10 files changed, 559 insertions, 202 deletions

diff --git a/drivers/xen/Makefile b/drivers/xen/Makefile
index 37af04f1ffd9..d2a8fdf0e191 100644
--- a/drivers/xen/Makefile
+++ b/drivers/xen/Makefile
@@ -1,4 +1,5 @@
-obj-y	+= grant-table.o features.o events.o
+obj-y	+= grant-table.o features.o events.o manage.o
 obj-y	+= xenbus/
+obj-$(CONFIG_HOTPLUG_CPU)	+= cpu_hotplug.o
 obj-$(CONFIG_XEN_XENCOMM)	+= xencomm.o
 obj-$(CONFIG_XEN_BALLOON)	+= balloon.o
diff --git a/drivers/xen/balloon.c b/drivers/xen/balloon.c
index ab25ba6cbbb9..8c83abc73400 100644
--- a/drivers/xen/balloon.c
+++ b/drivers/xen/balloon.c
@@ -53,14 +53,13 @@
 #include <asm/tlb.h>
 
 #include <xen/interface/memory.h>
-#include <xen/balloon.h>
 #include <xen/xenbus.h>
 #include <xen/features.h>
 #include <xen/page.h>
 
 #define PAGES2KB(_p) ((_p)<<(PAGE_SHIFT-10))
 
-#define BALLOON_CLASS_NAME "memory"
+#define BALLOON_CLASS_NAME "xen_memory"
 
 struct balloon_stats {
 	/* We aim for 'current allocation' == 'target allocation'. */
@@ -225,10 +224,9 @@ static int increase_reservation(unsigned long nr_pages)
 		page = balloon_next_page(page);
 	}
 
-	reservation.extent_start = (unsigned long)frame_list;
-	reservation.nr_extents   = nr_pages;
-	rc = HYPERVISOR_memory_op(
-		XENMEM_populate_physmap, &reservation);
+	set_xen_guest_handle(reservation.extent_start, frame_list);
+	reservation.nr_extents = nr_pages;
+	rc = HYPERVISOR_memory_op(XENMEM_populate_physmap, &reservation);
 	if (rc < nr_pages) {
 		if (rc > 0) {
 			int ret;
@@ -236,7 +234,7 @@ static int increase_reservation(unsigned long nr_pages)
 			/* We hit the Xen hard limit: reprobe. */
 			reservation.nr_extents = rc;
 			ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation,
-					&reservation);
+						   &reservation);
 			BUG_ON(ret != rc);
 		}
 		if (rc >= 0)
@@ -321,7 +319,7 @@ static int decrease_reservation(unsigned long nr_pages)
 		balloon_append(pfn_to_page(pfn));
 	}
 
-	reservation.extent_start = (unsigned long)frame_list;
+	set_xen_guest_handle(reservation.extent_start, frame_list);
 	reservation.nr_extents   = nr_pages;
 	ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation, &reservation);
 	BUG_ON(ret != nr_pages);
@@ -368,7 +366,7 @@ static void balloon_process(struct work_struct *work)
 }
 
 /* Resets the Xen limit, sets new target, and kicks off processing. */
-void balloon_set_new_target(unsigned long target)
+static void balloon_set_new_target(unsigned long target)
 {
 	/* No need for lock. Not read-modify-write updates. */
 	balloon_stats.hard_limit   = ~0UL;
@@ -420,7 +418,7 @@ static int __init balloon_init(void)
 	unsigned long pfn;
 	struct page *page;
 
-	if (!is_running_on_xen())
+	if (!xen_pv_domain())
 		return -ENODEV;
 
 	pr_info("xen_balloon: Initialising balloon driver.\n");
@@ -464,136 +462,13 @@ static void balloon_exit(void)
 
 module_exit(balloon_exit);
 
