transfer vmscan pageout works to the flusher thread

@@ -716,6 +716,23 @@ get_next_work_item(struct backing_dev_in
 	return work;
 }
 
+static long wb_check_background_flush(struct bdi_writeback *wb)
+{
+	if (over_bground_thresh()) {
+
+		struct wb_writeback_work work = {
+			.nr_pages	= LONG_MAX,
+			.sync_mode	= WB_SYNC_NONE,
+			.for_background	= 1,
+			.range_cyclic	= 1,
+		};
+
+		return wb_writeback(wb, &work);
+	}
+
+	return 0;
+}
+
 static long wb_check_old_data_flush(struct bdi_writeback *wb)
 {
 	unsigned long expired;

@@ -787,6 +804,7 @@ long wb_do_writeback(struct bdi_writebac
 	 * Check for periodic writeback, kupdated() style
 	 */
 	wrote += wb_check_old_data_flush(wb);
+	wrote += wb_check_background_flush(wb);
 	clear_bit(BDI_writeback_running, &wb->bdi->state);
 
 	return wrote;

@@ -643,6 +643,14 @@ static long wb_writeback(struct bdi_writ
 			break;
 
 		/*
+		 * Background writeout and kupdate-style writeback are
+		 * easily livelockable. Stop them if there is other work
+		 * to do so that e.g. sync can proceed.
+		 */
+		if ((work->for_background || work->for_kupdate) &&
+		    !list_empty(&wb->bdi->work_list))
+			break;
+		/*
 		 * For background writeout, stop when we are below the
 		 * background dirty threshold
 		 */

@@ -79,13 +79,9 @@ static inline struct backing_dev_info *i
 	return sb->s_bdi;
 }
 
-static void bdi_queue_work(struct backing_dev_info *bdi,
-		struct wb_writeback_work *work)
+/* Wakeup flusher thread or forker thread to fork it. Requires bdi->wb_lock. */
+static void _bdi_wakeup_flusher(struct backing_dev_info *bdi)
 {
-	trace_writeback_queue(bdi, work);
-
-	spin_lock_bh(&bdi->wb_lock);
-	list_add_tail(&work->list, &bdi->work_list);
 	if (bdi->wb.task) {
 		wake_up_process(bdi->wb.task);
 	} else {

@@ -93,15 +89,26 @@ static void bdi_queue_work(struct backin
 		 * The bdi thread isn't there, wake up the forker thread which
 		 * will create and run it.
 		 */
-		trace_writeback_nothread(bdi, work);
 		wake_up_process(default_backing_dev_info.wb.task);
 	}
+}
+
+static void bdi_queue_work(struct backing_dev_info *bdi,
+			   struct wb_writeback_work *work)
+{
+	trace_writeback_queue(bdi, work);
+
+	spin_lock_bh(&bdi->wb_lock);
+	list_add_tail(&work->list, &bdi->work_list);
+	if (!bdi->wb.task)
+		trace_writeback_nothread(bdi, work);
+	_bdi_wakeup_flusher(bdi);
 	spin_unlock_bh(&bdi->wb_lock);
 }
 
 static void
 __bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages,
-		bool range_cyclic, bool for_background)
+		      bool range_cyclic)
 {
 	struct wb_writeback_work *work;
 

@@ -121,7 +128,6 @@ __bdi_start_writeback(struct backing_dev
 	work->sync_mode	= WB_SYNC_NONE;
 	work->nr_pages	= nr_pages;
 	work->range_cyclic = range_cyclic;
-	work->for_background = for_background;
 
 	bdi_queue_work(bdi, work);
 }

@@ -139,7 +145,7 @@ __bdi_start_writeback(struct backing_dev
  */
 void bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages)
 {
-	__bdi_start_writeback(bdi, nr_pages, true, false);
+	__bdi_start_writeback(bdi, nr_pages, true);
 }
 
