RFR: 8259349: -XX:AvgMonitorsPerThreadEstimate=1 does not work right [v5]

Daniel D.Daugherty dcubed at openjdk.java.net
Mon Jan 11 23:41:59 UTC 2021


On Mon, 11 Jan 2021 21:38:25 GMT, Daniel D. Daugherty <dcubed at openjdk.org> wrote:

>> I'm really tempted to back out the change that made AvgMonitorsPerThreadEstimate
>> an `int` instead of an `intx` and all the associated follow-on changes. Fixing that
>> type wasn't the intent of this fix and I'm spending way too much time on stuff that
>> has absolutely nothing to do with the bug that I'm trying to solve.
>
> In the "learn something new every day" category, I just learned that
> command line options can be of type `size_t`. I've updated the patch
> again. Please take another look. Mach5 Tier[1-3] testing is in process.

Mach5 Tier[1-3] testing looks good; no Tier[12] failures; 2 unrelated
Tier3 failures.

@coleenp and @dholmes-ora - a re-review would be appreciated.

I believe this fix will also address:

    https://bugs.openjdk.java.net/browse/JDK-8256307 cleanup AvgMonitorsPerThreadEstimate and _in_use_list_ceiling types

I'm sorry that I forgot about that bug...

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

PR: https://git.openjdk.java.net/jdk/pull/1992


More information about the hotspot-runtime-dev mailing list