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

Pull ext4 updates from Ted Ts'o:
 "Major changes for 3.14 include support for the newly added ZERO_RANGE
  and COLLAPSE_RANGE fallocate operations, and scalability improvements
  in the jbd2 layer and in xattr handling when the extended attributes
  spill over into an external block.

  Other than that, the usual clean ups and minor bug fixes"

* tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (42 commits)
  ext4: fix premature freeing of partial clusters split across leaf blocks
  ext4: remove unneeded test of ret variable
  ext4: fix comment typo
  ext4: make ext4_block_zero_page_range static
  ext4: atomically set inode->i_flags in ext4_set_inode_flags()
  ext4: optimize Hurd tests when reading/writing inodes
  ext4: kill i_version support for Hurd-castrated file systems
  ext4: each filesystem creates and uses its own mb_cache
  fs/mbcache.c: doucple the locking of local from global data
  fs/mbcache.c: change block and index hash chain to hlist_bl_node
  ext4: Introduce FALLOC_FL_ZERO_RANGE flag for fallocate
  ext4: refactor ext4_fallocate code
  ext4: Update inode i_size after the preallocation
  ext4: fix partial cluster handling for bigalloc file systems
  ext4: delete path dealloc code in ext4_ext_handle_uninitialized_extents
  ext4: only call sync_filesystm() when remounting read-only
  fs: push sync_filesystem() down to the file system's remount_fs()
  jbd2: improve error messages for inconsistent journal heads
  jbd2: minimize region locked by j_list_lock in jbd2_journal_forget()
  jbd2: minimize region locked by j_list_lock in journal_get_create_access()
  ...

3 files changed, 65 insertions, 68 deletions

diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c
index cf2fc0594063..5f26139a165a 100644
--- a/fs/jbd2/commit.c
+++ b/fs/jbd2/commit.c
@@ -555,7 +555,6 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	blk_start_plug(&plug);
 	jbd2_journal_write_revoke_records(journal, commit_transaction,
 					  &log_bufs, WRITE_SYNC);
-	blk_finish_plug(&plug);
 
 	jbd_debug(3, "JBD2: commit phase 2b\n");
 
@@ -582,7 +581,6 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	err = 0;
 	bufs = 0;
 	descriptor = NULL;
-	blk_start_plug(&plug);
 	while (commit_transaction->t_buffers) {
 
 		/* Find the next buffer to be journaled... */
@@ -1067,6 +1065,25 @@ restart_loop:
 		goto restart_loop;
 	}
 
+	/* Add the transaction to the checkpoint list
+	 * __journal_remove_checkpoint() can not destroy transaction
+	 * under us because it is not marked as T_FINISHED yet */
+	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;
+	}
+	spin_unlock(&journal->j_list_lock);
+
 	/* Done with this transaction! */
 
 	jbd_debug(3, "JBD2: commit phase 7\n");
@@ -1085,24 +1102,7 @@ restart_loop:
 		atomic_read(&commit_transaction->t_handle_count);
 	trace_jbd2_run_stats(journal->j_fs_dev->bd_dev,
 			     commit_transaction->t_tid, &stats.run);
-
-	/*
-	 * Calculate overall stats
-	 */
-	spin_lock(&journal->j_history_lock);
-	journal->j_stats.ts_tid++;
-	if (commit_transaction->t_requested)
-		journal->j_stats.ts_requested++;
-	journal->j_stats.run.rs_wait += stats.run.rs_wait;
-	journal->j_stats.run.rs_request_delay += stats.run.rs_request_delay;
-	journal->j_stats.run.rs_running += stats.run.rs_running;
-	journal->j_stats.run.rs_locked += stats.run.rs_locked;
-	journal->j_stats.run.rs_flushing += stats.run.rs_flushing;
-	journal->j_stats.run.rs_logging += stats.run.rs_logging;
-	journal->j_stats.run.rs_handle_count += stats.run.rs_handle_count;
-	journal->j_stats.run.rs_blocks += stats.run.rs_blocks;
-	journal->j_stats.run.rs_blocks_logged += stats.run.rs_blocks_logged;
-	spin_unlock(&journal->j_history_lock);
+	stats.ts_requested = (commit_transaction->t_requested) ? 1 : 0;
 
 	commit_transaction->t_state = T_COMMIT_CALLBACK;
 	J_ASSERT(commit_transaction == journal->j_committing_transaction);
@@ -1122,24 +1122,6 @@ restart_loop:
 
 	write_unlock(&journal->j_state_lock);
 
-	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;
-	}
-	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);
 
