source/libs/gmp/gmp-src/mpn/arm/mod_34lsub1.asm

   1 dnl  ARM mpn_mod_34lsub1 -- remainder modulo 2^24-1.
   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      1.33
  39 C Cortex-A15     1.33
  40
  41 define(`ap',    r0)
  42 define(`n',     r1)
  43
  44 C mp_limb_t mpn_mod_34lsub1 (mp_srcptr up, mp_size_t n)
  45
  46 C TODO
  47 C  * Write cleverer summation code.
  48 C  * Consider loading 6 64-bit aligned registers at a time, to approach 1 c/l.
  49
  50 ASM_START()
  51         TEXT
  52         ALIGN(32)
  53 PROLOGUE(mpn_mod_34lsub1)
  54         push    { r4, r5, r6, r7 }
  55
  56         subs    n, n, #3
  57         mov     r7, #0
  58         blt     L(le2)                  C n <= 2
  59
  60         ldmia   ap!, { r2, r3, r12 }
  61         subs    n, n, #3
  62         blt     L(sum)                  C n <= 5
  63         cmn     r0, #0                  C clear carry
  64         sub     n, n, #3
  65         b       L(mid)
  66
  67 L(top): adcs    r2, r2, r4
  68         adcs    r3, r3, r5
  69         adcs    r12, r12, r6
  70 L(mid): ldmia   ap!, { r4, r5, r6 }
  71         tst     n, n
  72         sub     n, n, #3
  73         bpl     L(top)
  74
  75         add     n, n, #3
  76
  77         adcs    r2, r2, r4
  78         adcs    r3, r3, r5
  79         adcs    r12, r12, r6
  80         movcs   r7, #1                  C r7 <= 1
  81
  82 L(sum): cmn     n, #2
  83         movlo   r4, #0
  84         ldrhs   r4, [ap], #4
  85         movls   r5, #0
  86         ldrhi   r5, [ap], #4
  87
  88         adds    r2, r2, r4
  89         adcs    r3, r3, r5
  90         adcs    r12, r12, #0
  91         adc     r7, r7, #0              C r7 <= 2
  92
  93 L(sum2):
  94         bic     r0, r2, #0xff000000
  95         add     r0, r0, r2, lsr #24
  96         add     r0, r0, r7
  97
  98         mov     r7, r3, lsl #8
  99         bic     r1, r7, #0xff000000
 100         add     r0, r0, r1
 101         add     r0, r0, r3, lsr #16
 102
 103         mov     r7, r12, lsl #16
 104         bic     r1, r7, #0xff000000
 105         add     r0, r0, r1
 106         add     r0, r0, r12, lsr #8
 107
 108         pop     { r4, r5, r6, r7 }
 109         bx      lr
 110
 111 L(le2): cmn     n, #1
 112         bne     L(1)
 113         ldmia   ap!, { r2, r3 }
 114         mov     r12, #0
 115         b       L(sum2)
 116 L(1):   ldr     r2, [ap]
 117         bic     r0, r2, #0xff000000
 118         add     r0, r0, r2, lsr #24
 119         pop     { r4, r5, r6, r7 }
 120         bx      lr
 121 EPILOGUE()