source/libs/gmp/gmp-src/mpn/x86_64/core2/rshift.asm

   1 dnl  x86-64 mpn_rshift optimized for "Core 2".
   2
   3 dnl  Copyright 2007, 2009, 2011, 2012 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
  34 C            cycles/limb
  35 C AMD K8,K9      4.25
  36 C AMD K10        4.25
  37 C Intel P4      14.7
  38 C Intel core2    1.27
  39 C Intel NHM      1.375  (up to about n = 260, then 1.5)
  40 C Intel SBR      1.77
  41 C Intel atom     ?
  42 C VIA nano       ?
  43
  44
  45 C INPUT PARAMETERS
  46 define(`rp',    `%rdi')
  47 define(`up',    `%rsi')
  48 define(`n',     `%rdx')
  49 define(`cnt',   `%rcx')
  50
  51 ABI_SUPPORT(DOS64)
  52 ABI_SUPPORT(STD64)
  53
  54 ASM_START()
  55         TEXT
  56         ALIGN(16)
  57 PROLOGUE(mpn_rshift)
  58         FUNC_ENTRY(4)
  59         mov     R32(%rdx), R32(%rax)
  60         and     $3, R32(%rax)
  61         jne     L(nb00)
  62 L(b00): C n = 4, 8, 12, ...
  63         mov     (up), %r10
  64         mov     8(up), %r11
  65         xor     R32(%rax), R32(%rax)
  66         shrd    R8(cnt), %r10, %rax
  67         mov     16(up), %r8
  68         lea     8(up), up
  69         lea     -24(rp), rp
  70         sub     $4, n
  71         jmp     L(00)
  72
  73 L(nb00):C n = 1, 5, 9, ...
  74         cmp     $2, R32(%rax)
  75         jae     L(nb01)
  76 L(b01): mov     (up), %r9
  77         xor     R32(%rax), R32(%rax)
  78         shrd    R8(cnt), %r9, %rax
  79         sub     $2, n
  80         jb      L(le1)
  81         mov     8(up), %r10
  82         mov     16(up), %r11
  83         lea     16(up), up
  84         lea     -16(rp), rp
  85         jmp     L(01)
  86 L(le1): shr     R8(cnt), %r9
  87         mov     %r9, (rp)
  88         FUNC_EXIT()
  89         ret
  90
  91 L(nb01):C n = 2, 6, 10, ...
  92         jne     L(b11)
  93 L(b10): mov     (up), %r8
  94         mov     8(up), %r9
  95         xor     R32(%rax), R32(%rax)
  96         shrd    R8(cnt), %r8, %rax
  97         sub     $3, n
  98         jb      L(le2)
  99         mov     16(up), %r10
 100         lea     24(up), up
 101         lea     -8(rp), rp
 102         jmp     L(10)
 103 L(le2): shrd    R8(cnt), %r9, %r8
 104         mov     %r8, (rp)
 105         shr     R8(cnt), %r9
 106         mov     %r9, 8(rp)
 107         FUNC_EXIT()
 108         ret
 109
 110         ALIGN(16)
 111 L(b11): C n = 3, 7, 11, ...
 112         mov     (up), %r11
 113         mov     8(up), %r8
 114         xor     R32(%rax), R32(%rax)
 115         shrd    R8(cnt), %r11, %rax
 116         mov     16(up), %r9
 117         lea     32(up), up
 118         sub     $4, n
 119         jb      L(end)
 120
 121         ALIGN(16)
 122 L(top): shrd    R8(cnt), %r8, %r11
 123         mov     -8(up), %r10
 124         mov     %r11, (rp)
 125 L(10):  shrd    R8(cnt), %r9, %r8
 126         mov     (up), %r11
 127         mov     %r8, 8(rp)
 128 L(01):  shrd    R8(cnt), %r10, %r9
 129         mov     8(up), %r8
 130         mov     %r9, 16(rp)
 131 L(00):  shrd    R8(cnt), %r11, %r10
 132         mov     16(up), %r9
 133         mov     %r10, 24(rp)
 134         add     $32, up
 135         lea     32(rp), rp
 136         sub     $4, n
 137         jnc     L(top)
 138
 139 L(end): shrd    R8(cnt), %r8, %r11
 140         mov     %r11, (rp)
 141         shrd    R8(cnt), %r9, %r8
 142         mov     %r8, 8(rp)
 143         shr     R8(cnt), %r9
 144         mov     %r9, 16(rp)
 145         FUNC_EXIT()
 146         ret
 147 EPILOGUE()