Linux-libre 4.8.2-gnupck-4.8.2-gnu

12 files changed, 2220 insertions, 118 deletions

diff --git a/virt/kvm/arm/vgic/vgic-init.c b/virt/kvm/arm/vgic/vgic-init.c
index a1442f7c9..83777c1cb 100644
--- a/virt/kvm/arm/vgic/vgic-init.c
+++ b/virt/kvm/arm/vgic/vgic-init.c
@@ -73,12 +73,8 @@ int kvm_vgic_create(struct kvm *kvm, u32 type)
 	int i, vcpu_lock_idx = -1, ret;
 	struct kvm_vcpu *vcpu;
 
-	mutex_lock(&kvm->lock);
-
-	if (irqchip_in_kernel(kvm)) {
-		ret = -EEXIST;
-		goto out;
-	}
+	if (irqchip_in_kernel(kvm))
+		return -EEXIST;
 
 	/*
 	 * This function is also called by the KVM_CREATE_IRQCHIP handler,
@@ -87,10 +83,8 @@ int kvm_vgic_create(struct kvm *kvm, u32 type)
 	 * the proper checks already.
 	 */
 	if (type == KVM_DEV_TYPE_ARM_VGIC_V2 &&
-		!kvm_vgic_global_state.can_emulate_gicv2) {
-		ret = -ENODEV;
-		goto out;
-	}
+		!kvm_vgic_global_state.can_emulate_gicv2)
+		return -ENODEV;
 
 	/*
 	 * Any time a vcpu is run, vcpu_load is called which tries to grab the
@@ -138,9 +132,6 @@ out_unlock:
 		vcpu = kvm_get_vcpu(kvm, vcpu_lock_idx);
 		mutex_unlock(&vcpu->mutex);
 	}
-
-out:
-	mutex_unlock(&kvm->lock);
 	return ret;
 }
 
@@ -157,6 +148,9 @@ static int kvm_vgic_dist_init(struct kvm *kvm, unsigned int nr_spis)
 	struct kvm_vcpu *vcpu0 = kvm_get_vcpu(kvm, 0);
 	int i;
 
+	INIT_LIST_HEAD(&dist->lpi_list_head);
+	spin_lock_init(&dist->lpi_list_lock);
+
 	dist->spis = kcalloc(nr_spis, sizeof(struct vgic_irq), GFP_KERNEL);
 	if (!dist->spis)
 		return  -ENOMEM;
@@ -177,6 +171,7 @@ static int kvm_vgic_dist_init(struct kvm *kvm, unsigned int nr_spis)
 		spin_lock_init(&irq->irq_lock);
 		irq->vcpu = NULL;
 		irq->target_vcpu = vcpu0;
+		kref_init(&irq->refcount);
 		if (dist->vgic_model == KVM_DEV_TYPE_ARM_VGIC_V2)
 			irq->targets = 0;
 		else
@@ -211,6 +206,7 @@ static void kvm_vgic_vcpu_init(struct kvm_vcpu *vcpu)
 		irq->vcpu = NULL;
 		irq->target_vcpu = vcpu;
 		irq->targets = 1U << vcpu->vcpu_id;
+		kref_init(&irq->refcount);
 		if (vgic_irq_is_sgi(i)) {
 			/* SGIs */
 			irq->enabled = 1;
@@ -253,9 +249,16 @@ int vgic_init(struct kvm *kvm)
 	if (ret)
 		goto out;
 
+	if (vgic_has_its(kvm))
+		dist->msis_require_devid = true;
+
 	kvm_for_each_vcpu(i, vcpu, kvm)
 		kvm_vgic_vcpu_init(vcpu);
 
+	ret = kvm_vgic_setup_default_irq_routing(kvm);
+	if (ret)
+		goto out;
+
 	dist->initialized = true;
 out:
 	return ret;
@@ -271,7 +274,6 @@ static void kvm_vgic_dist_destroy(struct kvm *kvm)
 	dist->initialized = false;
 
 	kfree(dist->spis);
-	kfree(dist->redist_iodevs);
 	dist->nr_spis = 0;
 
 	mutex_unlock(&kvm->lock);
@@ -353,32 +355,19 @@ out:
 
 /* GENERIC PROBE */
 
-static void vgic_init_maintenance_interrupt(void *info)
+static int vgic_init_cpu_starting(unsigned int cpu)
 {
 	enable_percpu_irq(kvm_vgic_global_state.maint_irq, 0);
+	return 0;
 }
 
-static int vgic_cpu_notify(struct notifier_block *self,
-			   unsigned long action, void *cpu)
-{
-	switch (action) {
-	case CPU_STARTING:
-	case CPU_STARTING_FROZEN:
-		vgic_init_maintenance_interrupt(NULL);
-		break;
-	case CPU_DYING:
-	case CPU_DYING_FROZEN:
-		disable_percpu_irq(kvm_vgic_global_state.maint_irq);
-		break;
-	}
 
-	return NOTIFY_OK;
+static int vgic_init_cpu_dying(unsigned int cpu)
+{
+	disable_percpu_irq(kvm_vgic_global_state.maint_irq);
+	return 0;
 }
 
-static struct notifier_block vgic_cpu_nb = {
-	.notifier_call = vgic_cpu_notify,
-};
-
 static irqreturn_t vgic_maintenance_handler(int irq, void *data)
 {
 	/*
@@ -434,14 +423,14 @@ int kvm_vgic_hyp_init(void)
 		return ret;
 	}
 
-	ret = __register_cpu_notifier(&vgic_cpu_nb);
+	ret = cpuhp_setup_state(CPUHP_AP_KVM_ARM_VGIC_INIT_STARTING,
+				"AP_KVM_ARM_VGIC_INIT_STARTING",
+				vgic_init_cpu_starting, vgic_init_cpu_dying);
 	if (ret) {
 		kvm_err("Cannot register vgic CPU notifier\n");
 		goto out_free_irq;
 	}
 
-	on_each_cpu(vgic_init_maintenance_interrupt, NULL, 1);
-
 	kvm_info("vgic interrupt IRQ%d\n", kvm_vgic_global_state.maint_irq);
 	return 0;
 
diff --git a/virt/kvm/arm/vgic/vgic-irqfd.c b/virt/kvm/arm/vgic/vgic-irqfd.c
index c67551327..b31a51a14 100644
--- a/virt/kvm/arm/vgic/vgic-irqfd.c
+++ b/virt/kvm/arm/vgic/vgic-irqfd.c
@@ -17,36 +17,116 @@
 #include <linux/kvm.h>
 #include <linux/kvm_host.h>
 #include <trace/events/kvm.h>
+#include <kvm/arm_vgic.h>
+#include "vgic.h"
 
-int kvm_irq_map_gsi(struct kvm *kvm,
-		    struct kvm_kernel_irq_routing_entry *entries,
-		    int gsi)
+/**
+ * vgic_irqfd_set_irq: inject the IRQ corresponding to the
+ * irqchip routing entry
+ *
+ * This is the entry point for irqfd IRQ injection
+ */
+static int vgic_irqfd_set_irq(struct kvm_kernel_irq_routing_entry *e,
+			struct kvm *kvm, int irq_source_id,
+			int level, bool line_status)
 {
-	return 0;
+	unsigned int spi_id = e->irqchip.pin + VGIC_NR_PRIVATE_IRQS;
+
+	if (!vgic_valid_spi(kvm, spi_id))
+		return -EINVAL;
+	return kvm_vgic_inject_irq(kvm, 0, spi_id, level);
 }
 
-int kvm_irq_map_chip_pin(struct kvm *kvm, unsigned int irqchip,
-			 unsigned int pin)
+/**
+ * kvm_set_routing_entry: populate a kvm routing entry
+ * from a user routing entry
+ *
+ * @kvm: the VM this entry is applied to
+ * @e: kvm kernel routing entry handle
+ * @ue: user api routing entry handle
+ * return 0 on success, -EINVAL on errors.
+ */
+#ifdef KVM_CAP_X2APIC_API
+int kvm_set_routing_entry(struct kvm *kvm,
+			  struct kvm_kernel_irq_routing_entry *e,
+			  const struct kvm_irq_routing_entry *ue)
+#else
+/* Remove this version and the ifdefery once merged into 4.8 */
+int kvm_set_routing_entry(struct kvm_kernel_irq_routing_entry *e,
+			  const struct kvm_irq_routing_entry *ue)
+#endif
 {
-	return pin;
+	int r = -EINVAL;
+
+	switch (ue->type) {
+	case KVM_IRQ_ROUTING_IRQCHIP:
+		e->set = vgic_irqfd_set_irq;
+		e->irqchip.irqchip = ue->u.irqchip.irqchip;
+		e->irqchip.pin = ue->u.irqchip.pin;
+		if ((e->irqchip.pin >= KVM_IRQCHIP_NUM_PINS) ||
+		    (e->irqchip.irqchip >= KVM_NR_IRQCHIPS))
+			goto out;
+		break;
+	case KVM_IRQ_ROUTING_MSI:
+		e->set = kvm_set_msi;
+		e->msi.address_lo = ue->u.msi.address_lo;
+		e->msi.address_hi = ue->u.msi.address_hi;
+		e->msi.data = ue->u.msi.data;
+		e->msi.flags = ue->flags;
+		e->msi.devid = ue->u.msi.devid;
+		break;
+	default:
+		goto out;
+	}
+	r = 0;
+out:
+	return r;
 }
 
-int kvm_set_irq(struct kvm *kvm, int irq_source_id,
-		u32 irq, int level, bool line_status)
+/**
+ * kvm_set_msi: inject the MSI corresponding to the
+ * MSI routing entry
+ *
+ * This is the entry point for irqfd MSI injection
+ * and userspace MSI injection.
+ */
+int kvm_set_msi(struct kvm_kernel_irq_routing_entry *e,
+		struct kvm *kvm, int irq_source_id,
+		int level, bool line_status)
 {
-	unsigned int spi = irq + VGIC_NR_PRIVATE_IRQS;
+	struct kvm_msi msi;
 
-	trace_kvm_set_irq(irq, level, irq_source_id);
+	msi.address_lo = e->msi.address_lo;
+	msi.address_hi = e->msi.address_hi;
+	msi.data = e->msi.data;
+	msi.flags = e->msi.flags;
+	msi.devid = e->msi.devid;
 
-	BUG_ON(!vgic_initialized(kvm));
+	if (!vgic_has_its(kvm))
+		return -ENODEV;
 
-	return kvm_vgic_inject_irq(kvm, 0, spi, level);
+	return vgic_its_inject_msi(kvm, &msi);
 }
 
