Merge tag 'kvm-x86-mmu-6.7' of https://github.com/kvm-x86/linux into HEAD

KVM x86 MMU changes for 6.7:

 - Clean up code that deals with honoring guest MTRRs when the VM has
   non-coherent DMA and host MTRRs are ignored, i.e. EPT is enabled.

 - Zap EPT entries when non-coherent DMA assignment stops/start to prevent
   using stale entries with the wrong memtype.

 - Don't ignore guest PAT for CR0.CD=1 && KVM_X86_QUIRK_CD_NW_CLEARED=y, as
   there's zero reason to ignore guest PAT if the effective MTRR memtype is WB.
   This will also allow for future optimizations of handling guest MTRR updates
   for VMs with non-coherent DMA and the quirk enabled.

 - Harden the fast page fault path to guard against encountering an invalid
   root when walking SPTEs.

5 files changed, 55 insertions, 21 deletions

diff --git a/arch/x86/kvm/mmu.h b/arch/x86/kvm/mmu.h
index 253fb2093d5d..bb8c86eefac0 100644
--- a/arch/x86/kvm/mmu.h
+++ b/arch/x86/kvm/mmu.h
@@ -237,6 +237,13 @@ static inline u8 permission_fault(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu,
 	return -(u32)fault & errcode;
 }
 
+bool __kvm_mmu_honors_guest_mtrrs(bool vm_has_noncoherent_dma);
+
+static inline bool kvm_mmu_honors_guest_mtrrs(struct kvm *kvm)
+{
+	return __kvm_mmu_honors_guest_mtrrs(kvm_arch_has_noncoherent_dma(kvm));
+}
+
 void kvm_zap_gfn_range(struct kvm *kvm, gfn_t gfn_start, gfn_t gfn_end);
 
 int kvm_arch_write_log_dirty(struct kvm_vcpu *vcpu);
diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
index f7901cb4d2fa..b0f01d605617 100644
--- a/arch/x86/kvm/mmu/mmu.c
+++ b/arch/x86/kvm/mmu/mmu.c
@@ -3425,8 +3425,8 @@ static int fast_page_fault(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault)
 {
 	struct kvm_mmu_page *sp;
 	int ret = RET_PF_INVALID;
-	u64 spte = 0ull;
-	u64 *sptep = NULL;
+	u64 spte;
+	u64 *sptep;
 	uint retry_count = 0;
 
 	if (!page_fault_can_be_fast(fault))
@@ -3442,6 +3442,14 @@ static int fast_page_fault(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault)
 		else
 			sptep = fast_pf_get_last_sptep(vcpu, fault->addr, &spte);
 
+		/*
+		 * It's entirely possible for the mapping to have been zapped
+		 * by a different task, but the root page should always be
+		 * available as the vCPU holds a reference to its root(s).
+		 */
+		if (WARN_ON_ONCE(!sptep))
+			spte = REMOVED_SPTE;
+
 		if (!is_shadow_present_pte(spte))
 			break;
 
@@ -4479,21 +4487,28 @@ out_unlock:
 }
 #endif
 
-int kvm_tdp_page_fault(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault)
+bool __kvm_mmu_honors_guest_mtrrs(bool vm_has_noncoherent_dma)
 {
 	/*
-	 * If the guest's MTRRs may be used to compute the "real" memtype,
-	 * restrict the mapping level to ensure KVM uses a consistent memtype
-	 * across the entire mapping.  If the host MTRRs are ignored by TDP
-	 * (shadow_memtype_mask is non-zero), and the VM has non-coherent DMA
-	 * (DMA doesn't snoop CPU caches), KVM's ABI is to honor the memtype
-	 * from the guest's MTRRs so that guest accesses to memory that is
-	 * DMA'd aren't cached against the guest's wishes.
+	 * If host MTRRs are ignored (shadow_memtype_mask is non-zero), and the
+	 * VM has non-coherent DMA (DMA doesn't snoop CPU caches), KVM's ABI is
+	 * to honor the memtype from the guest's MTRRs so that guest accesses
+	 * to memory that is DMA'd aren't cached against the guest's wishes.
 	 *
 	 * Note, KVM may still ultimately ignore guest MTRRs for certain PFNs,
 	 * e.g. KVM will force UC memtype for host MMIO.
 	 */
-	if (shadow_memtype_mask && kvm_arch_has_noncoherent_dma(vcpu->kvm)) {
+	return vm_has_noncoherent_dma && shadow_memtype_mask;
+}
+
+int kvm_tdp_page_fault(struct kvm_vcpu *vcpu, struct kvm_page_fault *fault)
+{
+	/*
+	 * If the guest's MTRRs may be used to compute the "real" memtype,
+	 * restrict the mapping level to ensure KVM uses a consistent memtype
+	 * across the entire mapping.
+	 */
+	if (kvm_mmu_honors_guest_mtrrs(vcpu->kvm)) {
 		for ( ; fault->max_level > PG_LEVEL_4K; --fault->max_level) {
 			int page_num = KVM_PAGES_PER_HPAGE(fault->max_level);
 			gfn_t base = gfn_round_for_level(fault->gfn,
diff --git a/arch/x86/kvm/mtrr.c b/arch/x86/kvm/mtrr.c
index 3eb6e7f47e96..a67c28a56417 100644
--- a/arch/x86/kvm/mtrr.c
+++ b/arch/x86/kvm/mtrr.c
@@ -320,7 +320,7 @@ static void update_mtrr(struct kvm_vcpu *vcpu, u32 msr)
 	struct kvm_mtrr *mtrr_state = &vcpu->arch.mtrr_state;
 	gfn_t start, end;
 
