Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm

Pull KVM fixes from Radim Krčmář:
 "ARM:
   - Fix a problem with GICv3 userspace save/restore
   - Clarify GICv2 userspace save/restore ABI
   - Be more careful in clearing GIC LRs
   - Add missing synchronization primitive to our MMU handling code

  PPC:
   - Check for a NULL return from kzalloc

  s390:
   - Prevent translation exception errors on valid page tables for the
     instruction-exection-protection support

  x86:
   - Fix Page-Modification Logging when running a nested guest"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
  KVM: PPC: Book3S HV: Check for kmalloc errors in ioctl
  KVM: nVMX: initialize PML fields in vmcs02
  KVM: nVMX: do not leak PML full vmexit to L1
  KVM: arm/arm64: vgic: Fix GICC_PMR uaccess on GICv3 and clarify ABI
  KVM: arm64: Ensure LRs are clear when they should be
  kvm: arm/arm64: Fix locking for kvm_free_stage2_pgd
  KVM: s390: remove change-recording override support
  arm/arm64: KVM: Take mmap_sem in kvm_arch_prepare_memory_region
  arm/arm64: KVM: Take mmap_sem in stage2_unmap_vm

12 files changed, 120 insertions, 15 deletions

diff --git a/Documentation/virtual/kvm/devices/arm-vgic.txt b/Documentation/virtual/kvm/devices/arm-vgic.txt
index 76e61c883347..b2f60ca8b60c 100644
--- a/Documentation/virtual/kvm/devices/arm-vgic.txt
+++ b/Documentation/virtual/kvm/devices/arm-vgic.txt
@@ -83,6 +83,12 @@ Groups:
 
     Bits for undefined preemption levels are RAZ/WI.
 
+    For historical reasons and to provide ABI compatibility with userspace we
+    export the GICC_PMR register in the format of the GICH_VMCR.VMPriMask
+    field in the lower 5 bits of a word, meaning that userspace must always
+    use the lower 5 bits to communicate with the KVM device and must shift the
+    value left by 3 places to obtain the actual priority mask level.
+
   Limitations:
     - Priorities are not implemented, and registers are RAZ/WI
     - Currently only implemented for KVM_DEV_TYPE_ARM_VGIC_V2.
diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c
index 96dba7cd8be7..314eb6abe1ff 100644
--- a/arch/arm/kvm/arm.c
+++ b/arch/arm/kvm/arm.c
@@ -1124,6 +1124,9 @@ static void cpu_hyp_reinit(void)
 		if (__hyp_get_vectors() == hyp_default_vectors)
 			cpu_init_hyp_mode(NULL);
 	}
+
+	if (vgic_present)
+		kvm_vgic_init_cpu_hardware();
 }
 
 static void cpu_hyp_reset(void)
diff --git a/arch/arm/kvm/mmu.c b/arch/arm/kvm/mmu.c
index 962616fd4ddd..582a972371cf 100644
--- a/arch/arm/kvm/mmu.c
+++ b/arch/arm/kvm/mmu.c
@@ -292,11 +292,18 @@ static void unmap_stage2_range(struct kvm *kvm, phys_addr_t start, u64 size)
 	phys_addr_t addr = start, end = start + size;
 	phys_addr_t next;
 
+	assert_spin_locked(&kvm->mmu_lock);
 	pgd = kvm->arch.pgd + stage2_pgd_index(addr);
 	do {
 		next = stage2_pgd_addr_end(addr, end);
 		if (!stage2_pgd_none(*pgd))
 			unmap_stage2_puds(kvm, pgd, addr, next);
+		/*
+		 * If the range is too large, release the kvm->mmu_lock
+		 * to prevent starvation and lockup detector warnings.
+		 */
+		if (next != end)
+			cond_resched_lock(&kvm->mmu_lock);
 	} while (pgd++, addr = next, addr != end);
 }
 