-/* MSI not implemented yet */
-int kvm_set_msi(struct kvm_kernel_irq_routing_entry *e,
-		struct kvm *kvm, int irq_source_id,
-		int level, bool line_status)
+int kvm_vgic_setup_default_irq_routing(struct kvm *kvm)
 {
-	return 0;
+	struct kvm_irq_routing_entry *entries;
+	struct vgic_dist *dist = &kvm->arch.vgic;
+	u32 nr = dist->nr_spis;
+	int i, ret;
+
+	entries = kcalloc(nr, sizeof(struct kvm_kernel_irq_routing_entry),
+			  GFP_KERNEL);
+	if (!entries)
+		return -ENOMEM;
+
+	for (i = 0; i < nr; i++) {
+		entries[i].gsi = i;
+		entries[i].type = KVM_IRQ_ROUTING_IRQCHIP;
+		entries[i].u.irqchip.irqchip = 0;
+		entries[i].u.irqchip.pin = i;
+	}
+	ret = kvm_set_irq_routing(kvm, entries, nr, 0);
+	kfree(entries);
+	return ret;
 }
diff --git a/virt/kvm/arm/vgic/vgic-its.c b/virt/kvm/arm/vgic/vgic-its.c
new file mode 100644
index 000000000..4660a7d04
--- /dev/null
+++ b/virt/kvm/arm/vgic/vgic-its.c
@@ -0,0 +1,1570 @@
+/*
+ * GICv3 ITS emulation
+ *
+ * Copyright (C) 2015,2016 ARM Ltd.
+ * Author: Andre Przywara <andre.przywara@arm.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <linux/cpu.h>
+#include <linux/kvm.h>
+#include <linux/kvm_host.h>
+#include <linux/interrupt.h>
+#include <linux/list.h>
+#include <linux/uaccess.h>
+
+#include <linux/irqchip/arm-gic-v3.h>
+
+#include <asm/kvm_emulate.h>
+#include <asm/kvm_arm.h>
+#include <asm/kvm_mmu.h>
+
+#include "vgic.h"
+#include "vgic-mmio.h"
+
+/*
+ * Creates a new (reference to a) struct vgic_irq for a given LPI.
+ * If this LPI is already mapped on another ITS, we increase its refcount
+ * and return a pointer to the existing structure.
+ * If this is a "new" LPI, we allocate and initialize a new struct vgic_irq.
+ * This function returns a pointer to the _unlocked_ structure.
+ */
+static struct vgic_irq *vgic_add_lpi(struct kvm *kvm, u32 intid)
+{
+	struct vgic_dist *dist = &kvm->arch.vgic;
+	struct vgic_irq *irq = vgic_get_irq(kvm, NULL, intid), *oldirq;
+
+	/* In this case there is no put, since we keep the reference. */
+	if (irq)
+		return irq;
+
+	irq = kzalloc(sizeof(struct vgic_irq), GFP_KERNEL);
+	if (!irq)
+		return ERR_PTR(-ENOMEM);
+
+	INIT_LIST_HEAD(&irq->lpi_list);
+	INIT_LIST_HEAD(&irq->ap_list);
+	spin_lock_init(&irq->irq_lock);
+
+	irq->config = VGIC_CONFIG_EDGE;
+	kref_init(&irq->refcount);
+	irq->intid = intid;
+
+	spin_lock(&dist->lpi_list_lock);
+
+	/*
+	 * There could be a race with another vgic_add_lpi(), so we need to
+	 * check that we don't add a second list entry with the same LPI.
+	 */
+	list_for_each_entry(oldirq, &dist->lpi_list_head, lpi_list) {
+		if (oldirq->intid != intid)
+			continue;
+
+		/* Someone was faster with adding this LPI, lets use that. */
+		kfree(irq);
+		irq = oldirq;
+
+		/*
+		 * This increases the refcount, the caller is expected to
+		 * call vgic_put_irq() on the returned pointer once it's
+		 * finished with the IRQ.
+		 */
+		vgic_get_irq_kref(irq);
+
+		goto out_unlock;
+	}
+
+	list_add_tail(&irq->lpi_list, &dist->lpi_list_head);
+	dist->lpi_list_count++;
+
+out_unlock:
+	spin_unlock(&dist->lpi_list_lock);
+
+	return irq;
+}
+
+struct its_device {
+	struct list_head dev_list;
+
+	/* the head for the list of ITTEs */
+	struct list_head itt_head;
+	u32 device_id;
+};
+
+#define COLLECTION_NOT_MAPPED ((u32)~0)
+
+struct its_collection {
+	struct list_head coll_list;
+
+	u32 collection_id;
+	u32 target_addr;
+};
+
+#define its_is_collection_mapped(coll) ((coll) && \
+				((coll)->target_addr != COLLECTION_NOT_MAPPED))
+
+struct its_itte {
+	struct list_head itte_list;
+
+	struct vgic_irq *irq;
+	struct its_collection *collection;
+	u32 lpi;
+	u32 event_id;
+};
+
+/*
+ * Find and returns a device in the device table for an ITS.
+ * Must be called with the its_lock mutex held.
+ */
+static struct its_device *find_its_device(struct vgic_its *its, u32 device_id)
+{
+	struct its_device *device;
+
+	list_for_each_entry(device, &its->device_list, dev_list)
+		if (device_id == device->device_id)
+			return device;
+
+	return NULL;
+}
+
+/*
+ * Find and returns an interrupt translation table entry (ITTE) for a given
+ * Device ID/Event ID pair on an ITS.
+ * Must be called with the its_lock mutex held.
+ */
+static struct its_itte *find_itte(struct vgic_its *its, u32 device_id,
+				  u32 event_id)
+{
+	struct its_device *device;
+	struct its_itte *itte;
+
+	device = find_its_device(its, device_id);
+	if (device == NULL)
+		return NULL;
+
+	list_for_each_entry(itte, &device->itt_head, itte_list)
+		if (itte->event_id == event_id)
+			return itte;
+
+	return NULL;
+}
+
+/* To be used as an iterator this macro misses the enclosing parentheses */
+#define for_each_lpi_its(dev, itte, its) \
+	list_for_each_entry(dev, &(its)->device_list, dev_list) \
+		list_for_each_entry(itte, &(dev)->itt_head, itte_list)
+
+/*
+ * We only implement 48 bits of PA at the moment, although the ITS
+ * supports more. Let's be restrictive here.
+ */
+#define BASER_ADDRESS(x)	((x) & GENMASK_ULL(47, 16))
+#define CBASER_ADDRESS(x)	((x) & GENMASK_ULL(47, 12))
+#define PENDBASER_ADDRESS(x)	((x) & GENMASK_ULL(47, 16))
+#define PROPBASER_ADDRESS(x)	((x) & GENMASK_ULL(47, 12))
+
+#define GIC_LPI_OFFSET 8192
+
+/*
+ * Finds and returns a collection in the ITS collection table.
+ * Must be called with the its_lock mutex held.
+ */
+static struct its_collection *find_collection(struct vgic_its *its, int coll_id)
+{
+	struct its_collection *collection;
+
+	list_for_each_entry(collection, &its->collection_list, coll_list) {
+		if (coll_id == collection->collection_id)
+			return collection;
+	}
+
+	return NULL;
+}
+
+#define LPI_PROP_ENABLE_BIT(p)	((p) & LPI_PROP_ENABLED)
+#define LPI_PROP_PRIORITY(p)	((p) & 0xfc)
+
+/*
+ * Reads the configuration data for a given LPI from guest memory and
+ * updates the fields in struct vgic_irq.
+ * If filter_vcpu is not NULL, applies only if the IRQ is targeting this
+ * VCPU. Unconditionally applies if filter_vcpu is NULL.
+ */
+static int update_lpi_config(struct kvm *kvm, struct vgic_irq *irq,
+			     struct kvm_vcpu *filter_vcpu)
+{
+	u64 propbase = PROPBASER_ADDRESS(kvm->arch.vgic.propbaser);
+	u8 prop;
+	int ret;
+
+	ret = kvm_read_guest(kvm, propbase + irq->intid - GIC_LPI_OFFSET,
+			     &prop, 1);
+
+	if (ret)
+		return ret;
+
+	spin_lock(&irq->irq_lock);
+
+	if (!filter_vcpu || filter_vcpu == irq->target_vcpu) {
+		irq->priority = LPI_PROP_PRIORITY(prop);
+		irq->enabled = LPI_PROP_ENABLE_BIT(prop);
+
+		vgic_queue_irq_unlock(kvm, irq);
+	} else {
+		spin_unlock(&irq->irq_lock);
+	}
+
+	return 0;
+}
+
+/*
+ * Create a snapshot of the current LPI list, so that we can enumerate all
+ * LPIs without holding any lock.
+ * Returns the array length and puts the kmalloc'ed array into intid_ptr.
+ */
+static int vgic_copy_lpi_list(struct kvm *kvm, u32 **intid_ptr)
+{
+	struct vgic_dist *dist = &kvm->arch.vgic;
+	struct vgic_irq *irq;
+	u32 *intids;
+	int irq_count = dist->lpi_list_count, i = 0;
+
+	/*
+	 * We use the current value of the list length, which may change
+	 * after the kmalloc. We don't care, because the guest shouldn't
+	 * change anything while the command handling is still running,
+	 * and in the worst case we would miss a new IRQ, which one wouldn't
+	 * expect to be covered by this command anyway.
+	 */
+	intids = kmalloc_array(irq_count, sizeof(intids[0]), GFP_KERNEL);
+	if (!intids)
+		return -ENOMEM;
+
+	spin_lock(&dist->lpi_list_lock);
+	list_for_each_entry(irq, &dist->lpi_list_head, lpi_list) {
+		/* We don't need to "get" the IRQ, as we hold the list lock. */
+		intids[i] = irq->intid;
+		if (++i == irq_count)
+			break;
+	}
+	spin_unlock(&dist->lpi_list_lock);
+
+	*intid_ptr = intids;
+	return irq_count;
+}
+
+/*
+ * Promotes the ITS view of affinity of an ITTE (which redistributor this LPI
+ * is targeting) to the VGIC's view, which deals with target VCPUs.
+ * Needs to be called whenever either the collection for a LPIs has
+ * changed or the collection itself got retargeted.
+ */
+static void update_affinity_itte(struct kvm *kvm, struct its_itte *itte)
+{
+	struct kvm_vcpu *vcpu;
+
+	if (!its_is_collection_mapped(itte->collection))
+		return;
+
+	vcpu = kvm_get_vcpu(kvm, itte->collection->target_addr);
+
+	spin_lock(&itte->irq->irq_lock);
+	itte->irq->target_vcpu = vcpu;
+	spin_unlock(&itte->irq->irq_lock);
+}
+
+/*
+ * Updates the target VCPU for every LPI targeting this collection.
+ * Must be called with the its_lock mutex held.
+ */
+static void update_affinity_collection(struct kvm *kvm, struct vgic_its *its,
+				       struct its_collection *coll)
+{
+	struct its_device *device;
+	struct its_itte *itte;
+
+	for_each_lpi_its(device, itte, its) {
+		if (!itte->collection || coll != itte->collection)
+			continue;
+
+		update_affinity_itte(kvm, itte);
+	}
+}
+
+static u32 max_lpis_propbaser(u64 propbaser)
+{
+	int nr_idbits = (propbaser & 0x1f) + 1;
+
+	return 1U << min(nr_idbits, INTERRUPT_ID_BITS_ITS);
+}
+
+/*
+ * Scan the whole LPI pending table and sync the pending bit in there
+ * with our own data structures. This relies on the LPI being
+ * mapped before.
+ */
+static int its_sync_lpi_pending_table(struct kvm_vcpu *vcpu)
+{
+	gpa_t pendbase = PENDBASER_ADDRESS(vcpu->arch.vgic_cpu.pendbaser);
+	struct vgic_irq *irq;
+	int last_byte_offset = -1;
+	int ret = 0;
+	u32 *intids;
+	int nr_irqs, i;
+
+	nr_irqs = vgic_copy_lpi_list(vcpu->kvm, &intids);
+	if (nr_irqs < 0)
+		return nr_irqs;
+
+	for (i = 0; i < nr_irqs; i++) {
+		int byte_offset, bit_nr;
+		u8 pendmask;
+
+		byte_offset = intids[i] / BITS_PER_BYTE;
+		bit_nr = intids[i] % BITS_PER_BYTE;
+
+		/*
+		 * For contiguously allocated LPIs chances are we just read
+		 * this very same byte in the last iteration. Reuse that.
+		 */
+		if (byte_offset != last_byte_offset) {
+			ret = kvm_read_guest(vcpu->kvm, pendbase + byte_offset,
+					     &pendmask, 1);
+			if (ret) {
+				kfree(intids);
+				return ret;
+			}
+			last_byte_offset = byte_offset;
+		}
+
+		irq = vgic_get_irq(vcpu->kvm, NULL, intids[i]);
+		spin_lock(&irq->irq_lock);
+		irq->pending = pendmask & (1U << bit_nr);
+		vgic_queue_irq_unlock(vcpu->kvm, irq);
+		vgic_put_irq(vcpu->kvm, irq);
+	}
+
+	kfree(intids);
+
+	return ret;
+}
+
+static unsigned long vgic_mmio_read_its_ctlr(struct kvm *vcpu,
+					     struct vgic_its *its,
+					     gpa_t addr, unsigned int len)
+{
+	u32 reg = 0;
+
+	mutex_lock(&its->cmd_lock);
+	if (its->creadr == its->cwriter)
+		reg |= GITS_CTLR_QUIESCENT;
+	if (its->enabled)
+		reg |= GITS_CTLR_ENABLE;
+	mutex_unlock(&its->cmd_lock);
+
+	return reg;
+}
+
+static void vgic_mmio_write_its_ctlr(struct kvm *kvm, struct vgic_its *its,
+				     gpa_t addr, unsigned int len,
+				     unsigned long val)
+{
+	its->enabled = !!(val & GITS_CTLR_ENABLE);
+}
+
+static unsigned long vgic_mmio_read_its_typer(struct kvm *kvm,
+					      struct vgic_its *its,
+					      gpa_t addr, unsigned int len)
+{
+	u64 reg = GITS_TYPER_PLPIS;
+
+	/*
+	 * We use linear CPU numbers for redistributor addressing,
+	 * so GITS_TYPER.PTA is 0.
+	 * Also we force all PROPBASER registers to be the same, so
+	 * CommonLPIAff is 0 as well.
+	 * To avoid memory waste in the guest, we keep the number of IDBits and
+	 * DevBits low - as least for the time being.
+	 */
+	reg |= 0x0f << GITS_TYPER_DEVBITS_SHIFT;
+	reg |= 0x0f << GITS_TYPER_IDBITS_SHIFT;
+
+	return extract_bytes(reg, addr & 7, len);
+}
+
+static unsigned long vgic_mmio_read_its_iidr(struct kvm *kvm,
+					     struct vgic_its *its,
+					     gpa_t addr, unsigned int len)
+{
+	return (PRODUCT_ID_KVM << 24) | (IMPLEMENTER_ARM << 0);
+}
+
+static unsigned long vgic_mmio_read_its_idregs(struct kvm *kvm,
+					       struct vgic_its *its,
+					       gpa_t addr, unsigned int len)
+{
+	switch (addr & 0xffff) {
+	case GITS_PIDR0:
+		return 0x92;	/* part number, bits[7:0] */
+	case GITS_PIDR1:
+		return 0xb4;	/* part number, bits[11:8] */
+	case GITS_PIDR2:
+		return GIC_PIDR2_ARCH_GICv3 | 0x0b;
+	case GITS_PIDR4:
+		return 0x40;	/* This is a 64K software visible page */
+	/* The following are the ID registers for (any) GIC. */
+	case GITS_CIDR0:
+		return 0x0d;
+	case GITS_CIDR1:
+		return 0xf0;
+	case GITS_CIDR2:
+		return 0x05;
+	case GITS_CIDR3:
+		return 0xb1;
+	}
+
+	return 0;
+}
+
+/*
+ * Find the target VCPU and the LPI number for a given devid/eventid pair
+ * and make this IRQ pending, possibly injecting it.
+ * Must be called with the its_lock mutex held.
+ * Returns 0 on success, a positive error value for any ITS mapping
+ * related errors and negative error values for generic errors.
+ */
+static int vgic_its_trigger_msi(struct kvm *kvm, struct vgic_its *its,
+				u32 devid, u32 eventid)
+{
+	struct kvm_vcpu *vcpu;
+	struct its_itte *itte;
+
+	if (!its->enabled)
+		return -EBUSY;
+
+	itte = find_itte(its, devid, eventid);
+	if (!itte || !its_is_collection_mapped(itte->collection))
+		return E_ITS_INT_UNMAPPED_INTERRUPT;
+
+	vcpu = kvm_get_vcpu(kvm, itte->collection->target_addr);
+	if (!vcpu)
+		return E_ITS_INT_UNMAPPED_INTERRUPT;
+
+	if (!vcpu->arch.vgic_cpu.lpis_enabled)
+		return -EBUSY;
+
+	spin_lock(&itte->irq->irq_lock);
+	itte->irq->pending = true;
+	vgic_queue_irq_unlock(kvm, itte->irq);
+
+	return 0;
+}
+
+static struct vgic_io_device *vgic_get_its_iodev(struct kvm_io_device *dev)
+{
+	struct vgic_io_device *iodev;
+
+	if (dev->ops != &kvm_io_gic_ops)
+		return NULL;
+
+	iodev = container_of(dev, struct vgic_io_device, dev);
+
+	if (iodev->iodev_type != IODEV_ITS)
+		return NULL;
+
+	return iodev;
+}
+
+/*
+ * Queries the KVM IO bus framework to get the ITS pointer from the given
+ * doorbell address.
+ * We then call vgic_its_trigger_msi() with the decoded data.
+ * According to the KVM_SIGNAL_MSI API description returns 1 on success.
+ */
+int vgic_its_inject_msi(struct kvm *kvm, struct kvm_msi *msi)
+{
+	u64 address;
+	struct kvm_io_device *kvm_io_dev;
+	struct vgic_io_device *iodev;
+	int ret;
+
+	if (!vgic_has_its(kvm))
+		return -ENODEV;
+
+	if (!(msi->flags & KVM_MSI_VALID_DEVID))
+		return -EINVAL;
+
+	address = (u64)msi->address_hi << 32 | msi->address_lo;
+
+	kvm_io_dev = kvm_io_bus_get_dev(kvm, KVM_MMIO_BUS, address);
+	if (!kvm_io_dev)
+		return -EINVAL;
+
+	iodev = vgic_get_its_iodev(kvm_io_dev);
+	if (!iodev)
+		return -EINVAL;
+
+	mutex_lock(&iodev->its->its_lock);
+	ret = vgic_its_trigger_msi(kvm, iodev->its, msi->devid, msi->data);
+	mutex_unlock(&iodev->its->its_lock);
+
+	if (ret < 0)
+		return ret;
+
+	/*
+	 * KVM_SIGNAL_MSI demands a return value > 0 for success and 0
+	 * if the guest has blocked the MSI. So we map any LPI mapping
