Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
From:   KarimAllah Ahmed <karahmed@amazon.de>
To:     kvm@vger.kernel.org, linux-kernel@vger.kernel.org, x86@kernel.org
Cc:     KarimAllah Ahmed <karahmed@amazon.de>,
        Asit Mallick <asit.k.mallick@intel.com>,
        Arjan Van De Ven <arjan.van.de.ven@intel.com>,
        Dave Hansen <dave.hansen@intel.com>,
        Andi Kleen <ak@linux.intel.com>,
        Andrea Arcangeli <aarcange@redhat.com>,
        Linus Torvalds <torvalds@linux-foundation.org>,
        Tim Chen <tim.c.chen@linux.intel.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Dan Williams <dan.j.williams@intel.com>,
        Jun Nakajima <jun.nakajima@intel.com>,
        Paolo Bonzini <pbonzini@redhat.com>,
        David Woodhouse <dwmw@amazon.co.uk>,
        Greg KH <gregkh@linuxfoundation.org>,
        Andy Lutomirski <luto@kernel.org>,
        Ashok Raj <ashok.raj@intel.com>
Subject: [PATCH v5 4/5] KVM: VMX: Allow direct access to MSR_IA32_SPEC_CTRL
Date:   Wed, 31 Jan 2018 20:37:46 +0100
Message-Id: <1517427467-28567-5-git-send-email-karahmed@amazon.de>
In-Reply-To: <1517427467-28567-1-git-send-email-karahmed@amazon.de>
References: <1517427467-28567-1-git-send-email-karahmed@amazon.de>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

[ 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 atomically saving and restoring the
MSR_IA32_SPEC_CTRL for guests that do not actually use the MSR, only
add_atomic_switch_msr 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]

Cc: Asit Mallick <asit.k.mallick@intel.com>
Cc: Arjan Van De Ven <arjan.van.de.ven@intel.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Tim Chen <tim.c.chen@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Jun Nakajima <jun.nakajima@intel.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: David Woodhouse <dwmw@amazon.co.uk>
Cc: Greg KH <gregkh@linuxfoundation.org>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: KarimAllah Ahmed <karahmed@amazon.de>
Signed-off-by: David Woodhouse <dwmw@amazon.co.uk>
---
v5:
- Also check for X86_FEATURE_SPEC_CTRL for the msr reads/writes
v4:
- Add IBRS to kvm_cpuid_8000_0008_ebx_x86_features
- Handling nested guests
v3:
- Save/restore manually
- Fix CPUID handling
- Fix a copy & paste error in the name of SPEC_CTRL MSR in
  disable_intercept.
- support !cpu_has_vmx_msr_bitmap()
v2:
- remove 'host_spec_ctrl' in favor of only a comment (dwmw@).
- special case writing '0' in SPEC_CTRL to avoid confusing live-migration
  when the instance never used the MSR (dwmw@).
- depend on X86_FEATURE_IBRS instead of X86_FEATURE_SPEC_CTRL (dwmw@).
- add MSR_IA32_SPEC_CTRL to the list of MSRs to save (dropped it by accident).
---
 arch/x86/kvm/cpuid.c |  9 ++++---
 arch/x86/kvm/vmx.c   | 73 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 arch/x86/kvm/x86.c   |  2 +-
 3 files changed, 80 insertions(+), 4 deletions(-)

diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
index 1909635..13f5d42 100644
--- a/arch/x86/kvm/cpuid.c
+++ b/arch/x86/kvm/cpuid.c
@@ -367,7 +367,7 @@ 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_cpuid_C000_0001_edx_x86_features =
@@ -394,7 +394,8 @@ 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(AVX512_4VNNIW) | F(AVX512_4FMAPS) | F(ARCH_CAPABILITIES);
+		F(AVX512_4VNNIW) | F(AVX512_4FMAPS) | F(SPEC_CTRL) |
+		F(ARCH_CAPABILITIES);
 
