VextaProject/ Vexta
Vexta Vexta-CPU-Miner Vexta-Android-Wallet
Watch
1
0
Fork
You've already forked Vexta
 0
Code Issues Pull requests Projects Packages Wiki Activity Actions

Initial Vexta Core v0.1.0 release

Browse source
This commit is contained in:
Capri 2026-07-15 14:07:28 +00:00
commit 25c454329d
2481 changed files with 689289 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

41
src/crypto/KeccakP-800-SnP.h Normal file
View file

@ -0,0 +1,41 @@
/*
Implementation by the Keccak, Keyak and Ketje Teams, namely, Guido Bertoni,
Joan Daemen, Michaël Peeters, Gilles Van Assche and Ronny Van Keer, hereby
denoted as "the implementer".
For more information, feedback or questions, please refer to our websites:
http://keccak.noekeon.org/
http://keyak.noekeon.org/
http://ketje.noekeon.org/
To the extent possible under law, the implementer has waived all copyright
and related or neighboring rights to the source code in this file.
http://creativecommons.org/publicdomain/zero/1.0/
*/
#ifndef _KeccakP_800_SnP_h_
#define _KeccakP_800_SnP_h_
/** For the documentation, see SnP-documentation.h.
*/
#define KeccakP800_implementation "32-bit reference implementation"
#define KeccakP800_stateSizeInBytes 100
#define KeccakP800_stateAlignment 4
#ifdef KeccakReference
void KeccakP800_StaticInitialize( void );
#else
#define KeccakP800_StaticInitialize()
#endif
void KeccakP800_Initialize(void *state);
void KeccakP800_AddByte(void *state, unsigned char data, unsigned int offset);
void KeccakP800_AddBytes(void *state, const unsigned char *data, unsigned int offset, unsigned int length);
void KeccakP800_OverwriteBytes(void *state, const unsigned char *data, unsigned int offset, unsigned int length);
void KeccakP800_OverwriteWithZeroes(void *state, unsigned int byteCount);
void KeccakP800_Permute_12rounds(void *state);
void KeccakP800_Permute_22rounds(void *state);
void KeccakP800_ExtractBytes(const void *state, unsigned char *data, unsigned int offset, unsigned int length);
void KeccakP800_ExtractAndAddBytes(const void *state, const unsigned char *input, unsigned char *output, unsigned int offset, unsigned int length);
#endif

421
src/crypto/KeccakP-800-reference.cpp Normal file
View file

@ -0,0 +1,421 @@
/*
Implementation by the Keccak Team, namely, Guido Bertoni, Joan Daemen,
Michaël Peeters, Gilles Van Assche and Ronny Van Keer,
hereby denoted as "the implementer".
For more information, feedback or questions, please refer to our website:
https://keccak.team/
To the extent possible under law, the implementer has waived all copyright
and related or neighboring rights to the source code in this file.
http://creativecommons.org/publicdomain/zero/1.0/
---
This file implements Keccak-p[800] in a SnP-compatible way.
Please refer to SnP-documentation.h for more details.
This implementation comes with KeccakP-800-SnP.h in the same folder.
Please refer to LowLevel.build for the exact list of other files it must be combined with.
*/
#include <assert.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include "brg_endian.h"
#ifdef KeccakReference
#include "displayIntermediateValues.h"
#endif
typedef unsigned char UINT8;
typedef unsigned int UINT32;
typedef UINT32 tKeccakLane;
#define maxNrRounds 22
#define nrLanes 25
#define index(x, y) (((x)%5)+5*((y)%5))
#ifdef KeccakReference
static tKeccakLane KeccakRoundConstants[maxNrRounds];
static unsigned int KeccakRhoOffsets[nrLanes];
/* ---------------------------------------------------------------- */
void KeccakP800_InitializeRoundConstants(void);
void KeccakP800_InitializeRhoOffsets(void);
static int LFSR86540(UINT8 *LFSR);
void KeccakP800_StaticInitialize(void)
{
if (sizeof(tKeccakLane) != 4) {
printf("tKeccakLane should be 32-bit wide\n");
abort();
}
KeccakP800_InitializeRoundConstants();
KeccakP800_InitializeRhoOffsets();
}
void KeccakP800_InitializeRoundConstants(void)
{
UINT8 LFSRstate = 0x01;
unsigned int i, j, bitPosition;
for(i=0; i<maxNrRounds; i++) {
KeccakRoundConstants[i] = 0;
for(j=0; j<7; j++) {
bitPosition = (1<<j)-1; /* 2^j-1 */
if (LFSR86540(&LFSRstate) && (bitPosition < (sizeof(tKeccakLane)*8)))
KeccakRoundConstants[i] ^= (tKeccakLane)(1<<bitPosition);
}
}
}
void KeccakP800_InitializeRhoOffsets(void)
{
unsigned int x, y, t, newX, newY;
KeccakRhoOffsets[index(0, 0)] = 0;
x = 1;
y = 0;
for(t=0; t<24; t++) {
KeccakRhoOffsets[index(x, y)] = ((t+1)*(t+2)/2) % (sizeof(tKeccakLane) * 8);
newX = (0*x+1*y) % 5;
newY = (2*x+3*y) % 5;
x = newX;
y = newY;
}
}
static int LFSR86540(UINT8 *LFSR)
{
int result = ((*LFSR) & 0x01) != 0;
if (((*LFSR) & 0x80) != 0)
/* Primitive polynomial over GF(2): x^8+x^6+x^5+x^4+1 */
(*LFSR) = ((*LFSR) << 1) ^ 0x71;
else
(*LFSR) <<= 1;
return result;
}
#else
static const tKeccakLane KeccakRoundConstants[maxNrRounds] =
{
0x00000001,
0x00008082,
0x0000808a,
0x80008000,
0x0000808b,
0x80000001,
0x80008081,
0x00008009,
0x0000008a,
0x00000088,
0x80008009,
0x8000000a,
0x8000808b,
0x0000008b,
0x00008089,
0x00008003,
0x00008002,
0x00000080,
0x0000800a,
0x8000000a,
0x80008081,
0x00008080,
};
static const unsigned int KeccakRhoOffsets[nrLanes] =
{
0, 1, 30, 28, 27, 4, 12, 6, 23, 20, 3, 10, 11, 25, 7, 9, 13, 15, 21, 8, 18, 2, 29, 24, 14
};
#endif
/* ---------------------------------------------------------------- */
void KeccakP800_Initialize(void *state)
{
memset(state, 0, 800/8);
}
/* ---------------------------------------------------------------- */
void KeccakP800_AddByte(void *state, unsigned char byte, unsigned int offset)
{
assert(offset < 100);
((unsigned char *)state)[offset] ^= byte;
}
/* ---------------------------------------------------------------- */
void KeccakP800_AddBytes(void *state, const unsigned char *data, unsigned int offset, unsigned int length)
{
unsigned int i;
assert(offset < 100);
assert(offset+length <= 100);
for(i=0; i<length; i++)
((unsigned char *)state)[offset+i] ^= data[i];
}
/* ---------------------------------------------------------------- */
void KeccakP800_OverwriteBytes(void *state, const unsigned char *data, unsigned int offset, unsigned int length)
{
assert(offset < 100);
assert(offset+length <= 100);
memcpy((unsigned char*)state+offset, data, length);
}
/* ---------------------------------------------------------------- */
void KeccakP800_OverwriteWithZeroes(void *state, unsigned int byteCount)
{
assert(byteCount <= 100);
memset(state, 0, byteCount);
}
/* ---------------------------------------------------------------- */
static void fromBytesToWords(tKeccakLane *stateAsWords, const unsigned char *state);
static void fromWordsToBytes(unsigned char *state, const tKeccakLane *stateAsWords);
void KeccakP800OnWords(tKeccakLane *state, unsigned int nrRounds);
void KeccakP800Round(tKeccakLane *state, unsigned int indexRound);
static void theta(tKeccakLane *A);
static void rho(tKeccakLane *A);
static void pi(tKeccakLane *A);
static void chi(tKeccakLane *A);
static void iota(tKeccakLane *A, unsigned int indexRound);
void KeccakP800_Permute_Nrounds(void *state, unsigned int nrounds)
{
#if (PLATFORM_BYTE_ORDER != IS_LITTLE_ENDIAN)
tKeccakLane stateAsWords[800/32];
#endif
#ifdef KeccakReference
displayStateAsBytes(1, "Input of permutation", (const unsigned char *)state, 800);
#endif
#if (PLATFORM_BYTE_ORDER == IS_LITTLE_ENDIAN)
KeccakP800OnWords((tKeccakLane*)state, nrounds);
#else
fromBytesToWords(stateAsWords, (const unsigned char *)state);
KeccakP800OnWords(stateAsWords, nrounds);
fromWordsToBytes((unsigned char *)state, stateAsWords);
#endif
#ifdef KeccakReference
displayStateAsBytes(1, "State after permutation", (const unsigned char *)state, 800);
#endif
}
void KeccakP800_Permute_12rounds(void *state)
{
#if (PLATFORM_BYTE_ORDER != IS_LITTLE_ENDIAN)
tKeccakLane stateAsWords[800/32];
#endif
#ifdef KeccakReference
displayStateAsBytes(1, "Input of permutation", (const unsigned char *)state, 800);
#endif
#if (PLATFORM_BYTE_ORDER == IS_LITTLE_ENDIAN)
KeccakP800OnWords((tKeccakLane*)state, 12);
#else
fromBytesToWords(stateAsWords, (const unsigned char *)state);
KeccakP800OnWords(stateAsWords, 12);
fromWordsToBytes((unsigned char *)state, stateAsWords);
#endif
#ifdef KeccakReference
displayStateAsBytes(1, "State after permutation", (const unsigned char *)state, 800);
#endif
}
void KeccakP800_Permute_22rounds(void *state)
{
#if (PLATFORM_BYTE_ORDER != IS_LITTLE_ENDIAN)
tKeccakLane stateAsWords[800/32];
#endif
#ifdef KeccakReference
displayStateAsBytes(1, "Input of permutation", (const unsigned char *)state, 800);
#endif
#if (PLATFORM_BYTE_ORDER == IS_LITTLE_ENDIAN)
KeccakP800OnWords((tKeccakLane*)state, 22);
#else
fromBytesToWords(stateAsWords, (const unsigned char *)state);
KeccakP800OnWords(stateAsWords, 22);
fromWordsToBytes((unsigned char *)state, stateAsWords);
#endif
#ifdef KeccakReference
displayStateAsBytes(1, "State after permutation", (const unsigned char *)state, 800);
#endif
}
static void fromBytesToWords(tKeccakLane *stateAsWords, const unsigned char *state)
{
unsigned int i, j;
for(i=0; i<nrLanes; i++) {
stateAsWords[i] = 0;
for(j=0; j<sizeof(tKeccakLane); j++)
stateAsWords[i] |= (tKeccakLane)(state[i*sizeof(tKeccakLane)+j]) << (8*j);
}
}
static void fromWordsToBytes(unsigned char *state, const tKeccakLane *stateAsWords)
{
unsigned int i, j;
for(i=0; i<nrLanes; i++)
for(j=0; j<sizeof(tKeccakLane); j++)
state[i*sizeof(tKeccakLane)+j] = (stateAsWords[i] >> (8*j)) & 0xFF;
}
void KeccakP800OnWords(tKeccakLane *state, unsigned int nrRounds)
{
unsigned int i;
#ifdef KeccakReference
displayStateAsLanes(3, "Same, with lanes as 32-bit words", state, 800);
#endif
for(i=(maxNrRounds-nrRounds); i<maxNrRounds; i++)
KeccakP800Round(state, i);
}
void KeccakP800Round(tKeccakLane *state, unsigned int indexRound)
{
#ifdef KeccakReference
displayRoundNumber(3, indexRound);
#endif
theta(state);
#ifdef KeccakReference
displayStateAsLanes(3, "After theta", state, 800);
#endif
rho(state);
#ifdef KeccakReference
displayStateAsLanes(3, "After rho", state, 800);
#endif
pi(state);
#ifdef KeccakReference
displayStateAsLanes(3, "After pi", state, 800);
#endif
chi(state);
#ifdef KeccakReference
displayStateAsLanes(3, "After chi", state, 800);
#endif
iota(state, indexRound);
#ifdef KeccakReference
displayStateAsLanes(3, "After iota", state, 800);
#endif
}
#define ROL32(a, offset) ((offset != 0) ? ((((tKeccakLane)a) << offset) ^ (((tKeccakLane)a) >> (sizeof(tKeccakLane)*8-offset))) : a)
static void theta(tKeccakLane *A)
{
unsigned int x, y;
tKeccakLane C[5], D[5];
for(x=0; x<5; x++) {
C[x] = 0;
for(y=0; y<5; y++)
C[x] ^= A[index(x, y)];
}
for(x=0; x<5; x++)
D[x] = ROL32(C[(x+1)%5], 1) ^ C[(x+4)%5];
for(x=0; x<5; x++)
for(y=0; y<5; y++)
A[index(x, y)] ^= D[x];
}
static void rho(tKeccakLane *A)
{
unsigned int x, y;
for(x=0; x<5; x++) for(y=0; y<5; y++)
A[index(x, y)] = ROL32(A[index(x, y)], KeccakRhoOffsets[index(x, y)]);
}
static void pi(tKeccakLane *A)
{
unsigned int x, y;
tKeccakLane tempA[25];
for(x=0; x<5; x++) for(y=0; y<5; y++)
tempA[index(x, y)] = A[index(x, y)];
for(x=0; x<5; x++) for(y=0; y<5; y++)
A[index(0*x+1*y, 2*x+3*y)] = tempA[index(x, y)];
}
static void chi(tKeccakLane *A)
{
unsigned int x, y;
tKeccakLane C[5];
for(y=0; y<5; y++) {
for(x=0; x<5; x++)
C[x] = A[index(x, y)] ^ ((~A[index(x+1, y)]) & A[index(x+2, y)]);
for(x=0; x<5; x++)
A[index(x, y)] = C[x];
}
}
static void iota(tKeccakLane *A, unsigned int indexRound)
{
A[index(0, 0)] ^= KeccakRoundConstants[indexRound];
}
/* ---------------------------------------------------------------- */
void KeccakP800_ExtractBytes(const void *state, unsigned char *data, unsigned int offset, unsigned int length)
{
assert(offset < 100);
assert(offset+length <= 100);
memcpy(data, (unsigned char*)state+offset, length);
}
/* ---------------------------------------------------------------- */
void KeccakP800_ExtractAndAddBytes(const void *state, const unsigned char *input, unsigned char *output, unsigned int offset, unsigned int length)
{
unsigned int i;
assert(offset < 100);
assert(offset+length <= 100);
for(i=0; i<length; i++)
output[i] = input[i] ^ ((unsigned char *)state)[offset+i];
}
/* ---------------------------------------------------------------- */
void KeccakP800_DisplayRoundConstants(FILE *f)
{
unsigned int i;
for(i=0; i<maxNrRounds; i++) {
fprintf(f, "RC[%02i][0][0] = ", i);
fprintf(f, "%08X", (unsigned int)(KeccakRoundConstants[i]));
fprintf(f, "\n");
}
fprintf(f, "\n");
}
void KeccakP800_DisplayRhoOffsets(FILE *f)
{
unsigned int x, y;
for(y=0; y<5; y++) for(x=0; x<5; x++) {
fprintf(f, "RhoOffset[%i][%i] = ", x, y);
fprintf(f, "%2i", KeccakRhoOffsets[index(x, y)]);
fprintf(f, "\n");
}
fprintf(f, "\n");
}

151
src/crypto/aes.cpp Normal file
View file

@ -0,0 +1,151 @@
// Copyright (c) 2009-2020 The Bitcoin Core developers
// Copyright (c) 2014-2020 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#include <crypto/aes.h>
#include <string.h>
#include <crypto/ctaes/ctaes.cpp>
AES256Encrypt::AES256Encrypt(const unsigned char key[32])
{
AES256_init(&ctx, key);
}
AES256Encrypt::~AES256Encrypt()
{
memset(&ctx, 0, sizeof(ctx));
}
void AES256Encrypt::Encrypt(unsigned char ciphertext[16], const unsigned char plaintext[16]) const
{
AES256_encrypt(&ctx, 1, ciphertext, plaintext);
}
AES256Decrypt::AES256Decrypt(const unsigned char key[32])
{
AES256_init(&ctx, key);
}
AES256Decrypt::~AES256Decrypt()
{
memset(&ctx, 0, sizeof(ctx));
}
void AES256Decrypt::Decrypt(unsigned char plaintext[16], const unsigned char ciphertext[16]) const
{
AES256_decrypt(&ctx, 1, plaintext, ciphertext);
}
template <typename T>
static int CBCEncrypt(const T& enc, const unsigned char iv[AES_BLOCKSIZE], const unsigned char* data, int size, bool pad, unsigned char* out)
{
int written = 0;
int padsize = size % AES_BLOCKSIZE;
unsigned char mixed[AES_BLOCKSIZE];
if (!data || !size || !out)
return 0;
if (!pad && padsize != 0)
return 0;
memcpy(mixed, iv, AES_BLOCKSIZE);
// Write all but the last block
while (written + AES_BLOCKSIZE <= size) {
for (int i = 0; i != AES_BLOCKSIZE; i++)
mixed[i] ^= *data++;
enc.Encrypt(out + written, mixed);
memcpy(mixed, out + written, AES_BLOCKSIZE);
written += AES_BLOCKSIZE;
}
if (pad) {
// For all that remains, pad each byte with the value of the remaining
// space. If there is none, pad by a full block.
for (int i = 0; i != padsize; i++)
mixed[i] ^= *data++;
for (int i = padsize; i != AES_BLOCKSIZE; i++)
mixed[i] ^= AES_BLOCKSIZE - padsize;
enc.Encrypt(out + written, mixed);
written += AES_BLOCKSIZE;
}
return written;
}
template <typename T>
static int CBCDecrypt(const T& dec, const unsigned char iv[AES_BLOCKSIZE], const unsigned char* data, int size, bool pad, unsigned char* out)
{
int written = 0;
bool fail = false;
const unsigned char* prev = iv;
if (!data || !size || !out)
return 0;
if (size % AES_BLOCKSIZE != 0)
return 0;
// Decrypt all data. Padding will be checked in the output.
while (written != size) {
dec.Decrypt(out, data + written);
for (int i = 0; i != AES_BLOCKSIZE; i++)
*out++ ^= prev[i];
prev = data + written;
written += AES_BLOCKSIZE;
}
// When decrypting padding, attempt to run in constant-time
if (pad) {
// If used, padding size is the value of the last decrypted byte. For
// it to be valid, It must be between 1 and AES_BLOCKSIZE.
unsigned char padsize = *--out;
fail = !padsize | (padsize > AES_BLOCKSIZE);
// If not well-formed, treat it as though there's no padding.
padsize *= !fail;
// All padding must equal the last byte otherwise it's not well-formed
for (int i = AES_BLOCKSIZE; i != 0; i--)
fail |= ((i > AES_BLOCKSIZE - padsize) & (*out-- != padsize));
written -= padsize;
}
return written * !fail;
}
AES256CBCEncrypt::AES256CBCEncrypt(const unsigned char key[AES256_KEYSIZE], const unsigned char ivIn[AES_BLOCKSIZE], bool padIn)
: enc(key), pad(padIn)
{
memcpy(iv, ivIn, AES_BLOCKSIZE);
}
int AES256CBCEncrypt::Encrypt(const unsigned char* data, int size, unsigned char* out) const
{
return CBCEncrypt(enc, iv, data, size, pad, out);
}
AES256CBCEncrypt::~AES256CBCEncrypt()
{
memset(iv, 0, sizeof(iv));
}
AES256CBCDecrypt::AES256CBCDecrypt(const unsigned char key[AES256_KEYSIZE], const unsigned char ivIn[AES_BLOCKSIZE], bool padIn)
: dec(key), pad(padIn)
{
memcpy(iv, ivIn, AES_BLOCKSIZE);
}
int AES256CBCDecrypt::Decrypt(const unsigned char* data, int size, unsigned char* out) const
{
return CBCDecrypt(dec, iv, data, size, pad, out);
}
AES256CBCDecrypt::~AES256CBCDecrypt()
{
memset(iv, 0, sizeof(iv));
}

68
src/crypto/aes.h Normal file
View file

@ -0,0 +1,68 @@
// Copyright (c) 2009-2020 The Bitcoin Core developers
// Copyright (c) 2014-2020 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
//
// C++ wrapper around ctaes, a constant-time AES implementation
#ifndef DIGIBYTE_CRYPTO_AES_H
#define DIGIBYTE_CRYPTO_AES_H
extern "C" {
#include <crypto/ctaes/ctaes.h>
}
static const int AES_BLOCKSIZE = 16;
static const int AES256_KEYSIZE = 32;
/** An encryption class for AES-256. */
class AES256Encrypt
{
private:
AES256_ctx ctx;
public:
explicit AES256Encrypt(const unsigned char key[32]);
~AES256Encrypt();
void Encrypt(unsigned char ciphertext[16], const unsigned char plaintext[16]) const;
};
/** A decryption class for AES-256. */
class AES256Decrypt
{
private:
AES256_ctx ctx;
public:
explicit AES256Decrypt(const unsigned char key[32]);
~AES256Decrypt();
void Decrypt(unsigned char plaintext[16], const unsigned char ciphertext[16]) const;
};
class AES256CBCEncrypt
{
public:
AES256CBCEncrypt(const unsigned char key[AES256_KEYSIZE], const unsigned char ivIn[AES_BLOCKSIZE], bool padIn);
~AES256CBCEncrypt();
int Encrypt(const unsigned char* data, int size, unsigned char* out) const;
private:
const AES256Encrypt enc;
const bool pad;
unsigned char iv[AES_BLOCKSIZE];
};
class AES256CBCDecrypt
{
public:
AES256CBCDecrypt(const unsigned char key[AES256_KEYSIZE], const unsigned char ivIn[AES_BLOCKSIZE], bool padIn);
~AES256CBCDecrypt();
int Decrypt(const unsigned char* data, int size, unsigned char* out) const;
private:
const AES256Decrypt dec;
const bool pad;
unsigned char iv[AES_BLOCKSIZE];
};
#endif // DIGIBYTE_CRYPTO_AES_H

392
src/crypto/aes_helper.cpp Normal file
View file

@ -0,0 +1,392 @@
/* $Id: aes_helper.c 220 2010-06-09 09:21:50Z tp $ */
/*
* AES tables. This file is not meant to be compiled by itself; it
* is included by some hash function implementations. It contains
* the precomputed tables and helper macros for evaluating an AES
* round, optionally with a final XOR with a subkey.
*
* By default, this file defines the tables and macros for little-endian
* processing (i.e. it is assumed that the input bytes have been read
* from memory and assembled with the little-endian convention). If
* the 'AES_BIG_ENDIAN' macro is defined (to a non-zero integer value)
* when this file is included, then the tables and macros for big-endian
* processing are defined instead. The big-endian tables and macros have
* names distinct from the little-endian tables and macros, hence it is
* possible to have both simultaneously, by including this file twice
* (with and without the AES_BIG_ENDIAN macro).
*
* ==========================(LICENSE BEGIN)============================
*
* Copyright (c) 2007-2010 Projet RNRT SAPHIR
*
* Permission is hereby granted, free of charge, to any person obtaining
* a copy of this software and associated documentation files (the
* "Software"), to deal in the Software without restriction, including
* without limitation the rights to use, copy, modify, merge, publish,
* distribute, sublicense, and/or sell copies of the Software, and to
* permit persons to whom the Software is furnished to do so, subject to
* the following conditions:
*
* The above copyright notice and this permission notice shall be
* included in all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*
* ===========================(LICENSE END)=============================
*
* @author Thomas Pornin <thomas.pornin@cryptolog.com>
*/
#include "sph_types.h"
#ifdef __cplusplus
extern "C"{
#endif
#if AES_BIG_ENDIAN
#define AESx(x) ( ((SPH_C32(x) >> 24) & SPH_C32(0x000000FF)) \
| ((SPH_C32(x) >> 8) & SPH_C32(0x0000FF00)) \
| ((SPH_C32(x) << 8) & SPH_C32(0x00FF0000)) \
| ((SPH_C32(x) << 24) & SPH_C32(0xFF000000)))
#define AES0 AES0_BE
#define AES1 AES1_BE
#define AES2 AES2_BE
#define AES3 AES3_BE
#define AES_ROUND_BE(X0, X1, X2, X3, K0, K1, K2, K3, Y0, Y1, Y2, Y3) do { \
(Y0) = AES0[((X0) >> 24) & 0xFF] \
^ AES1[((X1) >> 16) & 0xFF] \
^ AES2[((X2) >> 8) & 0xFF] \
^ AES3[(X3) & 0xFF] ^ (K0); \
(Y1) = AES0[((X1) >> 24) & 0xFF] \
^ AES1[((X2) >> 16) & 0xFF] \
^ AES2[((X3) >> 8) & 0xFF] \
^ AES3[(X0) & 0xFF] ^ (K1); \
(Y2) = AES0[((X2) >> 24) & 0xFF] \
^ AES1[((X3) >> 16) & 0xFF] \
^ AES2[((X0) >> 8) & 0xFF] \
^ AES3[(X1) & 0xFF] ^ (K2); \
(Y3) = AES0[((X3) >> 24) & 0xFF] \
^ AES1[((X0) >> 16) & 0xFF] \
^ AES2[((X1) >> 8) & 0xFF] \
^ AES3[(X2) & 0xFF] ^ (K3); \
} while (0)
#define AES_ROUND_NOKEY_BE(X0, X1, X2, X3, Y0, Y1, Y2, Y3) \
AES_ROUND_BE(X0, X1, X2, X3, 0, 0, 0, 0, Y0, Y1, Y2, Y3)
#else
#define AESx(x) SPH_C32(x)
#define AES0 AES0_LE
#define AES1 AES1_LE
#define AES2 AES2_LE
#define AES3 AES3_LE
#define AES_ROUND_LE(X0, X1, X2, X3, K0, K1, K2, K3, Y0, Y1, Y2, Y3) do { \
(Y0) = AES0[(X0) & 0xFF] \
^ AES1[((X1) >> 8) & 0xFF] \
^ AES2[((X2) >> 16) & 0xFF] \
^ AES3[((X3) >> 24) & 0xFF] ^ (K0); \
(Y1) = AES0[(X1) & 0xFF] \
^ AES1[((X2) >> 8) & 0xFF] \
^ AES2[((X3) >> 16) & 0xFF] \
^ AES3[((X0) >> 24) & 0xFF] ^ (K1); \
(Y2) = AES0[(X2) & 0xFF] \
^ AES1[((X3) >> 8) & 0xFF] \
^ AES2[((X0) >> 16) & 0xFF] \
^ AES3[((X1) >> 24) & 0xFF] ^ (K2); \
(Y3) = AES0[(X3) & 0xFF] \
^ AES1[((X0) >> 8) & 0xFF] \
^ AES2[((X1) >> 16) & 0xFF] \
^ AES3[((X2) >> 24) & 0xFF] ^ (K3); \
} while (0)
#define AES_ROUND_NOKEY_LE(X0, X1, X2, X3, Y0, Y1, Y2, Y3) \
AES_ROUND_LE(X0, X1, X2, X3, 0, 0, 0, 0, Y0, Y1, Y2, Y3)
#endif
/*
* The AES*[] tables allow us to perform a fast evaluation of an AES
* round; table AESi[] combines SubBytes for a byte at row i, and
* MixColumns for the column where that byte goes after ShiftRows.
*/
static const sph_u32 AES0[256] = {
AESx(0xA56363C6), AESx(0x847C7CF8), AESx(0x997777EE), AESx(0x8D7B7BF6),
AESx(0x0DF2F2FF), AESx(0xBD6B6BD6), AESx(0xB16F6FDE), AESx(0x54C5C591),
AESx(0x50303060), AESx(0x03010102), AESx(0xA96767CE), AESx(0x7D2B2B56),
AESx(0x19FEFEE7), AESx(0x62D7D7B5), AESx(0xE6ABAB4D), AESx(0x9A7676EC),
AESx(0x45CACA8F), AESx(0x9D82821F), AESx(0x40C9C989), AESx(0x877D7DFA),
AESx(0x15FAFAEF), AESx(0xEB5959B2), AESx(0xC947478E), AESx(0x0BF0F0FB),
AESx(0xECADAD41), AESx(0x67D4D4B3), AESx(0xFDA2A25F), AESx(0xEAAFAF45),
AESx(0xBF9C9C23), AESx(0xF7A4A453), AESx(0x967272E4), AESx(0x5BC0C09B),
AESx(0xC2B7B775), AESx(0x1CFDFDE1), AESx(0xAE93933D), AESx(0x6A26264C),
AESx(0x5A36366C), AESx(0x413F3F7E), AESx(0x02F7F7F5), AESx(0x4FCCCC83),
AESx(0x5C343468), AESx(0xF4A5A551), AESx(0x34E5E5D1), AESx(0x08F1F1F9),
AESx(0x937171E2), AESx(0x73D8D8AB), AESx(0x53313162), AESx(0x3F15152A),
AESx(0x0C040408), AESx(0x52C7C795), AESx(0x65232346), AESx(0x5EC3C39D),
AESx(0x28181830), AESx(0xA1969637), AESx(0x0F05050A), AESx(0xB59A9A2F),
AESx(0x0907070E), AESx(0x36121224), AESx(0x9B80801B), AESx(0x3DE2E2DF),
AESx(0x26EBEBCD), AESx(0x6927274E), AESx(0xCDB2B27F), AESx(0x9F7575EA),
AESx(0x1B090912), AESx(0x9E83831D), AESx(0x742C2C58), AESx(0x2E1A1A34),
AESx(0x2D1B1B36), AESx(0xB26E6EDC), AESx(0xEE5A5AB4), AESx(0xFBA0A05B),
AESx(0xF65252A4), AESx(0x4D3B3B76), AESx(0x61D6D6B7), AESx(0xCEB3B37D),
AESx(0x7B292952), AESx(0x3EE3E3DD), AESx(0x712F2F5E), AESx(0x97848413),
AESx(0xF55353A6), AESx(0x68D1D1B9), AESx(0x00000000), AESx(0x2CEDEDC1),
AESx(0x60202040), AESx(0x1FFCFCE3), AESx(0xC8B1B179), AESx(0xED5B5BB6),
AESx(0xBE6A6AD4), AESx(0x46CBCB8D), AESx(0xD9BEBE67), AESx(0x4B393972),
AESx(0xDE4A4A94), AESx(0xD44C4C98), AESx(0xE85858B0), AESx(0x4ACFCF85),
AESx(0x6BD0D0BB), AESx(0x2AEFEFC5), AESx(0xE5AAAA4F), AESx(0x16FBFBED),
AESx(0xC5434386), AESx(0xD74D4D9A), AESx(0x55333366), AESx(0x94858511),
AESx(0xCF45458A), AESx(0x10F9F9E9), AESx(0x06020204), AESx(0x817F7FFE),
AESx(0xF05050A0), AESx(0x443C3C78), AESx(0xBA9F9F25), AESx(0xE3A8A84B),
AESx(0xF35151A2), AESx(0xFEA3A35D), AESx(0xC0404080), AESx(0x8A8F8F05),
AESx(0xAD92923F), AESx(0xBC9D9D21), AESx(0x48383870), AESx(0x04F5F5F1),
AESx(0xDFBCBC63), AESx(0xC1B6B677), AESx(0x75DADAAF), AESx(0x63212142),
AESx(0x30101020), AESx(0x1AFFFFE5), AESx(0x0EF3F3FD), AESx(0x6DD2D2BF),
AESx(0x4CCDCD81), AESx(0x140C0C18), AESx(0x35131326), AESx(0x2FECECC3),
AESx(0xE15F5FBE), AESx(0xA2979735), AESx(0xCC444488), AESx(0x3917172E),
AESx(0x57C4C493), AESx(0xF2A7A755), AESx(0x827E7EFC), AESx(0x473D3D7A),
AESx(0xAC6464C8), AESx(0xE75D5DBA), AESx(0x2B191932), AESx(0x957373E6),
AESx(0xA06060C0), AESx(0x98818119), AESx(0xD14F4F9E), AESx(0x7FDCDCA3),
AESx(0x66222244), AESx(0x7E2A2A54), AESx(0xAB90903B), AESx(0x8388880B),
AESx(0xCA46468C), AESx(0x29EEEEC7), AESx(0xD3B8B86B), AESx(0x3C141428),
AESx(0x79DEDEA7), AESx(0xE25E5EBC), AESx(0x1D0B0B16), AESx(0x76DBDBAD),
AESx(0x3BE0E0DB), AESx(0x56323264), AESx(0x4E3A3A74), AESx(0x1E0A0A14),
AESx(0xDB494992), AESx(0x0A06060C), AESx(0x6C242448), AESx(0xE45C5CB8),
AESx(0x5DC2C29F), AESx(0x6ED3D3BD), AESx(0xEFACAC43), AESx(0xA66262C4),
AESx(0xA8919139), AESx(0xA4959531), AESx(0x37E4E4D3), AESx(0x8B7979F2),
AESx(0x32E7E7D5), AESx(0x43C8C88B), AESx(0x5937376E), AESx(0xB76D6DDA),
AESx(0x8C8D8D01), AESx(0x64D5D5B1), AESx(0xD24E4E9C), AESx(0xE0A9A949),
AESx(0xB46C6CD8), AESx(0xFA5656AC), AESx(0x07F4F4F3), AESx(0x25EAEACF),
AESx(0xAF6565CA), AESx(0x8E7A7AF4), AESx(0xE9AEAE47), AESx(0x18080810),
AESx(0xD5BABA6F), AESx(0x887878F0), AESx(0x6F25254A), AESx(0x722E2E5C),
AESx(0x241C1C38), AESx(0xF1A6A657), AESx(0xC7B4B473), AESx(0x51C6C697),
AESx(0x23E8E8CB), AESx(0x7CDDDDA1), AESx(0x9C7474E8), AESx(0x211F1F3E),
AESx(0xDD4B4B96), AESx(0xDCBDBD61), AESx(0x868B8B0D), AESx(0x858A8A0F),
AESx(0x907070E0), AESx(0x423E3E7C), AESx(0xC4B5B571), AESx(0xAA6666CC),
AESx(0xD8484890), AESx(0x05030306), AESx(0x01F6F6F7), AESx(0x120E0E1C),
AESx(0xA36161C2), AESx(0x5F35356A), AESx(0xF95757AE), AESx(0xD0B9B969),
AESx(0x91868617), AESx(0x58C1C199), AESx(0x271D1D3A), AESx(0xB99E9E27),
AESx(0x38E1E1D9), AESx(0x13F8F8EB), AESx(0xB398982B), AESx(0x33111122),
AESx(0xBB6969D2), AESx(0x70D9D9A9), AESx(0x898E8E07), AESx(0xA7949433),
AESx(0xB69B9B2D), AESx(0x221E1E3C), AESx(0x92878715), AESx(0x20E9E9C9),
AESx(0x49CECE87), AESx(0xFF5555AA), AESx(0x78282850), AESx(0x7ADFDFA5),
AESx(0x8F8C8C03), AESx(0xF8A1A159), AESx(0x80898909), AESx(0x170D0D1A),
AESx(0xDABFBF65), AESx(0x31E6E6D7), AESx(0xC6424284), AESx(0xB86868D0),
AESx(0xC3414182), AESx(0xB0999929), AESx(0x772D2D5A), AESx(0x110F0F1E),
AESx(0xCBB0B07B), AESx(0xFC5454A8), AESx(0xD6BBBB6D), AESx(0x3A16162C)
};
static const sph_u32 AES1[256] = {
AESx(0x6363C6A5), AESx(0x7C7CF884), AESx(0x7777EE99), AESx(0x7B7BF68D),
AESx(0xF2F2FF0D), AESx(0x6B6BD6BD), AESx(0x6F6FDEB1), AESx(0xC5C59154),
AESx(0x30306050), AESx(0x01010203), AESx(0x6767CEA9), AESx(0x2B2B567D),
AESx(0xFEFEE719), AESx(0xD7D7B562), AESx(0xABAB4DE6), AESx(0x7676EC9A),
AESx(0xCACA8F45), AESx(0x82821F9D), AESx(0xC9C98940), AESx(0x7D7DFA87),
AESx(0xFAFAEF15), AESx(0x5959B2EB), AESx(0x47478EC9), AESx(0xF0F0FB0B),
AESx(0xADAD41EC), AESx(0xD4D4B367), AESx(0xA2A25FFD), AESx(0xAFAF45EA),
AESx(0x9C9C23BF), AESx(0xA4A453F7), AESx(0x7272E496), AESx(0xC0C09B5B),
AESx(0xB7B775C2), AESx(0xFDFDE11C), AESx(0x93933DAE), AESx(0x26264C6A),
AESx(0x36366C5A), AESx(0x3F3F7E41), AESx(0xF7F7F502), AESx(0xCCCC834F),
AESx(0x3434685C), AESx(0xA5A551F4), AESx(0xE5E5D134), AESx(0xF1F1F908),
AESx(0x7171E293), AESx(0xD8D8AB73), AESx(0x31316253), AESx(0x15152A3F),
AESx(0x0404080C), AESx(0xC7C79552), AESx(0x23234665), AESx(0xC3C39D5E),
AESx(0x18183028), AESx(0x969637A1), AESx(0x05050A0F), AESx(0x9A9A2FB5),
AESx(0x07070E09), AESx(0x12122436), AESx(0x80801B9B), AESx(0xE2E2DF3D),
AESx(0xEBEBCD26), AESx(0x27274E69), AESx(0xB2B27FCD), AESx(0x7575EA9F),
AESx(0x0909121B), AESx(0x83831D9E), AESx(0x2C2C5874), AESx(0x1A1A342E),
AESx(0x1B1B362D), AESx(0x6E6EDCB2), AESx(0x5A5AB4EE), AESx(0xA0A05BFB),
AESx(0x5252A4F6), AESx(0x3B3B764D), AESx(0xD6D6B761), AESx(0xB3B37DCE),
AESx(0x2929527B), AESx(0xE3E3DD3E), AESx(0x2F2F5E71), AESx(0x84841397),
AESx(0x5353A6F5), AESx(0xD1D1B968), AESx(0x00000000), AESx(0xEDEDC12C),
AESx(0x20204060), AESx(0xFCFCE31F), AESx(0xB1B179C8), AESx(0x5B5BB6ED),
AESx(0x6A6AD4BE), AESx(0xCBCB8D46), AESx(0xBEBE67D9), AESx(0x3939724B),
AESx(0x4A4A94DE), AESx(0x4C4C98D4), AESx(0x5858B0E8), AESx(0xCFCF854A),
AESx(0xD0D0BB6B), AESx(0xEFEFC52A), AESx(0xAAAA4FE5), AESx(0xFBFBED16),
AESx(0x434386C5), AESx(0x4D4D9AD7), AESx(0x33336655), AESx(0x85851194),
AESx(0x45458ACF), AESx(0xF9F9E910), AESx(0x02020406), AESx(0x7F7FFE81),
AESx(0x5050A0F0), AESx(0x3C3C7844), AESx(0x9F9F25BA), AESx(0xA8A84BE3),
AESx(0x5151A2F3), AESx(0xA3A35DFE), AESx(0x404080C0), AESx(0x8F8F058A),
AESx(0x92923FAD), AESx(0x9D9D21BC), AESx(0x38387048), AESx(0xF5F5F104),
AESx(0xBCBC63DF), AESx(0xB6B677C1), AESx(0xDADAAF75), AESx(0x21214263),
AESx(0x10102030), AESx(0xFFFFE51A), AESx(0xF3F3FD0E), AESx(0xD2D2BF6D),
AESx(0xCDCD814C), AESx(0x0C0C1814), AESx(0x13132635), AESx(0xECECC32F),
AESx(0x5F5FBEE1), AESx(0x979735A2), AESx(0x444488CC), AESx(0x17172E39),
AESx(0xC4C49357), AESx(0xA7A755F2), AESx(0x7E7EFC82), AESx(0x3D3D7A47),
AESx(0x6464C8AC), AESx(0x5D5DBAE7), AESx(0x1919322B), AESx(0x7373E695),
AESx(0x6060C0A0), AESx(0x81811998), AESx(0x4F4F9ED1), AESx(0xDCDCA37F),
AESx(0x22224466), AESx(0x2A2A547E), AESx(0x90903BAB), AESx(0x88880B83),
AESx(0x46468CCA), AESx(0xEEEEC729), AESx(0xB8B86BD3), AESx(0x1414283C),
AESx(0xDEDEA779), AESx(0x5E5EBCE2), AESx(0x0B0B161D), AESx(0xDBDBAD76),
AESx(0xE0E0DB3B), AESx(0x32326456), AESx(0x3A3A744E), AESx(0x0A0A141E),
AESx(0x494992DB), AESx(0x06060C0A), AESx(0x2424486C), AESx(0x5C5CB8E4),
AESx(0xC2C29F5D), AESx(0xD3D3BD6E), AESx(0xACAC43EF), AESx(0x6262C4A6),
AESx(0x919139A8), AESx(0x959531A4), AESx(0xE4E4D337), AESx(0x7979F28B),
AESx(0xE7E7D532), AESx(0xC8C88B43), AESx(0x37376E59), AESx(0x6D6DDAB7),
AESx(0x8D8D018C), AESx(0xD5D5B164), AESx(0x4E4E9CD2), AESx(0xA9A949E0),
AESx(0x6C6CD8B4), AESx(0x5656ACFA), AESx(0xF4F4F307), AESx(0xEAEACF25),
AESx(0x6565CAAF), AESx(0x7A7AF48E), AESx(0xAEAE47E9), AESx(0x08081018),
AESx(0xBABA6FD5), AESx(0x7878F088), AESx(0x25254A6F), AESx(0x2E2E5C72),
AESx(0x1C1C3824), AESx(0xA6A657F1), AESx(0xB4B473C7), AESx(0xC6C69751),
AESx(0xE8E8CB23), AESx(0xDDDDA17C), AESx(0x7474E89C), AESx(0x1F1F3E21),
AESx(0x4B4B96DD), AESx(0xBDBD61DC), AESx(0x8B8B0D86), AESx(0x8A8A0F85),
AESx(0x7070E090), AESx(0x3E3E7C42), AESx(0xB5B571C4), AESx(0x6666CCAA),
AESx(0x484890D8), AESx(0x03030605), AESx(0xF6F6F701), AESx(0x0E0E1C12),
AESx(0x6161C2A3), AESx(0x35356A5F), AESx(0x5757AEF9), AESx(0xB9B969D0),
AESx(0x86861791), AESx(0xC1C19958), AESx(0x1D1D3A27), AESx(0x9E9E27B9),
AESx(0xE1E1D938), AESx(0xF8F8EB13), AESx(0x98982BB3), AESx(0x11112233),
AESx(0x6969D2BB), AESx(0xD9D9A970), AESx(0x8E8E0789), AESx(0x949433A7),
AESx(0x9B9B2DB6), AESx(0x1E1E3C22), AESx(0x87871592), AESx(0xE9E9C920),
AESx(0xCECE8749), AESx(0x5555AAFF), AESx(0x28285078), AESx(0xDFDFA57A),
AESx(0x8C8C038F), AESx(0xA1A159F8), AESx(0x89890980), AESx(0x0D0D1A17),
AESx(0xBFBF65DA), AESx(0xE6E6D731), AESx(0x424284C6), AESx(0x6868D0B8),
AESx(0x414182C3), AESx(0x999929B0), AESx(0x2D2D5A77), AESx(0x0F0F1E11),
AESx(0xB0B07BCB), AESx(0x5454A8FC), AESx(0xBBBB6DD6), AESx(0x16162C3A)
};
static const sph_u32 AES2[256] = {
AESx(0x63C6A563), AESx(0x7CF8847C), AESx(0x77EE9977), AESx(0x7BF68D7B),
AESx(0xF2FF0DF2), AESx(0x6BD6BD6B), AESx(0x6FDEB16F), AESx(0xC59154C5),
AESx(0x30605030), AESx(0x01020301), AESx(0x67CEA967), AESx(0x2B567D2B),
AESx(0xFEE719FE), AESx(0xD7B562D7), AESx(0xAB4DE6AB), AESx(0x76EC9A76),
AESx(0xCA8F45CA), AESx(0x821F9D82), AESx(0xC98940C9), AESx(0x7DFA877D),
AESx(0xFAEF15FA), AESx(0x59B2EB59), AESx(0x478EC947), AESx(0xF0FB0BF0),
AESx(0xAD41ECAD), AESx(0xD4B367D4), AESx(0xA25FFDA2), AESx(0xAF45EAAF),
AESx(0x9C23BF9C), AESx(0xA453F7A4), AESx(0x72E49672), AESx(0xC09B5BC0),
AESx(0xB775C2B7), AESx(0xFDE11CFD), AESx(0x933DAE93), AESx(0x264C6A26),
AESx(0x366C5A36), AESx(0x3F7E413F), AESx(0xF7F502F7), AESx(0xCC834FCC),
AESx(0x34685C34), AESx(0xA551F4A5), AESx(0xE5D134E5), AESx(0xF1F908F1),
AESx(0x71E29371), AESx(0xD8AB73D8), AESx(0x31625331), AESx(0x152A3F15),
AESx(0x04080C04), AESx(0xC79552C7), AESx(0x23466523), AESx(0xC39D5EC3),
AESx(0x18302818), AESx(0x9637A196), AESx(0x050A0F05), AESx(0x9A2FB59A),
AESx(0x070E0907), AESx(0x12243612), AESx(0x801B9B80), AESx(0xE2DF3DE2),
AESx(0xEBCD26EB), AESx(0x274E6927), AESx(0xB27FCDB2), AESx(0x75EA9F75),
AESx(0x09121B09), AESx(0x831D9E83), AESx(0x2C58742C), AESx(0x1A342E1A),
AESx(0x1B362D1B), AESx(0x6EDCB26E), AESx(0x5AB4EE5A), AESx(0xA05BFBA0),
AESx(0x52A4F652), AESx(0x3B764D3B), AESx(0xD6B761D6), AESx(0xB37DCEB3),
AESx(0x29527B29), AESx(0xE3DD3EE3), AESx(0x2F5E712F), AESx(0x84139784),
AESx(0x53A6F553), AESx(0xD1B968D1), AESx(0x00000000), AESx(0xEDC12CED),
AESx(0x20406020), AESx(0xFCE31FFC), AESx(0xB179C8B1), AESx(0x5BB6ED5B),
AESx(0x6AD4BE6A), AESx(0xCB8D46CB), AESx(0xBE67D9BE), AESx(0x39724B39),
AESx(0x4A94DE4A), AESx(0x4C98D44C), AESx(0x58B0E858), AESx(0xCF854ACF),
AESx(0xD0BB6BD0), AESx(0xEFC52AEF), AESx(0xAA4FE5AA), AESx(0xFBED16FB),
AESx(0x4386C543), AESx(0x4D9AD74D), AESx(0x33665533), AESx(0x85119485),
AESx(0x458ACF45), AESx(0xF9E910F9), AESx(0x02040602), AESx(0x7FFE817F),
AESx(0x50A0F050), AESx(0x3C78443C), AESx(0x9F25BA9F), AESx(0xA84BE3A8),
AESx(0x51A2F351), AESx(0xA35DFEA3), AESx(0x4080C040), AESx(0x8F058A8F),
AESx(0x923FAD92), AESx(0x9D21BC9D), AESx(0x38704838), AESx(0xF5F104F5),
AESx(0xBC63DFBC), AESx(0xB677C1B6), AESx(0xDAAF75DA), AESx(0x21426321),
AESx(0x10203010), AESx(0xFFE51AFF), AESx(0xF3FD0EF3), AESx(0xD2BF6DD2),
AESx(0xCD814CCD), AESx(0x0C18140C), AESx(0x13263513), AESx(0xECC32FEC),
AESx(0x5FBEE15F), AESx(0x9735A297), AESx(0x4488CC44), AESx(0x172E3917),
AESx(0xC49357C4), AESx(0xA755F2A7), AESx(0x7EFC827E), AESx(0x3D7A473D),
AESx(0x64C8AC64), AESx(0x5DBAE75D), AESx(0x19322B19), AESx(0x73E69573),
AESx(0x60C0A060), AESx(0x81199881), AESx(0x4F9ED14F), AESx(0xDCA37FDC),
AESx(0x22446622), AESx(0x2A547E2A), AESx(0x903BAB90), AESx(0x880B8388),
AESx(0x468CCA46), AESx(0xEEC729EE), AESx(0xB86BD3B8), AESx(0x14283C14),
AESx(0xDEA779DE), AESx(0x5EBCE25E), AESx(0x0B161D0B), AESx(0xDBAD76DB),
AESx(0xE0DB3BE0), AESx(0x32645632), AESx(0x3A744E3A), AESx(0x0A141E0A),
AESx(0x4992DB49), AESx(0x060C0A06), AESx(0x24486C24), AESx(0x5CB8E45C),
AESx(0xC29F5DC2), AESx(0xD3BD6ED3), AESx(0xAC43EFAC), AESx(0x62C4A662),
AESx(0x9139A891), AESx(0x9531A495), AESx(0xE4D337E4), AESx(0x79F28B79),
AESx(0xE7D532E7), AESx(0xC88B43C8), AESx(0x376E5937), AESx(0x6DDAB76D),
AESx(0x8D018C8D), AESx(0xD5B164D5), AESx(0x4E9CD24E), AESx(0xA949E0A9),
AESx(0x6CD8B46C), AESx(0x56ACFA56), AESx(0xF4F307F4), AESx(0xEACF25EA),
AESx(0x65CAAF65), AESx(0x7AF48E7A), AESx(0xAE47E9AE), AESx(0x08101808),
AESx(0xBA6FD5BA), AESx(0x78F08878), AESx(0x254A6F25), AESx(0x2E5C722E),
AESx(0x1C38241C), AESx(0xA657F1A6), AESx(0xB473C7B4), AESx(0xC69751C6),
AESx(0xE8CB23E8), AESx(0xDDA17CDD), AESx(0x74E89C74), AESx(0x1F3E211F),
AESx(0x4B96DD4B), AESx(0xBD61DCBD), AESx(0x8B0D868B), AESx(0x8A0F858A),
AESx(0x70E09070), AESx(0x3E7C423E), AESx(0xB571C4B5), AESx(0x66CCAA66),
AESx(0x4890D848), AESx(0x03060503), AESx(0xF6F701F6), AESx(0x0E1C120E),
AESx(0x61C2A361), AESx(0x356A5F35), AESx(0x57AEF957), AESx(0xB969D0B9),
AESx(0x86179186), AESx(0xC19958C1), AESx(0x1D3A271D), AESx(0x9E27B99E),
AESx(0xE1D938E1), AESx(0xF8EB13F8), AESx(0x982BB398), AESx(0x11223311),
AESx(0x69D2BB69), AESx(0xD9A970D9), AESx(0x8E07898E), AESx(0x9433A794),
AESx(0x9B2DB69B), AESx(0x1E3C221E), AESx(0x87159287), AESx(0xE9C920E9),
AESx(0xCE8749CE), AESx(0x55AAFF55), AESx(0x28507828), AESx(0xDFA57ADF),
AESx(0x8C038F8C), AESx(0xA159F8A1), AESx(0x89098089), AESx(0x0D1A170D),
AESx(0xBF65DABF), AESx(0xE6D731E6), AESx(0x4284C642), AESx(0x68D0B868),
AESx(0x4182C341), AESx(0x9929B099), AESx(0x2D5A772D), AESx(0x0F1E110F),
AESx(0xB07BCBB0), AESx(0x54A8FC54), AESx(0xBB6DD6BB), AESx(0x162C3A16)
};
static const sph_u32 AES3[256] = {
AESx(0xC6A56363), AESx(0xF8847C7C), AESx(0xEE997777), AESx(0xF68D7B7B),
AESx(0xFF0DF2F2), AESx(0xD6BD6B6B), AESx(0xDEB16F6F), AESx(0x9154C5C5),
AESx(0x60503030), AESx(0x02030101), AESx(0xCEA96767), AESx(0x567D2B2B),
AESx(0xE719FEFE), AESx(0xB562D7D7), AESx(0x4DE6ABAB), AESx(0xEC9A7676),
AESx(0x8F45CACA), AESx(0x1F9D8282), AESx(0x8940C9C9), AESx(0xFA877D7D),
AESx(0xEF15FAFA), AESx(0xB2EB5959), AESx(0x8EC94747), AESx(0xFB0BF0F0),
AESx(0x41ECADAD), AESx(0xB367D4D4), AESx(0x5FFDA2A2), AESx(0x45EAAFAF),
AESx(0x23BF9C9C), AESx(0x53F7A4A4), AESx(0xE4967272), AESx(0x9B5BC0C0),
AESx(0x75C2B7B7), AESx(0xE11CFDFD), AESx(0x3DAE9393), AESx(0x4C6A2626),
AESx(0x6C5A3636), AESx(0x7E413F3F), AESx(0xF502F7F7), AESx(0x834FCCCC),
AESx(0x685C3434), AESx(0x51F4A5A5), AESx(0xD134E5E5), AESx(0xF908F1F1),
AESx(0xE2937171), AESx(0xAB73D8D8), AESx(0x62533131), AESx(0x2A3F1515),
AESx(0x080C0404), AESx(0x9552C7C7), AESx(0x46652323), AESx(0x9D5EC3C3),
AESx(0x30281818), AESx(0x37A19696), AESx(0x0A0F0505), AESx(0x2FB59A9A),
AESx(0x0E090707), AESx(0x24361212), AESx(0x1B9B8080), AESx(0xDF3DE2E2),
AESx(0xCD26EBEB), AESx(0x4E692727), AESx(0x7FCDB2B2), AESx(0xEA9F7575),
AESx(0x121B0909), AESx(0x1D9E8383), AESx(0x58742C2C), AESx(0x342E1A1A),
AESx(0x362D1B1B), AESx(0xDCB26E6E), AESx(0xB4EE5A5A), AESx(0x5BFBA0A0),
AESx(0xA4F65252), AESx(0x764D3B3B), AESx(0xB761D6D6), AESx(0x7DCEB3B3),
AESx(0x527B2929), AESx(0xDD3EE3E3), AESx(0x5E712F2F), AESx(0x13978484),
AESx(0xA6F55353), AESx(0xB968D1D1), AESx(0x00000000), AESx(0xC12CEDED),
AESx(0x40602020), AESx(0xE31FFCFC), AESx(0x79C8B1B1), AESx(0xB6ED5B5B),
AESx(0xD4BE6A6A), AESx(0x8D46CBCB), AESx(0x67D9BEBE), AESx(0x724B3939),
AESx(0x94DE4A4A), AESx(0x98D44C4C), AESx(0xB0E85858), AESx(0x854ACFCF),
AESx(0xBB6BD0D0), AESx(0xC52AEFEF), AESx(0x4FE5AAAA), AESx(0xED16FBFB),
AESx(0x86C54343), AESx(0x9AD74D4D), AESx(0x66553333), AESx(0x11948585),
AESx(0x8ACF4545), AESx(0xE910F9F9), AESx(0x04060202), AESx(0xFE817F7F),
AESx(0xA0F05050), AESx(0x78443C3C), AESx(0x25BA9F9F), AESx(0x4BE3A8A8),
AESx(0xA2F35151), AESx(0x5DFEA3A3), AESx(0x80C04040), AESx(0x058A8F8F),
AESx(0x3FAD9292), AESx(0x21BC9D9D), AESx(0x70483838), AESx(0xF104F5F5),
AESx(0x63DFBCBC), AESx(0x77C1B6B6), AESx(0xAF75DADA), AESx(0x42632121),
AESx(0x20301010), AESx(0xE51AFFFF), AESx(0xFD0EF3F3), AESx(0xBF6DD2D2),
AESx(0x814CCDCD), AESx(0x18140C0C), AESx(0x26351313), AESx(0xC32FECEC),
AESx(0xBEE15F5F), AESx(0x35A29797), AESx(0x88CC4444), AESx(0x2E391717),
AESx(0x9357C4C4), AESx(0x55F2A7A7), AESx(0xFC827E7E), AESx(0x7A473D3D),
AESx(0xC8AC6464), AESx(0xBAE75D5D), AESx(0x322B1919), AESx(0xE6957373),
AESx(0xC0A06060), AESx(0x19988181), AESx(0x9ED14F4F), AESx(0xA37FDCDC),
AESx(0x44662222), AESx(0x547E2A2A), AESx(0x3BAB9090), AESx(0x0B838888),
AESx(0x8CCA4646), AESx(0xC729EEEE), AESx(0x6BD3B8B8), AESx(0x283C1414),
AESx(0xA779DEDE), AESx(0xBCE25E5E), AESx(0x161D0B0B), AESx(0xAD76DBDB),
AESx(0xDB3BE0E0), AESx(0x64563232), AESx(0x744E3A3A), AESx(0x141E0A0A),
AESx(0x92DB4949), AESx(0x0C0A0606), AESx(0x486C2424), AESx(0xB8E45C5C),
AESx(0x9F5DC2C2), AESx(0xBD6ED3D3), AESx(0x43EFACAC), AESx(0xC4A66262),
AESx(0x39A89191), AESx(0x31A49595), AESx(0xD337E4E4), AESx(0xF28B7979),
AESx(0xD532E7E7), AESx(0x8B43C8C8), AESx(0x6E593737), AESx(0xDAB76D6D),
AESx(0x018C8D8D), AESx(0xB164D5D5), AESx(0x9CD24E4E), AESx(0x49E0A9A9),
AESx(0xD8B46C6C), AESx(0xACFA5656), AESx(0xF307F4F4), AESx(0xCF25EAEA),
AESx(0xCAAF6565), AESx(0xF48E7A7A), AESx(0x47E9AEAE), AESx(0x10180808),
AESx(0x6FD5BABA), AESx(0xF0887878), AESx(0x4A6F2525), AESx(0x5C722E2E),
AESx(0x38241C1C), AESx(0x57F1A6A6), AESx(0x73C7B4B4), AESx(0x9751C6C6),
AESx(0xCB23E8E8), AESx(0xA17CDDDD), AESx(0xE89C7474), AESx(0x3E211F1F),
AESx(0x96DD4B4B), AESx(0x61DCBDBD), AESx(0x0D868B8B), AESx(0x0F858A8A),
AESx(0xE0907070), AESx(0x7C423E3E), AESx(0x71C4B5B5), AESx(0xCCAA6666),
AESx(0x90D84848), AESx(0x06050303), AESx(0xF701F6F6), AESx(0x1C120E0E),
AESx(0xC2A36161), AESx(0x6A5F3535), AESx(0xAEF95757), AESx(0x69D0B9B9),
AESx(0x17918686), AESx(0x9958C1C1), AESx(0x3A271D1D), AESx(0x27B99E9E),
AESx(0xD938E1E1), AESx(0xEB13F8F8), AESx(0x2BB39898), AESx(0x22331111),
AESx(0xD2BB6969), AESx(0xA970D9D9), AESx(0x07898E8E), AESx(0x33A79494),
AESx(0x2DB69B9B), AESx(0x3C221E1E), AESx(0x15928787), AESx(0xC920E9E9),
AESx(0x8749CECE), AESx(0xAAFF5555), AESx(0x50782828), AESx(0xA57ADFDF),
AESx(0x038F8C8C), AESx(0x59F8A1A1), AESx(0x09808989), AESx(0x1A170D0D),
AESx(0x65DABFBF), AESx(0xD731E6E6), AESx(0x84C64242), AESx(0xD0B86868),
AESx(0x82C34141), AESx(0x29B09999), AESx(0x5A772D2D), AESx(0x1E110F0F),
AESx(0x7BCBB0B0), AESx(0xA8FC5454), AESx(0x6DD6BBBB), AESx(0x2C3A1616)
};
#ifdef __cplusplus
}
#endif

