/* md4.c

   The MD4 hash function, described in RFC 1320.

   Copyright (C) 2003 Niels Möller, Marcus Comstedt

   This file is part of GNU Nettle.

   GNU Nettle is free software: you can redistribute it and/or
   modify it under the terms of either:

     * the GNU Lesser General Public License as published by the Free
       Software Foundation; either version 3 of the License, or (at your
       option) any later version.

   or

     * the GNU General Public License as published by the Free
       Software Foundation; either version 2 of the License, or (at your
       option) any later version.

   or both in parallel, as here.

   GNU Nettle 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 copies of the GNU General Public License and
   the GNU Lesser General Public License along with this program.  If
   not, see http://www.gnu.org/licenses/.
*/

/* Based on the public domain md5 code, and modified by Marcus
   Comstedt */

#if HAVE_CONFIG_H
# include "config.h"
#endif

#include <assert.h>
#include <string.h>

#include "md4.h"

#include "macros.h"
#include "nettle-write.h"

/* A block, treated as a sequence of 32-bit words. */
#define MD4_DATA_LENGTH 16

static void
md4_transform(uint32_t *digest, const uint32_t *data);

static void
md4_compress(struct md4_ctx *ctx, const uint8_t *block);

/* FIXME: Could be an alias for md5_init */
void
md4_init(struct md4_ctx *ctx)
{
  /* Same constants as for md5. */
  const uint32_t iv[_MD4_DIGEST_LENGTH] =
    {
      0x67452301,
      0xefcdab89,
      0x98badcfe,
      0x10325476,
    };
  memcpy(ctx->state, iv, sizeof(ctx->state));
  
  ctx->count = 0;
  ctx->index = 0;
}

void
md4_update(struct md4_ctx *ctx,
	   size_t length,
	   const uint8_t *data)
{
  MD_UPDATE(ctx, length, data, md4_compress, ctx->count++);
}

void
md4_digest(struct md4_ctx *ctx,
	   size_t length,
	   uint8_t *digest)
{
  uint64_t bit_count;
  uint32_t data[MD4_DATA_LENGTH];
  unsigned i;

  assert(length <= MD4_DIGEST_SIZE);

  MD_PAD(ctx, 8, md4_compress);
  for (i = 0; i < MD4_DATA_LENGTH - 2; i++)
    data[i] = LE_READ_UINT32(ctx->block + 4*i);

  /* There are 512 = 2^9 bits in one block 
   * Little-endian order => Least significant word first */
  bit_count = (ctx->count << 9) | (ctx->index << 3);
  data[MD4_DATA_LENGTH-2] = bit_count;
  data[MD4_DATA_LENGTH-1] = bit_count >> 32;
  md4_transform(ctx->state, data);

  _nettle_write_le32(length, digest, ctx->state);
  md4_init(ctx);
}

/* MD4 functions */
#define F(x, y, z) (((y) & (x)) | ((z) & ~(x)))
#define G(x, y, z) (((y) & (x)) | ((z) & (x)) | ((y) & (z)))
#define H(x, y, z) ((x) ^ (y) ^ (z))

#define ROUND(f, w, x, y, z, data, s) \
( w += f(x, y, z) + data,  w = w<<s | w>>(32-s) )

/* Perform the MD4 transformation on one full block of 16 32-bit words. */
   
static void
md4_transform(uint32_t *digest, const uint32_t *data)
{
  uint32_t a, b, c, d;
  a = digest[0];
  b = digest[1];
  c = digest[2];
  d = digest[3];

