[201] in linux-net channel archive
Re: Patch to fix NE2000 hangs
daemon@ATHENA.MIT.EDU (Ulrich Callmeier)
Fri Apr 14 13:48:38 1995
To: submit-linux-dev-net@ratatosk.yggdrasil.com
From: uc@brian.lunetix.de (Ulrich Callmeier)
Date: 14 Apr 1995 16:07:59 GMT
Paul Gortmaker (paul@rasty.anu.edu.au) wrote:
: Okay, here is my latest attempt at solving everybody's problems
: with ne2000 cards. These patches are against 1.2.5, and will *fail*
: Please test it with ne cards (and other 8390 based cards) and let me
: know the results via here or e-mail.
Congratulations! I think you finally found the bug. A test that
always crashed my ne2000 machine before now runs without crashing.
Here's what I do (brian is my machine, it's got 2 NE2000 clones, pythia
is another machine with a 3c509):
[501] test@brian (~) $ rsh pythia 'cat /dev/zero' | buffer -S 100k >/dev/null
I just use buffer because it shows a running total of bytes transferred.
This used to crash my machine until including stock 1.2.5 after
a maximum of 200Megabyte transferred.
Now I'm running this for quite a while and it has passed a Gigabyte
and still didn't crash. I killed syslogd and klogd before the test
so I'd see any messages on the console. The
eth0: timed out waiting for RDC, kicking board.
message shows up quite frequently (perhaps once a minute)... According
to Paul's info the machine would have crashed then without the patches...
Good work!
Ulrich
--
--------------(-------------------------------------------------------------
Ulrich Callmeier uc@brian.lunetix.de
(op-t (munge-grunge (six-two (eq make-sense nil) (shoot-yourself (foot))))
--------)-----------------------------------------------------------------)-
