aacdec_mips.h 11.1 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 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
/*
 * Copyright (c) 2012
 *      MIPS Technologies, Inc., California.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 * 3. Neither the name of the MIPS Technologies, Inc., nor the names of its
 *    contributors may be used to endorse or promote products derived from
 *    this software without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE MIPS TECHNOLOGIES, INC. ``AS IS'' AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED.  IN NO EVENT SHALL THE MIPS TECHNOLOGIES, INC. BE LIABLE
 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 * SUCH DAMAGE.
 *
 * Authors:  Darko Laus      (darko@mips.com)
 *           Djordje Pesut   (djordje@mips.com)
 *           Mirjana Vulin   (mvulin@mips.com)
 *
 * AAC Spectral Band Replication decoding functions optimized for MIPS
 *
 * This file is part of FFmpeg.
 *
 * FFmpeg is free software; you can redistribute it and/or
 * 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.
 *
 * FFmpeg is distributed in the hope that it will be useful,
 * 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
 * License along with FFmpeg; if not, write to the Free Software
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

/**
 * @file
 * Reference: libavcodec/aacdec.c
 */

57 58
#ifndef AVCODEC_MIPS_AACDEC_MIPS_H
#define AVCODEC_MIPS_AACDEC_MIPS_H
59 60

#include "libavcodec/aac.h"
61
#include "libavutil/mips/asmdefs.h"
62 63

#if HAVE_INLINE_ASM && HAVE_MIPSFPU
64
#if !HAVE_MIPS32R6 && !HAVE_MIPS64R6
65 66 67 68 69 70 71 72
static inline float *VMUL2_mips(float *dst, const float *v, unsigned idx,
                           const float *scale)
{
    float temp0, temp1, temp2;
    int temp3, temp4;
    float *ret;

    __asm__ volatile(
73 74
        "andi    %[temp3],  %[idx],       0x0F         \n\t"
        "andi    %[temp4],  %[idx],       0xF0         \n\t"
75
        "sll     %[temp3],  %[temp3],     2            \n\t"
76
        "srl     %[temp4],  %[temp4],     2            \n\t"
77 78 79 80 81
        "lwc1    %[temp2],  0(%[scale])                \n\t"
        "lwxc1   %[temp0],  %[temp3](%[v])             \n\t"
        "lwxc1   %[temp1],  %[temp4](%[v])             \n\t"
        "mul.s   %[temp0],  %[temp0],     %[temp2]     \n\t"
        "mul.s   %[temp1],  %[temp1],     %[temp2]     \n\t"
82
        PTR_ADDIU "%[ret],  %[dst],       8            \n\t"
83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103
        "swc1    %[temp0],  0(%[dst])                  \n\t"
        "swc1    %[temp1],  4(%[dst])                  \n\t"

        : [temp0]"=&f"(temp0), [temp1]"=&f"(temp1),
          [temp2]"=&f"(temp2), [temp3]"=&r"(temp3),
          [temp4]"=&r"(temp4), [ret]"=&r"(ret)
        : [idx]"r"(idx), [scale]"r"(scale), [v]"r"(v),
          [dst]"r"(dst)
        : "memory"
    );
    return ret;
}

static inline float *VMUL4_mips(float *dst, const float *v, unsigned idx,
                           const float *scale)
{
    int temp0, temp1, temp2, temp3;
    float temp4, temp5, temp6, temp7, temp8;
    float *ret;

    __asm__ volatile(
104 105 106 107
        "andi    %[temp0],  %[idx],       0x03        \n\t"
        "andi    %[temp1],  %[idx],       0x0C        \n\t"
        "andi    %[temp2],  %[idx],       0x30        \n\t"
        "andi    %[temp3],  %[idx],       0xC0        \n\t"
108
        "sll     %[temp0],  %[temp0],     2           \n\t"
109 110
        "srl     %[temp2],  %[temp2],     2           \n\t"
        "srl     %[temp3],  %[temp3],     4           \n\t"
111 112 113 114 115 116 117 118 119
        "lwc1    %[temp4],  0(%[scale])               \n\t"
        "lwxc1   %[temp5],  %[temp0](%[v])            \n\t"
        "lwxc1   %[temp6],  %[temp1](%[v])            \n\t"
        "lwxc1   %[temp7],  %[temp2](%[v])            \n\t"
        "lwxc1   %[temp8],  %[temp3](%[v])            \n\t"
        "mul.s   %[temp5],  %[temp5],     %[temp4]    \n\t"
        "mul.s   %[temp6],  %[temp6],     %[temp4]    \n\t"
        "mul.s   %[temp7],  %[temp7],     %[temp4]    \n\t"
        "mul.s   %[temp8],  %[temp8],     %[temp4]    \n\t"
120
        PTR_ADDIU "%[ret],  %[dst],       16          \n\t"
121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145
        "swc1    %[temp5],  0(%[dst])                 \n\t"
        "swc1    %[temp6],  4(%[dst])                 \n\t"
        "swc1    %[temp7],  8(%[dst])                 \n\t"
        "swc1    %[temp8],  12(%[dst])                \n\t"

