14 files changed, 1632 insertions, 626 deletions
diff --git a/drivers/xen/Kconfig b/drivers/xen/Kconfig
index 6e6180ccd72..07bec09d1da 100644
--- a/drivers/xen/Kconfig
+++ b/drivers/xen/Kconfig
@@ -29,6 +29,14 @@ config XEN_DEV_EVTCHN
 	  firing.
 	  If in doubt, say yes.
 
+config XEN_BACKEND
+	bool "Backend driver support"
+	depends on XEN_DOM0
+	default y
+	help
+	  Support for backend device drivers that provide I/O services
+	  to other virtual machines.
+
 config XENFS
 	tristate "Xen filesystem"
 	default y
@@ -62,9 +70,19 @@ config XEN_SYS_HYPERVISOR
 	 virtual environment, /sys/hypervisor will still be present,
 	 but will have no xen contents.
 
+config XEN_XENBUS_FRONTEND
+	tristate
+
+config XEN_GNTDEV
+	tristate "userspace grant access device driver"
+	depends on XEN
+	select MMU_NOTIFIER
+	help
+	  Allows userspace processes to use grants.
+
 config XEN_PLATFORM_PCI
 	tristate "xen platform pci device driver"
-	depends on XEN_PVHVM
+	depends on XEN_PVHVM && PCI
 	default m
 	help
 	  Driver for the Xen PCI Platform device: it is responsible for
diff --git a/drivers/xen/Makefile b/drivers/xen/Makefile
index 533a199e7a3..5088cc2e6fe 100644
--- a/drivers/xen/Makefile
+++ b/drivers/xen/Makefile
@@ -9,11 +9,14 @@ obj-$(CONFIG_HOTPLUG_CPU)	+= cpu_hotplug.o
 obj-$(CONFIG_XEN_XENCOMM)	+= xencomm.o
 obj-$(CONFIG_XEN_BALLOON)	+= balloon.o
 obj-$(CONFIG_XEN_DEV_EVTCHN)	+= xen-evtchn.o
+obj-$(CONFIG_XEN_GNTDEV)	+= xen-gntdev.o
 obj-$(CONFIG_XENFS)		+= xenfs/
 obj-$(CONFIG_XEN_SYS_HYPERVISOR)	+= sys-hypervisor.o
-obj-$(CONFIG_XEN_PLATFORM_PCI)	+= platform-pci.o
+obj-$(CONFIG_XEN_PLATFORM_PCI)	+= xen-platform-pci.o
 obj-$(CONFIG_SWIOTLB_XEN)	+= swiotlb-xen.o
 obj-$(CONFIG_XEN_DOM0)		+= pci.o
 
 xen-evtchn-y			:= evtchn.o
+xen-gntdev-y				:= gntdev.o
 
+xen-platform-pci-y		:= platform-pci.o
diff --git a/drivers/xen/balloon.c b/drivers/xen/balloon.c
index 43f9f02c7db..718050ace08 100644
--- a/drivers/xen/balloon.c
+++ b/drivers/xen/balloon.c
@@ -232,7 +232,7 @@ static int increase_reservation(unsigned long nr_pages)
 		set_phys_to_machine(pfn, frame_list[i]);
 
 		/* Link back into the page tables if not highmem. */
-		if (pfn < max_low_pfn) {
+		if (!xen_hvm_domain() && pfn < max_low_pfn) {
 			int ret;
 			ret = HYPERVISOR_update_va_mapping(
 				(unsigned long)__va(pfn << PAGE_SHIFT),
@@ -280,7 +280,7 @@ static int decrease_reservation(unsigned long nr_pages)
 
 		scrub_page(page);
 
-		if (!PageHighMem(page)) {
+		if (!xen_hvm_domain() && !PageHighMem(page)) {
 			ret = HYPERVISOR_update_va_mapping(
 				(unsigned long)__va(pfn << PAGE_SHIFT),
 				__pte_ma(0), 0);
@@ -296,7 +296,7 @@ static int decrease_reservation(unsigned long nr_pages)
 	/* No more mappings: invalidate P2M and add to balloon. */
 	for (i = 0; i < nr_pages; i++) {
 		pfn = mfn_to_pfn(frame_list[i]);
-		set_phys_to_machine(pfn, INVALID_P2M_ENTRY);
+		__set_phys_to_machine(pfn, INVALID_P2M_ENTRY);
 		balloon_append(pfn_to_page(pfn));
 	}
 
@@ -392,15 +392,19 @@ static struct notifier_block xenstore_notifier;
 
 static int __init balloon_init(void)
 {
-	unsigned long pfn, extra_pfn_end;
+ 	unsigned long pfn, nr_pages, extra_pfn_end;
 	struct page *page;
 
-	if (!xen_pv_domain())
+	if (!xen_domain())
 		return -ENODEV;
 
 	pr_info("xen_balloon: Initialising balloon driver.\n");
 
-	balloon_stats.current_pages = min(xen_start_info->nr_pages, max_pfn);
+ 	if (xen_pv_domain())
+ 		nr_pages = xen_start_info->nr_pages;
+ 	else
+ 		nr_pages = max_pfn;
+ 	balloon_stats.current_pages = min(nr_pages, max_pfn);
 	balloon_stats.target_pages  = balloon_stats.current_pages;
 	balloon_stats.balloon_low   = 0;
 	balloon_stats.balloon_high  = 0;
diff --git a/drivers/xen/events.c b/drivers/xen/events.c
index 31af0ac31a9..0ad1699a1b3 100644
--- a/drivers/xen/events.c
+++ b/drivers/xen/events.c
@@ -114,7 +114,7 @@ struct cpu_evtchn_s {
 static __initdata struct cpu_evtchn_s init_evtchn_mask = {
 	.bits[0 ... (NR_EVENT_CHANNELS/BITS_PER_LONG)-1] = ~0ul,
 };
-static struct cpu_evtchn_s *cpu_evtchn_mask_p = &init_evtchn_mask;
+static struct cpu_evtchn_s __refdata *cpu_evtchn_mask_p = &init_evtchn_mask;
 
 static inline unsigned long *cpu_evtchn_mask(int cpu)
 {
@@ -170,6 +170,9 @@ static struct irq_info *info_for_irq(unsigned irq)
 
 static unsigned int evtchn_from_irq(unsigned irq)
 {
+	if (unlikely(WARN(irq < 0 || irq >= nr_irqs, "Invalid irq %d!\n", irq)))
+		return 0;
+
 	return info_for_irq(irq)->evtchn;
 }
 
@@ -274,7 +277,7 @@ static void bind_evtchn_to_cpu(unsigned int chn, unsigned int cpu)
 
 	BUG_ON(irq == -1);
 #ifdef CONFIG_SMP
-	cpumask_copy(irq_to_desc(irq)->affinity, cpumask_of(cpu));
+	cpumask_copy(irq_to_desc(irq)->irq_data.affinity, cpumask_of(cpu));
 #endif
 
 	clear_bit(chn, cpu_evtchn_mask(cpu_from_irq(irq)));
@@ -291,7 +294,7 @@ static void init_evtchn_cpu_bindings(void)
 
 	/* By default all event channels notify CPU#0. */
 	for_each_irq_desc(i, desc) {
-		cpumask_copy(desc->affinity, cpumask_of(0));
+		cpumask_copy(desc->irq_data.affinity, cpumask_of(0));
 	}
 #endif
 
@@ -355,7 +358,7 @@ static void unmask_evtchn(int port)
 		struct evtchn_unmask unmask = { .port = port };
 		(void)HYPERVISOR_event_channel_op(EVTCHNOP_unmask, &unmask);
 	} else {
-		struct vcpu_info *vcpu_info = __get_cpu_var(xen_vcpu);
+		struct vcpu_info *vcpu_info = __this_cpu_read(xen_vcpu);
 
 		sync_clear_bit(port, &s->evtchn_mask[0]);
 
@@ -373,75 +376,69 @@ static void unmask_evtchn(int port)
 	put_cpu();
 }
 
-static int get_nr_hw_irqs(void)
+static int xen_allocate_irq_dynamic(void)
 {
-	int ret = 1;
+	int first = 0;
+	int irq;
 
 #ifdef CONFIG_X86_IO_APIC
-	ret = get_nr_irqs_gsi();
+	/*
+	 * For an HVM guest or domain 0 which see "real" (emulated or
+	 * actual repectively) GSIs we allocate dynamic IRQs
+	 * e.g. those corresponding to event channels or MSIs
+	 * etc. from the range above those "real" GSIs to avoid
+	 * collisions.
+	 */
+	if (xen_initial_domain() || xen_hvm_domain())
+		first = get_nr_irqs_gsi();
 #endif
 
-	return ret;
-}
+retry:
+	irq = irq_alloc_desc_from(first, -1);
 
-static int find_unbound_pirq(int type)
-{
-	int rc, i;
-	struct physdev_get_free_pirq op_get_free_pirq;
-	op_get_free_pirq.type = type;
+	if (irq == -ENOMEM && first > NR_IRQS_LEGACY) {
+		printk(KERN_ERR "Out of dynamic IRQ space and eating into GSI space. You should increase nr_irqs\n");
+		first = max(NR_IRQS_LEGACY, first - NR_IRQS_LEGACY);
+		goto retry;
+	}
 
-	rc = HYPERVISOR_physdev_op(PHYSDEVOP_get_free_pirq, &op_get_free_pirq);
-	if (!rc)
-		return op_get_free_pirq.pirq;
+	if (irq < 0)
+		panic("No available IRQ to bind to: increase nr_irqs!\n");
 
-	for (i = 0; i < nr_irqs; i++) {
-		if (pirq_to_irq[i] < 0)
-			return i;
-	}
-	return -1;
+	return irq;
 }
 
-static int find_unbound_irq(void)
+static int xen_allocate_irq_gsi(unsigned gsi)
 {
-	struct irq_data *data;
-	int irq, res;
-	int start = get_nr_hw_irqs();
-
-	if (start == nr_irqs)
-		goto no_irqs;
-
-	/* nr_irqs is a magic value. Must not use it.*/
-	for (irq = nr_irqs-1; irq > start; irq--) {
-		data = irq_get_irq_data(irq);
-		/* only 0->15 have init'd desc; handle irq > 16 */
-		if (!data)
-			break;
-		if (data->chip == &no_irq_chip)
-			break;
-		if (data->chip != &xen_dynamic_chip)
-			continue;
-		if (irq_info[irq].type == IRQT_UNBOUND)
-			return irq;
-	}
+	int irq;
 
-	if (irq == start)
-		goto no_irqs;
+	/*
+	 * A PV guest has no concept of a GSI (since it has no ACPI
+	 * nor access to/knowledge of the physical APICs). Therefore
+	 * all IRQs are dynamically allocated from the entire IRQ
+	 * space.
+	 */
+	if (xen_pv_domain() && !xen_initial_domain())
+		return xen_allocate_irq_dynamic();
 
-	res = irq_alloc_desc_at(irq, -1);
+	/* Legacy IRQ descriptors are already allocated by the arch. */
+	if (gsi < NR_IRQS_LEGACY)
+		return gsi;
 
-	if (WARN_ON(res != irq))
-		return -1;
+	irq = irq_alloc_desc_at(gsi, -1);
+	if (irq < 0)
+		panic("Unable to allocate to IRQ%d (%d)\n", gsi, irq);
 
 	return irq;
-
-no_irqs:
-	panic("No available IRQ to bind to: increase nr_irqs!\n");
 }
 
-static bool identity_mapped_irq(unsigned irq)
+static void xen_free_irq(unsigned irq)
 {
-	/* identity map all the hardware irqs */
-	return irq < get_nr_hw_irqs();
+	/* Legacy IRQ descriptors are managed by the arch. */
+	if (irq < NR_IRQS_LEGACY)
+		return;
+
+	irq_free_desc(irq);
 }
 
 static void pirq_unmask_notify(int irq)
@@ -477,7 +474,7 @@ static bool probing_irq(int irq)
 	return desc && desc->action == NULL;
 }
 
-static unsigned int startup_pirq(unsigned int irq)
+static unsigned int __startup_pirq(unsigned int irq)
 {
 	struct evtchn_bind_pirq bind_pirq;
 	struct irq_info *info = info_for_irq(irq);
@@ -515,9 +512,15 @@ out:
 	return 0;
 }
 
-static void shutdown_pirq(unsigned int irq)
+static unsigned int startup_pirq(struct irq_data *data)
+{
+	return __startup_pirq(data->irq);
+}
+
+static void shutdown_pirq(struct irq_data *data)
 {
 	struct evtchn_close close;
+	unsigned int irq = data->irq;
 	struct irq_info *info = info_for_irq(irq);
 	int evtchn = evtchn_from_irq(irq);
 
@@ -537,20 +540,20 @@ static void shutdown_pirq(unsigned int irq)
 	info->evtchn = 0;
 }
 
-static void enable_pirq(unsigned int irq)
+static void enable_pirq(struct irq_data *data)
 {
-	startup_pirq(irq);
+	startup_pirq(data);
 }
 
-static void disable_pirq(unsigned int irq)
+static void disable_pirq(struct irq_data *data)
 {
 }
 
-static void ack_pirq(unsigned int irq)
+static void ack_pirq(struct irq_data *data)
 {
-	int evtchn = evtchn_from_irq(irq);
+	int evtchn = evtchn_from_irq(data->irq);
 
