Skip to contentSkip to conversation

nilfs2: fix use-after-free bug of ns_writer on remount

PATCH In review Ryusuke Konishi 2 participants 2 messages Nov 21, 2022 – Nov 22, 2022 Stable branch

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 nilfs2: fix use-after-free bug of ns_writer on remount

commit 8cccf05fe857a18ee26e20d11a8455a73ffd4efd upstream.

If a nilfs2 filesystem is downgraded to read-only due to metadata corruption on disk and is remounted read/write, or if emergency read-only remount is performed, detaching a log writer and synchronizing the filesystem can be done at the same time.

In these cases, use-after-free of the log writer (hereinafter nilfs->ns_writer) can happen as shown in the scenario below:

Task1 Task2
-------------------------------- ------------------------------
nilfs_construct_segment
nilfs_segctor_sync
init_wait
init_waitqueue_entry
add_wait_queue
schedule
nilfs_remount (R/W remount case)
nilfs_attach_log_writer
nilfs_detach_log_writer
nilfs_segctor_destroy
kfree
finish_wait
_raw_spin_lock_irqsave
__raw_spin_lock_irqsave
do_raw_spin_lock
debug_spin_lock_before <-- use-after-free

While Task1 is sleeping, nilfs->ns_writer is freed by Task2. After Task1 waked up, Task1 accesses nilfs->ns_writer which is already freed. This scenario diagram is based on the Shigeru Yoshida's post [1].

This patch fixes the issue by not detaching nilfs->ns_writer on remount so that this UAF race doesn't happen. Along with this change, this patch also inserts a few necessary read-only checks with superblock instance where only the ns_writer pointer was used to check if the filesystem is read-only.

---
Please apply this patch to stable-4.9 instead of the patch that could not be applied to it last time.

A rejected hunk has been manually resolved, and replaced sb_rdonly() with its equivalent bitwise operation since the function does not yet exist in this kernel. Also, retested against that stable tree.

 fs/nilfs2/segment.c | 15 ++++++++-------
 fs/nilfs2/super.c   |  2 --
 2 files changed, 8 insertions(+), 9 deletions(-)
fs/nilfs2/segment.c +8 -7
index 450bfdee513b..86769ecfe61f 100644
@@ -329,7 +329,7 @@ void nilfs_relax_pressure_in_lock(struct super_block *sb)
 	struct the_nilfs *nilfs = sb->s_fs_info;
 	struct nilfs_sc_info *sci = nilfs->ns_writer;
 
-	if (!sci || !sci->sc_flush_request)
+	if (sb->s_flags & MS_RDONLY || unlikely(!sci) || !sci->sc_flush_request)
 		return;
 
 	set_bit(NILFS_SC_PRIOR_FLUSH, &sci->sc_flags);
@@ -2252,7 +2252,7 @@ int nilfs_construct_segment(struct super_block *sb)
 	struct nilfs_transaction_info *ti;
 	int err;
 
-	if (!sci)
+	if (sb->s_flags & MS_RDONLY || unlikely(!sci))
 		return -EROFS;
 
 	/* A call inside transactions causes a deadlock. */
@@ -2291,7 +2291,7 @@ int nilfs_construct_dsync_segment(struct super_block *sb, struct inode *inode,
 	struct nilfs_transaction_info ti;
 	int err = 0;
 
-	if (!sci)
+	if (sb->s_flags & MS_RDONLY || unlikely(!sci))
 		return -EROFS;
 
 	nilfs_transaction_lock(sb, &ti, 0);
@@ -2788,11 +2788,12 @@ int nilfs_attach_log_writer(struct super_block *sb, struct nilfs_root *root)
 
 	if (nilfs->ns_writer) {
 		/*
-		 * This happens if the filesystem was remounted
-		 * read/write after nilfs_error degenerated it into a
-		 * read-only mount.
+		 * This happens if the filesystem is made read-only by
+		 * __nilfs_error or nilfs_remount and then remounted
+		 * read/write.  In these cases, reuse the existing
+		 * writer.
 		 */
-		nilfs_detach_log_writer(sb);
+		return 0;
 	}
 
 	nilfs->ns_writer = nilfs_segctor_new(sb, root);
fs/nilfs2/super.c +0 -2
index c95d369e90aa..9e8f186b14d1 100644
@@ -1147,8 +1147,6 @@ static int nilfs_remount(struct super_block *sb, int *flags, char *data)
 	if ((*flags & MS_RDONLY) == (sb->s_flags & MS_RDONLY))
 		goto out;
 	if (*flags & MS_RDONLY) {
-		/* Shutting down log writer */
-		nilfs_detach_log_writer(sb);
 		sb->s_flags |= MS_RDONLY;
 
 		/*
Signature · 1 line
2.31.1
#1 lore raw
Quoted 48 lines · Ryusuke Konishi

commit 8cccf05fe857a18ee26e20d11a8455a73ffd4efd upstream.

If a nilfs2 filesystem is downgraded to read-only due to metadata corruption on disk and is remounted read/write, or if emergency read-only remount is performed, detaching a log writer and synchronizing the filesystem can be done at the same time.

In these cases, use-after-free of the log writer (hereinafter nilfs->ns_writer) can happen as shown in the scenario below:

Task1 Task2
-------------------------------- ------------------------------
nilfs_construct_segment
nilfs_segctor_sync
init_wait
init_waitqueue_entry
add_wait_queue
schedule
nilfs_remount (R/W remount case)
nilfs_attach_log_writer
nilfs_detach_log_writer
nilfs_segctor_destroy
kfree
finish_wait
_raw_spin_lock_irqsave
__raw_spin_lock_irqsave
do_raw_spin_lock
debug_spin_lock_before <-- use-after-free

While Task1 is sleeping, nilfs->ns_writer is freed by Task2. After Task1 waked up, Task1 accesses nilfs->ns_writer which is already freed. This scenario diagram is based on the Shigeru Yoshida's post [1].

This patch fixes the issue by not detaching nilfs->ns_writer on remount so that this UAF race doesn't happen. Along with this change, this patch also inserts a few necessary read-only checks with superblock instance where only the ns_writer pointer was used to check if the filesystem is read-only.

Quoted 7 lines

---
Please apply this patch to stable-4.9 instead of the patch that could not be applied to it last time.

A rejected hunk has been manually resolved, and replaced sb_rdonly() with its equivalent bitwise operation since the function does not yet exist in this kernel. Also, retested against that stable tree.

Now queued up, thanks.

greg k-h