On Tue, Jun 07, 2022 at 02:26:17PM -0500, will schmidt wrote: > On Mon, 2022-06-06 at 20:31 -0400, Michael Meissner wrote: > > (define_insn "vsx_xxspltd_<mode>" > > [(set (match_operand:VSX_D 0 "vsx_register_operand" "=wa") > > - (unspec:VSX_D [(match_operand:VSX_D 1 "vsx_register_operand" > > "wa")
Someone (you?) uses format=flawed. You cannot reply to emails that contain patches that way, it messes up everything :-( > > - (match_operand:QI 2 "u5bit_cint_operand" "i")] > > - UNSPEC_VSX_XXSPLTD))] > > + (vec_duplicate:VSX_D > > + (vec_select:<VS_scalar> > > + (match_operand:VSX_D 1 "gpc_reg_operand" "wa") > > + (parallel [(match_operand:QI 2 "const_0_to_1_operand" > > "i")]))))] > > "VECTOR_MEM_VSX_P (<MODE>mode)" > > Noting that > (define_mode_iterator VSX_D [V2DF V2DI]) > (define_mode_attr VS_scalar [(V1TI "TI") > (V2DF "DF") > (V2DI "DI") > (V4SF "SF") > (V4SI "SI") > (V8HI "HI") > (V16QI "QI")]) Yeah, the comment ;; Map the scalar mode for a vector type is misleading, in more ways than one :-( And the whole thing is just the same as VEC_base anyway, so it is much better to just use that. Segher
