> Inaccurate since you proceeded to bash the user who had a problem, and then
> made a comment that linux/sparc doesn't have that problem, while failing
> to actually define 'problem' (or whether NetBSD even had it, which it
> didn't).
 
Actually, he did define the problem, if you grabed Linux kernel source
code you could fix the problem in a snap, I think this is a good
enough description of the problem:

> Aaron doesn't even attempt to check if the le is using a tpe or aui
> connection, nor does he ask the prom for what burst sizes the bus is
> capable of handling for dvma.  Therefore I'm surprised the le ever

You can get the patch that fixes this for Linux from:
vger.rutgers.edu:/pub/linux/Sparc/kernel

Look at drivers/net/sunlance.c

Author for the patch is ecd@Pool.Informatik.RWTH-Aachen.DE (Christian
Dost).

Miguel.