Re: [PATCH] x86/sev: Return an error on a returned non-zero SW_EXITINFO1[31:0]

[Tom Lendacky]

On 10/1/21 4:43 AM, Borislav Petkov wrote:
> On Thu, Sep 30, 2021 at 11:42:01PM -0500, Tom Lendacky wrote:
> > After returning from a VMGEXIT NAE event, SW_EXITINFO1[31:0] is checked
> > for a value of 1, which indicates an error and that SW_EXITINFO2 contains
> > exception information. However, future versions of the GHCB specification
> > may define new values for SW_EXITINFO1[31:0], so really any non-zero value
> > should be treated as an error.
>
> So I wanna do this ontop. Might wanna apply it and look at the result -
> it shows better what the changes are.

Yup, looks good to me.

> ---
> From: Borislav Petkov <bp@xxxxxxx>
> Date: Fri, 1 Oct 2021 11:41:05 +0200
> Subject: [PATCH] x86/sev: Carve out HV call return value verification
>
> Carve out the verification of the HV call return value into a separate
> helper and make it more readable.
>
> No it more readable.

I'm assuming you don't want this last sentence...

Thanks,
Tom

>