11 files changed, 194 insertions, 80 deletions

diff --git a/drivers/xen/Makefile b/drivers/xen/Makefile
index babdca808861..c3621b9f4012 100644
--- a/drivers/xen/Makefile
+++ b/drivers/xen/Makefile
@@ -21,7 +21,7 @@ obj-$(CONFIG_XEN_GNTDEV)		+= xen-gntdev.o
 obj-$(CONFIG_XEN_GRANT_DEV_ALLOC)	+= xen-gntalloc.o
 obj-$(CONFIG_XENFS)			+= xenfs/
 obj-$(CONFIG_XEN_SYS_HYPERVISOR)	+= sys-hypervisor.o
-obj-$(CONFIG_XEN_PVHVM)			+= platform-pci.o
+obj-$(CONFIG_XEN_PVHVM_GUEST)		+= platform-pci.o
 obj-$(CONFIG_SWIOTLB_XEN)		+= swiotlb-xen.o
 obj-$(CONFIG_XEN_MCE_LOG)		+= mcelog.o
 obj-$(CONFIG_XEN_PCIDEV_BACKEND)	+= xen-pciback/
diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c
index 523dcdf39cc9..3729bea0c989 100644
--- a/drivers/xen/grant-table.c
+++ b/drivers/xen/grant-table.c
@@ -813,6 +813,129 @@ int gnttab_alloc_pages(int nr_pages, struct page **pages)
 }
 EXPORT_SYMBOL_GPL(gnttab_alloc_pages);
 
+#ifdef CONFIG_XEN_UNPOPULATED_ALLOC
+static inline void cache_init(struct gnttab_page_cache *cache)
+{
+	cache->pages = NULL;
+}
+
+static inline bool cache_empty(struct gnttab_page_cache *cache)
+{
+	return !cache->pages;
+}
+
+static inline struct page *cache_deq(struct gnttab_page_cache *cache)
+{
+	struct page *page;
+
+	page = cache->pages;
+	cache->pages = page->zone_device_data;
+
+	return page;
+}
+
+static inline void cache_enq(struct gnttab_page_cache *cache, struct page *page)
+{
+	page->zone_device_data = cache->pages;
+	cache->pages = page;
+}
+#else
+static inline void cache_init(struct gnttab_page_cache *cache)
+{
+	INIT_LIST_HEAD(&cache->pages);
+}
+
+static inline bool cache_empty(struct gnttab_page_cache *cache)
+{
+	return list_empty(&cache->pages);
+}
+
+static inline struct page *cache_deq(struct gnttab_page_cache *cache)
+{
+	struct page *page;
+
+	page = list_first_entry(&cache->pages, struct page, lru);
+	list_del(&page->lru);
+
+	return page;
+}
+
+static inline void cache_enq(struct gnttab_page_cache *cache, struct page *page)
+{
+	list_add(&page->lru, &cache->pages);
+}
+#endif
+
+void gnttab_page_cache_init(struct gnttab_page_cache *cache)
+{
+	spin_lock_init(&cache->lock);
+	cache_init(cache);
+	cache->num_pages = 0;
+}
+EXPORT_SYMBOL_GPL(gnttab_page_cache_init);
+
+int gnttab_page_cache_get(struct gnttab_page_cache *cache, struct page **page)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&cache->lock, flags);
+
+	if (cache_empty(cache)) {
+		spin_unlock_irqrestore(&cache->lock, flags);
+		return gnttab_alloc_pages(1, page);
+	}
+
+	page[0] = cache_deq(cache);
+	cache->num_pages--;
+
+	spin_unlock_irqrestore(&cache->lock, flags);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(gnttab_page_cache_get);
+
+void gnttab_page_cache_put(struct gnttab_page_cache *cache, struct page **page,
+			   unsigned int num)
+{
+	unsigned long flags;
+	unsigned int i;
+
+	spin_lock_irqsave(&cache->lock, flags);
+
+	for (i = 0; i < num; i++)
+		cache_enq(cache, page[i]);
+	cache->num_pages += num;
+
+	spin_unlock_irqrestore(&cache->lock, flags);
+}
+EXPORT_SYMBOL_GPL(gnttab_page_cache_put);
+
+void gnttab_page_cache_shrink(struct gnttab_page_cache *cache, unsigned int num)
+{
+	struct page *page[10];
+	unsigned int i = 0;
+	unsigned long flags;
+
+	spin_lock_irqsave(&cache->lock, flags);
+
+	while (cache->num_pages > num) {
+		page[i] = cache_deq(cache);
+		cache->num_pages--;
+		if (++i == ARRAY_SIZE(page)) {
+			spin_unlock_irqrestore(&cache->lock, flags);
+			gnttab_free_pages(i, page);
+			i = 0;
+			spin_lock_irqsave(&cache->lock, flags);
+		}
+	}
+
+	spin_unlock_irqrestore(&cache->lock, flags);
+
+	if (i != 0)
+		gnttab_free_pages(i, page);
+}
+EXPORT_SYMBOL_GPL(gnttab_page_cache_shrink);
+
 void gnttab_pages_clear_private(int nr_pages, struct page **pages)
 {
 	int i;
diff --git a/drivers/xen/manage.c b/drivers/xen/manage.c
index cd046684e0d1..374d36de7f5a 100644
--- a/drivers/xen/manage.c
+++ b/drivers/xen/manage.c
@@ -179,6 +179,7 @@ static int poweroff_nb(struct notifier_block *cb, unsigned long code, void *unus
 	case SYS_HALT:
 	case SYS_POWER_OFF:
 		shutting_down = SHUTDOWN_POWEROFF;
+		break;
 	default:
 		break;
 	}
diff --git a/drivers/xen/unpopulated-alloc.c b/drivers/xen/unpopulated-alloc.c
index 8c512ea550bb..e64e6befc63b 100644
--- a/drivers/xen/unpopulated-alloc.c
+++ b/drivers/xen/unpopulated-alloc.c
@@ -12,7 +12,7 @@
 #include <xen/xen.h>
 
