DXR will be turned off on Tuesday, December 29th. It will redirect to Searchfox.
See the announcement on Discourse.

DXR is a code search and navigation tool aimed at making sense of large projects. It supports full-text and regex searches as well as structural queries.

Mercurial (d38398e5144e)

VCS Links

Line Code
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 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225
/*
 *  Copyright (c) 2015 The WebM project authors. All Rights Reserved.
 *
 *  Use of this source code is governed by a BSD-style license
 *  that can be found in the LICENSE file in the root of the source
 *  tree. An additional intellectual property rights grant can be found
 *  in the file PATENTS.  All contributing project authors may
 *  be found in the AUTHORS file in the root of the source tree.
 */

#include "vpx_dsp/mips/common_dspr2.h"

#if HAVE_DSPR2
void vpx_h_predictor_4x4_dspr2(uint8_t *dst, ptrdiff_t stride,
                               const uint8_t *above, const uint8_t *left) {
  int32_t tmp1, tmp2, tmp3, tmp4;
  (void)above;

  __asm__ __volatile__(
      "lb         %[tmp1],      (%[left])                    \n\t"
      "lb         %[tmp2],      1(%[left])                   \n\t"
      "lb         %[tmp3],      2(%[left])                   \n\t"
      "lb         %[tmp4],      3(%[left])                   \n\t"
      "replv.qb   %[tmp1],      %[tmp1]                      \n\t"
      "replv.qb   %[tmp2],      %[tmp2]                      \n\t"
      "replv.qb   %[tmp3],      %[tmp3]                      \n\t"
      "replv.qb   %[tmp4],      %[tmp4]                      \n\t"
      "sw         %[tmp1],      (%[dst])                     \n\t"
      "add        %[dst],       %[dst],         %[stride]    \n\t"
      "sw         %[tmp2],      (%[dst])                     \n\t"
      "add        %[dst],       %[dst],         %[stride]    \n\t"
      "sw         %[tmp3],      (%[dst])                     \n\t"
      "add        %[dst],       %[dst],         %[stride]    \n\t"
      "sw         %[tmp4],      (%[dst])                     \n\t"

      : [tmp1] "=&r"(tmp1), [tmp2] "=&r"(tmp2), [tmp3] "=&r"(tmp3),
        [tmp4] "=&r"(tmp4)
      : [left] "r"(left), [dst] "r"(dst), [stride] "r"(stride));
}

void vpx_dc_predictor_4x4_dspr2(uint8_t *dst, ptrdiff_t stride,
                                const uint8_t *above, const uint8_t *left) {
  int32_t expected_dc;
  int32_t average;
  int32_t tmp, above_c, above_l, above_r, left_c, left_r, left_l;

  __asm__ __volatile__(
      "lw              %[above_c],         (%[above])                    \n\t"
      "lw              %[left_c],          (%[left])                     \n\t"

      "preceu.ph.qbl   %[above_l],         %[above_c]                    \n\t"
      "preceu.ph.qbr   %[above_r],         %[above_c]                    \n\t"
      "preceu.ph.qbl   %[left_l],          %[left_c]                     \n\t"
      "preceu.ph.qbr   %[left_r],          %[left_c]                     \n\t"

      "addu.ph         %[average],         %[above_r],       %[above_l]  \n\t"
      "addu.ph         %[average],         %[average],       %[left_l]   \n\t"
      "addu.ph         %[average],         %[average],       %[left_r]   \n\t"
      "addiu           %[average],         %[average],       4           \n\t"
      "srl             %[tmp],             %[average],       16          \n\t"
      "addu.ph         %[average],         %[tmp],           %[average]  \n\t"
      "srl             %[expected_dc],     %[average],       3           \n\t"
      "replv.qb        %[expected_dc],     %[expected_dc]                \n\t"

