Merge tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4

Pull ext4 updates for 3.4 from Ted Ts'o:
 "Ext4 commits for 3.3 merge window; mostly cleanups and bug fixes

  The changes to export dirty_writeback_interval are from Artem's s_dirt
  cleanup patch series.  The same is true of the change to remove the
  s_dirt helper functions which never got used by anyone in-tree.  I've
  run these changes by Al Viro, and am carrying them so that Artem can
  more easily fix up the rest of the file systems during the next merge
  window.  (Originally we had hopped to remove the use of s_dirt from
  ext4 during this merge window, but his patches had some bugs, so I
  ultimately ended dropping them from the ext4 tree.)"

* tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (66 commits)
  vfs: remove unused superblock helpers
  mm: export dirty_writeback_interval
  ext4: remove useless s_dirt assignment
  ext4: write superblock only once on unmount
  ext4: do not mark superblock as dirty unnecessarily
  ext4: correct ext4_punch_hole return codes
  ext4: remove restrictive checks for EOFBLOCKS_FL
  ext4: always set then trimmed blocks count into len
  ext4: fix trimmed block count accunting
  ext4: fix start and len arguments handling in ext4_trim_fs()
  ext4: update s_free_{inodes,blocks}_count during online resize
  ext4: change some printk() calls to use ext4_msg() instead
  ext4: avoid output message interleaving in ext4_error_<foo>()
  ext4: remove trailing newlines from ext4_msg() and ext4_error() messages
  ext4: add no_printk argument validation, fix fallout
  ext4: remove redundant "EXT4-fs: " from uses of ext4_msg
  ext4: give more helpful error message in ext4_ext_rm_leaf()
  ext4: remove unused code from ext4_ext_map_blocks()
  ext4: rewrite punch hole to use ext4_ext_remove_space()
  jbd2: cleanup journal tail after transaction commit
  ...

6 files changed, 343 insertions, 270 deletions

diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c
index d49d202903f..c78841ee81c 100644
--- a/fs/jbd2/checkpoint.c
+++ b/fs/jbd2/checkpoint.c
@@ -88,14 +88,13 @@ static inline void __buffer_relink_io(struct journal_head *jh)
  * whole transaction.
  *
  * Requires j_list_lock
- * Called under jbd_lock_bh_state(jh2bh(jh)), and drops it
  */
 static int __try_to_free_cp_buf(struct journal_head *jh)
 {
 	int ret = 0;
 	struct buffer_head *bh = jh2bh(jh);
 
-	if (jh->b_jlist == BJ_None && !buffer_locked(bh) &&
+	if (jh->b_transaction == NULL && !buffer_locked(bh) &&
 	    !buffer_dirty(bh) && !buffer_write_io_error(bh)) {
 		/*
 		 * Get our reference so that bh cannot be freed before
@@ -104,11 +103,8 @@ static int __try_to_free_cp_buf(struct journal_head *jh)
 		get_bh(bh);
 		JBUFFER_TRACE(jh, "remove from checkpoint list");
 		ret = __jbd2_journal_remove_checkpoint(jh) + 1;
-		jbd_unlock_bh_state(bh);
 		BUFFER_TRACE(bh, "release");
 		__brelse(bh);
-	} else {
-		jbd_unlock_bh_state(bh);
 	}
 	return ret;
 }
@@ -180,21 +176,6 @@ void __jbd2_log_wait_for_space(journal_t *journal)
 }
 
 /*
- * We were unable to perform jbd_trylock_bh_state() inside j_list_lock.
- * The caller must restart a list walk.  Wait for someone else to run
- * jbd_unlock_bh_state().
- */
-static void jbd_sync_bh(journal_t *journal, struct buffer_head *bh)
-	__releases(journal->j_list_lock)
-{
-	get_bh(bh);
-	spin_unlock(&journal->j_list_lock);
-	jbd_lock_bh_state(bh);
-	jbd_unlock_bh_state(bh);
-	put_bh(bh);
-}
-
-/*
  * Clean up transaction's list of buffers submitted for io.
  * We wait for any pending IO to complete and remove any clean
  * buffers. Note that we take the buffers in the opposite ordering
@@ -222,15 +203,9 @@ restart:
 	while (!released && transaction->t_checkpoint_io_list) {
 		jh = transaction->t_checkpoint_io_list;
 		bh = jh2bh(jh);
-		if (!jbd_trylock_bh_state(bh)) {
-			jbd_sync_bh(journal, bh);
-			spin_lock(&journal->j_list_lock);
-			goto restart;
-		}
 		get_bh(bh);
 		if (buffer_locked(bh)) {
 			spin_unlock(&journal->j_list_lock);
-			jbd_unlock_bh_state(bh);
 			wait_on_buffer(bh);
 			/* the journal_head may have gone by now */
 			BUFFER_TRACE(bh, "brelse");
@@ -246,7 +221,6 @@ restart:
 		 * it has been written out and so we can drop it from the list
 		 */
 		released = __jbd2_journal_remove_checkpoint(jh);
-		jbd_unlock_bh_state(bh);
 		__brelse(bh);
 	}
 
