Re: hackbench regression with kernel 2.6.32-rc1

[Peter Zijlstra <a.p.zijlstra@chello.nl>]

On Mon, 2009-10-12 at 15:05 +0800, Zhang, Yanmin wrote:

> > So hackbench is a multi-cast, with one sender spraying multiple
> > receivers, who in their turn don't spray back, right?

> Right. volanoMark has about 9% regression on stoakley and 50% regression
> on tigerton. If I revert the original patches, volanoMark regression on stoakley
> disappears, but still has about 45% on tigerton.

> > /me ponders a bit
> > 
> > Does this make it any better?

> I apply this patch and another one you sent on tbench email thread.
> On stoakley, hackbench is recovered. If reverting the original 2 patches,
> we get 8% improvement.
> On tigerton, with your 2 patches, there is still about 45% regression.

[ and here I got confused because this 45% seemed to match the 45%
above, but then I saw it was hackbench vs volano ]

> As for volanoMark, with your 2 patches, regression disappears on staokley
> and it becomes about 35% on tigerton.

So hackbench on tigerton is worse, but volano on tigerton is better with
this patch vs reverting bits?

> The good news is only tbench has about 6% regression on Nehalem machines.
> Other regressions such like hackbench/aim7/volanoMark is not clear/big on
> Nehalem. But reverting the original 2 patches don't fix the tbench regression
> on Nehalem machines.

Right, so Mike's suggestion of doing: 
  echo NEXT_BUDDY > /debug/sched_features

Seems like the next thing to try..

Mike, did we ever figure out _why_ NEXT_BUDDY introduced latencies?

Buddies shouldn't make latencies worse than regular while(1); loops
would.