On Thu, Jul 04, 2002 at 01:06:35PM +1200, Gregory McGarry wrote:

> I implemented restartable atomic sequences (RAS) to compare their
> performance with atomic test-and-set instructions.

Cool!

> Since I don't have an R4000+ cpu to compare the performance, I tested
> on a DX4 (i386).

Is you test code available?  I can try it on a variety of CPUs that have
ll/sc.


> The only other interesting fact is that during the 3.8 seconds
> and 10E6 locks, 52 restarts were performed.  Not much statistical
> variation either; suprisingly determinstic in fact.

Related in some way to the number of clock interrupts?  That's only
13-odd HZ, but the only thing that comes to mind...

Simon.
--
Simon Burge                                   <simonb@wasabisystems.com>
NetBSD Development, Support and Service:   http://www.wasabisystems.com/