File size: 10,997 Bytes
8b7c501 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 |
// Copyright 2022 Google LLC
//
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree.
$assert SAMPLE_TILE >= 1
#include <assert.h>
#include <stddef.h>
#include <stdint.h>
#include <xnnpack/math.h>
#include <xnnpack/fft.h>
void xnn_cs16_bfly4_ukernel__scalar_x${SAMPLE_TILE}(
size_t batch,
size_t samples,
int16_t* data,
const int16_t* twiddle,
size_t stride)
{
assert(batch != 0);
assert(samples != 0);
assert(samples % (sizeof(int16_t) * 2) == 0);
assert(data != NULL);
assert(stride != 0);
assert(twiddle != NULL);
int16_t* data3 = data;
do {
int16_t* data0 = data3;
int16_t* data1 = (int16_t*) ((uintptr_t) data0 + samples);
int16_t* data2 = (int16_t*) ((uintptr_t) data1 + samples);
data3 = (int16_t*) ((uintptr_t) data2 + samples);
const int16_t* tw1 = twiddle;
const int16_t* tw2 = twiddle;
const int16_t* tw3 = twiddle;
tw1 = (const int16_t*) ((uintptr_t) tw1 + stride);
tw2 = (const int16_t*) ((uintptr_t) tw2 + stride * 2);
tw3 = (const int16_t*) ((uintptr_t) tw3 + stride * 3);
size_t s = samples - sizeof(int16_t) * 2;
// First sample skips twiddle.
// Same code as samples=1 but supports stride
{
int32_t vout0r = (int32_t) data0[0];
int32_t vout0i = (int32_t) data0[1];
int32_t vout1r = (int32_t) data1[0];
int32_t vout1i = (int32_t) data1[1];
int32_t vout2r = (int32_t) data2[0];
int32_t vout2i = (int32_t) data2[1];
int32_t vout3r = (int32_t) data3[0];
int32_t vout3i = (int32_t) data3[1];
// Note 32767 / 4 = 8191. Should be 8192.
vout0r = math_asr_s32(vout0r * 8191 + 16384, 15);
vout0i = math_asr_s32(vout0i * 8191 + 16384, 15);
vout1r = math_asr_s32(vout1r * 8191 + 16384, 15);
vout1i = math_asr_s32(vout1i * 8191 + 16384, 15);
vout2r = math_asr_s32(vout2r * 8191 + 16384, 15);
vout2i = math_asr_s32(vout2i * 8191 + 16384, 15);
vout3r = math_asr_s32(vout3r * 8191 + 16384, 15);
vout3i = math_asr_s32(vout3i * 8191 + 16384, 15);
const int32_t vtmp5r = vout0r - vout2r;
const int32_t vtmp5i = vout0i - vout2i;
vout0r += vout2r;
vout0i += vout2i;
const int32_t vtmp3r = vout1r + vout3r;
const int32_t vtmp3i = vout1i + vout3i;
const int32_t vtmp4r = vout1i - vout3i;
const int32_t vtmp4i = -(vout1r - vout3r); // swap r,i and neg i
vout2r = vout0r - vtmp3r;
vout2i = vout0i - vtmp3i;
vout0r += vtmp3r;
vout0i += vtmp3i;
vout1r = vtmp5r + vtmp4r;
vout1i = vtmp5i + vtmp4i;
vout3r = vtmp5r - vtmp4r;
vout3i = vtmp5i - vtmp4i;
data0[0] = (int16_t) vout0r;
data0[1] = (int16_t) vout0i;
data1[0] = (int16_t) vout1r;
data1[1] = (int16_t) vout1i;
data2[0] = (int16_t) vout2r;
data2[1] = (int16_t) vout2i;
data3[0] = (int16_t) vout3r;
data3[1] = (int16_t) vout3i;
data0 += 2;
data1 += 2;
data2 += 2;
data3 += 2;
}
$if SAMPLE_TILE > 1:
for (; s >= ${SAMPLE_TILE} * sizeof(int16_t) * 2; s -= ${SAMPLE_TILE} * sizeof(int16_t) * 2) {
$for C in range(SAMPLE_TILE):
int32_t vout0r${C} = (int32_t) data0[${C * 2 + 0}];
int32_t vout0i${C} = (int32_t) data0[${C * 2 + 1}];
$for C in range(SAMPLE_TILE):
int32_t vout1r${C} = (int32_t) data1[${C * 2 + 0}];
int32_t vout1i${C} = (int32_t) data1[${C * 2 + 1}];
$for C in range(SAMPLE_TILE):
int32_t vout2r${C} = (int32_t) data2[${C * 2 + 0}];
int32_t vout2i${C} = (int32_t) data2[${C * 2 + 1}];
$for C in range(SAMPLE_TILE):
int32_t vout3r${C} = (int32_t) data3[${C * 2 + 0}];
int32_t vout3i${C} = (int32_t) data3[${C * 2 + 1}];
$for C in range(SAMPLE_TILE):
const int32_t vtw1r${C} = (const int32_t) tw1[0];
const int32_t vtw1i${C} = (const int32_t) tw1[1];
tw1 = (const int16_t*) ((uintptr_t) tw1 + stride);
$for C in range(SAMPLE_TILE):
const int32_t vtw2r${C} = (const int32_t) tw2[0];
const int32_t vtw2i${C} = (const int32_t) tw2[1];
tw2 = (const int16_t*) ((uintptr_t) tw2 + stride * 2);
$for C in range(SAMPLE_TILE):
const int32_t vtw3r${C} = (const int32_t) tw3[0];
const int32_t vtw3i${C} = (const int32_t) tw3[1];
tw3 = (const int16_t*) ((uintptr_t) tw3 + stride * 3);
// Note 32767 / 4 = 8191. Should be 8192.
