RFR(s): 8247248: JVM TI might create JNI locals in another thread when using handshakes.

[David Holmes]

Looks good!

Thanks for making the change.

On a positive note I think this would now show that the conversion from 
VMop to direct handshake was actually much simpler than might have been 
thought. :)

I'm not sure when the repo fork is happening, so am unclear whether this 
will head to the right repo in time. :)

Thanks,
David
-----

On 10/06/2020 11:57 pm, Robbin Ehn wrote:
> Hi David and Serguei, (Dan feel free to chime in)
> 
>> Honestly I think I'd like to see things reverted to the use of 
>> calling_thread as done for the VMOperation previously. We know it is 
>> functionally correct and it should also have the same performance 
>> profile.
> 
> Done:
> http://cr.openjdk.java.net/~rehn/8247248/v2/webrev/
> 
> Passes: hotspot jdi/jvmti testing, running mach5.
> 
> I'll push tomorrow morning if test is ok and you all are happy (+- 
> nits). (and no objection to break the 24h rule)
> I started this patch with reverting "8242425: JVMTI monitor operations 
> should use Thread-Local Handshakes".
> And work my way forward.
> 
> Thanks, Robbin
> 
>>
>> Thanks,
>> David
>>
>>> Thanks, Robbin
>>>
>>>>
>>>> Thanks,
>>>> David
>>>> -----
>>>>
>>>>> Issue:
>>>>> https://bugs.openjdk.java.net/browse/JDK-8247248
>>>>>
>>>>> Local testing of JDI/JVMTI and t1-5.
>>>>> (no real crash so there is nothing to reproduce)
>>>>>
>>>>> Thanks, Robbin