G-NIC II Issues.

[Chris Johns]

Keith Underwood wrote:
> 
> The list is archived at http://beowulf.gsfc.nasa.gov
> 

Thanks.

> A newer driver can be found at
> http://www.nscl.msu.edu/~kasten/perf/hamachi/
> but I'm making no promises that it is necessarily "better".  It did seem
> to fix a problem I was having though...
> 

Version 0.14 does not fix my problem. I still get the skput overflow
panic. The length of 65532 is (0 - 4) which to me is a little sus.

Checking the rx code looking for a -4 I noticed :

  u16 pkt_len = frame_status - 4; /* Implicit truncate, omit CRC */

How does pkt_len relate to the frame_status ? What happened to data_size
?

I would expect frame_status to be 0 on a good frame, ie my bug.

What I am finding strange is I have had the driver working, and if the
above is a bug, how did the driver work ?

I am yet to test the change.

-- 
 Chris Johns, mailto:ccj@acm.org
 | To unsubscribe, send mail to Majordomo@cesdis.gsfc.nasa.gov, and within the
 |  body of the mail, include only the text:
 |   unsubscribe this-list-name youraddress@wherever.org
 | You will be unsubscribed as speedily as possible.