[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20160920025504.72524-2-jaegeuk@kernel.org>
Date: Mon, 19 Sep 2016 19:55:04 -0700
From: Jaegeuk Kim <jaegeuk@...nel.org>
To: linux-kernel@...r.kernel.org, linux-fsdevel@...r.kernel.org,
linux-f2fs-devel@...ts.sourceforge.net
Cc: Jaegeuk Kim <jaegeuk@...nel.org>
Subject: [PATCH 2/2] f2fs: put directory inodes before checkpoint in roll-forward recovery
Before checkpoint, we'd be better drop any inodes.
Signed-off-by: Jaegeuk Kim <jaegeuk@...nel.org>
---
fs/f2fs/recovery.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/fs/f2fs/recovery.c b/fs/f2fs/recovery.c
index 2b8a56d..509273a 100644
--- a/fs/f2fs/recovery.c
+++ b/fs/f2fs/recovery.c
@@ -630,6 +630,9 @@ out:
set_ckpt_flags(sbi->ckpt, CP_ERROR_FLAG);
mutex_unlock(&sbi->cp_mutex);
+ /* let's drop all the directory inodes for clean checkpoint */
+ destroy_fsync_dnodes(&dir_list);
+
if (!err && need_writecp) {
struct cp_control cpc = {
.reason = CP_RECOVERY,
@@ -637,7 +640,6 @@ out:
err = write_checkpoint(sbi, &cpc);
}
- destroy_fsync_dnodes(&dir_list);
kmem_cache_destroy(fsync_entry_slab);
return ret ? ret: err;
}
--
2.8.3
Powered by blists - more mailing lists