This file is indexed.

/usr/include/volk/volk_32u_byteswap.h is in libvolk1-dev 1.2.1-1.

This file is owned by root:root, with mode 0o644.

The actual contents of the file can be viewed below.

  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
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
/* -*- c++ -*- */
/*
 * Copyright 2012, 2014 Free Software Foundation, Inc.
 *
 * This file is part of GNU Radio
 *
 * GNU Radio is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 3, or (at your option)
 * any later version.
 *
 * GNU Radio 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 General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with GNU Radio; see the file COPYING.  If not, write to
 * the Free Software Foundation, Inc., 51 Franklin Street,
 * Boston, MA 02110-1301, USA.
 */

/*!
 * \page volk_32u_byteswap
 *
 * \b Overview
 *
 * Byteswaps (in-place) an aligned vector of int32_t's.
 *
 * <b>Dispatcher Prototype</b>
 * \code
 * void volk_32u_byteswap(uint32_t* intsToSwap, unsigned int num_points)
 * \endcode
 *
 * \b Inputs
 * \li intsToSwap: The vector of data to byte swap.
 * \li num_points: The number of data points.
 *
 * \b Outputs
 * \li intsToSwap: returns as an in-place calculation.
 *
 * \b Example
 * \code
 *   int N = 10;
 *   unsigned int alignment = volk_get_alignment();
 *
 *   uint32_t bitstring[] = {0x0, 0x1, 0xf, 0xffffffff,
 *       0x5a5a5a5a, 0xa5a5a5a5, 0x2a2a2a2a,
 *       0xffffffff, 0x32, 0x64};
 *   uint32_t hamming_distance = 0;
 *
 *   printf("byteswap vector =\n");
 *   for(unsigned int ii=0; ii<N; ++ii){
 *       printf("    %.8x\n", bitstring[ii]);
 *   }
 *
 *   volk_32u_byteswap(bitstring, N);
 *
 *   printf("byteswapped vector =\n");
 *   for(unsigned int ii=0; ii<N; ++ii){
 *       printf("    %.8x\n", bitstring[ii]);
 *   }
 * \endcode
 */

#ifndef INCLUDED_volk_32u_byteswap_u_H
#define INCLUDED_volk_32u_byteswap_u_H

#include <inttypes.h>
#include <stdio.h>

#if LV_HAVE_AVX2
#include <immintrin.h>
static inline void volk_32u_byteswap_u_avx2(uint32_t* intsToSwap, unsigned int num_points){

  unsigned int number;

  const unsigned int nPerSet = 8;
  const uint64_t     nSets   = num_points / nPerSet;

  uint32_t* inputPtr = intsToSwap;

  const uint8_t shuffleVector[32] = { 3, 2, 1, 0, 7, 6, 5, 4, 11, 10, 9, 8, 15, 14, 13, 12, 19, 18, 17, 16, 23, 22, 21, 20, 27, 26, 25, 24, 31, 30, 29, 28 };

  const __m256i myShuffle = _mm256_loadu_si256((__m256i*) &shuffleVector);

  for (number = 0 ;number < nSets; number++) {

    // Load the 32t values, increment inputPtr later since we're doing it in-place.
    const __m256i input  = _mm256_loadu_si256((__m256i*)inputPtr);
    const __m256i output = _mm256_shuffle_epi8(input,myShuffle);

    // Store the results
    _mm256_storeu_si256((__m256i*)inputPtr, output);
    inputPtr += nPerSet;
  }
  _mm256_zeroupper();

  // Byteswap any remaining points:
  for(number = nSets * nPerSet; number < num_points; number++){
    uint32_t outputVal = *inputPtr;
    outputVal = (((outputVal >> 24) & 0xff) | ((outputVal >> 8) & 0x0000ff00) | ((outputVal << 8) & 0x00ff0000) | ((outputVal << 24) & 0xff000000));
    *inputPtr = outputVal;
    inputPtr++;
  }
}
#endif /* LV_HAVE_AVX2 */