-static void balloon_update_driver_allowance(long delta)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&balloon_lock, flags);
-	balloon_stats.driver_pages += delta;
-	spin_unlock_irqrestore(&balloon_lock, flags);
-}
-
-static int dealloc_pte_fn(
-	pte_t *pte, struct page *pmd_page, unsigned long addr, void *data)
-{
-	unsigned long mfn = pte_mfn(*pte);
-	int ret;
-	struct xen_memory_reservation reservation = {
-		.nr_extents   = 1,
-		.extent_order = 0,
-		.domid        = DOMID_SELF
-	};
-	reservation.extent_start = (unsigned long)&mfn;
-	set_pte_at(&init_mm, addr, pte, __pte_ma(0ull));
-	set_phys_to_machine(__pa(addr) >> PAGE_SHIFT, INVALID_P2M_ENTRY);
-	ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation, &reservation);
-	BUG_ON(ret != 1);
-	return 0;
-}
-
-static struct page **alloc_empty_pages_and_pagevec(int nr_pages)
-{
-	unsigned long vaddr, flags;
-	struct page *page, **pagevec;
-	int i, ret;
-
-	pagevec = kmalloc(sizeof(page) * nr_pages, GFP_KERNEL);
-	if (pagevec == NULL)
-		return NULL;
-
-	for (i = 0; i < nr_pages; i++) {
-		page = pagevec[i] = alloc_page(GFP_KERNEL);
-		if (page == NULL)
-			goto err;
-
-		vaddr = (unsigned long)page_address(page);
-
-		scrub_page(page);
-
-		spin_lock_irqsave(&balloon_lock, flags);
-
-		if (xen_feature(XENFEAT_auto_translated_physmap)) {
-			unsigned long gmfn = page_to_pfn(page);
-			struct xen_memory_reservation reservation = {
-				.nr_extents   = 1,
-				.extent_order = 0,
-				.domid        = DOMID_SELF
-			};
-			reservation.extent_start = (unsigned long)&gmfn;
-			ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation,
-						   &reservation);
-			if (ret == 1)
-				ret = 0; /* success */
-		} else {
-			ret = apply_to_page_range(&init_mm, vaddr, PAGE_SIZE,
-						  dealloc_pte_fn, NULL);
-		}
-
-		if (ret != 0) {
-			spin_unlock_irqrestore(&balloon_lock, flags);
-			__free_page(page);
-			goto err;
-		}
-
-		totalram_pages = --balloon_stats.current_pages;
-
-		spin_unlock_irqrestore(&balloon_lock, flags);
-	}
-
- out:
-	schedule_work(&balloon_worker);
-	flush_tlb_all();
-	return pagevec;
-
- err:
-	spin_lock_irqsave(&balloon_lock, flags);
-	while (--i >= 0)
-		balloon_append(pagevec[i]);
-	spin_unlock_irqrestore(&balloon_lock, flags);
-	kfree(pagevec);
-	pagevec = NULL;
-	goto out;
-}
-
-static void free_empty_pages_and_pagevec(struct page **pagevec, int nr_pages)
-{
-	unsigned long flags;
-	int i;
-
-	if (pagevec == NULL)
-		return;
-
-	spin_lock_irqsave(&balloon_lock, flags);
-	for (i = 0; i < nr_pages; i++) {
-		BUG_ON(page_count(pagevec[i]) != 1);
-		balloon_append(pagevec[i]);
-	}
-	spin_unlock_irqrestore(&balloon_lock, flags);
-
-	kfree(pagevec);
-
-	schedule_work(&balloon_worker);
-}
-
-static void balloon_release_driver_page(struct page *page)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&balloon_lock, flags);
-	balloon_append(page);
-	balloon_stats.driver_pages--;
-	spin_unlock_irqrestore(&balloon_lock, flags);
-
-	schedule_work(&balloon_worker);
-}
-
-
-#define BALLOON_SHOW(name, format, args...)			\
-	static ssize_t show_##name(struct sys_device *dev,	\
-				   char *buf)			\
-	{							\
-		return sprintf(buf, format, ##args);		\
-	}							\
+#define BALLOON_SHOW(name, format, args...)				\
+	static ssize_t show_##name(struct sys_device *dev,		\
+				   struct sysdev_attribute *attr,	\
+				   char *buf)				\
+	{								\
+		return sprintf(buf, format, ##args);			\
+	}								\
 	static SYSDEV_ATTR(name, S_IRUGO, show_##name, NULL)
 
 BALLOON_SHOW(current_kb, "%lu\n", PAGES2KB(balloon_stats.current_pages));
@@ -604,28 +479,25 @@ BALLOON_SHOW(hard_limit_kb,
 	     (balloon_stats.hard_limit!=~0UL) ? PAGES2KB(balloon_stats.hard_limit) : 0);
 BALLOON_SHOW(driver_kb, "%lu\n", PAGES2KB(balloon_stats.driver_pages));
 
-static ssize_t show_target_kb(struct sys_device *dev, char *buf)
+static ssize_t show_target_kb(struct sys_device *dev, struct sysdev_attribute *attr,
+			      char *buf)
 {
 	return sprintf(buf, "%lu\n", PAGES2KB(balloon_stats.target_pages));
 }
 
 static ssize_t store_target_kb(struct sys_device *dev,
+			       struct sysdev_attribute *attr,
 			       const char *buf,
 			       size_t count)
 {
-	char memstring[64], *endchar;
+	char *endchar;
 	unsigned long long target_bytes;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
-	if (count <= 1)
-		return -EBADMSG; /* runt */
-	if (count > sizeof(memstring))
-		return -EFBIG;   /* too long */
-	strcpy(memstring, buf);
+	target_bytes = memparse(buf, &endchar);
 
