byterun/md5.c

   1 /***********************************************************************/
   2 /*                                                                     */
   3 /*                           Objective Caml                            */
   4 /*                                                                     */
   5 /*            Xavier Leroy, projet Cristal, INRIA Rocquencourt         */
   6 /*                                                                     */
   7 /*  Copyright 1996 Institut National de Recherche en Informatique et   */
   8 /*  en Automatique.  All rights reserved.  This file is distributed    */
   9 /*  under the terms of the GNU Library General Public License, with    */
  10 /*  the special exception on linking described in file ../LICENSE.     */
  11 /*                                                                     */
  12 /***********************************************************************/
  13
  14 /* $Id$ */
  15
  16 #include <string.h>
  17 #include "alloc.h"
  18 #include "fail.h"
  19 #include "md5.h"
  20 #include "memory.h"
  21 #include "mlvalues.h"
  22 #include "io.h"
  23 #include "reverse.h"
  24
  25 /* MD5 message digest */
  26
  27 CAMLprim value caml_md5_string(value str, value ofs, value len)
  28 {
  29   struct MD5Context ctx;
  30   value res;
  31   caml_MD5Init(&ctx);
  32   caml_MD5Update(&ctx, &Byte_u(str, Long_val(ofs)), Long_val(len));
  33   res = caml_alloc_string(16);
  34   caml_MD5Final(&Byte_u(res, 0), &ctx);
  35   return res;
  36 }
  37
  38 CAMLprim value caml_md5_chan(value vchan, value len)
  39 {
  40   CAMLparam2 (vchan, len);
  41   struct channel * chan = Channel(vchan);
  42   struct MD5Context ctx;
  43   value res;
  44   intnat toread, read;
  45   char buffer[4096];
  46
  47   Lock(chan);
  48   caml_MD5Init(&ctx);
  49   toread = Long_val(len);
  50   if (toread < 0){
  51     while (1){
  52       read = caml_getblock (chan, buffer, sizeof(buffer));
  53       if (read == 0) break;
  54       caml_MD5Update (&ctx, (unsigned char *) buffer, read);
  55     }
  56   }else{
  57     while (toread > 0) {
  58       read = caml_getblock(chan, buffer,
  59                            toread > sizeof(buffer) ? sizeof(buffer) : toread);
  60       if (read == 0) caml_raise_end_of_file();
  61       caml_MD5Update(&ctx, (unsigned char *) buffer, read);
  62       toread -= read;
  63     }
  64   }
  65   res = caml_alloc_string(16);
  66   caml_MD5Final(&Byte_u(res, 0), &ctx);
  67   Unlock(chan);
  68   CAMLreturn (res);
  69 }
  70
  71 /*
  72  * This code implements the MD5 message-digest algorithm.
  73  * The algorithm is due to Ron Rivest.  This code was
  74  * written by Colin Plumb in 1993, no copyright is claimed.
  75  * This code is in the public domain; do with it what you wish.
  76  *
  77  * Equivalent code is available from RSA Data Security, Inc.
  78  * This code has been tested against that, and is equivalent,
  79  * except that you don't need to include two pages of legalese
  80  * with every copy.
  81  *
  82  * To compute the message digest of a chunk of bytes, declare an
  83  * MD5Context structure, pass it to caml_MD5Init, call caml_MD5Update as
  84  * needed on buffers full of bytes, and then call caml_MD5Final, which
  85  * will fill a supplied 16-byte array with the digest.
  86  */
  87
  88 #ifndef ARCH_BIG_ENDIAN
  89 #define byteReverse(buf, len)   /* Nothing */
  90 #else
  91 static void byteReverse(unsigned char * buf, unsigned longs)
  92 {
  93     uint32 t;
  94     do {
  95         t = (uint32) ((unsigned) buf[3] << 8 | buf[2]) << 16 |
  96             ((unsigned) buf[1] << 8 | buf[0]);
  97         *(uint32 *) buf = t;
  98         buf += 4;
  99     } while (--longs);
 100 }
 101 #endif
 102
 103 /*
 104  * Start MD5 accumulation.  Set bit count to 0 and buffer to mysterious
 105  * initialization constants.
