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

Pull ext4 updates from Ted Ts'o:
 "The big new feature added this time is supporting online resizing
  using the meta_bg feature.  This allows us to resize file systems
  which are greater than 16TB.  In addition, the speed of online
  resizing has been improved in general.

  We also fix a number of races, some of which could lead to deadlocks,
  in ext4's Asynchronous I/O and online defrag support, thanks to good
  work by Dmitry Monakhov.

  There are also a large number of more minor bug fixes and cleanups
  from a number of other ext4 contributors, quite of few of which have
  submitted fixes for the first time."

* tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (69 commits)
  ext4: fix ext4_flush_completed_IO wait semantics
  ext4: fix mtime update in nodelalloc mode
  ext4: fix ext_remove_space for punch_hole case
  ext4: punch_hole should wait for DIO writers
  ext4: serialize truncate with owerwrite DIO workers
  ext4: endless truncate due to nonlocked dio readers
  ext4: serialize unlocked dio reads with truncate
  ext4: serialize dio nonlocked reads with defrag workers
  ext4: completed_io locking cleanup
  ext4: fix unwritten counter leakage
  ext4: give i_aiodio_unwritten a more appropriate name
  ext4: ext4_inode_info diet
  ext4: convert to use leXX_add_cpu()
  ext4: ext4_bread usage audit
  fs: reserve fallocate flag codepoint
  ext4: remove redundant offset check in mext_check_arguments()
  ext4: don't clear orphan list on ro mount with errors
  jbd2: fix assertion failure in commit code due to lacking transaction credits
  ext4: release donor reference when EXT4_IOC_MOVE_EXT ioctl fails
  ext4: enable FITRIM ioctl on bigalloc file system
  ...

4 files changed, 84 insertions, 33 deletions

diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c
index af5280fb579b..3091d42992f0 100644
--- a/fs/jbd2/commit.c
+++ b/fs/jbd2/commit.c
@@ -1014,17 +1014,35 @@ restart_loop:
 		 * there's no point in keeping a checkpoint record for
 		 * it. */
 
-		/* A buffer which has been freed while still being
-		 * journaled by a previous transaction may end up still
-		 * being dirty here, but we want to avoid writing back
-		 * that buffer in the future after the "add to orphan"
-		 * operation been committed,  That's not only a performance
-		 * gain, it also stops aliasing problems if the buffer is
-		 * left behind for writeback and gets reallocated for another
-		 * use in a different page. */
-		if (buffer_freed(bh) && !jh->b_next_transaction) {
-			clear_buffer_freed(bh);
-			clear_buffer_jbddirty(bh);
+		/*
+		* A buffer which has been freed while still being journaled by
+		* a previous transaction.
+		*/
+		if (buffer_freed(bh)) {
+			/*
+			 * If the running transaction is the one containing
+			 * "add to orphan" operation (b_next_transaction !=
+			 * NULL), we have to wait for that transaction to
+			 * commit before we can really get rid of the buffer.
+			 * So just clear b_modified to not confuse transaction
+			 * credit accounting and refile the buffer to
+			 * BJ_Forget of the running transaction. If the just
+			 * committed transaction contains "add to orphan"
+			 * operation, we can completely invalidate the buffer
+			 * now. We are rather through in that since the
+			 * buffer may be still accessible when blocksize <
+			 * pagesize and it is attached to the last partial
+			 * page.
+			 */
+			jh->b_modified = 0;
+			if (!jh->b_next_transaction) {
+				clear_buffer_freed(bh);
+				clear_buffer_jbddirty(bh);
+				clear_buffer_mapped(bh);
+				clear_buffer_new(bh);
+				clear_buffer_req(bh);
+				bh->b_bdev = NULL;
+			}
 		}
 
 		if (buffer_jbddirty(bh)) {
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index e149b99a7ffb..484b8d1c6cb6 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -1354,6 +1354,11 @@ static void jbd2_mark_journal_empty(journal_t *journal)
 
 	BUG_ON(!mutex_is_locked(&journal->j_checkpoint_mutex));
 	read_lock(&journal->j_state_lock);
+	/* Is it already empty? */
+	if (sb->s_start == 0) {
+		read_unlock(&journal->j_state_lock);
+		return;
+	}
 	jbd_debug(1, "JBD2: Marking journal as empty (seq %d)\n",
 		  journal->j_tail_sequence);
 
diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c
index 0131e4362534..626846bac32f 100644
--- a/fs/jbd2/recovery.c
+++ b/fs/jbd2/recovery.c
@@ -289,8 +289,11 @@ int jbd2_journal_recover(journal_t *journal)
 	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);
+	if (journal->j_flags & JBD2_BARRIER) {
+		err2 = blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL, NULL);
+		if (!err)
+			err = err2;
+	}
 	return err;
 }
 
diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c
index fb1ab9533b67..a74ba4659549 100644
--- a/fs/jbd2/transaction.c
+++ b/fs/jbd2/transaction.c
@@ -1841,15 +1841,16 @@ static int __dispose_buffer(struct journal_head *jh, transaction_t *transaction)
  * We're outside-transaction here.  Either or both of j_running_transaction
  * and j_committing_transaction may be NULL.
  */
-static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh)
+static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh,
+				int partial_page)
 {
 	transaction_t *transaction;
 	struct journal_head *jh;
 	int may_free = 1;
-	int ret;
 