-	move_native_irq(irq);
+	move_native_irq(data->irq);
 
 	if (VALID_EVTCHN(evtchn)) {
 		mask_evtchn(evtchn);
@@ -558,23 +561,6 @@ static void ack_pirq(unsigned int irq)
 	}
 }
 
-static void end_pirq(unsigned int irq)
-{
-	int evtchn = evtchn_from_irq(irq);
-	struct irq_desc *desc = irq_to_desc(irq);
-
-	if (WARN_ON(!desc))
-		return;
-
-	if ((desc->status & (IRQ_DISABLED|IRQ_PENDING)) ==
-	    (IRQ_DISABLED|IRQ_PENDING)) {
-		shutdown_pirq(irq);
-	} else if (VALID_EVTCHN(evtchn)) {
-		unmask_evtchn(evtchn);
-		pirq_unmask_notify(irq);
-	}
-}
-
 static int find_irq_by_gsi(unsigned gsi)
 {
 	int irq;
@@ -629,14 +615,7 @@ int xen_map_pirq_gsi(unsigned pirq, unsigned gsi, int shareable, char *name)
 		goto out;	/* XXX need refcount? */
 	}
 
-	/* If we are a PV guest, we don't have GSIs (no ACPI passed). Therefore
-	 * we are using the !xen_initial_domain() to drop in the function.*/
-	if (identity_mapped_irq(gsi) || (!xen_initial_domain() &&
-				xen_pv_domain())) {
-		irq = gsi;
-		irq_alloc_desc_at(irq, -1);
-	} else
-		irq = find_unbound_irq();
+	irq = xen_allocate_irq_gsi(gsi);
 
 	set_irq_chip_and_handler_name(irq, &xen_pirq_chip,
 				      handle_level_irq, name);
@@ -649,7 +628,7 @@ int xen_map_pirq_gsi(unsigned pirq, unsigned gsi, int shareable, char *name)
 	 * this in the priv domain. */
 	if (xen_initial_domain() &&
 	    HYPERVISOR_physdev_op(PHYSDEVOP_alloc_irq_vector, &irq_op)) {
-		irq_free_desc(irq);
+		xen_free_irq(irq);
 		irq = -ENOSPC;
 		goto out;
 	}
@@ -665,87 +644,46 @@ out:
 }
 
 #ifdef CONFIG_PCI_MSI
-#include <linux/msi.h>
-#include "../pci/msi.h"
-
-void xen_allocate_pirq_msi(char *name, int *irq, int *pirq, int alloc)
+int xen_allocate_pirq_msi(struct pci_dev *dev, struct msi_desc *msidesc)
 {
-	spin_lock(&irq_mapping_update_lock);
-
-	if (alloc & XEN_ALLOC_IRQ) {
-		*irq = find_unbound_irq();
-		if (*irq == -1)
-			goto out;
-	}
-
-	if (alloc & XEN_ALLOC_PIRQ) {
-		*pirq = find_unbound_pirq(MAP_PIRQ_TYPE_MSI);
-		if (*pirq == -1)
-			goto out;
-	}
+	int rc;
+	struct physdev_get_free_pirq op_get_free_pirq;
 
-	set_irq_chip_and_handler_name(*irq, &xen_pirq_chip,
-				      handle_level_irq, name);
+	op_get_free_pirq.type = MAP_PIRQ_TYPE_MSI;
+	rc = HYPERVISOR_physdev_op(PHYSDEVOP_get_free_pirq, &op_get_free_pirq);
 
-	irq_info[*irq] = mk_pirq_info(0, *pirq, 0, 0);
-	pirq_to_irq[*pirq] = *irq;
+	WARN_ONCE(rc == -ENOSYS,
+		  "hypervisor does not support the PHYSDEVOP_get_free_pirq interface\n");
 
-out:
-	spin_unlock(&irq_mapping_update_lock);
+	return rc ? -1 : op_get_free_pirq.pirq;
 }
 
-int xen_create_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int type)
+int xen_bind_pirq_msi_to_irq(struct pci_dev *dev, struct msi_desc *msidesc,
+			     int pirq, int vector, const char *name)
 {
-	int irq = -1;
-	struct physdev_map_pirq map_irq;
-	int rc;
-	int pos;
-	u32 table_offset, bir;
-
-	memset(&map_irq, 0, sizeof(map_irq));
-	map_irq.domid = DOMID_SELF;
-	map_irq.type = MAP_PIRQ_TYPE_MSI;
-	map_irq.index = -1;
-	map_irq.pirq = -1;
-	map_irq.bus = dev->bus->number;
-	map_irq.devfn = dev->devfn;
-
-	if (type == PCI_CAP_ID_MSIX) {
-		pos = pci_find_capability(dev, PCI_CAP_ID_MSIX);
-
-		pci_read_config_dword(dev, msix_table_offset_reg(pos),
-					&table_offset);
-		bir = (u8)(table_offset & PCI_MSIX_FLAGS_BIRMASK);
-
-		map_irq.table_base = pci_resource_start(dev, bir);
-		map_irq.entry_nr = msidesc->msi_attrib.entry_nr;
-	}
+	int irq, ret;
 
 	spin_lock(&irq_mapping_update_lock);
 
-	irq = find_unbound_irq();
-
+	irq = xen_allocate_irq_dynamic();
 	if (irq == -1)
 		goto out;
 
-	rc = HYPERVISOR_physdev_op(PHYSDEVOP_map_pirq, &map_irq);
-	if (rc) {
-		printk(KERN_WARNING "xen map irq failed %d\n", rc);
-
-		irq_free_desc(irq);
-
-		irq = -1;
-		goto out;
-	}
-	irq_info[irq] = mk_pirq_info(0, map_irq.pirq, 0, map_irq.index);
-
 	set_irq_chip_and_handler_name(irq, &xen_pirq_chip,
-			handle_level_irq,
-			(type == PCI_CAP_ID_MSIX) ? "msi-x":"msi");
+				      handle_level_irq, name);
 
+	irq_info[irq] = mk_pirq_info(0, pirq, 0, vector);
+	pirq_to_irq[pirq] = irq;
+	ret = irq_set_msi_desc(irq, msidesc);
+	if (ret < 0)
+		goto error_irq;
 out:
 	spin_unlock(&irq_mapping_update_lock);
 	return irq;
+error_irq:
+	spin_unlock(&irq_mapping_update_lock);
+	xen_free_irq(irq);
+	return -1;
 }
 #endif
 
@@ -770,11 +708,12 @@ int xen_destroy_irq(int irq)
 			printk(KERN_WARNING "unmap irq failed %d\n", rc);
 			goto out;
 		}
-		pirq_to_irq[info->u.pirq.pirq] = -1;
 	}
+	pirq_to_irq[info->u.pirq.pirq] = -1;
+
 	irq_info[irq] = mk_unbound_info();
 
-	irq_free_desc(irq);
+	xen_free_irq(irq);
 
 out:
 	spin_unlock(&irq_mapping_update_lock);
@@ -805,7 +744,7 @@ int bind_evtchn_to_irq(unsigned int evtchn)
 	irq = evtchn_to_irq[evtchn];
 
 	if (irq == -1) {
-		irq = find_unbound_irq();
+		irq = xen_allocate_irq_dynamic();
 
 		set_irq_chip_and_handler_name(irq, &xen_dynamic_chip,
 					      handle_fasteoi_irq, "event");
@@ -830,7 +769,7 @@ static int bind_ipi_to_irq(unsigned int ipi, unsigned int cpu)
 	irq = per_cpu(ipi_to_irq, cpu)[ipi];
 
 	if (irq == -1) {
-		irq = find_unbound_irq();
+		irq = xen_allocate_irq_dynamic();
 		if (irq < 0)
 			goto out;
 
@@ -866,7 +805,7 @@ int bind_virq_to_irq(unsigned int virq, unsigned int cpu)
 	irq = per_cpu(virq_to_irq, cpu)[virq];
 
 	if (irq == -1) {
-		irq = find_unbound_irq();
+		irq = xen_allocate_irq_dynamic();
 
 		set_irq_chip_and_handler_name(irq, &xen_percpu_chip,
 					      handle_percpu_irq, "virq");
@@ -925,7 +864,7 @@ static void unbind_from_irq(unsigned int irq)
 	if (irq_info[irq].type != IRQT_UNBOUND) {
 		irq_info[irq] = mk_unbound_info();
 
-		irq_free_desc(irq);
+		xen_free_irq(irq);
 	}
 
 	spin_unlock(&irq_mapping_update_lock);
@@ -981,7 +920,7 @@ int bind_ipi_to_irqhandler(enum ipi_vector ipi,
 	if (irq < 0)
 		return irq;
 
-	irqflags |= IRQF_NO_SUSPEND;
+	irqflags |= IRQF_NO_SUSPEND | IRQF_FORCE_RESUME;
 	retval = request_irq(irq, handler, irqflags, devname, dev_id);
 	if (retval != 0) {
 		unbind_from_irq(irq);
@@ -1101,7 +1040,7 @@ static void __xen_evtchn_do_upcall(void)
 {
 	int cpu = get_cpu();
 	struct shared_info *s = HYPERVISOR_shared_info;
-	struct vcpu_info *vcpu_info = __get_cpu_var(xen_vcpu);
+	struct vcpu_info *vcpu_info = __this_cpu_read(xen_vcpu);
  	unsigned count;
 
 	do {
@@ -1109,7 +1048,7 @@ static void __xen_evtchn_do_upcall(void)
 
 		vcpu_info->evtchn_upcall_pending = 0;
 
-		if (__get_cpu_var(xed_nesting_count)++)
+		if (__this_cpu_inc_return(xed_nesting_count) - 1)
 			goto out;
 
 #ifndef CONFIG_X86 /* No need for a barrier -- XCHG is a barrier on x86. */
@@ -1141,8 +1080,8 @@ static void __xen_evtchn_do_upcall(void)
 
 		BUG_ON(!irqs_disabled());
 
-		count = __get_cpu_var(xed_nesting_count);
-		__get_cpu_var(xed_nesting_count) = 0;
+		count = __this_cpu_read(xed_nesting_count);
+		__this_cpu_write(xed_nesting_count, 0);
 	} while (count != 1 || vcpu_info->evtchn_upcall_pending);
 
 out:
@@ -1225,11 +1164,12 @@ static int rebind_irq_to_cpu(unsigned irq, unsigned tcpu)
 	return 0;
 }
 
-static int set_affinity_irq(unsigned irq, const struct cpumask *dest)
+static int set_affinity_irq(struct irq_data *data, const struct cpumask *dest,
+			    bool force)
 {
 	unsigned tcpu = cpumask_first(dest);
 
-	return rebind_irq_to_cpu(irq, tcpu);
+	return rebind_irq_to_cpu(data->irq, tcpu);
 }
 
 int resend_irq_on_evtchn(unsigned int irq)
@@ -1248,35 +1188,35 @@ int resend_irq_on_evtchn(unsigned int irq)
 	return 1;
 }
 
-static void enable_dynirq(unsigned int irq)
+static void enable_dynirq(struct irq_data *data)
 {
-	int evtchn = evtchn_from_irq(irq);
+	int evtchn = evtchn_from_irq(data->irq);
 
 	if (VALID_EVTCHN(evtchn))
 		unmask_evtchn(evtchn);
 }
 
-static void disable_dynirq(unsigned int irq)
+static void disable_dynirq(struct irq_data *data)
 {
-	int evtchn = evtchn_from_irq(irq);
+	int evtchn = evtchn_from_irq(data->irq);
 
 	if (VALID_EVTCHN(evtchn))
 		mask_evtchn(evtchn);
 }
 
-static void ack_dynirq(unsigned int irq)
+static void ack_dynirq(struct irq_data *data)
 {
-	int evtchn = evtchn_from_irq(irq);
+	int evtchn = evtchn_from_irq(data->irq);
 
-	move_masked_irq(irq);
+	move_masked_irq(data->irq);
 
 	if (VALID_EVTCHN(evtchn))
 		unmask_evtchn(evtchn);
 }
 
-static int retrigger_dynirq(unsigned int irq)
+static int retrigger_dynirq(struct irq_data *data)
 {
-	int evtchn = evtchn_from_irq(irq);
+	int evtchn = evtchn_from_irq(data->irq);
 	struct shared_info *sh = HYPERVISOR_shared_info;
 	int ret = 0;
 
@@ -1325,7 +1265,7 @@ static void restore_cpu_pirqs(void)
 
 		printk(KERN_DEBUG "xen: --> irq=%d, pirq=%d\n", irq, map_irq.pirq);
 
-		startup_pirq(irq);
+		__startup_pirq(irq);
 	}
 }
 
@@ -1436,7 +1376,6 @@ void xen_poll_irq(int irq)
 void xen_irq_resume(void)
 {
 	unsigned int cpu, irq, evtchn;
-	struct irq_desc *desc;
 
 	init_evtchn_cpu_bindings();
 
