A very unique bytecode flow (described in the test) results in an invalid stack map frame. The fix clears stale stack map frame locals that would otherwise cause problems in subsequent stack map generation rounds.
Please review. Thank you, Adam --------- - [x] I confirm that I make this contribution in accordance with the [OpenJDK Interim AI Policy](https://openjdk.org/legal/ai). ------------- Commit messages: - 8386700: Class-File API: StackMapGenerator.setLocalsFromArg leaves stale locals, generating invalid stack maps Changes: https://git.openjdk.org/jdk/pull/31533/files Webrev: https://webrevs.openjdk.org/?repo=jdk&pr=31533&range=00 Issue: https://bugs.openjdk.org/browse/JDK-8386700 Stats: 31 lines in 2 files changed: 28 ins; 0 del; 3 mod Patch: https://git.openjdk.org/jdk/pull/31533.diff Fetch: git fetch https://git.openjdk.org/jdk.git pull/31533/head:pull/31533 PR: https://git.openjdk.org/jdk/pull/31533
