KVM/VMX: Allow direct access to MSR_IA32_SPEC_CTRL

commit d28b387fb74da95d69d2615732f50cceb38e9a4d upstream.

[ Based on a patch from Ashok Raj <ashok.raj@intel.com> ]

Add direct access to MSR_IA32_SPEC_CTRL for guests. This is needed for
guests that will only mitigate Spectre V2 through IBRS+IBPB and will not
be using a retpoline+IBPB based approach.

To avoid the overhead of saving and restoring the MSR_IA32_SPEC_CTRL for
guests that do not actually use the MSR, only start saving and restoring
when a non-zero is written to it.

No attempt is made to handle STIBP here, intentionally. Filtering STIBP
may be added in a future patch, which may require trapping all writes
if we don't want to pass it through directly to the guest.

[dwmw2: Clean up CPUID bits, save/restore manually, handle reset]

Signed-off-by: KarimAllah Ahmed <karahmed@amazon.de>
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Darren Kenny <darren.kenny@oracle.com>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reviewed-by: Jim Mattson <jmattson@google.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jun Nakajima <jun.nakajima@intel.com>
Cc: kvm@vger.kernel.org
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Tim Chen <tim.c.chen@linux.intel.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Asit Mallick <asit.k.mallick@intel.com>
Cc: Arjan Van De Ven <arjan.van.de.ven@intel.com>
Cc: Greg KH <gregkh@linuxfoundation.org>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ashok Raj <ashok.raj@intel.com>
Link: https://lkml.kernel.org/r/1517522386-18410-5-git-send-email-karahmed@amazon.de
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[bwh: Backported to 3.16:
 - No support for nested MSR bitmaps
 - Adjust context]
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>

4 files changed, 81 insertions, 4 deletions

diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
index 17b732346cec..c911af5d0cf5 100644
--- a/arch/x86/kvm/cpuid.c
+++ b/arch/x86/kvm/cpuid.c
@@ -302,7 +302,7 @@ static inline int __do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function,
 
 	/* cpuid 0x80000008.ebx */
 	const u32 kvm_cpuid_8000_0008_ebx_x86_features =
-		F(IBPB);
+		F(IBPB) | F(IBRS);
 
 	/* cpuid 0xC0000001.edx */
 	const u32 kvm_supported_word5_x86_features =
@@ -318,7 +318,7 @@ static inline int __do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function,
 
 	/* cpuid 7.0.edx*/
 	const u32 kvm_cpuid_7_0_edx_x86_features =
-		F(ARCH_CAPABILITIES);
+		F(SPEC_CTRL) | F(ARCH_CAPABILITIES);
 
 	/* all calls to cpuid_count() should be made on the same cpu */
 	get_cpu();
@@ -524,9 +524,11 @@ static inline int __do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function,
 			g_phys_as = phys_as;
 		entry->eax = g_phys_as | (virt_as << 8);
 		entry->edx = 0;
-		/* IBPB isn't necessarily present in hardware cpuid */
+		/* IBRS and IBPB aren't necessarily present in hardware cpuid */
 		if (boot_cpu_has(X86_FEATURE_IBPB))
 			entry->ebx |= F(IBPB);
+		if (boot_cpu_has(X86_FEATURE_IBRS))
+			entry->ebx |= F(IBRS);
 		entry->ebx &= kvm_cpuid_8000_0008_ebx_x86_features;
 		cpuid_mask(&entry->ebx, 11);
 		break;
diff --git a/arch/x86/kvm/cpuid.h b/arch/x86/kvm/cpuid.h
index 3e901fa4940d..aff98b185e87 100644
--- a/arch/x86/kvm/cpuid.h
+++ b/arch/x86/kvm/cpuid.h
@@ -115,6 +115,17 @@ static inline bool guest_cpuid_has_ibpb(struct kvm_vcpu *vcpu)
 	return best && (best->edx & bit(X86_FEATURE_SPEC_CTRL));
 }
 
+static inline bool guest_cpuid_has_ibrs(struct kvm_vcpu *vcpu)
+{
+	struct kvm_cpuid_entry2 *best;
+
+	best = kvm_find_cpuid_entry(vcpu, 0x80000008, 0);
+	if (best && (best->ebx & bit(X86_FEATURE_IBRS)))
+		return true;
+	best = kvm_find_cpuid_entry(vcpu, 7, 0);
+	return best && (best->edx & bit(X86_FEATURE_SPEC_CTRL));
+}
+
 static inline bool guest_cpuid_has_arch_capabilities(struct kvm_vcpu *vcpu)
 {
 	struct kvm_cpuid_entry2 *best;
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 059d5e8f795c..805eb0cfc613 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -433,6 +433,7 @@ struct vcpu_vmx {
 #endif
 
 	u64 		      arch_capabilities;
+	u64 		      spec_ctrl;
 