1120
src/crypto/blake.cpp Normal file
View file

File diff suppressed because it is too large Load diff

965
src/crypto/bmw.cpp Normal file
View file

@ -0,0 +1,965 @@
/* $Id: bmw.c 227 2010-06-16 17:28:38Z tp $ */
/*
* BMW implementation.
*
* ==========================(LICENSE BEGIN)============================
*
* Copyright (c) 2007-2010 Projet RNRT SAPHIR
*
* Permission is hereby granted, free of charge, to any person obtaining
* a copy of this software and associated documentation files (the
* "Software"), to deal in the Software without restriction, including
* without limitation the rights to use, copy, modify, merge, publish,
* distribute, sublicense, and/or sell copies of the Software, and to
* permit persons to whom the Software is furnished to do so, subject to
* the following conditions:
*
* The above copyright notice and this permission notice shall be
* included in all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*
* ===========================(LICENSE END)=============================
*
* @author Thomas Pornin <thomas.pornin@cryptolog.com>
*/
#include <stddef.h>
#include <string.h>
#include <limits.h>
#ifdef __cplusplus
extern "C"{
#endif
#include "crypto/sph_bmw.h"
#if SPH_SMALL_FOOTPRINT && !defined SPH_SMALL_FOOTPRINT_BMW
#define SPH_SMALL_FOOTPRINT_BMW 1
#endif
#ifdef _MSC_VER
#pragma warning (disable: 4146)
#endif
static const sph_u32 IV224[] = {
SPH_C32(0x00010203), SPH_C32(0x04050607),
SPH_C32(0x08090A0B), SPH_C32(0x0C0D0E0F),
SPH_C32(0x10111213), SPH_C32(0x14151617),
SPH_C32(0x18191A1B), SPH_C32(0x1C1D1E1F),
SPH_C32(0x20212223), SPH_C32(0x24252627),
SPH_C32(0x28292A2B), SPH_C32(0x2C2D2E2F),
SPH_C32(0x30313233), SPH_C32(0x34353637),
SPH_C32(0x38393A3B), SPH_C32(0x3C3D3E3F)
};
static const sph_u32 IV256[] = {
SPH_C32(0x40414243), SPH_C32(0x44454647),
SPH_C32(0x48494A4B), SPH_C32(0x4C4D4E4F),
SPH_C32(0x50515253), SPH_C32(0x54555657),
SPH_C32(0x58595A5B), SPH_C32(0x5C5D5E5F),
SPH_C32(0x60616263), SPH_C32(0x64656667),
SPH_C32(0x68696A6B), SPH_C32(0x6C6D6E6F),
SPH_C32(0x70717273), SPH_C32(0x74757677),
SPH_C32(0x78797A7B), SPH_C32(0x7C7D7E7F)
};
#if SPH_64
static const sph_u64 IV384[] = {
SPH_C64(0x0001020304050607), SPH_C64(0x08090A0B0C0D0E0F),
SPH_C64(0x1011121314151617), SPH_C64(0x18191A1B1C1D1E1F),
SPH_C64(0x2021222324252627), SPH_C64(0x28292A2B2C2D2E2F),
SPH_C64(0x3031323334353637), SPH_C64(0x38393A3B3C3D3E3F),
SPH_C64(0x4041424344454647), SPH_C64(0x48494A4B4C4D4E4F),
SPH_C64(0x5051525354555657), SPH_C64(0x58595A5B5C5D5E5F),
SPH_C64(0x6061626364656667), SPH_C64(0x68696A6B6C6D6E6F),
SPH_C64(0x7071727374757677), SPH_C64(0x78797A7B7C7D7E7F)
};
static const sph_u64 IV512[] = {
SPH_C64(0x8081828384858687), SPH_C64(0x88898A8B8C8D8E8F),
SPH_C64(0x9091929394959697), SPH_C64(0x98999A9B9C9D9E9F),
SPH_C64(0xA0A1A2A3A4A5A6A7), SPH_C64(0xA8A9AAABACADAEAF),
SPH_C64(0xB0B1B2B3B4B5B6B7), SPH_C64(0xB8B9BABBBCBDBEBF),
SPH_C64(0xC0C1C2C3C4C5C6C7), SPH_C64(0xC8C9CACBCCCDCECF),
SPH_C64(0xD0D1D2D3D4D5D6D7), SPH_C64(0xD8D9DADBDCDDDEDF),
SPH_C64(0xE0E1E2E3E4E5E6E7), SPH_C64(0xE8E9EAEBECEDEEEF),
SPH_C64(0xF0F1F2F3F4F5F6F7), SPH_C64(0xF8F9FAFBFCFDFEFF)
};
#endif
#define XCAT(x, y) XCAT_(x, y)
#define XCAT_(x, y) x ## y
#define LPAR (
#define I16_16 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
#define I16_17 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16
#define I16_18 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17
#define I16_19 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18
#define I16_20 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19
#define I16_21 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20
#define I16_22 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21
#define I16_23 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22
#define I16_24 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23
#define I16_25 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24
#define I16_26 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25
#define I16_27 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26
#define I16_28 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27
#define I16_29 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28
#define I16_30 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29
#define I16_31 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30
#define M16_16 0, 1, 3, 4, 7, 10, 11
#define M16_17 1, 2, 4, 5, 8, 11, 12
#define M16_18 2, 3, 5, 6, 9, 12, 13
#define M16_19 3, 4, 6, 7, 10, 13, 14
#define M16_20 4, 5, 7, 8, 11, 14, 15
#define M16_21 5, 6, 8, 9, 12, 15, 16
#define M16_22 6, 7, 9, 10, 13, 0, 1
#define M16_23 7, 8, 10, 11, 14, 1, 2
#define M16_24 8, 9, 11, 12, 15, 2, 3
#define M16_25 9, 10, 12, 13, 0, 3, 4
#define M16_26 10, 11, 13, 14, 1, 4, 5
#define M16_27 11, 12, 14, 15, 2, 5, 6
#define M16_28 12, 13, 15, 16, 3, 6, 7
#define M16_29 13, 14, 0, 1, 4, 7, 8
#define M16_30 14, 15, 1, 2, 5, 8, 9
#define M16_31 15, 16, 2, 3, 6, 9, 10
#define ss0(x) (((x) >> 1) ^ SPH_T32((x) << 3) \
^ SPH_ROTL32(x, 4) ^ SPH_ROTL32(x, 19))
#define ss1(x) (((x) >> 1) ^ SPH_T32((x) << 2) \
^ SPH_ROTL32(x, 8) ^ SPH_ROTL32(x, 23))
#define ss2(x) (((x) >> 2) ^ SPH_T32((x) << 1) \
^ SPH_ROTL32(x, 12) ^ SPH_ROTL32(x, 25))
#define ss3(x) (((x) >> 2) ^ SPH_T32((x) << 2) \
^ SPH_ROTL32(x, 15) ^ SPH_ROTL32(x, 29))
#define ss4(x) (((x) >> 1) ^ (x))
#define ss5(x) (((x) >> 2) ^ (x))
#define rs1(x) SPH_ROTL32(x, 3)
#define rs2(x) SPH_ROTL32(x, 7)
#define rs3(x) SPH_ROTL32(x, 13)
#define rs4(x) SPH_ROTL32(x, 16)
#define rs5(x) SPH_ROTL32(x, 19)
#define rs6(x) SPH_ROTL32(x, 23)
#define rs7(x) SPH_ROTL32(x, 27)
#define Ks(j) SPH_T32((sph_u32)(j) * SPH_C32(0x05555555))
#define add_elt_s(mf, hf, j0m, j1m, j3m, j4m, j7m, j10m, j11m, j16) \
(SPH_T32(SPH_ROTL32(mf(j0m), j1m) + SPH_ROTL32(mf(j3m), j4m) \
- SPH_ROTL32(mf(j10m), j11m) + Ks(j16)) ^ hf(j7m))
#define expand1s_inner(qf, mf, hf, i16, \
i0, i1, i2, i3, i4, i5, i6, i7, i8, \
i9, i10, i11, i12, i13, i14, i15, \
i0m, i1m, i3m, i4m, i7m, i10m, i11m) \
SPH_T32(ss1(qf(i0)) + ss2(qf(i1)) + ss3(qf(i2)) + ss0(qf(i3)) \
+ ss1(qf(i4)) + ss2(qf(i5)) + ss3(qf(i6)) + ss0(qf(i7)) \
+ ss1(qf(i8)) + ss2(qf(i9)) + ss3(qf(i10)) + ss0(qf(i11)) \
+ ss1(qf(i12)) + ss2(qf(i13)) + ss3(qf(i14)) + ss0(qf(i15)) \
+ add_elt_s(mf, hf, i0m, i1m, i3m, i4m, i7m, i10m, i11m, i16))
#define expand1s(qf, mf, hf, i16) \
expand1s_(qf, mf, hf, i16, I16_ ## i16, M16_ ## i16)
#define expand1s_(qf, mf, hf, i16, ix, iy) \
expand1s_inner LPAR qf, mf, hf, i16, ix, iy)
#define expand2s_inner(qf, mf, hf, i16, \
i0, i1, i2, i3, i4, i5, i6, i7, i8, \
i9, i10, i11, i12, i13, i14, i15, \
i0m, i1m, i3m, i4m, i7m, i10m, i11m) \
SPH_T32(qf(i0) + rs1(qf(i1)) + qf(i2) + rs2(qf(i3)) \
+ qf(i4) + rs3(qf(i5)) + qf(i6) + rs4(qf(i7)) \
+ qf(i8) + rs5(qf(i9)) + qf(i10) + rs6(qf(i11)) \
+ qf(i12) + rs7(qf(i13)) + ss4(qf(i14)) + ss5(qf(i15)) \
+ add_elt_s(mf, hf, i0m, i1m, i3m, i4m, i7m, i10m, i11m, i16))
#define expand2s(qf, mf, hf, i16) \
expand2s_(qf, mf, hf, i16, I16_ ## i16, M16_ ## i16)
#define expand2s_(qf, mf, hf, i16, ix, iy) \
expand2s_inner LPAR qf, mf, hf, i16, ix, iy)
#if SPH_64
#define sb0(x) (((x) >> 1) ^ SPH_T64((x) << 3) \
^ SPH_ROTL64(x, 4) ^ SPH_ROTL64(x, 37))
#define sb1(x) (((x) >> 1) ^ SPH_T64((x) << 2) \
^ SPH_ROTL64(x, 13) ^ SPH_ROTL64(x, 43))
#define sb2(x) (((x) >> 2) ^ SPH_T64((x) << 1) \
^ SPH_ROTL64(x, 19) ^ SPH_ROTL64(x, 53))
#define sb3(x) (((x) >> 2) ^ SPH_T64((x) << 2) \
^ SPH_ROTL64(x, 28) ^ SPH_ROTL64(x, 59))
#define sb4(x) (((x) >> 1) ^ (x))
#define sb5(x) (((x) >> 2) ^ (x))
#define rb1(x) SPH_ROTL64(x, 5)
#define rb2(x) SPH_ROTL64(x, 11)
#define rb3(x) SPH_ROTL64(x, 27)
#define rb4(x) SPH_ROTL64(x, 32)
#define rb5(x) SPH_ROTL64(x, 37)
#define rb6(x) SPH_ROTL64(x, 43)
#define rb7(x) SPH_ROTL64(x, 53)
#define Kb(j) SPH_T64((sph_u64)(j) * SPH_C64(0x0555555555555555))
#if SPH_SMALL_FOOTPRINT_BMW
static const sph_u64 Kb_tab[] = {
Kb(16), Kb(17), Kb(18), Kb(19), Kb(20), Kb(21), Kb(22), Kb(23),
Kb(24), Kb(25), Kb(26), Kb(27), Kb(28), Kb(29), Kb(30), Kb(31)
};
#define rol_off(mf, j, off) \
SPH_ROTL64(mf(((j) + (off)) & 15), (((j) + (off)) & 15) + 1)
#define add_elt_b(mf, hf, j) \
(SPH_T64(rol_off(mf, j, 0) + rol_off(mf, j, 3) \
- rol_off(mf, j, 10) + Kb_tab[j]) ^ hf(((j) + 7) & 15))
#define expand1b(qf, mf, hf, i) \
SPH_T64(sb1(qf((i) - 16)) + sb2(qf((i) - 15)) \
+ sb3(qf((i) - 14)) + sb0(qf((i) - 13)) \
+ sb1(qf((i) - 12)) + sb2(qf((i) - 11)) \
+ sb3(qf((i) - 10)) + sb0(qf((i) - 9)) \
+ sb1(qf((i) - 8)) + sb2(qf((i) - 7)) \
+ sb3(qf((i) - 6)) + sb0(qf((i) - 5)) \
+ sb1(qf((i) - 4)) + sb2(qf((i) - 3)) \
+ sb3(qf((i) - 2)) + sb0(qf((i) - 1)) \
+ add_elt_b(mf, hf, (i) - 16))
#define expand2b(qf, mf, hf, i) \
SPH_T64(qf((i) - 16) + rb1(qf((i) - 15)) \
+ qf((i) - 14) + rb2(qf((i) - 13)) \
+ qf((i) - 12) + rb3(qf((i) - 11)) \
+ qf((i) - 10) + rb4(qf((i) - 9)) \
+ qf((i) - 8) + rb5(qf((i) - 7)) \
+ qf((i) - 6) + rb6(qf((i) - 5)) \
+ qf((i) - 4) + rb7(qf((i) - 3)) \
+ sb4(qf((i) - 2)) + sb5(qf((i) - 1)) \
+ add_elt_b(mf, hf, (i) - 16))
#else
#define add_elt_b(mf, hf, j0m, j1m, j3m, j4m, j7m, j10m, j11m, j16) \
(SPH_T64(SPH_ROTL64(mf(j0m), j1m) + SPH_ROTL64(mf(j3m), j4m) \
- SPH_ROTL64(mf(j10m), j11m) + Kb(j16)) ^ hf(j7m))
#define expand1b_inner(qf, mf, hf, i16, \
i0, i1, i2, i3, i4, i5, i6, i7, i8, \
i9, i10, i11, i12, i13, i14, i15, \
i0m, i1m, i3m, i4m, i7m, i10m, i11m) \
SPH_T64(sb1(qf(i0)) + sb2(qf(i1)) + sb3(qf(i2)) + sb0(qf(i3)) \
+ sb1(qf(i4)) + sb2(qf(i5)) + sb3(qf(i6)) + sb0(qf(i7)) \
+ sb1(qf(i8)) + sb2(qf(i9)) + sb3(qf(i10)) + sb0(qf(i11)) \
+ sb1(qf(i12)) + sb2(qf(i13)) + sb3(qf(i14)) + sb0(qf(i15)) \
+ add_elt_b(mf, hf, i0m, i1m, i3m, i4m, i7m, i10m, i11m, i16))
#define expand1b(qf, mf, hf, i16) \
expand1b_(qf, mf, hf, i16, I16_ ## i16, M16_ ## i16)
#define expand1b_(qf, mf, hf, i16, ix, iy) \
expand1b_inner LPAR qf, mf, hf, i16, ix, iy)
#define expand2b_inner(qf, mf, hf, i16, \
i0, i1, i2, i3, i4, i5, i6, i7, i8, \
i9, i10, i11, i12, i13, i14, i15, \
i0m, i1m, i3m, i4m, i7m, i10m, i11m) \
SPH_T64(qf(i0) + rb1(qf(i1)) + qf(i2) + rb2(qf(i3)) \
+ qf(i4) + rb3(qf(i5)) + qf(i6) + rb4(qf(i7)) \
+ qf(i8) + rb5(qf(i9)) + qf(i10) + rb6(qf(i11)) \
+ qf(i12) + rb7(qf(i13)) + sb4(qf(i14)) + sb5(qf(i15)) \
+ add_elt_b(mf, hf, i0m, i1m, i3m, i4m, i7m, i10m, i11m, i16))
#define expand2b(qf, mf, hf, i16) \
expand2b_(qf, mf, hf, i16, I16_ ## i16, M16_ ## i16)
#define expand2b_(qf, mf, hf, i16, ix, iy) \
expand2b_inner LPAR qf, mf, hf, i16, ix, iy)
#endif
#endif
#define MAKE_W(tt, i0, op01, i1, op12, i2, op23, i3, op34, i4) \
tt((M(i0) ^ H(i0)) op01 (M(i1) ^ H(i1)) op12 (M(i2) ^ H(i2)) \
op23 (M(i3) ^ H(i3)) op34 (M(i4) ^ H(i4)))
#define Ws0 MAKE_W(SPH_T32, 5, -, 7, +, 10, +, 13, +, 14)
#define Ws1 MAKE_W(SPH_T32, 6, -, 8, +, 11, +, 14, -, 15)
#define Ws2 MAKE_W(SPH_T32, 0, +, 7, +, 9, -, 12, +, 15)
#define Ws3 MAKE_W(SPH_T32, 0, -, 1, +, 8, -, 10, +, 13)
#define Ws4 MAKE_W(SPH_T32, 1, +, 2, +, 9, -, 11, -, 14)
#define Ws5 MAKE_W(SPH_T32, 3, -, 2, +, 10, -, 12, +, 15)
#define Ws6 MAKE_W(SPH_T32, 4, -, 0, -, 3, -, 11, +, 13)
#define Ws7 MAKE_W(SPH_T32, 1, -, 4, -, 5, -, 12, -, 14)
#define Ws8 MAKE_W(SPH_T32, 2, -, 5, -, 6, +, 13, -, 15)
#define Ws9 MAKE_W(SPH_T32, 0, -, 3, +, 6, -, 7, +, 14)
#define Ws10 MAKE_W(SPH_T32, 8, -, 1, -, 4, -, 7, +, 15)
#define Ws11 MAKE_W(SPH_T32, 8, -, 0, -, 2, -, 5, +, 9)
#define Ws12 MAKE_W(SPH_T32, 1, +, 3, -, 6, -, 9, +, 10)
#define Ws13 MAKE_W(SPH_T32, 2, +, 4, +, 7, +, 10, +, 11)
#define Ws14 MAKE_W(SPH_T32, 3, -, 5, +, 8, -, 11, -, 12)
#define Ws15 MAKE_W(SPH_T32, 12, -, 4, -, 6, -, 9, +, 13)
#if SPH_SMALL_FOOTPRINT_BMW
#define MAKE_Qas do { \
unsigned u; \
sph_u32 Ws[16]; \
Ws[ 0] = Ws0; \
Ws[ 1] = Ws1; \
Ws[ 2] = Ws2; \
Ws[ 3] = Ws3; \
Ws[ 4] = Ws4; \
Ws[ 5] = Ws5; \
Ws[ 6] = Ws6; \
Ws[ 7] = Ws7; \
Ws[ 8] = Ws8; \
Ws[ 9] = Ws9; \
Ws[10] = Ws10; \
Ws[11] = Ws11; \
Ws[12] = Ws12; \
Ws[13] = Ws13; \
Ws[14] = Ws14; \
Ws[15] = Ws15; \
for (u = 0; u < 15; u += 5) { \
qt[u + 0] = SPH_T32(ss0(Ws[u + 0]) + H(u + 1)); \
qt[u + 1] = SPH_T32(ss1(Ws[u + 1]) + H(u + 2)); \
qt[u + 2] = SPH_T32(ss2(Ws[u + 2]) + H(u + 3)); \
qt[u + 3] = SPH_T32(ss3(Ws[u + 3]) + H(u + 4)); \
qt[u + 4] = SPH_T32(ss4(Ws[u + 4]) + H(u + 5)); \
} \
qt[15] = SPH_T32(ss0(Ws[15]) + H(0)); \
} while (0)
#define MAKE_Qbs do { \
qt[16] = expand1s(Qs, M, H, 16); \
qt[17] = expand1s(Qs, M, H, 17); \
qt[18] = expand2s(Qs, M, H, 18); \
qt[19] = expand2s(Qs, M, H, 19); \
qt[20] = expand2s(Qs, M, H, 20); \
qt[21] = expand2s(Qs, M, H, 21); \
qt[22] = expand2s(Qs, M, H, 22); \
qt[23] = expand2s(Qs, M, H, 23); \
qt[24] = expand2s(Qs, M, H, 24); \
qt[25] = expand2s(Qs, M, H, 25); \
qt[26] = expand2s(Qs, M, H, 26); \
qt[27] = expand2s(Qs, M, H, 27); \
qt[28] = expand2s(Qs, M, H, 28); \
qt[29] = expand2s(Qs, M, H, 29); \
qt[30] = expand2s(Qs, M, H, 30); \
qt[31] = expand2s(Qs, M, H, 31); \
} while (0)
#else
#define MAKE_Qas do { \
qt[ 0] = SPH_T32(ss0(Ws0 ) + H( 1)); \
qt[ 1] = SPH_T32(ss1(Ws1 ) + H( 2)); \
qt[ 2] = SPH_T32(ss2(Ws2 ) + H( 3)); \
qt[ 3] = SPH_T32(ss3(Ws3 ) + H( 4)); \
qt[ 4] = SPH_T32(ss4(Ws4 ) + H( 5)); \
qt[ 5] = SPH_T32(ss0(Ws5 ) + H( 6)); \
qt[ 6] = SPH_T32(ss1(Ws6 ) + H( 7)); \
qt[ 7] = SPH_T32(ss2(Ws7 ) + H( 8)); \
qt[ 8] = SPH_T32(ss3(Ws8 ) + H( 9)); \
qt[ 9] = SPH_T32(ss4(Ws9 ) + H(10)); \
qt[10] = SPH_T32(ss0(Ws10) + H(11)); \
qt[11] = SPH_T32(ss1(Ws11) + H(12)); \
qt[12] = SPH_T32(ss2(Ws12) + H(13)); \
qt[13] = SPH_T32(ss3(Ws13) + H(14)); \
qt[14] = SPH_T32(ss4(Ws14) + H(15)); \
qt[15] = SPH_T32(ss0(Ws15) + H( 0)); \
} while (0)
#define MAKE_Qbs do { \
qt[16] = expand1s(Qs, M, H, 16); \
qt[17] = expand1s(Qs, M, H, 17); \
qt[18] = expand2s(Qs, M, H, 18); \
qt[19] = expand2s(Qs, M, H, 19); \
qt[20] = expand2s(Qs, M, H, 20); \
qt[21] = expand2s(Qs, M, H, 21); \
qt[22] = expand2s(Qs, M, H, 22); \
qt[23] = expand2s(Qs, M, H, 23); \
qt[24] = expand2s(Qs, M, H, 24); \
qt[25] = expand2s(Qs, M, H, 25); \
qt[26] = expand2s(Qs, M, H, 26); \
qt[27] = expand2s(Qs, M, H, 27); \
qt[28] = expand2s(Qs, M, H, 28); \
qt[29] = expand2s(Qs, M, H, 29); \
qt[30] = expand2s(Qs, M, H, 30); \
qt[31] = expand2s(Qs, M, H, 31); \
} while (0)
#endif
#define MAKE_Qs do { \
MAKE_Qas; \
MAKE_Qbs; \
} while (0)
#define Qs(j) (qt[j])
#if SPH_64
#define Wb0 MAKE_W(SPH_T64, 5, -, 7, +, 10, +, 13, +, 14)
#define Wb1 MAKE_W(SPH_T64, 6, -, 8, +, 11, +, 14, -, 15)
#define Wb2 MAKE_W(SPH_T64, 0, +, 7, +, 9, -, 12, +, 15)
#define Wb3 MAKE_W(SPH_T64, 0, -, 1, +, 8, -, 10, +, 13)
#define Wb4 MAKE_W(SPH_T64, 1, +, 2, +, 9, -, 11, -, 14)
#define Wb5 MAKE_W(SPH_T64, 3, -, 2, +, 10, -, 12, +, 15)
#define Wb6 MAKE_W(SPH_T64, 4, -, 0, -, 3, -, 11, +, 13)
#define Wb7 MAKE_W(SPH_T64, 1, -, 4, -, 5, -, 12, -, 14)
#define Wb8 MAKE_W(SPH_T64, 2, -, 5, -, 6, +, 13, -, 15)
#define Wb9 MAKE_W(SPH_T64, 0, -, 3, +, 6, -, 7, +, 14)
#define Wb10 MAKE_W(SPH_T64, 8, -, 1, -, 4, -, 7, +, 15)
#define Wb11 MAKE_W(SPH_T64, 8, -, 0, -, 2, -, 5, +, 9)
#define Wb12 MAKE_W(SPH_T64, 1, +, 3, -, 6, -, 9, +, 10)
#define Wb13 MAKE_W(SPH_T64, 2, +, 4, +, 7, +, 10, +, 11)
#define Wb14 MAKE_W(SPH_T64, 3, -, 5, +, 8, -, 11, -, 12)
#define Wb15 MAKE_W(SPH_T64, 12, -, 4, -, 6, -, 9, +, 13)
#if SPH_SMALL_FOOTPRINT_BMW
#define MAKE_Qab do { \
unsigned u; \
sph_u64 Wb[16]; \
Wb[ 0] = Wb0; \
Wb[ 1] = Wb1; \
Wb[ 2] = Wb2; \
Wb[ 3] = Wb3; \
Wb[ 4] = Wb4; \
Wb[ 5] = Wb5; \
Wb[ 6] = Wb6; \
Wb[ 7] = Wb7; \
Wb[ 8] = Wb8; \
Wb[ 9] = Wb9; \
Wb[10] = Wb10; \
Wb[11] = Wb11; \
Wb[12] = Wb12; \
Wb[13] = Wb13; \
Wb[14] = Wb14; \
Wb[15] = Wb15; \
for (u = 0; u < 15; u += 5) { \
qt[u + 0] = SPH_T64(sb0(Wb[u + 0]) + H(u + 1)); \
qt[u + 1] = SPH_T64(sb1(Wb[u + 1]) + H(u + 2)); \
qt[u + 2] = SPH_T64(sb2(Wb[u + 2]) + H(u + 3)); \
qt[u + 3] = SPH_T64(sb3(Wb[u + 3]) + H(u + 4)); \
qt[u + 4] = SPH_T64(sb4(Wb[u + 4]) + H(u + 5)); \
} \
qt[15] = SPH_T64(sb0(Wb[15]) + H(0)); \
} while (0)
#define MAKE_Qbb do { \
unsigned u; \
for (u = 16; u < 18; u ++) \
qt[u] = expand1b(Qb, M, H, u); \
for (u = 18; u < 32; u ++) \
qt[u] = expand2b(Qb, M, H, u); \
} while (0)
#else
#define MAKE_Qab do { \
qt[ 0] = SPH_T64(sb0(Wb0 ) + H( 1)); \
qt[ 1] = SPH_T64(sb1(Wb1 ) + H( 2)); \
qt[ 2] = SPH_T64(sb2(Wb2 ) + H( 3)); \
qt[ 3] = SPH_T64(sb3(Wb3 ) + H( 4)); \
qt[ 4] = SPH_T64(sb4(Wb4 ) + H( 5)); \
qt[ 5] = SPH_T64(sb0(Wb5 ) + H( 6)); \
qt[ 6] = SPH_T64(sb1(Wb6 ) + H( 7)); \
qt[ 7] = SPH_T64(sb2(Wb7 ) + H( 8)); \
qt[ 8] = SPH_T64(sb3(Wb8 ) + H( 9)); \
qt[ 9] = SPH_T64(sb4(Wb9 ) + H(10)); \
qt[10] = SPH_T64(sb0(Wb10) + H(11)); \
qt[11] = SPH_T64(sb1(Wb11) + H(12)); \
qt[12] = SPH_T64(sb2(Wb12) + H(13)); \
qt[13] = SPH_T64(sb3(Wb13) + H(14)); \
qt[14] = SPH_T64(sb4(Wb14) + H(15)); \
qt[15] = SPH_T64(sb0(Wb15) + H( 0)); \
} while (0)
#define MAKE_Qbb do { \
qt[16] = expand1b(Qb, M, H, 16); \
qt[17] = expand1b(Qb, M, H, 17); \
qt[18] = expand2b(Qb, M, H, 18); \
qt[19] = expand2b(Qb, M, H, 19); \
qt[20] = expand2b(Qb, M, H, 20); \
qt[21] = expand2b(Qb, M, H, 21); \
qt[22] = expand2b(Qb, M, H, 22); \
qt[23] = expand2b(Qb, M, H, 23); \
qt[24] = expand2b(Qb, M, H, 24); \
qt[25] = expand2b(Qb, M, H, 25); \
qt[26] = expand2b(Qb, M, H, 26); \
qt[27] = expand2b(Qb, M, H, 27); \
qt[28] = expand2b(Qb, M, H, 28); \
qt[29] = expand2b(Qb, M, H, 29); \
qt[30] = expand2b(Qb, M, H, 30); \
qt[31] = expand2b(Qb, M, H, 31); \
} while (0)
#endif
#define MAKE_Qb do { \
MAKE_Qab; \
MAKE_Qbb; \
} while (0)
#define Qb(j) (qt[j])
#endif
#define FOLD(type, mkQ, tt, rol, mf, qf, dhf) do { \
type qt[32], xl, xh; \
mkQ; \
xl = qf(16) ^ qf(17) ^ qf(18) ^ qf(19) \
^ qf(20) ^ qf(21) ^ qf(22) ^ qf(23); \
xh = xl ^ qf(24) ^ qf(25) ^ qf(26) ^ qf(27) \
^ qf(28) ^ qf(29) ^ qf(30) ^ qf(31); \
dhf( 0) = tt(((xh << 5) ^ (qf(16) >> 5) ^ mf( 0)) \
+ (xl ^ qf(24) ^ qf( 0))); \
dhf( 1) = tt(((xh >> 7) ^ (qf(17) << 8) ^ mf( 1)) \
+ (xl ^ qf(25) ^ qf( 1))); \
dhf( 2) = tt(((xh >> 5) ^ (qf(18) << 5) ^ mf( 2)) \
+ (xl ^ qf(26) ^ qf( 2))); \
dhf( 3) = tt(((xh >> 1) ^ (qf(19) << 5) ^ mf( 3)) \
+ (xl ^ qf(27) ^ qf( 3))); \
dhf( 4) = tt(((xh >> 3) ^ (qf(20) << 0) ^ mf( 4)) \
+ (xl ^ qf(28) ^ qf( 4))); \
dhf( 5) = tt(((xh << 6) ^ (qf(21) >> 6) ^ mf( 5)) \
+ (xl ^ qf(29) ^ qf( 5))); \
dhf( 6) = tt(((xh >> 4) ^ (qf(22) << 6) ^ mf( 6)) \
+ (xl ^ qf(30) ^ qf( 6))); \
dhf( 7) = tt(((xh >> 11) ^ (qf(23) << 2) ^ mf( 7)) \
+ (xl ^ qf(31) ^ qf( 7))); \
dhf( 8) = tt(rol(dhf(4), 9) + (xh ^ qf(24) ^ mf( 8)) \
+ ((xl << 8) ^ qf(23) ^ qf( 8))); \
dhf( 9) = tt(rol(dhf(5), 10) + (xh ^ qf(25) ^ mf( 9)) \
+ ((xl >> 6) ^ qf(16) ^ qf( 9))); \
dhf(10) = tt(rol(dhf(6), 11) + (xh ^ qf(26) ^ mf(10)) \
+ ((xl << 6) ^ qf(17) ^ qf(10))); \
dhf(11) = tt(rol(dhf(7), 12) + (xh ^ qf(27) ^ mf(11)) \
+ ((xl << 4) ^ qf(18) ^ qf(11))); \
dhf(12) = tt(rol(dhf(0), 13) + (xh ^ qf(28) ^ mf(12)) \
+ ((xl >> 3) ^ qf(19) ^ qf(12))); \
dhf(13) = tt(rol(dhf(1), 14) + (xh ^ qf(29) ^ mf(13)) \
+ ((xl >> 4) ^ qf(20) ^ qf(13))); \
dhf(14) = tt(rol(dhf(2), 15) + (xh ^ qf(30) ^ mf(14)) \
+ ((xl >> 7) ^ qf(21) ^ qf(14))); \
dhf(15) = tt(rol(dhf(3), 16) + (xh ^ qf(31) ^ mf(15)) \
+ ((xl >> 2) ^ qf(22) ^ qf(15))); \
} while (0)
#define FOLDs FOLD(sph_u32, MAKE_Qs, SPH_T32, SPH_ROTL32, M, Qs, dH)
#if SPH_64
#define FOLDb FOLD(sph_u64, MAKE_Qb, SPH_T64, SPH_ROTL64, M, Qb, dH)
#endif
static void
compress_small(const unsigned char *data, const sph_u32 h[16], sph_u32 dh[16])
{
#if SPH_LITTLE_FAST
#define M(x) sph_dec32le_aligned(data + 4 * (x))
#else
sph_u32 mv[16];
mv[ 0] = sph_dec32le_aligned(data + 0);
mv[ 1] = sph_dec32le_aligned(data + 4);
mv[ 2] = sph_dec32le_aligned(data + 8);
mv[ 3] = sph_dec32le_aligned(data + 12);
mv[ 4] = sph_dec32le_aligned(data + 16);
mv[ 5] = sph_dec32le_aligned(data + 20);
mv[ 6] = sph_dec32le_aligned(data + 24);
mv[ 7] = sph_dec32le_aligned(data + 28);
mv[ 8] = sph_dec32le_aligned(data + 32);
mv[ 9] = sph_dec32le_aligned(data + 36);
mv[10] = sph_dec32le_aligned(data + 40);
mv[11] = sph_dec32le_aligned(data + 44);
mv[12] = sph_dec32le_aligned(data + 48);
mv[13] = sph_dec32le_aligned(data + 52);
mv[14] = sph_dec32le_aligned(data + 56);
mv[15] = sph_dec32le_aligned(data + 60);
#define M(x) (mv[x])
#endif
#define H(x) (h[x])
#define dH(x) (dh[x])
FOLDs;
#undef M
#undef H
#undef dH
}
static const sph_u32 final_s[16] = {
SPH_C32(0xaaaaaaa0), SPH_C32(0xaaaaaaa1), SPH_C32(0xaaaaaaa2),
SPH_C32(0xaaaaaaa3), SPH_C32(0xaaaaaaa4), SPH_C32(0xaaaaaaa5),
SPH_C32(0xaaaaaaa6), SPH_C32(0xaaaaaaa7), SPH_C32(0xaaaaaaa8),
SPH_C32(0xaaaaaaa9), SPH_C32(0xaaaaaaaa), SPH_C32(0xaaaaaaab),
SPH_C32(0xaaaaaaac), SPH_C32(0xaaaaaaad), SPH_C32(0xaaaaaaae),
SPH_C32(0xaaaaaaaf)
};
static void
bmw32_init(sph_bmw_small_context *sc, const sph_u32 *iv)
{
memcpy(sc->H, iv, sizeof sc->H);
sc->ptr = 0;
#if SPH_64
sc->bit_count = 0;
#else
sc->bit_count_high = 0;
sc->bit_count_low = 0;
#endif
}
static void
bmw32(sph_bmw_small_context *sc, const void *data, size_t len)
{
unsigned char *buf;
size_t ptr;
sph_u32 htmp[16];
sph_u32 *h1, *h2;
#if !SPH_64
sph_u32 tmp;
#endif
#if SPH_64
sc->bit_count += (sph_u64)len << 3;
#else
tmp = sc->bit_count_low;
sc->bit_count_low = SPH_T32(tmp + ((sph_u32)len << 3));
if (sc->bit_count_low < tmp)
sc->bit_count_high ++;
sc->bit_count_high += len >> 29;
#endif
buf = sc->buf;
ptr = sc->ptr;
h1 = sc->H;
h2 = htmp;
while (len > 0) {
size_t clen;
clen = (sizeof sc->buf) - ptr;
if (clen > len)
clen = len;
memcpy(buf + ptr, data, clen);
data = (const unsigned char *)data + clen;
len -= clen;
ptr += clen;
if (ptr == sizeof sc->buf) {
sph_u32 *ht;
compress_small(buf, h1, h2);
ht = h1;
h1 = h2;
h2 = ht;
ptr = 0;
}
}
sc->ptr = ptr;
if (h1 != sc->H)
memcpy(sc->H, h1, sizeof sc->H);
}
static void
bmw32_close(sph_bmw_small_context *sc, unsigned ub, unsigned n,
void *dst, size_t out_size_w32)
{
unsigned char *buf, *out;
size_t ptr, u, v;
unsigned z;
sph_u32 h1[16], h2[16], *h;
buf = sc->buf;
ptr = sc->ptr;
z = 0x80 >> n;
buf[ptr ++] = ((ub & -z) | z) & 0xFF;
h = sc->H;
if (ptr > (sizeof sc->buf) - 8) {
memset(buf + ptr, 0, (sizeof sc->buf) - ptr);
compress_small(buf, h, h1);
ptr = 0;
h = h1;
}
memset(buf + ptr, 0, (sizeof sc->buf) - 8 - ptr);
#if SPH_64
sph_enc64le_aligned(buf + (sizeof sc->buf) - 8,
SPH_T64(sc->bit_count + n));
#else
sph_enc32le_aligned(buf + (sizeof sc->buf) - 8,
sc->bit_count_low + n);
sph_enc32le_aligned(buf + (sizeof sc->buf) - 4,
SPH_T32(sc->bit_count_high));
#endif
compress_small(buf, h, h2);
for (u = 0; u < 16; u ++)
sph_enc32le_aligned(buf + 4 * u, h2[u]);
compress_small(buf, final_s, h1);
out = (unsigned char*) dst;
for (u = 0, v = 16 - out_size_w32; u < out_size_w32; u ++, v ++)
sph_enc32le(out + 4 * u, h1[v]);
}
#if SPH_64
static void
compress_big(const unsigned char *data, const sph_u64 h[16], sph_u64 dh[16])
{
#if SPH_LITTLE_FAST
#define M(x) sph_dec64le_aligned(data + 8 * (x))
#else
sph_u64 mv[16];
mv[ 0] = sph_dec64le_aligned(data + 0);
mv[ 1] = sph_dec64le_aligned(data + 8);
mv[ 2] = sph_dec64le_aligned(data + 16);
mv[ 3] = sph_dec64le_aligned(data + 24);
mv[ 4] = sph_dec64le_aligned(data + 32);
mv[ 5] = sph_dec64le_aligned(data + 40);
mv[ 6] = sph_dec64le_aligned(data + 48);
mv[ 7] = sph_dec64le_aligned(data + 56);
mv[ 8] = sph_dec64le_aligned(data + 64);
mv[ 9] = sph_dec64le_aligned(data + 72);
mv[10] = sph_dec64le_aligned(data + 80);
mv[11] = sph_dec64le_aligned(data + 88);
mv[12] = sph_dec64le_aligned(data + 96);
mv[13] = sph_dec64le_aligned(data + 104);
mv[14] = sph_dec64le_aligned(data + 112);
mv[15] = sph_dec64le_aligned(data + 120);
#define M(x) (mv[x])
#endif
#define H(x) (h[x])
#define dH(x) (dh[x])
FOLDb;
#undef M
#undef H
#undef dH
}
static const sph_u64 final_b[16] = {
SPH_C64(0xaaaaaaaaaaaaaaa0), SPH_C64(0xaaaaaaaaaaaaaaa1),
SPH_C64(0xaaaaaaaaaaaaaaa2), SPH_C64(0xaaaaaaaaaaaaaaa3),
SPH_C64(0xaaaaaaaaaaaaaaa4), SPH_C64(0xaaaaaaaaaaaaaaa5),
SPH_C64(0xaaaaaaaaaaaaaaa6), SPH_C64(0xaaaaaaaaaaaaaaa7),
SPH_C64(0xaaaaaaaaaaaaaaa8), SPH_C64(0xaaaaaaaaaaaaaaa9),
SPH_C64(0xaaaaaaaaaaaaaaaa), SPH_C64(0xaaaaaaaaaaaaaaab),
SPH_C64(0xaaaaaaaaaaaaaaac), SPH_C64(0xaaaaaaaaaaaaaaad),
SPH_C64(0xaaaaaaaaaaaaaaae), SPH_C64(0xaaaaaaaaaaaaaaaf)
};
static void
bmw64_init(sph_bmw_big_context *sc, const sph_u64 *iv)
{
memcpy(sc->H, iv, sizeof sc->H);
sc->ptr = 0;
sc->bit_count = 0;
}
static void
bmw64(sph_bmw_big_context *sc, const void *data, size_t len)
{
unsigned char *buf;
size_t ptr;
sph_u64 htmp[16];
sph_u64 *h1, *h2;
sc->bit_count += (sph_u64)len << 3;
buf = sc->buf;
ptr = sc->ptr;
h1 = sc->H;
h2 = htmp;
while (len > 0) {
size_t clen;
clen = (sizeof sc->buf) - ptr;
if (clen > len)
clen = len;
memcpy(buf + ptr, data, clen);
data = (const unsigned char *)data + clen;
len -= clen;
ptr += clen;
if (ptr == sizeof sc->buf) {
sph_u64 *ht;
compress_big(buf, h1, h2);
ht = h1;
h1 = h2;
h2 = ht;
ptr = 0;
}
}
sc->ptr = ptr;
if (h1 != sc->H)
memcpy(sc->H, h1, sizeof sc->H);
}
static void
bmw64_close(sph_bmw_big_context *sc, unsigned ub, unsigned n,
void *dst, size_t out_size_w64)
{
unsigned char *buf, *out;
size_t ptr, u, v;
unsigned z;
sph_u64 h1[16], h2[16], *h;
buf = sc->buf;
ptr = sc->ptr;
z = 0x80 >> n;
buf[ptr ++] = ((ub & -z) | z) & 0xFF;
h = sc->H;
if (ptr > (sizeof sc->buf) - 8) {
memset(buf + ptr, 0, (sizeof sc->buf) - ptr);
compress_big(buf, h, h1);
ptr = 0;
h = h1;
}
memset(buf + ptr, 0, (sizeof sc->buf) - 8 - ptr);
sph_enc64le_aligned(buf + (sizeof sc->buf) - 8,
SPH_T64(sc->bit_count + n));
compress_big(buf, h, h2);
for (u = 0; u < 16; u ++)
sph_enc64le_aligned(buf + 8 * u, h2[u]);
compress_big(buf, final_b, h1);
out = (unsigned char*) dst;
for (u = 0, v = 16 - out_size_w64; u < out_size_w64; u ++, v ++)
sph_enc64le(out + 8 * u, h1[v]);
}
#endif
/* see sph_bmw.h */
void
sph_bmw224_init(void *cc)
{
bmw32_init((sph_bmw_small_context*) cc, IV224);
}
/* see sph_bmw.h */
void
sph_bmw224(void *cc, const void *data, size_t len)
{
bmw32((sph_bmw_small_context*) cc, data, len);
}
/* see sph_bmw.h */
void
sph_bmw224_close(void *cc, void *dst)
{
sph_bmw224_addbits_and_close((sph_bmw_small_context*) cc, 0, 0, dst);
}
/* see sph_bmw.h */
void
sph_bmw224_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst)
{
bmw32_close((sph_bmw_small_context*) cc, ub, n, dst, 7);
sph_bmw224_init(cc);
}
/* see sph_bmw.h */
void
sph_bmw256_init(void *cc)
{
bmw32_init((sph_bmw_small_context*) cc, IV256);
}
/* see sph_bmw.h */
void
sph_bmw256(void *cc, const void *data, size_t len)
{
bmw32((sph_bmw_small_context*) cc, data, len);
}
/* see sph_bmw.h */
void
sph_bmw256_close(void *cc, void *dst)
{
sph_bmw256_addbits_and_close((sph_bmw_small_context*) cc, 0, 0, dst);
}
/* see sph_bmw.h */
void
sph_bmw256_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst)
{
bmw32_close((sph_bmw_small_context*) cc, ub, n, dst, 8);
sph_bmw256_init(cc);
}
#if SPH_64
/* see sph_bmw.h */
void
sph_bmw384_init(void *cc)
{
bmw64_init((sph_bmw_big_context*) cc, IV384);
}
/* see sph_bmw.h */
void
sph_bmw384(void *cc, const void *data, size_t len)
{
bmw64((sph_bmw_big_context*) cc, data, len);
}
/* see sph_bmw.h */
void
sph_bmw384_close(void *cc, void *dst)
{
sph_bmw384_addbits_and_close((sph_bmw_big_context*) cc, 0, 0, dst);
}
/* see sph_bmw.h */
void
sph_bmw384_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst)
{
bmw64_close((sph_bmw_big_context*) cc, ub, n, dst, 6);
sph_bmw384_init(cc);
}
/* see sph_bmw.h */
void
sph_bmw512_init(void *cc)
{
bmw64_init((sph_bmw_big_context*) cc, IV512);
}
/* see sph_bmw.h */
void
sph_bmw512(void *cc, const void *data, size_t len)
{
bmw64((sph_bmw_big_context*) cc, data, len);
}
/* see sph_bmw.h */
void
sph_bmw512_close(void *cc, void *dst)
{
sph_bmw512_addbits_and_close((sph_bmw_big_context*) cc, 0, 0, dst);
}
/* see sph_bmw.h */
void
sph_bmw512_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst)
{
bmw64_close((sph_bmw_big_context*) cc, ub, n, dst, 8);
sph_bmw512_init(cc);
}
#endif
#ifdef __cplusplus
}
#endif

142
src/crypto/brg_endian.h Normal file
View file

@ -0,0 +1,142 @@
/*
---------------------------------------------------------------------------
Copyright (c) 1998-2008, Brian Gladman, Worcester, UK. All rights reserved.
LICENSE TERMS
The redistribution and use of this software (with or without changes)
is allowed without the payment of fees or royalties provided that:
1. source code distributions include the above copyright notice, this
list of conditions and the following disclaimer;
2. binary distributions include the above copyright notice, this list
of conditions and the following disclaimer in their documentation;
3. the name of the copyright holder is not used to endorse products
built using this software without specific written permission.
DISCLAIMER
This software is provided 'as is' with no explicit or implied warranties
in respect of its properties, including, but not limited to, correctness
and/or fitness for purpose.
---------------------------------------------------------------------------
Issue Date: 20/12/2007
Changes for ARM 9/9/2010
*/
#ifndef _BRG_ENDIAN_H
#define _BRG_ENDIAN_H
#define IS_BIG_ENDIAN 4321 /* byte 0 is most significant (mc68k) */
#define IS_LITTLE_ENDIAN 1234 /* byte 0 is least significant (i386) */
#if 0
/* Include files where endian defines and byteswap functions may reside */
#if defined( __sun )
# include <sys/isa_defs.h>
#elif defined( __FreeBSD__ ) || defined( __OpenBSD__ ) || defined( __NetBSD__ )
# include <sys/endian.h>
#elif defined( BSD ) && ( BSD >= 199103 ) || defined( __APPLE__ ) || \
defined( __CYGWIN32__ ) || defined( __DJGPP__ ) || defined( __osf__ )
# include <machine/endian.h>
#elif defined( __linux__ ) || defined( __GNUC__ ) || defined( __GNU_LIBRARY__ )
# if !defined( __MINGW32__ ) && !defined( _AIX )
# include <endian.h>
# if !defined( __BEOS__ )
# include <byteswap.h>
# endif
# endif
#endif
#endif
/* Now attempt to set the define for platform byte order using any */
/* of the four forms SYMBOL, _SYMBOL, __SYMBOL & __SYMBOL__, which */
/* seem to encompass most endian symbol definitions */
#if defined( BIG_ENDIAN ) && defined( LITTLE_ENDIAN )
# if defined( BYTE_ORDER ) && BYTE_ORDER == BIG_ENDIAN
# define PLATFORM_BYTE_ORDER IS_BIG_ENDIAN
# elif defined( BYTE_ORDER ) && BYTE_ORDER == LITTLE_ENDIAN
# define PLATFORM_BYTE_ORDER IS_LITTLE_ENDIAN
# endif
#elif defined( BIG_ENDIAN )
# define PLATFORM_BYTE_ORDER IS_BIG_ENDIAN
#elif defined( LITTLE_ENDIAN )
# define PLATFORM_BYTE_ORDER IS_LITTLE_ENDIAN
#endif
#if defined( _BIG_ENDIAN ) && defined( _LITTLE_ENDIAN )
# if defined( _BYTE_ORDER ) && _BYTE_ORDER == _BIG_ENDIAN
# define PLATFORM_BYTE_ORDER IS_BIG_ENDIAN
# elif defined( _BYTE_ORDER ) && _BYTE_ORDER == _LITTLE_ENDIAN
# define PLATFORM_BYTE_ORDER IS_LITTLE_ENDIAN
# endif
#elif defined( _BIG_ENDIAN )
# define PLATFORM_BYTE_ORDER IS_BIG_ENDIAN
#elif defined( _LITTLE_ENDIAN )
# define PLATFORM_BYTE_ORDER IS_LITTLE_ENDIAN
#endif
#if defined( __BIG_ENDIAN ) && defined( __LITTLE_ENDIAN )
# if defined( __BYTE_ORDER ) && __BYTE_ORDER == __BIG_ENDIAN
# define PLATFORM_BYTE_ORDER IS_BIG_ENDIAN
# elif defined( __BYTE_ORDER ) && __BYTE_ORDER == __LITTLE_ENDIAN
# define PLATFORM_BYTE_ORDER IS_LITTLE_ENDIAN
# endif
#elif defined( __BIG_ENDIAN )
# define PLATFORM_BYTE_ORDER IS_BIG_ENDIAN
#elif defined( __LITTLE_ENDIAN )
# define PLATFORM_BYTE_ORDER IS_LITTLE_ENDIAN
#endif
#if defined( __BIG_ENDIAN__ ) && defined( __LITTLE_ENDIAN__ )
# if defined( __BYTE_ORDER__ ) && __BYTE_ORDER__ == __BIG_ENDIAN__
# define PLATFORM_BYTE_ORDER IS_BIG_ENDIAN
# elif defined( __BYTE_ORDER__ ) && __BYTE_ORDER__ == __LITTLE_ENDIAN__
# define PLATFORM_BYTE_ORDER IS_LITTLE_ENDIAN
# endif
#elif defined( __BIG_ENDIAN__ )
# define PLATFORM_BYTE_ORDER IS_BIG_ENDIAN
#elif defined( __LITTLE_ENDIAN__ )
# define PLATFORM_BYTE_ORDER IS_LITTLE_ENDIAN
#endif
/* if the platform byte order could not be determined, then try to */
/* set this define using common machine defines */
#if !defined(PLATFORM_BYTE_ORDER)
#if defined( __alpha__ ) || defined( __alpha ) || defined( i386 ) || \
defined( __i386__ ) || defined( _M_I86 ) || defined( _M_IX86 ) || \
defined( __OS2__ ) || defined( sun386 ) || defined( __TURBOC__ ) || \
defined( vax ) || defined( vms ) || defined( VMS ) || \
defined( __VMS ) || defined( _M_X64 )
# define PLATFORM_BYTE_ORDER IS_LITTLE_ENDIAN
#elif defined( AMIGA ) || defined( applec ) || defined( __AS400__ ) || \
defined( _CRAY ) || defined( __hppa ) || defined( __hp9000 ) || \
defined( ibm370 ) || defined( mc68000 ) || defined( m68k ) || \
defined( __MRC__ ) || defined( __MVS__ ) || defined( __MWERKS__ ) || \
defined( sparc ) || defined( __sparc) || defined( SYMANTEC_C ) || \
defined( __VOS__ ) || defined( __TIGCC__ ) || defined( __TANDEM ) || \
defined( THINK_C ) || defined( __VMCMS__ ) || defined( _AIX )
# define PLATFORM_BYTE_ORDER IS_BIG_ENDIAN
#elif defined(__arm__)
# ifdef __BIG_ENDIAN
# define PLATFORM_BYTE_ORDER IS_BIG_ENDIAN
# else
# define PLATFORM_BYTE_ORDER IS_LITTLE_ENDIAN
# endif
#elif 1 /* **** EDIT HERE IF NECESSARY **** */
# define PLATFORM_BYTE_ORDER IS_LITTLE_ENDIAN
#elif 0 /* **** EDIT HERE IF NECESSARY **** */
# define PLATFORM_BYTE_ORDER IS_BIG_ENDIAN
#else
# error Please edit lines 132 or 134 in brg_endian.h to set the platform byte order
#endif
#endif
#endif

311
src/crypto/chacha20.cpp Normal file
View file

@ -0,0 +1,311 @@
// Copyright (c) 2017-2020 The Bitcoin Core developers
// Copyright (c) 2017-2020 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
// Based on the public domain implementation 'merged' by D. J. Bernstein
// See https://cr.yp.to/chacha.html.
#include <crypto/common.h>
#include <crypto/chacha20.h>
#include <string.h>
constexpr static inline uint32_t rotl32(uint32_t v, int c) { return (v << c) | (v >> (32 - c)); }
#define QUARTERROUND(a,b,c,d) \
a += b; d = rotl32(d ^ a, 16); \
c += d; b = rotl32(b ^ c, 12); \
a += b; d = rotl32(d ^ a, 8); \
c += d; b = rotl32(b ^ c, 7);
static const unsigned char sigma[] = "expand 32-byte k";
static const unsigned char tau[] = "expand 16-byte k";
void ChaCha20::SetKey(const unsigned char* k, size_t keylen)
{
const unsigned char *constants;
input[4] = ReadLE32(k + 0);
input[5] = ReadLE32(k + 4);
input[6] = ReadLE32(k + 8);
input[7] = ReadLE32(k + 12);
if (keylen == 32) { /* recommended */
k += 16;
constants = sigma;
} else { /* keylen == 16 */
constants = tau;
}
input[8] = ReadLE32(k + 0);
input[9] = ReadLE32(k + 4);
input[10] = ReadLE32(k + 8);
input[11] = ReadLE32(k + 12);
input[0] = ReadLE32(constants + 0);
input[1] = ReadLE32(constants + 4);
input[2] = ReadLE32(constants + 8);
input[3] = ReadLE32(constants + 12);
input[12] = 0;
input[13] = 0;
input[14] = 0;
input[15] = 0;
}
ChaCha20::ChaCha20()
{
memset(input, 0, sizeof(input));
}
ChaCha20::ChaCha20(const unsigned char* k, size_t keylen)
{
SetKey(k, keylen);
}
void ChaCha20::SetIV(uint64_t iv)
{
input[14] = iv;
input[15] = iv >> 32;
}
void ChaCha20::Seek(uint64_t pos)
{
input[12] = pos;
input[13] = pos >> 32;
}
void ChaCha20::Keystream(unsigned char* c, size_t bytes)
{
uint32_t x0, x1, x2, x3, x4, x5, x6, x7, x8, x9, x10, x11, x12, x13, x14, x15;
uint32_t j0, j1, j2, j3, j4, j5, j6, j7, j8, j9, j10, j11, j12, j13, j14, j15;
unsigned char *ctarget = nullptr;
unsigned char tmp[64];
unsigned int i;
if (!bytes) return;
j0 = input[0];
j1 = input[1];
j2 = input[2];
j3 = input[3];
j4 = input[4];
j5 = input[5];
j6 = input[6];
j7 = input[7];
j8 = input[8];
j9 = input[9];
j10 = input[10];
j11 = input[11];
j12 = input[12];
j13 = input[13];
j14 = input[14];
j15 = input[15];
for (;;) {
if (bytes < 64) {
ctarget = c;
c = tmp;
}
x0 = j0;
x1 = j1;
x2 = j2;
x3 = j3;
x4 = j4;
x5 = j5;
x6 = j6;
x7 = j7;
x8 = j8;
x9 = j9;
x10 = j10;
x11 = j11;
x12 = j12;
x13 = j13;
x14 = j14;
x15 = j15;
for (i = 20;i > 0;i -= 2) {
QUARTERROUND( x0, x4, x8,x12)
QUARTERROUND( x1, x5, x9,x13)
QUARTERROUND( x2, x6,x10,x14)
QUARTERROUND( x3, x7,x11,x15)
QUARTERROUND( x0, x5,x10,x15)
QUARTERROUND( x1, x6,x11,x12)
QUARTERROUND( x2, x7, x8,x13)
QUARTERROUND( x3, x4, x9,x14)
}
x0 += j0;
x1 += j1;
x2 += j2;
x3 += j3;
x4 += j4;
x5 += j5;
x6 += j6;
x7 += j7;
x8 += j8;
x9 += j9;
x10 += j10;
x11 += j11;
x12 += j12;
x13 += j13;
x14 += j14;
x15 += j15;
++j12;
if (!j12) ++j13;
WriteLE32(c + 0, x0);
WriteLE32(c + 4, x1);
WriteLE32(c + 8, x2);
WriteLE32(c + 12, x3);
WriteLE32(c + 16, x4);
WriteLE32(c + 20, x5);
WriteLE32(c + 24, x6);
WriteLE32(c + 28, x7);
WriteLE32(c + 32, x8);
WriteLE32(c + 36, x9);
WriteLE32(c + 40, x10);
WriteLE32(c + 44, x11);
WriteLE32(c + 48, x12);
WriteLE32(c + 52, x13);
WriteLE32(c + 56, x14);
WriteLE32(c + 60, x15);
if (bytes <= 64) {
if (bytes < 64) {
for (i = 0;i < bytes;++i) ctarget[i] = c[i];
}
input[12] = j12;
input[13] = j13;
return;
}
bytes -= 64;
c += 64;
}
}
void ChaCha20::Crypt(const unsigned char* m, unsigned char* c, size_t bytes)
{
uint32_t x0, x1, x2, x3, x4, x5, x6, x7, x8, x9, x10, x11, x12, x13, x14, x15;
uint32_t j0, j1, j2, j3, j4, j5, j6, j7, j8, j9, j10, j11, j12, j13, j14, j15;
unsigned char *ctarget = nullptr;
unsigned char tmp[64];
unsigned int i;
if (!bytes) return;
j0 = input[0];
j1 = input[1];
j2 = input[2];
j3 = input[3];
j4 = input[4];
j5 = input[5];
j6 = input[6];
j7 = input[7];
j8 = input[8];
j9 = input[9];
j10 = input[10];
j11 = input[11];
j12 = input[12];
j13 = input[13];
j14 = input[14];
j15 = input[15];
for (;;) {
if (bytes < 64) {
// if m has fewer than 64 bytes available, copy m to tmp and
// read from tmp instead
for (i = 0;i < bytes;++i) tmp[i] = m[i];
m = tmp;
ctarget = c;
c = tmp;
}
x0 = j0;
x1 = j1;
x2 = j2;
x3 = j3;
x4 = j4;
x5 = j5;
x6 = j6;
x7 = j7;
x8 = j8;
x9 = j9;
x10 = j10;
x11 = j11;
x12 = j12;
x13 = j13;
x14 = j14;
x15 = j15;
for (i = 20;i > 0;i -= 2) {
QUARTERROUND( x0, x4, x8,x12)
QUARTERROUND( x1, x5, x9,x13)
QUARTERROUND( x2, x6,x10,x14)
QUARTERROUND( x3, x7,x11,x15)
QUARTERROUND( x0, x5,x10,x15)
QUARTERROUND( x1, x6,x11,x12)
QUARTERROUND( x2, x7, x8,x13)
QUARTERROUND( x3, x4, x9,x14)
}
x0 += j0;
x1 += j1;
x2 += j2;
x3 += j3;
x4 += j4;
x5 += j5;
x6 += j6;
x7 += j7;
x8 += j8;
x9 += j9;
x10 += j10;
x11 += j11;
x12 += j12;
x13 += j13;
x14 += j14;
x15 += j15;
x0 ^= ReadLE32(m + 0);
x1 ^= ReadLE32(m + 4);
x2 ^= ReadLE32(m + 8);
x3 ^= ReadLE32(m + 12);
x4 ^= ReadLE32(m + 16);
x5 ^= ReadLE32(m + 20);
x6 ^= ReadLE32(m + 24);
x7 ^= ReadLE32(m + 28);
x8 ^= ReadLE32(m + 32);
x9 ^= ReadLE32(m + 36);
x10 ^= ReadLE32(m + 40);
x11 ^= ReadLE32(m + 44);
x12 ^= ReadLE32(m + 48);
x13 ^= ReadLE32(m + 52);
x14 ^= ReadLE32(m + 56);
x15 ^= ReadLE32(m + 60);
++j12;
if (!j12) ++j13;
WriteLE32(c + 0, x0);
WriteLE32(c + 4, x1);
WriteLE32(c + 8, x2);
WriteLE32(c + 12, x3);
WriteLE32(c + 16, x4);
WriteLE32(c + 20, x5);
WriteLE32(c + 24, x6);
WriteLE32(c + 28, x7);
WriteLE32(c + 32, x8);
WriteLE32(c + 36, x9);
WriteLE32(c + 40, x10);
WriteLE32(c + 44, x11);
WriteLE32(c + 48, x12);
WriteLE32(c + 52, x13);
WriteLE32(c + 56, x14);
WriteLE32(c + 60, x15);
if (bytes <= 64) {
if (bytes < 64) {
for (i = 0;i < bytes;++i) ctarget[i] = c[i];
}
input[12] = j12;
input[13] = j13;
return;
}
bytes -= 64;
c += 64;
m += 64;
}
}

35
src/crypto/chacha20.h Normal file
View file

@ -0,0 +1,35 @@
// Copyright (c) 2017-2020 The Bitcoin Core developers
// Copyright (c) 2017-2020 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#ifndef DIGIBYTE_CRYPTO_CHACHA20_H
#define DIGIBYTE_CRYPTO_CHACHA20_H
#include <stdint.h>
#include <stdlib.h>
/** A class for ChaCha20 256-bit stream cipher developed by Daniel J. Bernstein
https://cr.yp.to/chacha/chacha-20080128.pdf */
class ChaCha20
{
private:
uint32_t input[16];
public:
ChaCha20();
ChaCha20(const unsigned char* key, size_t keylen);
void SetKey(const unsigned char* key, size_t keylen); //!< set key with flexible keylength; 256bit recommended */
void SetIV(uint64_t iv); // set the 64bit nonce
void Seek(uint64_t pos); // set the 64bit block counter
/** outputs the keystream of size <bytes> into <c> */
void Keystream(unsigned char* c, size_t bytes);
/** enciphers the message <input> of length <bytes> and write the enciphered representation into <output>
* Used for encryption and decryption (XOR)
*/
void Crypt(const unsigned char* input, unsigned char* output, size_t bytes);
};
#endif // DIGIBYTE_CRYPTO_CHACHA20_H

