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

Pull ext4 updates from Ted Ts'o:
 "A very large number of cleanups and bug fixes --- in particular for
  the ext4 encryption patches, which is a new feature added in the last
  merge window.  Also fix a number of long-standing xfstest failures.
  (Quota writes failing due to ENOSPC, a race between truncate and
  writepage in data=journalled mode that was causing generic/068 to
  fail, and other corner cases.)

  Also add support for FALLOC_FL_INSERT_RANGE, and improve jbd2
  performance eliminating locking when a buffer is modified more than
  once during a transaction (which is very common for allocation
  bitmaps, for example), in which case the state of the journalled
  buffer head doesn't need to change"

[ I renamed "ext4_follow_link()" to "ext4_encrypted_follow_link()" in
  the merge resolution, to make it clear that that function is _only_
  used for encrypted symlinks.  The function doesn't actually work for
  non-encrypted symlinks at all, and they use the generic helpers
                                         - Linus ]

* tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (52 commits)
  ext4: set lazytime on remount if MS_LAZYTIME is set by mount
  ext4: only call ext4_truncate when size <= isize
  ext4: make online defrag error reporting consistent
  ext4: minor cleanup of ext4_da_reserve_space()
  ext4: don't retry file block mapping on bigalloc fs with non-extent file
  ext4: prevent ext4_quota_write() from failing due to ENOSPC
  ext4: call sync_blockdev() before invalidate_bdev() in put_super()
  jbd2: speedup jbd2_journal_dirty_metadata()
  jbd2: get rid of open coded allocation retry loop
  ext4: improve warning directory handling messages
  jbd2: fix ocfs2 corrupt when updating journal superblock fails
  ext4: mballoc: avoid 20-argument function call
  ext4: wait for existing dio workers in ext4_alloc_file_blocks()
  ext4: recalculate journal credits as inode depth changes
  jbd2: use GFP_NOFS in jbd2_cleanup_journal_tail()
  ext4: use swap() in mext_page_double_lock()
  ext4: use swap() in memswap()
  ext4: fix race between truncate and __ext4_journalled_writepage()
  ext4 crypto: fail the mount if blocksize != pagesize
  ext4: Add support FALLOC_FL_INSERT_RANGE for fallocate
  ...

1 files changed, 35 insertions, 22 deletions

diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index b96bd8076b70..179d7d8733f2 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -371,16 +371,7 @@ int jbd2_journal_write_metadata_buffer(transaction_t *transaction,
 	 */
 	J_ASSERT_BH(bh_in, buffer_jbddirty(bh_in));
 
-retry_alloc:
-	new_bh = alloc_buffer_head(GFP_NOFS);
-	if (!new_bh) {
-		/*
-		 * Failure is not an option, but __GFP_NOFAIL is going
-		 * away; so we retry ourselves here.
-		 */
-		congestion_wait(BLK_RW_ASYNC, HZ/50);
-		goto retry_alloc;
-	}
+	new_bh = alloc_buffer_head(GFP_NOFS|__GFP_NOFAIL);
 
 	/* keep subsequent assertions sane */
 	atomic_set(&new_bh->b_count, 1);
@@ -885,9 +876,10 @@ int jbd2_journal_get_log_tail(journal_t *journal, tid_t *tid,
  *
  * Requires j_checkpoint_mutex
  */
-void __jbd2_update_log_tail(journal_t *journal, tid_t tid, unsigned long block)
+int __jbd2_update_log_tail(journal_t *journal, tid_t tid, unsigned long block)
 {
 	unsigned long freed;
+	int ret;
 
 	BUG_ON(!mutex_is_locked(&journal->j_checkpoint_mutex));
 
@@ -897,7 +889,10 @@ void __jbd2_update_log_tail(journal_t *journal, tid_t tid, unsigned long block)
 	 * 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);
+	ret = jbd2_journal_update_sb_log_tail(journal, tid, block, WRITE_FUA);
+	if (ret)
+		goto out;
+
 	write_lock(&journal->j_state_lock);
 	freed = block - journal->j_tail;
 	if (block < journal->j_tail)
@@ -913,6 +908,9 @@ void __jbd2_update_log_tail(journal_t *journal, tid_t tid, unsigned long block)
 	journal->j_tail_sequence = tid;
 	journal->j_tail = block;
 	write_unlock(&journal->j_state_lock);
+
+out:
+	return ret;
 }
 
