sysdeps/vax/mul_1.s

   1 # VAX __mpn_mul_1 -- Multiply a limb vector with a limb and store
   2 # the result in a second limb vector.
   3
   4 # Copyright (C) 1992, 1994 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 Library General Public License as published by
  10 # the Free Software Foundation; either version 2 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 Library General Public
  16 # License for more details.
  17
  18 # You should have received a copy of the GNU Library 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       (sp + 4)
  26 # s1_ptr        (sp + 8)
  27 # size          (sp + 12)
  28 # s2_limb       (sp + 16)
  29
  30 .text
  31         .align 1
  32 .globl ___mpn_mul_1
  33 ___mpn_mul_1:
  34         .word   0xfc0
  35         movl    12(ap),r4
  36         movl    8(ap),r8
  37         movl    4(ap),r9
  38         movl    16(ap),r6
  39         jlss    s2_big
  40
  41 # One might want to combine the addl2 and the store below, but that
  42 # is actually just slower according to my timing tests.  (VAX 3600)
  43
  44         clrl    r3
  45         incl    r4
  46         ashl    $-1,r4,r7
  47         jlbc    r4,L1
  48         clrl    r11
  49
  50 # Loop for S2_LIMB < 0x80000000
  51 Loop1:  movl    (r8)+,r1
  52         jlss    L1n0
  53         emul    r1,r6,$0,r2
  54         addl2   r11,r2
  55         adwc    $0,r3
  56         movl    r2,(r9)+
  57 L1:     movl    (r8)+,r1
  58         jlss    L1n1
  59 L1p1:   emul    r1,r6,$0,r10
  60         addl2   r3,r10
  61         adwc    $0,r11
  62         movl    r10,(r9)+
  63
  64         jsobgtr r7,Loop1
  65         movl    r11,r0
  66         ret
  67
  68 L1n0:   emul    r1,r6,$0,r2
  69         addl2   r11,r2
  70         adwc    r6,r3
  71         movl    r2,(r9)+
  72         movl    (r8)+,r1
  73         jgeq    L1p1
  74 L1n1:   emul    r1,r6,$0,r10
  75         addl2   r3,r10
  76         adwc    r6,r11
  77         movl    r10,(r9)+
  78
  79         jsobgtr r7,Loop1
  80         movl    r11,r0
  81         ret
  82
  83
  84 s2_big: clrl    r3
  85         incl    r4
  86         ashl    $-1,r4,r7
  87         jlbc    r4,L2
  88         clrl    r11
  89
  90 # Loop for S2_LIMB >= 0x80000000
  91 Loop2:  movl    (r8)+,r1
  92         jlss    L2n0
  93         emul    r1,r6,$0,r2
  94         addl2   r11,r2
  95         adwc    r1,r3
  96         movl    r2,(r9)+
  97 L2:     movl    (r8)+,r1
  98         jlss    L2n1
  99 L2p1:   emul    r1,r6,$0,r10
 100         addl2   r3,r10
 101         adwc    r1,r11
 102         movl    r10,(r9)+
 103
 104         jsobgtr r7,Loop2
 105         movl    r11,r0
 106         ret
 107
 108 L2n0:   emul    r1,r6,$0,r2
 109         addl2   r1,r3
 110         addl2   r11,r2
 111         adwc    r6,r3
 112         movl    r2,(r9)+
 113         movl    (r8)+,r1
 114         jgeq    L2p1
 115 L2n1:   emul    r1,r6,$0,r10
 116         addl2   r1,r11
 117         addl2   r3,r10
 118         adwc    r6,r11
 119         movl    r10,(r9)+
 120
 121         jsobgtr r7,Loop2
 122         movl    r11,r0
 123         ret