1 files changed, 221 insertions, 145 deletions

diff --git a/drivers/xen/balloon.c b/drivers/xen/balloon.c
index ceb5048de9a7..49c3f9926394 100644
--- a/drivers/xen/balloon.c
+++ b/drivers/xen/balloon.c
@@ -43,6 +43,8 @@
 #include <linux/sched.h>
 #include <linux/cred.h>
 #include <linux/errno.h>
+#include <linux/freezer.h>
+#include <linux/kthread.h>
 #include <linux/mm.h>
 #include <linux/memblock.h>
 #include <linux/pagemap.h>
@@ -56,10 +58,10 @@
 #include <linux/percpu-defs.h>
 #include <linux/slab.h>
 #include <linux/sysctl.h>
+#include <linux/moduleparam.h>
+#include <linux/jiffies.h>
 
 #include <asm/page.h>
-#include <asm/pgalloc.h>
-#include <asm/pgtable.h>
 #include <asm/tlb.h>
 
 #include <asm/xen/hypervisor.h>
@@ -73,44 +75,29 @@
 #include <xen/page.h>
 #include <xen/mem-reservation.h>
 
-static int xen_hotplug_unpopulated;
+#undef MODULE_PARAM_PREFIX
+#define MODULE_PARAM_PREFIX "xen."
 
-#ifdef CONFIG_XEN_BALLOON_MEMORY_HOTPLUG
+static uint __read_mostly balloon_boot_timeout = 180;
+module_param(balloon_boot_timeout, uint, 0444);
 
-static int zero;
-static int one = 1;
+#ifdef CONFIG_XEN_BALLOON_MEMORY_HOTPLUG
+static int xen_hotplug_unpopulated;
 
-static struct ctl_table balloon_table[] = {
+static const struct ctl_table balloon_table[] = {
 	{
 		.procname	= "hotplug_unpopulated",
 		.data		= &xen_hotplug_unpopulated,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec_minmax,
-		.extra1         = &zero,
-		.extra2         = &one,
+		.extra1         = SYSCTL_ZERO,
+		.extra2         = SYSCTL_ONE,
 	},
-	{ }
-};
-
-static struct ctl_table balloon_root[] = {
-	{
-		.procname	= "balloon",
-		.mode		= 0555,
-		.child		= balloon_table,
-	},
-	{ }
-};
-
-static struct ctl_table xen_root[] = {
-	{
-		.procname	= "xen",
-		.mode		= 0555,
-		.child		= balloon_root,
-	},
-	{ }
 };
 
+#else
+#define xen_hotplug_unpopulated 0
 #endif
 
 /*
@@ -120,7 +107,7 @@ static struct ctl_table xen_root[] = {
 #define EXTENT_ORDER (fls(XEN_PFN_PER_PAGE) - 1)
 
 /*
- * balloon_process() state:
+ * balloon_thread() state:
  *
  * BP_DONE: done or nothing to do,
  * BP_WAIT: wait to be rescheduled,
@@ -128,13 +115,15 @@ static struct ctl_table xen_root[] = {
  * BP_ECANCELED: error, balloon operation canceled.
  */
 
-enum bp_state {
+static enum bp_state {
 	BP_DONE,
 	BP_WAIT,
 	BP_EAGAIN,
 	BP_ECANCELED
-};
+} balloon_state = BP_DONE;
 
+/* Main waiting point for xen-balloon thread. */
+static DECLARE_WAIT_QUEUE_HEAD(balloon_thread_wq);
 
 static DEFINE_MUTEX(balloon_mutex);
 
@@ -149,18 +138,17 @@ static xen_pfn_t frame_list[PAGE_SIZE / sizeof(xen_pfn_t)];
 static LIST_HEAD(ballooned_pages);
 static DECLARE_WAIT_QUEUE_HEAD(balloon_wq);
 