@@ -1456,66 +1395,48 @@ void xen_irq_resume(void)
 		restore_cpu_ipis(cpu);
 	}
 
-	/*
-	 * Unmask any IRQF_NO_SUSPEND IRQs which are enabled. These
-	 * are not handled by the IRQ core.
-	 */
-	for_each_irq_desc(irq, desc) {
-		if (!desc->action || !(desc->action->flags & IRQF_NO_SUSPEND))
-			continue;
-		if (desc->status & IRQ_DISABLED)
-			continue;
-
-		evtchn = evtchn_from_irq(irq);
-		if (evtchn == -1)
-			continue;
-
-		unmask_evtchn(evtchn);
-	}
-
 	restore_cpu_pirqs();
 }
 
 static struct irq_chip xen_dynamic_chip __read_mostly = {
-	.name		= "xen-dyn",
+	.name			= "xen-dyn",
 
-	.disable	= disable_dynirq,
-	.mask		= disable_dynirq,
-	.unmask		= enable_dynirq,
+	.irq_disable		= disable_dynirq,
+	.irq_mask		= disable_dynirq,
+	.irq_unmask		= enable_dynirq,
 
-	.eoi		= ack_dynirq,
-	.set_affinity	= set_affinity_irq,
-	.retrigger	= retrigger_dynirq,
+	.irq_eoi		= ack_dynirq,
+	.irq_set_affinity	= set_affinity_irq,
+	.irq_retrigger		= retrigger_dynirq,
 };
 
 static struct irq_chip xen_pirq_chip __read_mostly = {
-	.name		= "xen-pirq",
+	.name			= "xen-pirq",
 
-	.startup	= startup_pirq,
-	.shutdown	= shutdown_pirq,
+	.irq_startup		= startup_pirq,
+	.irq_shutdown		= shutdown_pirq,
 
-	.enable		= enable_pirq,
-	.unmask		= enable_pirq,
+	.irq_enable		= enable_pirq,
+	.irq_unmask		= enable_pirq,
 
-	.disable	= disable_pirq,
-	.mask		= disable_pirq,
+	.irq_disable		= disable_pirq,
+	.irq_mask		= disable_pirq,
 
-	.ack		= ack_pirq,
-	.end		= end_pirq,
+	.irq_ack		= ack_pirq,
 
-	.set_affinity	= set_affinity_irq,
+	.irq_set_affinity	= set_affinity_irq,
 
-	.retrigger	= retrigger_dynirq,
+	.irq_retrigger		= retrigger_dynirq,
 };
 
 static struct irq_chip xen_percpu_chip __read_mostly = {
-	.name		= "xen-percpu",
+	.name			= "xen-percpu",
 
-	.disable	= disable_dynirq,
-	.mask		= disable_dynirq,
-	.unmask		= enable_dynirq,
+	.irq_disable		= disable_dynirq,
+	.irq_mask		= disable_dynirq,
+	.irq_unmask		= enable_dynirq,
 
-	.ack		= ack_dynirq,
+	.irq_ack		= ack_dynirq,
 };
 