 static DEFINE_MUTEX(list_lock);
-static LIST_HEAD(page_list);
+static struct page *page_list;
 static unsigned int list_count;
 
 static int fill_list(unsigned int nr_pages)
@@ -27,11 +27,6 @@ static int fill_list(unsigned int nr_pages)
 	if (!res)
 		return -ENOMEM;
 
-	pgmap = kzalloc(sizeof(*pgmap), GFP_KERNEL);
-	if (!pgmap)
-		goto err_pgmap;
-
-	pgmap->type = MEMORY_DEVICE_GENERIC;
 	res->name = "Xen scratch";
 	res->flags = IORESOURCE_MEM | IORESOURCE_BUSY;
 
@@ -43,6 +38,11 @@ static int fill_list(unsigned int nr_pages)
 		goto err_resource;
 	}
 
+	pgmap = kzalloc(sizeof(*pgmap), GFP_KERNEL);
+	if (!pgmap)
+		goto err_pgmap;
+
+	pgmap->type = MEMORY_DEVICE_GENERIC;
 	pgmap->range = (struct range) {
 		.start = res->start,
 		.end = res->end,
@@ -84,17 +84,18 @@ static int fill_list(unsigned int nr_pages)
 		struct page *pg = virt_to_page(vaddr + PAGE_SIZE * i);
 
 		BUG_ON(!virt_addr_valid(vaddr + PAGE_SIZE * i));
-		list_add(&pg->lru, &page_list);
+		pg->zone_device_data = page_list;
+		page_list = pg;
 		list_count++;
 	}
 
 	return 0;
 
 err_memremap:
-	release_resource(res);
-err_resource:
 	kfree(pgmap);
 err_pgmap:
+	release_resource(res);
+err_resource:
 	kfree(res);
 	return ret;
 }
@@ -118,12 +119,10 @@ int xen_alloc_unpopulated_pages(unsigned int nr_pages, struct page **pages)
 	}
 
 	for (i = 0; i < nr_pages; i++) {
-		struct page *pg = list_first_entry_or_null(&page_list,
-							   struct page,
-							   lru);
+		struct page *pg = page_list;
 
 		BUG_ON(!pg);
-		list_del(&pg->lru);
+		page_list = pg->zone_device_data;
 		list_count--;
 		pages[i] = pg;
 
@@ -134,7 +133,8 @@ int xen_alloc_unpopulated_pages(unsigned int nr_pages, struct page **pages)
 				unsigned int j;
 
 				for (j = 0; j <= i; j++) {
-					list_add(&pages[j]->lru, &page_list);
+					pages[j]->zone_device_data = page_list;
+					page_list = pages[j];
 					list_count++;
 				}
 				goto out;
@@ -160,7 +160,8 @@ void xen_free_unpopulated_pages(unsigned int nr_pages, struct page **pages)
 
 	mutex_lock(&list_lock);
 	for (i = 0; i < nr_pages; i++) {
-		list_add(&pages[i]->lru, &page_list);
+		pages[i]->zone_device_data = page_list;
+		page_list = pages[i];
 		list_count++;
 	}
 	mutex_unlock(&list_lock);
@@ -189,7 +190,8 @@ static int __init init(void)
 			struct page *pg =
 				pfn_to_page(xen_extra_mem[i].start_pfn + j);
 
