[PATCH v7 1/3] efi_var_file: refactor to move buffer functions
Ilias Apalodimas
ilias.apalodimas at linaro.org
Wed Feb 11 16:16:34 CET 2026
On Wed, 11 Feb 2026 at 17:13, Tom Rini <trini at konsulko.com> wrote:
>
> On Wed, Feb 11, 2026 at 05:04:00PM +0200, Ilias Apalodimas wrote:
> > So this fails for a weird reason
> >
> > Some boards don't have FAT_WRTE. It seems that the Kconfig order
> > matters and those boards pick the new option if SPI is enabled.
> >
> > This fixes it
> >
> > diff --git a/lib/efi_loader/Kconfig b/lib/efi_loader/Kconfig
> > index c13cb8952caa..0f6f927ddaf1 100644
> > --- a/lib/efi_loader/Kconfig
> > +++ b/lib/efi_loader/Kconfig
> > @@ -112,7 +112,8 @@ menu "UEFI Variables"
> >
> > choice
> > prompt "Store for non-volatile UEFI variables"
> > - default EFI_VARIABLE_FILE_STORE
> > + default EFI_VARIABLE_FILE_STORE if FAT_WRITE
> > + default EFI_VARIABLE_NO_STORE
> > help
> > Select where non-volatile UEFI variables shall be stored.
>
> The behavior of a choice statement is such that if there's no default
> (and in this case, platforms don't set FAT_WRITE, so there's effectively
> no default) they select the first presented valid option. Before this
> patch, that ends up being EFI_VARIABLE_NO_STORE but after this patch it
> ends up being the SPI option. So making it clear that the fallback is
> NO_STORE is a correct general clarification / bugfix. It can be its own
> patch even :)\
Yep, we already discussed this with Michal over IRC. He'll send it as
a separate patch with your Suggested-by etc
Thanks
/Ilias
>
> --
> Tom
More information about the U-Boot
mailing list