Adding upstream version 6.7.7.upstream/6.7.7

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

14 files changed, 762 insertions, 378 deletions

diff --git a/drivers/xen/Kconfig b/drivers/xen/Kconfig
index d43153fec1..d5989871dd 100644
--- a/drivers/xen/Kconfig
+++ b/drivers/xen/Kconfig
@@ -269,12 +269,12 @@ config XEN_PRIVCMD
 	  disaggregated Xen setups this driver might be needed for other
 	  domains, too.
 
-config XEN_PRIVCMD_IRQFD
-	bool "Xen irqfd support"
+config XEN_PRIVCMD_EVENTFD
+	bool "Xen Ioeventfd and irqfd support"
 	depends on XEN_PRIVCMD && XEN_VIRTIO && EVENTFD
 	help
-	  Using the irqfd mechanism a virtio backend running in a daemon can
-	  speed up interrupt injection into a guest.
+	  Using the ioeventfd / irqfd mechanism a virtio backend running in a
+	  daemon can speed up interrupt delivery from / to a guest.
 
 config XEN_ACPI_PROCESSOR
 	tristate "Xen ACPI processor"
diff --git a/drivers/xen/balloon.c b/drivers/xen/balloon.c
index 586a167345..976c6cdf9e 100644
--- a/drivers/xen/balloon.c
+++ b/drivers/xen/balloon.c
@@ -94,7 +94,6 @@ static struct ctl_table balloon_table[] = {
 		.extra1         = SYSCTL_ZERO,
 		.extra2         = SYSCTL_ONE,
 	},
-	{ }
 };
 
 #else
diff --git a/drivers/xen/events/events_2l.c b/drivers/xen/events/events_2l.c
index b8f2f971c2..e3585330cf 100644
--- a/drivers/xen/events/events_2l.c
+++ b/drivers/xen/events/events_2l.c
@@ -171,11 +171,11 @@ static void evtchn_2l_handle_events(unsigned cpu, struct evtchn_loop_ctrl *ctrl)
 	int i;
 	struct shared_info *s = HYPERVISOR_shared_info;
 	struct vcpu_info *vcpu_info = __this_cpu_read(xen_vcpu);
+	evtchn_port_t evtchn;
 
 	/* Timer interrupt has highest priority. */
-	irq = irq_from_virq(cpu, VIRQ_TIMER);
+	irq = irq_evtchn_from_virq(cpu, VIRQ_TIMER, &evtchn);
 	if (irq != -1) {
-		evtchn_port_t evtchn = evtchn_from_irq(irq);
 		word_idx = evtchn / BITS_PER_LONG;
 		bit_idx = evtchn % BITS_PER_LONG;
 		if (active_evtchns(cpu, s, word_idx) & (1ULL << bit_idx))
@@ -328,9 +328,9 @@ irqreturn_t xen_debug_interrupt(int irq, void *dev_id)
 	for (i = 0; i < EVTCHN_2L_NR_CHANNELS; i++) {
 		if (sync_test_bit(i, BM(sh->evtchn_pending))) {
 			int word_idx = i / BITS_PER_EVTCHN_WORD;
-			printk("  %d: event %d -> irq %d%s%s%s\n",
+			printk("  %d: event %d -> irq %u%s%s%s\n",
 			       cpu_from_evtchn(i), i,
-			       get_evtchn_to_irq(i),
+			       irq_from_evtchn(i),
 			       sync_test_bit(word_idx, BM(&v->evtchn_pending_sel))
 			       ? "" : " l2-clear",
 			       !sync_test_bit(i, BM(sh->evtchn_mask))
diff --git a/drivers/xen/events/events_base.c b/drivers/xen/events/events_base.c
index c50419638a..3b9f080109 100644
--- a/drivers/xen/events/events_base.c
+++ b/drivers/xen/events/events_base.c
@@ -174,7 +174,7 @@ static int **evtchn_to_irq;
 #ifdef CONFIG_X86
 static unsigned long *pirq_eoi_map;
 #endif
-static bool (*pirq_needs_eoi)(unsigned irq);
+static bool (*pirq_needs_eoi)(struct irq_info *info);
 
 #define EVTCHN_ROW(e)  (e / (PAGE_SIZE/sizeof(**evtchn_to_irq)))
 #define EVTCHN_COL(e)  (e % (PAGE_SIZE/sizeof(**evtchn_to_irq)))
@@ -190,7 +190,6 @@ static struct irq_chip xen_lateeoi_chip;
 static struct irq_chip xen_percpu_chip;
 static struct irq_chip xen_pirq_chip;
 static void enable_dynirq(struct irq_data *data);
-static void disable_dynirq(struct irq_data *data);
 
 static DEFINE_PER_CPU(unsigned int, irq_epoch);
 
@@ -248,15 +247,6 @@ static int set_evtchn_to_irq(evtchn_port_t evtchn, unsigned int irq)
 	return 0;
 }
 
-int get_evtchn_to_irq(evtchn_port_t evtchn)
-{
-	if (evtchn >= xen_evtchn_max_channels())
-		return -1;
-	if (evtchn_to_irq[EVTCHN_ROW(evtchn)] == NULL)
-		return -1;
-	return READ_ONCE(evtchn_to_irq[EVTCHN_ROW(evtchn)][EVTCHN_COL(evtchn)]);
-}
-
 /* Get info for IRQ */
 static struct irq_info *info_for_irq(unsigned irq)
 {
@@ -274,6 +264,19 @@ static void set_info_for_irq(unsigned int irq, struct irq_info *info)
 		irq_set_chip_data(irq, info);
 }
 
+static struct irq_info *evtchn_to_info(evtchn_port_t evtchn)
+{
+	int irq;
+
+	if (evtchn >= xen_evtchn_max_channels())
+		return NULL;
+	if (evtchn_to_irq[EVTCHN_ROW(evtchn)] == NULL)
+		return NULL;
+	irq = READ_ONCE(evtchn_to_irq[EVTCHN_ROW(evtchn)][EVTCHN_COL(evtchn)]);
+
+	return (irq < 0) ? NULL : info_for_irq(irq);
+}
+
 /* Per CPU channel accounting */
 static void channels_on_cpu_dec(struct irq_info *info)
 {
@@ -300,6 +303,13 @@ static void channels_on_cpu_inc(struct irq_info *info)
 	info->is_accounted = 1;
 }
 
+static void xen_irq_free_desc(unsigned int irq)
+{
+	/* Legacy IRQ descriptors are managed by the arch. */
+	if (irq >= nr_legacy_irqs())
+		irq_free_desc(irq);
+}
+
 static void delayed_free_irq(struct work_struct *work)
 {
 	struct irq_info *info = container_of(to_rcu_work(work), struct irq_info,
@@ -311,14 +321,11 @@ static void delayed_free_irq(struct work_struct *work)
 
 	kfree(info);
 
-	/* Legacy IRQ descriptors are managed by the arch. */
-	if (irq >= nr_legacy_irqs())
-		irq_free_desc(irq);
+	xen_irq_free_desc(irq);
 }
 
 /* Constructors for packed IRQ information. */
 static int xen_irq_info_common_setup(struct irq_info *info,
-				     unsigned irq,
 				     enum xen_irq_type type,
 				     evtchn_port_t evtchn,
 				     unsigned short cpu)
@@ -328,29 +335,27 @@ static int xen_irq_info_common_setup(struct irq_info *info,
 	BUG_ON(info->type != IRQT_UNBOUND && info->type != type);
 
 	info->type = type;
-	info->irq = irq;
 	info->evtchn = evtchn;
 	info->cpu = cpu;
 	info->mask_reason = EVT_MASK_REASON_EXPLICIT;
 	raw_spin_lock_init(&info->lock);
 
-	ret = set_evtchn_to_irq(evtchn, irq);
+	ret = set_evtchn_to_irq(evtchn, info->irq);
 	if (ret < 0)
 		return ret;
 
-	irq_clear_status_flags(irq, IRQ_NOREQUEST|IRQ_NOAUTOEN);
+	irq_clear_status_flags(info->irq, IRQ_NOREQUEST | IRQ_NOAUTOEN);
 
 	return xen_evtchn_port_setup(evtchn);
 }
 
-static int xen_irq_info_evtchn_setup(unsigned irq,
+static int xen_irq_info_evtchn_setup(struct irq_info *info,
 				     evtchn_port_t evtchn,
 				     struct xenbus_device *dev)
 {
-	struct irq_info *info = info_for_irq(irq);
 	int ret;
 
-	ret = xen_irq_info_common_setup(info, irq, IRQT_EVTCHN, evtchn, 0);
+	ret = xen_irq_info_common_setup(info, IRQT_EVTCHN, evtchn, 0);
 	info->u.interdomain = dev;
 	if (dev)
 		atomic_inc(&dev->event_channels);
@@ -358,50 +363,37 @@ static int xen_irq_info_evtchn_setup(unsigned irq,
 	return ret;
 }
 
-static int xen_irq_info_ipi_setup(unsigned cpu,
-				  unsigned irq,
-				  evtchn_port_t evtchn,
-				  enum ipi_vector ipi)
+static int xen_irq_info_ipi_setup(struct irq_info *info, unsigned int cpu,
+				  evtchn_port_t evtchn, enum ipi_vector ipi)
 {
-	struct irq_info *info = info_for_irq(irq);
-
 	info->u.ipi = ipi;
 
-	per_cpu(ipi_to_irq, cpu)[ipi] = irq;
+	per_cpu(ipi_to_irq, cpu)[ipi] = info->irq;
 	per_cpu(ipi_to_evtchn, cpu)[ipi] = evtchn;
 
-	return xen_irq_info_common_setup(info, irq, IRQT_IPI, evtchn, 0);
+	return xen_irq_info_common_setup(info, IRQT_IPI, evtchn, 0);
 }
 
-static int xen_irq_info_virq_setup(unsigned cpu,
-				   unsigned irq,
-				   evtchn_port_t evtchn,
-				   unsigned virq)
+static int xen_irq_info_virq_setup(struct irq_info *info, unsigned int cpu,
+				   evtchn_port_t evtchn, unsigned int virq)
 {
-	struct irq_info *info = info_for_irq(irq);
-
 	info->u.virq = virq;
 
-	per_cpu(virq_to_irq, cpu)[virq] = irq;
+	per_cpu(virq_to_irq, cpu)[virq] = info->irq;
 
-	return xen_irq_info_common_setup(info, irq, IRQT_VIRQ, evtchn, 0);
+	return xen_irq_info_common_setup(info, IRQT_VIRQ, evtchn, 0);
 }
 