-/* Main work function, always executed in process context. */
-static void balloon_process(struct work_struct *work);
-static DECLARE_DELAYED_WORK(balloon_worker, balloon_process);
-
 /* When ballooning out (allocating memory to return to Xen) we don't really
    want the kernel to try too hard since that can trigger the oom killer. */
 #define GFP_BALLOON \
 	(GFP_HIGHUSER | __GFP_NOWARN | __GFP_NORETRY | __GFP_NOMEMALLOC)
 
 /* balloon_append: add the given page to the balloon. */
-static void __balloon_append(struct page *page)
+static void balloon_append(struct page *page)
 {
+	if (!PageOffline(page))
+		__SetPageOffline(page);
+
 	/* Lowmem is re-populated first, so highmem pages go at list tail. */
 	if (PageHighMem(page)) {
 		list_add_tail(&page->lru, &ballooned_pages);
@@ -169,12 +157,9 @@ static void __balloon_append(struct page *page)
 		list_add(&page->lru, &ballooned_pages);
 		balloon_stats.balloon_low++;
 	}
-	wake_up(&balloon_wq);
-}
+	inc_node_page_state(page, NR_BALLOON_PAGES);
 
-static void balloon_append(struct page *page)
-{
-	__balloon_append(page);
+	wake_up(&balloon_wq);
 }
 
 /* balloon_retrieve: rescue a page from the balloon, if it is not empty. */
@@ -195,6 +180,9 @@ static struct page *balloon_retrieve(bool require_lowmem)
 	else
 		balloon_stats.balloon_low--;
 
+	__ClearPageOffline(page);
+	dec_node_page_state(page, NR_BALLOON_PAGES);
+
 	return page;
 }
 
@@ -206,18 +194,15 @@ static struct page *balloon_next_page(struct page *page)
 	return list_entry(next, struct page, lru);
 }
 
-static enum bp_state update_schedule(enum bp_state state)
+static void update_schedule(void)
 {
-	if (state == BP_WAIT)
-		return BP_WAIT;
-
-	if (state == BP_ECANCELED)
-		return BP_ECANCELED;
+	if (balloon_state == BP_WAIT || balloon_state == BP_ECANCELED)
+		return;
 
-	if (state == BP_DONE) {
+	if (balloon_state == BP_DONE) {
 		balloon_stats.schedule_delay = 1;
 		balloon_stats.retry_count = 1;
-		return BP_DONE;
+		return;
 	}
 
 	++balloon_stats.retry_count;
@@ -226,7 +211,8 @@ static enum bp_state update_schedule(enum bp_state state)
 			balloon_stats.retry_count > balloon_stats.max_retry_count) {
 		balloon_stats.schedule_delay = 1;
 		balloon_stats.retry_count = 1;
-		return BP_ECANCELED;
+		balloon_state = BP_ECANCELED;
+		return;
 	}
 
 	balloon_stats.schedule_delay <<= 1;
@@ -234,7 +220,7 @@ static enum bp_state update_schedule(enum bp_state state)
 	if (balloon_stats.schedule_delay > balloon_stats.max_schedule_delay)
 		balloon_stats.schedule_delay = balloon_stats.max_schedule_delay;
 
-	return BP_EAGAIN;
+	balloon_state = BP_EAGAIN;
 }
 
 #ifdef CONFIG_XEN_BALLOON_MEMORY_HOTPLUG
@@ -272,20 +258,6 @@ static struct resource *additional_memory_resource(phys_addr_t size)
 		return NULL;
 	}
 
-#ifdef CONFIG_SPARSEMEM
-	{
-		unsigned long limit = 1UL << (MAX_PHYSMEM_BITS - PAGE_SHIFT);
-		unsigned long pfn = res->start >> PAGE_SHIFT;
-
-		if (pfn > limit) {
-			pr_err("New System RAM resource outside addressable RAM (%lu > %lu)\n",
-			       pfn, limit);
-			release_memory_resource(res);
-			return NULL;
-		}
-	}
-#endif
-
 	return res;
 }
 