-	target_bytes = memparse(memstring, &endchar);
 	balloon_set_new_target(target_bytes >> PAGE_SHIFT);
 
 	return count;
@@ -693,20 +565,4 @@ static int register_balloon(struct sys_device *sysdev)
 	return error;
 }
 
-static void unregister_balloon(struct sys_device *sysdev)
-{
-	int i;
-
-	sysfs_remove_group(&sysdev->kobj, &balloon_info_group);
-	for (i = 0; i < ARRAY_SIZE(balloon_attrs); i++)
-		sysdev_remove_file(sysdev, balloon_attrs[i]);
-	sysdev_unregister(sysdev);
-	sysdev_class_unregister(&balloon_sysdev_class);
-}
-
-static void balloon_sysfs_exit(void)
-{
-	unregister_balloon(&balloon_sysdev);
-}
-
 MODULE_LICENSE("GPL");
diff --git a/drivers/xen/cpu_hotplug.c b/drivers/xen/cpu_hotplug.c
new file mode 100644
index 000000000000..565280ec1c6a
--- /dev/null
+++ b/drivers/xen/cpu_hotplug.c
@@ -0,0 +1,90 @@
+#include <linux/notifier.h>
+
+#include <xen/xenbus.h>
+
+#include <asm-x86/xen/hypervisor.h>
+#include <asm/cpu.h>
+
+static void enable_hotplug_cpu(int cpu)
+{
+	if (!cpu_present(cpu))
+		arch_register_cpu(cpu);
+
+	cpu_set(cpu, cpu_present_map);
+}
+
+static void disable_hotplug_cpu(int cpu)
+{
+	if (cpu_present(cpu))
+		arch_unregister_cpu(cpu);
+
+	cpu_clear(cpu, cpu_present_map);
+}
+
+static void vcpu_hotplug(unsigned int cpu)
+{
+	int err;
+	char dir[32], state[32];
+
+	if (!cpu_possible(cpu))
+		return;
+
+	sprintf(dir, "cpu/%u", cpu);
+	err = xenbus_scanf(XBT_NIL, dir, "availability", "%s", state);
+	if (err != 1) {
+		printk(KERN_ERR "XENBUS: Unable to read cpu state\n");
+		return;
+	}
+
+	if (strcmp(state, "online") == 0) {
+		enable_hotplug_cpu(cpu);
+	} else if (strcmp(state, "offline") == 0) {
+		(void)cpu_down(cpu);
+		disable_hotplug_cpu(cpu);
+	} else {
+		printk(KERN_ERR "XENBUS: unknown state(%s) on CPU%d\n",
+		       state, cpu);
+	}
+}
+
+static void handle_vcpu_hotplug_event(struct xenbus_watch *watch,
+					const char **vec, unsigned int len)
+{
+	unsigned int cpu;
+	char *cpustr;
+	const char *node = vec[XS_WATCH_PATH];
+
+	cpustr = strstr(node, "cpu/");
+	if (cpustr != NULL) {
+		sscanf(cpustr, "cpu/%u", &cpu);
+		vcpu_hotplug(cpu);
+	}
+}
+
+static int setup_cpu_watcher(struct notifier_block *notifier,
+			      unsigned long event, void *data)
+{
+	static struct xenbus_watch cpu_watch = {
+		.node = "cpu",
+		.callback = handle_vcpu_hotplug_event};
+
+	(void)register_xenbus_watch(&cpu_watch);
+
+	return NOTIFY_DONE;
+}
+
+static int __init setup_vcpu_hotplug_event(void)
+{
+	static struct notifier_block xsn_cpu = {
+		.notifier_call = setup_cpu_watcher };
+
+	if (!xen_pv_domain())
+		return -ENODEV;
+
+	register_xenstore_notifier(&xsn_cpu);
+
+	return 0;
+}
+
+arch_initcall(setup_vcpu_hotplug_event);
+
diff --git a/drivers/xen/events.c b/drivers/xen/events.c
index 76e5b7386af9..c3290bc186a0 100644
--- a/drivers/xen/events.c
+++ b/drivers/xen/events.c
@@ -84,17 +84,6 @@ static int irq_bindcount[NR_IRQS];
 /* Xen will never allocate port zero for any purpose. */
 #define VALID_EVTCHN(chn)	((chn) != 0)
 