125
src/crypto/chacha_poly_aead.cpp Normal file
View file

@ -0,0 +1,125 @@
// Copyright (c) 2019 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#include <crypto/chacha_poly_aead.h>
#include <crypto/poly1305.h>
#include <support/cleanse.h>
#include <assert.h>
#include <string.h>
#include <cstdio>
#include <limits>
#ifndef HAVE_TIMINGSAFE_BCMP
int timingsafe_bcmp(const unsigned char* b1, const unsigned char* b2, size_t n)
{
const unsigned char *p1 = b1, *p2 = b2;
int ret = 0;
for (; n > 0; n--)
ret |= *p1++ ^ *p2++;
return (ret != 0);
}
#endif // TIMINGSAFE_BCMP
ChaCha20Poly1305AEAD::ChaCha20Poly1305AEAD(const unsigned char* K_1, size_t K_1_len, const unsigned char* K_2, size_t K_2_len)
{
assert(K_1_len == CHACHA20_POLY1305_AEAD_KEY_LEN);
assert(K_2_len == CHACHA20_POLY1305_AEAD_KEY_LEN);
m_chacha_main.SetKey(K_1, CHACHA20_POLY1305_AEAD_KEY_LEN);
m_chacha_header.SetKey(K_2, CHACHA20_POLY1305_AEAD_KEY_LEN);
// set the cached sequence number to uint64 max which hints for an unset cache.
// we can't hit uint64 max since the rekey rule (which resets the sequence number) is 1GB
m_cached_aad_seqnr = std::numeric_limits<uint64_t>::max();
}
bool ChaCha20Poly1305AEAD::Crypt(uint64_t seqnr_payload, uint64_t seqnr_aad, int aad_pos, unsigned char* dest, size_t dest_len /* length of the output buffer for sanity checks */, const unsigned char* src, size_t src_len, bool is_encrypt)
{
// check buffer boundaries
if (
// if we encrypt, make sure the source contains at least the expected AAD and the destination has at least space for the source + MAC
(is_encrypt && (src_len < CHACHA20_POLY1305_AEAD_AAD_LEN || dest_len < src_len + POLY1305_TAGLEN)) ||
// if we decrypt, make sure the source contains at least the expected AAD+MAC and the destination has at least space for the source - MAC
(!is_encrypt && (src_len < CHACHA20_POLY1305_AEAD_AAD_LEN + POLY1305_TAGLEN || dest_len < src_len - POLY1305_TAGLEN))) {
return false;
}
unsigned char expected_tag[POLY1305_TAGLEN], poly_key[POLY1305_KEYLEN];
memset(poly_key, 0, sizeof(poly_key));
m_chacha_main.SetIV(seqnr_payload);
// block counter 0 for the poly1305 key
// use lower 32bytes for the poly1305 key
// (throws away 32 unused bytes (upper 32) from this ChaCha20 round)
m_chacha_main.Seek(0);
m_chacha_main.Crypt(poly_key, poly_key, sizeof(poly_key));
// if decrypting, verify the tag prior to decryption
if (!is_encrypt) {
const unsigned char* tag = src + src_len - POLY1305_TAGLEN;
poly1305_auth(expected_tag, src, src_len - POLY1305_TAGLEN, poly_key);
// constant time compare the calculated MAC with the provided MAC
if (timingsafe_bcmp(expected_tag, tag, POLY1305_TAGLEN) != 0) {
memory_cleanse(expected_tag, sizeof(expected_tag));
memory_cleanse(poly_key, sizeof(poly_key));
return false;
}
memory_cleanse(expected_tag, sizeof(expected_tag));
// MAC has been successfully verified, make sure we don't covert it in decryption
src_len -= POLY1305_TAGLEN;
}
// calculate and cache the next 64byte keystream block if requested sequence number is not yet the cache
if (m_cached_aad_seqnr != seqnr_aad) {
m_cached_aad_seqnr = seqnr_aad;
m_chacha_header.SetIV(seqnr_aad);
m_chacha_header.Seek(0);
m_chacha_header.Keystream(m_aad_keystream_buffer, CHACHA20_ROUND_OUTPUT);
}
// crypt the AAD (3 bytes message length) with given position in AAD cipher instance keystream
dest[0] = src[0] ^ m_aad_keystream_buffer[aad_pos];
dest[1] = src[1] ^ m_aad_keystream_buffer[aad_pos + 1];
dest[2] = src[2] ^ m_aad_keystream_buffer[aad_pos + 2];
// Set the playload ChaCha instance block counter to 1 and crypt the payload
m_chacha_main.Seek(1);
m_chacha_main.Crypt(src + CHACHA20_POLY1305_AEAD_AAD_LEN, dest + CHACHA20_POLY1305_AEAD_AAD_LEN, src_len - CHACHA20_POLY1305_AEAD_AAD_LEN);
// If encrypting, calculate and append tag
if (is_encrypt) {
// the poly1305 tag expands over the AAD (3 bytes length) & encrypted payload
poly1305_auth(dest + src_len, dest, src_len, poly_key);
}
// cleanse no longer required MAC and polykey
memory_cleanse(poly_key, sizeof(poly_key));
return true;
}
bool ChaCha20Poly1305AEAD::GetLength(uint32_t* len24_out, uint64_t seqnr_aad, int aad_pos, const uint8_t* ciphertext)
{
// enforce valid aad position to avoid accessing outside of the 64byte keystream cache
// (there is space for 21 times 3 bytes)
assert(aad_pos >= 0 && aad_pos < CHACHA20_ROUND_OUTPUT - CHACHA20_POLY1305_AEAD_AAD_LEN);
if (m_cached_aad_seqnr != seqnr_aad) {
// we need to calculate the 64 keystream bytes since we reached a new aad sequence number
m_cached_aad_seqnr = seqnr_aad;
m_chacha_header.SetIV(seqnr_aad); // use LE for the nonce
m_chacha_header.Seek(0); // block counter 0
m_chacha_header.Keystream(m_aad_keystream_buffer, CHACHA20_ROUND_OUTPUT); // write keystream to the cache
}
// decrypt the ciphertext length by XORing the right position of the 64byte keystream cache with the ciphertext
*len24_out = (ciphertext[0] ^ m_aad_keystream_buffer[aad_pos + 0]) |
(ciphertext[1] ^ m_aad_keystream_buffer[aad_pos + 1]) << 8 |
(ciphertext[2] ^ m_aad_keystream_buffer[aad_pos + 2]) << 16;
return true;
}

146
src/crypto/chacha_poly_aead.h Normal file
View file

@ -0,0 +1,146 @@
// Copyright (c) 2019 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#ifndef DIGIBYTE_CRYPTO_CHACHA_POLY_AEAD_H
#define DIGIBYTE_CRYPTO_CHACHA_POLY_AEAD_H
#include <crypto/chacha20.h>
#include <cmath>
static constexpr int CHACHA20_POLY1305_AEAD_KEY_LEN = 32;
static constexpr int CHACHA20_POLY1305_AEAD_AAD_LEN = 3; /* 3 bytes length */
static constexpr int CHACHA20_ROUND_OUTPUT = 64; /* 64 bytes per round */
static constexpr int AAD_PACKAGES_PER_ROUND = 21; /* 64 / 3 round down*/
/* A AEAD class for ChaCha20-Poly1305@digibyte.
*
* ChaCha20 is a stream cipher designed by Daniel Bernstein and described in
* <ref>[https://cr.yp.to/chacha/chacha-20080128.pdf ChaCha20]</ref>. It operates
* by permuting 128 fixed bits, 128 or 256 bits of key, a 64 bit nonce and a 64
* bit counter into 64 bytes of output. This output is used as a keystream, with
* any unused bytes simply discarded.
*
* Poly1305 <ref>[https://cr.yp.to/mac/poly1305-20050329.pdf Poly1305]</ref>, also
* by Daniel Bernstein, is a one-time Carter-Wegman MAC that computes a 128 bit
* integrity tag given a message and a single-use 256 bit secret key.
*
* The chacha20-poly1305@digibyte combines these two primitives into an
* authenticated encryption mode. The construction used is based on that proposed
* for TLS by Adam Langley in
* <ref>[http://tools.ietf.org/html/draft-agl-tls-chacha20poly1305-03 "ChaCha20
* and Poly1305 based Cipher Suites for TLS", Adam Langley]</ref>, but differs in
* the layout of data passed to the MAC and in the addition of encryption of the
* packet lengths.
*
* ==== Detailed Construction ====
*
* The chacha20-poly1305@digibyte cipher requires two 256 bits of key material as
* output from the key exchange. Each key (K_1 and K_2) are used by two separate
* instances of chacha20.
*
* The instance keyed by K_1 is a stream cipher that is used only to encrypt the 3
* byte packet length field and has its own sequence number. The second instance,
* keyed by K_2, is used in conjunction with poly1305 to build an AEAD
* (Authenticated Encryption with Associated Data) that is used to encrypt and
* authenticate the entire packet.
*
* Two separate cipher instances are used here so as to keep the packet lengths
* confidential but not create an oracle for the packet payload cipher by
* decrypting and using the packet length prior to checking the MAC. By using an
* independently-keyed cipher instance to encrypt the length, an active attacker
* seeking to exploit the packet input handling as a decryption oracle can learn
* nothing about the payload contents or its MAC (assuming key derivation,
* ChaCha20 and Poly1305 are secure).
*
* The AEAD is constructed as follows: for each packet, generate a Poly1305 key by
* taking the first 256 bits of ChaCha20 stream output generated using K_2, an IV
* consisting of the packet sequence number encoded as an LE uint64 and a ChaCha20
* block counter of zero. The K_2 ChaCha20 block counter is then set to the
* little-endian encoding of 1 (i.e. {1, 0, 0, 0, 0, 0, 0, 0}) and this instance
* is used for encryption of the packet payload.
*
* ==== Packet Handling ====
*
* When receiving a packet, the length must be decrypted first. When 3 bytes of
* ciphertext length have been received, they may be decrypted.
*
* A ChaCha20 round always calculates 64bytes which is sufficient to crypt 21
* times a 3 bytes length field (21*3 = 63). The length field sequence number can
* thus be used 21 times (keystream caching).
*
* The length field must be enc-/decrypted with the ChaCha20 keystream keyed with
* K_1 defined by block counter 0, the length field sequence number in little
* endian and a keystream position from 0 to 60.
*
* Once the entire packet has been received, the MAC MUST be checked before
* decryption. A per-packet Poly1305 key is generated as described above and the
* MAC tag calculated using Poly1305 with this key over the ciphertext of the
* packet length and the payload together. The calculated MAC is then compared in
* constant time with the one appended to the packet and the packet decrypted
* using ChaCha20 as described above (with K_2, the packet sequence number as
* nonce and a starting block counter of 1).
*
* Detection of an invalid MAC MUST lead to immediate connection termination.
*
* To send a packet, first encode the 3 byte length and encrypt it using K_1 as
* described above. Encrypt the packet payload (using K_2) and append it to the
* encrypted length. Finally, calculate a MAC tag and append it.
*
* The initiating peer MUST use <code>K_1_A, K_2_A</code> to encrypt messages on
* the send channel, <code>K_1_B, K_2_B</code> MUST be used to decrypt messages on
* the receive channel.
*
* The responding peer MUST use <code>K_1_A, K_2_A</code> to decrypt messages on
* the receive channel, <code>K_1_B, K_2_B</code> MUST be used to encrypt messages
* on the send channel.
*
* Optimized implementations of ChaCha20-Poly1305@digibyte are relatively fast in
* general, therefore it is very likely that encrypted messages require not more
* CPU cycles per bytes then the current unencrypted p2p message format
* (ChaCha20/Poly1305 versus double SHA256).
*
* The initial packet sequence numbers are 0.
*
* K_2 ChaCha20 cipher instance (payload) must never reuse a {key, nonce} for
* encryption nor may it be used to encrypt more than 2^70 bytes under the same
* {key, nonce}.
*
* K_1 ChaCha20 cipher instance (length field/AAD) must never reuse a {key, nonce,
* position-in-keystream} for encryption nor may it be used to encrypt more than
* 2^70 bytes under the same {key, nonce}.
*
* We use message sequence numbers for both communication directions.
*/
class ChaCha20Poly1305AEAD
{
private:
ChaCha20 m_chacha_main; // payload and poly1305 key-derivation cipher instance
ChaCha20 m_chacha_header; // AAD cipher instance (encrypted length)
unsigned char m_aad_keystream_buffer[CHACHA20_ROUND_OUTPUT]; // aad keystream cache
uint64_t m_cached_aad_seqnr; // aad keystream cache hint
public:
ChaCha20Poly1305AEAD(const unsigned char* K_1, size_t K_1_len, const unsigned char* K_2, size_t K_2_len);
explicit ChaCha20Poly1305AEAD(const ChaCha20Poly1305AEAD&) = delete;
/** Encrypts/decrypts a packet
seqnr_payload, the message sequence number
seqnr_aad, the messages AAD sequence number which allows reuse of the AAD keystream
aad_pos, position to use in the AAD keystream to encrypt the AAD
dest, output buffer, must be of a size equal or larger then CHACHA20_POLY1305_AEAD_AAD_LEN + payload (+ POLY1305_TAG_LEN in encryption) bytes
destlen, length of the destination buffer
src, the AAD+payload to encrypt or the AAD+payload+MAC to decrypt
src_len, the length of the source buffer
is_encrypt, set to true if we encrypt (creates and appends the MAC instead of verifying it)
*/
bool Crypt(uint64_t seqnr_payload, uint64_t seqnr_aad, int aad_pos, unsigned char* dest, size_t dest_len, const unsigned char* src, size_t src_len, bool is_encrypt);
/** decrypts the 3 bytes AAD data and decodes it into a uint32_t field */
bool GetLength(uint32_t* len24_out, uint64_t seqnr_aad, int aad_pos, const uint8_t* ciphertext);
};
#endif // DIGIBYTE_CRYPTO_CHACHA_POLY_AEAD_H

111
src/crypto/common.h Normal file
View file

@ -0,0 +1,111 @@
// Copyright (c) 2009-2020 The Bitcoin Core developers
// Copyright (c) 2014-2020 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#ifndef DIGIBYTE_CRYPTO_COMMON_H
#define DIGIBYTE_CRYPTO_COMMON_H
#if defined(HAVE_CONFIG_H)
#include <config/vexta-config.h>
#endif
#include <stdint.h>
#include <string.h>
#include <compat/endian.h>
uint16_t static inline ReadLE16(const unsigned char* ptr)
{
uint16_t x;
memcpy((char*)&x, ptr, 2);
return le16toh(x);
}
uint32_t static inline ReadLE32(const unsigned char* ptr)
{
uint32_t x;
memcpy((char*)&x, ptr, 4);
return le32toh(x);
}
uint64_t static inline ReadLE64(const unsigned char* ptr)
{
uint64_t x;
memcpy((char*)&x, ptr, 8);
return le64toh(x);
}
void static inline WriteLE16(unsigned char* ptr, uint16_t x)
{
uint16_t v = htole16(x);
memcpy(ptr, (char*)&v, 2);
}
void static inline WriteLE32(unsigned char* ptr, uint32_t x)
{
uint32_t v = htole32(x);
memcpy(ptr, (char*)&v, 4);
}
void static inline WriteLE64(unsigned char* ptr, uint64_t x)
{
uint64_t v = htole64(x);
memcpy(ptr, (char*)&v, 8);
}
uint16_t static inline ReadBE16(const unsigned char* ptr)
{
uint16_t x;
memcpy((char*)&x, ptr, 2);
return be16toh(x);
}
uint32_t static inline ReadBE32(const unsigned char* ptr)
{
uint32_t x;
memcpy((char*)&x, ptr, 4);
return be32toh(x);
}
uint64_t static inline ReadBE64(const unsigned char* ptr)
{
uint64_t x;
memcpy((char*)&x, ptr, 8);
return be64toh(x);
}
void static inline WriteBE32(unsigned char* ptr, uint32_t x)
{
uint32_t v = htobe32(x);
memcpy(ptr, (char*)&v, 4);
}
void static inline WriteBE64(unsigned char* ptr, uint64_t x)
{
uint64_t v = htobe64(x);
memcpy(ptr, (char*)&v, 8);
}
/** Return the smallest number n such that (x >> n) == 0 (or 64 if the highest bit in x is set. */
uint64_t static inline CountBits(uint64_t x)
{
#if HAVE_BUILTIN_CLZL
if (sizeof(unsigned long) >= sizeof(uint64_t)) {
return x ? 8 * sizeof(unsigned long) - __builtin_clzl(x) : 0;
}
#endif
#if HAVE_BUILTIN_CLZLL
if (sizeof(unsigned long long) >= sizeof(uint64_t)) {
return x ? 8 * sizeof(unsigned long long) - __builtin_clzll(x) : 0;
}
#endif
int ret = 0;
while (x) {
x >>= 1;
++ret;
}
return ret;
}
#endif // DIGIBYTE_CRYPTO_COMMON_H

21
src/crypto/ctaes/COPYING Normal file
View file

@ -0,0 +1,21 @@
The MIT License (MIT)
Copyright (c) 2016 Pieter Wuille
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.

41
src/crypto/ctaes/README.md Normal file
View file

@ -0,0 +1,41 @@
ctaes
=====
Simple C module for constant-time AES encryption and decryption.
Features:
* Simple, pure C code without any dependencies.
* No tables or data-dependent branches whatsoever, but using bit sliced approach from https://eprint.iacr.org/2009/129.pdf.
* Very small object code: slightly over 4k of executable code when compiled with -Os.
* Slower than implementations based on precomputed tables or specialized instructions, but can do ~15 MB/s on modern CPUs.
Performance
-----------
Compiled with GCC 5.3.1 with -O3, on an Intel(R) Core(TM) i7-4800MQ CPU, numbers in CPU cycles:
| Algorithm | Key schedule | Encryption per byte | Decryption per byte |
| --------- | ------------:| -------------------:| -------------------:|
| AES-128 | 2.8k | 154 | 161 |
| AES-192 | 3.1k | 169 | 181 |
| AES-256 | 4.0k | 191 | 203 |
Build steps
-----------
Object code:
$ gcc -O3 ctaes.c -c -o ctaes.o
Tests:
$ gcc -O3 ctaes.c test.c -o test
Benchmark:
$ gcc -O3 ctaes.c bench.c -o bench
Review
------
Results of a formal review of the code can be found in http://digibyte.sipa.be/ctaes/review.zip

170
src/crypto/ctaes/bench.c Normal file
View file

@ -0,0 +1,170 @@
#include <stdio.h>
#include <math.h>
#include "sys/time.h"
#include "ctaes.h"
static double gettimedouble(void) {
struct timeval tv;
gettimeofday(&tv, NULL);
return tv.tv_usec * 0.000001 + tv.tv_sec;
}
static void print_number(double x) {
double y = x;
int c = 0;
if (y < 0.0) {
y = -y;
}
while (y < 100.0) {
y *= 10.0;
c++;
}
printf("%.*f", c, x);
}
static void run_benchmark(char *name, void (*benchmark)(void*), void (*setup)(void*), void (*teardown)(void*), void* data, int count, int iter) {
int i;
double min = HUGE_VAL;
double sum = 0.0;
double max = 0.0;
for (i = 0; i < count; i++) {
double begin, total;
if (setup != NULL) {
setup(data);
}
begin = gettimedouble();
benchmark(data);
total = gettimedouble() - begin;
if (teardown != NULL) {
teardown(data);
}
if (total < min) {
min = total;
}
if (total > max) {
max = total;
}
sum += total;
}
printf("%s: min ", name);
print_number(min * 1000000000.0 / iter);
printf("ns / avg ");
print_number((sum / count) * 1000000000.0 / iter);
printf("ns / max ");
print_number(max * 1000000000.0 / iter);
printf("ns\n");
}
static void bench_AES128_init(void* data) {
AES128_ctx* ctx = (AES128_ctx*)data;
int i;
for (i = 0; i < 50000; i++) {
AES128_init(ctx, (unsigned char*)ctx);
}
}
static void bench_AES128_encrypt_setup(void* data) {
AES128_ctx* ctx = (AES128_ctx*)data;
static const unsigned char key[16] = {0};
AES128_init(ctx, key);
}
static void bench_AES128_encrypt(void* data) {
const AES128_ctx* ctx = (const AES128_ctx*)data;
unsigned char scratch[16] = {0};
int i;
for (i = 0; i < 4000000 / 16; i++) {
AES128_encrypt(ctx, 1, scratch, scratch);
}
}
static void bench_AES128_decrypt(void* data) {
const AES128_ctx* ctx = (const AES128_ctx*)data;
unsigned char scratch[16] = {0};
int i;
for (i = 0; i < 4000000 / 16; i++) {
AES128_decrypt(ctx, 1, scratch, scratch);
}
}
static void bench_AES192_init(void* data) {
AES192_ctx* ctx = (AES192_ctx*)data;
int i;
for (i = 0; i < 50000; i++) {
AES192_init(ctx, (unsigned char*)ctx);
}
}
static void bench_AES192_encrypt_setup(void* data) {
AES192_ctx* ctx = (AES192_ctx*)data;
static const unsigned char key[16] = {0};
AES192_init(ctx, key);
}
static void bench_AES192_encrypt(void* data) {
const AES192_ctx* ctx = (const AES192_ctx*)data;
unsigned char scratch[16] = {0};
int i;
for (i = 0; i < 4000000 / 16; i++) {
AES192_encrypt(ctx, 1, scratch, scratch);
}
}
static void bench_AES192_decrypt(void* data) {
const AES192_ctx* ctx = (const AES192_ctx*)data;
unsigned char scratch[16] = {0};
int i;
for (i = 0; i < 4000000 / 16; i++) {
AES192_decrypt(ctx, 1, scratch, scratch);
}
}
static void bench_AES256_init(void* data) {
AES256_ctx* ctx = (AES256_ctx*)data;
int i;
for (i = 0; i < 50000; i++) {
AES256_init(ctx, (unsigned char*)ctx);
}
}
static void bench_AES256_encrypt_setup(void* data) {
AES256_ctx* ctx = (AES256_ctx*)data;
static const unsigned char key[16] = {0};
AES256_init(ctx, key);
}
static void bench_AES256_encrypt(void* data) {
const AES256_ctx* ctx = (const AES256_ctx*)data;
unsigned char scratch[16] = {0};
int i;
for (i = 0; i < 4000000 / 16; i++) {
AES256_encrypt(ctx, 1, scratch, scratch);
}
}
static void bench_AES256_decrypt(void* data) {
const AES256_ctx* ctx = (const AES256_ctx*)data;
unsigned char scratch[16] = {0};
int i;
for (i = 0; i < 4000000 / 16; i++) {
AES256_decrypt(ctx, 1, scratch, scratch);
}
}
int main(void) {
AES128_ctx ctx128;
AES192_ctx ctx192;
AES256_ctx ctx256;
run_benchmark("aes128_init", bench_AES128_init, NULL, NULL, &ctx128, 20, 50000);
run_benchmark("aes128_encrypt_byte", bench_AES128_encrypt, bench_AES128_encrypt_setup, NULL, &ctx128, 20, 4000000);
run_benchmark("aes128_decrypt_byte", bench_AES128_decrypt, bench_AES128_encrypt_setup, NULL, &ctx128, 20, 4000000);
run_benchmark("aes192_init", bench_AES192_init, NULL, NULL, &ctx192, 20, 50000);
run_benchmark("aes192_encrypt_byte", bench_AES192_encrypt, bench_AES192_encrypt_setup, NULL, &ctx192, 20, 4000000);
run_benchmark("aes192_decrypt_byte", bench_AES192_decrypt, bench_AES192_encrypt_setup, NULL, &ctx192, 20, 4000000);
run_benchmark("aes256_init", bench_AES256_init, NULL, NULL, &ctx256, 20, 50000);
run_benchmark("aes256_encrypt_byte", bench_AES256_encrypt, bench_AES256_encrypt_setup, NULL, &ctx256, 20, 4000000);
run_benchmark("aes256_decrypt_byte", bench_AES256_decrypt, bench_AES256_encrypt_setup, NULL, &ctx256, 20, 4000000);
return 0;
}

556
src/crypto/ctaes/ctaes.cpp Normal file
View file

@ -0,0 +1,556 @@
/*********************************************************************
* Copyright (c) 2016 Pieter Wuille *
* Distributed under the MIT software license, see the accompanying *
* file COPYING or http://www.opensource.org/licenses/mit-license.php.*
**********************************************************************/
/* Constant time, unoptimized, concise, plain C, AES implementation
* Based On:
* Emilia Kasper and Peter Schwabe, Faster and Timing-Attack Resistant AES-GCM
* http://www.iacr.org/archive/ches2009/57470001/57470001.pdf
* But using 8 16-bit integers representing a single AES state rather than 8 128-bit
* integers representing 8 AES states.
*/
#include "ctaes.h"
/* Slice variable slice_i contains the i'th bit of the 16 state variables in this order:
* 0 1 2 3
* 4 5 6 7
* 8 9 10 11
* 12 13 14 15
*/
/** Convert a byte to sliced form, storing it corresponding to given row and column in s */
static void LoadByte(AES_state* s, unsigned char byte, int r, int c) {
int i;
for (i = 0; i < 8; i++) {
s->slice[i] |= (byte & 1) << (r * 4 + c);
byte >>= 1;
}
}
/** Load 16 bytes of data into 8 sliced integers */
static void LoadBytes(AES_state *s, const unsigned char* data16) {
int c;
for (c = 0; c < 4; c++) {
int r;
for (r = 0; r < 4; r++) {
LoadByte(s, *(data16++), r, c);
}
}
}
/** Convert 8 sliced integers into 16 bytes of data */
static void SaveBytes(unsigned char* data16, const AES_state *s) {
int c;
for (c = 0; c < 4; c++) {
int r;
for (r = 0; r < 4; r++) {
int b;
uint8_t v = 0;
for (b = 0; b < 8; b++) {
v |= ((s->slice[b] >> (r * 4 + c)) & 1) << b;
}
*(data16++) = v;
}
}
}
/* S-box implementation based on the gate logic from:
* Joan Boyar and Rene Peralta, A depth-16 circuit for the AES S-box.
* https://eprint.iacr.org/2011/332.pdf
*/
static void SubBytes(AES_state *s, int inv) {
/* Load the bit slices */
uint16_t U0 = s->slice[7], U1 = s->slice[6], U2 = s->slice[5], U3 = s->slice[4];
uint16_t U4 = s->slice[3], U5 = s->slice[2], U6 = s->slice[1], U7 = s->slice[0];
uint16_t T1, T2, T3, T4, T5, T6, T7, T8, T9, T10, T11, T12, T13, T14, T15, T16;
uint16_t T17, T18, T19, T20, T21, T22, T23, T24, T25, T26, T27, D;
uint16_t M1, M6, M11, M13, M15, M20, M21, M22, M23, M25, M37, M38, M39, M40;
uint16_t M41, M42, M43, M44, M45, M46, M47, M48, M49, M50, M51, M52, M53, M54;
uint16_t M55, M56, M57, M58, M59, M60, M61, M62, M63;
if (inv) {
uint16_t R5, R13, R17, R18, R19;
/* Undo linear postprocessing */
T23 = U0 ^ U3;
T22 = ~(U1 ^ U3);
T2 = ~(U0 ^ U1);
T1 = U3 ^ U4;
T24 = ~(U4 ^ U7);
R5 = U6 ^ U7;
T8 = ~(U1 ^ T23);
T19 = T22 ^ R5;
T9 = ~(U7 ^ T1);
T10 = T2 ^ T24;
T13 = T2 ^ R5;
T3 = T1 ^ R5;
T25 = ~(U2 ^ T1);
R13 = U1 ^ U6;
T17 = ~(U2 ^ T19);
T20 = T24 ^ R13;
T4 = U4 ^ T8;
R17 = ~(U2 ^ U5);
R18 = ~(U5 ^ U6);
R19 = ~(U2 ^ U4);
D = U0 ^ R17;
T6 = T22 ^ R17;
T16 = R13 ^ R19;
T27 = T1 ^ R18;
T15 = T10 ^ T27;
T14 = T10 ^ R18;
T26 = T3 ^ T16;
} else {
/* Linear preprocessing. */
T1 = U0 ^ U3;
T2 = U0 ^ U5;
T3 = U0 ^ U6;
T4 = U3 ^ U5;
T5 = U4 ^ U6;
T6 = T1 ^ T5;
T7 = U1 ^ U2;
T8 = U7 ^ T6;
T9 = U7 ^ T7;
T10 = T6 ^ T7;
T11 = U1 ^ U5;
T12 = U2 ^ U5;
T13 = T3 ^ T4;
T14 = T6 ^ T11;
T15 = T5 ^ T11;
T16 = T5 ^ T12;
T17 = T9 ^ T16;
T18 = U3 ^ U7;
T19 = T7 ^ T18;
T20 = T1 ^ T19;
T21 = U6 ^ U7;
T22 = T7 ^ T21;
T23 = T2 ^ T22;
T24 = T2 ^ T10;
T25 = T20 ^ T17;
T26 = T3 ^ T16;
T27 = T1 ^ T12;
D = U7;
}
/* Non-linear transformation (shared between the forward and backward case) */
M1 = T13 & T6;
M6 = T3 & T16;
M11 = T1 & T15;
M13 = (T4 & T27) ^ M11;
M15 = (T2 & T10) ^ M11;
M20 = T14 ^ M1 ^ (T23 & T8) ^ M13;
M21 = (T19 & D) ^ M1 ^ T24 ^ M15;
M22 = T26 ^ M6 ^ (T22 & T9) ^ M13;
M23 = (T20 & T17) ^ M6 ^ M15 ^ T25;
M25 = M22 & M20;
M37 = M21 ^ ((M20 ^ M21) & (M23 ^ M25));
M38 = M20 ^ M25 ^ (M21 | (M20 & M23));
M39 = M23 ^ ((M22 ^ M23) & (M21 ^ M25));
M40 = M22 ^ M25 ^ (M23 | (M21 & M22));
M41 = M38 ^ M40;
M42 = M37 ^ M39;
M43 = M37 ^ M38;
M44 = M39 ^ M40;
M45 = M42 ^ M41;
M46 = M44 & T6;
M47 = M40 & T8;
M48 = M39 & D;
M49 = M43 & T16;
M50 = M38 & T9;
M51 = M37 & T17;
M52 = M42 & T15;
M53 = M45 & T27;
M54 = M41 & T10;
M55 = M44 & T13;
M56 = M40 & T23;
M57 = M39 & T19;
M58 = M43 & T3;
M59 = M38 & T22;
M60 = M37 & T20;
M61 = M42 & T1;
M62 = M45 & T4;
M63 = M41 & T2;
if (inv){
/* Undo linear preprocessing */
uint16_t P0 = M52 ^ M61;
uint16_t P1 = M58 ^ M59;
uint16_t P2 = M54 ^ M62;
uint16_t P3 = M47 ^ M50;
uint16_t P4 = M48 ^ M56;
uint16_t P5 = M46 ^ M51;
uint16_t P6 = M49 ^ M60;
uint16_t P7 = P0 ^ P1;
uint16_t P8 = M50 ^ M53;
uint16_t P9 = M55 ^ M63;
uint16_t P10 = M57 ^ P4;
uint16_t P11 = P0 ^ P3;
uint16_t P12 = M46 ^ M48;
uint16_t P13 = M49 ^ M51;
uint16_t P14 = M49 ^ M62;
uint16_t P15 = M54 ^ M59;
uint16_t P16 = M57 ^ M61;
uint16_t P17 = M58 ^ P2;
uint16_t P18 = M63 ^ P5;
uint16_t P19 = P2 ^ P3;
uint16_t P20 = P4 ^ P6;
uint16_t P22 = P2 ^ P7;
uint16_t P23 = P7 ^ P8;
uint16_t P24 = P5 ^ P7;
uint16_t P25 = P6 ^ P10;
uint16_t P26 = P9 ^ P11;
uint16_t P27 = P10 ^ P18;
uint16_t P28 = P11 ^ P25;
uint16_t P29 = P15 ^ P20;
s->slice[7] = P13 ^ P22;
s->slice[6] = P26 ^ P29;
s->slice[5] = P17 ^ P28;
s->slice[4] = P12 ^ P22;
s->slice[3] = P23 ^ P27;
s->slice[2] = P19 ^ P24;
s->slice[1] = P14 ^ P23;
s->slice[0] = P9 ^ P16;
} else {
/* Linear postprocessing */
uint16_t L0 = M61 ^ M62;
uint16_t L1 = M50 ^ M56;
uint16_t L2 = M46 ^ M48;
uint16_t L3 = M47 ^ M55;
uint16_t L4 = M54 ^ M58;
uint16_t L5 = M49 ^ M61;
uint16_t L6 = M62 ^ L5;
uint16_t L7 = M46 ^ L3;
uint16_t L8 = M51 ^ M59;
uint16_t L9 = M52 ^ M53;
uint16_t L10 = M53 ^ L4;
uint16_t L11 = M60 ^ L2;
uint16_t L12 = M48 ^ M51;
uint16_t L13 = M50 ^ L0;
uint16_t L14 = M52 ^ M61;
uint16_t L15 = M55 ^ L1;
uint16_t L16 = M56 ^ L0;
uint16_t L17 = M57 ^ L1;
uint16_t L18 = M58 ^ L8;
uint16_t L19 = M63 ^ L4;
uint16_t L20 = L0 ^ L1;
uint16_t L21 = L1 ^ L7;
uint16_t L22 = L3 ^ L12;
uint16_t L23 = L18 ^ L2;
uint16_t L24 = L15 ^ L9;
uint16_t L25 = L6 ^ L10;
uint16_t L26 = L7 ^ L9;
uint16_t L27 = L8 ^ L10;
uint16_t L28 = L11 ^ L14;
uint16_t L29 = L11 ^ L17;
s->slice[7] = L6 ^ L24;
s->slice[6] = ~(L16 ^ L26);
s->slice[5] = ~(L19 ^ L28);
s->slice[4] = L6 ^ L21;
s->slice[3] = L20 ^ L22;
s->slice[2] = L25 ^ L29;
s->slice[1] = ~(L13 ^ L27);
s->slice[0] = ~(L6 ^ L23);
}
}
#define BIT_RANGE(from,to) (((1 << ((to) - (from))) - 1) << (from))
#define BIT_RANGE_LEFT(x,from,to,shift) (((x) & BIT_RANGE((from), (to))) << (shift))
#define BIT_RANGE_RIGHT(x,from,to,shift) (((x) & BIT_RANGE((from), (to))) >> (shift))
static void ShiftRows(AES_state* s) {
int i;
for (i = 0; i < 8; i++) {
uint16_t v = s->slice[i];
s->slice[i] =
(v & BIT_RANGE(0, 4)) |
BIT_RANGE_LEFT(v, 4, 5, 3) | BIT_RANGE_RIGHT(v, 5, 8, 1) |
BIT_RANGE_LEFT(v, 8, 10, 2) | BIT_RANGE_RIGHT(v, 10, 12, 2) |
BIT_RANGE_LEFT(v, 12, 15, 1) | BIT_RANGE_RIGHT(v, 15, 16, 3);
}
}
static void InvShiftRows(AES_state* s) {
int i;
for (i = 0; i < 8; i++) {
uint16_t v = s->slice[i];
s->slice[i] =
(v & BIT_RANGE(0, 4)) |
BIT_RANGE_LEFT(v, 4, 7, 1) | BIT_RANGE_RIGHT(v, 7, 8, 3) |
BIT_RANGE_LEFT(v, 8, 10, 2) | BIT_RANGE_RIGHT(v, 10, 12, 2) |
BIT_RANGE_LEFT(v, 12, 13, 3) | BIT_RANGE_RIGHT(v, 13, 16, 1);
}
}
#define ROT(x,b) (((x) >> ((b) * 4)) | ((x) << ((4-(b)) * 4)))
static void MixColumns(AES_state* s, int inv) {
/* The MixColumns transform treats the bytes of the columns of the state as
* coefficients of a 3rd degree polynomial over GF(2^8) and multiplies them
* by the fixed polynomial a(x) = {03}x^3 + {01}x^2 + {01}x + {02}, modulo
* x^4 + {01}.
*
* In the inverse transform, we multiply by the inverse of a(x),
* a^-1(x) = {0b}x^3 + {0d}x^2 + {09}x + {0e}. This is equal to
* a(x) * ({04}x^2 + {05}), so we can reuse the forward transform's code
* (found in OpenSSL's bsaes-x86_64.pl, attributed to Jussi Kivilinna)
*
* In the bitsliced representation, a multiplication of every column by x
* mod x^4 + 1 is simply a right rotation.
*/
/* Shared for both directions is a multiplication by a(x), which can be
* rewritten as (x^3 + x^2 + x) + {02}*(x^3 + {01}).
*
* First compute s into the s? variables, (x^3 + {01}) * s into the s?_01
* variables and (x^3 + x^2 + x)*s into the s?_123 variables.
*/
uint16_t s0 = s->slice[0], s1 = s->slice[1], s2 = s->slice[2], s3 = s->slice[3];
uint16_t s4 = s->slice[4], s5 = s->slice[5], s6 = s->slice[6], s7 = s->slice[7];
uint16_t s0_01 = s0 ^ ROT(s0, 1), s0_123 = ROT(s0_01, 1) ^ ROT(s0, 3);
uint16_t s1_01 = s1 ^ ROT(s1, 1), s1_123 = ROT(s1_01, 1) ^ ROT(s1, 3);
uint16_t s2_01 = s2 ^ ROT(s2, 1), s2_123 = ROT(s2_01, 1) ^ ROT(s2, 3);
uint16_t s3_01 = s3 ^ ROT(s3, 1), s3_123 = ROT(s3_01, 1) ^ ROT(s3, 3);
uint16_t s4_01 = s4 ^ ROT(s4, 1), s4_123 = ROT(s4_01, 1) ^ ROT(s4, 3);
uint16_t s5_01 = s5 ^ ROT(s5, 1), s5_123 = ROT(s5_01, 1) ^ ROT(s5, 3);
uint16_t s6_01 = s6 ^ ROT(s6, 1), s6_123 = ROT(s6_01, 1) ^ ROT(s6, 3);
uint16_t s7_01 = s7 ^ ROT(s7, 1), s7_123 = ROT(s7_01, 1) ^ ROT(s7, 3);
/* Now compute s = s?_123 + {02} * s?_01. */
s->slice[0] = s7_01 ^ s0_123;
s->slice[1] = s7_01 ^ s0_01 ^ s1_123;
s->slice[2] = s1_01 ^ s2_123;
s->slice[3] = s7_01 ^ s2_01 ^ s3_123;
s->slice[4] = s7_01 ^ s3_01 ^ s4_123;
s->slice[5] = s4_01 ^ s5_123;
s->slice[6] = s5_01 ^ s6_123;
s->slice[7] = s6_01 ^ s7_123;
if (inv) {
/* In the reverse direction, we further need to multiply by
* {04}x^2 + {05}, which can be written as {04} * (x^2 + {01}) + {01}.
*
* First compute (x^2 + {01}) * s into the t?_02 variables: */
uint16_t t0_02 = s->slice[0] ^ ROT(s->slice[0], 2);
uint16_t t1_02 = s->slice[1] ^ ROT(s->slice[1], 2);
uint16_t t2_02 = s->slice[2] ^ ROT(s->slice[2], 2);
uint16_t t3_02 = s->slice[3] ^ ROT(s->slice[3], 2);
uint16_t t4_02 = s->slice[4] ^ ROT(s->slice[4], 2);
uint16_t t5_02 = s->slice[5] ^ ROT(s->slice[5], 2);
uint16_t t6_02 = s->slice[6] ^ ROT(s->slice[6], 2);
uint16_t t7_02 = s->slice[7] ^ ROT(s->slice[7], 2);
/* And then update s += {04} * t?_02 */
s->slice[0] ^= t6_02;
s->slice[1] ^= t6_02 ^ t7_02;
s->slice[2] ^= t0_02 ^ t7_02;
s->slice[3] ^= t1_02 ^ t6_02;
s->slice[4] ^= t2_02 ^ t6_02 ^ t7_02;
s->slice[5] ^= t3_02 ^ t7_02;
s->slice[6] ^= t4_02;
s->slice[7] ^= t5_02;
}
}
static void AddRoundKey(AES_state* s, const AES_state* round) {
int b;
for (b = 0; b < 8; b++) {
s->slice[b] ^= round->slice[b];
}
}
/** column_0(s) = column_c(a) */
static void GetOneColumn(AES_state* s, const AES_state* a, int c) {
int b;
for (b = 0; b < 8; b++) {
s->slice[b] = (a->slice[b] >> c) & 0x1111;
}
}
/** column_c1(r) |= (column_0(s) ^= column_c2(a)) */
static void KeySetupColumnMix(AES_state* s, AES_state* r, const AES_state* a, int c1, int c2) {
int b;
for (b = 0; b < 8; b++) {
r->slice[b] |= ((s->slice[b] ^= ((a->slice[b] >> c2) & 0x1111)) & 0x1111) << c1;
}
}
/** Rotate the rows in s one position upwards, and xor in r */
static void KeySetupTransform(AES_state* s, const AES_state* r) {
int b;
for (b = 0; b < 8; b++) {
s->slice[b] = ((s->slice[b] >> 4) | (s->slice[b] << 12)) ^ r->slice[b];
}
}
/* Multiply the cells in s by x, as polynomials over GF(2) mod x^8 + x^4 + x^3 + x + 1 */
static void MultX(AES_state* s) {
uint16_t top = s->slice[7];
s->slice[7] = s->slice[6];
s->slice[6] = s->slice[5];
s->slice[5] = s->slice[4];
s->slice[4] = s->slice[3] ^ top;
s->slice[3] = s->slice[2] ^ top;
s->slice[2] = s->slice[1];
s->slice[1] = s->slice[0] ^ top;
s->slice[0] = top;
}
/** Expand the cipher key into the key schedule.
*
* state must be a pointer to an array of size nrounds + 1.
* key must be a pointer to 4 * nkeywords bytes.
*
* AES128 uses nkeywords = 4, nrounds = 10
* AES192 uses nkeywords = 6, nrounds = 12
* AES256 uses nkeywords = 8, nrounds = 14
*/
static void AES_setup(AES_state* rounds, const uint8_t* key, int nkeywords, int nrounds)
{
int i;
/* The one-byte round constant */
AES_state rcon = {{1,0,0,0,0,0,0,0}};
/* The number of the word being generated, modulo nkeywords */
int pos = 0;
/* The column representing the word currently being processed */
AES_state column;
for (i = 0; i < nrounds + 1; i++) {
int b;
for (b = 0; b < 8; b++) {
rounds[i].slice[b] = 0;
}
}
/* The first nkeywords round columns are just taken from the key directly. */
for (i = 0; i < nkeywords; i++) {
int r;
for (r = 0; r < 4; r++) {
LoadByte(&rounds[i >> 2], *(key++), r, i & 3);
}
}
GetOneColumn(&column, &rounds[(nkeywords - 1) >> 2], (nkeywords - 1) & 3);
for (i = nkeywords; i < 4 * (nrounds + 1); i++) {
/* Transform column */
if (pos == 0) {
SubBytes(&column, 0);
KeySetupTransform(&column, &rcon);
MultX(&rcon);
} else if (nkeywords > 6 && pos == 4) {
SubBytes(&column, 0);
}
if (++pos == nkeywords) pos = 0;
KeySetupColumnMix(&column, &rounds[i >> 2], &rounds[(i - nkeywords) >> 2], i & 3, (i - nkeywords) & 3);
}
}
static void AES_encrypt(const AES_state* rounds, int nrounds, unsigned char* cipher16, const unsigned char* plain16) {
AES_state s = {{0}};
int round;
LoadBytes(&s, plain16);
AddRoundKey(&s, rounds++);
for (round = 1; round < nrounds; round++) {
SubBytes(&s, 0);
ShiftRows(&s);
MixColumns(&s, 0);
AddRoundKey(&s, rounds++);
}
SubBytes(&s, 0);
ShiftRows(&s);
AddRoundKey(&s, rounds);
SaveBytes(cipher16, &s);
}
static void AES_decrypt(const AES_state* rounds, int nrounds, unsigned char* plain16, const unsigned char* cipher16) {
/* Most AES decryption implementations use the alternate scheme
* (the Equivalent Inverse Cipher), which allows for more code reuse between
* the encryption and decryption code, but requires separate setup for both.
*/
AES_state s = {{0}};
int round;
rounds += nrounds;
LoadBytes(&s, cipher16);
AddRoundKey(&s, rounds--);
for (round = 1; round < nrounds; round++) {
InvShiftRows(&s);
SubBytes(&s, 1);
AddRoundKey(&s, rounds--);
MixColumns(&s, 1);
}
InvShiftRows(&s);
SubBytes(&s, 1);
AddRoundKey(&s, rounds);
SaveBytes(plain16, &s);
}
void AES128_init(AES128_ctx* ctx, const unsigned char* key16) {
AES_setup(ctx->rk, key16, 4, 10);
}
void AES128_encrypt(const AES128_ctx* ctx, size_t blocks, unsigned char* cipher16, const unsigned char* plain16) {
while (blocks--) {
AES_encrypt(ctx->rk, 10, cipher16, plain16);
cipher16 += 16;
plain16 += 16;
}
}
void AES128_decrypt(const AES128_ctx* ctx, size_t blocks, unsigned char* plain16, const unsigned char* cipher16) {
while (blocks--) {
AES_decrypt(ctx->rk, 10, plain16, cipher16);
cipher16 += 16;
plain16 += 16;
}
}
void AES192_init(AES192_ctx* ctx, const unsigned char* key24) {
AES_setup(ctx->rk, key24, 6, 12);
}
void AES192_encrypt(const AES192_ctx* ctx, size_t blocks, unsigned char* cipher16, const unsigned char* plain16) {
while (blocks--) {
AES_encrypt(ctx->rk, 12, cipher16, plain16);
cipher16 += 16;
plain16 += 16;
}
}
void AES192_decrypt(const AES192_ctx* ctx, size_t blocks, unsigned char* plain16, const unsigned char* cipher16) {
while (blocks--) {
AES_decrypt(ctx->rk, 12, plain16, cipher16);
cipher16 += 16;
plain16 += 16;
}
}
void AES256_init(AES256_ctx* ctx, const unsigned char* key32) {
AES_setup(ctx->rk, key32, 8, 14);
}
void AES256_encrypt(const AES256_ctx* ctx, size_t blocks, unsigned char* cipher16, const unsigned char* plain16) {
while (blocks--) {
AES_encrypt(ctx->rk, 14, cipher16, plain16);
cipher16 += 16;
plain16 += 16;
}
}
void AES256_decrypt(const AES256_ctx* ctx, size_t blocks, unsigned char* plain16, const unsigned char* cipher16) {
while (blocks--) {
AES_decrypt(ctx->rk, 14, plain16, cipher16);
cipher16 += 16;
plain16 += 16;
}
}

41
src/crypto/ctaes/ctaes.h Normal file
View file

@ -0,0 +1,41 @@
/*********************************************************************
* Copyright (c) 2016 Pieter Wuille *
* Distributed under the MIT software license, see the accompanying *
* file COPYING or http://www.opensource.org/licenses/mit-license.php.*
**********************************************************************/
#ifndef _CTAES_H_
#define _CTAES_H_ 1
#include <stdint.h>
#include <stdlib.h>
typedef struct {
uint16_t slice[8];
} AES_state;
typedef struct {
AES_state rk[11];
} AES128_ctx;
typedef struct {
AES_state rk[13];
} AES192_ctx;
typedef struct {
AES_state rk[15];
} AES256_ctx;
void AES128_init(AES128_ctx* ctx, const unsigned char* key16);
void AES128_encrypt(const AES128_ctx* ctx, size_t blocks, unsigned char* cipher16, const unsigned char* plain16);
void AES128_decrypt(const AES128_ctx* ctx, size_t blocks, unsigned char* plain16, const unsigned char* cipher16);
void AES192_init(AES192_ctx* ctx, const unsigned char* key24);
void AES192_encrypt(const AES192_ctx* ctx, size_t blocks, unsigned char* cipher16, const unsigned char* plain16);
void AES192_decrypt(const AES192_ctx* ctx, size_t blocks, unsigned char* plain16, const unsigned char* cipher16);
void AES256_init(AES256_ctx* ctx, const unsigned char* key32);
void AES256_encrypt(const AES256_ctx* ctx, size_t blocks, unsigned char* cipher16, const unsigned char* plain16);
void AES256_decrypt(const AES256_ctx* ctx, size_t blocks, unsigned char* plain16, const unsigned char* cipher16);
#endif

110
src/crypto/ctaes/test.c Normal file
View file

@ -0,0 +1,110 @@
/*********************************************************************
* Copyright (c) 2016 Pieter Wuille *
* Distributed under the MIT software license, see the accompanying *
* file COPYING or http://www.opensource.org/licenses/mit-license.php.*
**********************************************************************/
#include "ctaes.h"
#include <stdio.h>
#include <string.h>
#include <assert.h>
typedef struct {
int keysize;
const char* key;
const char* plain;
const char* cipher;
} ctaes_test;
static const ctaes_test ctaes_tests[] = {
/* AES test vectors from FIPS 197. */
{128, "000102030405060708090a0b0c0d0e0f", "00112233445566778899aabbccddeeff", "69c4e0d86a7b0430d8cdb78070b4c55a"},
{192, "000102030405060708090a0b0c0d0e0f1011121314151617", "00112233445566778899aabbccddeeff", "dda97ca4864cdfe06eaf70a0ec0d7191"},
{256, "000102030405060708090a0b0c0d0e0f101112131415161718191a1b1c1d1e1f", "00112233445566778899aabbccddeeff", "8ea2b7ca516745bfeafc49904b496089"},
/* AES-ECB test vectors from NIST sp800-38a. */
{128, "2b7e151628aed2a6abf7158809cf4f3c", "6bc1bee22e409f96e93d7e117393172a", "3ad77bb40d7a3660a89ecaf32466ef97"},
{128, "2b7e151628aed2a6abf7158809cf4f3c", "ae2d8a571e03ac9c9eb76fac45af8e51", "f5d3d58503b9699de785895a96fdbaaf"},
{128, "2b7e151628aed2a6abf7158809cf4f3c", "30c81c46a35ce411e5fbc1191a0a52ef", "43b1cd7f598ece23881b00e3ed030688"},
{128, "2b7e151628aed2a6abf7158809cf4f3c", "f69f2445df4f9b17ad2b417be66c3710", "7b0c785e27e8ad3f8223207104725dd4"},
{192, "8e73b0f7da0e6452c810f32b809079e562f8ead2522c6b7b", "6bc1bee22e409f96e93d7e117393172a", "bd334f1d6e45f25ff712a214571fa5cc"},
{192, "8e73b0f7da0e6452c810f32b809079e562f8ead2522c6b7b", "ae2d8a571e03ac9c9eb76fac45af8e51", "974104846d0ad3ad7734ecb3ecee4eef"},
{192, "8e73b0f7da0e6452c810f32b809079e562f8ead2522c6b7b", "30c81c46a35ce411e5fbc1191a0a52ef", "ef7afd2270e2e60adce0ba2face6444e"},
{192, "8e73b0f7da0e6452c810f32b809079e562f8ead2522c6b7b", "f69f2445df4f9b17ad2b417be66c3710", "9a4b41ba738d6c72fb16691603c18e0e"},
{256, "603deb1015ca71be2b73aef0857d77811f352c073b6108d72d9810a30914dff4", "6bc1bee22e409f96e93d7e117393172a", "f3eed1bdb5d2a03c064b5a7e3db181f8"},
{256, "603deb1015ca71be2b73aef0857d77811f352c073b6108d72d9810a30914dff4", "ae2d8a571e03ac9c9eb76fac45af8e51", "591ccb10d410ed26dc5ba74a31362870"},
{256, "603deb1015ca71be2b73aef0857d77811f352c073b6108d72d9810a30914dff4", "30c81c46a35ce411e5fbc1191a0a52ef", "b6ed21b99ca6f4f9f153e7b1beafed1d"},
{256, "603deb1015ca71be2b73aef0857d77811f352c073b6108d72d9810a30914dff4", "f69f2445df4f9b17ad2b417be66c3710", "23304b7a39f9f3ff067d8d8f9e24ecc7"}
};
static void from_hex(unsigned char* data, int len, const char* hex) {
int p;
for (p = 0; p < len; p++) {
int v = 0;
int n;
for (n = 0; n < 2; n++) {
assert((*hex >= '0' && *hex <= '9') || (*hex >= 'a' && *hex <= 'f'));
if (*hex >= '0' && *hex <= '9') {
v |= (*hex - '0') << (4 * (1 - n));
} else {
v |= (*hex - 'a' + 10) << (4 * (1 - n));
}
hex++;
}
*(data++) = v;
}
assert(*hex == 0);
}
int main(void) {
int i;
int fail = 0;
for (i = 0; i < sizeof(ctaes_tests) / sizeof(ctaes_tests[0]); i++) {
unsigned char key[32], plain[16], cipher[16], ciphered[16], deciphered[16];
const ctaes_test* test = &ctaes_tests[i];
assert(test->keysize == 128 || test->keysize == 192 || test->keysize == 256);
from_hex(plain, 16, test->plain);
from_hex(cipher, 16, test->cipher);
switch (test->keysize) {
case 128: {
AES128_ctx ctx;
from_hex(key, 16, test->key);
AES128_init(&ctx, key);
AES128_encrypt(&ctx, 1, ciphered, plain);
AES128_decrypt(&ctx, 1, deciphered, cipher);
break;
}
case 192: {
AES192_ctx ctx;
from_hex(key, 24, test->key);
AES192_init(&ctx, key);
AES192_encrypt(&ctx, 1, ciphered, plain);
AES192_decrypt(&ctx, 1, deciphered, cipher);
break;
}
case 256: {
AES256_ctx ctx;
from_hex(key, 32, test->key);
AES256_init(&ctx, key);
AES256_encrypt(&ctx, 1, ciphered, plain);
AES256_decrypt(&ctx, 1, deciphered, cipher);
break;
}
}
if (memcmp(cipher, ciphered, 16)) {
fprintf(stderr, "E(key=\"%s\", plain=\"%s\") != \"%s\"\n", test->key, test->plain, test->cipher);
fail++;
}
if (memcmp(plain, deciphered, 16)) {
fprintf(stderr, "D(key=\"%s\", cipher=\"%s\") != \"%s\"\n", test->key, test->cipher, test->plain);
fail++;
}
}
if (fail == 0) {
fprintf(stderr, "All tests successful\n");
} else {
fprintf(stderr, "%i tests failed\n", fail);
}
return (fail != 0);
}

723
src/crypto/cubehash.cpp Normal file
View file

@ -0,0 +1,723 @@
/* $Id: cubehash.c 227 2010-06-16 17:28:38Z tp $ */
/*
* CubeHash implementation.
*
* ==========================(LICENSE BEGIN)============================
*
* Copyright (c) 2007-2010 Projet RNRT SAPHIR
*
* Permission is hereby granted, free of charge, to any person obtaining
* a copy of this software and associated documentation files (the
* "Software"), to deal in the Software without restriction, including
* without limitation the rights to use, copy, modify, merge, publish,
* distribute, sublicense, and/or sell copies of the Software, and to
* permit persons to whom the Software is furnished to do so, subject to
* the following conditions:
*
* The above copyright notice and this permission notice shall be
* included in all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*
* ===========================(LICENSE END)=============================
*
* @author Thomas Pornin <thomas.pornin@cryptolog.com>
*/
#include <stddef.h>
#include <string.h>
#include <limits.h>
#include "sph_cubehash.h"
#ifdef __cplusplus
extern "C"{
#endif
#if SPH_SMALL_FOOTPRINT && !defined SPH_SMALL_FOOTPRINT_CUBEHASH
#define SPH_SMALL_FOOTPRINT_CUBEHASH 1
#endif
/*
* Some tests were conducted on an Intel Core2 Q6600 (32-bit and 64-bit
* mode), a PowerPC G3, and a MIPS-compatible CPU (Broadcom BCM3302).
* It appears that the optimal settings are:
* -- full unroll, no state copy on the "big" systems (x86, PowerPC)
* -- unroll to 4 or 8, state copy on the "small" system (MIPS)
*/
#if SPH_SMALL_FOOTPRINT_CUBEHASH
#if !defined SPH_CUBEHASH_UNROLL
#define SPH_CUBEHASH_UNROLL 4
#endif
#if !defined SPH_CUBEHASH_NOCOPY
#define SPH_CUBEHASH_NOCOPY 1
#endif
#else
#if !defined SPH_CUBEHASH_UNROLL
#define SPH_CUBEHASH_UNROLL 0
#endif
#if !defined SPH_CUBEHASH_NOCOPY
#define SPH_CUBEHASH_NOCOPY 0
#endif
#endif
#ifdef _MSC_VER
#pragma warning (disable: 4146)
#endif
static const sph_u32 IV224[] = {
SPH_C32(0xB0FC8217), SPH_C32(0x1BEE1A90), SPH_C32(0x829E1A22),
SPH_C32(0x6362C342), SPH_C32(0x24D91C30), SPH_C32(0x03A7AA24),
SPH_C32(0xA63721C8), SPH_C32(0x85B0E2EF), SPH_C32(0xF35D13F3),
SPH_C32(0x41DA807D), SPH_C32(0x21A70CA6), SPH_C32(0x1F4E9774),
SPH_C32(0xB3E1C932), SPH_C32(0xEB0A79A8), SPH_C32(0xCDDAAA66),
SPH_C32(0xE2F6ECAA), SPH_C32(0x0A713362), SPH_C32(0xAA3080E0),
SPH_C32(0xD8F23A32), SPH_C32(0xCEF15E28), SPH_C32(0xDB086314),
SPH_C32(0x7F709DF7), SPH_C32(0xACD228A4), SPH_C32(0x704D6ECE),
SPH_C32(0xAA3EC95F), SPH_C32(0xE387C214), SPH_C32(0x3A6445FF),
SPH_C32(0x9CAB81C3), SPH_C32(0xC73D4B98), SPH_C32(0xD277AEBE),
SPH_C32(0xFD20151C), SPH_C32(0x00CB573E)
};
static const sph_u32 IV256[] = {
SPH_C32(0xEA2BD4B4), SPH_C32(0xCCD6F29F), SPH_C32(0x63117E71),
SPH_C32(0x35481EAE), SPH_C32(0x22512D5B), SPH_C32(0xE5D94E63),
SPH_C32(0x7E624131), SPH_C32(0xF4CC12BE), SPH_C32(0xC2D0B696),
SPH_C32(0x42AF2070), SPH_C32(0xD0720C35), SPH_C32(0x3361DA8C),
SPH_C32(0x28CCECA4), SPH_C32(0x8EF8AD83), SPH_C32(0x4680AC00),
SPH_C32(0x40E5FBAB), SPH_C32(0xD89041C3), SPH_C32(0x6107FBD5),
SPH_C32(0x6C859D41), SPH_C32(0xF0B26679), SPH_C32(0x09392549),
SPH_C32(0x5FA25603), SPH_C32(0x65C892FD), SPH_C32(0x93CB6285),
SPH_C32(0x2AF2B5AE), SPH_C32(0x9E4B4E60), SPH_C32(0x774ABFDD),
SPH_C32(0x85254725), SPH_C32(0x15815AEB), SPH_C32(0x4AB6AAD6),
SPH_C32(0x9CDAF8AF), SPH_C32(0xD6032C0A)
};
static const sph_u32 IV384[] = {
SPH_C32(0xE623087E), SPH_C32(0x04C00C87), SPH_C32(0x5EF46453),
SPH_C32(0x69524B13), SPH_C32(0x1A05C7A9), SPH_C32(0x3528DF88),
SPH_C32(0x6BDD01B5), SPH_C32(0x5057B792), SPH_C32(0x6AA7A922),
SPH_C32(0x649C7EEE), SPH_C32(0xF426309F), SPH_C32(0xCB629052),
SPH_C32(0xFC8E20ED), SPH_C32(0xB3482BAB), SPH_C32(0xF89E5E7E),
SPH_C32(0xD83D4DE4), SPH_C32(0x44BFC10D), SPH_C32(0x5FC1E63D),
SPH_C32(0x2104E6CB), SPH_C32(0x17958F7F), SPH_C32(0xDBEAEF70),
SPH_C32(0xB4B97E1E), SPH_C32(0x32C195F6), SPH_C32(0x6184A8E4),
SPH_C32(0x796C2543), SPH_C32(0x23DE176D), SPH_C32(0xD33BBAEC),
SPH_C32(0x0C12E5D2), SPH_C32(0x4EB95A7B), SPH_C32(0x2D18BA01),
SPH_C32(0x04EE475F), SPH_C32(0x1FC5F22E)
};
static const sph_u32 IV512[] = {
SPH_C32(0x2AEA2A61), SPH_C32(0x50F494D4), SPH_C32(0x2D538B8B),
SPH_C32(0x4167D83E), SPH_C32(0x3FEE2313), SPH_C32(0xC701CF8C),
SPH_C32(0xCC39968E), SPH_C32(0x50AC5695), SPH_C32(0x4D42C787),
SPH_C32(0xA647A8B3), SPH_C32(0x97CF0BEF), SPH_C32(0x825B4537),
SPH_C32(0xEEF864D2), SPH_C32(0xF22090C4), SPH_C32(0xD0E5CD33),
SPH_C32(0xA23911AE), SPH_C32(0xFCD398D9), SPH_C32(0x148FE485),
SPH_C32(0x1B017BEF), SPH_C32(0xB6444532), SPH_C32(0x6A536159),
SPH_C32(0x2FF5781C), SPH_C32(0x91FA7934), SPH_C32(0x0DBADEA9),
SPH_C32(0xD65C8A2B), SPH_C32(0xA5A70E75), SPH_C32(0xB1C62456),
SPH_C32(0xBC796576), SPH_C32(0x1921C8F7), SPH_C32(0xE7989AF1),
SPH_C32(0x7795D246), SPH_C32(0xD43E3B44)
};
#define T32 SPH_T32
#define ROTL32 SPH_ROTL32
#if SPH_CUBEHASH_NOCOPY
#define DECL_STATE
#define READ_STATE(cc)
#define WRITE_STATE(cc)
#define x0 ((sc)->state[ 0])
#define x1 ((sc)->state[ 1])
#define x2 ((sc)->state[ 2])
#define x3 ((sc)->state[ 3])
#define x4 ((sc)->state[ 4])
#define x5 ((sc)->state[ 5])
#define x6 ((sc)->state[ 6])
#define x7 ((sc)->state[ 7])
#define x8 ((sc)->state[ 8])
#define x9 ((sc)->state[ 9])
#define xa ((sc)->state[10])
#define xb ((sc)->state[11])
#define xc ((sc)->state[12])
#define xd ((sc)->state[13])
#define xe ((sc)->state[14])
#define xf ((sc)->state[15])
#define xg ((sc)->state[16])
#define xh ((sc)->state[17])
#define xi ((sc)->state[18])
#define xj ((sc)->state[19])
#define xk ((sc)->state[20])
#define xl ((sc)->state[21])
#define xm ((sc)->state[22])
#define xn ((sc)->state[23])
#define xo ((sc)->state[24])
#define xp ((sc)->state[25])
#define xq ((sc)->state[26])
#define xr ((sc)->state[27])
#define xs ((sc)->state[28])
#define xt ((sc)->state[29])
#define xu ((sc)->state[30])
#define xv ((sc)->state[31])
#else
#define DECL_STATE \
sph_u32 x0, x1, x2, x3, x4, x5, x6, x7; \
sph_u32 x8, x9, xa, xb, xc, xd, xe, xf; \
sph_u32 xg, xh, xi, xj, xk, xl, xm, xn; \
sph_u32 xo, xp, xq, xr, xs, xt, xu, xv;
#define READ_STATE(cc) do { \
x0 = (cc)->state[ 0]; \
x1 = (cc)->state[ 1]; \
x2 = (cc)->state[ 2]; \
x3 = (cc)->state[ 3]; \
x4 = (cc)->state[ 4]; \
x5 = (cc)->state[ 5]; \
x6 = (cc)->state[ 6]; \
x7 = (cc)->state[ 7]; \
x8 = (cc)->state[ 8]; \
x9 = (cc)->state[ 9]; \
xa = (cc)->state[10]; \
xb = (cc)->state[11]; \
xc = (cc)->state[12]; \
xd = (cc)->state[13]; \
xe = (cc)->state[14]; \
xf = (cc)->state[15]; \
xg = (cc)->state[16]; \
xh = (cc)->state[17]; \
xi = (cc)->state[18]; \
xj = (cc)->state[19]; \
xk = (cc)->state[20]; \
xl = (cc)->state[21]; \
xm = (cc)->state[22]; \
xn = (cc)->state[23]; \
xo = (cc)->state[24]; \
xp = (cc)->state[25]; \
xq = (cc)->state[26]; \
xr = (cc)->state[27]; \
xs = (cc)->state[28]; \
xt = (cc)->state[29]; \
xu = (cc)->state[30]; \
xv = (cc)->state[31]; \
} while (0)
#define WRITE_STATE(cc) do { \
(cc)->state[ 0] = x0; \
(cc)->state[ 1] = x1; \
(cc)->state[ 2] = x2; \
(cc)->state[ 3] = x3; \
(cc)->state[ 4] = x4; \
(cc)->state[ 5] = x5; \
(cc)->state[ 6] = x6; \
(cc)->state[ 7] = x7; \
(cc)->state[ 8] = x8; \
(cc)->state[ 9] = x9; \
(cc)->state[10] = xa; \
(cc)->state[11] = xb; \
(cc)->state[12] = xc; \
(cc)->state[13] = xd; \
(cc)->state[14] = xe; \
(cc)->state[15] = xf; \
(cc)->state[16] = xg; \
(cc)->state[17] = xh; \
(cc)->state[18] = xi; \
(cc)->state[19] = xj; \
(cc)->state[20] = xk; \
(cc)->state[21] = xl; \
(cc)->state[22] = xm; \
(cc)->state[23] = xn; \
(cc)->state[24] = xo; \
(cc)->state[25] = xp; \
(cc)->state[26] = xq; \
(cc)->state[27] = xr; \
(cc)->state[28] = xs; \
(cc)->state[29] = xt; \
(cc)->state[30] = xu; \
(cc)->state[31] = xv; \
} while (0)
#endif
#define INPUT_BLOCK do { \
x0 ^= sph_dec32le_aligned(buf + 0); \
x1 ^= sph_dec32le_aligned(buf + 4); \
x2 ^= sph_dec32le_aligned(buf + 8); \
x3 ^= sph_dec32le_aligned(buf + 12); \
x4 ^= sph_dec32le_aligned(buf + 16); \
x5 ^= sph_dec32le_aligned(buf + 20); \
x6 ^= sph_dec32le_aligned(buf + 24); \
x7 ^= sph_dec32le_aligned(buf + 28); \
} while (0)
#define ROUND_EVEN do { \
xg = T32(x0 + xg); \
x0 = ROTL32(x0, 7); \
xh = T32(x1 + xh); \
x1 = ROTL32(x1, 7); \
xi = T32(x2 + xi); \
x2 = ROTL32(x2, 7); \
xj = T32(x3 + xj); \
x3 = ROTL32(x3, 7); \
xk = T32(x4 + xk); \
x4 = ROTL32(x4, 7); \
xl = T32(x5 + xl); \
x5 = ROTL32(x5, 7); \
xm = T32(x6 + xm); \
x6 = ROTL32(x6, 7); \
xn = T32(x7 + xn); \
x7 = ROTL32(x7, 7); \
xo = T32(x8 + xo); \
x8 = ROTL32(x8, 7); \
xp = T32(x9 + xp); \
x9 = ROTL32(x9, 7); \
xq = T32(xa + xq); \
xa = ROTL32(xa, 7); \
xr = T32(xb + xr); \
xb = ROTL32(xb, 7); \
xs = T32(xc + xs); \
xc = ROTL32(xc, 7); \
xt = T32(xd + xt); \
xd = ROTL32(xd, 7); \
xu = T32(xe + xu); \
xe = ROTL32(xe, 7); \
xv = T32(xf + xv); \
xf = ROTL32(xf, 7); \
x8 ^= xg; \
x9 ^= xh; \
xa ^= xi; \
xb ^= xj; \
xc ^= xk; \
xd ^= xl; \
xe ^= xm; \
xf ^= xn; \
x0 ^= xo; \
x1 ^= xp; \
x2 ^= xq; \
x3 ^= xr; \
x4 ^= xs; \
x5 ^= xt; \
x6 ^= xu; \
x7 ^= xv; \
xi = T32(x8 + xi); \
x8 = ROTL32(x8, 11); \
xj = T32(x9 + xj); \
x9 = ROTL32(x9, 11); \
xg = T32(xa + xg); \
xa = ROTL32(xa, 11); \
xh = T32(xb + xh); \
xb = ROTL32(xb, 11); \
xm = T32(xc + xm); \
xc = ROTL32(xc, 11); \
xn = T32(xd + xn); \
xd = ROTL32(xd, 11); \
xk = T32(xe + xk); \
xe = ROTL32(xe, 11); \
xl = T32(xf + xl); \
xf = ROTL32(xf, 11); \
xq = T32(x0 + xq); \
x0 = ROTL32(x0, 11); \
xr = T32(x1 + xr); \
x1 = ROTL32(x1, 11); \
xo = T32(x2 + xo); \
x2 = ROTL32(x2, 11); \
xp = T32(x3 + xp); \
x3 = ROTL32(x3, 11); \
xu = T32(x4 + xu); \
x4 = ROTL32(x4, 11); \
xv = T32(x5 + xv); \
x5 = ROTL32(x5, 11); \
xs = T32(x6 + xs); \
x6 = ROTL32(x6, 11); \
xt = T32(x7 + xt); \
x7 = ROTL32(x7, 11); \
xc ^= xi; \
xd ^= xj; \
xe ^= xg; \
xf ^= xh; \
x8 ^= xm; \
x9 ^= xn; \
xa ^= xk; \
xb ^= xl; \
x4 ^= xq; \
x5 ^= xr; \
x6 ^= xo; \
x7 ^= xp; \
x0 ^= xu; \
x1 ^= xv; \
x2 ^= xs; \
x3 ^= xt; \
} while (0)
#define ROUND_ODD do { \
xj = T32(xc + xj); \
xc = ROTL32(xc, 7); \
xi = T32(xd + xi); \
xd = ROTL32(xd, 7); \
xh = T32(xe + xh); \
xe = ROTL32(xe, 7); \
xg = T32(xf + xg); \
xf = ROTL32(xf, 7); \
xn = T32(x8 + xn); \
x8 = ROTL32(x8, 7); \
xm = T32(x9 + xm); \
x9 = ROTL32(x9, 7); \
xl = T32(xa + xl); \
xa = ROTL32(xa, 7); \
xk = T32(xb + xk); \
xb = ROTL32(xb, 7); \
xr = T32(x4 + xr); \
x4 = ROTL32(x4, 7); \
xq = T32(x5 + xq); \
x5 = ROTL32(x5, 7); \
xp = T32(x6 + xp); \
x6 = ROTL32(x6, 7); \
xo = T32(x7 + xo); \
x7 = ROTL32(x7, 7); \
xv = T32(x0 + xv); \
x0 = ROTL32(x0, 7); \
xu = T32(x1 + xu); \
x1 = ROTL32(x1, 7); \
xt = T32(x2 + xt); \
x2 = ROTL32(x2, 7); \
xs = T32(x3 + xs); \
x3 = ROTL32(x3, 7); \
x4 ^= xj; \
x5 ^= xi; \
x6 ^= xh; \
x7 ^= xg; \
x0 ^= xn; \
x1 ^= xm; \
x2 ^= xl; \
x3 ^= xk; \
xc ^= xr; \
xd ^= xq; \
xe ^= xp; \
xf ^= xo; \
x8 ^= xv; \
x9 ^= xu; \
xa ^= xt; \
xb ^= xs; \
xh = T32(x4 + xh); \
x4 = ROTL32(x4, 11); \
xg = T32(x5 + xg); \
x5 = ROTL32(x5, 11); \
xj = T32(x6 + xj); \
x6 = ROTL32(x6, 11); \
xi = T32(x7 + xi); \
x7 = ROTL32(x7, 11); \
xl = T32(x0 + xl); \
x0 = ROTL32(x0, 11); \
xk = T32(x1 + xk); \
x1 = ROTL32(x1, 11); \
xn = T32(x2 + xn); \
x2 = ROTL32(x2, 11); \
xm = T32(x3 + xm); \
x3 = ROTL32(x3, 11); \
xp = T32(xc + xp); \
xc = ROTL32(xc, 11); \
xo = T32(xd + xo); \
xd = ROTL32(xd, 11); \
xr = T32(xe + xr); \
xe = ROTL32(xe, 11); \
xq = T32(xf + xq); \
xf = ROTL32(xf, 11); \
xt = T32(x8 + xt); \
x8 = ROTL32(x8, 11); \
xs = T32(x9 + xs); \
x9 = ROTL32(x9, 11); \
xv = T32(xa + xv); \
xa = ROTL32(xa, 11); \
xu = T32(xb + xu); \
xb = ROTL32(xb, 11); \
x0 ^= xh; \
x1 ^= xg; \
x2 ^= xj; \
x3 ^= xi; \
x4 ^= xl; \
x5 ^= xk; \
x6 ^= xn; \
x7 ^= xm; \
x8 ^= xp; \
x9 ^= xo; \
xa ^= xr; \
xb ^= xq; \
xc ^= xt; \
xd ^= xs; \
xe ^= xv; \
xf ^= xu; \
} while (0)
/*
* There is no need to unroll all 16 rounds. The word-swapping permutation
* is an involution, so we need to unroll an even number of rounds. On
* "big" systems, unrolling 4 rounds yields about 97% of the speed
* achieved with full unrolling; and it keeps the code more compact
* for small architectures.
*/
#if SPH_CUBEHASH_UNROLL == 2
#define SIXTEEN_ROUNDS do { \
int j; \
for (j = 0; j < 8; j ++) { \
ROUND_EVEN; \
ROUND_ODD; \
} \
} while (0)
#elif SPH_CUBEHASH_UNROLL == 4
#define SIXTEEN_ROUNDS do { \
int j; \
for (j = 0; j < 4; j ++) { \
ROUND_EVEN; \
ROUND_ODD; \
ROUND_EVEN; \
ROUND_ODD; \
} \
} while (0)
#elif SPH_CUBEHASH_UNROLL == 8
#define SIXTEEN_ROUNDS do { \
int j; \
for (j = 0; j < 2; j ++) { \
ROUND_EVEN; \
ROUND_ODD; \
ROUND_EVEN; \
ROUND_ODD; \
ROUND_EVEN; \
ROUND_ODD; \
ROUND_EVEN; \
ROUND_ODD; \
} \
} while (0)
#else
#define SIXTEEN_ROUNDS do { \
ROUND_EVEN; \
ROUND_ODD; \
ROUND_EVEN; \
ROUND_ODD; \
ROUND_EVEN; \
ROUND_ODD; \
ROUND_EVEN; \
ROUND_ODD; \
ROUND_EVEN; \
ROUND_ODD; \
ROUND_EVEN; \
ROUND_ODD; \
ROUND_EVEN; \
ROUND_ODD; \
ROUND_EVEN; \
ROUND_ODD; \
} while (0)
#endif
static void
cubehash_init(sph_cubehash_context *sc, const sph_u32 *iv)
{
memcpy(sc->state, iv, sizeof sc->state);
sc->ptr = 0;
}
static void
cubehash_core(sph_cubehash_context *sc, const void *data, size_t len)
{
unsigned char *buf;
size_t ptr;
DECL_STATE
buf = sc->buf;
ptr = sc->ptr;
if (len < (sizeof sc->buf) - ptr) {
memcpy(buf + ptr, data, len);
ptr += len;
sc->ptr = ptr;
return;
}
READ_STATE(sc);
while (len > 0) {
size_t clen;
clen = (sizeof sc->buf) - ptr;
if (clen > len)
clen = len;
memcpy(buf + ptr, data, clen);
ptr += clen;
data = (const unsigned char *)data + clen;
len -= clen;
if (ptr == sizeof sc->buf) {
INPUT_BLOCK;
SIXTEEN_ROUNDS;
ptr = 0;
}
}
WRITE_STATE(sc);
sc->ptr = ptr;
}
static void
cubehash_close(sph_cubehash_context *sc, unsigned ub, unsigned n,
void *dst, size_t out_size_w32)
{
unsigned char *buf, *out;
size_t ptr;
unsigned z;
int i;
DECL_STATE
buf = sc->buf;
ptr = sc->ptr;
z = 0x80 >> n;
buf[ptr ++] = ((ub & -z) | z) & 0xFF;
memset(buf + ptr, 0, (sizeof sc->buf) - ptr);
READ_STATE(sc);
INPUT_BLOCK;
for (i = 0; i < 11; i ++) {
SIXTEEN_ROUNDS;
if (i == 0)
xv ^= SPH_C32(1);
}
WRITE_STATE(sc);
out = (unsigned char*) dst;
for (z = 0; z < out_size_w32; z ++)
sph_enc32le(out + (z << 2), sc->state[z]);
}
/* see sph_cubehash.h */
void
sph_cubehash224_init(void *cc)
{
cubehash_init((sph_cubehash_context*) cc, IV224);
}
/* see sph_cubehash.h */
void
sph_cubehash224(void *cc, const void *data, size_t len)
{
cubehash_core((sph_cubehash_context*) cc, data, len);
}
/* see sph_cubehash.h */
void
sph_cubehash224_close(void *cc, void *dst)
{
sph_cubehash224_addbits_and_close((sph_cubehash_context*) cc, 0, 0, dst);
}
/* see sph_cubehash.h */
void
sph_cubehash224_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst)
{
cubehash_close((sph_cubehash_context*) cc, ub, n, dst, 7);
sph_cubehash224_init(cc);
}
/* see sph_cubehash.h */
void
sph_cubehash256_init(void *cc)
{
cubehash_init((sph_cubehash_context*) cc, IV256);
}
/* see sph_cubehash.h */
void
sph_cubehash256(void *cc, const void *data, size_t len)
{
cubehash_core((sph_cubehash_context*) cc, data, len);
}
/* see sph_cubehash.h */
void
sph_cubehash256_close(void *cc, void *dst)
{
sph_cubehash256_addbits_and_close((sph_cubehash_context*) cc, 0, 0, dst);
}
/* see sph_cubehash.h */
void
sph_cubehash256_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst)
{
cubehash_close((sph_cubehash_context*) cc, ub, n, dst, 8);
sph_cubehash256_init(cc);
}
/* see sph_cubehash.h */
void
sph_cubehash384_init(void *cc)
{
cubehash_init((sph_cubehash_context*) cc, IV384);
}
/* see sph_cubehash.h */
void
sph_cubehash384(void *cc, const void *data, size_t len)
{
cubehash_core((sph_cubehash_context*) cc, data, len);
}
/* see sph_cubehash.h */
void
sph_cubehash384_close(void *cc, void *dst)
{
sph_cubehash384_addbits_and_close((sph_cubehash_context*) cc, 0, 0, dst);
}
/* see sph_cubehash.h */
void
sph_cubehash384_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst)
{
cubehash_close((sph_cubehash_context*) cc, ub, n, dst, 12);
sph_cubehash384_init(cc);
}
/* see sph_cubehash.h */
void
sph_cubehash512_init(void *cc)
{
cubehash_init((sph_cubehash_context*) cc, IV512);
}
/* see sph_cubehash.h */
void
sph_cubehash512(void *cc, const void *data, size_t len)
{
cubehash_core((sph_cubehash_context*) cc, data, len);
}
/* see sph_cubehash.h */
void
sph_cubehash512_close(void *cc, void *dst)
{
sph_cubehash512_addbits_and_close((sph_cubehash_context*) cc, 0, 0, dst);
}
/* see sph_cubehash.h */
void
sph_cubehash512_addbits_and_close(void *cc, unsigned ub, unsigned n, void *dst)
{
cubehash_close((sph_cubehash_context*) cc, ub, n, dst, 16);
sph_cubehash512_init(cc);
}
#ifdef __cplusplus
}
#endif

1031
src/crypto/echo.cpp Normal file
View file

File diff suppressed because it is too large Load diff

21
src/crypto/hkdf_sha256_32.cpp Normal file
View file

@ -0,0 +1,21 @@
// Copyright (c) 2018-2019 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#include <crypto/hkdf_sha256_32.h>
#include <assert.h>
#include <string.h>
CHKDF_HMAC_SHA256_L32::CHKDF_HMAC_SHA256_L32(const unsigned char* ikm, size_t ikmlen, const std::string& salt)
{
CHMAC_SHA256((const unsigned char*)salt.data(), salt.size()).Write(ikm, ikmlen).Finalize(m_prk);
}
void CHKDF_HMAC_SHA256_L32::Expand32(const std::string& info, unsigned char hash[OUTPUT_SIZE])
{
// expand a 32byte key (single round)
assert(info.size() <= 128);
static const unsigned char one[1] = {1};
CHMAC_SHA256(m_prk, 32).Write((const unsigned char*)info.data(), info.size()).Write(one, 1).Finalize(hash);
}

25
src/crypto/hkdf_sha256_32.h Normal file
View file

@ -0,0 +1,25 @@
// Copyright (c) 2018 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#ifndef DIGIBYTE_CRYPTO_HKDF_SHA256_32_H
#define DIGIBYTE_CRYPTO_HKDF_SHA256_32_H
#include <crypto/hmac_sha256.h>
#include <stdint.h>
#include <stdlib.h>
/** A rfc5869 HKDF implementation with HMAC_SHA256 and fixed key output length of 32 bytes (L=32) */
class CHKDF_HMAC_SHA256_L32
{
private:
unsigned char m_prk[32];
static const size_t OUTPUT_SIZE = 32;
public:
CHKDF_HMAC_SHA256_L32(const unsigned char* ikm, size_t ikmlen, const std::string& salt);
void Expand32(const std::string& info, unsigned char hash[OUTPUT_SIZE]);
};
#endif // DIGIBYTE_CRYPTO_HKDF_SHA256_32_H

40
src/crypto/hmac_sha256.cpp Normal file
View file

@ -0,0 +1,40 @@
// Copyright (c) 2009-2020 The Bitcoin Core developers
// Copyright (c) 2014-2020 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#include <crypto/hmac_sha256.h>
#include <string.h>
CHMAC_SHA256::CHMAC_SHA256(const unsigned char* key, size_t keylen)
{
unsigned char rkey[64];
if (keylen <= 64) {
memcpy(rkey, key, keylen);
memset(rkey + keylen, 0, 64 - keylen);
} else {
CSHA256().Write(key, keylen).Finalize(rkey);
memset(rkey + 32, 0, 32);
}
for (int n = 0; n < 64; n++)
rkey[n] ^= 0x5c;
outer.Write(rkey, 64);
for (int n = 0; n < 64; n++)
rkey[n] ^= 0x5c ^ 0x36;
inner.Write(rkey, 64);
}
void CHMAC_SHA256::Reset() {
inner.Reset();
outer.Reset();
}
void CHMAC_SHA256::Finalize(unsigned char hash[OUTPUT_SIZE])
{
unsigned char temp[32];
inner.Finalize(temp);
outer.Write(temp, 32).Finalize(hash);
}

34
src/crypto/hmac_sha256.h Normal file
View file

@ -0,0 +1,34 @@
// Copyright (c) 2009-2020 The Bitcoin Core developers
// Copyright (c) 2014-2020 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#ifndef DIGIBYTE_CRYPTO_HMAC_SHA256_H
#define DIGIBYTE_CRYPTO_HMAC_SHA256_H
#include <crypto/sha256.h>
#include <stdint.h>
#include <stdlib.h>
/** A hasher class for HMAC-SHA-256. */
class CHMAC_SHA256
{
private:
CSHA256 outer;
CSHA256 inner;
public:
static const size_t OUTPUT_SIZE = 32;
CHMAC_SHA256(const unsigned char* key, size_t keylen);
CHMAC_SHA256& Write(const unsigned char* data, size_t len)
{
inner.Write(data, len);
return *this;
}
void Reset();
void Finalize(unsigned char hash[OUTPUT_SIZE]);
};
#endif // DIGIBYTE_CRYPTO_HMAC_SHA256_H

35
src/crypto/hmac_sha512.cpp Normal file
View file

@ -0,0 +1,35 @@
// Copyright (c) 2009-2020 The Bitcoin Core developers
// Copyright (c) 2014-2020 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#include <crypto/hmac_sha512.h>
#include <string.h>
CHMAC_SHA512::CHMAC_SHA512(const unsigned char* key, size_t keylen)
{
unsigned char rkey[128];
if (keylen <= 128) {
memcpy(rkey, key, keylen);
memset(rkey + keylen, 0, 128 - keylen);
} else {
CSHA512().Write(key, keylen).Finalize(rkey);
memset(rkey + 64, 0, 64);
}
for (int n = 0; n < 128; n++)
rkey[n] ^= 0x5c;
outer.Write(rkey, 128);
for (int n = 0; n < 128; n++)
rkey[n] ^= 0x5c ^ 0x36;
inner.Write(rkey, 128);
}
void CHMAC_SHA512::Finalize(unsigned char hash[OUTPUT_SIZE])
{
unsigned char temp[64];
inner.Finalize(temp);
outer.Write(temp, 64).Finalize(hash);
}

33
src/crypto/hmac_sha512.h Normal file
View file

@ -0,0 +1,33 @@
// Copyright (c) 2009-2020 The Bitcoin Core developers
// Copyright (c) 2014-2020 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#ifndef DIGIBYTE_CRYPTO_HMAC_SHA512_H
#define DIGIBYTE_CRYPTO_HMAC_SHA512_H
#include <crypto/sha512.h>
#include <stdint.h>
#include <stdlib.h>
/** A hasher class for HMAC-SHA-512. */
class CHMAC_SHA512
{
private:
CSHA512 outer;
CSHA512 inner;
public:
static const size_t OUTPUT_SIZE = 64;
CHMAC_SHA512(const unsigned char* key, size_t keylen);
CHMAC_SHA512& Write(const unsigned char* data, size_t len)
{
inner.Write(data, len);
return *this;
}
void Finalize(unsigned char hash[OUTPUT_SIZE]);
};
#endif // DIGIBYTE_CRYPTO_HMAC_SHA512_H

1116
src/crypto/jh.cpp Normal file
View file

File diff suppressed because it is too large Load diff

1824
src/crypto/keccak.cpp Normal file
View file

File diff suppressed because it is too large Load diff

1428
src/crypto/luffa.cpp Normal file
View file

File diff suppressed because it is too large Load diff

346
src/crypto/muhash.cpp Normal file
View file

@ -0,0 +1,346 @@
// Copyright (c) 2017-2020 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#include <crypto/muhash.h>
#include <crypto/chacha20.h>
#include <crypto/common.h>
#include <hash.h>
#include <cassert>
#include <cstdio>
#include <limits>
namespace {
using limb_t = Num3072::limb_t;
using double_limb_t = Num3072::double_limb_t;
constexpr int LIMB_SIZE = Num3072::LIMB_SIZE;
/** 2^3072 - 1103717, the largest 3072-bit safe prime number, is used as the modulus. */
constexpr limb_t MAX_PRIME_DIFF = 1103717;
/** Extract the lowest limb of [c0,c1,c2] into n, and left shift the number by 1 limb. */
inline void extract3(limb_t& c0, limb_t& c1, limb_t& c2, limb_t& n)
{
n = c0;
c0 = c1;
c1 = c2;
c2 = 0;
}
/** [c0,c1] = a * b */
inline void mul(limb_t& c0, limb_t& c1, const limb_t& a, const limb_t& b)
{
double_limb_t t = (double_limb_t)a * b;
c1 = t >> LIMB_SIZE;
c0 = t;
}
/* [c0,c1,c2] += n * [d0,d1,d2]. c2 is 0 initially */
inline void mulnadd3(limb_t& c0, limb_t& c1, limb_t& c2, limb_t& d0, limb_t& d1, limb_t& d2, const limb_t& n)
{
double_limb_t t = (double_limb_t)d0 * n + c0;
c0 = t;
t >>= LIMB_SIZE;
t += (double_limb_t)d1 * n + c1;
c1 = t;
t >>= LIMB_SIZE;
c2 = t + d2 * n;
}
/* [c0,c1] *= n */
inline void muln2(limb_t& c0, limb_t& c1, const limb_t& n)
{
double_limb_t t = (double_limb_t)c0 * n;
c0 = t;
t >>= LIMB_SIZE;
t += (double_limb_t)c1 * n;
c1 = t;
}
/** [c0,c1,c2] += a * b */
inline void muladd3(limb_t& c0, limb_t& c1, limb_t& c2, const limb_t& a, const limb_t& b)
{
double_limb_t t = (double_limb_t)a * b;
limb_t th = t >> LIMB_SIZE;
limb_t tl = t;
c0 += tl;
th += (c0 < tl) ? 1 : 0;
c1 += th;
c2 += (c1 < th) ? 1 : 0;
}
/** [c0,c1,c2] += 2 * a * b */
inline void muldbladd3(limb_t& c0, limb_t& c1, limb_t& c2, const limb_t& a, const limb_t& b)
{
double_limb_t t = (double_limb_t)a * b;
limb_t th = t >> LIMB_SIZE;
limb_t tl = t;
c0 += tl;
limb_t tt = th + ((c0 < tl) ? 1 : 0);
c1 += tt;
c2 += (c1 < tt) ? 1 : 0;
c0 += tl;
th += (c0 < tl) ? 1 : 0;
c1 += th;
c2 += (c1 < th) ? 1 : 0;
}
/**
* Add limb a to [c0,c1]: [c0,c1] += a. Then extract the lowest
* limb of [c0,c1] into n, and left shift the number by 1 limb.
* */
inline void addnextract2(limb_t& c0, limb_t& c1, const limb_t& a, limb_t& n)
{
limb_t c2 = 0;
// add
c0 += a;
if (c0 < a) {
c1 += 1;
// Handle case when c1 has overflown
if (c1 == 0)
c2 = 1;
}
// extract
n = c0;
c0 = c1;
c1 = c2;
}
/** in_out = in_out^(2^sq) * mul */
inline void square_n_mul(Num3072& in_out, const int sq, const Num3072& mul)
{
for (int j = 0; j < sq; ++j) in_out.Square();
in_out.Multiply(mul);
}
} // namespace
/** Indicates whether d is larger than the modulus. */
bool Num3072::IsOverflow() const
{
if (this->limbs[0] <= std::numeric_limits<limb_t>::max() - MAX_PRIME_DIFF) return false;
for (int i = 1; i < LIMBS; ++i) {
if (this->limbs[i] != std::numeric_limits<limb_t>::max()) return false;
}
return true;
}
void Num3072::FullReduce()
{
limb_t c0 = MAX_PRIME_DIFF;
limb_t c1 = 0;
for (int i = 0; i < LIMBS; ++i) {
addnextract2(c0, c1, this->limbs[i], this->limbs[i]);
}
}
Num3072 Num3072::GetInverse() const
{
// For fast exponentiation a sliding window exponentiation with repunit
// precomputation is utilized. See "Fast Point Decompression for Standard
// Elliptic Curves" (Brumley, Järvinen, 2008).
Num3072 p[12]; // p[i] = a^(2^(2^i)-1)
Num3072 out;
p[0] = *this;
for (int i = 0; i < 11; ++i) {
p[i + 1] = p[i];
for (int j = 0; j < (1 << i); ++j) p[i + 1].Square();
p[i + 1].Multiply(p[i]);
}
out = p[11];
square_n_mul(out, 512, p[9]);
square_n_mul(out, 256, p[8]);
square_n_mul(out, 128, p[7]);
square_n_mul(out, 64, p[6]);
square_n_mul(out, 32, p[5]);
square_n_mul(out, 8, p[3]);
square_n_mul(out, 2, p[1]);
square_n_mul(out, 1, p[0]);
square_n_mul(out, 5, p[2]);
square_n_mul(out, 3, p[0]);
square_n_mul(out, 2, p[0]);
square_n_mul(out, 4, p[0]);
square_n_mul(out, 4, p[1]);
square_n_mul(out, 3, p[0]);
return out;
}
void Num3072::Multiply(const Num3072& a)
{
limb_t c0 = 0, c1 = 0, c2 = 0;
Num3072 tmp;
/* Compute limbs 0..N-2 of this*a into tmp, including one reduction. */
for (int j = 0; j < LIMBS - 1; ++j) {
limb_t d0 = 0, d1 = 0, d2 = 0;
mul(d0, d1, this->limbs[1 + j], a.limbs[LIMBS + j - (1 + j)]);
for (int i = 2 + j; i < LIMBS; ++i) muladd3(d0, d1, d2, this->limbs[i], a.limbs[LIMBS + j - i]);
mulnadd3(c0, c1, c2, d0, d1, d2, MAX_PRIME_DIFF);
for (int i = 0; i < j + 1; ++i) muladd3(c0, c1, c2, this->limbs[i], a.limbs[j - i]);
extract3(c0, c1, c2, tmp.limbs[j]);
}
/* Compute limb N-1 of a*b into tmp. */
assert(c2 == 0);
for (int i = 0; i < LIMBS; ++i) muladd3(c0, c1, c2, this->limbs[i], a.limbs[LIMBS - 1 - i]);
extract3(c0, c1, c2, tmp.limbs[LIMBS - 1]);
/* Perform a second reduction. */
muln2(c0, c1, MAX_PRIME_DIFF);
for (int j = 0; j < LIMBS; ++j) {
addnextract2(c0, c1, tmp.limbs[j], this->limbs[j]);
}
assert(c1 == 0);
assert(c0 == 0 || c0 == 1);
/* Perform up to two more reductions if the internal state has already
* overflown the MAX of Num3072 or if it is larger than the modulus or
* if both are the case.
* */
if (this->IsOverflow()) this->FullReduce();
if (c0) this->FullReduce();
}
void Num3072::Square()
{
limb_t c0 = 0, c1 = 0, c2 = 0;
Num3072 tmp;
/* Compute limbs 0..N-2 of this*this into tmp, including one reduction. */
for (int j = 0; j < LIMBS - 1; ++j) {
limb_t d0 = 0, d1 = 0, d2 = 0;
for (int i = 0; i < (LIMBS - 1 - j) / 2; ++i) muldbladd3(d0, d1, d2, this->limbs[i + j + 1], this->limbs[LIMBS - 1 - i]);
if ((j + 1) & 1) muladd3(d0, d1, d2, this->limbs[(LIMBS - 1 - j) / 2 + j + 1], this->limbs[LIMBS - 1 - (LIMBS - 1 - j) / 2]);
mulnadd3(c0, c1, c2, d0, d1, d2, MAX_PRIME_DIFF);
for (int i = 0; i < (j + 1) / 2; ++i) muldbladd3(c0, c1, c2, this->limbs[i], this->limbs[j - i]);
if ((j + 1) & 1) muladd3(c0, c1, c2, this->limbs[(j + 1) / 2], this->limbs[j - (j + 1) / 2]);
extract3(c0, c1, c2, tmp.limbs[j]);
}
assert(c2 == 0);
for (int i = 0; i < LIMBS / 2; ++i) muldbladd3(c0, c1, c2, this->limbs[i], this->limbs[LIMBS - 1 - i]);
extract3(c0, c1, c2, tmp.limbs[LIMBS - 1]);
/* Perform a second reduction. */
muln2(c0, c1, MAX_PRIME_DIFF);
for (int j = 0; j < LIMBS; ++j) {
addnextract2(c0, c1, tmp.limbs[j], this->limbs[j]);
}
assert(c1 == 0);
assert(c0 == 0 || c0 == 1);
/* Perform up to two more reductions if the internal state has already
* overflown the MAX of Num3072 or if it is larger than the modulus or
* if both are the case.
* */
if (this->IsOverflow()) this->FullReduce();
if (c0) this->FullReduce();
}
void Num3072::SetToOne()
{
this->limbs[0] = 1;
for (int i = 1; i < LIMBS; ++i) this->limbs[i] = 0;
}
void Num3072::Divide(const Num3072& a)
{
if (this->IsOverflow()) this->FullReduce();
Num3072 inv{};
if (a.IsOverflow()) {
Num3072 b = a;
b.FullReduce();
inv = b.GetInverse();
} else {
inv = a.GetInverse();
}
this->Multiply(inv);
if (this->IsOverflow()) this->FullReduce();
}
Num3072::Num3072(const unsigned char (&data)[BYTE_SIZE]) {
for (int i = 0; i < LIMBS; ++i) {
if (sizeof(limb_t) == 4) {
this->limbs[i] = ReadLE32(data + 4 * i);
} else if (sizeof(limb_t) == 8) {
this->limbs[i] = ReadLE64(data + 8 * i);
}
}
}
void Num3072::ToBytes(unsigned char (&out)[BYTE_SIZE]) {
for (int i = 0; i < LIMBS; ++i) {
if (sizeof(limb_t) == 4) {
WriteLE32(out + i * 4, this->limbs[i]);
} else if (sizeof(limb_t) == 8) {
WriteLE64(out + i * 8, this->limbs[i]);
}
}
}
Num3072 MuHash3072::ToNum3072(Span<const unsigned char> in) {
unsigned char tmp[Num3072::BYTE_SIZE];
uint256 hashed_in = (CHashWriter(SER_DISK, 0) << in).GetSHA256();
ChaCha20(hashed_in.data(), hashed_in.size()).Keystream(tmp, Num3072::BYTE_SIZE);
Num3072 out{tmp};
return out;
}
MuHash3072::MuHash3072(Span<const unsigned char> in) noexcept
{
m_numerator = ToNum3072(in);
}
void MuHash3072::Finalize(uint256& out) noexcept
{
m_numerator.Divide(m_denominator);
m_denominator.SetToOne(); // Needed to keep the MuHash object valid
unsigned char data[Num3072::BYTE_SIZE];
m_numerator.ToBytes(data);
out = (CHashWriter(SER_DISK, 0) << data).GetSHA256();
}
MuHash3072& MuHash3072::operator*=(const MuHash3072& mul) noexcept
{
m_numerator.Multiply(mul.m_numerator);
m_denominator.Multiply(mul.m_denominator);
return *this;
}
MuHash3072& MuHash3072::operator/=(const MuHash3072& div) noexcept
{
m_numerator.Multiply(div.m_denominator);
m_denominator.Multiply(div.m_numerator);
return *this;
}
MuHash3072& MuHash3072::Insert(Span<const unsigned char> in) noexcept {
m_numerator.Multiply(ToNum3072(in));
return *this;
}
MuHash3072& MuHash3072::Remove(Span<const unsigned char> in) noexcept {
m_denominator.Multiply(ToNum3072(in));
return *this;
}

131
src/crypto/muhash.h Normal file
View file

@ -0,0 +1,131 @@
// Copyright (c) 2017-2020 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#ifndef DIGIBYTE_CRYPTO_MUHASH_H
#define DIGIBYTE_CRYPTO_MUHASH_H
#if defined(HAVE_CONFIG_H)
#include <config/vexta-config.h>
#endif
#include <serialize.h>
#include <uint256.h>
#include <stdint.h>
class Num3072
{
private:
void FullReduce();
bool IsOverflow() const;
Num3072 GetInverse() const;
public:
static constexpr size_t BYTE_SIZE = 384;
#ifdef HAVE___INT128
typedef unsigned __int128 double_limb_t;
typedef uint64_t limb_t;
static constexpr int LIMBS = 48;
static constexpr int LIMB_SIZE = 64;
#else
typedef uint64_t double_limb_t;
typedef uint32_t limb_t;
static constexpr int LIMBS = 96;
static constexpr int LIMB_SIZE = 32;
#endif
limb_t limbs[LIMBS];
// Sanity check for Num3072 constants
static_assert(LIMB_SIZE * LIMBS == 3072, "Num3072 isn't 3072 bits");
static_assert(sizeof(double_limb_t) == sizeof(limb_t) * 2, "bad size for double_limb_t");
static_assert(sizeof(limb_t) * 8 == LIMB_SIZE, "LIMB_SIZE is incorrect");
// Hard coded values in MuHash3072 constructor and Finalize
static_assert(sizeof(limb_t) == 4 || sizeof(limb_t) == 8, "bad size for limb_t");
void Multiply(const Num3072& a);
void Divide(const Num3072& a);
void SetToOne();
void Square();
void ToBytes(unsigned char (&out)[BYTE_SIZE]);
Num3072() { this->SetToOne(); };
Num3072(const unsigned char (&data)[BYTE_SIZE]);
SERIALIZE_METHODS(Num3072, obj)
{
for (auto& limb : obj.limbs) {
READWRITE(limb);
}
}
};
/** A class representing MuHash sets
*
* MuHash is a hashing algorithm that supports adding set elements in any
* order but also deleting in any order. As a result, it can maintain a
* running sum for a set of data as a whole, and add/remove when data
* is added to or removed from it. A downside of MuHash is that computing
* an inverse is relatively expensive. This is solved by representing
* the running value as a fraction, and multiplying added elements into
* the numerator and removed elements into the denominator. Only when the
* final hash is desired, a single modular inverse and multiplication is
* needed to combine the two. The combination is also run on serialization
* to allow for space-efficient storage on disk.
*
* As the update operations are also associative, H(a)+H(b)+H(c)+H(d) can
* in fact be computed as (H(a)+H(b)) + (H(c)+H(d)). This implies that
* all of this is perfectly parallellizable: each thread can process an
* arbitrary subset of the update operations, allowing them to be
* efficiently combined later.
*
* MuHash does not support checking if an element is already part of the
* set. That is why this class does not enforce the use of a set as the
* data it represents because there is no efficient way to do so.
* It is possible to add elements more than once and also to remove
* elements that have not been added before. However, this implementation
* is intended to represent a set of elements.
*
* See also https://cseweb.ucsd.edu/~mihir/papers/inchash.pdf and
* https://lists.linuxfoundation.org/pipermail/digibyte-dev/2017-May/014337.html.
*/
class MuHash3072
{
private:
Num3072 m_numerator;
Num3072 m_denominator;
Num3072 ToNum3072(Span<const unsigned char> in);
public:
/* The empty set. */
MuHash3072() noexcept {};
/* A singleton with variable sized data in it. */
explicit MuHash3072(Span<const unsigned char> in) noexcept;
/* Insert a single piece of data into the set. */
MuHash3072& Insert(Span<const unsigned char> in) noexcept;
/* Remove a single piece of data from the set. */
MuHash3072& Remove(Span<const unsigned char> in) noexcept;
/* Multiply (resulting in a hash for the union of the sets) */
MuHash3072& operator*=(const MuHash3072& mul) noexcept;
/* Divide (resulting in a hash for the difference of the sets) */
MuHash3072& operator/=(const MuHash3072& div) noexcept;
/* Finalize into a 32-byte hash. Does not change this object's value. */
void Finalize(uint256& out) noexcept;
SERIALIZE_METHODS(MuHash3072, obj)
{
READWRITE(obj.m_numerator);
READWRITE(obj.m_denominator);
}
};
#endif // DIGIBYTE_CRYPTO_MUHASH_H

141
src/crypto/poly1305.cpp Normal file
View file

@ -0,0 +1,141 @@
// Copyright (c) 2019 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
// Based on the public domain implementation by Andrew Moon
// poly1305-donna-unrolled.c from https://github.com/floodyberry/poly1305-donna
#include <crypto/common.h>
#include <crypto/poly1305.h>
#include <string.h>
#define mul32x32_64(a,b) ((uint64_t)(a) * (b))
void poly1305_auth(unsigned char out[POLY1305_TAGLEN], const unsigned char *m, size_t inlen, const unsigned char key[POLY1305_KEYLEN]) {
uint32_t t0,t1,t2,t3;
uint32_t h0,h1,h2,h3,h4;
uint32_t r0,r1,r2,r3,r4;
uint32_t s1,s2,s3,s4;
uint32_t b, nb;
size_t j;
uint64_t t[5];
uint64_t f0,f1,f2,f3;
uint64_t g0,g1,g2,g3,g4;
uint64_t c;
unsigned char mp[16];
/* clamp key */
t0 = ReadLE32(key+0);
t1 = ReadLE32(key+4);
t2 = ReadLE32(key+8);
t3 = ReadLE32(key+12);
/* precompute multipliers */
r0 = t0 & 0x3ffffff; t0 >>= 26; t0 |= t1 << 6;
r1 = t0 & 0x3ffff03; t1 >>= 20; t1 |= t2 << 12;
r2 = t1 & 0x3ffc0ff; t2 >>= 14; t2 |= t3 << 18;
r3 = t2 & 0x3f03fff; t3 >>= 8;
r4 = t3 & 0x00fffff;
s1 = r1 * 5;
s2 = r2 * 5;
s3 = r3 * 5;
s4 = r4 * 5;
/* init state */
h0 = 0;
h1 = 0;
h2 = 0;
h3 = 0;
h4 = 0;
/* full blocks */
if (inlen < 16) goto poly1305_donna_atmost15bytes;
poly1305_donna_16bytes:
m += 16;
inlen -= 16;
t0 = ReadLE32(m-16);
t1 = ReadLE32(m-12);
t2 = ReadLE32(m-8);
t3 = ReadLE32(m-4);
h0 += t0 & 0x3ffffff;
h1 += ((((uint64_t)t1 << 32) | t0) >> 26) & 0x3ffffff;
h2 += ((((uint64_t)t2 << 32) | t1) >> 20) & 0x3ffffff;
h3 += ((((uint64_t)t3 << 32) | t2) >> 14) & 0x3ffffff;
h4 += (t3 >> 8) | (1 << 24);
poly1305_donna_mul:
t[0] = mul32x32_64(h0,r0) + mul32x32_64(h1,s4) + mul32x32_64(h2,s3) + mul32x32_64(h3,s2) + mul32x32_64(h4,s1);
t[1] = mul32x32_64(h0,r1) + mul32x32_64(h1,r0) + mul32x32_64(h2,s4) + mul32x32_64(h3,s3) + mul32x32_64(h4,s2);
t[2] = mul32x32_64(h0,r2) + mul32x32_64(h1,r1) + mul32x32_64(h2,r0) + mul32x32_64(h3,s4) + mul32x32_64(h4,s3);
t[3] = mul32x32_64(h0,r3) + mul32x32_64(h1,r2) + mul32x32_64(h2,r1) + mul32x32_64(h3,r0) + mul32x32_64(h4,s4);
t[4] = mul32x32_64(h0,r4) + mul32x32_64(h1,r3) + mul32x32_64(h2,r2) + mul32x32_64(h3,r1) + mul32x32_64(h4,r0);
h0 = (uint32_t)t[0] & 0x3ffffff; c = (t[0] >> 26);
t[1] += c; h1 = (uint32_t)t[1] & 0x3ffffff; b = (uint32_t)(t[1] >> 26);
t[2] += b; h2 = (uint32_t)t[2] & 0x3ffffff; b = (uint32_t)(t[2] >> 26);
t[3] += b; h3 = (uint32_t)t[3] & 0x3ffffff; b = (uint32_t)(t[3] >> 26);
t[4] += b; h4 = (uint32_t)t[4] & 0x3ffffff; b = (uint32_t)(t[4] >> 26);
h0 += b * 5;
if (inlen >= 16) goto poly1305_donna_16bytes;
/* final bytes */
poly1305_donna_atmost15bytes:
if (!inlen) goto poly1305_donna_finish;
for (j = 0; j < inlen; j++) mp[j] = m[j];
mp[j++] = 1;
for (; j < 16; j++) mp[j] = 0;
inlen = 0;
t0 = ReadLE32(mp+0);
t1 = ReadLE32(mp+4);
t2 = ReadLE32(mp+8);
t3 = ReadLE32(mp+12);
h0 += t0 & 0x3ffffff;
h1 += ((((uint64_t)t1 << 32) | t0) >> 26) & 0x3ffffff;
h2 += ((((uint64_t)t2 << 32) | t1) >> 20) & 0x3ffffff;
h3 += ((((uint64_t)t3 << 32) | t2) >> 14) & 0x3ffffff;
h4 += (t3 >> 8);
goto poly1305_donna_mul;
poly1305_donna_finish:
b = h0 >> 26; h0 = h0 & 0x3ffffff;
h1 += b; b = h1 >> 26; h1 = h1 & 0x3ffffff;
h2 += b; b = h2 >> 26; h2 = h2 & 0x3ffffff;
h3 += b; b = h3 >> 26; h3 = h3 & 0x3ffffff;
h4 += b; b = h4 >> 26; h4 = h4 & 0x3ffffff;
h0 += b * 5; b = h0 >> 26; h0 = h0 & 0x3ffffff;
h1 += b;
g0 = h0 + 5; b = g0 >> 26; g0 &= 0x3ffffff;
g1 = h1 + b; b = g1 >> 26; g1 &= 0x3ffffff;
g2 = h2 + b; b = g2 >> 26; g2 &= 0x3ffffff;
g3 = h3 + b; b = g3 >> 26; g3 &= 0x3ffffff;
g4 = h4 + b - (1 << 26);
b = (g4 >> 31) - 1;
nb = ~b;
h0 = (h0 & nb) | (g0 & b);
h1 = (h1 & nb) | (g1 & b);
h2 = (h2 & nb) | (g2 & b);
h3 = (h3 & nb) | (g3 & b);
h4 = (h4 & nb) | (g4 & b);
f0 = ((h0 ) | (h1 << 26)) + (uint64_t)ReadLE32(&key[16]);
f1 = ((h1 >> 6) | (h2 << 20)) + (uint64_t)ReadLE32(&key[20]);
f2 = ((h2 >> 12) | (h3 << 14)) + (uint64_t)ReadLE32(&key[24]);
f3 = ((h3 >> 18) | (h4 << 8)) + (uint64_t)ReadLE32(&key[28]);
WriteLE32(&out[ 0], f0); f1 += (f0 >> 32);
WriteLE32(&out[ 4], f1); f2 += (f1 >> 32);
WriteLE32(&out[ 8], f2); f3 += (f2 >> 32);
WriteLE32(&out[12], f3);
}

17
src/crypto/poly1305.h Normal file
View file

@ -0,0 +1,17 @@
// Copyright (c) 2019 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#ifndef DIGIBYTE_CRYPTO_POLY1305_H
#define DIGIBYTE_CRYPTO_POLY1305_H
#include <stdint.h>
#include <stdlib.h>
#define POLY1305_KEYLEN 32
#define POLY1305_TAGLEN 16
void poly1305_auth(unsigned char out[POLY1305_TAGLEN], const unsigned char *m, size_t inlen,
const unsigned char key[POLY1305_KEYLEN]);
#endif // DIGIBYTE_CRYPTO_POLY1305_H

293
src/crypto/ripemd160.cpp Normal file
View file

@ -0,0 +1,293 @@
// Copyright (c) 2009-2020 The Bitcoin Core developers
// Copyright (c) 2014-2020 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#include <crypto/ripemd160.h>
#include <crypto/common.h>
#include <string.h>
// Internal implementation code.
namespace
{
/// Internal RIPEMD-160 implementation.
namespace ripemd160
{
uint32_t inline f1(uint32_t x, uint32_t y, uint32_t z) { return x ^ y ^ z; }
uint32_t inline f2(uint32_t x, uint32_t y, uint32_t z) { return (x & y) | (~x & z); }
uint32_t inline f3(uint32_t x, uint32_t y, uint32_t z) { return (x | ~y) ^ z; }
uint32_t inline f4(uint32_t x, uint32_t y, uint32_t z) { return (x & z) | (y & ~z); }
uint32_t inline f5(uint32_t x, uint32_t y, uint32_t z) { return x ^ (y | ~z); }
/** Initialize RIPEMD-160 state. */
void inline Initialize(uint32_t* s)
{
s[0] = 0x67452301ul;
s[1] = 0xEFCDAB89ul;
s[2] = 0x98BADCFEul;
s[3] = 0x10325476ul;
s[4] = 0xC3D2E1F0ul;
}
uint32_t inline rol(uint32_t x, int i) { return (x << i) | (x >> (32 - i)); }
void inline Round(uint32_t& a, uint32_t b, uint32_t& c, uint32_t d, uint32_t e, uint32_t f, uint32_t x, uint32_t k, int r)
{
a = rol(a + f + x + k, r) + e;
c = rol(c, 10);
}
void inline R11(uint32_t& a, uint32_t b, uint32_t& c, uint32_t d, uint32_t e, uint32_t x, int r) { Round(a, b, c, d, e, f1(b, c, d), x, 0, r); }
void inline R21(uint32_t& a, uint32_t b, uint32_t& c, uint32_t d, uint32_t e, uint32_t x, int r) { Round(a, b, c, d, e, f2(b, c, d), x, 0x5A827999ul, r); }
void inline R31(uint32_t& a, uint32_t b, uint32_t& c, uint32_t d, uint32_t e, uint32_t x, int r) { Round(a, b, c, d, e, f3(b, c, d), x, 0x6ED9EBA1ul, r); }
void inline R41(uint32_t& a, uint32_t b, uint32_t& c, uint32_t d, uint32_t e, uint32_t x, int r) { Round(a, b, c, d, e, f4(b, c, d), x, 0x8F1BBCDCul, r); }
void inline R51(uint32_t& a, uint32_t b, uint32_t& c, uint32_t d, uint32_t e, uint32_t x, int r) { Round(a, b, c, d, e, f5(b, c, d), x, 0xA953FD4Eul, r); }
void inline R12(uint32_t& a, uint32_t b, uint32_t& c, uint32_t d, uint32_t e, uint32_t x, int r) { Round(a, b, c, d, e, f5(b, c, d), x, 0x50A28BE6ul, r); }
void inline R22(uint32_t& a, uint32_t b, uint32_t& c, uint32_t d, uint32_t e, uint32_t x, int r) { Round(a, b, c, d, e, f4(b, c, d), x, 0x5C4DD124ul, r); }
void inline R32(uint32_t& a, uint32_t b, uint32_t& c, uint32_t d, uint32_t e, uint32_t x, int r) { Round(a, b, c, d, e, f3(b, c, d), x, 0x6D703EF3ul, r); }
void inline R42(uint32_t& a, uint32_t b, uint32_t& c, uint32_t d, uint32_t e, uint32_t x, int r) { Round(a, b, c, d, e, f2(b, c, d), x, 0x7A6D76E9ul, r); }
void inline R52(uint32_t& a, uint32_t b, uint32_t& c, uint32_t d, uint32_t e, uint32_t x, int r) { Round(a, b, c, d, e, f1(b, c, d), x, 0, r); }
/** Perform a RIPEMD-160 transformation, processing a 64-byte chunk. */
void Transform(uint32_t* s, const unsigned char* chunk)
{
uint32_t a1 = s[0], b1 = s[1], c1 = s[2], d1 = s[3], e1 = s[4];
uint32_t a2 = a1, b2 = b1, c2 = c1, d2 = d1, e2 = e1;
uint32_t w0 = ReadLE32(chunk + 0), w1 = ReadLE32(chunk + 4), w2 = ReadLE32(chunk + 8), w3 = ReadLE32(chunk + 12);
uint32_t w4 = ReadLE32(chunk + 16), w5 = ReadLE32(chunk + 20), w6 = ReadLE32(chunk + 24), w7 = ReadLE32(chunk + 28);
uint32_t w8 = ReadLE32(chunk + 32), w9 = ReadLE32(chunk + 36), w10 = ReadLE32(chunk + 40), w11 = ReadLE32(chunk + 44);
uint32_t w12 = ReadLE32(chunk + 48), w13 = ReadLE32(chunk + 52), w14 = ReadLE32(chunk + 56), w15 = ReadLE32(chunk + 60);
R11(a1, b1, c1, d1, e1, w0, 11);
R12(a2, b2, c2, d2, e2, w5, 8);
R11(e1, a1, b1, c1, d1, w1, 14);
R12(e2, a2, b2, c2, d2, w14, 9);
R11(d1, e1, a1, b1, c1, w2, 15);
R12(d2, e2, a2, b2, c2, w7, 9);
R11(c1, d1, e1, a1, b1, w3, 12);
R12(c2, d2, e2, a2, b2, w0, 11);
R11(b1, c1, d1, e1, a1, w4, 5);
R12(b2, c2, d2, e2, a2, w9, 13);
R11(a1, b1, c1, d1, e1, w5, 8);
R12(a2, b2, c2, d2, e2, w2, 15);
R11(e1, a1, b1, c1, d1, w6, 7);
R12(e2, a2, b2, c2, d2, w11, 15);
R11(d1, e1, a1, b1, c1, w7, 9);
R12(d2, e2, a2, b2, c2, w4, 5);
R11(c1, d1, e1, a1, b1, w8, 11);
R12(c2, d2, e2, a2, b2, w13, 7);
R11(b1, c1, d1, e1, a1, w9, 13);
R12(b2, c2, d2, e2, a2, w6, 7);
R11(a1, b1, c1, d1, e1, w10, 14);
R12(a2, b2, c2, d2, e2, w15, 8);
R11(e1, a1, b1, c1, d1, w11, 15);
R12(e2, a2, b2, c2, d2, w8, 11);
R11(d1, e1, a1, b1, c1, w12, 6);
R12(d2, e2, a2, b2, c2, w1, 14);
R11(c1, d1, e1, a1, b1, w13, 7);
R12(c2, d2, e2, a2, b2, w10, 14);
R11(b1, c1, d1, e1, a1, w14, 9);
R12(b2, c2, d2, e2, a2, w3, 12);
R11(a1, b1, c1, d1, e1, w15, 8);
R12(a2, b2, c2, d2, e2, w12, 6);
R21(e1, a1, b1, c1, d1, w7, 7);
R22(e2, a2, b2, c2, d2, w6, 9);
R21(d1, e1, a1, b1, c1, w4, 6);
R22(d2, e2, a2, b2, c2, w11, 13);
R21(c1, d1, e1, a1, b1, w13, 8);
R22(c2, d2, e2, a2, b2, w3, 15);
R21(b1, c1, d1, e1, a1, w1, 13);
R22(b2, c2, d2, e2, a2, w7, 7);
R21(a1, b1, c1, d1, e1, w10, 11);
R22(a2, b2, c2, d2, e2, w0, 12);
R21(e1, a1, b1, c1, d1, w6, 9);
R22(e2, a2, b2, c2, d2, w13, 8);
R21(d1, e1, a1, b1, c1, w15, 7);
R22(d2, e2, a2, b2, c2, w5, 9);
R21(c1, d1, e1, a1, b1, w3, 15);
R22(c2, d2, e2, a2, b2, w10, 11);
R21(b1, c1, d1, e1, a1, w12, 7);
R22(b2, c2, d2, e2, a2, w14, 7);
R21(a1, b1, c1, d1, e1, w0, 12);
R22(a2, b2, c2, d2, e2, w15, 7);
R21(e1, a1, b1, c1, d1, w9, 15);
R22(e2, a2, b2, c2, d2, w8, 12);
R21(d1, e1, a1, b1, c1, w5, 9);
R22(d2, e2, a2, b2, c2, w12, 7);
R21(c1, d1, e1, a1, b1, w2, 11);
R22(c2, d2, e2, a2, b2, w4, 6);
R21(b1, c1, d1, e1, a1, w14, 7);
R22(b2, c2, d2, e2, a2, w9, 15);
R21(a1, b1, c1, d1, e1, w11, 13);
R22(a2, b2, c2, d2, e2, w1, 13);
R21(e1, a1, b1, c1, d1, w8, 12);
R22(e2, a2, b2, c2, d2, w2, 11);
R31(d1, e1, a1, b1, c1, w3, 11);
R32(d2, e2, a2, b2, c2, w15, 9);
R31(c1, d1, e1, a1, b1, w10, 13);
R32(c2, d2, e2, a2, b2, w5, 7);
R31(b1, c1, d1, e1, a1, w14, 6);
R32(b2, c2, d2, e2, a2, w1, 15);
R31(a1, b1, c1, d1, e1, w4, 7);
R32(a2, b2, c2, d2, e2, w3, 11);
R31(e1, a1, b1, c1, d1, w9, 14);
R32(e2, a2, b2, c2, d2, w7, 8);
R31(d1, e1, a1, b1, c1, w15, 9);
R32(d2, e2, a2, b2, c2, w14, 6);
R31(c1, d1, e1, a1, b1, w8, 13);
R32(c2, d2, e2, a2, b2, w6, 6);
R31(b1, c1, d1, e1, a1, w1, 15);
R32(b2, c2, d2, e2, a2, w9, 14);
R31(a1, b1, c1, d1, e1, w2, 14);
R32(a2, b2, c2, d2, e2, w11, 12);
R31(e1, a1, b1, c1, d1, w7, 8);
R32(e2, a2, b2, c2, d2, w8, 13);
R31(d1, e1, a1, b1, c1, w0, 13);
R32(d2, e2, a2, b2, c2, w12, 5);
R31(c1, d1, e1, a1, b1, w6, 6);
R32(c2, d2, e2, a2, b2, w2, 14);
R31(b1, c1, d1, e1, a1, w13, 5);
R32(b2, c2, d2, e2, a2, w10, 13);
R31(a1, b1, c1, d1, e1, w11, 12);
R32(a2, b2, c2, d2, e2, w0, 13);
R31(e1, a1, b1, c1, d1, w5, 7);
R32(e2, a2, b2, c2, d2, w4, 7);
R31(d1, e1, a1, b1, c1, w12, 5);
R32(d2, e2, a2, b2, c2, w13, 5);
R41(c1, d1, e1, a1, b1, w1, 11);
R42(c2, d2, e2, a2, b2, w8, 15);
R41(b1, c1, d1, e1, a1, w9, 12);
R42(b2, c2, d2, e2, a2, w6, 5);
R41(a1, b1, c1, d1, e1, w11, 14);
R42(a2, b2, c2, d2, e2, w4, 8);
R41(e1, a1, b1, c1, d1, w10, 15);
R42(e2, a2, b2, c2, d2, w1, 11);
R41(d1, e1, a1, b1, c1, w0, 14);
R42(d2, e2, a2, b2, c2, w3, 14);
R41(c1, d1, e1, a1, b1, w8, 15);
R42(c2, d2, e2, a2, b2, w11, 14);
R41(b1, c1, d1, e1, a1, w12, 9);
R42(b2, c2, d2, e2, a2, w15, 6);
R41(a1, b1, c1, d1, e1, w4, 8);
R42(a2, b2, c2, d2, e2, w0, 14);
R41(e1, a1, b1, c1, d1, w13, 9);
R42(e2, a2, b2, c2, d2, w5, 6);
R41(d1, e1, a1, b1, c1, w3, 14);
R42(d2, e2, a2, b2, c2, w12, 9);
R41(c1, d1, e1, a1, b1, w7, 5);
R42(c2, d2, e2, a2, b2, w2, 12);
R41(b1, c1, d1, e1, a1, w15, 6);
R42(b2, c2, d2, e2, a2, w13, 9);
R41(a1, b1, c1, d1, e1, w14, 8);
R42(a2, b2, c2, d2, e2, w9, 12);
R41(e1, a1, b1, c1, d1, w5, 6);
R42(e2, a2, b2, c2, d2, w7, 5);
R41(d1, e1, a1, b1, c1, w6, 5);
R42(d2, e2, a2, b2, c2, w10, 15);
R41(c1, d1, e1, a1, b1, w2, 12);
R42(c2, d2, e2, a2, b2, w14, 8);
R51(b1, c1, d1, e1, a1, w4, 9);
R52(b2, c2, d2, e2, a2, w12, 8);
R51(a1, b1, c1, d1, e1, w0, 15);
R52(a2, b2, c2, d2, e2, w15, 5);
R51(e1, a1, b1, c1, d1, w5, 5);
R52(e2, a2, b2, c2, d2, w10, 12);
R51(d1, e1, a1, b1, c1, w9, 11);
R52(d2, e2, a2, b2, c2, w4, 9);
R51(c1, d1, e1, a1, b1, w7, 6);
R52(c2, d2, e2, a2, b2, w1, 12);
R51(b1, c1, d1, e1, a1, w12, 8);
R52(b2, c2, d2, e2, a2, w5, 5);
R51(a1, b1, c1, d1, e1, w2, 13);
R52(a2, b2, c2, d2, e2, w8, 14);
R51(e1, a1, b1, c1, d1, w10, 12);
R52(e2, a2, b2, c2, d2, w7, 6);
R51(d1, e1, a1, b1, c1, w14, 5);
R52(d2, e2, a2, b2, c2, w6, 8);
R51(c1, d1, e1, a1, b1, w1, 12);
R52(c2, d2, e2, a2, b2, w2, 13);
R51(b1, c1, d1, e1, a1, w3, 13);
R52(b2, c2, d2, e2, a2, w13, 6);
R51(a1, b1, c1, d1, e1, w8, 14);
R52(a2, b2, c2, d2, e2, w14, 5);
R51(e1, a1, b1, c1, d1, w11, 11);
R52(e2, a2, b2, c2, d2, w0, 15);
R51(d1, e1, a1, b1, c1, w6, 8);
R52(d2, e2, a2, b2, c2, w3, 13);
R51(c1, d1, e1, a1, b1, w15, 5);
R52(c2, d2, e2, a2, b2, w9, 11);
R51(b1, c1, d1, e1, a1, w13, 6);
R52(b2, c2, d2, e2, a2, w11, 11);
uint32_t t = s[0];
s[0] = s[1] + c1 + d2;
s[1] = s[2] + d1 + e2;
s[2] = s[3] + e1 + a2;
s[3] = s[4] + a1 + b2;
s[4] = t + b1 + c2;
}
} // namespace ripemd160
} // namespace
////// RIPEMD160
CRIPEMD160::CRIPEMD160() : bytes(0)
{
ripemd160::Initialize(s);
}
CRIPEMD160& CRIPEMD160::Write(const unsigned char* data, size_t len)
{
const unsigned char* end = data + len;
size_t bufsize = bytes % 64;
if (bufsize && bufsize + len >= 64) {
// Fill the buffer, and process it.
memcpy(buf + bufsize, data, 64 - bufsize);
bytes += 64 - bufsize;
data += 64 - bufsize;
ripemd160::Transform(s, buf);
bufsize = 0;
}
while (end - data >= 64) {
// Process full chunks directly from the source.
ripemd160::Transform(s, data);
bytes += 64;
data += 64;
}
if (end > data) {
// Fill the buffer with what remains.
memcpy(buf + bufsize, data, end - data);
bytes += end - data;
}
return *this;
}
void CRIPEMD160::Finalize(unsigned char hash[OUTPUT_SIZE])
{
static const unsigned char pad[64] = {0x80};
unsigned char sizedesc[8];
WriteLE64(sizedesc, bytes << 3);
Write(pad, 1 + ((119 - (bytes % 64)) % 64));
Write(sizedesc, 8);
WriteLE32(hash, s[0]);
WriteLE32(hash + 4, s[1]);
WriteLE32(hash + 8, s[2]);
WriteLE32(hash + 12, s[3]);
WriteLE32(hash + 16, s[4]);
}
CRIPEMD160& CRIPEMD160::Reset()
{
bytes = 0;
ripemd160::Initialize(s);
return *this;
}

28
src/crypto/ripemd160.h Normal file
View file

@ -0,0 +1,28 @@
// Copyright (c) 2014-2016 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#ifndef DIGIBYTE_CRYPTO_RIPEMD160_H
#define DIGIBYTE_CRYPTO_RIPEMD160_H
#include <stdint.h>
#include <stdlib.h>
/** A hasher class for RIPEMD-160. */
class CRIPEMD160
{
private:
uint32_t s[5];
unsigned char buf[64];
size_t bytes;
public:
static const size_t OUTPUT_SIZE = 20;
CRIPEMD160();
CRIPEMD160& Write(const unsigned char* data, size_t len);
void Finalize(unsigned char hash[OUTPUT_SIZE]);
CRIPEMD160& Reset();
};
#endif // DIGIBYTE_CRYPTO_RIPEMD160_H

