groovy-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From David Clark <plotinussm...@gmail.com>
Subject Re: Very Odd/Random JVM Slowdown With Indy
Date Sat, 19 Mar 2016 22:04:59 GMT
I believe I have a simple application which shows the problem. You can 
find it here:

https://github.com/dwclark/deopt-storm

The README should cover how to use it. The original application involved 
a dynamic state machine. I have tried to preserve the basics of how the 
program would execute while removing the application specific details. 
Please let me know if there is something I can do to improve this or if 
you have any questions.

Thanks,

David

On 03/18/2016 07:26 AM, Jochen Theodorou wrote:
>
>
> On 16.03.2016 20:31, David Clark wrote:
>> If I can think of a way to reproduce the problem with a small script or
>> application I will definitely post it to the list. I really have no idea
>> what would cause this problem. If you have any suggestions about what
>> types of application behavior would tend to lead to this issue I will
>> give them a try.
>>
>> Thank for the link to the JDK bug, that does seem to be what we are 
>> seeing.
>
> If it is not that bug, then it is most probably... well it is the most 
> probable reason anyway... our non-existing ability to handle multiple 
> targets for the method call in the callsite. It is non-existing, 
> because we don't do it for the non-indy case, and while this can 
> increase Groovy's performance quite a bit, it is additional complexity 
> to deal with and so I was waiting for a real life use case.... And it 
> might be, that your case is exactly what I was looking for.
>
> bye Jochen


Mime
View raw message