[f2fs-dev] [PATCH 1/2] f2fs: fix return value of f2fs_recover_fsync_data()

[f2fs-dev] [PATCH 1/2] f2fs: fix return value of f2fs_recover_fsync_data()

[Chao Yu via Linux-f2fs-devel]

With below scripts, it will trigger panic in f2fs:

mkfs.f2fs -f /dev/vdd
mount /dev/vdd /mnt/f2fs
touch /mnt/f2fs/foo
sync
echo 111 >> /mnt/f2fs/foo
f2fs_io fsync /mnt/f2fs/foo
f2fs_io shutdown 2 /mnt/f2fs
umount /mnt/f2fs
mount -o ro,norecovery /dev/vdd /mnt/f2fs
or
mount -o ro,disable_roll_forward /dev/vdd /mnt/f2fs

F2FS-fs (vdd): f2fs_recover_fsync_data: recovery fsync data, check_only: 0
F2FS-fs (vdd): Mounted with checkpoint version = 7f5c361f
F2FS-fs (vdd): Stopped filesystem due to reason: 0
F2FS-fs (vdd): f2fs_recover_fsync_data: recovery fsync data, check_only: 1
Filesystem f2fs get_tree() didn't set fc->root, returned 1
------------[ cut here ]------------
kernel BUG at fs/super.c:1761!
Oops: invalid opcode: 0000 [#1] SMP PTI
CPU: 3 UID: 0 PID: 722 Comm: mount Not tainted 6.18.0-rc2+ #721 PREEMPT(voluntary)
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.3-debian-1.16.3-2 04/01/2014
RIP: 0010:vfs_get_tree.cold+0x18/0x1a
Call Trace:
 <TASK>
 fc_mount+0x13/0xa0
 path_mount+0x34e/0xc50
 __x64_sys_mount+0x121/0x150
 do_syscall_64+0x84/0x800
 entry_SYSCALL_64_after_hwframe+0x76/0x7e
RIP: 0033:0x7fa6cc126cfe

The root cause is we missed to handle error number returned from
f2fs_recover_fsync_data() when mounting image w/ ro,norecovery or
ro,disable_roll_forward mount option, result in returning a positive
error number to vfs_get_tree(), fix it.

Cc: stable@kernel.org
Fixes: 6781eabba1bd ("f2fs: give -EINVAL for norecovery and rw mount")
Signed-off-by: Chao Yu <chao@kernel.org>
---
 fs/f2fs/super.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
index 0b0ef8ba243b..8cf98c40b160 100644
--- a/fs/f2fs/super.c
+++ b/fs/f2fs/super.c
@@ -5211,11 +5211,15 @@ static int f2fs_fill_super(struct super_block *sb, struct fs_context *fc)
 		}
 	} else {
 		err = f2fs_recover_fsync_data(sbi, true);
-
-		if (!f2fs_readonly(sb) && err > 0) {
-			err = -EINVAL;
-			f2fs_err(sbi, "Need to recover fsync data");
-			goto free_meta;
+		if (err > 0) {
+			if (!f2fs_readonly(sb)) {
+				f2fs_err(sbi, "Need to recover fsync data");
+				err = -EINVAL;
+				goto free_meta;
+			} else {
+				f2fs_info(sbi, "drop all fsynced data");
+				err = 0;
+			}
 		}
 	}
 
-- 
2.49.0



_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

[f2fs-dev] [PATCH 2/2] f2fs: fix to detect recoverable inode during dryrun of find_fsync_dnodes()

[Chao Yu via Linux-f2fs-devel]

mkfs.f2fs -f /dev/vdd
mount /dev/vdd /mnt/f2fs
touch /mnt/f2fs/foo
sync		# avoid CP_UMOUNT_FLAG in last f2fs_checkpoint.ckpt_flags
touch /mnt/f2fs/bar
f2fs_io fsync /mnt/f2fs/bar
f2fs_io shutdown 2 /mnt/f2fs
umount /mnt/f2fs
blockdev --setro /dev/vdd
mount /dev/vdd /mnt/f2fs
mount: /mnt/f2fs: WARNING: source write-protected, mounted read-only.

For the case if we create and fsync a new inode before sudden power-cut,
without norecovery or disable_roll_forward mount option, the following
mount will succeed w/o recovering last fsynced inode.

The problem here is that we only check inode_list list after
find_fsync_dnodes() in f2fs_recover_fsync_data() to find out whether
there is recoverable data in the iamge, but there is a missed case, if
last fsynced inode is not existing in last checkpoint, then, we will
fail to get its inode due to nat of inode node is not existing in last
checkpoint, so the inode won't be linked in inode_list.

