RFR: 8137022: Concurrent refinement thread adjustment and (de-)activation suboptimal [v3]

[Kim Barrett]

On Tue, 27 Sep 2022 15:49:36 GMT, Kim Barrett <kbarrett at openjdk.org> wrote:

>> src/hotspot/share/gc/g1/g1Policy.cpp line 963:
>> 
>>> 961:             hcc_cards,
>>> 962:             merge_hcc_time_ms,
>>> 963:             (exceeded_goal ? " (exceeded goal)" : ""));
>> 
>> Since the pending cards target is initialized to `SIZE_MAX`, the first GC this will always print: 
>> 
>> GC(0) GC refinement: goal: 18446744073709551615 ...
>> 
>> 
>> Not sure if printing 0 is much better but it looks more "buggy" to print `SIZE_MAX` imo.
>
> Oh, I forgot about that.  I was going to try to come up with something better there, like somehow printing "unlimited" or something like that.

Stefan and I talked about this offline, and decided to leave it as is for now.
- Printing a string like "unlimited" for that case makes parsing more complicated.
- Printing 0 can be confusing because that's a valid value.  Neither of us likes log messages that contain "lies".
- Suppressing the message in this case loses the information used to calculate the first normal value.

-------------

PR: https://git.openjdk.org/jdk/pull/10256