-static int xen_irq_info_pirq_setup(unsigned irq,
-				   evtchn_port_t evtchn,
-				   unsigned pirq,
-				   unsigned gsi,
-				   uint16_t domid,
-				   unsigned char flags)
+static int xen_irq_info_pirq_setup(struct irq_info *info, evtchn_port_t evtchn,
+				   unsigned int pirq, unsigned int gsi,
+				   uint16_t domid, unsigned char flags)
 {
-	struct irq_info *info = info_for_irq(irq);
-
 	info->u.pirq.pirq = pirq;
 	info->u.pirq.gsi = gsi;
 	info->u.pirq.domid = domid;
 	info->u.pirq.flags = flags;
 
-	return xen_irq_info_common_setup(info, irq, IRQT_PIRQ, evtchn, 0);
+	return xen_irq_info_common_setup(info, IRQT_PIRQ, evtchn, 0);
 }
 
 static void xen_irq_info_cleanup(struct irq_info *info)
@@ -415,7 +407,7 @@ static void xen_irq_info_cleanup(struct irq_info *info)
 /*
  * Accessors for packed IRQ information.
  */
-evtchn_port_t evtchn_from_irq(unsigned irq)
+static evtchn_port_t evtchn_from_irq(unsigned int irq)
 {
 	const struct irq_info *info = NULL;
 
@@ -429,64 +421,51 @@ evtchn_port_t evtchn_from_irq(unsigned irq)
 
 unsigned int irq_from_evtchn(evtchn_port_t evtchn)
 {
-	return get_evtchn_to_irq(evtchn);
+	struct irq_info *info = evtchn_to_info(evtchn);
+
+	return info ? info->irq : -1;
 }
 EXPORT_SYMBOL_GPL(irq_from_evtchn);
 
-int irq_from_virq(unsigned int cpu, unsigned int virq)
+int irq_evtchn_from_virq(unsigned int cpu, unsigned int virq,
+			 evtchn_port_t *evtchn)
 {
-	return per_cpu(virq_to_irq, cpu)[virq];
+	int irq = per_cpu(virq_to_irq, cpu)[virq];
+
+	*evtchn = evtchn_from_irq(irq);
+
+	return irq;
 }
 
-static enum ipi_vector ipi_from_irq(unsigned irq)
+static enum ipi_vector ipi_from_irq(struct irq_info *info)
 {
-	struct irq_info *info = info_for_irq(irq);
-
 	BUG_ON(info == NULL);
 	BUG_ON(info->type != IRQT_IPI);
 
 	return info->u.ipi;
 }
 
-static unsigned virq_from_irq(unsigned irq)
+static unsigned int virq_from_irq(struct irq_info *info)
 {
-	struct irq_info *info = info_for_irq(irq);
-
 	BUG_ON(info == NULL);
 	BUG_ON(info->type != IRQT_VIRQ);
 
 	return info->u.virq;
 }
 
-static unsigned pirq_from_irq(unsigned irq)
+static unsigned int pirq_from_irq(struct irq_info *info)
 {
-	struct irq_info *info = info_for_irq(irq);
-
 	BUG_ON(info == NULL);
 	BUG_ON(info->type != IRQT_PIRQ);
 
 	return info->u.pirq.pirq;
 }
 
-static enum xen_irq_type type_from_irq(unsigned irq)
-{
-	return info_for_irq(irq)->type;
-}
-
-static unsigned cpu_from_irq(unsigned irq)
-{
-	return info_for_irq(irq)->cpu;
-}
-
 unsigned int cpu_from_evtchn(evtchn_port_t evtchn)
 {
-	int irq = get_evtchn_to_irq(evtchn);
-	unsigned ret = 0;
-
-	if (irq != -1)
-		ret = cpu_from_irq(irq);
+	struct irq_info *info = evtchn_to_info(evtchn);
 
-	return ret;
+	return info ? info->cpu : 0;
 }
 
 static void do_mask(struct irq_info *info, u8 reason)
@@ -518,36 +497,30 @@ static void do_unmask(struct irq_info *info, u8 reason)
 }
 
 #ifdef CONFIG_X86
-static bool pirq_check_eoi_map(unsigned irq)
+static bool pirq_check_eoi_map(struct irq_info *info)
 {
-	return test_bit(pirq_from_irq(irq), pirq_eoi_map);
+	return test_bit(pirq_from_irq(info), pirq_eoi_map);
 }
 #endif
 
-static bool pirq_needs_eoi_flag(unsigned irq)
+static bool pirq_needs_eoi_flag(struct irq_info *info)
 {
-	struct irq_info *info = info_for_irq(irq);
 	BUG_ON(info->type != IRQT_PIRQ);
 
 	return info->u.pirq.flags & PIRQ_NEEDS_EOI;
 }
 
-static void bind_evtchn_to_cpu(evtchn_port_t evtchn, unsigned int cpu,
+static void bind_evtchn_to_cpu(struct irq_info *info, unsigned int cpu,
 			       bool force_affinity)
 {
-	int irq = get_evtchn_to_irq(evtchn);
-	struct irq_info *info = info_for_irq(irq);
-
-	BUG_ON(irq == -1);
-
 	if (IS_ENABLED(CONFIG_SMP) && force_affinity) {
-		struct irq_data *data = irq_get_irq_data(irq);
+		struct irq_data *data = irq_get_irq_data(info->irq);
 
 		irq_data_update_affinity(data, cpumask_of(cpu));
 		irq_data_update_effective_affinity(data, cpumask_of(cpu));
 	}
 
-	xen_evtchn_port_bind_to_cpu(evtchn, cpu, info->cpu);
+	xen_evtchn_port_bind_to_cpu(info->evtchn, cpu, info->cpu);
 
 	channels_on_cpu_dec(info);
 	info->cpu = cpu;
@@ -737,50 +710,49 @@ void xen_irq_lateeoi(unsigned int irq, unsigned int eoi_flags)
 }
 EXPORT_SYMBOL_GPL(xen_irq_lateeoi);
 
-static void xen_irq_init(unsigned irq)
+static struct irq_info *xen_irq_init(unsigned int irq)
 {
 	struct irq_info *info;
 
 	info = kzalloc(sizeof(*info), GFP_KERNEL);
-	if (info == NULL)
-		panic("Unable to allocate metadata for IRQ%d\n", irq);
+	if (info) {
+		info->irq = irq;
+		info->type = IRQT_UNBOUND;
+		info->refcnt = -1;
+		INIT_RCU_WORK(&info->rwork, delayed_free_irq);
 
-	info->type = IRQT_UNBOUND;
-	info->refcnt = -1;
-	INIT_RCU_WORK(&info->rwork, delayed_free_irq);
+		set_info_for_irq(irq, info);
+		/*
+		 * Interrupt affinity setting can be immediate. No point
+		 * in delaying it until an interrupt is handled.
+		 */
+		irq_set_status_flags(irq, IRQ_MOVE_PCNTXT);
 
-	set_info_for_irq(irq, info);
-	/*
-	 * Interrupt affinity setting can be immediate. No point
-	 * in delaying it until an interrupt is handled.
-	 */
-	irq_set_status_flags(irq, IRQ_MOVE_PCNTXT);
+		INIT_LIST_HEAD(&info->eoi_list);
+		list_add_tail(&info->list, &xen_irq_list_head);
+	}
 
-	INIT_LIST_HEAD(&info->eoi_list);
-	list_add_tail(&info->list, &xen_irq_list_head);
+	return info;
 }
 
-static int __must_check xen_allocate_irqs_dynamic(int nvec)
+static struct irq_info *xen_allocate_irq_dynamic(void)
 {
-	int i, irq = irq_alloc_descs(-1, 0, nvec, -1);
+	int irq = irq_alloc_desc_from(0, -1);
+	struct irq_info *info = NULL;
 
 	if (irq >= 0) {
-		for (i = 0; i < nvec; i++)
-			xen_irq_init(irq + i);
+		info = xen_irq_init(irq);
+		if (!info)
+			xen_irq_free_desc(irq);
 	}
 
-	return irq;
-}
-
-static inline int __must_check xen_allocate_irq_dynamic(void)
-{
-
-	return xen_allocate_irqs_dynamic(1);
+	return info;
 }
 
-static int __must_check xen_allocate_irq_gsi(unsigned gsi)
+static struct irq_info *xen_allocate_irq_gsi(unsigned int gsi)
 {
 	int irq;
+	struct irq_info *info;
 
 	/*
 	 * A PV guest has no concept of a GSI (since it has no ACPI
@@ -797,15 +769,15 @@ static int __must_check xen_allocate_irq_gsi(unsigned gsi)
 	else
 		irq = irq_alloc_desc_at(gsi, -1);
 
-	xen_irq_init(irq);
+	info = xen_irq_init(irq);
+	if (!info)
+		xen_irq_free_desc(irq);
 
-	return irq;
+	return info;
 }
 
-static void xen_free_irq(unsigned irq)
+static void xen_free_irq(struct irq_info *info)
 {
-	struct irq_info *info = info_for_irq(irq);
-
 	if (WARN_ON(!info))
 		return;
 
@@ -826,14 +798,11 @@ static void event_handler_exit(struct irq_info *info)
 	clear_evtchn(info->evtchn);
 }
 
-static void pirq_query_unmask(int irq)
+static void pirq_query_unmask(struct irq_info *info)
 {
 	struct physdev_irq_status_query irq_status;
-	struct irq_info *info = info_for_irq(irq);
 
-	BUG_ON(info->type != IRQT_PIRQ);
-
-	irq_status.irq = pirq_from_irq(irq);
+	irq_status.irq = pirq_from_irq(info);
 	if (HYPERVISOR_physdev_op(PHYSDEVOP_irq_status_query, &irq_status))
 		irq_status.flags = 0;
 
@@ -842,61 +811,81 @@ static void pirq_query_unmask(int irq)
 		info->u.pirq.flags |= PIRQ_NEEDS_EOI;
 }
 
-static void eoi_pirq(struct irq_data *data)
+static void do_eoi_pirq(struct irq_info *info)
 {
-	struct irq_info *info = info_for_irq(data->irq);
-	evtchn_port_t evtchn = info ? info->evtchn : 0;
-	struct physdev_eoi eoi = { .irq = pirq_from_irq(data->irq) };
+	struct physdev_eoi eoi = { .irq = pirq_from_irq(info) };
 	int rc = 0;
 
-	if (!VALID_EVTCHN(evtchn))
+	if (!VALID_EVTCHN(info->evtchn))
 		return;
 
 	event_handler_exit(info);
 
-	if (pirq_needs_eoi(data->irq)) {
+	if (pirq_needs_eoi(info)) {
 		rc = HYPERVISOR_physdev_op(PHYSDEVOP_eoi, &eoi);
 		WARN_ON(rc);
 	}
 }
 