Let's detect such case in dyrun mode to fix this issue.

After this change, mount will fail as expected below:
mount: /mnt/f2fs: cannot mount /dev/vdd read-only.
       dmesg(1) may have more information after failed mount system call.
demsg:
F2FS-fs (vdd): Need to recover fsync data, but write access unavailable, please try mount w/ disable_roll_forward or norecovery

Cc: stable@kernel.org
Fixes: 6781eabba1bd ("f2fs: give -EINVAL for norecovery and rw mount")
Signed-off-by: Chao Yu <chao@kernel.org>
---
 fs/f2fs/recovery.c | 20 ++++++++++++--------
 1 file changed, 12 insertions(+), 8 deletions(-)

diff --git a/fs/f2fs/recovery.c b/fs/f2fs/recovery.c
index 215e442db72c..d7faebaa3c6b 100644
--- a/fs/f2fs/recovery.c
+++ b/fs/f2fs/recovery.c
@@ -399,7 +399,7 @@ static int sanity_check_node_chain(struct f2fs_sb_info *sbi, block_t blkaddr,
 }
 
 static int find_fsync_dnodes(struct f2fs_sb_info *sbi, struct list_head *head,
-				bool check_only)
+				bool check_only, bool *new_inode)
 {
 	struct curseg_info *curseg;
 	block_t blkaddr, blkaddr_fast;
@@ -447,16 +447,19 @@ static int find_fsync_dnodes(struct f2fs_sb_info *sbi, struct list_head *head,
 				quota_inode = true;
 			}
 
-			/*
-			 * CP | dnode(F) | inode(DF)
-			 * For this case, we should not give up now.
-			 */
 			entry = add_fsync_inode(sbi, head, ino_of_node(folio),
 								quota_inode);
 			if (IS_ERR(entry)) {
 				err = PTR_ERR(entry);
-				if (err == -ENOENT)
+				/*
+				 * CP | dnode(F) | inode(DF)
+				 * For this case, we should not give up now.
+				 */
+				if (err == -ENOENT) {
+					if (check_only)
+						*new_inode = true;
 					goto next;
+				}
 				f2fs_folio_put(folio, true);
 				break;
 			}
@@ -875,6 +878,7 @@ int f2fs_recover_fsync_data(struct f2fs_sb_info *sbi, bool check_only)
 	int ret = 0;
 	unsigned long s_flags = sbi->sb->s_flags;
 	bool need_writecp = false;
+	bool new_inode = false;
 
 	f2fs_notice(sbi, "f2fs_recover_fsync_data: recovery fsync data, "
 					"check_only: %d", check_only);
@@ -890,8 +894,8 @@ int f2fs_recover_fsync_data(struct f2fs_sb_info *sbi, bool check_only)
 	f2fs_down_write(&sbi->cp_global_sem);
 
 	/* step #1: find fsynced inode numbers */
-	err = find_fsync_dnodes(sbi, &inode_list, check_only);
-	if (err || list_empty(&inode_list))
+	err = find_fsync_dnodes(sbi, &inode_list, check_only, &new_inode);
+	if (err < 0 || (list_empty(&inode_list) && (!check_only || !new_inode)))
 		goto skip;
 
 	if (check_only) {
-- 
2.49.0



_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

Re: [f2fs-dev] [PATCH 1/2] f2fs: fix return value of f2fs_recover_fsync_data()

[patchwork-bot+f2fs--- via Linux-f2fs-devel]

Hello:

This series was applied to jaegeuk/f2fs.git (dev)
by Jaegeuk Kim <jaegeuk@kernel.org>:

On Wed,  5 Nov 2025 14:50:22 +0800 you wrote:
> With below scripts, it will trigger panic in f2fs:
> 
> mkfs.f2fs -f /dev/vdd
> mount /dev/vdd /mnt/f2fs
> touch /mnt/f2fs/foo
> sync
> echo 111 >> /mnt/f2fs/foo
> f2fs_io fsync /mnt/f2fs/foo
> f2fs_io shutdown 2 /mnt/f2fs
> umount /mnt/f2fs
> mount -o ro,norecovery /dev/vdd /mnt/f2fs
> or
> mount -o ro,disable_roll_forward /dev/vdd /mnt/f2fs
> 
> [...]

Here is the summary with links:
  - [f2fs-dev,1/2] f2fs: fix return value of f2fs_recover_fsync_data()
    https://git.kernel.org/jaegeuk/f2fs/c/d05faa07e7da
  - [f2fs-dev,2/2] f2fs: fix to detect recoverable inode during dryrun of find_fsync_dnodes()
    https://git.kernel.org/jaegeuk/f2fs/c/ca244e9242a1

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html




_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel