Subj : Re: Memory visibility and MS Interlocked instructions
To   : comp.programming.threads
From : Alexander Terekhov
Date : Thu Aug 25 2005 10:42 pm


Scott Meyers wrote:
[...]
> May I assume then that my understanding is correct that both writer and
> reader must participate in a handshake to ensure that writer changes to
> memory are visible to readers in a relaxed memory architecture?  

Not really (if you mean assembly level and/or atomic<>).

>                                                                 If so,
> would it be reasonable to conclude that the semantics of the interlocked
> instructions are not generally implementable, though they may be
> implementable on particular architectures?

But MS interlocked stuff is brain-dead anyway.

regards,
alexander.

.