[PATCH V4] KVM, SEV: Add KVM_EXIT_SYSTEM_EVENT metadata for SEV-ES

[PATCH V4] KVM, SEV: Add KVM_EXIT_SYSTEM_EVENT metadata for SEV-ES

[Peter Gonda]

SEV-ES guests can request termination using the GHCB's MSR protocol. See
AMD's GHCB spec section '4.1.13 Termination Request'. Currently when a
guest does this the userspace VMM sees an KVM_EXIT_UNKNOWN (-EVINAL)
return code from KVM_RUN. By adding a KVM_EXIT_SYSTEM_EVENT to kvm_run
struct the userspace VMM can clearly see the guest has requested a SEV-ES
termination including the termination reason code set and reason code.

Signed-off-by: Peter Gonda <pgonda@google.com>
Suggested-by: Sean Christopherson <seanjc@google.com>
Cc: Vitaly Kuznetsov <vkuznets@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Cc: Brijesh Singh <brijesh.singh@amd.com>
Cc: Joerg Roedel <jroedel@suse.de>
Cc: Marc Orr <marcorr@google.com>
Cc: kvm@vger.kernel.org
Cc: linux-kernel@vger.kernel.org

---

V4
 * Switch to using KVM_SYSTEM_EVENT exit reason.

V3
 * Add Documentation/ update.
 * Updated other KVM_EXIT_SHUTDOWN exits to clear ndata and set reason
   to KVM_SHUTDOWN_REQ.

V2
 * Add KVM_CAP_EXIT_SHUTDOWN_REASON check for KVM_CHECK_EXTENSION.

Tested by making an SEV-ES guest call sev_es_terminate() with hardcoded
reason code set and reason code and then observing the codes from the
userspace VMM in the kvm_run.system_event fields.

---
 arch/x86/kvm/svm/sev.c   | 7 +++++--
 include/uapi/linux/kvm.h | 1 +
 2 files changed, 6 insertions(+), 2 deletions(-)

diff --git a/arch/x86/kvm/svm/sev.c b/arch/x86/kvm/svm/sev.c
index 75fa6dd268f0..039b241a9fb5 100644
--- a/arch/x86/kvm/svm/sev.c
+++ b/arch/x86/kvm/svm/sev.c
@@ -2735,8 +2735,11 @@ static int sev_handle_vmgexit_msr_protocol(struct vcpu_svm *svm)
 		pr_info("SEV-ES guest requested termination: %#llx:%#llx\n",
 			reason_set, reason_code);
 
-		ret = -EINVAL;
-		break;
+		vcpu->run->exit_reason = KVM_EXIT_SHUTDOWN;
+		vcpu->run->system_event.type = KVM_SYSTEM_EVENT_SEV_TERM;
+		vcpu->run->system_event.flags = control->ghcb_gpa;
+
+		return 0;
 	}
 	default:
 		/* Error, keep GHCB MSR value as-is */
diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h
index 8616af85dc5d..d9d24db12930 100644
--- a/include/uapi/linux/kvm.h
+++ b/include/uapi/linux/kvm.h
@@ -444,6 +444,7 @@ struct kvm_run {
 #define KVM_SYSTEM_EVENT_SHUTDOWN       1
 #define KVM_SYSTEM_EVENT_RESET          2
 #define KVM_SYSTEM_EVENT_CRASH          3
+#define KVM_SYSTEM_EVENT_SEV_TERM       4
 			__u32 type;
 			__u64 flags;
 		} system_event;
-- 
2.35.1.1094.g7c7d902a7c-goog

Re: [PATCH V4] KVM, SEV: Add KVM_EXIT_SYSTEM_EVENT metadata for SEV-ES

[Sean Christopherson]

On Tue, Apr 05, 2022, Peter Gonda wrote:
> SEV-ES guests can request termination using the GHCB's MSR protocol. See
> AMD's GHCB spec section '4.1.13 Termination Request'. Currently when a
> guest does this the userspace VMM sees an KVM_EXIT_UNKNOWN (-EVINAL)

s/EVINAL/EINVAL 

> return code from KVM_RUN. By adding a KVM_EXIT_SYSTEM_EVENT to kvm_run
> struct the userspace VMM can clearly see the guest has requested a SEV-ES
> termination including the termination reason code set and reason code.

Nit, phrase that last part as a command, nowhere in the changelog is it actually
stated that the patch converts to use KVM_EXIT_SYSTEM_EVENT.

