On Thu, 12 Jan 2023 09:23:49 GMT, Daniel Jeliński <djelin...@openjdk.org> wrote:
> Please review this patch that fixes a few memory leaks in JNI code. > > [The latest > documentation](https://docs.oracle.com/en/java/javase/17/docs/specs/jni/functions.html#releaseprimitivetypearrayelements-routines) > of JNI functions makes an explicit note about the use of JNI_COMMIT: > >> If `JNI_COMMIT` is passed as the `mode` argument when `elems` is a copy of >> the elements in `array`, then a final call to >> *Release<PrimitiveType\>ArrayElements* passing a `mode` argument of "0" or >> `JNI_ABORT`, should be made to free the `elems` buffer > > No new regression test. I manually verified the Linux fix using ClhdsbPstack > test in root mode. Also, tier1-2 tests on mach5 continue to pass. There are occurrences of JNI_COMMIT on macos in libawt and libsaproc. Is there a reason you did not fix these also? ------------- PR: https://git.openjdk.org/jdk/pull/11963