@@ -266,7 +240,6 @@ __flush_batch(journal_t *journal, int *batch_count)
 
 	for (i = 0; i < *batch_count; i++) {
 		struct buffer_head *bh = journal->j_chkpt_bhs[i];
-		clear_buffer_jwrite(bh);
 		BUFFER_TRACE(bh, "brelse");
 		__brelse(bh);
 	}
@@ -281,7 +254,6 @@ __flush_batch(journal_t *journal, int *batch_count)
  * be written out.
  *
  * Called with j_list_lock held and drops it if 1 is returned
- * Called under jbd_lock_bh_state(jh2bh(jh)), and drops it
  */
 static int __process_buffer(journal_t *journal, struct journal_head *jh,
 			    int *batch_count, transaction_t *transaction)
@@ -292,7 +264,6 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh,
 	if (buffer_locked(bh)) {
 		get_bh(bh);
 		spin_unlock(&journal->j_list_lock);
-		jbd_unlock_bh_state(bh);
 		wait_on_buffer(bh);
 		/* the journal_head may have gone by now */
 		BUFFER_TRACE(bh, "brelse");
@@ -304,7 +275,6 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh,
 
 		transaction->t_chp_stats.cs_forced_to_close++;
 		spin_unlock(&journal->j_list_lock);
-		jbd_unlock_bh_state(bh);
 		if (unlikely(journal->j_flags & JBD2_UNMOUNT))
 			/*
 			 * The journal thread is dead; so starting and
@@ -323,11 +293,9 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh,
 		if (unlikely(buffer_write_io_error(bh)))
 			ret = -EIO;
 		get_bh(bh);
-		J_ASSERT_JH(jh, !buffer_jbddirty(bh));
 		BUFFER_TRACE(bh, "remove from checkpoint");
 		__jbd2_journal_remove_checkpoint(jh);
 		spin_unlock(&journal->j_list_lock);
-		jbd_unlock_bh_state(bh);
 		__brelse(bh);
 	} else {
 		/*
@@ -340,10 +308,8 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh,
 		BUFFER_TRACE(bh, "queue");
 		get_bh(bh);
 		J_ASSERT_BH(bh, !buffer_jwrite(bh));
-		set_buffer_jwrite(bh);
 		journal->j_chkpt_bhs[*batch_count] = bh;
 		__buffer_relink_io(jh);
-		jbd_unlock_bh_state(bh);
 		transaction->t_chp_stats.cs_written++;
 		(*batch_count)++;
 		if (*batch_count == JBD2_NR_BATCH) {
@@ -407,15 +373,7 @@ restart:
 		int retry = 0, err;
 
 		while (!retry && transaction->t_checkpoint_list) {
-			struct buffer_head *bh;
-
 			jh = transaction->t_checkpoint_list;
-			bh = jh2bh(jh);
-			if (!jbd_trylock_bh_state(bh)) {
-				jbd_sync_bh(journal, bh);
-				retry = 1;
-				break;
-			}
 			retry = __process_buffer(journal, jh, &batch_count,
 						 transaction);
 			if (retry < 0 && !result)
@@ -478,79 +436,28 @@ out:
 
 int jbd2_cleanup_journal_tail(journal_t *journal)
 {
-	transaction_t * transaction;
 	tid_t		first_tid;
-	unsigned long	blocknr, freed;
+	unsigned long	blocknr;
 
 	if (is_journal_aborted(journal))
 		return 1;
 
-	/* OK, work out the oldest transaction remaining in the log, and
-	 * the log block it starts at.
-	 *
-	 * If the log is now empty, we need to work out which is the
-	 * next transaction ID we will write, and where it will
-	 * start. */
-
-	write_lock(&journal->j_state_lock);
-	spin_lock(&journal->j_list_lock);
-	transaction = journal->j_checkpoint_transactions;
-	if (transaction) {
-		first_tid = transaction->t_tid;
-		blocknr = transaction->t_log_start;
-	} else if ((transaction = journal->j_committing_transaction) != NULL) {
-		first_tid = transaction->t_tid;
-		blocknr = transaction->t_log_start;
-	} else if ((transaction = journal->j_running_transaction) != NULL) {
-		first_tid = transaction->t_tid;
-		blocknr = journal->j_head;
-	} else {
-		first_tid = journal->j_transaction_sequence;
-		blocknr = journal->j_head;
-	}
-	spin_unlock(&journal->j_list_lock);
-	J_ASSERT(blocknr != 0);
-
-	/* If the oldest pinned transaction is at the tail of the log
-           already then there's not much we can do right now. */
-	if (journal->j_tail_sequence == first_tid) {
-		write_unlock(&journal->j_state_lock);
+	if (!jbd2_journal_get_log_tail(journal, &first_tid, &blocknr))
 		return 1;
-	}
-
-	/* OK, update the superblock to recover the freed space.
-	 * Physical blocks come first: have we wrapped beyond the end of
-	 * the log?  */
-	freed = blocknr - journal->j_tail;
-	if (blocknr < journal->j_tail)
-		freed = freed + journal->j_last - journal->j_first;
-
-	trace_jbd2_cleanup_journal_tail(journal, first_tid, blocknr, freed);
-	jbd_debug(1,
-		  "Cleaning journal tail from %d to %d (offset %lu), "
-		  "freeing %lu\n",
-		  journal->j_tail_sequence, first_tid, blocknr, freed);
-
-	journal->j_free += freed;
-	journal->j_tail_sequence = first_tid;
-	journal->j_tail = blocknr;
-	write_unlock(&journal->j_state_lock);
+	J_ASSERT(blocknr != 0);
 
