RFR: 8305236: Some LoadLoad barriers in the interpreter are unnecessary after JDK-8220051 [v3]

SUN Guoyun duke at openjdk.org
Thu Apr 6 01:11:20 UTC 2023


On Tue, 4 Apr 2023 07:37:53 GMT, SUN Guoyun <duke at openjdk.org> wrote:

>> That looks reasonable to me. I am performing some tests on linux-riscv64 platform.
>
>> That looks reasonable to me. I am performing some tests on linux-riscv64 platform.
> 
> Some StoreStore barriers is unnecessary also in interpreter. I had create a PR https://github.com/openjdk/jdk/pull/13320, please review it together.

> Please do the corresponding stress tests on LoongArch, @sunny868 .

The jcstress test runned OK on LoongArch.

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

PR Comment: https://git.openjdk.org/jdk/pull/13244#issuecomment-1498359189


More information about the hotspot-runtime-dev mailing list