RFR: 8318479: [jmh] the test security.CacheBench failed for multiple threads run
Daniel Jeliński
djelinski at openjdk.org
Fri Nov 10 12:51:56 UTC 2023
On Mon, 23 Oct 2023 16:23:55 GMT, Daniel Jeliński <djelinski at openjdk.org> wrote:
> Remove the hardcoded maximum heap size.
>
> Verified that the benchmark now passes with 100 threads given sufficient heap:
>
> make test TEST=micro:CacheBench MICRO="OPTIONS=-t 100;JAVA_OPTIONS=-Xmx16g"
I introduced the limit originally; I think I needed it because the default limit on my machine was too low to run the benchmark with the largest cache size.
-------------
PR Comment: https://git.openjdk.org/jdk/pull/16315#issuecomment-1805675721
More information about the security-dev
mailing list