#ifdef LV_HAVE_SSE2
#include <emmintrin.h>

static inline void volk_32u_byteswap_u_sse2(uint32_t* intsToSwap, unsigned int num_points){
  unsigned int number = 0;

  uint32_t* inputPtr = intsToSwap;
  __m128i input, byte1, byte2, byte3, byte4, output;
  __m128i byte2mask = _mm_set1_epi32(0x00FF0000);
  __m128i byte3mask = _mm_set1_epi32(0x0000FF00);

  const uint64_t quarterPoints = num_points / 4;
  for(;number < quarterPoints; number++){
    // Load the 32t values, increment inputPtr later since we're doing it in-place.
    input = _mm_loadu_si128((__m128i*)inputPtr);
    // Do the four shifts
    byte1 = _mm_slli_epi32(input, 24);
    byte2 = _mm_slli_epi32(input, 8);
    byte3 = _mm_srli_epi32(input, 8);
    byte4 = _mm_srli_epi32(input, 24);
    // Or bytes together
    output = _mm_or_si128(byte1, byte4);
    byte2 = _mm_and_si128(byte2, byte2mask);
    output = _mm_or_si128(output, byte2);
    byte3 = _mm_and_si128(byte3, byte3mask);
    output = _mm_or_si128(output, byte3);
    // Store the results
    _mm_storeu_si128((__m128i*)inputPtr, output);
    inputPtr += 4;
  }

  // Byteswap any remaining points:
  number = quarterPoints*4;
  for(; number < num_points; number++){
    uint32_t outputVal = *inputPtr;
    outputVal = (((outputVal >> 24) & 0xff) | ((outputVal >> 8) & 0x0000ff00) | ((outputVal << 8) & 0x00ff0000) | ((outputVal << 24) & 0xff000000));
    *inputPtr = outputVal;
    inputPtr++;
  }
}
#endif /* LV_HAVE_SSE2 */


#ifdef LV_HAVE_NEON
#include <arm_neon.h>

static inline void volk_32u_byteswap_neon(uint32_t* intsToSwap, unsigned int num_points){
  uint32_t* inputPtr = intsToSwap;
  unsigned int number = 0;
  unsigned int n8points = num_points / 8;

  uint8x8x4_t input_table;
  uint8x8_t int_lookup01, int_lookup23, int_lookup45, int_lookup67;
  uint8x8_t swapped_int01, swapped_int23, swapped_int45, swapped_int67;

  /* these magic numbers are used as byte-indeces in the LUT.
     they are pre-computed to save time. A simple C program
     can calculate them; for example for lookup01:
    uint8_t chars[8] = {24, 16, 8, 0, 25, 17, 9, 1};
    for(ii=0; ii < 8; ++ii) {
        index += ((uint64_t)(*(chars+ii))) << (ii*8);
    }
  */
  int_lookup01 = vcreate_u8(74609667900706840);
  int_lookup23 = vcreate_u8(219290013576860186);
  int_lookup45 = vcreate_u8(363970359253013532);
  int_lookup67 = vcreate_u8(508650704929166878);

  for(number = 0; number < n8points; ++number){
    input_table = vld4_u8((uint8_t*) inputPtr);
    swapped_int01 = vtbl4_u8(input_table, int_lookup01);
    swapped_int23 = vtbl4_u8(input_table, int_lookup23);
    swapped_int45 = vtbl4_u8(input_table, int_lookup45);
    swapped_int67 = vtbl4_u8(input_table, int_lookup67);
    vst1_u8((uint8_t*) inputPtr, swapped_int01);
    vst1_u8((uint8_t*) (inputPtr+2), swapped_int23);
    vst1_u8((uint8_t*) (inputPtr+4), swapped_int45);
    vst1_u8((uint8_t*) (inputPtr+6), swapped_int67);

    inputPtr += 8;
  }

  for(number = n8points * 8; number < num_points; ++number){
    uint32_t output = *inputPtr;
    output = (((output >> 24) & 0xff) | ((output >> 8) & 0x0000ff00) | ((output << 8) & 0x00ff0000) | ((output << 24) & 0xff000000));

    *inputPtr = output;
    inputPtr++;
  }
}
#endif /* LV_HAVE_NEON */


#ifdef LV_HAVE_GENERIC

static inline void volk_32u_byteswap_generic(uint32_t* intsToSwap, unsigned int num_points){
  uint32_t* inputPtr = intsToSwap;

  unsigned int point;
  for(point = 0; point < num_points; point++){
    uint32_t output = *inputPtr;
    output = (((output >> 24) & 0xff) | ((output >> 8) & 0x0000ff00) | ((output << 8) & 0x00ff0000) | ((output << 24) & 0xff000000));

    *inputPtr = output;
    inputPtr++;
  }
}
#endif /* LV_HAVE_GENERIC */


#endif /* INCLUDED_volk_32u_byteswap_u_H */
#ifndef INCLUDED_volk_32u_byteswap_a_H
#define INCLUDED_volk_32u_byteswap_a_H