+	 * related error to that.
+	 */
+	if (ret)
+		return 0;
+	else
+		return 1;
+}
+
+/* Requires the its_lock to be held. */
+static void its_free_itte(struct kvm *kvm, struct its_itte *itte)
+{
+	list_del(&itte->itte_list);
+
+	/* This put matches the get in vgic_add_lpi. */
+	if (itte->irq)
+		vgic_put_irq(kvm, itte->irq);
+
+	kfree(itte);
+}
+
+static u64 its_cmd_mask_field(u64 *its_cmd, int word, int shift, int size)
+{
+	return (le64_to_cpu(its_cmd[word]) >> shift) & (BIT_ULL(size) - 1);
+}
+
+#define its_cmd_get_command(cmd)	its_cmd_mask_field(cmd, 0,  0,  8)
+#define its_cmd_get_deviceid(cmd)	its_cmd_mask_field(cmd, 0, 32, 32)
+#define its_cmd_get_id(cmd)		its_cmd_mask_field(cmd, 1,  0, 32)
+#define its_cmd_get_physical_id(cmd)	its_cmd_mask_field(cmd, 1, 32, 32)
+#define its_cmd_get_collection(cmd)	its_cmd_mask_field(cmd, 2,  0, 16)
+#define its_cmd_get_target_addr(cmd)	its_cmd_mask_field(cmd, 2, 16, 32)
+#define its_cmd_get_validbit(cmd)	its_cmd_mask_field(cmd, 2, 63,  1)
+
+/*
+ * The DISCARD command frees an Interrupt Translation Table Entry (ITTE).
+ * Must be called with the its_lock mutex held.
+ */
+static int vgic_its_cmd_handle_discard(struct kvm *kvm, struct vgic_its *its,
+				       u64 *its_cmd)
+{
+	u32 device_id = its_cmd_get_deviceid(its_cmd);
+	u32 event_id = its_cmd_get_id(its_cmd);
+	struct its_itte *itte;
+
+
+	itte = find_itte(its, device_id, event_id);
+	if (itte && itte->collection) {
+		/*
+		 * Though the spec talks about removing the pending state, we
+		 * don't bother here since we clear the ITTE anyway and the
+		 * pending state is a property of the ITTE struct.
+		 */
+		its_free_itte(kvm, itte);
+		return 0;
+	}
+
+	return E_ITS_DISCARD_UNMAPPED_INTERRUPT;
+}
+
+/*
+ * The MOVI command moves an ITTE to a different collection.
+ * Must be called with the its_lock mutex held.
+ */
+static int vgic_its_cmd_handle_movi(struct kvm *kvm, struct vgic_its *its,
+				    u64 *its_cmd)
+{
+	u32 device_id = its_cmd_get_deviceid(its_cmd);
+	u32 event_id = its_cmd_get_id(its_cmd);
+	u32 coll_id = its_cmd_get_collection(its_cmd);
+	struct kvm_vcpu *vcpu;
+	struct its_itte *itte;
+	struct its_collection *collection;
+
+	itte = find_itte(its, device_id, event_id);
+	if (!itte)
+		return E_ITS_MOVI_UNMAPPED_INTERRUPT;
+
+	if (!its_is_collection_mapped(itte->collection))
+		return E_ITS_MOVI_UNMAPPED_COLLECTION;
+
+	collection = find_collection(its, coll_id);
+	if (!its_is_collection_mapped(collection))
+		return E_ITS_MOVI_UNMAPPED_COLLECTION;
+
+	itte->collection = collection;
+	vcpu = kvm_get_vcpu(kvm, collection->target_addr);
+
+	spin_lock(&itte->irq->irq_lock);
+	itte->irq->target_vcpu = vcpu;
+	spin_unlock(&itte->irq->irq_lock);
+
+	return 0;
+}
+
+/*
+ * Check whether an ID can be stored into the corresponding guest table.
+ * For a direct table this is pretty easy, but gets a bit nasty for
+ * indirect tables. We check whether the resulting guest physical address
+ * is actually valid (covered by a memslot and guest accessbible).
+ * For this we have to read the respective first level entry.
+ */
+static bool vgic_its_check_id(struct vgic_its *its, u64 baser, int id)
+{
+	int l1_tbl_size = GITS_BASER_NR_PAGES(baser) * SZ_64K;
+	int index;
+	u64 indirect_ptr;
+	gfn_t gfn;
+
+	if (!(baser & GITS_BASER_INDIRECT)) {
+		phys_addr_t addr;
+
+		if (id >= (l1_tbl_size / GITS_BASER_ENTRY_SIZE(baser)))
+			return false;
+
+		addr = BASER_ADDRESS(baser) + id * GITS_BASER_ENTRY_SIZE(baser);
+		gfn = addr >> PAGE_SHIFT;
+
+		return kvm_is_visible_gfn(its->dev->kvm, gfn);
+	}
+
+	/* calculate and check the index into the 1st level */
+	index = id / (SZ_64K / GITS_BASER_ENTRY_SIZE(baser));
+	if (index >= (l1_tbl_size / sizeof(u64)))
+		return false;
+
+	/* Each 1st level entry is represented by a 64-bit value. */
+	if (kvm_read_guest(its->dev->kvm,
+			   BASER_ADDRESS(baser) + index * sizeof(indirect_ptr),
+			   &indirect_ptr, sizeof(indirect_ptr)))
+		return false;
+
+	indirect_ptr = le64_to_cpu(indirect_ptr);
+
+	/* check the valid bit of the first level entry */
+	if (!(indirect_ptr & BIT_ULL(63)))
+		return false;
+
+	/*
+	 * Mask the guest physical address and calculate the frame number.
+	 * Any address beyond our supported 48 bits of PA will be caught
+	 * by the actual check in the final step.
+	 */
+	indirect_ptr &= GENMASK_ULL(51, 16);
+
+	/* Find the address of the actual entry */
+	index = id % (SZ_64K / GITS_BASER_ENTRY_SIZE(baser));
+	indirect_ptr += index * GITS_BASER_ENTRY_SIZE(baser);
+	gfn = indirect_ptr >> PAGE_SHIFT;
+
+	return kvm_is_visible_gfn(its->dev->kvm, gfn);
+}
+
+static int vgic_its_alloc_collection(struct vgic_its *its,
+				     struct its_collection **colp,
+				     u32 coll_id)
+{
+	struct its_collection *collection;
+
+	if (!vgic_its_check_id(its, its->baser_coll_table, coll_id))
+		return E_ITS_MAPC_COLLECTION_OOR;
+
+	collection = kzalloc(sizeof(*collection), GFP_KERNEL);
+
+	collection->collection_id = coll_id;
+	collection->target_addr = COLLECTION_NOT_MAPPED;
+
+	list_add_tail(&collection->coll_list, &its->collection_list);
+	*colp = collection;
+
+	return 0;
+}
+
+static void vgic_its_free_collection(struct vgic_its *its, u32 coll_id)
+{
+	struct its_collection *collection;
+	struct its_device *device;
+	struct its_itte *itte;
+
+	/*
+	 * Clearing the mapping for that collection ID removes the
+	 * entry from the list. If there wasn't any before, we can
+	 * go home early.
+	 */
+	collection = find_collection(its, coll_id);
+	if (!collection)
+		return;
+
+	for_each_lpi_its(device, itte, its)
+		if (itte->collection &&
+		    itte->collection->collection_id == coll_id)
+			itte->collection = NULL;
+
+	list_del(&collection->coll_list);
+	kfree(collection);
+}
+
+/*
+ * The MAPTI and MAPI commands map LPIs to ITTEs.
+ * Must be called with its_lock mutex held.
+ */
+static int vgic_its_cmd_handle_mapi(struct kvm *kvm, struct vgic_its *its,
+				    u64 *its_cmd)
+{
+	u32 device_id = its_cmd_get_deviceid(its_cmd);
+	u32 event_id = its_cmd_get_id(its_cmd);
+	u32 coll_id = its_cmd_get_collection(its_cmd);
+	struct its_itte *itte;
+	struct its_device *device;
+	struct its_collection *collection, *new_coll = NULL;
+	int lpi_nr;
+	struct vgic_irq *irq;
+
+	device = find_its_device(its, device_id);
+	if (!device)
+		return E_ITS_MAPTI_UNMAPPED_DEVICE;
+
+	if (its_cmd_get_command(its_cmd) == GITS_CMD_MAPTI)
+		lpi_nr = its_cmd_get_physical_id(its_cmd);
+	else
+		lpi_nr = event_id;
+	if (lpi_nr < GIC_LPI_OFFSET ||
+	    lpi_nr >= max_lpis_propbaser(kvm->arch.vgic.propbaser))
+		return E_ITS_MAPTI_PHYSICALID_OOR;
+
+	/* If there is an existing mapping, behavior is UNPREDICTABLE. */
+	if (find_itte(its, device_id, event_id))
+		return 0;
+
+	collection = find_collection(its, coll_id);
+	if (!collection) {
+		int ret = vgic_its_alloc_collection(its, &collection, coll_id);
+		if (ret)
+			return ret;
+		new_coll = collection;
+	}
+
+	itte = kzalloc(sizeof(struct its_itte), GFP_KERNEL);
+	if (!itte) {
+		if (new_coll)
+			vgic_its_free_collection(its, coll_id);
+		return -ENOMEM;
+	}
+
+	itte->event_id	= event_id;
+	list_add_tail(&itte->itte_list, &device->itt_head);
+
+	itte->collection = collection;
+	itte->lpi = lpi_nr;
+
+	irq = vgic_add_lpi(kvm, lpi_nr);
+	if (IS_ERR(irq)) {
+		if (new_coll)
+			vgic_its_free_collection(its, coll_id);
+		its_free_itte(kvm, itte);
+		return PTR_ERR(irq);
+	}
+	itte->irq = irq;
+
+	update_affinity_itte(kvm, itte);
+
+	/*
+	 * We "cache" the configuration table entries in out struct vgic_irq's.
+	 * However we only have those structs for mapped IRQs, so we read in
+	 * the respective config data from memory here upon mapping the LPI.
+	 */
+	update_lpi_config(kvm, itte->irq, NULL);
+
+	return 0;
+}
+
+/* Requires the its_lock to be held. */
+static void vgic_its_unmap_device(struct kvm *kvm, struct its_device *device)
+{
+	struct its_itte *itte, *temp;
+
+	/*
+	 * The spec says that unmapping a device with still valid
+	 * ITTEs associated is UNPREDICTABLE. We remove all ITTEs,
+	 * since we cannot leave the memory unreferenced.
+	 */
+	list_for_each_entry_safe(itte, temp, &device->itt_head, itte_list)
+		its_free_itte(kvm, itte);
+
+	list_del(&device->dev_list);
+	kfree(device);
+}
+
+/*
+ * MAPD maps or unmaps a device ID to Interrupt Translation Tables (ITTs).
+ * Must be called with the its_lock mutex held.
+ */
+static int vgic_its_cmd_handle_mapd(struct kvm *kvm, struct vgic_its *its,
+				    u64 *its_cmd)
+{
+	u32 device_id = its_cmd_get_deviceid(its_cmd);
+	bool valid = its_cmd_get_validbit(its_cmd);
+	struct its_device *device;
+
+	if (!vgic_its_check_id(its, its->baser_device_table, device_id))
+		return E_ITS_MAPD_DEVICE_OOR;
+
+	device = find_its_device(its, device_id);
+
+	/*
+	 * The spec says that calling MAPD on an already mapped device
+	 * invalidates all cached data for this device. We implement this
+	 * by removing the mapping and re-establishing it.
+	 */
+	if (device)
+		vgic_its_unmap_device(kvm, device);
+
+	/*
+	 * The spec does not say whether unmapping a not-mapped device
+	 * is an error, so we are done in any case.
+	 */
+	if (!valid)
+		return 0;
+
+	device = kzalloc(sizeof(struct its_device), GFP_KERNEL);
+	if (!device)
+		return -ENOMEM;
+
+	device->device_id = device_id;
+	INIT_LIST_HEAD(&device->itt_head);
+
+	list_add_tail(&device->dev_list, &its->device_list);
+
+	return 0;
+}
+
+/*
+ * The MAPC command maps collection IDs to redistributors.
+ * Must be called with the its_lock mutex held.
+ */
+static int vgic_its_cmd_handle_mapc(struct kvm *kvm, struct vgic_its *its,
+				    u64 *its_cmd)
+{
+	u16 coll_id;
+	u32 target_addr;
+	struct its_collection *collection;
+	bool valid;
+
+	valid = its_cmd_get_validbit(its_cmd);
+	coll_id = its_cmd_get_collection(its_cmd);
+	target_addr = its_cmd_get_target_addr(its_cmd);
+
+	if (target_addr >= atomic_read(&kvm->online_vcpus))
+		return E_ITS_MAPC_PROCNUM_OOR;
+
+	if (!valid) {
+		vgic_its_free_collection(its, coll_id);
+	} else {
+		collection = find_collection(its, coll_id);
+
+		if (!collection) {
+			int ret;
+
+			ret = vgic_its_alloc_collection(its, &collection,
+							coll_id);
+			if (ret)
+				return ret;
+			collection->target_addr = target_addr;
+		} else {
+			collection->target_addr = target_addr;
+			update_affinity_collection(kvm, its, collection);
+		}
+	}
+
+	return 0;
+}
+
+/*
+ * The CLEAR command removes the pending state for a particular LPI.
+ * Must be called with the its_lock mutex held.
+ */
+static int vgic_its_cmd_handle_clear(struct kvm *kvm, struct vgic_its *its,
+				     u64 *its_cmd)
+{
+	u32 device_id = its_cmd_get_deviceid(its_cmd);
+	u32 event_id = its_cmd_get_id(its_cmd);
+	struct its_itte *itte;
+
+
+	itte = find_itte(its, device_id, event_id);
+	if (!itte)
+		return E_ITS_CLEAR_UNMAPPED_INTERRUPT;
+
+	itte->irq->pending = false;
+
+	return 0;
+}
+
+/*
+ * The INV command syncs the configuration bits from the memory table.
+ * Must be called with the its_lock mutex held.
+ */
+static int vgic_its_cmd_handle_inv(struct kvm *kvm, struct vgic_its *its,
+				   u64 *its_cmd)
+{
+	u32 device_id = its_cmd_get_deviceid(its_cmd);
+	u32 event_id = its_cmd_get_id(its_cmd);
+	struct its_itte *itte;
+
+
+	itte = find_itte(its, device_id, event_id);
+	if (!itte)
+		return E_ITS_INV_UNMAPPED_INTERRUPT;
+
+	return update_lpi_config(kvm, itte->irq, NULL);
+}
+
+/*
+ * The INVALL command requests flushing of all IRQ data in this collection.
+ * Find the VCPU mapped to that collection, then iterate over the VM's list
+ * of mapped LPIs and update the configuration for each IRQ which targets
+ * the specified vcpu. The configuration will be read from the in-memory
+ * configuration table.
+ * Must be called with the its_lock mutex held.
+ */
+static int vgic_its_cmd_handle_invall(struct kvm *kvm, struct vgic_its *its,
+				      u64 *its_cmd)
+{
+	u32 coll_id = its_cmd_get_collection(its_cmd);
+	struct its_collection *collection;
+	struct kvm_vcpu *vcpu;
+	struct vgic_irq *irq;
+	u32 *intids;
+	int irq_count, i;
+
+	collection = find_collection(its, coll_id);
+	if (!its_is_collection_mapped(collection))
+		return E_ITS_INVALL_UNMAPPED_COLLECTION;
+
+	vcpu = kvm_get_vcpu(kvm, collection->target_addr);
+
+	irq_count = vgic_copy_lpi_list(kvm, &intids);
+	if (irq_count < 0)
+		return irq_count;
+
+	for (i = 0; i < irq_count; i++) {
+		irq = vgic_get_irq(kvm, NULL, intids[i]);
+		if (!irq)
+			continue;
+		update_lpi_config(kvm, irq, vcpu);
+		vgic_put_irq(kvm, irq);
+	}
+
+	kfree(intids);
+
+	return 0;
+}
+
+/*
+ * The MOVALL command moves the pending state of all IRQs targeting one
+ * redistributor to another. We don't hold the pending state in the VCPUs,
+ * but in the IRQs instead, so there is really not much to do for us here.
+ * However the spec says that no IRQ must target the old redistributor
+ * afterwards, so we make sure that no LPI is using the associated target_vcpu.
+ * This command affects all LPIs in the system that target that redistributor.
+ */
+static int vgic_its_cmd_handle_movall(struct kvm *kvm, struct vgic_its *its,
+				      u64 *its_cmd)
+{
+	struct vgic_dist *dist = &kvm->arch.vgic;
+	u32 target1_addr = its_cmd_get_target_addr(its_cmd);
+	u32 target2_addr = its_cmd_mask_field(its_cmd, 3, 16, 32);
+	struct kvm_vcpu *vcpu1, *vcpu2;
+	struct vgic_irq *irq;
+
+	if (target1_addr >= atomic_read(&kvm->online_vcpus) ||
+	    target2_addr >= atomic_read(&kvm->online_vcpus))
+		return E_ITS_MOVALL_PROCNUM_OOR;
+
+	if (target1_addr == target2_addr)
+		return 0;
+
+	vcpu1 = kvm_get_vcpu(kvm, target1_addr);
+	vcpu2 = kvm_get_vcpu(kvm, target2_addr);
+
+	spin_lock(&dist->lpi_list_lock);
+
+	list_for_each_entry(irq, &dist->lpi_list_head, lpi_list) {
+		spin_lock(&irq->irq_lock);
+
+		if (irq->target_vcpu == vcpu1)
+			irq->target_vcpu = vcpu2;
+
+		spin_unlock(&irq->irq_lock);
+	}
+
+	spin_unlock(&dist->lpi_list_lock);
+
+	return 0;
+}
+
+/*
+ * The INT command injects the LPI associated with that DevID/EvID pair.
+ * Must be called with the its_lock mutex held.
+ */
+static int vgic_its_cmd_handle_int(struct kvm *kvm, struct vgic_its *its,
+				   u64 *its_cmd)
+{
+	u32 msi_data = its_cmd_get_id(its_cmd);
+	u64 msi_devid = its_cmd_get_deviceid(its_cmd);
+
+	return vgic_its_trigger_msi(kvm, its, msi_devid, msi_data);
+}
+
+/*
+ * This function is called with the its_cmd lock held, but the ITS data
+ * structure lock dropped.
+ */
+static int vgic_its_handle_command(struct kvm *kvm, struct vgic_its *its,
+				   u64 *its_cmd)
+{
+	int ret = -ENODEV;
+
+	mutex_lock(&its->its_lock);