-			list_add(&pg->lru, &page_list);
+			pg->zone_device_data = page_list;
+			page_list = pg;
 			list_count++;
 		}
 	}
diff --git a/drivers/xen/xen-pciback/xenbus.c b/drivers/xen/xen-pciback/xenbus.c
index 4b99ec3dec58..e7c692cfb2cf 100644
--- a/drivers/xen/xen-pciback/xenbus.c
+++ b/drivers/xen/xen-pciback/xenbus.c
@@ -689,7 +689,7 @@ static int xen_pcibk_xenbus_probe(struct xenbus_device *dev,
 
 	/* watch the backend node for backend configuration information */
 	err = xenbus_watch_path(dev, dev->nodename, &pdev->be_watch,
-				xen_pcibk_be_watch);
+				NULL, xen_pcibk_be_watch);
 	if (err)
 		goto out;
 
diff --git a/drivers/xen/xen-scsiback.c b/drivers/xen/xen-scsiback.c
index 4acc4e899600..862162dca33c 100644
--- a/drivers/xen/xen-scsiback.c
+++ b/drivers/xen/xen-scsiback.c
@@ -99,6 +99,8 @@ struct vscsibk_info {
 	struct list_head v2p_entry_lists;
 
 	wait_queue_head_t waiting_to_free;
+
+	struct gnttab_page_cache free_pages;
 };
 
 /* theoretical maximum of grants for one request */
@@ -188,10 +190,6 @@ module_param_named(max_buffer_pages, scsiback_max_buffer_pages, int, 0644);
 MODULE_PARM_DESC(max_buffer_pages,
 "Maximum number of free pages to keep in backend buffer");
 
-static DEFINE_SPINLOCK(free_pages_lock);
-static int free_pages_num;
-static LIST_HEAD(scsiback_free_pages);
-
 /* Global spinlock to protect scsiback TPG list */
 static DEFINE_MUTEX(scsiback_mutex);
 static LIST_HEAD(scsiback_list);
@@ -207,41 +205,6 @@ static void scsiback_put(struct vscsibk_info *info)
 		wake_up(&info->waiting_to_free);
 }
 
-static void put_free_pages(struct page **page, int num)
-{
-	unsigned long flags;
-	int i = free_pages_num + num, n = num;
-
-	if (num == 0)
-		return;
-	if (i > scsiback_max_buffer_pages) {
-		n = min(num, i - scsiback_max_buffer_pages);
-		gnttab_free_pages(n, page + num - n);
-		n = num - n;
-	}
-	spin_lock_irqsave(&free_pages_lock, flags);
-	for (i = 0; i < n; i++)
-		list_add(&page[i]->lru, &scsiback_free_pages);
-	free_pages_num += n;
-	spin_unlock_irqrestore(&free_pages_lock, flags);
-}
-
-static int get_free_page(struct page **page)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&free_pages_lock, flags);
-	if (list_empty(&scsiback_free_pages)) {
-		spin_unlock_irqrestore(&free_pages_lock, flags);
-		return gnttab_alloc_pages(1, page);
-	}
-	page[0] = list_first_entry(&scsiback_free_pages, struct page, lru);
-	list_del(&page[0]->lru);
-	free_pages_num--;
-	spin_unlock_irqrestore(&free_pages_lock, flags);
-	return 0;
-}
-
 static unsigned long vaddr_page(struct page *page)
 {
 	unsigned long pfn = page_to_pfn(page);
@@ -302,7 +265,8 @@ static void scsiback_fast_flush_area(struct vscsibk_pend *req)
 		BUG_ON(err);
 	}
 
-	put_free_pages(req->pages, req->n_grants);
+	gnttab_page_cache_put(&req->info->free_pages, req->pages,
+			      req->n_grants);
 	req->n_grants = 0;
 }
 
@@ -445,8 +409,8 @@ static int scsiback_gnttab_data_map_list(struct vscsibk_pend *pending_req,
 	struct vscsibk_info *info = pending_req->info;
 
 	for (i = 0; i < cnt; i++) {
-		if (get_free_page(pg + mapcount)) {
-			put_free_pages(pg, mapcount);
+		if (gnttab_page_cache_get(&info->free_pages, pg + mapcount)) {
+			gnttab_page_cache_put(&info->free_pages, pg, mapcount);
 			pr_err("no grant page\n");
 			return -ENOMEM;
 		}
@@ -796,6 +760,8 @@ static int scsiback_do_cmd_fn(struct vscsibk_info *info,
 		cond_resched();
 	}
 
