RFR: JDK-8277178: Reduce the priority of data dependent nodes when OptoScheduling enabled

SUN Guoyun duke at openjdk.java.net
Thu Jan 13 06:30:27 UTC 2022


On Tue, 30 Nov 2021 08:37:27 GMT, Tobias Hartmann <thartmann at openjdk.org> wrote:

> I'm not too familiar with this code but I gave it a quick run through our performance testing and all results look good except for the MonteCarlo benchmark from SPECjvm2008 with G1 which shows a 1% regression. It could just be run-to-run variance but better double check.
> 
> Also, I think it would be good to add a JMH benchmark for this.

I have add a new benmark test file (InstructionScheduling.java). I had make a test for LoongArch and MIPS64 architecture , and can significantly observe the performance improvement(about 30%). However, the performance improvement on aarch64 architecture is not obvious, and the test jitter is very large.

I don't have Silvermont or Centerton x86 machine, so can you make a performance testing again for this patch? Thank you very much.

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

PR: https://git.openjdk.java.net/jdk/pull/6407


More information about the hotspot-compiler-dev mailing list