+	switch (its_cmd_get_command(its_cmd)) {
+	case GITS_CMD_MAPD:
+		ret = vgic_its_cmd_handle_mapd(kvm, its, its_cmd);
+		break;
+	case GITS_CMD_MAPC:
+		ret = vgic_its_cmd_handle_mapc(kvm, its, its_cmd);
+		break;
+	case GITS_CMD_MAPI:
+		ret = vgic_its_cmd_handle_mapi(kvm, its, its_cmd);
+		break;
+	case GITS_CMD_MAPTI:
+		ret = vgic_its_cmd_handle_mapi(kvm, its, its_cmd);
+		break;
+	case GITS_CMD_MOVI:
+		ret = vgic_its_cmd_handle_movi(kvm, its, its_cmd);
+		break;
+	case GITS_CMD_DISCARD:
+		ret = vgic_its_cmd_handle_discard(kvm, its, its_cmd);
+		break;
+	case GITS_CMD_CLEAR:
+		ret = vgic_its_cmd_handle_clear(kvm, its, its_cmd);
+		break;
+	case GITS_CMD_MOVALL:
+		ret = vgic_its_cmd_handle_movall(kvm, its, its_cmd);
+		break;
+	case GITS_CMD_INT:
+		ret = vgic_its_cmd_handle_int(kvm, its, its_cmd);
+		break;
+	case GITS_CMD_INV:
+		ret = vgic_its_cmd_handle_inv(kvm, its, its_cmd);
+		break;
+	case GITS_CMD_INVALL:
+		ret = vgic_its_cmd_handle_invall(kvm, its, its_cmd);
+		break;
+	case GITS_CMD_SYNC:
+		/* we ignore this command: we are in sync all of the time */
+		ret = 0;
+		break;
+	}
+	mutex_unlock(&its->its_lock);
+
+	return ret;
+}
+
+static u64 vgic_sanitise_its_baser(u64 reg)
+{
+	reg = vgic_sanitise_field(reg, GITS_BASER_SHAREABILITY_MASK,
+				  GITS_BASER_SHAREABILITY_SHIFT,
+				  vgic_sanitise_shareability);
+	reg = vgic_sanitise_field(reg, GITS_BASER_INNER_CACHEABILITY_MASK,
+				  GITS_BASER_INNER_CACHEABILITY_SHIFT,
+				  vgic_sanitise_inner_cacheability);
+	reg = vgic_sanitise_field(reg, GITS_BASER_OUTER_CACHEABILITY_MASK,
+				  GITS_BASER_OUTER_CACHEABILITY_SHIFT,
+				  vgic_sanitise_outer_cacheability);
+
+	/* Bits 15:12 contain bits 51:48 of the PA, which we don't support. */
+	reg &= ~GENMASK_ULL(15, 12);
+
+	/* We support only one (ITS) page size: 64K */
+	reg = (reg & ~GITS_BASER_PAGE_SIZE_MASK) | GITS_BASER_PAGE_SIZE_64K;
+
+	return reg;
+}
+
+static u64 vgic_sanitise_its_cbaser(u64 reg)
+{
+	reg = vgic_sanitise_field(reg, GITS_CBASER_SHAREABILITY_MASK,
+				  GITS_CBASER_SHAREABILITY_SHIFT,
+				  vgic_sanitise_shareability);
+	reg = vgic_sanitise_field(reg, GITS_CBASER_INNER_CACHEABILITY_MASK,
+				  GITS_CBASER_INNER_CACHEABILITY_SHIFT,
+				  vgic_sanitise_inner_cacheability);
+	reg = vgic_sanitise_field(reg, GITS_CBASER_OUTER_CACHEABILITY_MASK,
+				  GITS_CBASER_OUTER_CACHEABILITY_SHIFT,
+				  vgic_sanitise_outer_cacheability);
+
+	/*
+	 * Sanitise the physical address to be 64k aligned.
+	 * Also limit the physical addresses to 48 bits.
+	 */
+	reg &= ~(GENMASK_ULL(51, 48) | GENMASK_ULL(15, 12));
+
+	return reg;
+}
+
+static unsigned long vgic_mmio_read_its_cbaser(struct kvm *kvm,
+					       struct vgic_its *its,
+					       gpa_t addr, unsigned int len)
+{
+	return extract_bytes(its->cbaser, addr & 7, len);
+}
+
+static void vgic_mmio_write_its_cbaser(struct kvm *kvm, struct vgic_its *its,
+				       gpa_t addr, unsigned int len,
+				       unsigned long val)
+{
+	/* When GITS_CTLR.Enable is 1, this register is RO. */
+	if (its->enabled)
+		return;
+
+	mutex_lock(&its->cmd_lock);
+	its->cbaser = update_64bit_reg(its->cbaser, addr & 7, len, val);
+	its->cbaser = vgic_sanitise_its_cbaser(its->cbaser);
+	its->creadr = 0;
+	/*
+	 * CWRITER is architecturally UNKNOWN on reset, but we need to reset
+	 * it to CREADR to make sure we start with an empty command buffer.
+	 */
+	its->cwriter = its->creadr;
+	mutex_unlock(&its->cmd_lock);
+}
+
+#define ITS_CMD_BUFFER_SIZE(baser)	((((baser) & 0xff) + 1) << 12)
+#define ITS_CMD_SIZE			32
+#define ITS_CMD_OFFSET(reg)		((reg) & GENMASK(19, 5))
+
+/*
+ * By writing to CWRITER the guest announces new commands to be processed.
+ * To avoid any races in the first place, we take the its_cmd lock, which
+ * protects our ring buffer variables, so that there is only one user
+ * per ITS handling commands at a given time.
+ */
+static void vgic_mmio_write_its_cwriter(struct kvm *kvm, struct vgic_its *its,
+					gpa_t addr, unsigned int len,
+					unsigned long val)
+{
+	gpa_t cbaser;
+	u64 cmd_buf[4];
+	u32 reg;
+
+	if (!its)
+		return;
+
+	mutex_lock(&its->cmd_lock);
+
+	reg = update_64bit_reg(its->cwriter, addr & 7, len, val);
+	reg = ITS_CMD_OFFSET(reg);
+	if (reg >= ITS_CMD_BUFFER_SIZE(its->cbaser)) {
+		mutex_unlock(&its->cmd_lock);
+		return;
+	}
+
+	its->cwriter = reg;
+	cbaser = CBASER_ADDRESS(its->cbaser);
+
+	while (its->cwriter != its->creadr) {
+		int ret = kvm_read_guest(kvm, cbaser + its->creadr,
+					 cmd_buf, ITS_CMD_SIZE);
+		/*
+		 * If kvm_read_guest() fails, this could be due to the guest
+		 * programming a bogus value in CBASER or something else going
+		 * wrong from which we cannot easily recover.
+		 * According to section 6.3.2 in the GICv3 spec we can just
+		 * ignore that command then.
+		 */
+		if (!ret)
+			vgic_its_handle_command(kvm, its, cmd_buf);
+
+		its->creadr += ITS_CMD_SIZE;
+		if (its->creadr == ITS_CMD_BUFFER_SIZE(its->cbaser))
+			its->creadr = 0;
+	}
+
+	mutex_unlock(&its->cmd_lock);
+}
+
+static unsigned long vgic_mmio_read_its_cwriter(struct kvm *kvm,
+						struct vgic_its *its,
+						gpa_t addr, unsigned int len)
+{
+	return extract_bytes(its->cwriter, addr & 0x7, len);
+}
+
+static unsigned long vgic_mmio_read_its_creadr(struct kvm *kvm,
+					       struct vgic_its *its,
+					       gpa_t addr, unsigned int len)
+{
+	return extract_bytes(its->creadr, addr & 0x7, len);
+}
+
+#define BASER_INDEX(addr) (((addr) / sizeof(u64)) & 0x7)
+static unsigned long vgic_mmio_read_its_baser(struct kvm *kvm,
+					      struct vgic_its *its,
+					      gpa_t addr, unsigned int len)
+{
+	u64 reg;
+
+	switch (BASER_INDEX(addr)) {
+	case 0:
+		reg = its->baser_device_table;
+		break;
+	case 1:
+		reg = its->baser_coll_table;
+		break;
+	default:
+		reg = 0;
+		break;
+	}
+
+	return extract_bytes(reg, addr & 7, len);
+}
+
+#define GITS_BASER_RO_MASK	(GENMASK_ULL(52, 48) | GENMASK_ULL(58, 56))
+static void vgic_mmio_write_its_baser(struct kvm *kvm,
+				      struct vgic_its *its,
+				      gpa_t addr, unsigned int len,
+				      unsigned long val)
+{
+	u64 entry_size, device_type;
+	u64 reg, *regptr, clearbits = 0;
+
+	/* When GITS_CTLR.Enable is 1, we ignore write accesses. */
+	if (its->enabled)
+		return;
+
+	switch (BASER_INDEX(addr)) {
+	case 0:
+		regptr = &its->baser_device_table;
+		entry_size = 8;
+		device_type = GITS_BASER_TYPE_DEVICE;
+		break;
+	case 1:
+		regptr = &its->baser_coll_table;
+		entry_size = 8;
+		device_type = GITS_BASER_TYPE_COLLECTION;
+		clearbits = GITS_BASER_INDIRECT;
+		break;
+	default:
+		return;
+	}
+
+	reg = update_64bit_reg(*regptr, addr & 7, len, val);
+	reg &= ~GITS_BASER_RO_MASK;
+	reg &= ~clearbits;
+
+	reg |= (entry_size - 1) << GITS_BASER_ENTRY_SIZE_SHIFT;
+	reg |= device_type << GITS_BASER_TYPE_SHIFT;
+	reg = vgic_sanitise_its_baser(reg);
+
+	*regptr = reg;
+}
+
+#define REGISTER_ITS_DESC(off, rd, wr, length, acc)		\
+{								\
+	.reg_offset = off,					\
+	.len = length,						\
+	.access_flags = acc,					\
+	.its_read = rd,						\
+	.its_write = wr,					\
+}
+
+static void its_mmio_write_wi(struct kvm *kvm, struct vgic_its *its,
+			      gpa_t addr, unsigned int len, unsigned long val)
+{
+	/* Ignore */
+}
+
+static struct vgic_register_region its_registers[] = {
+	REGISTER_ITS_DESC(GITS_CTLR,
+		vgic_mmio_read_its_ctlr, vgic_mmio_write_its_ctlr, 4,
+		VGIC_ACCESS_32bit),
+	REGISTER_ITS_DESC(GITS_IIDR,
+		vgic_mmio_read_its_iidr, its_mmio_write_wi, 4,
+		VGIC_ACCESS_32bit),
+	REGISTER_ITS_DESC(GITS_TYPER,
+		vgic_mmio_read_its_typer, its_mmio_write_wi, 8,
+		VGIC_ACCESS_64bit | VGIC_ACCESS_32bit),
+	REGISTER_ITS_DESC(GITS_CBASER,
+		vgic_mmio_read_its_cbaser, vgic_mmio_write_its_cbaser, 8,
+		VGIC_ACCESS_64bit | VGIC_ACCESS_32bit),
+	REGISTER_ITS_DESC(GITS_CWRITER,
+		vgic_mmio_read_its_cwriter, vgic_mmio_write_its_cwriter, 8,
+		VGIC_ACCESS_64bit | VGIC_ACCESS_32bit),
+	REGISTER_ITS_DESC(GITS_CREADR,
+		vgic_mmio_read_its_creadr, its_mmio_write_wi, 8,
+		VGIC_ACCESS_64bit | VGIC_ACCESS_32bit),
+	REGISTER_ITS_DESC(GITS_BASER,
+		vgic_mmio_read_its_baser, vgic_mmio_write_its_baser, 0x40,
+		VGIC_ACCESS_64bit | VGIC_ACCESS_32bit),
+	REGISTER_ITS_DESC(GITS_IDREGS_BASE,
+		vgic_mmio_read_its_idregs, its_mmio_write_wi, 0x30,
+		VGIC_ACCESS_32bit),
+};
+
+/* This is called on setting the LPI enable bit in the redistributor. */
+void vgic_enable_lpis(struct kvm_vcpu *vcpu)
+{
+	if (!(vcpu->arch.vgic_cpu.pendbaser & GICR_PENDBASER_PTZ))
+		its_sync_lpi_pending_table(vcpu);
+}
+
+static int vgic_register_its_iodev(struct kvm *kvm, struct vgic_its *its)
+{
+	struct vgic_io_device *iodev = &its->iodev;
+	int ret;
+
+	if (!its->initialized)
+		return -EBUSY;
+
+	if (IS_VGIC_ADDR_UNDEF(its->vgic_its_base))
+		return -ENXIO;
+
+	iodev->regions = its_registers;
+	iodev->nr_regions = ARRAY_SIZE(its_registers);
+	kvm_iodevice_init(&iodev->dev, &kvm_io_gic_ops);
+
+	iodev->base_addr = its->vgic_its_base;
+	iodev->iodev_type = IODEV_ITS;
+	iodev->its = its;
+	mutex_lock(&kvm->slots_lock);
+	ret = kvm_io_bus_register_dev(kvm, KVM_MMIO_BUS, iodev->base_addr,
+				      KVM_VGIC_V3_ITS_SIZE, &iodev->dev);
+	mutex_unlock(&kvm->slots_lock);
+
+	return ret;
+}
+
+#define INITIAL_BASER_VALUE						  \
+	(GIC_BASER_CACHEABILITY(GITS_BASER, INNER, RaWb)		| \
+	 GIC_BASER_CACHEABILITY(GITS_BASER, OUTER, SameAsInner)		| \
+	 GIC_BASER_SHAREABILITY(GITS_BASER, InnerShareable)		| \
+	 ((8ULL - 1) << GITS_BASER_ENTRY_SIZE_SHIFT)			| \
+	 GITS_BASER_PAGE_SIZE_64K)
+
+#define INITIAL_PROPBASER_VALUE						  \
+	(GIC_BASER_CACHEABILITY(GICR_PROPBASER, INNER, RaWb)		| \
+	 GIC_BASER_CACHEABILITY(GICR_PROPBASER, OUTER, SameAsInner)	| \
+	 GIC_BASER_SHAREABILITY(GICR_PROPBASER, InnerShareable))
+
+static int vgic_its_create(struct kvm_device *dev, u32 type)
+{
+	struct vgic_its *its;
+
+	if (type != KVM_DEV_TYPE_ARM_VGIC_ITS)
+		return -ENODEV;
+
+	its = kzalloc(sizeof(struct vgic_its), GFP_KERNEL);
+	if (!its)
+		return -ENOMEM;
+
+	mutex_init(&its->its_lock);
+	mutex_init(&its->cmd_lock);
+
+	its->vgic_its_base = VGIC_ADDR_UNDEF;
+
+	INIT_LIST_HEAD(&its->device_list);
+	INIT_LIST_HEAD(&its->collection_list);
+
+	dev->kvm->arch.vgic.has_its = true;
+	its->initialized = false;
+	its->enabled = false;
+	its->dev = dev;
+
+	its->baser_device_table = INITIAL_BASER_VALUE			|
+		((u64)GITS_BASER_TYPE_DEVICE << GITS_BASER_TYPE_SHIFT);
+	its->baser_coll_table = INITIAL_BASER_VALUE |
+		((u64)GITS_BASER_TYPE_COLLECTION << GITS_BASER_TYPE_SHIFT);
+	dev->kvm->arch.vgic.propbaser = INITIAL_PROPBASER_VALUE;
+
+	dev->private = its;
+
+	return 0;
+}
+
+static void vgic_its_destroy(struct kvm_device *kvm_dev)
+{
+	struct kvm *kvm = kvm_dev->kvm;
+	struct vgic_its *its = kvm_dev->private;
+	struct its_device *dev;
+	struct its_itte *itte;
+	struct list_head *dev_cur, *dev_temp;
+	struct list_head *cur, *temp;
+
+	/*
+	 * We may end up here without the lists ever having been initialized.
+	 * Check this and bail out early to avoid dereferencing a NULL pointer.
+	 */
+	if (!its->device_list.next)
+		return;
+
+	mutex_lock(&its->its_lock);
+	list_for_each_safe(dev_cur, dev_temp, &its->device_list) {
+		dev = container_of(dev_cur, struct its_device, dev_list);
+		list_for_each_safe(cur, temp, &dev->itt_head) {
+			itte = (container_of(cur, struct its_itte, itte_list));
+			its_free_itte(kvm, itte);
+		}
+		list_del(dev_cur);
+		kfree(dev);
+	}
+
+	list_for_each_safe(cur, temp, &its->collection_list) {
+		list_del(cur);
+		kfree(container_of(cur, struct its_collection, coll_list));
+	}
+	mutex_unlock(&its->its_lock);
+
+	kfree(its);
+}
+
+static int vgic_its_has_attr(struct kvm_device *dev,
+			     struct kvm_device_attr *attr)
+{
+	switch (attr->group) {
+	case KVM_DEV_ARM_VGIC_GRP_ADDR:
+		switch (attr->attr) {
+		case KVM_VGIC_ITS_ADDR_TYPE:
+			return 0;
+		}
+		break;
+	case KVM_DEV_ARM_VGIC_GRP_CTRL:
+		switch (attr->attr) {
+		case KVM_DEV_ARM_VGIC_CTRL_INIT:
+			return 0;
+		}
+		break;
+	}
+	return -ENXIO;
+}
+
+static int vgic_its_set_attr(struct kvm_device *dev,
+			     struct kvm_device_attr *attr)
+{
+	struct vgic_its *its = dev->private;
+	int ret;
+
+	switch (attr->group) {
+	case KVM_DEV_ARM_VGIC_GRP_ADDR: {
+		u64 __user *uaddr = (u64 __user *)(long)attr->addr;
+		unsigned long type = (unsigned long)attr->attr;
+		u64 addr;
+
+		if (type != KVM_VGIC_ITS_ADDR_TYPE)
+			return -ENODEV;
+
+		if (copy_from_user(&addr, uaddr, sizeof(addr)))
+			return -EFAULT;
+
+		ret = vgic_check_ioaddr(dev->kvm, &its->vgic_its_base,
+					addr, SZ_64K);
+		if (ret)
+			return ret;
+
+		its->vgic_its_base = addr;
+
+		return 0;
+	}
+	case KVM_DEV_ARM_VGIC_GRP_CTRL:
+		switch (attr->attr) {
+		case KVM_DEV_ARM_VGIC_CTRL_INIT:
+			its->initialized = true;
+
+			return 0;
+		}
+		break;
+	}
+	return -ENXIO;
+}
+
+static int vgic_its_get_attr(struct kvm_device *dev,
+			     struct kvm_device_attr *attr)
+{
+	switch (attr->group) {
+	case KVM_DEV_ARM_VGIC_GRP_ADDR: {
+		struct vgic_its *its = dev->private;
+		u64 addr = its->vgic_its_base;
+		u64 __user *uaddr = (u64 __user *)(long)attr->addr;
+		unsigned long type = (unsigned long)attr->attr;
+
+		if (type != KVM_VGIC_ITS_ADDR_TYPE)
+			return -ENODEV;
+
+		if (copy_to_user(uaddr, &addr, sizeof(addr)))
+			return -EFAULT;
+		break;
+	default:
+		return -ENXIO;
+	}
+	}
+
+	return 0;
+}
+
+static struct kvm_device_ops kvm_arm_vgic_its_ops = {
+	.name = "kvm-arm-vgic-its",
+	.create = vgic_its_create,
+	.destroy = vgic_its_destroy,
+	.set_attr = vgic_its_set_attr,
+	.get_attr = vgic_its_get_attr,
+	.has_attr = vgic_its_has_attr,
+};
+
+int kvm_vgic_register_its_device(void)
+{
+	return kvm_register_device_ops(&kvm_arm_vgic_its_ops,
+				       KVM_DEV_TYPE_ARM_VGIC_ITS);
+}
+
+/*
+ * Registers all ITSes with the kvm_io_bus framework.
+ * To follow the existing VGIC initialization sequence, this has to be
+ * done as late as possible, just before the first VCPU runs.
+ */
+int vgic_register_its_iodevs(struct kvm *kvm)
+{
+	struct kvm_device *dev;
+	int ret = 0;
+
+	list_for_each_entry(dev, &kvm->devices, vm_node) {
+		if (dev->ops != &kvm_arm_vgic_its_ops)
+			continue;
+
+		ret = vgic_register_its_iodev(kvm, dev->private);
+		if (ret)
+			return ret;
+		/*
+		 * We don't need to care about tearing down previously
+		 * registered ITSes, as the kvm_io_bus framework removes
+		 * them for us if the VM gets destroyed.
+		 */
+	}
+
+	return ret;
+}
diff --git a/virt/kvm/arm/vgic/vgic-kvm-device.c b/virt/kvm/arm/vgic/vgic-kvm-device.c
index 0130c4b14..1813f93b5 100644
--- a/virt/kvm/arm/vgic/vgic-kvm-device.c
+++ b/virt/kvm/arm/vgic/vgic-kvm-device.c
@@ -21,8 +21,8 @@
 
