KVM: x86/speculation: Disable Fill buffer clear within guests

commit 027bbb8 upstream

The enumeration of MD_CLEAR in CPUID(EAX=7,ECX=0).EDX{bit 10} is not an
accurate indicator on all CPUs of whether the VERW instruction will
overwrite fill buffers. FB_CLEAR enumeration in
IA32_ARCH_CAPABILITIES{bit 17} covers the case of CPUs that are not
vulnerable to MDS/TAA, indicating that microcode does overwrite fill
buffers.

Guests running in VMM environments may not be aware of all the
capabilities/vulnerabilities of the host CPU. Specifically, a guest may
apply MDS/TAA mitigations when a virtual CPU is enumerated as vulnerable
to MDS/TAA even when the physical CPU is not. On CPUs that enumerate
FB_CLEAR_CTRL the VMM may set FB_CLEAR_DIS to skip overwriting of fill
buffers by the VERW instruction. This is done by setting FB_CLEAR_DIS
during VMENTER and resetting on VMEXIT. For guests that enumerate
FB_CLEAR (explicitly asking for fill buffer clear capability) the VMM
will not use FB_CLEAR_DIS.

Irrespective of guest state, host overwrites CPU buffers before VMENTER
to protect itself from an MMIO capable guest, as part of mitigation for
MMIO Stale Data vulnerabilities.

Signed-off-by: Pawan Gupta <pawan.kumar.gupta@linux.intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
[cascardo: arch/x86/kvm/vmx.c has been split and context adjustment at
 vmx_vcpu_run]
[cascardo: moved functions so they are after struct vcpu_vmx definition]
[cascardo: fb_clear is disabled/enabled around __vmx_vcpu_run]
[cascardo: conflict context fixups]
Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@canonical.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

arch/x86/include/asm/msr-index.h

@@ -115,6 +115,11 @@
 						 * VERW clears CPU fill buffer
 						 * even on MDS_NO CPUs.
 						 */
+#define ARCH_CAP_FB_CLEAR_CTRL		BIT(18)	/*
+						 * MSR_IA32_MCU_OPT_CTRL[FB_CLEAR_DIS]
+						 * bit available to control VERW
+						 * behavior.
+						 */
 
 #define MSR_IA32_FLUSH_CMD		0x0000010b
 #define L1D_FLUSH			BIT(0)	/*
@@ -132,6 +137,7 @@
 /* SRBDS support */
 #define MSR_IA32_MCU_OPT_CTRL		0x00000123
 #define RNGDS_MITG_DIS			BIT(0)
+#define FB_CLEAR_DIS			BIT(3)	/* CPU Fill buffer clear disable */
 
 #define MSR_IA32_SYSENTER_CS		0x00000174
 #define MSR_IA32_SYSENTER_ESP		0x00000175

arch/x86/kvm/vmx.c

@@ -214,6 +214,9 @@ static const struct {
 #define L1D_CACHE_ORDER 4
 static void *vmx_l1d_flush_pages;
 
+/* Control for disabling CPU Fill buffer clear */
+static bool __read_mostly vmx_fb_clear_ctrl_available;
+
 static int vmx_setup_l1d_flush(enum vmx_l1d_flush_state l1tf)
 {
 	struct page *page;
@@ -820,6 +823,8 @@ struct vcpu_vmx {
 	 */
 	u64 msr_ia32_feature_control;
 	u64 msr_ia32_feature_control_valid_bits;
+	u64 msr_ia32_mcu_opt_ctrl;
+	bool disable_fb_clear;
 };
 
 enum segment_cache_field {
@@ -1628,6 +1633,60 @@ static inline void __invept(unsigned long ext, u64 eptp, gpa_t gpa)
 			: : "a" (&operand), "c" (ext) : "cc", "memory");
 }
 