@@ -803,6 +810,7 @@ void stage2_unmap_vm(struct kvm *kvm)
 	int idx;
 
 	idx = srcu_read_lock(&kvm->srcu);
+	down_read(&current->mm->mmap_sem);
 	spin_lock(&kvm->mmu_lock);
 
 	slots = kvm_memslots(kvm);
@@ -810,6 +818,7 @@ void stage2_unmap_vm(struct kvm *kvm)
 		stage2_unmap_memslot(kvm, memslot);
 
 	spin_unlock(&kvm->mmu_lock);
+	up_read(&current->mm->mmap_sem);
 	srcu_read_unlock(&kvm->srcu, idx);
 }
 
@@ -829,7 +838,10 @@ void kvm_free_stage2_pgd(struct kvm *kvm)
 	if (kvm->arch.pgd == NULL)
 		return;
 
+	spin_lock(&kvm->mmu_lock);
 	unmap_stage2_range(kvm, 0, KVM_PHYS_SIZE);
+	spin_unlock(&kvm->mmu_lock);
+
 	/* Free the HW pgd, one page at a time */
 	free_pages_exact(kvm->arch.pgd, S2_PGD_SIZE);
 	kvm->arch.pgd = NULL;
@@ -1801,6 +1813,7 @@ int kvm_arch_prepare_memory_region(struct kvm *kvm,
 	    (KVM_PHYS_SIZE >> PAGE_SHIFT))
 		return -EFAULT;
 
+	down_read(&current->mm->mmap_sem);
 	/*
 	 * A memory region could potentially cover multiple VMAs, and any holes
 	 * between them, so iterate over all of them to find out if we can map
@@ -1844,8 +1857,10 @@ int kvm_arch_prepare_memory_region(struct kvm *kvm,
 			pa += vm_start - vma->vm_start;
 
 			/* IO region dirty page logging not allowed */
-			if (memslot->flags & KVM_MEM_LOG_DIRTY_PAGES)
-				return -EINVAL;
+			if (memslot->flags & KVM_MEM_LOG_DIRTY_PAGES) {
+				ret = -EINVAL;
+				goto out;
+			}
 
 			ret = kvm_phys_addr_ioremap(kvm, gpa, pa,
 						    vm_end - vm_start,
@@ -1857,7 +1872,7 @@ int kvm_arch_prepare_memory_region(struct kvm *kvm,
 	} while (hva < reg_end);
 
 	if (change == KVM_MR_FLAGS_ONLY)
-		return ret;
+		goto out;
 
 	spin_lock(&kvm->mmu_lock);
 	if (ret)
@@ -1865,6 +1880,8 @@ int kvm_arch_prepare_memory_region(struct kvm *kvm,
 	else
 		stage2_flush_memslot(kvm, memslot);
 	spin_unlock(&kvm->mmu_lock);
+out:
+	up_read(&current->mm->mmap_sem);
 	return ret;
 }
 
diff --git a/arch/powerpc/kvm/book3s_64_mmu_hv.c b/arch/powerpc/kvm/book3s_64_mmu_hv.c
index 8c68145ba1bd..710e491206ed 100644
--- a/arch/powerpc/kvm/book3s_64_mmu_hv.c
+++ b/arch/powerpc/kvm/book3s_64_mmu_hv.c
@@ -1487,6 +1487,10 @@ long kvm_vm_ioctl_resize_hpt_prepare(struct kvm *kvm,
 	/* start new resize */
 
 	resize = kzalloc(sizeof(*resize), GFP_KERNEL);
+	if (!resize) {
+		ret = -ENOMEM;
+		goto out;
+	}
 	resize->order = shift;
 	resize->kvm = kvm;
 	INIT_WORK(&resize->work, resize_hpt_prepare_work);
diff --git a/arch/s390/kvm/gaccess.c b/arch/s390/kvm/gaccess.c
index d55c829a5944..ddbffb715b40 100644
--- a/arch/s390/kvm/gaccess.c
+++ b/arch/s390/kvm/gaccess.c
@@ -168,8 +168,7 @@ union page_table_entry {
 		unsigned long z  : 1; /* Zero Bit */
 		unsigned long i  : 1; /* Page-Invalid Bit */
 		unsigned long p  : 1; /* DAT-Protection Bit */
-		unsigned long co : 1; /* Change-Recording Override */
-		unsigned long	 : 8;
+		unsigned long	 : 9;
 	};
 };
 
