source/libs/gmp/gmp-src/mpn/alpha/invert_limb.asm

   1 dnl  Alpha mpn_invert_limb -- Invert a normalized limb.
   2
   3 dnl  Copyright 1996, 2000-2003, 2007, 2011, 2013 Free Software Foundation, Inc.
   4
   5 dnl  This file is part of the GNU MP Library.
   6 dnl
   7 dnl  The GNU MP Library is free software; you can redistribute it and/or modify
   8 dnl  it under the terms of either:
   9 dnl
  10 dnl    * the GNU Lesser General Public License as published by the Free
  11 dnl      Software Foundation; either version 3 of the License, or (at your
  12 dnl      option) any later version.
  13 dnl
  14 dnl  or
  15 dnl
  16 dnl    * the GNU General Public License as published by the Free Software
  17 dnl      Foundation; either version 2 of the License, or (at your option) any
  18 dnl      later version.
  19 dnl
  20 dnl  or both in parallel, as here.
  21 dnl
  22 dnl  The GNU MP Library is distributed in the hope that it will be useful, but
  23 dnl  WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
  24 dnl  or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
  25 dnl  for more details.
  26 dnl
  27 dnl  You should have received copies of the GNU General Public License and the
  28 dnl  GNU Lesser General Public License along with the GNU MP Library.  If not,
  29 dnl  see https://www.gnu.org/licenses/.
  30
  31 include(`../config.m4')
  32
  33 C      cycles/limb
  34 C EV4:     ?
  35 C EV5:   137/140  (with BWX/without BWX)
  36 C EV6:    71/72   (with BWX/without BWX)
  37
  38 C This was compiler generated, with minimal manual edits.  Surely several
  39 C cycles could be cut with some thought.
  40
  41 ASM_START()
  42 PROLOGUE(mpn_invert_limb,gp)
  43         LEA(    r2, approx_tab)
  44         srl     r16, 54, r1
  45         srl     r16, 24, r4
  46         and     r16, 1, r5
  47         bic     r1, 1, r7
  48         lda     r4, 1(r4)
  49         srl     r16, 1, r3
  50         addq    r7, r2, r1
  51 ifelse(bwx_available_p,1,`
  52         ldwu    r0, -512(r1)
  53 ',`
  54         ldq_u   r0, -512(r1)
  55         extwl   r0, r7, r0
  56 ')
  57         addq    r3, r5, r3
  58         mull    r0, r0, r1
  59         sll     r0, 11, r0
  60         mulq    r1, r4, r1
  61         srl     r1, 40, r1
  62         subq    r0, r1, r0
  63         lda     r0, -1(r0)
  64         mulq    r0, r0, r2
  65         sll     r0, 60, r1
  66         sll     r0, 13, r0
  67         mulq    r2, r4, r2
  68         subq    r1, r2, r1
  69         srl     r1, 47, r1
  70         addq    r0, r1, r0
  71         mulq    r0, r3, r3
  72         srl     r0, 1, r1
  73         cmoveq  r5, 0, r1
  74         subq    r1, r3, r1
  75         umulh   r1, r0, r3
  76         sll     r0, 31, r0
  77         srl     r3, 1, r1
  78         addq    r0, r1, r0
  79         mulq    r0, r16, r2
  80         umulh   r0, r16, r3
  81         addq    r2, r16, r1
  82         addq    r3, r16, r16
  83         cmpult  r1, r2, r1
  84         addq    r16, r1, r3
  85         subq    r0, r3, r0
  86         ret     r31, (r26), 1
  87 EPILOGUE()
  88 DATASTART(approx_tab,8)
  89 forloop(i,256,512-1,dnl
  90 `       .word   eval(0x7fd00/i)
  91 ')dnl
  92         SIZE(approx_tab, 512)
  93         TYPE(approx_tab, object)
  94 DATAEND()
  95 ASM_END()