+static void vmx_setup_fb_clear_ctrl(void)
+{
+	u64 msr;
+
+	if (boot_cpu_has(X86_FEATURE_ARCH_CAPABILITIES) &&
+	    !boot_cpu_has_bug(X86_BUG_MDS) &&
+	    !boot_cpu_has_bug(X86_BUG_TAA)) {
+		rdmsrl(MSR_IA32_ARCH_CAPABILITIES, msr);
+		if (msr & ARCH_CAP_FB_CLEAR_CTRL)
+			vmx_fb_clear_ctrl_available = true;
+	}
+}
+
+static __always_inline void vmx_disable_fb_clear(struct vcpu_vmx *vmx)
+{
+	u64 msr;
+
+	if (!vmx->disable_fb_clear)
+		return;
+
+	rdmsrl(MSR_IA32_MCU_OPT_CTRL, msr);
+	msr |= FB_CLEAR_DIS;
+	wrmsrl(MSR_IA32_MCU_OPT_CTRL, msr);
+	/* Cache the MSR value to avoid reading it later */
+	vmx->msr_ia32_mcu_opt_ctrl = msr;
+}
+
+static __always_inline void vmx_enable_fb_clear(struct vcpu_vmx *vmx)
+{
+	if (!vmx->disable_fb_clear)
+		return;
+
+	vmx->msr_ia32_mcu_opt_ctrl &= ~FB_CLEAR_DIS;
+	wrmsrl(MSR_IA32_MCU_OPT_CTRL, vmx->msr_ia32_mcu_opt_ctrl);
+}
+
+static void vmx_update_fb_clear_dis(struct kvm_vcpu *vcpu, struct vcpu_vmx *vmx)
+{
+	vmx->disable_fb_clear = vmx_fb_clear_ctrl_available;
+
+	/*
+	 * If guest will not execute VERW, there is no need to set FB_CLEAR_DIS
+	 * at VMEntry. Skip the MSR read/write when a guest has no use case to
+	 * execute VERW.
+	 */
+	if ((vcpu->arch.arch_capabilities & ARCH_CAP_FB_CLEAR) ||
+	   ((vcpu->arch.arch_capabilities & ARCH_CAP_MDS_NO) &&
+	    (vcpu->arch.arch_capabilities & ARCH_CAP_TAA_NO) &&
+	    (vcpu->arch.arch_capabilities & ARCH_CAP_PSDP_NO) &&
+	    (vcpu->arch.arch_capabilities & ARCH_CAP_FBSDP_NO) &&
+	    (vcpu->arch.arch_capabilities & ARCH_CAP_SBDR_SSDP_NO)))
+		vmx->disable_fb_clear = false;
+}
+
 static struct shared_msr_entry *find_msr_entry(struct vcpu_vmx *vmx, u32 msr)
 {
 	int i;
@@ -3700,9 +3759,13 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 			}
 			break;
 		}
-		ret = kvm_set_msr_common(vcpu, msr_info);
+			ret = kvm_set_msr_common(vcpu, msr_info);
 	}
 
+	/* FB_CLEAR may have changed, also update the FB_CLEAR_DIS behavior */
+	if (msr_index == MSR_IA32_ARCH_CAPABILITIES)
+		vmx_update_fb_clear_dis(vcpu, vmx);
+
 	return ret;
 }
 
@@ -6008,6 +6071,8 @@ static void vmx_vcpu_reset(struct kvm_vcpu *vcpu, bool init_event)
 	update_exception_bitmap(vcpu);
 
 	vpid_sync_context(vmx->vpid);
+
+	vmx_update_fb_clear_dis(vcpu, vmx);
 }
 
 /*
@@ -9783,6 +9848,8 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu)
 		 kvm_arch_has_assigned_device(vcpu->kvm))
 		mds_clear_cpu_buffers();
 
+	vmx_disable_fb_clear(vmx);
+
 	asm(
 		/* Store host registers */
 		"push %%" _ASM_DX "; push %%" _ASM_BP ";"
@@ -9900,6 +9967,8 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu)
 #endif
 	      );
 
+	vmx_enable_fb_clear(vmx);
+
 	/*
 	 * 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
@@ -12924,8 +12993,11 @@ static int __init vmx_init(void)
 		}
 	}
 
+	vmx_setup_fb_clear_ctrl();
+
 	for_each_possible_cpu(cpu) {
 		INIT_LIST_HEAD(&per_cpu(loaded_vmcss_on_cpu, cpu));
+
 		INIT_LIST_HEAD(&per_cpu(blocked_vcpu_on_cpu, cpu));
 		spin_lock_init(&per_cpu(blocked_vcpu_on_cpu_lock, cpu));
 	}

arch/x86/kvm/x86.c

@@ -1127,6 +1127,10 @@ u64 kvm_get_arch_capabilities(void)
 
 	/* KVM does not emulate MSR_IA32_TSX_CTRL.  */
 	data &= ~ARCH_CAP_TSX_CTRL_MSR;
+
+	/* Guests don't need to know "Fill buffer clear control" exists */
+	data &= ~ARCH_CAP_FB_CLEAR_CTRL;
+
 	return data;
 }