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

Pull ext4 updates from Ted Ts'o:
 "Mostly performance and bug fixes, plus some cleanups.  The one new
  feature this merge window is a new ioctl EXT4_IOC_SWAP_BOOT which
  allows installation of a hidden inode designed for boot loaders."

* tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (50 commits)
  ext4: fix type-widening bug in inode table readahead code
  ext4: add check for inodes_count overflow in new resize ioctl
  ext4: fix Kconfig documentation for CONFIG_EXT4_DEBUG
  ext4: fix online resizing for ext3-compat file systems
  jbd2: trace when lock_buffer in do_get_write_access takes a long time
  ext4: mark metadata blocks using bh flags
  buffer: add BH_Prio and BH_Meta flags
  ext4: mark all metadata I/O with REQ_META
  ext4: fix readdir error in case inline_data+^dir_index.
  ext4: fix readdir error in the case of inline_data+dir_index
  jbd2: use kmem_cache_zalloc instead of kmem_cache_alloc/memset
  ext4: mext_insert_extents should update extent block checksum
  ext4: move quota initialization out of inode allocation transaction
  ext4: reserve xattr index for Rich ACL support
  jbd2: reduce journal_head size
  ext4: clear buffer_uninit flag when submitting IO
  ext4: use io_end for multiple bios
  ext4: make ext4_bio_write_page() use BH_Async_Write flags
  ext4: Use kstrtoul() instead of parse_strtoul()
  ext4: defragmentation code cleanup
  ...

3 files changed, 67 insertions, 23 deletions

diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c
index 750c70148eff..0f53946f13c1 100644
--- a/fs/jbd2/commit.c
+++ b/fs/jbd2/commit.c
@@ -382,7 +382,7 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	int space_left = 0;
 	int first_tag = 0;
 	int tag_flag;
-	int i, to_free = 0;
+	int i;
 	int tag_bytes = journal_tag_bytes(journal);
 	struct buffer_head *cbh = NULL; /* For transactional checksums */
 	__u32 crc32_sum = ~0;
@@ -1134,7 +1134,7 @@ restart_loop:
 	journal->j_stats.run.rs_blocks_logged += stats.run.rs_blocks_logged;
 	spin_unlock(&journal->j_history_lock);
 
-	commit_transaction->t_state = T_FINISHED;
+	commit_transaction->t_state = T_COMMIT_CALLBACK;
 	J_ASSERT(commit_transaction == journal->j_committing_transaction);
 	journal->j_commit_sequence = commit_transaction->t_tid;
 	journal->j_committing_transaction = NULL;
@@ -1149,38 +1149,44 @@ restart_loop:
 				journal->j_average_commit_time*3) / 4;
 	else
 		journal->j_average_commit_time = commit_time;
+
 	write_unlock(&journal->j_state_lock);
 
-	if (commit_transaction->t_checkpoint_list == NULL &&
-	    commit_transaction->t_checkpoint_io_list == NULL) {
-		__jbd2_journal_drop_transaction(journal, commit_transaction);
-		to_free = 1;
+	if (journal->j_checkpoint_transactions == NULL) {
+		journal->j_checkpoint_transactions = commit_transaction;
+		commit_transaction->t_cpnext = commit_transaction;
+		commit_transaction->t_cpprev = commit_transaction;
 	} else {
-		if (journal->j_checkpoint_transactions == NULL) {
-			journal->j_checkpoint_transactions = commit_transaction;
-			commit_transaction->t_cpnext = commit_transaction;
-			commit_transaction->t_cpprev = commit_transaction;
-		} else {
-			commit_transaction->t_cpnext =
-				journal->j_checkpoint_transactions;
-			commit_transaction->t_cpprev =
-				commit_transaction->t_cpnext->t_cpprev;
-			commit_transaction->t_cpnext->t_cpprev =
-				commit_transaction;
-			commit_transaction->t_cpprev->t_cpnext =
+		commit_transaction->t_cpnext =
+			journal->j_checkpoint_transactions;
+		commit_transaction->t_cpprev =
+			commit_transaction->t_cpnext->t_cpprev;
+		commit_transaction->t_cpnext->t_cpprev =
+			commit_transaction;
+		commit_transaction->t_cpprev->t_cpnext =
 				commit_transaction;
-		}
 	}
 	spin_unlock(&journal->j_list_lock);
-
+	/* Drop all spin_locks because commit_callback may be block.
+	 * __journal_remove_checkpoint() can not destroy transaction
+	 * under us because it is not marked as T_FINISHED yet */
 	if (journal->j_commit_callback)
 		journal->j_commit_callback(journal, commit_transaction);
 
