Re: managing memory attributes in PEI

["Ard Biesheuvel" <ardb@kernel.org>]

On Tue, 23 May 2023 at 10:05, Gerd Hoffmann <kraxel@redhat.com> wrote:
>
>   Hi,
>
> > >> Before we enabled SMM for OVMF, we had never really used IA32X64 OVMF --
> > >> SMM-less ACPI S3 resume had just worked fine with X64-only OVMF. IA32X64
> > >> only proved a great platform option to fall back to, when we realized
> > >> that on X64 OVMF, ACPI S3 resume wouldn't just seamlessly extend to SMM.
> > >
> > > I don't quite understand. So, what's the conclusion of IA32X64 OVMF? Keep it? Remove it?
> >
> > As long as edk2 (core modules) will continue supporting IA32X64 firmware
> > platforms, I think keeping OVMF IA32X64 is useful, minimally as a test
> > bed for those core modules / PCDs / boot paths.
>
> Agree.
>
> I'll go switch downstream SMM-enabled builds from OvmfPkgIa32X64.dsc to
> OvmfPkgX64.dsc ASAP, so for virtualization use cases OvmfPkgIa32X64.dsc
> will not be needed any more.
>
> But it indeed makes sense to keep OvmfPkgIa32X64.dsc for testing and CI
> purposes.  So the question what the future of OvmfPkgIa32X64.dsc (and
> OvmfPkgIa32.dsc) should be is is more a question of what the overall
> edk2 plans for IA32 are.
>

Indeed. Let's make it very clear that it only remains as a validation
target, so keeping it alive only makes sense if other supported
IA32X64 based platforms still exist as well.