 	/*
-	 * If there is an external journal, we need to make sure that
-	 * any data blocks that were recently written out --- perhaps
-	 * by jbd2_log_do_checkpoint() --- are flushed out before we
-	 * drop the transactions from the external journal.  It's
-	 * unlikely this will be necessary, especially with a
-	 * appropriately sized journal, but we need this to guarantee
-	 * correctness.  Fortunately jbd2_cleanup_journal_tail()
-	 * doesn't get called all that often.
+	 * We need to make sure that any blocks that were recently written out
+	 * --- perhaps by jbd2_log_do_checkpoint() --- are flushed out before
+	 * we drop the transactions from the journal. It's unlikely this will
+	 * be necessary, especially with an appropriately sized journal, but we
+	 * need this to guarantee correctness.  Fortunately
+	 * jbd2_cleanup_journal_tail() doesn't get called all that often.
 	 */
-	if ((journal->j_fs_dev != journal->j_dev) &&
-	    (journal->j_flags & JBD2_BARRIER))
+	if (journal->j_flags & JBD2_BARRIER)
 		blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL, NULL);
-	if (!(journal->j_flags & JBD2_ABORT))
-		jbd2_journal_update_superblock(journal, 1);
+
+	__jbd2_update_log_tail(journal, first_tid, blocknr);
 	return 0;
 }
 
@@ -582,15 +489,12 @@ static int journal_clean_one_cp_list(struct journal_head *jh, int *released)
 	do {
 		jh = next_jh;
 		next_jh = jh->b_cpnext;
-		/* Use trylock because of the ranking */
-		if (jbd_trylock_bh_state(jh2bh(jh))) {
-			ret = __try_to_free_cp_buf(jh);
-			if (ret) {
-				freed++;
-				if (ret == 2) {
-					*released = 1;
-					return freed;
-				}
+		ret = __try_to_free_cp_buf(jh);
+		if (ret) {
+			freed++;
+			if (ret == 2) {
+				*released = 1;
+				return freed;
 			}
 		}
 		/*
@@ -673,9 +577,7 @@ out:
  * The function can free jh and bh.
  *
  * This function is called with j_list_lock held.
- * This function is called with jbd_lock_bh_state(jh2bh(jh))
  */
-
 int __jbd2_journal_remove_checkpoint(struct journal_head *jh)
 {
 	struct transaction_chp_stats_s *stats;
@@ -722,7 +624,7 @@ int __jbd2_journal_remove_checkpoint(struct journal_head *jh)
 				    transaction->t_tid, stats);
 
 	__jbd2_journal_drop_transaction(journal, transaction);
-	kfree(transaction);
+	jbd2_journal_free_transaction(transaction);
 
 	/* Just in case anybody was waiting for more transactions to be
            checkpointed... */
@@ -797,5 +699,7 @@ void __jbd2_journal_drop_transaction(journal_t *journal, transaction_t *transact
 	J_ASSERT(journal->j_committing_transaction != transaction);
 	J_ASSERT(journal->j_running_transaction != transaction);
 
+	trace_jbd2_drop_transaction(journal, transaction);
+
 	jbd_debug(1, "Dropping transaction %d, all done\n", transaction->t_tid);
 }
diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c
index c067a8cae63..17f557f01cf 100644
--- a/fs/jbd2/commit.c
+++ b/fs/jbd2/commit.c
@@ -331,6 +331,10 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	struct buffer_head *cbh = NULL; /* For transactional checksums */
 	__u32 crc32_sum = ~0;
 	struct blk_plug plug;
+	/* Tail of the journal */
+	unsigned long first_block;
+	tid_t first_tid;
+	int update_tail;
 
 	/*
 	 * First job: lock down the current transaction and wait for
@@ -340,7 +344,18 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	/* Do we need to erase the effects of a prior jbd2_journal_flush? */
 	if (journal->j_flags & JBD2_FLUSHED) {
 		jbd_debug(3, "super block updated\n");
-		jbd2_journal_update_superblock(journal, 1);
+		mutex_lock(&journal->j_checkpoint_mutex);
+		/*
+		 * We hold j_checkpoint_mutex so tail cannot change under us.
+		 * We don't need any special data guarantees for writing sb
+		 * since journal is empty and it is ok for write to be
+		 * flushed only with transaction commit.
+		 */
+		jbd2_journal_update_sb_log_tail(journal,
+						journal->j_tail_sequence,
+						journal->j_tail,
+						WRITE_SYNC);
+		mutex_unlock(&journal->j_checkpoint_mutex);
 	} else {
 		jbd_debug(3, "superblock not updated\n");
 	}
@@ -677,10 +692,30 @@ start_journal_io:
 		err = 0;
 	}
 
+	/*
+	 * Get current oldest transaction in the log before we issue flush
+	 * to the filesystem device. After the flush we can be sure that
+	 * blocks of all older transactions are checkpointed to persistent
+	 * storage and we will be safe to update journal start in the
+	 * superblock with the numbers we get here.
+	 */
+	update_tail =
+		jbd2_journal_get_log_tail(journal, &first_tid, &first_block);
+
 	write_lock(&journal->j_state_lock);
+	if (update_tail) {
+		long freed = first_block - journal->j_tail;
+
+		if (first_block < journal->j_tail)
+			freed += journal->j_last - journal->j_first;
+		/* Update tail only if we free significant amount of space */
+		if (freed < journal->j_maxlen / 4)
+			update_tail = 0;
+	}
 	J_ASSERT(commit_transaction->t_state == T_COMMIT);
 	commit_transaction->t_state = T_COMMIT_DFLUSH;
 	write_unlock(&journal->j_state_lock);
+
 	/* 
 	 * If the journal is not located on the file system device,
 	 * then we must flush the file system device before we issue
@@ -831,6 +866,14 @@ wait_for_iobuf:
 	if (err)
 		jbd2_journal_abort(journal, err);
 
