RFR: 8343704: Bad GC parallelism with processing Cleaner queues [v11]

[Brent Christian]

On Mon, 18 Nov 2024 09:46:50 GMT, Aleksey Shipilev <shade at openjdk.org> wrote:

>> See the bug for more discussion and reproducer. This PR replaces the ad-hoc linked list with segmented list of arrays. Arrays are easy targets for GC. There are possible improvements here, most glaring is parallelism that is currently knee-capped by global synchronization. The synchronization scheme follows what we have in original code, and I think it is safer to continue with it right now.
>> 
>> I'll put performance data in a separate comment.
>> 
>> Additional testing:
>>  - [x] Original reproducer improves drastically
>>  - [x] New microbenchmark shows no regression on "churning" tests, which covers insertion/removal perf
>>  - [x] New microbenchmark shows improvement on Full GC times (crude, but repeatable), serves as a proxy for reproducer
>>  - [x] `java/lang/ref` tests in release 
>>  - [x] `all` tests in fastdebug
>
> Aleksey Shipilev has updated the pull request incrementally with one additional commit since the last revision:
> 
>   Drop --add-exports from the test

Changes requested by bchristi (Reviewer).

test/jdk/jdk/internal/ref/Cleaner/CleanableListTest.java line 46:

> 44: public class CleanableListTest {
> 45: 
> 46:     static final int SINGLE_NODE_CAPACITY = CleanableListTestHelper.NODE_CAPACITY - 1;

If `CleanableListTestHelper` uses a list without the `CleanerCleanable`, shouldn't we be able to test the full `NODE_CAPACITY` ?

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

PR Review: https://git.openjdk.org/jdk/pull/22043#pullrequestreview-2439985994
PR Review Comment: https://git.openjdk.org/jdk/pull/22043#discussion_r1847495395