 /*
@@ -1331,7 +1329,7 @@ static int journal_reset(journal_t *journal)
 	return jbd2_journal_start_thread(journal);
 }
 
-static void jbd2_write_superblock(journal_t *journal, int write_op)
+static int jbd2_write_superblock(journal_t *journal, int write_op)
 {
 	struct buffer_head *bh = journal->j_sb_buffer;
 	journal_superblock_t *sb = journal->j_superblock;
@@ -1370,7 +1368,10 @@ static void jbd2_write_superblock(journal_t *journal, int write_op)
 		printk(KERN_ERR "JBD2: Error %d detected when updating "
 		       "journal superblock for %s.\n", ret,
 		       journal->j_devname);
+		jbd2_journal_abort(journal, ret);
 	}
+
+	return ret;
 }
 
 /**
@@ -1383,10 +1384,11 @@ static void jbd2_write_superblock(journal_t *journal, int write_op)
  * 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,
+int 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;
+	int ret;
 
 	BUG_ON(!mutex_is_locked(&journal->j_checkpoint_mutex));
 	jbd_debug(1, "JBD2: updating superblock (start %lu, seq %u)\n",
@@ -1395,13 +1397,18 @@ void jbd2_journal_update_sb_log_tail(journal_t *journal, tid_t tail_tid,
 	sb->s_sequence = cpu_to_be32(tail_tid);
 	sb->s_start    = cpu_to_be32(tail_block);
 
-	jbd2_write_superblock(journal, write_op);
+	ret = jbd2_write_superblock(journal, write_op);
+	if (ret)
+		goto out;
 
 	/* 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);
+
+out:
+	return ret;
 }
 
 /**
@@ -1950,7 +1957,14 @@ int jbd2_journal_flush(journal_t *journal)
 		return -EIO;
 
 	mutex_lock(&journal->j_checkpoint_mutex);
-	jbd2_cleanup_journal_tail(journal);
+	if (!err) {
+		err = jbd2_cleanup_journal_tail(journal);
+		if (err < 0) {
+			mutex_unlock(&journal->j_checkpoint_mutex);
+			goto out;
+		}
+		err = 0;
+	}
 
 	/* Finally, mark the journal as really needing no recovery.
 	 * This sets s_start==0 in the underlying superblock, which is
@@ -1966,7 +1980,8 @@ int jbd2_journal_flush(journal_t *journal)
 	J_ASSERT(journal->j_head == journal->j_tail);
 	J_ASSERT(journal->j_tail_sequence == journal->j_transaction_sequence);
 	write_unlock(&journal->j_state_lock);
-	return 0;
+out:
+	return err;
 }
 
 /**
@@ -2330,7 +2345,7 @@ static int jbd2_journal_init_journal_head_cache(void)
 	jbd2_journal_head_cache = kmem_cache_create("jbd2_journal_head",
 				sizeof(struct journal_head),
 				0,		/* offset */
-				SLAB_TEMPORARY,	/* flags */
+				SLAB_TEMPORARY | SLAB_DESTROY_BY_RCU,
 				NULL);		/* ctor */
 	retval = 0;
 	if (!jbd2_journal_head_cache) {
@@ -2362,10 +2377,8 @@ static struct journal_head *journal_alloc_journal_head(void)
 	if (!ret) {
 		jbd_debug(1, "out of memory for journal_head\n");
 		pr_notice_ratelimited("ENOMEM in %s, retrying.\n", __func__);
-		while (!ret) {
-			yield();
-			ret = kmem_cache_zalloc(jbd2_journal_head_cache, GFP_NOFS);
-		}
+		ret = kmem_cache_zalloc(jbd2_journal_head_cache,
+				GFP_NOFS | __GFP_NOFAIL);
 	}
 	return ret;
 }