        : [temp0]"=&r"(temp0), [temp1]"=&r"(temp1),
          [temp2]"=&r"(temp2), [temp3]"=&r"(temp3),
          [temp4]"=&f"(temp4), [temp5]"=&f"(temp5),
          [temp6]"=&f"(temp6), [temp7]"=&f"(temp7),
          [temp8]"=&f"(temp8), [ret]"=&r"(ret)
        : [idx]"r"(idx), [scale]"r"(scale), [v]"r"(v),
          [dst]"r"(dst)
        : "memory"
    );
    return ret;
}

static inline float *VMUL2S_mips(float *dst, const float *v, unsigned idx,
                            unsigned sign, const float *scale)
{
    int temp0, temp1, temp2, temp3, temp4, temp5;
    float temp6, temp7, temp8, temp9;
    float *ret;

    __asm__ volatile(
146 147
        "andi    %[temp0],  %[idx],       0x0F       \n\t"
        "andi    %[temp1],  %[idx],       0xF0       \n\t"
148 149 150 151 152
        "lw      %[temp4],  0(%[scale])              \n\t"
        "srl     %[temp2],  %[sign],      1          \n\t"
        "sll     %[temp3],  %[sign],      31         \n\t"
        "sll     %[temp2],  %[temp2],     31         \n\t"
        "sll     %[temp0],  %[temp0],     2          \n\t"
153
        "srl     %[temp1],  %[temp1],     2          \n\t"
154 155 156 157 158 159 160 161
        "lwxc1   %[temp8],  %[temp0](%[v])           \n\t"
        "lwxc1   %[temp9],  %[temp1](%[v])           \n\t"
        "xor     %[temp5],  %[temp4],     %[temp2]   \n\t"
        "xor     %[temp4],  %[temp4],     %[temp3]   \n\t"
        "mtc1    %[temp5],  %[temp6]                 \n\t"
        "mtc1    %[temp4],  %[temp7]                 \n\t"
        "mul.s   %[temp8],  %[temp8],     %[temp6]   \n\t"
        "mul.s   %[temp9],  %[temp9],     %[temp7]   \n\t"
162
        PTR_ADDIU "%[ret],  %[dst],       8          \n\t"
163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188
        "swc1    %[temp8],  0(%[dst])                \n\t"
        "swc1    %[temp9],  4(%[dst])                \n\t"

        : [temp0]"=&r"(temp0), [temp1]"=&r"(temp1),
          [temp2]"=&r"(temp2), [temp3]"=&r"(temp3),
          [temp4]"=&r"(temp4), [temp5]"=&r"(temp5),
          [temp6]"=&f"(temp6), [temp7]"=&f"(temp7),
          [temp8]"=&f"(temp8), [temp9]"=&f"(temp9),
          [ret]"=&r"(ret)
        : [idx]"r"(idx), [scale]"r"(scale), [v]"r"(v),
          [dst]"r"(dst), [sign]"r"(sign)
        : "memory"
    );
    return ret;
}

static inline float *VMUL4S_mips(float *dst, const float *v, unsigned idx,
                            unsigned sign, const float *scale)
{
    int temp0, temp1, temp2, temp3, temp4;
    float temp10, temp11, temp12, temp13, temp14, temp15, temp16, temp17;
    float *ret;
    unsigned int mask = 1U << 31;

