ccminer/heavy.cu

#include <string.h>
#include <openssl/sha.h>
#include <cuda.h>
#include "cuda_runtime.h"
#include "device_launch_parameters.h"
#include <map>

#ifndef _WIN32
#include <unistd.h>
#endif

// include thrust
#include <thrust/version.h>
#include <thrust/remove.h>
#include <thrust/device_vector.h>
#include <thrust/iterator/constant_iterator.h>

#include "miner.h"

#include "hefty1.h"
#include "sph_keccak.h"
#include "sph_blake.h"
#include "sph_groestl.h"

#include "cuda_hefty1.h"
#include "cuda_sha256.h"
#include "cuda_keccak512.h"
#include "cuda_groestl512.h"
#include "cuda_blake512.h"
#include "cuda_combine.h"

extern uint32_t *d_hash2output[8];
extern uint32_t *d_hash3output[8];
extern uint32_t *d_hash4output[8];
extern uint32_t *d_hash5output[8];

#define HEAVYCOIN_BLKHDR_SZ        84

// nonce-array für die threads
uint32_t *d_nonceVector[8];

/* Combines top 64-bits from each hash into a single hash */
static void combine_hashes(uint32_t *out, const uint32_t *hash1, const uint32_t *hash2, const uint32_t *hash3, const uint32_t *hash4)
{
    const uint32_t *hash[4] = { hash1, hash2, hash3, hash4 };
    int bits;
    unsigned int i;
    uint32_t mask;
    unsigned int k;

    /* Transpose first 64 bits of each hash into out */
    memset(out, 0, 32);
    bits = 0;
    for (i = 7; i >= 6; i--) {
        for (mask = 0x80000000; mask; mask >>= 1) {
            for (k = 0; k < 4; k++) {
                out[(255 - bits)/32] <<= 1;
                if ((hash[k][i] & mask) != 0)
                    out[(255 - bits)/32] |= 1;
                bits++;
            }
        }
    }
}

#ifdef _MSC_VER
#include <intrin.h>
static uint32_t __inline bitsset( uint32_t x )
{
   DWORD r = 0;
   _BitScanReverse(&r, x);
   return r;
}
#else
static uint32_t bitsset( uint32_t x )
{
    return 31-__builtin_clz(x);
}
#endif

// Finde das high bit in einem Multiword-Integer.
static int findhighbit(const uint32_t *ptarget, int words)
{
    int i;
    int highbit = 0;
    for (i=words-1; i >= 0; --i)
    {
        if (ptarget[i] != 0) {
            highbit = i*32 + bitsset(ptarget[i])+1;
                break;
        }
    }
    return highbit;
}

// Generiere ein Multiword-Integer das die Zahl
// (2 << highbit) - 1 repräsentiert.
static void genmask(uint32_t *ptarget, int words, int highbit)
{
    int i;
    for (i=words-1; i >= 0; --i)
    {
        if ((i+1)*32 <= highbit)
            ptarget[i] = 0xffffffff;
        else if (i*32 > highbit)
            ptarget[i] = 0x00000000;
        else
            ptarget[i] = (1 << (highbit-i*32)) - 1;
    }
}

struct check_nonce_for_remove
{    
    check_nonce_for_remove(uint64_t target, uint32_t *hashes, uint32_t hashlen, uint32_t startNonce) :
        m_target(target),
        m_hashes(hashes),
        m_hashlen(hashlen),
        m_startNonce(startNonce) { }

    __device__
    bool operator()(const uint32_t x)
    {
        // Position im Hash Buffer
        uint32_t hashIndex = x - m_startNonce;
        // Wert des Hashes (als uint64_t) auslesen.
        // Steht im 6. und 7. Wort des Hashes (jeder dieser Hashes hat 512 Bits)
        uint64_t hashValue = *((uint64_t*)(&m_hashes[m_hashlen*hashIndex + 6]));
        // gegen das Target prüfen. Es dürfen nur Bits aus dem Target gesetzt sein.
        return (hashValue & m_target) != hashValue;
    }

    uint64_t  m_target;
    uint32_t *m_hashes;
    uint32_t  m_hashlen;
    uint32_t  m_startNonce;
};