 106  */
 107 CAMLexport void caml_MD5Init(struct MD5Context *ctx)
 108 {
 109     ctx->buf[0] = 0x67452301;
 110     ctx->buf[1] = 0xefcdab89;
 111     ctx->buf[2] = 0x98badcfe;
 112     ctx->buf[3] = 0x10325476;
 113
 114     ctx->bits[0] = 0;
 115     ctx->bits[1] = 0;
 116 }
 117
 118 /*
 119  * Update context to reflect the concatenation of another buffer full
 120  * of bytes.
 121  */
 122 CAMLexport void caml_MD5Update(struct MD5Context *ctx, unsigned char *buf,
 123                                uintnat len)
 124 {
 125     uint32 t;
 126
 127     /* Update bitcount */
 128
 129     t = ctx->bits[0];
 130     if ((ctx->bits[0] = t + ((uint32) len << 3)) < t)
 131         ctx->bits[1]++;         /* Carry from low to high */
 132     ctx->bits[1] += len >> 29;
 133
 134     t = (t >> 3) & 0x3f;        /* Bytes already in shsInfo->data */
 135
 136     /* Handle any leading odd-sized chunks */
 137
 138     if (t) {
 139         unsigned char *p = (unsigned char *) ctx->in + t;
 140
 141         t = 64 - t;
 142         if (len < t) {
 143             memcpy(p, buf, len);
 144             return;
 145         }
 146         memcpy(p, buf, t);
 147         byteReverse(ctx->in, 16);
 148         caml_MD5Transform(ctx->buf, (uint32 *) ctx->in);
 149         buf += t;
 150         len -= t;
 151     }
 152     /* Process data in 64-byte chunks */
 153
 154     while (len >= 64) {
 155         memcpy(ctx->in, buf, 64);
 156         byteReverse(ctx->in, 16);
 157         caml_MD5Transform(ctx->buf, (uint32 *) ctx->in);
 158         buf += 64;
 159         len -= 64;
 160     }
 161
 162     /* Handle any remaining bytes of data. */
 163
 164     memcpy(ctx->in, buf, len);
 165 }
 166
 167 /*
 168  * Final wrapup - pad to 64-byte boundary with the bit pattern
 169  * 1 0* (64-bit count of bits processed, MSB-first)
 170  */
 171 CAMLexport void caml_MD5Final(unsigned char *digest, struct MD5Context *ctx)
 172 {
 173     unsigned count;
 174     unsigned char *p;
 175
 176     /* Compute number of bytes mod 64 */
 177     count = (ctx->bits[0] >> 3) & 0x3F;
 178
 179     /* Set the first char of padding to 0x80.  This is safe since there is
 180        always at least one byte free */
 181     p = ctx->in + count;
 182     *p++ = 0x80;
 183
 184     /* Bytes of padding needed to make 64 bytes */
 185     count = 64 - 1 - count;
 186
 187     /* Pad out to 56 mod 64 */
 188     if (count < 8) {
 189         /* Two lots of padding:  Pad the first block to 64 bytes */
 190         memset(p, 0, count);
 191         byteReverse(ctx->in, 16);
 192         caml_MD5Transform(ctx->buf, (uint32 *) ctx->in);
 193
 194         /* Now fill the next block with 56 bytes */
 195         memset(ctx->in, 0, 56);
 196     } else {
 197         /* Pad block to 56 bytes */
 198         memset(p, 0, count - 8);
 199     }
 200     byteReverse(ctx->in, 14);
 201
 202     /* Append length in bits and transform */
 203     ((uint32 *) ctx->in)[14] = ctx->bits[0];
 204     ((uint32 *) ctx->in)[15] = ctx->bits[1];
 205
 206     caml_MD5Transform(ctx->buf, (uint32 *) ctx->in);
 207     byteReverse((unsigned char *) ctx->buf, 4);
 208     memcpy(digest, ctx->buf, 16);
 209     memset(ctx, 0, sizeof(ctx));        /* In case it's sensitive */
 210 }
 211
 212 /* The four core functions - F1 is optimized somewhat */
 213
 214 /* #define F1(x, y, z) (x & y | ~x & z) */
 215 #define F1(x, y, z) (z ^ (x & (y ^ z)))
 216 #define F2(x, y, z) F1(z, x, y)
 217 #define F3(x, y, z) (x ^ y ^ z)
 218 #define F4(x, y, z) (y ^ (x | ~z))
 219
 220 /* This is the central step in the MD5 algorithm. */
 221 #define MD5STEP(f, w, x, y, z, data, s) \
 222         ( w += f(x, y, z) + data,  w = w<<s | w>>(32-s),  w += x )
 223
 224 /*
 225  * The core of the MD5 algorithm, this alters an existing MD5 hash to
 226  * reflect the addition of 16 longwords of new data.  caml_MD5Update blocks
 227  * the data and converts bytes into longwords for this routine.
