mpegaudiodsp_fixed_armv6.S 4.9 KB
Newer Older
1 2 3
/*
 * Copyright (c) 2011 Mans Rullgard <mans@mansr.com>
 *
4
 * This file is part of FFmpeg.
5
 *
6
 * FFmpeg is free software; you can redistribute it and/or
7 8 9 10
 * modify it under the terms of the GNU Lesser General Public
 * License as published by the Free Software Foundation; either
 * version 2.1 of the License, or (at your option) any later version.
 *
11
 * FFmpeg is distributed in the hope that it will be useful,
12 13 14 15 16
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * Lesser General Public License for more details.
 *
 * You should have received a copy of the GNU Lesser General Public
17
 * License along with FFmpeg; if not, write to the Free Software
18 19 20
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

21
#include "libavutil/arm/asm.S"
22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77

.macro  skip            args:vararg
.endm

.macro  sum8            lo,  hi,  w, p, t1, t2, t3, t4, rsb=skip, offs=0
        ldr             \t1, [\w, #4*\offs]
        ldr             \t2, [\p, #4]!
        \rsb            \t1, \t1, #0
  .irpc i, 135
        ldr             \t3, [\w, #4*64*\i+4*\offs]
        ldr             \t4, [\p, #4*64*\i]
        smlal           \lo, \hi, \t1, \t2
        \rsb            \t3, \t3, #0
        ldr             \t1, [\w, #4*64*(\i+1)+4*\offs]
        ldr             \t2, [\p, #4*64*(\i+1)]
        smlal           \lo, \hi, \t3, \t4
        \rsb            \t1, \t1, #0
  .endr
        ldr             \t3, [\w, #4*64*7+4*\offs]
        ldr             \t4, [\p, #4*64*7]
        smlal           \lo, \hi, \t1, \t2
        \rsb            \t3, \t3, #0
        smlal           \lo, \hi, \t3, \t4
.endm

.macro  round           rd,  lo,  hi
        lsr             \rd, \lo, #24
        bic             \lo, \lo, #0xff000000
        orr             \rd, \rd, \hi, lsl #8
        mov             \hi, #0
        ssat            \rd, #16, \rd
.endm

function ff_mpadsp_apply_window_fixed_armv6, export=1
        push            {r2,r4-r11,lr}

        add             r4,  r0,  #4*512        @ synth_buf + 512
    .rept 4
        ldm             r0!, {r5-r12}
        stm             r4!, {r5-r12}
    .endr

        ldr             r4,  [sp, #40]          @ incr
        sub             r0,  r0,  #4*17         @ synth_buf + 16
        ldr             r8,  [r2]               @ sum:low
        add             r2,  r0,  #4*32         @ synth_buf + 48
        rsb             r5,  r4,  r4,  lsl #5   @ 31 * incr
        lsl             r4,  r4,  #1
        asr             r9,  r8,  #31           @ sum:high
        add             r5,  r3,  r5,  lsl #1   @ samples2
        add             r6,  r1,  #4*32         @ w2
        str             r4,  [sp, #40]

        sum8            r8,  r9,  r1,  r0,  r10, r11, r12, lr
        sum8            r8,  r9,  r1,  r2,  r10, r11, r12, lr, rsb, 32
        round           r10, r8,  r9
78
        strh_post       r10, r3,  r4
79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129

        mov             lr,  #15
1:
        ldr             r12, [r0, #4]!
        ldr             r11, [r6, #-4]!
        ldr             r10, [r1, #4]!
  .irpc i, 0246
    .if \i
        ldr             r11, [r6, #4*64*\i]
        ldr             r10, [r1, #4*64*\i]
    .endif
        rsb             r11, r11, #0
        smlal           r8,  r9,  r10, r12
        ldr             r10, [r0, #4*64*(\i+1)]
    .ifeq \i
        smull           r4, r7, r11, r12
    .else
        smlal           r4, r7, r11, r12
    .endif
        ldr             r11, [r6, #4*64*(\i+1)]
        ldr             r12, [r1, #4*64*(\i+1)]
        rsb             r11, r11, #0
        smlal           r8,  r9,  r12, r10
    .iflt \i-6
        ldr             r12, [r0, #4*64*(\i+2)]
    .else
        ldr             r12, [r2, #-4]!
    .endif
        smlal           r4,  r7,  r11, r10
  .endr
  .irpc i, 0246
        ldr             r10, [r1, #4*64*\i+4*32]
        rsb             r12, r12, #0
        ldr             r11, [r6, #4*64*\i+4*32]
        smlal           r8,  r9,  r10, r12
        ldr             r10, [r2, #4*64*(\i+1)]
        smlal           r4,  r7,  r11, r12
        ldr             r12, [r1, #4*64*(\i+1)+4*32]
        rsb             r10, r10, #0
        ldr             r11, [r6, #4*64*(\i+1)+4*32]
        smlal           r8,  r9,  r12, r10
    .iflt \i-6
        ldr             r12, [r2, #4*64*(\i+2)]
    .else
        ldr             r12, [sp, #40]
    .endif
        smlal           r4,  r7,  r11, r10
  .endr
        round           r10, r8,  r9
        adds            r8,  r8,  r4
        adc             r9,  r9,  r7
130
        strh_post       r10, r3,  r12
131 132
        round           r11, r8,  r9
        subs            lr,  lr,  #1
133
        strh_dpost      r11, r5, r12
134 135 136 137 138 139 140 141 142 143
        bgt             1b

        sum8            r8,  r9,  r1,  r0,  r10, r11, r12, lr, rsb, 33
        pop             {r4}
        round           r10, r8,  r9
        str             r8,  [r4]
        strh            r10, [r3]

        pop             {r4-r11,pc}
endfunc