xapian-applications/omega/md5.cc

   1 /*
   2  * This code implements the MD5 message-digest algorithm.
   3  * The algorithm is due to Ron Rivest.  This code was
   4  * written by Colin Plumb in 1993, no copyright is claimed.
   5  * This code is in the public domain; do with it what you wish.
   6  *
   7  * Equivalent code is available from RSA Data Security, Inc.
   8  * This code has been tested against that, and is equivalent,
   9  * except that you don't need to include two pages of legalese
  10  * with every copy.
  11  *
  12  * To compute the message digest of a chunk of bytes, declare an
  13  * MD5Context structure, pass it to MD5Init, call MD5Update as
  14  * needed on buffers full of bytes, and then call MD5Final, which
  15  * will fill a supplied 16-byte array with the digest.
  16  */
  17
  18 /* Minor tweaks for use from Omega:
  19  *
  20  * + Use WORDS_BIGENDIAN (probed by autoconf) to check endianness.
  21  * + uint32 -> uint32_t (as defined by netinet/in.h).
  22  * + Internal functions are now static.
  23  * + Fix sizeof(ctx) to sizeof(*ctx) (bug picked up by Stanford checker on
  24  *   this code incorporated into the Linux kernel).
  25  * + Renamed to .cc and compiled as C++.
  26  * + Changed MD5Context.in to uint32_t instead of unsigned char, and
  27  *   byteReverse to take "uint32_t *" instead of "unsigned char *".
  28  */
  29
  30 #include <config.h>
  31
  32 #include <cstring>              /* for memcpy() */
  33 #include "md5.h"
  34
  35 using namespace std;
  36
  37 #ifndef WORDS_BIGENDIAN
  38 #define byteReverse(buf, len)   /* Nothing */
  39 #else
  40 static void byteReverse(uint32_t *buf, unsigned longs);
  41
  42 #ifndef ASM_MD5
  43 /*
  44  * Note: this code is harmless on little-endian machines.
  45  */
  46 static void byteReverse(uint32_t *buf, unsigned longs)
  47 {
  48     uint32_t t;
  49     do {
  50         unsigned char * p = (unsigned char *)buf;
  51         t = uint32_t(unsigned(p[3]) << 8 | p[2]) << 16 |
  52             (unsigned(p[1]) << 8 | p[0]);
  53         *buf++ = t;
  54     } while (--longs);
  55 }
  56 #endif
  57 #endif
  58
  59 static void MD5Transform(uint32_t buf[4], uint32_t const in[16]);
  60
  61 /*
  62  * Start MD5 accumulation.  Set bit count to 0 and buffer to mysterious
  63  * initialization constants.
  64  */
  65 void MD5Init(struct MD5Context *ctx)
  66 {
  67     ctx->buf[0] = 0x67452301;
  68     ctx->buf[1] = 0xefcdab89;
  69     ctx->buf[2] = 0x98badcfe;
  70     ctx->buf[3] = 0x10325476;
  71
  72     ctx->bits[0] = 0;
  73     ctx->bits[1] = 0;
  74 }
  75
  76 /*
  77  * Update context to reflect the concatenation of another buffer full
  78  * of bytes.
  79  */
  80 void MD5Update(struct MD5Context *ctx, unsigned char const *buf, unsigned len)
  81 {
  82     uint32_t t;
  83
  84     /* Update bitcount */
  85
  86     t = ctx->bits[0];
  87     if ((ctx->bits[0] = t + (uint32_t(len) << 3)) < t)
  88         ctx->bits[1]++;         /* Carry from low to high */
  89     ctx->bits[1] += len >> 29;
  90
  91     t = (t >> 3) & 0x3f;        /* Bytes already in shsInfo->data */
  92
  93     /* Handle any leading odd-sized chunks */
  94
  95     if (t) {
  96         unsigned char *p =
  97            reinterpret_cast<unsigned char *>(ctx->in) + t;
  98
  99         t = 64 - t;
 100         if (len < t) {
 101             memcpy(p, buf, len);
 102             return;
 103         }
 104         memcpy(p, buf, t);
 105         byteReverse(ctx->in, 16);
