sysdeps/alpha/addmul_1.s

   1  # Alpha 21064 __mpn_addmul_1 -- Multiply a limb vector with a limb and add
   2  # the result to a second limb vector.
   3
   4  # Copyright (C) 1992, 1994, 1995 Free Software Foundation, Inc.
   5
   6  # This file is part of the GNU MP Library.
   7
   8  # The GNU MP Library is free software; you can redistribute it and/or modify
   9  # it under the terms of the GNU Lesser General Public License as published by
  10  # the Free Software Foundation; either version 2.1 of the License, or (at your
  11  # option) any later version.
  12
  13  # The GNU MP Library is distributed in the hope that it will be useful, but
  14  # WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
  15  # or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public
  16  # License for more details.
  17
  18  # You should have received a copy of the GNU Lesser General Public License
  19  # along with the GNU MP Library; see the file COPYING.LIB.  If not, write to
  20  # the Free Software Foundation, Inc., 59 Temple Place - Suite 330, Boston,
  21  # MA 02111-1307, USA.
  22
  23
  24  # INPUT PARAMETERS
  25  # res_ptr      r16
  26  # s1_ptr       r17
  27  # size         r18
  28  # s2_limb      r19
  29
  30  # This code runs at 42 cycles/limb on EV4 and 18 cycles/limb on EV5.
  31
  32         .set    noreorder
  33         .set    noat
  34 .text
  35         .align  3
  36         .globl  __mpn_addmul_1
  37         .ent    __mpn_addmul_1 2
  38 __mpn_addmul_1:
  39         .frame  $30,0,$26
  40
  41         ldq     $2,0($17)       # $2 = s1_limb
  42         addq    $17,8,$17       # s1_ptr++
  43         subq    $18,1,$18       # size--
  44         mulq    $2,$19,$3       # $3 = prod_low
  45         ldq     $5,0($16)       # $5 = *res_ptr
  46         umulh   $2,$19,$0       # $0 = prod_high
  47         beq     $18,.Lend1      # jump if size was == 1
  48         ldq     $2,0($17)       # $2 = s1_limb
  49         addq    $17,8,$17       # s1_ptr++
  50         subq    $18,1,$18       # size--
  51         addq    $5,$3,$3
  52         cmpult  $3,$5,$4
  53         stq     $3,0($16)
  54         addq    $16,8,$16       # res_ptr++
  55         beq     $18,.Lend2      # jump if size was == 2
  56
  57         .align  3
  58 .Loop:  mulq    $2,$19,$3       # $3 = prod_low
  59         ldq     $5,0($16)       # $5 = *res_ptr
  60         addq    $4,$0,$0        # cy_limb = cy_limb + 'cy'
  61         subq    $18,1,$18       # size--
  62         umulh   $2,$19,$4       # $4 = cy_limb
  63         ldq     $2,0($17)       # $2 = s1_limb
  64         addq    $17,8,$17       # s1_ptr++
  65         addq    $3,$0,$3        # $3 = cy_limb + prod_low
  66         cmpult  $3,$0,$0        # $0 = carry from (cy_limb + prod_low)
  67         addq    $5,$3,$3
  68         cmpult  $3,$5,$5
  69         stq     $3,0($16)
  70         addq    $16,8,$16       # res_ptr++
  71         addq    $5,$0,$0        # combine carries
  72         bne     $18,.Loop
  73
  74 .Lend2: mulq    $2,$19,$3       # $3 = prod_low
  75         ldq     $5,0($16)       # $5 = *res_ptr
  76         addq    $4,$0,$0        # cy_limb = cy_limb + 'cy'
  77         umulh   $2,$19,$4       # $4 = cy_limb
  78         addq    $3,$0,$3        # $3 = cy_limb + prod_low
  79         cmpult  $3,$0,$0        # $0 = carry from (cy_limb + prod_low)
  80         addq    $5,$3,$3
  81         cmpult  $3,$5,$5
  82         stq     $3,0($16)
  83         addq    $5,$0,$0        # combine carries
  84         addq    $4,$0,$0        # cy_limb = prod_high + cy
  85         ret     $31,($26),1
  86 .Lend1: addq    $5,$3,$3
  87         cmpult  $3,$5,$5
  88         stq     $3,0($16)
  89         addq    $0,$5,$0
  90         ret     $31,($26),1
  91
  92         .end    __mpn_addmul_1