+static void eoi_pirq(struct irq_data *data)
+{
+	struct irq_info *info = info_for_irq(data->irq);
+
+	do_eoi_pirq(info);
+}
+
+static void do_disable_dynirq(struct irq_info *info)
+{
+	if (VALID_EVTCHN(info->evtchn))
+		do_mask(info, EVT_MASK_REASON_EXPLICIT);
+}
+
+static void disable_dynirq(struct irq_data *data)
+{
+	struct irq_info *info = info_for_irq(data->irq);
+
+	if (info)
+		do_disable_dynirq(info);
+}
+
 static void mask_ack_pirq(struct irq_data *data)
 {
-	disable_dynirq(data);
-	eoi_pirq(data);
+	struct irq_info *info = info_for_irq(data->irq);
+
+	if (info) {
+		do_disable_dynirq(info);
+		do_eoi_pirq(info);
+	}
 }
 
-static unsigned int __startup_pirq(unsigned int irq)
+static unsigned int __startup_pirq(struct irq_info *info)
 {
 	struct evtchn_bind_pirq bind_pirq;
-	struct irq_info *info = info_for_irq(irq);
-	evtchn_port_t evtchn = evtchn_from_irq(irq);
+	evtchn_port_t evtchn = info->evtchn;
 	int rc;
 
-	BUG_ON(info->type != IRQT_PIRQ);
-
 	if (VALID_EVTCHN(evtchn))
 		goto out;
 
-	bind_pirq.pirq = pirq_from_irq(irq);
+	bind_pirq.pirq = pirq_from_irq(info);
 	/* NB. We are happy to share unless we are probing. */
 	bind_pirq.flags = info->u.pirq.flags & PIRQ_SHAREABLE ?
 					BIND_PIRQ__WILL_SHARE : 0;
 	rc = HYPERVISOR_event_channel_op(EVTCHNOP_bind_pirq, &bind_pirq);
 	if (rc != 0) {
-		pr_warn("Failed to obtain physical IRQ %d\n", irq);
+		pr_warn("Failed to obtain physical IRQ %d\n", info->irq);
 		return 0;
 	}
 	evtchn = bind_pirq.port;
 
-	pirq_query_unmask(irq);
+	pirq_query_unmask(info);
 
-	rc = set_evtchn_to_irq(evtchn, irq);
+	rc = set_evtchn_to_irq(evtchn, info->irq);
 	if (rc)
 		goto err;
 
 	info->evtchn = evtchn;
-	bind_evtchn_to_cpu(evtchn, 0, false);
+	bind_evtchn_to_cpu(info, 0, false);
 
 	rc = xen_evtchn_port_setup(evtchn);
 	if (rc)
@@ -905,26 +894,28 @@ static unsigned int __startup_pirq(unsigned int irq)
 out:
 	do_unmask(info, EVT_MASK_REASON_EXPLICIT);
 
-	eoi_pirq(irq_get_irq_data(irq));
+	do_eoi_pirq(info);
 
 	return 0;
 
 err:
-	pr_err("irq%d: Failed to set port to irq mapping (%d)\n", irq, rc);
+	pr_err("irq%d: Failed to set port to irq mapping (%d)\n", info->irq,
+	       rc);
 	xen_evtchn_close(evtchn);
 	return 0;
 }
 
 static unsigned int startup_pirq(struct irq_data *data)
 {
-	return __startup_pirq(data->irq);
+	struct irq_info *info = info_for_irq(data->irq);
+
+	return __startup_pirq(info);
 }
 
 static void shutdown_pirq(struct irq_data *data)
 {
-	unsigned int irq = data->irq;
-	struct irq_info *info = info_for_irq(irq);
-	evtchn_port_t evtchn = evtchn_from_irq(irq);
+	struct irq_info *info = info_for_irq(data->irq);
+	evtchn_port_t evtchn = info->evtchn;
 
 	BUG_ON(info->type != IRQT_PIRQ);
 
@@ -932,8 +923,8 @@ static void shutdown_pirq(struct irq_data *data)
 		return;
 
 	do_mask(info, EVT_MASK_REASON_EXPLICIT);
-	xen_evtchn_close(evtchn);
 	xen_irq_info_cleanup(info);
+	xen_evtchn_close(evtchn);
 }
 
 static void enable_pirq(struct irq_data *data)
@@ -962,10 +953,15 @@ int xen_irq_from_gsi(unsigned gsi)
 }
 EXPORT_SYMBOL_GPL(xen_irq_from_gsi);
 
-static void __unbind_from_irq(unsigned int irq)
+static void __unbind_from_irq(struct irq_info *info, unsigned int irq)
 {
-	evtchn_port_t evtchn = evtchn_from_irq(irq);
-	struct irq_info *info = info_for_irq(irq);
+	evtchn_port_t evtchn;
+	bool close_evtchn = false;
+
+	if (!info) {
+		xen_irq_free_desc(irq);
+		return;
+	}
 
 	if (info->refcnt > 0) {
 		info->refcnt--;
@@ -973,20 +969,22 @@ static void __unbind_from_irq(unsigned int irq)
 			return;
 	}
 
+	evtchn = info->evtchn;
+
 	if (VALID_EVTCHN(evtchn)) {
-		unsigned int cpu = cpu_from_irq(irq);
+		unsigned int cpu = info->cpu;
 		struct xenbus_device *dev;
 
 		if (!info->is_static)
-			xen_evtchn_close(evtchn);
+			close_evtchn = true;
 
-		switch (type_from_irq(irq)) {
+		switch (info->type) {
 		case IRQT_VIRQ:
-			per_cpu(virq_to_irq, cpu)[virq_from_irq(irq)] = -1;
+			per_cpu(virq_to_irq, cpu)[virq_from_irq(info)] = -1;
 			break;
 		case IRQT_IPI:
-			per_cpu(ipi_to_irq, cpu)[ipi_from_irq(irq)] = -1;
-			per_cpu(ipi_to_evtchn, cpu)[ipi_from_irq(irq)] = 0;
+			per_cpu(ipi_to_irq, cpu)[ipi_from_irq(info)] = -1;
+			per_cpu(ipi_to_evtchn, cpu)[ipi_from_irq(info)] = 0;
 			break;
 		case IRQT_EVTCHN:
 			dev = info->u.interdomain;
@@ -998,9 +996,12 @@ static void __unbind_from_irq(unsigned int irq)
 		}
 
 		xen_irq_info_cleanup(info);
+
+		if (close_evtchn)
+			xen_evtchn_close(evtchn);
 	}
 
-	xen_free_irq(irq);
+	xen_free_irq(info);
 }
 
 /*
@@ -1016,24 +1017,24 @@ static void __unbind_from_irq(unsigned int irq)
 int xen_bind_pirq_gsi_to_irq(unsigned gsi,
 			     unsigned pirq, int shareable, char *name)
 {
-	int irq;
+	struct irq_info *info;
 	struct physdev_irq irq_op;
 	int ret;
 
 	mutex_lock(&irq_mapping_update_lock);
 
-	irq = xen_irq_from_gsi(gsi);
-	if (irq != -1) {
+	ret = xen_irq_from_gsi(gsi);
+	if (ret != -1) {
 		pr_info("%s: returning irq %d for gsi %u\n",
-			__func__, irq, gsi);
+			__func__, ret, gsi);
 		goto out;
 	}
 
-	irq = xen_allocate_irq_gsi(gsi);
-	if (irq < 0)
+	info = xen_allocate_irq_gsi(gsi);
+	if (!info)
 		goto out;
 
-	irq_op.irq = irq;
+	irq_op.irq = info->irq;
 	irq_op.vector = 0;
 
 	/* Only the privileged domain can do this. For non-priv, the pcifront
@@ -1041,20 +1042,19 @@ int xen_bind_pirq_gsi_to_irq(unsigned gsi,
 	 * this in the priv domain. */
 	if (xen_initial_domain() &&
 	    HYPERVISOR_physdev_op(PHYSDEVOP_alloc_irq_vector, &irq_op)) {
-		xen_free_irq(irq);
-		irq = -ENOSPC;
+		xen_free_irq(info);
+		ret = -ENOSPC;
 		goto out;
 	}
 
-	ret = xen_irq_info_pirq_setup(irq, 0, pirq, gsi, DOMID_SELF,
+	ret = xen_irq_info_pirq_setup(info, 0, pirq, gsi, DOMID_SELF,
 			       shareable ? PIRQ_SHAREABLE : 0);
 	if (ret < 0) {
-		__unbind_from_irq(irq);
-		irq = ret;
+		__unbind_from_irq(info, info->irq);
 		goto out;
 	}
 
-	pirq_query_unmask(irq);
+	pirq_query_unmask(info);
 	/* We try to use the handler with the appropriate semantic for the
 	 * type of interrupt: if the interrupt is an edge triggered
 	 * interrupt we use handle_edge_irq.
@@ -1071,16 +1071,18 @@ int xen_bind_pirq_gsi_to_irq(unsigned gsi,
 	 * is the right choice either way.
 	 */
 	if (shareable)
-		irq_set_chip_and_handler_name(irq, &xen_pirq_chip,
+		irq_set_chip_and_handler_name(info->irq, &xen_pirq_chip,
 				handle_fasteoi_irq, name);
 	else
-		irq_set_chip_and_handler_name(irq, &xen_pirq_chip,
+		irq_set_chip_and_handler_name(info->irq, &xen_pirq_chip,
 				handle_edge_irq, name);
 
+	ret = info->irq;
+
 out:
 	mutex_unlock(&irq_mapping_update_lock);
 
-	return irq;
+	return ret;
 }
 
 #ifdef CONFIG_PCI_MSI
@@ -1102,17 +1104,24 @@ int xen_bind_pirq_msi_to_irq(struct pci_dev *dev, struct msi_desc *msidesc,
 			     int pirq, int nvec, const char *name, domid_t domid)
 {
 	int i, irq, ret;
+	struct irq_info *info;
 
 	mutex_lock(&irq_mapping_update_lock);
 
-	irq = xen_allocate_irqs_dynamic(nvec);
+	irq = irq_alloc_descs(-1, 0, nvec, -1);
 	if (irq < 0)
 		goto out;
 
 	for (i = 0; i < nvec; i++) {
+		info = xen_irq_init(irq + i);
+		if (!info) {
+			ret = -ENOMEM;
+			goto error_irq;
+		}
+
 		irq_set_chip_and_handler_name(irq + i, &xen_pirq_chip, handle_edge_irq, name);
 
-		ret = xen_irq_info_pirq_setup(irq + i, 0, pirq + i, 0, domid,
+		ret = xen_irq_info_pirq_setup(info, 0, pirq + i, 0, domid,
 					      i == 0 ? 0 : PIRQ_MSI_GROUP);
 		if (ret < 0)
 			goto error_irq;
@@ -1124,9 +1133,12 @@ int xen_bind_pirq_msi_to_irq(struct pci_dev *dev, struct msi_desc *msidesc,
 out:
 	mutex_unlock(&irq_mapping_update_lock);
 	return irq;
+
 error_irq:
-	while (nvec--)
-		__unbind_from_irq(irq + nvec);
+	while (nvec--) {
+		info = info_for_irq(irq + nvec);
+		__unbind_from_irq(info, irq + nvec);
+	}
 	mutex_unlock(&irq_mapping_update_lock);
 	return ret;
 }
@@ -1162,67 +1174,45 @@ int xen_destroy_irq(int irq)
 		}
 	}
 
