On Fri, 9 Oct 2026 04:14:42 GMT, Serguei Spitsyn <[email protected]> wrote:
> > Talked to Patricio, going with the middle one, OPAQUE_FRAME from doit when > > the frozen caller is compiled, the thaw path stays for an interpreted > > caller. The deopt in doit for that case would never have found the frame > > anyway > > Sounds reasonable but it would be mice to look at the fix. Tried the compiled caller case as well and it works, so it is in. doit deopts the frozen compiled caller the way set_frame_pop does it, thaw leaves the preserved args to unpack_on_stack. Interpreted path as before, except the return slot is patched after push_return_frame instead of rebuilding the frame, builds on ppc and s390 too. popframes003 covers the compiled case with -Xcomp and a WhiteBox check, popframes002 the interpreted one, both run popframes001's classes with properties. A frozen compiled frame being popped still stays opaque. ------------- PR Comment: https://git.openjdk.org/jdk/pull/32661#issuecomment-6130132225
