summaryrefslogtreecommitdiffstats
path: root/src/cksum_pclmul.c
blob: 691302d74678daeebd2abebb20864bb5eff84812 (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
/* cksum -- calculate and print POSIX checksums and sizes of files
   Copyright (C) 1992-2022 Free Software Foundation, Inc.

   This program 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 of the License, or
   (at your option) any later version.

   This program 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 this program.  If not, see <https://www.gnu.org/licenses/>.  */

#include <config.h>

#include <stdio.h>
#include <sys/types.h>
#include <stdint.h>
#include <x86intrin.h>
#include "system.h"

/* Number of bytes to read at once.  */
#define BUFLEN (1 << 16)

extern uint_fast32_t const crctab[8][256];

extern bool
cksum_pclmul (FILE *fp, uint_fast32_t *crc_out, uintmax_t *length_out);

/* Calculate CRC32 using PCLMULQDQ CPU instruction found in x86/x64 CPUs */

bool
cksum_pclmul (FILE *fp, uint_fast32_t *crc_out, uintmax_t *length_out)
{
  __m128i buf[BUFLEN / sizeof (__m128i)];
  uint_fast32_t crc = 0;
  uintmax_t length = 0;
  size_t bytes_read;
  __m128i single_mult_constant;
  __m128i four_mult_constant;
  __m128i shuffle_constant;

  if (!fp || !crc_out || !length_out)
    return false;

  /* These constants and general algorithms are taken from the Intel whitepaper
     "Fast CRC Computation for Generic Polynomials Using PCLMULQDQ Instruction"
  */
  single_mult_constant = _mm_set_epi64x (0xC5B9CD4C, 0xE8A45605);
  four_mult_constant = _mm_set_epi64x (0x8833794C, 0xE6228B11);

  /* Constant to byteswap a full SSE register */
  shuffle_constant = _mm_set_epi8 (0, 1, 2, 3, 4, 5, 6, 7, 8,
                                   9, 10, 11, 12, 13, 14, 15);

  while ((bytes_read = fread (buf, 1, BUFLEN, fp)) > 0)
    {
      __m128i *datap;
      __m128i data;
      __m128i data2;
      __m128i data3;
      __m128i data4;
      __m128i data5;
      __m128i data6;
      __m128i data7;
      __m128i data8;
      __m128i fold_data;
      __m128i xor_crc;

      if (length + bytes_read < length)
        {
          errno = EOVERFLOW;
          return false;
        }
      length += bytes_read;

      if (bytes_read == 0)
        {
          if (ferror (fp))
            return false;
        }

      datap = (__m128i *)buf;

      /* Fold in parallel eight 16-byte blocks into four 16-byte blocks */
      if (bytes_read >= 16 * 8)
        {
          data = _mm_loadu_si128 (datap);
          data = _mm_shuffle_epi8 (data, shuffle_constant);
          /* XOR in initial CRC value (for us 0 so no effect), or CRC value
             calculated for previous BUFLEN buffer from fread */
          xor_crc = _mm_set_epi32 (crc, 0, 0, 0);
          crc = 0;
          data = _mm_xor_si128 (data, xor_crc);
          data3 = _mm_loadu_si128 (datap + 1);
          data3 = _mm_shuffle_epi8 (data3, shuffle_constant);
          data5 = _mm_loadu_si128 (datap + 2);
          data5 = _mm_shuffle_epi8 (data5, shuffle_constant);
          data7 = _mm_loadu_si128 (datap + 3);
          data7 = _mm_shuffle_epi8 (data7, shuffle_constant);


          while (bytes_read >= 16 * 8)
            {
              datap += 4;

              /* Do multiplication here for four consecutive 16 byte blocks */
              data2 = _mm_clmulepi64_si128 (data, four_mult_constant, 0x00);
              data = _mm_clmulepi64_si128 (data, four_mult_constant, 0x11);
              data4 = _mm_clmulepi64_si128 (data3, four_mult_constant, 0x00);
              data3 = _mm_clmulepi64_si128 (data3, four_mult_constant, 0x11);
              data6 = _mm_clmulepi64_si128 (data5, four_mult_constant, 0x00);
              data5 = _mm_clmulepi64_si128 (data5, four_mult_constant, 0x11);
              data8 = _mm_clmulepi64_si128 (data7, four_mult_constant, 0x00);
              data7 = _mm_clmulepi64_si128 (data7, four_mult_constant, 0x11);

              /* Now multiplication results for the four blocks is xor:ed with
                 next four 16 byte blocks from the buffer. This effectively
                 "consumes" the first four blocks from the buffer.
                 Keep xor result in variables for multiplication in next
                 round of loop. */
              data = _mm_xor_si128 (data, data2);
              data2 = _mm_loadu_si128 (datap);
              data2 = _mm_shuffle_epi8 (data2, shuffle_constant);
              data = _mm_xor_si128 (data, data2);

              data3 = _mm_xor_si128 (data3, data4);
              data4 = _mm_loadu_si128 (datap + 1);
              data4 = _mm_shuffle_epi8 (data4, shuffle_constant);
              data3 = _mm_xor_si128 (data3, data4);

              data5 = _mm_xor_si128 (data5, data6);
              data6 = _mm_loadu_si128 (datap + 2);
              data6 = _mm_shuffle_epi8 (data6, shuffle_constant);
              data5 = _mm_xor_si128 (data5, data6);

              data7 = _mm_xor_si128 (data7, data8);
              data8 = _mm_loadu_si128 (datap + 3);
              data8 = _mm_shuffle_epi8 (data8, shuffle_constant);
              data7 = _mm_xor_si128 (data7, data8);

              bytes_read -= (16 * 4);
            }
          /* At end of loop we write out results from variables back into
             the buffer, for use in single fold loop */
          data = _mm_shuffle_epi8 (data, shuffle_constant);
          _mm_storeu_si128 (datap, data);
          data3 = _mm_shuffle_epi8 (data3, shuffle_constant);
          _mm_storeu_si128 (datap + 1, data3);
          data5 = _mm_shuffle_epi8 (data5, shuffle_constant);
          _mm_storeu_si128 (datap + 2, data5);
          data7 = _mm_shuffle_epi8 (data7, shuffle_constant);
          _mm_storeu_si128 (datap + 3, data7);
        }

      /* Fold two 16-byte blocks into one 16-byte block */
      if (bytes_read >= 32)
        {
          data = _mm_loadu_si128 (datap);
          data = _mm_shuffle_epi8 (data, shuffle_constant);
          xor_crc = _mm_set_epi32 (crc, 0, 0, 0);
          crc = 0;
          data = _mm_xor_si128 (data, xor_crc);
          while (bytes_read >= 32)
            {
              datap++;

              data2 = _mm_clmulepi64_si128 (data, single_mult_constant, 0x00);
              data = _mm_clmulepi64_si128 (data, single_mult_constant, 0x11);
              fold_data = _mm_loadu_si128 (datap);
              fold_data = _mm_shuffle_epi8 (fold_data, shuffle_constant);
              data = _mm_xor_si128 (data, data2);
              data = _mm_xor_si128 (data, fold_data);
              bytes_read -= 16;
            }
          data = _mm_shuffle_epi8 (data, shuffle_constant);
          _mm_storeu_si128 (datap, data);
        }

      /* And finish up last 0-31 bytes in a byte by byte fashion */
      unsigned char *cp = (unsigned char *)datap;
      while (bytes_read--)
        crc = (crc << 8) ^ crctab[0][((crc >> 24) ^ *cp++) & 0xFF];
      if (feof (fp))
        break;
    }

  *crc_out = crc;
  *length_out = length;

  return true;
}