Hi Ben, Mike, Chris and all :)

Just saw this thread and had to reply on it allthough i must admit i've
been a bit too quiet on it ... :

On Sat, 13 Jan 2001, Ben Harris wrote:
> > I noticed that. I had a quick look at the bootloader and that would need 
> > some hacking if the kernel was rebased as it assumes that the kernel starts
> > at 0xf0000000. I cannot speak for the new bootloader written by Reinoud 
> > Zandijk but it may need similar tweakings.

Yep, just some simple tweakings... I must admit that my first try was to
get a good working system without optimising the memory map at first. I'll
do the modifications on my new bootloader as soon as possible to try to
squeeze some extra memory out of it.

As far as i know, the fixed address was put there for the Shark/EBSA
family. Just hadn't considered yet a change for the RiscPC. It has to be
tested on these machines too ... 

The problem i've seen mentioned in this thread is that some ubc chunks
can't be mapped due to their size ? Is that confirmed that it is really
the size ? Where do these critters get mapped anyway ... above the kernel
?

For the RiscPC I can't see any reason why the address can't be lowered
... say to 0xe0000000 .... Nothing there in the memory map anyway... and
the VIDC/MEMC/IOC can be reduced indeed... Will look at that too .

Its good to see that the SA1x00 port is making progress :) ...

Will roll out a new version of the updated bootloader soon ... I've also
got some clues as to why the debug-images were crashing ...

Cheers,
Reinoud