 106         MD5Transform(ctx->buf, ctx->in);
 107         buf += t;
 108         len -= t;
 109     }
 110     /* Process data in 64-byte chunks */
 111
 112     while (len >= 64) {
 113         memcpy(ctx->in, buf, 64);
 114         byteReverse(ctx->in, 16);
 115         MD5Transform(ctx->buf, ctx->in);
 116         buf += 64;
 117         len -= 64;
 118     }
 119
 120     /* Handle any remaining bytes of data. */
 121
 122     memcpy(ctx->in, buf, len);
 123 }
 124
 125 /*
 126  * Final wrapup - pad to 64-byte boundary with the bit pattern
 127  * 1 0* (64-bit count of bits processed, MSB-first)
 128  */
 129 void MD5Final(unsigned char digest[16], struct MD5Context *ctx)
 130 {
 131     unsigned count;
 132     unsigned char *p;
 133
 134     /* Compute number of bytes mod 64 */
 135     count = (ctx->bits[0] >> 3) & 0x3F;
 136
 137     /* Set the first char of padding to 0x80.  This is safe since there is
 138        always at least one byte free */
 139     p = reinterpret_cast<unsigned char *>(ctx->in) + count;
 140     *p++ = 0x80;
 141
 142     /* Bytes of padding needed to make 64 bytes */
 143     count = 64 - 1 - count;
 144
 145     /* Pad out to 56 mod 64 */
 146     if (count < 8) {
 147         /* Two lots of padding:  Pad the first block to 64 bytes */
 148         memset(p, 0, count);
 149         byteReverse(ctx->in, 16);
 150         MD5Transform(ctx->buf, ctx->in);
 151
 152         /* Now fill the next block with 56 bytes */
 153         memset(ctx->in, 0, 56);
 154     } else {
 155         /* Pad block to 56 bytes */
 156         memset(p, 0, count - 8);
 157         byteReverse(ctx->in, 14);
 158     }
 159
 160     /* Append length in bits and transform */
 161     ctx->in[14] = ctx->bits[0];
 162     ctx->in[15] = ctx->bits[1];
 163
 164     MD5Transform(ctx->buf, ctx->in);
 165     byteReverse(ctx->buf, 4);
 166     memcpy(digest, ctx->buf, 16);
 167     memset(ctx, 0, sizeof(*ctx));       /* In case it's sensitive */
 168 }
 169
 170 #ifndef ASM_MD5
 171
 172 /* The four core functions - F1 is optimized somewhat */
 173
 174 /* #define F1(x, y, z) (x & y | ~x & z) */
 175 #define F1(x, y, z) (z ^ (x & (y ^ z)))
 176 #define F2(x, y, z) F1(z, x, y)
 177 #define F3(x, y, z) (x ^ y ^ z)
 178 #define F4(x, y, z) (y ^ (x | ~z))
 179
 180 /* This is the central step in the MD5 algorithm. */
 181 #define MD5STEP(f, w, x, y, z, data, s) \
 182         ( w += f(x, y, z) + data,  w = w<<s | w>>(32-s),  w += x )
 183
 184 /*
 185  * The core of the MD5 algorithm, this alters an existing MD5 hash to
 186  * reflect the addition of 16 longwords of new data.  MD5Update blocks
 187  * the data and converts bytes into longwords for this routine.
 188  */
 189 #ifdef __clang__
 190 // Unsigned integer overflow is defined behaviour, but still something that it
 191 // is desirable to avoid in most code.  The code here makes use of it, so
 192 // turn off UBSan's optional "unsigned-integer-overflow" check for this
 193 // function.
 194 [[clang::no_sanitize("unsigned-integer-overflow")]]
 195 #endif
 196 static void MD5Transform(uint32_t buf[4], uint32_t const in[16])
 197 {
 198     uint32_t a, b, c, d;
 199
 200     a = buf[0];
 201     b = buf[1];
 202     c = buf[2];
 203     d = buf[3];
 204
 205     MD5STEP(F1, a, b, c, d, in[0] + 0xd76aa478, 7);
 206     MD5STEP(F1, d, a, b, c, in[1] + 0xe8c7b756, 12);
 207     MD5STEP(F1, c, d, a, b, in[2] + 0x242070db, 17);
 208     MD5STEP(F1, b, c, d, a, in[3] + 0xc1bdceee, 22);
 209     MD5STEP(F1, a, b, c, d, in[4] + 0xf57c0faf, 7);
