On Mar 13, 2007, at 8:53 AM, Brian Ginsbach wrote:
Yah, that's basically what I'm doing. I just don't want to be forced
to return the old value if the swap does not occur, nor do I
particularly want to return the old value at all (partially because it
means more instructions in the inner-loop of most load-locked/store-
conditional type systems).
-- thorpej