sha256比特币原版c++代码

2024-01-04

sha256是一个计算hash 的过程,比特币的算法核心就是计算sha256。

sha256.h

// Copyright (c) 2014-2022 The Bitcoin Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.

#ifndef BITCOIN_CRYPTO_SHA256_H
#define BITCOIN_CRYPTO_SHA256_H

#include <cstdlib>
#include <stdint.h>
#include <string>

/** A hasher class for SHA-256. */
class CSHA256
{
private:
    uint32_t s[8];
    unsigned char buf[64];
    uint64_t bytes{0};

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();
};

namespace sha256_implementation {
enum UseImplementation : uint8_t {
    STANDARD = 0,
    USE_SSE4 = 1 << 0,
    USE_AVX2 = 1 << 1,
    USE_SHANI = 1 << 2,
    USE_SSE4_AND_AVX2 = USE_SSE4 | USE_AVX2,
    USE_SSE4_AND_SHANI = USE_SSE4 | USE_SHANI,
    USE_ALL = USE_SSE4 | USE_AVX2 | USE_SHANI,
};
}

/** Autodetect the best available SHA256 implementation.
 *  Returns the name of the implementation.
 */
std::string SHA256AutoDetect(sha256_implementation::UseImplementation use_implementation = sha256_implementation::USE_ALL);

/** 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 // BITCOIN_CRYPTO_SHA256_H

sha256.cpp

// Copyright (c) 2014-2022 The Bitcoin Core developers
// Distributed under the MIT software license, see the accompanying
// file COPYING or http://www.opensource.org/licenses/mit-license.php.

#include "sha256.h"
//#include <crypto/common.h>
#include <assert.h>
#include <string.h>

#include "common.h"
//#include "cpuid.h"

#if defined(__linux__) && defined(ENABLE_ARM_SHANI) && !defined(BUILD_BITCOIN_INTERNAL)
#include <sys/auxv.h>
#include <asm/hwcap.h>
#endif

#if defined(MAC_OSX) && defined(ENABLE_ARM_SHANI) && !defined(BUILD_BITCOIN_INTERNAL)
#include <sys/types.h>
#include <sys/sysctl.h>
#endif

#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_x86_shani
{
void Transform_2way(unsigned char* out, const unsigned char* in);
}

namespace sha256_x86_shani
{
void Transform(uint32_t* s, const unsigned char* chunk, size_t blocks);
}

namespace sha256_arm_shani
{
void Transform(uint32_t* s, const unsigned char* chunk, size_t blocks);
}

namespace sha256d64_arm_shani
{
void Transform_2way(unsigned char* out, const unsigned char* in);
}

// 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(sha256_implementation::UseImplementation use_implementation)
{
    std::string ret = "standard";
    Transform = sha256::Transform;
    TransformD64 = sha256::TransformD64;
    TransformD64_2way = nullptr;
    TransformD64_4way = nullptr;
    TransformD64_8way = nullptr;

#if defined(USE_ASM) && defined(HAVE_GETCPUID)
    bool have_sse4 = false;
    bool have_xsave = false;
    bool have_avx = false;
    [[maybe_unused]] bool have_avx2 = false;
    [[maybe_unused]] bool have_x86_shani = false;
    [[maybe_unused]] bool enabled_avx = false;

    uint32_t eax, ebx, ecx, edx;
    GetCPUID(1, 0, eax, ebx, ecx, edx);
    if (use_implementation & sha256_implementation::USE_SSE4) {
        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);
        if (use_implementation & sha256_implementation::USE_AVX2) {
            have_avx2 = (ebx >> 5) & 1;
        }
        if (use_implementation & sha256_implementation::USE_SHANI) {
            have_x86_shani = (ebx >> 29) & 1;
        }
    }

#if defined(ENABLE_X86_SHANI) && !defined(BUILD_BITCOIN_INTERNAL)
    if (have_x86_shani) {
        Transform = sha256_x86_shani::Transform;
        TransformD64 = TransformD64Wrapper<sha256_x86_shani::Transform>;
        TransformD64_2way = sha256d64_x86_shani::Transform_2way;
        ret = "x86_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_BITCOIN_INTERNAL)
        TransformD64_4way = sha256d64_sse41::Transform_4way;
        ret += ",sse41(4way)";
#endif
    }

#if defined(ENABLE_AVX2) && !defined(BUILD_BITCOIN_INTERNAL)
    if (have_avx2 && have_avx && enabled_avx) {
        TransformD64_8way = sha256d64_avx2::Transform_8way;
        ret += ",avx2(8way)";
    }
#endif
#endif // defined(USE_ASM) && defined(HAVE_GETCPUID)

#if defined(ENABLE_ARM_SHANI) && !defined(BUILD_BITCOIN_INTERNAL)
    bool have_arm_shani = false;
    if (use_implementation & sha256_implementation::USE_SHANI) {
#if defined(__linux__)
#if defined(__arm__) // 32-bit
        if (getauxval(AT_HWCAP2) & HWCAP2_SHA2) {
            have_arm_shani = true;
        }
#endif
#if defined(__aarch64__) // 64-bit
        if (getauxval(AT_HWCAP) & HWCAP_SHA2) {
            have_arm_shani = true;
        }
#endif
#endif

#if defined(MAC_OSX)
        int val = 0;
        size_t len = sizeof(val);
        if (sysctlbyname("hw.optional.arm.FEAT_SHA256", &val, &len, nullptr, 0) == 0) {
            have_arm_shani = val != 0;
        }
#endif
    }

    if (have_arm_shani) {
        Transform = sha256_arm_shani::Transform;
        TransformD64 = TransformD64Wrapper<sha256_arm_shani::Transform>;
        TransformD64_2way = sha256d64_arm_shani::Transform_2way;
        ret = "arm_shani(1way,2way)";
    }
#endif

    assert(SelfTest());
    return ret;
}

// SHA-256

CSHA256::CSHA256()
{
    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;
    }
}

common.h

#include <endian.h>




uint32_t static inline ReadBE32(const unsigned char* ptr)
{
    uint32_t x;
    memcpy(&x, ptr, 4);
    return be32toh(x);
}

uint64_t static inline ReadBE64(const unsigned char* ptr)
{
    uint64_t x;
    memcpy(&x, ptr, 8);
    return be64toh(x);
}

void static inline WriteBE32(unsigned char* ptr, uint32_t x)
{
    uint32_t v = htobe32(x);
    memcpy(ptr, &v, 4);
}

void static inline WriteBE64(unsigned char* ptr, uint64_t x)
{
    uint64_t v = htobe64(x);
    memcpy(ptr, &v, 8);
}

main.cpp

#include "sha256.h"
//#include "uint256.h"


#include <iostream>
#include <string>
#include <vector>
#include <array>

using namespace std;


void printArray(array<uint8_t, 32> &result)
{
    for(int i=0; i<result.size();i++)
    {
        printf("%02x", result[i]);
    }
   
}



int main(int argc,  char** argv)
{
    string str = "123";
    if(argc > 1){
        str = argv[1];
    }


    cout<<"str:"<< str <<endl;

    //vector<uint8_t> result;
    array<uint8_t, 32> result{};
    //result.resize(32);


    CSHA256().Write((const unsigned char*) str.c_str(), str.size()).Finalize(result.data());
    printArray(result);
    printf("\n");

    int N = 1000000000;
    for(int i=0;i<N;i++)
    {
        string s  = str+ to_string(i);
        array<uint8_t, 32> result{};
        CSHA256().Write((const unsigned char*) s.c_str(), s.size()).Finalize(result.data());
        if(result[0] == 0x00 && result[1] == 0x00 && result[2] == 0x00)
        {
            cout<<"s: "<<s <<"," <<i << endl;
            printArray(result);
            printf("\n");
            
            break;
        }
    }







    cout<<endl;

    return 0;
}

build

g++ -O3 main.cpp sha256.cpp   -o main

run

./main  

模拟比特币挖矿时的hash计算过程。

本文内容由网友自发贡献,版权归原作者所有,本站不承担相应法律责任。如您发现有涉嫌抄袭侵权的内容,请联系:hwhale#tublm.com(使用前将#替换为@)

sha256比特币原版c++代码 的相关文章

  • 具有不同大小结构的结构数组的 malloc()