+	gnttab_page_cache_shrink(&info->free_pages, scsiback_max_buffer_pages);
+
 	RING_FINAL_CHECK_FOR_REQUESTS(&info->ring, more_to_do);
 	return more_to_do;
 }
@@ -1233,6 +1199,8 @@ static int scsiback_remove(struct xenbus_device *dev)
 
 	scsiback_release_translation_entry(info);
 
+	gnttab_page_cache_shrink(&info->free_pages, 0);
+
 	dev_set_drvdata(&dev->dev, NULL);
 
 	return 0;
@@ -1263,6 +1231,7 @@ static int scsiback_probe(struct xenbus_device *dev,
 	info->irq = 0;
 	INIT_LIST_HEAD(&info->v2p_entry_lists);
 	spin_lock_init(&info->v2p_lock);
+	gnttab_page_cache_init(&info->free_pages);
 
 	err = xenbus_printf(XBT_NIL, dev->nodename, "feature-sg-grant", "%u",
 			    SG_ALL);
@@ -1879,13 +1848,6 @@ out:
 
 static void __exit scsiback_exit(void)
 {
-	struct page *page;
-
-	while (free_pages_num) {
-		if (get_free_page(&page))
-			BUG();
-		gnttab_free_pages(1, &page);
-	}
 	target_unregister_template(&scsiback_ops);
 	xenbus_unregister_driver(&scsiback_driver);
 }
diff --git a/drivers/xen/xenbus/xenbus.h b/drivers/xen/xenbus/xenbus.h
index 5f5b8a7d5b80..2a93b7c9c159 100644
--- a/drivers/xen/xenbus/xenbus.h
+++ b/drivers/xen/xenbus/xenbus.h
@@ -44,6 +44,8 @@ struct xen_bus_type {
 	int (*get_bus_id)(char bus_id[XEN_BUS_ID_SIZE], const char *nodename);
 	int (*probe)(struct xen_bus_type *bus, const char *type,
 		     const char *dir);
+	bool (*otherend_will_handle)(struct xenbus_watch *watch,
+				     const char *path, const char *token);
 	void (*otherend_changed)(struct xenbus_watch *watch, const char *path,
 				 const char *token);
 	struct bus_type bus;
diff --git a/drivers/xen/xenbus/xenbus_client.c b/drivers/xen/xenbus/xenbus_client.c
index fd80e318b99c..0cd728961fce 100644
--- a/drivers/xen/xenbus/xenbus_client.c
+++ b/drivers/xen/xenbus/xenbus_client.c
@@ -127,18 +127,22 @@ EXPORT_SYMBOL_GPL(xenbus_strstate);
  */
 int xenbus_watch_path(struct xenbus_device *dev, const char *path,
 		      struct xenbus_watch *watch,
+		      bool (*will_handle)(struct xenbus_watch *,
+					  const char *, const char *),
 		      void (*callback)(struct xenbus_watch *,
 				       const char *, const char *))
 {
 	int err;
 
 	watch->node = path;
+	watch->will_handle = will_handle;
 	watch->callback = callback;
 
 	err = register_xenbus_watch(watch);
 
 	if (err) {
 		watch->node = NULL;
+		watch->will_handle = NULL;
 		watch->callback = NULL;
 		xenbus_dev_fatal(dev, err, "adding watch on %s", path);
 	}
@@ -165,6 +169,8 @@ EXPORT_SYMBOL_GPL(xenbus_watch_path);
  */
 int xenbus_watch_pathfmt(struct xenbus_device *dev,
 			 struct xenbus_watch *watch,
+			 bool (*will_handle)(struct xenbus_watch *,
+					const char *, const char *),
 			 void (*callback)(struct xenbus_watch *,
 					  const char *, const char *),
 			 const char *pathfmt, ...)
@@ -181,7 +187,7 @@ int xenbus_watch_pathfmt(struct xenbus_device *dev,
 		xenbus_dev_fatal(dev, -ENOMEM, "allocating path for watch");
 		return -ENOMEM;
 	}
-	err = xenbus_watch_path(dev, path, watch, callback);
+	err = xenbus_watch_path(dev, path, watch, will_handle, callback);
 
 	if (err)
 		kfree(path);
diff --git a/drivers/xen/xenbus/xenbus_probe.c b/drivers/xen/xenbus/xenbus_probe.c
index 38725d97d909..44634d970a5c 100644
--- a/drivers/xen/xenbus/xenbus_probe.c
+++ b/drivers/xen/xenbus/xenbus_probe.c
@@ -136,6 +136,7 @@ static int watch_otherend(struct xenbus_device *dev)
 		container_of(dev->dev.bus, struct xen_bus_type, bus);
 