+	/*
+	 * Now disk caches for filesystem device are flushed so we are safe to
+	 * erase checkpointed transactions from the log by updating journal
+	 * superblock.
+	 */
+	if (update_tail)
+		jbd2_update_log_tail(journal, first_tid, first_block);
+
 	/* End of a transaction!  Finally, we can do checkpoint
            processing: any buffers committed as a result of this
            transaction can be removed from any checkpoint list it was on
@@ -1048,7 +1091,7 @@ restart_loop:
 	jbd_debug(1, "JBD2: commit %d complete, head %d\n",
 		  journal->j_commit_sequence, journal->j_tail_sequence);
 	if (to_free)
-		kfree(commit_transaction);
+		jbd2_journal_free_transaction(commit_transaction);
 
 	wake_up(&journal->j_wait_done_commit);
 }
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index 839377e3d62..98ed6dbfe38 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -71,7 +71,6 @@ EXPORT_SYMBOL(jbd2_journal_revoke);
 
 EXPORT_SYMBOL(jbd2_journal_init_dev);
 EXPORT_SYMBOL(jbd2_journal_init_inode);
-EXPORT_SYMBOL(jbd2_journal_update_format);
 EXPORT_SYMBOL(jbd2_journal_check_used_features);
 EXPORT_SYMBOL(jbd2_journal_check_available_features);
 EXPORT_SYMBOL(jbd2_journal_set_features);
@@ -96,7 +95,6 @@ EXPORT_SYMBOL(jbd2_journal_release_jbd_inode);
 EXPORT_SYMBOL(jbd2_journal_begin_ordered_truncate);
 EXPORT_SYMBOL(jbd2_inode_cache);
 
-static int journal_convert_superblock_v1(journal_t *, journal_superblock_t *);
 static void __journal_abort_soft (journal_t *journal, int errno);
 static int jbd2_journal_create_slab(size_t slab_size);
 
@@ -746,6 +744,98 @@ struct journal_head *jbd2_journal_get_descriptor_buffer(journal_t *journal)
 	return jbd2_journal_add_journal_head(bh);
 }
 
+/*
+ * Return tid of the oldest transaction in the journal and block in the journal
+ * where the transaction starts.
+ *
+ * If the journal is now empty, return which will be the next transaction ID
+ * we will write and where will that transaction start.
+ *
+ * The return value is 0 if journal tail cannot be pushed any further, 1 if
+ * it can.
+ */
+int jbd2_journal_get_log_tail(journal_t *journal, tid_t *tid,
+			      unsigned long *block)
+{
+	transaction_t *transaction;
+	int ret;
+
+	read_lock(&journal->j_state_lock);
+	spin_lock(&journal->j_list_lock);
+	transaction = journal->j_checkpoint_transactions;
+	if (transaction) {
+		*tid = transaction->t_tid;
+		*block = transaction->t_log_start;
+	} else if ((transaction = journal->j_committing_transaction) != NULL) {
+		*tid = transaction->t_tid;
+		*block = transaction->t_log_start;
+	} else if ((transaction = journal->j_running_transaction) != NULL) {
+		*tid = transaction->t_tid;
+		*block = journal->j_head;
+	} else {
+		*tid = journal->j_transaction_sequence;
+		*block = journal->j_head;
+	}
+	ret = tid_gt(*tid, journal->j_tail_sequence);
+	spin_unlock(&journal->j_list_lock);
+	read_unlock(&journal->j_state_lock);
+
+	return ret;
+}
+
+/*
+ * Update information in journal structure and in on disk journal superblock
+ * about log tail. This function does not check whether information passed in
+ * really pushes log tail further. It's responsibility of the caller to make
+ * sure provided log tail information is valid (e.g. by holding
+ * j_checkpoint_mutex all the time between computing log tail and calling this
+ * function as is the case with jbd2_cleanup_journal_tail()).
+ *
+ * Requires j_checkpoint_mutex
+ */
+void __jbd2_update_log_tail(journal_t *journal, tid_t tid, unsigned long block)
+{
+	unsigned long freed;
+
+	BUG_ON(!mutex_is_locked(&journal->j_checkpoint_mutex));
+
+	/*
+	 * We cannot afford for write to remain in drive's caches since as
+	 * soon as we update j_tail, next transaction can start reusing journal
+	 * space and if we lose sb update during power failure we'd replay
+	 * old transaction with possibly newly overwritten data.
+	 */
+	jbd2_journal_update_sb_log_tail(journal, tid, block, WRITE_FUA);
+	write_lock(&journal->j_state_lock);
+	freed = block - journal->j_tail;
+	if (block < journal->j_tail)
+		freed += journal->j_last - journal->j_first;
+
+	trace_jbd2_update_log_tail(journal, tid, block, freed);
+	jbd_debug(1,
+		  "Cleaning journal tail from %d to %d (offset %lu), "
+		  "freeing %lu\n",
+		  journal->j_tail_sequence, tid, block, freed);
+
+	journal->j_free += freed;
+	journal->j_tail_sequence = tid;
+	journal->j_tail = block;
+	write_unlock(&journal->j_state_lock);
+}
+
+/*
+ * This is a variaon of __jbd2_update_log_tail which checks for validity of
+ * provided log tail and locks j_checkpoint_mutex. So it is safe against races
+ * with other threads updating log tail.
+ */
+void jbd2_update_log_tail(journal_t *journal, tid_t tid, unsigned long block)
+{
+	mutex_lock(&journal->j_checkpoint_mutex);
+	if (tid_gt(tid, journal->j_tail_sequence))
+		__jbd2_update_log_tail(journal, tid, block);
+	mutex_unlock(&journal->j_checkpoint_mutex);
+}
+
 struct jbd2_stats_proc_session {
 	journal_t *journal;
 	struct transaction_stats_s *stats;
@@ -1114,40 +1204,45 @@ static int journal_reset(journal_t *journal)
 
 	journal->j_max_transaction_buffers = journal->j_maxlen / 4;
 