 int xen_set_callback_via(uint64_t via)
diff --git a/drivers/xen/gntdev.c b/drivers/xen/gntdev.c
new file mode 100644
index 00000000000..1e31cdcdae1
--- /dev/null
+++ b/drivers/xen/gntdev.c
@@ -0,0 +1,665 @@
+/******************************************************************************
+ * gntdev.c
+ *
+ * Device for accessing (in user-space) pages that have been granted by other
+ * domains.
+ *
+ * Copyright (c) 2006-2007, D G Murray.
+ *           (c) 2009 Gerd Hoffmann <kraxel@redhat.com>
+ *
+ * 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, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ */
+
+#undef DEBUG
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/miscdevice.h>
+#include <linux/fs.h>
+#include <linux/mm.h>
+#include <linux/mman.h>
+#include <linux/mmu_notifier.h>
+#include <linux/types.h>
+#include <linux/uaccess.h>
+#include <linux/sched.h>
+#include <linux/spinlock.h>
+#include <linux/slab.h>
+
+#include <xen/xen.h>
+#include <xen/grant_table.h>
+#include <xen/gntdev.h>
+#include <asm/xen/hypervisor.h>
+#include <asm/xen/hypercall.h>
+#include <asm/xen/page.h>
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Derek G. Murray <Derek.Murray@cl.cam.ac.uk>, "
+	      "Gerd Hoffmann <kraxel@redhat.com>");
+MODULE_DESCRIPTION("User-space granted page access driver");
+
+static int limit = 1024;
+module_param(limit, int, 0644);
+MODULE_PARM_DESC(limit, "Maximum number of grants that may be mapped at "
+		"once by a gntdev instance");
+
+struct gntdev_priv {
+	struct list_head maps;
+	uint32_t used;
+	uint32_t limit;
+	/* lock protects maps from concurrent changes */
+	spinlock_t lock;
+	struct mm_struct *mm;
+	struct mmu_notifier mn;
+};
+
+struct grant_map {
+	struct list_head next;
+	struct gntdev_priv *priv;
+	struct vm_area_struct *vma;
+	int index;
+	int count;
+	int flags;
+	int is_mapped;
+	struct ioctl_gntdev_grant_ref *grants;
+	struct gnttab_map_grant_ref   *map_ops;
+	struct gnttab_unmap_grant_ref *unmap_ops;
+	struct page **pages;
+};
+
+/* ------------------------------------------------------------------ */
+
+static void gntdev_print_maps(struct gntdev_priv *priv,
+			      char *text, int text_index)
+{
+#ifdef DEBUG
+	struct grant_map *map;
+
+	pr_debug("maps list (priv %p, usage %d/%d)\n",
+	       priv, priv->used, priv->limit);
+
+	list_for_each_entry(map, &priv->maps, next)
+		pr_debug("  index %2d, count %2d %s\n",
+		       map->index, map->count,
+		       map->index == text_index && text ? text : "");
+#endif
+}
+
+static struct grant_map *gntdev_alloc_map(struct gntdev_priv *priv, int count)
+{
+	struct grant_map *add;
+	int i;
+
+	add = kzalloc(sizeof(struct grant_map), GFP_KERNEL);
+	if (NULL == add)
+		return NULL;
+
+	add->grants    = kzalloc(sizeof(add->grants[0])    * count, GFP_KERNEL);
+	add->map_ops   = kzalloc(sizeof(add->map_ops[0])   * count, GFP_KERNEL);
+	add->unmap_ops = kzalloc(sizeof(add->unmap_ops[0]) * count, GFP_KERNEL);
+	add->pages     = kzalloc(sizeof(add->pages[0])     * count, GFP_KERNEL);
+	if (NULL == add->grants    ||
+	    NULL == add->map_ops   ||
+	    NULL == add->unmap_ops ||
+	    NULL == add->pages)
+		goto err;
+
+	for (i = 0; i < count; i++) {
+		add->pages[i] = alloc_page(GFP_KERNEL | __GFP_HIGHMEM);
+		if (add->pages[i] == NULL)
+			goto err;
+	}
+
+	add->index = 0;
+	add->count = count;
+	add->priv  = priv;
+
+	if (add->count + priv->used > priv->limit)
+		goto err;
+
+	return add;
+
+err:
+	if (add->pages)
+		for (i = 0; i < count; i++) {
+			if (add->pages[i])
+				__free_page(add->pages[i]);
+		}
+	kfree(add->pages);
+	kfree(add->grants);
+	kfree(add->map_ops);
+	kfree(add->unmap_ops);
+	kfree(add);
+	return NULL;
+}
+
+static void gntdev_add_map(struct gntdev_priv *priv, struct grant_map *add)
+{
+	struct grant_map *map;
+
+	list_for_each_entry(map, &priv->maps, next) {
+		if (add->index + add->count < map->index) {
+			list_add_tail(&add->next, &map->next);
+			goto done;
+		}
+		add->index = map->index + map->count;
+	}
+	list_add_tail(&add->next, &priv->maps);
+
+done:
+	priv->used += add->count;
+	gntdev_print_maps(priv, "[new]", add->index);
+}
+
+static struct grant_map *gntdev_find_map_index(struct gntdev_priv *priv,
+		int index, int count)
+{
+	struct grant_map *map;
+
+	list_for_each_entry(map, &priv->maps, next) {
+		if (map->index != index)
+			continue;
+		if (map->count != count)
+			continue;
+		return map;
+	}
+	return NULL;
+}
+
+static struct grant_map *gntdev_find_map_vaddr(struct gntdev_priv *priv,
+					       unsigned long vaddr)
+{
+	struct grant_map *map;
+
+	list_for_each_entry(map, &priv->maps, next) {
+		if (!map->vma)
+			continue;
+		if (vaddr < map->vma->vm_start)
+			continue;
+		if (vaddr >= map->vma->vm_end)
+			continue;
+		return map;
+	}
+	return NULL;
+}
+
+static int gntdev_del_map(struct grant_map *map)
+{
+	int i;
+
+	if (map->vma)
+		return -EBUSY;
+	for (i = 0; i < map->count; i++)
+		if (map->unmap_ops[i].handle)
+			return -EBUSY;
+
+	map->priv->used -= map->count;
+	list_del(&map->next);
+	return 0;
+}
+
+static void gntdev_free_map(struct grant_map *map)
+{
+	int i;
+
+	if (!map)
+		return;
+
+	if (map->pages)
+		for (i = 0; i < map->count; i++) {
+			if (map->pages[i])
+				__free_page(map->pages[i]);
+		}
+	kfree(map->pages);
+	kfree(map->grants);
+	kfree(map->map_ops);
+	kfree(map->unmap_ops);
+	kfree(map);
+}
+
+/* ------------------------------------------------------------------ */
+
+static int find_grant_ptes(pte_t *pte, pgtable_t token,
+		unsigned long addr, void *data)
+{
+	struct grant_map *map = data;
+	unsigned int pgnr = (addr - map->vma->vm_start) >> PAGE_SHIFT;
+	u64 pte_maddr;
+
+	BUG_ON(pgnr >= map->count);
+	pte_maddr = arbitrary_virt_to_machine(pte).maddr;
+
+	gnttab_set_map_op(&map->map_ops[pgnr], pte_maddr,
+			  GNTMAP_contains_pte | map->flags,
+			  map->grants[pgnr].ref,
+			  map->grants[pgnr].domid);
+	gnttab_set_unmap_op(&map->unmap_ops[pgnr], pte_maddr,
+			    GNTMAP_contains_pte | map->flags,
+			    0 /* handle */);
+	return 0;
+}
+
+static int map_grant_pages(struct grant_map *map)
+{
+	int i, err = 0;
+
+	pr_debug("map %d+%d\n", map->index, map->count);
+	err = gnttab_map_refs(map->map_ops, map->pages, map->count);
+	if (err)
+		return err;
+
+	for (i = 0; i < map->count; i++) {
+		if (map->map_ops[i].status)
+			err = -EINVAL;
+		map->unmap_ops[i].handle = map->map_ops[i].handle;
+	}
+	return err;
+}
+
+static int unmap_grant_pages(struct grant_map *map, int offset, int pages)
+{
+	int i, err = 0;
+
+	pr_debug("map %d+%d [%d+%d]\n", map->index, map->count, offset, pages);
+	err = gnttab_unmap_refs(map->unmap_ops + offset, map->pages, pages);
+	if (err)
+		return err;
+
+	for (i = 0; i < pages; i++) {
+		if (map->unmap_ops[offset+i].status)
+			err = -EINVAL;
+		map->unmap_ops[offset+i].handle = 0;
+	}
+	return err;
+}
+
+/* ------------------------------------------------------------------ */
+
+static void gntdev_vma_close(struct vm_area_struct *vma)
+{
+	struct grant_map *map = vma->vm_private_data;
+
+	pr_debug("close %p\n", vma);
+	map->is_mapped = 0;
+	map->vma = NULL;
+	vma->vm_private_data = NULL;
+}
+
+static int gntdev_vma_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
+{
+	pr_debug("vaddr %p, pgoff %ld (shouldn't happen)\n",
+			vmf->virtual_address, vmf->pgoff);
+	vmf->flags = VM_FAULT_ERROR;
+	return 0;
+}
+
+static struct vm_operations_struct gntdev_vmops = {
+	.close = gntdev_vma_close,
+	.fault = gntdev_vma_fault,
+};
+
+/* ------------------------------------------------------------------ */
+
+static void mn_invl_range_start(struct mmu_notifier *mn,
+				struct mm_struct *mm,
+				unsigned long start, unsigned long end)
+{
+	struct gntdev_priv *priv = container_of(mn, struct gntdev_priv, mn);
+	struct grant_map *map;
+	unsigned long mstart, mend;
+	int err;
+
+	spin_lock(&priv->lock);
+	list_for_each_entry(map, &priv->maps, next) {
+		if (!map->vma)
+			continue;
+		if (!map->is_mapped)
+			continue;
+		if (map->vma->vm_start >= end)
+			continue;
+		if (map->vma->vm_end <= start)
+			continue;
+		mstart = max(start, map->vma->vm_start);
+		mend   = min(end,   map->vma->vm_end);
+		pr_debug("map %d+%d (%lx %lx), range %lx %lx, mrange %lx %lx\n",
+				map->index, map->count,
+				map->vma->vm_start, map->vma->vm_end,
+				start, end, mstart, mend);
+		err = unmap_grant_pages(map,
+					(mstart - map->vma->vm_start) >> PAGE_SHIFT,
+					(mend - mstart) >> PAGE_SHIFT);
+		WARN_ON(err);
+	}
+	spin_unlock(&priv->lock);
+}
+
+static void mn_invl_page(struct mmu_notifier *mn,
+			 struct mm_struct *mm,
+			 unsigned long address)
+{
+	mn_invl_range_start(mn, mm, address, address + PAGE_SIZE);
+}
+
+static void mn_release(struct mmu_notifier *mn,
+		       struct mm_struct *mm)
+{
+	struct gntdev_priv *priv = container_of(mn, struct gntdev_priv, mn);
+	struct grant_map *map;
+	int err;
+
+	spin_lock(&priv->lock);
+	list_for_each_entry(map, &priv->maps, next) {
+		if (!map->vma)
+			continue;
+		pr_debug("map %d+%d (%lx %lx)\n",
+				map->index, map->count,
+				map->vma->vm_start, map->vma->vm_end);
+		err = unmap_grant_pages(map, /* offset */ 0, map->count);
+		WARN_ON(err);
+	}
+	spin_unlock(&priv->lock);
+}
+
+struct mmu_notifier_ops gntdev_mmu_ops = {
+	.release                = mn_release,
+	.invalidate_page        = mn_invl_page,
+	.invalidate_range_start = mn_invl_range_start,
+};
+
+/* ------------------------------------------------------------------ */
+
+static int gntdev_open(struct inode *inode, struct file *flip)
+{
+	struct gntdev_priv *priv;
+	int ret = 0;
+
+	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+
+	INIT_LIST_HEAD(&priv->maps);
+	spin_lock_init(&priv->lock);
+	priv->limit = limit;
+
+	priv->mm = get_task_mm(current);
+	if (!priv->mm) {
+		kfree(priv);
+		return -ENOMEM;
+	}
+	priv->mn.ops = &gntdev_mmu_ops;
+	ret = mmu_notifier_register(&priv->mn, priv->mm);
+	mmput(priv->mm);
+
+	if (ret) {
+		kfree(priv);
+		return ret;
+	}
+
+	flip->private_data = priv;
+	pr_debug("priv %p\n", priv);
+
+	return 0;
+}
+
+static int gntdev_release(struct inode *inode, struct file *flip)
+{
+	struct gntdev_priv *priv = flip->private_data;
+	struct grant_map *map;
+	int err;
+
+	pr_debug("priv %p\n", priv);
+
+	spin_lock(&priv->lock);
+	while (!list_empty(&priv->maps)) {
+		map = list_entry(priv->maps.next, struct grant_map, next);
+		err = gntdev_del_map(map);
+		if (WARN_ON(err))
+			gntdev_free_map(map);
+
+	}
+	spin_unlock(&priv->lock);
+
+	mmu_notifier_unregister(&priv->mn, priv->mm);
+	kfree(priv);
+	return 0;
+}
+
+static long gntdev_ioctl_map_grant_ref(struct gntdev_priv *priv,
+				       struct ioctl_gntdev_map_grant_ref __user *u)
+{
+	struct ioctl_gntdev_map_grant_ref op;
+	struct grant_map *map;
+	int err;
+
+	if (copy_from_user(&op, u, sizeof(op)) != 0)
+		return -EFAULT;
+	pr_debug("priv %p, add %d\n", priv, op.count);
+	if (unlikely(op.count <= 0))
+		return -EINVAL;
+	if (unlikely(op.count > priv->limit))
+		return -EINVAL;
+
+	err = -ENOMEM;
+	map = gntdev_alloc_map(priv, op.count);
+	if (!map)
+		return err;
+	if (copy_from_user(map->grants, &u->refs,
+			   sizeof(map->grants[0]) * op.count) != 0) {
+		gntdev_free_map(map);
+		return err;
+	}
+
+	spin_lock(&priv->lock);
+	gntdev_add_map(priv, map);
+	op.index = map->index << PAGE_SHIFT;
+	spin_unlock(&priv->lock);
+
+	if (copy_to_user(u, &op, sizeof(op)) != 0) {
+		spin_lock(&priv->lock);
+		gntdev_del_map(map);
+		spin_unlock(&priv->lock);
+		gntdev_free_map(map);
+		return err;
+	}
+	return 0;
+}
+
+static long gntdev_ioctl_unmap_grant_ref(struct gntdev_priv *priv,
+					 struct ioctl_gntdev_unmap_grant_ref __user *u)
+{
+	struct ioctl_gntdev_unmap_grant_ref op;
+	struct grant_map *map;
+	int err = -ENOENT;
+
+	if (copy_from_user(&op, u, sizeof(op)) != 0)
+		return -EFAULT;
+	pr_debug("priv %p, del %d+%d\n", priv, (int)op.index, (int)op.count);
+
+	spin_lock(&priv->lock);
+	map = gntdev_find_map_index(priv, op.index >> PAGE_SHIFT, op.count);
+	if (map)
+		err = gntdev_del_map(map);
+	spin_unlock(&priv->lock);
+	if (!err)
+		gntdev_free_map(map);
+	return err;
+}
+
+static long gntdev_ioctl_get_offset_for_vaddr(struct gntdev_priv *priv,
+					      struct ioctl_gntdev_get_offset_for_vaddr __user *u)
+{
+	struct ioctl_gntdev_get_offset_for_vaddr op;
+	struct grant_map *map;
+
+	if (copy_from_user(&op, u, sizeof(op)) != 0)
+		return -EFAULT;
+	pr_debug("priv %p, offset for vaddr %lx\n", priv, (unsigned long)op.vaddr);
+
+	spin_lock(&priv->lock);
+	map = gntdev_find_map_vaddr(priv, op.vaddr);
+	if (map == NULL ||
+	    map->vma->vm_start != op.vaddr) {
+		spin_unlock(&priv->lock);
+		return -EINVAL;
+	}
+	op.offset = map->index << PAGE_SHIFT;
+	op.count = map->count;
+	spin_unlock(&priv->lock);
+
+	if (copy_to_user(u, &op, sizeof(op)) != 0)
+		return -EFAULT;
+	return 0;
+}
+
+static long gntdev_ioctl_set_max_grants(struct gntdev_priv *priv,
+					struct ioctl_gntdev_set_max_grants __user *u)
+{
+	struct ioctl_gntdev_set_max_grants op;
+
+	if (copy_from_user(&op, u, sizeof(op)) != 0)
+		return -EFAULT;
+	pr_debug("priv %p, limit %d\n", priv, op.count);
+	if (op.count > limit)
+		return -E2BIG;
+
+	spin_lock(&priv->lock);
+	priv->limit = op.count;
+	spin_unlock(&priv->lock);
+	return 0;
+}
+
+static long gntdev_ioctl(struct file *flip,
+			 unsigned int cmd, unsigned long arg)
+{
+	struct gntdev_priv *priv = flip->private_data;
+	void __user *ptr = (void __user *)arg;
+
+	switch (cmd) {
+	case IOCTL_GNTDEV_MAP_GRANT_REF:
+		return gntdev_ioctl_map_grant_ref(priv, ptr);
+
+	case IOCTL_GNTDEV_UNMAP_GRANT_REF:
+		return gntdev_ioctl_unmap_grant_ref(priv, ptr);
+
+	case IOCTL_GNTDEV_GET_OFFSET_FOR_VADDR:
+		return gntdev_ioctl_get_offset_for_vaddr(priv, ptr);
+
+	case IOCTL_GNTDEV_SET_MAX_GRANTS:
+		return gntdev_ioctl_set_max_grants(priv, ptr);
+
+	default:
+		pr_debug("priv %p, unknown cmd %x\n", priv, cmd);
+		return -ENOIOCTLCMD;
+	}
+
+	return 0;
+}
+
+static int gntdev_mmap(struct file *flip, struct vm_area_struct *vma)
+{
+	struct gntdev_priv *priv = flip->private_data;
+	int index = vma->vm_pgoff;
+	int count = (vma->vm_end - vma->vm_start) >> PAGE_SHIFT;
+	struct grant_map *map;
+	int err = -EINVAL;
+
+	if ((vma->vm_flags & VM_WRITE) && !(vma->vm_flags & VM_SHARED))
+		return -EINVAL;
+
+	pr_debug("map %d+%d at %lx (pgoff %lx)\n",
+			index, count, vma->vm_start, vma->vm_pgoff);
+
+	spin_lock(&priv->lock);
+	map = gntdev_find_map_index(priv, index, count);
+	if (!map)
+		goto unlock_out;
+	if (map->vma)
+		goto unlock_out;
+	if (priv->mm != vma->vm_mm) {
+		printk(KERN_WARNING "Huh? Other mm?\n");
+		goto unlock_out;
+	}
+
+	vma->vm_ops = &gntdev_vmops;
+
+	vma->vm_flags |= VM_RESERVED|VM_DONTCOPY|VM_DONTEXPAND|VM_PFNMAP;
+
+	vma->vm_private_data = map;
+	map->vma = vma;
+
+	map->flags = GNTMAP_host_map | GNTMAP_application_map;
+	if (!(vma->vm_flags & VM_WRITE))
+		map->flags |= GNTMAP_readonly;
+
+	spin_unlock(&priv->lock);
+
+	err = apply_to_page_range(vma->vm_mm, vma->vm_start,
+				  vma->vm_end - vma->vm_start,
+				  find_grant_ptes, map);
+	if (err) {
+		printk(KERN_WARNING "find_grant_ptes() failure.\n");
+		return err;
+	}
+
+	err = map_grant_pages(map);
+	if (err) {
+		printk(KERN_WARNING "map_grant_pages() failure.\n");
+		return err;
+	}
+
+	map->is_mapped = 1;
+
+	return 0;
+
+unlock_out:
+	spin_unlock(&priv->lock);
+	return err;
+}
+
+static const struct file_operations gntdev_fops = {
+	.owner = THIS_MODULE,
+	.open = gntdev_open,
+	.release = gntdev_release,
+	.mmap = gntdev_mmap,
+	.unlocked_ioctl = gntdev_ioctl
+};
+
+static struct miscdevice gntdev_miscdev = {
+	.minor        = MISC_DYNAMIC_MINOR,
+	.name         = "xen/gntdev",
+	.fops         = &gntdev_fops,
+};
+
+/* ------------------------------------------------------------------ */
+
+static int __init gntdev_init(void)
+{
+	int err;
+
+	if (!xen_domain())
+		return -ENODEV;
+
+	err = misc_register(&gntdev_miscdev);
+	if (err != 0) {
+		printk(KERN_ERR "Could not register gntdev device\n");
+		return err;
+	}
+	return 0;
+}
+
+static void __exit gntdev_exit(void)
+{
+	misc_deregister(&gntdev_miscdev);
+}
+
+module_init(gntdev_init);
+module_exit(gntdev_exit);
+
+/* ------------------------------------------------------------------ */
diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c
index 6c453181649..9ef54ebc119 100644
--- a/drivers/xen/grant-table.c
+++ b/drivers/xen/grant-table.c
@@ -447,6 +447,52 @@ unsigned int gnttab_max_grant_frames(void)
 }
 EXPORT_SYMBOL_GPL(gnttab_max_grant_frames);
 
+int gnttab_map_refs(struct gnttab_map_grant_ref *map_ops,
+		    struct page **pages, unsigned int count)
+{
+	int i, ret;
+	pte_t *pte;
+	unsigned long mfn;
+
+	ret = HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, map_ops, count);
+	if (ret)
+		return ret;
+
+	for (i = 0; i < count; i++) {
+		/* m2p override only supported for GNTMAP_contains_pte mappings */
+		if (!(map_ops[i].flags & GNTMAP_contains_pte))
+			continue;
+		pte = (pte_t *) (mfn_to_virt(PFN_DOWN(map_ops[i].host_addr)) +
+				(map_ops[i].host_addr & ~PAGE_MASK));
+		mfn = pte_mfn(*pte);
+		ret = m2p_add_override(mfn, pages[i]);
+		if (ret)
+			return ret;
+	}
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(gnttab_map_refs);
+
+int gnttab_unmap_refs(struct gnttab_unmap_grant_ref *unmap_ops,
+		struct page **pages, unsigned int count)
+{
+	int i, ret;
+
+	ret = HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, unmap_ops, count);
+	if (ret)
+		return ret;
+
+	for (i = 0; i < count; i++) {
+		ret = m2p_remove_override(pages[i]);
+		if (ret)
+			return ret;
+	}
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(gnttab_unmap_refs);
+
 static int gnttab_map(unsigned int start_idx, unsigned int end_idx)
 {
 	struct gnttab_setup_table setup;
diff --git a/drivers/xen/manage.c b/drivers/xen/manage.c
index db8c4c4ac88..ebb292859b5 100644
--- a/drivers/xen/manage.c
+++ b/drivers/xen/manage.c
@@ -34,32 +34,38 @@ enum shutdown_state {
 /* Ignore multiple shutdown requests. */
 static enum shutdown_state shutting_down = SHUTDOWN_INVALID;
 
-#ifdef CONFIG_PM_SLEEP
-static int xen_hvm_suspend(void *data)
-{
-	struct sched_shutdown r = { .reason = SHUTDOWN_suspend };
-	int *cancelled = data;
-
-	BUG_ON(!irqs_disabled());
-
-	*cancelled = HYPERVISOR_sched_op(SCHEDOP_shutdown, &r);
+struct suspend_info {
+	int cancelled;
+	unsigned long arg; /* extra hypercall argument */
+	void (*pre)(void);
+	void (*post)(int cancelled);
+};
 
