Re: [PATCH v5 3/6] x86/sev: Split off startup code from core code

[Borislav Petkov <bp@alien8.de>]

On April 23, 2025 6:50:00 PM GMT+03:00, Ard Biesheuvel <ardb@kernel.org> wrote:
>On Wed, 23 Apr 2025 at 17:22, Tom Lendacky <thomas.lendacky@amd.com> wrote:
>>
>> On 4/18/25 09:12, Ard Biesheuvel wrote:
>> > From: Ard Biesheuvel <ardb@kernel.org>
>> >
>> > Disentangle the SEV core code and the SEV code that is called during
>> > early boot. The latter piece will be moved into startup/ in a subsequent
>> > patch.
>> >
>> > Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
>>
>> This patch breaks SNP guests. The SNP guest boots, but no longer has
>> access to the VMPCK keys needed to communicate with the ASP, which is
>> used, for example, to obtain an attestation report.
>>
>> It looks like the secrets_pa is defined as static in both startup.c and
>> core.c. It is set by a function in startup.c and so when used in core.c
>> its value will be 0.
>>
>> The following fixed the issue for me. Let me know if it can be squashed
>> in or a full patch is needed. Although, it likely should be named
>> sev_secrets_pa since it is no longer static.
>>
>
>Thanks for the fix, and apologies for using you as a guinea pig - I've
>been struggling to get access to SEV-SNP capable hardware, although a
>suitable EPYC based machine should be arriving in a month or 2.
>
>I'd assume a proper patch is better, and renaming it to sev_secrets_pa
>doesn't seem that intrusive. But it is ultimately Ingo's call.

Those patches should be applied *after* they have been tested. mingo knows that but does it anyway. :-(

This rushed patch application and hurried testing and sticking fixes ontop is completely counterproductive and annoying.

-- 
Sent from a small device: formatting sucks and brevity is inevitable.