$for C in range(SAMPLE_TILE):
vout0r${C} = math_asr_s32(vout0r${C} * 8191 + 16384, 15);
vout0i${C} = math_asr_s32(vout0i${C} * 8191 + 16384, 15);
$for C in range(SAMPLE_TILE):
vout1r${C} = math_asr_s32(vout1r${C} * 8191 + 16384, 15);
vout1i${C} = math_asr_s32(vout1i${C} * 8191 + 16384, 15);
$for C in range(SAMPLE_TILE):
vout2r${C} = math_asr_s32(vout2r${C} * 8191 + 16384, 15);
vout2i${C} = math_asr_s32(vout2i${C} * 8191 + 16384, 15);
$for C in range(SAMPLE_TILE):
vout3r${C} = math_asr_s32(vout3r${C} * 8191 + 16384, 15);
vout3i${C} = math_asr_s32(vout3i${C} * 8191 + 16384, 15);
$for C in range(SAMPLE_TILE):
const int32_t vtmp0r${C} = math_asr_s32(vout1r${C} * vtw1r${C} - vout1i${C} * vtw1i${C} + 16384, 15);
const int32_t vtmp0i${C} = math_asr_s32(vout1r${C} * vtw1i${C} + vout1i${C} * vtw1r${C} + 16384, 15);
$for C in range(SAMPLE_TILE):
const int32_t vtmp1r${C} = math_asr_s32(vout2r${C} * vtw2r${C} - vout2i${C} * vtw2i${C} + 16384, 15);
const int32_t vtmp1i${C} = math_asr_s32(vout2r${C} * vtw2i${C} + vout2i${C} * vtw2r${C} + 16384, 15);
$for C in range(SAMPLE_TILE):
const int32_t vtmp2r${C} = math_asr_s32(vout3r${C} * vtw3r${C} - vout3i${C} * vtw3i${C} + 16384, 15);
const int32_t vtmp2i${C} = math_asr_s32(vout3r${C} * vtw3i${C} + vout3i${C} * vtw3r${C} + 16384, 15);
$for C in range(SAMPLE_TILE):
const int32_t vtmp5r${C} = vout0r${C} - vtmp1r${C};
const int32_t vtmp5i${C} = vout0i${C} - vtmp1i${C};
$for C in range(SAMPLE_TILE):
vout0r${C} += vtmp1r${C};
vout0i${C} += vtmp1i${C};
$for C in range(SAMPLE_TILE):
const int32_t vtmp3r${C} = vtmp0r${C} + vtmp2r${C};
const int32_t vtmp3i${C} = vtmp0i${C} + vtmp2i${C};
$for C in range(SAMPLE_TILE):
const int32_t vtmp4r${C} = vtmp0i${C} - vtmp2i${C};
const int32_t vtmp4i${C} = -(vtmp0r${C} - vtmp2r${C}); // swap r,i and neg i
$for C in range(SAMPLE_TILE):
vout2r${C} = vout0r${C} - vtmp3r${C};
vout2i${C} = vout0i${C} - vtmp3i${C};
$for C in range(SAMPLE_TILE):
vout0r${C} += vtmp3r${C};
vout0i${C} += vtmp3i${C};
$for C in range(SAMPLE_TILE):
vout1r${C} = vtmp5r${C} + vtmp4r${C};
vout1i${C} = vtmp5i${C} + vtmp4i${C};
$for C in range(SAMPLE_TILE):
vout3r${C} = vtmp5r${C} - vtmp4r${C};
vout3i${C} = vtmp5i${C} - vtmp4i${C};
$for C in range(SAMPLE_TILE):
data0[${C * 2 + 0}] = (int16_t) vout0r${C};
data0[${C * 2 + 1}] = (int16_t) vout0i${C};
data0 += ${SAMPLE_TILE} * 2;
$for C in range(SAMPLE_TILE):
data1[${C * 2 + 0}] = (int16_t) vout1r${C};
data1[${C * 2 + 1}] = (int16_t) vout1i${C};
data1 += ${SAMPLE_TILE} * 2;
$for C in range(SAMPLE_TILE):
data2[${C * 2 + 0}] = (int16_t) vout2r${C};
data2[${C * 2 + 1}] = (int16_t) vout2i${C};
data2 += ${SAMPLE_TILE} * 2;
$for C in range(SAMPLE_TILE):