-	xen_free_irq(irq);
+	xen_free_irq(info);
 
 out:
 	mutex_unlock(&irq_mapping_update_lock);
 	return rc;
 }
 
-int xen_irq_from_pirq(unsigned pirq)
-{
-	int irq;
-
-	struct irq_info *info;
-
-	mutex_lock(&irq_mapping_update_lock);
-
-	list_for_each_entry(info, &xen_irq_list_head, list) {
-		if (info->type != IRQT_PIRQ)
-			continue;
-		irq = info->irq;
-		if (info->u.pirq.pirq == pirq)
-			goto out;
-	}
-	irq = -1;
-out:
-	mutex_unlock(&irq_mapping_update_lock);
-
-	return irq;
-}
-
-
 int xen_pirq_from_irq(unsigned irq)
 {
-	return pirq_from_irq(irq);
+	struct irq_info *info = info_for_irq(irq);
+
+	return pirq_from_irq(info);
 }
 EXPORT_SYMBOL_GPL(xen_pirq_from_irq);
 
 static int bind_evtchn_to_irq_chip(evtchn_port_t evtchn, struct irq_chip *chip,
 				   struct xenbus_device *dev)
 {
-	int irq;
-	int ret;
+	int ret = -ENOMEM;
+	struct irq_info *info;
 
 	if (evtchn >= xen_evtchn_max_channels())
 		return -ENOMEM;
 
 	mutex_lock(&irq_mapping_update_lock);
 
-	irq = get_evtchn_to_irq(evtchn);
+	info = evtchn_to_info(evtchn);
 
-	if (irq == -1) {
-		irq = xen_allocate_irq_dynamic();
-		if (irq < 0)
+	if (!info) {
+		info = xen_allocate_irq_dynamic();
+		if (!info)
 			goto out;
 
-		irq_set_chip_and_handler_name(irq, chip,
+		irq_set_chip_and_handler_name(info->irq, chip,
 					      handle_edge_irq, "event");
 
-		ret = xen_irq_info_evtchn_setup(irq, evtchn, dev);
+		ret = xen_irq_info_evtchn_setup(info, evtchn, dev);
 		if (ret < 0) {
-			__unbind_from_irq(irq);
-			irq = ret;
+			__unbind_from_irq(info, info->irq);
 			goto out;
 		}
 		/*
@@ -1232,16 +1222,17 @@ static int bind_evtchn_to_irq_chip(evtchn_port_t evtchn, struct irq_chip *chip,
 		 * affinity setting is not invoked on them so nothing would
 		 * bind the channel.
 		 */
-		bind_evtchn_to_cpu(evtchn, 0, false);
-	} else {
-		struct irq_info *info = info_for_irq(irq);
-		WARN_ON(info == NULL || info->type != IRQT_EVTCHN);
+		bind_evtchn_to_cpu(info, 0, false);
+	} else if (!WARN_ON(info->type != IRQT_EVTCHN)) {
+		info->refcnt++;
 	}
 
+	ret = info->irq;
+
 out:
 	mutex_unlock(&irq_mapping_update_lock);
 
-	return irq;
+	return ret;
 }
 
 int bind_evtchn_to_irq(evtchn_port_t evtchn)
@@ -1260,18 +1251,19 @@ static int bind_ipi_to_irq(unsigned int ipi, unsigned int cpu)
 {
 	struct evtchn_bind_ipi bind_ipi;
 	evtchn_port_t evtchn;
-	int ret, irq;
+	struct irq_info *info;
+	int ret;
 
 	mutex_lock(&irq_mapping_update_lock);
 
-	irq = per_cpu(ipi_to_irq, cpu)[ipi];
+	ret = per_cpu(ipi_to_irq, cpu)[ipi];
 
-	if (irq == -1) {
-		irq = xen_allocate_irq_dynamic();
-		if (irq < 0)
+	if (ret == -1) {
+		info = xen_allocate_irq_dynamic();
+		if (!info)
 			goto out;
 
-		irq_set_chip_and_handler_name(irq, &xen_percpu_chip,
+		irq_set_chip_and_handler_name(info->irq, &xen_percpu_chip,
 					      handle_percpu_irq, "ipi");
 
 		bind_ipi.vcpu = xen_vcpu_nr(cpu);
@@ -1280,25 +1272,25 @@ static int bind_ipi_to_irq(unsigned int ipi, unsigned int cpu)
 			BUG();
 		evtchn = bind_ipi.port;
 
-		ret = xen_irq_info_ipi_setup(cpu, irq, evtchn, ipi);
+		ret = xen_irq_info_ipi_setup(info, cpu, evtchn, ipi);
 		if (ret < 0) {
-			__unbind_from_irq(irq);
-			irq = ret;
+			__unbind_from_irq(info, info->irq);
 			goto out;
 		}
 		/*
 		 * Force the affinity mask to the target CPU so proc shows
 		 * the correct target.
 		 */
-		bind_evtchn_to_cpu(evtchn, cpu, true);
+		bind_evtchn_to_cpu(info, cpu, true);
+		ret = info->irq;
 	} else {
-		struct irq_info *info = info_for_irq(irq);
+		info = info_for_irq(ret);
 		WARN_ON(info == NULL || info->type != IRQT_IPI);
 	}
 
  out:
 	mutex_unlock(&irq_mapping_update_lock);
-	return irq;
+	return ret;
 }
 
 static int bind_interdomain_evtchn_to_irq_chip(struct xenbus_device *dev,
@@ -1366,22 +1358,23 @@ int bind_virq_to_irq(unsigned int virq, unsigned int cpu, bool percpu)
 {
 	struct evtchn_bind_virq bind_virq;
 	evtchn_port_t evtchn = 0;
-	int irq, ret;
+	struct irq_info *info;
+	int ret;
 
 	mutex_lock(&irq_mapping_update_lock);
 
-	irq = per_cpu(virq_to_irq, cpu)[virq];
+	ret = per_cpu(virq_to_irq, cpu)[virq];
 
-	if (irq == -1) {
-		irq = xen_allocate_irq_dynamic();
-		if (irq < 0)
+	if (ret == -1) {
+		info = xen_allocate_irq_dynamic();
+		if (!info)
 			goto out;
 
 		if (percpu)
-			irq_set_chip_and_handler_name(irq, &xen_percpu_chip,
+			irq_set_chip_and_handler_name(info->irq, &xen_percpu_chip,
 						      handle_percpu_irq, "virq");
 		else
-			irq_set_chip_and_handler_name(irq, &xen_dynamic_chip,
+			irq_set_chip_and_handler_name(info->irq, &xen_dynamic_chip,
 						      handle_edge_irq, "virq");
 
 		bind_virq.virq = virq;
@@ -1396,10 +1389,9 @@ int bind_virq_to_irq(unsigned int virq, unsigned int cpu, bool percpu)
 			BUG_ON(ret < 0);
 		}
 
-		ret = xen_irq_info_virq_setup(cpu, irq, evtchn, virq);
+		ret = xen_irq_info_virq_setup(info, cpu, evtchn, virq);
 		if (ret < 0) {
-			__unbind_from_irq(irq);
-			irq = ret;
+			__unbind_from_irq(info, info->irq);
 			goto out;
 		}
 
@@ -1407,22 +1399,26 @@ int bind_virq_to_irq(unsigned int virq, unsigned int cpu, bool percpu)
 		 * Force the affinity mask for percpu interrupts so proc
 		 * shows the correct target.
 		 */
-		bind_evtchn_to_cpu(evtchn, cpu, percpu);
+		bind_evtchn_to_cpu(info, cpu, percpu);
+		ret = info->irq;
 	} else {
-		struct irq_info *info = info_for_irq(irq);
+		info = info_for_irq(ret);
 		WARN_ON(info == NULL || info->type != IRQT_VIRQ);
 	}
 
 out:
 	mutex_unlock(&irq_mapping_update_lock);
 
-	return irq;
+	return ret;
 }
 
 static void unbind_from_irq(unsigned int irq)
 {
+	struct irq_info *info;
+
 	mutex_lock(&irq_mapping_update_lock);
-	__unbind_from_irq(irq);
+	info = info_for_irq(irq);
+	__unbind_from_irq(info, irq);
 	mutex_unlock(&irq_mapping_update_lock);
 }
 
@@ -1573,13 +1569,7 @@ EXPORT_SYMBOL_GPL(xen_set_irq_priority);
 
 int evtchn_make_refcounted(evtchn_port_t evtchn, bool is_static)
 {
-	int irq = get_evtchn_to_irq(evtchn);
-	struct irq_info *info;
-
-	if (irq == -1)
-		return -ENOENT;
-
-	info = info_for_irq(irq);
+	struct irq_info *info = evtchn_to_info(evtchn);
 
 	if (!info)
 		return -ENOENT;
@@ -1595,7 +1585,6 @@ EXPORT_SYMBOL_GPL(evtchn_make_refcounted);
 
 int evtchn_get(evtchn_port_t evtchn)
 {
-	int irq;
 	struct irq_info *info;
 	int err = -ENOENT;
 
@@ -1604,11 +1593,7 @@ int evtchn_get(evtchn_port_t evtchn)
 
 	mutex_lock(&irq_mapping_update_lock);
 
-	irq = get_evtchn_to_irq(evtchn);
-	if (irq == -1)
-		goto done;
-
-	info = info_for_irq(irq);
+	info = evtchn_to_info(evtchn);
 
 	if (!info)
 		goto done;
@@ -1628,10 +1613,11 @@ EXPORT_SYMBOL_GPL(evtchn_get);
 
 void evtchn_put(evtchn_port_t evtchn)
 {
-	int irq = get_evtchn_to_irq(evtchn);
-	if (WARN_ON(irq == -1))
+	struct irq_info *info = evtchn_to_info(evtchn);
+
+	if (WARN_ON(!info))
 		return;
-	unbind_from_irq(irq);
+	unbind_from_irq(info->irq);
 }
 EXPORT_SYMBOL_GPL(evtchn_put);
 
@@ -1661,12 +1647,10 @@ struct evtchn_loop_ctrl {
 
 void handle_irq_for_port(evtchn_port_t port, struct evtchn_loop_ctrl *ctrl)
 {
-	int irq;
-	struct irq_info *info;
+	struct irq_info *info = evtchn_to_info(port);
 	struct xenbus_device *dev;
 
-	irq = get_evtchn_to_irq(port);
-	if (irq == -1)
+	if (!info)
 		return;
 
 	/*
@@ -1691,7 +1675,6 @@ void handle_irq_for_port(evtchn_port_t port, struct evtchn_loop_ctrl *ctrl)
 		}
 	}
 