-	xen_hvm_post_suspend(*cancelled);
+static void xen_hvm_post_suspend(int cancelled)
+{
+	xen_arch_hvm_post_suspend(cancelled);
 	gnttab_resume();
+}
 
-	if (!*cancelled) {
-		xen_irq_resume();
-		xen_console_resume();
-		xen_timer_resume();
-	}
+static void xen_pre_suspend(void)
+{
+	xen_mm_pin_all();
+	gnttab_suspend();
+	xen_arch_pre_suspend();
+}
 
-	return 0;
+static void xen_post_suspend(int cancelled)
+{
+	xen_arch_post_suspend(cancelled);
+	gnttab_resume();
+	xen_mm_unpin_all();
 }
 
+#ifdef CONFIG_PM_SLEEP
 static int xen_suspend(void *data)
 {
+	struct suspend_info *si = data;
 	int err;
-	int *cancelled = data;
 
 	BUG_ON(!irqs_disabled());
 
@@ -70,22 +76,20 @@ static int xen_suspend(void *data)
 		return err;
 	}
 
-	xen_mm_pin_all();
-	gnttab_suspend();
-	xen_pre_suspend();
+	if (si->pre)
+		si->pre();
 
 	/*
 	 * This hypercall returns 1 if suspend was cancelled
 	 * or the domain was merely checkpointed, and 0 if it
 	 * is resuming in a new domain.
 	 */
-	*cancelled = HYPERVISOR_suspend(virt_to_mfn(xen_start_info));
+	si->cancelled = HYPERVISOR_suspend(si->arg);
 
-	xen_post_suspend(*cancelled);
-	gnttab_resume();
-	xen_mm_unpin_all();
+	if (si->post)
+		si->post(si->cancelled);
 
-	if (!*cancelled) {
+	if (!si->cancelled) {
 		xen_irq_resume();
 		xen_console_resume();
 		xen_timer_resume();
@@ -99,7 +103,7 @@ static int xen_suspend(void *data)
 static void do_suspend(void)
 {
 	int err;
-	int cancelled = 1;
+	struct suspend_info si;
 
 	shutting_down = SHUTDOWN_SUSPEND;
 
@@ -129,20 +133,29 @@ static void do_suspend(void)
 		goto out_resume;
 	}
 
-	if (xen_hvm_domain())
-		err = stop_machine(xen_hvm_suspend, &cancelled, cpumask_of(0));
-	else
-		err = stop_machine(xen_suspend, &cancelled, cpumask_of(0));
+	si.cancelled = 1;
+
+	if (xen_hvm_domain()) {
+		si.arg = 0UL;
+		si.pre = NULL;
+		si.post = &xen_hvm_post_suspend;
+	} else {
+		si.arg = virt_to_mfn(xen_start_info);
+		si.pre = &xen_pre_suspend;
+		si.post = &xen_post_suspend;
+	}
+
+	err = stop_machine(xen_suspend, &si, cpumask_of(0));
 
 	dpm_resume_noirq(PMSG_RESUME);
 
 	if (err) {
 		printk(KERN_ERR "failed to start xen_suspend: %d\n", err);
-		cancelled = 1;
+		si.cancelled = 1;
 	}
 
 out_resume:
-	if (!cancelled) {
+	if (!si.cancelled) {
 		xen_arch_resume();
 		xs_resume();
 	} else
@@ -162,12 +175,39 @@ out:
 }
 #endif	/* CONFIG_PM_SLEEP */
 
+struct shutdown_handler {
+	const char *command;
+	void (*cb)(void);
+};
+
+static void do_poweroff(void)
+{
+	shutting_down = SHUTDOWN_POWEROFF;
+	orderly_poweroff(false);
+}
+
+static void do_reboot(void)
+{
+	shutting_down = SHUTDOWN_POWEROFF; /* ? */
+	ctrl_alt_del();
+}
+
 static void shutdown_handler(struct xenbus_watch *watch,
 			     const char **vec, unsigned int len)
 {
 	char *str;
 	struct xenbus_transaction xbt;
 	int err;
+	static struct shutdown_handler handlers[] = {
+		{ "poweroff",	do_poweroff },
+		{ "halt",	do_poweroff },
+		{ "reboot",	do_reboot   },
+#ifdef CONFIG_PM_SLEEP
+		{ "suspend",	do_suspend  },
+#endif
+		{NULL, NULL},
+	};
+	static struct shutdown_handler *handler;
 
 	if (shutting_down != SHUTDOWN_INVALID)
 		return;
@@ -184,7 +224,14 @@ static void shutdown_handler(struct xenbus_watch *watch,
 		return;
 	}
 
-	xenbus_write(xbt, "control", "shutdown", "");
+	for (handler = &handlers[0]; handler->command; handler++) {
+		if (strcmp(str, handler->command) == 0)
+			break;
+	}
+
+	/* Only acknowledge commands which we are prepared to handle. */
+	if (handler->cb)
+		xenbus_write(xbt, "control", "shutdown", "");
 
 	err = xenbus_transaction_end(xbt, 0);
 	if (err == -EAGAIN) {
@@ -192,17 +239,8 @@ static void shutdown_handler(struct xenbus_watch *watch,
 		goto again;
 	}
 
-	if (strcmp(str, "poweroff") == 0 ||
-	    strcmp(str, "halt") == 0) {
-		shutting_down = SHUTDOWN_POWEROFF;
-		orderly_poweroff(false);
-	} else if (strcmp(str, "reboot") == 0) {
-		shutting_down = SHUTDOWN_POWEROFF; /* ? */
-		ctrl_alt_del();
-#ifdef CONFIG_PM_SLEEP
-	} else if (strcmp(str, "suspend") == 0) {
-		do_suspend();
-#endif
+	if (handler->cb) {
+		handler->cb();
 	} else {
 		printk(KERN_INFO "Ignoring shutdown request: %s\n", str);
 		shutting_down = SHUTDOWN_INVALID;
@@ -281,27 +319,18 @@ static int shutdown_event(struct notifier_block *notifier,
 	return NOTIFY_DONE;
 }
 
-static int __init __setup_shutdown_event(void)
-{
-	/* Delay initialization in the PV on HVM case */
-	if (xen_hvm_domain())
-		return 0;
-
-	if (!xen_pv_domain())
-		return -ENODEV;
-
-	return xen_setup_shutdown_event();
-}
-
 int xen_setup_shutdown_event(void)
 {
 	static struct notifier_block xenstore_notifier = {
 		.notifier_call = shutdown_event
 	};
+
+	if (!xen_domain())
+		return -ENODEV;
 	register_xenstore_notifier(&xenstore_notifier);
 
 	return 0;
 }
 EXPORT_SYMBOL_GPL(xen_setup_shutdown_event);
 
-subsys_initcall(__setup_shutdown_event);
+subsys_initcall(xen_setup_shutdown_event);
diff --git a/drivers/xen/platform-pci.c b/drivers/xen/platform-pci.c
index c01b5ddce52..319dd0a94d5 100644
--- a/drivers/xen/platform-pci.c
+++ b/drivers/xen/platform-pci.c
@@ -105,7 +105,7 @@ static int __devinit platform_pci_init(struct pci_dev *pdev,
 				       const struct pci_device_id *ent)
 {
 	int i, ret;
-	long ioaddr, iolen;
+	long ioaddr;
 	long mmio_addr, mmio_len;
 	unsigned int max_nr_gframes;
 
@@ -114,7 +114,6 @@ static int __devinit platform_pci_init(struct pci_dev *pdev,
 		return i;
 
 	ioaddr = pci_resource_start(pdev, 0);
-	iolen = pci_resource_len(pdev, 0);
 
 	mmio_addr = pci_resource_start(pdev, 1);
 	mmio_len = pci_resource_len(pdev, 1);
@@ -125,19 +124,13 @@ static int __devinit platform_pci_init(struct pci_dev *pdev,
 		goto pci_out;
 	}
 
-	if (request_mem_region(mmio_addr, mmio_len, DRV_NAME) == NULL) {
-		dev_err(&pdev->dev, "MEM I/O resource 0x%lx @ 0x%lx busy\n",
-		       mmio_addr, mmio_len);
-		ret = -EBUSY;
+	ret = pci_request_region(pdev, 1, DRV_NAME);
+	if (ret < 0)
 		goto pci_out;
-	}
 
-	if (request_region(ioaddr, iolen, DRV_NAME) == NULL) {
-		dev_err(&pdev->dev, "I/O resource 0x%lx @ 0x%lx busy\n",
-		       iolen, ioaddr);
-		ret = -EBUSY;
+	ret = pci_request_region(pdev, 0, DRV_NAME);
+	if (ret < 0)
 		goto mem_out;
-	}
 
 	platform_mmio = mmio_addr;
 	platform_mmiolen = mmio_len;
@@ -163,15 +156,12 @@ static int __devinit platform_pci_init(struct pci_dev *pdev,
 	if (ret)
 		goto out;
 	xenbus_probe(NULL);
-	ret = xen_setup_shutdown_event();
-	if (ret)
-		goto out;
 	return 0;
 
 out:
-	release_region(ioaddr, iolen);
+	pci_release_region(pdev, 0);
 mem_out:
-	release_mem_region(mmio_addr, mmio_len);
+	pci_release_region(pdev, 1);
 pci_out:
 	pci_disable_device(pdev);
 	return ret;
diff --git a/drivers/xen/xenbus/Makefile b/drivers/xen/xenbus/Makefile
index 5571f5b8422..8dca685358b 100644
--- a/drivers/xen/xenbus/Makefile
+++ b/drivers/xen/xenbus/Makefile
@@ -5,3 +5,8 @@ xenbus-objs += xenbus_client.o
 xenbus-objs += xenbus_comms.o
 xenbus-objs += xenbus_xs.o
 xenbus-objs += xenbus_probe.o
+
+xenbus-be-objs-$(CONFIG_XEN_BACKEND) += xenbus_probe_backend.o
+xenbus-objs += $(xenbus-be-objs-y)
+
+obj-$(CONFIG_XEN_XENBUS_FRONTEND) += xenbus_probe_frontend.o
diff --git a/drivers/xen/xenbus/xenbus_probe.c b/drivers/xen/xenbus/xenbus_probe.c
index deb9c4ba3a9..baa65e7fbbc 100644
--- a/drivers/xen/xenbus/xenbus_probe.c
+++ b/drivers/xen/xenbus/xenbus_probe.c
@@ -56,7 +56,6 @@
 #include <xen/events.h>
 #include <xen/page.h>
 
-#include <xen/platform_pci.h>
 #include <xen/hvm.h>
 
 #include "xenbus_comms.h"
@@ -73,15 +72,6 @@ static unsigned long xen_store_mfn;
 
 static BLOCKING_NOTIFIER_HEAD(xenstore_chain);
 
-static void wait_for_devices(struct xenbus_driver *xendrv);
-
-static int xenbus_probe_frontend(const char *type, const char *name);
-
-static void xenbus_dev_shutdown(struct device *_dev);
-
-static int xenbus_dev_suspend(struct device *dev, pm_message_t state);
-static int xenbus_dev_resume(struct device *dev);
-
 /* If something in array of ids matches this device, return it. */
 static const struct xenbus_device_id *
 match_device(const struct xenbus_device_id *arr, struct xenbus_device *dev)
@@ -102,34 +92,7 @@ int xenbus_match(struct device *_dev, struct device_driver *_drv)
 
 	return match_device(drv->ids, to_xenbus_device(_dev)) != NULL;
 }
-
-static int xenbus_uevent(struct device *_dev, struct kobj_uevent_env *env)
-{
-	struct xenbus_device *dev = to_xenbus_device(_dev);
-
-	if (add_uevent_var(env, "MODALIAS=xen:%s", dev->devicetype))
-		return -ENOMEM;
-
-	return 0;
-}
-
-/* device/<type>/<id> => <type>-<id> */
-static int frontend_bus_id(char bus_id[XEN_BUS_ID_SIZE], const char *nodename)
-{
-	nodename = strchr(nodename, '/');
-	if (!nodename || strlen(nodename + 1) >= XEN_BUS_ID_SIZE) {
-		printk(KERN_WARNING "XENBUS: bad frontend %s\n", nodename);
-		return -EINVAL;
-	}
-
-	strlcpy(bus_id, nodename + 1, XEN_BUS_ID_SIZE);
-	if (!strchr(bus_id, '/')) {
-		printk(KERN_WARNING "XENBUS: bus_id %s no slash\n", bus_id);
-		return -EINVAL;
-	}
-	*strchr(bus_id, '/') = '-';
-	return 0;
-}
+EXPORT_SYMBOL_GPL(xenbus_match);
 
 
 static void free_otherend_details(struct xenbus_device *dev)
@@ -149,7 +112,30 @@ static void free_otherend_watch(struct xenbus_device *dev)
 }
 