 	trace_jbd2_end_commit(journal, commit_transaction);
 	jbd_debug(1, "JBD2: commit %d complete, head %d\n",
 		  journal->j_commit_sequence, journal->j_tail_sequence);
-	if (to_free)
-		jbd2_journal_free_transaction(commit_transaction);
 
+	write_lock(&journal->j_state_lock);
+	spin_lock(&journal->j_list_lock);
+	commit_transaction->t_state = T_FINISHED;
+	/* Recheck checkpoint lists after j_list_lock was dropped */
+	if (commit_transaction->t_checkpoint_list == NULL &&
+	    commit_transaction->t_checkpoint_io_list == NULL) {
+		__jbd2_journal_drop_transaction(journal, commit_transaction);
+		jbd2_journal_free_transaction(commit_transaction);
+	}
+	spin_unlock(&journal->j_list_lock);
+	write_unlock(&journal->j_state_lock);
 	wake_up(&journal->j_wait_done_commit);
 }
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index 8b220f1ab54f..f6c5ba027f4f 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -708,6 +708,37 @@ int jbd2_log_wait_commit(journal_t *journal, tid_t tid)
 }
 
 /*
+ * When this function returns the transaction corresponding to tid
+ * will be completed.  If the transaction has currently running, start
+ * committing that transaction before waiting for it to complete.  If
+ * the transaction id is stale, it is by definition already completed,
+ * so just return SUCCESS.
+ */
+int jbd2_complete_transaction(journal_t *journal, tid_t tid)
+{
+	int	need_to_wait = 1;
+
+	read_lock(&journal->j_state_lock);
+	if (journal->j_running_transaction &&
+	    journal->j_running_transaction->t_tid == tid) {
+		if (journal->j_commit_request != tid) {
+			/* transaction not yet started, so request it */
+			read_unlock(&journal->j_state_lock);
+			jbd2_log_start_commit(journal, tid);
+			goto wait_commit;
+		}
+	} else if (!(journal->j_committing_transaction &&
+		     journal->j_committing_transaction->t_tid == tid))
+		need_to_wait = 0;
+	read_unlock(&journal->j_state_lock);
+	if (!need_to_wait)
+		return 0;
+wait_commit:
+	return jbd2_log_wait_commit(journal, tid);
+}
+EXPORT_SYMBOL(jbd2_complete_transaction);
+
+/*
  * Log buffer allocation routines:
  */
 
diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c
index 325bc019ed88..10f524c59ea8 100644
--- a/fs/jbd2/transaction.c
+++ b/fs/jbd2/transaction.c
@@ -332,7 +332,6 @@ static handle_t *new_handle(int nblocks)
 	handle_t *handle = jbd2_alloc_handle(GFP_NOFS);
 	if (!handle)
 		return NULL;
-	memset(handle, 0, sizeof(*handle));
 	handle->h_buffer_credits = nblocks;
 	handle->h_ref = 1;
 
@@ -640,6 +639,7 @@ do_get_write_access(handle_t *handle, struct journal_head *jh,
 	int error;
 	char *frozen_buffer = NULL;
 	int need_copy = 0;
+	unsigned long start_lock, time_lock;
 
 	if (is_handle_aborted(handle))
 		return -EROFS;
@@ -655,9 +655,16 @@ repeat:
 
 	/* @@@ Need to check for errors here at some point. */
 
+ 	start_lock = jiffies;
 	lock_buffer(bh);
 	jbd_lock_bh_state(bh);
 
+	/* If it takes too long to lock the buffer, trace it */
+	time_lock = jbd2_time_diff(start_lock, jiffies);
+	if (time_lock > HZ/10)
+		trace_jbd2_lock_buffer_stall(bh->b_bdev->bd_dev,
+			jiffies_to_msecs(time_lock));
+
 	/* We now hold the buffer lock so it is safe to query the buffer
 	 * state.  Is the buffer dirty?
 	 *