On Tue, 2019-10-29 at 14:39 +0800, Ian Kent wrote: > > After working on this patch I'm even more inclined to let the kernel > do it's propagation thing and set the autofs mounts, either silently > by default or explicitly by map entry option. > > Because it's the propagation setting of the parent mount that > controls > the propagation of its children there shouldn't be any chance of a > race so this should be reliable. > > Anyway, here is a patch, compile tested only, and without the > changelog > hunk I normally add to save you possible conflicts. But unless there > are any problems found this is what I will eventually commit to the > repo. > > If there are any changes your not aware of I'll let you know. > > Clearly this depends on the other two related patches for this issue. Oh, forget to mention, this is compile tested only so far. Please let me know how it goes. Ian