-	/* Add the dynamic fields and write it to disk. */
-	jbd2_journal_update_superblock(journal, 1);
-	return jbd2_journal_start_thread(journal);
-}
-
-/**
- * void jbd2_journal_update_superblock() - Update journal sb on disk.
- * @journal: The journal to update.
- * @wait: Set to '0' if you don't want to wait for IO completion.
- *
- * Update a journal's dynamic superblock fields and write it to disk,
- * optionally waiting for the IO to complete.
- */
-void jbd2_journal_update_superblock(journal_t *journal, int wait)
-{
-	journal_superblock_t *sb = journal->j_superblock;
-	struct buffer_head *bh = journal->j_sb_buffer;
-
 	/*
 	 * As a special case, if the on-disk copy is already marked as needing
-	 * no recovery (s_start == 0) and there are no outstanding transactions
-	 * in the filesystem, then we can safely defer the superblock update
-	 * until the next commit by setting JBD2_FLUSHED.  This avoids
+	 * no recovery (s_start == 0), then we can safely defer the superblock
+	 * update until the next commit by setting JBD2_FLUSHED.  This avoids
 	 * attempting a write to a potential-readonly device.
 	 */
-	if (sb->s_start == 0 && journal->j_tail_sequence ==
-				journal->j_transaction_sequence) {
+	if (sb->s_start == 0) {
 		jbd_debug(1, "JBD2: Skipping superblock update on recovered sb "
 			"(start %ld, seq %d, errno %d)\n",
 			journal->j_tail, journal->j_tail_sequence,
 			journal->j_errno);
-		goto out;
+		journal->j_flags |= JBD2_FLUSHED;
+	} else {
+		/* Lock here to make assertions happy... */
+		mutex_lock(&journal->j_checkpoint_mutex);
+		/*
+		 * Update log tail information. We use WRITE_FUA since new
+		 * transaction will start reusing journal space and so we
+		 * must make sure information about current log tail is on
+		 * disk before that.
+		 */
+		jbd2_journal_update_sb_log_tail(journal,
+						journal->j_tail_sequence,
+						journal->j_tail,
+						WRITE_FUA);
+		mutex_unlock(&journal->j_checkpoint_mutex);
 	}
+	return jbd2_journal_start_thread(journal);
+}
 
+static void jbd2_write_superblock(journal_t *journal, int write_op)
+{
+	struct buffer_head *bh = journal->j_sb_buffer;
+	int ret;
+
+	trace_jbd2_write_superblock(journal, write_op);
+	if (!(journal->j_flags & JBD2_BARRIER))
+		write_op &= ~(REQ_FUA | REQ_FLUSH);
+	lock_buffer(bh);
 	if (buffer_write_io_error(bh)) {
 		/*
 		 * Oh, dear.  A previous attempt to write the journal
@@ -1163,48 +1258,106 @@ void jbd2_journal_update_superblock(journal_t *journal, int wait)
 		clear_buffer_write_io_error(bh);
 		set_buffer_uptodate(bh);
 	}
+	get_bh(bh);
+	bh->b_end_io = end_buffer_write_sync;
+	ret = submit_bh(write_op, bh);
+	wait_on_buffer(bh);
+	if (buffer_write_io_error(bh)) {
+		clear_buffer_write_io_error(bh);
+		set_buffer_uptodate(bh);
+		ret = -EIO;
+	}
+	if (ret) {
+		printk(KERN_ERR "JBD2: Error %d detected when updating "
+		       "journal superblock for %s.\n", ret,
+		       journal->j_devname);
+	}
+}
+
+/**
+ * jbd2_journal_update_sb_log_tail() - Update log tail in journal sb on disk.
+ * @journal: The journal to update.
+ * @tail_tid: TID of the new transaction at the tail of the log
+ * @tail_block: The first block of the transaction at the tail of the log
+ * @write_op: With which operation should we write the journal sb
+ *
+ * Update a journal's superblock information about log tail and write it to
+ * disk, waiting for the IO to complete.
+ */
+void jbd2_journal_update_sb_log_tail(journal_t *journal, tid_t tail_tid,
+				     unsigned long tail_block, int write_op)
+{
+	journal_superblock_t *sb = journal->j_superblock;
+
+	BUG_ON(!mutex_is_locked(&journal->j_checkpoint_mutex));
+	jbd_debug(1, "JBD2: updating superblock (start %lu, seq %u)\n",
+		  tail_block, tail_tid);
+
+	sb->s_sequence = cpu_to_be32(tail_tid);
+	sb->s_start    = cpu_to_be32(tail_block);
+
+	jbd2_write_superblock(journal, write_op);
+
+	/* Log is no longer empty */
+	write_lock(&journal->j_state_lock);
+	WARN_ON(!sb->s_sequence);
+	journal->j_flags &= ~JBD2_FLUSHED;
+	write_unlock(&journal->j_state_lock);
+}
+
+/**
+ * jbd2_mark_journal_empty() - Mark on disk journal as empty.
+ * @journal: The journal to update.
+ *
+ * Update a journal's dynamic superblock fields to show that journal is empty.
+ * Write updated superblock to disk waiting for IO to complete.
+ */
+static void jbd2_mark_journal_empty(journal_t *journal)
+{
+	journal_superblock_t *sb = journal->j_superblock;
 
