Skip to contentSkip to conversation

KVM: x86: Advertise SUCCOR and OVERFLOW_RECOV cpuid bits

PATCH Superseded John Allen 2 participants 5 messages Jul 30, 2024

Subsystems (3)

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore

John Allen

author
#0 lore raw

PATCH 1/1 KVM: x86: Advertise SUCCOR and OVERFLOW_RECOV cpuid bits

Handling deferred, uncorrected MCEs on AMD guests is now possible with additional support in qemu. Ensure that the SUCCOR and OVERFLOW_RECOV bits are advertised to the guest in KVM.

---
 arch/x86/kvm/cpuid.c   | 2 +-
 arch/x86/kvm/svm/svm.c | 7 +++++++
 2 files changed, 8 insertions(+), 1 deletion(-)
arch/x86/kvm/cpuid.c +1 -1
index 2617be544480..4745098416c3 100644
@@ -1241,7 +1241,7 @@ static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function)
 
 		/* mask against host */
 		entry->edx &= boot_cpu_data.x86_power;
-		entry->eax = entry->ebx = entry->ecx = 0;
+		entry->eax = entry->ecx = 0;
 		break;
 	case 0x80000008: {
 		/*
arch/x86/kvm/svm/svm.c +7 -0
index c115d26844f7..a6820b0915db 100644
@@ -5199,6 +5199,13 @@ static __init void svm_set_cpu_caps(void)
 		kvm_cpu_cap_set(X86_FEATURE_SVME_ADDR_CHK);
 	}
 
+	/* CPUID 0x80000007 */
+	if (boot_cpu_has(X86_FEATURE_SUCCOR))
+		kvm_cpu_cap_set(X86_FEATURE_SUCCOR);
+
+	if (boot_cpu_has(X86_FEATURE_OVERFLOW_RECOV))
+		kvm_cpu_cap_set(X86_FEATURE_OVERFLOW_RECOV);
+
 	/* CPUID 0x80000008 */
 	if (boot_cpu_has(X86_FEATURE_LS_CFG_SSBD) ||
 	    boot_cpu_has(X86_FEATURE_AMD_SSBD))
Signature · 1 line
2.34.1
#1 lore raw
Quoted 16 lines · John Allen

Handling deferred, uncorrected MCEs on AMD guests is now possible with additional support in qemu. Ensure that the SUCCOR and OVERFLOW_RECOV bits are advertised to the guest in KVM.

---
 arch/x86/kvm/cpuid.c   | 2 +-
 arch/x86/kvm/svm/svm.c | 7 +++++++
 2 files changed, 8 insertions(+), 1 deletion(-)
arch/x86/kvm/cpuid.c +0 -0
index 2617be544480..4745098416c3 100644

arch/x86/kvm/cpuid.c static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function) · line 1244

Quoted 5 lines
arch/x86/kvm/cpuid.c +1 -1
@@ -1241,7 +1241,7 @@ static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function)
 
 		/* mask against host */
 		entry->edx &= boot_cpu_data.x86_power;
-		entry->eax = entry->ebx = entry->ecx = 0;
+		entry->eax = entry->ecx = 0;

Needs an override to prevent reporting all of EBX to userspace.

		cpuid_entry_override(entry, CPUID_8000_0007_EBX);
Quoted 8 lines
arch/x86/kvm/cpuid.c +0 -0
 		break;
 	case 0x80000008: {
 		/*
arch/x86/kvm/svm/svm.c +0 -0
index c115d26844f7..a6820b0915db 100644

arch/x86/kvm/svm/svm.c static __init void svm_set_cpu_caps(void) · line 5207

Quoted 9 lines
arch/x86/kvm/svm/svm.c +6 -0
@@ -5199,6 +5199,13 @@ static __init void svm_set_cpu_caps(void)
 		kvm_cpu_cap_set(X86_FEATURE_SVME_ADDR_CHK);
 	}
 
+	/* CPUID 0x80000007 */
+	if (boot_cpu_has(X86_FEATURE_SUCCOR))
+		kvm_cpu_cap_set(X86_FEATURE_SUCCOR);
+
+	if (boot_cpu_has(X86_FEATURE_OVERFLOW_RECOV))
+		kvm_cpu_cap_set(X86_FEATURE_OVERFLOW_RECOV);

This _could_ use kvm_cpu_cap_check_and_set(), but given that this an AMD specific leaf and unlikely to ever be used by Intel, I'm inclined to handle this in cpuid.c, with an opporunustic "conversion" to one feature per line[*]:

arch/x86/kvm/cpuid.c +5 -0
index 2617be544480..ea11a7e45174 100644
@@ -743,6 +743,11 @@ void kvm_set_cpu_caps(void)
        if (!tdp_enabled && IS_ENABLED(CONFIG_X86_64))
                kvm_cpu_cap_set(X86_FEATURE_GBPAGES);
 
+       kvm_cpu_cap_mask(CPUID_8000_0007_EBX,
+               F(OVERFLOW_RECOV) |
+               F(SUCCOR)
+       );
+
        kvm_cpu_cap_init_kvm_defined(CPUID_8000_0007_EDX,
                SF(CONSTANT_TSC)
        );

[*] https://lore.kernel.org/all/[email protected]

author
#2 lore raw
Quoted 18 lines · Sean Christopherson
Quoted 17 lines · John Allen

Handling deferred, uncorrected MCEs on AMD guests is now possible with additional support in qemu. Ensure that the SUCCOR and OVERFLOW_RECOV bits are advertised to the guest in KVM.

---
 arch/x86/kvm/cpuid.c   | 2 +-
 arch/x86/kvm/svm/svm.c | 7 +++++++
 2 files changed, 8 insertions(+), 1 deletion(-)
arch/x86/kvm/cpuid.c +0 -0
index 2617be544480..4745098416c3 100644
@@ -1241,7 +1241,7 @@ static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function)
 
Quoted 8 lines

arch/x86/kvm/cpuid.c static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function) · line 1244

Quoted 4 lines
arch/x86/kvm/cpuid.c +1 -1
 		/* mask against host */
 		entry->edx &= boot_cpu_data.x86_power;
-		entry->eax = entry->ebx = entry->ecx = 0;
+		entry->eax = entry->ecx = 0;

Needs an override to prevent reporting all of EBX to userspace.

		cpuid_entry_override(entry, CPUID_8000_0007_EBX);

Right, I see what you mean. We just want to expose these specific bits and not all of EBX. I think with the patch as it is along with the change you suggest below, this should resolve this as the above case already has the cpuid_entry_override just above where it cuts off. Or is there another place we need it?

Thanks,
John

Quoted 42 lines
Quoted 9 lines
arch/x86/kvm/cpuid.c +0 -0
 		break;
 	case 0x80000008: {
 		/*
arch/x86/kvm/svm/svm.c +0 -0
index c115d26844f7..a6820b0915db 100644

arch/x86/kvm/svm/svm.c static __init void svm_set_cpu_caps(void) · line 5207

Quoted 9 lines
arch/x86/kvm/svm/svm.c +6 -0
@@ -5199,6 +5199,13 @@ static __init void svm_set_cpu_caps(void)
 		kvm_cpu_cap_set(X86_FEATURE_SVME_ADDR_CHK);
 	}
 
+	/* CPUID 0x80000007 */
+	if (boot_cpu_has(X86_FEATURE_SUCCOR))
+		kvm_cpu_cap_set(X86_FEATURE_SUCCOR);
+
+	if (boot_cpu_has(X86_FEATURE_OVERFLOW_RECOV))
+		kvm_cpu_cap_set(X86_FEATURE_OVERFLOW_RECOV);

This _could_ use kvm_cpu_cap_check_and_set(), but given that this an AMD specific leaf and unlikely to ever be used by Intel, I'm inclined to handle this in cpuid.c, with an opporunustic "conversion" to one feature per line[*]:

arch/x86/kvm/cpuid.c +5 -0
index 2617be544480..ea11a7e45174 100644
@@ -743,6 +743,11 @@ void kvm_set_cpu_caps(void)
        if (!tdp_enabled && IS_ENABLED(CONFIG_X86_64))
                kvm_cpu_cap_set(X86_FEATURE_GBPAGES);
 
+       kvm_cpu_cap_mask(CPUID_8000_0007_EBX,
+               F(OVERFLOW_RECOV) |
+               F(SUCCOR)
+       );
+
        kvm_cpu_cap_init_kvm_defined(CPUID_8000_0007_EDX,
                SF(CONSTANT_TSC)
        );

[*] https://lore.kernel.org/all/[email protected]

#3 lore raw
Quoted 26 lines · John Allen
Quoted 25 lines · Sean Christopherson
Quoted 16 lines · John Allen

Handling deferred, uncorrected MCEs on AMD guests is now possible with additional support in qemu. Ensure that the SUCCOR and OVERFLOW_RECOV bits are advertised to the guest in KVM.

---
 arch/x86/kvm/cpuid.c   | 2 +-
 arch/x86/kvm/svm/svm.c | 7 +++++++
 2 files changed, 8 insertions(+), 1 deletion(-)
arch/x86/kvm/cpuid.c +0 -0
index 2617be544480..4745098416c3 100644

arch/x86/kvm/cpuid.c static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function) · line 1244

Quoted 5 lines
arch/x86/kvm/cpuid.c +1 -1
@@ -1241,7 +1241,7 @@ static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function)
 
 		/* mask against host */
 		entry->edx &= boot_cpu_data.x86_power;
-		entry->eax = entry->ebx = entry->ecx = 0;
+		entry->eax = entry->ecx = 0;

Needs an override to prevent reporting all of EBX to userspace.

Quoted 6 lines
Quoted 1 line
		cpuid_entry_override(entry, CPUID_8000_0007_EBX);

Right, I see what you mean. We just want to expose these specific bits and not all of EBX. I think with the patch as it is along with the change you suggest below, this should resolve this as the above case already has the cpuid_entry_override just above where it cuts off.

Heh, nope, it doesn't. The existing override is for EDX, this needs one for EBX.

author
#4 lore raw
Quoted 27 lines · Sean Christopherson
Quoted 26 lines · John Allen
Quoted 25 lines · Sean Christopherson
Quoted 16 lines · John Allen

Handling deferred, uncorrected MCEs on AMD guests is now possible with additional support in qemu. Ensure that the SUCCOR and OVERFLOW_RECOV bits are advertised to the guest in KVM.

---
 arch/x86/kvm/cpuid.c   | 2 +-
 arch/x86/kvm/svm/svm.c | 7 +++++++
 2 files changed, 8 insertions(+), 1 deletion(-)
arch/x86/kvm/cpuid.c +0 -0
index 2617be544480..4745098416c3 100644

arch/x86/kvm/cpuid.c static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function) · line 1244

Quoted 5 lines
arch/x86/kvm/cpuid.c +1 -1
@@ -1241,7 +1241,7 @@ static inline int __do_cpuid_func(struct kvm_cpuid_array *array, u32 function)
 
 		/* mask against host */
 		entry->edx &= boot_cpu_data.x86_power;
-		entry->eax = entry->ebx = entry->ecx = 0;
+		entry->eax = entry->ecx = 0;

Needs an override to prevent reporting all of EBX to userspace.

Quoted 8 lines
Quoted 6 lines
Quoted 1 line
		cpuid_entry_override(entry, CPUID_8000_0007_EBX);

Right, I see what you mean. We just want to expose these specific bits and not all of EBX. I think with the patch as it is along with the change you suggest below, this should resolve this as the above case already has the cpuid_entry_override just above where it cuts off.

Heh, nope, it doesn't. The existing override is for EDX, this needs one for EBX.

Ah, yes you're right. Sorry for the noise!