Re: [PATCH] x86/mm: Print the encryption features correctly when a paravisor is present

[Borislav Petkov <bp@alien8.de>]

On Fri, Oct 20, 2023 at 08:00:13PM +0000, Dexuan Cui wrote:
> Currently arch/x86/mm/mem_encrypt.c: print_mem_encrypt_feature_info()
> prints an incorrect and confusing message 
> "Memory Encryption Features active: AMD SEV".
> when an Intel TDX VM with a paravisor runs on Hyper-V.
> 
> So I think a kernel patch is needed.

So I'm trying to parse this:

"Hyper-V provides two modes for running a TDX/SNP VM:

1) In TD Partitioning mode (TDX) or vTOM mode (SNP) with a paravisor;
2) In "fully enlightened" mode with the normal TDX shared bit or SNP C-bit
   control over page encryption, and no paravisor."

and it all sounds like word salad to me.

The fact that you've managed to advertize a salad of CPUID bits to the
guest to lead to such confusing statement, sounds like a major insanity.

> the native TDX/SNP CPUID capability is hidden from the VM

Why do you wonder then that it detects wrong?! You're hiding it!

> but cc_platform_has(CC_ATTR_MEM_ENCRYPT) and
> cc_platform_has(CC_ATTR_GUEST_MEM_ENCRYPT) are true;

I guess you need to go to talk to Michael:

812b0597fb40 ("x86/hyperv: Change vTOM handling to use standard coco mechanisms")

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette