5 files changed, 60 insertions, 28 deletions

diff --git a/arch/x86/kvm/assigned-dev.c b/arch/x86/kvm/assigned-dev.c
index 6eb5c20..d090ecf 100644
--- a/arch/x86/kvm/assigned-dev.c
+++ b/arch/x86/kvm/assigned-dev.c
@@ -666,7 +666,7 @@ static int probe_sysfs_permissions(struct pci_dev *dev)
 		if (r)
 			return r;
 
-		inode = path.dentry->d_inode;
+		inode = d_backing_inode(path.dentry);
 
 		r = inode_permission(inode, MAY_READ | MAY_WRITE | MAY_ACCESS);
 		path_put(&path);
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index d67206a..629af0f 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -683,8 +683,7 @@ bool kvm_irq_delivery_to_apic_fast(struct kvm *kvm, struct kvm_lapic *src,
 	unsigned long bitmap = 1;
 	struct kvm_lapic **dst;
 	int i;
-	bool ret = false;
-	bool x2apic_ipi = src && apic_x2apic_mode(src);
+	bool ret, x2apic_ipi;
 
 	*r = -1;
 
@@ -696,16 +695,18 @@ bool kvm_irq_delivery_to_apic_fast(struct kvm *kvm, struct kvm_lapic *src,
 	if (irq->shorthand)
 		return false;
 
+	x2apic_ipi = src && apic_x2apic_mode(src);
 	if (irq->dest_id == (x2apic_ipi ? X2APIC_BROADCAST : APIC_BROADCAST))
 		return false;
 
+	ret = true;
 	rcu_read_lock();
 	map = rcu_dereference(kvm->arch.apic_map);
 
-	if (!map)
+	if (!map) {
+		ret = false;
 		goto out;
-
-	ret = true;
+	}
 
 	if (irq->dest_mode == APIC_DEST_PHYSICAL) {
 		if (irq->dest_id >= ARRAY_SIZE(map->phys_map))
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 146f295..d43867c 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -4481,9 +4481,11 @@ static bool kvm_mmu_zap_collapsible_spte(struct kvm *kvm,
 		pfn = spte_to_pfn(*sptep);
 
 		/*
-		 * Only EPT supported for now; otherwise, one would need to
-		 * find out efficiently whether the guest page tables are
-		 * also using huge pages.
+		 * We cannot do huge page mapping for indirect shadow pages,
+		 * which are found on the last rmap (level = 1) when not using
+		 * tdp; such shadow pages are synced with the page table in
+		 * the guest, and the guest page table is using 4K page size
+		 * mapping if the indirect sp has level = 1.
 		 */
 		if (sp->role.direct &&
 			!kvm_is_reserved_pfn(pfn) &&
@@ -4504,19 +4506,12 @@ void kvm_mmu_zap_collapsible_sptes(struct kvm *kvm,
 	bool flush = false;
 	unsigned long *rmapp;
 	unsigned long last_index, index;
-	gfn_t gfn_start, gfn_end;
 
 	spin_lock(&kvm->mmu_lock);
 
-	gfn_start = memslot->base_gfn;
-	gfn_end = memslot->base_gfn + memslot->npages - 1;
-
-	if (gfn_start >= gfn_end)
-		goto out;
-
 	rmapp = memslot->arch.rmap[0];
-	last_index = gfn_to_index(gfn_end, memslot->base_gfn,
-					PT_PAGE_TABLE_LEVEL);
+	last_index = gfn_to_index(memslot->base_gfn + memslot->npages - 1,
+				memslot->base_gfn, PT_PAGE_TABLE_LEVEL);
 
 	for (index = 0; index <= last_index; ++index, ++rmapp) {
 		if (*rmapp)
@@ -4534,7 +4529,6 @@ void kvm_mmu_zap_collapsible_sptes(struct kvm *kvm,
 	if (flush)
 		kvm_flush_remote_tlbs(kvm);
 