-	info = info_for_irq(irq);
 	if (xchg_acquire(&info->is_active, 1))
 		return;
 
@@ -1705,7 +1688,7 @@ void handle_irq_for_port(evtchn_port_t port, struct evtchn_loop_ctrl *ctrl)
 		info->eoi_time = get_jiffies_64() + event_eoi_delay;
 	}
 
-	generic_handle_irq(irq);
+	generic_handle_irq(info->irq);
 }
 
 int xen_evtchn_do_upcall(void)
@@ -1763,16 +1746,17 @@ void rebind_evtchn_irq(evtchn_port_t evtchn, int irq)
 	mutex_lock(&irq_mapping_update_lock);
 
 	/* After resume the irq<->evtchn mappings are all cleared out */
-	BUG_ON(get_evtchn_to_irq(evtchn) != -1);
+	BUG_ON(evtchn_to_info(evtchn));
 	/* Expect irq to have been bound before,
 	   so there should be a proper type */
 	BUG_ON(info->type == IRQT_UNBOUND);
 
-	(void)xen_irq_info_evtchn_setup(irq, evtchn, NULL);
+	info->irq = irq;
+	(void)xen_irq_info_evtchn_setup(info, evtchn, NULL);
 
 	mutex_unlock(&irq_mapping_update_lock);
 
-	bind_evtchn_to_cpu(evtchn, info->cpu, false);
+	bind_evtchn_to_cpu(info, info->cpu, false);
 
 	/* Unmask the event channel. */
 	enable_irq(irq);
@@ -1806,7 +1790,7 @@ static int xen_rebind_evtchn_to_cpu(struct irq_info *info, unsigned int tcpu)
 	 * it, but don't do the xenlinux-level rebind in that case.
 	 */
 	if (HYPERVISOR_event_channel_op(EVTCHNOP_bind_vcpu, &bind_vcpu) >= 0)
-		bind_evtchn_to_cpu(evtchn, tcpu, false);
+		bind_evtchn_to_cpu(info, tcpu, false);
 
 	do_unmask(info, EVT_MASK_REASON_TEMPORARY);
 
@@ -1863,28 +1847,30 @@ static void enable_dynirq(struct irq_data *data)
 		do_unmask(info, EVT_MASK_REASON_EXPLICIT);
 }
 
-static void disable_dynirq(struct irq_data *data)
+static void do_ack_dynirq(struct irq_info *info)
 {
-	struct irq_info *info = info_for_irq(data->irq);
-	evtchn_port_t evtchn = info ? info->evtchn : 0;
+	evtchn_port_t evtchn = info->evtchn;
 
 	if (VALID_EVTCHN(evtchn))
-		do_mask(info, EVT_MASK_REASON_EXPLICIT);
+		event_handler_exit(info);
 }
 
 static void ack_dynirq(struct irq_data *data)
 {
 	struct irq_info *info = info_for_irq(data->irq);
-	evtchn_port_t evtchn = info ? info->evtchn : 0;
 
-	if (VALID_EVTCHN(evtchn))
-		event_handler_exit(info);
+	if (info)
+		do_ack_dynirq(info);
 }
 
 static void mask_ack_dynirq(struct irq_data *data)
 {
-	disable_dynirq(data);
-	ack_dynirq(data);
+	struct irq_info *info = info_for_irq(data->irq);
+
+	if (info) {
+		do_disable_dynirq(info);
+		do_ack_dynirq(info);
+	}
 }
 
 static void lateeoi_ack_dynirq(struct irq_data *data)
@@ -1957,13 +1943,13 @@ static void restore_pirqs(void)
 		if (rc) {
 			pr_warn("xen map irq failed gsi=%d irq=%d pirq=%d rc=%d\n",
 				gsi, irq, pirq, rc);
-			xen_free_irq(irq);
+			xen_free_irq(info);
 			continue;
 		}
 
 		printk(KERN_DEBUG "xen: --> irq=%d, pirq=%d\n", irq, map_irq.pirq);
 
-		__startup_pirq(irq);
+		__startup_pirq(info);
 	}
 }
 
@@ -1971,13 +1957,15 @@ static void restore_cpu_virqs(unsigned int cpu)
 {
 	struct evtchn_bind_virq bind_virq;
 	evtchn_port_t evtchn;
+	struct irq_info *info;
 	int virq, irq;
 
 	for (virq = 0; virq < NR_VIRQS; virq++) {
 		if ((irq = per_cpu(virq_to_irq, cpu)[virq]) == -1)
 			continue;
+		info = info_for_irq(irq);
 
-		BUG_ON(virq_from_irq(irq) != virq);
+		BUG_ON(virq_from_irq(info) != virq);
 
 		/* Get a new binding from Xen. */
 		bind_virq.virq = virq;
@@ -1988,9 +1976,9 @@ static void restore_cpu_virqs(unsigned int cpu)
 		evtchn = bind_virq.port;
 
 		/* Record the new mapping. */
-		(void)xen_irq_info_virq_setup(cpu, irq, evtchn, virq);
+		xen_irq_info_virq_setup(info, cpu, evtchn, virq);
 		/* The affinity mask is still valid */
-		bind_evtchn_to_cpu(evtchn, cpu, false);
+		bind_evtchn_to_cpu(info, cpu, false);
 	}
 }
 
@@ -1998,13 +1986,15 @@ static void restore_cpu_ipis(unsigned int cpu)
 {
 	struct evtchn_bind_ipi bind_ipi;
 	evtchn_port_t evtchn;
+	struct irq_info *info;
 	int ipi, irq;
 
 	for (ipi = 0; ipi < XEN_NR_IPIS; ipi++) {
 		if ((irq = per_cpu(ipi_to_irq, cpu)[ipi]) == -1)
 			continue;
+		info = info_for_irq(irq);
 
-		BUG_ON(ipi_from_irq(irq) != ipi);
+		BUG_ON(ipi_from_irq(info) != ipi);
 
 		/* Get a new binding from Xen. */
 		bind_ipi.vcpu = xen_vcpu_nr(cpu);
@@ -2014,9 +2004,9 @@ static void restore_cpu_ipis(unsigned int cpu)
 		evtchn = bind_ipi.port;
 
 		/* Record the new mapping. */
-		(void)xen_irq_info_ipi_setup(cpu, irq, evtchn, ipi);
+		xen_irq_info_ipi_setup(info, cpu, evtchn, ipi);
 		/* The affinity mask is still valid */
-		bind_evtchn_to_cpu(evtchn, cpu, false);
+		bind_evtchn_to_cpu(info, cpu, false);
 	}
 }
 
@@ -2030,13 +2020,6 @@ void xen_clear_irq_pending(int irq)
 		event_handler_exit(info);
 }
 EXPORT_SYMBOL(xen_clear_irq_pending);
-void xen_set_irq_pending(int irq)
-{
-	evtchn_port_t evtchn = evtchn_from_irq(irq);
-
-	if (VALID_EVTCHN(evtchn))
-		set_evtchn(evtchn);
-}
 
 bool xen_test_irq_pending(int irq)
 {
diff --git a/drivers/xen/events/events_fifo.c b/drivers/xen/events/events_fifo.c
index ad9fe51d3f..655775db7c 100644
--- a/drivers/xen/events/events_fifo.c
+++ b/drivers/xen/events/events_fifo.c
@@ -226,21 +226,20 @@ static bool evtchn_fifo_is_masked(evtchn_port_t port)
  */
 static bool clear_masked_cond(volatile event_word_t *word)
 {
-	event_word_t new, old, w;
+	event_word_t new, old;
 
-	w = *word;
+	old = *word;
 
 	do {
-		if (!(w & (1 << EVTCHN_FIFO_MASKED)))
+		if (!(old & (1 << EVTCHN_FIFO_MASKED)))
 			return true;
 
-		if (w & (1 << EVTCHN_FIFO_PENDING))
+		if (old & (1 << EVTCHN_FIFO_PENDING))
 			return false;
 
-		old = w & ~(1 << EVTCHN_FIFO_BUSY);
+		old = old & ~(1 << EVTCHN_FIFO_BUSY);
 		new = old & ~(1 << EVTCHN_FIFO_MASKED);
-		w = sync_cmpxchg(word, old, new);
-	} while (w != old);
+	} while (!sync_try_cmpxchg(word, &old, new));
 
 	return true;
 }
@@ -259,17 +258,16 @@ static void evtchn_fifo_unmask(evtchn_port_t port)
 
 static uint32_t clear_linked(volatile event_word_t *word)
 {
-	event_word_t new, old, w;
+	event_word_t new, old;
 
-	w = *word;
+	old = *word;
 
 	do {
-		old = w;
-		new = (w & ~((1 << EVTCHN_FIFO_LINKED)
-			     | EVTCHN_FIFO_LINK_MASK));
-	} while ((w = sync_cmpxchg(word, old, new)) != old);
+		new = (old & ~((1 << EVTCHN_FIFO_LINKED)
+			       | EVTCHN_FIFO_LINK_MASK));
+	} while (!sync_try_cmpxchg(word, &old, new));
 
-	return w & EVTCHN_FIFO_LINK_MASK;
+	return old & EVTCHN_FIFO_LINK_MASK;
 }
 
 static void consume_one_event(unsigned cpu, struct evtchn_loop_ctrl *ctrl,
diff --git a/drivers/xen/events/events_internal.h b/drivers/xen/events/events_internal.h
index 4d3398eff9..19ae31695e 100644
--- a/drivers/xen/events/events_internal.h
+++ b/drivers/xen/events/events_internal.h
@@ -33,7 +33,6 @@ struct evtchn_ops {
 
 extern const struct evtchn_ops *evtchn_ops;
 
-int get_evtchn_to_irq(evtchn_port_t evtchn);
 void handle_irq_for_port(evtchn_port_t port, struct evtchn_loop_ctrl *ctrl);
 
 unsigned int cpu_from_evtchn(evtchn_port_t evtchn);
diff --git a/drivers/xen/evtchn.c b/drivers/xen/evtchn.c
index 9139a7364d..59717628ca 100644
--- a/drivers/xen/evtchn.c
+++ b/drivers/xen/evtchn.c
@@ -397,7 +397,7 @@ static int evtchn_bind_to_user(struct per_user_data *u, evtchn_port_t port,
 	if (rc < 0)
 		goto err;
 
-	rc = bind_evtchn_to_irqhandler_lateeoi(port, evtchn_interrupt, 0,
+	rc = bind_evtchn_to_irqhandler_lateeoi(port, evtchn_interrupt, IRQF_SHARED,
 					       u->name, evtchn);
 	if (rc < 0)
 		goto err;
diff --git a/drivers/xen/gntdev-dmabuf.c b/drivers/xen/gntdev-dmabuf.c
index 4440e626b7..42adc2c1e0 100644
--- a/drivers/xen/gntdev-dmabuf.c
+++ b/drivers/xen/gntdev-dmabuf.c
@@ -11,6 +11,7 @@
 #include <linux/kernel.h>
 #include <linux/errno.h>
 #include <linux/dma-buf.h>
+#include <linux/dma-direct.h>
 #include <linux/slab.h>
 #include <linux/types.h>
 #include <linux/uaccess.h>
@@ -50,7 +51,7 @@ struct gntdev_dmabuf {
 