 /* common helpers */
 
-static int vgic_check_ioaddr(struct kvm *kvm, phys_addr_t *ioaddr,
-			     phys_addr_t addr, phys_addr_t alignment)
+int vgic_check_ioaddr(struct kvm *kvm, phys_addr_t *ioaddr,
+		      phys_addr_t addr, phys_addr_t alignment)
 {
 	if (addr & ~KVM_PHYS_MASK)
 		return -E2BIG;
@@ -210,20 +210,27 @@ static void vgic_destroy(struct kvm_device *dev)
 	kfree(dev);
 }
 
-void kvm_register_vgic_device(unsigned long type)
+int kvm_register_vgic_device(unsigned long type)
 {
+	int ret = -ENODEV;
+
 	switch (type) {
 	case KVM_DEV_TYPE_ARM_VGIC_V2:
-		kvm_register_device_ops(&kvm_arm_vgic_v2_ops,
-					KVM_DEV_TYPE_ARM_VGIC_V2);
+		ret = kvm_register_device_ops(&kvm_arm_vgic_v2_ops,
+					      KVM_DEV_TYPE_ARM_VGIC_V2);
 		break;
 #ifdef CONFIG_KVM_ARM_VGIC_V3
 	case KVM_DEV_TYPE_ARM_VGIC_V3:
-		kvm_register_device_ops(&kvm_arm_vgic_v3_ops,
-					KVM_DEV_TYPE_ARM_VGIC_V3);
+		ret = kvm_register_device_ops(&kvm_arm_vgic_v3_ops,
+					      KVM_DEV_TYPE_ARM_VGIC_V3);
+		if (ret)
+			break;
+		ret = kvm_vgic_register_its_device();
 		break;
 #endif
 	}
+
+	return ret;
 }
 
 /** vgic_attr_regs_access: allows user space to read/write VGIC registers
@@ -428,4 +435,3 @@ struct kvm_device_ops kvm_arm_vgic_v3_ops = {
 };
 
 #endif /* CONFIG_KVM_ARM_VGIC_V3 */
-
diff --git a/virt/kvm/arm/vgic/vgic-mmio-v2.c b/virt/kvm/arm/vgic/vgic-mmio-v2.c
index a21393637..b44b359cb 100644
--- a/virt/kvm/arm/vgic/vgic-mmio-v2.c
+++ b/virt/kvm/arm/vgic/vgic-mmio-v2.c
@@ -102,6 +102,7 @@ static void vgic_mmio_write_sgir(struct kvm_vcpu *source_vcpu,
 		irq->source |= 1U << source_vcpu->vcpu_id;
 
 		vgic_queue_irq_unlock(source_vcpu->kvm, irq);
+		vgic_put_irq(source_vcpu->kvm, irq);
 	}
 }
 
@@ -116,6 +117,8 @@ static unsigned long vgic_mmio_read_target(struct kvm_vcpu *vcpu,
 		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
 
 		val |= (u64)irq->targets << (i * 8);
+
+		vgic_put_irq(vcpu->kvm, irq);
 	}
 
 	return val;
@@ -143,6 +146,7 @@ static void vgic_mmio_write_target(struct kvm_vcpu *vcpu,
 		irq->target_vcpu = kvm_get_vcpu(vcpu->kvm, target);
 
 		spin_unlock(&irq->irq_lock);
+		vgic_put_irq(vcpu->kvm, irq);
 	}
 }
 
@@ -157,6 +161,8 @@ static unsigned long vgic_mmio_read_sgipend(struct kvm_vcpu *vcpu,
 		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
 
 		val |= (u64)irq->source << (i * 8);
+
+		vgic_put_irq(vcpu->kvm, irq);
 	}
 	return val;
 }
@@ -178,6 +184,7 @@ static void vgic_mmio_write_sgipendc(struct kvm_vcpu *vcpu,
 			irq->pending = false;
 
 		spin_unlock(&irq->irq_lock);
+		vgic_put_irq(vcpu->kvm, irq);
 	}
 }
 
@@ -201,6 +208,7 @@ static void vgic_mmio_write_sgipends(struct kvm_vcpu *vcpu,
 		} else {
 			spin_unlock(&irq->irq_lock);
 		}
+		vgic_put_irq(vcpu->kvm, irq);
 	}
 }
 
@@ -429,6 +437,7 @@ int vgic_v2_cpuif_uaccess(struct kvm_vcpu *vcpu, bool is_write,
 	struct vgic_io_device dev = {
 		.regions = vgic_v2_cpu_registers,
 		.nr_regions = ARRAY_SIZE(vgic_v2_cpu_registers),
+		.iodev_type = IODEV_CPUIF,
 	};
 
 	return vgic_uaccess(vcpu, &dev, is_write, offset, val);
@@ -440,6 +449,7 @@ int vgic_v2_dist_uaccess(struct kvm_vcpu *vcpu, bool is_write,
 	struct vgic_io_device dev = {
 		.regions = vgic_v2_dist_registers,
 		.nr_regions = ARRAY_SIZE(vgic_v2_dist_registers),
+		.iodev_type = IODEV_DIST,
 	};
 
 	return vgic_uaccess(vcpu, &dev, is_write, offset, val);
diff --git a/virt/kvm/arm/vgic/vgic-mmio-v3.c b/virt/kvm/arm/vgic/vgic-mmio-v3.c
index a0c515a41..90d81811f 100644
--- a/virt/kvm/arm/vgic/vgic-mmio-v3.c
+++ b/virt/kvm/arm/vgic/vgic-mmio-v3.c
@@ -23,12 +23,35 @@
 #include "vgic-mmio.h"
 
 /* extract @num bytes at @offset bytes offset in data */
-static unsigned long extract_bytes(unsigned long data, unsigned int offset,
-				   unsigned int num)
+unsigned long extract_bytes(unsigned long data, unsigned int offset,
+			    unsigned int num)
 {
 	return (data >> (offset * 8)) & GENMASK_ULL(num * 8 - 1, 0);
 }
 
