Subject: Re: port-next68k
To: Soren S. Jorvang <soren@t.DK>
From: Darrin B. Jewell <jewell@MIT.EDU>
Date: 07/09/1998 14:03:17
"Soren S. Jorvang" <soren@t.DK> writes:
> Okay. Do you have any suggestions for things I could try?
Well, you could try compiling the bootstrap with EN_DEBUG
and perhaps playing with next68k/stand/boot/en.c a little.
I'm currently a little suspicious of the "bmap" chip which controls
memory accesses and has a couple of ethernet related bits. I'd like
to know what bits might be useful there and whether they might be
different on different machines. I suspect the memory locations of
devices might be slightly different on the '030 machines due to
different handling of SLOT_ID_BMAP in the nextstep cpu.h header files.
Similarly, I have no idea if the turbo machines are different.
Also, the nextstep kernel seems to continuously check against the
"313" in the "dma_chip" global kernel variable. I know I have
a fujitsu chip with the #313 on it in the center of my motherboard.
Maybe the turbo machines have a different dma_chip value in the
nextstep kernel. I should probably write a nextstep LKM to probe
some of these values, but it is likely I won't be able to do so
until this weekend at the earliest. Maybe someone else already
has a way to probe the turbo machines handy.