RFR: 8343704: Bad GC parallelism with processing Cleaner queues [v14]
Aleksey Shipilev
shade at openjdk.org
Thu Nov 21 19:33:02 UTC 2024
On Thu, 21 Nov 2024 18:29:24 GMT, Aleksey Shipilev <shade at openjdk.org> wrote:
>> Sorry for being dense, but where is a value being set to `CleanerGC.prev` ?
>
> Ah (facepalms), I see, I thought I assigned it! Damn. Let me fix and re-measure.
Redone the benchmark. We actually want all targets to be reachable, and only expose the Cleaner-side linked lists to the test. New performance data is updated here: https://github.com/openjdk/jdk/pull/22043#issuecomment-2470928360
-------------
PR Review Comment: https://git.openjdk.org/jdk/pull/22043#discussion_r1852730179
More information about the core-libs-dev
mailing list