      "sw              %[expected_dc],     (%[dst])                      \n\t"
      "add             %[dst],              %[dst],          %[stride]   \n\t"
      "sw              %[expected_dc],     (%[dst])                      \n\t"
      "add             %[dst],              %[dst],          %[stride]   \n\t"
      "sw              %[expected_dc],     (%[dst])                      \n\t"
      "add             %[dst],              %[dst],          %[stride]   \n\t"
      "sw              %[expected_dc],     (%[dst])                      \n\t"

      : [above_c] "=&r"(above_c), [above_l] "=&r"(above_l),
        [above_r] "=&r"(above_r), [left_c] "=&r"(left_c),
        [left_l] "=&r"(left_l), [left_r] "=&r"(left_r),
        [average] "=&r"(average), [tmp] "=&r"(tmp),
        [expected_dc] "=&r"(expected_dc)
      : [above] "r"(above), [left] "r"(left), [dst] "r"(dst),
        [stride] "r"(stride));
}

void vpx_tm_predictor_4x4_dspr2(uint8_t *dst, ptrdiff_t stride,
                                const uint8_t *above, const uint8_t *left) {
  int32_t abovel, abover;
  int32_t left0, left1, left2, left3;
  int32_t res0, res1;
  int32_t resl;
  int32_t resr;
  int32_t top_left;
  uint8_t *cm = vpx_ff_cropTbl;

  __asm__ __volatile__(
      "ulw             %[resl],       (%[above])                         \n\t"

      "lbu             %[left0],       (%[left])                         \n\t"
      "lbu             %[left1],       1(%[left])                        \n\t"
      "lbu             %[left2],       2(%[left])                        \n\t"
      "lbu             %[left3],       3(%[left])                        \n\t"

      "lbu             %[top_left],    -1(%[above])                      \n\t"

      "preceu.ph.qbl   %[abovel],      %[resl]                           \n\t"
      "preceu.ph.qbr   %[abover],      %[resl]                           \n\t"

      "replv.ph        %[left0],       %[left0]                          \n\t"
      "replv.ph        %[left1],       %[left1]                          \n\t"
      "replv.ph        %[left2],       %[left2]                          \n\t"
      "replv.ph        %[left3],       %[left3]                          \n\t"

      "replv.ph        %[top_left],    %[top_left]                       \n\t"

      "addu.ph         %[resl],        %[abovel],         %[left0]       \n\t"
      "subu.ph         %[resl],        %[resl],           %[top_left]    \n\t"

      "addu.ph         %[resr],        %[abover],         %[left0]       \n\t"
      "subu.ph         %[resr],        %[resr],           %[top_left]    \n\t"

      "sll             %[res0],        %[resr],           16             \n\t"
      "sra             %[res0],        %[res0],           16             \n\t"
      "lbux            %[res0],        %[res0](%[cm])                    \n\t"

      "sra             %[res1],        %[resr],           16             \n\t"
      "lbux            %[res1],        %[res1](%[cm])                    \n\t"
      "sb              %[res0],        (%[dst])                          \n\t"

      "sll             %[res0],        %[resl],           16             \n\t"
      "sra             %[res0],        %[res0],           16             \n\t"
      "lbux            %[res0],        %[res0](%[cm])                    \n\t"
      "sb              %[res1],        1(%[dst])                         \n\t"

      "sra             %[res1],        %[resl],           16             \n\t"
      "lbux            %[res1],        %[res1](%[cm])                    \n\t"

      "addu.ph         %[resl],        %[abovel],         %[left1]       \n\t"
      "subu.ph         %[resl],        %[resl],           %[top_left]    \n\t"

      "addu.ph         %[resr],        %[abover],         %[left1]       \n\t"
      "subu.ph         %[resr],        %[resr],           %[top_left]    \n\t"

      "sb              %[res0],        2(%[dst])                         \n\t"
      "sb              %[res1],        3(%[dst])                         \n\t"

      "add             %[dst],          %[dst],           %[stride]      \n\t"

      "sll             %[res0],        %[resr],           16             \n\t"
      "sra             %[res0],        %[res0],           16             \n\t"
      "lbux            %[res0],        %[res0](%[cm])                    \n\t"