 	/* all calls to cpuid_count() should be made on the same cpu */
 	get_cpu();
@@ -630,9 +631,11 @@ 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, CPUID_8000_0008_EBX);
 		break;
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index a0b2bd1..4ee93cb 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -593,6 +593,8 @@ struct vcpu_vmx {
 	u64 		      msr_guest_kernel_gs_base;
 #endif
 	u64 		      arch_capabilities;
+	u64 		      spec_ctrl;
+	bool		      save_spec_ctrl_on_exit;
 
 	u32 vm_entry_controls_shadow;
 	u32 vm_exit_controls_shadow;
@@ -938,6 +940,8 @@ static void vmx_set_nmi_mask(struct kvm_vcpu *vcpu, bool masked);
 static bool nested_vmx_is_page_fault_vmexit(struct vmcs12 *vmcs12,
 					    u16 error_code);
 static void vmx_update_msr_bitmap(struct kvm_vcpu *vcpu);
+static void __always_inline vmx_disable_intercept_for_msr(unsigned long *msr_bitmap,
+							  u32 msr, int type);
 
 static DEFINE_PER_CPU(struct vmcs *, vmxarea);
 static DEFINE_PER_CPU(struct vmcs *, current_vmcs);
@@ -3238,6 +3242,14 @@ static int vmx_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 	case MSR_IA32_TSC:
 		msr_info->data = guest_read_tsc(vcpu);
 		break;
+	case MSR_IA32_SPEC_CTRL:
+		if (!msr_info->host_initiated &&
+		    !guest_cpuid_has(vcpu, X86_FEATURE_IBRS) &&
+		    !guest_cpuid_has(vcpu, X86_FEATURE_SPEC_CTRL))
+			return 1;
+
+		msr_info->data = to_vmx(vcpu)->spec_ctrl;
+		break;
 	case MSR_IA32_ARCH_CAPABILITIES:
 		if (!msr_info->host_initiated &&
 		    !guest_cpuid_has(vcpu, X86_FEATURE_ARCH_CAPABILITIES))
@@ -3351,6 +3363,36 @@ 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(vcpu, X86_FEATURE_IBRS) &&
+		    !guest_cpuid_has(vcpu, X86_FEATURE_SPEC_CTRL))
+			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;
+
+		/*
+		 * When it's written (to non-zero) for the first time, pass
+		 * it through. This means we don't have to take the perf
+		 * hit of saving it on vmexit for the common case of guests
+		 * that don't use it.
+		 */
+		if (cpu_has_vmx_msr_bitmap() && data &&
+		    !vmx->save_spec_ctrl_on_exit) {
+			vmx->save_spec_ctrl_on_exit = true;
+
+			if (is_guest_mode(vcpu))
+				break;
+
+			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(vcpu, X86_FEATURE_IBPB) &&
@@ -5668,6 +5710,7 @@ static void vmx_vcpu_reset(struct kvm_vcpu *vcpu, bool init_event)
 	u64 cr0;
 
 	vmx->rmode.vm86_active = 0;
+	vmx->spec_ctrl = 0;
 
 	vmx->vcpu.arch.regs[VCPU_REGS_RDX] = get_rdx_init_val();
 	kvm_set_cr8(vcpu, 0);
@@ -9339,6 +9382,15 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu)
 
 	vmx_arm_hv_timer(vcpu);
 
+	/*
+	 * 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 */
@@ -9457,6 +9509,19 @@ 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 (vmx->save_spec_ctrl_on_exit)
+		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();
 
@@ -10115,6 +10180,14 @@ static inline bool nested_vmx_merge_msr_bitmap(struct kvm_vcpu *vcpu,
 				MSR_TYPE_W);
 		}
 	}
+
+	if (to_vmx(vcpu)->save_spec_ctrl_on_exit) {
+		nested_vmx_disable_intercept_for_msr(
+				msr_bitmap_l1, msr_bitmap_l0,
+				MSR_IA32_SPEC_CTRL,
+				MSR_TYPE_R | MSR_TYPE_W);
+	}
+
 	kunmap(page);
 	kvm_release_page_clean(page);
 
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 4ec142e..ac38143 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -1009,7 +1009,7 @@ 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_TSC_AUX,
-	MSR_IA32_ARCH_CAPABILITIES
+	MSR_IA32_SPEC_CTRL, MSR_IA32_ARCH_CAPABILITIES
 };
 
 static unsigned num_msrs_to_save;
-- 
2.7.4