On Tue, Nov 09 1999, OKUJI Yoshinori wrote:
Hi,
I have managed to fix my problems. I was debugging stage1 with SoftIce
(it was easier than writing my own version of stage1) and I noticed
the 0xFF value at offset 0x40. And I thought - hey what if I got wrong
value there ? (This occured to me after seeing that DX had a value of
1 when booting from hard disk with SoftIce). So I wrote dumb "stage1"
which just told me whether the value in DX is >= 80 or not. And it
wasn't. So I simply put 0x80 at offset 0x40 and that fixed it (I know
that GRUB would have done the same had I used the 'd' flag during the
installation).
Thanks to everyone for your assistance. I hope I wasn't too annoying
(it's not my fault that HP & Intel can't make a decent BIOS).
PS. Maybe it's worth putting this into docs so that another user with
problematic BIOS won't run into the same problem.
Regards,
Goran
> From: Goran Koruga <[EMAIL PROTECTED]>
> Subject: Re: Troubles with grub on my SCSI drive
> Date: Mon, 8 Nov 1999 19:49:05 +0100
>
> > Hmm, I didn't notice anything like that in GRUB 0.5 sources - can you please
> > enlighten me ? I wanted to add more of my own debugging messages but I was
> > getting internal errors from gas (I simply copied the relevant lines from
> > other places in stage1.S).
>
> Sorry, I said a lie. The current stage1 is too tight to have any
> debugging feature, so it does not have such a message.
>
> However, you can omit the LBA support code temporarily so that you
> can add some debugging functions into the stage1. Can you write
> assembly? If you can, print out all the registers just before the INT
> 13 call and check if the information is correct. Otherwise, I'll write
> a special version of stage1 and send it to you.
>
> ----------------------------------------------------------------------
> OKUJI Yoshinori <[EMAIL PROTECTED]> ^o-o^
> http://duff.kuicr.kyoto-u.ac.jp/~okuji (in English) m /
--
Writing about music is like dancing about architecture.
-- Frank Zappa
