Re: condvar performance in .59 vs .60

Ulrich Drepper wrote:
> > First variation of the test runs about 10 times longer on 60 
> > compared to 59. At the same time decrease in rate is less than
> > 3 times.
> 
> This is an extreme case and probably not a very realistic pattern.  The
> worst slowdown I've seen was, I think, 5%.  Yes, some things got slower
> because the kernel is now, on SMP machines, too efficient and releases
> the waiter so fast, that it runs into the blocked internal mutex.
> Resulting in more context switches etc.

Would a synchronous wakeup help?  Synchronous wakeup means that the
woken task doesn't preempt the waker, as it is assumed the waker will
sleep soon.  I'm not sure if this example is appropriate for a
synchronous wakeup, but the mechanism is there in the kernel if it's useful.

--Jamie