    如果每个结构都包含一个大小不同的字符串数组 那么如何正确地 malloc 一个结构数组 因此每个结构可能有不同的大小 并且不可能 realloc 结构体数量 sizeof 结构体名称 after malloc 初始大小 sizeof 结构名
  • 分段错误(核心转储)错误

    我的程序编译罚款 但在输入文件时出现 分段错误 核心转储 错误 我没有正确处理 ostream 吗 include
  • 内联函数/方法

    声明 内联函数必须在调用之前定义 这个说法正确吗 EDIT 该问题最初是德语 内联功能穆森 弗 伊赫雷姆 奥夫鲁夫定义 sein 也许它对任何人都有帮助 是的 它是正确的 但只是部分正确 它可能正确地重新构建如下 内联函数必须在每个翻译单位
  • 将字节数组转换为托管结构

    更新 这个问题的答案帮助我编写了开源项目GitHub 上的 AlicanC 现代战争 2 工具 https github com AlicanC AlicanC s Modern Warfare 2 Tool 你可以看到我是如何阅读这些数据
  • C 程序从连接到系统的 USB 设备读取数据

    我正在尝试从连接到系统 USB 端口的 USB 设备 例如随身碟 获取数据 在这里 我可以打开设备文件并读取一些随机原始数据 但我想获取像 minicom teraterm 这样的数据 请让我知道我可以使用哪些方法和库来成功完成此操作以及如
  • SSL/TLS/HTTPS 站点在 C#/.NET WebBrowser 控件中非常慢,但在 Internet Explorer 中则很好

    背景 我正在修改自动维基浏览器 http en wikipedia org wiki Wikipedia AutoWikiBrowser使用托管在安全服务器上的 MediaWiki 站点 我允许用户通过 C 应用程序中的 WebBrowse
  • C# 正则表达式用于查找 中具有特定结尾的链接

    我需要一个正则表达式模式来查找字符串 带有 HTML 代码 中的链接 以获取文件结尾如 gif 或 png 的链接 示例字符串 a href site com folder picture png target blank picture
  • C++中delete和delete[]的区别[重复]

    这个问题在这里已经有答案了 可能的重复 C 中的删除与删除 运算符 https stackoverflow com questions 2425728 delete vs delete operators in c 我写了一个包含两个指针的
  • mprotect 之后 malloc 导致分段错误

    在使用 mprotect 保护内存区域后第一次调用 malloc 时 我遇到分段错误 这是执行内存分配和保护的代码片段 define PAGESIZE 4096 void paalloc int size Allocates and ali
  • 如何将带有自定义分配器的 std::vector 传递给需要带有 std::allocator 的函数?

    我正在使用外部库 pcl 因此我需要一个不会更改现有函数原型的解决方案 我正在使用的一个函数生成一个std vector
  • TcpClient 在异步读取期间断开连接

    我有几个关于完成 tcp 连接的问题 客户端使用 Tcp 连接到我的服务器 在接受客户端后listener BeginAcceptTcpClient ConnectionEstabilishedCallback null 我开始阅读netw
  • 如何在 C++ 中将 CString 转换为 double?

    我如何转换CString to a double在 C 中 Unicode 支持也很好 Thanks A CString可以转换为LPCTSTR 这基本上是一个const char const wchar t 在 Unicode 版本中 知
  • 为什么 clang 使用 -O0 生成低效的 asm(对于这个简单的浮点和)?

