/head/sys/dev/msk/if_msk.c?r1=199012&r2=199413&view=patch

I did get the file from HEAD, and I have verified that:

|    if ((sc_if->msk_flags & MSK_FLAG_LINK) == 0) |
msk_miibus_statchg(sc_if->msk_if_dev);

is in the patched kernel I built.

> > On a related note, last night, when the system did boot, I would
> > also run into a problem where the following message would be logged:
> > "msk0: Rx FIFO overrun!". Once logged, the NIC seemed to be
> > completely wedged
> 
> At least this indicates you didn't use latest msk(4) in HEAD because
> the message was removed there.

Sorry I was unclear on the chronology; I was seeing this behavior before
I honed in on the link-state problem and built the msk(4) from HEAD.

> > and unusable. Doing ifconfig down/up did not help things. At the
> > time, I hadn't discovered the physical down/up workaround, so I
> > can't speak to whether that would have helped (and this error
        > > condition hasn't recurred (knock on wood)). I don't know if
> > the issues are related or
> 
> I think it would be different issue, let's fix link state issue first.

I think you're right that there are two issues here; I think I may have
confused matters a little by intertwining my observations from both of
them. Given that I've tried that patch and experienced the same link
state behavior, is there something else I can try?

Thanks,

-- 
Chris Cowart
  http://www.timesinks.net/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 834 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/freebsd-net/attachments/20091207/10d91020/attachment.pgp


More information about the freebsd-net mailing list