[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Debian Installer GRUB test image available

On 5/5/19 3:59 PM, Thomas Schmitt wrote:
> It's name on hard disk was "grub.chrp".
> (Its purpose and format is beyond my knowledge.)

It's a CHRP bootscript for PowerMacs.

> If a file CD1/System/Library/CoreServices/BootX had existed on hard disk,
> the pathspec "CD1" would have overwritten the previously inserted IsoNode
> object which held the instruction to copy from "grub.chrp".
> Afterwards, the content of file "BootX" would have been copied into
> the ISO.

But there is no file CD1/System/Library/CoreServices/BootX on the harddisk,
the whole folder structure below /System is solely created for GRUB and
only because I am mimicking what grub-mkrescue does:

> http://git.savannah.gnu.org/cgit/grub.git/tree/util/grub-mkrescue.c

> If not prohibited for other reasons, it would be clearer and safer to copy
> the desired files to their places underneath CD1 already on hard disk.
>   cp /home/glaubitz/tmp/sid/CD1/../CD1/boot/grub/powerpc-ieee1275/grub.chrp \
>      CD1/System/Library/CoreServices/BootX
>   cp /home/glaubitz/tmp/sid/CD1/../CD1/boot/grub/powerpc.elf \
>      CD1/System/Library/CoreServices/grub.elf

I don't see how this is better over using the pathspecs. The original powerpc.elf
is placed in the \\boot\grub subdirectory and with the pathspecs, I'm just linking
the powerpc.elf into the correct places for IBM CHRP machines and PowerMacs, just
like grub-mkrescue does it which produces a nice bootable image. I prefer using
pathspecs over copy commands as this is also what is done for the other architectures.

>   xorriso -as mkisofs ...no.-graft-points.and.no.long.pathspecs... CD1

Why no -graft-points and -long-pathspecs?


 .''`.  John Paul Adrian Glaubitz
: :' :  Debian Developer - glaubitz@debian.org
`. `'   Freie Universitaet Berlin - glaubitz@physik.fu-berlin.de
  `-    GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913

Reply to: