Skip to contentSkip to conversation

KVM: SVM: Add a helper to get LBR field pointer to dedup MSR accesses

PATCH Applied Sean Christopherson 2 participants 8 messages Mar 10 – Apr 3

Subsystems (3)
lore
Share & tools
lore

Sean Christopherson

author maintainer
#0 lore raw

PATCH 1/1 KVM: SVM: Add a helper to get LBR field pointer to dedup MSR accesses

Add a helper to get a pointer to the corresponding VMCB field given an LBR MSR index, and use it to dedup the handling in svm_{g,s}et_msr().

No functional change intended.

---
 arch/x86/kvm/svm/svm.c | 48 +++++++++++++++++-------------------------
 1 file changed, 19 insertions(+), 29 deletions(-)
arch/x86/kvm/svm/svm.c +19 -29
index 3407deac90bd..88999b6fc8a3 100644
@@ -2750,6 +2750,23 @@ static int svm_get_feature_msr(u32 msr, u64 *data)
 	return 0;
 }
 
+static __always_inline u64 *svm_vmcb_lbr(struct vcpu_svm *svm, u32 msr)
+{
+	switch (msr) {
+	case MSR_IA32_LASTBRANCHFROMIP:
+		return &svm->vmcb->save.br_from;
+	case MSR_IA32_LASTBRANCHTOIP:
+		return &svm->vmcb->save.br_to;
+	case MSR_IA32_LASTINTFROMIP:
+		return &svm->vmcb->save.last_excp_from;
+	case MSR_IA32_LASTINTTOIP:
+		return &svm->vmcb->save.last_excp_to;
+	default:
+		break;
+	}
+	BUILD_BUG();
+}
+
 static bool sev_es_prevent_msr_access(struct kvm_vcpu *vcpu,
 				      struct msr_data *msr_info)
 {
@@ -2826,16 +2843,10 @@ static int svm_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 		msr_info->data = lbrv ? svm->vmcb->save.dbgctl : 0;
 		break;
 	case MSR_IA32_LASTBRANCHFROMIP:
-		msr_info->data = lbrv ? svm->vmcb->save.br_from : 0;
-		break;
 	case MSR_IA32_LASTBRANCHTOIP:
-		msr_info->data = lbrv ? svm->vmcb->save.br_to : 0;
-		break;
 	case MSR_IA32_LASTINTFROMIP:
-		msr_info->data = lbrv ? svm->vmcb->save.last_excp_from : 0;
-		break;
 	case MSR_IA32_LASTINTTOIP:
-		msr_info->data = lbrv ? svm->vmcb->save.last_excp_to : 0;
+		msr_info->data = lbrv ? *svm_vmcb_lbr(svm, msr_info->index) : 0;
 		break;
 	case MSR_VM_HSAVE_PA:
 		msr_info->data = svm->nested.hsave_msr;
@@ -3111,35 +3122,14 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr)
 		svm_update_lbrv(vcpu);
 		break;
 	case MSR_IA32_LASTBRANCHFROMIP:
-		if (!lbrv)
-			return KVM_MSR_RET_UNSUPPORTED;
-		if (!msr->host_initiated)
-			return 1;
-		svm->vmcb->save.br_from = data;
-		vmcb_mark_dirty(svm->vmcb, VMCB_LBR);
-		break;
 	case MSR_IA32_LASTBRANCHTOIP:
-		if (!lbrv)
-			return KVM_MSR_RET_UNSUPPORTED;
-		if (!msr->host_initiated)
-			return 1;
-		svm->vmcb->save.br_to = data;
-		vmcb_mark_dirty(svm->vmcb, VMCB_LBR);
-		break;
 	case MSR_IA32_LASTINTFROMIP:
-		if (!lbrv)
-			return KVM_MSR_RET_UNSUPPORTED;
-		if (!msr->host_initiated)
-			return 1;
-		svm->vmcb->save.last_excp_from = data;
-		vmcb_mark_dirty(svm->vmcb, VMCB_LBR);
-		break;
 	case MSR_IA32_LASTINTTOIP:
 		if (!lbrv)
 			return KVM_MSR_RET_UNSUPPORTED;
 		if (!msr->host_initiated)
 			return 1;
-		svm->vmcb->save.last_excp_to = data;
+		*svm_vmcb_lbr(svm, ecx) = data;
 		vmcb_mark_dirty(svm->vmcb, VMCB_LBR);
 		break;
 	case MSR_VM_HSAVE_PA:

base-commit: 5128b972fb2801ad9aca54d990a75611ab5283a9

Signature · 1 line
2.53.0.473.g4a7958ca14-goog
#2 lore raw
In reply to Sean Christopherson
Quoted 26 lines · Sean Christopherson

Add a helper to get a pointer to the corresponding VMCB field given an LBR MSR index, and use it to dedup the handling in svm_{g,s}et_msr().

No functional change intended.