    __asm__ volatile(
        "lw      %[temp0],   0(%[scale])               \n\t"
189 190 191 192 193 194 195
        "andi    %[temp1],  %[idx],       0x03         \n\t"
        "andi    %[temp2],  %[idx],       0x0C         \n\t"
        "andi    %[temp3],  %[idx],       0x30         \n\t"
        "andi    %[temp4],  %[idx],       0xC0         \n\t"
        "sll     %[temp1],  %[temp1],     2            \n\t"
        "srl     %[temp3],  %[temp3],     2            \n\t"
        "srl     %[temp4],  %[temp4],     4            \n\t"
196 197 198 199 200
        "lwxc1   %[temp10],  %[temp1](%[v])            \n\t"
        "lwxc1   %[temp11],  %[temp2](%[v])            \n\t"
        "lwxc1   %[temp12],  %[temp3](%[v])            \n\t"
        "lwxc1   %[temp13],  %[temp4](%[v])            \n\t"
        "and     %[temp1],   %[sign],      %[mask]     \n\t"
201 202 203 204 205 206
        "srl     %[temp2],   %[idx],       12          \n\t"
        "srl     %[temp3],   %[idx],       13          \n\t"
        "srl     %[temp4],   %[idx],       14          \n\t"
        "andi    %[temp2],   %[temp2],     1           \n\t"
        "andi    %[temp3],   %[temp3],     1           \n\t"
        "andi    %[temp4],   %[temp4],     1           \n\t"
207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224
        "sllv    %[sign],    %[sign],      %[temp2]    \n\t"
        "xor     %[temp1],   %[temp0],     %[temp1]    \n\t"
        "and     %[temp2],   %[sign],      %[mask]     \n\t"
        "mtc1    %[temp1],   %[temp14]                 \n\t"
        "xor     %[temp2],   %[temp0],     %[temp2]    \n\t"
        "sllv    %[sign],    %[sign],      %[temp3]    \n\t"
        "mtc1    %[temp2],   %[temp15]                 \n\t"
        "and     %[temp3],   %[sign],      %[mask]     \n\t"
        "sllv    %[sign],    %[sign],      %[temp4]    \n\t"
        "xor     %[temp3],   %[temp0],     %[temp3]    \n\t"
        "and     %[temp4],   %[sign],      %[mask]     \n\t"
        "mtc1    %[temp3],   %[temp16]                 \n\t"
        "xor     %[temp4],   %[temp0],     %[temp4]    \n\t"
        "mtc1    %[temp4],   %[temp17]                 \n\t"
        "mul.s   %[temp10],  %[temp10],    %[temp14]   \n\t"
        "mul.s   %[temp11],  %[temp11],    %[temp15]   \n\t"
        "mul.s   %[temp12],  %[temp12],    %[temp16]   \n\t"
        "mul.s   %[temp13],  %[temp13],    %[temp17]   \n\t"
225
        PTR_ADDIU "%[ret],   %[dst],       16          \n\t"
226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249
        "swc1    %[temp10],  0(%[dst])                 \n\t"
        "swc1    %[temp11],  4(%[dst])                 \n\t"
        "swc1    %[temp12],  8(%[dst])                 \n\t"
        "swc1    %[temp13],  12(%[dst])                \n\t"

        : [temp0]"=&r"(temp0), [temp1]"=&r"(temp1),
          [temp2]"=&r"(temp2), [temp3]"=&r"(temp3),
          [temp4]"=&r"(temp4), [temp10]"=&f"(temp10),
          [temp11]"=&f"(temp11), [temp12]"=&f"(temp12),
          [temp13]"=&f"(temp13), [temp14]"=&f"(temp14),
          [temp15]"=&f"(temp15), [temp16]"=&f"(temp16),
          [temp17]"=&f"(temp17), [ret]"=&r"(ret),
          [sign]"+r"(sign)
        : [idx]"r"(idx), [scale]"r"(scale), [v]"r"(v),
          [dst]"r"(dst), [mask]"r"(mask)
        : "memory"
    );
    return ret;
}

#define VMUL2 VMUL2_mips
#define VMUL4 VMUL4_mips
#define VMUL2S VMUL2S_mips
#define VMUL4S VMUL4S_mips
250
#endif /* !HAVE_MIPS32R6 && !HAVE_MIPS64R6 */
251 252
#endif /* HAVE_INLINE_ASM && HAVE_MIPSFPU */

253
#endif /* AVCODEC_MIPS_AACDEC_MIPS_H */