On Tue, 22 Sep 2020 14:38:05 GMT, Daniel D. Daugherty <dcu...@openjdk.org> wrote:
> @fisk - You are missing testing information for this PR. Sorry about that. I have performed testing with ZGC enabled from tier1-7 (multiple times), and standard testing tier1-5. I have also stress tested the change with gc-test-suite, with full verification on (there is a bunch of it), and the GC interval reduced to very small (GC all the time). Recent versions of this patch I also tested from tier1-8 (2 times). I have also perf tested this in dev-submit to check for regressions (none found), as well as stress tested SPECjbb2015 on a 3 TB machine with ~2200 threads. I have also manually stress tested the changes on an AArch64 machine, using gc-test-suite. ------------- PR: https://git.openjdk.java.net/jdk/pull/296