+/* allows updates of any half of a 64-bit register (or the whole thing) */
+u64 update_64bit_reg(u64 reg, unsigned int offset, unsigned int len,
+		     unsigned long val)
+{
+	int lower = (offset & 4) * 8;
+	int upper = lower + 8 * len - 1;
+
+	reg &= ~GENMASK_ULL(upper, lower);
+	val &= GENMASK_ULL(len * 8 - 1, 0);
+
+	return reg | ((u64)val << lower);
+}
+
+bool vgic_has_its(struct kvm *kvm)
+{
+	struct vgic_dist *dist = &kvm->arch.vgic;
+
+	if (dist->vgic_model != KVM_DEV_TYPE_ARM_VGIC_V3)
+		return false;
+
+	return dist->has_its;
+}
+
 static unsigned long vgic_mmio_read_v3_misc(struct kvm_vcpu *vcpu,
 					    gpa_t addr, unsigned int len)
 {
@@ -43,7 +66,12 @@ static unsigned long vgic_mmio_read_v3_misc(struct kvm_vcpu *vcpu,
 	case GICD_TYPER:
 		value = vcpu->kvm->arch.vgic.nr_spis + VGIC_NR_PRIVATE_IRQS;
 		value = (value >> 5) - 1;
-		value |= (INTERRUPT_ID_BITS_SPIS - 1) << 19;
+		if (vgic_has_its(vcpu->kvm)) {
+			value |= (INTERRUPT_ID_BITS_ITS - 1) << 19;
+			value |= GICD_TYPER_LPIS;
+		} else {
+			value |= (INTERRUPT_ID_BITS_SPIS - 1) << 19;
+		}
 		break;
 	case GICD_IIDR:
 		value = (PRODUCT_ID_KVM << 24) | (IMPLEMENTER_ARM << 0);
@@ -80,15 +108,17 @@ static unsigned long vgic_mmio_read_irouter(struct kvm_vcpu *vcpu,
 {
 	int intid = VGIC_ADDR_TO_INTID(addr, 64);
 	struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, NULL, intid);
+	unsigned long ret = 0;
 
 	if (!irq)
 		return 0;
 
 	/* The upper word is RAZ for us. */
-	if (addr & 4)
-		return 0;
+	if (!(addr & 4))
+		ret = extract_bytes(READ_ONCE(irq->mpidr), addr & 7, len);
 
-	return extract_bytes(READ_ONCE(irq->mpidr), addr & 7, len);
+	vgic_put_irq(vcpu->kvm, irq);
+	return ret;
 }
 
 static void vgic_mmio_write_irouter(struct kvm_vcpu *vcpu,
@@ -96,15 +126,17 @@ static void vgic_mmio_write_irouter(struct kvm_vcpu *vcpu,
 				    unsigned long val)
 {
 	int intid = VGIC_ADDR_TO_INTID(addr, 64);
-	struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, NULL, intid);
-
-	if (!irq)
-		return;
+	struct vgic_irq *irq;
 
 	/* The upper word is WI for us since we don't implement Aff3. */
 	if (addr & 4)
 		return;
 
+	irq = vgic_get_irq(vcpu->kvm, NULL, intid);
+
+	if (!irq)
+		return;
+
 	spin_lock(&irq->irq_lock);
 
 	/* We only care about and preserve Aff0, Aff1 and Aff2. */
@@ -112,6 +144,32 @@ static void vgic_mmio_write_irouter(struct kvm_vcpu *vcpu,
 	irq->target_vcpu = kvm_mpidr_to_vcpu(vcpu->kvm, irq->mpidr);
 
 	spin_unlock(&irq->irq_lock);
+	vgic_put_irq(vcpu->kvm, irq);
+}
+
+static unsigned long vgic_mmio_read_v3r_ctlr(struct kvm_vcpu *vcpu,
+					     gpa_t addr, unsigned int len)
+{
+	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
+
+	return vgic_cpu->lpis_enabled ? GICR_CTLR_ENABLE_LPIS : 0;
+}
+
+
+static void vgic_mmio_write_v3r_ctlr(struct kvm_vcpu *vcpu,
+				     gpa_t addr, unsigned int len,
+				     unsigned long val)
+{
+	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
+	bool was_enabled = vgic_cpu->lpis_enabled;
+
+	if (!vgic_has_its(vcpu->kvm))
+		return;
+
+	vgic_cpu->lpis_enabled = val & GICR_CTLR_ENABLE_LPIS;
+
+	if (!was_enabled && vgic_cpu->lpis_enabled)
+		vgic_enable_lpis(vcpu);
 }
 
 static unsigned long vgic_mmio_read_v3r_typer(struct kvm_vcpu *vcpu,
@@ -125,6 +183,8 @@ static unsigned long vgic_mmio_read_v3r_typer(struct kvm_vcpu *vcpu,
 	value |= ((target_vcpu_id & 0xffff) << 8);
 	if (target_vcpu_id == atomic_read(&vcpu->kvm->online_vcpus) - 1)
 		value |= GICR_TYPER_LAST;
+	if (vgic_has_its(vcpu->kvm))
+		value |= GICR_TYPER_PLPIS;
 
 	return extract_bytes(value, addr & 7, len);
 }
@@ -147,6 +207,148 @@ static unsigned long vgic_mmio_read_v3_idregs(struct kvm_vcpu *vcpu,
 	return 0;
 }
 
+/* We want to avoid outer shareable. */
+u64 vgic_sanitise_shareability(u64 field)
+{
+	switch (field) {
+	case GIC_BASER_OuterShareable:
+		return GIC_BASER_InnerShareable;
+	default:
+		return field;
+	}
+}
+
+/* Avoid any inner non-cacheable mapping. */
+u64 vgic_sanitise_inner_cacheability(u64 field)
+{
+	switch (field) {
+	case GIC_BASER_CACHE_nCnB:
+	case GIC_BASER_CACHE_nC:
+		return GIC_BASER_CACHE_RaWb;
+	default:
+		return field;
+	}
+}
+
+/* Non-cacheable or same-as-inner are OK. */
+u64 vgic_sanitise_outer_cacheability(u64 field)
+{
+	switch (field) {
+	case GIC_BASER_CACHE_SameAsInner:
+	case GIC_BASER_CACHE_nC:
+		return field;
+	default:
+		return GIC_BASER_CACHE_nC;
+	}
+}
+
+u64 vgic_sanitise_field(u64 reg, u64 field_mask, int field_shift,
+			u64 (*sanitise_fn)(u64))
+{
+	u64 field = (reg & field_mask) >> field_shift;
+
+	field = sanitise_fn(field) << field_shift;
+	return (reg & ~field_mask) | field;
+}
+
+#define PROPBASER_RES0_MASK						\
+	(GENMASK_ULL(63, 59) | GENMASK_ULL(55, 52) | GENMASK_ULL(6, 5))
+#define PENDBASER_RES0_MASK						\
+	(BIT_ULL(63) | GENMASK_ULL(61, 59) | GENMASK_ULL(55, 52) |	\
+	 GENMASK_ULL(15, 12) | GENMASK_ULL(6, 0))
+
+static u64 vgic_sanitise_pendbaser(u64 reg)
+{
+	reg = vgic_sanitise_field(reg, GICR_PENDBASER_SHAREABILITY_MASK,
+				  GICR_PENDBASER_SHAREABILITY_SHIFT,
+				  vgic_sanitise_shareability);
+	reg = vgic_sanitise_field(reg, GICR_PENDBASER_INNER_CACHEABILITY_MASK,
+				  GICR_PENDBASER_INNER_CACHEABILITY_SHIFT,
+				  vgic_sanitise_inner_cacheability);
+	reg = vgic_sanitise_field(reg, GICR_PENDBASER_OUTER_CACHEABILITY_MASK,
+				  GICR_PENDBASER_OUTER_CACHEABILITY_SHIFT,
+				  vgic_sanitise_outer_cacheability);
+
+	reg &= ~PENDBASER_RES0_MASK;
+	reg &= ~GENMASK_ULL(51, 48);
+
+	return reg;
+}
+
+static u64 vgic_sanitise_propbaser(u64 reg)
+{
+	reg = vgic_sanitise_field(reg, GICR_PROPBASER_SHAREABILITY_MASK,
+				  GICR_PROPBASER_SHAREABILITY_SHIFT,
+				  vgic_sanitise_shareability);
+	reg = vgic_sanitise_field(reg, GICR_PROPBASER_INNER_CACHEABILITY_MASK,
+				  GICR_PROPBASER_INNER_CACHEABILITY_SHIFT,
+				  vgic_sanitise_inner_cacheability);
+	reg = vgic_sanitise_field(reg, GICR_PROPBASER_OUTER_CACHEABILITY_MASK,
+				  GICR_PROPBASER_OUTER_CACHEABILITY_SHIFT,
+				  vgic_sanitise_outer_cacheability);
+
+	reg &= ~PROPBASER_RES0_MASK;
+	reg &= ~GENMASK_ULL(51, 48);
+	return reg;
+}
+
+static unsigned long vgic_mmio_read_propbase(struct kvm_vcpu *vcpu,
+					     gpa_t addr, unsigned int len)
+{
+	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
+
+	return extract_bytes(dist->propbaser, addr & 7, len);
+}
+
+static void vgic_mmio_write_propbase(struct kvm_vcpu *vcpu,
+				     gpa_t addr, unsigned int len,
+				     unsigned long val)
+{
+	struct vgic_dist *dist = &vcpu->kvm->arch.vgic;
+	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
+	u64 old_propbaser, propbaser;
+
+	/* Storing a value with LPIs already enabled is undefined */
+	if (vgic_cpu->lpis_enabled)
+		return;
+
+	do {
+		old_propbaser = dist->propbaser;
+		propbaser = old_propbaser;
+		propbaser = update_64bit_reg(propbaser, addr & 4, len, val);
+		propbaser = vgic_sanitise_propbaser(propbaser);
+	} while (cmpxchg64(&dist->propbaser, old_propbaser,
+			   propbaser) != old_propbaser);
+}
+
+static unsigned long vgic_mmio_read_pendbase(struct kvm_vcpu *vcpu,
+					     gpa_t addr, unsigned int len)
+{
+	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
+
+	return extract_bytes(vgic_cpu->pendbaser, addr & 7, len);
+}
+
+static void vgic_mmio_write_pendbase(struct kvm_vcpu *vcpu,
+				     gpa_t addr, unsigned int len,
+				     unsigned long val)
+{
+	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
+	u64 old_pendbaser, pendbaser;
+
+	/* Storing a value with LPIs already enabled is undefined */
+	if (vgic_cpu->lpis_enabled)
+		return;
+
+	do {
+		old_pendbaser = vgic_cpu->pendbaser;
+		pendbaser = old_pendbaser;
+		pendbaser = update_64bit_reg(pendbaser, addr & 4, len, val);
+		pendbaser = vgic_sanitise_pendbaser(pendbaser);
+	} while (cmpxchg64(&vgic_cpu->pendbaser, old_pendbaser,
+			   pendbaser) != old_pendbaser);
+}
+
 /*
  * The GICv3 per-IRQ registers are split to control PPIs and SGIs in the
  * redistributors, while SPIs are covered by registers in the distributor
@@ -218,7 +420,7 @@ static const struct vgic_register_region vgic_v3_dist_registers[] = {
 
 static const struct vgic_register_region vgic_v3_rdbase_registers[] = {
 	REGISTER_DESC_WITH_LENGTH(GICR_CTLR,
-		vgic_mmio_read_raz, vgic_mmio_write_wi, 4,
+		vgic_mmio_read_v3r_ctlr, vgic_mmio_write_v3r_ctlr, 4,
 		VGIC_ACCESS_32bit),
 	REGISTER_DESC_WITH_LENGTH(GICR_IIDR,
 		vgic_mmio_read_v3r_iidr, vgic_mmio_write_wi, 4,
@@ -227,10 +429,10 @@ static const struct vgic_register_region vgic_v3_rdbase_registers[] = {
 		vgic_mmio_read_v3r_typer, vgic_mmio_write_wi, 8,
 		VGIC_ACCESS_64bit | VGIC_ACCESS_32bit),
 	REGISTER_DESC_WITH_LENGTH(GICR_PROPBASER,
-		vgic_mmio_read_raz, vgic_mmio_write_wi, 8,
+		vgic_mmio_read_propbase, vgic_mmio_write_propbase, 8,
 		VGIC_ACCESS_64bit | VGIC_ACCESS_32bit),
 	REGISTER_DESC_WITH_LENGTH(GICR_PENDBASER,
-		vgic_mmio_read_raz, vgic_mmio_write_wi, 8,
+		vgic_mmio_read_pendbase, vgic_mmio_write_pendbase, 8,
 		VGIC_ACCESS_64bit | VGIC_ACCESS_32bit),
 	REGISTER_DESC_WITH_LENGTH(GICR_IDREGS,
 		vgic_mmio_read_v3_idregs, vgic_mmio_write_wi, 48,
@@ -285,24 +487,18 @@ unsigned int vgic_v3_init_dist_iodev(struct vgic_io_device *dev)
 
 int vgic_register_redist_iodevs(struct kvm *kvm, gpa_t redist_base_address)
 {
-	int nr_vcpus = atomic_read(&kvm->online_vcpus);
 	struct kvm_vcpu *vcpu;
-	struct vgic_io_device *devices;
 	int c, ret = 0;
 
-	devices = kmalloc(sizeof(struct vgic_io_device) * nr_vcpus * 2,
-			  GFP_KERNEL);
-	if (!devices)
-		return -ENOMEM;
-
 	kvm_for_each_vcpu(c, vcpu, kvm) {
 		gpa_t rd_base = redist_base_address + c * SZ_64K * 2;
 		gpa_t sgi_base = rd_base + SZ_64K;
-		struct vgic_io_device *rd_dev = &devices[c * 2];
-		struct vgic_io_device *sgi_dev = &devices[c * 2 + 1];
+		struct vgic_io_device *rd_dev = &vcpu->arch.vgic_cpu.rd_iodev;
+		struct vgic_io_device *sgi_dev = &vcpu->arch.vgic_cpu.sgi_iodev;
 
 		kvm_iodevice_init(&rd_dev->dev, &kvm_io_gic_ops);
 		rd_dev->base_addr = rd_base;
+		rd_dev->iodev_type = IODEV_REDIST;
 		rd_dev->regions = vgic_v3_rdbase_registers;
 		rd_dev->nr_regions = ARRAY_SIZE(vgic_v3_rdbase_registers);
 		rd_dev->redist_vcpu = vcpu;
@@ -317,6 +513,7 @@ int vgic_register_redist_iodevs(struct kvm *kvm, gpa_t redist_base_address)
 
 		kvm_iodevice_init(&sgi_dev->dev, &kvm_io_gic_ops);
 		sgi_dev->base_addr = sgi_base;
+		sgi_dev->iodev_type = IODEV_REDIST;
 		sgi_dev->regions = vgic_v3_sgibase_registers;
 		sgi_dev->nr_regions = ARRAY_SIZE(vgic_v3_sgibase_registers);
 		sgi_dev->redist_vcpu = vcpu;
@@ -335,14 +532,15 @@ int vgic_register_redist_iodevs(struct kvm *kvm, gpa_t redist_base_address)
 	if (ret) {
 		/* The current c failed, so we start with the previous one. */
 		for (c--; c >= 0; c--) {
+			struct vgic_cpu *vgic_cpu;
+
+			vcpu = kvm_get_vcpu(kvm, c);
+			vgic_cpu = &vcpu->arch.vgic_cpu;
 			kvm_io_bus_unregister_dev(kvm, KVM_MMIO_BUS,
-						  &devices[c * 2].dev);
+						  &vgic_cpu->rd_iodev.dev);
 			kvm_io_bus_unregister_dev(kvm, KVM_MMIO_BUS,
-						  &devices[c * 2 + 1].dev);
+						  &vgic_cpu->sgi_iodev.dev);
 		}
-		kfree(devices);
-	} else {
-		kvm->arch.vgic.redist_iodevs = devices;
 	}
 
 	return ret;
@@ -451,5 +649,6 @@ void vgic_v3_dispatch_sgi(struct kvm_vcpu *vcpu, u64 reg)
 		irq->pending = true;
 
 		vgic_queue_irq_unlock(vcpu->kvm, irq);
+		vgic_put_irq(vcpu->kvm, irq);
 	}
 }
diff --git a/virt/kvm/arm/vgic/vgic-mmio.c b/virt/kvm/arm/vgic/vgic-mmio.c
index 9f6fab74d..3bad3c5ed 100644
--- a/virt/kvm/arm/vgic/vgic-mmio.c
+++ b/virt/kvm/arm/vgic/vgic-mmio.c
@@ -56,6 +56,8 @@ unsigned long vgic_mmio_read_enable(struct kvm_vcpu *vcpu,
 
 		if (irq->enabled)
 			value |= (1U << i);
+
+		vgic_put_irq(vcpu->kvm, irq);
 	}
 
 	return value;
@@ -74,6 +76,8 @@ void vgic_mmio_write_senable(struct kvm_vcpu *vcpu,
 		spin_lock(&irq->irq_lock);
 		irq->enabled = true;
 		vgic_queue_irq_unlock(vcpu->kvm, irq);
+
+		vgic_put_irq(vcpu->kvm, irq);
 	}
 }
 