 	BUFFER_TRACE(bh, "entry");
 
+retry:
 	/*
 	 * It is safe to proceed here without the j_list_lock because the
 	 * buffers cannot be stolen by try_to_free_buffers as long as we are
@@ -1878,10 +1879,18 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh)
 	 * clear the buffer dirty bit at latest at the moment when the
 	 * transaction marking the buffer as freed in the filesystem
 	 * structures is committed because from that moment on the
-	 * buffer can be reallocated and used by a different page.
+	 * block can be reallocated and used by a different page.
 	 * Since the block hasn't been freed yet but the inode has
 	 * already been added to orphan list, it is safe for us to add
 	 * the buffer to BJ_Forget list of the newest transaction.
+	 *
+	 * Also we have to clear buffer_mapped flag of a truncated buffer
+	 * because the buffer_head may be attached to the page straddling
+	 * i_size (can happen only when blocksize < pagesize) and thus the
+	 * buffer_head can be reused when the file is extended again. So we end
+	 * up keeping around invalidated buffers attached to transactions'
+	 * BJ_Forget list just to stop checkpointing code from cleaning up
+	 * the transaction this buffer was modified in.
 	 */
 	transaction = jh->b_transaction;
 	if (transaction == NULL) {
@@ -1908,13 +1917,9 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh)
 			 * committed, the buffer won't be needed any
 			 * longer. */
 			JBUFFER_TRACE(jh, "checkpointed: add to BJ_Forget");
-			ret = __dispose_buffer(jh,
+			may_free = __dispose_buffer(jh,
 					journal->j_running_transaction);
-			jbd2_journal_put_journal_head(jh);
-			spin_unlock(&journal->j_list_lock);
-			jbd_unlock_bh_state(bh);
-			write_unlock(&journal->j_state_lock);
-			return ret;
+			goto zap_buffer;
 		} else {
 			/* There is no currently-running transaction. So the
 			 * orphan record which we wrote for this file must have
@@ -1922,13 +1927,9 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh)
 			 * the committing transaction, if it exists. */
 			if (journal->j_committing_transaction) {
 				JBUFFER_TRACE(jh, "give to committing trans");
-				ret = __dispose_buffer(jh,
+				may_free = __dispose_buffer(jh,
 					journal->j_committing_transaction);
-				jbd2_journal_put_journal_head(jh);
-				spin_unlock(&journal->j_list_lock);
-				jbd_unlock_bh_state(bh);
-				write_unlock(&journal->j_state_lock);
-				return ret;
+				goto zap_buffer;
 			} else {
 				/* The orphan record's transaction has
 				 * committed.  We can cleanse this buffer */
@@ -1940,10 +1941,24 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh)
 		JBUFFER_TRACE(jh, "on committing transaction");
 		/*
 		 * The buffer is committing, we simply cannot touch
-		 * it. So we just set j_next_transaction to the
-		 * running transaction (if there is one) and mark
-		 * buffer as freed so that commit code knows it should
-		 * clear dirty bits when it is done with the buffer.
+		 * it. If the page is straddling i_size we have to wait
+		 * for commit and try again.
+		 */
+		if (partial_page) {
+			tid_t tid = journal->j_committing_transaction->t_tid;
+
+			jbd2_journal_put_journal_head(jh);
+			spin_unlock(&journal->j_list_lock);
+			jbd_unlock_bh_state(bh);
+			write_unlock(&journal->j_state_lock);
+			jbd2_log_wait_commit(journal, tid);
+			goto retry;
+		}
+		/*
+		 * OK, buffer won't be reachable after truncate. We just set
+		 * j_next_transaction to the running transaction (if there is
+		 * one) and mark buffer as freed so that commit code knows it
+		 * should clear dirty bits when it is done with the buffer.
 		 */
 		set_buffer_freed(bh);
 		if (journal->j_running_transaction && buffer_jbddirty(bh))
@@ -1966,6 +1981,15 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh)
 	}
 
 zap_buffer:
+	/*
+	 * This is tricky. Although the buffer is truncated, it may be reused
+	 * if blocksize < pagesize and it is attached to the page straddling
+	 * EOF. Since the buffer might have been added to BJ_Forget list of the
+	 * running transaction, journal_get_write_access() won't clear
+	 * b_modified and credit accounting gets confused. So clear b_modified
+	 * here.
+	 */
+	jh->b_modified = 0;
 	jbd2_journal_put_journal_head(jh);
 zap_buffer_no_jh:
 	spin_unlock(&journal->j_list_lock);
@@ -2017,7 +2041,8 @@ void jbd2_journal_invalidatepage(journal_t *journal,
 		if (offset <= curr_off) {
 			/* This block is wholly outside the truncation point */
 			lock_buffer(bh);
-			may_free &= journal_unmap_buffer(journal, bh);
+			may_free &= journal_unmap_buffer(journal, bh,
+							 offset > 0);
 			unlock_buffer(bh);
 		}
 		curr_off = next_off;