summaryrefslogtreecommitdiff
path: root/crc/xxhash.c
blob: db7890b027826938e8aa61a8c49420c39332a6b2 (plain)
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
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
/*
xxHash - Fast Hash algorithm
Copyright (C) 2012-2014, Yann Collet.
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)

Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:

* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* 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.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"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 COPYRIGHT
OWNER OR CONTRIBUTORS 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.

You can contact the author at :
- xxHash source repository : http://code.google.com/p/xxhash/
*/


//**************************************
// Tuning parameters
//**************************************
// Unaligned memory access is automatically enabled for "common" CPU, such as x86.
// For others CPU, the compiler will be more cautious, and insert extra code to ensure aligned access is respected.
// If you know your target CPU supports unaligned memory access, you want to force this option manually to improve performance.
// You can also enable this parameter if you know your input data will always be aligned (boundaries of 4, for uint32_t).
#if defined(__ARM_FEATURE_UNALIGNED) || defined(__i386) || defined(_M_IX86) || defined(__x86_64__) || defined(_M_X64)
#  define XXH_USE_UNALIGNED_ACCESS 1
#endif

// XXH_ACCEPT_NULL_INPUT_POINTER :
// If the input pointer is a null pointer, xxHash default behavior is to trigger a memory access error, since it is a bad pointer.
// When this option is enabled, xxHash output for null input pointers will be the same as a null-length input.
// This option has a very small performance cost (only measurable on small inputs).
// By default, this option is disabled. To enable it, uncomment below define :
//#define XXH_ACCEPT_NULL_INPUT_POINTER 1

// XXH_FORCE_NATIVE_FORMAT :
// By default, xxHash library provides endian-independant Hash values, based on little-endian convention.
// Results are therefore identical for little-endian and big-endian CPU.
// This comes at a performance cost for big-endian CPU, since some swapping is required to emulate little-endian format.
// Should endian-independance be of no importance for your application, you may set the #define below to 1.
// It will improve speed for Big-endian CPU.
// This option has no impact on Little_Endian CPU.
#define XXH_FORCE_NATIVE_FORMAT 0


//**************************************
// Includes & Memory related functions
//**************************************
#include "xxhash.h"
#include <stdlib.h>
#include <string.h>


#if defined(__GNUC__)  && !defined(XXH_USE_UNALIGNED_ACCESS)
#  define _PACKED __attribute__ ((packed))
#else
#  define _PACKED
#endif

#if !defined(XXH_USE_UNALIGNED_ACCESS) && !defined(__GNUC__)
#  ifdef __IBMC__
#    pragma pack(1)
#  else
#    pragma pack(push, 1)
#  endif
#endif

typedef struct _uint32_t_S { uint32_t v; } _PACKED uint32_t_S;

#if !defined(XXH_USE_UNALIGNED_ACCESS) && !defined(__GNUC__)
#  pragma pack(pop)
#endif

#define A32(x) (((uint32_t_S *)(x))->v)


//***************************************
// Compiler-specific Functions and Macros
//***************************************
#define GCC_VERSION (__GNUC__ * 100 + __GNUC_MINOR__)

// Note : although _rotl exists for minGW (GCC under windows), performance seems poor
#if defined(_MSC_VER)
#  define XXH_rotl32(x,r) _rotl(x,r)
#else
#  define XXH_rotl32(x,r) ((x << r) | (x >> (32 - r)))
#endif

#if defined(_MSC_VER)     // Visual Studio
#  define XXH_swap32 _byteswap_ulong
#elif GCC_VERSION >= 403
#  define XXH_swap32 __builtin_bswap32
#else
static inline uint32_t XXH_swap32 (uint32_t x)
{
    return  ((x << 24) & 0xff000000 ) |
        ((x <<  8) & 0x00ff0000 ) |
        ((x >>  8) & 0x0000ff00 ) |
        ((x >> 24) & 0x000000ff );
}
#endif


