Merge branch 'l1tf-final' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Merge L1 Terminal Fault fixes from Thomas Gleixner:
 "L1TF, aka L1 Terminal Fault, is yet another speculative hardware
  engineering trainwreck. It's a hardware vulnerability which allows
  unprivileged speculative access to data which is available in the
  Level 1 Data Cache when the page table entry controlling the virtual
  address, which is used for the access, has the Present bit cleared or
  other reserved bits set.

  If an instruction accesses a virtual address for which the relevant
  page table entry (PTE) has the Present bit cleared or other reserved
  bits set, then speculative execution ignores the invalid PTE and loads
  the referenced data if it is present in the Level 1 Data Cache, as if
  the page referenced by the address bits in the PTE was still present
  and accessible.

  While this is a purely speculative mechanism and the instruction will
  raise a page fault when it is retired eventually, the pure act of
  loading the data and making it available to other speculative
  instructions opens up the opportunity for side channel attacks to
  unprivileged malicious code, similar to the Meltdown attack.

  While Meltdown breaks the user space to kernel space protection, L1TF
  allows to attack any physical memory address in the system and the
  attack works across all protection domains. It allows an attack of SGX
  and also works from inside virtual machines because the speculation
  bypasses the extended page table (EPT) protection mechanism.

  The assoicated CVEs are: CVE-2018-3615, CVE-2018-3620, CVE-2018-3646

  The mitigations provided by this pull request include:

   - Host side protection by inverting the upper address bits of a non
     present page table entry so the entry points to uncacheable memory.

   - Hypervisor protection by flushing L1 Data Cache on VMENTER.

   - SMT (HyperThreading) control knobs, which allow to 'turn off' SMT
     by offlining the sibling CPU threads. The knobs are available on
     the kernel command line and at runtime via sysfs

   - Control knobs for the hypervisor mitigation, related to L1D flush
     and SMT control. The knobs are available on the kernel command line
     and at runtime via sysfs

   - Extensive documentation about L1TF including various degrees of
     mitigations.

  Thanks to all people who have contributed to this in various ways -
  patches, review, testing, backporting - and the fruitful, sometimes
  heated, but at the end constructive discussions.

  There is work in progress to provide other forms of mitigations, which
  might be less horrible performance wise for a particular kind of
  workloads, but this is not yet ready for consumption due to their
  complexity and limitations"

* 'l1tf-final' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (75 commits)
  x86/microcode: Allow late microcode loading with SMT disabled
  tools headers: Synchronise x86 cpufeatures.h for L1TF additions
  x86/mm/kmmio: Make the tracer robust against L1TF
  x86/mm/pat: Make set_memory_np() L1TF safe
  x86/speculation/l1tf: Make pmd/pud_mknotpresent() invert
  x86/speculation/l1tf: Invert all not present mappings
  cpu/hotplug: Fix SMT supported evaluation
  KVM: VMX: Tell the nested hypervisor to skip L1D flush on vmentry
  x86/speculation: Use ARCH_CAPABILITIES to skip L1D flush on vmentry
  x86/speculation: Simplify sysfs report of VMX L1TF vulnerability
  Documentation/l1tf: Remove Yonah processors from not vulnerable list
  x86/KVM/VMX: Don't set l1tf_flush_l1d from vmx_handle_external_intr()
  x86/irq: Let interrupt handlers set kvm_cpu_l1tf_flush_l1d
  x86: Don't include linux/irq.h from asm/hardirq.h
  x86/KVM/VMX: Introduce per-host-cpu analogue of l1tf_flush_l1d
  x86/irq: Demote irq_cpustat_t::__softirq_pending to u16
  x86/KVM/VMX: Move the l1tf_flush_l1d test to vmx_l1d_flush()
  x86/KVM/VMX: Replace 'vmx_l1d_flush_always' with 'vmx_l1d_flush_cond'
  x86/KVM/VMX: Don't set l1tf_flush_l1d to true from vmx_l1d_flush()
  cpu/hotplug: detect SMT disabled by BIOS
  ...

3 files changed, 411 insertions, 79 deletions

diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 6b8f11521c41..a44e568363a4 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -3840,6 +3840,7 @@ int kvm_handle_page_fault(struct kvm_vcpu *vcpu, u64 error_code,
 {
 	int r = 1;
 
+	vcpu->arch.l1tf_flush_l1d = true;
 	switch (vcpu->arch.apf.host_apf_reason) {
 	default:
 		trace_kvm_page_fault(fault_address, error_code);
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 5d8e317c2b04..46b428c0990e 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -188,6 +188,150 @@ module_param(ple_window_max, uint, 0444);
 
 extern const ulong vmx_return;
 
+static DEFINE_STATIC_KEY_FALSE(vmx_l1d_should_flush);
+static DEFINE_STATIC_KEY_FALSE(vmx_l1d_flush_cond);
+static DEFINE_MUTEX(vmx_l1d_flush_mutex);
+
+/* Storage for pre module init parameter parsing */
+static enum vmx_l1d_flush_state __read_mostly vmentry_l1d_flush_param = VMENTER_L1D_FLUSH_AUTO;
+
+static const struct {
+	const char *option;
+	enum vmx_l1d_flush_state cmd;
+} vmentry_l1d_param[] = {
+	{"auto",	VMENTER_L1D_FLUSH_AUTO},
+	{"never",	VMENTER_L1D_FLUSH_NEVER},
+	{"cond",	VMENTER_L1D_FLUSH_COND},
+	{"always",	VMENTER_L1D_FLUSH_ALWAYS},
+};
+
+#define L1D_CACHE_ORDER 4
+static void *vmx_l1d_flush_pages;
+
+static int vmx_setup_l1d_flush(enum vmx_l1d_flush_state l1tf)
+{
+	struct page *page;
+	unsigned int i;
+
+	if (!enable_ept) {
+		l1tf_vmx_mitigation = VMENTER_L1D_FLUSH_EPT_DISABLED;
+		return 0;
+	}
+
+       if (boot_cpu_has(X86_FEATURE_ARCH_CAPABILITIES)) {
+	       u64 msr;
+
+	       rdmsrl(MSR_IA32_ARCH_CAPABILITIES, msr);
+	       if (msr & ARCH_CAP_SKIP_VMENTRY_L1DFLUSH) {
+		       l1tf_vmx_mitigation = VMENTER_L1D_FLUSH_NOT_REQUIRED;
+		       return 0;
+	       }
+       }
+
+	/* If set to auto use the default l1tf mitigation method */
+	if (l1tf == VMENTER_L1D_FLUSH_AUTO) {
+		switch (l1tf_mitigation) {
+		case L1TF_MITIGATION_OFF:
+			l1tf = VMENTER_L1D_FLUSH_NEVER;
+			break;
+		case L1TF_MITIGATION_FLUSH_NOWARN:
+		case L1TF_MITIGATION_FLUSH:
+		case L1TF_MITIGATION_FLUSH_NOSMT:
+			l1tf = VMENTER_L1D_FLUSH_COND;
+			break;
+		case L1TF_MITIGATION_FULL:
+		case L1TF_MITIGATION_FULL_FORCE:
+			l1tf = VMENTER_L1D_FLUSH_ALWAYS;
+			break;
+		}
+	} else if (l1tf_mitigation == L1TF_MITIGATION_FULL_FORCE) {
+		l1tf = VMENTER_L1D_FLUSH_ALWAYS;
+	}
+
+	if (l1tf != VMENTER_L1D_FLUSH_NEVER && !vmx_l1d_flush_pages &&
+	    !boot_cpu_has(X86_FEATURE_FLUSH_L1D)) {
+		page = alloc_pages(GFP_KERNEL, L1D_CACHE_ORDER);
+		if (!page)
+			return -ENOMEM;
+		vmx_l1d_flush_pages = page_address(page);
+
+		/*
+		 * Initialize each page with a different pattern in
+		 * order to protect against KSM in the nested
+		 * virtualization case.
+		 */
+		for (i = 0; i < 1u << L1D_CACHE_ORDER; ++i) {
+			memset(vmx_l1d_flush_pages + i * PAGE_SIZE, i + 1,
+			       PAGE_SIZE);
+		}
+	}
+
+	l1tf_vmx_mitigation = l1tf;
+
+	if (l1tf != VMENTER_L1D_FLUSH_NEVER)
+		static_branch_enable(&vmx_l1d_should_flush);
+	else
+		static_branch_disable(&vmx_l1d_should_flush);
+
+	if (l1tf == VMENTER_L1D_FLUSH_COND)
+		static_branch_enable(&vmx_l1d_flush_cond);
+	else
+		static_branch_disable(&vmx_l1d_flush_cond);
+	return 0;
+}
+
+static int vmentry_l1d_flush_parse(const char *s)
+{
+	unsigned int i;
+
+	if (s) {
+		for (i = 0; i < ARRAY_SIZE(vmentry_l1d_param); i++) {
+			if (sysfs_streq(s, vmentry_l1d_param[i].option))
+				return vmentry_l1d_param[i].cmd;
+		}
+	}
+	return -EINVAL;
+}
+
+static int vmentry_l1d_flush_set(const char *s, const struct kernel_param *kp)
+{
+	int l1tf, ret;
+
+	if (!boot_cpu_has(X86_BUG_L1TF))
+		return 0;
+
+	l1tf = vmentry_l1d_flush_parse(s);
+	if (l1tf < 0)
+		return l1tf;
+
+	/*
+	 * Has vmx_init() run already? If not then this is the pre init
+	 * parameter parsing. In that case just store the value and let
+	 * vmx_init() do the proper setup after enable_ept has been
+	 * established.
+	 */
+	if (l1tf_vmx_mitigation == VMENTER_L1D_FLUSH_AUTO) {
+		vmentry_l1d_flush_param = l1tf;
+		return 0;
+	}
+
+	mutex_lock(&vmx_l1d_flush_mutex);
+	ret = vmx_setup_l1d_flush(l1tf);
+	mutex_unlock(&vmx_l1d_flush_mutex);
+	return ret;
+}
+
+static int vmentry_l1d_flush_get(char *s, const struct kernel_param *kp)
+{
+	return sprintf(s, "%s\n", vmentry_l1d_param[l1tf_vmx_mitigation].option);
+}
+
+static const struct kernel_param_ops vmentry_l1d_flush_ops = {
+	.set = vmentry_l1d_flush_set,
+	.get = vmentry_l1d_flush_get,
+};
+module_param_cb(vmentry_l1d_flush, &vmentry_l1d_flush_ops, NULL, 0644);
+
 struct kvm_vmx {
 	struct kvm kvm;
 
@@ -757,6 +901,11 @@ static inline int pi_test_sn(struct pi_desc *pi_desc)
 			(unsigned long *)&pi_desc->control);
 }
 
+struct vmx_msrs {
+	unsigned int		nr;
+	struct vmx_msr_entry	val[NR_AUTOLOAD_MSRS];
+};
+
 struct vcpu_vmx {
 	struct kvm_vcpu       vcpu;
 	unsigned long         host_rsp;
@@ -790,9 +939,8 @@ struct vcpu_vmx {
 	struct loaded_vmcs   *loaded_vmcs;
 	bool                  __launched; /* temporary, used in vmx_vcpu_run */
 	struct msr_autoload {
-		unsigned nr;
-		struct vmx_msr_entry guest[NR_AUTOLOAD_MSRS];
-		struct vmx_msr_entry host[NR_AUTOLOAD_MSRS];
+		struct vmx_msrs guest;
+		struct vmx_msrs host;
 	} msr_autoload;
 	struct {
 		int           loaded;
@@ -2377,9 +2525,20 @@ static void clear_atomic_switch_msr_special(struct vcpu_vmx *vmx,
 	vm_exit_controls_clearbit(vmx, exit);
 }
 
+static int find_msr(struct vmx_msrs *m, unsigned int msr)
+{
+	unsigned int i;
+
+	for (i = 0; i < m->nr; ++i) {
+		if (m->val[i].index == msr)
+			return i;
+	}
+	return -ENOENT;
+}
+
 static void clear_atomic_switch_msr(struct vcpu_vmx *vmx, unsigned msr)
 {
-	unsigned i;
+	int i;
 	struct msr_autoload *m = &vmx->msr_autoload;
 
 	switch (msr) {
@@ -2400,18 +2559,21 @@ static void clear_atomic_switch_msr(struct vcpu_vmx *vmx, unsigned msr)
 		}
 		break;
 	}
+	i = find_msr(&m->guest, msr);
+	if (i < 0)
+		goto skip_guest;
+	--m->guest.nr;
+	m->guest.val[i] = m->guest.val[m->guest.nr];
+	vmcs_write32(VM_ENTRY_MSR_LOAD_COUNT, m->guest.nr);
 