@@ -330,7 +302,7 @@ static enum bp_state reserve_additional_memory(void)
          * are not restored since this region is now known not to
          * conflict with any devices.
          */ 
-	if (!xen_feature(XENFEAT_auto_translated_physmap)) {
+	if (xen_pv_domain()) {
 		unsigned long pfn, i;
 
 		pfn = PFN_DOWN(resource->start);
@@ -352,7 +324,7 @@ static enum bp_state reserve_additional_memory(void)
 	mutex_unlock(&balloon_mutex);
 	/* add_memory_resource() requires the device_hotplug lock */
 	lock_device_hotplug();
-	rc = add_memory_resource(nid, resource);
+	rc = add_memory_resource(nid, resource, MHP_MERGE_RESOURCE);
 	unlock_device_hotplug();
 	mutex_lock(&balloon_mutex);
 
@@ -369,21 +341,25 @@ static enum bp_state reserve_additional_memory(void)
 	return BP_ECANCELED;
 }
 
-static void xen_online_page(struct page *page)
+static void xen_online_page(struct page *page, unsigned int order)
 {
-	__online_page_set_limits(page);
+	unsigned long i, size = (1 << order);
+	unsigned long start_pfn = page_to_pfn(page);
+	struct page *p;
 
+	pr_debug("Online %lu pages starting at pfn 0x%lx\n", size, start_pfn);
 	mutex_lock(&balloon_mutex);
-
-	__balloon_append(page);
-
+	for (i = 0; i < size; i++) {
+		p = pfn_to_page(start_pfn + i);
+		balloon_append(p);
+	}
 	mutex_unlock(&balloon_mutex);
 }
 
 static int xen_memory_notifier(struct notifier_block *nb, unsigned long val, void *v)
 {
 	if (val == MEM_ONLINE)
-		schedule_delayed_work(&balloon_worker, 0);
+		wake_up(&balloon_thread_wq);
 
 	return NOTIFY_OK;
 }
@@ -395,7 +371,8 @@ static struct notifier_block xen_memory_nb = {
 #else
 static enum bp_state reserve_additional_memory(void)
 {
-	balloon_stats.target_pages = balloon_stats.current_pages;
+	balloon_stats.target_pages = balloon_stats.current_pages +
+				     balloon_stats.target_unpopulated;
 	return BP_ECANCELED;
 }
 #endif /* CONFIG_XEN_BALLOON_MEMORY_HOTPLUG */
@@ -440,7 +417,11 @@ static enum bp_state increase_reservation(unsigned long nr_pages)
 
 		xenmem_reservation_va_mapping_update(1, &page, &frame_list[i]);
 
-		/* Relinquish the page back to the allocator. */
+		/*
+		 * Relinquish the page back to the allocator. Note that
+		 * some pages, including ones added via xen_online_page(), might
+		 * not be marked reserved; free_reserved_page() will handle that.
+		 */
 		free_reserved_page(page);
 	}
 
@@ -507,43 +488,79 @@ static enum bp_state decrease_reservation(unsigned long nr_pages, gfp_t gfp)
 }
 
 /*
- * As this is a work item it is guaranteed to run as a single instance only.
+ * Stop waiting if either state is BP_DONE and ballooning action is
+ * needed, or if the credit has changed while state is not BP_DONE.
+ */
+static bool balloon_thread_cond(long credit)
+{
+	if (balloon_state == BP_DONE)
+		credit = 0;
+
+	return current_credit() != credit || kthread_should_stop();
+}
+
+/*
+ * As this is a kthread it is guaranteed to run as a single instance only.
  * We may of course race updates of the target counts (which are protected
  * by the balloon lock), or with changes to the Xen hard limit, but we will
  * recover from these in time.
  */
-static void balloon_process(struct work_struct *work)
+static int balloon_thread(void *unused)
 {
-	enum bp_state state = BP_DONE;
 	long credit;
+	unsigned long timeout;
+
+	set_freezable();
+	for (;;) {
+		switch (balloon_state) {
+		case BP_DONE:
+		case BP_ECANCELED:
+			timeout = 3600 * HZ;
+			break;
+		case BP_EAGAIN:
+			timeout = balloon_stats.schedule_delay * HZ;
+			break;
+		case BP_WAIT:
+			timeout = HZ;
+			break;
+		}
+
+		credit = current_credit();
+
+		wait_event_freezable_timeout(balloon_thread_wq,
+			balloon_thread_cond(credit), timeout);
 
