On Mon, Jun 16, 2025 at 08:14:58PM +0100, Al Viro wrote: > On Mon, Jun 16, 2025 at 04:40:14PM +0200, Christian Brauner wrote: > > On Sat, Jun 14, 2025 at 07:02:25AM +0100, Al Viro wrote: > > > ... and fix an old deadlock on spufs_mkdir() failures to populate > > > subdirectory - spufs_rmdir() had always been taking lock on the > > > victim, so doing it while the victim is locked is a bad idea. > > > > > > Signed-off-by: Al Viro <viro@xxxxxxxxxxxxxxxxxx> > > > --- > > > > Fwiw, I think simple_recursive_removal_locked() might be better. > > It's longer and arguably uglier but it clearer communicates that its the > > same helper as simple_recursive_removal() just with the assumption that > > the caller already holds the lock. > > Not sure... TBH, I'm somewhat tempted to rename simple_recursive_removal() > to simple_remove()... Sounds good.