    我正在 llvm clang Apple LLVM 版本 8 0 0 clang 800 0 42 1 上反汇编此代码 int main float a 0 151234 float b 0 2 float c a b printf f c
  • C++11 动态线程池

    最近 我一直在尝试寻找一个用于线程并发任务的库 理想情况下 是一个在线程上调用函数的简单接口 任何时候都有 n 个线程 有些线程比其他线程完成得更快 并且到达的时间不同 首先我尝试了 Rx 它在 C 中非常棒 我还研究了 Blocks 和
  • .NET 客户端中 Google 表格中的条件格式请求

    我知道如何在 Google Sheets API 中对值和其他格式进行批量电子表格更新请求 但条件格式似乎有所不同 我已正确设置请求 AddConditionalFormatRuleRequest formatRequest new Add
  • 初始化 LPCTSTR /LPCWSTR [重复]

    这个问题在这里已经有答案了 我很难理解并使其正常工作 基本上归结为我无法成功初始化这种类型的变量 它需要有说的内容7 2E25DC9D 0 USB003 有人可以解释 展示这种类型的正确初始化和类似的值吗 我已查看此站点上的所有帮助 将项目
  • 受限 AppDomain 中的代码访问安全异常

    Goal 我需要在权限非常有限的 AppDomain 中运行一些代码 它不应该访问任何花哨或不安全的内容 except对于我在其他地方定义的一些辅助方法 我做了什么 我正在创建一个具有所需基本权限的沙箱 AppDomain 并创建一个运行代
  • 为什么文件更新时“如果较新则复制”不复制文件?

    我在 Visual Studio Express 中有一个解决方案 如下所示 The LogicSchemaC 中的类 将在运行时解析指定的 XML 文件 以下是在main的方法Program cs LogicSchema ls new L
  • 带有私有设置器的 EFCore Base 实体模型属性 - 迁移奇怪的行为

    实体模型继承的类内的私有设置器似乎会导致 EFCore 迁移出现奇怪的问题 考虑以下示例 其中有多个类 Bar and Baz 继承自Foo 跑步时Add Migration多次命令 添加 删除private修饰符 生成的模式在多个方面都是
  • C#中为线程指定特殊的cpu

    我有 2 个线程 我想告诉其中一个在第一个 cpu 上运行 第二个在第二个 cpu 上运行 例如在具有两个 cpu 的机器中 我怎样才能做到这一点 这是我的代码 UCI UCIMain new UCI Thread UCIThread ne

随机推荐

  • 使用pytorch构建图卷积网络预测化学分子性质

    在本文中 我们将通过化学的视角探索图卷积网络 我们将尝试将网络的特征与自然科学中的传统模型进行比较 并思考为什么它的工作效果要比传统的方法好 图和图神经网络 化学或物理中的模型通常是一个连续函数 例如y f x x x x 其中x x x
  • CAD Exchanger SDK 3.24 for Android Crack

    CAD Exchanger SDK Software Libraries to Read Write and Visualize 3D CAD files Quickly and easily enrich your web server
  • 蜜罐溯源以及蜜罐HFish的使用

    一 蜜罐是什么 蜜罐技术本质上是一种对攻击方进行欺骗的技术 通过布置一些作为诱饵的主机 网络服务或者信息 诱使攻击方对它们实施攻击 从而可以对攻击行为进行捕获和分析 了解攻击方所使用的工具与方法 推测攻击意图和动机 能够让防御方清晰地了解他
  • Matlab图像处理系列——图像复原之噪声模型仿真

    微信公众号上线 搜索公众号 小灰灰的FPGA 关注可获取相关源码 定期更新有关FPGA的项目以及开源项目源码 包括但不限于各类检测芯片驱动 低速接口驱动 高速接口驱动 数据信号处理 图像处理以及AXI总线等 本节目录 一 图像复原的模型 二
  • VS Code 自动选择Python3 venv

    我们使用VS Code写Python代码时 往往希望这个项目的依赖和其他项目或者全局的python环境隔离开 VS Code不像PyCharm那样自动完成 但是我们也可以快速的进行设置 首先我们需要把python项目所在的目录添加为VS C
  • CAD Exchanger SDK 3.24 FOR WIN Crack