// Zahl der CUDA Devices im System bestimmen
extern "C" int cuda_num_devices()
{
    int version;
    cudaError_t err = cudaDriverGetVersion(&version);
    if (err != cudaSuccess)
    {
        applog(LOG_ERR, "Unable to query CUDA driver version! Is an nVidia driver installed?");
        exit(1);
    }

    int maj = version / 1000, min = version % 100; // same as in deviceQuery sample
    if (maj < 5 || (maj == 5 && min < 5))
    {
        applog(LOG_ERR, "Driver does not support CUDA %d.%d API! Update your nVidia driver!", 5, 5);
        exit(1);
    }

    int GPU_N;
    err = cudaGetDeviceCount(&GPU_N);
    if (err != cudaSuccess)
    {
        applog(LOG_ERR, "Unable to query number of CUDA devices! Is an nVidia driver installed?");
        exit(1);
    }
    return GPU_N;
}

static bool substringsearch(const char *haystack, const char *needle, int &match)
{
    int hlen = strlen(haystack);
    int nlen = strlen(needle);
    for (int i=0; i < hlen; ++i)
    {
        if (haystack[i] == ' ') continue;
        int j=0, x = 0;
        while(j < nlen)
        {
            if (haystack[i+x] == ' ') {++x; continue;}
            if (needle[j] == ' ') {++j; continue;}
            if (needle[j] == '#') return ++match == needle[j+1]-'0';
            if (tolower(haystack[i+x]) != tolower(needle[j])) break;
            ++j; ++x;
        }
        if (j == nlen) return true;
    }
    return false;
}

// CUDA Gerät nach Namen finden (gibt Geräte-Index zurück oder -1)
extern "C" int cuda_finddevice(char *name)
{
    int num = cuda_num_devices();
    int match = 0;
    for (int i=0; i < num; ++i)
    {
        cudaDeviceProp props;
        if (cudaGetDeviceProperties(&props, i) == cudaSuccess)
            if (substringsearch(props.name, name, match)) return i;
    }
    return -1;
}

// Zeitsynchronisations-Routine von cudaminer mit CPU sleep
typedef struct { double value[8]; } tsumarray;
cudaError_t MyStreamSynchronize(cudaStream_t stream, int situation, int thr_id)
{
    cudaError_t result = cudaSuccess;
    if (situation >= 0)
    {   
        static std::map<int, tsumarray> tsum;

        double a = 0.95, b = 0.05;
        if (tsum.find(situation) == tsum.end()) { a = 0.5; b = 0.5; } // faster initial convergence

        double tsync = 0.0;
        double tsleep = 0.95 * tsum[situation].value[thr_id];
        if (cudaStreamQuery(stream) == cudaErrorNotReady)
        {
            usleep((useconds_t)(1e6*tsleep));
            struct timeval tv_start, tv_end;
            gettimeofday(&tv_start, NULL);
            result = cudaStreamSynchronize(stream);
            gettimeofday(&tv_end, NULL);
            tsync = 1e-6 * (tv_end.tv_usec-tv_start.tv_usec) + (tv_end.tv_sec-tv_start.tv_sec);
        }
        if (tsync >= 0) tsum[situation].value[thr_id] = a * tsum[situation].value[thr_id] + b * (tsleep+tsync);
    }
    else
        result = cudaStreamSynchronize(stream);
    return result;
}

int scanhash_heavy_cpp(int thr_id, uint32_t *pdata,
 const uint32_t *ptarget, uint32_t max_nonce,
 unsigned long *hashes_done, uint32_t maxvote);

extern "C"
int scanhash_heavy(int thr_id, uint32_t *pdata,
 const uint32_t *ptarget, uint32_t max_nonce,
 unsigned long *hashes_done, uint32_t maxvote)
{
 return scanhash_heavy_cpp(thr_id, pdata,
  ptarget, max_nonce, hashes_done, maxvote);
}

int scanhash_heavy_cpp(int thr_id, uint32_t *pdata,
 const uint32_t *ptarget, uint32_t max_nonce,
 unsigned long *hashes_done, uint32_t maxvote)
{
    // CUDA will process thousands of threads.
    const int throughput = 4096 * 128;

    int rc = 0;
    uint32_t *hash = NULL;
    cudaMallocHost(&hash, throughput*8*sizeof(uint32_t));
    uint32_t *cpu_nonceVector = NULL;
    cudaMallocHost(&cpu_nonceVector, throughput*sizeof(uint32_t));

    int nrmCalls[6];
    memset(nrmCalls, 0, sizeof(int) * 6);

    uint32_t start_nonce = pdata[19];    
    uint16_t *ext = (uint16_t *)&pdata[20];