-	for (i = 0; i < m->nr; ++i)
-		if (m->guest[i].index == msr)
-			break;
-
-	if (i == m->nr)
+skip_guest:
+	i = find_msr(&m->host, msr);
+	if (i < 0)
 		return;
-	--m->nr;
-	m->guest[i] = m->guest[m->nr];
-	m->host[i] = m->host[m->nr];
-	vmcs_write32(VM_ENTRY_MSR_LOAD_COUNT, m->nr);
-	vmcs_write32(VM_EXIT_MSR_LOAD_COUNT, m->nr);
+
+	--m->host.nr;
+	m->host.val[i] = m->host.val[m->host.nr];
+	vmcs_write32(VM_EXIT_MSR_LOAD_COUNT, m->host.nr);
 }
 
 static void add_atomic_switch_msr_special(struct vcpu_vmx *vmx,
@@ -2426,9 +2588,9 @@ static void add_atomic_switch_msr_special(struct vcpu_vmx *vmx,
 }
 
 static void add_atomic_switch_msr(struct vcpu_vmx *vmx, unsigned msr,
-				  u64 guest_val, u64 host_val)
+				  u64 guest_val, u64 host_val, bool entry_only)
 {
-	unsigned i;
+	int i, j = 0;
 	struct msr_autoload *m = &vmx->msr_autoload;
 
 	switch (msr) {
@@ -2463,24 +2625,31 @@ static void add_atomic_switch_msr(struct vcpu_vmx *vmx, unsigned msr,
 		wrmsrl(MSR_IA32_PEBS_ENABLE, 0);
 	}
 
-	for (i = 0; i < m->nr; ++i)
-		if (m->guest[i].index == msr)
-			break;
+	i = find_msr(&m->guest, msr);
+	if (!entry_only)
+		j = find_msr(&m->host, msr);
 
-	if (i == NR_AUTOLOAD_MSRS) {
+	if (i == NR_AUTOLOAD_MSRS || j == NR_AUTOLOAD_MSRS) {
 		printk_once(KERN_WARNING "Not enough msr switch entries. "
 				"Can't add msr %x\n", msr);
 		return;
-	} else if (i == m->nr) {
-		++m->nr;
-		vmcs_write32(VM_ENTRY_MSR_LOAD_COUNT, m->nr);
-		vmcs_write32(VM_EXIT_MSR_LOAD_COUNT, m->nr);
 	}
+	if (i < 0) {
+		i = m->guest.nr++;
+		vmcs_write32(VM_ENTRY_MSR_LOAD_COUNT, m->guest.nr);
+	}
+	m->guest.val[i].index = msr;
+	m->guest.val[i].value = guest_val;
+
+	if (entry_only)
+		return;
 
-	m->guest[i].index = msr;
-	m->guest[i].value = guest_val;
-	m->host[i].index = msr;
-	m->host[i].value = host_val;
+	if (j < 0) {
+		j = m->host.nr++;
+		vmcs_write32(VM_EXIT_MSR_LOAD_COUNT, m->host.nr);
+	}
+	m->host.val[j].index = msr;
+	m->host.val[j].value = host_val;
 }
 
 static bool update_transition_efer(struct vcpu_vmx *vmx, int efer_offset)
@@ -2524,7 +2693,7 @@ static bool update_transition_efer(struct vcpu_vmx *vmx, int efer_offset)
 			guest_efer &= ~EFER_LME;
 		if (guest_efer != host_efer)
 			add_atomic_switch_msr(vmx, MSR_EFER,
-					      guest_efer, host_efer);
+					      guest_efer, host_efer, false);
 		return false;
 	} else {
 		guest_efer &= ~ignore_bits;
@@ -3987,7 +4156,7 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 		vcpu->arch.ia32_xss = data;
 		if (vcpu->arch.ia32_xss != host_xss)
 			add_atomic_switch_msr(vmx, MSR_IA32_XSS,
-				vcpu->arch.ia32_xss, host_xss);
+				vcpu->arch.ia32_xss, host_xss, false);
 		else
 			clear_atomic_switch_msr(vmx, MSR_IA32_XSS);
 		break;
@@ -6274,9 +6443,9 @@ static void vmx_vcpu_setup(struct vcpu_vmx *vmx)
 
 	vmcs_write32(VM_EXIT_MSR_STORE_COUNT, 0);
 	vmcs_write32(VM_EXIT_MSR_LOAD_COUNT, 0);
-	vmcs_write64(VM_EXIT_MSR_LOAD_ADDR, __pa(vmx->msr_autoload.host));
+	vmcs_write64(VM_EXIT_MSR_LOAD_ADDR, __pa(vmx->msr_autoload.host.val));
 	vmcs_write32(VM_ENTRY_MSR_LOAD_COUNT, 0);
-	vmcs_write64(VM_ENTRY_MSR_LOAD_ADDR, __pa(vmx->msr_autoload.guest));
+	vmcs_write64(VM_ENTRY_MSR_LOAD_ADDR, __pa(vmx->msr_autoload.guest.val));
 
 	if (vmcs_config.vmentry_ctrl & VM_ENTRY_LOAD_IA32_PAT)
 		vmcs_write64(GUEST_IA32_PAT, vmx->vcpu.arch.pat);
@@ -6296,8 +6465,7 @@ static void vmx_vcpu_setup(struct vcpu_vmx *vmx)
 		++vmx->nmsrs;
 	}
 
