On 06/06/2009, Rene Gollent <anevilyak@xxxxxxxxx> wrote: > If someone > wants to jump in and has the requisite hardware, they're more than > welcome to. Actually, I'd discourage making larger changes to either codebase ("simple" or "affine"). I mean, Real Life (tm) did rear its ugly head, but I'm close to getting my game back and it would be a waste of energy to have someone else work on the scheduler. But of course competition is healthy and everything, and I'm not territorial about it. It's just that there are plenty of areas needing the manpower, and I'm already committed to the scheduler, have made a lot of research, a lot of experiments, have attacked the problems from numerous approaches, and so on. > By and large I'd be somewhat surprised if the results > differed greatly from scheduler_simple in this case though, I suspect > there's probably lock contention issues outside of the scheduler > somewhere that are causing most of the problems. Well, either "affine" is uncovering problems that don't manifest themselves under "simple", or it has caused some of its own. Still, when it does work, I like it better than "simple". Cheers, A. -- One last piece of advice: "ice".