---
 arch/x86/kvm/svm/svm.c | 48 +++++++++++++++++-------------------------
 1 file changed, 19 insertions(+), 29 deletions(-)
arch/x86/kvm/svm/svm.c +6 -0
index 3407deac90bd..88999b6fc8a3 100644
@@ -2750,6 +2750,23 @@ static int svm_get_feature_msr(u32 msr, u64 *data)
        return 0;
 }
 
+static __always_inline u64 *svm_vmcb_lbr(struct vcpu_svm *svm, u32 msr)
+{
+       switch (msr) {
+       case MSR_IA32_LASTBRANCHFROMIP:
+               return &svm->vmcb->save.br_from;
+       case MSR_IA32_LASTBRANCHTOIP:

arch/x86/kvm/svm/svm.c static int svm_get_feature_msr(u32 msr, u64 *data) · line 2768

Quoted 10 lines
arch/x86/kvm/svm/svm.c +10 -0
+               return &svm->vmcb->save.br_to;
+       case MSR_IA32_LASTINTFROMIP:
+               return &svm->vmcb->save.last_excp_from;
+       case MSR_IA32_LASTINTTOIP:
+               return &svm->vmcb->save.last_excp_to;
+       default:
+               break;
+       }
+       BUILD_BUG();
+}

This actually fires for me with clang (from both callsites):

/usr/local/google/home/yosryahmed/dev/kernel.org/nsvm_fixups/arch/x86/kvm/svm/svm.c:2766:2:
error: call to '__compiletime_assert_1058' declared with 'error'
attribute: BUILD_BUG failed

 2766 |         BUILD_BUG();
      |         ^

/usr/local/google/home/yosryahmed/dev/kernel.org/nsvm_fixups/include/linux/build_bug.h:59:21:
note: expanded from macro 'BUILD_BUG'
59 | #define BUILD_BUG() BUILD_BUG_ON_MSG(1, "BUILD_BUG failed")
| ^
/usr/local/google/home/yosryahmed/dev/kernel.org/nsvm_fixups/include/linux/build_bug.h:39:37:
note: expanded from macro 'BUILD_BUG_ON_MSG'
39 | #define BUILD_BUG_ON_MSG(cond, msg) compiletime_assert(!(cond), msg)
| ^
/usr/local/google/home/yosryahmed/dev/kernel.org/nsvm_fixups/include/linux/compiler_types.h:706:2:
note: expanded from macro 'compiletime_assert'
706 | _compiletime_assert(condition, msg,
__compiletime_assert_, __COUNTER__)
| ^
/usr/local/google/home/yosryahmed/dev/kernel.org/nsvm_fixups/include/linux/compiler_types.h:694:2:
note: expanded from macro '_compiletime_assert'

  694 |         __compiletime_assert(condition, msg, prefix, suffix)
      |         ^
/usr/local/google/home/yosryahmed/dev/kernel.org/nsvm_fixups/include/linux/compiler_types.h:687:4:
note: expanded from macro '__compiletime_assert'
  687 |                         prefix ## suffix();
         \
      |                         ^
<scratch space>:102:1: note: expanded from here
  102 | __compiletime_assert_1058
      | ^
author maintainer
#3 lore raw
Quoted 60 lines · Yosry Ahmed
Quoted 26 lines · Sean Christopherson

Add a helper to get a pointer to the corresponding VMCB field given an LBR MSR index, and use it to dedup the handling in svm_{g,s}et_msr().

No functional change intended.

---
 arch/x86/kvm/svm/svm.c | 48 +++++++++++++++++-------------------------
 1 file changed, 19 insertions(+), 29 deletions(-)
arch/x86/kvm/svm/svm.c +6 -0
index 3407deac90bd..88999b6fc8a3 100644
@@ -2750,6 +2750,23 @@ static int svm_get_feature_msr(u32 msr, u64 *data)
        return 0;
 }
 
+static __always_inline u64 *svm_vmcb_lbr(struct vcpu_svm *svm, u32 msr)
+{
+       switch (msr) {
+       case MSR_IA32_LASTBRANCHFROMIP:
+               return &svm->vmcb->save.br_from;
+       case MSR_IA32_LASTBRANCHTOIP:

arch/x86/kvm/svm/svm.c static int svm_get_feature_msr(u32 msr, u64 *data) · line 2768

Quoted 10 lines
arch/x86/kvm/svm/svm.c +10 -0
+               return &svm->vmcb->save.br_to;
+       case MSR_IA32_LASTINTFROMIP:
+               return &svm->vmcb->save.last_excp_from;
+       case MSR_IA32_LASTINTTOIP:
+               return &svm->vmcb->save.last_excp_to;
+       default:
+               break;
+       }
+       BUILD_BUG();
+}

This actually fires for me with clang (from both callsites):

/usr/local/google/home/yosryahmed/dev/kernel.org/nsvm_fixups/arch/x86/kvm/svm/svm.c:2766:2:
error: call to '__compiletime_assert_1058' declared with 'error'
attribute: BUILD_BUG failed

