Java 7 Groovy twice as fast as Java 8 and 9

[Chris Newland]

Hi Richard,

I've not been able to reproduce the difference in scores you encountered
although I'd be interested to know what platform and Java commands you
used to get them.

If you like, maybe we can take this over to
https://groups.google.com/forum/#!forum/jitwatch where I think a few
Groovy users would have some ideas.

Cheers,

Chris

On Tue, February 17, 2015 11:42, Richard Kolb wrote:
> Thanks Chris and Martijn.
> Much appreciated.
>
>
> Regards,
> Richard.
> On 17 Feb 2015 13:16, "Martijn Verburg" <martijnverburg at gmail.com> wrote:
>
>
>> Hi Richard,
>>
>>
>> Yes, that and other related issues.  Basically there was a lot changing
>> in Hotspot at the time (G1 GC interactions as well) and it wasn't really
>>  deemed a stable improvement until 8
>>
>> Cheers,
>> Martijn
>>
>>
>> On 17 February 2015 at 10:59, Richard Kolb <rjdkolb at gmail.com> wrote:
>>
>>
>>> Hi Chris
>>>
>>>
>>> Looks like you are correct.
>>> If I run the beanchmarks though Java 8 :'java -XX:TieredStopAtLevel=1
>>> -jar
>>> benchmarks.jar' I get significant improvement
>>> (-XX:+TieredCompilation does not seem to have as much effect by
>>> itself)
>>>
>>> I assume it is turned off because of this
>>> <http://bugs.java.com/view_bug.do?bug_id=7159766> ?
>>>
>>>
>>> regards, Richard
>>> .
>>>
>>>
>>> On 17 February 2015 at 11:54, Chris Newland
>>> <cnewland at chrisnewland.com>
>>> wrote:
>>>
>>>
>>>> Hi Richard,
>>>>
>>>>
>>>> I'll run this through JITWatch [1] later and see if there is
>>>> anything obvious.
>>>>
>>>> Have you considered that with 8 and 9 you're probably hitting
>>>> tiered compilation (disabled by default in 7) which means you're
>>>> running on "quick-and-dirty" C1 optimisation for part of your
>>>> benchmark.
>>>>
>>>> If your benchmark doesn't run for long enough for C2 to kick in
>>>> then
>>> that
>>>> could be a simple reason.
>>>>
>>>> Cheers,
>>>>
>>>>
>>>> Chris
>>>>
>>>>
>>>> [1] https://github.com/AdoptOpenJDK/jitwatch/
>>>>
>>>>
>>>> On Tue, February 17, 2015 06:18, Richard Kolb wrote:
>>>>
>>>>> Hello All,
>>>>>
>>>>>
>>>>>
>>>>> I've created my own micro beanchmarks with jmh.
>>>>> It's forked to here :
>>>>> https://github.com/AdoptOpenJDK/vmbenchmarks
>>>>>
>>>>>
>>>>>
>>>>> Most of the benchmarks clearly show how performance is increasing
>>>>>
>>> except
>>>>> for Groovy.
>>>>>
>>>>
>>> https://github.com/AdoptOpenJDK/vmbenchmarks/blob/master/src/main/jav
>>> a/co
>>>>> m/github/vmbenchmarks/DynamicLanguages.java
>>>>>
>>>>> Perhaps my benchmark is wrong, but it looks like 8 and 9 are
>>>>> significantly slower than 7.
>>>>>
>>>>> Java 7 (build 1.7.0_60-b19):
>>>>> c.g.v.DynamicLanguages.groovy 665.826 ops/s
>>>>>
>>>>> Java 8 (build 1.8.0_31-b13):
>>>>> c.g.v.DynamicLanguages.groovy 366.274
>>>>>
>>>>> Java 9 (build 1.9.0-ea-b42) :
>>>>> c.g.v.DynamicLanguages.groovy 375.593 ops/s
>>>>>
>>>>> regards, Richard.
>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>
>>
>>
>