 	u32 vm_entry_controls_shadow;
 	u32 vm_exit_controls_shadow;
@@ -2524,6 +2525,13 @@ static int vmx_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 	case MSR_IA32_TSC:
 		msr_info->data = guest_read_tsc();
 		break;
+	case MSR_IA32_SPEC_CTRL:
+		if (!msr_info->host_initiated &&
+		    !guest_cpuid_has_ibrs(vcpu))
+			return 1;
+
+		msr_info->data = to_vmx(vcpu)->spec_ctrl;
+		break;
 	case MSR_IA32_ARCH_CAPABILITIES:
 		if (!msr_info->host_initiated &&
 		    !guest_cpuid_has_arch_capabilities(vcpu))
@@ -2622,6 +2630,36 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 	case MSR_IA32_TSC:
 		kvm_write_tsc(vcpu, msr_info);
 		break;
+	case MSR_IA32_SPEC_CTRL:
+		if (!msr_info->host_initiated &&
+		    !guest_cpuid_has_ibrs(vcpu))
+			return 1;
+
+		/* The STIBP bit doesn't fault even if it's not advertised */
+		if (data & ~(SPEC_CTRL_IBRS | SPEC_CTRL_STIBP))
+			return 1;
+
+		vmx->spec_ctrl = data;
+
+		if (!data)
+			break;
+
+		/*
+		 * For non-nested:
+		 * When it's written (to non-zero) for the first time, pass
+		 * it through.
+		 *
+		 * For nested:
+		 * The handling of the MSR bitmap for L2 guests is done in
+		 * nested_vmx_merge_msr_bitmap. We should not touch the
+		 * vmcs02.msr_bitmap here since it gets completely overwritten
+		 * in the merging. We update the vmcs01 here for L1 as well
+		 * since it will end up touching the MSR anyway now.
+		 */
+		vmx_disable_intercept_for_msr(vmx->vmcs01.msr_bitmap,
+					      MSR_IA32_SPEC_CTRL,
+					      MSR_TYPE_RW);
+		break;
 	case MSR_IA32_PRED_CMD:
 		if (!msr_info->host_initiated &&
 		    !guest_cpuid_has_ibpb(vcpu))
@@ -4617,6 +4655,7 @@ static void vmx_vcpu_reset(struct kvm_vcpu *vcpu)
 	struct msr_data apic_base_msr;
 
 	vmx->rmode.vm86_active = 0;
+	vmx->spec_ctrl = 0;
 
 	vmx->soft_vnmi_blocked = 0;
 
@@ -7496,6 +7535,15 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu)
 	atomic_switch_perf_msrs(vmx);
 	debugctlmsr = get_debugctlmsr();
 
+	/*
+	 * If this vCPU has touched SPEC_CTRL, restore the guest's value if
+	 * it's non-zero. Since vmentry is serialising on affected CPUs, there
+	 * is no need to worry about the conditional branch over the wrmsr
+	 * being speculatively taken.
+	 */
+	if (vmx->spec_ctrl)
+		wrmsrl(MSR_IA32_SPEC_CTRL, vmx->spec_ctrl);
+
 	vmx->__launched = vmx->loaded_vmcs->launched;
 	asm(
 		/* Store host registers */
@@ -7614,6 +7662,22 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu)
 #endif
 	      );
 
+	/*
+	 * We do not use IBRS in the kernel. If this vCPU has used the
+	 * SPEC_CTRL MSR it may have left it on; save the value and
+	 * turn it off. This is much more efficient than blindly adding
+	 * it to the atomic save/restore list. Especially as the former
+	 * (Saving guest MSRs on vmexit) doesn't even exist in KVM.
+	 *
+	 * If the L01 MSR bitmap does not intercept the MSR, then we need to
+	 * save it.
+	 */
+	if (!msr_write_intercepted_l01(vcpu, MSR_IA32_SPEC_CTRL))
+		rdmsrl(MSR_IA32_SPEC_CTRL, vmx->spec_ctrl);
+
+	if (vmx->spec_ctrl)
+		wrmsrl(MSR_IA32_SPEC_CTRL, 0);
+
 	/* Eliminate branch target predictions from guest mode */
 	vmexit_fill_RSB();
 
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 4cabaa3ee7d5..e959897a0f44 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -894,7 +894,7 @@ static u32 msrs_to_save[] = {
 #endif
 	MSR_IA32_TSC, MSR_IA32_CR_PAT, MSR_VM_HSAVE_PA,
 	MSR_IA32_FEATURE_CONTROL, MSR_IA32_BNDCFGS,
-	MSR_IA32_ARCH_CAPABILITIES
+	MSR_IA32_SPEC_CTRL, MSR_IA32_ARCH_CAPABILITIES
 };
 
 static unsigned num_msrs_to_save;