Patrick Baggett wrote:
The atomic operations themselves aren't "performance critical", but they are
> used within performance critical code -- after all, if mutexes were all that, > lockless primitives would be obsolete. I can't see them being a bottleneck, > really, but a bit extra performance at the cost of someone else's time isn't > such a bad deal, especially when they aren't getting paid for it. ;)
I'm not suggesting the work is useless; I'm suggesting what might be the reason that a less-naive version hasn't been yet provided. By all means, improve away.
_______________________________________________ opensolaris-code mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/opensolaris-code