 /**

@@ -147,13 +153,20 @@ void bdi_start_writeback(struct backing_
  * @bdi: the backing device to write from
  *
  * Description:
- *   This does WB_SYNC_NONE background writeback. The IO is only
- *   started when this function returns, we make no guarentees on
- *   completion. Caller need not hold sb s_umount semaphore.
+ *   This makes sure WB_SYNC_NONE background writeback happens. When
+ *   this function returns, it is only guaranteed that for given BDI
+ *   some IO is happening if we are over background dirty threshold.
+ *   Caller need not hold sb s_umount semaphore.
  */
 void bdi_start_background_writeback(struct backing_dev_info *bdi)
 {
-	__bdi_start_writeback(bdi, LONG_MAX, true, true);
+	/*
+	 * We just wake up the flusher thread. It will perform background
+	 * writeback as soon as there is no other work to do.
+	 */
+	spin_lock_bh(&bdi->wb_lock);
+	_bdi_wakeup_flusher(bdi);
+	spin_unlock_bh(&bdi->wb_lock);
 }
 
 /*

@@ -724,6 +737,23 @@ get_next_work_item(struct backing_dev_in
 	return work;
 }
 
+static long wb_check_background_flush(struct bdi_writeback *wb)
+{
+	if (over_bground_thresh()) {
+
+		struct wb_writeback_work work = {
+			.nr_pages	= LONG_MAX,
+			.sync_mode	= WB_SYNC_NONE,
+			.for_background	= 1,
+			.range_cyclic	= 1,
+		};
+
+		return wb_writeback(wb, &work);
+	}
+
+	return 0;
+}
+
 static long wb_check_old_data_flush(struct bdi_writeback *wb)
 {
 	unsigned long expired;

@@ -795,6 +825,7 @@ long wb_do_writeback(struct bdi_writebac
 	 * Check for periodic writeback, kupdated() style
 	 */
 	wrote += wb_check_old_data_flush(wb);
+	wrote += wb_check_background_flush(wb);
 	clear_bit(BDI_writeback_running, &wb->bdi->state);
 
 	return wrote;

@@ -881,7 +912,7 @@ void wakeup_flusher_threads(long nr_page
 	list_for_each_entry_rcu(bdi, &bdi_list, bdi_list) {
 		if (!bdi_has_dirty_io(bdi))
 			continue;
-		__bdi_start_writeback(bdi, nr_pages, false, false);
+		__bdi_start_writeback(bdi, nr_pages, false);
 	}
 	rcu_read_unlock();
 }

@@ -664,6 +664,15 @@ static long wb_writeback(struct bdi_writ
 			break;
 
 		/*
+		 * Background writeout and kupdate-style writeback are
+		 * easily livelockable. Stop them if there is other work
+		 * to do so that e.g. sync can proceed.
+		 */
+		if ((work->for_background || work->for_kupdate) &&
+		    !list_empty(&wb->bdi->work_list))
+			break;
+
+		/*
 		 * For background writeout, stop when we are below the
 		 * background dirty threshold
 		 */

@@ -81,6 +81,7 @@ DEFINE_EVENT(writeback_class, name, \
 	TP_ARGS(bdi))
 
 DEFINE_WRITEBACK_EVENT(writeback_nowork);
+DEFINE_WRITEBACK_EVENT(writeback_wake_background);
 DEFINE_WRITEBACK_EVENT(writeback_wake_thread);
 DEFINE_WRITEBACK_EVENT(writeback_wake_forker_thread);
 DEFINE_WRITEBACK_EVENT(writeback_bdi_register);

@@ -164,6 +164,7 @@ void bdi_start_background_writeback(stru
 	 * We just wake up the flusher thread. It will perform background
 	 * writeback as soon as there is no other work to do.
 	 */
+	trace_writeback_wake_background(bdi);
 	spin_lock_bh(&bdi->wb_lock);
 	bdi_wakeup_flusher(bdi);
 	spin_unlock_bh(&bdi->wb_lock);

@@ -79,13 +79,9 @@ static inline struct backing_dev_info *i
 	return sb->s_bdi;
 }
 
-static void bdi_queue_work(struct backing_dev_info *bdi,
-		struct wb_writeback_work *work)
+/* Wakeup flusher thread or forker thread to fork it. Requires bdi->wb_lock. */
+static void bdi_wakeup_flusher(struct backing_dev_info *bdi)
 {
-	trace_writeback_queue(bdi, work);
-
-	spin_lock_bh(&bdi->wb_lock);
-	list_add_tail(&work->list, &bdi->work_list);
 	if (bdi->wb.task) {
 		wake_up_process(bdi->wb.task);
 	} else {