200
src/crypto/sha1.cpp Normal file
View file

@ -0,0 +1,200 @@
// Copyright (c) 2009-2020 The Bitcoin Core developers
// Copyright (c) 2014-2020 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#include <crypto/sha1.h>
#include <crypto/common.h>
#include <string.h>
// Internal implementation code.
namespace
{
/// Internal SHA-1 implementation.
namespace sha1
{
/** One round of SHA-1. */
void inline Round(uint32_t a, uint32_t& b, uint32_t c, uint32_t d, uint32_t& e, uint32_t f, uint32_t k, uint32_t w)
{
e += ((a << 5) | (a >> 27)) + f + k + w;
b = (b << 30) | (b >> 2);
}
uint32_t inline f1(uint32_t b, uint32_t c, uint32_t d) { return d ^ (b & (c ^ d)); }
uint32_t inline f2(uint32_t b, uint32_t c, uint32_t d) { return b ^ c ^ d; }
uint32_t inline f3(uint32_t b, uint32_t c, uint32_t d) { return (b & c) | (d & (b | c)); }
uint32_t inline left(uint32_t x) { return (x << 1) | (x >> 31); }
/** Initialize SHA-1 state. */
void inline Initialize(uint32_t* s)
{
s[0] = 0x67452301ul;
s[1] = 0xEFCDAB89ul;
s[2] = 0x98BADCFEul;
s[3] = 0x10325476ul;
s[4] = 0xC3D2E1F0ul;
}
const uint32_t k1 = 0x5A827999ul;
const uint32_t k2 = 0x6ED9EBA1ul;
const uint32_t k3 = 0x8F1BBCDCul;
const uint32_t k4 = 0xCA62C1D6ul;
/** Perform a SHA-1 transformation, processing a 64-byte chunk. */
void Transform(uint32_t* s, const unsigned char* chunk)
{
uint32_t a = s[0], b = s[1], c = s[2], d = s[3], e = s[4];
uint32_t w0, w1, w2, w3, w4, w5, w6, w7, w8, w9, w10, w11, w12, w13, w14, w15;
Round(a, b, c, d, e, f1(b, c, d), k1, w0 = ReadBE32(chunk + 0));
Round(e, a, b, c, d, f1(a, b, c), k1, w1 = ReadBE32(chunk + 4));
Round(d, e, a, b, c, f1(e, a, b), k1, w2 = ReadBE32(chunk + 8));
Round(c, d, e, a, b, f1(d, e, a), k1, w3 = ReadBE32(chunk + 12));
Round(b, c, d, e, a, f1(c, d, e), k1, w4 = ReadBE32(chunk + 16));
Round(a, b, c, d, e, f1(b, c, d), k1, w5 = ReadBE32(chunk + 20));
Round(e, a, b, c, d, f1(a, b, c), k1, w6 = ReadBE32(chunk + 24));
Round(d, e, a, b, c, f1(e, a, b), k1, w7 = ReadBE32(chunk + 28));
Round(c, d, e, a, b, f1(d, e, a), k1, w8 = ReadBE32(chunk + 32));
Round(b, c, d, e, a, f1(c, d, e), k1, w9 = ReadBE32(chunk + 36));
Round(a, b, c, d, e, f1(b, c, d), k1, w10 = ReadBE32(chunk + 40));
Round(e, a, b, c, d, f1(a, b, c), k1, w11 = ReadBE32(chunk + 44));
Round(d, e, a, b, c, f1(e, a, b), k1, w12 = ReadBE32(chunk + 48));
Round(c, d, e, a, b, f1(d, e, a), k1, w13 = ReadBE32(chunk + 52));
Round(b, c, d, e, a, f1(c, d, e), k1, w14 = ReadBE32(chunk + 56));
Round(a, b, c, d, e, f1(b, c, d), k1, w15 = ReadBE32(chunk + 60));
Round(e, a, b, c, d, f1(a, b, c), k1, w0 = left(w0 ^ w13 ^ w8 ^ w2));
Round(d, e, a, b, c, f1(e, a, b), k1, w1 = left(w1 ^ w14 ^ w9 ^ w3));
Round(c, d, e, a, b, f1(d, e, a), k1, w2 = left(w2 ^ w15 ^ w10 ^ w4));
Round(b, c, d, e, a, f1(c, d, e), k1, w3 = left(w3 ^ w0 ^ w11 ^ w5));
Round(a, b, c, d, e, f2(b, c, d), k2, w4 = left(w4 ^ w1 ^ w12 ^ w6));
Round(e, a, b, c, d, f2(a, b, c), k2, w5 = left(w5 ^ w2 ^ w13 ^ w7));
Round(d, e, a, b, c, f2(e, a, b), k2, w6 = left(w6 ^ w3 ^ w14 ^ w8));
Round(c, d, e, a, b, f2(d, e, a), k2, w7 = left(w7 ^ w4 ^ w15 ^ w9));
Round(b, c, d, e, a, f2(c, d, e), k2, w8 = left(w8 ^ w5 ^ w0 ^ w10));
Round(a, b, c, d, e, f2(b, c, d), k2, w9 = left(w9 ^ w6 ^ w1 ^ w11));
Round(e, a, b, c, d, f2(a, b, c), k2, w10 = left(w10 ^ w7 ^ w2 ^ w12));
Round(d, e, a, b, c, f2(e, a, b), k2, w11 = left(w11 ^ w8 ^ w3 ^ w13));
Round(c, d, e, a, b, f2(d, e, a), k2, w12 = left(w12 ^ w9 ^ w4 ^ w14));
Round(b, c, d, e, a, f2(c, d, e), k2, w13 = left(w13 ^ w10 ^ w5 ^ w15));
Round(a, b, c, d, e, f2(b, c, d), k2, w14 = left(w14 ^ w11 ^ w6 ^ w0));
Round(e, a, b, c, d, f2(a, b, c), k2, w15 = left(w15 ^ w12 ^ w7 ^ w1));
Round(d, e, a, b, c, f2(e, a, b), k2, w0 = left(w0 ^ w13 ^ w8 ^ w2));
Round(c, d, e, a, b, f2(d, e, a), k2, w1 = left(w1 ^ w14 ^ w9 ^ w3));
Round(b, c, d, e, a, f2(c, d, e), k2, w2 = left(w2 ^ w15 ^ w10 ^ w4));
Round(a, b, c, d, e, f2(b, c, d), k2, w3 = left(w3 ^ w0 ^ w11 ^ w5));
Round(e, a, b, c, d, f2(a, b, c), k2, w4 = left(w4 ^ w1 ^ w12 ^ w6));
Round(d, e, a, b, c, f2(e, a, b), k2, w5 = left(w5 ^ w2 ^ w13 ^ w7));
Round(c, d, e, a, b, f2(d, e, a), k2, w6 = left(w6 ^ w3 ^ w14 ^ w8));
Round(b, c, d, e, a, f2(c, d, e), k2, w7 = left(w7 ^ w4 ^ w15 ^ w9));
Round(a, b, c, d, e, f3(b, c, d), k3, w8 = left(w8 ^ w5 ^ w0 ^ w10));
Round(e, a, b, c, d, f3(a, b, c), k3, w9 = left(w9 ^ w6 ^ w1 ^ w11));
Round(d, e, a, b, c, f3(e, a, b), k3, w10 = left(w10 ^ w7 ^ w2 ^ w12));
Round(c, d, e, a, b, f3(d, e, a), k3, w11 = left(w11 ^ w8 ^ w3 ^ w13));
Round(b, c, d, e, a, f3(c, d, e), k3, w12 = left(w12 ^ w9 ^ w4 ^ w14));
Round(a, b, c, d, e, f3(b, c, d), k3, w13 = left(w13 ^ w10 ^ w5 ^ w15));
Round(e, a, b, c, d, f3(a, b, c), k3, w14 = left(w14 ^ w11 ^ w6 ^ w0));
Round(d, e, a, b, c, f3(e, a, b), k3, w15 = left(w15 ^ w12 ^ w7 ^ w1));
Round(c, d, e, a, b, f3(d, e, a), k3, w0 = left(w0 ^ w13 ^ w8 ^ w2));
Round(b, c, d, e, a, f3(c, d, e), k3, w1 = left(w1 ^ w14 ^ w9 ^ w3));
Round(a, b, c, d, e, f3(b, c, d), k3, w2 = left(w2 ^ w15 ^ w10 ^ w4));
Round(e, a, b, c, d, f3(a, b, c), k3, w3 = left(w3 ^ w0 ^ w11 ^ w5));
Round(d, e, a, b, c, f3(e, a, b), k3, w4 = left(w4 ^ w1 ^ w12 ^ w6));
Round(c, d, e, a, b, f3(d, e, a), k3, w5 = left(w5 ^ w2 ^ w13 ^ w7));
Round(b, c, d, e, a, f3(c, d, e), k3, w6 = left(w6 ^ w3 ^ w14 ^ w8));
Round(a, b, c, d, e, f3(b, c, d), k3, w7 = left(w7 ^ w4 ^ w15 ^ w9));
Round(e, a, b, c, d, f3(a, b, c), k3, w8 = left(w8 ^ w5 ^ w0 ^ w10));
Round(d, e, a, b, c, f3(e, a, b), k3, w9 = left(w9 ^ w6 ^ w1 ^ w11));
Round(c, d, e, a, b, f3(d, e, a), k3, w10 = left(w10 ^ w7 ^ w2 ^ w12));
Round(b, c, d, e, a, f3(c, d, e), k3, w11 = left(w11 ^ w8 ^ w3 ^ w13));
Round(a, b, c, d, e, f2(b, c, d), k4, w12 = left(w12 ^ w9 ^ w4 ^ w14));
Round(e, a, b, c, d, f2(a, b, c), k4, w13 = left(w13 ^ w10 ^ w5 ^ w15));
Round(d, e, a, b, c, f2(e, a, b), k4, w14 = left(w14 ^ w11 ^ w6 ^ w0));
Round(c, d, e, a, b, f2(d, e, a), k4, w15 = left(w15 ^ w12 ^ w7 ^ w1));
Round(b, c, d, e, a, f2(c, d, e), k4, w0 = left(w0 ^ w13 ^ w8 ^ w2));
Round(a, b, c, d, e, f2(b, c, d), k4, w1 = left(w1 ^ w14 ^ w9 ^ w3));
Round(e, a, b, c, d, f2(a, b, c), k4, w2 = left(w2 ^ w15 ^ w10 ^ w4));
Round(d, e, a, b, c, f2(e, a, b), k4, w3 = left(w3 ^ w0 ^ w11 ^ w5));
Round(c, d, e, a, b, f2(d, e, a), k4, w4 = left(w4 ^ w1 ^ w12 ^ w6));
Round(b, c, d, e, a, f2(c, d, e), k4, w5 = left(w5 ^ w2 ^ w13 ^ w7));
Round(a, b, c, d, e, f2(b, c, d), k4, w6 = left(w6 ^ w3 ^ w14 ^ w8));
Round(e, a, b, c, d, f2(a, b, c), k4, w7 = left(w7 ^ w4 ^ w15 ^ w9));
Round(d, e, a, b, c, f2(e, a, b), k4, w8 = left(w8 ^ w5 ^ w0 ^ w10));
Round(c, d, e, a, b, f2(d, e, a), k4, w9 = left(w9 ^ w6 ^ w1 ^ w11));
Round(b, c, d, e, a, f2(c, d, e), k4, w10 = left(w10 ^ w7 ^ w2 ^ w12));
Round(a, b, c, d, e, f2(b, c, d), k4, w11 = left(w11 ^ w8 ^ w3 ^ w13));
Round(e, a, b, c, d, f2(a, b, c), k4, w12 = left(w12 ^ w9 ^ w4 ^ w14));
Round(d, e, a, b, c, f2(e, a, b), k4, left(w13 ^ w10 ^ w5 ^ w15));
Round(c, d, e, a, b, f2(d, e, a), k4, left(w14 ^ w11 ^ w6 ^ w0));
Round(b, c, d, e, a, f2(c, d, e), k4, left(w15 ^ w12 ^ w7 ^ w1));
s[0] += a;
s[1] += b;
s[2] += c;
s[3] += d;
s[4] += e;
}
} // namespace sha1
} // namespace
////// SHA1
CSHA1::CSHA1() : bytes(0)
{
sha1::Initialize(s);
}
CSHA1& CSHA1::Write(const unsigned char* data, size_t len)
{
const unsigned char* end = data + len;
size_t bufsize = bytes % 64;
if (bufsize && bufsize + len >= 64) {
// Fill the buffer, and process it.
memcpy(buf + bufsize, data, 64 - bufsize);
bytes += 64 - bufsize;
data += 64 - bufsize;
sha1::Transform(s, buf);
bufsize = 0;
}
while (end - data >= 64) {
// Process full chunks directly from the source.
sha1::Transform(s, data);
bytes += 64;
data += 64;
}
if (end > data) {
// Fill the buffer with what remains.
memcpy(buf + bufsize, data, end - data);
bytes += end - data;
}
return *this;
}
void CSHA1::Finalize(unsigned char hash[OUTPUT_SIZE])
{
static const unsigned char pad[64] = {0x80};
unsigned char sizedesc[8];
WriteBE64(sizedesc, bytes << 3);
Write(pad, 1 + ((119 - (bytes % 64)) % 64));
Write(sizedesc, 8);
WriteBE32(hash, s[0]);
WriteBE32(hash + 4, s[1]);
WriteBE32(hash + 8, s[2]);
WriteBE32(hash + 12, s[3]);
WriteBE32(hash + 16, s[4]);
}
CSHA1& CSHA1::Reset()
{
bytes = 0;
sha1::Initialize(s);
return *this;
}

28
src/crypto/sha1.h Normal file
View file

@ -0,0 +1,28 @@
// Copyright (c) 2014-2016 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#ifndef DIGIBYTE_CRYPTO_SHA1_H
#define DIGIBYTE_CRYPTO_SHA1_H
#include <stdint.h>
#include <stdlib.h>
/** A hasher class for SHA1. */
class CSHA1
{
private:
uint32_t s[5];
unsigned char buf[64];
size_t bytes;
public:
static const size_t OUTPUT_SIZE = 20;
CSHA1();
CSHA1& Write(const unsigned char* data, size_t len);
void Finalize(unsigned char hash[OUTPUT_SIZE]);
CSHA1& Reset();
};
#endif // DIGIBYTE_CRYPTO_SHA1_H

720
src/crypto/sha256.cpp Normal file
View file

@ -0,0 +1,720 @@
// Copyright (c) 2009-2020 The Bitcoin Core developers
// Copyright (c) 2014-2020 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#include <crypto/sha256.h>
#include <crypto/common.h>
#include <assert.h>
#include <string.h>
#include <compat/cpuid.h>
#if defined(__x86_64__) || defined(__amd64__) || defined(__i386__)
#if defined(USE_ASM)
namespace sha256_sse4
{
void Transform(uint32_t* s, const unsigned char* chunk, size_t blocks);
}
#endif
#endif
namespace sha256d64_sse41
{
void Transform_4way(unsigned char* out, const unsigned char* in);
}
namespace sha256d64_avx2
{
void Transform_8way(unsigned char* out, const unsigned char* in);
}
namespace sha256d64_shani
{
void Transform_2way(unsigned char* out, const unsigned char* in);
}
namespace sha256_shani
{
void Transform(uint32_t* s, const unsigned char* chunk, size_t blocks);
}
// Internal implementation code.
namespace
{
/// Internal SHA-256 implementation.
namespace sha256
{
uint32_t inline Ch(uint32_t x, uint32_t y, uint32_t z) { return z ^ (x & (y ^ z)); }
uint32_t inline Maj(uint32_t x, uint32_t y, uint32_t z) { return (x & y) | (z & (x | y)); }
uint32_t inline Sigma0(uint32_t x) { return (x >> 2 | x << 30) ^ (x >> 13 | x << 19) ^ (x >> 22 | x << 10); }
uint32_t inline Sigma1(uint32_t x) { return (x >> 6 | x << 26) ^ (x >> 11 | x << 21) ^ (x >> 25 | x << 7); }
uint32_t inline sigma0(uint32_t x) { return (x >> 7 | x << 25) ^ (x >> 18 | x << 14) ^ (x >> 3); }
uint32_t inline sigma1(uint32_t x) { return (x >> 17 | x << 15) ^ (x >> 19 | x << 13) ^ (x >> 10); }
/** One round of SHA-256. */
void inline Round(uint32_t a, uint32_t b, uint32_t c, uint32_t& d, uint32_t e, uint32_t f, uint32_t g, uint32_t& h, uint32_t k)
{
uint32_t t1 = h + Sigma1(e) + Ch(e, f, g) + k;
uint32_t t2 = Sigma0(a) + Maj(a, b, c);
d += t1;
h = t1 + t2;
}
/** Initialize SHA-256 state. */
void inline Initialize(uint32_t* s)
{
s[0] = 0x6a09e667ul;
s[1] = 0xbb67ae85ul;
s[2] = 0x3c6ef372ul;
s[3] = 0xa54ff53aul;
s[4] = 0x510e527ful;
s[5] = 0x9b05688cul;
s[6] = 0x1f83d9abul;
s[7] = 0x5be0cd19ul;
}
/** Perform a number of SHA-256 transformations, processing 64-byte chunks. */
void Transform(uint32_t* s, const unsigned char* chunk, size_t blocks)
{
while (blocks--) {
uint32_t a = s[0], b = s[1], c = s[2], d = s[3], e = s[4], f = s[5], g = s[6], h = s[7];
uint32_t w0, w1, w2, w3, w4, w5, w6, w7, w8, w9, w10, w11, w12, w13, w14, w15;
Round(a, b, c, d, e, f, g, h, 0x428a2f98 + (w0 = ReadBE32(chunk + 0)));
Round(h, a, b, c, d, e, f, g, 0x71374491 + (w1 = ReadBE32(chunk + 4)));
Round(g, h, a, b, c, d, e, f, 0xb5c0fbcf + (w2 = ReadBE32(chunk + 8)));
Round(f, g, h, a, b, c, d, e, 0xe9b5dba5 + (w3 = ReadBE32(chunk + 12)));
Round(e, f, g, h, a, b, c, d, 0x3956c25b + (w4 = ReadBE32(chunk + 16)));
Round(d, e, f, g, h, a, b, c, 0x59f111f1 + (w5 = ReadBE32(chunk + 20)));
Round(c, d, e, f, g, h, a, b, 0x923f82a4 + (w6 = ReadBE32(chunk + 24)));
Round(b, c, d, e, f, g, h, a, 0xab1c5ed5 + (w7 = ReadBE32(chunk + 28)));
Round(a, b, c, d, e, f, g, h, 0xd807aa98 + (w8 = ReadBE32(chunk + 32)));
Round(h, a, b, c, d, e, f, g, 0x12835b01 + (w9 = ReadBE32(chunk + 36)));
Round(g, h, a, b, c, d, e, f, 0x243185be + (w10 = ReadBE32(chunk + 40)));
Round(f, g, h, a, b, c, d, e, 0x550c7dc3 + (w11 = ReadBE32(chunk + 44)));
Round(e, f, g, h, a, b, c, d, 0x72be5d74 + (w12 = ReadBE32(chunk + 48)));
Round(d, e, f, g, h, a, b, c, 0x80deb1fe + (w13 = ReadBE32(chunk + 52)));
Round(c, d, e, f, g, h, a, b, 0x9bdc06a7 + (w14 = ReadBE32(chunk + 56)));
Round(b, c, d, e, f, g, h, a, 0xc19bf174 + (w15 = ReadBE32(chunk + 60)));
Round(a, b, c, d, e, f, g, h, 0xe49b69c1 + (w0 += sigma1(w14) + w9 + sigma0(w1)));
Round(h, a, b, c, d, e, f, g, 0xefbe4786 + (w1 += sigma1(w15) + w10 + sigma0(w2)));
Round(g, h, a, b, c, d, e, f, 0x0fc19dc6 + (w2 += sigma1(w0) + w11 + sigma0(w3)));
Round(f, g, h, a, b, c, d, e, 0x240ca1cc + (w3 += sigma1(w1) + w12 + sigma0(w4)));
Round(e, f, g, h, a, b, c, d, 0x2de92c6f + (w4 += sigma1(w2) + w13 + sigma0(w5)));
Round(d, e, f, g, h, a, b, c, 0x4a7484aa + (w5 += sigma1(w3) + w14 + sigma0(w6)));
Round(c, d, e, f, g, h, a, b, 0x5cb0a9dc + (w6 += sigma1(w4) + w15 + sigma0(w7)));
Round(b, c, d, e, f, g, h, a, 0x76f988da + (w7 += sigma1(w5) + w0 + sigma0(w8)));
Round(a, b, c, d, e, f, g, h, 0x983e5152 + (w8 += sigma1(w6) + w1 + sigma0(w9)));
Round(h, a, b, c, d, e, f, g, 0xa831c66d + (w9 += sigma1(w7) + w2 + sigma0(w10)));
Round(g, h, a, b, c, d, e, f, 0xb00327c8 + (w10 += sigma1(w8) + w3 + sigma0(w11)));
Round(f, g, h, a, b, c, d, e, 0xbf597fc7 + (w11 += sigma1(w9) + w4 + sigma0(w12)));
Round(e, f, g, h, a, b, c, d, 0xc6e00bf3 + (w12 += sigma1(w10) + w5 + sigma0(w13)));
Round(d, e, f, g, h, a, b, c, 0xd5a79147 + (w13 += sigma1(w11) + w6 + sigma0(w14)));
Round(c, d, e, f, g, h, a, b, 0x06ca6351 + (w14 += sigma1(w12) + w7 + sigma0(w15)));
Round(b, c, d, e, f, g, h, a, 0x14292967 + (w15 += sigma1(w13) + w8 + sigma0(w0)));
Round(a, b, c, d, e, f, g, h, 0x27b70a85 + (w0 += sigma1(w14) + w9 + sigma0(w1)));
Round(h, a, b, c, d, e, f, g, 0x2e1b2138 + (w1 += sigma1(w15) + w10 + sigma0(w2)));
Round(g, h, a, b, c, d, e, f, 0x4d2c6dfc + (w2 += sigma1(w0) + w11 + sigma0(w3)));
Round(f, g, h, a, b, c, d, e, 0x53380d13 + (w3 += sigma1(w1) + w12 + sigma0(w4)));
Round(e, f, g, h, a, b, c, d, 0x650a7354 + (w4 += sigma1(w2) + w13 + sigma0(w5)));
Round(d, e, f, g, h, a, b, c, 0x766a0abb + (w5 += sigma1(w3) + w14 + sigma0(w6)));
Round(c, d, e, f, g, h, a, b, 0x81c2c92e + (w6 += sigma1(w4) + w15 + sigma0(w7)));
Round(b, c, d, e, f, g, h, a, 0x92722c85 + (w7 += sigma1(w5) + w0 + sigma0(w8)));
Round(a, b, c, d, e, f, g, h, 0xa2bfe8a1 + (w8 += sigma1(w6) + w1 + sigma0(w9)));
Round(h, a, b, c, d, e, f, g, 0xa81a664b + (w9 += sigma1(w7) + w2 + sigma0(w10)));
Round(g, h, a, b, c, d, e, f, 0xc24b8b70 + (w10 += sigma1(w8) + w3 + sigma0(w11)));
Round(f, g, h, a, b, c, d, e, 0xc76c51a3 + (w11 += sigma1(w9) + w4 + sigma0(w12)));
Round(e, f, g, h, a, b, c, d, 0xd192e819 + (w12 += sigma1(w10) + w5 + sigma0(w13)));
Round(d, e, f, g, h, a, b, c, 0xd6990624 + (w13 += sigma1(w11) + w6 + sigma0(w14)));
Round(c, d, e, f, g, h, a, b, 0xf40e3585 + (w14 += sigma1(w12) + w7 + sigma0(w15)));
Round(b, c, d, e, f, g, h, a, 0x106aa070 + (w15 += sigma1(w13) + w8 + sigma0(w0)));
Round(a, b, c, d, e, f, g, h, 0x19a4c116 + (w0 += sigma1(w14) + w9 + sigma0(w1)));
Round(h, a, b, c, d, e, f, g, 0x1e376c08 + (w1 += sigma1(w15) + w10 + sigma0(w2)));
Round(g, h, a, b, c, d, e, f, 0x2748774c + (w2 += sigma1(w0) + w11 + sigma0(w3)));
Round(f, g, h, a, b, c, d, e, 0x34b0bcb5 + (w3 += sigma1(w1) + w12 + sigma0(w4)));
Round(e, f, g, h, a, b, c, d, 0x391c0cb3 + (w4 += sigma1(w2) + w13 + sigma0(w5)));
Round(d, e, f, g, h, a, b, c, 0x4ed8aa4a + (w5 += sigma1(w3) + w14 + sigma0(w6)));
Round(c, d, e, f, g, h, a, b, 0x5b9cca4f + (w6 += sigma1(w4) + w15 + sigma0(w7)));
Round(b, c, d, e, f, g, h, a, 0x682e6ff3 + (w7 += sigma1(w5) + w0 + sigma0(w8)));
Round(a, b, c, d, e, f, g, h, 0x748f82ee + (w8 += sigma1(w6) + w1 + sigma0(w9)));
Round(h, a, b, c, d, e, f, g, 0x78a5636f + (w9 += sigma1(w7) + w2 + sigma0(w10)));
Round(g, h, a, b, c, d, e, f, 0x84c87814 + (w10 += sigma1(w8) + w3 + sigma0(w11)));
Round(f, g, h, a, b, c, d, e, 0x8cc70208 + (w11 += sigma1(w9) + w4 + sigma0(w12)));
Round(e, f, g, h, a, b, c, d, 0x90befffa + (w12 += sigma1(w10) + w5 + sigma0(w13)));
Round(d, e, f, g, h, a, b, c, 0xa4506ceb + (w13 += sigma1(w11) + w6 + sigma0(w14)));
Round(c, d, e, f, g, h, a, b, 0xbef9a3f7 + (w14 + sigma1(w12) + w7 + sigma0(w15)));
Round(b, c, d, e, f, g, h, a, 0xc67178f2 + (w15 + sigma1(w13) + w8 + sigma0(w0)));
s[0] += a;
s[1] += b;
s[2] += c;
s[3] += d;
s[4] += e;
s[5] += f;
s[6] += g;
s[7] += h;
chunk += 64;
}
}
void TransformD64(unsigned char* out, const unsigned char* in)
{
// Transform 1
uint32_t a = 0x6a09e667ul;
uint32_t b = 0xbb67ae85ul;
uint32_t c = 0x3c6ef372ul;
uint32_t d = 0xa54ff53aul;
uint32_t e = 0x510e527ful;
uint32_t f = 0x9b05688cul;
uint32_t g = 0x1f83d9abul;
uint32_t h = 0x5be0cd19ul;
uint32_t w0, w1, w2, w3, w4, w5, w6, w7, w8, w9, w10, w11, w12, w13, w14, w15;
Round(a, b, c, d, e, f, g, h, 0x428a2f98ul + (w0 = ReadBE32(in + 0)));
Round(h, a, b, c, d, e, f, g, 0x71374491ul + (w1 = ReadBE32(in + 4)));
Round(g, h, a, b, c, d, e, f, 0xb5c0fbcful + (w2 = ReadBE32(in + 8)));
Round(f, g, h, a, b, c, d, e, 0xe9b5dba5ul + (w3 = ReadBE32(in + 12)));
Round(e, f, g, h, a, b, c, d, 0x3956c25bul + (w4 = ReadBE32(in + 16)));
Round(d, e, f, g, h, a, b, c, 0x59f111f1ul + (w5 = ReadBE32(in + 20)));
Round(c, d, e, f, g, h, a, b, 0x923f82a4ul + (w6 = ReadBE32(in + 24)));
Round(b, c, d, e, f, g, h, a, 0xab1c5ed5ul + (w7 = ReadBE32(in + 28)));
Round(a, b, c, d, e, f, g, h, 0xd807aa98ul + (w8 = ReadBE32(in + 32)));
Round(h, a, b, c, d, e, f, g, 0x12835b01ul + (w9 = ReadBE32(in + 36)));
Round(g, h, a, b, c, d, e, f, 0x243185beul + (w10 = ReadBE32(in + 40)));
Round(f, g, h, a, b, c, d, e, 0x550c7dc3ul + (w11 = ReadBE32(in + 44)));
Round(e, f, g, h, a, b, c, d, 0x72be5d74ul + (w12 = ReadBE32(in + 48)));
Round(d, e, f, g, h, a, b, c, 0x80deb1feul + (w13 = ReadBE32(in + 52)));
Round(c, d, e, f, g, h, a, b, 0x9bdc06a7ul + (w14 = ReadBE32(in + 56)));
Round(b, c, d, e, f, g, h, a, 0xc19bf174ul + (w15 = ReadBE32(in + 60)));
Round(a, b, c, d, e, f, g, h, 0xe49b69c1ul + (w0 += sigma1(w14) + w9 + sigma0(w1)));
Round(h, a, b, c, d, e, f, g, 0xefbe4786ul + (w1 += sigma1(w15) + w10 + sigma0(w2)));
Round(g, h, a, b, c, d, e, f, 0x0fc19dc6ul + (w2 += sigma1(w0) + w11 + sigma0(w3)));
Round(f, g, h, a, b, c, d, e, 0x240ca1ccul + (w3 += sigma1(w1) + w12 + sigma0(w4)));
Round(e, f, g, h, a, b, c, d, 0x2de92c6ful + (w4 += sigma1(w2) + w13 + sigma0(w5)));
Round(d, e, f, g, h, a, b, c, 0x4a7484aaul + (w5 += sigma1(w3) + w14 + sigma0(w6)));
Round(c, d, e, f, g, h, a, b, 0x5cb0a9dcul + (w6 += sigma1(w4) + w15 + sigma0(w7)));
Round(b, c, d, e, f, g, h, a, 0x76f988daul + (w7 += sigma1(w5) + w0 + sigma0(w8)));
Round(a, b, c, d, e, f, g, h, 0x983e5152ul + (w8 += sigma1(w6) + w1 + sigma0(w9)));
Round(h, a, b, c, d, e, f, g, 0xa831c66dul + (w9 += sigma1(w7) + w2 + sigma0(w10)));
Round(g, h, a, b, c, d, e, f, 0xb00327c8ul + (w10 += sigma1(w8) + w3 + sigma0(w11)));
Round(f, g, h, a, b, c, d, e, 0xbf597fc7ul + (w11 += sigma1(w9) + w4 + sigma0(w12)));
Round(e, f, g, h, a, b, c, d, 0xc6e00bf3ul + (w12 += sigma1(w10) + w5 + sigma0(w13)));
Round(d, e, f, g, h, a, b, c, 0xd5a79147ul + (w13 += sigma1(w11) + w6 + sigma0(w14)));
Round(c, d, e, f, g, h, a, b, 0x06ca6351ul + (w14 += sigma1(w12) + w7 + sigma0(w15)));
Round(b, c, d, e, f, g, h, a, 0x14292967ul + (w15 += sigma1(w13) + w8 + sigma0(w0)));
Round(a, b, c, d, e, f, g, h, 0x27b70a85ul + (w0 += sigma1(w14) + w9 + sigma0(w1)));
Round(h, a, b, c, d, e, f, g, 0x2e1b2138ul + (w1 += sigma1(w15) + w10 + sigma0(w2)));
Round(g, h, a, b, c, d, e, f, 0x4d2c6dfcul + (w2 += sigma1(w0) + w11 + sigma0(w3)));
Round(f, g, h, a, b, c, d, e, 0x53380d13ul + (w3 += sigma1(w1) + w12 + sigma0(w4)));
Round(e, f, g, h, a, b, c, d, 0x650a7354ul + (w4 += sigma1(w2) + w13 + sigma0(w5)));
Round(d, e, f, g, h, a, b, c, 0x766a0abbul + (w5 += sigma1(w3) + w14 + sigma0(w6)));
Round(c, d, e, f, g, h, a, b, 0x81c2c92eul + (w6 += sigma1(w4) + w15 + sigma0(w7)));
Round(b, c, d, e, f, g, h, a, 0x92722c85ul + (w7 += sigma1(w5) + w0 + sigma0(w8)));
Round(a, b, c, d, e, f, g, h, 0xa2bfe8a1ul + (w8 += sigma1(w6) + w1 + sigma0(w9)));
Round(h, a, b, c, d, e, f, g, 0xa81a664bul + (w9 += sigma1(w7) + w2 + sigma0(w10)));
Round(g, h, a, b, c, d, e, f, 0xc24b8b70ul + (w10 += sigma1(w8) + w3 + sigma0(w11)));
Round(f, g, h, a, b, c, d, e, 0xc76c51a3ul + (w11 += sigma1(w9) + w4 + sigma0(w12)));
Round(e, f, g, h, a, b, c, d, 0xd192e819ul + (w12 += sigma1(w10) + w5 + sigma0(w13)));
Round(d, e, f, g, h, a, b, c, 0xd6990624ul + (w13 += sigma1(w11) + w6 + sigma0(w14)));
Round(c, d, e, f, g, h, a, b, 0xf40e3585ul + (w14 += sigma1(w12) + w7 + sigma0(w15)));
Round(b, c, d, e, f, g, h, a, 0x106aa070ul + (w15 += sigma1(w13) + w8 + sigma0(w0)));
Round(a, b, c, d, e, f, g, h, 0x19a4c116ul + (w0 += sigma1(w14) + w9 + sigma0(w1)));
Round(h, a, b, c, d, e, f, g, 0x1e376c08ul + (w1 += sigma1(w15) + w10 + sigma0(w2)));
Round(g, h, a, b, c, d, e, f, 0x2748774cul + (w2 += sigma1(w0) + w11 + sigma0(w3)));
Round(f, g, h, a, b, c, d, e, 0x34b0bcb5ul + (w3 += sigma1(w1) + w12 + sigma0(w4)));
Round(e, f, g, h, a, b, c, d, 0x391c0cb3ul + (w4 += sigma1(w2) + w13 + sigma0(w5)));
Round(d, e, f, g, h, a, b, c, 0x4ed8aa4aul + (w5 += sigma1(w3) + w14 + sigma0(w6)));
Round(c, d, e, f, g, h, a, b, 0x5b9cca4ful + (w6 += sigma1(w4) + w15 + sigma0(w7)));
Round(b, c, d, e, f, g, h, a, 0x682e6ff3ul + (w7 += sigma1(w5) + w0 + sigma0(w8)));
Round(a, b, c, d, e, f, g, h, 0x748f82eeul + (w8 += sigma1(w6) + w1 + sigma0(w9)));
Round(h, a, b, c, d, e, f, g, 0x78a5636ful + (w9 += sigma1(w7) + w2 + sigma0(w10)));
Round(g, h, a, b, c, d, e, f, 0x84c87814ul + (w10 += sigma1(w8) + w3 + sigma0(w11)));
Round(f, g, h, a, b, c, d, e, 0x8cc70208ul + (w11 += sigma1(w9) + w4 + sigma0(w12)));
Round(e, f, g, h, a, b, c, d, 0x90befffaul + (w12 += sigma1(w10) + w5 + sigma0(w13)));
Round(d, e, f, g, h, a, b, c, 0xa4506cebul + (w13 += sigma1(w11) + w6 + sigma0(w14)));
Round(c, d, e, f, g, h, a, b, 0xbef9a3f7ul + (w14 + sigma1(w12) + w7 + sigma0(w15)));
Round(b, c, d, e, f, g, h, a, 0xc67178f2ul + (w15 + sigma1(w13) + w8 + sigma0(w0)));
a += 0x6a09e667ul;
b += 0xbb67ae85ul;
c += 0x3c6ef372ul;
d += 0xa54ff53aul;
e += 0x510e527ful;
f += 0x9b05688cul;
g += 0x1f83d9abul;
h += 0x5be0cd19ul;
uint32_t t0 = a, t1 = b, t2 = c, t3 = d, t4 = e, t5 = f, t6 = g, t7 = h;
// Transform 2
Round(a, b, c, d, e, f, g, h, 0xc28a2f98ul);
Round(h, a, b, c, d, e, f, g, 0x71374491ul);
Round(g, h, a, b, c, d, e, f, 0xb5c0fbcful);
Round(f, g, h, a, b, c, d, e, 0xe9b5dba5ul);
Round(e, f, g, h, a, b, c, d, 0x3956c25bul);
Round(d, e, f, g, h, a, b, c, 0x59f111f1ul);
Round(c, d, e, f, g, h, a, b, 0x923f82a4ul);
Round(b, c, d, e, f, g, h, a, 0xab1c5ed5ul);
Round(a, b, c, d, e, f, g, h, 0xd807aa98ul);
Round(h, a, b, c, d, e, f, g, 0x12835b01ul);
Round(g, h, a, b, c, d, e, f, 0x243185beul);
Round(f, g, h, a, b, c, d, e, 0x550c7dc3ul);
Round(e, f, g, h, a, b, c, d, 0x72be5d74ul);
Round(d, e, f, g, h, a, b, c, 0x80deb1feul);
Round(c, d, e, f, g, h, a, b, 0x9bdc06a7ul);
Round(b, c, d, e, f, g, h, a, 0xc19bf374ul);
Round(a, b, c, d, e, f, g, h, 0x649b69c1ul);
Round(h, a, b, c, d, e, f, g, 0xf0fe4786ul);
Round(g, h, a, b, c, d, e, f, 0x0fe1edc6ul);
Round(f, g, h, a, b, c, d, e, 0x240cf254ul);
Round(e, f, g, h, a, b, c, d, 0x4fe9346ful);
Round(d, e, f, g, h, a, b, c, 0x6cc984beul);
Round(c, d, e, f, g, h, a, b, 0x61b9411eul);
Round(b, c, d, e, f, g, h, a, 0x16f988faul);
Round(a, b, c, d, e, f, g, h, 0xf2c65152ul);
Round(h, a, b, c, d, e, f, g, 0xa88e5a6dul);
Round(g, h, a, b, c, d, e, f, 0xb019fc65ul);
Round(f, g, h, a, b, c, d, e, 0xb9d99ec7ul);
Round(e, f, g, h, a, b, c, d, 0x9a1231c3ul);
Round(d, e, f, g, h, a, b, c, 0xe70eeaa0ul);
Round(c, d, e, f, g, h, a, b, 0xfdb1232bul);
Round(b, c, d, e, f, g, h, a, 0xc7353eb0ul);
Round(a, b, c, d, e, f, g, h, 0x3069bad5ul);
Round(h, a, b, c, d, e, f, g, 0xcb976d5ful);
Round(g, h, a, b, c, d, e, f, 0x5a0f118ful);
Round(f, g, h, a, b, c, d, e, 0xdc1eeefdul);
Round(e, f, g, h, a, b, c, d, 0x0a35b689ul);
Round(d, e, f, g, h, a, b, c, 0xde0b7a04ul);
Round(c, d, e, f, g, h, a, b, 0x58f4ca9dul);
Round(b, c, d, e, f, g, h, a, 0xe15d5b16ul);
Round(a, b, c, d, e, f, g, h, 0x007f3e86ul);
Round(h, a, b, c, d, e, f, g, 0x37088980ul);
Round(g, h, a, b, c, d, e, f, 0xa507ea32ul);
Round(f, g, h, a, b, c, d, e, 0x6fab9537ul);
Round(e, f, g, h, a, b, c, d, 0x17406110ul);
Round(d, e, f, g, h, a, b, c, 0x0d8cd6f1ul);
Round(c, d, e, f, g, h, a, b, 0xcdaa3b6dul);
Round(b, c, d, e, f, g, h, a, 0xc0bbbe37ul);
Round(a, b, c, d, e, f, g, h, 0x83613bdaul);
Round(h, a, b, c, d, e, f, g, 0xdb48a363ul);
Round(g, h, a, b, c, d, e, f, 0x0b02e931ul);
Round(f, g, h, a, b, c, d, e, 0x6fd15ca7ul);
Round(e, f, g, h, a, b, c, d, 0x521afacaul);
Round(d, e, f, g, h, a, b, c, 0x31338431ul);
Round(c, d, e, f, g, h, a, b, 0x6ed41a95ul);
Round(b, c, d, e, f, g, h, a, 0x6d437890ul);
Round(a, b, c, d, e, f, g, h, 0xc39c91f2ul);
Round(h, a, b, c, d, e, f, g, 0x9eccabbdul);
Round(g, h, a, b, c, d, e, f, 0xb5c9a0e6ul);
Round(f, g, h, a, b, c, d, e, 0x532fb63cul);
Round(e, f, g, h, a, b, c, d, 0xd2c741c6ul);
Round(d, e, f, g, h, a, b, c, 0x07237ea3ul);
Round(c, d, e, f, g, h, a, b, 0xa4954b68ul);
Round(b, c, d, e, f, g, h, a, 0x4c191d76ul);
w0 = t0 + a;
w1 = t1 + b;
w2 = t2 + c;
w3 = t3 + d;
w4 = t4 + e;
w5 = t5 + f;
w6 = t6 + g;
w7 = t7 + h;
// Transform 3
a = 0x6a09e667ul;
b = 0xbb67ae85ul;
c = 0x3c6ef372ul;
d = 0xa54ff53aul;
e = 0x510e527ful;
f = 0x9b05688cul;
g = 0x1f83d9abul;
h = 0x5be0cd19ul;
Round(a, b, c, d, e, f, g, h, 0x428a2f98ul + w0);
Round(h, a, b, c, d, e, f, g, 0x71374491ul + w1);
Round(g, h, a, b, c, d, e, f, 0xb5c0fbcful + w2);
Round(f, g, h, a, b, c, d, e, 0xe9b5dba5ul + w3);
Round(e, f, g, h, a, b, c, d, 0x3956c25bul + w4);
Round(d, e, f, g, h, a, b, c, 0x59f111f1ul + w5);
Round(c, d, e, f, g, h, a, b, 0x923f82a4ul + w6);
Round(b, c, d, e, f, g, h, a, 0xab1c5ed5ul + w7);
Round(a, b, c, d, e, f, g, h, 0x5807aa98ul);
Round(h, a, b, c, d, e, f, g, 0x12835b01ul);
Round(g, h, a, b, c, d, e, f, 0x243185beul);
Round(f, g, h, a, b, c, d, e, 0x550c7dc3ul);
Round(e, f, g, h, a, b, c, d, 0x72be5d74ul);
Round(d, e, f, g, h, a, b, c, 0x80deb1feul);
Round(c, d, e, f, g, h, a, b, 0x9bdc06a7ul);
Round(b, c, d, e, f, g, h, a, 0xc19bf274ul);
Round(a, b, c, d, e, f, g, h, 0xe49b69c1ul + (w0 += sigma0(w1)));
Round(h, a, b, c, d, e, f, g, 0xefbe4786ul + (w1 += 0xa00000ul + sigma0(w2)));
Round(g, h, a, b, c, d, e, f, 0x0fc19dc6ul + (w2 += sigma1(w0) + sigma0(w3)));
Round(f, g, h, a, b, c, d, e, 0x240ca1ccul + (w3 += sigma1(w1) + sigma0(w4)));
Round(e, f, g, h, a, b, c, d, 0x2de92c6ful + (w4 += sigma1(w2) + sigma0(w5)));
Round(d, e, f, g, h, a, b, c, 0x4a7484aaul + (w5 += sigma1(w3) + sigma0(w6)));
Round(c, d, e, f, g, h, a, b, 0x5cb0a9dcul + (w6 += sigma1(w4) + 0x100ul + sigma0(w7)));
Round(b, c, d, e, f, g, h, a, 0x76f988daul + (w7 += sigma1(w5) + w0 + 0x11002000ul));
Round(a, b, c, d, e, f, g, h, 0x983e5152ul + (w8 = 0x80000000ul + sigma1(w6) + w1));
Round(h, a, b, c, d, e, f, g, 0xa831c66dul + (w9 = sigma1(w7) + w2));
Round(g, h, a, b, c, d, e, f, 0xb00327c8ul + (w10 = sigma1(w8) + w3));
Round(f, g, h, a, b, c, d, e, 0xbf597fc7ul + (w11 = sigma1(w9) + w4));
Round(e, f, g, h, a, b, c, d, 0xc6e00bf3ul + (w12 = sigma1(w10) + w5));
Round(d, e, f, g, h, a, b, c, 0xd5a79147ul + (w13 = sigma1(w11) + w6));
Round(c, d, e, f, g, h, a, b, 0x06ca6351ul + (w14 = sigma1(w12) + w7 + 0x400022ul));
Round(b, c, d, e, f, g, h, a, 0x14292967ul + (w15 = 0x100ul + sigma1(w13) + w8 + sigma0(w0)));
Round(a, b, c, d, e, f, g, h, 0x27b70a85ul + (w0 += sigma1(w14) + w9 + sigma0(w1)));
Round(h, a, b, c, d, e, f, g, 0x2e1b2138ul + (w1 += sigma1(w15) + w10 + sigma0(w2)));
Round(g, h, a, b, c, d, e, f, 0x4d2c6dfcul + (w2 += sigma1(w0) + w11 + sigma0(w3)));
Round(f, g, h, a, b, c, d, e, 0x53380d13ul + (w3 += sigma1(w1) + w12 + sigma0(w4)));
Round(e, f, g, h, a, b, c, d, 0x650a7354ul + (w4 += sigma1(w2) + w13 + sigma0(w5)));
Round(d, e, f, g, h, a, b, c, 0x766a0abbul + (w5 += sigma1(w3) + w14 + sigma0(w6)));
Round(c, d, e, f, g, h, a, b, 0x81c2c92eul + (w6 += sigma1(w4) + w15 + sigma0(w7)));
Round(b, c, d, e, f, g, h, a, 0x92722c85ul + (w7 += sigma1(w5) + w0 + sigma0(w8)));
Round(a, b, c, d, e, f, g, h, 0xa2bfe8a1ul + (w8 += sigma1(w6) + w1 + sigma0(w9)));
Round(h, a, b, c, d, e, f, g, 0xa81a664bul + (w9 += sigma1(w7) + w2 + sigma0(w10)));
Round(g, h, a, b, c, d, e, f, 0xc24b8b70ul + (w10 += sigma1(w8) + w3 + sigma0(w11)));
Round(f, g, h, a, b, c, d, e, 0xc76c51a3ul + (w11 += sigma1(w9) + w4 + sigma0(w12)));
Round(e, f, g, h, a, b, c, d, 0xd192e819ul + (w12 += sigma1(w10) + w5 + sigma0(w13)));
Round(d, e, f, g, h, a, b, c, 0xd6990624ul + (w13 += sigma1(w11) + w6 + sigma0(w14)));
Round(c, d, e, f, g, h, a, b, 0xf40e3585ul + (w14 += sigma1(w12) + w7 + sigma0(w15)));
Round(b, c, d, e, f, g, h, a, 0x106aa070ul + (w15 += sigma1(w13) + w8 + sigma0(w0)));
Round(a, b, c, d, e, f, g, h, 0x19a4c116ul + (w0 += sigma1(w14) + w9 + sigma0(w1)));
Round(h, a, b, c, d, e, f, g, 0x1e376c08ul + (w1 += sigma1(w15) + w10 + sigma0(w2)));
Round(g, h, a, b, c, d, e, f, 0x2748774cul + (w2 += sigma1(w0) + w11 + sigma0(w3)));
Round(f, g, h, a, b, c, d, e, 0x34b0bcb5ul + (w3 += sigma1(w1) + w12 + sigma0(w4)));
Round(e, f, g, h, a, b, c, d, 0x391c0cb3ul + (w4 += sigma1(w2) + w13 + sigma0(w5)));
Round(d, e, f, g, h, a, b, c, 0x4ed8aa4aul + (w5 += sigma1(w3) + w14 + sigma0(w6)));
Round(c, d, e, f, g, h, a, b, 0x5b9cca4ful + (w6 += sigma1(w4) + w15 + sigma0(w7)));
Round(b, c, d, e, f, g, h, a, 0x682e6ff3ul + (w7 += sigma1(w5) + w0 + sigma0(w8)));
Round(a, b, c, d, e, f, g, h, 0x748f82eeul + (w8 += sigma1(w6) + w1 + sigma0(w9)));
Round(h, a, b, c, d, e, f, g, 0x78a5636ful + (w9 += sigma1(w7) + w2 + sigma0(w10)));
Round(g, h, a, b, c, d, e, f, 0x84c87814ul + (w10 += sigma1(w8) + w3 + sigma0(w11)));
Round(f, g, h, a, b, c, d, e, 0x8cc70208ul + (w11 += sigma1(w9) + w4 + sigma0(w12)));
Round(e, f, g, h, a, b, c, d, 0x90befffaul + (w12 += sigma1(w10) + w5 + sigma0(w13)));
Round(d, e, f, g, h, a, b, c, 0xa4506cebul + (w13 += sigma1(w11) + w6 + sigma0(w14)));
Round(c, d, e, f, g, h, a, b, 0xbef9a3f7ul + (w14 + sigma1(w12) + w7 + sigma0(w15)));
Round(b, c, d, e, f, g, h, a, 0xc67178f2ul + (w15 + sigma1(w13) + w8 + sigma0(w0)));
// Output
WriteBE32(out + 0, a + 0x6a09e667ul);
WriteBE32(out + 4, b + 0xbb67ae85ul);
WriteBE32(out + 8, c + 0x3c6ef372ul);
WriteBE32(out + 12, d + 0xa54ff53aul);
WriteBE32(out + 16, e + 0x510e527ful);
WriteBE32(out + 20, f + 0x9b05688cul);
WriteBE32(out + 24, g + 0x1f83d9abul);
WriteBE32(out + 28, h + 0x5be0cd19ul);
}
} // namespace sha256
typedef void (*TransformType)(uint32_t*, const unsigned char*, size_t);
typedef void (*TransformD64Type)(unsigned char*, const unsigned char*);
template<TransformType tr>
void TransformD64Wrapper(unsigned char* out, const unsigned char* in)
{
uint32_t s[8];
static const unsigned char padding1[64] = {
0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0
};
unsigned char buffer2[64] = {
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0
};
sha256::Initialize(s);
tr(s, in, 1);
tr(s, padding1, 1);
WriteBE32(buffer2 + 0, s[0]);
WriteBE32(buffer2 + 4, s[1]);
WriteBE32(buffer2 + 8, s[2]);
WriteBE32(buffer2 + 12, s[3]);
WriteBE32(buffer2 + 16, s[4]);
WriteBE32(buffer2 + 20, s[5]);
WriteBE32(buffer2 + 24, s[6]);
WriteBE32(buffer2 + 28, s[7]);
sha256::Initialize(s);
tr(s, buffer2, 1);
WriteBE32(out + 0, s[0]);
WriteBE32(out + 4, s[1]);
WriteBE32(out + 8, s[2]);
WriteBE32(out + 12, s[3]);
WriteBE32(out + 16, s[4]);
WriteBE32(out + 20, s[5]);
WriteBE32(out + 24, s[6]);
WriteBE32(out + 28, s[7]);
}
TransformType Transform = sha256::Transform;
TransformD64Type TransformD64 = sha256::TransformD64;
TransformD64Type TransformD64_2way = nullptr;
TransformD64Type TransformD64_4way = nullptr;
TransformD64Type TransformD64_8way = nullptr;
bool SelfTest() {
// Input state (equal to the initial SHA256 state)
static const uint32_t init[8] = {
0x6a09e667ul, 0xbb67ae85ul, 0x3c6ef372ul, 0xa54ff53aul, 0x510e527ful, 0x9b05688cul, 0x1f83d9abul, 0x5be0cd19ul
};
// Some random input data to test with
static const unsigned char data[641] = "-" // Intentionally not aligned
"Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do "
"eiusmod tempor incididunt ut labore et dolore magna aliqua. Et m"
"olestie ac feugiat sed lectus vestibulum mattis ullamcorper. Mor"
"bi blandit cursus risus at ultrices mi tempus imperdiet nulla. N"
"unc congue nisi vita suscipit tellus mauris. Imperdiet proin fer"
"mentum leo vel orci. Massa tempor nec feugiat nisl pretium fusce"
" id velit. Telus in metus vulputate eu scelerisque felis. Mi tem"
"pus imperdiet nulla malesuada pellentesque. Tristique magna sit.";
// Expected output state for hashing the i*64 first input bytes above (excluding SHA256 padding).
static const uint32_t result[9][8] = {
{0x6a09e667ul, 0xbb67ae85ul, 0x3c6ef372ul, 0xa54ff53aul, 0x510e527ful, 0x9b05688cul, 0x1f83d9abul, 0x5be0cd19ul},
{0x91f8ec6bul, 0x4da10fe3ul, 0x1c9c292cul, 0x45e18185ul, 0x435cc111ul, 0x3ca26f09ul, 0xeb954caeul, 0x402a7069ul},
{0xcabea5acul, 0x374fb97cul, 0x182ad996ul, 0x7bd69cbful, 0x450ff900ul, 0xc1d2be8aul, 0x6a41d505ul, 0xe6212dc3ul},
{0xbcff09d6ul, 0x3e76f36eul, 0x3ecb2501ul, 0x78866e97ul, 0xe1c1e2fdul, 0x32f4eafful, 0x8aa6c4e5ul, 0xdfc024bcul},
{0xa08c5d94ul, 0x0a862f93ul, 0x6b7f2f40ul, 0x8f9fae76ul, 0x6d40439ful, 0x79dcee0cul, 0x3e39ff3aul, 0xdc3bdbb1ul},
{0x216a0895ul, 0x9f1a3662ul, 0xe99946f9ul, 0x87ba4364ul, 0x0fb5db2cul, 0x12bed3d3ul, 0x6689c0c7ul, 0x292f1b04ul},
{0xca3067f8ul, 0xbc8c2656ul, 0x37cb7e0dul, 0x9b6b8b0ful, 0x46dc380bul, 0xf1287f57ul, 0xc42e4b23ul, 0x3fefe94dul},
{0x3e4c4039ul, 0xbb6fca8cul, 0x6f27d2f7ul, 0x301e44a4ul, 0x8352ba14ul, 0x5769ce37ul, 0x48a1155ful, 0xc0e1c4c6ul},
{0xfe2fa9ddul, 0x69d0862bul, 0x1ae0db23ul, 0x471f9244ul, 0xf55c0145ul, 0xc30f9c3bul, 0x40a84ea0ul, 0x5b8a266cul},
};
// Expected output for each of the individual 8 64-byte messages under full double SHA256 (including padding).
static const unsigned char result_d64[256] = {
0x09, 0x3a, 0xc4, 0xd0, 0x0f, 0xf7, 0x57, 0xe1, 0x72, 0x85, 0x79, 0x42, 0xfe, 0xe7, 0xe0, 0xa0,
0xfc, 0x52, 0xd7, 0xdb, 0x07, 0x63, 0x45, 0xfb, 0x53, 0x14, 0x7d, 0x17, 0x22, 0x86, 0xf0, 0x52,
0x48, 0xb6, 0x11, 0x9e, 0x6e, 0x48, 0x81, 0x6d, 0xcc, 0x57, 0x1f, 0xb2, 0x97, 0xa8, 0xd5, 0x25,
0x9b, 0x82, 0xaa, 0x89, 0xe2, 0xfd, 0x2d, 0x56, 0xe8, 0x28, 0x83, 0x0b, 0xe2, 0xfa, 0x53, 0xb7,
0xd6, 0x6b, 0x07, 0x85, 0x83, 0xb0, 0x10, 0xa2, 0xf5, 0x51, 0x3c, 0xf9, 0x60, 0x03, 0xab, 0x45,
0x6c, 0x15, 0x6e, 0xef, 0xb5, 0xac, 0x3e, 0x6c, 0xdf, 0xb4, 0x92, 0x22, 0x2d, 0xce, 0xbf, 0x3e,
0xe9, 0xe5, 0xf6, 0x29, 0x0e, 0x01, 0x4f, 0xd2, 0xd4, 0x45, 0x65, 0xb3, 0xbb, 0xf2, 0x4c, 0x16,
0x37, 0x50, 0x3c, 0x6e, 0x49, 0x8c, 0x5a, 0x89, 0x2b, 0x1b, 0xab, 0xc4, 0x37, 0xd1, 0x46, 0xe9,
0x3d, 0x0e, 0x85, 0xa2, 0x50, 0x73, 0xa1, 0x5e, 0x54, 0x37, 0xd7, 0x94, 0x17, 0x56, 0xc2, 0xd8,
0xe5, 0x9f, 0xed, 0x4e, 0xae, 0x15, 0x42, 0x06, 0x0d, 0x74, 0x74, 0x5e, 0x24, 0x30, 0xce, 0xd1,
0x9e, 0x50, 0xa3, 0x9a, 0xb8, 0xf0, 0x4a, 0x57, 0x69, 0x78, 0x67, 0x12, 0x84, 0x58, 0xbe, 0xc7,
0x36, 0xaa, 0xee, 0x7c, 0x64, 0xa3, 0x76, 0xec, 0xff, 0x55, 0x41, 0x00, 0x2a, 0x44, 0x68, 0x4d,
0xb6, 0x53, 0x9e, 0x1c, 0x95, 0xb7, 0xca, 0xdc, 0x7f, 0x7d, 0x74, 0x27, 0x5c, 0x8e, 0xa6, 0x84,
0xb5, 0xac, 0x87, 0xa9, 0xf3, 0xff, 0x75, 0xf2, 0x34, 0xcd, 0x1a, 0x3b, 0x82, 0x2c, 0x2b, 0x4e,
0x6a, 0x46, 0x30, 0xa6, 0x89, 0x86, 0x23, 0xac, 0xf8, 0xa5, 0x15, 0xe9, 0x0a, 0xaa, 0x1e, 0x9a,
0xd7, 0x93, 0x6b, 0x28, 0xe4, 0x3b, 0xfd, 0x59, 0xc6, 0xed, 0x7c, 0x5f, 0xa5, 0x41, 0xcb, 0x51
};
// Test Transform() for 0 through 8 transformations.
for (size_t i = 0; i <= 8; ++i) {
uint32_t state[8];
std::copy(init, init + 8, state);
Transform(state, data + 1, i);
if (!std::equal(state, state + 8, result[i])) return false;
}
// Test TransformD64
unsigned char out[32];
TransformD64(out, data + 1);
if (!std::equal(out, out + 32, result_d64)) return false;
// Test TransformD64_2way, if available.
if (TransformD64_2way) {
unsigned char out[64];
TransformD64_2way(out, data + 1);
if (!std::equal(out, out + 64, result_d64)) return false;
}
// Test TransformD64_4way, if available.
if (TransformD64_4way) {
unsigned char out[128];
TransformD64_4way(out, data + 1);
if (!std::equal(out, out + 128, result_d64)) return false;
}
// Test TransformD64_8way, if available.
if (TransformD64_8way) {
unsigned char out[256];
TransformD64_8way(out, data + 1);
if (!std::equal(out, out + 256, result_d64)) return false;
}
return true;
}
#if defined(USE_ASM) && (defined(__x86_64__) || defined(__amd64__) || defined(__i386__))
/** Check whether the OS has enabled AVX registers. */
bool AVXEnabled()
{
uint32_t a, d;
__asm__("xgetbv" : "=a"(a), "=d"(d) : "c"(0));
return (a & 6) == 6;
}
#endif
} // namespace
std::string SHA256AutoDetect()
{
std::string ret = "standard";
#if defined(USE_ASM) && defined(HAVE_GETCPUID)
bool have_sse4 = false;
bool have_xsave = false;
bool have_avx = false;
bool have_avx2 = false;
bool have_shani = false;
bool enabled_avx = false;
(void)AVXEnabled;
(void)have_sse4;
(void)have_avx;
(void)have_xsave;
(void)have_avx2;
(void)have_shani;
(void)enabled_avx;
uint32_t eax, ebx, ecx, edx;
GetCPUID(1, 0, eax, ebx, ecx, edx);
have_sse4 = (ecx >> 19) & 1;
have_xsave = (ecx >> 27) & 1;
have_avx = (ecx >> 28) & 1;
if (have_xsave && have_avx) {
enabled_avx = AVXEnabled();
}
if (have_sse4) {
GetCPUID(7, 0, eax, ebx, ecx, edx);
have_avx2 = (ebx >> 5) & 1;
have_shani = (ebx >> 29) & 1;
}
#if defined(ENABLE_SHANI) && !defined(BUILD_VEXTA_INTERNAL)
if (have_shani) {
Transform = sha256_shani::Transform;
TransformD64 = TransformD64Wrapper<sha256_shani::Transform>;
TransformD64_2way = sha256d64_shani::Transform_2way;
ret = "shani(1way,2way)";
have_sse4 = false; // Disable SSE4/AVX2;
have_avx2 = false;
}
#endif
if (have_sse4) {
#if defined(__x86_64__) || defined(__amd64__)
Transform = sha256_sse4::Transform;
TransformD64 = TransformD64Wrapper<sha256_sse4::Transform>;
ret = "sse4(1way)";
#endif
#if defined(ENABLE_SSE41) && !defined(BUILD_VEXTA_INTERNAL)
TransformD64_4way = sha256d64_sse41::Transform_4way;
ret += ",sse41(4way)";
#endif
}
#if defined(ENABLE_AVX2) && !defined(BUILD_VEXTA_INTERNAL)
if (have_avx2 && have_avx && enabled_avx) {
TransformD64_8way = sha256d64_avx2::Transform_8way;
ret += ",avx2(8way)";
}
#endif
#endif
assert(SelfTest());
return ret;
}
////// SHA-256
CSHA256::CSHA256() : bytes(0)
{
sha256::Initialize(s);
}
CSHA256& CSHA256::Write(const unsigned char* data, size_t len)
{
const unsigned char* end = data + len;
size_t bufsize = bytes % 64;
if (bufsize && bufsize + len >= 64) {
// Fill the buffer, and process it.
memcpy(buf + bufsize, data, 64 - bufsize);
bytes += 64 - bufsize;
data += 64 - bufsize;
Transform(s, buf, 1);
bufsize = 0;
}
if (end - data >= 64) {
size_t blocks = (end - data) / 64;
Transform(s, data, blocks);
data += 64 * blocks;
bytes += 64 * blocks;
}
if (end > data) {
// Fill the buffer with what remains.
memcpy(buf + bufsize, data, end - data);
bytes += end - data;
}
return *this;
}
void CSHA256::Finalize(unsigned char hash[OUTPUT_SIZE])
{
static const unsigned char pad[64] = {0x80};
unsigned char sizedesc[8];
WriteBE64(sizedesc, bytes << 3);
Write(pad, 1 + ((119 - (bytes % 64)) % 64));
Write(sizedesc, 8);
WriteBE32(hash, s[0]);
WriteBE32(hash + 4, s[1]);
WriteBE32(hash + 8, s[2]);
WriteBE32(hash + 12, s[3]);
WriteBE32(hash + 16, s[4]);
WriteBE32(hash + 20, s[5]);
WriteBE32(hash + 24, s[6]);
WriteBE32(hash + 28, s[7]);
}
CSHA256& CSHA256::Reset()
{
bytes = 0;
sha256::Initialize(s);
return *this;
}
void SHA256D64(unsigned char* out, const unsigned char* in, size_t blocks)
{
if (TransformD64_8way) {
while (blocks >= 8) {
TransformD64_8way(out, in);
out += 256;
in += 512;
blocks -= 8;
}
}
if (TransformD64_4way) {
while (blocks >= 4) {
TransformD64_4way(out, in);
out += 128;
in += 256;
blocks -= 4;
}
}
if (TransformD64_2way) {
while (blocks >= 2) {
TransformD64_2way(out, in);
out += 64;
in += 128;
blocks -= 2;
}
}
while (blocks) {
TransformD64(out, in);
out += 32;
in += 64;
--blocks;
}
}

