Re: A few Decstation questions.
On Tue, Apr 08, 2003 at 08:07:51PM +0200, Karsten Merker wrote:
> Guido Guenther (agx@d.o.) has done some work on backporting the ll/sc emu to
> the kernel-patch-2.4.19-mips package in unstable, but AFAIK it is not yet
> complete. You can try a current 2.4 kernel cvs from cvs.linux-mips.org,
> although a lot of rather radical changes are committed to the cvs currently,
> so the cvs might be broken from time to time.
I don't think I want to playt with the CVS stuff yet. I looked at the
logs and a few diffs from it and it doesn't seem liek the ll/sc
emulation stuff is that simple, given I don't know mips much. I also
don't get the impression that 2.4 cvs had working ll/sc emulation yet
(well at least not the 2.4.21-pre3 branch I grabbed. It is different
than the latest traps.c which seemed to contain the emulation code).
> ttyS0 and ttyS1 can be mouse and keyboard, depending on your DECstation
> model. Which DECstation do you have? Neither my 5000/1xx nor my 5000/240
> have a serial port labeled "1" and the console port is labeled "3" there.
Oops. Yeah of course. Console on 3 (ttyS2), modem on 2 (ttyS1?).
It is a 5000/133.
Any idea what the max port speed is, given I was trying to run a 56k
modem of it at 115200, although 38400 and 9600 didn't help yet.
> We have a preliminary driver, but there are still some issues with the
> general framebuffer driver framework as we found out that a PMAD-AA
> behaves a bit different from what you expect from a standard mono
> framebuffer. Besides we only have one such card which makes distributed
> work on the driver quite difficult. Joey (joey@d.o.) has the lastest
> driver source. I hope we can tackle the driver again on the next developer
> meeting.
I suppose I could ask for a copy to play with given I have a card and
monitor here, currently in a 5000/240 running NetBSD 1.6.
Looking at the NetBSD code, it certainly sounds like a weird card.
Colormap chip, and a couple of sprite generators as far as I understand
it, and two bit planes. I get hte impression from some stuff so far,
that the card actually does grey scale (4 bit perhaps), but is address
as if it did 8bit memory wise. Weird.
I think you meant PMAG-AA btw. The PMAD-AA doesn't work either in linux
as far as I can tell (being a network card), which I also have one of,
which also works fine in NetBSD at the moment. I will deal with it
later.
Len Sorensen
Reply to: