[Bug c++/80635] std::optional and bogus -Wmaybe-uninitialized warning

[glisse at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80635

--- Comment #7 from Marc Glisse <glisse at gcc dot gnu.org> ---
The warning comes from
  _Z3setiD.6701 (maybe_a$D6763$m_dummy_6);
which is protected by
  _9 = VIEW_CONVERT_EXPR<boolD.2220>(maybe_a$4_7);
  if (_9 != 0)
with
  # maybe_a$D6763$m_dummy_6 = PHI <maybe_a$D6763$m_dummy_4(D)(6), _5(4)>
  # maybe_a$4_7 = PHI <0(6), 1(4)>

In this case, more aggressive threading would kill the possibility to call set
on something undefined (I believe Jeff was already looking into it for other
Wmaybe-uninitialized testcases). The warning is unstable because it depends on
fragile optimization results.

This isn't solvable in general anyway, Wmaybe-uninitialized has "maybe" for a
good reason.