[PATCH 1/2] f2fs: fix leakage of inmemory atomic pages

[PATCH 1/2] f2fs: fix leakage of inmemory atomic pages

[Jaegeuk Kim]

If we got failure during commit_atomic_write, abort_volatile_write will be
called, but will not drop the inmemory pages due to no FI_ATOMIC_FILE.
Actually, there is no reason to check the flag in abort_volatile_write.

Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/file.c | 10 +++-------
 1 file changed, 3 insertions(+), 7 deletions(-)

diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
index 91c51a6..a197215 100644
--- a/fs/f2fs/file.c
+++ b/fs/f2fs/file.c
@@ -1443,13 +1443,9 @@ static int f2fs_ioc_abort_volatile_write(struct file *filp)
 
 	f2fs_balance_fs(F2FS_I_SB(inode));
 
-	if (f2fs_is_atomic_file(inode)) {
-		clear_inode_flag(F2FS_I(inode), FI_ATOMIC_FILE);
-		commit_inmem_pages(inode, true);
-	}
-
-	if (f2fs_is_volatile_file(inode))
-		clear_inode_flag(F2FS_I(inode), FI_VOLATILE_FILE);
+	clear_inode_flag(F2FS_I(inode), FI_ATOMIC_FILE);
+	clear_inode_flag(F2FS_I(inode), FI_VOLATILE_FILE);
+	commit_inmem_pages(inode, true);
 
 	mnt_drop_write_file(filp);
 	return ret;
-- 
2.1.1


------------------------------------------------------------------------------

[PATCH 2/2] f2fs: don't need to submit bio on error case

[Jaegeuk Kim]

If commit_atomic_write is failed, we don't need to submit any bio.

Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/segment.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index 7835e41..7efd96ad 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -249,11 +249,11 @@ int commit_inmem_pages(struct inode *inode, bool abort)
 				trace_f2fs_commit_inmem_page(cur->page, INMEM);
 				fio.page = cur->page;
 				err = do_write_data_page(&fio);
-				submit_bio = true;
 				if (err) {
 					unlock_page(cur->page);
 					break;
 				}
+				submit_bio = true;
 			}
 		} else {
 			trace_f2fs_commit_inmem_page(cur->page, INMEM_DROP);
-- 
2.1.1


------------------------------------------------------------------------------