Re: Question about MILO


Oleg Gusev (oleg@usm.uni-muenchen.de)
Wed, 07 Apr 1999 12:45:15 +0200


Hyung Min SEO wrote:
>
> The relocate move the other parts( osfpal, stub, milo ) to 0x20 0000 and
> the PC jumps to osfpal.
> The osfpal execute and the PC jumps to stub.
> The stub move the milo to 0xd0 0000 and the milo start to work.
> Is it right?

Yes. The annoying thing here is that if you put MILO at 12MB you need
16MB
to be able to run it. This is terrible waste of memory ( the guys with
4-8GB RAM don't shoot me, i'd like linux to run in 8MB )
 
>
> I don't know why the relocate move the other parts to 0x20 0000.

Because everybody is not supposed to use micros~1 boot loader.
You can use SRM,failsafe firmware update, failsafe SRM bootblock floppy,
Debug Monitor depending on the hardware you have. So it is necessary to
relocate
MILO to some place, acceptable by all, from its initial "boot" position.

> If the block of composed osfpal, stub, milo remains at same place(0x30
> 0b00) then there is any problem?

There are also HWRPB and MILO RPB position, page table position, etc.
Linux kernel most be put at 3MB, i.e. a lot of hardcoded things.
Carefully rewritten MILO would be more compact and position independent,
but nobody has done that so far...

 Oleg.
 

-- 
AlphaBIOS message: 
"You have selected to switch to a different operating system.
 This computer will no longer be able to run Windows NT."

-- 
To unsubscribe: send e-mail to axp-list-request@redhat.com with
'unsubscribe' as the subject.  Do not send it to axp-list@redhat.com



This archive was generated by hypermail 2.0b3 on Wed Apr 07 1999 - 04:00:06 PDT