RFR: 8185525: [Event Request] Add Tracing event for DictionarySizes

[coleen.phillimore at oracle.com]

Looks like another change has leaked in:

http://cr.openjdk.java.net/~gziemski/8185525_rev9/src/java.base/share/native/libjava/jdk_util.c.udiff.html

Needs @bug 8185525

in test: 
http://cr.openjdk.java.net/~gziemski/8185525_rev9/test/jdk/jdk/jfr/event/runtime/TestTableStatisticsEvent.java.html

Otherwise looks good to me.  I don't need to see another webrev for 
this.  Thank you for finding a creative solution to the try_lock problem.

Thanks,
Coleen

On 5/2/19 2:27 PM, gerard ziemski wrote:
> Thank you Erik!
>
> I changed the bug title (as well as my other JFR event bug title)
>
>
> cheers
>
> On 5/2/19 12:47 PM, Erik Gahlin wrote:
>> Looks good.
>>
>> Since the tracing framework has been replaced with JFR and there is a 
>> JIRA tag for "jfr-event-request", we could perhaps remove "[Event 
>> Request]" and replace "tracing" with "JFR" in the bug title.
>>
>> Thanks
>> Erik
>>
>>> hi all,
>>>
>>> Please review this, hopefully final, fix. After some offline 
>>> discussion, we agreed to:
>>>
>>> #1 Keep the rate attributes (can't synthesize otherwise because we 
>>> don't track additions/deletions on their own)
>>>
>>> #2 Use data from last event if we can't lock a table (fulfills JFR 
>>> contract in which we need to deliver event at specified period)
>>>
>>> Bug:     https://bugs.openjdk.java.net/browse/JDK-8185525
>>> Webrev:  http://cr.openjdk.java.net/~gziemski/8185525_rev9
>>> Testing: Mach5 tier1,2,3 (another Mach5 tier1,2,3,4,5,6,7 in progress…)
>>>
>>> cheers
>>
>>
>