Re: [PATCH] fs/ext4: fix ABBA deadlock in ext4_rmdir()

[Baokun Li]

On 2026/8/27 20:50, Sergey Senozhatsky wrote:
> We observe the following ABBA deadlock (casefolding enabled):
>
>  INFO: task jbd2/dm-13-8:447 blocked for more than 720 seconds.
>  task:jbd2/dm-13-8    state:D stack:0     pid:447   tgid:447   ppid:2
>  Call Trace:
>   <TASK>
>   __schedule+0xc2c/0xe60
>   schedule+0x40/0xe0
>   jbd2_journal_wait_updates+0x8f/0xf0
>   jbd2_journal_commit_transaction+0x321/0x1850
>   kjournald2+0xa9/0x230
>   kthread+0x226/0x2a0
>   ret_from_fork+0x101/0x1e0
>   ret_from_fork_asm+0x1a/0x30
>
>  INFO: task Thread-10:4161 blocked for more than 720 seconds.
>  task:Thread-10       state:D stack:0     pid:4161  tgid:3848  ppid:540
>  Call Trace:
>   <TASK>
>   __schedule+0xc2c/0xe60
>   schedule+0x40/0xe0
>   wait_transaction_locked+0x8c/0xd0
>   start_this_handle+0x252/0x870
>   jbd2__journal_start+0x120/0x280
>   __ext4_journal_start_sb+0x11a/0x250
>   ext4_evict_inode+0x208/0x760
>   iput+0x222/0x5c0
>   dput+0x293/0x690
>   ____fput+0x145/0x2b0
>   task_work_run+0x7a/0xb0
>   exit_to_user_mode_loop+0xc0/0xd0
>   do_syscall_64+0x14b/0xf10
>   entry_SYSCALL_64_after_hwframe+0x76/0x7e
>
>  NMI backtrace for cpu 0
>  CPU: 0 UID: 1010216 PID: 4160 Comm: Thread-7
>  RIP: 0010:d_walk+0x0/0x290
> [..]
>  Call Trace:
>  <TASK>
>   shrink_dcache_parent+0xb2/0x100
>   d_invalidate+0x50/0x110
>   ext4_rmdir+0x3a2/0x3d0
>   vfs_rmdir+0x9d/0x1d0
>   do_rmdir+0xf0/0x330
>   __x64_sys_unlinkat+0x34/0x50
>   do_syscall_64+0x61/0xf10
>
> The problem is that ext4_rmdir() calls d_invalidate() before stopping
> the jbd2 transaction handle.  Because d_invalidate() traverses child
> dentries it can encounter dying or in-use dentries.  If a concurrent
> thread executing __dentry_kill() on a child dentry invokes ext4_evict_inode(),
> it requests a new transaction handle via ext4_journal_start(), which
> waits for all running handles to close.  However, the thread executing
> ext4_rmdir() still holds its active transaction handle while looping
> in shrink_dcache_parent() waiting for the dying child dentry to complete
> eviction.
>
> Stop transaction handle in ext4_rmdir() before calling d_invalidate().
>
> Fixes: b886ee3e778e ("ext4: Support case-insensitive file name lookups")
> Co-developed-by: Sarthak Kukreti <sarthakkukreti@xxxxxxxxxx>
> Signed-off-by: Sarthak Kukreti <sarthakkukreti@xxxxxxxxxx>
> Signed-off-by: Sergey Senozhatsky <senozhatsky@xxxxxxxxxxxx>

Looks good!

Reviewed-by: Baokun Li <libaokun@xxxxxxxxxxxxxxxxx>

> ---
>  fs/ext4/namei.c | 11 ++++++-----
>  1 file changed, 6 insertions(+), 5 deletions(-)
>
> diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c
> index a6386c1d237f..231a274b814e 100644
> --- a/fs/ext4/namei.c
> +++ b/fs/ext4/namei.c
> @@ -3209,19 +3209,20 @@ static int ext4_rmdir(struct inode *dir, struct dentry *dentry)
>  	ext4_fc_track_unlink(handle, dentry);
>  	retval = ext4_mark_inode_dirty(handle, dir);
>  
> +end_rmdir:
> +	brelse(bh);
> +	if (handle)
> +		ext4_journal_stop(handle);
> +
>  	/* VFS negative dentries are incompatible with Encoding and
>  	 * Case-insensitiveness. Eventually we'll want avoid
>  	 * invalidating the dentries here, alongside with returning the
>  	 * negative dentries at ext4_lookup(), when it is better
>  	 * supported by the VFS for the CI case.
>  	 */
> -	if (IS_ENABLED(CONFIG_UNICODE) && IS_CASEFOLDED(dir))
> +	if (!retval && IS_ENABLED(CONFIG_UNICODE) && IS_CASEFOLDED(dir))
>  		d_invalidate(dentry);
>  
> -end_rmdir:
> -	brelse(bh);
> -	if (handle)
> -		ext4_journal_stop(handle);
>  	return retval;
>  }
>