[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <20210622202217.3864743-1-jaegeuk@kernel.org>
Date: Tue, 22 Jun 2021 13:22:17 -0700
From: Jaegeuk Kim <jaegeuk@...nel.org>
To: linux-kernel@...r.kernel.org,
linux-f2fs-devel@...ts.sourceforge.net
Cc: Jaegeuk Kim <jaegeuk@...nel.org>
Subject: [PATCH] f2fs: remove false alarm on iget failure during GC
This patch removes setting SBI_NEED_FSCK when GC gets an error on f2fs_iget,
since f2fs_iget can give ENOMEM and others by race condition.
If we set this critical fsck flag, we'll get EIO during fsync via the below
code path.
In f2fs_inplace_write_data(),
if (is_sbi_flag_set(sbi, SBI_NEED_FSCK) || f2fs_cp_error(sbi)) {
err = -EIO;
goto drop_bio;
}
Fixes: 9557727876674 ("f2fs: drop inplace IO if fs status is abnormal")
Signed-off-by: Jaegeuk Kim <jaegeuk@...nel.org>
---
fs/f2fs/gc.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
index da5947b30142..0e42ee5f7770 100644
--- a/fs/f2fs/gc.c
+++ b/fs/f2fs/gc.c
@@ -1451,10 +1451,8 @@ static int gc_data_segment(struct f2fs_sb_info *sbi, struct f2fs_summary *sum,
if (phase == 3) {
inode = f2fs_iget(sb, dni.ino);
- if (IS_ERR(inode) || is_bad_inode(inode)) {
- set_sbi_flag(sbi, SBI_NEED_FSCK);
+ if (IS_ERR(inode) || is_bad_inode(inode))
continue;
- }
if (!down_write_trylock(
&F2FS_I(inode)->i_gc_rwsem[WRITE])) {
--
2.32.0.288.g62a8d224e6-goog
Powered by blists - more mailing lists