 	/* Number of pages this buffer has. */
 	int nr_pages;
-	/* Pages of this buffer. */
+	/* Pages of this buffer (only for dma-buf export). */
 	struct page **pages;
 };
 
@@ -484,7 +485,7 @@ out:
 /* DMA buffer import support. */
 
 static int
-dmabuf_imp_grant_foreign_access(struct page **pages, u32 *refs,
+dmabuf_imp_grant_foreign_access(unsigned long *gfns, u32 *refs,
 				int count, int domid)
 {
 	grant_ref_t priv_gref_head;
@@ -507,7 +508,7 @@ dmabuf_imp_grant_foreign_access(struct page **pages, u32 *refs,
 		}
 
 		gnttab_grant_foreign_access_ref(cur_ref, domid,
-						xen_page_to_gfn(pages[i]), 0);
+						gfns[i], 0);
 		refs[i] = cur_ref;
 	}
 
@@ -529,7 +530,6 @@ static void dmabuf_imp_end_foreign_access(u32 *refs, int count)
 
 static void dmabuf_imp_free_storage(struct gntdev_dmabuf *gntdev_dmabuf)
 {
-	kfree(gntdev_dmabuf->pages);
 	kfree(gntdev_dmabuf->u.imp.refs);
 	kfree(gntdev_dmabuf);
 }
@@ -549,12 +549,6 @@ static struct gntdev_dmabuf *dmabuf_imp_alloc_storage(int count)
 	if (!gntdev_dmabuf->u.imp.refs)
 		goto fail;
 
-	gntdev_dmabuf->pages = kcalloc(count,
-				       sizeof(gntdev_dmabuf->pages[0]),
-				       GFP_KERNEL);
-	if (!gntdev_dmabuf->pages)
-		goto fail;
-
 	gntdev_dmabuf->nr_pages = count;
 
 	for (i = 0; i < count; i++)
@@ -576,7 +570,8 @@ dmabuf_imp_to_refs(struct gntdev_dmabuf_priv *priv, struct device *dev,
 	struct dma_buf *dma_buf;
 	struct dma_buf_attachment *attach;
 	struct sg_table *sgt;
-	struct sg_page_iter sg_iter;
+	struct sg_dma_page_iter sg_iter;
+	unsigned long *gfns;
 	int i;
 
 	dma_buf = dma_buf_get(fd);
@@ -624,26 +619,31 @@ dmabuf_imp_to_refs(struct gntdev_dmabuf_priv *priv, struct device *dev,
 
 	gntdev_dmabuf->u.imp.sgt = sgt;
 
-	/* Now convert sgt to array of pages and check for page validity. */
+	gfns = kcalloc(count, sizeof(*gfns), GFP_KERNEL);
+	if (!gfns) {
+		ret = ERR_PTR(-ENOMEM);
+		goto fail_unmap;
+	}
+
+	/*
+	 * Now convert sgt to array of gfns without accessing underlying pages.
+	 * It is not allowed to access the underlying struct page of an sg table
+	 * exported by DMA-buf, but since we deal with special Xen dma device here
+	 * (not a normal physical one) look at the dma addresses in the sg table
+	 * and then calculate gfns directly from them.
+	 */
 	i = 0;
-	for_each_sgtable_page(sgt, &sg_iter, 0) {
-		struct page *page = sg_page_iter_page(&sg_iter);
-		/*
-		 * Check if page is valid: this can happen if we are given
-		 * a page from VRAM or other resources which are not backed
-		 * by a struct page.
-		 */
-		if (!pfn_valid(page_to_pfn(page))) {
-			ret = ERR_PTR(-EINVAL);
-			goto fail_unmap;
-		}
+	for_each_sgtable_dma_page(sgt, &sg_iter, 0) {
+		dma_addr_t addr = sg_page_iter_dma_address(&sg_iter);
+		unsigned long pfn = bfn_to_pfn(XEN_PFN_DOWN(dma_to_phys(dev, addr)));
 
-		gntdev_dmabuf->pages[i++] = page;
+		gfns[i++] = pfn_to_gfn(pfn);
 	}
 
-	ret = ERR_PTR(dmabuf_imp_grant_foreign_access(gntdev_dmabuf->pages,
+	ret = ERR_PTR(dmabuf_imp_grant_foreign_access(gfns,
 						      gntdev_dmabuf->u.imp.refs,
 						      count, domid));
+	kfree(gfns);
 	if (IS_ERR(ret))
 		goto fail_end_access;
 
diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c
index 35659bf707..04a6b470b1 100644
--- a/drivers/xen/grant-table.c
+++ b/drivers/xen/grant-table.c
@@ -427,16 +427,14 @@ EXPORT_SYMBOL_GPL(gnttab_grant_foreign_access);
 
 static int gnttab_end_foreign_access_ref_v1(grant_ref_t ref)
 {
-	u16 flags, nflags;
-	u16 *pflags;
+	u16 *pflags = &gnttab_shared.v1[ref].flags;
+	u16 flags;
 
-	pflags = &gnttab_shared.v1[ref].flags;
-	nflags = *pflags;
+	flags = *pflags;
 	do {
-		flags = nflags;
 		if (flags & (GTF_reading|GTF_writing))
 			return 0;
-	} while ((nflags = sync_cmpxchg(pflags, flags, 0)) != flags);
+	} while (!sync_try_cmpxchg(pflags, &flags, 0));
 
 	return 1;
 }
diff --git a/drivers/xen/privcmd.c b/drivers/xen/privcmd.c
index da88173bac..0eb337a8ec 100644
--- a/drivers/xen/privcmd.c
+++ b/drivers/xen/privcmd.c
@@ -29,15 +29,18 @@
 #include <linux/seq_file.h>
 #include <linux/miscdevice.h>
 #include <linux/moduleparam.h>
+#include <linux/virtio_mmio.h>
 
 #include <asm/xen/hypervisor.h>
 #include <asm/xen/hypercall.h>
 
 #include <xen/xen.h>
+#include <xen/events.h>
 #include <xen/privcmd.h>
 #include <xen/interface/xen.h>
 #include <xen/interface/memory.h>
 #include <xen/interface/hvm/dm_op.h>
+#include <xen/interface/hvm/ioreq.h>
 #include <xen/features.h>
 #include <xen/page.h>
 #include <xen/xen-ops.h>
@@ -782,6 +785,7 @@ static long privcmd_ioctl_mmap_resource(struct file *file,
 			goto out;
 
 		pages = vma->vm_private_data;
+
 		for (i = 0; i < kdata.num; i++) {
 			xen_pfn_t pfn =
 				page_to_xen_pfn(pages[i / XEN_PFN_PER_PAGE]);
@@ -838,7 +842,7 @@ out:
 	return rc;
 }
 
-#ifdef CONFIG_XEN_PRIVCMD_IRQFD
+#ifdef CONFIG_XEN_PRIVCMD_EVENTFD
 /* Irqfd support */
 static struct workqueue_struct *irqfd_cleanup_wq;
 static DEFINE_MUTEX(irqfds_lock);
@@ -1079,6 +1083,389 @@ static void privcmd_irqfd_exit(void)
 
