Messages in this thread |  | | Date | Wed, 12 Feb 1997 15:32:12 -0800 (PST) | From | "Craig S. Bell" <> | Subject | Re: 2.0.27 major problems #1 -- 3c59x driver. |
| |
Philip Blundell writes:
> A transmitter access conflict is not disaster. There is no need to > reinitialise the controller
Perhaps so... all I know is that when I see transmitter access conflict errors on my 3C59x and 3C90x cards, all of that host's net interfaces promptly wedge. v0.30-all and a couple of odd patches haven't helped. So, there's a need for something. >=-)>
I'm no NIC driver expert, but I tell ya something is indeed amiss. Lest I come off as a mere whiner, I've asked Mr. Becker for advice several times, and have received no response. DejaNews reveals no recent articles.
That's not bad, I suppose, but it does mean that I'm going to vocalise about the problem in an attempt to dig up information. Queries to this and other mailing lists (vortex and vortex-bug included) haven't yielded a solution, although it's comforting to know that others have similar symptoms.
Other problems I have experienced with this driver include mis-identification of cards (3C597-TX showing up as 3C592), resource allocation wierdness (loads once with IRQ5, loads again with IRQ3 and doesn't work).
I don't want to put any undue pressure on him, but any sort of missive from Mr. Becker (even if it's just "acknowlege, be patient") would be welcome. Perhaps he's on holiday...
In the meantime, we're getting an ethernet switch and going back to 3c509's. I consider this a full retreat from the problem, a rare thing in my lengthy experience with linux. >=-(>
cheers, Craig S. Bell
--------------------------------------------------------------- aracnet.com -- Portland's loudest electrons ++ info@aracnet.com Ring +1 503 626.6873 V.34 (33.6K) voice: 626-7696 fax: 626-8675 <a href="http://www.aracnet.com/">aracnet.com info & prices</a>
|  |