lkml.org 
[lkml]   [2014]   [Aug]   [15]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
From
Subject[PATCH 2/4] f2fs: prevent checkpoint during roll-forward
Date
Any checkpoint should not be done during the core roll-forward procedure.
Especially, it includes error cases too.

Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
fs/f2fs/recovery.c | 8 ++++++++
1 file changed, 8 insertions(+)

diff --git a/fs/f2fs/recovery.c b/fs/f2fs/recovery.c
index 7ca7aad..d36ef35 100644
--- a/fs/f2fs/recovery.c
+++ b/fs/f2fs/recovery.c
@@ -459,6 +459,9 @@ int recover_fsync_data(struct f2fs_sb_info *sbi)
/* step #1: find fsynced inode numbers */
sbi->por_doing = true;

+ /* prevent checkpoint */
+ mutex_lock(&sbi->cp_mutex);
+
blkaddr = NEXT_FREE_BLKADDR(sbi, curseg);

err = find_fsync_dnodes(sbi, &inode_list);
@@ -490,8 +493,13 @@ out:
/* Flush all the NAT/SIT pages */
while (get_pages(sbi, F2FS_DIRTY_META))
sync_meta_pages(sbi, META, LONG_MAX);
+ set_ckpt_flags(sbi->ckpt, CP_ERROR_FLAG);
+ mutex_unlock(&sbi->cp_mutex);
} else if (need_writecp) {
+ mutex_unlock(&sbi->cp_mutex);
write_checkpoint(sbi, false);
+ } else {
+ mutex_unlock(&sbi->cp_mutex);
}
return err;
}
--
1.8.5.2 (Apple Git-48)


\
 
 \ /
  Last update: 2014-08-16 00:21    [W:0.206 / U:0.684 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site