 	destroy_workqueue(irqfd_cleanup_wq);
 }
+
+/* Ioeventfd Support */
+#define QUEUE_NOTIFY_VQ_MASK 0xFFFF
+
+static DEFINE_MUTEX(ioreq_lock);
+static LIST_HEAD(ioreq_list);
+
+/* per-eventfd structure */
+struct privcmd_kernel_ioeventfd {
+	struct eventfd_ctx *eventfd;
+	struct list_head list;
+	u64 addr;
+	unsigned int addr_len;
+	unsigned int vq;
+};
+
+/* per-guest CPU / port structure */
+struct ioreq_port {
+	int vcpu;
+	unsigned int port;
+	struct privcmd_kernel_ioreq *kioreq;
+};
+
+/* per-guest structure */
+struct privcmd_kernel_ioreq {
+	domid_t dom;
+	unsigned int vcpus;
+	u64 uioreq;
+	struct ioreq *ioreq;
+	spinlock_t lock; /* Protects ioeventfds list */
+	struct list_head ioeventfds;
+	struct list_head list;
+	struct ioreq_port ports[] __counted_by(vcpus);
+};
+
+static irqreturn_t ioeventfd_interrupt(int irq, void *dev_id)
+{
+	struct ioreq_port *port = dev_id;
+	struct privcmd_kernel_ioreq *kioreq = port->kioreq;
+	struct ioreq *ioreq = &kioreq->ioreq[port->vcpu];
+	struct privcmd_kernel_ioeventfd *kioeventfd;
+	unsigned int state = STATE_IOREQ_READY;
+
+	if (ioreq->state != STATE_IOREQ_READY ||
+	    ioreq->type != IOREQ_TYPE_COPY || ioreq->dir != IOREQ_WRITE)
+		return IRQ_NONE;
+
+	/*
+	 * We need a barrier, smp_mb(), here to ensure reads are finished before
+	 * `state` is updated. Since the lock implementation ensures that
+	 * appropriate barrier will be added anyway, we can avoid adding
+	 * explicit barrier here.
+	 *
+	 * Ideally we don't need to update `state` within the locks, but we do
+	 * that here to avoid adding explicit barrier.
+	 */
+
+	spin_lock(&kioreq->lock);
+	ioreq->state = STATE_IOREQ_INPROCESS;
+
+	list_for_each_entry(kioeventfd, &kioreq->ioeventfds, list) {
+		if (ioreq->addr == kioeventfd->addr + VIRTIO_MMIO_QUEUE_NOTIFY &&
+		    ioreq->size == kioeventfd->addr_len &&
+		    (ioreq->data & QUEUE_NOTIFY_VQ_MASK) == kioeventfd->vq) {
+			eventfd_signal(kioeventfd->eventfd, 1);
+			state = STATE_IORESP_READY;
+			break;
+		}
+	}
+	spin_unlock(&kioreq->lock);
+
+	/*
+	 * We need a barrier, smp_mb(), here to ensure writes are finished
+	 * before `state` is updated. Since the lock implementation ensures that
+	 * appropriate barrier will be added anyway, we can avoid adding
+	 * explicit barrier here.
+	 */
+
+	ioreq->state = state;
+
+	if (state == STATE_IORESP_READY) {
+		notify_remote_via_evtchn(port->port);
+		return IRQ_HANDLED;
+	}
+
+	return IRQ_NONE;
+}
+
+static void ioreq_free(struct privcmd_kernel_ioreq *kioreq)
+{
+	struct ioreq_port *ports = kioreq->ports;
+	int i;
+
+	lockdep_assert_held(&ioreq_lock);
+
+	list_del(&kioreq->list);
+
+	for (i = kioreq->vcpus - 1; i >= 0; i--)
+		unbind_from_irqhandler(irq_from_evtchn(ports[i].port), &ports[i]);
+
+	kfree(kioreq);
+}
+
+static
+struct privcmd_kernel_ioreq *alloc_ioreq(struct privcmd_ioeventfd *ioeventfd)
+{
+	struct privcmd_kernel_ioreq *kioreq;
+	struct mm_struct *mm = current->mm;
+	struct vm_area_struct *vma;
+	struct page **pages;
+	unsigned int *ports;
+	int ret, size, i;
+
+	lockdep_assert_held(&ioreq_lock);
+
+	size = struct_size(kioreq, ports, ioeventfd->vcpus);
+	kioreq = kzalloc(size, GFP_KERNEL);
+	if (!kioreq)
+		return ERR_PTR(-ENOMEM);
+
+	kioreq->dom = ioeventfd->dom;
+	kioreq->vcpus = ioeventfd->vcpus;
+	kioreq->uioreq = ioeventfd->ioreq;
+	spin_lock_init(&kioreq->lock);
+	INIT_LIST_HEAD(&kioreq->ioeventfds);
+
+	/* The memory for ioreq server must have been mapped earlier */
+	mmap_write_lock(mm);
+	vma = find_vma(mm, (unsigned long)ioeventfd->ioreq);
+	if (!vma) {
+		pr_err("Failed to find vma for ioreq page!\n");
+		mmap_write_unlock(mm);
+		ret = -EFAULT;
+		goto error_kfree;
+	}
+
+	pages = vma->vm_private_data;
+	kioreq->ioreq = (struct ioreq *)(page_to_virt(pages[0]));
+	mmap_write_unlock(mm);
+
+	size = sizeof(*ports) * kioreq->vcpus;
+	ports = kzalloc(size, GFP_KERNEL);
+	if (!ports) {
+		ret = -ENOMEM;
+		goto error_kfree;
+	}
+
+	if (copy_from_user(ports, u64_to_user_ptr(ioeventfd->ports), size)) {
+		ret = -EFAULT;
+		goto error_kfree_ports;
+	}
+
+	for (i = 0; i < kioreq->vcpus; i++) {
+		kioreq->ports[i].vcpu = i;
+		kioreq->ports[i].port = ports[i];
+		kioreq->ports[i].kioreq = kioreq;
+
+		ret = bind_evtchn_to_irqhandler_lateeoi(ports[i],
+				ioeventfd_interrupt, IRQF_SHARED, "ioeventfd",
+				&kioreq->ports[i]);
+		if (ret < 0)
+			goto error_unbind;
+	}
+
+	kfree(ports);
+
+	list_add_tail(&kioreq->list, &ioreq_list);
+
+	return kioreq;
+
+error_unbind:
+	while (--i >= 0)
+		unbind_from_irqhandler(irq_from_evtchn(ports[i]), &kioreq->ports[i]);
+error_kfree_ports:
+	kfree(ports);
+error_kfree:
+	kfree(kioreq);
+	return ERR_PTR(ret);
+}
+
+static struct privcmd_kernel_ioreq *
+get_ioreq(struct privcmd_ioeventfd *ioeventfd, struct eventfd_ctx *eventfd)
+{
+	struct privcmd_kernel_ioreq *kioreq;
+	unsigned long flags;
+
+	list_for_each_entry(kioreq, &ioreq_list, list) {
+		struct privcmd_kernel_ioeventfd *kioeventfd;
+
+		/*
+		 * kioreq fields can be accessed here without a lock as they are
+		 * never updated after being added to the ioreq_list.
+		 */
+		if (kioreq->uioreq != ioeventfd->ioreq) {
+			continue;
+		} else if (kioreq->dom != ioeventfd->dom ||
+			   kioreq->vcpus != ioeventfd->vcpus) {
+			pr_err("Invalid ioeventfd configuration mismatch, dom (%u vs %u), vcpus (%u vs %u)\n",
+			       kioreq->dom, ioeventfd->dom, kioreq->vcpus,
+			       ioeventfd->vcpus);
+			return ERR_PTR(-EINVAL);
+		}
+
+		/* Look for a duplicate eventfd for the same guest */
+		spin_lock_irqsave(&kioreq->lock, flags);
+		list_for_each_entry(kioeventfd, &kioreq->ioeventfds, list) {
+			if (eventfd == kioeventfd->eventfd) {
+				spin_unlock_irqrestore(&kioreq->lock, flags);
+				return ERR_PTR(-EBUSY);
+			}
+		}
+		spin_unlock_irqrestore(&kioreq->lock, flags);
+
+		return kioreq;
+	}
+
+	/* Matching kioreq isn't found, allocate a new one */
+	return alloc_ioreq(ioeventfd);
+}
+
+static void ioeventfd_free(struct privcmd_kernel_ioeventfd *kioeventfd)
+{
+	list_del(&kioeventfd->list);
+	eventfd_ctx_put(kioeventfd->eventfd);
+	kfree(kioeventfd);
+}
+
+static int privcmd_ioeventfd_assign(struct privcmd_ioeventfd *ioeventfd)
+{
+	struct privcmd_kernel_ioeventfd *kioeventfd;
+	struct privcmd_kernel_ioreq *kioreq;
+	unsigned long flags;
+	struct fd f;
+	int ret;
+
+	/* Check for range overflow */
+	if (ioeventfd->addr + ioeventfd->addr_len < ioeventfd->addr)
+		return -EINVAL;
+
+	/* Vhost requires us to support length 1, 2, 4, and 8 */
+	if (!(ioeventfd->addr_len == 1 || ioeventfd->addr_len == 2 ||
+	      ioeventfd->addr_len == 4 || ioeventfd->addr_len == 8))
+		return -EINVAL;
+
+	/* 4096 vcpus limit enough ? */
+	if (!ioeventfd->vcpus || ioeventfd->vcpus > 4096)
+		return -EINVAL;
+
+	kioeventfd = kzalloc(sizeof(*kioeventfd), GFP_KERNEL);
+	if (!kioeventfd)
+		return -ENOMEM;
+
+	f = fdget(ioeventfd->event_fd);
+	if (!f.file) {
+		ret = -EBADF;
+		goto error_kfree;
+	}
+
+	kioeventfd->eventfd = eventfd_ctx_fileget(f.file);
+	fdput(f);
+
+	if (IS_ERR(kioeventfd->eventfd)) {
+		ret = PTR_ERR(kioeventfd->eventfd);
+		goto error_kfree;
+	}
+
+	kioeventfd->addr = ioeventfd->addr;
+	kioeventfd->addr_len = ioeventfd->addr_len;
+	kioeventfd->vq = ioeventfd->vq;
+
+	mutex_lock(&ioreq_lock);
+	kioreq = get_ioreq(ioeventfd, kioeventfd->eventfd);
+	if (IS_ERR(kioreq)) {
+		mutex_unlock(&ioreq_lock);
+		ret = PTR_ERR(kioreq);
+		goto error_eventfd;
+	}
+
+	spin_lock_irqsave(&kioreq->lock, flags);
+	list_add_tail(&kioeventfd->list, &kioreq->ioeventfds);
+	spin_unlock_irqrestore(&kioreq->lock, flags);
+
+	mutex_unlock(&ioreq_lock);
+
+	return 0;
+
+error_eventfd:
+	eventfd_ctx_put(kioeventfd->eventfd);
+
+error_kfree:
+	kfree(kioeventfd);
+	return ret;
+}
+
+static int privcmd_ioeventfd_deassign(struct privcmd_ioeventfd *ioeventfd)
+{
+	struct privcmd_kernel_ioreq *kioreq, *tkioreq;
+	struct eventfd_ctx *eventfd;
+	unsigned long flags;
+	int ret = 0;
+
+	eventfd = eventfd_ctx_fdget(ioeventfd->event_fd);
+	if (IS_ERR(eventfd))
+		return PTR_ERR(eventfd);
+
+	mutex_lock(&ioreq_lock);
+	list_for_each_entry_safe(kioreq, tkioreq, &ioreq_list, list) {
+		struct privcmd_kernel_ioeventfd *kioeventfd, *tmp;
+		/*
+		 * kioreq fields can be accessed here without a lock as they are
+		 * never updated after being added to the ioreq_list.
+		 */
+		if (kioreq->dom != ioeventfd->dom ||
+		    kioreq->uioreq != ioeventfd->ioreq ||
+		    kioreq->vcpus != ioeventfd->vcpus)
+			continue;
+
+		spin_lock_irqsave(&kioreq->lock, flags);
+		list_for_each_entry_safe(kioeventfd, tmp, &kioreq->ioeventfds, list) {
+			if (eventfd == kioeventfd->eventfd) {
+				ioeventfd_free(kioeventfd);
+				spin_unlock_irqrestore(&kioreq->lock, flags);
+
+				if (list_empty(&kioreq->ioeventfds))
+					ioreq_free(kioreq);
+				goto unlock;
+			}
+		}
+		spin_unlock_irqrestore(&kioreq->lock, flags);
+		break;
+	}
+
+	pr_err("Ioeventfd isn't already assigned, dom: %u, addr: %llu\n",
+	       ioeventfd->dom, ioeventfd->addr);
+	ret = -ENODEV;
+
+unlock:
+	mutex_unlock(&ioreq_lock);
+	eventfd_ctx_put(eventfd);
+
+	return ret;
+}
+
+static long privcmd_ioctl_ioeventfd(struct file *file, void __user *udata)
+{
+	struct privcmd_data *data = file->private_data;
+	struct privcmd_ioeventfd ioeventfd;
+
+	if (copy_from_user(&ioeventfd, udata, sizeof(ioeventfd)))
+		return -EFAULT;
+
+	/* No other flags should be set */
+	if (ioeventfd.flags & ~PRIVCMD_IOEVENTFD_FLAG_DEASSIGN)
+		return -EINVAL;
+
+	/* If restriction is in place, check the domid matches */
+	if (data->domid != DOMID_INVALID && data->domid != ioeventfd.dom)
+		return -EPERM;
+
+	if (ioeventfd.flags & PRIVCMD_IOEVENTFD_FLAG_DEASSIGN)
+		return privcmd_ioeventfd_deassign(&ioeventfd);
+
+	return privcmd_ioeventfd_assign(&ioeventfd);
+}
+
+static void privcmd_ioeventfd_exit(void)
+{
+	struct privcmd_kernel_ioreq *kioreq, *tmp;
+	unsigned long flags;
+
+	mutex_lock(&ioreq_lock);
+	list_for_each_entry_safe(kioreq, tmp, &ioreq_list, list) {
+		struct privcmd_kernel_ioeventfd *kioeventfd, *tmp;
+
+		spin_lock_irqsave(&kioreq->lock, flags);
+		list_for_each_entry_safe(kioeventfd, tmp, &kioreq->ioeventfds, list)
+			ioeventfd_free(kioeventfd);
+		spin_unlock_irqrestore(&kioreq->lock, flags);
+
+		ioreq_free(kioreq);
+	}
+	mutex_unlock(&ioreq_lock);
+}
 #else
 static inline long privcmd_ioctl_irqfd(struct file *file, void __user *udata)
 {
@@ -1093,7 +1480,16 @@ static inline int privcmd_irqfd_init(void)
 static inline void privcmd_irqfd_exit(void)
 {
 }
-#endif /* CONFIG_XEN_PRIVCMD_IRQFD */
+
+static inline long privcmd_ioctl_ioeventfd(struct file *file, void __user *udata)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline void privcmd_ioeventfd_exit(void)
+{
+}
+#endif /* CONFIG_XEN_PRIVCMD_EVENTFD */
 
 static long privcmd_ioctl(struct file *file,
 			  unsigned int cmd, unsigned long data)
@@ -1134,6 +1530,10 @@ static long privcmd_ioctl(struct file *file,
 		ret = privcmd_ioctl_irqfd(file, udata);
 		break;
 