-/*
- * Force a proper event-channel callback from Xen after clearing the
- * callback mask. We do this in a very simple manner, by making a call
- * down into Xen. The pending flag will be checked by Xen on return.
- */
-void force_evtchn_callback(void)
-{
-	(void)HYPERVISOR_xen_version(0, NULL);
-}
-EXPORT_SYMBOL_GPL(force_evtchn_callback);
-
 static struct irq_chip xen_dynamic_chip;
 
 /* Constructor for packed IRQ information. */
@@ -175,6 +164,12 @@ static inline void set_evtchn(int port)
 	sync_set_bit(port, &s->evtchn_pending[0]);
 }
 
+static inline int test_evtchn(int port)
+{
+	struct shared_info *s = HYPERVISOR_shared_info;
+	return sync_test_bit(port, &s->evtchn_pending[0]);
+}
+
 
 /**
  * notify_remote_via_irq - send event to remote end of event channel via irq
@@ -355,7 +350,7 @@ static void unbind_from_irq(unsigned int irq)
 
 	spin_lock(&irq_mapping_update_lock);
 
-	if (VALID_EVTCHN(evtchn) && (--irq_bindcount[irq] == 0)) {
+	if ((--irq_bindcount[irq] == 0) && VALID_EVTCHN(evtchn)) {
 		close.port = evtchn;
 		if (HYPERVISOR_event_channel_op(EVTCHNOP_close, &close) != 0)
 			BUG();
@@ -365,6 +360,10 @@ static void unbind_from_irq(unsigned int irq)
 			per_cpu(virq_to_irq, cpu_from_evtchn(evtchn))
 				[index_from_irq(irq)] = -1;
 			break;
+		case IRQT_IPI:
+			per_cpu(ipi_to_irq, cpu_from_evtchn(evtchn))
+				[index_from_irq(irq)] = -1;
+			break;
 		default:
 			break;
 		}
@@ -375,7 +374,7 @@ static void unbind_from_irq(unsigned int irq)
 		evtchn_to_irq[evtchn] = -1;
 		irq_info[irq] = IRQ_UNBOUND;
 
-		dynamic_irq_init(irq);
+		dynamic_irq_cleanup(irq);
 	}
 
 	spin_unlock(&irq_mapping_update_lock);
@@ -557,6 +556,33 @@ out:
 	put_cpu();
 }
 
+/* Rebind a new event channel to an existing irq. */
+void rebind_evtchn_irq(int evtchn, int irq)
+{
+	/* Make sure the irq is masked, since the new event channel
+	   will also be masked. */
+	disable_irq(irq);
+
+	spin_lock(&irq_mapping_update_lock);
+
+	/* After resume the irq<->evtchn mappings are all cleared out */
+	BUG_ON(evtchn_to_irq[evtchn] != -1);
+	/* Expect irq to have been bound before,
+	   so the bindcount should be non-0 */
+	BUG_ON(irq_bindcount[irq] == 0);
+
+	evtchn_to_irq[evtchn] = irq;
+	irq_info[irq] = mk_irq_info(IRQT_EVTCHN, 0, evtchn);
+
+	spin_unlock(&irq_mapping_update_lock);
+
+	/* new event channels are always bound to cpu 0 */
+	irq_set_affinity(irq, cpumask_of_cpu(0));
+
+	/* Unmask the event channel. */
+	enable_irq(irq);
+}
+
 /* Rebind an evtchn so that it gets delivered to a specific cpu */
 static void rebind_irq_to_cpu(unsigned irq, unsigned tcpu)
 {
@@ -647,6 +673,135 @@ static int retrigger_dynirq(unsigned int irq)
 	return ret;
 }
 
