https://gcc.gnu.org/bugzilla/show_bug.cgi?id=119492
Andrew Pinski <pinskia at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |ASSIGNED
Assignee|unassigned at gcc dot gnu.org |pinskia at gcc dot
gnu.org
--- Comment #5 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
So looking into this further, I think PR 15387 got fixed correctly via PR 49519
(and PR51323 and PR 50074).
Which case the check in tree-tailcall.cc for taking the argument address might
be able to be removed and let the rest of the alias analysis part take care of
it.
I am going to test that for GCC 16 (make sure I test 32bit too since that is
one target which uses the stack for almost all argument passing) and
https://gcc.gnu.org/pipermail/gcc-patches/2025-March/679182.html is there for
musttail .