+		if (kthread_should_stop())
+			return 0;
 
-	do {
 		mutex_lock(&balloon_mutex);
 
 		credit = current_credit();
 
 		if (credit > 0) {
 			if (balloon_is_inflated())
-				state = increase_reservation(credit);
+				balloon_state = increase_reservation(credit);
 			else
-				state = reserve_additional_memory();
+				balloon_state = reserve_additional_memory();
 		}
 
-		if (credit < 0)
-			state = decrease_reservation(-credit, GFP_BALLOON);
+		if (credit < 0) {
+			long n_pages;
+
+			n_pages = min(-credit, si_mem_available());
+			balloon_state = decrease_reservation(n_pages,
+							     GFP_BALLOON);
+			if (balloon_state == BP_DONE && n_pages != -credit &&
+			    n_pages < totalreserve_pages)
+				balloon_state = BP_EAGAIN;
+		}
 
-		state = update_schedule(state);
+		update_schedule();
 
 		mutex_unlock(&balloon_mutex);
 
 		cond_resched();
-
-	} while (credit && state == BP_DONE);
-
-	/* Schedule more work if there is some still to be done. */
-	if (state == BP_EAGAIN)
-		schedule_delayed_work(&balloon_worker, balloon_stats.schedule_delay * HZ);
+	}
 }
 
 /* Resets the Xen limit, sets new target, and kicks off processing. */
@@ -551,25 +568,30 @@ void balloon_set_new_target(unsigned long target)
 {
 	/* No need for lock. Not read-modify-write updates. */
 	balloon_stats.target_pages = target;
-	schedule_delayed_work(&balloon_worker, 0);
+	wake_up(&balloon_thread_wq);
 }
 EXPORT_SYMBOL_GPL(balloon_set_new_target);
 
-static int add_ballooned_pages(int nr_pages)
+static int add_ballooned_pages(unsigned int nr_pages)
 {
 	enum bp_state st;
 
 	if (xen_hotplug_unpopulated) {
 		st = reserve_additional_memory();
 		if (st != BP_ECANCELED) {
+			int rc;
+
 			mutex_unlock(&balloon_mutex);
-			wait_event(balloon_wq,
+			rc = wait_event_interruptible(balloon_wq,
 				   !list_empty(&ballooned_pages));
 			mutex_lock(&balloon_mutex);
-			return 0;
+			return rc ? -ENOMEM : 0;
 		}
 	}
 
+	if (si_mem_available() < nr_pages)
+		return -ENOMEM;
+
 	st = decrease_reservation(nr_pages, GFP_USER);
 	if (st != BP_DONE)
 		return -ENOMEM;
@@ -578,14 +600,14 @@ static int add_ballooned_pages(int nr_pages)
 }
 
 /**
- * alloc_xenballooned_pages - get pages that have been ballooned out
+ * xen_alloc_ballooned_pages - get pages that have been ballooned out
  * @nr_pages: Number of pages to get
  * @pages: pages returned
  * @return 0 on success, error otherwise
  */
-int alloc_xenballooned_pages(int nr_pages, struct page **pages)
+int xen_alloc_ballooned_pages(unsigned int nr_pages, struct page **pages)
 {
-	int pgno = 0;
+	unsigned int pgno = 0;
 	struct page *page;
 	int ret;
 
@@ -604,7 +626,7 @@ int alloc_xenballooned_pages(int nr_pages, struct page **pages)
 			 */
 			BUILD_BUG_ON(XEN_PAGE_SIZE != PAGE_SIZE);
 