And my personal preference is to lead with the "what", especially when there's
already a fair amount of assumed knowledge, e.g. someone that's familiar with
SEV-ES probably already knows the guest can request termination, or at least won't
be surprised by the news, whereas leading with the SEV-ES and GHCB info is just
going to add to the confusion of someone who's clueless about SEV-ES.

  If an SEV-ES guest requests termination, exit to userspace with
  KVM_EXIT_SYSTEM_EVENT and a dedicated SEV_TERM type instead of -EINVAL
  so that userspace can take appropriate action.

  See AMD's GHCB spec section '4.1.13 Termination Request' for more details.

> Signed-off-by: Peter Gonda <pgonda@google.com>
> Suggested-by: Sean Christopherson <seanjc@google.com>
> Cc: Vitaly Kuznetsov <vkuznets@redhat.com>
> Cc: Borislav Petkov <bp@alien8.de>
> Cc: Tom Lendacky <thomas.lendacky@amd.com>
> Cc: Brijesh Singh <brijesh.singh@amd.com>
> Cc: Joerg Roedel <jroedel@suse.de>
> Cc: Marc Orr <marcorr@google.com>
> Cc: kvm@vger.kernel.org
> Cc: linux-kernel@vger.kernel.org
> 
> ---
> 
> V4
>  * Switch to using KVM_SYSTEM_EVENT exit reason.
> 
> V3
>  * Add Documentation/ update.
>  * Updated other KVM_EXIT_SHUTDOWN exits to clear ndata and set reason
>    to KVM_SHUTDOWN_REQ.
> 
> V2
>  * Add KVM_CAP_EXIT_SHUTDOWN_REASON check for KVM_CHECK_EXTENSION.
> 
> Tested by making an SEV-ES guest call sev_es_terminate() with hardcoded
> reason code set and reason code and then observing the codes from the
> userspace VMM in the kvm_run.system_event fields.
> 
> ---
>  arch/x86/kvm/svm/sev.c   | 7 +++++--
>  include/uapi/linux/kvm.h | 1 +
>  2 files changed, 6 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/x86/kvm/svm/sev.c b/arch/x86/kvm/svm/sev.c
> index 75fa6dd268f0..039b241a9fb5 100644
> --- a/arch/x86/kvm/svm/sev.c
> +++ b/arch/x86/kvm/svm/sev.c
> @@ -2735,8 +2735,11 @@ static int sev_handle_vmgexit_msr_protocol(struct vcpu_svm *svm)
>  		pr_info("SEV-ES guest requested termination: %#llx:%#llx\n",
>  			reason_set, reason_code);
>  
> -		ret = -EINVAL;
> -		break;
> +		vcpu->run->exit_reason = KVM_EXIT_SHUTDOWN;

Wrong exit reason.

> +		vcpu->run->system_event.type = KVM_SYSTEM_EVENT_SEV_TERM;
> +		vcpu->run->system_event.flags = control->ghcb_gpa;
> +
> +		return 0;
>  	}
>  	default:
>  		/* Error, keep GHCB MSR value as-is */
> diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h
> index 8616af85dc5d..d9d24db12930 100644
> --- a/include/uapi/linux/kvm.h
> +++ b/include/uapi/linux/kvm.h
> @@ -444,6 +444,7 @@ struct kvm_run {
>  #define KVM_SYSTEM_EVENT_SHUTDOWN       1
>  #define KVM_SYSTEM_EVENT_RESET          2
>  #define KVM_SYSTEM_EVENT_CRASH          3
> +#define KVM_SYSTEM_EVENT_SEV_TERM       4
>  			__u32 type;
>  			__u64 flags;

@type isn't properly padded, so this needs to be changed when using flags.   And
we definitely want to grab more room than just a single u64.

Per Paolo and I's combined powers[*], use bit 31 of the type to enumerate that ndata
is valid, and then change the sub-struct to:

		struct {
#define KVM_SYSTEM_EVENT_SHUTDOWN       1
#define KVM_SYSTEM_EVENT_RESET          2
#define KVM_SYSTEM_EVENT_CRASH          3
#define KVM_SYSTEM_EVENT_SEV_TERM	4
#define KVM_SYSTEM_EVENT_NDATA_VALID	(1u << 31)
			__u32 type;
			__u32 ndata;
			__u64 data[16];
		} system_event;

[*] https://lore.kernel.org/all/e0285020-49d9-8168-be4d-90940a30a048@redhat.com


>  		} system_event;
> -- 
> 2.35.1.1094.g7c7d902a7c-goog
>