Messages in this thread | | | Date | Thu, 16 Feb 2023 13:19:22 +0100 | From | Juergen Gross <> | Subject | Re: [PATCH v2 2/8] x86/mtrr: support setting MTRR state for software defined MTRRs |
| |
On 16.02.23 12:25, Borislav Petkov wrote: > On Thu, Feb 16, 2023 at 10:32:28AM +0100, Juergen Gross wrote: >> Is that flag _really_ meant to indicate we are running as a SEV-SNP guest? > > Yes. > >> Given that the referenced patch is part of the SEV-SNP host support series, >> I'm inclined to suspect it won't be set for sure in HyperV SEV-SNP guests. > > It better be. If it is a modified guest - no matter how modified - it > should set that flag. The vTOM thing is still being discussed. > >> And who is setting it for KVM SEV-SNP guests? > > That same patch does.
Hmm, I must be blind. I can't spot it.
I'm seeing only the feature bit #define and a call of setup_clear_cpu_cap(X86_FEATURE_SEV_SNP) in this patch.
Or is it done by hardware or the hypervisor?
Juergen [unhandled content-type:application/pgp-keys][unhandled content-type:application/pgp-signature] | |