From patchwork Wed Jan 30 05:39:28 2013 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Subject: [1/2] jbd2: don't wake kjournald unnecessarily Date: Tue, 29 Jan 2013 19:39:28 -0000 From: Theodore Ts'o X-Patchwork-Id: 216768 Message-Id: <1359524369-31234-1-git-send-email-tytso@mit.edu> To: Ext4 Developers List Cc: Theodore Ts'o , Eric Sandeen , Sedat Dilek , mszeredi@suse.cz Don't send an extra wakeup to kjournald in the case where we already have the proper target in j_commit_request, i.e. that transaction has already been requested for commit. commit deeeaf13 "jbd2: fix fsync() tid wraparound bug" changed the logic leading to a wakeup, but it caused some extra wakeups which were found to lead to a measurable performance regression. Reported-by: Eric Sandeen Signed-off-by: "Theodore Ts'o" Cc: Eric Sandeen Cc: Sedat Dilek Cc: mszeredi@suse.cz --- Eric, I think this might be clearer than the patch which you proposed; what do you think? fs/jbd2/journal.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index dbf41f9..1a80e31 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -513,6 +513,10 @@ int __jbd2_log_space_left(journal_t *journal) */ int __jbd2_log_start_commit(journal_t *journal, tid_t target) { + /* Return if the txn has already requested to be committed */ + if (journal->j_commit_request == target) + return 0; + /* * The only transaction we can possibly wait upon is the * currently running transaction (if it exists). Otherwise,