Revert "writeback: do not sync data dirtied after sync start"

This reverts commit c4a391b53a72d2df4ee97f96f78c1d5971b47489. Dave
Chinner <david@fromorbit.com> has reported the commit may cause some
inodes to be left out from sync(2). This is because we can call
redirty_tail() for some inode (which sets i_dirtied_when to current time)
after sync(2) has started or similarly requeue_inode() can set
i_dirtied_when to current time if writeback had to skip some pages. The
real problem is in the functions clobbering i_dirtied_when but fixing
that isn't trivial so revert is a safer choice for now.

CC: stable@vger.kernel.org # >= 3.13
Signed-off-by: Jan Kara <jack@suse.cz>

3 files changed, 18 insertions, 32 deletions

diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index e0259a163f98..d754e3cf99a8 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -40,18 +40,13 @@
 struct wb_writeback_work {
 	long nr_pages;
 	struct super_block *sb;
-	/*
-	 * Write only inodes dirtied before this time. Don't forget to set
-	 * older_than_this_is_set when you set this.
-	 */
-	unsigned long older_than_this;
+	unsigned long *older_than_this;
 	enum writeback_sync_modes sync_mode;
 	unsigned int tagged_writepages:1;
 	unsigned int for_kupdate:1;
 	unsigned int range_cyclic:1;
 	unsigned int for_background:1;
 	unsigned int for_sync:1;	/* sync(2) WB_SYNC_ALL writeback */
-	unsigned int older_than_this_is_set:1;
 	enum wb_reason reason;		/* why was writeback initiated? */
 
 	struct list_head list;		/* pending work list */
@@ -252,10 +247,10 @@ static int move_expired_inodes(struct list_head *delaying_queue,
 	int do_sb_sort = 0;
 	int moved = 0;
 
-	WARN_ON_ONCE(!work->older_than_this_is_set);
 	while (!list_empty(delaying_queue)) {
 		inode = wb_inode(delaying_queue->prev);
-		if (inode_dirtied_after(inode, work->older_than_this))
+		if (work->older_than_this &&
+		    inode_dirtied_after(inode, *work->older_than_this))
 			break;
 		list_move(&inode->i_wb_list, &tmp);
 		moved++;
@@ -742,8 +737,6 @@ static long writeback_inodes_wb(struct bdi_writeback *wb, long nr_pages,
 		.sync_mode	= WB_SYNC_NONE,
 		.range_cyclic	= 1,
 		.reason		= reason,
-		.older_than_this = jiffies,
-		.older_than_this_is_set = 1,
 	};
 
 	spin_lock(&wb->list_lock);
@@ -802,13 +795,12 @@ static long wb_writeback(struct bdi_writeback *wb,
 {
 	unsigned long wb_start = jiffies;
 	long nr_pages = work->nr_pages;
+	unsigned long oldest_jif;
 	struct inode *inode;
 	long progress;
 
-	if (!work->older_than_this_is_set) {
-		work->older_than_this = jiffies;
-		work->older_than_this_is_set = 1;
-	}
+	oldest_jif = jiffies;
+	work->older_than_this = &oldest_jif;
 
 	spin_lock(&wb->list_lock);
 	for (;;) {
@@ -842,10 +834,10 @@ static long wb_writeback(struct bdi_writeback *wb,
 		 * safe.
 		 */
 		if (work->for_kupdate) {
-			work->older_than_this = jiffies -
+			oldest_jif = jiffies -
 				msecs_to_jiffies(dirty_expire_interval * 10);
 		} else if (work->for_background)
-			work->older_than_this = jiffies;
+			oldest_jif = jiffies;
 
 		trace_writeback_start(wb->bdi, work);
 		if (list_empty(&wb->b_io))
@@ -1357,21 +1349,18 @@ EXPORT_SYMBOL(try_to_writeback_inodes_sb);
 
 /**
  * sync_inodes_sb	-	sync sb inode pages
- * @sb:			the superblock
- * @older_than_this:	timestamp
+ * @sb: the superblock
  *
  * This function writes and waits on any dirty inode belonging to this
- * superblock that has been dirtied before given timestamp.
+ * super_block.
  */
-void sync_inodes_sb(struct super_block *sb, unsigned long older_than_this)
+void sync_inodes_sb(struct super_block *sb)
 {
 	DECLARE_COMPLETION_ONSTACK(done);
 	struct wb_writeback_work work = {
 		.sb		= sb,
 		.sync_mode	= WB_SYNC_ALL,
 		.nr_pages	= LONG_MAX,
-		.older_than_this = older_than_this,
-		.older_than_this_is_set = 1,
 		.range_cyclic	= 0,
 		.done		= &done,
 		.reason		= WB_REASON_SYNC,
diff --git a/fs/sync.c b/fs/sync.c
index e8ba024a055b..b28d1dd10e8b 100644
--- a/fs/sync.c
+++ b/fs/sync.c
@@ -27,11 +27,10 @@
  * wait == 1 case since in that case write_inode() functions do
  * sync_dirty_buffer() and thus effectively write one block at a time.
  */
-static int __sync_filesystem(struct super_block *sb, int wait,
-			     unsigned long start)
+static int __sync_filesystem(struct super_block *sb, int wait)
 {
 	if (wait)
-		sync_inodes_sb(sb, start);
+		sync_inodes_sb(sb);
 	else
 		writeback_inodes_sb(sb, WB_REASON_SYNC);
 
@@ -48,7 +47,6 @@ static int __sync_filesystem(struct super_block *sb, int wait,
 int sync_filesystem(struct super_block *sb)
 {
 	int ret;
-	unsigned long start = jiffies;
 
 	/*
 	 * We need to be protected against the filesystem going from
@@ -62,17 +60,17 @@ int sync_filesystem(struct super_block *sb)
 	if (sb->s_flags & MS_RDONLY)
 		return 0;
 
-	ret = __sync_filesystem(sb, 0, start);
+	ret = __sync_filesystem(sb, 0);
 	if (ret < 0)
 		return ret;
-	return __sync_filesystem(sb, 1, start);
+	return __sync_filesystem(sb, 1);
 }
 EXPORT_SYMBOL_GPL(sync_filesystem);
 
 static void sync_inodes_one_sb(struct super_block *sb, void *arg)
 {
 	if (!(sb->s_flags & MS_RDONLY))
-		sync_inodes_sb(sb, *((unsigned long *)arg));
+		sync_inodes_sb(sb);
 }
 
 static void sync_fs_one_sb(struct super_block *sb, void *arg)
@@ -104,10 +102,9 @@ static void fdatawait_one_bdev(struct block_device *bdev, void *arg)
 SYSCALL_DEFINE0(sync)
 {
 	int nowait = 0, wait = 1;
-	unsigned long start = jiffies;
 
 	wakeup_flusher_threads(0, WB_REASON_SYNC);
-	iterate_supers(sync_inodes_one_sb, &start);
+	iterate_supers(sync_inodes_one_sb, NULL);
 	iterate_supers(sync_fs_one_sb, &nowait);
 	iterate_supers(sync_fs_one_sb, &wait);
 	iterate_bdevs(fdatawrite_one_bdev, NULL);
diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c
index f317488263dd..d971f4932b5d 100644
--- a/fs/xfs/xfs_super.c
+++ b/fs/xfs/xfs_super.c
@@ -913,7 +913,7 @@ xfs_flush_inodes(
 	struct super_block	*sb = mp->m_super;
 
 	if (down_read_trylock(&sb->s_umount)) {
-		sync_inodes_sb(sb, jiffies);
+		sync_inodes_sb(sb);
 		up_read(&sb->s_umount);
 	}
 }