 
-int read_otherend_details(struct xenbus_device *xendev,
+static int talk_to_otherend(struct xenbus_device *dev)
+{
+	struct xenbus_driver *drv = to_xenbus_driver(dev->dev.driver);
+
+	free_otherend_watch(dev);
+	free_otherend_details(dev);
+
+	return drv->read_otherend_details(dev);
+}
+
+
+
+static int watch_otherend(struct xenbus_device *dev)
+{
+	struct xen_bus_type *bus =
+		container_of(dev->dev.bus, struct xen_bus_type, bus);
+
+	return xenbus_watch_pathfmt(dev, &dev->otherend_watch,
+				    bus->otherend_changed,
+				    "%s/%s", dev->otherend, "state");
+}
+
+
+int xenbus_read_otherend_details(struct xenbus_device *xendev,
 				 char *id_node, char *path_node)
 {
 	int err = xenbus_gather(XBT_NIL, xendev->nodename,
@@ -174,39 +160,11 @@ int read_otherend_details(struct xenbus_device *xendev,
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(xenbus_read_otherend_details);
 
-
-static int read_backend_details(struct xenbus_device *xendev)
-{
-	return read_otherend_details(xendev, "backend-id", "backend");
-}
-
-static struct device_attribute xenbus_dev_attrs[] = {
-	__ATTR_NULL
-};
-
-/* Bus type for frontend drivers. */
-static struct xen_bus_type xenbus_frontend = {
-	.root = "device",
-	.levels = 2, 		/* device/type/<id> */
-	.get_bus_id = frontend_bus_id,
-	.probe = xenbus_probe_frontend,
-	.bus = {
-		.name      = "xen",
-		.match     = xenbus_match,
-		.uevent    = xenbus_uevent,
-		.probe     = xenbus_dev_probe,
-		.remove    = xenbus_dev_remove,
-		.shutdown  = xenbus_dev_shutdown,
-		.dev_attrs = xenbus_dev_attrs,
-
-		.suspend   = xenbus_dev_suspend,
-		.resume    = xenbus_dev_resume,
-	},
-};
-
-static void otherend_changed(struct xenbus_watch *watch,
-			     const char **vec, unsigned int len)
+void xenbus_otherend_changed(struct xenbus_watch *watch,
+			     const char **vec, unsigned int len,
+			     int ignore_on_shutdown)
 {
 	struct xenbus_device *dev =
 		container_of(watch, struct xenbus_device, otherend_watch);
@@ -234,11 +192,7 @@ static void otherend_changed(struct xenbus_watch *watch,
 	 * work that can fail e.g., when the rootfs is gone.
 	 */
 	if (system_state > SYSTEM_RUNNING) {
-		struct xen_bus_type *bus = bus;
-		bus = container_of(dev->dev.bus, struct xen_bus_type, bus);
-		/* If we're frontend, drive the state machine to Closed. */
-		/* This should cause the backend to release our resources. */
-		if ((bus == &xenbus_frontend) && (state == XenbusStateClosing))
+		if (ignore_on_shutdown && (state == XenbusStateClosing))
 			xenbus_frontend_closed(dev);
 		return;
 	}
@@ -246,25 +200,7 @@ static void otherend_changed(struct xenbus_watch *watch,
 	if (drv->otherend_changed)
 		drv->otherend_changed(dev, state);
 }
-
-
-static int talk_to_otherend(struct xenbus_device *dev)
-{
-	struct xenbus_driver *drv = to_xenbus_driver(dev->dev.driver);
-
-	free_otherend_watch(dev);
-	free_otherend_details(dev);
-
-	return drv->read_otherend_details(dev);
-}
-
-
-static int watch_otherend(struct xenbus_device *dev)
-{
-	return xenbus_watch_pathfmt(dev, &dev->otherend_watch, otherend_changed,
-				    "%s/%s", dev->otherend, "state");
-}
-
+EXPORT_SYMBOL_GPL(xenbus_otherend_changed);
 
 int xenbus_dev_probe(struct device *_dev)
 {
@@ -308,8 +244,9 @@ int xenbus_dev_probe(struct device *_dev)
 fail:
 	xenbus_dev_error(dev, err, "xenbus_dev_probe on %s", dev->nodename);
 	xenbus_switch_state(dev, XenbusStateClosed);
-	return -ENODEV;
+	return err;
 }
+EXPORT_SYMBOL_GPL(xenbus_dev_probe);
 
 int xenbus_dev_remove(struct device *_dev)
 {
@@ -327,8 +264,9 @@ int xenbus_dev_remove(struct device *_dev)
 	xenbus_switch_state(dev, XenbusStateClosed);
 	return 0;
 }
+EXPORT_SYMBOL_GPL(xenbus_dev_remove);
 
-static void xenbus_dev_shutdown(struct device *_dev)
+void xenbus_dev_shutdown(struct device *_dev)
 {
 	struct xenbus_device *dev = to_xenbus_device(_dev);
 	unsigned long timeout = 5*HZ;
@@ -349,6 +287,7 @@ static void xenbus_dev_shutdown(struct device *_dev)
  out:
 	put_device(&dev->dev);
 }
+EXPORT_SYMBOL_GPL(xenbus_dev_shutdown);
 
 int xenbus_register_driver_common(struct xenbus_driver *drv,
 				  struct xen_bus_type *bus,
@@ -362,25 +301,7 @@ int xenbus_register_driver_common(struct xenbus_driver *drv,
 
 	return driver_register(&drv->driver);
 }
-
-int __xenbus_register_frontend(struct xenbus_driver *drv,
-			       struct module *owner, const char *mod_name)
-{
-	int ret;
-
-	drv->read_otherend_details = read_backend_details;
-
-	ret = xenbus_register_driver_common(drv, &xenbus_frontend,
-					    owner, mod_name);
-	if (ret)
-		return ret;
-
-	/* If this driver is loaded as a module wait for devices to attach. */
-	wait_for_devices(drv);
-
-	return 0;
-}
-EXPORT_SYMBOL_GPL(__xenbus_register_frontend);
+EXPORT_SYMBOL_GPL(xenbus_register_driver_common);
 
 void xenbus_unregister_driver(struct xenbus_driver *drv)
 {
@@ -551,24 +472,7 @@ fail:
 	kfree(xendev);
 	return err;
 }
-
-/* device/<typename>/<name> */
-static int xenbus_probe_frontend(const char *type, const char *name)
-{
-	char *nodename;
-	int err;
-
-	nodename = kasprintf(GFP_KERNEL, "%s/%s/%s",
-			     xenbus_frontend.root, type, name);
-	if (!nodename)
-		return -ENOMEM;
-
-	DPRINTK("%s", nodename);
-
-	err = xenbus_probe_node(&xenbus_frontend, type, nodename);
-	kfree(nodename);
-	return err;
-}
+EXPORT_SYMBOL_GPL(xenbus_probe_node);
 
 static int xenbus_probe_device_type(struct xen_bus_type *bus, const char *type)
 {
@@ -582,10 +486,11 @@ static int xenbus_probe_device_type(struct xen_bus_type *bus, const char *type)
 		return PTR_ERR(dir);
 
 	for (i = 0; i < dir_n; i++) {
-		err = bus->probe(type, dir[i]);
+		err = bus->probe(bus, type, dir[i]);
 		if (err)
 			break;
 	}
+
 	kfree(dir);
 	return err;
 }
@@ -605,9 +510,11 @@ int xenbus_probe_devices(struct xen_bus_type *bus)
 		if (err)
 			break;
 	}
+
 	kfree(dir);
 	return err;
 }
+EXPORT_SYMBOL_GPL(xenbus_probe_devices);
 
 static unsigned int char_count(const char *str, char c)
 {
@@ -670,32 +577,18 @@ void xenbus_dev_changed(const char *node, struct xen_bus_type *bus)
 }
 EXPORT_SYMBOL_GPL(xenbus_dev_changed);
 
-static void frontend_changed(struct xenbus_watch *watch,
-			     const char **vec, unsigned int len)
-{
-	DPRINTK("");
-
-	xenbus_dev_changed(vec[XS_WATCH_PATH], &xenbus_frontend);
-}
-
-/* We watch for devices appearing and vanishing. */
-static struct xenbus_watch fe_watch = {
-	.node = "device",
-	.callback = frontend_changed,
-};
-
-static int xenbus_dev_suspend(struct device *dev, pm_message_t state)
+int xenbus_dev_suspend(struct device *dev, pm_message_t state)
 {
 	int err = 0;
 	struct xenbus_driver *drv;
-	struct xenbus_device *xdev;
+	struct xenbus_device *xdev
+		= container_of(dev, struct xenbus_device, dev);
 
-	DPRINTK("");
+	DPRINTK("%s", xdev->nodename);
 
 	if (dev->driver == NULL)
 		return 0;
 	drv = to_xenbus_driver(dev->driver);
-	xdev = container_of(dev, struct xenbus_device, dev);
 	if (drv->suspend)
 		err = drv->suspend(xdev, state);
 	if (err)
@@ -703,21 +596,20 @@ static int xenbus_dev_suspend(struct device *dev, pm_message_t state)
 		       "xenbus: suspend %s failed: %i\n", dev_name(dev), err);
 	return 0;
 }
+EXPORT_SYMBOL_GPL(xenbus_dev_suspend);
 
-static int xenbus_dev_resume(struct device *dev)
+int xenbus_dev_resume(struct device *dev)
 {
 	int err;
 	struct xenbus_driver *drv;
-	struct xenbus_device *xdev;
+	struct xenbus_device *xdev
+		= container_of(dev, struct xenbus_device, dev);
 
-	DPRINTK("");
+	DPRINTK("%s", xdev->nodename);
 
 	if (dev->driver == NULL)
 		return 0;
-
 	drv = to_xenbus_driver(dev->driver);
-	xdev = container_of(dev, struct xenbus_device, dev);
-
 	err = talk_to_otherend(xdev);
 	if (err) {
 		printk(KERN_WARNING
@@ -748,6 +640,7 @@ static int xenbus_dev_resume(struct device *dev)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(xenbus_dev_resume);
 
 /* A flag to determine if xenstored is 'ready' (i.e. has started) */
 int xenstored_ready = 0;
@@ -776,11 +669,6 @@ void xenbus_probe(struct work_struct *unused)
 {
 	xenstored_ready = 1;
 
-	/* Enumerate devices in xenstore and watch for changes. */
-	xenbus_probe_devices(&xenbus_frontend);
-	register_xenbus_watch(&fe_watch);
-	xenbus_backend_probe_and_watch();
-
 	/* Notify others that xenstore is up */
 	blocking_notifier_call_chain(&xenstore_chain, 0, NULL);
 }
@@ -809,16 +697,7 @@ static int __init xenbus_init(void)
 
 	err = -ENODEV;
 	if (!xen_domain())
-		goto out_error;
-
-	/* Register ourselves with the kernel bus subsystem */
-	err = bus_register(&xenbus_frontend.bus);
-	if (err)
-		goto out_error;
-
-	err = xenbus_backend_bus_register();
-	if (err)
-		goto out_unreg_front;
+		return err;
 
 	/*
 	 * Domain0 doesn't have a store_evtchn or store_mfn yet.
@@ -874,7 +753,7 @@ static int __init xenbus_init(void)
 	if (err) {
 		printk(KERN_WARNING
 		       "XENBUS: Error initializing xenstore comms: %i\n", err);
-		goto out_unreg_back;
+		goto out_error;
 	}
 
 #ifdef CONFIG_XEN_COMPAT_XENFS
@@ -887,133 +766,13 @@ static int __init xenbus_init(void)
 
 	return 0;
 
-  out_unreg_back:
-	xenbus_backend_bus_unregister();
-
-  out_unreg_front:
-	bus_unregister(&xenbus_frontend.bus);
-
   out_error:
 	if (page != 0)
 		free_page(page);
+
 	return err;
 }
 
 postcore_initcall(xenbus_init);
 
