RFR: 8343468: GenShen: Enable relocation of remembered set card tables [v3]

[Kelvin Nilsen]

On Tue, 21 Jan 2025 23:25:27 GMT, Y. Srinivas Ramakrishna <ysr at openjdk.org> wrote:

>> src/hotspot/share/gc/shenandoah/shenandoahGeneration.cpp line 228:
>> 
>>> 226: // This avoids the need to synchronize reads of the table by the GC workers doing
>>> 227: // remset scanning, on the one hand, with the dirtying of the table by mutators
>>> 228: // and by the GC workers doing remset scans, on the other.
>> 
>> Remove "and by the the GC workers doing remset scans" from this comment, so it reads:
>> "This avoids the need to synchronize reads of the table by the GC workers doing
>> remset scanning, on the one hand, with the dirtying of the table by mutators on the other.
>
> @kdnilsen , when are the existing intergenerational pointers (represented by (a subset of) the dirty cards in the read table) transferred to the write table?

That happens during remembered set scanning.  If the remembered set scan (from the read table) finds that certain cards still hold interesting pointers, the card will be dirtied in the write table.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/23170#discussion_r1931077697