-	if (boot_cpu_has(X86_FEATURE_ARCH_CAPABILITIES))
-		rdmsrl(MSR_IA32_ARCH_CAPABILITIES, vmx->arch_capabilities);
+	vmx->arch_capabilities = kvm_get_arch_capabilities();
 
 	vm_exit_controls_init(vmx, vmcs_config.vmexit_ctrl);
 
@@ -9548,6 +9716,79 @@ static int vmx_handle_exit(struct kvm_vcpu *vcpu)
 	}
 }
 
+/*
+ * Software based L1D cache flush which is used when microcode providing
+ * the cache control MSR is not loaded.
+ *
+ * The L1D cache is 32 KiB on Nehalem and later microarchitectures, but to
+ * flush it is required to read in 64 KiB because the replacement algorithm
+ * is not exactly LRU. This could be sized at runtime via topology
+ * information but as all relevant affected CPUs have 32KiB L1D cache size
+ * there is no point in doing so.
+ */
+#define L1D_CACHE_ORDER 4
+static void *vmx_l1d_flush_pages;
+
+static void vmx_l1d_flush(struct kvm_vcpu *vcpu)
+{
+	int size = PAGE_SIZE << L1D_CACHE_ORDER;
+
+	/*
+	 * This code is only executed when the the flush mode is 'cond' or
+	 * 'always'
+	 */
+	if (static_branch_likely(&vmx_l1d_flush_cond)) {
+		bool flush_l1d;
+
+		/*
+		 * Clear the per-vcpu flush bit, it gets set again
+		 * either from vcpu_run() or from one of the unsafe
+		 * VMEXIT handlers.
+		 */
+		flush_l1d = vcpu->arch.l1tf_flush_l1d;
+		vcpu->arch.l1tf_flush_l1d = false;
+
+		/*
+		 * Clear the per-cpu flush bit, it gets set again from
+		 * the interrupt handlers.
+		 */
+		flush_l1d |= kvm_get_cpu_l1tf_flush_l1d();
+		kvm_clear_cpu_l1tf_flush_l1d();
+
+		if (!flush_l1d)
+			return;
+	}
+
+	vcpu->stat.l1d_flush++;
+
+	if (static_cpu_has(X86_FEATURE_FLUSH_L1D)) {
+		wrmsrl(MSR_IA32_FLUSH_CMD, L1D_FLUSH);
+		return;
+	}
+
+	asm volatile(
+		/* First ensure the pages are in the TLB */
+		"xorl	%%eax, %%eax\n"
+		".Lpopulate_tlb:\n\t"
+		"movzbl	(%[flush_pages], %%" _ASM_AX "), %%ecx\n\t"
+		"addl	$4096, %%eax\n\t"
+		"cmpl	%%eax, %[size]\n\t"
+		"jne	.Lpopulate_tlb\n\t"
+		"xorl	%%eax, %%eax\n\t"
+		"cpuid\n\t"
+		/* Now fill the cache */
+		"xorl	%%eax, %%eax\n"
+		".Lfill_cache:\n"
+		"movzbl	(%[flush_pages], %%" _ASM_AX "), %%ecx\n\t"
+		"addl	$64, %%eax\n\t"
+		"cmpl	%%eax, %[size]\n\t"
+		"jne	.Lfill_cache\n\t"
+		"lfence\n"
+		:: [flush_pages] "r" (vmx_l1d_flush_pages),
+		    [size] "r" (size)
+		: "eax", "ebx", "ecx", "edx");
+}
+
 static void update_cr8_intercept(struct kvm_vcpu *vcpu, int tpr, int irr)
 {
 	struct vmcs12 *vmcs12 = get_vmcs12(vcpu);
@@ -9949,7 +10190,7 @@ static void atomic_switch_perf_msrs(struct vcpu_vmx *vmx)
 			clear_atomic_switch_msr(vmx, msrs[i].msr);
 		else
 			add_atomic_switch_msr(vmx, msrs[i].msr, msrs[i].guest,
-					msrs[i].host);
+					msrs[i].host, false);
 }
 
 static void vmx_arm_hv_timer(struct kvm_vcpu *vcpu)
@@ -10044,6 +10285,9 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu)
 	evmcs_rsp = static_branch_unlikely(&enable_evmcs) ?
 		(unsigned long)&current_evmcs->host_rsp : 0;
 
