Re: GRUB unexpected trap in Itanium

[Frank Scheiner]

Dear Vladimir,

On 11.08.23 03:01, Vladimir 'phcoder' Serbinenko wrote:
> Le ven. 11 août 2023, 02:17, Pedro Miguel Justo <pmsjt@texair.net 
> <mailto:pmsjt@texair.net>> a écrit :
>
>     I have bisected the issue and it resulted into the following change:
>
>     ```
>     06edd40db76bb78457ac26156ed5f7b62381bbe8 is the first bad commit
>     commit 06edd40db76bb78457ac26156ed5f7b62381bbe8
>     Author: Oliver Steffen <osteffen@redhat.com
>     <mailto:osteffen@redhat.com>>
>     Date:   Fri May 26 13:35:43 2023 +0200
>
>          guid: Unify GUID types
>
>          There are 3 implementations of a GUID in GRUB. Replace them with
>          a common one, placed in types.h.
>
>          It uses the "packed" flavor of the GUID structs, the alignment
>     attribute
>          is dropped, since it is not required.
>
>          Signed-off-by: Oliver Steffen <osteffen@redhat.com
>     <mailto:osteffen@redhat.com>>
>          Reviewed-by: Daniel Kiper <daniel.kiper@oracle.com
>     <mailto:daniel.kiper@oracle.com>>
>     ```
>
>     The issue is most likely related with an unaligned memory access
>     exception resulting from may global GUID variables being changed
>     from `grub_efi_guid_t` (which are decorated with `__attribute__
>     ((aligned(8)))`) to the new and unified `grub_guid_t` type (now
>     decorated with `__attribute__ ((packed))`).
>
>     Do you concur?
>
> That makes perfect sense. Thank you for the analysis. When passing GUID 
> to EFI we need it to be aligned. I propose to partially revert the 
> commit and keep efi_guid either completely separate or to include common 
> guid type as the only structure in it.

Linux actually uses a minimum alignment of 32 bits, see for example [1].

[1]: 
https://github.com/torvalds/linux/blob/master/include/linux/efi.h#L60-L78

Would that make sense for GRUB, too?

Cheers,
Frank