arch/c6x/lib/mpyll.S

   1 ;;  Copyright (C) 2010 Texas Instruments Incorporated
   2 ;;  Contributed by Mark Salter <msalter@redhat.com>.
   3 ;;
   4 ;; This program is free software; you can redistribute it and/or modify
   5 ;; it under the terms of the GNU General Public License as published by
   6 ;; the Free Software Foundation; either version 2 of the License, or
   7 ;; (at your option) any later version.
   8 ;;
   9 ;; This program is distributed in the hope that it will be useful,
  10 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
  11 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  12 ;; GNU General Public License for more details.
  13 ;;
  14 ;; You should have received a copy of the GNU General Public License
  15 ;; along with this program; if not, write to the Free Software
  16 ;; Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
  17
  18 #include <linux/linkage.h>
  19
  20         ;; uint64_t __c6xabi_mpyll(uint64_t x, uint64_t y)
  21         ;;
  22         ;; 64x64 multiply
  23         ;; First compute partial results using 32-bit parts of x and y:
  24         ;;
  25         ;;   b63         b32 b31          b0
  26         ;;    -----------------------------
  27         ;;    |      1      |      0      |
  28         ;;    -----------------------------
  29         ;;
  30         ;;   P0 = X0*Y0
  31         ;;   P1 = X0*Y1 + X1*Y0
  32         ;;   P2 = X1*Y1
  33         ;;
  34         ;;   result = (P2 << 64) + (P1 << 32) + P0
  35         ;;
  36         ;; Since the result is also 64-bit, we can skip the P2 term.
  37
  38         .text
  39 ENTRY(__c6xabi_mpyll)
  40         mpy32u  .m1x    A4,B4,A1:A0     ; X0*Y0
  41         b       .s2     B3
  42  ||     mpy32u  .m2x    B5,A4,B1:B0     ; X0*Y1 (don't need upper 32-bits)
  43  ||     mpy32u  .m1x    A5,B4,A3:A2     ; X1*Y0 (don't need upper 32-bits)
  44         nop
  45         nop
  46         mv      .s1     A0,A4
  47         add     .l1x    A2,B0,A5
  48         add     .s1     A1,A5,A5
  49 ENDPROC(__c6xabi_mpyll)