RFR: 8256265 G1: Improve parallelism in regions that failed evacuation

Hamlin Li mli at openjdk.java.net
Thu Dec 2 02:14:26 UTC 2021


On Thu, 2 Dec 2021 01:58:51 GMT, Kirk Pepperdine <kirk at kodewerk.com> wrote:

> Am I to understand that these benchmarks were running for 3-5 seconds?

No, the time is just for `Post Evacuate Cleanup 1` pause time.
For the test, it takes about dozens of seconds for each round.

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

PR: https://git.openjdk.java.net/jdk/pull/6627



More information about the hotspot-gc-dev mailing list