Re: Native CD test results

["Vesa Jääskeläinen" <chaac@nic.fi>]

Pavel Roskin wrote:
> And while "looking deeper", it turned out that there is a scary timebomb
> in kern/i386/pc/startup.S.  Only the initial part of that file is
> supposed to be uncompressed, but in fact, the error handlers of the
> decompression function itself were spilling into the compressed area.
> Adding even minimal debugging code would push the main part of
> lzo1x_decompress() beyond the boundary defined by
> GRUB_KERNEL_MACHINE_RAW_SIZE, which would crash even before "Welcome to
> GRUB".
> 
> So I've applied another patch that assures that to uncompressed code
> stays below GRUB_KERNEL_MACHINE_RAW_SIZE.  It would be nice to avoid any
> constant here and make grub-mkimage.c use a label in the image, but it
> would need more work and more testing.

Yeah... I noticed that too when I was playing with IDT's some time ago 
and Bean found out the reason. I think we need to update this so that it 
will never happen. Perhaps divide code to sections or so? We could 
provide this in LD variable... hmm... I could investigate what would be 
required for it... Good place to learn more about LD scripts.