Subject: Re: recent ep0 problems
To: None <firstname.lastname@example.org>
From: Wolfgang Rupprecht <email@example.com>
Date: 05/18/2000 12:57:05
firstname.lastname@example.org (Wolfgang Rupprecht) writes:
> I just noticed a problem with ep0 on -current when run on my most
> ancient x86 machine, a toshiba 4800.
> Did something change and I need to turn on some special "fixup" flags
> to get the old behavior?
[I just want to make sure the answer makes it into the archives in case
someone else is running ancient junk like I am. -wsr]
I think I figured it out. A long time ago I needed to run a
slightly hacked kernel on the t4800 to account for an apparent lack of
some address lines going to the PCMCIA chip.
# T4800CT is busted with only 10bits of address wired to PCMCIA.
options PCIC_ISA_ALLOC_IOBASE=0x380 # 0x400 doesn't work--address lines?
Then at a later time I could run the same kernel as on all my other
x86 machines. I guess the slightly older PCMCIA drivers just
naturally packed things into the "good" address space. I hate it when
hardware designers save a few wires like this. So much pain for so
A big thanks to Martin Husemann for helping me eliminate recent
changes to the elink code as the culprit. It cut down my search space
Wolfgang Rupprecht <email@example.com>
DGPS signals via the Internet http://www.wsrcc.com/wolfgang/gps/dgps-ip.html