Re: Ultra5 hard lockups with 4.0_STABLE and hme status messages

[Julian Coleman <jdc%coris.org.uk@localhost>]

Hi,

> I don't know if its related but once up, I noticed a series of messages
> like the following:
> 
>   hme0: status=8<CCNTEXP>
>   hme0: status=40<CVCNTEXP>
>   hme0: status=10009<GOTFRAME,CCNTEXP,RXTOHOST>
>   hme0: status=108<CCNTEXP,SENTFRAME>

> Looking at "hmereg.h" the descriptions of these status messages sound
> rather benign.  Unless I'm just naive.

It looks like you are getting a number of receive packet errors.  Looking at
the HME User's Guide [1], these counters only hold 256 entries so you will
get an expired (overflowed) status every 256 errors.  The usual cause of
CRC errors (CCNTEXP) is a duplex mismatch (one end half- and one end full-
duplex).  Could this be the case here?

In the gem driver, the receive error counters are now counted as input errors.
However, we don't appear to use the individual counters in the hme driver.  It
would probably make sense to do so.

Anyway, I don't think these errors are to do with your lock ups, unless there
is something else that causes the hme driver to lock up.

J

-- 
  My other computer also runs NetBSD    /        Sailing at Newbiggin
        http://www.netbsd.org/        /   http://www.newbigginsailingclub.org/