https://gcc.gnu.org/bugzilla/show_bug.cgi?id=120870
--- Comment #30 from H.J. Lu <hjl.tools at gmail dot com> --- (In reply to Sam James from comment #29) > I think the patch in CPython adds an optimisation and it is latent in GCC. > Trying to understand it more. If CPython manipulates stack pointer assuming certain compiler behaviors, it may not work. We need a standalone testcase to show the compiler bug.