 228  */
 229 CAMLexport void caml_MD5Transform(uint32 *buf, uint32 *in)
 230 {
 231     register uint32 a, b, c, d;
 232
 233     a = buf[0];
 234     b = buf[1];
 235     c = buf[2];
 236     d = buf[3];
 237
 238     MD5STEP(F1, a, b, c, d, in[0] + 0xd76aa478, 7);
 239     MD5STEP(F1, d, a, b, c, in[1] + 0xe8c7b756, 12);
 240     MD5STEP(F1, c, d, a, b, in[2] + 0x242070db, 17);
 241     MD5STEP(F1, b, c, d, a, in[3] + 0xc1bdceee, 22);
 242     MD5STEP(F1, a, b, c, d, in[4] + 0xf57c0faf, 7);
 243     MD5STEP(F1, d, a, b, c, in[5] + 0x4787c62a, 12);
 244     MD5STEP(F1, c, d, a, b, in[6] + 0xa8304613, 17);
 245     MD5STEP(F1, b, c, d, a, in[7] + 0xfd469501, 22);
 246     MD5STEP(F1, a, b, c, d, in[8] + 0x698098d8, 7);
 247     MD5STEP(F1, d, a, b, c, in[9] + 0x8b44f7af, 12);
 248     MD5STEP(F1, c, d, a, b, in[10] + 0xffff5bb1, 17);
 249     MD5STEP(F1, b, c, d, a, in[11] + 0x895cd7be, 22);
 250     MD5STEP(F1, a, b, c, d, in[12] + 0x6b901122, 7);
 251     MD5STEP(F1, d, a, b, c, in[13] + 0xfd987193, 12);
 252     MD5STEP(F1, c, d, a, b, in[14] + 0xa679438e, 17);
 253     MD5STEP(F1, b, c, d, a, in[15] + 0x49b40821, 22);
 254
 255     MD5STEP(F2, a, b, c, d, in[1] + 0xf61e2562, 5);
 256     MD5STEP(F2, d, a, b, c, in[6] + 0xc040b340, 9);
 257     MD5STEP(F2, c, d, a, b, in[11] + 0x265e5a51, 14);
 258     MD5STEP(F2, b, c, d, a, in[0] + 0xe9b6c7aa, 20);
 259     MD5STEP(F2, a, b, c, d, in[5] + 0xd62f105d, 5);
 260     MD5STEP(F2, d, a, b, c, in[10] + 0x02441453, 9);
 261     MD5STEP(F2, c, d, a, b, in[15] + 0xd8a1e681, 14);
 262     MD5STEP(F2, b, c, d, a, in[4] + 0xe7d3fbc8, 20);
 263     MD5STEP(F2, a, b, c, d, in[9] + 0x21e1cde6, 5);
 264     MD5STEP(F2, d, a, b, c, in[14] + 0xc33707d6, 9);
 265     MD5STEP(F2, c, d, a, b, in[3] + 0xf4d50d87, 14);
 266     MD5STEP(F2, b, c, d, a, in[8] + 0x455a14ed, 20);
 267     MD5STEP(F2, a, b, c, d, in[13] + 0xa9e3e905, 5);
 268     MD5STEP(F2, d, a, b, c, in[2] + 0xfcefa3f8, 9);
 269     MD5STEP(F2, c, d, a, b, in[7] + 0x676f02d9, 14);