//**************************************
// Constants
//**************************************
#define PRIME32_1   2654435761U
#define PRIME32_2   2246822519U
#define PRIME32_3   3266489917U
#define PRIME32_4    668265263U
#define PRIME32_5    374761393U


//**************************************
// Architecture Macros
//**************************************
typedef enum { XXH_bigEndian=0, XXH_littleEndian=1 } XXH_endianess;
#ifndef XXH_CPU_LITTLE_ENDIAN   // It is possible to define XXH_CPU_LITTLE_ENDIAN externally, for example using a compiler switch
    static const int one = 1;
#   define XXH_CPU_LITTLE_ENDIAN   (*(char*)(&one))
#endif


//**************************************
// Macros
//**************************************
#define XXH_STATIC_ASSERT(c)   { enum { XXH_static_assert = 1/(!!(c)) }; }    // use only *after* variable declarations


//****************************
// Memory reads
//****************************
typedef enum { XXH_aligned, XXH_unaligned } XXH_alignment;

uint32_t XXH_readLE32_align(const uint32_t* ptr, XXH_endianess endian, XXH_alignment align)
{
    if (align==XXH_unaligned)
        return endian==XXH_littleEndian ? A32(ptr) : XXH_swap32(A32(ptr));
    else
        return endian==XXH_littleEndian ? *ptr : XXH_swap32(*ptr);
}

uint32_t XXH_readLE32(const uint32_t* ptr, XXH_endianess endian) { return XXH_readLE32_align(ptr, endian, XXH_unaligned); }


//****************************
// Simple Hash Functions
//****************************
uint32_t XXH32_endian_align(const void* input, int len, uint32_t seed, XXH_endianess endian, XXH_alignment align)
{
    const uint8_t *p = (const uint8_t *)input;
    const uint8_t * const bEnd = p + len;
    uint32_t h32;

#ifdef XXH_ACCEPT_NULL_INPUT_POINTER
    if (p==NULL) { len=0; p=(const uint8_t *)(size_t)16; }
#endif

    if (len>=16)
    {
        const uint8_t * const limit = bEnd - 16;
        uint32_t v1 = seed + PRIME32_1 + PRIME32_2;
        uint32_t v2 = seed + PRIME32_2;
        uint32_t v3 = seed + 0;
        uint32_t v4 = seed - PRIME32_1;

        do
        {
            v1 += XXH_readLE32_align((const uint32_t*)p, endian, align) * PRIME32_2; v1 = XXH_rotl32(v1, 13); v1 *= PRIME32_1; p+=4;
            v2 += XXH_readLE32_align((const uint32_t*)p, endian, align) * PRIME32_2; v2 = XXH_rotl32(v2, 13); v2 *= PRIME32_1; p+=4;
            v3 += XXH_readLE32_align((const uint32_t*)p, endian, align) * PRIME32_2; v3 = XXH_rotl32(v3, 13); v3 *= PRIME32_1; p+=4;
            v4 += XXH_readLE32_align((const uint32_t*)p, endian, align) * PRIME32_2; v4 = XXH_rotl32(v4, 13); v4 *= PRIME32_1; p+=4;
        } while (p<=limit);

        h32 = XXH_rotl32(v1, 1) + XXH_rotl32(v2, 7) + XXH_rotl32(v3, 12) + XXH_rotl32(v4, 18);
    }
    else
    {
        h32  = seed + PRIME32_5;
    }

    h32 += (uint32_t) len;

    while (p<=bEnd-4)
    {
        h32 += XXH_readLE32_align((const uint32_t*)p, endian, align) * PRIME32_3;
        h32  = XXH_rotl32(h32, 17) * PRIME32_4 ;
        p+=4;
    }

    while (p<bEnd)
    {
        h32 += (*p) * PRIME32_5;
        h32 = XXH_rotl32(h32, 11) * PRIME32_1 ;
        p++;
    }

    h32 ^= h32 >> 15;
    h32 *= PRIME32_2;
    h32 ^= h32 >> 13;
    h32 *= PRIME32_3;
    h32 ^= h32 >> 16;

    return h32;
}


uint32_t XXH32(const void* input, int len, uint32_t seed)
{
#if 0
    // Simple version, good for code maintenance, but unfortunately slow for small inputs
    void* state = XXH32_init(seed);
    XXH32_update(state, input, len);
    return XXH32_digest(state);
#else
    XXH_endianess endian_detected = (XXH_endianess)XXH_CPU_LITTLE_ENDIAN;

#  if !defined(XXH_USE_UNALIGNED_ACCESS)
    if ((((size_t)input) & 3))   // Input is aligned, let's leverage the speed advantage
    {
        if ((endian_detected==XXH_littleEndian) || XXH_FORCE_NATIVE_FORMAT)
            return XXH32_endian_align(input, len, seed, XXH_littleEndian, XXH_aligned);
        else
            return XXH32_endian_align(input, len, seed, XXH_bigEndian, XXH_aligned);
    }
#  endif

    if ((endian_detected==XXH_littleEndian) || XXH_FORCE_NATIVE_FORMAT)
        return XXH32_endian_align(input, len, seed, XXH_littleEndian, XXH_unaligned);
    else
        return XXH32_endian_align(input, len, seed, XXH_bigEndian, XXH_unaligned);
#endif
}


//****************************
// Advanced Hash Functions
//****************************

int XXH32_sizeofState(void)
{
    XXH_STATIC_ASSERT(XXH32_SIZEOFSTATE >= sizeof(struct XXH_state32_t));   // A compilation error here means XXH32_SIZEOFSTATE is not large enough
    return sizeof(struct XXH_state32_t);
}


XXH_errorcode XXH32_resetState(void* state_in, uint32_t seed)
{
    struct XXH_state32_t * state = (struct XXH_state32_t *) state_in;
    state->seed = seed;
    state->v1 = seed + PRIME32_1 + PRIME32_2;
    state->v2 = seed + PRIME32_2;
    state->v3 = seed + 0;
    state->v4 = seed - PRIME32_1;
    state->total_len = 0;
    state->memsize = 0;
    return XXH_OK;
}


void* XXH32_init (uint32_t seed)
{
    void *state = malloc (sizeof(struct XXH_state32_t));
    XXH32_resetState(state, seed);
    return state;
}


XXH_errorcode XXH32_update_endian (void* state_in, const void* input, int len, XXH_endianess endian)
{
    struct XXH_state32_t * state = (struct XXH_state32_t *) state_in;
    const uint8_t *p = (const uint8_t *)input;
    const uint8_t * const bEnd = p + len;

#ifdef XXH_ACCEPT_NULL_INPUT_POINTER
    if (input==NULL) return XXH_ERROR;
#endif

    state->total_len += len;

    if (state->memsize + len < 16)   // fill in tmp buffer
    {
        memcpy(state->memory + state->memsize, input, len);
        state->memsize +=  len;
        return XXH_OK;
    }

    if (state->memsize)   // some data left from previous update
    {
        memcpy(state->memory + state->memsize, input, 16-state->memsize);
        {
            const uint32_t* p32 = (const uint32_t*)state->memory;
            state->v1 += XXH_readLE32(p32, endian) * PRIME32_2; state->v1 = XXH_rotl32(state->v1, 13); state->v1 *= PRIME32_1; p32++;
            state->v2 += XXH_readLE32(p32, endian) * PRIME32_2; state->v2 = XXH_rotl32(state->v2, 13); state->v2 *= PRIME32_1; p32++;
            state->v3 += XXH_readLE32(p32, endian) * PRIME32_2; state->v3 = XXH_rotl32(state->v3, 13); state->v3 *= PRIME32_1; p32++;
            state->v4 += XXH_readLE32(p32, endian) * PRIME32_2; state->v4 = XXH_rotl32(state->v4, 13); state->v4 *= PRIME32_1; p32++;
        }
        p += 16-state->memsize;
        state->memsize = 0;
    }