-out:
 	spin_unlock(&kvm->mmu_lock);
 }
 
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index f5e8dce..f7b6168 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -3622,8 +3622,16 @@ static void vmx_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3)
 
 static int vmx_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
 {
-	unsigned long hw_cr4 = cr4 | (to_vmx(vcpu)->rmode.vm86_active ?
-		    KVM_RMODE_VM_CR4_ALWAYS_ON : KVM_PMODE_VM_CR4_ALWAYS_ON);
+	/*
+	 * Pass through host's Machine Check Enable value to hw_cr4, which
+	 * is in force while we are in guest mode.  Do not let guests control
+	 * this bit, even if host CR4.MCE == 0.
+	 */
+	unsigned long hw_cr4 =
+		(cr4_read_shadow() & X86_CR4_MCE) |
+		(cr4 & ~X86_CR4_MCE) |
+		(to_vmx(vcpu)->rmode.vm86_active ?
+		 KVM_RMODE_VM_CR4_ALWAYS_ON : KVM_PMODE_VM_CR4_ALWAYS_ON);
 
 	if (cr4 & X86_CR4_VMXE) {
 		/*
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index e1a8126..c73efcd 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -1669,12 +1669,28 @@ static int kvm_guest_time_update(struct kvm_vcpu *v)
 		&guest_hv_clock, sizeof(guest_hv_clock))))
 		return 0;
 
-	/*
-	 * The interface expects us to write an even number signaling that the
-	 * update is finished. Since the guest won't see the intermediate
-	 * state, we just increase by 2 at the end.
+	/* This VCPU is paused, but it's legal for a guest to read another
+	 * VCPU's kvmclock, so we really have to follow the specification where
+	 * it says that version is odd if data is being modified, and even after
+	 * it is consistent.
+	 *
+	 * Version field updates must be kept separate.  This is because
+	 * kvm_write_guest_cached might use a "rep movs" instruction, and
+	 * writes within a string instruction are weakly ordered.  So there
+	 * are three writes overall.
+	 *
+	 * As a small optimization, only write the version field in the first
+	 * and third write.  The vcpu->pv_time cache is still valid, because the
+	 * version field is the first in the struct.
 	 */
-	vcpu->hv_clock.version = guest_hv_clock.version + 2;
+	BUILD_BUG_ON(offsetof(struct pvclock_vcpu_time_info, version) != 0);
+
+	vcpu->hv_clock.version = guest_hv_clock.version + 1;
+	kvm_write_guest_cached(v->kvm, &vcpu->pv_time,
+				&vcpu->hv_clock,
+				sizeof(vcpu->hv_clock.version));
+
+	smp_wmb();
 
 	/* retain PVCLOCK_GUEST_STOPPED if set in guest copy */
 	pvclock_flags = (guest_hv_clock.flags & PVCLOCK_GUEST_STOPPED);
@@ -1695,6 +1711,13 @@ static int kvm_guest_time_update(struct kvm_vcpu *v)
 	kvm_write_guest_cached(v->kvm, &vcpu->pv_time,
 				&vcpu->hv_clock,
 				sizeof(vcpu->hv_clock));
+
+	smp_wmb();
+
+	vcpu->hv_clock.version++;
+	kvm_write_guest_cached(v->kvm, &vcpu->pv_time,
+				&vcpu->hv_clock,
+				sizeof(vcpu->hv_clock.version));
 	return 0;
 }
 
@@ -5799,7 +5822,6 @@ int kvm_arch_init(void *opaque)
 	kvm_set_mmio_spte_mask();
 
 	kvm_x86_ops = ops;
-	kvm_init_msr_list();
 
 	kvm_mmu_set_mask_ptes(PT_USER_MASK, PT_ACCESSED_MASK,
 			PT_DIRTY_MASK, PT64_NX_MASK, 0);
@@ -7253,7 +7275,14 @@ void kvm_arch_hardware_disable(void)
 
 int kvm_arch_hardware_setup(void)
 {
-	return kvm_x86_ops->hardware_setup();
+	int r;
+
+	r = kvm_x86_ops->hardware_setup();
+	if (r != 0)
+		return r;
+
+	kvm_init_msr_list();
+	return 0;
 }
 
 void kvm_arch_hardware_unsetup(void)