On Wed, 28 Jul 1999 23:46:13 -0400 (EDT) 
 "Gordon W. Ross" <gwr@mediaone.net> wrote:

 > Why invent a new API for the mutex locks?

I'm not.  I'm using the same one that's been in BSD since it incorporated
Mach VM, and which was further refined in 4.4BSD-Lite2.  My message merely
defined the machine-dependent atomic goop that simple_lock(), etc. need to
function.

 > Using mutex_t, mutex_lock(), mutex_unlock(), mutex_trylock, etc.
 > in the kernel seems easier to document, at least...
 > (and otherwise apparently has identical semantics)

        -- Jason R. Thorpe <thorpej@nas.nasa.gov>