RFR: 8137022: Concurrent refinement thread adjustment and (de-)activation suboptimal [v2]

[Stefan Johansson]

On Thu, 22 Sep 2022 07:07:59 GMT, Kim Barrett <kbarrett at openjdk.org> wrote:

>> src/hotspot/share/gc/g1/g1ConcurrentRefine.cpp line 267:
>> 
>>> 265:     // If target is "unbounded" then wait forever (until explicitly
>>> 266:     // activated).  This happens during startup, when we don't bother with
>>> 267:     // refinement.
>> 
>> We use `SIZE_MAX` to signal that the target is "unbounded", like the comment says. We do this comparison a few times and I think it would make sense to define a constant with a more descriptive name for this.
>
> Added `is_pending_cards_target_initialize` and `PendingCardsTargetUninitialized`.

Good, was thinking about suggesting a predicate as well.

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

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