RFR: Differentiate young generation marking. [v2]

[Bernd Mathiske]

On Tue, 8 Dec 2020 19:28:36 GMT, Charlie Gracie <cgracie at openjdk.org> wrote:

>> Bernd Mathiske has updated the pull request with a new target base due to a merge or a rebase. The pull request now contains one commit:
>> 
>>   Differentiate young generation marking.
>
> src/hotspot/share/gc/shenandoah/shenandoahHeap.cpp line 1625:
> 
>> 1623:   OrderAccess::fence();
>> 1624: 
>> 1625:   generation->concurrent_mark()->mark_roots(ShenandoahPhaseTimings::scan_roots);
> 
> I am wondering if it makes sense to have an instance field / an API to get the "active generation". It may reduce API changes.

We have that in later commits and we will use it in other phases, just not in marking. We aspire to having two marking phases running concurrently with each other, later on, past milestone 6.

> src/hotspot/share/gc/shenandoah/shenandoahHeap.hpp line 523:
> 
>> 521:   // Prepare concurrent root processing
>> 522:   void prepare_concurrent_roots();
>> 523: 
> 
> (nit) extra whitespace change

Fixed. Thanks! "git diff" did not show this insertion. So I'll check with other means as well going forward.

> src/hotspot/share/gc/shenandoah/shenandoahHeap.hpp line 526:
> 
>> 524:   // Prepare and finish concurrent unloading
>> 525:   void prepare_concurrent_unloading();
>> 526: 
> 
> (nit) extra whitespace change

Fixed.

> src/hotspot/share/gc/shenandoah/shenandoahHeap.hpp line 675:
> 
>> 673: // ---------- Evacuation support
>> 674: //
>> 675: 
> 
> (nit) extra whitespace change

Fixed.

> src/hotspot/share/gc/shenandoah/shenandoahHeap.hpp line 730:
> 
>> 728: // ---------- Testing helpers functions
>> 729: //
>> 730: 
> 
> (nit) extra whitespace change

Fixed.

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

PR: https://git.openjdk.java.net/shenandoah/pull/8