+	if (static_branch_unlikely(&vmx_l1d_should_flush))
+		vmx_l1d_flush(vcpu);
+
 	asm(
 		/* Store host registers */
 		"push %%" _ASM_DX "; push %%" _ASM_BP ";"
@@ -10403,10 +10647,37 @@ free_vcpu:
 	return ERR_PTR(err);
 }
 
+#define L1TF_MSG_SMT "L1TF CPU bug present and SMT on, data leak possible. See CVE-2018-3646 and https://www.kernel.org/doc/html/latest/admin-guide/l1tf.html for details.\n"
+#define L1TF_MSG_L1D "L1TF CPU bug present and virtualization mitigation disabled, data leak possible. See CVE-2018-3646 and https://www.kernel.org/doc/html/latest/admin-guide/l1tf.html for details.\n"
+
 static int vmx_vm_init(struct kvm *kvm)
 {
 	if (!ple_gap)
 		kvm->arch.pause_in_guest = true;
+
+	if (boot_cpu_has(X86_BUG_L1TF) && enable_ept) {
+		switch (l1tf_mitigation) {
+		case L1TF_MITIGATION_OFF:
+		case L1TF_MITIGATION_FLUSH_NOWARN:
+			/* 'I explicitly don't care' is set */
+			break;
+		case L1TF_MITIGATION_FLUSH:
+		case L1TF_MITIGATION_FLUSH_NOSMT:
+		case L1TF_MITIGATION_FULL:
+			/*
+			 * Warn upon starting the first VM in a potentially
+			 * insecure environment.
+			 */
+			if (cpu_smt_control == CPU_SMT_ENABLED)
+				pr_warn_once(L1TF_MSG_SMT);
+			if (l1tf_vmx_mitigation == VMENTER_L1D_FLUSH_NEVER)
+				pr_warn_once(L1TF_MSG_L1D);
+			break;
+		case L1TF_MITIGATION_FULL_FORCE:
+			/* Flush is enforced */
+			break;
+		}
+	}
 	return 0;
 }
 
@@ -11260,10 +11531,10 @@ static void prepare_vmcs02_full(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12)
 	 * Set the MSR load/store lists to match L0's settings.
 	 */
 	vmcs_write32(VM_EXIT_MSR_STORE_COUNT, 0);
-	vmcs_write32(VM_EXIT_MSR_LOAD_COUNT, vmx->msr_autoload.nr);
-	vmcs_write64(VM_EXIT_MSR_LOAD_ADDR, __pa(vmx->msr_autoload.host));
-	vmcs_write32(VM_ENTRY_MSR_LOAD_COUNT, vmx->msr_autoload.nr);
-	vmcs_write64(VM_ENTRY_MSR_LOAD_ADDR, __pa(vmx->msr_autoload.guest));
+	vmcs_write32(VM_EXIT_MSR_LOAD_COUNT, vmx->msr_autoload.host.nr);
+	vmcs_write64(VM_EXIT_MSR_LOAD_ADDR, __pa(vmx->msr_autoload.host.val));
+	vmcs_write32(VM_ENTRY_MSR_LOAD_COUNT, vmx->msr_autoload.guest.nr);
+	vmcs_write64(VM_ENTRY_MSR_LOAD_ADDR, __pa(vmx->msr_autoload.guest.val));
 
 	set_cr4_guest_host_mask(vmx);
 
@@ -11899,6 +12170,9 @@ static int nested_vmx_run(struct kvm_vcpu *vcpu, bool launch)
 		return ret;
 	}
 
+	/* Hide L1D cache contents from the nested guest.  */
+	vmx->vcpu.arch.l1tf_flush_l1d = true;
+
 	/*
 	 * If we're entering a halted L2 vcpu and the L2 vcpu won't be woken
 	 * by event injection, halt vcpu.
@@ -12419,8 +12693,8 @@ static void nested_vmx_vmexit(struct kvm_vcpu *vcpu, u32 exit_reason,
 	vmx_segment_cache_clear(vmx);
 