 210     MD5STEP(F1, d, a, b, c, in[5] + 0x4787c62a, 12);
 211     MD5STEP(F1, c, d, a, b, in[6] + 0xa8304613, 17);
 212     MD5STEP(F1, b, c, d, a, in[7] + 0xfd469501, 22);
 213     MD5STEP(F1, a, b, c, d, in[8] + 0x698098d8, 7);
 214     MD5STEP(F1, d, a, b, c, in[9] + 0x8b44f7af, 12);
 215     MD5STEP(F1, c, d, a, b, in[10] + 0xffff5bb1, 17);
 216     MD5STEP(F1, b, c, d, a, in[11] + 0x895cd7be, 22);
 217     MD5STEP(F1, a, b, c, d, in[12] + 0x6b901122, 7);
 218     MD5STEP(F1, d, a, b, c, in[13] + 0xfd987193, 12);
 219     MD5STEP(F1, c, d, a, b, in[14] + 0xa679438e, 17);
 220     MD5STEP(F1, b, c, d, a, in[15] + 0x49b40821, 22);
 221
 222     MD5STEP(F2, a, b, c, d, in[1] + 0xf61e2562, 5);
 223     MD5STEP(F2, d, a, b, c, in[6] + 0xc040b340, 9);
 224     MD5STEP(F2, c, d, a, b, in[11] + 0x265e5a51, 14);
 225     MD5STEP(F2, b, c, d, a, in[0] + 0xe9b6c7aa, 20);
 226     MD5STEP(F2, a, b, c, d, in[5] + 0xd62f105d, 5);
 227     MD5STEP(F2, d, a, b, c, in[10] + 0x02441453, 9);
 228     MD5STEP(F2, c, d, a, b, in[15] + 0xd8a1e681, 14);
 229     MD5STEP(F2, b, c, d, a, in[4] + 0xe7d3fbc8, 20);
 230     MD5STEP(F2, a, b, c, d, in[9] + 0x21e1cde6, 5);
 231     MD5STEP(F2, d, a, b, c, in[14] + 0xc33707d6, 9);
 232     MD5STEP(F2, c, d, a, b, in[3] + 0xf4d50d87, 14);
 233     MD5STEP(F2, b, c, d, a, in[8] + 0x455a14ed, 20);
 234     MD5STEP(F2, a, b, c, d, in[13] + 0xa9e3e905, 5);
 235     MD5STEP(F2, d, a, b, c, in[2] + 0xfcefa3f8, 9);
 236     MD5STEP(F2, c, d, a, b, in[7] + 0x676f02d9, 14);
 237     MD5STEP(F2, b, c, d, a, in[12] + 0x8d2a4c8a, 20);
 238
 239     MD5STEP(F3, a, b, c, d, in[5] + 0xfffa3942, 4);
 240     MD5STEP(F3, d, a, b, c, in[8] + 0x8771f681, 11);
 241     MD5STEP(F3, c, d, a, b, in[11] + 0x6d9d6122, 16);
 242     MD5STEP(F3, b, c, d, a, in[14] + 0xfde5380c, 23);
 243     MD5STEP(F3, a, b, c, d, in[1] + 0xa4beea44, 4);
 244     MD5STEP(F3, d, a, b, c, in[4] + 0x4bdecfa9, 11);
 245     MD5STEP(F3, c, d, a, b, in[7] + 0xf6bb4b60, 16);
 246     MD5STEP(F3, b, c, d, a, in[10] + 0xbebfbc70, 23);
 247     MD5STEP(F3, a, b, c, d, in[13] + 0x289b7ec6, 4);
 248     MD5STEP(F3, d, a, b, c, in[0] + 0xeaa127fa, 11);
 249     MD5STEP(F3, c, d, a, b, in[3] + 0xd4ef3085, 16);
 250     MD5STEP(F3, b, c, d, a, in[6] + 0x04881d05, 23);
 251     MD5STEP(F3, a, b, c, d, in[9] + 0xd9d4d039, 4);
 252     MD5STEP(F3, d, a, b, c, in[12] + 0xe6db99e5, 11);
 253     MD5STEP(F3, c, d, a, b, in[15] + 0x1fa27cf8, 16);
 254     MD5STEP(F3, b, c, d, a, in[2] + 0xc4ac5665, 23);
 255
 256     MD5STEP(F4, a, b, c, d, in[0] + 0xf4292244, 6);
 257     MD5STEP(F4, d, a, b, c, in[7] + 0x432aff97, 10);
 258     MD5STEP(F4, c, d, a, b, in[14] + 0xab9423a7, 15);
 259     MD5STEP(F4, b, c, d, a, in[5] + 0xfc93a039, 21);
 260     MD5STEP(F4, a, b, c, d, in[12] + 0x655b59c3, 6);
 261     MD5STEP(F4, d, a, b, c, in[3] + 0x8f0ccc92, 10);
 262     MD5STEP(F4, c, d, a, b, in[10] + 0xffeff47d, 15);
 263     MD5STEP(F4, b, c, d, a, in[1] + 0x85845dd1, 21);
 264     MD5STEP(F4, a, b, c, d, in[8] + 0x6fa87e4f, 6);
 265     MD5STEP(F4, d, a, b, c, in[15] + 0xfe2ce6e0, 10);
 266     MD5STEP(F4, c, d, a, b, in[6] + 0xa3014314, 15);
 267     MD5STEP(F4, b, c, d, a, in[13] + 0x4e0811a1, 21);
 268     MD5STEP(F4, a, b, c, d, in[4] + 0xf7537e82, 6);
 269     MD5STEP(F4, d, a, b, c, in[11] + 0xbd3af235, 10);
 270     MD5STEP(F4, c, d, a, b, in[2] + 0x2ad7d2bb, 15);
 271     MD5STEP(F4, b, c, d, a, in[9] + 0xeb86d391, 21);
 272
 273     buf[0] += a;
 274     buf[1] += b;
 275     buf[2] += c;
 276     buf[3] += d;
 277 }
 278
 279 #endif