 270     MD5STEP(F2, b, c, d, a, in[12] + 0x8d2a4c8a, 20);
 271
 272     MD5STEP(F3, a, b, c, d, in[5] + 0xfffa3942, 4);
 273     MD5STEP(F3, d, a, b, c, in[8] + 0x8771f681, 11);
 274     MD5STEP(F3, c, d, a, b, in[11] + 0x6d9d6122, 16);
 275     MD5STEP(F3, b, c, d, a, in[14] + 0xfde5380c, 23);
 276     MD5STEP(F3, a, b, c, d, in[1] + 0xa4beea44, 4);
 277     MD5STEP(F3, d, a, b, c, in[4] + 0x4bdecfa9, 11);
 278     MD5STEP(F3, c, d, a, b, in[7] + 0xf6bb4b60, 16);
 279     MD5STEP(F3, b, c, d, a, in[10] + 0xbebfbc70, 23);
 280     MD5STEP(F3, a, b, c, d, in[13] + 0x289b7ec6, 4);
 281     MD5STEP(F3, d, a, b, c, in[0] + 0xeaa127fa, 11);
 282     MD5STEP(F3, c, d, a, b, in[3] + 0xd4ef3085, 16);
 283     MD5STEP(F3, b, c, d, a, in[6] + 0x04881d05, 23);
 284     MD5STEP(F3, a, b, c, d, in[9] + 0xd9d4d039, 4);
 285     MD5STEP(F3, d, a, b, c, in[12] + 0xe6db99e5, 11);
 286     MD5STEP(F3, c, d, a, b, in[15] + 0x1fa27cf8, 16);
 287     MD5STEP(F3, b, c, d, a, in[2] + 0xc4ac5665, 23);
 288
 289     MD5STEP(F4, a, b, c, d, in[0] + 0xf4292244, 6);
 290     MD5STEP(F4, d, a, b, c, in[7] + 0x432aff97, 10);
 291     MD5STEP(F4, c, d, a, b, in[14] + 0xab9423a7, 15);
 292     MD5STEP(F4, b, c, d, a, in[5] + 0xfc93a039, 21);
 293     MD5STEP(F4, a, b, c, d, in[12] + 0x655b59c3, 6);
 294     MD5STEP(F4, d, a, b, c, in[3] + 0x8f0ccc92, 10);
 295     MD5STEP(F4, c, d, a, b, in[10] + 0xffeff47d, 15);
 296     MD5STEP(F4, b, c, d, a, in[1] + 0x85845dd1, 21);
 297     MD5STEP(F4, a, b, c, d, in[8] + 0x6fa87e4f, 6);
 298     MD5STEP(F4, d, a, b, c, in[15] + 0xfe2ce6e0, 10);
 299     MD5STEP(F4, c, d, a, b, in[6] + 0xa3014314, 15);
 300     MD5STEP(F4, b, c, d, a, in[13] + 0x4e0811a1, 21);
 301     MD5STEP(F4, a, b, c, d, in[4] + 0xf7537e82, 6);
 302     MD5STEP(F4, d, a, b, c, in[11] + 0xbd3af235, 10);
 303     MD5STEP(F4, c, d, a, b, in[2] + 0x2ad7d2bb, 15);
 304     MD5STEP(F4, b, c, d, a, in[9] + 0xeb86d391, 21);
 305
 306     buf[0] += a;
 307     buf[1] += b;
 308     buf[2] += c;
 309     buf[3] += d;
 310 }
 311