@@ -1150,7 +1132,7 @@ restart_loop:
 	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 */
+	/* Check if the transaction can be dropped now that we are finished */
 	if (commit_transaction->t_checkpoint_list == NULL &&
 	    commit_transaction->t_checkpoint_io_list == NULL) {
 		__jbd2_journal_drop_transaction(journal, commit_transaction);
@@ -1159,4 +1141,21 @@ restart_loop:
 	spin_unlock(&journal->j_list_lock);
 	write_unlock(&journal->j_state_lock);
 	wake_up(&journal->j_wait_done_commit);
+
+	/*
+	 * Calculate overall stats
+	 */
+	spin_lock(&journal->j_history_lock);
+	journal->j_stats.ts_tid++;
+	journal->j_stats.ts_requested += stats.ts_requested;
+	journal->j_stats.run.rs_wait += stats.run.rs_wait;
+	journal->j_stats.run.rs_request_delay += stats.run.rs_request_delay;
+	journal->j_stats.run.rs_running += stats.run.rs_running;
+	journal->j_stats.run.rs_locked += stats.run.rs_locked;
+	journal->j_stats.run.rs_flushing += stats.run.rs_flushing;
+	journal->j_stats.run.rs_logging += stats.run.rs_logging;
+	journal->j_stats.run.rs_handle_count += stats.run.rs_handle_count;
+	journal->j_stats.run.rs_blocks += stats.run.rs_blocks;
+	journal->j_stats.run.rs_blocks_logged += stats.run.rs_blocks_logged;
+	spin_unlock(&journal->j_history_lock);
 }
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index 5fa344afb49a..67b8e303946c 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -122,7 +122,7 @@ EXPORT_SYMBOL(__jbd2_debug);
 #endif
 
 /* Checksumming functions */
-int jbd2_verify_csum_type(journal_t *j, journal_superblock_t *sb)
+static int jbd2_verify_csum_type(journal_t *j, journal_superblock_t *sb)
 {
 	if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2))
 		return 1;
@@ -143,7 +143,7 @@ static __be32 jbd2_superblock_csum(journal_t *j, journal_superblock_t *sb)
 	return cpu_to_be32(csum);
 }
 
-int jbd2_superblock_csum_verify(journal_t *j, journal_superblock_t *sb)
+static int jbd2_superblock_csum_verify(journal_t *j, journal_superblock_t *sb)
 {
 	if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2))
 		return 1;
@@ -151,7 +151,7 @@ int jbd2_superblock_csum_verify(journal_t *j, journal_superblock_t *sb)
 	return sb->s_checksum == jbd2_superblock_csum(j, sb);
 }
 
-void jbd2_superblock_csum_set(journal_t *j, journal_superblock_t *sb)
+static void jbd2_superblock_csum_set(journal_t *j, journal_superblock_t *sb)
 {
 	if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2))
 		return;
@@ -302,8 +302,8 @@ static void journal_kill_thread(journal_t *journal)
 	journal->j_flags |= JBD2_UNMOUNT;
 
 	while (journal->j_task) {
-		wake_up(&journal->j_wait_commit);
 		write_unlock(&journal->j_state_lock);
+		wake_up(&journal->j_wait_commit);
 		wait_event(journal->j_wait_done_commit, journal->j_task == NULL);
 		write_lock(&journal->j_state_lock);
 	}
@@ -710,8 +710,8 @@ int jbd2_log_wait_commit(journal_t *journal, tid_t tid)
 	while (tid_gt(tid, journal->j_commit_sequence)) {
 		jbd_debug(1, "JBD2: want %d, j_commit_sequence=%d\n",
 				  tid, journal->j_commit_sequence);
-		wake_up(&journal->j_wait_commit);
 		read_unlock(&journal->j_state_lock);
+		wake_up(&journal->j_wait_commit);
 		wait_event(journal->j_wait_done_commit,
 				!tid_gt(tid, journal->j_commit_sequence));
 		read_lock(&journal->j_state_lock);
diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c
index 60bb365f54a5..38cfcf5f6fce 100644
--- a/fs/jbd2/transaction.c
+++ b/fs/jbd2/transaction.c
@@ -1073,7 +1073,6 @@ int jbd2_journal_get_create_access(handle_t *handle, struct buffer_head *bh)
 	 * reused here.
 	 */
 	jbd_lock_bh_state(bh);
-	spin_lock(&journal->j_list_lock);
 	J_ASSERT_JH(jh, (jh->b_transaction == transaction ||
 		jh->b_transaction == NULL ||
 		(jh->b_transaction == journal->j_committing_transaction &&
@@ -1096,12 +1095,14 @@ int jbd2_journal_get_create_access(handle_t *handle, struct buffer_head *bh)
 		jh->b_modified = 0;
 
 		JBUFFER_TRACE(jh, "file as BJ_Reserved");
+		spin_lock(&journal->j_list_lock);
 		__jbd2_journal_file_buffer(jh, transaction, BJ_Reserved);
 	} else if (jh->b_transaction == journal->j_committing_transaction) {
 		/* first access by this transaction */
 		jh->b_modified = 0;
 
