monocypher.c

// Monocypher version 3.1.2
//
// This file is dual-licensed.  Choose whichever licence you want from
// the two licences listed below.
//
// The first licence is a regular 2-clause BSD licence.  The second licence
// is the CC-0 from Creative Commons. It is intended to release Monocypher
// to the public domain.  The BSD licence serves as a fallback option.
//
// SPDX-License-Identifier: BSD-2-Clause OR CC0-1.0
//
// ------------------------------------------------------------------------
//
// Copyright (c) 2017-2020, Loup Vaillant
// 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 COPYRIGHT HOLDERS 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 COPYRIGHT
// HOLDER 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.
//
// ------------------------------------------------------------------------
//
// Written in 2017-2020 by Loup Vaillant
//
// To the extent possible under law, the author(s) have dedicated all copyright
// and related neighboring rights to this software to the public domain
// worldwide.  This software is distributed without any warranty.
//
// You should have received a copy of the CC0 Public Domain Dedication along
// with this software.  If not, see
// <https://creativecommons.org/publicdomain/zero/1.0/>

#include "monocypher.h"

/////////////////
/// Utilities ///
/////////////////
#define FOR_T(type, i, start, end) for (type i = (start); i < (end); i++)
#define FOR(i, start, end)         FOR_T(size_t, i, start, end)
#define COPY(dst, src, size)       FOR(i, 0, size) (dst)[i] = (src)[i]
#define ZERO(buf, size)            FOR(i, 0, size) (buf)[i] = 0
#define WIPE_CTX(ctx)              crypto_wipe(ctx   , sizeof(*(ctx)))
#define WIPE_BUFFER(buffer)        crypto_wipe(buffer, sizeof(buffer))
#define MIN(a, b)                  ((a) <= (b) ? (a) : (b))
#define MAX(a, b)                  ((a) >= (b) ? (a) : (b))

typedef int8_t   i8;
typedef uint8_t  u8;
typedef int16_t  i16;
typedef uint32_t u32;
typedef int32_t  i32;
typedef int64_t  i64;
typedef uint64_t u64;

static const u8 zero[128] = {0};

// returns the smallest positive integer y such that
// (x + y) % pow_2  == 0
// Basically, it's how many bytes we need to add to "align" x.
// Only works when pow_2 is a power of 2.
// Note: we use ~x+1 instead of -x to avoid compiler warnings
static size_t align(size_t x, size_t pow_2)
{
    return (~x + 1) & (pow_2 - 1);
}

static u32 load24_le(const u8 s[3])
{
    return (u32)s[0]
        | ((u32)s[1] <<  8)
        | ((u32)s[2] << 16);
}

static u32 load32_le(const u8 s[4])
{
    return (u32)s[0]
        | ((u32)s[1] <<  8)
        | ((u32)s[2] << 16)
        | ((u32)s[3] << 24);
}

static u64 load64_le(const u8 s[8])
{
    return load32_le(s) | ((u64)load32_le(s+4) << 32);
}

static void store32_le(u8 out[4], u32 in)
{
    out[0] =  in        & 0xff;
    out[1] = (in >>  8) & 0xff;
    out[2] = (in >> 16) & 0xff;
    out[3] = (in >> 24) & 0xff;
}

static void store64_le(u8 out[8], u64 in)
{
    store32_le(out    , (u32)in );
    store32_le(out + 4, in >> 32);
}

static void load32_le_buf (u32 *dst, const u8 *src, size_t size) {
    FOR(i, 0, size) { dst[i] = load32_le(src + i*4); }
}
static void load64_le_buf (u64 *dst, const u8 *src, size_t size) {
    FOR(i, 0, size) { dst[i] = load64_le(src + i*8); }
}
static void store32_le_buf(u8 *dst, const u32 *src, size_t size) {
    FOR(i, 0, size) { store32_le(dst + i*4, src[i]); }
}
static void store64_le_buf(u8 *dst, const u64 *src, size_t size) {
    FOR(i, 0, size) { store64_le(dst + i*8, src[i]); }
}

static u64 rotr64(u64 x, u64 n) { return (x >> n) ^ (x << (64 - n)); }
static u32 rotl32(u32 x, u32 n) { return (x << n) ^ (x >> (32 - n)); }

static int neq0(u64 diff)
{   // constant time comparison to zero
    // return diff != 0 ? -1 : 0
    u64 half = (diff >> 32) | ((u32)diff);
    return (1 & ((half - 1) >> 32)) - 1;
}

static u64 x16(const u8 a[16], const u8 b[16])
{
    return (load64_le(a + 0) ^ load64_le(b + 0))
        |  (load64_le(a + 8) ^ load64_le(b + 8));
}
static u64 x32(const u8 a[32],const u8 b[32]){return x16(a,b)| x16(a+16, b+16);}
static u64 x64(const u8 a[64],const u8 b[64]){return x32(a,b)| x32(a+32, b+32);}
int crypto_verify16(const u8 a[16], const u8 b[16]){ return neq0(x16(a, b)); }
int crypto_verify32(const u8 a[32], const u8 b[32]){ return neq0(x32(a, b)); }
int crypto_verify64(const u8 a[64], const u8 b[64]){ return neq0(x64(a, b)); }

void crypto_wipe(void *secret, size_t size)
{
    volatile u8 *v_secret = (u8*)secret;
    ZERO(v_secret, size);
}

/////////////////
/// Chacha 20 ///
/////////////////
#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 void chacha20_rounds(u32 out[16], const u32 in[16])
{
    // The temporary variables make Chacha20 10% faster.
    u32 t0  = in[ 0];  u32 t1  = in[ 1];  u32 t2  = in[ 2];  u32 t3  = in[ 3];
    u32 t4  = in[ 4];  u32 t5  = in[ 5];  u32 t6  = in[ 6];  u32 t7  = in[ 7];
    u32 t8  = in[ 8];  u32 t9  = in[ 9];  u32 t10 = in[10];  u32 t11 = in[11];
    u32 t12 = in[12];  u32 t13 = in[13];  u32 t14 = in[14];  u32 t15 = in[15];

    FOR (i, 0, 10) { // 20 rounds, 2 rounds per loop.
        QUARTERROUND(t0, t4, t8 , t12); // column 0
        QUARTERROUND(t1, t5, t9 , t13); // column 1
        QUARTERROUND(t2, t6, t10, t14); // column 2
        QUARTERROUND(t3, t7, t11, t15); // column 3
        QUARTERROUND(t0, t5, t10, t15); // diagonal 0
        QUARTERROUND(t1, t6, t11, t12); // diagonal 1
        QUARTERROUND(t2, t7, t8 , t13); // diagonal 2
        QUARTERROUND(t3, t4, t9 , t14); // diagonal 3
    }
    out[ 0] = t0;   out[ 1] = t1;   out[ 2] = t2;   out[ 3] = t3;
    out[ 4] = t4;   out[ 5] = t5;   out[ 6] = t6;   out[ 7] = t7;
    out[ 8] = t8;   out[ 9] = t9;   out[10] = t10;  out[11] = t11;
    out[12] = t12;  out[13] = t13;  out[14] = t14;  out[15] = t15;
}

static void chacha20_init_key(u32 block[16], const u8 key[32])
{
    load32_le_buf(block  , (const u8*)"expand 32-byte k", 4); // constant
    load32_le_buf(block+4, key                          , 8); // key
}