42
src/crypto/sha256.h Normal file
View file

@ -0,0 +1,42 @@
// Copyright (c) 2009-2020 The Bitcoin Core developers
// Copyright (c) 2014-2020 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#ifndef DIGIBYTE_CRYPTO_SHA256_H
#define DIGIBYTE_CRYPTO_SHA256_H
#include <stdint.h>
#include <stdlib.h>
#include <string>
/** A hasher class for SHA-256. */
class CSHA256
{
private:
uint32_t s[8];
unsigned char buf[64];
size_t bytes;
public:
static const size_t OUTPUT_SIZE = 32;
CSHA256();
CSHA256& Write(const unsigned char* data, size_t len);
void Finalize(unsigned char hash[OUTPUT_SIZE]);
CSHA256& Reset();
};
/** Autodetect the best available SHA256 implementation.
* Returns the name of the implementation.
*/
std::string SHA256AutoDetect();
/** Compute multiple double-SHA256's of 64-byte blobs.
* output: pointer to a blocks*32 byte output buffer
* input: pointer to a blocks*64 byte input buffer
* blocks: the number of hashes to compute.
*/
void SHA256D64(unsigned char* output, const unsigned char* input, size_t blocks);
#endif // DIGIBYTE_CRYPTO_SHA256_H

411
src/crypto/sha256_Y.cpp Normal file
View file

@ -0,0 +1,411 @@
/*-
* Copyright 2005,2007,2009 Colin Percival
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. 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 AUTHOR 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 AUTHOR 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.
*/
#include <sys/types.h>
#include <stdint.h>
#include <string.h>
#include "sysendian.h"
#include "sha256_Y.h"
/*
* Encode a length len/4 vector of (uint32_t) into a length len vector of
* (unsigned char) in big-endian form. Assumes len is a multiple of 4.
*/
static void
be32enc_vect(unsigned char *dst, const uint32_t *src, size_t len)
{
size_t i;
for (i = 0; i < len / 4; i++)
be32enc(dst + i * 4, src[i]);
}
/*
* Decode a big-endian length len vector of (unsigned char) into a length
* len/4 vector of (uint32_t). Assumes len is a multiple of 4.
*/
static void
be32dec_vect(uint32_t *dst, const unsigned char *src, size_t len)
{
size_t i;
for (i = 0; i < len / 4; i++)
dst[i] = be32dec(src + i * 4);
}
/* Elementary functions used by SHA256 */
#define Ch(x, y, z) ((x & (y ^ z)) ^ z)
#define Maj(x, y, z) ((x & (y | z)) | (y & z))
#define SHR(x, n) (x >> n)
#define ROTR(x, n) ((x >> n) | (x << (32 - n)))
#define S0(x) (ROTR(x, 2) ^ ROTR(x, 13) ^ ROTR(x, 22))
#define S1(x) (ROTR(x, 6) ^ ROTR(x, 11) ^ ROTR(x, 25))
#define s0(x) (ROTR(x, 7) ^ ROTR(x, 18) ^ SHR(x, 3))
#define s1(x) (ROTR(x, 17) ^ ROTR(x, 19) ^ SHR(x, 10))
/* SHA256 round function */
#define RND(a, b, c, d, e, f, g, h, k) \
t0 = h + S1(e) + Ch(e, f, g) + k; \
t1 = S0(a) + Maj(a, b, c); \
d += t0; \
h = t0 + t1;
/* Adjusted round function for rotating state */
#define RNDr(S, W, i, k) \
RND(S[(64 - i) % 8], S[(65 - i) % 8], \
S[(66 - i) % 8], S[(67 - i) % 8], \
S[(68 - i) % 8], S[(69 - i) % 8], \
S[(70 - i) % 8], S[(71 - i) % 8], \
W[i] + k)
/*
* SHA256 block compression function. The 256-bit state is transformed via
* the 512-bit input block to produce a new state.
*/
static void
SHA256_Transform(uint32_t * state, const unsigned char block[64])
{
uint32_t W[64];
uint32_t S[8];
uint32_t t0, t1;
int i;
/* 1. Prepare message schedule W. */
be32dec_vect(W, block, 64);
for (i = 16; i < 64; i++)
W[i] = s1(W[i - 2]) + W[i - 7] + s0(W[i - 15]) + W[i - 16];
/* 2. Initialize working variables. */
memcpy(S, state, 32);
/* 3. Mix. */
RNDr(S, W, 0, 0x428a2f98);
RNDr(S, W, 1, 0x71374491);
RNDr(S, W, 2, 0xb5c0fbcf);
RNDr(S, W, 3, 0xe9b5dba5);
RNDr(S, W, 4, 0x3956c25b);
RNDr(S, W, 5, 0x59f111f1);
RNDr(S, W, 6, 0x923f82a4);
RNDr(S, W, 7, 0xab1c5ed5);
RNDr(S, W, 8, 0xd807aa98);
RNDr(S, W, 9, 0x12835b01);
RNDr(S, W, 10, 0x243185be);
RNDr(S, W, 11, 0x550c7dc3);
RNDr(S, W, 12, 0x72be5d74);
RNDr(S, W, 13, 0x80deb1fe);
RNDr(S, W, 14, 0x9bdc06a7);
RNDr(S, W, 15, 0xc19bf174);
RNDr(S, W, 16, 0xe49b69c1);
RNDr(S, W, 17, 0xefbe4786);
RNDr(S, W, 18, 0x0fc19dc6);
RNDr(S, W, 19, 0x240ca1cc);
RNDr(S, W, 20, 0x2de92c6f);
RNDr(S, W, 21, 0x4a7484aa);
RNDr(S, W, 22, 0x5cb0a9dc);
RNDr(S, W, 23, 0x76f988da);
RNDr(S, W, 24, 0x983e5152);
RNDr(S, W, 25, 0xa831c66d);
RNDr(S, W, 26, 0xb00327c8);
RNDr(S, W, 27, 0xbf597fc7);
RNDr(S, W, 28, 0xc6e00bf3);
RNDr(S, W, 29, 0xd5a79147);
RNDr(S, W, 30, 0x06ca6351);
RNDr(S, W, 31, 0x14292967);
RNDr(S, W, 32, 0x27b70a85);
RNDr(S, W, 33, 0x2e1b2138);
RNDr(S, W, 34, 0x4d2c6dfc);
RNDr(S, W, 35, 0x53380d13);
RNDr(S, W, 36, 0x650a7354);
RNDr(S, W, 37, 0x766a0abb);
RNDr(S, W, 38, 0x81c2c92e);
RNDr(S, W, 39, 0x92722c85);
RNDr(S, W, 40, 0xa2bfe8a1);
RNDr(S, W, 41, 0xa81a664b);
RNDr(S, W, 42, 0xc24b8b70);
RNDr(S, W, 43, 0xc76c51a3);
RNDr(S, W, 44, 0xd192e819);
RNDr(S, W, 45, 0xd6990624);
RNDr(S, W, 46, 0xf40e3585);
RNDr(S, W, 47, 0x106aa070);
RNDr(S, W, 48, 0x19a4c116);
RNDr(S, W, 49, 0x1e376c08);
RNDr(S, W, 50, 0x2748774c);
RNDr(S, W, 51, 0x34b0bcb5);
RNDr(S, W, 52, 0x391c0cb3);
RNDr(S, W, 53, 0x4ed8aa4a);
RNDr(S, W, 54, 0x5b9cca4f);
RNDr(S, W, 55, 0x682e6ff3);
RNDr(S, W, 56, 0x748f82ee);
RNDr(S, W, 57, 0x78a5636f);
RNDr(S, W, 58, 0x84c87814);
RNDr(S, W, 59, 0x8cc70208);
RNDr(S, W, 60, 0x90befffa);
RNDr(S, W, 61, 0xa4506ceb);
RNDr(S, W, 62, 0xbef9a3f7);
RNDr(S, W, 63, 0xc67178f2);
/* 4. Mix local working variables into global state */
for (i = 0; i < 8; i++)
state[i] += S[i];
/* Clean the stack. */
memset(W, 0, 256);
memset(S, 0, 32);
t0 = t1 = 0;
}
static unsigned char PAD[64] = {
0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
};
/* Add padding and terminating bit-count. */
static void
SHA256_Pad(SHA256_CTX_Y * ctx)
{
unsigned char len[8];
uint32_t r, plen;
/*
* Convert length to a vector of bytes -- we do this now rather
* than later because the length will change after we pad.
*/
be32enc_vect(len, ctx->count, 8);
/* Add 1--64 bytes so that the resulting length is 56 mod 64 */
r = (ctx->count[1] >> 3) & 0x3f;
plen = (r < 56) ? (56 - r) : (120 - r);
SHA256_Update_Y(ctx, PAD, (size_t)plen);
/* Add the terminating bit-count */
SHA256_Update_Y(ctx, len, 8);
}
/* SHA-256 initialization. Begins a SHA-256 operation. */
void
SHA256_Init_Y(SHA256_CTX_Y * ctx)
{
/* Zero bits processed so far */
ctx->count[0] = ctx->count[1] = 0;
/* Magic initialization constants */
ctx->state[0] = 0x6A09E667;
ctx->state[1] = 0xBB67AE85;
ctx->state[2] = 0x3C6EF372;
ctx->state[3] = 0xA54FF53A;
ctx->state[4] = 0x510E527F;
ctx->state[5] = 0x9B05688C;
ctx->state[6] = 0x1F83D9AB;
ctx->state[7] = 0x5BE0CD19;
}
/* Add bytes into the hash */
void
SHA256_Update_Y(SHA256_CTX_Y * ctx, const void *in, size_t len)
{
uint32_t bitlen[2];
uint32_t r;
const unsigned char *src = in;
/* Number of bytes left in the buffer from previous updates */
r = (ctx->count[1] >> 3) & 0x3f;
/* Convert the length into a number of bits */
bitlen[1] = ((uint32_t)len) << 3;
bitlen[0] = (uint32_t)(len >> 29);
/* Update number of bits */
if ((ctx->count[1] += bitlen[1]) < bitlen[1])
ctx->count[0]++;
ctx->count[0] += bitlen[0];
/* Handle the case where we don't need to perform any transforms */
if (len < 64 - r) {
memcpy(&ctx->buf[r], src, len);
return;
}
/* Finish the current block */
memcpy(&ctx->buf[r], src, 64 - r);
SHA256_Transform(ctx->state, ctx->buf);
src += 64 - r;
len -= 64 - r;
/* Perform complete blocks */
while (len >= 64) {
SHA256_Transform(ctx->state, src);
src += 64;
len -= 64;
}
/* Copy left over data into buffer */
memcpy(ctx->buf, src, len);
}
/*
* SHA-256 finalization. Pads the input data, exports the hash value,
* and clears the context state.
*/
void
SHA256_Final_Y(unsigned char digest[32], SHA256_CTX_Y * ctx)
{
/* Add padding */
SHA256_Pad(ctx);
/* Write the hash */
be32enc_vect(digest, ctx->state, 32);
/* Clear the context state */
memset((void *)ctx, 0, sizeof(*ctx));
}
/* Initialize an HMAC-SHA256 operation with the given key. */
void
HMAC_SHA256_Init_Y(HMAC_SHA256_CTX_Y * ctx, const void * _K, size_t Klen)
{
unsigned char pad[64];
unsigned char khash[32];
const unsigned char * K = _K;
size_t i;
/* If Klen > 64, the key is really SHA256(K). */
if (Klen > 64) {
SHA256_Init_Y(&ctx->ictx);
SHA256_Update_Y(&ctx->ictx, K, Klen);
SHA256_Final_Y(khash, &ctx->ictx);
K = khash;
Klen = 32;
}
/* Inner SHA256 operation is SHA256(K xor [block of 0x36] || data). */
SHA256_Init_Y(&ctx->ictx);
memset(pad, 0x36, 64);
for (i = 0; i < Klen; i++)
pad[i] ^= K[i];
SHA256_Update_Y(&ctx->ictx, pad, 64);
/* Outer SHA256 operation is SHA256(K xor [block of 0x5c] || hash). */
SHA256_Init_Y(&ctx->octx);
memset(pad, 0x5c, 64);
for (i = 0; i < Klen; i++)
pad[i] ^= K[i];
SHA256_Update_Y(&ctx->octx, pad, 64);
/* Clean the stack. */
memset(khash, 0, 32);
}
/* Add bytes to the HMAC-SHA256 operation. */
void
HMAC_SHA256_Update_Y(HMAC_SHA256_CTX_Y * ctx, const void *in, size_t len)
{
/* Feed data to the inner SHA256 operation. */
SHA256_Update_Y(&ctx->ictx, in, len);
}
/* Finish an HMAC-SHA256 operation. */
void
HMAC_SHA256_Final_Y(unsigned char digest[32], HMAC_SHA256_CTX_Y * ctx)
{
unsigned char ihash[32];
/* Finish the inner SHA256 operation. */
SHA256_Final_Y(ihash, &ctx->ictx);
/* Feed the inner hash to the outer SHA256 operation. */
SHA256_Update_Y(&ctx->octx, ihash, 32);
/* Finish the outer SHA256 operation. */
SHA256_Final_Y(digest, &ctx->octx);
/* Clean the stack. */
memset(ihash, 0, 32);
}
/**
* PBKDF2_SHA256(passwd, passwdlen, salt, saltlen, c, buf, dkLen):
* Compute PBKDF2(passwd, salt, c, dkLen) using HMAC-SHA256 as the PRF, and
* write the output to buf. The value dkLen must be at most 32 * (2^32 - 1).
*/
void
PBKDF2_SHA256(const uint8_t * passwd, size_t passwdlen, const uint8_t * salt,
size_t saltlen, uint64_t c, uint8_t * buf, size_t dkLen)
{
HMAC_SHA256_CTX_Y PShctx, hctx;
size_t i;
uint8_t ivec[4];
uint8_t U[32];
uint8_t T[32];
uint64_t j;
int k;
size_t clen;
/* Compute HMAC state after processing P and S. */
HMAC_SHA256_Init_Y(&PShctx, passwd, passwdlen);
HMAC_SHA256_Update_Y(&PShctx, salt, saltlen);
/* Iterate through the blocks. */
for (i = 0; i * 32 < dkLen; i++) {
/* Generate INT(i + 1). */
be32enc(ivec, (uint32_t)(i + 1));
/* Compute U_1 = PRF(P, S || INT(i)). */
memcpy(&hctx, &PShctx, sizeof(HMAC_SHA256_CTX_Y));
HMAC_SHA256_Update_Y(&hctx, ivec, 4);
HMAC_SHA256_Final_Y(U, &hctx);
/* T_i = U_1 ... */
memcpy(T, U, 32);
for (j = 2; j <= c; j++) {
/* Compute U_j. */
HMAC_SHA256_Init_Y(&hctx, passwd, passwdlen);
HMAC_SHA256_Update_Y(&hctx, U, 32);
HMAC_SHA256_Final_Y(U, &hctx);
/* ... xor U_j ... */
for (k = 0; k < 32; k++)
T[k] ^= U[k];
}
/* Copy as many bytes as necessary into buf. */
clen = dkLen - i * 32;
if (clen > 32)
clen = 32;
memcpy(&buf[i * 32], T, clen);
}
/* Clean PShctx, since we never called _Final on it. */
memset(&PShctx, 0, sizeof(HMAC_SHA256_CTX_Y));
}

62
src/crypto/sha256_Y.h Normal file
View file

@ -0,0 +1,62 @@
/*-
* Copyright 2005,2007,2009 Colin Percival
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. 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 AUTHOR 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 AUTHOR 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.
*
* $FreeBSD: src/lib/libmd/sha256_Y.h,v 1.2 2006/01/17 15:35:56 phk Exp $
*/
#ifndef _SHA256_H_
#define _SHA256_H_
#include <sys/types.h>
#include <stdint.h>
typedef struct SHA256Context {
uint32_t state[8];
uint32_t count[2];
unsigned char buf[64];
} SHA256_CTX_Y;
typedef struct HMAC_SHA256Context {
SHA256_CTX_Y ictx;
SHA256_CTX_Y octx;
} HMAC_SHA256_CTX_Y;
void SHA256_Init_Y(SHA256_CTX_Y *);
void SHA256_Update_Y(SHA256_CTX_Y *, const void *, size_t);
void SHA256_Final_Y(unsigned char [32], SHA256_CTX_Y *);
void HMAC_SHA256_Init_Y(HMAC_SHA256_CTX_Y *, const void *, size_t);
void HMAC_SHA256_Update_Y(HMAC_SHA256_CTX_Y *, const void *, size_t);
void HMAC_SHA256_Final_Y(unsigned char [32], HMAC_SHA256_CTX_Y *);
/**
* PBKDF2_SHA256(passwd, passwdlen, salt, saltlen, c, buf, dkLen):
* Compute PBKDF2(passwd, salt, c, dkLen) using HMAC-SHA256 as the PRF, and
* write the output to buf. The value dkLen must be at most 32 * (2^32 - 1).
*/
void PBKDF2_SHA256(const uint8_t *, size_t, const uint8_t *, size_t,
uint64_t, uint8_t *, size_t);
#endif /* !_SHA256_H_ */

328
src/crypto/sha256_avx2.cpp Normal file
View file

@ -0,0 +1,328 @@
// Copyright (c) 2017-2019 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#ifdef ENABLE_AVX2
#include <stdint.h>
#include <immintrin.h>
#include <crypto/common.h>
namespace sha256d64_avx2 {
namespace {
__m256i inline K(uint32_t x) { return _mm256_set1_epi32(x); }
__m256i inline Add(__m256i x, __m256i y) { return _mm256_add_epi32(x, y); }
__m256i inline Add(__m256i x, __m256i y, __m256i z) { return Add(Add(x, y), z); }
__m256i inline Add(__m256i x, __m256i y, __m256i z, __m256i w) { return Add(Add(x, y), Add(z, w)); }
__m256i inline Add(__m256i x, __m256i y, __m256i z, __m256i w, __m256i v) { return Add(Add(x, y, z), Add(w, v)); }
__m256i inline Inc(__m256i& x, __m256i y) { x = Add(x, y); return x; }
__m256i inline Inc(__m256i& x, __m256i y, __m256i z) { x = Add(x, y, z); return x; }
__m256i inline Inc(__m256i& x, __m256i y, __m256i z, __m256i w) { x = Add(x, y, z, w); return x; }
__m256i inline Xor(__m256i x, __m256i y) { return _mm256_xor_si256(x, y); }
__m256i inline Xor(__m256i x, __m256i y, __m256i z) { return Xor(Xor(x, y), z); }
__m256i inline Or(__m256i x, __m256i y) { return _mm256_or_si256(x, y); }
__m256i inline And(__m256i x, __m256i y) { return _mm256_and_si256(x, y); }
__m256i inline ShR(__m256i x, int n) { return _mm256_srli_epi32(x, n); }
__m256i inline ShL(__m256i x, int n) { return _mm256_slli_epi32(x, n); }
__m256i inline Ch(__m256i x, __m256i y, __m256i z) { return Xor(z, And(x, Xor(y, z))); }
__m256i inline Maj(__m256i x, __m256i y, __m256i z) { return Or(And(x, y), And(z, Or(x, y))); }
__m256i inline Sigma0(__m256i x) { return Xor(Or(ShR(x, 2), ShL(x, 30)), Or(ShR(x, 13), ShL(x, 19)), Or(ShR(x, 22), ShL(x, 10))); }
__m256i inline Sigma1(__m256i x) { return Xor(Or(ShR(x, 6), ShL(x, 26)), Or(ShR(x, 11), ShL(x, 21)), Or(ShR(x, 25), ShL(x, 7))); }
__m256i inline sigma0(__m256i x) { return Xor(Or(ShR(x, 7), ShL(x, 25)), Or(ShR(x, 18), ShL(x, 14)), ShR(x, 3)); }
__m256i inline sigma1(__m256i x) { return Xor(Or(ShR(x, 17), ShL(x, 15)), Or(ShR(x, 19), ShL(x, 13)), ShR(x, 10)); }
/** One round of SHA-256. */
void inline __attribute__((always_inline)) Round(__m256i a, __m256i b, __m256i c, __m256i& d, __m256i e, __m256i f, __m256i g, __m256i& h, __m256i k)
{
__m256i t1 = Add(h, Sigma1(e), Ch(e, f, g), k);
__m256i t2 = Add(Sigma0(a), Maj(a, b, c));
d = Add(d, t1);
h = Add(t1, t2);
}
__m256i inline Read8(const unsigned char* chunk, int offset) {
__m256i ret = _mm256_set_epi32(
ReadLE32(chunk + 0 + offset),
ReadLE32(chunk + 64 + offset),
ReadLE32(chunk + 128 + offset),
ReadLE32(chunk + 192 + offset),
ReadLE32(chunk + 256 + offset),
ReadLE32(chunk + 320 + offset),
ReadLE32(chunk + 384 + offset),
ReadLE32(chunk + 448 + offset)
);
return _mm256_shuffle_epi8(ret, _mm256_set_epi32(0x0C0D0E0FUL, 0x08090A0BUL, 0x04050607UL, 0x00010203UL, 0x0C0D0E0FUL, 0x08090A0BUL, 0x04050607UL, 0x00010203UL));
}
void inline Write8(unsigned char* out, int offset, __m256i v) {
v = _mm256_shuffle_epi8(v, _mm256_set_epi32(0x0C0D0E0FUL, 0x08090A0BUL, 0x04050607UL, 0x00010203UL, 0x0C0D0E0FUL, 0x08090A0BUL, 0x04050607UL, 0x00010203UL));
WriteLE32(out + 0 + offset, _mm256_extract_epi32(v, 7));
WriteLE32(out + 32 + offset, _mm256_extract_epi32(v, 6));
WriteLE32(out + 64 + offset, _mm256_extract_epi32(v, 5));
WriteLE32(out + 96 + offset, _mm256_extract_epi32(v, 4));
WriteLE32(out + 128 + offset, _mm256_extract_epi32(v, 3));
WriteLE32(out + 160 + offset, _mm256_extract_epi32(v, 2));
WriteLE32(out + 192 + offset, _mm256_extract_epi32(v, 1));
WriteLE32(out + 224 + offset, _mm256_extract_epi32(v, 0));
}
}
void Transform_8way(unsigned char* out, const unsigned char* in)
{
// Transform 1
__m256i a = K(0x6a09e667ul);
__m256i b = K(0xbb67ae85ul);
__m256i c = K(0x3c6ef372ul);
__m256i d = K(0xa54ff53aul);
__m256i e = K(0x510e527ful);
__m256i f = K(0x9b05688cul);
__m256i g = K(0x1f83d9abul);
__m256i h = K(0x5be0cd19ul);
__m256i w0, w1, w2, w3, w4, w5, w6, w7, w8, w9, w10, w11, w12, w13, w14, w15;
Round(a, b, c, d, e, f, g, h, Add(K(0x428a2f98ul), w0 = Read8(in, 0)));
Round(h, a, b, c, d, e, f, g, Add(K(0x71374491ul), w1 = Read8(in, 4)));
Round(g, h, a, b, c, d, e, f, Add(K(0xb5c0fbcful), w2 = Read8(in, 8)));
Round(f, g, h, a, b, c, d, e, Add(K(0xe9b5dba5ul), w3 = Read8(in, 12)));
Round(e, f, g, h, a, b, c, d, Add(K(0x3956c25bul), w4 = Read8(in, 16)));
Round(d, e, f, g, h, a, b, c, Add(K(0x59f111f1ul), w5 = Read8(in, 20)));
Round(c, d, e, f, g, h, a, b, Add(K(0x923f82a4ul), w6 = Read8(in, 24)));
Round(b, c, d, e, f, g, h, a, Add(K(0xab1c5ed5ul), w7 = Read8(in, 28)));
Round(a, b, c, d, e, f, g, h, Add(K(0xd807aa98ul), w8 = Read8(in, 32)));
Round(h, a, b, c, d, e, f, g, Add(K(0x12835b01ul), w9 = Read8(in, 36)));
Round(g, h, a, b, c, d, e, f, Add(K(0x243185beul), w10 = Read8(in, 40)));
Round(f, g, h, a, b, c, d, e, Add(K(0x550c7dc3ul), w11 = Read8(in, 44)));
Round(e, f, g, h, a, b, c, d, Add(K(0x72be5d74ul), w12 = Read8(in, 48)));
Round(d, e, f, g, h, a, b, c, Add(K(0x80deb1feul), w13 = Read8(in, 52)));
Round(c, d, e, f, g, h, a, b, Add(K(0x9bdc06a7ul), w14 = Read8(in, 56)));
Round(b, c, d, e, f, g, h, a, Add(K(0xc19bf174ul), w15 = Read8(in, 60)));
Round(a, b, c, d, e, f, g, h, Add(K(0xe49b69c1ul), Inc(w0, sigma1(w14), w9, sigma0(w1))));
Round(h, a, b, c, d, e, f, g, Add(K(0xefbe4786ul), Inc(w1, sigma1(w15), w10, sigma0(w2))));
Round(g, h, a, b, c, d, e, f, Add(K(0x0fc19dc6ul), Inc(w2, sigma1(w0), w11, sigma0(w3))));
Round(f, g, h, a, b, c, d, e, Add(K(0x240ca1ccul), Inc(w3, sigma1(w1), w12, sigma0(w4))));
Round(e, f, g, h, a, b, c, d, Add(K(0x2de92c6ful), Inc(w4, sigma1(w2), w13, sigma0(w5))));
Round(d, e, f, g, h, a, b, c, Add(K(0x4a7484aaul), Inc(w5, sigma1(w3), w14, sigma0(w6))));
Round(c, d, e, f, g, h, a, b, Add(K(0x5cb0a9dcul), Inc(w6, sigma1(w4), w15, sigma0(w7))));
Round(b, c, d, e, f, g, h, a, Add(K(0x76f988daul), Inc(w7, sigma1(w5), w0, sigma0(w8))));
Round(a, b, c, d, e, f, g, h, Add(K(0x983e5152ul), Inc(w8, sigma1(w6), w1, sigma0(w9))));
Round(h, a, b, c, d, e, f, g, Add(K(0xa831c66dul), Inc(w9, sigma1(w7), w2, sigma0(w10))));
Round(g, h, a, b, c, d, e, f, Add(K(0xb00327c8ul), Inc(w10, sigma1(w8), w3, sigma0(w11))));
Round(f, g, h, a, b, c, d, e, Add(K(0xbf597fc7ul), Inc(w11, sigma1(w9), w4, sigma0(w12))));
Round(e, f, g, h, a, b, c, d, Add(K(0xc6e00bf3ul), Inc(w12, sigma1(w10), w5, sigma0(w13))));
Round(d, e, f, g, h, a, b, c, Add(K(0xd5a79147ul), Inc(w13, sigma1(w11), w6, sigma0(w14))));
Round(c, d, e, f, g, h, a, b, Add(K(0x06ca6351ul), Inc(w14, sigma1(w12), w7, sigma0(w15))));
Round(b, c, d, e, f, g, h, a, Add(K(0x14292967ul), Inc(w15, sigma1(w13), w8, sigma0(w0))));
Round(a, b, c, d, e, f, g, h, Add(K(0x27b70a85ul), Inc(w0, sigma1(w14), w9, sigma0(w1))));
Round(h, a, b, c, d, e, f, g, Add(K(0x2e1b2138ul), Inc(w1, sigma1(w15), w10, sigma0(w2))));
Round(g, h, a, b, c, d, e, f, Add(K(0x4d2c6dfcul), Inc(w2, sigma1(w0), w11, sigma0(w3))));
Round(f, g, h, a, b, c, d, e, Add(K(0x53380d13ul), Inc(w3, sigma1(w1), w12, sigma0(w4))));
Round(e, f, g, h, a, b, c, d, Add(K(0x650a7354ul), Inc(w4, sigma1(w2), w13, sigma0(w5))));
Round(d, e, f, g, h, a, b, c, Add(K(0x766a0abbul), Inc(w5, sigma1(w3), w14, sigma0(w6))));
Round(c, d, e, f, g, h, a, b, Add(K(0x81c2c92eul), Inc(w6, sigma1(w4), w15, sigma0(w7))));
Round(b, c, d, e, f, g, h, a, Add(K(0x92722c85ul), Inc(w7, sigma1(w5), w0, sigma0(w8))));
Round(a, b, c, d, e, f, g, h, Add(K(0xa2bfe8a1ul), Inc(w8, sigma1(w6), w1, sigma0(w9))));
Round(h, a, b, c, d, e, f, g, Add(K(0xa81a664bul), Inc(w9, sigma1(w7), w2, sigma0(w10))));
Round(g, h, a, b, c, d, e, f, Add(K(0xc24b8b70ul), Inc(w10, sigma1(w8), w3, sigma0(w11))));
Round(f, g, h, a, b, c, d, e, Add(K(0xc76c51a3ul), Inc(w11, sigma1(w9), w4, sigma0(w12))));
Round(e, f, g, h, a, b, c, d, Add(K(0xd192e819ul), Inc(w12, sigma1(w10), w5, sigma0(w13))));
Round(d, e, f, g, h, a, b, c, Add(K(0xd6990624ul), Inc(w13, sigma1(w11), w6, sigma0(w14))));
Round(c, d, e, f, g, h, a, b, Add(K(0xf40e3585ul), Inc(w14, sigma1(w12), w7, sigma0(w15))));
Round(b, c, d, e, f, g, h, a, Add(K(0x106aa070ul), Inc(w15, sigma1(w13), w8, sigma0(w0))));
Round(a, b, c, d, e, f, g, h, Add(K(0x19a4c116ul), Inc(w0, sigma1(w14), w9, sigma0(w1))));
Round(h, a, b, c, d, e, f, g, Add(K(0x1e376c08ul), Inc(w1, sigma1(w15), w10, sigma0(w2))));
Round(g, h, a, b, c, d, e, f, Add(K(0x2748774cul), Inc(w2, sigma1(w0), w11, sigma0(w3))));
Round(f, g, h, a, b, c, d, e, Add(K(0x34b0bcb5ul), Inc(w3, sigma1(w1), w12, sigma0(w4))));
Round(e, f, g, h, a, b, c, d, Add(K(0x391c0cb3ul), Inc(w4, sigma1(w2), w13, sigma0(w5))));
Round(d, e, f, g, h, a, b, c, Add(K(0x4ed8aa4aul), Inc(w5, sigma1(w3), w14, sigma0(w6))));
Round(c, d, e, f, g, h, a, b, Add(K(0x5b9cca4ful), Inc(w6, sigma1(w4), w15, sigma0(w7))));
Round(b, c, d, e, f, g, h, a, Add(K(0x682e6ff3ul), Inc(w7, sigma1(w5), w0, sigma0(w8))));
Round(a, b, c, d, e, f, g, h, Add(K(0x748f82eeul), Inc(w8, sigma1(w6), w1, sigma0(w9))));
Round(h, a, b, c, d, e, f, g, Add(K(0x78a5636ful), Inc(w9, sigma1(w7), w2, sigma0(w10))));
Round(g, h, a, b, c, d, e, f, Add(K(0x84c87814ul), Inc(w10, sigma1(w8), w3, sigma0(w11))));
Round(f, g, h, a, b, c, d, e, Add(K(0x8cc70208ul), Inc(w11, sigma1(w9), w4, sigma0(w12))));
Round(e, f, g, h, a, b, c, d, Add(K(0x90befffaul), Inc(w12, sigma1(w10), w5, sigma0(w13))));
Round(d, e, f, g, h, a, b, c, Add(K(0xa4506cebul), Inc(w13, sigma1(w11), w6, sigma0(w14))));
Round(c, d, e, f, g, h, a, b, Add(K(0xbef9a3f7ul), Inc(w14, sigma1(w12), w7, sigma0(w15))));
Round(b, c, d, e, f, g, h, a, Add(K(0xc67178f2ul), Inc(w15, sigma1(w13), w8, sigma0(w0))));
a = Add(a, K(0x6a09e667ul));
b = Add(b, K(0xbb67ae85ul));
c = Add(c, K(0x3c6ef372ul));
d = Add(d, K(0xa54ff53aul));
e = Add(e, K(0x510e527ful));
f = Add(f, K(0x9b05688cul));
g = Add(g, K(0x1f83d9abul));
h = Add(h, K(0x5be0cd19ul));
__m256i t0 = a, t1 = b, t2 = c, t3 = d, t4 = e, t5 = f, t6 = g, t7 = h;
// Transform 2
Round(a, b, c, d, e, f, g, h, K(0xc28a2f98ul));
Round(h, a, b, c, d, e, f, g, K(0x71374491ul));
Round(g, h, a, b, c, d, e, f, K(0xb5c0fbcful));
Round(f, g, h, a, b, c, d, e, K(0xe9b5dba5ul));
Round(e, f, g, h, a, b, c, d, K(0x3956c25bul));
Round(d, e, f, g, h, a, b, c, K(0x59f111f1ul));
Round(c, d, e, f, g, h, a, b, K(0x923f82a4ul));
Round(b, c, d, e, f, g, h, a, K(0xab1c5ed5ul));
Round(a, b, c, d, e, f, g, h, K(0xd807aa98ul));
Round(h, a, b, c, d, e, f, g, K(0x12835b01ul));
Round(g, h, a, b, c, d, e, f, K(0x243185beul));
Round(f, g, h, a, b, c, d, e, K(0x550c7dc3ul));
Round(e, f, g, h, a, b, c, d, K(0x72be5d74ul));
Round(d, e, f, g, h, a, b, c, K(0x80deb1feul));
Round(c, d, e, f, g, h, a, b, K(0x9bdc06a7ul));
Round(b, c, d, e, f, g, h, a, K(0xc19bf374ul));
Round(a, b, c, d, e, f, g, h, K(0x649b69c1ul));
Round(h, a, b, c, d, e, f, g, K(0xf0fe4786ul));
Round(g, h, a, b, c, d, e, f, K(0x0fe1edc6ul));
Round(f, g, h, a, b, c, d, e, K(0x240cf254ul));
Round(e, f, g, h, a, b, c, d, K(0x4fe9346ful));
Round(d, e, f, g, h, a, b, c, K(0x6cc984beul));
Round(c, d, e, f, g, h, a, b, K(0x61b9411eul));
Round(b, c, d, e, f, g, h, a, K(0x16f988faul));
Round(a, b, c, d, e, f, g, h, K(0xf2c65152ul));
Round(h, a, b, c, d, e, f, g, K(0xa88e5a6dul));
Round(g, h, a, b, c, d, e, f, K(0xb019fc65ul));
Round(f, g, h, a, b, c, d, e, K(0xb9d99ec7ul));
Round(e, f, g, h, a, b, c, d, K(0x9a1231c3ul));
Round(d, e, f, g, h, a, b, c, K(0xe70eeaa0ul));
Round(c, d, e, f, g, h, a, b, K(0xfdb1232bul));
Round(b, c, d, e, f, g, h, a, K(0xc7353eb0ul));
Round(a, b, c, d, e, f, g, h, K(0x3069bad5ul));
Round(h, a, b, c, d, e, f, g, K(0xcb976d5ful));
Round(g, h, a, b, c, d, e, f, K(0x5a0f118ful));
Round(f, g, h, a, b, c, d, e, K(0xdc1eeefdul));
Round(e, f, g, h, a, b, c, d, K(0x0a35b689ul));
Round(d, e, f, g, h, a, b, c, K(0xde0b7a04ul));
Round(c, d, e, f, g, h, a, b, K(0x58f4ca9dul));
Round(b, c, d, e, f, g, h, a, K(0xe15d5b16ul));
Round(a, b, c, d, e, f, g, h, K(0x007f3e86ul));
Round(h, a, b, c, d, e, f, g, K(0x37088980ul));
Round(g, h, a, b, c, d, e, f, K(0xa507ea32ul));
Round(f, g, h, a, b, c, d, e, K(0x6fab9537ul));
Round(e, f, g, h, a, b, c, d, K(0x17406110ul));
Round(d, e, f, g, h, a, b, c, K(0x0d8cd6f1ul));
Round(c, d, e, f, g, h, a, b, K(0xcdaa3b6dul));
Round(b, c, d, e, f, g, h, a, K(0xc0bbbe37ul));
Round(a, b, c, d, e, f, g, h, K(0x83613bdaul));
Round(h, a, b, c, d, e, f, g, K(0xdb48a363ul));
Round(g, h, a, b, c, d, e, f, K(0x0b02e931ul));
Round(f, g, h, a, b, c, d, e, K(0x6fd15ca7ul));
Round(e, f, g, h, a, b, c, d, K(0x521afacaul));
Round(d, e, f, g, h, a, b, c, K(0x31338431ul));
Round(c, d, e, f, g, h, a, b, K(0x6ed41a95ul));
Round(b, c, d, e, f, g, h, a, K(0x6d437890ul));
Round(a, b, c, d, e, f, g, h, K(0xc39c91f2ul));
Round(h, a, b, c, d, e, f, g, K(0x9eccabbdul));
Round(g, h, a, b, c, d, e, f, K(0xb5c9a0e6ul));
Round(f, g, h, a, b, c, d, e, K(0x532fb63cul));
Round(e, f, g, h, a, b, c, d, K(0xd2c741c6ul));
Round(d, e, f, g, h, a, b, c, K(0x07237ea3ul));
Round(c, d, e, f, g, h, a, b, K(0xa4954b68ul));
Round(b, c, d, e, f, g, h, a, K(0x4c191d76ul));
w0 = Add(t0, a);
w1 = Add(t1, b);
w2 = Add(t2, c);
w3 = Add(t3, d);
w4 = Add(t4, e);
w5 = Add(t5, f);
w6 = Add(t6, g);
w7 = Add(t7, h);
// Transform 3
a = K(0x6a09e667ul);
b = K(0xbb67ae85ul);
c = K(0x3c6ef372ul);
d = K(0xa54ff53aul);
e = K(0x510e527ful);
f = K(0x9b05688cul);
g = K(0x1f83d9abul);
h = K(0x5be0cd19ul);
Round(a, b, c, d, e, f, g, h, Add(K(0x428a2f98ul), w0));
Round(h, a, b, c, d, e, f, g, Add(K(0x71374491ul), w1));
Round(g, h, a, b, c, d, e, f, Add(K(0xb5c0fbcful), w2));
Round(f, g, h, a, b, c, d, e, Add(K(0xe9b5dba5ul), w3));
Round(e, f, g, h, a, b, c, d, Add(K(0x3956c25bul), w4));
Round(d, e, f, g, h, a, b, c, Add(K(0x59f111f1ul), w5));
Round(c, d, e, f, g, h, a, b, Add(K(0x923f82a4ul), w6));
Round(b, c, d, e, f, g, h, a, Add(K(0xab1c5ed5ul), w7));
Round(a, b, c, d, e, f, g, h, K(0x5807aa98ul));
Round(h, a, b, c, d, e, f, g, K(0x12835b01ul));
Round(g, h, a, b, c, d, e, f, K(0x243185beul));
Round(f, g, h, a, b, c, d, e, K(0x550c7dc3ul));
Round(e, f, g, h, a, b, c, d, K(0x72be5d74ul));
Round(d, e, f, g, h, a, b, c, K(0x80deb1feul));
Round(c, d, e, f, g, h, a, b, K(0x9bdc06a7ul));
Round(b, c, d, e, f, g, h, a, K(0xc19bf274ul));
Round(a, b, c, d, e, f, g, h, Add(K(0xe49b69c1ul), Inc(w0, sigma0(w1))));
Round(h, a, b, c, d, e, f, g, Add(K(0xefbe4786ul), Inc(w1, K(0xa00000ul), sigma0(w2))));
Round(g, h, a, b, c, d, e, f, Add(K(0x0fc19dc6ul), Inc(w2, sigma1(w0), sigma0(w3))));
Round(f, g, h, a, b, c, d, e, Add(K(0x240ca1ccul), Inc(w3, sigma1(w1), sigma0(w4))));
Round(e, f, g, h, a, b, c, d, Add(K(0x2de92c6ful), Inc(w4, sigma1(w2), sigma0(w5))));
Round(d, e, f, g, h, a, b, c, Add(K(0x4a7484aaul), Inc(w5, sigma1(w3), sigma0(w6))));
Round(c, d, e, f, g, h, a, b, Add(K(0x5cb0a9dcul), Inc(w6, sigma1(w4), K(0x100ul), sigma0(w7))));
Round(b, c, d, e, f, g, h, a, Add(K(0x76f988daul), Inc(w7, sigma1(w5), w0, K(0x11002000ul))));
Round(a, b, c, d, e, f, g, h, Add(K(0x983e5152ul), w8 = Add(K(0x80000000ul), sigma1(w6), w1)));
Round(h, a, b, c, d, e, f, g, Add(K(0xa831c66dul), w9 = Add(sigma1(w7), w2)));
Round(g, h, a, b, c, d, e, f, Add(K(0xb00327c8ul), w10 = Add(sigma1(w8), w3)));
Round(f, g, h, a, b, c, d, e, Add(K(0xbf597fc7ul), w11 = Add(sigma1(w9), w4)));
Round(e, f, g, h, a, b, c, d, Add(K(0xc6e00bf3ul), w12 = Add(sigma1(w10), w5)));
Round(d, e, f, g, h, a, b, c, Add(K(0xd5a79147ul), w13 = Add(sigma1(w11), w6)));
Round(c, d, e, f, g, h, a, b, Add(K(0x06ca6351ul), w14 = Add(sigma1(w12), w7, K(0x400022ul))));
Round(b, c, d, e, f, g, h, a, Add(K(0x14292967ul), w15 = Add(K(0x100ul), sigma1(w13), w8, sigma0(w0))));
Round(a, b, c, d, e, f, g, h, Add(K(0x27b70a85ul), Inc(w0, sigma1(w14), w9, sigma0(w1))));
Round(h, a, b, c, d, e, f, g, Add(K(0x2e1b2138ul), Inc(w1, sigma1(w15), w10, sigma0(w2))));
Round(g, h, a, b, c, d, e, f, Add(K(0x4d2c6dfcul), Inc(w2, sigma1(w0), w11, sigma0(w3))));
Round(f, g, h, a, b, c, d, e, Add(K(0x53380d13ul), Inc(w3, sigma1(w1), w12, sigma0(w4))));
Round(e, f, g, h, a, b, c, d, Add(K(0x650a7354ul), Inc(w4, sigma1(w2), w13, sigma0(w5))));
Round(d, e, f, g, h, a, b, c, Add(K(0x766a0abbul), Inc(w5, sigma1(w3), w14, sigma0(w6))));
Round(c, d, e, f, g, h, a, b, Add(K(0x81c2c92eul), Inc(w6, sigma1(w4), w15, sigma0(w7))));
Round(b, c, d, e, f, g, h, a, Add(K(0x92722c85ul), Inc(w7, sigma1(w5), w0, sigma0(w8))));
Round(a, b, c, d, e, f, g, h, Add(K(0xa2bfe8a1ul), Inc(w8, sigma1(w6), w1, sigma0(w9))));
Round(h, a, b, c, d, e, f, g, Add(K(0xa81a664bul), Inc(w9, sigma1(w7), w2, sigma0(w10))));
Round(g, h, a, b, c, d, e, f, Add(K(0xc24b8b70ul), Inc(w10, sigma1(w8), w3, sigma0(w11))));
Round(f, g, h, a, b, c, d, e, Add(K(0xc76c51a3ul), Inc(w11, sigma1(w9), w4, sigma0(w12))));
Round(e, f, g, h, a, b, c, d, Add(K(0xd192e819ul), Inc(w12, sigma1(w10), w5, sigma0(w13))));
Round(d, e, f, g, h, a, b, c, Add(K(0xd6990624ul), Inc(w13, sigma1(w11), w6, sigma0(w14))));
Round(c, d, e, f, g, h, a, b, Add(K(0xf40e3585ul), Inc(w14, sigma1(w12), w7, sigma0(w15))));
Round(b, c, d, e, f, g, h, a, Add(K(0x106aa070ul), Inc(w15, sigma1(w13), w8, sigma0(w0))));
Round(a, b, c, d, e, f, g, h, Add(K(0x19a4c116ul), Inc(w0, sigma1(w14), w9, sigma0(w1))));
Round(h, a, b, c, d, e, f, g, Add(K(0x1e376c08ul), Inc(w1, sigma1(w15), w10, sigma0(w2))));
Round(g, h, a, b, c, d, e, f, Add(K(0x2748774cul), Inc(w2, sigma1(w0), w11, sigma0(w3))));
Round(f, g, h, a, b, c, d, e, Add(K(0x34b0bcb5ul), Inc(w3, sigma1(w1), w12, sigma0(w4))));
Round(e, f, g, h, a, b, c, d, Add(K(0x391c0cb3ul), Inc(w4, sigma1(w2), w13, sigma0(w5))));
Round(d, e, f, g, h, a, b, c, Add(K(0x4ed8aa4aul), Inc(w5, sigma1(w3), w14, sigma0(w6))));
Round(c, d, e, f, g, h, a, b, Add(K(0x5b9cca4ful), Inc(w6, sigma1(w4), w15, sigma0(w7))));
Round(b, c, d, e, f, g, h, a, Add(K(0x682e6ff3ul), Inc(w7, sigma1(w5), w0, sigma0(w8))));
Round(a, b, c, d, e, f, g, h, Add(K(0x748f82eeul), Inc(w8, sigma1(w6), w1, sigma0(w9))));
Round(h, a, b, c, d, e, f, g, Add(K(0x78a5636ful), Inc(w9, sigma1(w7), w2, sigma0(w10))));
Round(g, h, a, b, c, d, e, f, Add(K(0x84c87814ul), Inc(w10, sigma1(w8), w3, sigma0(w11))));
Round(f, g, h, a, b, c, d, e, Add(K(0x8cc70208ul), Inc(w11, sigma1(w9), w4, sigma0(w12))));
Round(e, f, g, h, a, b, c, d, Add(K(0x90befffaul), Inc(w12, sigma1(w10), w5, sigma0(w13))));
Round(d, e, f, g, h, a, b, c, Add(K(0xa4506cebul), Inc(w13, sigma1(w11), w6, sigma0(w14))));
Round(c, d, e, f, g, h, a, b, Add(K(0xbef9a3f7ul), w14, sigma1(w12), w7, sigma0(w15)));
Round(b, c, d, e, f, g, h, a, Add(K(0xc67178f2ul), w15, sigma1(w13), w8, sigma0(w0)));
// Output
Write8(out, 0, Add(a, K(0x6a09e667ul)));
Write8(out, 4, Add(b, K(0xbb67ae85ul)));
Write8(out, 8, Add(c, K(0x3c6ef372ul)));
Write8(out, 12, Add(d, K(0xa54ff53aul)));
Write8(out, 16, Add(e, K(0x510e527ful)));
Write8(out, 20, Add(f, K(0x9b05688cul)));
Write8(out, 24, Add(g, K(0x1f83d9abul)));
Write8(out, 28, Add(h, K(0x5be0cd19ul)));
}
}
#endif

357
src/crypto/sha256_shani.cpp Normal file
View file

@ -0,0 +1,357 @@
// Copyright (c) 2018-2020 The Bitcoin Core developers
// Copyright (c) 2018-2020 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
//
// Based on https://github.com/noloader/SHA-Intrinsics/blob/master/sha256-x86.c,
// Written and placed in public domain by Jeffrey Walton.
// Based on code from Intel, and by Sean Gulley for the miTLS project.
#ifdef ENABLE_SHANI
#include <stdint.h>
#include <immintrin.h>
namespace {
alignas(__m128i) const uint8_t MASK[16] = {0x03, 0x02, 0x01, 0x00, 0x07, 0x06, 0x05, 0x04, 0x0b, 0x0a, 0x09, 0x08, 0x0f, 0x0e, 0x0d, 0x0c};
alignas(__m128i) const uint8_t INIT0[16] = {0x8c, 0x68, 0x05, 0x9b, 0x7f, 0x52, 0x0e, 0x51, 0x85, 0xae, 0x67, 0xbb, 0x67, 0xe6, 0x09, 0x6a};
alignas(__m128i) const uint8_t INIT1[16] = {0x19, 0xcd, 0xe0, 0x5b, 0xab, 0xd9, 0x83, 0x1f, 0x3a, 0xf5, 0x4f, 0xa5, 0x72, 0xf3, 0x6e, 0x3c};
void inline __attribute__((always_inline)) QuadRound(__m128i& state0, __m128i& state1, uint64_t k1, uint64_t k0)
{
const __m128i msg = _mm_set_epi64x(k1, k0);
state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
state0 = _mm_sha256rnds2_epu32(state0, state1, _mm_shuffle_epi32(msg, 0x0e));
}
void inline __attribute__((always_inline)) QuadRound(__m128i& state0, __m128i& state1, __m128i m, uint64_t k1, uint64_t k0)
{
const __m128i msg = _mm_add_epi32(m, _mm_set_epi64x(k1, k0));
state1 = _mm_sha256rnds2_epu32(state1, state0, msg);
state0 = _mm_sha256rnds2_epu32(state0, state1, _mm_shuffle_epi32(msg, 0x0e));
}
void inline __attribute__((always_inline)) ShiftMessageA(__m128i& m0, __m128i m1)
{
m0 = _mm_sha256msg1_epu32(m0, m1);
}
void inline __attribute__((always_inline)) ShiftMessageC(__m128i& m0, __m128i m1, __m128i& m2)
{
m2 = _mm_sha256msg2_epu32(_mm_add_epi32(m2, _mm_alignr_epi8(m1, m0, 4)), m1);
}
void inline __attribute__((always_inline)) ShiftMessageB(__m128i& m0, __m128i m1, __m128i& m2)
{
ShiftMessageC(m0, m1, m2);
ShiftMessageA(m0, m1);
}
void inline __attribute__((always_inline)) Shuffle(__m128i& s0, __m128i& s1)
{
const __m128i t1 = _mm_shuffle_epi32(s0, 0xB1);
const __m128i t2 = _mm_shuffle_epi32(s1, 0x1B);
s0 = _mm_alignr_epi8(t1, t2, 0x08);
s1 = _mm_blend_epi16(t2, t1, 0xF0);
}
void inline __attribute__((always_inline)) Unshuffle(__m128i& s0, __m128i& s1)
{
const __m128i t1 = _mm_shuffle_epi32(s0, 0x1B);
const __m128i t2 = _mm_shuffle_epi32(s1, 0xB1);
s0 = _mm_blend_epi16(t1, t2, 0xF0);
s1 = _mm_alignr_epi8(t2, t1, 0x08);
}
__m128i inline __attribute__((always_inline)) Load(const unsigned char* in)
{
return _mm_shuffle_epi8(_mm_loadu_si128((const __m128i*)in), _mm_load_si128((const __m128i*)MASK));
}
void inline __attribute__((always_inline)) Save(unsigned char* out, __m128i s)
{
_mm_storeu_si128((__m128i*)out, _mm_shuffle_epi8(s, _mm_load_si128((const __m128i*)MASK)));
}
}
namespace sha256_shani {
void Transform(uint32_t* s, const unsigned char* chunk, size_t blocks)
{
__m128i m0, m1, m2, m3, s0, s1, so0, so1;
/* Load state */
s0 = _mm_loadu_si128((const __m128i*)s);
s1 = _mm_loadu_si128((const __m128i*)(s + 4));
Shuffle(s0, s1);
while (blocks--) {
/* Remember old state */
so0 = s0;
so1 = s1;
/* Load data and transform */
m0 = Load(chunk);
QuadRound(s0, s1, m0, 0xe9b5dba5b5c0fbcfull, 0x71374491428a2f98ull);
m1 = Load(chunk + 16);
QuadRound(s0, s1, m1, 0xab1c5ed5923f82a4ull, 0x59f111f13956c25bull);
ShiftMessageA(m0, m1);
m2 = Load(chunk + 32);
QuadRound(s0, s1, m2, 0x550c7dc3243185beull, 0x12835b01d807aa98ull);
ShiftMessageA(m1, m2);
m3 = Load(chunk + 48);
QuadRound(s0, s1, m3, 0xc19bf1749bdc06a7ull, 0x80deb1fe72be5d74ull);
ShiftMessageB(m2, m3, m0);
QuadRound(s0, s1, m0, 0x240ca1cc0fc19dc6ull, 0xefbe4786E49b69c1ull);
ShiftMessageB(m3, m0, m1);
QuadRound(s0, s1, m1, 0x76f988da5cb0a9dcull, 0x4a7484aa2de92c6full);
ShiftMessageB(m0, m1, m2);
QuadRound(s0, s1, m2, 0xbf597fc7b00327c8ull, 0xa831c66d983e5152ull);
ShiftMessageB(m1, m2, m3);
QuadRound(s0, s1, m3, 0x1429296706ca6351ull, 0xd5a79147c6e00bf3ull);
ShiftMessageB(m2, m3, m0);
QuadRound(s0, s1, m0, 0x53380d134d2c6dfcull, 0x2e1b213827b70a85ull);
ShiftMessageB(m3, m0, m1);
QuadRound(s0, s1, m1, 0x92722c8581c2c92eull, 0x766a0abb650a7354ull);
ShiftMessageB(m0, m1, m2);
QuadRound(s0, s1, m2, 0xc76c51A3c24b8b70ull, 0xa81a664ba2bfe8a1ull);
ShiftMessageB(m1, m2, m3);
QuadRound(s0, s1, m3, 0x106aa070f40e3585ull, 0xd6990624d192e819ull);
ShiftMessageB(m2, m3, m0);
QuadRound(s0, s1, m0, 0x34b0bcb52748774cull, 0x1e376c0819a4c116ull);
ShiftMessageB(m3, m0, m1);
QuadRound(s0, s1, m1, 0x682e6ff35b9cca4full, 0x4ed8aa4a391c0cb3ull);
ShiftMessageC(m0, m1, m2);
QuadRound(s0, s1, m2, 0x8cc7020884c87814ull, 0x78a5636f748f82eeull);
ShiftMessageC(m1, m2, m3);
QuadRound(s0, s1, m3, 0xc67178f2bef9A3f7ull, 0xa4506ceb90befffaull);
/* Combine with old state */
s0 = _mm_add_epi32(s0, so0);
s1 = _mm_add_epi32(s1, so1);
/* Advance */
chunk += 64;
}
Unshuffle(s0, s1);
_mm_storeu_si128((__m128i*)s, s0);
_mm_storeu_si128((__m128i*)(s + 4), s1);
}
}
namespace sha256d64_shani {
void Transform_2way(unsigned char* out, const unsigned char* in)
{
__m128i am0, am1, am2, am3, as0, as1, aso0, aso1;
__m128i bm0, bm1, bm2, bm3, bs0, bs1, bso0, bso1;
/* Transform 1 */
bs0 = as0 = _mm_load_si128((const __m128i*)INIT0);
bs1 = as1 = _mm_load_si128((const __m128i*)INIT1);
am0 = Load(in);
bm0 = Load(in + 64);
QuadRound(as0, as1, am0, 0xe9b5dba5b5c0fbcfull, 0x71374491428a2f98ull);
QuadRound(bs0, bs1, bm0, 0xe9b5dba5b5c0fbcfull, 0x71374491428a2f98ull);
am1 = Load(in + 16);
bm1 = Load(in + 80);
QuadRound(as0, as1, am1, 0xab1c5ed5923f82a4ull, 0x59f111f13956c25bull);
QuadRound(bs0, bs1, bm1, 0xab1c5ed5923f82a4ull, 0x59f111f13956c25bull);
ShiftMessageA(am0, am1);
ShiftMessageA(bm0, bm1);
am2 = Load(in + 32);
bm2 = Load(in + 96);
QuadRound(as0, as1, am2, 0x550c7dc3243185beull, 0x12835b01d807aa98ull);
QuadRound(bs0, bs1, bm2, 0x550c7dc3243185beull, 0x12835b01d807aa98ull);
ShiftMessageA(am1, am2);
ShiftMessageA(bm1, bm2);
am3 = Load(in + 48);
bm3 = Load(in + 112);
QuadRound(as0, as1, am3, 0xc19bf1749bdc06a7ull, 0x80deb1fe72be5d74ull);
QuadRound(bs0, bs1, bm3, 0xc19bf1749bdc06a7ull, 0x80deb1fe72be5d74ull);
ShiftMessageB(am2, am3, am0);
ShiftMessageB(bm2, bm3, bm0);
QuadRound(as0, as1, am0, 0x240ca1cc0fc19dc6ull, 0xefbe4786E49b69c1ull);
QuadRound(bs0, bs1, bm0, 0x240ca1cc0fc19dc6ull, 0xefbe4786E49b69c1ull);
ShiftMessageB(am3, am0, am1);
ShiftMessageB(bm3, bm0, bm1);
QuadRound(as0, as1, am1, 0x76f988da5cb0a9dcull, 0x4a7484aa2de92c6full);
QuadRound(bs0, bs1, bm1, 0x76f988da5cb0a9dcull, 0x4a7484aa2de92c6full);
ShiftMessageB(am0, am1, am2);
ShiftMessageB(bm0, bm1, bm2);
QuadRound(as0, as1, am2, 0xbf597fc7b00327c8ull, 0xa831c66d983e5152ull);
QuadRound(bs0, bs1, bm2, 0xbf597fc7b00327c8ull, 0xa831c66d983e5152ull);
ShiftMessageB(am1, am2, am3);
ShiftMessageB(bm1, bm2, bm3);
QuadRound(as0, as1, am3, 0x1429296706ca6351ull, 0xd5a79147c6e00bf3ull);
QuadRound(bs0, bs1, bm3, 0x1429296706ca6351ull, 0xd5a79147c6e00bf3ull);
ShiftMessageB(am2, am3, am0);
ShiftMessageB(bm2, bm3, bm0);
QuadRound(as0, as1, am0, 0x53380d134d2c6dfcull, 0x2e1b213827b70a85ull);
QuadRound(bs0, bs1, bm0, 0x53380d134d2c6dfcull, 0x2e1b213827b70a85ull);
ShiftMessageB(am3, am0, am1);
ShiftMessageB(bm3, bm0, bm1);
QuadRound(as0, as1, am1, 0x92722c8581c2c92eull, 0x766a0abb650a7354ull);
QuadRound(bs0, bs1, bm1, 0x92722c8581c2c92eull, 0x766a0abb650a7354ull);
ShiftMessageB(am0, am1, am2);
ShiftMessageB(bm0, bm1, bm2);
QuadRound(as0, as1, am2, 0xc76c51A3c24b8b70ull, 0xa81a664ba2bfe8a1ull);
QuadRound(bs0, bs1, bm2, 0xc76c51A3c24b8b70ull, 0xa81a664ba2bfe8a1ull);
ShiftMessageB(am1, am2, am3);
ShiftMessageB(bm1, bm2, bm3);
QuadRound(as0, as1, am3, 0x106aa070f40e3585ull, 0xd6990624d192e819ull);
QuadRound(bs0, bs1, bm3, 0x106aa070f40e3585ull, 0xd6990624d192e819ull);
ShiftMessageB(am2, am3, am0);
ShiftMessageB(bm2, bm3, bm0);
QuadRound(as0, as1, am0, 0x34b0bcb52748774cull, 0x1e376c0819a4c116ull);
QuadRound(bs0, bs1, bm0, 0x34b0bcb52748774cull, 0x1e376c0819a4c116ull);
ShiftMessageB(am3, am0, am1);
ShiftMessageB(bm3, bm0, bm1);
QuadRound(as0, as1, am1, 0x682e6ff35b9cca4full, 0x4ed8aa4a391c0cb3ull);
QuadRound(bs0, bs1, bm1, 0x682e6ff35b9cca4full, 0x4ed8aa4a391c0cb3ull);
ShiftMessageC(am0, am1, am2);
ShiftMessageC(bm0, bm1, bm2);
QuadRound(as0, as1, am2, 0x8cc7020884c87814ull, 0x78a5636f748f82eeull);
QuadRound(bs0, bs1, bm2, 0x8cc7020884c87814ull, 0x78a5636f748f82eeull);
ShiftMessageC(am1, am2, am3);
ShiftMessageC(bm1, bm2, bm3);
QuadRound(as0, as1, am3, 0xc67178f2bef9A3f7ull, 0xa4506ceb90befffaull);
QuadRound(bs0, bs1, bm3, 0xc67178f2bef9A3f7ull, 0xa4506ceb90befffaull);
as0 = _mm_add_epi32(as0, _mm_load_si128((const __m128i*)INIT0));
bs0 = _mm_add_epi32(bs0, _mm_load_si128((const __m128i*)INIT0));
as1 = _mm_add_epi32(as1, _mm_load_si128((const __m128i*)INIT1));
bs1 = _mm_add_epi32(bs1, _mm_load_si128((const __m128i*)INIT1));
/* Transform 2 */
aso0 = as0;
bso0 = bs0;
aso1 = as1;
bso1 = bs1;
QuadRound(as0, as1, 0xe9b5dba5b5c0fbcfull, 0x71374491c28a2f98ull);
QuadRound(bs0, bs1, 0xe9b5dba5b5c0fbcfull, 0x71374491c28a2f98ull);
QuadRound(as0, as1, 0xab1c5ed5923f82a4ull, 0x59f111f13956c25bull);
QuadRound(bs0, bs1, 0xab1c5ed5923f82a4ull, 0x59f111f13956c25bull);
QuadRound(as0, as1, 0x550c7dc3243185beull, 0x12835b01d807aa98ull);
QuadRound(bs0, bs1, 0x550c7dc3243185beull, 0x12835b01d807aa98ull);
QuadRound(as0, as1, 0xc19bf3749bdc06a7ull, 0x80deb1fe72be5d74ull);
QuadRound(bs0, bs1, 0xc19bf3749bdc06a7ull, 0x80deb1fe72be5d74ull);
QuadRound(as0, as1, 0x240cf2540fe1edc6ull, 0xf0fe4786649b69c1ull);
QuadRound(bs0, bs1, 0x240cf2540fe1edc6ull, 0xf0fe4786649b69c1ull);
QuadRound(as0, as1, 0x16f988fa61b9411eull, 0x6cc984be4fe9346full);
QuadRound(bs0, bs1, 0x16f988fa61b9411eull, 0x6cc984be4fe9346full);
QuadRound(as0, as1, 0xb9d99ec7b019fc65ull, 0xa88e5a6df2c65152ull);
QuadRound(bs0, bs1, 0xb9d99ec7b019fc65ull, 0xa88e5a6df2c65152ull);
QuadRound(as0, as1, 0xc7353eb0fdb1232bull, 0xe70eeaa09a1231c3ull);
QuadRound(bs0, bs1, 0xc7353eb0fdb1232bull, 0xe70eeaa09a1231c3ull);
QuadRound(as0, as1, 0xdc1eeefd5a0f118full, 0xcb976d5f3069bad5ull);
QuadRound(bs0, bs1, 0xdc1eeefd5a0f118full, 0xcb976d5f3069bad5ull);
QuadRound(as0, as1, 0xe15d5b1658f4ca9dull, 0xde0b7a040a35b689ull);
QuadRound(bs0, bs1, 0xe15d5b1658f4ca9dull, 0xde0b7a040a35b689ull);
QuadRound(as0, as1, 0x6fab9537a507ea32ull, 0x37088980007f3e86ull);
QuadRound(bs0, bs1, 0x6fab9537a507ea32ull, 0x37088980007f3e86ull);
QuadRound(as0, as1, 0xc0bbbe37cdaa3b6dull, 0x0d8cd6f117406110ull);
QuadRound(bs0, bs1, 0xc0bbbe37cdaa3b6dull, 0x0d8cd6f117406110ull);
QuadRound(as0, as1, 0x6fd15ca70b02e931ull, 0xdb48a36383613bdaull);
QuadRound(bs0, bs1, 0x6fd15ca70b02e931ull, 0xdb48a36383613bdaull);
QuadRound(as0, as1, 0x6d4378906ed41a95ull, 0x31338431521afacaull);
QuadRound(bs0, bs1, 0x6d4378906ed41a95ull, 0x31338431521afacaull);
QuadRound(as0, as1, 0x532fb63cb5c9a0e6ull, 0x9eccabbdc39c91f2ull);
QuadRound(bs0, bs1, 0x532fb63cb5c9a0e6ull, 0x9eccabbdc39c91f2ull);
QuadRound(as0, as1, 0x4c191d76a4954b68ull, 0x07237ea3d2c741c6ull);
QuadRound(bs0, bs1, 0x4c191d76a4954b68ull, 0x07237ea3d2c741c6ull);
as0 = _mm_add_epi32(as0, aso0);
bs0 = _mm_add_epi32(bs0, bso0);
as1 = _mm_add_epi32(as1, aso1);
bs1 = _mm_add_epi32(bs1, bso1);
/* Extract hash */
Unshuffle(as0, as1);
Unshuffle(bs0, bs1);
am0 = as0;
bm0 = bs0;
am1 = as1;
bm1 = bs1;
/* Transform 3 */
bs0 = as0 = _mm_load_si128((const __m128i*)INIT0);
bs1 = as1 = _mm_load_si128((const __m128i*)INIT1);
QuadRound(as0, as1, am0, 0xe9b5dba5B5c0fbcfull, 0x71374491428a2f98ull);
QuadRound(bs0, bs1, bm0, 0xe9b5dba5B5c0fbcfull, 0x71374491428a2f98ull);
QuadRound(as0, as1, am1, 0xab1c5ed5923f82a4ull, 0x59f111f13956c25bull);
QuadRound(bs0, bs1, bm1, 0xab1c5ed5923f82a4ull, 0x59f111f13956c25bull);
ShiftMessageA(am0, am1);
ShiftMessageA(bm0, bm1);
bm2 = am2 = _mm_set_epi64x(0x0ull, 0x80000000ull);
QuadRound(as0, as1, 0x550c7dc3243185beull, 0x12835b015807aa98ull);
QuadRound(bs0, bs1, 0x550c7dc3243185beull, 0x12835b015807aa98ull);
ShiftMessageA(am1, am2);
ShiftMessageA(bm1, bm2);
bm3 = am3 = _mm_set_epi64x(0x10000000000ull, 0x0ull);
QuadRound(as0, as1, 0xc19bf2749bdc06a7ull, 0x80deb1fe72be5d74ull);
QuadRound(bs0, bs1, 0xc19bf2749bdc06a7ull, 0x80deb1fe72be5d74ull);
ShiftMessageB(am2, am3, am0);
ShiftMessageB(bm2, bm3, bm0);
QuadRound(as0, as1, am0, 0x240ca1cc0fc19dc6ull, 0xefbe4786e49b69c1ull);
QuadRound(bs0, bs1, bm0, 0x240ca1cc0fc19dc6ull, 0xefbe4786e49b69c1ull);
ShiftMessageB(am3, am0, am1);
ShiftMessageB(bm3, bm0, bm1);
QuadRound(as0, as1, am1, 0x76f988da5cb0a9dcull, 0x4a7484aa2de92c6full);
QuadRound(bs0, bs1, bm1, 0x76f988da5cb0a9dcull, 0x4a7484aa2de92c6full);
ShiftMessageB(am0, am1, am2);
ShiftMessageB(bm0, bm1, bm2);
QuadRound(as0, as1, am2, 0xbf597fc7b00327c8ull, 0xa831c66d983e5152ull);
QuadRound(bs0, bs1, bm2, 0xbf597fc7b00327c8ull, 0xa831c66d983e5152ull);
ShiftMessageB(am1, am2, am3);
ShiftMessageB(bm1, bm2, bm3);
QuadRound(as0, as1, am3, 0x1429296706ca6351ull, 0xd5a79147c6e00bf3ull);
QuadRound(bs0, bs1, bm3, 0x1429296706ca6351ull, 0xd5a79147c6e00bf3ull);
ShiftMessageB(am2, am3, am0);
ShiftMessageB(bm2, bm3, bm0);
QuadRound(as0, as1, am0, 0x53380d134d2c6dfcull, 0x2e1b213827b70a85ull);
QuadRound(bs0, bs1, bm0, 0x53380d134d2c6dfcull, 0x2e1b213827b70a85ull);
ShiftMessageB(am3, am0, am1);
ShiftMessageB(bm3, bm0, bm1);
QuadRound(as0, as1, am1, 0x92722c8581c2c92eull, 0x766a0abb650a7354ull);
QuadRound(bs0, bs1, bm1, 0x92722c8581c2c92eull, 0x766a0abb650a7354ull);
ShiftMessageB(am0, am1, am2);
ShiftMessageB(bm0, bm1, bm2);
QuadRound(as0, as1, am2, 0xc76c51a3c24b8b70ull, 0xa81a664ba2bfe8A1ull);
QuadRound(bs0, bs1, bm2, 0xc76c51a3c24b8b70ull, 0xa81a664ba2bfe8A1ull);
ShiftMessageB(am1, am2, am3);
ShiftMessageB(bm1, bm2, bm3);
QuadRound(as0, as1, am3, 0x106aa070f40e3585ull, 0xd6990624d192e819ull);
QuadRound(bs0, bs1, bm3, 0x106aa070f40e3585ull, 0xd6990624d192e819ull);
ShiftMessageB(am2, am3, am0);
ShiftMessageB(bm2, bm3, bm0);
QuadRound(as0, as1, am0, 0x34b0bcb52748774cull, 0x1e376c0819a4c116ull);
QuadRound(bs0, bs1, bm0, 0x34b0bcb52748774cull, 0x1e376c0819a4c116ull);
ShiftMessageB(am3, am0, am1);
ShiftMessageB(bm3, bm0, bm1);
QuadRound(as0, as1, am1, 0x682e6ff35b9cca4full, 0x4ed8aa4a391c0cb3ull);
QuadRound(bs0, bs1, bm1, 0x682e6ff35b9cca4full, 0x4ed8aa4a391c0cb3ull);
ShiftMessageC(am0, am1, am2);
ShiftMessageC(bm0, bm1, bm2);
QuadRound(as0, as1, am2, 0x8cc7020884c87814ull, 0x78a5636f748f82eeull);
QuadRound(bs0, bs1, bm2, 0x8cc7020884c87814ull, 0x78a5636f748f82eeull);
ShiftMessageC(am1, am2, am3);
ShiftMessageC(bm1, bm2, bm3);
QuadRound(as0, as1, am3, 0xc67178f2bef9a3f7ull, 0xa4506ceb90befffaull);
QuadRound(bs0, bs1, bm3, 0xc67178f2bef9a3f7ull, 0xa4506ceb90befffaull);
as0 = _mm_add_epi32(as0, _mm_load_si128((const __m128i*)INIT0));
bs0 = _mm_add_epi32(bs0, _mm_load_si128((const __m128i*)INIT0));
as1 = _mm_add_epi32(as1, _mm_load_si128((const __m128i*)INIT1));
bs1 = _mm_add_epi32(bs1, _mm_load_si128((const __m128i*)INIT1));
/* Extract hash into out */
Unshuffle(as0, as1);
Unshuffle(bs0, bs1);
Save(out, as0);
Save(out + 16, as1);
Save(out + 32, bs0);
Save(out + 48, bs1);
}
}
#endif

