mah@dec1.wu-wien.ac.at (Michael Haberler) (03/08/91)
I found the solution to the problem with the NE1000 packet driver. Actually its not a problem with the software *at all*. This trace shows it. Well, I didnt see it right away, either: Wed Mar 06 12:49:15 1991 - ec0 recv: Ether: len 60 08:00:2b:14:18:f6->47:54:43:00:05:9b type IP IP: len 32 137.208.1.5->137.208.1.128 ihl 20 ttl 255 prot ICMP ICMP: type Echo Reply id 65535 seq 0 If you look at the ethernet destination address, you'll find 47:54:43 as the manufacturer code, which is hex for the string 'GTC'. Unfortunately, the first bit of 'G' is 1, so to KA9Q looks like a multicast or broadcast adress (can somebody enlighten me on this bit's meaning? I just inferred that from a list of ethernet type fields, multicast adresses and manufacturer number culled from the net). The card `manual' reads 'Local Area Network GT-1000 Ethernet Card, Users Manual'; no further hints about the company are given, so watch out. Those turkeys dont have a registered manufacturer number, and dont know about its meaning. I could imagine that several other software packages break with such a card. I think for me that means an option to the attach command to set the hardware adress, or mask it to a sane value :-( - michael
nelson@sun.soe.clarkson.edu (Russ Nelson) (03/18/91)
In article <1991Mar07.180439.5382@nestroy.wu-wien.ac.at> mah@dec1.wu-wien.ac.at (Michael Haberler) writes:
If you look at the ethernet destination address, you'll find
47:54:43 as the manufacturer code, which is hex for the string
'GTC'. Unfortunately, the first bit of 'G' is 1, so to KA9Q looks
like a multicast or broadcast adress (can somebody enlighten me on
this bit's meaning?
Multicast (one of the many possible multicast addresses is the broadcast
address). I'm going to put a check for multicast addresses into the ne1000
and ne2000 drivers for the 9.x release.
--
--russ <nelson@clutx.clarkson.edu> I'm proud to be a humble Quaker.
It's better to get mugged than to live a life of fear -- Freeman Dyson
I joined the League for Programming Freedom, and I hope you'll join too.rpw3@rigden.wpd.sgi.com (Rob Warnock) (03/20/91)
In article <NELSON.91Mar18120225@sun.clarkson.edu> nelson@clutx.clarkson.edu (aka NELSON@CLUTX.BITNET) writes: +--------------- | Multicast (one of the many possible multicast addresses is the broadcast | address). I'm going to put a check for multicast addresses into the ne1000 | and ne2000 drivers for the 9.x release. +--------------- But please don't *disable* receiving multicast packets! We live and breath sending multicast around here! Have you ever heard of "dog", the multi-user version of SGI's "flight" simulator? It uses UDP multicast. Also, XTP is largely used in multicast mode... -Rob ----- Rob Warnock, MS-1L/515 rpw3@sgi.com rpw3@pei.com Silicon Graphics, Inc. (415)335-1673 Protocol Engines, Inc. 2011 N. Shoreline Blvd. Mountain View, CA 94039-7311
nelson@sun.soe.clarkson.edu (Russ Nelson) (03/21/91)
In article <92430@sgi.sgi.com> rpw3@rigden.wpd.sgi.com (Rob Warnock) writes: In article <NELSON.91Mar18120225@sun.clarkson.edu> nelson@clutx.clarkson.edu (aka NELSON@CLUTX.BITNET) writes: +--------------- | Multicast (one of the many possible multicast addresses is the | broadcast address). I'm going to put a check for multicast | addresses into the ne1000 and ne2000 drivers for the 9.x release. +--------------- But please don't *disable* receiving multicast packets! Oh goodness, no! I'm going to check to see if the board's *own* address has the multicast bit set. I think you can see how that might cause problems. :-) -- --russ <nelson@clutx.clarkson.edu> I'm proud to be a humble Quaker. It's better to get mugged than to live a life of fear -- Freeman Dyson I joined the League for Programming Freedom, and I hope you'll join too.