wd7193

[Martin Murray]

> I got it compiled under 2.2.10, but it doesn't work (see above). 
> 
> I will try to implement "manual" download of the firmware instead of download by
> the RISC. If this doesn't help, we all have to buy new cards.
> 
> Martin, how much have you changed in the code? I have seen a issue in the
> wd719x_detect function, so that interrupt-stuff is not given back when
> initialization of the card doesn't conclude. Perhaps we have to do some sort of
> merge?

	I've pushed, proded, changed, hacked, and made the code look like
spaghetti, however, have nothing conclusive. Nothing worth showing off at
least. I Dunno, right now, I downloaded all the 2.1.x patches and have
been working my way along. I had some trouble with the early 2.1.x kernels
and a new compiler, so I started up with 119, 109, 99, 89 and finally then
79. Had a fun time trying to figure out why egcs didn't like doing a btl
with %dl, but, wanted %edx instead. *grin* Anyways. I got as far as 79
without it working. Since I have yet to find a kernel where it does work,
I think I may be doing something wrong. I decided to get closer to .54,
the last kernel for which you provide a Kernel-Diff, and I assume works.
However, .59 had considerably more trouble, but, I think thats something I
can fix. I don't know the SCSI standard, and don't have a clear picture of
how this all fits together, but, I'm hoping figuring out where the driver
breaks will help me out. Have you had any success?

Martin

Martin Murray  -  Nil carborundum illigitimi!
(B8,00,4C,CD,21)