 MODULE_LICENSE("GPL");
-
-static int is_device_connecting(struct device *dev, void *data)
-{
-	struct xenbus_device *xendev = to_xenbus_device(dev);
-	struct device_driver *drv = data;
-	struct xenbus_driver *xendrv;
-
-	/*
-	 * A device with no driver will never connect. We care only about
-	 * devices which should currently be in the process of connecting.
-	 */
-	if (!dev->driver)
-		return 0;
-
-	/* Is this search limited to a particular driver? */
-	if (drv && (dev->driver != drv))
-		return 0;
-
-	xendrv = to_xenbus_driver(dev->driver);
-	return (xendev->state < XenbusStateConnected ||
-		(xendev->state == XenbusStateConnected &&
-		 xendrv->is_ready && !xendrv->is_ready(xendev)));
-}
-
-static int exists_connecting_device(struct device_driver *drv)
-{
-	return bus_for_each_dev(&xenbus_frontend.bus, NULL, drv,
-				is_device_connecting);
-}
-
-static int print_device_status(struct device *dev, void *data)
-{
-	struct xenbus_device *xendev = to_xenbus_device(dev);
-	struct device_driver *drv = data;
-
-	/* Is this operation limited to a particular driver? */
-	if (drv && (dev->driver != drv))
-		return 0;
-
-	if (!dev->driver) {
-		/* Information only: is this too noisy? */
-		printk(KERN_INFO "XENBUS: Device with no driver: %s\n",
-		       xendev->nodename);
-	} else if (xendev->state < XenbusStateConnected) {
-		enum xenbus_state rstate = XenbusStateUnknown;
-		if (xendev->otherend)
-			rstate = xenbus_read_driver_state(xendev->otherend);
-		printk(KERN_WARNING "XENBUS: Timeout connecting "
-		       "to device: %s (local state %d, remote state %d)\n",
-		       xendev->nodename, xendev->state, rstate);
-	}
-
-	return 0;
-}
-
-/* We only wait for device setup after most initcalls have run. */
-static int ready_to_wait_for_devices;
-
-/*
- * On a 5-minute timeout, wait for all devices currently configured.  We need
- * to do this to guarantee that the filesystems and / or network devices
- * needed for boot are available, before we can allow the boot to proceed.
- *
- * This needs to be on a late_initcall, to happen after the frontend device
- * drivers have been initialised, but before the root fs is mounted.
- *
- * A possible improvement here would be to have the tools add a per-device
- * flag to the store entry, indicating whether it is needed at boot time.
- * This would allow people who knew what they were doing to accelerate their
- * boot slightly, but of course needs tools or manual intervention to set up
- * those flags correctly.
- */
-static void wait_for_devices(struct xenbus_driver *xendrv)
-{
-	unsigned long start = jiffies;
-	struct device_driver *drv = xendrv ? &xendrv->driver : NULL;
-	unsigned int seconds_waited = 0;
-
-	if (!ready_to_wait_for_devices || !xen_domain())
-		return;
-
-	while (exists_connecting_device(drv)) {
-		if (time_after(jiffies, start + (seconds_waited+5)*HZ)) {
-			if (!seconds_waited)
-				printk(KERN_WARNING "XENBUS: Waiting for "
-				       "devices to initialise: ");
-			seconds_waited += 5;
-			printk("%us...", 300 - seconds_waited);
-			if (seconds_waited == 300)
-				break;
-		}
-
-		schedule_timeout_interruptible(HZ/10);
-	}
-
-	if (seconds_waited)
-		printk("\n");
-
-	bus_for_each_dev(&xenbus_frontend.bus, NULL, drv,
-			 print_device_status);
-}
-
-#ifndef MODULE
-static int __init boot_wait_for_devices(void)
-{
-	if (xen_hvm_domain() && !xen_platform_pci_unplug)
-		return -ENODEV;
-
-	ready_to_wait_for_devices = 1;
-	wait_for_devices(NULL);
-	return 0;
-}
-
-late_initcall(boot_wait_for_devices);
-#endif
diff --git a/drivers/xen/xenbus/xenbus_probe.h b/drivers/xen/xenbus/xenbus_probe.h
index 6c5e3185a6a..24665812316 100644
--- a/drivers/xen/xenbus/xenbus_probe.h
+++ b/drivers/xen/xenbus/xenbus_probe.h
@@ -36,26 +36,15 @@
 
 #define XEN_BUS_ID_SIZE			20
 
-#ifdef CONFIG_XEN_BACKEND
-extern void xenbus_backend_suspend(int (*fn)(struct device *, void *));
-extern void xenbus_backend_resume(int (*fn)(struct device *, void *));
-extern void xenbus_backend_probe_and_watch(void);
-extern int xenbus_backend_bus_register(void);
-extern void xenbus_backend_bus_unregister(void);
-#else
-static inline void xenbus_backend_suspend(int (*fn)(struct device *, void *)) {}
-static inline void xenbus_backend_resume(int (*fn)(struct device *, void *)) {}
-static inline void xenbus_backend_probe_and_watch(void) {}
-static inline int xenbus_backend_bus_register(void) { return 0; }
-static inline void xenbus_backend_bus_unregister(void) {}
-#endif
-
 struct xen_bus_type
 {
 	char *root;
 	unsigned int levels;
 	int (*get_bus_id)(char bus_id[XEN_BUS_ID_SIZE], const char *nodename);
-	int (*probe)(const char *type, const char *dir);
+	int (*probe)(struct xen_bus_type *bus, const char *type,
+		     const char *dir);
+	void (*otherend_changed)(struct xenbus_watch *watch, const char **vec,
+				 unsigned int len);
 	struct bus_type bus;
 };
 
@@ -73,4 +62,16 @@ extern int xenbus_probe_devices(struct xen_bus_type *bus);
 
 extern void xenbus_dev_changed(const char *node, struct xen_bus_type *bus);
 
