RFR: Fix satb barrier for object array copy

[Kelvin Nilsen]

On Thu, 30 Mar 2023 08:13:51 GMT, Aleksey Shipilev <shade at openjdk.org> wrote:

>> The satb barrier for overwritten arrays residing in old-gen must be unconditional during young GC.
>> 
>> During old GC, the satb barrier is only needed for overwritten old-gen arrays if the array is below TAMS within its old-gen region.
>
> src/hotspot/share/gc/shenandoah/shenandoahFullGC.cpp line 325:
> 
>> 323: 
>> 324:     phase4_compact_objects(worker_slices);
>> 325: 
> 
> This is the part of #235, is it? Should not be in this PR, I think.

I included #235 redundantly here so that I could pass some regression tests.  I've integrated #235 already, so the history shows that commit before this one...

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

PR Review Comment: https://git.openjdk.org/shenandoah/pull/237#discussion_r1153869428