@@ -745,8 +744,6 @@ static unsigned long guest_translate(struct kvm_vcpu *vcpu, unsigned long gva,
 		return PGM_PAGE_TRANSLATION;
 	if (pte.z)
 		return PGM_TRANSLATION_SPEC;
-	if (pte.co && !edat1)
-		return PGM_TRANSLATION_SPEC;
 	dat_protection |= pte.p;
 	raddr.pfra = pte.pfra;
 real_address:
@@ -1182,7 +1179,7 @@ int kvm_s390_shadow_fault(struct kvm_vcpu *vcpu, struct gmap *sg,
 		rc = gmap_read_table(sg->parent, pgt + vaddr.px * 8, &pte.val);
 	if (!rc && pte.i)
 		rc = PGM_PAGE_TRANSLATION;
-	if (!rc && (pte.z || (pte.co && sg->edat_level < 1)))
+	if (!rc && pte.z)
 		rc = PGM_TRANSLATION_SPEC;
 shadow_page:
 	pte.p |= dat_protection;
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 2ee00dbbbd51..259e9b28ccf8 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -8198,6 +8198,9 @@ static bool nested_vmx_exit_handled(struct kvm_vcpu *vcpu)
 		return nested_cpu_has2(vmcs12, SECONDARY_EXEC_XSAVES);
 	case EXIT_REASON_PREEMPTION_TIMER:
 		return false;
+	case EXIT_REASON_PML_FULL:
+		/* We don't expose PML support to L1. */
+		return false;
 	default:
 		return true;
 	}
@@ -10267,6 +10270,18 @@ static int prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12,
 
 	}
 
+	if (enable_pml) {
+		/*
+		 * Conceptually we want to copy the PML address and index from
+		 * vmcs01 here, and then back to vmcs01 on nested vmexit. But,
+		 * since we always flush the log on each vmexit, this happens
+		 * to be equivalent to simply resetting the fields in vmcs02.
+		 */
+		ASSERT(vmx->pml_pg);
+		vmcs_write64(PML_ADDRESS, page_to_phys(vmx->pml_pg));
+		vmcs_write16(GUEST_PML_INDEX, PML_ENTITY_NUM - 1);
+	}
+
 	if (nested_cpu_has_ept(vmcs12)) {
 		kvm_mmu_unload(vcpu);
 		nested_ept_init_mmu_context(vcpu);
diff --git a/include/kvm/arm_vgic.h b/include/kvm/arm_vgic.h
index b72dd2ad5f44..c0b3d999c266 100644
--- a/include/kvm/arm_vgic.h
+++ b/include/kvm/arm_vgic.h
@@ -295,6 +295,7 @@ void kvm_vgic_vcpu_early_init(struct kvm_vcpu *vcpu);
 void kvm_vgic_vcpu_destroy(struct kvm_vcpu *vcpu);
 int kvm_vgic_map_resources(struct kvm *kvm);
 int kvm_vgic_hyp_init(void);
+void kvm_vgic_init_cpu_hardware(void);
 
 int kvm_vgic_inject_irq(struct kvm *kvm, int cpuid, unsigned int intid,
 			bool level);
diff --git a/include/linux/irqchip/arm-gic.h b/include/linux/irqchip/arm-gic.h
index eafc965b3eb8..dc30f3d057eb 100644
--- a/include/linux/irqchip/arm-gic.h
+++ b/include/linux/irqchip/arm-gic.h
@@ -96,6 +96,9 @@
 #define GICH_MISR_EOI			(1 << 0)
 #define GICH_MISR_U			(1 << 1)
 
+#define GICV_PMR_PRIORITY_SHIFT		3
+#define GICV_PMR_PRIORITY_MASK		(0x1f << GICV_PMR_PRIORITY_SHIFT)
+
 #ifndef __ASSEMBLY__
 
