* Eric W Biederman <[EMAIL PROTECTED]> [020715 14:03]:
> Stefan Reinauer <[EMAIL PROTECTED]> writes:
>
> There was a fair bit that was Tsunami specific. But it was written to
> be as well factored, and general as possible. That just isn't very
> possible on the alpha.
I know this from milo code. but it should be acceptable to #ifdef
different alpha platforms or even better have seperate files for
each of those platforms.
> I won't get in the way. This does assume people have time to do the work,
> and no market potential makes those people harder to find.
agreed. Though the bootloader situation for all non-srm alphas is severe
anyways. it should be easier to find people willing to look into
linuxbios than keeping up the milo waste of time.
> > I.e. do I have to flash the
> > LinuxBIOS image to a certain position and take care not to overwrite
> > other areas in flash?
> Basically the SROM scans for either the first or the second valid
> boot header. This is actually one of the slower parts of the boot.
So the way to go is to determine some free space on the flash and put
a header to it that is known to srom as a boot header?
I guess the problem with my former try was that I just overwrote the
beginning of the flash device, thus just erasing the SRM without
taking care of a second image on the chip.
> Ron or I should be able to give you a working image to start with.
Does this contain an SRM backup image I can go back to? Switching
between 2 LinuxBIOS versions should be sufficient if one is working.
If you could provide such an image this would be very useful.
Looks like it is time that I fix the VGA stuff - one of the DS10s is my
workstation here so having that working would be kind of nice.
Stefan
--
The x86 isn't all that complex - it just doesn't make a lot of
sense. -- Mike Johnson, Leader of 80x86 Design at AMD
Microprocessor Report (1994)
