[ANN] ccminer 2.3 - opensource - GPL (tpruvot)

[joblo]

To inside the ' work_decode ' function, so it gets checked on new jobs, to provide compatibility with profit switching.

Simple problem, simple fix but what's the connection to profit swicthing?

[cryptofarmer]

The connection:
When mining SPDR ( coin without roots ) all ok
Then switch to LUX or GEX (coins with roots ) all ok roots are enabled in ccminer.
Then switch back to SPDR ( coin without roots) BIG trouble.
From this moment ccminer is still in enabled roots mode and cant produce valid shares anymore.

Therefore the "has roots" check needs to be performed on every received job and set to false by default.
I think that's the best approach?

[joblo]

Then switch back to SPDR ( coin without roots) BIG trouble.

Thanks, I should have realized that. I'll make the change in cpuminer-opt.

Edit: phi2-lux now makes sense, it's just a workaround for phi2 with roots enabled for miners
that can't handle them dynamically.

[startsts]

Any idea how to compile it on Ubuntu 19.04?  Will the dev update source code for compatibility?

[cryptofarmer]

Then switch back to SPDR ( coin without roots) BIG trouble.

Edit: phi2-lux now makes sense, it's just a workaround for phi2 with roots enabled for miners
that can't handle them dynamically.

Thank you kindly !

[sp_]

This miner is not updated anymore. I did a quick test of the x17 algo on my RTX 2060 SUPER. The latest ccminer 2.3 does around 9MHASH,The popular alexis 1.0.0 fork is doing around 12 MHASH My free opensource suprminer/SPMOD-GIT is doing 19MHASH. The private miner Enemy/T-rex & co is doing around 26MHASH.
But it all doesn't really matter because there is no profit in 26MHASH.

Is X17 FPGA/ASIC minable, or is there a faster Private GPU kernel out there that harvest all of the profit?

[scryptr]

This miner is not updated anymore. I did a quick test of the x17 algo on my RTX 2060 SUPER. The latest ccminer 2.3 does around 9MHASH,The popular alexis 1.0.0 fork is doing around 12 MHASH My free opensource suprminer/SPMOD-GIT is doing 19MHASH. The private miner Enemy/T-rex & co is doing around 26MHASH.
But it all doesn't really matter because there is no profit in 26MHASH.

Is X17 FPGA/ASIC minable, or is there a faster Private GPU kernel out there that harvest all of the profit?

X17 IS FPGA MINABLE--

There are a couple posts, in sequence, on the ZergPool thread.  Here is the link:

  Link to ZergPool thread- https://bitcointalk.org/index.php?topic=2759935.msg52643630#msg52643630

FPGA mining is a real thumbscrew for GPU miners.  A short while ago, $50/day for a single FPGA on a few algos, but now just $1-3/day at best.  And, FPGAs cost 3X-5X as much as a GPU rig of 6 or more cards.

--scryptr

[ralayi7348]

Just checked with 1.7.4 - both x86 and x64 versions
same behaviour, telnet connects, on first jey pressed exits to prompt :S
I tried changing port to 4069. no change

Tried running as admin/normal - same

[JayDDee]

For anyone who's interested ccminer no longer compiles on Ubuntu-21.10 (Cuda 11, GCC 11).

Cuda 11 requires all lines referencing sm_30 or sm_35 be removed from Makefile.am.

GCC 11 requires a fix ported from cpuminer and applied to 2 files: sph/blake2s.c, equi/blake2/blake2bx.cpp
https://github.com/tpruvot/cpuminer-multi/commit/7411020cefb5809202dd0db346df947a4561db4b

[andy2572192561]

http://ccminer.org/img/ccminer-banner.png

Welcome to the discussion thread for my ccminer fork.

It contains most algorithms, and was adapted to be usable on both linux and windows.
My releases are generally tuned with the Geforce GTX 750Ti, which was, and still is a good reference for tests (Linux + Windows) and has less hashrate drop in activity (unlike the 9xx and pascal series).

For a comprehensive list of the algorithms supported as well as details about the various configuration parameters, please check the ReadMe file.

Direct linksWindows Binaries : https://github.com/tpruvot/ccminer/releases
Source Code : https://github.com/tpruvot/ccminer
Linux, read first : https://github.com/tpruvot/ccminer/wiki/Compatibility
Sample Command Line : ccminer -a bitcore -o stratum+tcp://yiimp.eu:3556 -u 1Hpjgz8i4aRdnmLXm9Eri8fFgje2dA9BaQ -p c=BTX


http://cryptomining-blog.com/wp-content/uploads/2017/03/ccminer-20-tpruvot-580x293.jpg

void hashx7(void *state, const void *input, uint64_t timestamp)
{
    uint32_t _ALIGN(64) hash[7][16];
    uint32_t *hash0 = hash[0];
    uint32_t *hash1 = hash[1];
    uint32_t *hash2 = hash[2];
    uint32_t *hash3 = hash[3];
    uint32_t *hash4 = hash[4];
    uint32_t *hash5 = hash[5];
    uint32_t *hash6 = hash[6];
   
    x7_context_overlay ctx;

    // Blake512
    sph_blake512_init(&ctx.blake);
    sph_blake512(&ctx.blake, &timestamp, sizeof(timestamp));
    sph_blake512(&ctx.blake, input, 80);
    sph_blake512_close(&ctx.blake, hash0);   

    // BMW512
    sph_bmw512_init(&ctx.bmw);
    sph_bmw512(&ctx.bmw, hash0, 64);
    sph_bmw512_close(&ctx.bmw, hash1);

    // XOR with hash0
    for (int i = 0; i < 64; ++i) {
        hash1 ^= hash0;
    }