@@ -93,15 +89,26 @@ static void bdi_queue_work(struct backin
 		 * The bdi thread isn't there, wake up the forker thread which
 		 * will create and run it.
 		 */
-		trace_writeback_nothread(bdi, work);
 		wake_up_process(default_backing_dev_info.wb.task);
 	}
+}
+
+static void bdi_queue_work(struct backing_dev_info *bdi,
+			   struct wb_writeback_work *work)
+{
+	trace_writeback_queue(bdi, work);
+
+	spin_lock_bh(&bdi->wb_lock);
+	list_add_tail(&work->list, &bdi->work_list);
+	if (!bdi->wb.task)
+		trace_writeback_nothread(bdi, work);
+	bdi_wakeup_flusher(bdi);
 	spin_unlock_bh(&bdi->wb_lock);
 }
 
 static void
 __bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages,
-		bool range_cyclic, bool for_background)
+		      bool range_cyclic)
 {
 	struct wb_writeback_work *work;
 

@@ -121,7 +128,6 @@ __bdi_start_writeback(struct backing_dev
 	work->sync_mode	= WB_SYNC_NONE;
 	work->nr_pages	= nr_pages;
 	work->range_cyclic = range_cyclic;
-	work->for_background = for_background;
 
 	bdi_queue_work(bdi, work);
 }

@@ -139,7 +145,7 @@ __bdi_start_writeback(struct backing_dev
  */
 void bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages)
 {
-	__bdi_start_writeback(bdi, nr_pages, true, false);
+	__bdi_start_writeback(bdi, nr_pages, true);
 }
 
 /**

@@ -147,13 +153,20 @@ void bdi_start_writeback(struct backing_
  * @bdi: the backing device to write from
  *
  * Description:
- *   This does WB_SYNC_NONE background writeback. The IO is only
- *   started when this function returns, we make no guarentees on
- *   completion. Caller need not hold sb s_umount semaphore.
+ *   This makes sure WB_SYNC_NONE background writeback happens. When
+ *   this function returns, it is only guaranteed that for given BDI
+ *   some IO is happening if we are over background dirty threshold.
+ *   Caller need not hold sb s_umount semaphore.
  */
 void bdi_start_background_writeback(struct backing_dev_info *bdi)
 {
-	__bdi_start_writeback(bdi, LONG_MAX, true, true);
+	/*
+	 * We just wake up the flusher thread. It will perform background
+	 * writeback as soon as there is no other work to do.
+	 */
+	spin_lock_bh(&bdi->wb_lock);
+	bdi_wakeup_flusher(bdi);
+	spin_unlock_bh(&bdi->wb_lock);
 }
 
 /*

@@ -724,6 +737,23 @@ get_next_work_item(struct backing_dev_in
 	return work;
 }
 
+static long wb_check_background_flush(struct bdi_writeback *wb)
+{
+	if (over_bground_thresh()) {
+
+		struct wb_writeback_work work = {
+			.nr_pages	= LONG_MAX,
+			.sync_mode	= WB_SYNC_NONE,
+			.for_background	= 1,
+			.range_cyclic	= 1,
+		};
+
+		return wb_writeback(wb, &work);
+	}
+
+	return 0;
+}
+
 static long wb_check_old_data_flush(struct bdi_writeback *wb)
 {
 	unsigned long expired;

@@ -795,6 +825,7 @@ long wb_do_writeback(struct bdi_writebac
 	 * Check for periodic writeback, kupdated() style
 	 */
 	wrote += wb_check_old_data_flush(wb);
+	wrote += wb_check_background_flush(wb);
 	clear_bit(BDI_writeback_running, &wb->bdi->state);
 
 	return wrote;

@@ -881,7 +912,7 @@ void wakeup_flusher_threads(long nr_page
 	list_for_each_entry_rcu(bdi, &bdi_list, bdi_list) {
 		if (!bdi_has_dirty_io(bdi))
 			continue;
-		__bdi_start_writeback(bdi, nr_pages, false, false);
+		__bdi_start_writeback(bdi, nr_pages, false);
 	}
 	rcu_read_unlock();
 }

@@ -30,11 +30,20 @@
 #include "internal.h"
 
