jbd2: optimize jbd2_log_do_checkpoint() a bit

When we discover written out buffer in transaction checkpoint list we
don't have to recheck validity of a transaction. Either this is the
last buffer in a transaction - and then we are done - or this isn't
and then we can just take another buffer from the checkpoint list
without dropping j_list_lock.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>

diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c
index 90d6091d7e187e2b77c6776764f8df9e2ef7560b..9ffb19cf376b47cc57eafd38b67a1d534048827f 100644 (file)
--- a/fs/jbd2/checkpoint.c
+++ b/fs/jbd2/checkpoint.c
@@ -285,9 +285,10 @@ restart:
                        if (unlikely(buffer_write_io_error(bh)) && !result)
                                result = -EIO;
                        BUFFER_TRACE(bh, "remove from checkpoint");
-                       __jbd2_journal_remove_checkpoint(jh);
-                       spin_unlock(&journal->j_list_lock);
-                       goto retry;
+                       if (__jbd2_journal_remove_checkpoint(jh))
+                               /* The transaction was released; we're done */
+                               goto out;
+                       continue;
                }
                /*
                 * Important: we are about to write the buffer, and