@@ -92,6 +96,7 @@ void vgic_mmio_write_cenable(struct kvm_vcpu *vcpu,
 		irq->enabled = false;
 
 		spin_unlock(&irq->irq_lock);
+		vgic_put_irq(vcpu->kvm, irq);
 	}
 }
 
@@ -108,6 +113,8 @@ unsigned long vgic_mmio_read_pending(struct kvm_vcpu *vcpu,
 
 		if (irq->pending)
 			value |= (1U << i);
+
+		vgic_put_irq(vcpu->kvm, irq);
 	}
 
 	return value;
@@ -129,6 +136,7 @@ void vgic_mmio_write_spending(struct kvm_vcpu *vcpu,
 			irq->soft_pending = true;
 
 		vgic_queue_irq_unlock(vcpu->kvm, irq);
+		vgic_put_irq(vcpu->kvm, irq);
 	}
 }
 
@@ -152,6 +160,7 @@ void vgic_mmio_write_cpending(struct kvm_vcpu *vcpu,
 		}
 
 		spin_unlock(&irq->irq_lock);
+		vgic_put_irq(vcpu->kvm, irq);
 	}
 }
 
@@ -168,6 +177,8 @@ unsigned long vgic_mmio_read_active(struct kvm_vcpu *vcpu,
 
 		if (irq->active)
 			value |= (1U << i);
+
+		vgic_put_irq(vcpu->kvm, irq);
 	}
 
 	return value;
@@ -242,6 +253,7 @@ void vgic_mmio_write_cactive(struct kvm_vcpu *vcpu,
 	for_each_set_bit(i, &val, len * 8) {
 		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
 		vgic_mmio_change_active(vcpu, irq, false);
+		vgic_put_irq(vcpu->kvm, irq);
 	}
 	vgic_change_active_finish(vcpu, intid);
 }
@@ -257,6 +269,7 @@ void vgic_mmio_write_sactive(struct kvm_vcpu *vcpu,
 	for_each_set_bit(i, &val, len * 8) {
 		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
 		vgic_mmio_change_active(vcpu, irq, true);
+		vgic_put_irq(vcpu->kvm, irq);
 	}
 	vgic_change_active_finish(vcpu, intid);
 }
@@ -272,6 +285,8 @@ unsigned long vgic_mmio_read_priority(struct kvm_vcpu *vcpu,
 		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
 
 		val |= (u64)irq->priority << (i * 8);
+
+		vgic_put_irq(vcpu->kvm, irq);
 	}
 
 	return val;
@@ -298,6 +313,8 @@ void vgic_mmio_write_priority(struct kvm_vcpu *vcpu,
 		/* Narrow the priority range to what we actually support */
 		irq->priority = (val >> (i * 8)) & GENMASK(7, 8 - VGIC_PRI_BITS);
 		spin_unlock(&irq->irq_lock);
+
+		vgic_put_irq(vcpu->kvm, irq);
 	}
 }
 
@@ -313,6 +330,8 @@ unsigned long vgic_mmio_read_config(struct kvm_vcpu *vcpu,
 
 		if (irq->config == VGIC_CONFIG_EDGE)
 			value |= (2U << (i * 2));
+
+		vgic_put_irq(vcpu->kvm, irq);
 	}
 
 	return value;
@@ -326,7 +345,7 @@ void vgic_mmio_write_config(struct kvm_vcpu *vcpu,
 	int i;
 
 	for (i = 0; i < len * 4; i++) {
-		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
+		struct vgic_irq *irq;
 
 		/*
 		 * The configuration cannot be changed for SGIs in general,
@@ -337,14 +356,18 @@ void vgic_mmio_write_config(struct kvm_vcpu *vcpu,
 		if (intid + i < VGIC_NR_PRIVATE_IRQS)
 			continue;
 
+		irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
 		spin_lock(&irq->irq_lock);
+
 		if (test_bit(i * 2 + 1, &val)) {
 			irq->config = VGIC_CONFIG_EDGE;
 		} else {
 			irq->config = VGIC_CONFIG_LEVEL;
 			irq->pending = irq->line_level | irq->soft_pending;
 		}
+
 		spin_unlock(&irq->irq_lock);
+		vgic_put_irq(vcpu->kvm, irq);
 	}
 }
 
@@ -450,8 +473,7 @@ static int dispatch_mmio_read(struct kvm_vcpu *vcpu, struct kvm_io_device *dev,
 {
 	struct vgic_io_device *iodev = kvm_to_vgic_iodev(dev);
 	const struct vgic_register_region *region;
-	struct kvm_vcpu *r_vcpu;
-	unsigned long data;
+	unsigned long data = 0;
 
 	region = vgic_find_mmio_region(iodev->regions, iodev->nr_regions,
 				       addr - iodev->base_addr);
@@ -460,8 +482,21 @@ static int dispatch_mmio_read(struct kvm_vcpu *vcpu, struct kvm_io_device *dev,
 		return 0;
 	}
 
-	r_vcpu = iodev->redist_vcpu ? iodev->redist_vcpu : vcpu;
-	data = region->read(r_vcpu, addr, len);
+	switch (iodev->iodev_type) {
+	case IODEV_CPUIF:
+		data = region->read(vcpu, addr, len);
+		break;
+	case IODEV_DIST:
+		data = region->read(vcpu, addr, len);
+		break;
+	case IODEV_REDIST:
+		data = region->read(iodev->redist_vcpu, addr, len);
+		break;
+	case IODEV_ITS:
+		data = region->its_read(vcpu->kvm, iodev->its, addr, len);
+		break;
+	}
+
 	vgic_data_host_to_mmio_bus(val, len, data);
 	return 0;
 }
@@ -471,7 +506,6 @@ static int dispatch_mmio_write(struct kvm_vcpu *vcpu, struct kvm_io_device *dev,
 {
 	struct vgic_io_device *iodev = kvm_to_vgic_iodev(dev);
 	const struct vgic_register_region *region;
-	struct kvm_vcpu *r_vcpu;
 	unsigned long data = vgic_data_mmio_bus_to_host(val, len);
 
 	region = vgic_find_mmio_region(iodev->regions, iodev->nr_regions,
@@ -482,8 +516,21 @@ static int dispatch_mmio_write(struct kvm_vcpu *vcpu, struct kvm_io_device *dev,
 	if (!check_region(region, addr, len))
 		return 0;
 
-	r_vcpu = iodev->redist_vcpu ? iodev->redist_vcpu : vcpu;
-	region->write(r_vcpu, addr, len, data);
+	switch (iodev->iodev_type) {
+	case IODEV_CPUIF:
+		region->write(vcpu, addr, len, data);
+		break;
+	case IODEV_DIST:
+		region->write(vcpu, addr, len, data);
+		break;
+	case IODEV_REDIST:
+		region->write(iodev->redist_vcpu, addr, len, data);
+		break;
+	case IODEV_ITS:
+		region->its_write(vcpu->kvm, iodev->its, addr, len, data);
+		break;
+	}
+
 	return 0;
 }
 
@@ -513,6 +560,7 @@ int vgic_register_dist_iodev(struct kvm *kvm, gpa_t dist_base_address,
 	}
 
 	io_device->base_addr = dist_base_address;
+	io_device->iodev_type = IODEV_DIST;
 	io_device->redist_vcpu = NULL;
 
 	mutex_lock(&kvm->slots_lock);
diff --git a/virt/kvm/arm/vgic/vgic-mmio.h b/virt/kvm/arm/vgic/vgic-mmio.h
index 850901482..0b3ecf9d1 100644
--- a/virt/kvm/arm/vgic/vgic-mmio.h
+++ b/virt/kvm/arm/vgic/vgic-mmio.h
@@ -21,10 +21,19 @@ struct vgic_register_region {
 	unsigned int len;
 	unsigned int bits_per_irq;
 	unsigned int access_flags;
-	unsigned long (*read)(struct kvm_vcpu *vcpu, gpa_t addr,
-			      unsigned int len);
-	void (*write)(struct kvm_vcpu *vcpu, gpa_t addr, unsigned int len,
-		      unsigned long val);
+	union {
+		unsigned long (*read)(struct kvm_vcpu *vcpu, gpa_t addr,
+				      unsigned int len);
+		unsigned long (*its_read)(struct kvm *kvm, struct vgic_its *its,
+					  gpa_t addr, unsigned int len);
+	};
+	union {
+		void (*write)(struct kvm_vcpu *vcpu, gpa_t addr,
+			      unsigned int len, unsigned long val);
+		void (*its_write)(struct kvm *kvm, struct vgic_its *its,
+				  gpa_t addr, unsigned int len,
+				  unsigned long val);
+	};
 };
 
 extern struct kvm_io_device_ops kvm_io_gic_ops;
@@ -87,6 +96,12 @@ unsigned long vgic_data_mmio_bus_to_host(const void *val, unsigned int len);
 void vgic_data_host_to_mmio_bus(void *buf, unsigned int len,
 				unsigned long data);
 
+unsigned long extract_bytes(unsigned long data, unsigned int offset,
+			    unsigned int num);
+
+u64 update_64bit_reg(u64 reg, unsigned int offset, unsigned int len,
+		     unsigned long val);
+
 unsigned long vgic_mmio_read_raz(struct kvm_vcpu *vcpu,
 				 gpa_t addr, unsigned int len);
 
@@ -147,4 +162,12 @@ unsigned int vgic_v2_init_dist_iodev(struct vgic_io_device *dev);
 
 unsigned int vgic_v3_init_dist_iodev(struct vgic_io_device *dev);
 
+#ifdef CONFIG_KVM_ARM_VGIC_V3
+u64 vgic_sanitise_outer_cacheability(u64 reg);
+u64 vgic_sanitise_inner_cacheability(u64 reg);
+u64 vgic_sanitise_shareability(u64 reg);
+u64 vgic_sanitise_field(u64 reg, u64 field_mask, int field_shift,
+			u64 (*sanitise_fn)(u64));
+#endif
+
 #endif
diff --git a/virt/kvm/arm/vgic/vgic-v2.c b/virt/kvm/arm/vgic/vgic-v2.c
index e31405ee5..0bf6709d1 100644
--- a/virt/kvm/arm/vgic/vgic-v2.c
+++ b/virt/kvm/arm/vgic/vgic-v2.c
@@ -124,6 +124,7 @@ void vgic_v2_fold_lr_state(struct kvm_vcpu *vcpu)
 		}
 
 		spin_unlock(&irq->irq_lock);
+		vgic_put_irq(vcpu->kvm, irq);
 	}
 }
 
@@ -332,20 +333,25 @@ int vgic_v2_probe(const struct gic_kvm_info *info)
 	vtr = readl_relaxed(kvm_vgic_global_state.vctrl_base + GICH_VTR);
 	kvm_vgic_global_state.nr_lr = (vtr & 0x3f) + 1;
 
+	ret = kvm_register_vgic_device(KVM_DEV_TYPE_ARM_VGIC_V2);
+	if (ret) {
+		kvm_err("Cannot register GICv2 KVM device\n");
+		iounmap(kvm_vgic_global_state.vctrl_base);
+		return ret;
+	}
+
 	ret = create_hyp_io_mappings(kvm_vgic_global_state.vctrl_base,
 				     kvm_vgic_global_state.vctrl_base +
 					 resource_size(&info->vctrl),
 				     info->vctrl.start);
-
 	if (ret) {
 		kvm_err("Cannot map VCTRL into hyp\n");
+		kvm_unregister_device_ops(KVM_DEV_TYPE_ARM_VGIC_V2);
 		iounmap(kvm_vgic_global_state.vctrl_base);
 		return ret;
 	}
 
 	kvm_vgic_global_state.can_emulate_gicv2 = true;
-	kvm_register_vgic_device(KVM_DEV_TYPE_ARM_VGIC_V2);
-
 	kvm_vgic_global_state.vcpu_base = info->vcpu.start;
 	kvm_vgic_global_state.type = VGIC_V2;
 	kvm_vgic_global_state.max_gic_vcpus = VGIC_V2_MAX_CPUS;
diff --git a/virt/kvm/arm/vgic/vgic-v3.c b/virt/kvm/arm/vgic/vgic-v3.c
index 346b4ad12..9f0dae397 100644
--- a/virt/kvm/arm/vgic/vgic-v3.c
+++ b/virt/kvm/arm/vgic/vgic-v3.c
@@ -81,6 +81,8 @@ void vgic_v3_fold_lr_state(struct kvm_vcpu *vcpu)
 		else
 			intid = val & GICH_LR_VIRTUALID;
 		irq = vgic_get_irq(vcpu->kvm, vcpu, intid);
+		if (!irq)	/* An LPI could have been unmapped. */
+			continue;
 
 		spin_lock(&irq->irq_lock);
 
@@ -113,6 +115,7 @@ void vgic_v3_fold_lr_state(struct kvm_vcpu *vcpu)
 		}
 
 		spin_unlock(&irq->irq_lock);
+		vgic_put_irq(vcpu->kvm, irq);
 	}
 }
 
@@ -190,6 +193,11 @@ void vgic_v3_get_vmcr(struct kvm_vcpu *vcpu, struct vgic_vmcr *vmcrp)
 	vmcrp->pmr  = (vmcr & ICH_VMCR_PMR_MASK) >> ICH_VMCR_PMR_SHIFT;
 }
 
+#define INITIAL_PENDBASER_VALUE						  \
+	(GIC_BASER_CACHEABILITY(GICR_PENDBASER, INNER, RaWb)		| \
+	GIC_BASER_CACHEABILITY(GICR_PENDBASER, OUTER, SameAsInner)	| \
+	GIC_BASER_SHAREABILITY(GICR_PENDBASER, InnerShareable))
+
 void vgic_v3_enable(struct kvm_vcpu *vcpu)
 {
 	struct vgic_v3_cpu_if *vgic_v3 = &vcpu->arch.vgic_cpu.vgic_v3;
@@ -207,10 +215,12 @@ void vgic_v3_enable(struct kvm_vcpu *vcpu)
 	 * way, so we force SRE to 1 to demonstrate this to the guest.
 	 * This goes with the spec allowing the value to be RAO/WI.
 	 */
-	if (vcpu->kvm->arch.vgic.vgic_model == KVM_DEV_TYPE_ARM_VGIC_V3)
+	if (vcpu->kvm->arch.vgic.vgic_model == KVM_DEV_TYPE_ARM_VGIC_V3) {
 		vgic_v3->vgic_sre = ICC_SRE_EL1_SRE;
-	else
+		vcpu->arch.vgic_cpu.pendbaser = INITIAL_PENDBASER_VALUE;
+	} else {
 		vgic_v3->vgic_sre = 0;
+	}
 
