[Bug target/115910] [15 Regression] ((unsigned)x)/3 with a range for (unsigned)x that does not have the sign bit set seems to produce much worse code

[jakub at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115910

--- Comment #6 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
So perhaps that
          if (CONSTANT_P (SET_SRC (x)))
            /* Constant costs assume a base value of COSTS_N_INSNS (1) and add
               a small value, possibly zero for cheap constants.  */
            src_cost += COSTS_N_INSNS (1);
should be done only for src_cost != 0?  Or src_cost < COSTS_N_INSNS (1)?
Or change pattern_cost not to do
  return cost > 0 ? cost : COSTS_N_INSNS (1);
but
  return cost > COSTS_N_INSNS (1) ? cost : COSTS_N_INSNS (1);
?