Is there any interest in providing a query lock to go with lock and unlock?

[David Woodhouse]

rjb at ayrnetworks.com said:
> We have a parochial need which I have met in a heavy handed way by
> adding a mtd vector and an ioctl.

That makes sense. Change the name to something like test_lock() or 
is_locked(), and perhaps declare that it shall only take an 'offset' 
argument which must be the start of a block, rather than doing all the 
range stuff -- the users can do that themselves if they need it, can't they?

--
dwmw2