      "sra             %[res1],        %[resr],           16             \n\t"
      "lbux            %[res1],        %[res1](%[cm])                    \n\t"
      "sb              %[res0],        (%[dst])                          \n\t"

      "sll             %[res0],        %[resl],           16             \n\t"
      "sra             %[res0],        %[res0],           16             \n\t"
      "lbux            %[res0],        %[res0](%[cm])                    \n\t"

      "sb              %[res1],        1(%[dst])                         \n\t"
      "sra             %[res1],        %[resl],           16             \n\t"
      "lbux            %[res1],        %[res1](%[cm])                    \n\t"

      "addu.ph         %[resl],        %[abovel],         %[left2]       \n\t"
      "subu.ph         %[resl],        %[resl],           %[top_left]    \n\t"

      "addu.ph         %[resr],        %[abover],         %[left2]       \n\t"
      "subu.ph         %[resr],        %[resr],           %[top_left]    \n\t"

      "sb              %[res0],        2(%[dst])                         \n\t"
      "sb              %[res1],        3(%[dst])                         \n\t"

      "add             %[dst],          %[dst],           %[stride]      \n\t"

      "sll             %[res0],        %[resr],           16             \n\t"
      "sra             %[res0],        %[res0],           16             \n\t"
      "lbux            %[res0],        %[res0](%[cm])                    \n\t"

      "sra             %[res1],        %[resr],           16             \n\t"
      "lbux            %[res1],        %[res1](%[cm])                    \n\t"
      "sb              %[res0],        (%[dst])                          \n\t"

      "sll             %[res0],        %[resl],           16             \n\t"
      "sra             %[res0],        %[res0],           16             \n\t"
      "lbux            %[res0],        %[res0](%[cm])                    \n\t"

      "sb              %[res1],        1(%[dst])                         \n\t"
      "sra             %[res1],        %[resl],           16             \n\t"
      "lbux            %[res1],        %[res1](%[cm])                    \n\t"

      "addu.ph         %[resl],        %[abovel],        %[left3]        \n\t"
      "subu.ph         %[resl],        %[resl],          %[top_left]     \n\t"

      "addu.ph         %[resr],        %[abover],        %[left3]        \n\t"
      "subu.ph         %[resr],        %[resr],          %[top_left]     \n\t"

      "sb              %[res0],        2(%[dst])                         \n\t"
      "sb              %[res1],        3(%[dst])                         \n\t"

      "add             %[dst],          %[dst],          %[stride]       \n\t"

      "sll             %[res0],        %[resr],           16             \n\t"
      "sra             %[res0],        %[res0],           16             \n\t"
      "lbux            %[res0],        %[res0](%[cm])                    \n\t"

      "sra             %[res1],        %[resr],           16             \n\t"
      "lbux            %[res1],        %[res1](%[cm])                    \n\t"
      "sb              %[res0],        (%[dst])                          \n\t"

      "sll             %[res0],        %[resl],           16             \n\t"
      "sra             %[res0],        %[res0],           16             \n\t"
      "lbux            %[res0],        %[res0](%[cm])                    \n\t"
      "sb              %[res1],        1(%[dst])                         \n\t"

      "sra             %[res1],        %[resl],           16             \n\t"
      "lbux            %[res1],        %[res1](%[cm])                    \n\t"

      "sb              %[res0],        2(%[dst])                         \n\t"
      "sb              %[res1],        3(%[dst])                         \n\t"

      : [abovel] "=&r"(abovel), [abover] "=&r"(abover), [left0] "=&r"(left0),
        [left1] "=&r"(left1), [left2] "=&r"(left2), [res0] "=&r"(res0),
        [res1] "=&r"(res1), [left3] "=&r"(left3), [resl] "=&r"(resl),
        [resr] "=&r"(resr), [top_left] "=&r"(top_left)
      : [above] "r"(above), [left] "r"(left), [dst] "r"(dst),
        [stride] "r"(stride), [cm] "r"(cm));
}
#endif  // #if HAVE_DSPR2