 /*
+ * When flushing an inode page (for page reclaim), try to piggy back more
+ * nearby pages for IO efficiency. These pages will have good opportunity
+ * to be in the same LRU list.
+ */
+#define WRITE_AROUND_PAGES	(1UL << (20 - PAGE_CACHE_SHIFT))
+
+/*
  * Passed into wb_writeback(), essentially a subset of writeback_control
  */
 struct wb_writeback_work {
 	long nr_pages;
 	struct super_block *sb;
+	struct inode *inode;
+	pgoff_t offset;
 	enum writeback_sync_modes sync_mode;
 	unsigned int for_kupdate:1;
 	unsigned int range_cyclic:1;

@@ -59,6 +68,27 @@ struct wb_writeback_work {
  */
 int nr_pdflush_threads;
 
+static mempool_t *wb_work_mempool;
+
+static void *wb_work_alloc(gfp_t gfp_mask, void *pool_data)
+{
+	/*
+	 * bdi_start_inode_writeback() may be called on page reclaim
+	 */
+	if (current->flags & PF_MEMALLOC)
+		return NULL;
+
+	return kmalloc(sizeof(struct wb_writeback_work), gfp_mask);
+}
+
+static __init int wb_work_init(void)
+{
+	wb_work_mempool = mempool_create(1024,
+					 wb_work_alloc, mempool_kfree, NULL);
+	return wb_work_mempool ? 0 : -ENOMEM;
+}
+fs_initcall(wb_work_init);
+
 /**
  * writeback_in_progress - determine whether there is writeback in progress
  * @bdi: the device's backing_dev_info structure.

@@ -101,7 +131,7 @@ __bdi_start_writeback(struct backing_dev
 	 * This is WB_SYNC_NONE writeback, so if allocation fails just
 	 * wakeup the thread for old dirty data writeback
 	 */
-	work = kzalloc(sizeof(*work), GFP_ATOMIC);
+	work = mempool_alloc(wb_work_mempool, GFP_NOWAIT);
 	if (!work) {
 		if (bdi->wb.task) {
 			trace_writeback_nowork(bdi);

@@ -110,6 +140,7 @@ __bdi_start_writeback(struct backing_dev
 		return;
 	}
 
+	memset(work, 0, sizeof(*work));
 	work->sync_mode	= WB_SYNC_NONE;
 	work->nr_pages	= nr_pages;
 	work->range_cyclic = range_cyclic;

@@ -148,6 +179,55 @@ void bdi_start_background_writeback(stru
 	__bdi_start_writeback(bdi, LONG_MAX, true, true);
 }
 
+int bdi_start_inode_writeback(struct backing_dev_info *bdi,
+			      struct inode *inode, pgoff_t offset)
+{
+	struct wb_writeback_work *work;
+
+	spin_lock_bh(&bdi->wb_lock);
+	list_for_each_entry_reverse(work, &bdi->work_list, list) {
+		unsigned long end;
+		if (work->inode != inode)
+			continue;
+		end = work->offset + work->nr_pages;
+		if (work->offset - offset < WRITE_AROUND_PAGES) {
+			work->nr_pages += work->offset - offset;
+			work->offset = offset;
+			inode = NULL;
+			break;
+		} else if (offset - end < WRITE_AROUND_PAGES) {
+			work->nr_pages += offset - end;
+			inode = NULL;
+			break;
+		} else if (offset > work->offset &&
+			   offset < end) {
+			inode = NULL;
+			break;
+		}
+	}
+	spin_unlock_bh(&bdi->wb_lock);
+
+	if (!inode)
+		return 0;
+
+	if (!igrab(inode))
+		return -ENOENT;
+
+	work = mempool_alloc(wb_work_mempool, GFP_NOWAIT);
+	if (!work)
+		return -ENOMEM;
+
+	memset(work, 0, sizeof(*work));
+	work->sync_mode		= WB_SYNC_NONE;
+	work->inode		= inode;
+	work->offset		= offset;
+	work->nr_pages		= 1;
+
+	bdi_queue_work(inode->i_sb->s_bdi, work);
+
+	return 0;
+}
+
 /*
  * Redirty an inode: set its when-it-was dirtied timestamp and move it to the
  * furthest end of its superblock's dirty-inode list.

