Issue 2699: Atomic multi-path locks (in FSFS)

I just added the following comment to issue #2699 in response to the problem
of trying to allow atomic locking of multiple paths:

   I wonder (aloud) ... since locks are sorta outside the flow of normal
   versioned data, would it make sense to use a different mutex file in
   FSFS for lock/unlock operations? That way lockers/unlockers could
   block other lockers/unlockers, but not committers.

It's not a well-baked idea, I admit. But can others offer any immediate
thoughts on the idea? Ignoring compat issues, are there glaring problems?

-- 
C. Michael Pilato <cmpilato@collab.net>
CollabNet   <>   www.collab.net   <>   Distributed Development On Demand