-			if (!xen_feature(XENFEAT_auto_translated_physmap)) {
+			if (xen_pv_domain()) {
 				ret = xen_alloc_p2m_entry(page_to_pfn(page));
 				if (ret < 0)
 					goto out_undo;
@@ -620,19 +642,25 @@ int alloc_xenballooned_pages(int nr_pages, struct page **pages)
 	return 0;
  out_undo:
 	mutex_unlock(&balloon_mutex);
-	free_xenballooned_pages(pgno, pages);
+	xen_free_ballooned_pages(pgno, pages);
+	/*
+	 * NB: xen_free_ballooned_pages will only subtract pgno pages, but since
+	 * target_unpopulated is incremented with nr_pages at the start we need
+	 * to remove the remaining ones also, or accounting will be screwed.
+	 */
+	balloon_stats.target_unpopulated -= nr_pages - pgno;
 	return ret;
 }
-EXPORT_SYMBOL(alloc_xenballooned_pages);
+EXPORT_SYMBOL(xen_alloc_ballooned_pages);
 
 /**
- * free_xenballooned_pages - return pages retrieved with get_ballooned_pages
+ * xen_free_ballooned_pages - return pages retrieved with get_ballooned_pages
  * @nr_pages: Number of pages
  * @pages: pages to return
  */
-void free_xenballooned_pages(int nr_pages, struct page **pages)
+void xen_free_ballooned_pages(unsigned int nr_pages, struct page **pages)
 {
-	int i;
+	unsigned int i;
 
 	mutex_lock(&balloon_mutex);
 
@@ -645,52 +673,70 @@ void free_xenballooned_pages(int nr_pages, struct page **pages)
 
 	/* The balloon may be too large now. Shrink it if needed. */
 	if (current_credit())
-		schedule_delayed_work(&balloon_worker, 0);
+		wake_up(&balloon_thread_wq);
 
 	mutex_unlock(&balloon_mutex);
 }
-EXPORT_SYMBOL(free_xenballooned_pages);
+EXPORT_SYMBOL(xen_free_ballooned_pages);
 
-#ifdef CONFIG_XEN_PV
-static void __init balloon_add_region(unsigned long start_pfn,
-				      unsigned long pages)
+static int __init balloon_add_regions(void)
 {
+	unsigned long start_pfn, pages;
 	unsigned long pfn, extra_pfn_end;
-	struct page *page;
+	unsigned int i;
 
-	/*
-	 * If the amount of usable memory has been limited (e.g., with
-	 * the 'mem' command line parameter), don't add pages beyond
-	 * this limit.
-	 */
-	extra_pfn_end = min(max_pfn, start_pfn + pages);
-
-	for (pfn = start_pfn; pfn < extra_pfn_end; pfn++) {
-		page = pfn_to_page(pfn);
-		/* totalram_pages and totalhigh_pages do not
-		   include the boot-time balloon extension, so
-		   don't subtract from it. */
-		__balloon_append(page);
+	for (i = 0; i < XEN_EXTRA_MEM_MAX_REGIONS; i++) {
+		pages = xen_extra_mem[i].n_pfns;
+		if (!pages)
+			continue;
+
+		start_pfn = xen_extra_mem[i].start_pfn;
+
+		/*
+		 * If the amount of usable memory has been limited (e.g., with
+		 * the 'mem' command line parameter), don't add pages beyond
+		 * this limit.
+		 */
+		extra_pfn_end = min(max_pfn, start_pfn + pages);
+
+		for (pfn = start_pfn; pfn < extra_pfn_end; pfn++)
+			balloon_append(pfn_to_page(pfn));
+
+		/*
+		 * Extra regions are accounted for in the physmap, but need
+		 * decreasing from current_pages and target_pages to balloon
+		 * down the initial allocation, because they are already
+		 * accounted for in total_pages.
+		 */
+		pages = extra_pfn_end - start_pfn;
+		if (pages >= balloon_stats.current_pages ||
+		    pages >= balloon_stats.target_pages) {
+			WARN(1, "Extra pages underflow current target");
+			return -ERANGE;
+		}
+		balloon_stats.current_pages -= pages;
+		balloon_stats.target_pages -= pages;
 	}
 