  ROUND(F, a, b, c, d, data[ 0], 3);
  ROUND(F, d, a, b, c, data[ 1], 7);
  ROUND(F, c, d, a, b, data[ 2], 11);
  ROUND(F, b, c, d, a, data[ 3], 19);
  ROUND(F, a, b, c, d, data[ 4], 3);
  ROUND(F, d, a, b, c, data[ 5], 7);
  ROUND(F, c, d, a, b, data[ 6], 11);
  ROUND(F, b, c, d, a, data[ 7], 19);
  ROUND(F, a, b, c, d, data[ 8], 3);
  ROUND(F, d, a, b, c, data[ 9], 7);
  ROUND(F, c, d, a, b, data[10], 11);
  ROUND(F, b, c, d, a, data[11], 19);
  ROUND(F, a, b, c, d, data[12], 3);
  ROUND(F, d, a, b, c, data[13], 7);
  ROUND(F, c, d, a, b, data[14], 11);
  ROUND(F, b, c, d, a, data[15], 19);

  ROUND(G, a, b, c, d, data[ 0] + 0x5a827999, 3);
  ROUND(G, d, a, b, c, data[ 4] + 0x5a827999, 5);
  ROUND(G, c, d, a, b, data[ 8] + 0x5a827999, 9);
  ROUND(G, b, c, d, a, data[12] + 0x5a827999, 13);
  ROUND(G, a, b, c, d, data[ 1] + 0x5a827999, 3);
  ROUND(G, d, a, b, c, data[ 5] + 0x5a827999, 5);
  ROUND(G, c, d, a, b, data[ 9] + 0x5a827999, 9);
  ROUND(G, b, c, d, a, data[13] + 0x5a827999, 13);
  ROUND(G, a, b, c, d, data[ 2] + 0x5a827999, 3);
  ROUND(G, d, a, b, c, data[ 6] + 0x5a827999, 5);
  ROUND(G, c, d, a, b, data[10] + 0x5a827999, 9);
  ROUND(G, b, c, d, a, data[14] + 0x5a827999, 13);
  ROUND(G, a, b, c, d, data[ 3] + 0x5a827999, 3);
  ROUND(G, d, a, b, c, data[ 7] + 0x5a827999, 5);
  ROUND(G, c, d, a, b, data[11] + 0x5a827999, 9);
  ROUND(G, b, c, d, a, data[15] + 0x5a827999, 13);

  ROUND(H, a, b, c, d, data[ 0] + 0x6ed9eba1, 3);
  ROUND(H, d, a, b, c, data[ 8] + 0x6ed9eba1, 9);
  ROUND(H, c, d, a, b, data[ 4] + 0x6ed9eba1, 11);
  ROUND(H, b, c, d, a, data[12] + 0x6ed9eba1, 15);
  ROUND(H, a, b, c, d, data[ 2] + 0x6ed9eba1, 3);
  ROUND(H, d, a, b, c, data[10] + 0x6ed9eba1, 9);
  ROUND(H, c, d, a, b, data[ 6] + 0x6ed9eba1, 11);
  ROUND(H, b, c, d, a, data[14] + 0x6ed9eba1, 15);
  ROUND(H, a, b, c, d, data[ 1] + 0x6ed9eba1, 3);
  ROUND(H, d, a, b, c, data[ 9] + 0x6ed9eba1, 9);
  ROUND(H, c, d, a, b, data[ 5] + 0x6ed9eba1, 11);
  ROUND(H, b, c, d, a, data[13] + 0x6ed9eba1, 15);
  ROUND(H, a, b, c, d, data[ 3] + 0x6ed9eba1, 3);
  ROUND(H, d, a, b, c, data[11] + 0x6ed9eba1, 9);
  ROUND(H, c, d, a, b, data[ 7] + 0x6ed9eba1, 11);
  ROUND(H, b, c, d, a, data[15] + 0x6ed9eba1, 15);

  digest[0] += a;
  digest[1] += b;
  digest[2] += c;
  digest[3] += d;
}

static void
md4_compress(struct md4_ctx *ctx, const uint8_t *block)
{
  uint32_t data[MD4_DATA_LENGTH];
  unsigned i;
  
  /* Endian independent conversion */
  for (i = 0; i<16; i++, block += 4)
    data[i] = LE_READ_UINT32(block);

  md4_transform(ctx->state, data);
}