On Thu, Apr 2, 2020 at 5:23 PM David Howells <dhowells@xxxxxxxxxx> wrote: > > Miklos Szeredi <miklos@xxxxxxxxxx> wrote: > > > > > Not sure what you mean. I think it shows precisely the information > > > > Karel asked for. > > > > > > It's not atomic. > > > > Yes it is. > > No, it really isn't - though it could be made so. > > ext4_show_mount(), for example, doesn't lock against "mount -o remount", so > the configuration can be changing whilst it's being rendered to text. Does s_umount nest inside namespace_sem? I really don't see the relation of those locks. Thanks, Miklos