 	/* Get the show on the road... */
 	vgic_v3->vgic_hcr = ICH_HCR_EN;
@@ -279,6 +289,14 @@ int vgic_v3_map_resources(struct kvm *kvm)
 		goto out;
 	}
 
+	if (vgic_has_its(kvm)) {
+		ret = vgic_register_its_iodevs(kvm);
+		if (ret) {
+			kvm_err("Unable to register VGIC ITS MMIO regions\n");
+			goto out;
+		}
+	}
+
 	dist->ready = true;
 
 out:
@@ -296,6 +314,7 @@ out:
 int vgic_v3_probe(const struct gic_kvm_info *info)
 {
 	u32 ich_vtr_el2 = kvm_call_hyp(__vgic_v3_get_ich_vtr_el2);
+	int ret;
 
 	/*
 	 * The ListRegs field is 5 bits, but there is a architectural
@@ -319,12 +338,22 @@ int vgic_v3_probe(const struct gic_kvm_info *info)
 	} else {
 		kvm_vgic_global_state.vcpu_base = info->vcpu.start;
 		kvm_vgic_global_state.can_emulate_gicv2 = true;
-		kvm_register_vgic_device(KVM_DEV_TYPE_ARM_VGIC_V2);
+		ret = kvm_register_vgic_device(KVM_DEV_TYPE_ARM_VGIC_V2);
+		if (ret) {
+			kvm_err("Cannot register GICv2 KVM device.\n");
+			return ret;
+		}
 		kvm_info("vgic-v2@%llx\n", info->vcpu.start);
 	}
+	ret = kvm_register_vgic_device(KVM_DEV_TYPE_ARM_VGIC_V3);
+	if (ret) {
+		kvm_err("Cannot register GICv3 KVM device.\n");
+		kvm_unregister_device_ops(KVM_DEV_TYPE_ARM_VGIC_V2);
+		return ret;
+	}
+
 	if (kvm_vgic_global_state.vcpu_base == 0)
 		kvm_info("disabling GICv2 emulation\n");
-	kvm_register_vgic_device(KVM_DEV_TYPE_ARM_VGIC_V3);
 
 	kvm_vgic_global_state.vctrl_base = NULL;
 	kvm_vgic_global_state.type = VGIC_V3;
diff --git a/virt/kvm/arm/vgic/vgic.c b/virt/kvm/arm/vgic/vgic.c
index 69b61abef..b465ac6d5 100644
--- a/virt/kvm/arm/vgic/vgic.c
+++ b/virt/kvm/arm/vgic/vgic.c
@@ -33,10 +33,17 @@ struct vgic_global __section(.hyp.text) kvm_vgic_global_state;
 
 /*
  * Locking order is always:
- *   vgic_cpu->ap_list_lock
- *     vgic_irq->irq_lock
+ * its->cmd_lock (mutex)
+ *   its->its_lock (mutex)
+ *     vgic_cpu->ap_list_lock
+ *       kvm->lpi_list_lock
+ *         vgic_irq->irq_lock
  *
- * (that is, always take the ap_list_lock before the struct vgic_irq lock).
+ * If you need to take multiple locks, always take the upper lock first,
+ * then the lower ones, e.g. first take the its_lock, then the irq_lock.
+ * If you are already holding a lock and need to take a higher one, you
+ * have to drop the lower ranking lock first and re-aquire it after having
+ * taken the upper one.
  *
  * When taking more than one ap_list_lock at the same time, always take the
  * lowest numbered VCPU's ap_list_lock first, so:
@@ -45,6 +52,41 @@ struct vgic_global __section(.hyp.text) kvm_vgic_global_state;
  *     spin_lock(vcpuY->arch.vgic_cpu.ap_list_lock);
  */
 
+/*
+ * Iterate over the VM's list of mapped LPIs to find the one with a
+ * matching interrupt ID and return a reference to the IRQ structure.
+ */
+static struct vgic_irq *vgic_get_lpi(struct kvm *kvm, u32 intid)
+{
+	struct vgic_dist *dist = &kvm->arch.vgic;
+	struct vgic_irq *irq = NULL;
+
+	spin_lock(&dist->lpi_list_lock);
+
+	list_for_each_entry(irq, &dist->lpi_list_head, lpi_list) {
+		if (irq->intid != intid)
+			continue;
+
+		/*
+		 * This increases the refcount, the caller is expected to
+		 * call vgic_put_irq() later once it's finished with the IRQ.
+		 */
+		vgic_get_irq_kref(irq);
+		goto out_unlock;
+	}
+	irq = NULL;
+
+out_unlock:
+	spin_unlock(&dist->lpi_list_lock);
+
+	return irq;
+}
+
+/*
+ * This looks up the virtual interrupt ID to get the corresponding
+ * struct vgic_irq. It also increases the refcount, so any caller is expected
+ * to call vgic_put_irq() once it's finished with this IRQ.
+ */
 struct vgic_irq *vgic_get_irq(struct kvm *kvm, struct kvm_vcpu *vcpu,
 			      u32 intid)
 {
@@ -56,14 +98,43 @@ struct vgic_irq *vgic_get_irq(struct kvm *kvm, struct kvm_vcpu *vcpu,
 	if (intid <= VGIC_MAX_SPI)
 		return &kvm->arch.vgic.spis[intid - VGIC_NR_PRIVATE_IRQS];
 
-	/* LPIs are not yet covered */
+	/* LPIs */
 	if (intid >= VGIC_MIN_LPI)
-		return NULL;
+		return vgic_get_lpi(kvm, intid);
 
 	WARN(1, "Looking up struct vgic_irq for reserved INTID");
 	return NULL;
 }
 
+/*
+ * We can't do anything in here, because we lack the kvm pointer to
+ * lock and remove the item from the lpi_list. So we keep this function
+ * empty and use the return value of kref_put() to trigger the freeing.
+ */
+static void vgic_irq_release(struct kref *ref)
+{
+}
+
+void vgic_put_irq(struct kvm *kvm, struct vgic_irq *irq)
+{
+	struct vgic_dist *dist = &kvm->arch.vgic;
+
+	if (irq->intid < VGIC_MIN_LPI)
+		return;
+
+	spin_lock(&dist->lpi_list_lock);
+	if (!kref_put(&irq->refcount, vgic_irq_release)) {
+		spin_unlock(&dist->lpi_list_lock);
+		return;
+	};
+
+	list_del(&irq->lpi_list);
+	dist->lpi_list_count--;
+	spin_unlock(&dist->lpi_list_lock);
+
+	kfree(irq);
+}
+
 /**
  * kvm_vgic_target_oracle - compute the target vcpu for an irq
  *
@@ -236,6 +307,11 @@ retry:
 		goto retry;
 	}
 
+	/*
+	 * Grab a reference to the irq to reflect the fact that it is
+	 * now in the ap_list.
+	 */
+	vgic_get_irq_kref(irq);
 	list_add_tail(&irq->ap_list, &vcpu->arch.vgic_cpu.ap_list_head);
 	irq->vcpu = vcpu;
 
@@ -269,14 +345,17 @@ static int vgic_update_irq_pending(struct kvm *kvm, int cpuid,
 	if (!irq)
 		return -EINVAL;
 
-	if (irq->hw != mapped_irq)
+	if (irq->hw != mapped_irq) {
+		vgic_put_irq(kvm, irq);
 		return -EINVAL;
+	}
 
 	spin_lock(&irq->irq_lock);
 
 	if (!vgic_validate_injection(irq, level)) {
 		/* Nothing to see here, move along... */
 		spin_unlock(&irq->irq_lock);
+		vgic_put_irq(kvm, irq);
 		return 0;
 	}
 
@@ -288,6 +367,7 @@ static int vgic_update_irq_pending(struct kvm *kvm, int cpuid,
 	}
 
 	vgic_queue_irq_unlock(kvm, irq);
+	vgic_put_irq(kvm, irq);
 
 	return 0;
 }
@@ -330,25 +410,28 @@ int kvm_vgic_map_phys_irq(struct kvm_vcpu *vcpu, u32 virt_irq, u32 phys_irq)
 	irq->hwintid = phys_irq;
 
 	spin_unlock(&irq->irq_lock);
+	vgic_put_irq(vcpu->kvm, irq);
 
 	return 0;
 }
 
 int kvm_vgic_unmap_phys_irq(struct kvm_vcpu *vcpu, unsigned int virt_irq)
 {
-	struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, virt_irq);
-
-	BUG_ON(!irq);
+	struct vgic_irq *irq;
 
 	if (!vgic_initialized(vcpu->kvm))
 		return -EAGAIN;
 
+	irq = vgic_get_irq(vcpu->kvm, vcpu, virt_irq);
+	BUG_ON(!irq);
+
 	spin_lock(&irq->irq_lock);
 
 	irq->hw = false;
 	irq->hwintid = 0;
 
 	spin_unlock(&irq->irq_lock);
+	vgic_put_irq(vcpu->kvm, irq);
 
 	return 0;
 }
@@ -386,6 +469,15 @@ retry:
 			list_del(&irq->ap_list);
 			irq->vcpu = NULL;
 			spin_unlock(&irq->irq_lock);
+
+			/*
+			 * This vgic_put_irq call matches the
+			 * vgic_get_irq_kref in vgic_queue_irq_unlock,
+			 * where we added the LPI to the ap_list. As
+			 * we remove the irq from the list, we drop
+			 * also drop the refcount.
+			 */
+			vgic_put_irq(vcpu->kvm, irq);
 			continue;
 		}
 
@@ -553,6 +645,9 @@ next:
 /* Sync back the hardware VGIC state into our emulation after a guest's run. */
 void kvm_vgic_sync_hwstate(struct kvm_vcpu *vcpu)
 {
+	if (unlikely(!vgic_initialized(vcpu->kvm)))
+		return;
+
 	vgic_process_maintenance_interrupt(vcpu);
 	vgic_fold_lr_state(vcpu);
 	vgic_prune_ap_list(vcpu);
@@ -561,6 +656,9 @@ void kvm_vgic_sync_hwstate(struct kvm_vcpu *vcpu)
 /* Flush our emulation state into the GIC hardware before entering the guest. */
 void kvm_vgic_flush_hwstate(struct kvm_vcpu *vcpu)
 {
+	if (unlikely(!vgic_initialized(vcpu->kvm)))
+		return;
+
 	spin_lock(&vcpu->arch.vgic_cpu.ap_list_lock);
 	vgic_flush_lr_state(vcpu);
 	spin_unlock(&vcpu->arch.vgic_cpu.ap_list_lock);
@@ -614,6 +712,8 @@ bool kvm_vgic_map_is_active(struct kvm_vcpu *vcpu, unsigned int virt_irq)
 	spin_lock(&irq->irq_lock);
 	map_is_active = irq->hw && irq->active;
 	spin_unlock(&irq->irq_lock);
+	vgic_put_irq(vcpu->kvm, irq);
 
 	return map_is_active;
 }
+
diff --git a/virt/kvm/arm/vgic/vgic.h b/virt/kvm/arm/vgic/vgic.h
index 7b300ca37..6c4625c46 100644
--- a/virt/kvm/arm/vgic/vgic.h
+++ b/virt/kvm/arm/vgic/vgic.h
@@ -25,6 +25,7 @@
 #define IS_VGIC_ADDR_UNDEF(_x)  ((_x) == VGIC_ADDR_UNDEF)
 
 #define INTERRUPT_ID_BITS_SPIS	10
+#define INTERRUPT_ID_BITS_ITS	16
 #define VGIC_PRI_BITS		5
 
 #define vgic_irq_is_sgi(intid) ((intid) < VGIC_NR_SGIS)
@@ -38,9 +39,13 @@ struct vgic_vmcr {
 
 struct vgic_irq *vgic_get_irq(struct kvm *kvm, struct kvm_vcpu *vcpu,
 			      u32 intid);
+void vgic_put_irq(struct kvm *kvm, struct vgic_irq *irq);
 bool vgic_queue_irq_unlock(struct kvm *kvm, struct vgic_irq *irq);
 void vgic_kick_vcpus(struct kvm *kvm);
 
+int vgic_check_ioaddr(struct kvm *kvm, phys_addr_t *ioaddr,
+		      phys_addr_t addr, phys_addr_t alignment);
+
 void vgic_v2_process_maintenance(struct kvm_vcpu *vcpu);
 void vgic_v2_fold_lr_state(struct kvm_vcpu *vcpu);
 void vgic_v2_populate_lr(struct kvm_vcpu *vcpu, struct vgic_irq *irq, int lr);
@@ -59,6 +64,14 @@ int vgic_v2_map_resources(struct kvm *kvm);
 int vgic_register_dist_iodev(struct kvm *kvm, gpa_t dist_base_address,
 			     enum vgic_type);
 
+static inline void vgic_get_irq_kref(struct vgic_irq *irq)
+{
+	if (irq->intid < VGIC_MIN_LPI)
+		return;
+
+	kref_get(&irq->refcount);
+}
+
 #ifdef CONFIG_KVM_ARM_VGIC_V3
 void vgic_v3_process_maintenance(struct kvm_vcpu *vcpu);
 void vgic_v3_fold_lr_state(struct kvm_vcpu *vcpu);
@@ -71,6 +84,11 @@ void vgic_v3_enable(struct kvm_vcpu *vcpu);
 int vgic_v3_probe(const struct gic_kvm_info *info);
 int vgic_v3_map_resources(struct kvm *kvm);
 int vgic_register_redist_iodevs(struct kvm *kvm, gpa_t dist_base_address);
+int vgic_register_its_iodevs(struct kvm *kvm);
+bool vgic_has_its(struct kvm *kvm);
+int kvm_vgic_register_its_device(void);
+void vgic_enable_lpis(struct kvm_vcpu *vcpu);
+int vgic_its_inject_msi(struct kvm *kvm, struct kvm_msi *msi);
 #else
 static inline void vgic_v3_process_maintenance(struct kvm_vcpu *vcpu)
 {
@@ -122,9 +140,33 @@ static inline int vgic_register_redist_iodevs(struct kvm *kvm,
 {
 	return -ENODEV;
 }
+
+static inline int vgic_register_its_iodevs(struct kvm *kvm)
+{
+	return -ENODEV;
+}
+
+static inline bool vgic_has_its(struct kvm *kvm)
+{
+	return false;
+}
+
+static inline int kvm_vgic_register_its_device(void)
+{
+	return -ENODEV;
+}
+
+static inline void vgic_enable_lpis(struct kvm_vcpu *vcpu)
+{
+}
+
+static inline int vgic_its_inject_msi(struct kvm *kvm, struct kvm_msi *msi)
+{
+	return -ENODEV;
+}
 #endif
 
-void kvm_register_vgic_device(unsigned long type);
+int kvm_register_vgic_device(unsigned long type);
 int vgic_lazy_init(struct kvm *kvm);
 int vgic_init(struct kvm *kvm);