https://gcc.gnu.org/bugzilla/show_bug.cgi?id=120378
--- Comment #5 from Richard Biener <rguenth at gcc dot gnu.org> --- (In reply to Robin Dapp from comment #4) > Does it make sense to have the vmax/vmin/truncate pattern as a fallback for > other targets? On riscv it would save one predicated instruction. I think if targets decide that they could implement the optab in this way, so I'd not do this initially.
