Re: problems with savedefault

[Jochen Hoenicke]

On Jan 31, Thierry Laronde wrote:
> On Wed, Jan 31, 2001 at 02:52:58PM +0100, Thierry Laronde wrote:
> > 
> >     Is grub able to write sectors ordinarily ? In at least one case, the
> >     answer is yes : with the install command. So this succeeds with
> >     install and fails with a function using the previous functions. So
> >     the question : what is the differences between the use of these
> >     functions in `savedefault' and `install' ?
> 
> There is in fact another difference between floppy and hard disk : in
> floppy, generaly speaking, there is no *stage1_5.
> 
> I have tried to suppress *stage1_5 for the grub on the MBR of an IDE drive :
> savedefault works...
> 
> So the problem is --- I guess... --- with the location of the first sector
> of stage2 when a *stage1_5 is used.
> 
> And I am afraid of one thing : if this location is wrong where by hell is
> writing `savedefault' ?

I looked over the code, and I'm quite certain, that the savedefault
mechanism can't work for stage1_5, since neither the saved_entry
variable, nor the install_second_sector is copied from stage1_5 to
stage2. 

What's worse, is that AFAICS save_default writes the saved_entry into
A RANDOM SECTOR, without any sanity checks if the sector could be
right.  This is because "install_second_sector", the sector where the
saved_entry is written to, is initialized from %ebp which is not set
in chain_stage2.

IMHO there should be a sanity check in savedefault_func, comparing
some magic bytes in the sector that it changes and giving an error
message if they are not present.

  Jochen