+static void restore_cpu_virqs(unsigned int cpu)
+{
+	struct evtchn_bind_virq bind_virq;
+	int virq, irq, evtchn;
+
+	for (virq = 0; virq < NR_VIRQS; virq++) {
+		if ((irq = per_cpu(virq_to_irq, cpu)[virq]) == -1)
+			continue;
+
+		BUG_ON(irq_info[irq].type != IRQT_VIRQ);
+		BUG_ON(irq_info[irq].index != virq);
+
+		/* Get a new binding from Xen. */
+		bind_virq.virq = virq;
+		bind_virq.vcpu = cpu;
+		if (HYPERVISOR_event_channel_op(EVTCHNOP_bind_virq,
+						&bind_virq) != 0)
+			BUG();
+		evtchn = bind_virq.port;
+
+		/* Record the new mapping. */
+		evtchn_to_irq[evtchn] = irq;
+		irq_info[irq] = mk_irq_info(IRQT_VIRQ, virq, evtchn);
+		bind_evtchn_to_cpu(evtchn, cpu);
+
+		/* Ready for use. */
+		unmask_evtchn(evtchn);
+	}
+}
+
+static void restore_cpu_ipis(unsigned int cpu)
+{
+	struct evtchn_bind_ipi bind_ipi;
+	int ipi, irq, evtchn;
+
+	for (ipi = 0; ipi < XEN_NR_IPIS; ipi++) {
+		if ((irq = per_cpu(ipi_to_irq, cpu)[ipi]) == -1)
+			continue;
+
+		BUG_ON(irq_info[irq].type != IRQT_IPI);
+		BUG_ON(irq_info[irq].index != ipi);
+
+		/* Get a new binding from Xen. */
+		bind_ipi.vcpu = cpu;
+		if (HYPERVISOR_event_channel_op(EVTCHNOP_bind_ipi,
+						&bind_ipi) != 0)
+			BUG();
+		evtchn = bind_ipi.port;
+
+		/* Record the new mapping. */
+		evtchn_to_irq[evtchn] = irq;
+		irq_info[irq] = mk_irq_info(IRQT_IPI, ipi, evtchn);
+		bind_evtchn_to_cpu(evtchn, cpu);
+
+		/* Ready for use. */
+		unmask_evtchn(evtchn);
+
+	}
+}
+
+/* Clear an irq's pending state, in preparation for polling on it */
+void xen_clear_irq_pending(int irq)
+{
+	int evtchn = evtchn_from_irq(irq);
+
+	if (VALID_EVTCHN(evtchn))
+		clear_evtchn(evtchn);
+}
+
+void xen_set_irq_pending(int irq)
+{
+	int evtchn = evtchn_from_irq(irq);
+
+	if (VALID_EVTCHN(evtchn))
+		set_evtchn(evtchn);
+}
+
+bool xen_test_irq_pending(int irq)
+{
+	int evtchn = evtchn_from_irq(irq);
+	bool ret = false;
+
+	if (VALID_EVTCHN(evtchn))
+		ret = test_evtchn(evtchn);
+
+	return ret;
+}
+
+/* Poll waiting for an irq to become pending.  In the usual case, the
+   irq will be disabled so it won't deliver an interrupt. */
+void xen_poll_irq(int irq)
+{
+	evtchn_port_t evtchn = evtchn_from_irq(irq);
+
+	if (VALID_EVTCHN(evtchn)) {
+		struct sched_poll poll;
+
+		poll.nr_ports = 1;
+		poll.timeout = 0;
+		poll.ports = &evtchn;
+
+		if (HYPERVISOR_sched_op(SCHEDOP_poll, &poll) != 0)
+			BUG();
+	}
+}
+
+void xen_irq_resume(void)
+{
+	unsigned int cpu, irq, evtchn;
+
+	init_evtchn_cpu_bindings();
+
+	/* New event-channel space is not 'live' yet. */
+	for (evtchn = 0; evtchn < NR_EVENT_CHANNELS; evtchn++)
+		mask_evtchn(evtchn);
+
+	/* No IRQ <-> event-channel mappings. */
+	for (irq = 0; irq < NR_IRQS; irq++)
+		irq_info[irq].evtchn = 0; /* zap event-channel binding */
+
+	for (evtchn = 0; evtchn < NR_EVENT_CHANNELS; evtchn++)
+		evtchn_to_irq[evtchn] = -1;
+
+	for_each_possible_cpu(cpu) {
+		restore_cpu_virqs(cpu);
+		restore_cpu_ipis(cpu);
+	}
+}
+
 static struct irq_chip xen_dynamic_chip __read_mostly = {
 	.name		= "xen-dyn",
 	.mask		= disable_dynirq,
diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c
index 52b6b41b909d..06592b9da83c 100644
--- a/drivers/xen/grant-table.c
+++ b/drivers/xen/grant-table.c
@@ -471,14 +471,14 @@ static int gnttab_map(unsigned int start_idx, unsigned int end_idx)
 	return 0;
 }
 
-static int gnttab_resume(void)
+int gnttab_resume(void)
 {
 	if (max_nr_grant_frames() < nr_grant_frames)
 		return -ENOSYS;
 	return gnttab_map(0, nr_grant_frames - 1);
 }
 
-static int gnttab_suspend(void)
+int gnttab_suspend(void)
 {
 	arch_gnttab_unmap_shared(shared, nr_grant_frames);
 	return 0;
@@ -508,7 +508,7 @@ static int __devinit gnttab_init(void)
 	unsigned int max_nr_glist_frames, nr_glist_frames;
 	unsigned int nr_init_grefs;
 
