Subj : Re: posix and lock-free algorithms
To   : comp.programming.threads
From : David Schwartz
Date : Sat Aug 13 2005 02:42 am


"David Hopwood" <david.nospam.hopwood@blueyonder.co.uk> wrote in message 
news:0zbLe.1363$P.391@fe3.news.blueyonder.co.uk...

> Right, that's part of it. DS also seems to be implicitly assuming that
> "synchronizations of memory" are totally ordered in global time. Without
> further specification of the memory model, there's no basis for that
> assumption.

    What else would it mean to "synchronize memory" with *no* caveats?

    DS

.