    // für jeden Hash ein individuelles Target erstellen basierend
    // auf dem höchsten Bit, das in ptarget gesetzt ist.
    int highbit = findhighbit(ptarget, 8);
    uint32_t target2[2], target3[2], target4[2], target5[2];
    genmask(target2, 2, highbit/4+(((highbit%4)>3)?1:0) ); // SHA256
    genmask(target3, 2, highbit/4+(((highbit%4)>2)?1:0) ); // keccak512
    genmask(target4, 2, highbit/4+(((highbit%4)>1)?1:0) ); // groestl512
    genmask(target5, 2, highbit/4+(((highbit%4)>0)?1:0) ); // blake512

    static bool init[8] = {0,0,0,0,0,0,0,0};
    if (!init[thr_id])
    {
        hefty_cpu_init(thr_id, throughput);
        sha256_cpu_init(thr_id, throughput);
        keccak512_cpu_init(thr_id, throughput);
        groestl512_cpu_init(thr_id, throughput);
        blake512_cpu_init(thr_id, throughput);
        combine_cpu_init(thr_id, throughput);
        init[thr_id] = true;
        cudaMalloc(&d_nonceVector[thr_id], sizeof(uint32_t) * throughput);
    }


    if (opt_vote > maxvote) {
        printf("Warning: Your block reward vote (%hu) exceeds "
                "the maxvote reported by the pool (%hu).\n",
                opt_vote, maxvote);
    }

    if (opt_trust_pool && opt_vote > maxvote) {
        printf("Warning: Capping block reward vote to maxvote reported by pool.\n");
        ext[0] = maxvote;
    }
    else
        ext[0] = opt_vote;

    // Setze die Blockdaten
    hefty_cpu_setBlock(thr_id, throughput, pdata);
    sha256_cpu_setBlock(pdata);
    keccak512_cpu_setBlock(pdata);
    groestl512_cpu_setBlock(pdata);
    blake512_cpu_setBlock(pdata);

    do {
        int i;

        ////// Compaction init
        thrust::device_ptr<uint32_t> devNoncePtr(d_nonceVector[thr_id]);
        thrust::device_ptr<uint32_t> devNoncePtrEnd((d_nonceVector[thr_id]) + throughput);
        uint32_t actualNumberOfValuesInNonceVectorGPU = throughput;

        hefty_cpu_hash(thr_id, throughput, pdata[19]);
        //cudaThreadSynchronize();
        sha256_cpu_hash(thr_id, throughput, pdata[19]);
        //cudaThreadSynchronize();

        // Hier ist die längste CPU Wartephase. Deshalb ein strategisches MyStreamSynchronize() hier.
        MyStreamSynchronize(NULL, 0, thr_id);

        ////// Compaction
        devNoncePtrEnd = thrust::remove_if(devNoncePtr, devNoncePtrEnd, check_nonce_for_remove(*((uint64_t*)target2), d_hash2output[thr_id], 8, pdata[19]));
        actualNumberOfValuesInNonceVectorGPU = (uint32_t)(devNoncePtrEnd - devNoncePtr);
        if(actualNumberOfValuesInNonceVectorGPU == 0)
            goto emptyNonceVector;
        
        keccak512_cpu_hash(thr_id, actualNumberOfValuesInNonceVectorGPU, pdata[19]);
        //cudaThreadSynchronize();

        ////// Compaction
        devNoncePtrEnd = thrust::remove_if(devNoncePtr, devNoncePtrEnd, check_nonce_for_remove(*((uint64_t*)target3), d_hash3output[thr_id], 16, pdata[19]));
        actualNumberOfValuesInNonceVectorGPU = (uint32_t)(devNoncePtrEnd - devNoncePtr);
        if(actualNumberOfValuesInNonceVectorGPU == 0)
            goto emptyNonceVector;

        blake512_cpu_hash(thr_id, actualNumberOfValuesInNonceVectorGPU, pdata[19]);
        //cudaThreadSynchronize();

        ////// Compaction
        devNoncePtrEnd = thrust::remove_if(devNoncePtr, devNoncePtrEnd, check_nonce_for_remove(*((uint64_t*)target5), d_hash5output[thr_id], 16, pdata[19]));
        actualNumberOfValuesInNonceVectorGPU = (uint32_t)(devNoncePtrEnd - devNoncePtr);
        if(actualNumberOfValuesInNonceVectorGPU == 0)
            goto emptyNonceVector;

        groestl512_cpu_hash(thr_id, actualNumberOfValuesInNonceVectorGPU, pdata[19]);
        //cudaThreadSynchronize();

        ////// Compaction
        devNoncePtrEnd = thrust::remove_if(devNoncePtr, devNoncePtrEnd, check_nonce_for_remove(*((uint64_t*)target4), d_hash4output[thr_id], 16, pdata[19]));
        actualNumberOfValuesInNonceVectorGPU = (uint32_t)(devNoncePtrEnd - devNoncePtr);
        if(actualNumberOfValuesInNonceVectorGPU == 0)
            goto emptyNonceVector;
        
        // combine
        combine_cpu_hash(thr_id, actualNumberOfValuesInNonceVectorGPU, pdata[19], hash);

        // Ergebnisse kopieren
        if(actualNumberOfValuesInNonceVectorGPU > 0)
        {
            cudaMemcpy(cpu_nonceVector, d_nonceVector[thr_id], sizeof(uint32_t) * actualNumberOfValuesInNonceVectorGPU, cudaMemcpyDeviceToHost);
        
            for (i=0; i<actualNumberOfValuesInNonceVectorGPU;++i)
            {
                uint32_t nonce = cpu_nonceVector[i];
                //uint32_t index = nonce - pdata[19];
                uint32_t index = i;
                uint32_t *foundhash = &hash[8*index];
                if (foundhash[7] <= ptarget[7]) {
                    if (fulltest(foundhash, ptarget)) {
                        uint32_t verification[8];
                        pdata[19] += nonce - pdata[19];
                        heavycoin_hash((unsigned char *)verification, (const unsigned char *)pdata, HEAVYCOIN_BLKHDR_SZ);
                        if (memcmp(verification, foundhash, 8*sizeof(uint32_t))) {
                            applog(LOG_ERR, "hash for nonce=$%08X does not validate on CPU!\n", nonce);
                        }
                        else
                        {
                            *hashes_done = pdata[19] - start_nonce;
                            rc = 1;
                            goto exit;
                        }
                    }
                }
            }
        }

emptyNonceVector:

        pdata[19] += throughput;

    } while (pdata[19] < max_nonce && !work_restart[thr_id].restart);
    *hashes_done = pdata[19] - start_nonce;

exit:
    cudaFreeHost(cpu_nonceVector);
    cudaFreeHost(hash);
    return rc;
}

void heavycoin_hash(unsigned char* output, const unsigned char* input, int len)
{
    unsigned char hash1[32];
    unsigned char hash2[32];
    uint32_t hash3[16];
    uint32_t hash4[16];
    uint32_t hash5[16];
    uint32_t *final;
    SHA256_CTX ctx;
    sph_keccak512_context keccakCtx;
    sph_groestl512_context groestlCtx;
    sph_blake512_context blakeCtx;

    HEFTY1(input, len, hash1);

    /* HEFTY1 is new, so take an extra security measure to eliminate
     * the possiblity of collisions:
     *
     *     Hash(x) = SHA256(x + HEFTY1(x))
     *
     * N.B. '+' is concatenation.
     */
    SHA256_Init(&ctx);
    SHA256_Update(&ctx, input, len);
    SHA256_Update(&ctx, hash1, sizeof(hash1));
    SHA256_Final(hash2, &ctx);

    /* Additional security: Do not rely on a single cryptographic hash
     * function.  Instead, combine the outputs of 4 of the most secure
     * cryptographic hash functions-- SHA256, KECCAK512, GROESTL512
     * and BLAKE512.
     */

    sph_keccak512_init(&keccakCtx);
    sph_keccak512(&keccakCtx, input, len);
    sph_keccak512(&keccakCtx, hash1, sizeof(hash1));
    sph_keccak512_close(&keccakCtx, (void *)&hash3);

    sph_groestl512_init(&groestlCtx);
    sph_groestl512(&groestlCtx, input, len);
    sph_groestl512(&groestlCtx, hash1, sizeof(hash1));
    sph_groestl512_close(&groestlCtx, (void *)&hash4);

    sph_blake512_init(&blakeCtx);
    sph_blake512(&blakeCtx, input, len);
    sph_blake512(&blakeCtx, (unsigned char *)&hash1, sizeof(hash1));
    sph_blake512_close(&blakeCtx, (void *)&hash5);

    final = (uint32_t *)output;
    combine_hashes(final, (uint32_t *)hash2, hash3, hash4, hash5);
}