source/libs/gmp/gmp-src/mpn/x86_64/bobcat/aors_n.asm

   1 dnl  AMD64 mpn_add_n, mpn_sub_n optimised for bobcat.
   2
   3 dnl  Copyright 2003-2005, 2007, 2008, 2010-2013 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 AMD K8,K9
  35 C AMD K10
  36 C AMD bd1
  37 C AMD bobcat     2.28
  38 C Intel P4
  39 C Intel core2
  40 C Intel NHM
  41 C Intel SBR
  42 C Intel IBR
  43 C Intel atom
  44 C VIA nano
  45
  46 C The loop of this code is the result of running a code generation and
  47 C optimization tool suite written by David Harvey and Torbjorn Granlund.
  48
  49 C INPUT PARAMETERS
  50 define(`rp',    `%rdi') C rcx
  51 define(`up',    `%rsi') C rdx
  52 define(`vp',    `%rdx') C r8
  53 define(`n',     `%rcx') C r9
  54 define(`cy',    `%r8')  C rsp+40    (mpn_add_nc and mpn_sub_nc)
  55
  56 ifdef(`OPERATION_add_n', `
  57         define(ADCSBB,        adc)
  58         define(func,          mpn_add_n)
  59         define(func_nc,       mpn_add_nc)')
  60 ifdef(`OPERATION_sub_n', `
  61         define(ADCSBB,        sbb)
  62         define(func,          mpn_sub_n)
  63         define(func_nc,       mpn_sub_nc)')
  64
  65 MULFUNC_PROLOGUE(mpn_add_n mpn_add_nc mpn_sub_n mpn_sub_nc)
  66
  67 ABI_SUPPORT(DOS64)
  68 ABI_SUPPORT(STD64)
  69
  70 ASM_START()
  71         TEXT
  72         ALIGN(16)
  73 PROLOGUE(func)
  74         FUNC_ENTRY(4)
  75         xor     %r8, %r8
  76 L(ent): test    $1, R8(n)
  77         jnz     L(bx1)
  78
  79 L(bx0): test    $2, R8(n)
  80         jnz     L(b10)
  81
  82 L(b00): shr     $2, n
  83         neg     %r8
  84         mov     $3, R32(%rax)
  85         mov     (up), %r10
  86         mov     8(up), %r11
  87         jmp     L(lo0)
  88
  89 L(b10): shr     $2, n
  90         neg     %r8
  91         mov     $1, R32(%rax)
  92         mov     (up), %r8
  93         mov     8(up), %r9
  94         jrcxz   L(cj2)
  95         jmp     L(top)
  96
  97 L(bx1): test    $2, R8(n)
  98         jnz     L(b11)
  99
 100 L(b01): shr     $2, n
 101         neg     %r8
 102         mov     $0, R32(%rax)
 103         mov     (up), %r9
 104         jrcxz   L(cj1)
 105         mov     8(up), %r10
 106         jmp     L(lo1)
 107
 108         ALIGN(8)
 109 L(b11): inc     n
 110         shr     $2, n
 111         neg     %r8
 112         mov     $2, R32(%rax)
 113         mov     (up), %r11
 114         jmp     L(lo3)
 115
 116         ALIGN(4)
 117 L(top): mov     8(up,%rax,8), %r10
 118         ADCSBB  -8(vp,%rax,8), %r8
 119         mov     %r8, -8(rp,%rax,8)
 120 L(lo1): mov     16(up,%rax,8), %r11
 121         ADCSBB  (vp,%rax,8), %r9
 122         lea     4(%rax), %rax
 123         mov     %r9, -32(rp,%rax,8)
 124 L(lo0): ADCSBB  -24(vp,%rax,8), %r10
 125         mov     %r10, -24(rp,%rax,8)
 126 L(lo3): ADCSBB  -16(vp,%rax,8), %r11
 127         dec     n
 128         mov     -8(up,%rax,8), %r8
 129         mov     %r11, -16(rp,%rax,8)
 130 L(lo2): mov     (up,%rax,8), %r9
 131         jnz     L(top)
 132
 133 L(cj2): ADCSBB  -8(vp,%rax,8), %r8
 134         mov     %r8, -8(rp,%rax,8)
 135 L(cj1): ADCSBB  (vp,%rax,8), %r9
 136         mov     %r9, (rp,%rax,8)
 137
 138         mov     $0, R32(%rax)
 139         adc     $0, R32(%rax)
 140
 141         FUNC_EXIT()
 142         ret
 143 EPILOGUE()
 144
 145         ALIGN(16)
 146 PROLOGUE(func_nc)
 147         FUNC_ENTRY(4)
 148 IFDOS(` mov     56(%rsp), %r8   ')
 149         jmp     L(ent)
 150 EPILOGUE()