+	BUG_ON(!mutex_is_locked(&journal->j_checkpoint_mutex));
 	read_lock(&journal->j_state_lock);
-	jbd_debug(1, "JBD2: updating superblock (start %ld, seq %d, errno %d)\n",
-		  journal->j_tail, journal->j_tail_sequence, journal->j_errno);
+	jbd_debug(1, "JBD2: Marking journal as empty (seq %d)\n",
+		  journal->j_tail_sequence);
 
 	sb->s_sequence = cpu_to_be32(journal->j_tail_sequence);
-	sb->s_start    = cpu_to_be32(journal->j_tail);
-	sb->s_errno    = cpu_to_be32(journal->j_errno);
+	sb->s_start    = cpu_to_be32(0);
 	read_unlock(&journal->j_state_lock);
 
-	BUFFER_TRACE(bh, "marking dirty");
-	mark_buffer_dirty(bh);
-	if (wait) {
-		sync_dirty_buffer(bh);
-		if (buffer_write_io_error(bh)) {
-			printk(KERN_ERR "JBD2: I/O error detected "
-			       "when updating journal superblock for %s.\n",
-			       journal->j_devname);
-			clear_buffer_write_io_error(bh);
-			set_buffer_uptodate(bh);
-		}
-	} else
-		write_dirty_buffer(bh, WRITE);
-
-out:
-	/* If we have just flushed the log (by marking s_start==0), then
-	 * any future commit will have to be careful to update the
-	 * superblock again to re-record the true start of the log. */
+	jbd2_write_superblock(journal, WRITE_FUA);
 
+	/* Log is no longer empty */
 	write_lock(&journal->j_state_lock);
-	if (sb->s_start)
-		journal->j_flags &= ~JBD2_FLUSHED;
-	else
-		journal->j_flags |= JBD2_FLUSHED;
+	journal->j_flags |= JBD2_FLUSHED;
 	write_unlock(&journal->j_state_lock);
 }
 
+
+/**
+ * jbd2_journal_update_sb_errno() - Update error in the journal.
+ * @journal: The journal to update.
+ *
+ * Update a journal's errno.  Write updated superblock to disk waiting for IO
+ * to complete.
+ */
+static void jbd2_journal_update_sb_errno(journal_t *journal)
+{
+	journal_superblock_t *sb = journal->j_superblock;
+
+	read_lock(&journal->j_state_lock);
+	jbd_debug(1, "JBD2: updating superblock error (errno %d)\n",
+		  journal->j_errno);
+	sb->s_errno    = cpu_to_be32(journal->j_errno);
+	read_unlock(&journal->j_state_lock);
+
+	jbd2_write_superblock(journal, WRITE_SYNC);
+}
+
 /*
  * Read the superblock for a given journal, performing initial
  * validation of the format.
  */
-
 static int journal_get_superblock(journal_t *journal)
 {
 	struct buffer_head *bh;
@@ -1398,14 +1551,11 @@ int jbd2_journal_destroy(journal_t *journal)
 
 	if (journal->j_sb_buffer) {
 		if (!is_journal_aborted(journal)) {
-			/* We can now mark the journal as empty. */
-			journal->j_tail = 0;
-			journal->j_tail_sequence =
-				++journal->j_transaction_sequence;
-			jbd2_journal_update_superblock(journal, 1);
-		} else {
+			mutex_lock(&journal->j_checkpoint_mutex);
+			jbd2_mark_journal_empty(journal);
+			mutex_unlock(&journal->j_checkpoint_mutex);
+		} else
 			err = -EIO;
-		}
 		brelse(journal->j_sb_buffer);
 	}
 
@@ -1552,61 +1702,6 @@ void jbd2_journal_clear_features(journal_t *journal, unsigned long compat,
 EXPORT_SYMBOL(jbd2_journal_clear_features);
 
 /**
- * int jbd2_journal_update_format () - Update on-disk journal structure.
- * @journal: Journal to act on.
- *
- * Given an initialised but unloaded journal struct, poke about in the
- * on-disk structure to update it to the most recent supported version.
- */
-int jbd2_journal_update_format (journal_t *journal)
-{
-	journal_superblock_t *sb;
-	int err;
-
-	err = journal_get_superblock(journal);
-	if (err)
-		return err;
-
-	sb = journal->j_superblock;
-
-	switch (be32_to_cpu(sb->s_header.h_blocktype)) {
-	case JBD2_SUPERBLOCK_V2:
-		return 0;
-	case JBD2_SUPERBLOCK_V1:
-		return journal_convert_superblock_v1(journal, sb);
-	default:
-		break;
-	}
-	return -EINVAL;
-}
-
-static int journal_convert_superblock_v1(journal_t *journal,
-					 journal_superblock_t *sb)
-{
-	int offset, blocksize;
-	struct buffer_head *bh;
-
-	printk(KERN_WARNING
-		"JBD2: Converting superblock from version 1 to 2.\n");
-
-	/* Pre-initialise new fields to zero */
-	offset = ((char *) &(sb->s_feature_compat)) - ((char *) sb);
-	blocksize = be32_to_cpu(sb->s_blocksize);
-	memset(&sb->s_feature_compat, 0, blocksize-offset);
-
-	sb->s_nr_users = cpu_to_be32(1);
-	sb->s_header.h_blocktype = cpu_to_be32(JBD2_SUPERBLOCK_V2);
-	journal->j_format_version = 2;
-
-	bh = journal->j_sb_buffer;
-	BUFFER_TRACE(bh, "marking dirty");
-	mark_buffer_dirty(bh);
-	sync_dirty_buffer(bh);
-	return 0;
-}
-
-
-/**
  * int jbd2_journal_flush () - Flush journal
  * @journal: Journal to act on.
  *
@@ -1619,7 +1714,6 @@ int jbd2_journal_flush(journal_t *journal)
 {
 	int err = 0;
 	transaction_t *transaction = NULL;
-	unsigned long old_tail;
 
 	write_lock(&journal->j_state_lock);
 
