On Fri, 30 Oct 2026 21:08:41 GMT, Patricio Chilano Mateo <[email protected]> wrote:
>> Shiv Shah has updated the pull request incrementally with one additional >> commit since the last revision: >> >> Add the invokeExact cases, the LambdaForm frame keeps the >> DirectMethodHandle in local0 > > I think > [this](https://github.com/openjdk/jdk/blob/6ae6ea84a5c1b3b65658a39f60f748b88c20e350/src/hotspot/cpu/x86/templateInterpreterGenerator_x86.cpp#L1440) > comment should now be updated to mention that we are also handling the > return barrier case for virtual threads, not just deoptimization. And that > the current check for an interpreter return pc already works for both cases. @pchilano Thanks for the feedback. Pushed the round. Own entry per port, mdp reset, your assert and the comments are in. ppc, s390 and arm only jump to the deopt entry, I have no way to run those, said so in the description. ------------- PR Comment: https://git.openjdk.org/jdk/pull/32661#issuecomment-6579707340
