gbura...@gmail.com wrote: >> I asked you for at least a usecase when it makes a difference but till >> now I've read only about theoretical advantages. >> > > Ok, my usecase is the following: > We are installing grub to EFI system partition and on the next load it must > load itself (rmadisk&kernel) from Windows partiton (where our main product is > installed). It need however to find the NTFS partition with our product, the > only ways is GPT GUID. > > Out of Linux world FS UUID is nothing, our main product works on Windows, and > the only way to identify the partition correctly - GUID. UEFI spec also use > GPT GUIDs in it's "device path", same as Windows. > > grub2 is not for Linux load only. > > I thought about this and decided that it's ok to have this functionality as long as it's implemented cleanly. In particular: - It shouldn't be named "GPT GUID" but something like PARTITION GUID. If partition is GPT then GPT UUIDs are used, if it's msdos then nt-like UUIDs should be used and so on. You don't have to implement non-GPT but arcitecture must be clean - It must not give any size increase of basic modules - grub-probe must support it (trivial) - Not used by default > -- > This message was sent on behalf of gbura...@gmail.com at openSubscriber.com > http://www.opensubscriber.com/message/grub-devel@gnu.org/13458994.html > > > _______________________________________________ > Grub-devel mailing list > Grub-devel@gnu.org > http://lists.gnu.org/mailman/listinfo/grub-devel > >
-- Regards Vladimir 'φ-coder/phcoder' Serbinenko
signature.asc
Description: OpenPGP digital signature
_______________________________________________ Grub-devel mailing list Grub-devel@gnu.org http://lists.gnu.org/mailman/listinfo/grub-devel
