From: Alexander Terekhov (terekhov_at_[hidden])
Date: 2003-06-04 16:34:19
Stefan Seefeld wrote:
>
> Alexander Terekhov wrote:
>
> > It is showing that semas (e.g. bin-semas aka "auto-reset events")
> > are really error-prone.
>
> you seem to equate microsoft's implementation of semaphores with
> the concept of semaphores (which is what I'd like to get feedback on).
No. I'm talking about the erroneous USE of a binary semaphore in
the Microsoft implementation of "metered section" silliness (which,
"conceptually" is nothing but a counting semaphore).
>
> If all that is wrong is that microsoft does a crappy job at implementing
> them, the response could be to provide a special implementation using
> mutexes and cv's *for the MS platforms*, and using native
> implementations when possible.
You don't need semaphores; neither binary nor counting semas are
needed for *threading*. Use mutexes for locking and condvars for
waiting. Modern semas are meant for things that need either async-
signal-safe "unlock" operation or memory isolation (no shared mem).
Threading has really nothing to do with that.
regards,
alexander.