    if (p <= bEnd-16)
    {
        const uint8_t * const limit = bEnd - 16;
        uint32_t v1 = state->v1;
        uint32_t v2 = state->v2;
        uint32_t v3 = state->v3;
        uint32_t v4 = state->v4;

        do
        {
            v1 += XXH_readLE32((const uint32_t*)p, endian) * PRIME32_2; v1 = XXH_rotl32(v1, 13); v1 *= PRIME32_1; p+=4;
            v2 += XXH_readLE32((const uint32_t*)p, endian) * PRIME32_2; v2 = XXH_rotl32(v2, 13); v2 *= PRIME32_1; p+=4;
            v3 += XXH_readLE32((const uint32_t*)p, endian) * PRIME32_2; v3 = XXH_rotl32(v3, 13); v3 *= PRIME32_1; p+=4;
            v4 += XXH_readLE32((const uint32_t*)p, endian) * PRIME32_2; v4 = XXH_rotl32(v4, 13); v4 *= PRIME32_1; p+=4;
        } while (p<=limit);

        state->v1 = v1;
        state->v2 = v2;
        state->v3 = v3;
        state->v4 = v4;
    }

    if (p < bEnd)
    {
        memcpy(state->memory, p, bEnd-p);
        state->memsize = (int)(bEnd-p);
    }

    return XXH_OK;
}

XXH_errorcode XXH32_update (void* state_in, const void* input, int len)
{
    XXH_endianess endian_detected = (XXH_endianess)XXH_CPU_LITTLE_ENDIAN;

    if ((endian_detected==XXH_littleEndian) || XXH_FORCE_NATIVE_FORMAT)
        return XXH32_update_endian(state_in, input, len, XXH_littleEndian);
    else
        return XXH32_update_endian(state_in, input, len, XXH_bigEndian);
}



uint32_t XXH32_intermediateDigest_endian (void* state_in, XXH_endianess endian)
{
    struct XXH_state32_t * state = (struct XXH_state32_t *) state_in;
    const uint8_t *p = (const uint8_t *)state->memory;
    uint8_t * bEnd = (uint8_t *)state->memory + state->memsize;
    uint32_t h32;

    if (state->total_len >= 16)
    {
        h32 = XXH_rotl32(state->v1, 1) + XXH_rotl32(state->v2, 7) + XXH_rotl32(state->v3, 12) + XXH_rotl32(state->v4, 18);
    }
    else
    {
        h32  = state->seed + PRIME32_5;
    }

    h32 += (uint32_t) state->total_len;

    while (p<=bEnd-4)
    {
        h32 += XXH_readLE32((const uint32_t*)p, endian) * PRIME32_3;
        h32  = XXH_rotl32(h32, 17) * PRIME32_4;
        p+=4;
    }

    while (p<bEnd)
    {
        h32 += (*p) * PRIME32_5;
        h32 = XXH_rotl32(h32, 11) * PRIME32_1;
        p++;
    }

    h32 ^= h32 >> 15;
    h32 *= PRIME32_2;
    h32 ^= h32 >> 13;
    h32 *= PRIME32_3;
    h32 ^= h32 >> 16;

    return h32;
}


uint32_t XXH32_intermediateDigest (void* state_in)
{
    XXH_endianess endian_detected = (XXH_endianess)XXH_CPU_LITTLE_ENDIAN;

    if ((endian_detected==XXH_littleEndian) || XXH_FORCE_NATIVE_FORMAT)
        return XXH32_intermediateDigest_endian(state_in, XXH_littleEndian);
    else
        return XXH32_intermediateDigest_endian(state_in, XXH_bigEndian);
}


uint32_t XXH32_digest (void* state_in)
{
    uint32_t h32 = XXH32_intermediateDigest(state_in);

    free(state_in);

    return h32;
}