Shenandoah performance problem

[Attila Axt]

Thanks for the fix and the quick response.  I've tried the first 
changeset locally against the benchmark, it seems to work fine.

I can try the full changeset earliest on monday in production 
environment. I will feedback on the results.

Greetings,
   axt

On 10/17/19 8:53 PM, Aleksey Shipilev wrote:
> On 10/17/19 6:29 PM, Aleksey Shipilev wrote:
>> On 10/17/19 12:49 PM, Aleksey Shipilev wrote:
>>> This workload runs:
>>>    Parallel GC, -t 1:     0.112 ± 0.006  ms/op
>>>    Parallel GC, -t max:   0.146 ± 0.002  ms/op
>>>    Shenandoah GC, -t 1:   0.120 ± 0.002  ms/op
>>>    Shenandoah GC, -t max: 1.595 ± 0.050  ms/op <--- scalability bottleneck
>>>
>>> And you are right, that's the contention on pinning/unpinning, I am looking what can be done about
>>> that...
>> This bunch of fixes makes it much better:
>>    https://cr.openjdk.java.net/~shade/shenandoah/async-pinning/
> Filed:
>    https://bugs.openjdk.java.net/browse/JDK-8232575
>
> ...would RFR it tomorrow.
>