+extern void xenbus_dev_shutdown(struct device *_dev);
+
+extern int xenbus_dev_suspend(struct device *dev, pm_message_t state);
+extern int xenbus_dev_resume(struct device *dev);
+
+extern void xenbus_otherend_changed(struct xenbus_watch *watch,
+				    const char **vec, unsigned int len,
+				    int ignore_on_shutdown);
+
+extern int xenbus_read_otherend_details(struct xenbus_device *xendev,
+					char *id_node, char *path_node);
+
 #endif
diff --git a/drivers/xen/xenbus/xenbus_probe_backend.c b/drivers/xen/xenbus/xenbus_probe_backend.c
new file mode 100644
index 00000000000..6cf467bf63e
--- /dev/null
+++ b/drivers/xen/xenbus/xenbus_probe_backend.c
@@ -0,0 +1,276 @@
+/******************************************************************************
+ * Talks to Xen Store to figure out what devices we have (backend half).
+ *
+ * Copyright (C) 2005 Rusty Russell, IBM Corporation
+ * Copyright (C) 2005 Mike Wray, Hewlett-Packard
+ * Copyright (C) 2005, 2006 XenSource Ltd
+ * Copyright (C) 2007 Solarflare Communications, Inc.
+ *
+ * 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; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#define DPRINTK(fmt, args...)				\
+	pr_debug("xenbus_probe (%s:%d) " fmt ".\n",	\
+		 __func__, __LINE__, ##args)
+
+#include <linux/kernel.h>
+#include <linux/err.h>
+#include <linux/string.h>
+#include <linux/ctype.h>
+#include <linux/fcntl.h>
+#include <linux/mm.h>
+#include <linux/notifier.h>
+
+#include <asm/page.h>
+#include <asm/pgtable.h>
+#include <asm/xen/hypervisor.h>
+#include <asm/hypervisor.h>
+#include <xen/xenbus.h>
+#include <xen/features.h>
+
+#include "xenbus_comms.h"
+#include "xenbus_probe.h"
+
+/* backend/<type>/<fe-uuid>/<id> => <type>-<fe-domid>-<id> */
+static int backend_bus_id(char bus_id[XEN_BUS_ID_SIZE], const char *nodename)
+{
+	int domid, err;
+	const char *devid, *type, *frontend;
+	unsigned int typelen;
+
+	type = strchr(nodename, '/');
+	if (!type)
+		return -EINVAL;
+	type++;
+	typelen = strcspn(type, "/");
+	if (!typelen || type[typelen] != '/')
+		return -EINVAL;
+
+	devid = strrchr(nodename, '/') + 1;
+
+	err = xenbus_gather(XBT_NIL, nodename, "frontend-id", "%i", &domid,
+			    "frontend", NULL, &frontend,
+			    NULL);
+	if (err)
+		return err;
+	if (strlen(frontend) == 0)
+		err = -ERANGE;
+	if (!err && !xenbus_exists(XBT_NIL, frontend, ""))
+		err = -ENOENT;
+	kfree(frontend);
+
+	if (err)
+		return err;
+
+	if (snprintf(bus_id, XEN_BUS_ID_SIZE, "%.*s-%i-%s",
+		     typelen, type, domid, devid) >= XEN_BUS_ID_SIZE)
+		return -ENOSPC;
+	return 0;
+}
+
+static int xenbus_uevent_backend(struct device *dev,
+				 struct kobj_uevent_env *env)
+{
+	struct xenbus_device *xdev;
+	struct xenbus_driver *drv;
+	struct xen_bus_type *bus;
+
+	DPRINTK("");
+
+	if (dev == NULL)
+		return -ENODEV;
+
+	xdev = to_xenbus_device(dev);
+	bus = container_of(xdev->dev.bus, struct xen_bus_type, bus);
+	if (xdev == NULL)
+		return -ENODEV;
+
+	/* stuff we want to pass to /sbin/hotplug */
+	if (add_uevent_var(env, "XENBUS_TYPE=%s", xdev->devicetype))
+		return -ENOMEM;
+
+	if (add_uevent_var(env, "XENBUS_PATH=%s", xdev->nodename))
+		return -ENOMEM;
+
+	if (add_uevent_var(env, "XENBUS_BASE_PATH=%s", bus->root))
+		return -ENOMEM;
+
+	if (dev->driver) {
+		drv = to_xenbus_driver(dev->driver);
+		if (drv && drv->uevent)
+			return drv->uevent(xdev, env);
+	}
+
+	return 0;
+}
+
+/* backend/<typename>/<frontend-uuid>/<name> */
+static int xenbus_probe_backend_unit(struct xen_bus_type *bus,
+				     const char *dir,
+				     const char *type,
+				     const char *name)
+{
+	char *nodename;
+	int err;
+
+	nodename = kasprintf(GFP_KERNEL, "%s/%s", dir, name);
+	if (!nodename)
+		return -ENOMEM;
+
+	DPRINTK("%s\n", nodename);
+
+	err = xenbus_probe_node(bus, type, nodename);
+	kfree(nodename);
+	return err;
+}
+
+/* backend/<typename>/<frontend-domid> */
+static int xenbus_probe_backend(struct xen_bus_type *bus, const char *type,
+				const char *domid)
+{
+	char *nodename;
+	int err = 0;
+	char **dir;
+	unsigned int i, dir_n = 0;
+
+	DPRINTK("");
+
+	nodename = kasprintf(GFP_KERNEL, "%s/%s/%s", bus->root, type, domid);
+	if (!nodename)
+		return -ENOMEM;
+
+	dir = xenbus_directory(XBT_NIL, nodename, "", &dir_n);
+	if (IS_ERR(dir)) {
+		kfree(nodename);
+		return PTR_ERR(dir);
+	}
+
+	for (i = 0; i < dir_n; i++) {
+		err = xenbus_probe_backend_unit(bus, nodename, type, dir[i]);
+		if (err)
+			break;
+	}
+	kfree(dir);
+	kfree(nodename);
+	return err;
+}
+
+static void frontend_changed(struct xenbus_watch *watch,
+			    const char **vec, unsigned int len)
+{
+	xenbus_otherend_changed(watch, vec, len, 0);
+}
+
+static struct device_attribute xenbus_backend_dev_attrs[] = {
+	__ATTR_NULL
+};
+
+static struct xen_bus_type xenbus_backend = {
+	.root = "backend",
+	.levels = 3,		/* backend/type/<frontend>/<id> */
+	.get_bus_id = backend_bus_id,
+	.probe = xenbus_probe_backend,
+	.otherend_changed = frontend_changed,
+	.bus = {
+		.name		= "xen-backend",
+		.match		= xenbus_match,
+		.uevent		= xenbus_uevent_backend,
+		.probe		= xenbus_dev_probe,
+		.remove		= xenbus_dev_remove,
+		.shutdown	= xenbus_dev_shutdown,
+		.dev_attrs	= xenbus_backend_dev_attrs,
+	},
+};
+
+static void backend_changed(struct xenbus_watch *watch,
+			    const char **vec, unsigned int len)
+{
+	DPRINTK("");
+
+	xenbus_dev_changed(vec[XS_WATCH_PATH], &xenbus_backend);
+}
+
+static struct xenbus_watch be_watch = {
+	.node = "backend",
+	.callback = backend_changed,
+};
+
+static int read_frontend_details(struct xenbus_device *xendev)
+{
+	return xenbus_read_otherend_details(xendev, "frontend-id", "frontend");
+}
+
+int xenbus_dev_is_online(struct xenbus_device *dev)
+{
+	int rc, val;
+
+	rc = xenbus_scanf(XBT_NIL, dev->nodename, "online", "%d", &val);
+	if (rc != 1)
+		val = 0; /* no online node present */
+
+	return val;
+}
+EXPORT_SYMBOL_GPL(xenbus_dev_is_online);
+
+int __xenbus_register_backend(struct xenbus_driver *drv,
+			      struct module *owner, const char *mod_name)
+{
+	drv->read_otherend_details = read_frontend_details;
+
+	return xenbus_register_driver_common(drv, &xenbus_backend,
+					     owner, mod_name);
+}
+EXPORT_SYMBOL_GPL(__xenbus_register_backend);
+
+static int backend_probe_and_watch(struct notifier_block *notifier,
+				   unsigned long event,
+				   void *data)
+{
+	/* Enumerate devices in xenstore and watch for changes. */
+	xenbus_probe_devices(&xenbus_backend);
+	register_xenbus_watch(&be_watch);
+
+	return NOTIFY_DONE;
+}
+
+static int __init xenbus_probe_backend_init(void)
+{
+	static struct notifier_block xenstore_notifier = {
+		.notifier_call = backend_probe_and_watch
+	};
+	int err;
+
+	DPRINTK("");
+
+	/* Register ourselves with the kernel bus subsystem */
+	err = bus_register(&xenbus_backend.bus);
+	if (err)
+		return err;
+
+	register_xenstore_notifier(&xenstore_notifier);
+
+	return 0;
+}
+subsys_initcall(xenbus_probe_backend_init);
diff --git a/drivers/xen/xenbus/xenbus_probe_frontend.c b/drivers/xen/xenbus/xenbus_probe_frontend.c
new file mode 100644
index 00000000000..5bcc2d6cf12
--- /dev/null
+++ b/drivers/xen/xenbus/xenbus_probe_frontend.c
@@ -0,0 +1,294 @@
+#define DPRINTK(fmt, args...)				\
+	pr_debug("xenbus_probe (%s:%d) " fmt ".\n",	\
+		 __func__, __LINE__, ##args)
+
+#include <linux/kernel.h>
+#include <linux/err.h>
+#include <linux/string.h>
+#include <linux/ctype.h>
+#include <linux/fcntl.h>
+#include <linux/mm.h>
+#include <linux/proc_fs.h>
+#include <linux/notifier.h>
+#include <linux/kthread.h>
+#include <linux/mutex.h>
+#include <linux/io.h>
+
+#include <asm/page.h>
+#include <asm/pgtable.h>
+#include <asm/xen/hypervisor.h>
+#include <xen/xenbus.h>
+#include <xen/events.h>
+#include <xen/page.h>
+
+#include <xen/platform_pci.h>
+
+#include "xenbus_comms.h"
+#include "xenbus_probe.h"
+
+
+/* device/<type>/<id> => <type>-<id> */
+static int frontend_bus_id(char bus_id[XEN_BUS_ID_SIZE], const char *nodename)
+{
+	nodename = strchr(nodename, '/');
+	if (!nodename || strlen(nodename + 1) >= XEN_BUS_ID_SIZE) {
+		printk(KERN_WARNING "XENBUS: bad frontend %s\n", nodename);
+		return -EINVAL;
+	}
+
+	strlcpy(bus_id, nodename + 1, XEN_BUS_ID_SIZE);
+	if (!strchr(bus_id, '/')) {
+		printk(KERN_WARNING "XENBUS: bus_id %s no slash\n", bus_id);
+		return -EINVAL;
+	}
+	*strchr(bus_id, '/') = '-';
+	return 0;
+}
+
+/* device/<typename>/<name> */
+static int xenbus_probe_frontend(struct xen_bus_type *bus, const char *type,
+				 const char *name)
+{
+	char *nodename;
+	int err;
+
+	nodename = kasprintf(GFP_KERNEL, "%s/%s/%s", bus->root, type, name);
+	if (!nodename)
+		return -ENOMEM;
+
+	DPRINTK("%s", nodename);
+
+	err = xenbus_probe_node(bus, type, nodename);
+	kfree(nodename);
+	return err;
+}
+
+static int xenbus_uevent_frontend(struct device *_dev,
+				  struct kobj_uevent_env *env)
+{
+	struct xenbus_device *dev = to_xenbus_device(_dev);
+
+	if (add_uevent_var(env, "MODALIAS=xen:%s", dev->devicetype))
+		return -ENOMEM;
+
+	return 0;
+}
+
+
+static void backend_changed(struct xenbus_watch *watch,
+			    const char **vec, unsigned int len)
+{
+	xenbus_otherend_changed(watch, vec, len, 1);
+}
+
+static struct device_attribute xenbus_frontend_dev_attrs[] = {
+	__ATTR_NULL
+};
+
+static struct xen_bus_type xenbus_frontend = {
+	.root = "device",
+	.levels = 2,		/* device/type/<id> */
+	.get_bus_id = frontend_bus_id,
+	.probe = xenbus_probe_frontend,
+	.otherend_changed = backend_changed,
+	.bus = {
+		.name		= "xen",
+		.match		= xenbus_match,
+		.uevent		= xenbus_uevent_frontend,
+		.probe		= xenbus_dev_probe,
+		.remove		= xenbus_dev_remove,
+		.shutdown	= xenbus_dev_shutdown,
+		.dev_attrs	= xenbus_frontend_dev_attrs,
+
+		.suspend	= xenbus_dev_suspend,
+		.resume		= xenbus_dev_resume,
+	},
+};
+
+static void frontend_changed(struct xenbus_watch *watch,
+			     const char **vec, unsigned int len)
+{
+	DPRINTK("");
+
+	xenbus_dev_changed(vec[XS_WATCH_PATH], &xenbus_frontend);
+}
+
+
+/* We watch for devices appearing and vanishing. */
+static struct xenbus_watch fe_watch = {
+	.node = "device",
+	.callback = frontend_changed,
+};
+
+static int read_backend_details(struct xenbus_device *xendev)
+{
+	return xenbus_read_otherend_details(xendev, "backend-id", "backend");
+}
+
+static int is_device_connecting(struct device *dev, void *data)
+{
+	struct xenbus_device *xendev = to_xenbus_device(dev);
+	struct device_driver *drv = data;
+	struct xenbus_driver *xendrv;
+
+	/*
+	 * A device with no driver will never connect. We care only about
+	 * devices which should currently be in the process of connecting.
+	 */
+	if (!dev->driver)
+		return 0;
+
+	/* Is this search limited to a particular driver? */
+	if (drv && (dev->driver != drv))
+		return 0;
+
+	xendrv = to_xenbus_driver(dev->driver);
+	return (xendev->state < XenbusStateConnected ||
+		(xendev->state == XenbusStateConnected &&
+		 xendrv->is_ready && !xendrv->is_ready(xendev)));
+}
+
+static int exists_connecting_device(struct device_driver *drv)
+{
+	return bus_for_each_dev(&xenbus_frontend.bus, NULL, drv,
+				is_device_connecting);
+}
+
+static int print_device_status(struct device *dev, void *data)
+{
+	struct xenbus_device *xendev = to_xenbus_device(dev);
+	struct device_driver *drv = data;
+
+	/* Is this operation limited to a particular driver? */
+	if (drv && (dev->driver != drv))
+		return 0;
+
+	if (!dev->driver) {
+		/* Information only: is this too noisy? */
+		printk(KERN_INFO "XENBUS: Device with no driver: %s\n",
+		       xendev->nodename);
+	} else if (xendev->state < XenbusStateConnected) {
+		enum xenbus_state rstate = XenbusStateUnknown;
+		if (xendev->otherend)
+			rstate = xenbus_read_driver_state(xendev->otherend);
+		printk(KERN_WARNING "XENBUS: Timeout connecting "
+		       "to device: %s (local state %d, remote state %d)\n",
+		       xendev->nodename, xendev->state, rstate);
+	}
+
+	return 0;
+}
+
+/* We only wait for device setup after most initcalls have run. */
+static int ready_to_wait_for_devices;
+
+/*
+ * On a 5-minute timeout, wait for all devices currently configured.  We need
+ * to do this to guarantee that the filesystems and / or network devices
+ * needed for boot are available, before we can allow the boot to proceed.
+ *
+ * This needs to be on a late_initcall, to happen after the frontend device
+ * drivers have been initialised, but before the root fs is mounted.
+ *
+ * A possible improvement here would be to have the tools add a per-device
+ * flag to the store entry, indicating whether it is needed at boot time.
+ * This would allow people who knew what they were doing to accelerate their
+ * boot slightly, but of course needs tools or manual intervention to set up
+ * those flags correctly.
+ */
+static void wait_for_devices(struct xenbus_driver *xendrv)
+{
+	unsigned long start = jiffies;
+	struct device_driver *drv = xendrv ? &xendrv->driver : NULL;
+	unsigned int seconds_waited = 0;
+
+	if (!ready_to_wait_for_devices || !xen_domain())
+		return;
+
+	while (exists_connecting_device(drv)) {
+		if (time_after(jiffies, start + (seconds_waited+5)*HZ)) {
+			if (!seconds_waited)
+				printk(KERN_WARNING "XENBUS: Waiting for "
+				       "devices to initialise: ");
+			seconds_waited += 5;
+			printk("%us...", 300 - seconds_waited);
+			if (seconds_waited == 300)
+				break;
+		}
+
+		schedule_timeout_interruptible(HZ/10);
+	}
+
+	if (seconds_waited)
+		printk("\n");
+
+	bus_for_each_dev(&xenbus_frontend.bus, NULL, drv,
+			 print_device_status);
+}
+
+int __xenbus_register_frontend(struct xenbus_driver *drv,
+			       struct module *owner, const char *mod_name)
+{
+	int ret;
+
+	drv->read_otherend_details = read_backend_details;
+
+	ret = xenbus_register_driver_common(drv, &xenbus_frontend,
+					    owner, mod_name);
+	if (ret)
+		return ret;
+
+	/* If this driver is loaded as a module wait for devices to attach. */
+	wait_for_devices(drv);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(__xenbus_register_frontend);
+
+static int frontend_probe_and_watch(struct notifier_block *notifier,
+				   unsigned long event,
+				   void *data)
+{
+	/* Enumerate devices in xenstore and watch for changes. */
+	xenbus_probe_devices(&xenbus_frontend);
+	register_xenbus_watch(&fe_watch);
+
+	return NOTIFY_DONE;
+}
+
+
+static int __init xenbus_probe_frontend_init(void)
+{
+	static struct notifier_block xenstore_notifier = {
+		.notifier_call = frontend_probe_and_watch
+	};
+	int err;
+
+	DPRINTK("");
+
+	/* Register ourselves with the kernel bus subsystem */
+	err = bus_register(&xenbus_frontend.bus);
+	if (err)
+		return err;
+
+	register_xenstore_notifier(&xenstore_notifier);
+
+	return 0;
+}
+subsys_initcall(xenbus_probe_frontend_init);
+
+#ifndef MODULE
+static int __init boot_wait_for_devices(void)
+{
+	if (xen_hvm_domain() && !xen_platform_pci_unplug)
+		return -ENODEV;
+
+	ready_to_wait_for_devices = 1;
+	wait_for_devices(NULL);
+	return 0;
+}
+
+late_initcall(boot_wait_for_devices);
+#endif
+
+MODULE_LICENSE("GPL");
diff --git a/drivers/xen/xenfs/xenbus.c b/drivers/xen/xenfs/xenbus.c
index 1c1236087f7..bbd000f88af 100644
--- a/drivers/xen/xenfs/xenbus.c
+++ b/drivers/xen/xenfs/xenbus.c
@@ -122,6 +122,7 @@ static ssize_t xenbus_file_read(struct file *filp,
 	int ret;
 
 	mutex_lock(&u->reply_mutex);
+again:
 	while (list_empty(&u->read_buffers)) {
 		mutex_unlock(&u->reply_mutex);
 		if (filp->f_flags & O_NONBLOCK)
@@ -144,7 +145,7 @@ static ssize_t xenbus_file_read(struct file *filp,
 		i += sz - ret;
 		rb->cons += sz - ret;
 
-		if (ret != sz) {
+		if (ret != 0) {
 			if (i == 0)
 				i = -EFAULT;
 			goto out;
@@ -160,6 +161,8 @@ static ssize_t xenbus_file_read(struct file *filp,
 					struct read_buffer, list);
 		}
 	}
+	if (i == 0)
+		goto again;
 
 out:
 	mutex_unlock(&u->reply_mutex);
@@ -407,6 +410,7 @@ static int xenbus_write_watch(unsigned msg_type, struct xenbus_file_priv *u)
 
 		mutex_lock(&u->reply_mutex);
 		rc = queue_reply(&u->read_buffers, &reply, sizeof(reply));
+		wake_up(&u->read_waitq);
 		mutex_unlock(&u->reply_mutex);
 	}
 
@@ -455,7 +459,7 @@ static ssize_t xenbus_file_write(struct file *filp,
 
 	ret = copy_from_user(u->u.buffer + u->len, ubuf, len);
 
-	if (ret == len) {
+	if (ret != 0) {
 		rc = -EFAULT;
 		goto out;
 	}
@@ -488,21 +492,6 @@ static ssize_t xenbus_file_write(struct file *filp,
 	msg_type = u->u.msg.type;
 
 	switch (msg_type) {
-	case XS_TRANSACTION_START:
-	case XS_TRANSACTION_END:
-	case XS_DIRECTORY:
-	case XS_READ:
-	case XS_GET_PERMS:
-	case XS_RELEASE:
-	case XS_GET_DOMAIN_PATH:
-	case XS_WRITE:
-	case XS_MKDIR:
-	case XS_RM:
-	case XS_SET_PERMS:
-		/* Send out a transaction */
-		ret = xenbus_write_transaction(msg_type, u);
-		break;
-
 	case XS_WATCH:
 	case XS_UNWATCH:
 		/* (Un)Ask for some path to be watched for changes */
@@ -510,7 +499,8 @@ static ssize_t xenbus_file_write(struct file *filp,
 		break;
 
 	default:
-		ret = -EINVAL;
+		/* Send out a transaction */
+		ret = xenbus_write_transaction(msg_type, u);
 		break;
 	}
 	if (ret != 0)
@@ -555,6 +545,7 @@ static int xenbus_file_release(struct inode *inode, struct file *filp)
 	struct xenbus_file_priv *u = filp->private_data;
 	struct xenbus_transaction_holder *trans, *tmp;
 	struct watch_adapter *watch, *tmp_watch;
+	struct read_buffer *rb, *tmp_rb;
 
 	/*
 	 * No need for locking here because there are no other users,
@@ -573,6 +564,10 @@ static int xenbus_file_release(struct inode *inode, struct file *filp)
 		free_watch_adapter(watch);
 	}
 
+	list_for_each_entry_safe(rb, tmp_rb, &u->read_buffers, list) {
+		list_del(&rb->list);
+		kfree(rb);
+	}
 	kfree(u);
 
 	return 0;