-	if (!is_running_on_xen())
+	if (!xen_domain())
 		return -ENODEV;
 
 	nr_grant_frames = 1;
diff --git a/drivers/xen/manage.c b/drivers/xen/manage.c
new file mode 100644
index 000000000000..d0e87cbe157c
--- /dev/null
+++ b/drivers/xen/manage.c
@@ -0,0 +1,254 @@
+/*
+ * Handle extern requests for shutdown, reboot and sysrq
+ */
+#include <linux/kernel.h>
+#include <linux/err.h>
+#include <linux/reboot.h>
+#include <linux/sysrq.h>
+#include <linux/stop_machine.h>
+#include <linux/freezer.h>
+
+#include <xen/xenbus.h>
+#include <xen/grant_table.h>
+#include <xen/events.h>
+#include <xen/hvc-console.h>
+#include <xen/xen-ops.h>
+
+#include <asm/xen/hypercall.h>
+#include <asm/xen/page.h>
+
+enum shutdown_state {
+	SHUTDOWN_INVALID = -1,
+	SHUTDOWN_POWEROFF = 0,
+	SHUTDOWN_SUSPEND = 2,
+	/* Code 3 is SHUTDOWN_CRASH, which we don't use because the domain can only
+	   report a crash, not be instructed to crash!
+	   HALT is the same as POWEROFF, as far as we're concerned.  The tools use
+	   the distinction when we return the reason code to them.  */
+	 SHUTDOWN_HALT = 4,
+};
+
+/* Ignore multiple shutdown requests. */
+static enum shutdown_state shutting_down = SHUTDOWN_INVALID;
+
+#ifdef CONFIG_PM_SLEEP
+static int xen_suspend(void *data)
+{
+	int *cancelled = data;
+	int err;
+
+	BUG_ON(!irqs_disabled());
+
+	load_cr3(swapper_pg_dir);
+
+	err = device_power_down(PMSG_SUSPEND);
+	if (err) {
+		printk(KERN_ERR "xen_suspend: device_power_down failed: %d\n",
+		       err);
+		return err;
+	}
+
+	xen_mm_pin_all();
+	gnttab_suspend();
+	xen_pre_suspend();
+
+	/*
+	 * 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));
+
+	xen_post_suspend(*cancelled);
+	gnttab_resume();
+	xen_mm_unpin_all();
+
+	device_power_up(PMSG_RESUME);
+
+	if (!*cancelled) {
+		xen_irq_resume();
+		xen_console_resume();
+		xen_timer_resume();
+	}
+
+	return 0;
+}
+
+static void do_suspend(void)
+{
+	int err;
+	int cancelled = 1;
+
+	shutting_down = SHUTDOWN_SUSPEND;
+
+#ifdef CONFIG_PREEMPT
+	/* If the kernel is preemptible, we need to freeze all the processes
+	   to prevent them from being in the middle of a pagetable update
+	   during suspend. */
+	err = freeze_processes();
+	if (err) {
+		printk(KERN_ERR "xen suspend: freeze failed %d\n", err);
+		return;
+	}
+#endif
+
+	err = device_suspend(PMSG_SUSPEND);
+	if (err) {
+		printk(KERN_ERR "xen suspend: device_suspend %d\n", err);
+		goto out;
+	}
+
+	printk("suspending xenbus...\n");
+	/* XXX use normal device tree? */
+	xenbus_suspend();
+
+	err = stop_machine(xen_suspend, &cancelled, &cpumask_of_cpu(0));
+	if (err) {
+		printk(KERN_ERR "failed to start xen_suspend: %d\n", err);
+		goto out;
+	}
+
+	if (!cancelled) {
+		xen_arch_resume();
+		xenbus_resume();
+	} else
+		xenbus_suspend_cancel();
+
+	device_resume(PMSG_RESUME);
+
+	/* Make sure timer events get retriggered on all CPUs */
+	clock_was_set();