@@ -724,6 +804,20 @@ get_next_work_item(struct backing_dev_in
 	return work;
 }
 
+static long wb_flush_inode(struct bdi_writeback *wb,
+			   struct wb_writeback_work *work)
+{
+	pgoff_t start = round_down(work->offset, WRITE_AROUND_PAGES);
+	pgoff_t end = round_up(work->offset + work->nr_pages,
+			       WRITE_AROUND_PAGES);
+	int wrote;
+
+	wrote = __filemap_fdatawrite_range(work->inode->i_mapping,
+					   start, end, WB_SYNC_NONE);
+	iput(work->inode);
+	return wrote;
+}
+
 static long wb_check_background_flush(struct bdi_writeback *wb)
 {
 	if (over_bground_thresh()) {

@@ -796,7 +890,10 @@ long wb_do_writeback(struct bdi_writebac
 
 		trace_writeback_exec(bdi, work);
 
-		wrote += wb_writeback(wb, work);
+		if (work->inode)
+			wrote += wb_flush_inode(wb, work);
+		else
+			wrote += wb_writeback(wb, work);
 
 		/*
 		 * Notify the caller of completion if this is a synchronous

@@ -805,7 +902,7 @@ long wb_do_writeback(struct bdi_writebac
 		if (work->done)
 			complete(work->done);
 		else
-			kfree(work);
+			mempool_free(work, wb_work_mempool);
 	}
 
 	/*

@@ -106,6 +106,8 @@ void bdi_unregister(struct backing_dev_i
 int bdi_setup_and_register(struct backing_dev_info *, char *, unsigned int);
 void bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages);
 void bdi_start_background_writeback(struct backing_dev_info *bdi);
+int bdi_start_inode_writeback(struct backing_dev_info *bdi,
+			      struct inode *inode, pgoff_t offset);
 int bdi_writeback_thread(void *data);
 int bdi_has_dirty_io(struct backing_dev_info *bdi);
 void bdi_arm_supers_timer(void);

@@ -752,6 +754,16 @@ static unsigned long shrink_page_list(st
 				}
 			}
 
+			if (page_is_file_cache(page) && mapping &&
+			    sync_writeback == PAGEOUT_IO_ASYNC) {
+				if (!bdi_start_inode_writeback(
+					mapping->backing_dev_info,
+					mapping->host, page_index(page))) {
+					SetPageReclaim(page);
+					goto keep_locked;
+				}
+			}
+
 			if (references == PAGEREF_RECLAIM_CLEAN)
 				goto keep_locked;
 			if (!may_enter_fs)

171 lines not on this page · raw message

@@ -107,6 +107,8 @@ void bdi_unregister(struct backing_dev_i
 int bdi_setup_and_register(struct backing_dev_info *, char *, unsigned int);
 void bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages);
 void bdi_start_background_writeback(struct backing_dev_info *bdi);
+int bdi_start_inode_writeback(struct backing_dev_info *bdi,
+			      struct inode *inode, pgoff_t offset);
 int bdi_writeback_thread(void *data);
 int bdi_has_dirty_io(struct backing_dev_info *bdi);
 void bdi_arm_supers_timer(void);

@@ -756,6 +756,19 @@ static unsigned long shrink_page_list(st
 				}
 			}
 
+			if (page_is_file_cache(page) && mapping &&
+			    sync_writeback == PAGEOUT_IO_ASYNC) {
+				if (!bdi_start_inode_writeback(
+					mapping->backing_dev_info,
+					mapping->host, page_index(page))) {
+					SetPageReclaim(page);
+					goto keep_locked;
+				} else if (!current_is_kswapd() &&
+					   printk_ratelimit()) {
+					printk(KERN_INFO "cannot pageout\n");
+				}
+			}
+
 			if (references == PAGEREF_RECLAIM_CLEAN)
 				goto keep_locked;
 			if (!may_enter_fs)

@@ -1232,6 +1232,7 @@ int set_page_dirty(struct page *page)
 {
 	struct address_space *mapping = page_mapping(page);
 
+	ClearPageReclaim(page);
 	if (likely(mapping)) {
 		int (*spd)(struct page *) = mapping->a_ops->set_page_dirty;
 #ifdef CONFIG_BLOCK

@@ -1289,7 +1290,6 @@ int clear_page_dirty_for_io(struct page 
 
 	BUG_ON(!PageLocked(page));
 
-	ClearPageReclaim(page);
 	if (mapping && mapping_cap_account_dirty(mapping)) {
 		/*
 		 * Yes, Virginia, this is indeed insane.