source/libs/gmp/gmp-src/mpn/arm/v6/submul_1.asm

   1 dnl  ARM mpn_submul_1.
   2
   3 dnl  Copyright 2012, 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 StrongARM:     -
  35 C XScale         -
  36 C Cortex-A7      ?
  37 C Cortex-A8      ?
  38 C Cortex-A9      3.75
  39 C Cortex-A15     4.0
  40
  41 C This loop complements U on the fly,
  42 C   U' = B^n - 1 - U
  43 C and then uses that
  44 C   R - U*v = R + U'*v + v - B^n v
  45
  46 C TODO
  47 C  * Micro-optimise feed-in code.
  48 C  * Optimise for n=1,2 by delaying register saving.
  49 C  * Try using ldm/stm.
  50
  51 define(`rp',`r0')
  52 define(`up',`r1')
  53 define(`n', `r2')
  54 define(`v0',`r3')
  55
  56 ASM_START()
  57 PROLOGUE(mpn_submul_1)
  58         stmfd   sp!, { r4, r5, r6, r7 }
  59
  60         ands    r6, n, #3
  61         mov     r12, v0
  62         beq     L(fi0)
  63         cmp     r6, #2
  64         bcc     L(fi1)
  65         beq     L(fi2)
  66
  67 L(fi3): ldr     r4, [up], #12
  68         mvn     r4, r4
  69         ldr     r6, [rp, #0]
  70         ldr     r5, [up, #-8]
  71         b       L(lo3)
  72
  73 L(fi0): ldr     r5, [up], #16
  74         mvn     r5, r5
  75         ldr     r7, [rp], #4
  76         ldr     r4, [up, #-12]
  77         b       L(lo0)
  78
  79 L(fi1): ldr     r4, [up], #4
  80         mvn     r4, r4
  81         ldr     r6, [rp], #8
  82         subs    n, n, #1
  83         beq     L(1)
  84         ldr     r5, [up]
  85         b       L(lo1)
  86
  87 L(fi2): ldr     r5, [up], #8
  88         mvn     r5, r5
  89         ldr     r7, [rp], #12
  90         ldr     r4, [up, #-4]
  91         b       L(lo2)
  92
  93         ALIGN(16)
  94 L(top): ldr     r6, [rp, #-8]
  95         ldr     r5, [up]
  96         str     r7, [rp, #-12]
  97 L(lo1): umaal   r6, r12, r4, v0
  98         add     up, up, #16
  99         mvn     r5, r5
 100         ldr     r7, [rp, #-4]
 101         ldr     r4, [up, #-12]
 102         str     r6, [rp, #-8]
 103 L(lo0): umaal   r7, r12, r5, v0
 104         mvn     r4, r4
 105         ldr     r6, [rp, #0]
 106         ldr     r5, [up, #-8]
 107         str     r7, [rp, #-4]
 108 L(lo3): umaal   r6, r12, r4, v0
 109         mvn     r5, r5
 110         ldr     r7, [rp, #4]
 111         ldr     r4, [up, #-4]
 112         str     r6, [rp], #16
 113 L(lo2): umaal   r7, r12, r5, v0
 114         mvn     r4, r4
 115         subs    n, n, #4
 116         bhi     L(top)
 117
 118         ldr     r6, [rp, #-8]
 119         str     r7, [rp, #-12]
 120 L(1):   umaal   r6, r12, r4, v0
 121         str     r6, [rp, #-8]
 122         sub     r0, v0, r12
 123         ldmfd   sp!, { r4, r5, r6, r7 }
 124         bx      lr
 125 EPILOGUE()