According to the APM, TF on VMRUN causes a #DB after VMRUN completes on
the _host_ side. However, KVM injects a #DB in L2 context instead (or
exits to userspace if KVM_GUESTDBG_SINGLESTEP is set) in
kvm_skip_emulated_instruction().

Avoid single-step handling on VMRUN by open-coding the rest of
kvm_skip_emulated_instruction() in nested_svm_vmrun(). This doesn't look
pretty, but following changes will need to open-code
kvm_pmu_instruction_retired() anyway, and will cleanup the code.  This
ignores TF on VMRUN instead of injecting a spurious exception into
L2. Document this virtualization hole with a FIXME.

Note that a failed VMRUN would have been correctly single-stepped, but
now TF is always ignored for consistency and simplicity purposes.  VMX
does not support TF on a successful VMLAUNCH/VMRESUME, so it's unlikely
that single-stepping VMRUN properly is important, especially if it's
only for failed VMRUNs.

Fixes: c8e16b78c614 ("x86: KVM: svm: eliminate hardcoded RIP advancement from vmrun_interception()")
Signed-off-by: Yosry Ahmed <yosry@kernel.org>
---
 arch/x86/kvm/svm/nested.c | 18 +++++++++++++++---
 arch/x86/kvm/svm/svm.c    |  2 +-
 arch/x86/kvm/svm/svm.h    |  2 ++
 3 files changed, 18 insertions(+), 4 deletions(-)

diff --git a/arch/x86/kvm/svm/nested.c b/arch/x86/kvm/svm/nested.c
index 28ac5d5c990dd..01e3e6fa8bbb1 100644
--- a/arch/x86/kvm/svm/nested.c
+++ b/arch/x86/kvm/svm/nested.c
@@ -30,6 +30,7 @@
 #include "lapic.h"
 #include "svm.h"
 #include "hyperv.h"
+#include "pmu.h"
 
 #define CC KVM_NESTED_VMENTER_CONSISTENCY_CHECK
 
@@ -1145,11 +1146,22 @@ int nested_svm_vmrun(struct kvm_vcpu *vcpu)
 			return kvm_handle_memory_failure(vcpu, X86EMUL_IO_NEEDED, NULL);
 
 		/* Advance RIP past VMRUN as part of the nested #VMEXIT. */
-		return kvm_skip_emulated_instruction(vcpu);
+		if (!svm_skip_emulated_instruction(vcpu))
+			return 0;
+
+		kvm_pmu_instruction_retired(vcpu);
+		return 1;
 	}
 
-	/* At this point, VMRUN is guaranteed to not fault; advance RIP. */
-	ret = kvm_skip_emulated_instruction(vcpu);
+	/*
+	 * At this point, VMRUN is guaranteed to not fault; advance RIP.
+	 *
+	 * FIXME: If TF is set on VMRUN should inject a #DB (or handle guest
+	 * debugging) right after #VMEXIT, right now it's just ignored.
+	 */
+	ret = svm_skip_emulated_instruction(vcpu);
+	if (ret)
+		kvm_pmu_instruction_retired(vcpu);
 
 	/*
 	 * Since vmcb01 is not in use, we can use it to store some of the L1
diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c
index e74fcde6155ec..183e577802301 100644
--- a/arch/x86/kvm/svm/svm.c
+++ b/arch/x86/kvm/svm/svm.c
@@ -333,7 +333,7 @@ static int __svm_skip_emulated_instruction(struct kvm_vcpu *vcpu,
 	return 1;
 }
 
-static int svm_skip_emulated_instruction(struct kvm_vcpu *vcpu)
+int svm_skip_emulated_instruction(struct kvm_vcpu *vcpu)
 {
 	return __svm_skip_emulated_instruction(vcpu, EMULTYPE_SKIP, true);
 }
diff --git a/arch/x86/kvm/svm/svm.h b/arch/x86/kvm/svm/svm.h
index 2b6733dffd76f..e5d9984ef6320 100644
--- a/arch/x86/kvm/svm/svm.h
+++ b/arch/x86/kvm/svm/svm.h
@@ -832,6 +832,8 @@ static inline void svm_enable_intercept_for_msr(struct kvm_vcpu *vcpu,
 	svm_set_intercept_for_msr(vcpu, msr, type, true);
 }
 
+int svm_skip_emulated_instruction(struct kvm_vcpu *vcpu);
+
 /* nested.c */
 
 #define NESTED_EXIT_HOST	0	/* Exit handled on host level */
-- 
2.54.0.794.g4f17f83d09-goog