https://gcc.gnu.org/bugzilla/show_bug.cgi?id=89256
Alexander Monakov <amonakov at gcc dot gnu.org> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |amonakov at gcc dot gnu.org --- Comment #1 from Alexander Monakov <amonakov at gcc dot gnu.org> --- It's not so easy, as this trick needs a widening multiply or a high-multiply operation: e.g. in func64, the compiler uses upper 64 bits of the 128-bit product of two 64-bit operands. In func128, it would need to obtain 128 upper bits of a 256-bit product, which no CPU offers as part of instruction set; so it would need to be a libcall for a TImode high-multiply operation.