Re: [PATCH Part1 RFC v2 06/20] x86/sev: Define SNP guest request NAE events

From: Borislav Petkov
Date: Tue May 18 2021 - 06:45:42 EST


On Fri, Apr 30, 2021 at 07:16:02AM -0500, Brijesh Singh wrote:
> Version 2 of the GHCB specification added the support for SNP guest
> request NAE events. The SEV-SNP guests will use this event to request
> the attestation report. See the GHCB specification for more details.
>
> Signed-off-by: Brijesh Singh <brijesh.singh@xxxxxxx>
> ---
> arch/x86/include/uapi/asm/svm.h | 4 ++++
> 1 file changed, 4 insertions(+)
>
> diff --git a/arch/x86/include/uapi/asm/svm.h b/arch/x86/include/uapi/asm/svm.h
> index f7bf12cad58c..7a45aa284530 100644
> --- a/arch/x86/include/uapi/asm/svm.h
> +++ b/arch/x86/include/uapi/asm/svm.h
> @@ -109,6 +109,8 @@
> #define SVM_VMGEXIT_SET_AP_JUMP_TABLE 0
> #define SVM_VMGEXIT_GET_AP_JUMP_TABLE 1
> #define SVM_VMGEXIT_SNP_PAGE_STATE_CHANGE 0x80000010
> +#define SVM_VMGEXIT_SNP_GUEST_REQUEST 0x80000011
> +#define SVM_VMGEXIT_SNP_EXT_GUEST_REQUEST 0x80000012

Why does this need the "VMGEXIT" *and* "SNP" prefixes?

Why not simply:

SVM_VMGEXIT_GUEST_REQ
SVM_VMGEXIT_EXT_GUEST_REQ

like the rest of the VMGEXIT defines in there?


> #define SVM_VMGEXIT_HYPERVISOR_FEATURES 0x8000fffd
> #define SVM_VMGEXIT_UNSUPPORTED_EVENT 0x8000ffff
>
> @@ -218,6 +220,8 @@
> { SVM_VMGEXIT_AP_HLT_LOOP, "vmgexit_ap_hlt_loop" }, \
> { SVM_VMGEXIT_AP_JUMP_TABLE, "vmgexit_ap_jump_table" }, \
> { SVM_VMGEXIT_SNP_PAGE_STATE_CHANGE, "vmgexit_page_state_change" }, \
> + { SVM_VMGEXIT_SNP_GUEST_REQUEST, "vmgexit_snp_guest_request" }, \
> + { SVM_VMGEXIT_SNP_EXT_GUEST_REQUEST, "vmgexit_snp_extended_guest_request" }, \
> { SVM_VMGEXIT_HYPERVISOR_FEATURES, "vmgexit_hypervisor_feature" }, \
> { SVM_EXIT_ERR, "invalid_guest_state" }

Ditto.

--
Regards/Gruss,
Boris.

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