-	if (!tdp_enabled || !kvm_arch_has_noncoherent_dma(vcpu->kvm))
+	if (!kvm_mmu_honors_guest_mtrrs(vcpu->kvm))
 		return;
 
 	if (!mtrr_is_enabled(mtrr_state) && msr != MSR_MTRRdefType)
diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index c721a6785909..610e37e215be 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -7579,8 +7579,6 @@ static int vmx_vm_init(struct kvm *kvm)
 
 static u8 vmx_get_mt_mask(struct kvm_vcpu *vcpu, gfn_t gfn, bool is_mmio)
 {
-	u8 cache;
-
 	/* We wanted to honor guest CD/MTRR/PAT, but doing so could result in
 	 * memory aliases with conflicting memory types and sometimes MCEs.
 	 * We have to be careful as to what are honored and when.
@@ -7607,11 +7605,10 @@ static u8 vmx_get_mt_mask(struct kvm_vcpu *vcpu, gfn_t gfn, bool is_mmio)
 
 	if (kvm_read_cr0_bits(vcpu, X86_CR0_CD)) {
 		if (kvm_check_has_quirk(vcpu->kvm, KVM_X86_QUIRK_CD_NW_CLEARED))
-			cache = MTRR_TYPE_WRBACK;
+			return MTRR_TYPE_WRBACK << VMX_EPT_MT_EPTE_SHIFT;
 		else
-			cache = MTRR_TYPE_UNCACHABLE;
-
-		return (cache << VMX_EPT_MT_EPTE_SHIFT) | VMX_EPT_IPAT_BIT;
+			return (MTRR_TYPE_UNCACHABLE << VMX_EPT_MT_EPTE_SHIFT) |
+				VMX_EPT_IPAT_BIT;
 	}
 
 	return kvm_mtrr_get_guest_memory_type(vcpu, gfn) << VMX_EPT_MT_EPTE_SHIFT;
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 4937aa50d40b..3584363b8fa6 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -962,7 +962,7 @@ void kvm_post_set_cr0(struct kvm_vcpu *vcpu, unsigned long old_cr0, unsigned lon
 		kvm_mmu_reset_context(vcpu);
 
 	if (((cr0 ^ old_cr0) & X86_CR0_CD) &&
-	    kvm_arch_has_noncoherent_dma(vcpu->kvm) &&
+	    kvm_mmu_honors_guest_mtrrs(vcpu->kvm) &&
 	    !kvm_check_has_quirk(vcpu->kvm, KVM_X86_QUIRK_CD_NW_CLEARED))
 		kvm_zap_gfn_range(vcpu->kvm, 0, ~0ULL);
 }
@@ -13313,15 +13313,30 @@ bool noinstr kvm_arch_has_assigned_device(struct kvm *kvm)
 }
 EXPORT_SYMBOL_GPL(kvm_arch_has_assigned_device);
 
+static void kvm_noncoherent_dma_assignment_start_or_stop(struct kvm *kvm)
+{
+	/*
+	 * Non-coherent DMA assignment and de-assignment will affect
+	 * whether KVM honors guest MTRRs and cause changes in memtypes
+	 * in TDP.
+	 * So, pass %true unconditionally to indicate non-coherent DMA was,
+	 * or will be involved, and that zapping SPTEs might be necessary.
+	 */
+	if (__kvm_mmu_honors_guest_mtrrs(true))
+		kvm_zap_gfn_range(kvm, gpa_to_gfn(0), gpa_to_gfn(~0ULL));
+}
+
 void kvm_arch_register_noncoherent_dma(struct kvm *kvm)
 {
-	atomic_inc(&kvm->arch.noncoherent_dma_count);
+	if (atomic_inc_return(&kvm->arch.noncoherent_dma_count) == 1)
+		kvm_noncoherent_dma_assignment_start_or_stop(kvm);
 }
 EXPORT_SYMBOL_GPL(kvm_arch_register_noncoherent_dma);
 
 void kvm_arch_unregister_noncoherent_dma(struct kvm *kvm)
 {
-	atomic_dec(&kvm->arch.noncoherent_dma_count);
+	if (!atomic_dec_return(&kvm->arch.noncoherent_dma_count))
+		kvm_noncoherent_dma_assignment_start_or_stop(kvm);
 }
 EXPORT_SYMBOL_GPL(kvm_arch_unregister_noncoherent_dma);