 #include <linux/irqdomain.h>
diff --git a/virt/kvm/arm/vgic/vgic-init.c b/virt/kvm/arm/vgic/vgic-init.c
index 276139a24e6f..702f8108608d 100644
--- a/virt/kvm/arm/vgic/vgic-init.c
+++ b/virt/kvm/arm/vgic/vgic-init.c
@@ -392,6 +392,25 @@ static irqreturn_t vgic_maintenance_handler(int irq, void *data)
 }
 
 /**
+ * kvm_vgic_init_cpu_hardware - initialize the GIC VE hardware
+ *
+ * For a specific CPU, initialize the GIC VE hardware.
+ */
+void kvm_vgic_init_cpu_hardware(void)
+{
+	BUG_ON(preemptible());
+
+	/*
+	 * We want to make sure the list registers start out clear so that we
+	 * only have the program the used registers.
+	 */
+	if (kvm_vgic_global_state.type == VGIC_V2)
+		vgic_v2_init_lrs();
+	else
+		kvm_call_hyp(__vgic_v3_init_lrs);
+}
+
+/**
  * kvm_vgic_hyp_init: populates the kvm_vgic_global_state variable
  * according to the host GIC model. Accordingly calls either
  * vgic_v2/v3_probe which registers the KVM_DEVICE that can be
diff --git a/virt/kvm/arm/vgic/vgic-mmio-v2.c b/virt/kvm/arm/vgic/vgic-mmio-v2.c
index a3ad7ff95c9b..0a4283ed9aa7 100644
--- a/virt/kvm/arm/vgic/vgic-mmio-v2.c
+++ b/virt/kvm/arm/vgic/vgic-mmio-v2.c
@@ -229,7 +229,15 @@ static unsigned long vgic_mmio_read_vcpuif(struct kvm_vcpu *vcpu,
 		val = vmcr.ctlr;
 		break;
 	case GIC_CPU_PRIMASK:
-		val = vmcr.pmr;
+		/*
+		 * Our KVM_DEV_TYPE_ARM_VGIC_V2 device ABI exports the
+		 * the PMR field as GICH_VMCR.VMPriMask rather than
+		 * GICC_PMR.Priority, so we expose the upper five bits of
+		 * priority mask to userspace using the lower bits in the
+		 * unsigned long.
+		 */
+		val = (vmcr.pmr & GICV_PMR_PRIORITY_MASK) >>
+			GICV_PMR_PRIORITY_SHIFT;
 		break;
 	case GIC_CPU_BINPOINT:
 		val = vmcr.bpr;
@@ -262,7 +270,15 @@ static void vgic_mmio_write_vcpuif(struct kvm_vcpu *vcpu,
 		vmcr.ctlr = val;
 		break;
 	case GIC_CPU_PRIMASK:
-		vmcr.pmr = val;
+		/*
+		 * Our KVM_DEV_TYPE_ARM_VGIC_V2 device ABI exports the
+		 * the PMR field as GICH_VMCR.VMPriMask rather than
+		 * GICC_PMR.Priority, so we expose the upper five bits of
+		 * priority mask to userspace using the lower bits in the
+		 * unsigned long.
+		 */
+		vmcr.pmr = (val << GICV_PMR_PRIORITY_SHIFT) &
+			GICV_PMR_PRIORITY_MASK;
 		break;
 	case GIC_CPU_BINPOINT:
 		vmcr.bpr = val;
diff --git a/virt/kvm/arm/vgic/vgic-v2.c b/virt/kvm/arm/vgic/vgic-v2.c
index b834ecdf3225..b637d9c7afe3 100644
--- a/virt/kvm/arm/vgic/vgic-v2.c
+++ b/virt/kvm/arm/vgic/vgic-v2.c
@@ -36,6 +36,21 @@ static unsigned long *u64_to_bitmask(u64 *val)
 	return (unsigned long *)val;
 }
 
+static inline void vgic_v2_write_lr(int lr, u32 val)
+{
+	void __iomem *base = kvm_vgic_global_state.vctrl_base;
+
+	writel_relaxed(val, base + GICH_LR0 + (lr * 4));
+}
+
+void vgic_v2_init_lrs(void)
+{
+	int i;
+
+	for (i = 0; i < kvm_vgic_global_state.nr_lr; i++)
+		vgic_v2_write_lr(i, 0);
+}
+
 void vgic_v2_process_maintenance(struct kvm_vcpu *vcpu)
 {
 	struct vgic_v2_cpu_if *cpuif = &vcpu->arch.vgic_cpu.vgic_v2;
@@ -191,8 +206,8 @@ void vgic_v2_set_vmcr(struct kvm_vcpu *vcpu, struct vgic_vmcr *vmcrp)
 		GICH_VMCR_ALIAS_BINPOINT_MASK;
 	vmcr |= (vmcrp->bpr << GICH_VMCR_BINPOINT_SHIFT) &
 		GICH_VMCR_BINPOINT_MASK;
-	vmcr |= (vmcrp->pmr << GICH_VMCR_PRIMASK_SHIFT) &
-		GICH_VMCR_PRIMASK_MASK;
+	vmcr |= ((vmcrp->pmr >> GICV_PMR_PRIORITY_SHIFT) <<
+		 GICH_VMCR_PRIMASK_SHIFT) & GICH_VMCR_PRIMASK_MASK;
 
 	vcpu->arch.vgic_cpu.vgic_v2.vgic_vmcr = vmcr;
 }
@@ -207,8 +222,8 @@ void vgic_v2_get_vmcr(struct kvm_vcpu *vcpu, struct vgic_vmcr *vmcrp)
 			GICH_VMCR_ALIAS_BINPOINT_SHIFT;
 	vmcrp->bpr  = (vmcr & GICH_VMCR_BINPOINT_MASK) >>
 			GICH_VMCR_BINPOINT_SHIFT;
-	vmcrp->pmr  = (vmcr & GICH_VMCR_PRIMASK_MASK) >>
-			GICH_VMCR_PRIMASK_SHIFT;
+	vmcrp->pmr  = ((vmcr & GICH_VMCR_PRIMASK_MASK) >>
+			GICH_VMCR_PRIMASK_SHIFT) << GICV_PMR_PRIORITY_SHIFT;
 }
 
 void vgic_v2_enable(struct kvm_vcpu *vcpu)
diff --git a/virt/kvm/arm/vgic/vgic.h b/virt/kvm/arm/vgic/vgic.h
index db28f7cadab2..6cf557e9f718 100644
--- a/virt/kvm/arm/vgic/vgic.h
+++ b/virt/kvm/arm/vgic/vgic.h
@@ -81,11 +81,18 @@ static inline bool irq_is_pending(struct vgic_irq *irq)
 		return irq->pending_latch || irq->line_level;
 }
 
+/*
+ * This struct provides an intermediate representation of the fields contained
+ * in the GICH_VMCR and ICH_VMCR registers, such that code exporting the GIC
+ * state to userspace can generate either GICv2 or GICv3 CPU interface
+ * registers regardless of the hardware backed GIC used.
+ */
 struct vgic_vmcr {
 	u32	ctlr;
 	u32	abpr;
 	u32	bpr;
-	u32	pmr;
+	u32	pmr;  /* Priority mask field in the GICC_PMR and
+		       * ICC_PMR_EL1 priority field format */
 	/* Below member variable are valid only for GICv3 */
 	u32	grpen0;
 	u32	grpen1;
@@ -130,6 +137,8 @@ int vgic_v2_map_resources(struct kvm *kvm);
 int vgic_register_dist_iodev(struct kvm *kvm, gpa_t dist_base_address,
 			     enum vgic_type);
 
+void vgic_v2_init_lrs(void);
+
 static inline void vgic_get_irq_kref(struct vgic_irq *irq)
 {
 	if (irq->intid < VGIC_MIN_LPI)