@@ -1654,6 +1748,7 @@ int jbd2_journal_flush(journal_t *journal)
 	if (is_journal_aborted(journal))
 		return -EIO;
 
+	mutex_lock(&journal->j_checkpoint_mutex);
 	jbd2_cleanup_journal_tail(journal);
 
 	/* Finally, mark the journal as really needing no recovery.
@@ -1661,14 +1756,9 @@ int jbd2_journal_flush(journal_t *journal)
 	 * the magic code for a fully-recovered superblock.  Any future
 	 * commits of data to the journal will restore the current
 	 * s_start value. */
+	jbd2_mark_journal_empty(journal);
+	mutex_unlock(&journal->j_checkpoint_mutex);
 	write_lock(&journal->j_state_lock);
-	old_tail = journal->j_tail;
-	journal->j_tail = 0;
-	write_unlock(&journal->j_state_lock);
-	jbd2_journal_update_superblock(journal, 1);
-	write_lock(&journal->j_state_lock);
-	journal->j_tail = old_tail;
-
 	J_ASSERT(!journal->j_running_transaction);
 	J_ASSERT(!journal->j_committing_transaction);
 	J_ASSERT(!journal->j_checkpoint_transactions);
@@ -1708,8 +1798,12 @@ int jbd2_journal_wipe(journal_t *journal, int write)
 		write ? "Clearing" : "Ignoring");
 
 	err = jbd2_journal_skip_recovery(journal);
-	if (write)
-		jbd2_journal_update_superblock(journal, 1);
+	if (write) {
+		/* Lock to make assertions happy... */
+		mutex_lock(&journal->j_checkpoint_mutex);
+		jbd2_mark_journal_empty(journal);
+		mutex_unlock(&journal->j_checkpoint_mutex);
+	}
 
  no_recovery:
 	return err;
@@ -1759,7 +1853,7 @@ static void __journal_abort_soft (journal_t *journal, int errno)
 	__jbd2_journal_abort_hard(journal);
 
 	if (errno)
-		jbd2_journal_update_superblock(journal, 1);
+		jbd2_journal_update_sb_errno(journal);
 }
 
 /**
@@ -2017,7 +2111,7 @@ static struct kmem_cache *jbd2_journal_head_cache;
 static atomic_t nr_journal_heads = ATOMIC_INIT(0);
 #endif
 
-static int journal_init_jbd2_journal_head_cache(void)
+static int jbd2_journal_init_journal_head_cache(void)
 {
 	int retval;
 
@@ -2035,7 +2129,7 @@ static int journal_init_jbd2_journal_head_cache(void)
 	return retval;
 }
 
-static void jbd2_journal_destroy_jbd2_journal_head_cache(void)
+static void jbd2_journal_destroy_journal_head_cache(void)
 {
 	if (jbd2_journal_head_cache) {
 		kmem_cache_destroy(jbd2_journal_head_cache);
@@ -2323,7 +2417,7 @@ static void __exit jbd2_remove_jbd_stats_proc_entry(void)
 
 struct kmem_cache *jbd2_handle_cache, *jbd2_inode_cache;
 
-static int __init journal_init_handle_cache(void)
+static int __init jbd2_journal_init_handle_cache(void)
 {
 	jbd2_handle_cache = KMEM_CACHE(jbd2_journal_handle, SLAB_TEMPORARY);
 	if (jbd2_handle_cache == NULL) {
@@ -2358,17 +2452,20 @@ static int __init journal_init_caches(void)
 
 	ret = jbd2_journal_init_revoke_caches();
 	if (ret == 0)
-		ret = journal_init_jbd2_journal_head_cache();
+		ret = jbd2_journal_init_journal_head_cache();
+	if (ret == 0)
+		ret = jbd2_journal_init_handle_cache();
 	if (ret == 0)
-		ret = journal_init_handle_cache();
+		ret = jbd2_journal_init_transaction_cache();
 	return ret;
 }
 
 static void jbd2_journal_destroy_caches(void)
 {
 	jbd2_journal_destroy_revoke_caches();
-	jbd2_journal_destroy_jbd2_journal_head_cache();
+	jbd2_journal_destroy_journal_head_cache();
 	jbd2_journal_destroy_handle_cache();
+	jbd2_journal_destroy_transaction_cache();
 	jbd2_journal_destroy_slabs();
 }
 
diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c
index da6d7baf139..c1a03354a22 100644
--- a/fs/jbd2/recovery.c
+++ b/fs/jbd2/recovery.c
@@ -21,6 +21,7 @@
 #include <linux/jbd2.h>
 #include <linux/errno.h>
 #include <linux/crc32.h>
+#include <linux/blkdev.h>
 #endif
 