+out:
+#ifdef CONFIG_PREEMPT
+	thaw_processes();
+#endif
+	shutting_down = SHUTDOWN_INVALID;
+}
+#endif	/* CONFIG_PM_SLEEP */
+
+static void shutdown_handler(struct xenbus_watch *watch,
+			     const char **vec, unsigned int len)
+{
+	char *str;
+	struct xenbus_transaction xbt;
+	int err;
+
+	if (shutting_down != SHUTDOWN_INVALID)
+		return;
+
+ again:
+	err = xenbus_transaction_start(&xbt);
+	if (err)
+		return;
+
+	str = (char *)xenbus_read(xbt, "control", "shutdown", NULL);
+	/* Ignore read errors and empty reads. */
+	if (XENBUS_IS_ERR_READ(str)) {
+		xenbus_transaction_end(xbt, 1);
+		return;
+	}
+
+	xenbus_write(xbt, "control", "shutdown", "");
+
+	err = xenbus_transaction_end(xbt, 0);
+	if (err == -EAGAIN) {
+		kfree(str);
+		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
+	} else {
+		printk(KERN_INFO "Ignoring shutdown request: %s\n", str);
+		shutting_down = SHUTDOWN_INVALID;
+	}
+
+	kfree(str);
+}
+
+static void sysrq_handler(struct xenbus_watch *watch, const char **vec,
+			  unsigned int len)
+{
+	char sysrq_key = '\0';
+	struct xenbus_transaction xbt;
+	int err;
+
+ again:
+	err = xenbus_transaction_start(&xbt);
+	if (err)
+		return;
+	if (!xenbus_scanf(xbt, "control", "sysrq", "%c", &sysrq_key)) {
+		printk(KERN_ERR "Unable to read sysrq code in "
+		       "control/sysrq\n");
+		xenbus_transaction_end(xbt, 1);
+		return;
+	}
+
+	if (sysrq_key != '\0')
+		xenbus_printf(xbt, "control", "sysrq", "%c", '\0');
+
+	err = xenbus_transaction_end(xbt, 0);
+	if (err == -EAGAIN)
+		goto again;
+
+	if (sysrq_key != '\0')
+		handle_sysrq(sysrq_key, NULL);
+}
+
+static struct xenbus_watch shutdown_watch = {
+	.node = "control/shutdown",
+	.callback = shutdown_handler
+};
+
+static struct xenbus_watch sysrq_watch = {
+	.node = "control/sysrq",
+	.callback = sysrq_handler
+};
+
+static int setup_shutdown_watcher(void)
+{
+	int err;
+
+	err = register_xenbus_watch(&shutdown_watch);
+	if (err) {
+		printk(KERN_ERR "Failed to set shutdown watcher\n");
+		return err;
+	}
+
+	err = register_xenbus_watch(&sysrq_watch);
+	if (err) {
+		printk(KERN_ERR "Failed to set sysrq watcher\n");
+		return err;
+	}
+
+	return 0;
+}
+
+static int shutdown_event(struct notifier_block *notifier,
+			  unsigned long event,
+			  void *data)
+{
+	setup_shutdown_watcher();
+	return NOTIFY_DONE;
+}
+
+static int __init setup_shutdown_event(void)
+{
+	static struct notifier_block xenstore_notifier = {
+		.notifier_call = shutdown_event
+	};
+	register_xenstore_notifier(&xenstore_notifier);
+
+	return 0;
+}
+
+subsys_initcall(setup_shutdown_event);
diff --git a/drivers/xen/xenbus/xenbus_client.c b/drivers/xen/xenbus/xenbus_client.c
index 0f86b0ff7879..9678b3e98c63 100644
--- a/drivers/xen/xenbus/xenbus_client.c
+++ b/drivers/xen/xenbus/xenbus_client.c
@@ -117,7 +117,7 @@ int xenbus_watch_pathfmt(struct xenbus_device *dev,
 	char *path;
 
