JFS: Improve sync barrier processing

Under heavy load, hot metadata pages are often locked by non-committed
transactions, making them difficult to flush to disk.  This prevents
the sync point from advancing past a transaction that had modified the
page.

There is a point during the sync barrier processing where all
outstanding transactions have been committed to disk, but no new
transaction have been allowed to proceed.  This is the best time
to write the metadata.

Signed-off-by: Dave Kleikamp <shaggy@austin.ibm.com>

1 files changed, 5 insertions, 5 deletions

diff --git a/fs/jfs/jfs_txnmgr.c b/fs/jfs/jfs_txnmgr.c
index 121c981ff453..3555acff12d0 100644
--- a/fs/jfs/jfs_txnmgr.c
+++ b/fs/jfs/jfs_txnmgr.c
@@ -552,6 +552,11 @@ void txEnd(tid_t tid)
                  * synchronize with logsync barrier
                  */
                 if (test_bit(log_SYNCBARRIER, &log->flag)) {
+                        TXN_UNLOCK();
+
+                        /* write dirty metadata & forward log syncpt */
+                        jfs_syncpt(log, 1);
+
                         jfs_info("log barrier off: 0x%x", log->lsn);
 
                         /* enable new transactions start */
@@ -560,11 +565,6 @@ void txEnd(tid_t tid)
                         /* wakeup all waitors for logsync barrier */
                         TXN_WAKEUP(&log->syncwait);
 
-                        TXN_UNLOCK();
-
-                        /* forward log syncpt */
-                        jfs_syncpt(log);
-
                         goto wakeup;
                 }
         }