 	/* Update any VMCS fields that might have changed while L2 ran */
-	vmcs_write32(VM_EXIT_MSR_LOAD_COUNT, vmx->msr_autoload.nr);
-	vmcs_write32(VM_ENTRY_MSR_LOAD_COUNT, vmx->msr_autoload.nr);
+	vmcs_write32(VM_EXIT_MSR_LOAD_COUNT, vmx->msr_autoload.host.nr);
+	vmcs_write32(VM_ENTRY_MSR_LOAD_COUNT, vmx->msr_autoload.guest.nr);
 	vmcs_write64(TSC_OFFSET, vcpu->arch.tsc_offset);
 	if (vmx->hv_deadline_tsc == -1)
 		vmcs_clear_bits(PIN_BASED_VM_EXEC_CONTROL,
@@ -13137,6 +13411,51 @@ static struct kvm_x86_ops vmx_x86_ops __ro_after_init = {
 	.enable_smi_window = enable_smi_window,
 };
 
+static void vmx_cleanup_l1d_flush(void)
+{
+	if (vmx_l1d_flush_pages) {
+		free_pages((unsigned long)vmx_l1d_flush_pages, L1D_CACHE_ORDER);
+		vmx_l1d_flush_pages = NULL;
+	}
+	/* Restore state so sysfs ignores VMX */
+	l1tf_vmx_mitigation = VMENTER_L1D_FLUSH_AUTO;
+}
+
+static void vmx_exit(void)
+{
+#ifdef CONFIG_KEXEC_CORE
+	RCU_INIT_POINTER(crash_vmclear_loaded_vmcss, NULL);
+	synchronize_rcu();
+#endif
+
+	kvm_exit();
+
+#if IS_ENABLED(CONFIG_HYPERV)
+	if (static_branch_unlikely(&enable_evmcs)) {
+		int cpu;
+		struct hv_vp_assist_page *vp_ap;
+		/*
+		 * Reset everything to support using non-enlightened VMCS
+		 * access later (e.g. when we reload the module with
+		 * enlightened_vmcs=0)
+		 */
+		for_each_online_cpu(cpu) {
+			vp_ap =	hv_get_vp_assist_page(cpu);
+
+			if (!vp_ap)
+				continue;
+
+			vp_ap->current_nested_vmcs = 0;
+			vp_ap->enlighten_vmentry = 0;
+		}
+
+		static_branch_disable(&enable_evmcs);
+	}
+#endif
+	vmx_cleanup_l1d_flush();
+}
+module_exit(vmx_exit);
+
 static int __init vmx_init(void)
 {
 	int r;
@@ -13171,10 +13490,25 @@ static int __init vmx_init(void)
 #endif
 
 	r = kvm_init(&vmx_x86_ops, sizeof(struct vcpu_vmx),
-                     __alignof__(struct vcpu_vmx), THIS_MODULE);
+		     __alignof__(struct vcpu_vmx), THIS_MODULE);
 	if (r)
 		return r;
 
+	/*
+	 * Must be called after kvm_init() so enable_ept is properly set
+	 * up. Hand the parameter mitigation value in which was stored in
+	 * the pre module init parser. If no parameter was given, it will
+	 * contain 'auto' which will be turned into the default 'cond'
+	 * mitigation mode.
+	 */
+	if (boot_cpu_has(X86_BUG_L1TF)) {
+		r = vmx_setup_l1d_flush(vmentry_l1d_flush_param);
+		if (r) {
+			vmx_exit();
+			return r;
+		}
+	}
+
 #ifdef CONFIG_KEXEC_CORE
 	rcu_assign_pointer(crash_vmclear_loaded_vmcss,
 			   crash_vmclear_local_loaded_vmcss);
@@ -13183,39 +13517,4 @@ static int __init vmx_init(void)
 
