[jdk8u] RFR: 8348228: Major performance regression in GetMethodDeclaringClass and other JVMTI Method functions

[Zhengyu Gu]

On Mon, 23 Jun 2025 13:31:38 GMT, Zhengyu Gu <zgu at openjdk.org> wrote:

> Please review this patch for the parity of Oracle's [JDK-8185348](https://bugs.openjdk.org/browse/JDK-8185348).
> 
> Oracle's patch is non-public. This patch is based on @sspitsyn's comments in JDK-8185348 and consistent with JDK9's implementation.
> 
> Before patch, slowjvmti test case from JDK-8185348:
> 16366ms
> 
> After patch:
> 3ms
> 
> PR for jdk8u-dev has been approved: https://github.com/openjdk/jdk8u-dev/pull/656

> Why is this being requested as a critical patch? The bug exists since `8u112` - so why the sudden urgency? I'd think this warrants more bake time in the development trees as an 8u only bug. Please close this pr and pursue [openjdk/jdk8u-dev#656](https://github.com/openjdk/jdk8u-dev/pull/656) instead. Thanks!

Well, I requested 8u-dev backport on 5/30/2025, three weeks ago, have it even considered? since it is significant performance regression.

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

PR Comment: https://git.openjdk.org/jdk8u/pull/76#issuecomment-2996645022