-	balloon_stats.total_pages += extra_pfn_end - start_pfn;
+	return 0;
 }
-#endif
 
 static int __init balloon_init(void)
 {
+	struct task_struct *task;
+	int rc;
+
 	if (!xen_domain())
 		return -ENODEV;
 
 	pr_info("Initialising balloon driver\n");
 
-#ifdef CONFIG_XEN_PV
-	balloon_stats.current_pages = xen_pv_domain()
-		? min(xen_start_info->nr_pages - xen_released_pages, max_pfn)
-		: get_num_physpages();
-#else
-	balloon_stats.current_pages = get_num_physpages();
-#endif
+	if (xen_released_pages >= get_num_physpages()) {
+		WARN(1, "Released pages underflow current target");
+		return -ERANGE;
+	}
+
+	balloon_stats.current_pages = get_num_physpages() - xen_released_pages;
 	balloon_stats.target_pages  = balloon_stats.current_pages;
 	balloon_stats.balloon_low   = 0;
 	balloon_stats.balloon_high  = 0;
@@ -699,28 +745,23 @@ static int __init balloon_init(void)
 	balloon_stats.schedule_delay = 1;
 	balloon_stats.max_schedule_delay = 32;
 	balloon_stats.retry_count = 1;
-	balloon_stats.max_retry_count = RETRY_UNLIMITED;
+	balloon_stats.max_retry_count = 4;
 
 #ifdef CONFIG_XEN_BALLOON_MEMORY_HOTPLUG
 	set_online_page_callback(&xen_online_page);
 	register_memory_notifier(&xen_memory_nb);
-	register_sysctl_table(xen_root);
+	register_sysctl_init("xen/balloon", balloon_table);
 #endif
 
-#ifdef CONFIG_XEN_PV
-	{
-		int i;
+	rc = balloon_add_regions();
+	if (rc)
+		return rc;
 
-		/*
-		 * Initialize the balloon with pages from the extra memory
-		 * regions (see arch/x86/xen/setup.c).
-		 */
-		for (i = 0; i < XEN_EXTRA_MEM_MAX_REGIONS; i++)
-			if (xen_extra_mem[i].n_pfns)
-				balloon_add_region(xen_extra_mem[i].start_pfn,
-						   xen_extra_mem[i].n_pfns);
+	task = kthread_run(balloon_thread, NULL, "xen-balloon");
+	if (IS_ERR(task)) {
+		pr_err("xen-balloon thread could not be started, ballooning will not work!\n");
+		return PTR_ERR(task);
 	}
-#endif
 
 	/* Init the xen-balloon driver. */
 	xen_balloon_init();
@@ -728,3 +769,38 @@ static int __init balloon_init(void)
 	return 0;
 }
 subsys_initcall(balloon_init);
+
+static int __init balloon_wait_finish(void)
+{
+	long credit, last_credit = 0;
+	unsigned long last_changed = 0;
+
+	if (!xen_domain())
+		return -ENODEV;
+
+	/* PV guests don't need to wait. */
+	if (xen_pv_domain() || !current_credit())
+		return 0;
+
+	pr_notice("Waiting for initial ballooning down having finished.\n");
+
+	while ((credit = current_credit()) < 0) {
+		if (credit != last_credit) {
+			last_changed = jiffies;
+			last_credit = credit;
+		}
+		if (balloon_state == BP_ECANCELED) {
+			pr_warn_once("Initial ballooning failed, %ld pages need to be freed.\n",
+				     -credit);
+			if (time_is_before_eq_jiffies(last_changed + HZ * balloon_boot_timeout))
+				panic("Initial ballooning failed!\n");
+		}
+
+		schedule_timeout_interruptible(HZ / 10);
+	}
+
+	pr_notice("Initial ballooning down finished.\n");
+
+	return 0;
+}
+late_initcall_sync(balloon_wait_finish);