 	va_start(ap, pathfmt);
-	path = kvasprintf(GFP_KERNEL, pathfmt, ap);
+	path = kvasprintf(GFP_NOIO | __GFP_HIGH, pathfmt, ap);
 	va_end(ap);
 
 	if (!path) {
diff --git a/drivers/xen/xenbus/xenbus_comms.c b/drivers/xen/xenbus/xenbus_comms.c
index 6efbe3f29ca5..090c61ee8fd0 100644
--- a/drivers/xen/xenbus/xenbus_comms.c
+++ b/drivers/xen/xenbus/xenbus_comms.c
@@ -203,7 +203,6 @@ int xb_read(void *data, unsigned len)
 int xb_init_comms(void)
 {
 	struct xenstore_domain_interface *intf = xen_store_interface;
-	int err;
 
 	if (intf->req_prod != intf->req_cons)
 		printk(KERN_ERR "XENBUS request ring is not quiescent "
@@ -216,18 +215,20 @@ int xb_init_comms(void)
 		intf->rsp_cons = intf->rsp_prod;
 	}
 
-	if (xenbus_irq)
-		unbind_from_irqhandler(xenbus_irq, &xb_waitq);
+	if (xenbus_irq) {
+		/* Already have an irq; assume we're resuming */
+		rebind_evtchn_irq(xen_store_evtchn, xenbus_irq);
+	} else {
+		int err;
+		err = bind_evtchn_to_irqhandler(xen_store_evtchn, wake_waiting,
+						0, "xenbus", &xb_waitq);
+		if (err <= 0) {
+			printk(KERN_ERR "XENBUS request irq failed %i\n", err);
+			return err;
+		}
 
-	err = bind_evtchn_to_irqhandler(
-		xen_store_evtchn, wake_waiting,
-		0, "xenbus", &xb_waitq);
-	if (err <= 0) {
-		printk(KERN_ERR "XENBUS request irq failed %i\n", err);
-		return err;
+		xenbus_irq = err;
 	}
 
-	xenbus_irq = err;
-
 	return 0;
 }
diff --git a/drivers/xen/xenbus/xenbus_probe.c b/drivers/xen/xenbus/xenbus_probe.c
index 57ceb5346b74..7f24a98a446f 100644
--- a/drivers/xen/xenbus/xenbus_probe.c
+++ b/drivers/xen/xenbus/xenbus_probe.c
@@ -814,7 +814,7 @@ static int __init xenbus_probe_init(void)
 	DPRINTK("");
 
 	err = -ENODEV;
-	if (!is_running_on_xen())
+	if (!xen_domain())
 		goto out_error;
 
 	/* Register ourselves with the kernel bus subsystem */
@@ -829,7 +829,7 @@ static int __init xenbus_probe_init(void)
 	/*
 	 * Domain0 doesn't have a store_evtchn or store_mfn yet.
 	 */
-	if (is_initial_xendomain()) {
+	if (xen_initial_domain()) {
 		/* dom0 not yet supported */
 	} else {
 		xenstored_ready = 1;
@@ -846,7 +846,7 @@ static int __init xenbus_probe_init(void)
 		goto out_unreg_back;
 	}
 
-	if (!is_initial_xendomain())
+	if (!xen_initial_domain())
 		xenbus_probe(NULL);
 
 	return 0;
@@ -937,7 +937,7 @@ static void wait_for_devices(struct xenbus_driver *xendrv)
 	unsigned long timeout = jiffies + 10*HZ;
 	struct device_driver *drv = xendrv ? &xendrv->driver : NULL;
 
-	if (!ready_to_wait_for_devices || !is_running_on_xen())
+	if (!ready_to_wait_for_devices || !xen_domain())
 		return;
 
 	while (exists_disconnected_device(drv)) {
diff --git a/drivers/xen/xenbus/xenbus_xs.c b/drivers/xen/xenbus/xenbus_xs.c
index 227d53b12a5c..7f2f91c0e11d 100644
--- a/drivers/xen/xenbus/xenbus_xs.c
+++ b/drivers/xen/xenbus/xenbus_xs.c
@@ -283,9 +283,9 @@ static char *join(const char *dir, const char *name)
 	char *buffer;
 
 	if (strlen(name) == 0)
-		buffer = kasprintf(GFP_KERNEL, "%s", dir);
+		buffer = kasprintf(GFP_NOIO | __GFP_HIGH, "%s", dir);
 	else
-		buffer = kasprintf(GFP_KERNEL, "%s/%s", dir, name);
+		buffer = kasprintf(GFP_NOIO | __GFP_HIGH, "%s/%s", dir, name);
 	return (!buffer) ? ERR_PTR(-ENOMEM) : buffer;
 }
 
@@ -297,7 +297,7 @@ static char **split(char *strings, unsigned int len, unsigned int *num)
 	*num = count_strings(strings, len);
 
 	/* Transfer to one big alloc for easy freeing. */
-	ret = kmalloc(*num * sizeof(char *) + len, GFP_KERNEL);
+	ret = kmalloc(*num * sizeof(char *) + len, GFP_NOIO | __GFP_HIGH);
 	if (!ret) {
 		kfree(strings);
 		return ERR_PTR(-ENOMEM);
@@ -751,7 +751,7 @@ static int process_msg(void)
 	}
 
 
-	msg = kmalloc(sizeof(*msg), GFP_KERNEL);
+	msg = kmalloc(sizeof(*msg), GFP_NOIO | __GFP_HIGH);
 	if (msg == NULL) {
 		err = -ENOMEM;
 		goto out;
@@ -763,7 +763,7 @@ static int process_msg(void)
 		goto out;
 	}
 
-	body = kmalloc(msg->hdr.len + 1, GFP_KERNEL);
+	body = kmalloc(msg->hdr.len + 1, GFP_NOIO | __GFP_HIGH);
 	if (body == NULL) {
 		kfree(msg);
 		err = -ENOMEM;