 2766 |         BUILD_BUG();
      |         ^

/usr/local/google/home/yosryahmed/dev/kernel.org/nsvm_fixups/include/linux/build_bug.h:59:21:
note: expanded from macro 'BUILD_BUG'
59 | #define BUILD_BUG() BUILD_BUG_ON_MSG(1, "BUILD_BUG failed")
| ^
/usr/local/google/home/yosryahmed/dev/kernel.org/nsvm_fixups/include/linux/build_bug.h:39:37:
note: expanded from macro 'BUILD_BUG_ON_MSG'
39 | #define BUILD_BUG_ON_MSG(cond, msg) compiletime_assert(!(cond), msg)
| ^
/usr/local/google/home/yosryahmed/dev/kernel.org/nsvm_fixups/include/linux/compiler_types.h:706:2:
note: expanded from macro 'compiletime_assert'
706 | _compiletime_assert(condition, msg,
__compiletime_assert_, __COUNTER__)
| ^
/usr/local/google/home/yosryahmed/dev/kernel.org/nsvm_fixups/include/linux/compiler_types.h:694:2:
note: expanded from macro '_compiletime_assert'

Quoted 10 lines
  694 |         __compiletime_assert(condition, msg, prefix, suffix)
      |         ^
/usr/local/google/home/yosryahmed/dev/kernel.org/nsvm_fixups/include/linux/compiler_types.h:687:4:
note: expanded from macro '__compiletime_assert'
  687 |                         prefix ## suffix();
         \
      |                         ^
<scratch space>:102:1: note: expanded from here
  102 | __compiletime_assert_1058
      | ^

Hrm, I was assuming the compiler would be smart to understand the other cases are unreachable. What clang version, and are you doing anything special with your config? Even with KASAN and other sanitizers enabled, clang-19 is a-ok with the code. And I haven't gotten yelled at by build bots (yet).

#4 lore raw
Quoted 3 lines

Hrm, I was assuming the compiler would be smart to understand the other cases are unreachable. What clang version, and are you doing anything special with your config?

Debian clang version 18.1.8 (20+build1)

Nothing special with my config AFAICT.

author maintainer
#5 lore raw
Quoted 7 lines · Yosry Ahmed
Quoted 3 lines

Hrm, I was assuming the compiler would be smart to understand the other cases are unreachable. What clang version, and are you doing anything special with your config?

Debian clang version 18.1.8 (20+build1)

Nothing special with my config AFAICT.

Lame. Given that it's Friday and I'm about to become a pumpkin, and it doesn't really matter if it's compile-timer assert or not, I'm going to sqaush it to this:

static u64 *svm_vmcb_lbr(struct vcpu_svm *svm, u32 msr)

{
	switch (msr) {
	case MSR_IA32_LASTBRANCHFROMIP:
		return &svm->vmcb->save.br_from;
	case MSR_IA32_LASTBRANCHTOIP:
		return &svm->vmcb->save.br_to;
	case MSR_IA32_LASTINTFROMIP:
		return &svm->vmcb->save.last_excp_from;
	case MSR_IA32_LASTINTTOIP:
		return &svm->vmcb->save.last_excp_to;
	default:
		break;
	}
	KVM_BUG_ON(1, svm->vcpu.kvm);
	return &svm->vmcb->save.br_from;
}

Can you double check that kvm-x86/nested builds for you? In the process of rebuilding kvm-x86/next...

#6 lore raw
Quoted 31 lines · Sean Christopherson
Quoted 7 lines · Yosry Ahmed
Quoted 3 lines

Hrm, I was assuming the compiler would be smart to understand the other cases are unreachable. What clang version, and are you doing anything special with your config?

Debian clang version 18.1.8 (20+build1)

Nothing special with my config AFAICT.

Lame. Given that it's Friday and I'm about to become a pumpkin, and it doesn't really matter if it's compile-timer assert or not, I'm going to sqaush it to this:

static u64 *svm_vmcb_lbr(struct vcpu_svm *svm, u32 msr)

{
        switch (msr) {
        case MSR_IA32_LASTBRANCHFROMIP:
                return &svm->vmcb->save.br_from;
        case MSR_IA32_LASTBRANCHTOIP:
                return &svm->vmcb->save.br_to;
        case MSR_IA32_LASTINTFROMIP:
                return &svm->vmcb->save.last_excp_from;
        case MSR_IA32_LASTINTTOIP:
                return &svm->vmcb->save.last_excp_to;
        default:
                break;
        }
        KVM_BUG_ON(1, svm->vcpu.kvm);
        return &svm->vmcb->save.br_from;
}
Quoted 2 lines

Can you double check that kvm-x86/nested builds for you? In the process of rebuilding kvm-x86/next...

It builds. Thanks for taking care of this.

author maintainer Applied
#7 lore raw
In reply to Sean Christopherson
Quoted 4 lines · Sean Christopherson

Add a helper to get a pointer to the corresponding VMCB field given an LBR MSR index, and use it to dedup the handling in svm_{g,s}et_msr().

No functional change intended.

Applied to kvm-x86 nested (on March 13th...).

[1/1] KVM: SVM: Add a helper to get LBR field pointer to dedup MSR accesses

      https://github.com/kvm-x86/linux/commit/0b4a043a5414

--
https://github.com/kvm-x86/linux/tree/next