data3[${C * 2 + 0}] = (int16_t) vout3r${C};
data3[${C * 2 + 1}] = (int16_t) vout3i${C};
data3 += ${SAMPLE_TILE} * 2;
}
if XNN_UNLIKELY(s != 0) {
do {
int32_t vout0r = (int32_t) data0[0];
int32_t vout0i = (int32_t) data0[1];
int32_t vout1r = (int32_t) data1[0];
int32_t vout1i = (int32_t) data1[1];
int32_t vout2r = (int32_t) data2[0];
int32_t vout2i = (int32_t) data2[1];
int32_t vout3r = (int32_t) data3[0];
int32_t vout3i = (int32_t) data3[1];
const int32_t vtw1r = (const int32_t) tw1[0];
const int32_t vtw1i = (const int32_t) tw1[1];
const int32_t vtw2r = (const int32_t) tw2[0];
const int32_t vtw2i = (const int32_t) tw2[1];
const int32_t vtw3r = (const int32_t) tw3[0];
const int32_t vtw3i = (const int32_t) tw3[1];
tw1 = (const int16_t*) ((uintptr_t) tw1 + stride);
tw2 = (const int16_t*) ((uintptr_t) tw2 + stride * 2);
tw3 = (const int16_t*) ((uintptr_t) tw3 + stride * 3);
// Note 32767 / 4 = 8191. Should be 8192.
vout0r = math_asr_s32(vout0r * 8191 + 16384, 15);
vout0i = math_asr_s32(vout0i * 8191 + 16384, 15);
vout1r = math_asr_s32(vout1r * 8191 + 16384, 15);
vout1i = math_asr_s32(vout1i * 8191 + 16384, 15);
vout2r = math_asr_s32(vout2r * 8191 + 16384, 15);
vout2i = math_asr_s32(vout2i * 8191 + 16384, 15);
vout3r = math_asr_s32(vout3r * 8191 + 16384, 15);
vout3i = math_asr_s32(vout3i * 8191 + 16384, 15);
const int32_t vtmp0r = math_asr_s32(vout1r * vtw1r - vout1i * vtw1i + 16384, 15);
const int32_t vtmp0i = math_asr_s32(vout1r * vtw1i + vout1i * vtw1r + 16384, 15);
const int32_t vtmp1r = math_asr_s32(vout2r * vtw2r - vout2i * vtw2i + 16384, 15);
const int32_t vtmp1i = math_asr_s32(vout2r * vtw2i + vout2i * vtw2r + 16384, 15);
const int32_t vtmp2r = math_asr_s32(vout3r * vtw3r - vout3i * vtw3i + 16384, 15);
const int32_t vtmp2i = math_asr_s32(vout3r * vtw3i + vout3i * vtw3r + 16384, 15);
const int32_t vtmp5r = vout0r - vtmp1r;
const int32_t vtmp5i = vout0i - vtmp1i;
vout0r += vtmp1r;
vout0i += vtmp1i;
const int32_t vtmp3r = vtmp0r + vtmp2r;
const int32_t vtmp3i = vtmp0i + vtmp2i;
const int32_t vtmp4r = vtmp0i - vtmp2i;
const int32_t vtmp4i = -(vtmp0r - vtmp2r); // swap r,i and neg i
vout2r = vout0r - vtmp3r;
vout2i = vout0i - vtmp3i;
vout0r += vtmp3r;
vout0i += vtmp3i;
vout1r = vtmp5r + vtmp4r;
vout1i = vtmp5i + vtmp4i;
vout3r = vtmp5r - vtmp4r;
vout3i = vtmp5i - vtmp4i;
data0[0] = (int16_t) vout0r;
data0[1] = (int16_t) vout0i;
data1[0] = (int16_t) vout1r;
data1[1] = (int16_t) vout1i;
data2[0] = (int16_t) vout2r;
data2[1] = (int16_t) vout2i;
data3[0] = (int16_t) vout3r;
data3[1] = (int16_t) vout3i;
data0 += 2;
data1 += 2;
data2 += 2;
data3 += 2;
s -= sizeof(int16_t) * 2;
} while (s != 0);
}
} while (--batch != 0);
}
|