1506
src/crypto/sha256_sse4.cpp Normal file
View file

File diff suppressed because it is too large Load diff

320
src/crypto/sha256_sse41.cpp Normal file
View file

@ -0,0 +1,320 @@
// Copyright (c) 2018-2019 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#ifdef ENABLE_SSE41
#include <stdint.h>
#include <immintrin.h>
#include <crypto/common.h>
namespace sha256d64_sse41 {
namespace {
__m128i inline K(uint32_t x) { return _mm_set1_epi32(x); }
__m128i inline Add(__m128i x, __m128i y) { return _mm_add_epi32(x, y); }
__m128i inline Add(__m128i x, __m128i y, __m128i z) { return Add(Add(x, y), z); }
__m128i inline Add(__m128i x, __m128i y, __m128i z, __m128i w) { return Add(Add(x, y), Add(z, w)); }
__m128i inline Add(__m128i x, __m128i y, __m128i z, __m128i w, __m128i v) { return Add(Add(x, y, z), Add(w, v)); }
__m128i inline Inc(__m128i& x, __m128i y) { x = Add(x, y); return x; }
__m128i inline Inc(__m128i& x, __m128i y, __m128i z) { x = Add(x, y, z); return x; }
__m128i inline Inc(__m128i& x, __m128i y, __m128i z, __m128i w) { x = Add(x, y, z, w); return x; }
__m128i inline Xor(__m128i x, __m128i y) { return _mm_xor_si128(x, y); }
__m128i inline Xor(__m128i x, __m128i y, __m128i z) { return Xor(Xor(x, y), z); }
__m128i inline Or(__m128i x, __m128i y) { return _mm_or_si128(x, y); }
__m128i inline And(__m128i x, __m128i y) { return _mm_and_si128(x, y); }
__m128i inline ShR(__m128i x, int n) { return _mm_srli_epi32(x, n); }
__m128i inline ShL(__m128i x, int n) { return _mm_slli_epi32(x, n); }
__m128i inline Ch(__m128i x, __m128i y, __m128i z) { return Xor(z, And(x, Xor(y, z))); }
__m128i inline Maj(__m128i x, __m128i y, __m128i z) { return Or(And(x, y), And(z, Or(x, y))); }
__m128i inline Sigma0(__m128i x) { return Xor(Or(ShR(x, 2), ShL(x, 30)), Or(ShR(x, 13), ShL(x, 19)), Or(ShR(x, 22), ShL(x, 10))); }
__m128i inline Sigma1(__m128i x) { return Xor(Or(ShR(x, 6), ShL(x, 26)), Or(ShR(x, 11), ShL(x, 21)), Or(ShR(x, 25), ShL(x, 7))); }
__m128i inline sigma0(__m128i x) { return Xor(Or(ShR(x, 7), ShL(x, 25)), Or(ShR(x, 18), ShL(x, 14)), ShR(x, 3)); }
__m128i inline sigma1(__m128i x) { return Xor(Or(ShR(x, 17), ShL(x, 15)), Or(ShR(x, 19), ShL(x, 13)), ShR(x, 10)); }
/** One round of SHA-256. */
void inline __attribute__((always_inline)) Round(__m128i a, __m128i b, __m128i c, __m128i& d, __m128i e, __m128i f, __m128i g, __m128i& h, __m128i k)
{
__m128i t1 = Add(h, Sigma1(e), Ch(e, f, g), k);
__m128i t2 = Add(Sigma0(a), Maj(a, b, c));
d = Add(d, t1);
h = Add(t1, t2);
}
__m128i inline Read4(const unsigned char* chunk, int offset) {
__m128i ret = _mm_set_epi32(
ReadLE32(chunk + 0 + offset),
ReadLE32(chunk + 64 + offset),
ReadLE32(chunk + 128 + offset),
ReadLE32(chunk + 192 + offset)
);
return _mm_shuffle_epi8(ret, _mm_set_epi32(0x0C0D0E0FUL, 0x08090A0BUL, 0x04050607UL, 0x00010203UL));
}
void inline Write4(unsigned char* out, int offset, __m128i v) {
v = _mm_shuffle_epi8(v, _mm_set_epi32(0x0C0D0E0FUL, 0x08090A0BUL, 0x04050607UL, 0x00010203UL));
WriteLE32(out + 0 + offset, _mm_extract_epi32(v, 3));
WriteLE32(out + 32 + offset, _mm_extract_epi32(v, 2));
WriteLE32(out + 64 + offset, _mm_extract_epi32(v, 1));
WriteLE32(out + 96 + offset, _mm_extract_epi32(v, 0));
}
}
void Transform_4way(unsigned char* out, const unsigned char* in)
{
// Transform 1
__m128i a = K(0x6a09e667ul);
__m128i b = K(0xbb67ae85ul);
__m128i c = K(0x3c6ef372ul);
__m128i d = K(0xa54ff53aul);
__m128i e = K(0x510e527ful);
__m128i f = K(0x9b05688cul);
__m128i g = K(0x1f83d9abul);
__m128i h = K(0x5be0cd19ul);
__m128i w0, w1, w2, w3, w4, w5, w6, w7, w8, w9, w10, w11, w12, w13, w14, w15;
Round(a, b, c, d, e, f, g, h, Add(K(0x428a2f98ul), w0 = Read4(in, 0)));
Round(h, a, b, c, d, e, f, g, Add(K(0x71374491ul), w1 = Read4(in, 4)));
Round(g, h, a, b, c, d, e, f, Add(K(0xb5c0fbcful), w2 = Read4(in, 8)));
Round(f, g, h, a, b, c, d, e, Add(K(0xe9b5dba5ul), w3 = Read4(in, 12)));
Round(e, f, g, h, a, b, c, d, Add(K(0x3956c25bul), w4 = Read4(in, 16)));
Round(d, e, f, g, h, a, b, c, Add(K(0x59f111f1ul), w5 = Read4(in, 20)));
Round(c, d, e, f, g, h, a, b, Add(K(0x923f82a4ul), w6 = Read4(in, 24)));
Round(b, c, d, e, f, g, h, a, Add(K(0xab1c5ed5ul), w7 = Read4(in, 28)));
Round(a, b, c, d, e, f, g, h, Add(K(0xd807aa98ul), w8 = Read4(in, 32)));
Round(h, a, b, c, d, e, f, g, Add(K(0x12835b01ul), w9 = Read4(in, 36)));
Round(g, h, a, b, c, d, e, f, Add(K(0x243185beul), w10 = Read4(in, 40)));
Round(f, g, h, a, b, c, d, e, Add(K(0x550c7dc3ul), w11 = Read4(in, 44)));
Round(e, f, g, h, a, b, c, d, Add(K(0x72be5d74ul), w12 = Read4(in, 48)));
Round(d, e, f, g, h, a, b, c, Add(K(0x80deb1feul), w13 = Read4(in, 52)));
Round(c, d, e, f, g, h, a, b, Add(K(0x9bdc06a7ul), w14 = Read4(in, 56)));
Round(b, c, d, e, f, g, h, a, Add(K(0xc19bf174ul), w15 = Read4(in, 60)));
Round(a, b, c, d, e, f, g, h, Add(K(0xe49b69c1ul), Inc(w0, sigma1(w14), w9, sigma0(w1))));
Round(h, a, b, c, d, e, f, g, Add(K(0xefbe4786ul), Inc(w1, sigma1(w15), w10, sigma0(w2))));
Round(g, h, a, b, c, d, e, f, Add(K(0x0fc19dc6ul), Inc(w2, sigma1(w0), w11, sigma0(w3))));
Round(f, g, h, a, b, c, d, e, Add(K(0x240ca1ccul), Inc(w3, sigma1(w1), w12, sigma0(w4))));
Round(e, f, g, h, a, b, c, d, Add(K(0x2de92c6ful), Inc(w4, sigma1(w2), w13, sigma0(w5))));
Round(d, e, f, g, h, a, b, c, Add(K(0x4a7484aaul), Inc(w5, sigma1(w3), w14, sigma0(w6))));
Round(c, d, e, f, g, h, a, b, Add(K(0x5cb0a9dcul), Inc(w6, sigma1(w4), w15, sigma0(w7))));
Round(b, c, d, e, f, g, h, a, Add(K(0x76f988daul), Inc(w7, sigma1(w5), w0, sigma0(w8))));
Round(a, b, c, d, e, f, g, h, Add(K(0x983e5152ul), Inc(w8, sigma1(w6), w1, sigma0(w9))));
Round(h, a, b, c, d, e, f, g, Add(K(0xa831c66dul), Inc(w9, sigma1(w7), w2, sigma0(w10))));
Round(g, h, a, b, c, d, e, f, Add(K(0xb00327c8ul), Inc(w10, sigma1(w8), w3, sigma0(w11))));
Round(f, g, h, a, b, c, d, e, Add(K(0xbf597fc7ul), Inc(w11, sigma1(w9), w4, sigma0(w12))));
Round(e, f, g, h, a, b, c, d, Add(K(0xc6e00bf3ul), Inc(w12, sigma1(w10), w5, sigma0(w13))));
Round(d, e, f, g, h, a, b, c, Add(K(0xd5a79147ul), Inc(w13, sigma1(w11), w6, sigma0(w14))));
Round(c, d, e, f, g, h, a, b, Add(K(0x06ca6351ul), Inc(w14, sigma1(w12), w7, sigma0(w15))));
Round(b, c, d, e, f, g, h, a, Add(K(0x14292967ul), Inc(w15, sigma1(w13), w8, sigma0(w0))));
Round(a, b, c, d, e, f, g, h, Add(K(0x27b70a85ul), Inc(w0, sigma1(w14), w9, sigma0(w1))));
Round(h, a, b, c, d, e, f, g, Add(K(0x2e1b2138ul), Inc(w1, sigma1(w15), w10, sigma0(w2))));
Round(g, h, a, b, c, d, e, f, Add(K(0x4d2c6dfcul), Inc(w2, sigma1(w0), w11, sigma0(w3))));
Round(f, g, h, a, b, c, d, e, Add(K(0x53380d13ul), Inc(w3, sigma1(w1), w12, sigma0(w4))));
Round(e, f, g, h, a, b, c, d, Add(K(0x650a7354ul), Inc(w4, sigma1(w2), w13, sigma0(w5))));
Round(d, e, f, g, h, a, b, c, Add(K(0x766a0abbul), Inc(w5, sigma1(w3), w14, sigma0(w6))));
Round(c, d, e, f, g, h, a, b, Add(K(0x81c2c92eul), Inc(w6, sigma1(w4), w15, sigma0(w7))));
Round(b, c, d, e, f, g, h, a, Add(K(0x92722c85ul), Inc(w7, sigma1(w5), w0, sigma0(w8))));
Round(a, b, c, d, e, f, g, h, Add(K(0xa2bfe8a1ul), Inc(w8, sigma1(w6), w1, sigma0(w9))));
Round(h, a, b, c, d, e, f, g, Add(K(0xa81a664bul), Inc(w9, sigma1(w7), w2, sigma0(w10))));
Round(g, h, a, b, c, d, e, f, Add(K(0xc24b8b70ul), Inc(w10, sigma1(w8), w3, sigma0(w11))));
Round(f, g, h, a, b, c, d, e, Add(K(0xc76c51a3ul), Inc(w11, sigma1(w9), w4, sigma0(w12))));
Round(e, f, g, h, a, b, c, d, Add(K(0xd192e819ul), Inc(w12, sigma1(w10), w5, sigma0(w13))));
Round(d, e, f, g, h, a, b, c, Add(K(0xd6990624ul), Inc(w13, sigma1(w11), w6, sigma0(w14))));
Round(c, d, e, f, g, h, a, b, Add(K(0xf40e3585ul), Inc(w14, sigma1(w12), w7, sigma0(w15))));
Round(b, c, d, e, f, g, h, a, Add(K(0x106aa070ul), Inc(w15, sigma1(w13), w8, sigma0(w0))));
Round(a, b, c, d, e, f, g, h, Add(K(0x19a4c116ul), Inc(w0, sigma1(w14), w9, sigma0(w1))));
Round(h, a, b, c, d, e, f, g, Add(K(0x1e376c08ul), Inc(w1, sigma1(w15), w10, sigma0(w2))));
Round(g, h, a, b, c, d, e, f, Add(K(0x2748774cul), Inc(w2, sigma1(w0), w11, sigma0(w3))));
Round(f, g, h, a, b, c, d, e, Add(K(0x34b0bcb5ul), Inc(w3, sigma1(w1), w12, sigma0(w4))));
Round(e, f, g, h, a, b, c, d, Add(K(0x391c0cb3ul), Inc(w4, sigma1(w2), w13, sigma0(w5))));
Round(d, e, f, g, h, a, b, c, Add(K(0x4ed8aa4aul), Inc(w5, sigma1(w3), w14, sigma0(w6))));
Round(c, d, e, f, g, h, a, b, Add(K(0x5b9cca4ful), Inc(w6, sigma1(w4), w15, sigma0(w7))));
Round(b, c, d, e, f, g, h, a, Add(K(0x682e6ff3ul), Inc(w7, sigma1(w5), w0, sigma0(w8))));
Round(a, b, c, d, e, f, g, h, Add(K(0x748f82eeul), Inc(w8, sigma1(w6), w1, sigma0(w9))));
Round(h, a, b, c, d, e, f, g, Add(K(0x78a5636ful), Inc(w9, sigma1(w7), w2, sigma0(w10))));
Round(g, h, a, b, c, d, e, f, Add(K(0x84c87814ul), Inc(w10, sigma1(w8), w3, sigma0(w11))));
Round(f, g, h, a, b, c, d, e, Add(K(0x8cc70208ul), Inc(w11, sigma1(w9), w4, sigma0(w12))));
Round(e, f, g, h, a, b, c, d, Add(K(0x90befffaul), Inc(w12, sigma1(w10), w5, sigma0(w13))));
Round(d, e, f, g, h, a, b, c, Add(K(0xa4506cebul), Inc(w13, sigma1(w11), w6, sigma0(w14))));
Round(c, d, e, f, g, h, a, b, Add(K(0xbef9a3f7ul), Inc(w14, sigma1(w12), w7, sigma0(w15))));
Round(b, c, d, e, f, g, h, a, Add(K(0xc67178f2ul), Inc(w15, sigma1(w13), w8, sigma0(w0))));
a = Add(a, K(0x6a09e667ul));
b = Add(b, K(0xbb67ae85ul));
c = Add(c, K(0x3c6ef372ul));
d = Add(d, K(0xa54ff53aul));
e = Add(e, K(0x510e527ful));
f = Add(f, K(0x9b05688cul));
g = Add(g, K(0x1f83d9abul));
h = Add(h, K(0x5be0cd19ul));
__m128i t0 = a, t1 = b, t2 = c, t3 = d, t4 = e, t5 = f, t6 = g, t7 = h;
// Transform 2
Round(a, b, c, d, e, f, g, h, K(0xc28a2f98ul));
Round(h, a, b, c, d, e, f, g, K(0x71374491ul));
Round(g, h, a, b, c, d, e, f, K(0xb5c0fbcful));
Round(f, g, h, a, b, c, d, e, K(0xe9b5dba5ul));
Round(e, f, g, h, a, b, c, d, K(0x3956c25bul));
Round(d, e, f, g, h, a, b, c, K(0x59f111f1ul));
Round(c, d, e, f, g, h, a, b, K(0x923f82a4ul));
Round(b, c, d, e, f, g, h, a, K(0xab1c5ed5ul));
Round(a, b, c, d, e, f, g, h, K(0xd807aa98ul));
Round(h, a, b, c, d, e, f, g, K(0x12835b01ul));
Round(g, h, a, b, c, d, e, f, K(0x243185beul));
Round(f, g, h, a, b, c, d, e, K(0x550c7dc3ul));
Round(e, f, g, h, a, b, c, d, K(0x72be5d74ul));
Round(d, e, f, g, h, a, b, c, K(0x80deb1feul));
Round(c, d, e, f, g, h, a, b, K(0x9bdc06a7ul));
Round(b, c, d, e, f, g, h, a, K(0xc19bf374ul));
Round(a, b, c, d, e, f, g, h, K(0x649b69c1ul));
Round(h, a, b, c, d, e, f, g, K(0xf0fe4786ul));
Round(g, h, a, b, c, d, e, f, K(0x0fe1edc6ul));
Round(f, g, h, a, b, c, d, e, K(0x240cf254ul));
Round(e, f, g, h, a, b, c, d, K(0x4fe9346ful));
Round(d, e, f, g, h, a, b, c, K(0x6cc984beul));
Round(c, d, e, f, g, h, a, b, K(0x61b9411eul));
Round(b, c, d, e, f, g, h, a, K(0x16f988faul));
Round(a, b, c, d, e, f, g, h, K(0xf2c65152ul));
Round(h, a, b, c, d, e, f, g, K(0xa88e5a6dul));
Round(g, h, a, b, c, d, e, f, K(0xb019fc65ul));
Round(f, g, h, a, b, c, d, e, K(0xb9d99ec7ul));
Round(e, f, g, h, a, b, c, d, K(0x9a1231c3ul));
Round(d, e, f, g, h, a, b, c, K(0xe70eeaa0ul));
Round(c, d, e, f, g, h, a, b, K(0xfdb1232bul));
Round(b, c, d, e, f, g, h, a, K(0xc7353eb0ul));
Round(a, b, c, d, e, f, g, h, K(0x3069bad5ul));
Round(h, a, b, c, d, e, f, g, K(0xcb976d5ful));
Round(g, h, a, b, c, d, e, f, K(0x5a0f118ful));
Round(f, g, h, a, b, c, d, e, K(0xdc1eeefdul));
Round(e, f, g, h, a, b, c, d, K(0x0a35b689ul));
Round(d, e, f, g, h, a, b, c, K(0xde0b7a04ul));
Round(c, d, e, f, g, h, a, b, K(0x58f4ca9dul));
Round(b, c, d, e, f, g, h, a, K(0xe15d5b16ul));
Round(a, b, c, d, e, f, g, h, K(0x007f3e86ul));
Round(h, a, b, c, d, e, f, g, K(0x37088980ul));
Round(g, h, a, b, c, d, e, f, K(0xa507ea32ul));
Round(f, g, h, a, b, c, d, e, K(0x6fab9537ul));
Round(e, f, g, h, a, b, c, d, K(0x17406110ul));
Round(d, e, f, g, h, a, b, c, K(0x0d8cd6f1ul));
Round(c, d, e, f, g, h, a, b, K(0xcdaa3b6dul));
Round(b, c, d, e, f, g, h, a, K(0xc0bbbe37ul));
Round(a, b, c, d, e, f, g, h, K(0x83613bdaul));
Round(h, a, b, c, d, e, f, g, K(0xdb48a363ul));
Round(g, h, a, b, c, d, e, f, K(0x0b02e931ul));
Round(f, g, h, a, b, c, d, e, K(0x6fd15ca7ul));
Round(e, f, g, h, a, b, c, d, K(0x521afacaul));
Round(d, e, f, g, h, a, b, c, K(0x31338431ul));
Round(c, d, e, f, g, h, a, b, K(0x6ed41a95ul));
Round(b, c, d, e, f, g, h, a, K(0x6d437890ul));
Round(a, b, c, d, e, f, g, h, K(0xc39c91f2ul));
Round(h, a, b, c, d, e, f, g, K(0x9eccabbdul));
Round(g, h, a, b, c, d, e, f, K(0xb5c9a0e6ul));
Round(f, g, h, a, b, c, d, e, K(0x532fb63cul));
Round(e, f, g, h, a, b, c, d, K(0xd2c741c6ul));
Round(d, e, f, g, h, a, b, c, K(0x07237ea3ul));
Round(c, d, e, f, g, h, a, b, K(0xa4954b68ul));
Round(b, c, d, e, f, g, h, a, K(0x4c191d76ul));
w0 = Add(t0, a);
w1 = Add(t1, b);
w2 = Add(t2, c);
w3 = Add(t3, d);
w4 = Add(t4, e);
w5 = Add(t5, f);
w6 = Add(t6, g);
w7 = Add(t7, h);
// Transform 3
a = K(0x6a09e667ul);
b = K(0xbb67ae85ul);
c = K(0x3c6ef372ul);
d = K(0xa54ff53aul);
e = K(0x510e527ful);
f = K(0x9b05688cul);
g = K(0x1f83d9abul);
h = K(0x5be0cd19ul);
Round(a, b, c, d, e, f, g, h, Add(K(0x428a2f98ul), w0));
Round(h, a, b, c, d, e, f, g, Add(K(0x71374491ul), w1));
Round(g, h, a, b, c, d, e, f, Add(K(0xb5c0fbcful), w2));
Round(f, g, h, a, b, c, d, e, Add(K(0xe9b5dba5ul), w3));
Round(e, f, g, h, a, b, c, d, Add(K(0x3956c25bul), w4));
Round(d, e, f, g, h, a, b, c, Add(K(0x59f111f1ul), w5));
Round(c, d, e, f, g, h, a, b, Add(K(0x923f82a4ul), w6));
Round(b, c, d, e, f, g, h, a, Add(K(0xab1c5ed5ul), w7));
Round(a, b, c, d, e, f, g, h, K(0x5807aa98ul));
Round(h, a, b, c, d, e, f, g, K(0x12835b01ul));
Round(g, h, a, b, c, d, e, f, K(0x243185beul));
Round(f, g, h, a, b, c, d, e, K(0x550c7dc3ul));
Round(e, f, g, h, a, b, c, d, K(0x72be5d74ul));
Round(d, e, f, g, h, a, b, c, K(0x80deb1feul));
Round(c, d, e, f, g, h, a, b, K(0x9bdc06a7ul));
Round(b, c, d, e, f, g, h, a, K(0xc19bf274ul));
Round(a, b, c, d, e, f, g, h, Add(K(0xe49b69c1ul), Inc(w0, sigma0(w1))));
Round(h, a, b, c, d, e, f, g, Add(K(0xefbe4786ul), Inc(w1, K(0xa00000ul), sigma0(w2))));
Round(g, h, a, b, c, d, e, f, Add(K(0x0fc19dc6ul), Inc(w2, sigma1(w0), sigma0(w3))));
Round(f, g, h, a, b, c, d, e, Add(K(0x240ca1ccul), Inc(w3, sigma1(w1), sigma0(w4))));
Round(e, f, g, h, a, b, c, d, Add(K(0x2de92c6ful), Inc(w4, sigma1(w2), sigma0(w5))));
Round(d, e, f, g, h, a, b, c, Add(K(0x4a7484aaul), Inc(w5, sigma1(w3), sigma0(w6))));
Round(c, d, e, f, g, h, a, b, Add(K(0x5cb0a9dcul), Inc(w6, sigma1(w4), K(0x100ul), sigma0(w7))));
Round(b, c, d, e, f, g, h, a, Add(K(0x76f988daul), Inc(w7, sigma1(w5), w0, K(0x11002000ul))));
Round(a, b, c, d, e, f, g, h, Add(K(0x983e5152ul), w8 = Add(K(0x80000000ul), sigma1(w6), w1)));
Round(h, a, b, c, d, e, f, g, Add(K(0xa831c66dul), w9 = Add(sigma1(w7), w2)));
Round(g, h, a, b, c, d, e, f, Add(K(0xb00327c8ul), w10 = Add(sigma1(w8), w3)));
Round(f, g, h, a, b, c, d, e, Add(K(0xbf597fc7ul), w11 = Add(sigma1(w9), w4)));
Round(e, f, g, h, a, b, c, d, Add(K(0xc6e00bf3ul), w12 = Add(sigma1(w10), w5)));
Round(d, e, f, g, h, a, b, c, Add(K(0xd5a79147ul), w13 = Add(sigma1(w11), w6)));
Round(c, d, e, f, g, h, a, b, Add(K(0x06ca6351ul), w14 = Add(sigma1(w12), w7, K(0x400022ul))));
Round(b, c, d, e, f, g, h, a, Add(K(0x14292967ul), w15 = Add(K(0x100ul), sigma1(w13), w8, sigma0(w0))));
Round(a, b, c, d, e, f, g, h, Add(K(0x27b70a85ul), Inc(w0, sigma1(w14), w9, sigma0(w1))));
Round(h, a, b, c, d, e, f, g, Add(K(0x2e1b2138ul), Inc(w1, sigma1(w15), w10, sigma0(w2))));
Round(g, h, a, b, c, d, e, f, Add(K(0x4d2c6dfcul), Inc(w2, sigma1(w0), w11, sigma0(w3))));
Round(f, g, h, a, b, c, d, e, Add(K(0x53380d13ul), Inc(w3, sigma1(w1), w12, sigma0(w4))));
Round(e, f, g, h, a, b, c, d, Add(K(0x650a7354ul), Inc(w4, sigma1(w2), w13, sigma0(w5))));
Round(d, e, f, g, h, a, b, c, Add(K(0x766a0abbul), Inc(w5, sigma1(w3), w14, sigma0(w6))));
Round(c, d, e, f, g, h, a, b, Add(K(0x81c2c92eul), Inc(w6, sigma1(w4), w15, sigma0(w7))));
Round(b, c, d, e, f, g, h, a, Add(K(0x92722c85ul), Inc(w7, sigma1(w5), w0, sigma0(w8))));
Round(a, b, c, d, e, f, g, h, Add(K(0xa2bfe8a1ul), Inc(w8, sigma1(w6), w1, sigma0(w9))));
Round(h, a, b, c, d, e, f, g, Add(K(0xa81a664bul), Inc(w9, sigma1(w7), w2, sigma0(w10))));
Round(g, h, a, b, c, d, e, f, Add(K(0xc24b8b70ul), Inc(w10, sigma1(w8), w3, sigma0(w11))));
Round(f, g, h, a, b, c, d, e, Add(K(0xc76c51a3ul), Inc(w11, sigma1(w9), w4, sigma0(w12))));
Round(e, f, g, h, a, b, c, d, Add(K(0xd192e819ul), Inc(w12, sigma1(w10), w5, sigma0(w13))));
Round(d, e, f, g, h, a, b, c, Add(K(0xd6990624ul), Inc(w13, sigma1(w11), w6, sigma0(w14))));
Round(c, d, e, f, g, h, a, b, Add(K(0xf40e3585ul), Inc(w14, sigma1(w12), w7, sigma0(w15))));
Round(b, c, d, e, f, g, h, a, Add(K(0x106aa070ul), Inc(w15, sigma1(w13), w8, sigma0(w0))));
Round(a, b, c, d, e, f, g, h, Add(K(0x19a4c116ul), Inc(w0, sigma1(w14), w9, sigma0(w1))));
Round(h, a, b, c, d, e, f, g, Add(K(0x1e376c08ul), Inc(w1, sigma1(w15), w10, sigma0(w2))));
Round(g, h, a, b, c, d, e, f, Add(K(0x2748774cul), Inc(w2, sigma1(w0), w11, sigma0(w3))));
Round(f, g, h, a, b, c, d, e, Add(K(0x34b0bcb5ul), Inc(w3, sigma1(w1), w12, sigma0(w4))));
Round(e, f, g, h, a, b, c, d, Add(K(0x391c0cb3ul), Inc(w4, sigma1(w2), w13, sigma0(w5))));
Round(d, e, f, g, h, a, b, c, Add(K(0x4ed8aa4aul), Inc(w5, sigma1(w3), w14, sigma0(w6))));
Round(c, d, e, f, g, h, a, b, Add(K(0x5b9cca4ful), Inc(w6, sigma1(w4), w15, sigma0(w7))));
Round(b, c, d, e, f, g, h, a, Add(K(0x682e6ff3ul), Inc(w7, sigma1(w5), w0, sigma0(w8))));
Round(a, b, c, d, e, f, g, h, Add(K(0x748f82eeul), Inc(w8, sigma1(w6), w1, sigma0(w9))));
Round(h, a, b, c, d, e, f, g, Add(K(0x78a5636ful), Inc(w9, sigma1(w7), w2, sigma0(w10))));
Round(g, h, a, b, c, d, e, f, Add(K(0x84c87814ul), Inc(w10, sigma1(w8), w3, sigma0(w11))));
Round(f, g, h, a, b, c, d, e, Add(K(0x8cc70208ul), Inc(w11, sigma1(w9), w4, sigma0(w12))));
Round(e, f, g, h, a, b, c, d, Add(K(0x90befffaul), Inc(w12, sigma1(w10), w5, sigma0(w13))));
Round(d, e, f, g, h, a, b, c, Add(K(0xa4506cebul), Inc(w13, sigma1(w11), w6, sigma0(w14))));
Round(c, d, e, f, g, h, a, b, Add(K(0xbef9a3f7ul), w14, sigma1(w12), w7, sigma0(w15)));
Round(b, c, d, e, f, g, h, a, Add(K(0xc67178f2ul), w15, sigma1(w13), w8, sigma0(w0)));
// Output
Write4(out, 0, Add(a, K(0x6a09e667ul)));
Write4(out, 4, Add(b, K(0xbb67ae85ul)));
Write4(out, 8, Add(c, K(0x3c6ef372ul)));
Write4(out, 12, Add(d, K(0xa54ff53aul)));
Write4(out, 16, Add(e, K(0x510e527ful)));
Write4(out, 20, Add(f, K(0x9b05688cul)));
Write4(out, 24, Add(g, K(0x1f83d9abul)));
Write4(out, 28, Add(h, K(0x5be0cd19ul)));
}
}
#endif

161
src/crypto/sha3.cpp Normal file
View file

@ -0,0 +1,161 @@
// Copyright (c) 2020 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
// Based on https://github.com/mjosaarinen/tiny_sha3/blob/master/sha3.c
// by Markku-Juhani O. Saarinen <mjos@iki.fi>
#include <crypto/sha3.h>
#include <crypto/common.h>
#include <span.h>
#include <algorithm>
#include <array> // For std::begin and std::end.
#include <stdint.h>
// Internal implementation code.
namespace
{
uint64_t Rotl(uint64_t x, int n) { return (x << n) | (x >> (64 - n)); }
} // namespace
void KeccakF(uint64_t (&st)[25])
{
static constexpr uint64_t RNDC[24] = {
0x0000000000000001, 0x0000000000008082, 0x800000000000808a, 0x8000000080008000,
0x000000000000808b, 0x0000000080000001, 0x8000000080008081, 0x8000000000008009,
0x000000000000008a, 0x0000000000000088, 0x0000000080008009, 0x000000008000000a,
0x000000008000808b, 0x800000000000008b, 0x8000000000008089, 0x8000000000008003,
0x8000000000008002, 0x8000000000000080, 0x000000000000800a, 0x800000008000000a,
0x8000000080008081, 0x8000000000008080, 0x0000000080000001, 0x8000000080008008
};
static constexpr int ROUNDS = 24;
for (int round = 0; round < ROUNDS; ++round) {
uint64_t bc0, bc1, bc2, bc3, bc4, t;
// Theta
bc0 = st[0] ^ st[5] ^ st[10] ^ st[15] ^ st[20];
bc1 = st[1] ^ st[6] ^ st[11] ^ st[16] ^ st[21];
bc2 = st[2] ^ st[7] ^ st[12] ^ st[17] ^ st[22];
bc3 = st[3] ^ st[8] ^ st[13] ^ st[18] ^ st[23];
bc4 = st[4] ^ st[9] ^ st[14] ^ st[19] ^ st[24];
t = bc4 ^ Rotl(bc1, 1); st[0] ^= t; st[5] ^= t; st[10] ^= t; st[15] ^= t; st[20] ^= t;
t = bc0 ^ Rotl(bc2, 1); st[1] ^= t; st[6] ^= t; st[11] ^= t; st[16] ^= t; st[21] ^= t;
t = bc1 ^ Rotl(bc3, 1); st[2] ^= t; st[7] ^= t; st[12] ^= t; st[17] ^= t; st[22] ^= t;
t = bc2 ^ Rotl(bc4, 1); st[3] ^= t; st[8] ^= t; st[13] ^= t; st[18] ^= t; st[23] ^= t;
t = bc3 ^ Rotl(bc0, 1); st[4] ^= t; st[9] ^= t; st[14] ^= t; st[19] ^= t; st[24] ^= t;
// Rho Pi
t = st[1];
bc0 = st[10]; st[10] = Rotl(t, 1); t = bc0;
bc0 = st[7]; st[7] = Rotl(t, 3); t = bc0;
bc0 = st[11]; st[11] = Rotl(t, 6); t = bc0;
bc0 = st[17]; st[17] = Rotl(t, 10); t = bc0;
bc0 = st[18]; st[18] = Rotl(t, 15); t = bc0;
bc0 = st[3]; st[3] = Rotl(t, 21); t = bc0;
bc0 = st[5]; st[5] = Rotl(t, 28); t = bc0;
bc0 = st[16]; st[16] = Rotl(t, 36); t = bc0;
bc0 = st[8]; st[8] = Rotl(t, 45); t = bc0;
bc0 = st[21]; st[21] = Rotl(t, 55); t = bc0;
bc0 = st[24]; st[24] = Rotl(t, 2); t = bc0;
bc0 = st[4]; st[4] = Rotl(t, 14); t = bc0;
bc0 = st[15]; st[15] = Rotl(t, 27); t = bc0;
bc0 = st[23]; st[23] = Rotl(t, 41); t = bc0;
bc0 = st[19]; st[19] = Rotl(t, 56); t = bc0;
bc0 = st[13]; st[13] = Rotl(t, 8); t = bc0;
bc0 = st[12]; st[12] = Rotl(t, 25); t = bc0;
bc0 = st[2]; st[2] = Rotl(t, 43); t = bc0;
bc0 = st[20]; st[20] = Rotl(t, 62); t = bc0;
bc0 = st[14]; st[14] = Rotl(t, 18); t = bc0;
bc0 = st[22]; st[22] = Rotl(t, 39); t = bc0;
bc0 = st[9]; st[9] = Rotl(t, 61); t = bc0;
bc0 = st[6]; st[6] = Rotl(t, 20); t = bc0;
st[1] = Rotl(t, 44);
// Chi Iota
bc0 = st[0]; bc1 = st[1]; bc2 = st[2]; bc3 = st[3]; bc4 = st[4];
st[0] = bc0 ^ (~bc1 & bc2) ^ RNDC[round];
st[1] = bc1 ^ (~bc2 & bc3);
st[2] = bc2 ^ (~bc3 & bc4);
st[3] = bc3 ^ (~bc4 & bc0);
st[4] = bc4 ^ (~bc0 & bc1);
bc0 = st[5]; bc1 = st[6]; bc2 = st[7]; bc3 = st[8]; bc4 = st[9];
st[5] = bc0 ^ (~bc1 & bc2);
st[6] = bc1 ^ (~bc2 & bc3);
st[7] = bc2 ^ (~bc3 & bc4);
st[8] = bc3 ^ (~bc4 & bc0);
st[9] = bc4 ^ (~bc0 & bc1);
bc0 = st[10]; bc1 = st[11]; bc2 = st[12]; bc3 = st[13]; bc4 = st[14];
st[10] = bc0 ^ (~bc1 & bc2);
st[11] = bc1 ^ (~bc2 & bc3);
st[12] = bc2 ^ (~bc3 & bc4);
st[13] = bc3 ^ (~bc4 & bc0);
st[14] = bc4 ^ (~bc0 & bc1);
bc0 = st[15]; bc1 = st[16]; bc2 = st[17]; bc3 = st[18]; bc4 = st[19];
st[15] = bc0 ^ (~bc1 & bc2);
st[16] = bc1 ^ (~bc2 & bc3);
st[17] = bc2 ^ (~bc3 & bc4);
st[18] = bc3 ^ (~bc4 & bc0);
st[19] = bc4 ^ (~bc0 & bc1);
bc0 = st[20]; bc1 = st[21]; bc2 = st[22]; bc3 = st[23]; bc4 = st[24];
st[20] = bc0 ^ (~bc1 & bc2);
st[21] = bc1 ^ (~bc2 & bc3);
st[22] = bc2 ^ (~bc3 & bc4);
st[23] = bc3 ^ (~bc4 & bc0);
st[24] = bc4 ^ (~bc0 & bc1);
}
}
SHA3_256& SHA3_256::Write(Span<const unsigned char> data)
{
if (m_bufsize && m_bufsize + data.size() >= sizeof(m_buffer)) {
// Fill the buffer and process it.
std::copy(data.begin(), data.begin() + sizeof(m_buffer) - m_bufsize, m_buffer + m_bufsize);
data = data.subspan(sizeof(m_buffer) - m_bufsize);
m_state[m_pos++] ^= ReadLE64(m_buffer);
m_bufsize = 0;
if (m_pos == RATE_BUFFERS) {
KeccakF(m_state);
m_pos = 0;
}
}
while (data.size() >= sizeof(m_buffer)) {
// Process chunks directly from the buffer.
m_state[m_pos++] ^= ReadLE64(data.data());
data = data.subspan(8);
if (m_pos == RATE_BUFFERS) {
KeccakF(m_state);
m_pos = 0;
}
}
if (data.size()) {
// Keep the remainder in the buffer.
std::copy(data.begin(), data.end(), m_buffer + m_bufsize);
m_bufsize += data.size();
}
return *this;
}
SHA3_256& SHA3_256::Finalize(Span<unsigned char> output)
{
assert(output.size() == OUTPUT_SIZE);
std::fill(m_buffer + m_bufsize, m_buffer + sizeof(m_buffer), 0);
m_buffer[m_bufsize] ^= 0x06;
m_state[m_pos] ^= ReadLE64(m_buffer);
m_state[RATE_BUFFERS - 1] ^= 0x8000000000000000;
KeccakF(m_state);
for (unsigned i = 0; i < 4; ++i) {
WriteLE64(output.data() + 8 * i, m_state[i]);
}
return *this;
}
SHA3_256& SHA3_256::Reset()
{
m_bufsize = 0;
m_pos = 0;
std::fill(std::begin(m_state), std::end(m_state), 0);
return *this;
}

41
src/crypto/sha3.h Normal file
View file

@ -0,0 +1,41 @@
// Copyright (c) 2020 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#ifndef DIGIBYTE_CRYPTO_SHA3_H
#define DIGIBYTE_CRYPTO_SHA3_H
#include <span.h>
#include <stdint.h>
#include <stdlib.h>
//! The Keccak-f[1600] transform.
void KeccakF(uint64_t (&st)[25]);
class SHA3_256
{
private:
uint64_t m_state[25] = {0};
unsigned char m_buffer[8];
unsigned m_bufsize = 0;
unsigned m_pos = 0;
//! Sponge rate in bits.
static constexpr unsigned RATE_BITS = 1088;
//! Sponge rate expressed as a multiple of the buffer size.
static constexpr unsigned RATE_BUFFERS = RATE_BITS / (8 * sizeof(m_buffer));
static_assert(RATE_BITS % (8 * sizeof(m_buffer)) == 0, "Rate must be a multiple of 8 bytes");
public:
static constexpr size_t OUTPUT_SIZE = 32;
SHA3_256() {}
SHA3_256& Write(Span<const unsigned char> data);
SHA3_256& Finalize(Span<unsigned char> output);
SHA3_256& Reset();
};
#endif // DIGIBYTE_CRYPTO_SHA3_H

208
src/crypto/sha512.cpp Normal file
View file

@ -0,0 +1,208 @@
// Copyright (c) 2009-2020 The Bitcoin Core developers
// Copyright (c) 2014-2020 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#include <crypto/sha512.h>
#include <crypto/common.h>
#include <string.h>
// Internal implementation code.
namespace
{
/// Internal SHA-512 implementation.
namespace sha512
{
uint64_t inline Ch(uint64_t x, uint64_t y, uint64_t z) { return z ^ (x & (y ^ z)); }
uint64_t inline Maj(uint64_t x, uint64_t y, uint64_t z) { return (x & y) | (z & (x | y)); }
uint64_t inline Sigma0(uint64_t x) { return (x >> 28 | x << 36) ^ (x >> 34 | x << 30) ^ (x >> 39 | x << 25); }
uint64_t inline Sigma1(uint64_t x) { return (x >> 14 | x << 50) ^ (x >> 18 | x << 46) ^ (x >> 41 | x << 23); }
uint64_t inline sigma0(uint64_t x) { return (x >> 1 | x << 63) ^ (x >> 8 | x << 56) ^ (x >> 7); }
uint64_t inline sigma1(uint64_t x) { return (x >> 19 | x << 45) ^ (x >> 61 | x << 3) ^ (x >> 6); }
/** One round of SHA-512. */
void inline Round(uint64_t a, uint64_t b, uint64_t c, uint64_t& d, uint64_t e, uint64_t f, uint64_t g, uint64_t& h, uint64_t k, uint64_t w)
{
uint64_t t1 = h + Sigma1(e) + Ch(e, f, g) + k + w;
uint64_t t2 = Sigma0(a) + Maj(a, b, c);
d += t1;
h = t1 + t2;
}
/** Initialize SHA-256 state. */
void inline Initialize(uint64_t* s)
{
s[0] = 0x6a09e667f3bcc908ull;
s[1] = 0xbb67ae8584caa73bull;
s[2] = 0x3c6ef372fe94f82bull;
s[3] = 0xa54ff53a5f1d36f1ull;
s[4] = 0x510e527fade682d1ull;
s[5] = 0x9b05688c2b3e6c1full;
s[6] = 0x1f83d9abfb41bd6bull;
s[7] = 0x5be0cd19137e2179ull;
}
/** Perform one SHA-512 transformation, processing a 128-byte chunk. */
void Transform(uint64_t* s, const unsigned char* chunk)
{
uint64_t a = s[0], b = s[1], c = s[2], d = s[3], e = s[4], f = s[5], g = s[6], h = s[7];
uint64_t w0, w1, w2, w3, w4, w5, w6, w7, w8, w9, w10, w11, w12, w13, w14, w15;
Round(a, b, c, d, e, f, g, h, 0x428a2f98d728ae22ull, w0 = ReadBE64(chunk + 0));
Round(h, a, b, c, d, e, f, g, 0x7137449123ef65cdull, w1 = ReadBE64(chunk + 8));
Round(g, h, a, b, c, d, e, f, 0xb5c0fbcfec4d3b2full, w2 = ReadBE64(chunk + 16));
Round(f, g, h, a, b, c, d, e, 0xe9b5dba58189dbbcull, w3 = ReadBE64(chunk + 24));
Round(e, f, g, h, a, b, c, d, 0x3956c25bf348b538ull, w4 = ReadBE64(chunk + 32));
Round(d, e, f, g, h, a, b, c, 0x59f111f1b605d019ull, w5 = ReadBE64(chunk + 40));
Round(c, d, e, f, g, h, a, b, 0x923f82a4af194f9bull, w6 = ReadBE64(chunk + 48));
Round(b, c, d, e, f, g, h, a, 0xab1c5ed5da6d8118ull, w7 = ReadBE64(chunk + 56));
Round(a, b, c, d, e, f, g, h, 0xd807aa98a3030242ull, w8 = ReadBE64(chunk + 64));
Round(h, a, b, c, d, e, f, g, 0x12835b0145706fbeull, w9 = ReadBE64(chunk + 72));
Round(g, h, a, b, c, d, e, f, 0x243185be4ee4b28cull, w10 = ReadBE64(chunk + 80));
Round(f, g, h, a, b, c, d, e, 0x550c7dc3d5ffb4e2ull, w11 = ReadBE64(chunk + 88));
Round(e, f, g, h, a, b, c, d, 0x72be5d74f27b896full, w12 = ReadBE64(chunk + 96));
Round(d, e, f, g, h, a, b, c, 0x80deb1fe3b1696b1ull, w13 = ReadBE64(chunk + 104));
Round(c, d, e, f, g, h, a, b, 0x9bdc06a725c71235ull, w14 = ReadBE64(chunk + 112));
Round(b, c, d, e, f, g, h, a, 0xc19bf174cf692694ull, w15 = ReadBE64(chunk + 120));
Round(a, b, c, d, e, f, g, h, 0xe49b69c19ef14ad2ull, w0 += sigma1(w14) + w9 + sigma0(w1));
Round(h, a, b, c, d, e, f, g, 0xefbe4786384f25e3ull, w1 += sigma1(w15) + w10 + sigma0(w2));
Round(g, h, a, b, c, d, e, f, 0x0fc19dc68b8cd5b5ull, w2 += sigma1(w0) + w11 + sigma0(w3));
Round(f, g, h, a, b, c, d, e, 0x240ca1cc77ac9c65ull, w3 += sigma1(w1) + w12 + sigma0(w4));
Round(e, f, g, h, a, b, c, d, 0x2de92c6f592b0275ull, w4 += sigma1(w2) + w13 + sigma0(w5));
Round(d, e, f, g, h, a, b, c, 0x4a7484aa6ea6e483ull, w5 += sigma1(w3) + w14 + sigma0(w6));
Round(c, d, e, f, g, h, a, b, 0x5cb0a9dcbd41fbd4ull, w6 += sigma1(w4) + w15 + sigma0(w7));
Round(b, c, d, e, f, g, h, a, 0x76f988da831153b5ull, w7 += sigma1(w5) + w0 + sigma0(w8));
Round(a, b, c, d, e, f, g, h, 0x983e5152ee66dfabull, w8 += sigma1(w6) + w1 + sigma0(w9));
Round(h, a, b, c, d, e, f, g, 0xa831c66d2db43210ull, w9 += sigma1(w7) + w2 + sigma0(w10));
Round(g, h, a, b, c, d, e, f, 0xb00327c898fb213full, w10 += sigma1(w8) + w3 + sigma0(w11));
Round(f, g, h, a, b, c, d, e, 0xbf597fc7beef0ee4ull, w11 += sigma1(w9) + w4 + sigma0(w12));
Round(e, f, g, h, a, b, c, d, 0xc6e00bf33da88fc2ull, w12 += sigma1(w10) + w5 + sigma0(w13));
Round(d, e, f, g, h, a, b, c, 0xd5a79147930aa725ull, w13 += sigma1(w11) + w6 + sigma0(w14));
Round(c, d, e, f, g, h, a, b, 0x06ca6351e003826full, w14 += sigma1(w12) + w7 + sigma0(w15));
Round(b, c, d, e, f, g, h, a, 0x142929670a0e6e70ull, w15 += sigma1(w13) + w8 + sigma0(w0));
Round(a, b, c, d, e, f, g, h, 0x27b70a8546d22ffcull, w0 += sigma1(w14) + w9 + sigma0(w1));
Round(h, a, b, c, d, e, f, g, 0x2e1b21385c26c926ull, w1 += sigma1(w15) + w10 + sigma0(w2));
Round(g, h, a, b, c, d, e, f, 0x4d2c6dfc5ac42aedull, w2 += sigma1(w0) + w11 + sigma0(w3));
Round(f, g, h, a, b, c, d, e, 0x53380d139d95b3dfull, w3 += sigma1(w1) + w12 + sigma0(w4));
Round(e, f, g, h, a, b, c, d, 0x650a73548baf63deull, w4 += sigma1(w2) + w13 + sigma0(w5));
Round(d, e, f, g, h, a, b, c, 0x766a0abb3c77b2a8ull, w5 += sigma1(w3) + w14 + sigma0(w6));
Round(c, d, e, f, g, h, a, b, 0x81c2c92e47edaee6ull, w6 += sigma1(w4) + w15 + sigma0(w7));
Round(b, c, d, e, f, g, h, a, 0x92722c851482353bull, w7 += sigma1(w5) + w0 + sigma0(w8));
Round(a, b, c, d, e, f, g, h, 0xa2bfe8a14cf10364ull, w8 += sigma1(w6) + w1 + sigma0(w9));
Round(h, a, b, c, d, e, f, g, 0xa81a664bbc423001ull, w9 += sigma1(w7) + w2 + sigma0(w10));
Round(g, h, a, b, c, d, e, f, 0xc24b8b70d0f89791ull, w10 += sigma1(w8) + w3 + sigma0(w11));
Round(f, g, h, a, b, c, d, e, 0xc76c51a30654be30ull, w11 += sigma1(w9) + w4 + sigma0(w12));
Round(e, f, g, h, a, b, c, d, 0xd192e819d6ef5218ull, w12 += sigma1(w10) + w5 + sigma0(w13));
Round(d, e, f, g, h, a, b, c, 0xd69906245565a910ull, w13 += sigma1(w11) + w6 + sigma0(w14));
Round(c, d, e, f, g, h, a, b, 0xf40e35855771202aull, w14 += sigma1(w12) + w7 + sigma0(w15));
Round(b, c, d, e, f, g, h, a, 0x106aa07032bbd1b8ull, w15 += sigma1(w13) + w8 + sigma0(w0));
Round(a, b, c, d, e, f, g, h, 0x19a4c116b8d2d0c8ull, w0 += sigma1(w14) + w9 + sigma0(w1));
Round(h, a, b, c, d, e, f, g, 0x1e376c085141ab53ull, w1 += sigma1(w15) + w10 + sigma0(w2));
Round(g, h, a, b, c, d, e, f, 0x2748774cdf8eeb99ull, w2 += sigma1(w0) + w11 + sigma0(w3));
Round(f, g, h, a, b, c, d, e, 0x34b0bcb5e19b48a8ull, w3 += sigma1(w1) + w12 + sigma0(w4));
Round(e, f, g, h, a, b, c, d, 0x391c0cb3c5c95a63ull, w4 += sigma1(w2) + w13 + sigma0(w5));
Round(d, e, f, g, h, a, b, c, 0x4ed8aa4ae3418acbull, w5 += sigma1(w3) + w14 + sigma0(w6));
Round(c, d, e, f, g, h, a, b, 0x5b9cca4f7763e373ull, w6 += sigma1(w4) + w15 + sigma0(w7));
Round(b, c, d, e, f, g, h, a, 0x682e6ff3d6b2b8a3ull, w7 += sigma1(w5) + w0 + sigma0(w8));
Round(a, b, c, d, e, f, g, h, 0x748f82ee5defb2fcull, w8 += sigma1(w6) + w1 + sigma0(w9));
Round(h, a, b, c, d, e, f, g, 0x78a5636f43172f60ull, w9 += sigma1(w7) + w2 + sigma0(w10));
Round(g, h, a, b, c, d, e, f, 0x84c87814a1f0ab72ull, w10 += sigma1(w8) + w3 + sigma0(w11));
Round(f, g, h, a, b, c, d, e, 0x8cc702081a6439ecull, w11 += sigma1(w9) + w4 + sigma0(w12));
Round(e, f, g, h, a, b, c, d, 0x90befffa23631e28ull, w12 += sigma1(w10) + w5 + sigma0(w13));
Round(d, e, f, g, h, a, b, c, 0xa4506cebde82bde9ull, w13 += sigma1(w11) + w6 + sigma0(w14));
Round(c, d, e, f, g, h, a, b, 0xbef9a3f7b2c67915ull, w14 += sigma1(w12) + w7 + sigma0(w15));
Round(b, c, d, e, f, g, h, a, 0xc67178f2e372532bull, w15 += sigma1(w13) + w8 + sigma0(w0));
Round(a, b, c, d, e, f, g, h, 0xca273eceea26619cull, w0 += sigma1(w14) + w9 + sigma0(w1));
Round(h, a, b, c, d, e, f, g, 0xd186b8c721c0c207ull, w1 += sigma1(w15) + w10 + sigma0(w2));
Round(g, h, a, b, c, d, e, f, 0xeada7dd6cde0eb1eull, w2 += sigma1(w0) + w11 + sigma0(w3));
Round(f, g, h, a, b, c, d, e, 0xf57d4f7fee6ed178ull, w3 += sigma1(w1) + w12 + sigma0(w4));
Round(e, f, g, h, a, b, c, d, 0x06f067aa72176fbaull, w4 += sigma1(w2) + w13 + sigma0(w5));
Round(d, e, f, g, h, a, b, c, 0x0a637dc5a2c898a6ull, w5 += sigma1(w3) + w14 + sigma0(w6));
Round(c, d, e, f, g, h, a, b, 0x113f9804bef90daeull, w6 += sigma1(w4) + w15 + sigma0(w7));
Round(b, c, d, e, f, g, h, a, 0x1b710b35131c471bull, w7 += sigma1(w5) + w0 + sigma0(w8));
Round(a, b, c, d, e, f, g, h, 0x28db77f523047d84ull, w8 += sigma1(w6) + w1 + sigma0(w9));
Round(h, a, b, c, d, e, f, g, 0x32caab7b40c72493ull, w9 += sigma1(w7) + w2 + sigma0(w10));
Round(g, h, a, b, c, d, e, f, 0x3c9ebe0a15c9bebcull, w10 += sigma1(w8) + w3 + sigma0(w11));
Round(f, g, h, a, b, c, d, e, 0x431d67c49c100d4cull, w11 += sigma1(w9) + w4 + sigma0(w12));
Round(e, f, g, h, a, b, c, d, 0x4cc5d4becb3e42b6ull, w12 += sigma1(w10) + w5 + sigma0(w13));
Round(d, e, f, g, h, a, b, c, 0x597f299cfc657e2aull, w13 += sigma1(w11) + w6 + sigma0(w14));
Round(c, d, e, f, g, h, a, b, 0x5fcb6fab3ad6faecull, w14 + sigma1(w12) + w7 + sigma0(w15));
Round(b, c, d, e, f, g, h, a, 0x6c44198c4a475817ull, w15 + sigma1(w13) + w8 + sigma0(w0));
s[0] += a;
s[1] += b;
s[2] += c;
s[3] += d;
s[4] += e;
s[5] += f;
s[6] += g;
s[7] += h;
}
} // namespace sha512
} // namespace
////// SHA-512
CSHA512::CSHA512() : bytes(0)
{
sha512::Initialize(s);
}
CSHA512& CSHA512::Write(const unsigned char* data, size_t len)
{
const unsigned char* end = data + len;
size_t bufsize = bytes % 128;
if (bufsize && bufsize + len >= 128) {
// Fill the buffer, and process it.
memcpy(buf + bufsize, data, 128 - bufsize);
bytes += 128 - bufsize;
data += 128 - bufsize;
sha512::Transform(s, buf);
bufsize = 0;
}
while (end - data >= 128) {
// Process full chunks directly from the source.
sha512::Transform(s, data);
data += 128;
bytes += 128;
}
if (end > data) {
// Fill the buffer with what remains.
memcpy(buf + bufsize, data, end - data);
bytes += end - data;
}
return *this;
}
void CSHA512::Finalize(unsigned char hash[OUTPUT_SIZE])
{
static const unsigned char pad[128] = {0x80};
unsigned char sizedesc[16] = {0x00};
WriteBE64(sizedesc + 8, bytes << 3);
Write(pad, 1 + ((239 - (bytes % 128)) % 128));
Write(sizedesc, 16);
WriteBE64(hash, s[0]);
WriteBE64(hash + 8, s[1]);
WriteBE64(hash + 16, s[2]);
WriteBE64(hash + 24, s[3]);
WriteBE64(hash + 32, s[4]);
WriteBE64(hash + 40, s[5]);
WriteBE64(hash + 48, s[6]);
WriteBE64(hash + 56, s[7]);
}
CSHA512& CSHA512::Reset()
{
bytes = 0;
sha512::Initialize(s);
return *this;
}