 		JBUFFER_TRACE(jh, "set next transaction");
+		spin_lock(&journal->j_list_lock);
 		jh->b_next_transaction = transaction;
 	}
 	spin_unlock(&journal->j_list_lock);
@@ -1312,7 +1313,7 @@ int jbd2_journal_dirty_metadata(handle_t *handle, struct buffer_head *bh)
 			     journal->j_running_transaction)) {
 			printk(KERN_ERR "JBD2: %s: "
 			       "jh->b_transaction (%llu, %p, %u) != "
-			       "journal->j_running_transaction (%p, %u)",
+			       "journal->j_running_transaction (%p, %u)\n",
 			       journal->j_devname,
 			       (unsigned long long) bh->b_blocknr,
 			       jh->b_transaction,
@@ -1335,30 +1336,25 @@ int jbd2_journal_dirty_metadata(handle_t *handle, struct buffer_head *bh)
 	 */
 	if (jh->b_transaction != transaction) {
 		JBUFFER_TRACE(jh, "already on other transaction");
-		if (unlikely(jh->b_transaction !=
-			     journal->j_committing_transaction)) {
-			printk(KERN_ERR "JBD2: %s: "
-			       "jh->b_transaction (%llu, %p, %u) != "
-			       "journal->j_committing_transaction (%p, %u)",
+		if (unlikely(((jh->b_transaction !=
+			       journal->j_committing_transaction)) ||
+			     (jh->b_next_transaction != transaction))) {
+			printk(KERN_ERR "jbd2_journal_dirty_metadata: %s: "
+			       "bad jh for block %llu: "
+			       "transaction (%p, %u), "
+			       "jh->b_transaction (%p, %u), "
+			       "jh->b_next_transaction (%p, %u), jlist %u\n",
 			       journal->j_devname,
 			       (unsigned long long) bh->b_blocknr,
+			       transaction, transaction->t_tid,
 			       jh->b_transaction,
-			       jh->b_transaction ? jh->b_transaction->t_tid : 0,
-			       journal->j_committing_transaction,
-			       journal->j_committing_transaction ?
-			       journal->j_committing_transaction->t_tid : 0);
-			ret = -EINVAL;
-		}
-		if (unlikely(jh->b_next_transaction != transaction)) {
-			printk(KERN_ERR "JBD2: %s: "
-			       "jh->b_next_transaction (%llu, %p, %u) != "
-			       "transaction (%p, %u)",
-			       journal->j_devname,
-			       (unsigned long long) bh->b_blocknr,
+			       jh->b_transaction ?
+			       jh->b_transaction->t_tid : 0,
 			       jh->b_next_transaction,
 			       jh->b_next_transaction ?
 			       jh->b_next_transaction->t_tid : 0,
-			       transaction, transaction->t_tid);
+			       jh->b_jlist);
+			WARN_ON(1);
 			ret = -EINVAL;
 		}
 		/* And this case is illegal: we can't reuse another
@@ -1415,7 +1411,6 @@ int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh)
 	BUFFER_TRACE(bh, "entry");
 
 	jbd_lock_bh_state(bh);
-	spin_lock(&journal->j_list_lock);
 
 	if (!buffer_jbd(bh))
 		goto not_jbd;
@@ -1468,6 +1463,7 @@ int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh)
 		 * we know to remove the checkpoint after we commit.
 		 */
 
+		spin_lock(&journal->j_list_lock);
 		if (jh->b_cp_transaction) {
 			__jbd2_journal_temp_unlink_buffer(jh);
 			__jbd2_journal_file_buffer(jh, transaction, BJ_Forget);
@@ -1480,6 +1476,7 @@ int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh)
 				goto drop;
 			}
 		}
+		spin_unlock(&journal->j_list_lock);
 	} else if (jh->b_transaction) {
 		J_ASSERT_JH(jh, (jh->b_transaction ==
 				 journal->j_committing_transaction));
@@ -1491,7 +1488,9 @@ int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh)
 
 		if (jh->b_next_transaction) {
 			J_ASSERT(jh->b_next_transaction == transaction);
+			spin_lock(&journal->j_list_lock);
 			jh->b_next_transaction = NULL;
+			spin_unlock(&journal->j_list_lock);
 
 			/*
 			 * only drop a reference if this transaction modified
@@ -1503,7 +1502,6 @@ int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh)
 	}
 
 not_jbd:
-	spin_unlock(&journal->j_list_lock);
 	jbd_unlock_bh_state(bh);
 	__brelse(bh);
 drop:
@@ -1821,11 +1819,11 @@ __journal_try_to_free_buffer(journal_t *journal, struct buffer_head *bh)
 	if (buffer_locked(bh) || buffer_dirty(bh))
 		goto out;
 
-	if (jh->b_next_transaction != NULL)
+	if (jh->b_next_transaction != NULL || jh->b_transaction != NULL)
 		goto out;
 
 	spin_lock(&journal->j_list_lock);
-	if (jh->b_cp_transaction != NULL && jh->b_transaction == NULL) {
+	if (jh->b_cp_transaction != NULL) {
 		/* written-back checkpointed metadata buffer */
 		JBUFFER_TRACE(jh, "remove from checkpoint list");
 		__jbd2_journal_remove_checkpoint(jh);