On Tue, Sep 26, 2017 at 11:36:14AM -0500, Segher Boessenkool wrote: > Hi! > > On Tue, Sep 26, 2017 at 10:34:44AM -0400, Michael Meissner wrote: > > * config/rs6000/rs6000.md (movsi_from_df): Optimize converting a > > DFmode to a SFmode, and then needing to move the SFmode to a GPR > > to use the XSCVDPSP instruction instead of FRSP and XSCVDPSPN. > > > --- gcc/config/rs6000/rs6000.md (revision 253170) > > +++ gcc/config/rs6000/rs6000.md (working copy) > > @@ -6919,6 +6919,26 @@ (define_insn_and_split "*movdi_from_sf_z > > "4, 4, 4, 4, 8, > > 8, 4")]) > > > > +;; Like movsi_from_sf, but combine a convert from DFmode to SFmode before > > +;; moving it to SImode. We can do a SFmode store without having to do the > > +;; conversion explicitly. If we are doing a register->register > > conversion, use > > +;; XSCVDPSP instead of XSCVDPSPN, since the former handles cases where the > > +;; input will not fit in a SFmode, and the later assumes the value has > > already > > +;; been rounded. > > +(define_insn "*movsi_from_df" > > + [(set (match_operand:SI 0 "nonimmediate_operand" "=wa,m,wY,Z") > > + (unspec:SI [(float_truncate:SF > > + (match_operand:DF 1 "gpc_reg_operand" "wa, f,wb,wa"))] > > + UNSPEC_SI_FROM_SF))] > > (The indentation is a bit broken here -- DF line is indented a space too > many, and the constraint strings do not line up).
That must something funky with patches and tabs. It looks ok after I apply the patch (the match_operand:DF is indented one space under float_truncate:SF). -- Michael Meissner, IBM IBM, M/S 2506R, 550 King Street, Littleton, MA 01460-6245, USA email: meiss...@linux.vnet.ibm.com, phone: +1 (978) 899-4797