30
src/crypto/sha512.h Normal file
View file

@ -0,0 +1,30 @@
// Copyright (c) 2014-2020 The Bitcoin Core developers
// Copyright (c) 2014-2020 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#ifndef DIGIBYTE_CRYPTO_SHA512_H
#define DIGIBYTE_CRYPTO_SHA512_H
#include <stdint.h>
#include <stdlib.h>
/** A hasher class for SHA-512. */
class CSHA512
{
private:
uint64_t s[8];
unsigned char buf[128];
size_t bytes;
public:
static constexpr size_t OUTPUT_SIZE = 64;
CSHA512();
CSHA512& Write(const unsigned char* data, size_t len);
void Finalize(unsigned char hash[OUTPUT_SIZE]);
CSHA512& Reset();
uint64_t Size() const { return bytes; }
};
#endif // DIGIBYTE_CRYPTO_SHA512_H

1764
src/crypto/shavite.cpp Normal file
View file

File diff suppressed because it is too large Load diff

1799
src/crypto/simd.cpp Normal file
View file

File diff suppressed because it is too large Load diff

173
src/crypto/siphash.cpp Normal file
View file

@ -0,0 +1,173 @@
// Copyright (c) 2016-2020 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#include <crypto/siphash.h>
#define ROTL(x, b) (uint64_t)(((x) << (b)) | ((x) >> (64 - (b))))
#define SIPROUND do { \
v0 += v1; v1 = ROTL(v1, 13); v1 ^= v0; \
v0 = ROTL(v0, 32); \
v2 += v3; v3 = ROTL(v3, 16); v3 ^= v2; \
v0 += v3; v3 = ROTL(v3, 21); v3 ^= v0; \
v2 += v1; v1 = ROTL(v1, 17); v1 ^= v2; \
v2 = ROTL(v2, 32); \
} while (0)
CSipHasher::CSipHasher(uint64_t k0, uint64_t k1)
{
v[0] = 0x736f6d6570736575ULL ^ k0;
v[1] = 0x646f72616e646f6dULL ^ k1;
v[2] = 0x6c7967656e657261ULL ^ k0;
v[3] = 0x7465646279746573ULL ^ k1;
count = 0;
tmp = 0;
}
CSipHasher& CSipHasher::Write(uint64_t data)
{
uint64_t v0 = v[0], v1 = v[1], v2 = v[2], v3 = v[3];
assert(count % 8 == 0);
v3 ^= data;
SIPROUND;
SIPROUND;
v0 ^= data;
v[0] = v0;
v[1] = v1;
v[2] = v2;
v[3] = v3;
count += 8;
return *this;
}
CSipHasher& CSipHasher::Write(const unsigned char* data, size_t size)
{
uint64_t v0 = v[0], v1 = v[1], v2 = v[2], v3 = v[3];
uint64_t t = tmp;
uint8_t c = count;
while (size--) {
t |= ((uint64_t)(*(data++))) << (8 * (c % 8));
c++;
if ((c & 7) == 0) {
v3 ^= t;
SIPROUND;
SIPROUND;
v0 ^= t;
t = 0;
}
}
v[0] = v0;
v[1] = v1;
v[2] = v2;
v[3] = v3;
count = c;
tmp = t;
return *this;
}
uint64_t CSipHasher::Finalize() const
{
uint64_t v0 = v[0], v1 = v[1], v2 = v[2], v3 = v[3];
uint64_t t = tmp | (((uint64_t)count) << 56);
v3 ^= t;
SIPROUND;
SIPROUND;
v0 ^= t;
v2 ^= 0xFF;
SIPROUND;
SIPROUND;
SIPROUND;
SIPROUND;
return v0 ^ v1 ^ v2 ^ v3;
}
uint64_t SipHashUint256(uint64_t k0, uint64_t k1, const uint256& val)
{
/* Specialized implementation for efficiency */
uint64_t d = val.GetUint64(0);
uint64_t v0 = 0x736f6d6570736575ULL ^ k0;
uint64_t v1 = 0x646f72616e646f6dULL ^ k1;
uint64_t v2 = 0x6c7967656e657261ULL ^ k0;
uint64_t v3 = 0x7465646279746573ULL ^ k1 ^ d;
SIPROUND;
SIPROUND;
v0 ^= d;
d = val.GetUint64(1);
v3 ^= d;
SIPROUND;
SIPROUND;
v0 ^= d;
d = val.GetUint64(2);
v3 ^= d;
SIPROUND;
SIPROUND;
v0 ^= d;
d = val.GetUint64(3);
v3 ^= d;
SIPROUND;
SIPROUND;
v0 ^= d;
v3 ^= ((uint64_t)4) << 59;
SIPROUND;
SIPROUND;
v0 ^= ((uint64_t)4) << 59;
v2 ^= 0xFF;
SIPROUND;
SIPROUND;
SIPROUND;
SIPROUND;
return v0 ^ v1 ^ v2 ^ v3;
}
uint64_t SipHashUint256Extra(uint64_t k0, uint64_t k1, const uint256& val, uint32_t extra)
{
/* Specialized implementation for efficiency */
uint64_t d = val.GetUint64(0);
uint64_t v0 = 0x736f6d6570736575ULL ^ k0;
uint64_t v1 = 0x646f72616e646f6dULL ^ k1;
uint64_t v2 = 0x6c7967656e657261ULL ^ k0;
uint64_t v3 = 0x7465646279746573ULL ^ k1 ^ d;
SIPROUND;
SIPROUND;
v0 ^= d;
d = val.GetUint64(1);
v3 ^= d;
SIPROUND;
SIPROUND;
v0 ^= d;
d = val.GetUint64(2);
v3 ^= d;
SIPROUND;
SIPROUND;
v0 ^= d;
d = val.GetUint64(3);
v3 ^= d;
SIPROUND;
SIPROUND;
v0 ^= d;
d = (((uint64_t)36) << 56) | extra;
v3 ^= d;
SIPROUND;
SIPROUND;
v0 ^= d;
v2 ^= 0xFF;
SIPROUND;
SIPROUND;
SIPROUND;
SIPROUND;
return v0 ^ v1 ^ v2 ^ v3;
}

47
src/crypto/siphash.h Normal file
View file

@ -0,0 +1,47 @@
// Copyright (c) 2016-2020 The DigiByte Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.
#ifndef DIGIBYTE_CRYPTO_SIPHASH_H
#define DIGIBYTE_CRYPTO_SIPHASH_H
#include <stdint.h>
#include <uint256.h>
/** SipHash-2-4 */
class CSipHasher
{
private:
uint64_t v[4];
uint64_t tmp;
uint8_t count; // Only the low 8 bits of the input size matter.
public:
/** Construct a SipHash calculator initialized with 128-bit key (k0, k1) */
CSipHasher(uint64_t k0, uint64_t k1);
/** Hash a 64-bit integer worth of data
* It is treated as if this was the little-endian interpretation of 8 bytes.
* This function can only be used when a multiple of 8 bytes have been written so far.
*/
CSipHasher& Write(uint64_t data);
/** Hash arbitrary bytes. */
CSipHasher& Write(const unsigned char* data, size_t size);
/** Compute the 64-bit SipHash-2-4 of the data written so far. The object remains untouched. */
uint64_t Finalize() const;
};
/** Optimized SipHash-2-4 implementation for uint256.
*
* It is identical to:
* SipHasher(k0, k1)
* .Write(val.GetUint64(0))
* .Write(val.GetUint64(1))
* .Write(val.GetUint64(2))
* .Write(val.GetUint64(3))
* .Finalize()
*/
uint64_t SipHashUint256(uint64_t k0, uint64_t k1, const uint256& val);
uint64_t SipHashUint256Extra(uint64_t k0, uint64_t k1, const uint256& val, uint32_t extra);
#endif // DIGIBYTE_CRYPTO_SIPHASH_H

327
src/crypto/sph_blake.h Normal file
View file

@ -0,0 +1,327 @@
/* $Id: sph_blake.h 252 2011-06-07 17:55:14Z tp $ */
/**
* BLAKE interface. BLAKE is a family of functions which differ by their
* output size; this implementation defines BLAKE for output sizes 224,
* 256, 384 and 512 bits. This implementation conforms to the "third
* round" specification.
*
* ==========================(LICENSE BEGIN)============================
*
* Copyright (c) 2007-2010 Projet RNRT SAPHIR
*
* Permission is hereby granted, free of charge, to any person obtaining
* a copy of this software and associated documentation files (the
* "Software"), to deal in the Software without restriction, including
* without limitation the rights to use, copy, modify, merge, publish,
* distribute, sublicense, and/or sell copies of the Software, and to
* permit persons to whom the Software is furnished to do so, subject to
* the following conditions:
*
* The above copyright notice and this permission notice shall be
* included in all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*
* ===========================(LICENSE END)=============================
*
* @file sph_blake.h
* @author Thomas Pornin <thomas.pornin@cryptolog.com>
*/
#ifndef SPH_BLAKE_H__
#define SPH_BLAKE_H__
#ifdef __cplusplus
extern "C"{
#endif
#include <stddef.h>
#include "sph_types.h"
/**
* Output size (in bits) for BLAKE-224.
*/
#define SPH_SIZE_blake224 224
/**
* Output size (in bits) for BLAKE-256.
*/
#define SPH_SIZE_blake256 256
#if SPH_64
/**
* Output size (in bits) for BLAKE-384.
*/
#define SPH_SIZE_blake384 384
/**
* Output size (in bits) for BLAKE-512.
*/
#define SPH_SIZE_blake512 512
#endif
/**
* This structure is a context for BLAKE-224 and BLAKE-256 computations:
* it contains the intermediate values and some data from the last
* entered block. Once a BLAKE computation has been performed, the
* context can be reused for another computation.
*
* The contents of this structure are private. A running BLAKE
* computation can be cloned by copying the context (e.g. with a simple
* <code>memcpy()</code>).
*/
typedef struct {
#ifndef DOXYGEN_IGNORE
unsigned char buf[64]; /* first field, for alignment */
size_t ptr;
sph_u32 H[8];
sph_u32 S[4];
sph_u32 T0, T1;
#endif
} sph_blake_small_context;
/**
* This structure is a context for BLAKE-224 computations. It is
* identical to the common <code>sph_blake_small_context</code>.
*/
typedef sph_blake_small_context sph_blake224_context;
/**
* This structure is a context for BLAKE-256 computations. It is
* identical to the common <code>sph_blake_small_context</code>.
*/
typedef sph_blake_small_context sph_blake256_context;
#if SPH_64
/**
* This structure is a context for BLAKE-384 and BLAKE-512 computations:
* it contains the intermediate values and some data from the last
* entered block. Once a BLAKE computation has been performed, the
* context can be reused for another computation.
*
* The contents of this structure are private. A running BLAKE
* computation can be cloned by copying the context (e.g. with a simple
* <code>memcpy()</code>).
*/
typedef struct {
#ifndef DOXYGEN_IGNORE
unsigned char buf[128]; /* first field, for alignment */
size_t ptr;
sph_u64 H[8];
sph_u64 S[4];
sph_u64 T0, T1;
#endif
} sph_blake_big_context;
/**
* This structure is a context for BLAKE-384 computations. It is
* identical to the common <code>sph_blake_small_context</code>.
*/
typedef sph_blake_big_context sph_blake384_context;
/**
* This structure is a context for BLAKE-512 computations. It is
* identical to the common <code>sph_blake_small_context</code>.
*/
typedef sph_blake_big_context sph_blake512_context;
#endif
/**
* Initialize a BLAKE-224 context. This process performs no memory allocation.
*
* @param cc the BLAKE-224 context (pointer to a
* <code>sph_blake224_context</code>)
*/
void sph_blake224_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the BLAKE-224 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_blake224(void *cc, const void *data, size_t len);
/**
* Terminate the current BLAKE-224 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (28 bytes). The context is automatically
* reinitialized.
*
* @param cc the BLAKE-224 context
* @param dst the destination buffer
*/
void sph_blake224_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (28 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the BLAKE-224 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_blake224_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize a BLAKE-256 context. This process performs no memory allocation.
*
* @param cc the BLAKE-256 context (pointer to a
* <code>sph_blake256_context</code>)
*/
void sph_blake256_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the BLAKE-256 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_blake256(void *cc, const void *data, size_t len);
/**
* Terminate the current BLAKE-256 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (32 bytes). The context is automatically
* reinitialized.
*
* @param cc the BLAKE-256 context
* @param dst the destination buffer
*/
void sph_blake256_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (32 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the BLAKE-256 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_blake256_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
#if SPH_64
/**
* Initialize a BLAKE-384 context. This process performs no memory allocation.
*
* @param cc the BLAKE-384 context (pointer to a
* <code>sph_blake384_context</code>)
*/
void sph_blake384_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the BLAKE-384 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_blake384(void *cc, const void *data, size_t len);
/**
* Terminate the current BLAKE-384 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (48 bytes). The context is automatically
* reinitialized.
*
* @param cc the BLAKE-384 context
* @param dst the destination buffer
*/
void sph_blake384_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (48 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the BLAKE-384 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_blake384_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize a BLAKE-512 context. This process performs no memory allocation.
*
* @param cc the BLAKE-512 context (pointer to a
* <code>sph_blake512_context</code>)
*/
void sph_blake512_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the BLAKE-512 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_blake512(void *cc, const void *data, size_t len);
/**
* Terminate the current BLAKE-512 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (64 bytes). The context is automatically
* reinitialized.
*
* @param cc the BLAKE-512 context
* @param dst the destination buffer
*/
void sph_blake512_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (64 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the BLAKE-512 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_blake512_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
#endif
#ifdef __cplusplus
}
#endif
#endif

328
src/crypto/sph_bmw.h Normal file
View file

@ -0,0 +1,328 @@
/* $Id: sph_bmw.h 216 2010-06-08 09:46:57Z tp $ */
/**
* BMW interface. BMW (aka "Blue Midnight Wish") is a family of
* functions which differ by their output size; this implementation
* defines BMW for output sizes 224, 256, 384 and 512 bits.
*
* ==========================(LICENSE BEGIN)============================
*
* Copyright (c) 2007-2010 Projet RNRT SAPHIR
*
* Permission is hereby granted, free of charge, to any person obtaining
* a copy of this software and associated documentation files (the
* "Software"), to deal in the Software without restriction, including
* without limitation the rights to use, copy, modify, merge, publish,
* distribute, sublicense, and/or sell copies of the Software, and to
* permit persons to whom the Software is furnished to do so, subject to
* the following conditions:
*
* The above copyright notice and this permission notice shall be
* included in all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*
* ===========================(LICENSE END)=============================
*
* @file sph_bmw.h
* @author Thomas Pornin <thomas.pornin@cryptolog.com>
*/
#ifndef SPH_BMW_H__
#define SPH_BMW_H__
#ifdef __cplusplus
extern "C"{
#endif
#include <stddef.h>
#include "sph_types.h"
/**
* Output size (in bits) for BMW-224.
*/
#define SPH_SIZE_bmw224 224
/**
* Output size (in bits) for BMW-256.
*/
#define SPH_SIZE_bmw256 256
#if SPH_64
/**
* Output size (in bits) for BMW-384.
*/
#define SPH_SIZE_bmw384 384
/**
* Output size (in bits) for BMW-512.
*/
#define SPH_SIZE_bmw512 512
#endif
/**
* This structure is a context for BMW-224 and BMW-256 computations:
* it contains the intermediate values and some data from the last
* entered block. Once a BMW computation has been performed, the
* context can be reused for another computation.
*
* The contents of this structure are private. A running BMW
* computation can be cloned by copying the context (e.g. with a simple
* <code>memcpy()</code>).
*/
typedef struct {
#ifndef DOXYGEN_IGNORE
unsigned char buf[64]; /* first field, for alignment */
size_t ptr;
sph_u32 H[16];
#if SPH_64
sph_u64 bit_count;
#else
sph_u32 bit_count_high, bit_count_low;
#endif
#endif
} sph_bmw_small_context;
/**
* This structure is a context for BMW-224 computations. It is
* identical to the common <code>sph_bmw_small_context</code>.
*/
typedef sph_bmw_small_context sph_bmw224_context;
/**
* This structure is a context for BMW-256 computations. It is
* identical to the common <code>sph_bmw_small_context</code>.
*/
typedef sph_bmw_small_context sph_bmw256_context;
#if SPH_64
/**
* This structure is a context for BMW-384 and BMW-512 computations:
* it contains the intermediate values and some data from the last
* entered block. Once a BMW computation has been performed, the
* context can be reused for another computation.
*
* The contents of this structure are private. A running BMW
* computation can be cloned by copying the context (e.g. with a simple
* <code>memcpy()</code>).
*/
typedef struct {
#ifndef DOXYGEN_IGNORE
unsigned char buf[128]; /* first field, for alignment */
size_t ptr;
sph_u64 H[16];
sph_u64 bit_count;
#endif
} sph_bmw_big_context;
/**
* This structure is a context for BMW-384 computations. It is
* identical to the common <code>sph_bmw_small_context</code>.
*/
typedef sph_bmw_big_context sph_bmw384_context;
/**
* This structure is a context for BMW-512 computations. It is
* identical to the common <code>sph_bmw_small_context</code>.
*/
typedef sph_bmw_big_context sph_bmw512_context;
#endif
/**
* Initialize a BMW-224 context. This process performs no memory allocation.
*
* @param cc the BMW-224 context (pointer to a
* <code>sph_bmw224_context</code>)
*/
void sph_bmw224_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the BMW-224 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_bmw224(void *cc, const void *data, size_t len);
/**
* Terminate the current BMW-224 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (28 bytes). The context is automatically
* reinitialized.
*
* @param cc the BMW-224 context
* @param dst the destination buffer
*/
void sph_bmw224_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (28 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the BMW-224 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_bmw224_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize a BMW-256 context. This process performs no memory allocation.
*
* @param cc the BMW-256 context (pointer to a
* <code>sph_bmw256_context</code>)
*/
void sph_bmw256_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the BMW-256 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_bmw256(void *cc, const void *data, size_t len);
/**
* Terminate the current BMW-256 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (32 bytes). The context is automatically
* reinitialized.
*
* @param cc the BMW-256 context
* @param dst the destination buffer
*/
void sph_bmw256_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (32 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the BMW-256 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_bmw256_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
#if SPH_64
/**
* Initialize a BMW-384 context. This process performs no memory allocation.
*
* @param cc the BMW-384 context (pointer to a
* <code>sph_bmw384_context</code>)
*/
void sph_bmw384_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the BMW-384 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_bmw384(void *cc, const void *data, size_t len);
/**
* Terminate the current BMW-384 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (48 bytes). The context is automatically
* reinitialized.
*
* @param cc the BMW-384 context
* @param dst the destination buffer
*/
void sph_bmw384_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (48 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the BMW-384 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_bmw384_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize a BMW-512 context. This process performs no memory allocation.
*
* @param cc the BMW-512 context (pointer to a
* <code>sph_bmw512_context</code>)
*/
void sph_bmw512_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the BMW-512 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_bmw512(void *cc, const void *data, size_t len);
/**
* Terminate the current BMW-512 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (64 bytes). The context is automatically
* reinitialized.
*
* @param cc the BMW-512 context
* @param dst the destination buffer
*/
void sph_bmw512_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (64 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the BMW-512 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_bmw512_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
#endif
#ifdef __cplusplus
}
#endif
#endif

292
src/crypto/sph_cubehash.h Normal file
View file

@ -0,0 +1,292 @@
/* $Id: sph_cubehash.h 180 2010-05-08 02:29:25Z tp $ */
/**
* CubeHash interface. CubeHash is a family of functions which differ by
* their output size; this implementation defines CubeHash for output
* sizes 224, 256, 384 and 512 bits, with the "standard parameters"
* (CubeHash16/32 with the CubeHash specification notations).
*
* ==========================(LICENSE BEGIN)============================
*
* Copyright (c) 2007-2010 Projet RNRT SAPHIR
*
* Permission is hereby granted, free of charge, to any person obtaining
* a copy of this software and associated documentation files (the
* "Software"), to deal in the Software without restriction, including
* without limitation the rights to use, copy, modify, merge, publish,
* distribute, sublicense, and/or sell copies of the Software, and to
* permit persons to whom the Software is furnished to do so, subject to
* the following conditions:
*
* The above copyright notice and this permission notice shall be
* included in all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*
* ===========================(LICENSE END)=============================
*
* @file sph_cubehash.h
* @author Thomas Pornin <thomas.pornin@cryptolog.com>
*/
#ifndef SPH_CUBEHASH_H__
#define SPH_CUBEHASH_H__
#ifdef __cplusplus
extern "C"{
#endif
#include <stddef.h>
#include "sph_types.h"
/**
* Output size (in bits) for CubeHash-224.
*/
#define SPH_SIZE_cubehash224 224
/**
* Output size (in bits) for CubeHash-256.
*/
#define SPH_SIZE_cubehash256 256
/**
* Output size (in bits) for CubeHash-384.
*/
#define SPH_SIZE_cubehash384 384
/**
* Output size (in bits) for CubeHash-512.
*/
#define SPH_SIZE_cubehash512 512
/**
* This structure is a context for CubeHash computations: it contains the
* intermediate values and some data from the last entered block. Once
* a CubeHash computation has been performed, the context can be reused for
* another computation.
*
* The contents of this structure are private. A running CubeHash computation
* can be cloned by copying the context (e.g. with a simple
* <code>memcpy()</code>).
*/
typedef struct {
#ifndef DOXYGEN_IGNORE
unsigned char buf[32]; /* first field, for alignment */
size_t ptr;
sph_u32 state[32];
#endif
} sph_cubehash_context;
/**
* Type for a CubeHash-224 context (identical to the common context).
*/
typedef sph_cubehash_context sph_cubehash224_context;
/**
* Type for a CubeHash-256 context (identical to the common context).
*/
typedef sph_cubehash_context sph_cubehash256_context;
/**
* Type for a CubeHash-384 context (identical to the common context).
*/
typedef sph_cubehash_context sph_cubehash384_context;
/**
* Type for a CubeHash-512 context (identical to the common context).
*/
typedef sph_cubehash_context sph_cubehash512_context;
/**
* Initialize a CubeHash-224 context. This process performs no memory
* allocation.
*
* @param cc the CubeHash-224 context (pointer to a
* <code>sph_cubehash224_context</code>)
*/
void sph_cubehash224_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the CubeHash-224 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_cubehash224(void *cc, const void *data, size_t len);
/**
* Terminate the current CubeHash-224 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (28 bytes). The context is automatically
* reinitialized.
*
* @param cc the CubeHash-224 context
* @param dst the destination buffer
*/
void sph_cubehash224_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (28 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the CubeHash-224 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_cubehash224_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize a CubeHash-256 context. This process performs no memory
* allocation.
*
* @param cc the CubeHash-256 context (pointer to a
* <code>sph_cubehash256_context</code>)
*/
void sph_cubehash256_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the CubeHash-256 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_cubehash256(void *cc, const void *data, size_t len);
/**
* Terminate the current CubeHash-256 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (32 bytes). The context is automatically
* reinitialized.
*
* @param cc the CubeHash-256 context
* @param dst the destination buffer
*/
void sph_cubehash256_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (32 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the CubeHash-256 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_cubehash256_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize a CubeHash-384 context. This process performs no memory
* allocation.
*
* @param cc the CubeHash-384 context (pointer to a
* <code>sph_cubehash384_context</code>)
*/
void sph_cubehash384_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the CubeHash-384 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_cubehash384(void *cc, const void *data, size_t len);
/**
* Terminate the current CubeHash-384 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (48 bytes). The context is automatically
* reinitialized.
*
* @param cc the CubeHash-384 context
* @param dst the destination buffer
*/
void sph_cubehash384_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (48 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the CubeHash-384 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_cubehash384_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize a CubeHash-512 context. This process performs no memory
* allocation.
*
* @param cc the CubeHash-512 context (pointer to a
* <code>sph_cubehash512_context</code>)
*/
void sph_cubehash512_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the CubeHash-512 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_cubehash512(void *cc, const void *data, size_t len);
/**
* Terminate the current CubeHash-512 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (64 bytes). The context is automatically
* reinitialized.
*
* @param cc the CubeHash-512 context
* @param dst the destination buffer
*/
void sph_cubehash512_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (64 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the CubeHash-512 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_cubehash512_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
#ifdef __cplusplus
}
#endif
#endif

320
src/crypto/sph_echo.h Normal file
View file

@ -0,0 +1,320 @@
/* $Id: sph_echo.h 216 2010-06-08 09:46:57Z tp $ */
/**
* ECHO interface. ECHO is a family of functions which differ by
* their output size; this implementation defines ECHO for output
* sizes 224, 256, 384 and 512 bits.
*
* ==========================(LICENSE BEGIN)============================
*
* Copyright (c) 2007-2010 Projet RNRT SAPHIR
*
* Permission is hereby granted, free of charge, to any person obtaining
* a copy of this software and associated documentation files (the
* "Software"), to deal in the Software without restriction, including
* without limitation the rights to use, copy, modify, merge, publish,
* distribute, sublicense, and/or sell copies of the Software, and to
* permit persons to whom the Software is furnished to do so, subject to
* the following conditions:
*
* The above copyright notice and this permission notice shall be
* included in all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*
* ===========================(LICENSE END)=============================
*
* @file sph_echo.h
* @author Thomas Pornin <thomas.pornin@cryptolog.com>
*/
#ifndef SPH_ECHO_H__
#define SPH_ECHO_H__
#ifdef __cplusplus
extern "C"{
#endif
#include <stddef.h>
#include "sph_types.h"
/**
* Output size (in bits) for ECHO-224.
*/
#define SPH_SIZE_echo224 224
/**
* Output size (in bits) for ECHO-256.
*/
#define SPH_SIZE_echo256 256
/**
* Output size (in bits) for ECHO-384.
*/
#define SPH_SIZE_echo384 384
/**
* Output size (in bits) for ECHO-512.
*/
#define SPH_SIZE_echo512 512
/**
* This structure is a context for ECHO computations: it contains the
* intermediate values and some data from the last entered block. Once
* an ECHO computation has been performed, the context can be reused for
* another computation. This specific structure is used for ECHO-224
* and ECHO-256.
*
* The contents of this structure are private. A running ECHO computation
* can be cloned by copying the context (e.g. with a simple
* <code>memcpy()</code>).
*/
typedef struct {
#ifndef DOXYGEN_IGNORE
unsigned char buf[192]; /* first field, for alignment */
size_t ptr;
union {
sph_u32 Vs[4][4];
#if SPH_64
sph_u64 Vb[4][2];
#endif
} u;
sph_u32 C0, C1, C2, C3;
#endif
} sph_echo_small_context;
/**
* This structure is a context for ECHO computations: it contains the
* intermediate values and some data from the last entered block. Once
* an ECHO computation has been performed, the context can be reused for
* another computation. This specific structure is used for ECHO-384
* and ECHO-512.
*
* The contents of this structure are private. A running ECHO computation
* can be cloned by copying the context (e.g. with a simple
* <code>memcpy()</code>).
*/
typedef struct {
#ifndef DOXYGEN_IGNORE
unsigned char buf[128]; /* first field, for alignment */
size_t ptr;
union {
sph_u32 Vs[8][4];
#if SPH_64
sph_u64 Vb[8][2];
#endif
} u;
sph_u32 C0, C1, C2, C3;
#endif
} sph_echo_big_context;
/**
* Type for a ECHO-224 context (identical to the common "small" context).
*/
typedef sph_echo_small_context sph_echo224_context;
/**
* Type for a ECHO-256 context (identical to the common "small" context).
*/
typedef sph_echo_small_context sph_echo256_context;
/**
* Type for a ECHO-384 context (identical to the common "big" context).
*/
typedef sph_echo_big_context sph_echo384_context;
/**
* Type for a ECHO-512 context (identical to the common "big" context).
*/
typedef sph_echo_big_context sph_echo512_context;
/**
* Initialize an ECHO-224 context. This process performs no memory allocation.
*
* @param cc the ECHO-224 context (pointer to a
* <code>sph_echo224_context</code>)
*/
void sph_echo224_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the ECHO-224 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_echo224(void *cc, const void *data, size_t len);
/**
* Terminate the current ECHO-224 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (28 bytes). The context is automatically
* reinitialized.
*
* @param cc the ECHO-224 context
* @param dst the destination buffer
*/
void sph_echo224_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (28 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the ECHO-224 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_echo224_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize an ECHO-256 context. This process performs no memory allocation.
*
* @param cc the ECHO-256 context (pointer to a
* <code>sph_echo256_context</code>)
*/
void sph_echo256_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the ECHO-256 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_echo256(void *cc, const void *data, size_t len);
/**
* Terminate the current ECHO-256 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (32 bytes). The context is automatically
* reinitialized.
*
* @param cc the ECHO-256 context
* @param dst the destination buffer
*/
void sph_echo256_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (32 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the ECHO-256 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_echo256_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize an ECHO-384 context. This process performs no memory allocation.
*
* @param cc the ECHO-384 context (pointer to a
* <code>sph_echo384_context</code>)
*/
void sph_echo384_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the ECHO-384 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_echo384(void *cc, const void *data, size_t len);
/**
* Terminate the current ECHO-384 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (48 bytes). The context is automatically
* reinitialized.
*
* @param cc the ECHO-384 context
* @param dst the destination buffer
*/
void sph_echo384_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (48 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the ECHO-384 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_echo384_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize an ECHO-512 context. This process performs no memory allocation.
*
* @param cc the ECHO-512 context (pointer to a
* <code>sph_echo512_context</code>)
*/
void sph_echo512_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the ECHO-512 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_echo512(void *cc, const void *data, size_t len);
/**
* Terminate the current ECHO-512 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (64 bytes). The context is automatically
* reinitialized.
*
* @param cc the ECHO-512 context
* @param dst the destination buffer
*/
void sph_echo512_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (64 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the ECHO-512 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_echo512_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
#ifdef __cplusplus
}
#endif
#endif

298
src/crypto/sph_jh.h Normal file
View file

@ -0,0 +1,298 @@
/* $Id: sph_jh.h 216 2010-06-08 09:46:57Z tp $ */
/**
* JH interface. JH is a family of functions which differ by
* their output size; this implementation defines JH for output
* sizes 224, 256, 384 and 512 bits.
*
* ==========================(LICENSE BEGIN)============================
*
* Copyright (c) 2007-2010 Projet RNRT SAPHIR
*
* Permission is hereby granted, free of charge, to any person obtaining
* a copy of this software and associated documentation files (the
* "Software"), to deal in the Software without restriction, including
* without limitation the rights to use, copy, modify, merge, publish,
* distribute, sublicense, and/or sell copies of the Software, and to
* permit persons to whom the Software is furnished to do so, subject to
* the following conditions:
*
* The above copyright notice and this permission notice shall be
* included in all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*
* ===========================(LICENSE END)=============================
*
* @file sph_jh.h
* @author Thomas Pornin <thomas.pornin@cryptolog.com>
*/
#ifndef SPH_JH_H__
#define SPH_JH_H__
#ifdef __cplusplus
extern "C"{
#endif
#include <stddef.h>
#include "sph_types.h"
/**
* Output size (in bits) for JH-224.
*/
#define SPH_SIZE_jh224 224
/**
* Output size (in bits) for JH-256.
*/
#define SPH_SIZE_jh256 256
/**
* Output size (in bits) for JH-384.
*/
#define SPH_SIZE_jh384 384
/**
* Output size (in bits) for JH-512.
*/
#define SPH_SIZE_jh512 512
/**
* This structure is a context for JH computations: it contains the
* intermediate values and some data from the last entered block. Once
* a JH computation has been performed, the context can be reused for
* another computation.
*
* The contents of this structure are private. A running JH computation
* can be cloned by copying the context (e.g. with a simple
* <code>memcpy()</code>).
*/
typedef struct {
#ifndef DOXYGEN_IGNORE
unsigned char buf[64]; /* first field, for alignment */
size_t ptr;
union {
#if SPH_64
sph_u64 wide[16];
#endif
sph_u32 narrow[32];
} H;
#if SPH_64
sph_u64 block_count;
#else
sph_u32 block_count_high, block_count_low;
#endif
#endif
} sph_jh_context;
/**
* Type for a JH-224 context (identical to the common context).
*/
typedef sph_jh_context sph_jh224_context;
/**
* Type for a JH-256 context (identical to the common context).
*/
typedef sph_jh_context sph_jh256_context;
/**
* Type for a JH-384 context (identical to the common context).
*/
typedef sph_jh_context sph_jh384_context;
/**
* Type for a JH-512 context (identical to the common context).
*/
typedef sph_jh_context sph_jh512_context;
/**
* Initialize a JH-224 context. This process performs no memory allocation.
*
* @param cc the JH-224 context (pointer to a
* <code>sph_jh224_context</code>)
*/
void sph_jh224_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the JH-224 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_jh224(void *cc, const void *data, size_t len);
/**
* Terminate the current JH-224 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (28 bytes). The context is automatically
* reinitialized.
*
* @param cc the JH-224 context
* @param dst the destination buffer
*/
void sph_jh224_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (28 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the JH-224 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_jh224_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize a JH-256 context. This process performs no memory allocation.
*
* @param cc the JH-256 context (pointer to a
* <code>sph_jh256_context</code>)
*/
void sph_jh256_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the JH-256 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_jh256(void *cc, const void *data, size_t len);
/**
* Terminate the current JH-256 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (32 bytes). The context is automatically
* reinitialized.
*
* @param cc the JH-256 context
* @param dst the destination buffer
*/
void sph_jh256_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (32 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the JH-256 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_jh256_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize a JH-384 context. This process performs no memory allocation.
*
* @param cc the JH-384 context (pointer to a
* <code>sph_jh384_context</code>)
*/
void sph_jh384_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the JH-384 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_jh384(void *cc, const void *data, size_t len);
/**
* Terminate the current JH-384 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (48 bytes). The context is automatically
* reinitialized.
*
* @param cc the JH-384 context
* @param dst the destination buffer
*/
void sph_jh384_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (48 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the JH-384 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_jh384_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize a JH-512 context. This process performs no memory allocation.
*
* @param cc the JH-512 context (pointer to a
* <code>sph_jh512_context</code>)
*/
void sph_jh512_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the JH-512 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_jh512(void *cc, const void *data, size_t len);
/**
* Terminate the current JH-512 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (64 bytes). The context is automatically
* reinitialized.
*
* @param cc the JH-512 context
* @param dst the destination buffer
*/
void sph_jh512_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (64 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the JH-512 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_jh512_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
#ifdef __cplusplus
}
#endif
#endif

293
src/crypto/sph_keccak.h Normal file
View file

@ -0,0 +1,293 @@
/* $Id: sph_keccak.h 216 2010-06-08 09:46:57Z tp $ */
/**
* Keccak interface. This is the interface for Keccak with the
* recommended parameters for SHA-3, with output lengths 224, 256,
* 384 and 512 bits.
*
* ==========================(LICENSE BEGIN)============================
*
* Copyright (c) 2007-2010 Projet RNRT SAPHIR
*
* Permission is hereby granted, free of charge, to any person obtaining
* a copy of this software and associated documentation files (the
* "Software"), to deal in the Software without restriction, including
* without limitation the rights to use, copy, modify, merge, publish,
* distribute, sublicense, and/or sell copies of the Software, and to
* permit persons to whom the Software is furnished to do so, subject to
* the following conditions:
*
* The above copyright notice and this permission notice shall be
* included in all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*
* ===========================(LICENSE END)=============================
*
* @file sph_keccak.h
* @author Thomas Pornin <thomas.pornin@cryptolog.com>
*/
#ifndef SPH_KECCAK_H__
#define SPH_KECCAK_H__
#ifdef __cplusplus
extern "C"{
#endif
#include <stddef.h>
#include "sph_types.h"
/**
* Output size (in bits) for Keccak-224.
*/
#define SPH_SIZE_keccak224 224
/**
* Output size (in bits) for Keccak-256.
*/
#define SPH_SIZE_keccak256 256
/**
* Output size (in bits) for Keccak-384.
*/
#define SPH_SIZE_keccak384 384
/**
* Output size (in bits) for Keccak-512.
*/
#define SPH_SIZE_keccak512 512
/**
* This structure is a context for Keccak computations: it contains the
* intermediate values and some data from the last entered block. Once a
* Keccak computation has been performed, the context can be reused for
* another computation.
*
* The contents of this structure are private. A running Keccak computation
* can be cloned by copying the context (e.g. with a simple
* <code>memcpy()</code>).
*/
typedef struct {
#ifndef DOXYGEN_IGNORE
unsigned char buf[144]; /* first field, for alignment */
size_t ptr, lim;
union {
#if SPH_64
sph_u64 wide[25];
#endif
sph_u32 narrow[50];
} u;
#endif
} sph_keccak_context;
/**
* Type for a Keccak-224 context (identical to the common context).
*/
typedef sph_keccak_context sph_keccak224_context;
/**
* Type for a Keccak-256 context (identical to the common context).
*/
typedef sph_keccak_context sph_keccak256_context;
/**
* Type for a Keccak-384 context (identical to the common context).
*/
typedef sph_keccak_context sph_keccak384_context;
/**
* Type for a Keccak-512 context (identical to the common context).
*/
typedef sph_keccak_context sph_keccak512_context;
/**
* Initialize a Keccak-224 context. This process performs no memory allocation.
*
* @param cc the Keccak-224 context (pointer to a
* <code>sph_keccak224_context</code>)
*/
void sph_keccak224_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the Keccak-224 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_keccak224(void *cc, const void *data, size_t len);
/**
* Terminate the current Keccak-224 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (28 bytes). The context is automatically
* reinitialized.
*
* @param cc the Keccak-224 context
* @param dst the destination buffer
*/
void sph_keccak224_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (28 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the Keccak-224 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_keccak224_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize a Keccak-256 context. This process performs no memory allocation.
*
* @param cc the Keccak-256 context (pointer to a
* <code>sph_keccak256_context</code>)
*/
void sph_keccak256_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the Keccak-256 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_keccak256(void *cc, const void *data, size_t len);
/**
* Terminate the current Keccak-256 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (32 bytes). The context is automatically
* reinitialized.
*
* @param cc the Keccak-256 context
* @param dst the destination buffer
*/
void sph_keccak256_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (32 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the Keccak-256 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_keccak256_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize a Keccak-384 context. This process performs no memory allocation.
*
* @param cc the Keccak-384 context (pointer to a
* <code>sph_keccak384_context</code>)
*/
void sph_keccak384_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the Keccak-384 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_keccak384(void *cc, const void *data, size_t len);
/**
* Terminate the current Keccak-384 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (48 bytes). The context is automatically
* reinitialized.
*
* @param cc the Keccak-384 context
* @param dst the destination buffer
*/
void sph_keccak384_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (48 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the Keccak-384 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_keccak384_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize a Keccak-512 context. This process performs no memory allocation.
*
* @param cc the Keccak-512 context (pointer to a
* <code>sph_keccak512_context</code>)
*/
void sph_keccak512_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the Keccak-512 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_keccak512(void *cc, const void *data, size_t len);
/**
* Terminate the current Keccak-512 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (64 bytes). The context is automatically
* reinitialized.
*
* @param cc the Keccak-512 context
* @param dst the destination buffer
*/
void sph_keccak512_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (64 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the Keccak-512 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_keccak512_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
#ifdef __cplusplus
}
#endif
#endif

296
src/crypto/sph_luffa.h Normal file
View file

@ -0,0 +1,296 @@
/* $Id: sph_luffa.h 154 2010-04-26 17:00:24Z tp $ */
/**
* Luffa interface. Luffa is a family of functions which differ by
* their output size; this implementation defines Luffa for output
* sizes 224, 256, 384 and 512 bits.
*
* ==========================(LICENSE BEGIN)============================
*
* Copyright (c) 2007-2010 Projet RNRT SAPHIR
*
* Permission is hereby granted, free of charge, to any person obtaining
* a copy of this software and associated documentation files (the
* "Software"), to deal in the Software without restriction, including
* without limitation the rights to use, copy, modify, merge, publish,
* distribute, sublicense, and/or sell copies of the Software, and to
* permit persons to whom the Software is furnished to do so, subject to
* the following conditions:
*
* The above copyright notice and this permission notice shall be
* included in all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*
* ===========================(LICENSE END)=============================
*
* @file sph_luffa.h
* @author Thomas Pornin <thomas.pornin@cryptolog.com>
*/
#ifndef SPH_LUFFA_H__
#define SPH_LUFFA_H__
#ifdef __cplusplus
extern "C"{
#endif
#include <stddef.h>
#include "sph_types.h"
/**
* Output size (in bits) for Luffa-224.
*/
#define SPH_SIZE_luffa224 224
/**
* Output size (in bits) for Luffa-256.
*/
#define SPH_SIZE_luffa256 256
/**
* Output size (in bits) for Luffa-384.
*/
#define SPH_SIZE_luffa384 384
/**
* Output size (in bits) for Luffa-512.
*/
#define SPH_SIZE_luffa512 512
/**
* This structure is a context for Luffa-224 computations: it contains
* the intermediate values and some data from the last entered block.
* Once a Luffa computation has been performed, the context can be
* reused for another computation.
*
* The contents of this structure are private. A running Luffa
* computation can be cloned by copying the context (e.g. with a simple
* <code>memcpy()</code>).
*/
typedef struct {
#ifndef DOXYGEN_IGNORE
unsigned char buf[32]; /* first field, for alignment */
size_t ptr;
sph_u32 V[3][8];
#endif
} sph_luffa224_context;
/**
* This structure is a context for Luffa-256 computations. It is
* identical to <code>sph_luffa224_context</code>.
*/
typedef sph_luffa224_context sph_luffa256_context;
/**
* This structure is a context for Luffa-384 computations.
*/
typedef struct {
#ifndef DOXYGEN_IGNORE
unsigned char buf[32]; /* first field, for alignment */
size_t ptr;
sph_u32 V[4][8];
#endif
} sph_luffa384_context;
/**
* This structure is a context for Luffa-512 computations.
*/
typedef struct {
#ifndef DOXYGEN_IGNORE
unsigned char buf[32]; /* first field, for alignment */
size_t ptr;
sph_u32 V[5][8];
#endif
} sph_luffa512_context;
/**
* Initialize a Luffa-224 context. This process performs no memory allocation.
*
* @param cc the Luffa-224 context (pointer to a
* <code>sph_luffa224_context</code>)
*/
void sph_luffa224_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the Luffa-224 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_luffa224(void *cc, const void *data, size_t len);
/**
* Terminate the current Luffa-224 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (28 bytes). The context is automatically
* reinitialized.
*
* @param cc the Luffa-224 context
* @param dst the destination buffer
*/
void sph_luffa224_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (28 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the Luffa-224 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_luffa224_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize a Luffa-256 context. This process performs no memory allocation.
*
* @param cc the Luffa-256 context (pointer to a
* <code>sph_luffa256_context</code>)
*/
void sph_luffa256_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the Luffa-256 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_luffa256(void *cc, const void *data, size_t len);
/**
* Terminate the current Luffa-256 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (32 bytes). The context is automatically
* reinitialized.
*
* @param cc the Luffa-256 context
* @param dst the destination buffer
*/
void sph_luffa256_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (32 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the Luffa-256 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_luffa256_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize a Luffa-384 context. This process performs no memory allocation.
*
* @param cc the Luffa-384 context (pointer to a
* <code>sph_luffa384_context</code>)
*/
void sph_luffa384_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the Luffa-384 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_luffa384(void *cc, const void *data, size_t len);
/**
* Terminate the current Luffa-384 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (48 bytes). The context is automatically
* reinitialized.
*
* @param cc the Luffa-384 context
* @param dst the destination buffer
*/
void sph_luffa384_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (48 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the Luffa-384 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_luffa384_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize a Luffa-512 context. This process performs no memory allocation.
*
* @param cc the Luffa-512 context (pointer to a
* <code>sph_luffa512_context</code>)
*/
void sph_luffa512_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the Luffa-512 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_luffa512(void *cc, const void *data, size_t len);
/**
* Terminate the current Luffa-512 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (64 bytes). The context is automatically
* reinitialized.
*
* @param cc the Luffa-512 context
* @param dst the destination buffer
*/
void sph_luffa512_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (64 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the Luffa-512 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_luffa512_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
#ifdef __cplusplus
}
#endif
#endif

314
src/crypto/sph_shavite.h Normal file
View file

@ -0,0 +1,314 @@
/* $Id: sph_shavite.h 208 2010-06-02 20:33:00Z tp $ */
/**
* SHAvite-3 interface. This code implements SHAvite-3 with the
* recommended parameters for SHA-3, with outputs of 224, 256, 384 and
* 512 bits. In the following, we call the function "SHAvite" (without
* the "-3" suffix), thus "SHAvite-224" is "SHAvite-3 with a 224-bit
* output".
*
* ==========================(LICENSE BEGIN)============================
*
* Copyright (c) 2007-2010 Projet RNRT SAPHIR
*
* Permission is hereby granted, free of charge, to any person obtaining
* a copy of this software and associated documentation files (the
* "Software"), to deal in the Software without restriction, including
* without limitation the rights to use, copy, modify, merge, publish,
* distribute, sublicense, and/or sell copies of the Software, and to
* permit persons to whom the Software is furnished to do so, subject to
* the following conditions:
*
* The above copyright notice and this permission notice shall be
* included in all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*
* ===========================(LICENSE END)=============================
*
* @file sph_shavite.h
* @author Thomas Pornin <thomas.pornin@cryptolog.com>
*/
#ifndef SPH_SHAVITE_H__
#define SPH_SHAVITE_H__
#include <stddef.h>
#include "sph_types.h"
#ifdef __cplusplus
extern "C"{
#endif
/**
* Output size (in bits) for SHAvite-224.
*/
#define SPH_SIZE_shavite224 224
/**
* Output size (in bits) for SHAvite-256.
*/
#define SPH_SIZE_shavite256 256
/**
* Output size (in bits) for SHAvite-384.
*/
#define SPH_SIZE_shavite384 384
/**
* Output size (in bits) for SHAvite-512.
*/
#define SPH_SIZE_shavite512 512
/**
* This structure is a context for SHAvite-224 and SHAvite-256 computations:
* it contains the intermediate values and some data from the last
* entered block. Once a SHAvite computation has been performed, the
* context can be reused for another computation.
*
* The contents of this structure are private. A running SHAvite
* computation can be cloned by copying the context (e.g. with a simple
* <code>memcpy()</code>).
*/
typedef struct {
#ifndef DOXYGEN_IGNORE
unsigned char buf[64]; /* first field, for alignment */
size_t ptr;
sph_u32 h[8];
sph_u32 count0, count1;
#endif
} sph_shavite_small_context;
/**
* This structure is a context for SHAvite-224 computations. It is
* identical to the common <code>sph_shavite_small_context</code>.
*/
typedef sph_shavite_small_context sph_shavite224_context;
/**
* This structure is a context for SHAvite-256 computations. It is
* identical to the common <code>sph_shavite_small_context</code>.
*/
typedef sph_shavite_small_context sph_shavite256_context;
/**
* This structure is a context for SHAvite-384 and SHAvite-512 computations:
* it contains the intermediate values and some data from the last
* entered block. Once a SHAvite computation has been performed, the
* context can be reused for another computation.
*
* The contents of this structure are private. A running SHAvite
* computation can be cloned by copying the context (e.g. with a simple
* <code>memcpy()</code>).
*/
typedef struct {
#ifndef DOXYGEN_IGNORE
unsigned char buf[128]; /* first field, for alignment */
size_t ptr;
sph_u32 h[16];
sph_u32 count0, count1, count2, count3;
#endif
} sph_shavite_big_context;
/**
* This structure is a context for SHAvite-384 computations. It is
* identical to the common <code>sph_shavite_small_context</code>.
*/
typedef sph_shavite_big_context sph_shavite384_context;
/**
* This structure is a context for SHAvite-512 computations. It is
* identical to the common <code>sph_shavite_small_context</code>.
*/
typedef sph_shavite_big_context sph_shavite512_context;
/**
* Initialize a SHAvite-224 context. This process performs no memory allocation.
*
* @param cc the SHAvite-224 context (pointer to a
* <code>sph_shavite224_context</code>)
*/
void sph_shavite224_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the SHAvite-224 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_shavite224(void *cc, const void *data, size_t len);
/**
* Terminate the current SHAvite-224 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (28 bytes). The context is automatically
* reinitialized.
*
* @param cc the SHAvite-224 context
* @param dst the destination buffer
*/
void sph_shavite224_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (28 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the SHAvite-224 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_shavite224_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize a SHAvite-256 context. This process performs no memory allocation.
*
* @param cc the SHAvite-256 context (pointer to a
* <code>sph_shavite256_context</code>)
*/
void sph_shavite256_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the SHAvite-256 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_shavite256(void *cc, const void *data, size_t len);
/**
* Terminate the current SHAvite-256 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (32 bytes). The context is automatically
* reinitialized.
*
* @param cc the SHAvite-256 context
* @param dst the destination buffer
*/
void sph_shavite256_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (32 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the SHAvite-256 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_shavite256_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize a SHAvite-384 context. This process performs no memory allocation.
*
* @param cc the SHAvite-384 context (pointer to a
* <code>sph_shavite384_context</code>)
*/
void sph_shavite384_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the SHAvite-384 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_shavite384(void *cc, const void *data, size_t len);
/**
* Terminate the current SHAvite-384 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (48 bytes). The context is automatically
* reinitialized.
*
* @param cc the SHAvite-384 context
* @param dst the destination buffer
*/
void sph_shavite384_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (48 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the SHAvite-384 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_shavite384_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize a SHAvite-512 context. This process performs no memory allocation.
*
* @param cc the SHAvite-512 context (pointer to a
* <code>sph_shavite512_context</code>)
*/
void sph_shavite512_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the SHAvite-512 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_shavite512(void *cc, const void *data, size_t len);
/**
* Terminate the current SHAvite-512 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (64 bytes). The context is automatically
* reinitialized.
*
* @param cc the SHAvite-512 context
* @param dst the destination buffer
*/
void sph_shavite512_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (64 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the SHAvite-512 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_shavite512_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
#ifdef __cplusplus
}
#endif
#endif

309
src/crypto/sph_simd.h Normal file
View file

@ -0,0 +1,309 @@
/* $Id: sph_simd.h 154 2010-04-26 17:00:24Z tp $ */
/**
* SIMD interface. SIMD is a family of functions which differ by
* their output size; this implementation defines SIMD for output
* sizes 224, 256, 384 and 512 bits.
*
* ==========================(LICENSE BEGIN)============================
*
* Copyright (c) 2007-2010 Projet RNRT SAPHIR
*
* Permission is hereby granted, free of charge, to any person obtaining
* a copy of this software and associated documentation files (the
* "Software"), to deal in the Software without restriction, including
* without limitation the rights to use, copy, modify, merge, publish,
* distribute, sublicense, and/or sell copies of the Software, and to
* permit persons to whom the Software is furnished to do so, subject to
* the following conditions:
*
* The above copyright notice and this permission notice shall be
* included in all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*
* ===========================(LICENSE END)=============================
*
* @file sph_simd.h
* @author Thomas Pornin <thomas.pornin@cryptolog.com>
*/
#ifndef SPH_SIMD_H__
#define SPH_SIMD_H__
#ifdef __cplusplus
extern "C"{
#endif
#include <stddef.h>
#include "sph_types.h"
/**
* Output size (in bits) for SIMD-224.
*/
#define SPH_SIZE_simd224 224
/**
* Output size (in bits) for SIMD-256.
*/
#define SPH_SIZE_simd256 256
/**
* Output size (in bits) for SIMD-384.
*/
#define SPH_SIZE_simd384 384
/**
* Output size (in bits) for SIMD-512.
*/
#define SPH_SIZE_simd512 512
/**
* This structure is a context for SIMD computations: it contains the
* intermediate values and some data from the last entered block. Once
* an SIMD computation has been performed, the context can be reused for
* another computation. This specific structure is used for SIMD-224
* and SIMD-256.
*
* The contents of this structure are private. A running SIMD computation
* can be cloned by copying the context (e.g. with a simple
* <code>memcpy()</code>).
*/
typedef struct {
#ifndef DOXYGEN_IGNORE
unsigned char buf[64]; /* first field, for alignment */
size_t ptr;
sph_u32 state[16];
sph_u32 count_low, count_high;
#endif
} sph_simd_small_context;
/**
* This structure is a context for SIMD computations: it contains the
* intermediate values and some data from the last entered block. Once
* an SIMD computation has been performed, the context can be reused for
* another computation. This specific structure is used for SIMD-384
* and SIMD-512.
*
* The contents of this structure are private. A running SIMD computation
* can be cloned by copying the context (e.g. with a simple
* <code>memcpy()</code>).
*/
typedef struct {
#ifndef DOXYGEN_IGNORE
unsigned char buf[128]; /* first field, for alignment */
size_t ptr;
sph_u32 state[32];
sph_u32 count_low, count_high;
#endif
} sph_simd_big_context;
/**
* Type for a SIMD-224 context (identical to the common "small" context).
*/
typedef sph_simd_small_context sph_simd224_context;
/**
* Type for a SIMD-256 context (identical to the common "small" context).
*/
typedef sph_simd_small_context sph_simd256_context;
/**
* Type for a SIMD-384 context (identical to the common "big" context).
*/
typedef sph_simd_big_context sph_simd384_context;
/**
* Type for a SIMD-512 context (identical to the common "big" context).
*/
typedef sph_simd_big_context sph_simd512_context;
/**
* Initialize an SIMD-224 context. This process performs no memory allocation.
*
* @param cc the SIMD-224 context (pointer to a
* <code>sph_simd224_context</code>)
*/
void sph_simd224_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the SIMD-224 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_simd224(void *cc, const void *data, size_t len);
/**
* Terminate the current SIMD-224 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (28 bytes). The context is automatically
* reinitialized.
*
* @param cc the SIMD-224 context
* @param dst the destination buffer
*/
void sph_simd224_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (28 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the SIMD-224 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_simd224_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize an SIMD-256 context. This process performs no memory allocation.
*
* @param cc the SIMD-256 context (pointer to a
* <code>sph_simd256_context</code>)
*/
void sph_simd256_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the SIMD-256 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_simd256(void *cc, const void *data, size_t len);
/**
* Terminate the current SIMD-256 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (32 bytes). The context is automatically
* reinitialized.
*
* @param cc the SIMD-256 context
* @param dst the destination buffer
*/
void sph_simd256_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (32 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the SIMD-256 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_simd256_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize an SIMD-384 context. This process performs no memory allocation.
*
* @param cc the SIMD-384 context (pointer to a
* <code>sph_simd384_context</code>)
*/
void sph_simd384_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the SIMD-384 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_simd384(void *cc, const void *data, size_t len);
/**
* Terminate the current SIMD-384 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (48 bytes). The context is automatically
* reinitialized.
*
* @param cc the SIMD-384 context
* @param dst the destination buffer
*/
void sph_simd384_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (48 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the SIMD-384 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_simd384_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
/**
* Initialize an SIMD-512 context. This process performs no memory allocation.
*
* @param cc the SIMD-512 context (pointer to a
* <code>sph_simd512_context</code>)
*/
void sph_simd512_init(void *cc);
/**
* Process some data bytes. It is acceptable that <code>len</code> is zero
* (in which case this function does nothing).
*
* @param cc the SIMD-512 context
* @param data the input data
* @param len the input data length (in bytes)
*/
void sph_simd512(void *cc, const void *data, size_t len);
/**
* Terminate the current SIMD-512 computation and output the result into
* the provided buffer. The destination buffer must be wide enough to
* accomodate the result (64 bytes). The context is automatically
* reinitialized.
*
* @param cc the SIMD-512 context
* @param dst the destination buffer
*/
void sph_simd512_close(void *cc, void *dst);
/**
* Add a few additional bits (0 to 7) to the current computation, then
* terminate it and output the result in the provided buffer, which must
* be wide enough to accomodate the result (64 bytes). If bit number i
* in <code>ub</code> has value 2^i, then the extra bits are those
* numbered 7 downto 8-n (this is the big-endian convention at the byte
* level). The context is automatically reinitialized.
*
* @param cc the SIMD-512 context
* @param ub the extra bits
* @param n the number of extra bits (0 to 7)
* @param dst the destination buffer
*/
void sph_simd512_addbits_and_close(
void *cc, unsigned ub, unsigned n, void *dst);
#ifdef __cplusplus
}
#endif
#endif

1976
src/crypto/sph_types.h Normal file
View file

File diff suppressed because it is too large Load diff

138
src/crypto/sysendian.h Normal file
View file

@ -0,0 +1,138 @@
/*-
* Copyright 2007-2009 Colin Percival
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. 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 AUTHOR 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 AUTHOR 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.
*
* This file was originally written by Colin Percival as part of the Tarsnap
* online backup system.
*/
#ifndef _SYSENDIAN_H_
#define _SYSENDIAN_H_
/* If we don't have be64enc, the <sys/endian.h> we have isn't usable. */
#if !HAVE_DECL_BE64ENC
#undef HAVE_SYS_ENDIAN_H
#endif
#ifdef HAVE_SYS_ENDIAN_H
#include <sys/endian.h>
#else
#include <stdint.h>
static inline uint32_t
be32dec(const void *pp)
{
const uint8_t *p = (uint8_t const *)pp;
return ((uint32_t)(p[3]) + ((uint32_t)(p[2]) << 8) +
((uint32_t)(p[1]) << 16) + ((uint32_t)(p[0]) << 24));
}
static inline void
be32enc(void *pp, uint32_t x)
{
uint8_t * p = (uint8_t *)pp;
p[3] = x & 0xff;
p[2] = (x >> 8) & 0xff;
p[1] = (x >> 16) & 0xff;
p[0] = (x >> 24) & 0xff;
}
static inline uint64_t
be64dec(const void *pp)
{
const uint8_t *p = (uint8_t const *)pp;
return ((uint64_t)(p[7]) + ((uint64_t)(p[6]) << 8) +
((uint64_t)(p[5]) << 16) + ((uint64_t)(p[4]) << 24) +
((uint64_t)(p[3]) << 32) + ((uint64_t)(p[2]) << 40) +
((uint64_t)(p[1]) << 48) + ((uint64_t)(p[0]) << 56));
}
static inline void
be64enc(void *pp, uint64_t x)
{
uint8_t * p = (uint8_t *)pp;
p[7] = x & 0xff;
p[6] = (x >> 8) & 0xff;
p[5] = (x >> 16) & 0xff;
p[4] = (x >> 24) & 0xff;
p[3] = (x >> 32) & 0xff;
p[2] = (x >> 40) & 0xff;
p[1] = (x >> 48) & 0xff;
p[0] = (x >> 56) & 0xff;
}
static inline uint32_t
le32dec(const void *pp)
{
const uint8_t *p = (uint8_t const *)pp;
return ((uint32_t)(p[0]) + ((uint32_t)(p[1]) << 8) +
((uint32_t)(p[2]) << 16) + ((uint32_t)(p[3]) << 24));
}
static inline void
le32enc(void *pp, uint32_t x)
{
uint8_t * p = (uint8_t *)pp;
p[0] = x & 0xff;
p[1] = (x >> 8) & 0xff;
p[2] = (x >> 16) & 0xff;
p[3] = (x >> 24) & 0xff;
}
static inline uint64_t
le64dec(const void *pp)
{
const uint8_t *p = (uint8_t const *)pp;
return ((uint64_t)(p[0]) + ((uint64_t)(p[1]) << 8) +
((uint64_t)(p[2]) << 16) + ((uint64_t)(p[3]) << 24) +
((uint64_t)(p[4]) << 32) + ((uint64_t)(p[5]) << 40) +
((uint64_t)(p[6]) << 48) + ((uint64_t)(p[7]) << 56));
}
static inline void
le64enc(void *pp, uint64_t x)
{
uint8_t * p = (uint8_t *)pp;
p[0] = x & 0xff;
p[1] = (x >> 8) & 0xff;
p[2] = (x >> 16) & 0xff;
p[3] = (x >> 24) & 0xff;
p[4] = (x >> 32) & 0xff;
p[5] = (x >> 40) & 0xff;
p[6] = (x >> 48) & 0xff;
p[7] = (x >> 56) & 0xff;
}
#endif /* !HAVE_SYS_ENDIAN_H */
#endif /* !_SYSENDIAN_H_ */