#include <inttypes.h>
#include <stdio.h>


#if LV_HAVE_AVX2
#include <immintrin.h>
static inline void volk_32u_byteswap_a_avx2(uint32_t* intsToSwap, unsigned int num_points){

  unsigned int number;

  const unsigned int nPerSet = 8;
  const uint64_t     nSets   = num_points / nPerSet;

  uint32_t* inputPtr = intsToSwap;

  const uint8_t shuffleVector[32] = { 3, 2, 1, 0, 7, 6, 5, 4, 11, 10, 9, 8, 15, 14, 13, 12, 19, 18, 17, 16, 23, 22, 21, 20, 27, 26, 25, 24, 31, 30, 29, 28 };

  const __m256i myShuffle = _mm256_loadu_si256((__m256i*) &shuffleVector);

  for (number = 0 ;number < nSets; number++) {

    // Load the 32t values, increment inputPtr later since we're doing it in-place.
    const __m256i input  = _mm256_load_si256((__m256i*)inputPtr);
    const __m256i output = _mm256_shuffle_epi8(input,myShuffle);

    // Store the results
    _mm256_store_si256((__m256i*)inputPtr, output);
    inputPtr += nPerSet;
  }
  _mm256_zeroupper();

  // Byteswap any remaining points:
  for(number = nSets * nPerSet; number < num_points; number++){
    uint32_t outputVal = *inputPtr;
    outputVal = (((outputVal >> 24) & 0xff) | ((outputVal >> 8) & 0x0000ff00) | ((outputVal << 8) & 0x00ff0000) | ((outputVal << 24) & 0xff000000));
    *inputPtr = outputVal;
    inputPtr++;
  }
}
#endif /* LV_HAVE_AVX2 */


#ifdef LV_HAVE_SSE2
#include <emmintrin.h>


static inline void volk_32u_byteswap_a_sse2(uint32_t* intsToSwap, unsigned int num_points){
  unsigned int number = 0;

  uint32_t* inputPtr = intsToSwap;
  __m128i input, byte1, byte2, byte3, byte4, output;
  __m128i byte2mask = _mm_set1_epi32(0x00FF0000);
  __m128i byte3mask = _mm_set1_epi32(0x0000FF00);

  const uint64_t quarterPoints = num_points / 4;
  for(;number < quarterPoints; number++){
    // Load the 32t values, increment inputPtr later since we're doing it in-place.
    input = _mm_load_si128((__m128i*)inputPtr);
    // Do the four shifts
    byte1 = _mm_slli_epi32(input, 24);
    byte2 = _mm_slli_epi32(input, 8);
    byte3 = _mm_srli_epi32(input, 8);
    byte4 = _mm_srli_epi32(input, 24);
    // Or bytes together
    output = _mm_or_si128(byte1, byte4);
    byte2 = _mm_and_si128(byte2, byte2mask);
    output = _mm_or_si128(output, byte2);
    byte3 = _mm_and_si128(byte3, byte3mask);
    output = _mm_or_si128(output, byte3);
    // Store the results
    _mm_store_si128((__m128i*)inputPtr, output);
    inputPtr += 4;
  }

  // Byteswap any remaining points:
  number = quarterPoints*4;
  for(; number < num_points; number++){
    uint32_t outputVal = *inputPtr;
    outputVal = (((outputVal >> 24) & 0xff) | ((outputVal >> 8) & 0x0000ff00) | ((outputVal << 8) & 0x00ff0000) | ((outputVal << 24) & 0xff000000));
    *inputPtr = outputVal;
    inputPtr++;
  }
}
#endif /* LV_HAVE_SSE2 */


#ifdef LV_HAVE_GENERIC

static inline void volk_32u_byteswap_a_generic(uint32_t* intsToSwap, unsigned int num_points){
  uint32_t* inputPtr = intsToSwap;

  unsigned int point;
  for(point = 0; point < num_points; point++){
    uint32_t output = *inputPtr;
    output = (((output >> 24) & 0xff) | ((output >> 8) & 0x0000ff00) | ((output << 8) & 0x00ff0000) | ((output << 24) & 0xff000000));

    *inputPtr = output;
    inputPtr++;
  }
}
#endif /* LV_HAVE_GENERIC */




#endif /* INCLUDED_volk_32u_byteswap_a_H */