    // groestl   
#if defined(__AES__)
    groestl512_full(&ctx.groestl, (char*)hash2, (const char*)hash1, 512);
#else
    sph_groestl512_init(&ctx.groestl);
    sph_groestl512(&ctx.groestl, hash1, 64);
    sph_groestl512_close(&ctx.groestl, hash2);
#endif

    // Skein512
    sph_skein512_init(&ctx.skein);
    sph_skein512(&ctx.skein, hash2, 64);
    sph_skein512_close(&ctx.skein, hash3);

    // XOR with hash2
    for (int i = 0; i < 64; ++i) {
        hash3 ^= hash2;
    }

    // Keccak512
    sph_keccak512_init(&ctx.keccak);
    sph_keccak512(&ctx.keccak, hash3, 64);
    sph_keccak512_close(&ctx.keccak, hash4);

#if defined(__aarch64__)
    sph_luffa512_init(&ctx.luffa);
    sph_luffa512(&ctx.luffa, hash4, 64);
    sph_luffa512_close(&ctx.luffa, hash5);
#else
    luffa_full(&ctx.luffa, hash5, 512, hash4, 64);
#endif

#if defined(__AES__)
    echo_full(&ctx.echo, (BitSequence *)hash6, 512, (const BitSequence *)hash5, 64);
#else
    sph_echo512_init(&ctx.echo);
    sph_echo512(&ctx.echo, hash5, 64);
    sph_echo512_close(&ctx.echo, hash6);
#endif

    // XOR with hash5
    for (int i = 0; i < 64; ++i) {
        hash6 ^= hash5;
    }

    memcpy(state, hash6, 32);   
}

Hello! How does ccminer handle the algorithm with timestamp in sph_blake512(&ctx.blake, &timestamp, sizeof(timestamp)); algorithm?

[andy2572192561]

http://ccminer.org/img/ccminer-banner.png

Welcome to the discussion thread for my ccminer fork.

It contains most algorithms, and was adapted to be usable on both linux and windows.
My releases are generally tuned with the Geforce GTX 750Ti, which was, and still is a good reference for tests (Linux + Windows) and has less hashrate drop in activity (unlike the 9xx and pascal series).

For a comprehensive list of the algorithms supported as well as details about the various configuration parameters, please check the ReadMe file.

Direct linksWindows Binaries : https://github.com/tpruvot/ccminer/releases
Source Code : https://github.com/tpruvot/ccminer
Linux, read first : https://github.com/tpruvot/ccminer/wiki/Compatibility
Sample Command Line : ccminer -a bitcore -o stratum+tcp://yiimp.eu:3556 -u 1Hpjgz8i4aRdnmLXm9Eri8fFgje2dA9BaQ -p c=BTX


http://cryptomining-blog.com/wp-content/uploads/2017/03/ccminer-20-tpruvot-580x293.jpg

void hashx7(void *state, const void *input, uint64_t timestamp)
{
    uint32_t _ALIGN(64) hash[7][16];
    uint32_t *hash0 = hash[0];
    uint32_t *hash1 = hash[1];
    uint32_t *hash2 = hash[2];
    uint32_t *hash3 = hash[3];
    uint32_t *hash4 = hash[4];
    uint32_t *hash5 = hash[5];
    uint32_t *hash6 = hash[6];
   
    x7_context_overlay ctx;

    // Blake512
    sph_blake512_init(&ctx.blake);
    sph_blake512(&ctx.blake, &timestamp, sizeof(timestamp));
    sph_blake512(&ctx.blake, input, 80);
    sph_blake512_close(&ctx.blake, hash0);   

    // BMW512
    sph_bmw512_init(&ctx.bmw);
    sph_bmw512(&ctx.bmw, hash0, 64);
    sph_bmw512_close(&ctx.bmw, hash1);

    // XOR with hash0
    for (int i = 0; i < 64; ++i) {
        hash1 ^= hash0;
    }

    // groestl   
#if defined(__AES__)
    groestl512_full(&ctx.groestl, (char*)hash2, (const char*)hash1, 512);
#else
    sph_groestl512_init(&ctx.groestl);
    sph_groestl512(&ctx.groestl, hash1, 64);
    sph_groestl512_close(&ctx.groestl, hash2);
#endif

    // Skein512
    sph_skein512_init(&ctx.skein);
    sph_skein512(&ctx.skein, hash2, 64);
    sph_skein512_close(&ctx.skein, hash3);

    // XOR with hash2
    for (int i = 0; i < 64; ++i) {
        hash3 ^= hash2;
    }

    // Keccak512
    sph_keccak512_init(&ctx.keccak);
    sph_keccak512(&ctx.keccak, hash3, 64);
    sph_keccak512_close(&ctx.keccak, hash4);

#if defined(__aarch64__)
    sph_luffa512_init(&ctx.luffa);
    sph_luffa512(&ctx.luffa, hash4, 64);
    sph_luffa512_close(&ctx.luffa, hash5);
#else
    luffa_full(&ctx.luffa, hash5, 512, hash4, 64);
#endif

#if defined(__AES__)
    echo_full(&ctx.echo, (BitSequence *)hash6, 512, (const BitSequence *)hash5, 64);
#else
    sph_echo512_init(&ctx.echo);
    sph_echo512(&ctx.echo, hash5, 64);
    sph_echo512_close(&ctx.echo, hash6);
#endif

    // XOR with hash5
    for (int i = 0; i < 64; ++i) {
        hash6 ^= hash5;
    }

    memcpy(state, hash6, 32);   
}

Hello! How does ccminer handle the algorithm with timestamp in sph_blake512(&ctx.blake, &timestamp, sizeof(timestamp)); algorithm?

All other algorithm problems have been solved, but I don't know how to deal with the timestamp.