Paulo Marques wrote:
James Painter wrote:
[...]
The attached patch repairs this. Apply it after the Brownell
patches. Let us know if it fixes your problem.
Unfortunately it doesn't :(
I can confirm that it does only happen in half-duplex mode.
Full-duplex connections seem to work fine.
What is even weirder is that it doesn't _always_ happen. Sometimes I
connect the adapter and it works fine. It even survives ping floods
with no packet loss at all. Other times I connect it and start a
regular ping (1 per second, etc.) and it stops responding after just a
few seconds.
By the way, there is more traffic on this network than the one I
generate myself. Maybe there is some broadcast packet that the adapter
is receiving that is making it go berserk, and that would explain the
random nature of the lockups...
Too bad.
Running ethereal or another sniffer on the interface when it is failing
might reveal something.
I'll look at this further. I see that we always have the FD flow
control bits turned on, even when runing HD. It might help if we turn
those bits off. If you see pause frames in the sniffer, that would be a
clue. I suppose we should look at the link partner capabilities to
decide whether to turn on flow control.
--Jamie
-------------------------------------------------------
SF.Net email is Sponsored by the Better Software Conference & EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
_______________________________________________
[email protected]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