 	return xenbus_watch_pathfmt(dev, &dev->otherend_watch,
+				    bus->otherend_will_handle,
 				    bus->otherend_changed,
 				    "%s/%s", dev->otherend, "state");
 }
diff --git a/drivers/xen/xenbus/xenbus_probe_backend.c b/drivers/xen/xenbus/xenbus_probe_backend.c
index 2ba699897e6d..5abded97e1a7 100644
--- a/drivers/xen/xenbus/xenbus_probe_backend.c
+++ b/drivers/xen/xenbus/xenbus_probe_backend.c
@@ -180,6 +180,12 @@ static int xenbus_probe_backend(struct xen_bus_type *bus, const char *type,
 	return err;
 }
 
+static bool frontend_will_handle(struct xenbus_watch *watch,
+				 const char *path, const char *token)
+{
+	return watch->nr_pending == 0;
+}
+
 static void frontend_changed(struct xenbus_watch *watch,
 			     const char *path, const char *token)
 {
@@ -191,6 +197,7 @@ static struct xen_bus_type xenbus_backend = {
 	.levels = 3,		/* backend/type/<frontend>/<id> */
 	.get_bus_id = backend_bus_id,
 	.probe = xenbus_probe_backend,
+	.otherend_will_handle = frontend_will_handle,
 	.otherend_changed = frontend_changed,
 	.bus = {
 		.name		= "xen-backend",
diff --git a/drivers/xen/xenbus/xenbus_xs.c b/drivers/xen/xenbus/xenbus_xs.c
index 3a06eb699f33..12e02eb01f59 100644
--- a/drivers/xen/xenbus/xenbus_xs.c
+++ b/drivers/xen/xenbus/xenbus_xs.c
@@ -705,9 +705,13 @@ int xs_watch_msg(struct xs_watch_event *event)
 
 	spin_lock(&watches_lock);
 	event->handle = find_watch(event->token);
-	if (event->handle != NULL) {
+	if (event->handle != NULL &&
+			(!event->handle->will_handle ||
+			 event->handle->will_handle(event->handle,
+				 event->path, event->token))) {
 		spin_lock(&watch_events_lock);
 		list_add_tail(&event->list, &watch_events);
+		event->handle->nr_pending++;
 		wake_up(&watch_events_waitq);
 		spin_unlock(&watch_events_lock);
 	} else
@@ -765,6 +769,8 @@ int register_xenbus_watch(struct xenbus_watch *watch)
 
 	sprintf(token, "%lX", (long)watch);
 
+	watch->nr_pending = 0;
+
 	down_read(&xs_watch_rwsem);
 
 	spin_lock(&watches_lock);
@@ -814,11 +820,14 @@ void unregister_xenbus_watch(struct xenbus_watch *watch)
 
 	/* Cancel pending watch events. */
 	spin_lock(&watch_events_lock);
-	list_for_each_entry_safe(event, tmp, &watch_events, list) {
-		if (event->handle != watch)
-			continue;
-		list_del(&event->list);
-		kfree(event);
+	if (watch->nr_pending) {
+		list_for_each_entry_safe(event, tmp, &watch_events, list) {
+			if (event->handle != watch)
+				continue;
+			list_del(&event->list);
+			kfree(event);
+		}
+		watch->nr_pending = 0;
 	}
 	spin_unlock(&watch_events_lock);
 
@@ -865,7 +874,6 @@ void xs_suspend_cancel(void)
 
 static int xenwatch_thread(void *unused)
 {
-	struct list_head *ent;
 	struct xs_watch_event *event;
 
 	xenwatch_pid = current->pid;
@@ -880,13 +888,15 @@ static int xenwatch_thread(void *unused)
 		mutex_lock(&xenwatch_mutex);
 
 		spin_lock(&watch_events_lock);
-		ent = watch_events.next;
-		if (ent != &watch_events)
-			list_del(ent);
+		event = list_first_entry_or_null(&watch_events,
+				struct xs_watch_event, list);
+		if (event) {
+			list_del(&event->list);
+			event->handle->nr_pending--;
+		}
 		spin_unlock(&watch_events_lock);
 
-		if (ent != &watch_events) {
-			event = list_entry(ent, struct xs_watch_event, list);
+		if (event) {
 			event->handle->callback(event->handle, event->path,
 						event->token);
 			kfree(event);