https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85143
Andrew Pinski <pinskia at gcc dot gnu.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |NEW Last reconfirmed| |2018-04-01 Ever confirmed|0 |1 --- Comment #4 from Andrew Pinski <pinskia at gcc dot gnu.org> --- Confirmed. Internal details to GCC: Note on aarch64 (and maybe other targets too), when we do the expansion of m_2 = MIN_EXPR <n_1(D), 1337>; _4 = m_2 > i_3(D); We want to convert it back to just: _1 = i_3(D) < n_4(D); _2 = i_3(D) <= 1336; _5 = _1 & _2; But that is only if we can't pull the MIN_EXPR out of the loop.