https://gcc.gnu.org/bugzilla/show_bug.cgi?id=117208
Andrew Pinski <pinskia at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Resolution|--- |INVALID
Status|WAITING |RESOLVED
--- Comment #5 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
(In reply to Richard Biener from comment #4)
> h should be zero-initialized. Can't reproduce either.
h might be zero-initialized but it might be assigned via:
k((l(3 && c), s), 0, 0, h = t, 0);
Plus the order if h is assigned first vs `(l(3 && c), s)` being evulated first
is also unspecified in C since the order of the arguments of a function call
evulation is unspecified.
That is on you can even get different answers even if t is initlaized to 1 but
it depends on the target. e.g. on aarch64, you will always get `1 -1` no
matter what since the assignment to h is evulated after the call to l.