+	case IOCTL_PRIVCMD_IOEVENTFD:
+		ret = privcmd_ioctl_ioeventfd(file, udata);
+		break;
+
 	default:
 		break;
 	}
@@ -1278,6 +1678,7 @@ err_privcmdbuf:
 
 static void __exit privcmd_exit(void)
 {
+	privcmd_ioeventfd_exit();
 	privcmd_irqfd_exit();
 	misc_deregister(&privcmd_dev);
 	misc_deregister(&xen_privcmdbuf_dev);
diff --git a/drivers/xen/xen-front-pgdir-shbuf.c b/drivers/xen/xen-front-pgdir-shbuf.c
index b52e0fa595..223870a011 100644
--- a/drivers/xen/xen-front-pgdir-shbuf.c
+++ b/drivers/xen/xen-front-pgdir-shbuf.c
@@ -21,7 +21,7 @@
 
 #include <xen/xen-front-pgdir-shbuf.h>
 
-/**
+/*
  * This structure represents the structure of a shared page
  * that contains grant references to the pages of the shared
  * buffer. This structure is common to many Xen para-virtualized
@@ -33,7 +33,7 @@ struct xen_page_directory {
 	grant_ref_t gref[]; /* Variable length */
 };
 
-/**
+/*
  * Shared buffer ops which are differently implemented
  * depending on the allocation mode, e.g. if the buffer
  * is allocated by the corresponding backend or frontend.
@@ -61,7 +61,7 @@ struct xen_front_pgdir_shbuf_ops {
 	int (*unmap)(struct xen_front_pgdir_shbuf *buf);
 };
 
-/**
+/*
  * Get granted reference to the very first page of the
  * page directory. Usually this is passed to the backend,
  * so it can find/fill the grant references to the buffer's
@@ -81,7 +81,7 @@ xen_front_pgdir_shbuf_get_dir_start(struct xen_front_pgdir_shbuf *buf)
 }
 EXPORT_SYMBOL_GPL(xen_front_pgdir_shbuf_get_dir_start);
 
-/**
+/*
  * Map granted references of the shared buffer.
  *
  * Depending on the shared buffer mode of allocation
@@ -102,7 +102,7 @@ int xen_front_pgdir_shbuf_map(struct xen_front_pgdir_shbuf *buf)
 }
 EXPORT_SYMBOL_GPL(xen_front_pgdir_shbuf_map);
 
-/**
+/*
  * Unmap granted references of the shared buffer.
  *
  * Depending on the shared buffer mode of allocation
@@ -123,7 +123,7 @@ int xen_front_pgdir_shbuf_unmap(struct xen_front_pgdir_shbuf *buf)
 }
 EXPORT_SYMBOL_GPL(xen_front_pgdir_shbuf_unmap);
 
-/**
+/*
  * Free all the resources of the shared buffer.
  *
  * \param buf shared buffer which resources to be freed.
@@ -150,7 +150,7 @@ EXPORT_SYMBOL_GPL(xen_front_pgdir_shbuf_free);
 				 offsetof(struct xen_page_directory, \
 					  gref)) / sizeof(grant_ref_t))
 
-/**
+/*
  * Get the number of pages the page directory consumes itself.
  *
  * \param buf shared buffer.
@@ -160,7 +160,7 @@ static int get_num_pages_dir(struct xen_front_pgdir_shbuf *buf)
 	return DIV_ROUND_UP(buf->num_pages, XEN_NUM_GREFS_PER_PAGE);
 }
 
-/**
+/*
  * Calculate the number of grant references needed to share the buffer
  * and its pages when backend allocates the buffer.
  *
@@ -172,7 +172,7 @@ static void backend_calc_num_grefs(struct xen_front_pgdir_shbuf *buf)
 	buf->num_grefs = get_num_pages_dir(buf);
 }
 
-/**
+/*
  * Calculate the number of grant references needed to share the buffer
  * and its pages when frontend allocates the buffer.
  *
@@ -190,7 +190,7 @@ static void guest_calc_num_grefs(struct xen_front_pgdir_shbuf *buf)
 #define xen_page_to_vaddr(page) \
 	((uintptr_t)pfn_to_kaddr(page_to_xen_pfn(page)))
 
-/**
+/*
  * Unmap the buffer previously mapped with grant references
  * provided by the backend.
  *
@@ -238,7 +238,7 @@ static int backend_unmap(struct xen_front_pgdir_shbuf *buf)
 	return ret;
 }
 
-/**
+/*
  * Map the buffer with grant references provided by the backend.
  *
  * \param buf shared buffer.
@@ -320,7 +320,7 @@ static int backend_map(struct xen_front_pgdir_shbuf *buf)
 	return ret;
 }
 
-/**
+/*
  * Fill page directory with grant references to the pages of the
  * page directory itself.
  *
@@ -350,7 +350,7 @@ static void backend_fill_page_dir(struct xen_front_pgdir_shbuf *buf)
 	page_dir->gref_dir_next_page = XEN_GREF_LIST_END;
 }
 
-/**
+/*
  * Fill page directory with grant references to the pages of the
  * page directory and the buffer we share with the backend.
  *
@@ -389,7 +389,7 @@ static void guest_fill_page_dir(struct xen_front_pgdir_shbuf *buf)
 	}
 }
 
-/**
+/*
  * Grant references to the frontend's buffer pages.
  *
  * These will be shared with the backend, so it can
@@ -418,7 +418,7 @@ static int guest_grant_refs_for_buffer(struct xen_front_pgdir_shbuf *buf,
 	return 0;
 }
 
-/**
+/*
  * Grant all the references needed to share the buffer.
  *
  * Grant references to the page directory pages and, if
@@ -466,7 +466,7 @@ static int grant_references(struct xen_front_pgdir_shbuf *buf)
 	return 0;
 }
 
-/**
+/*
  * Allocate all required structures to mange shared buffer.
  *
  * \param buf shared buffer.
@@ -506,7 +506,7 @@ static const struct xen_front_pgdir_shbuf_ops local_ops = {
 	.grant_refs_for_buffer = guest_grant_refs_for_buffer,
 };
 
-/**
+/*
  * Allocate a new instance of a shared buffer.
  *
  * \param cfg configuration to be used while allocating a new shared buffer.
diff --git a/drivers/xen/xen-scsiback.c b/drivers/xen/xen-scsiback.c
index 8b77e4c06e..0c51edfd13 100644
--- a/drivers/xen/xen-scsiback.c
+++ b/drivers/xen/xen-scsiback.c
@@ -1832,6 +1832,9 @@ static const struct target_core_fabric_ops scsiback_ops = {
 	.tfc_wwn_attrs			= scsiback_wwn_attrs,
 	.tfc_tpg_base_attrs		= scsiback_tpg_attrs,
 	.tfc_tpg_param_attrs		= scsiback_param_attrs,
+
+	.default_submit_type		= TARGET_DIRECT_SUBMIT,
+	.direct_submit_supp		= 1,
 };
 
 static const struct xenbus_device_id scsiback_ids[] = {
diff --git a/drivers/xen/xenbus/xenbus_dev_frontend.c b/drivers/xen/xenbus/xenbus_dev_frontend.c
index 0792fda49a..6f56640092 100644
--- a/drivers/xen/xenbus/xenbus_dev_frontend.c
+++ b/drivers/xen/xenbus/xenbus_dev_frontend.c
@@ -82,7 +82,7 @@ struct read_buffer {
 	struct list_head list;
 	unsigned int cons;
 	unsigned int len;
-	char msg[];
+	char msg[] __counted_by(len);
 };
 
 struct xenbus_file_priv {
@@ -195,7 +195,7 @@ static int queue_reply(struct list_head *queue, const void *data, size_t len)
 	if (len > XENSTORE_PAYLOAD_MAX)
 		return -EINVAL;
 
-	rb = kmalloc(sizeof(*rb) + len, GFP_KERNEL);
+	rb = kmalloc(struct_size(rb, msg, len), GFP_KERNEL);
 	if (rb == NULL)
 		return -ENOMEM;
 
diff --git a/drivers/xen/xenbus/xenbus_probe_backend.c b/drivers/xen/xenbus/xenbus_probe_backend.c
index da96c260e2..5ebb723307 100644
--- a/drivers/xen/xenbus/xenbus_probe_backend.c
+++ b/drivers/xen/xenbus/xenbus_probe_backend.c
@@ -284,13 +284,9 @@ static unsigned long backend_shrink_memory_count(struct shrinker *shrinker,
 	return 0;
 }
 
-static struct shrinker backend_memory_shrinker = {
-	.count_objects = backend_shrink_memory_count,
-	.seeks = DEFAULT_SEEKS,
-};
-
 static int __init xenbus_probe_backend_init(void)
 {
+	struct shrinker *backend_memory_shrinker;
 	static struct notifier_block xenstore_notifier = {
 		.notifier_call = backend_probe_and_watch
 	};
@@ -305,8 +301,15 @@ static int __init xenbus_probe_backend_init(void)
 
 	register_xenstore_notifier(&xenstore_notifier);
 
-	if (register_shrinker(&backend_memory_shrinker, "xen-backend"))
-		pr_warn("shrinker registration failed\n");
+	backend_memory_shrinker = shrinker_alloc(0, "xen-backend");
+	if (!backend_memory_shrinker) {
+		pr_warn("shrinker allocation failed\n");
+		return 0;
+	}
+
+	backend_memory_shrinker->count_objects = backend_shrink_memory_count;
+
+	shrinker_register(backend_memory_shrinker);
 
 	return 0;
 }