[boost] Re: no semaphores in boost::thread

From: Alexander Terekhov (terekhov_at_[hidden])
Date: 2003-06-10 10:14:55

"Victor A. Wagner, Jr." wrote:
> 
> well, unless (likely given this biz) the words have changed meaning again,
> "naked semaphore" was described by Dijkstra way back when (1964??)

http://www.cs.utexas.edu/users/EWD/index01xx.html

But just a few years later (in 1971) Sir Tony Hoare (ironically, his 
"current" homepage is @ <http://www.research.microsoft.com/~thoare>)
wrote the following in the ACKNOWLEDGMENTS section of his paper named 
"TOWARDS A THEORY OF PARALLEL PROGRAMMING":

"...
 It will be obvious to all how much this paper owes to the thought 
 and writings of Professor E. W. Dijkstra, to whom I owe the concept
 of the critical region, and the semaphore, the deadly embracy and
 ...
 Less obvious but equally invaluable has been his constant 
 encouragement in the search for a concept to 'replace' the semaphore
 ..."             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

[...]
> btw, note that IF my definition is correct, then at a minimum a mutex must
> do more work than a semaphore (since it must establish "ownership" of the
> mutex).

That's needed for recursive or error-check mutexes and things like 
"priority inheritance" protocol to fight the problem of unbounded
priority inversion in the "realtime" appls. It isn't needed for the
"normal" mutexes with or without "priority protection" protocol.

regards,
alexander.