Re: MILO size problems
On 3 Mar 2000, David Huggins-Daines wrote:
> Rich Payne <email@example.com> writes:
> > Don't forget about the UX boards with ARCSBios, MILO will still be
> Doh! forgot about those, thanks.
> > necessary for them. Rumor has it that Q has SRM console for the XL series
> > (as all the testing was done on OpenVMS) the problem is getting them to
> > release it. Also at this point the UP1000 (Nautilus) is dependent on APB,
> > though that will be changing.
> Yes, I've added a note about the UP1000 to the documentation.
> However, it would be nicer if we could provide APB images and add APB
> support to dbootstrap.
> Can I download APB images from somewhere, and is it possible (read:
> legal) for us to distribute them?
I'll send you the latest copy, yes can redistribute them (do you need the
palcode file as well?). They are usually available from the API website
(www.alpha-processor.com) however I happen to know the copy up there is
slightly older and the apb contined in there has a few known issues.
> Also, in the event that the UP1000
> does not require APB in the future, is there any way to tell whether a
> UP1000 has been booted from APB or something else? For MILO we just
> look in /proc/cpuinfo (i.e. in the HWRPB :)
Well, at the moment you can tell it booted from APB because the CPU serial
number will be set to MILO-0000, systems booted from SRM display nothing.
Is that enough?
> Actually I rather like the idea of APB - separating the PALCode from
> the bootloader is a good idea...
Yes, though with AlphaBIOS having no future......
> > DP264 and Nautilus didn't compile with 2.2.14?
> Yeah, at least Nautilus didn't when I tried. Blame Henderson :)
2.2.14 should compile w/o any trouble on Nautilus, that's what I have
running here in the lab.