    Manufacturing Toolkit and Web Toolkit enhancements Unity performance optimization renaming and rotating SDK examples in
  • vscode插件离线安装地址

    因内网开发 编辑器不可联网 插件需要离线安装 vscode插件商店 Extensions for Visual Studio family of products Visual Studio Marketplace
  • SpringCloud+saToken实现登录及权限认证

    SpringCloud saToken实现登录及权限认证 文章目录 SpringCloud saToken实现登录及权限认证 1 为什么要用sa Token 2 sa Token功能 3 springcloud集成sa token 3 1
  • CAD Exchanger SDK 3.24 for Linux Crack

    CAD Exchanger SDK Software Libraries to Read Write and Visualize 3D CAD files Quickly and easily enrich your web server
  • Jlink V9刷入自动升级固件

    Jlink V9刷入自动升级固件 1 所需工具 一个可用的jlink 一个待刷jlink 2 接线如图 3 查看待刷Jlink的主控芯片型号 我的型号为stm32f205rc 4 刷入固件 固件下载地址 https download csd
  • 一批J-link V9变砖拯救

    一批J link V9变砖拯救 weixin 51547258 于 2023 05 05 16 05 09 发布 阅读量282 收藏 点赞数 文章标签 单片机 stm32 嵌入式硬件 版权 手里有一批J link V9版本 由于误操作升级固
  • 指尖互鉴APP 毕业设计源码48084

    赠送源码 毕业设计 SSM指尖互鉴app https www bilibili com video BV15t4y1Z7Gs vd source 72970c26ba7734ebd1a34aa537ef5301 题目 SSM 指尖互鉴APP
  • 14.9-时序和组合的混合逻辑——使用非阻塞赋值

    时序和组合的混合逻辑 使用非阻塞赋值 1 在一个always块中同时实现组合逻辑和时序逻辑 2 将组合和时序逻辑分别写入两个always块中 原则4 在同一个always块中描述时序和组合逻辑混合电路时 用非阻塞赋值 1 在一个always
  • 14.11-对同一变量进行多次赋值

    对同一变量进行多次赋值 1 同一变量多次赋值 即便是非阻塞赋值 也会存在竞争冒险 原则6 严禁在多个always块中对同一个变量赋值 包括阻塞和非阻塞赋值 1 同一变量多次赋值 即便是非阻塞赋值 也会存在竞争冒险 两个always块都对输出
  • zzz666

    6
  • 拓展:vue 父组件调用子组件方法ref(且父组件可通过ref调用的方法传值给子组件)

    1 ref被用来给元素或子组件注册引用信息 引用信息将会注册在父组件的 refs对象上 一 ref被用来给元素或子组件注册引用信息 引用信息将会注册在父组件的 refs对象上 div class formBtn fl 111 div div
  • haiku定义简单的模型并初始化参数

    Haiku 是一个基于 JAX 的深度学习库 旨在提供简洁 灵活且易于使用的 API 以构建和训练神经网络模型 import haiku as hk import jax import jax numpy as jnp 1 定义简单的二层神
  • 学Python到底能干点什么?我为什么不建议去学Python

    学习了Python到底都能做些什么 好学编程今天带大家来看一看 Python 作为编程语言因其简单易学而受到很多开发者及初学者的青睐 那么 学了Python学了以后到底有什么用 它的应用领域有哪些呢 Python 的应用领域非常广泛 目前几
  • 基于SpringBoot的流浪动物救助网站

    文章目录 项目介绍 主要功能截图 部分代码展示 设计总结 项目获取方式 作者主页 超级无敌暴龙战士塔塔开 简介 Java领域优质创作者 简历模板 学习资料 面试题库 关注我 都给你 文末获取源码联系 项目介绍 基于SpringBoot的流浪
  • sha256比特币原版c++代码

    sha256是一个计算hash 的过程 比特币的算法核心就是计算sha256 sha256 h Copyright c 2014 2022 The Bitcoin Core developers Distributed under the