 /*
@@ -265,7 +266,9 @@ int jbd2_journal_recover(journal_t *journal)
 	err2 = sync_blockdev(journal->j_fs_dev);
 	if (!err)
 		err = err2;
-
+	/* Make sure all replayed data is on permanent storage */
+	if (journal->j_flags & JBD2_BARRIER)
+		blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL, NULL);
 	return err;
 }
 
diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c
index 30b2867d6cc..6973705d6a3 100644
--- a/fs/jbd2/revoke.c
+++ b/fs/jbd2/revoke.c
@@ -208,17 +208,13 @@ int __init jbd2_journal_init_revoke_caches(void)
 	J_ASSERT(!jbd2_revoke_record_cache);
 	J_ASSERT(!jbd2_revoke_table_cache);
 
-	jbd2_revoke_record_cache = kmem_cache_create("jbd2_revoke_record",
-					   sizeof(struct jbd2_revoke_record_s),
-					   0,
-					   SLAB_HWCACHE_ALIGN|SLAB_TEMPORARY,
-					   NULL);
+	jbd2_revoke_record_cache = KMEM_CACHE(jbd2_revoke_record_s,
+					SLAB_HWCACHE_ALIGN|SLAB_TEMPORARY);
 	if (!jbd2_revoke_record_cache)
 		goto record_cache_failure;
 
-	jbd2_revoke_table_cache = kmem_cache_create("jbd2_revoke_table",
-					   sizeof(struct jbd2_revoke_table_s),
-					   0, SLAB_TEMPORARY, NULL);
+	jbd2_revoke_table_cache = KMEM_CACHE(jbd2_revoke_table_s,
+					     SLAB_TEMPORARY);
 	if (!jbd2_revoke_table_cache)
 		goto table_cache_failure;
 	return 0;
diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c
index e5aba56e1fd..ddcd3549c6c 100644
--- a/fs/jbd2/transaction.c
+++ b/fs/jbd2/transaction.c
@@ -33,6 +33,35 @@
 static void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh);
 static void __jbd2_journal_unfile_buffer(struct journal_head *jh);
 
+static struct kmem_cache *transaction_cache;
+int __init jbd2_journal_init_transaction_cache(void)
+{
+	J_ASSERT(!transaction_cache);
+	transaction_cache = kmem_cache_create("jbd2_transaction_s",
+					sizeof(transaction_t),
+					0,
+					SLAB_HWCACHE_ALIGN|SLAB_TEMPORARY,
+					NULL);
+	if (transaction_cache)
+		return 0;
+	return -ENOMEM;
+}
+
+void jbd2_journal_destroy_transaction_cache(void)
+{
+	if (transaction_cache) {
+		kmem_cache_destroy(transaction_cache);
+		transaction_cache = NULL;
+	}
+}
+
+void jbd2_journal_free_transaction(transaction_t *transaction)
+{
+	if (unlikely(ZERO_OR_NULL_PTR(transaction)))
+		return;
+	kmem_cache_free(transaction_cache, transaction);
+}
+
 /*
  * jbd2_get_transaction: obtain a new transaction_t object.
  *
@@ -133,7 +162,8 @@ static int start_this_handle(journal_t *journal, handle_t *handle,
 
 alloc_transaction:
 	if (!journal->j_running_transaction) {
-		new_transaction = kzalloc(sizeof(*new_transaction), gfp_mask);
+		new_transaction = kmem_cache_alloc(transaction_cache,
+						   gfp_mask | __GFP_ZERO);
 		if (!new_transaction) {
 			/*
 			 * If __GFP_FS is not present, then we may be
@@ -162,7 +192,7 @@ repeat:
 	if (is_journal_aborted(journal) ||
 	    (journal->j_errno != 0 && !(journal->j_flags & JBD2_ACK_ERR))) {
 		read_unlock(&journal->j_state_lock);
-		kfree(new_transaction);
+		jbd2_journal_free_transaction(new_transaction);
 		return -EROFS;
 	}
 
@@ -284,7 +314,7 @@ repeat:
 	read_unlock(&journal->j_state_lock);
 
 	lock_map_acquire(&handle->h_lockdep_map);
-	kfree(new_transaction);
+	jbd2_journal_free_transaction(new_transaction);
 	return 0;
 }
 
@@ -1549,9 +1579,9 @@ __blist_del_buffer(struct journal_head **list, struct journal_head *jh)
  * of these pointers, it could go bad.  Generally the caller needs to re-read
  * the pointer from the transaction_t.
  *
- * Called under j_list_lock.  The journal may not be locked.
+ * Called under j_list_lock.
  */
-void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh)
+static void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh)
 {
 	struct journal_head **list = NULL;
 	transaction_t *transaction;
@@ -1646,10 +1676,8 @@ __journal_try_to_free_buffer(journal_t *journal, struct buffer_head *bh)
 	spin_lock(&journal->j_list_lock);
 	if (jh->b_cp_transaction != NULL && jh->b_transaction == NULL) {
 		/* written-back checkpointed metadata buffer */
-		if (jh->b_jlist == BJ_None) {
-			JBUFFER_TRACE(jh, "remove from checkpoint list");
-			__jbd2_journal_remove_checkpoint(jh);
-		}
+		JBUFFER_TRACE(jh, "remove from checkpoint list");
+		__jbd2_journal_remove_checkpoint(jh);
 	}
 	spin_unlock(&journal->j_list_lock);
 out:
@@ -1949,6 +1977,8 @@ zap_buffer_unlocked:
 	clear_buffer_mapped(bh);
 	clear_buffer_req(bh);
 	clear_buffer_new(bh);
+	clear_buffer_delay(bh);
+	clear_buffer_unwritten(bh);
 	bh->b_bdev = NULL;
 	return may_free;
 }