RFR: 8343704: Bad GC parallelism with processing Cleaner queues [v5]

[Chen Liang]

On Wed, 13 Nov 2024 20:33:25 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 the `ArrayList` wrapper that manages synchronization, search and replacements efficiently. 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:
> 
>   Review feedback: make sure trimming actually works, stylistic changes

src/java.base/share/classes/jdk/internal/ref/CleanerImpl.java line 236:

> 234:     static final class PhantomCleanableList {
> 235:         private static final int MIN_CAPACITY = 16;
> 236:         private final Object lock = new Object();

Why use another lock object when identity of this list is not escaped elsewhere?

src/java.base/share/classes/jdk/internal/ref/CleanerImpl.java line 311:

> 309:                 // want to cause an immediate resize on next insertion.
> 310:                 if ((size < arr.length / 4) && (size > MIN_CAPACITY)) {
> 311:                     int newLen = ArraysSupport.newLength(size, 1, size);

This code isn't immediately clear that it's trying to half the array size, as you are using the size computation method for array growth.  Can you add a remark for so?

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

PR Review Comment: https://git.openjdk.org/jdk/pull/22043#discussion_r1841375400
PR Review Comment: https://git.openjdk.org/jdk/pull/22043#discussion_r1841378770