 	return 0;
 }
-
-static void __exit vmx_exit(void)
-{
-#ifdef CONFIG_KEXEC_CORE
-	RCU_INIT_POINTER(crash_vmclear_loaded_vmcss, NULL);
-	synchronize_rcu();
-#endif
-
-	kvm_exit();
-
-#if IS_ENABLED(CONFIG_HYPERV)
-	if (static_branch_unlikely(&enable_evmcs)) {
-		int cpu;
-		struct hv_vp_assist_page *vp_ap;
-		/*
-		 * Reset everything to support using non-enlightened VMCS
-		 * access later (e.g. when we reload the module with
-		 * enlightened_vmcs=0)
-		 */
-		for_each_online_cpu(cpu) {
-			vp_ap =	hv_get_vp_assist_page(cpu);
-
-			if (!vp_ap)
-				continue;
-
-			vp_ap->current_nested_vmcs = 0;
-			vp_ap->enlighten_vmentry = 0;
-		}
-
-		static_branch_disable(&enable_evmcs);
-	}
-#endif
-}
-
-module_init(vmx_init)
-module_exit(vmx_exit)
+module_init(vmx_init);
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 2b812b3c5088..a5caa5e5480c 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -195,6 +195,7 @@ struct kvm_stats_debugfs_item debugfs_entries[] = {
 	{ "irq_injections", VCPU_STAT(irq_injections) },
 	{ "nmi_injections", VCPU_STAT(nmi_injections) },
 	{ "req_event", VCPU_STAT(req_event) },
+	{ "l1d_flush", VCPU_STAT(l1d_flush) },
 	{ "mmu_shadow_zapped", VM_STAT(mmu_shadow_zapped) },
 	{ "mmu_pte_write", VM_STAT(mmu_pte_write) },
 	{ "mmu_pte_updated", VM_STAT(mmu_pte_updated) },
@@ -1102,11 +1103,35 @@ static u32 msr_based_features[] = {
 
 static unsigned int num_msr_based_features;
 
+u64 kvm_get_arch_capabilities(void)
+{
+	u64 data;
+
+	rdmsrl_safe(MSR_IA32_ARCH_CAPABILITIES, &data);
+
+	/*
+	 * If we're doing cache flushes (either "always" or "cond")
+	 * we will do one whenever the guest does a vmlaunch/vmresume.
+	 * If an outer hypervisor is doing the cache flush for us
+	 * (VMENTER_L1D_FLUSH_NESTED_VM), we can safely pass that
+	 * capability to the guest too, and if EPT is disabled we're not
+	 * vulnerable.  Overall, only VMENTER_L1D_FLUSH_NEVER will
+	 * require a nested hypervisor to do a flush of its own.
+	 */
+	if (l1tf_vmx_mitigation != VMENTER_L1D_FLUSH_NEVER)
+		data |= ARCH_CAP_SKIP_VMENTRY_L1DFLUSH;
+
+	return data;
+}
+EXPORT_SYMBOL_GPL(kvm_get_arch_capabilities);
+
 static int kvm_get_msr_feature(struct kvm_msr_entry *msr)
 {
 	switch (msr->index) {
-	case MSR_IA32_UCODE_REV:
 	case MSR_IA32_ARCH_CAPABILITIES:
+		msr->data = kvm_get_arch_capabilities();
+		break;
+	case MSR_IA32_UCODE_REV:
 		rdmsrl_safe(msr->index, &msr->data);
 		break;
 	default:
@@ -4876,6 +4901,9 @@ static int emulator_write_std(struct x86_emulate_ctxt *ctxt, gva_t addr, void *v
 int kvm_write_guest_virt_system(struct kvm_vcpu *vcpu, gva_t addr, void *val,
 				unsigned int bytes, struct x86_exception *exception)
 {
+	/* kvm_write_guest_virt_system can pull in tons of pages. */
+	vcpu->arch.l1tf_flush_l1d = true;
+
 	return kvm_write_guest_virt_helper(addr, val, bytes, vcpu,
 					   PFERR_WRITE_MASK, exception);
 }
@@ -6052,6 +6080,8 @@ int x86_emulate_instruction(struct kvm_vcpu *vcpu,
 	bool writeback = true;
 	bool write_fault_to_spt = vcpu->arch.write_fault_to_shadow_pgtable;
 
+	vcpu->arch.l1tf_flush_l1d = true;
+
 	/*
 	 * Clear write_fault_to_shadow_pgtable here to ensure it is
 	 * never reused.
@@ -7581,6 +7611,7 @@ static int vcpu_run(struct kvm_vcpu *vcpu)
 	struct kvm *kvm = vcpu->kvm;
 
 	vcpu->srcu_idx = srcu_read_lock(&kvm->srcu);
+	vcpu->arch.l1tf_flush_l1d = true;
 
 	for (;;) {
 		if (kvm_vcpu_running(vcpu)) {
@@ -8700,6 +8731,7 @@ void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
 
 void kvm_arch_sched_in(struct kvm_vcpu *vcpu, int cpu)
 {
+	vcpu->arch.l1tf_flush_l1d = true;
 	kvm_x86_ops->sched_in(vcpu, cpu);
 }