RFR: Use peak occupancy to optimize/pessimize free_threshold in adaptive heuristics

[Aleksey Shipilev]

On 05/04/2017 07:46 PM, Roman Kennke wrote:
> Am 04.05.2017 um 19:41 schrieb Aleksey Shipilev:
>> On 05/04/2017 07:38 PM, Roman Kennke wrote:
>> Which seems to tell different story: with UR enabled, we do not adjust after CM.
>> And actually, I think it is correct to compute the max over all phases in the
>> cycle to decide heuristics move. We only need to max, not destructively overwrite.
> 
> ok. Like this?
> 
> http://cr.openjdk.java.net/~rkennke/adaptive-minfree/webrev.01/

Yes! Looks and runs good.

This deserves the backporting to 8u, because our adopters are running with
adaptive by default. Let's see what overnight runs show.

Thanks,
-Aleksey