ocminer / nheqminer-djezo

Fork of Nicehash's djezo equihash miner optimized for linux

Build instructions:

Dependencies:

• Boost 1.54+

Windows:

Windows builds made by us are available here: https://github.com/nicehash/nheqminer/releases

Download and install:

• AMD APP SDK (if not needed remove USE_OCL_XMP from nheqminer Preprocessor definitions under Properties > C/C++ > Preprocessor)
• CUDA SDK (if not needed remove USE_CUDA_TROMP from nheqminer Preprocessor definitions under Properties > C/C++ > Preprocessor)
• Visual Studio 2013 Community: https://www.visualstudio.com/en-us/news/releasenotes/vs2013-community-vs
• Visual Studio Update 5 installed
• 64 bit version only

Open nheqminer.sln under nheqminer/nheqminer.sln and build. You will have to build ReleaseSSE2 cpu_tromp project first, then Release7.5 cuda_tromp project, then select Release and build all.

Linux

Work in progress.

Working solvers CPU_TROMP, CPU_XENONCAT, CUDA_TROMP, OCL_XMP, OCL_SILENTARMY

Linux (Ubuntu 14.04 / 16.04) Build CPU_XENONCAT:

• Open terminal and run the following commands:
  • sudo apt-get install cmake build-essential libboost-all-dev
  • git clone -b Linux https://github.com/nicehash/nheqminer.git
  • cd nheqminer/cpu_xenoncat/Linux/asm/
  • sh assemble.sh
  • cd ../../../Linux_cmake/nheqminer_cpu_xenoncat
  • cmake .
  • make -j $(nproc)

Linux (Ubuntu 14.04 / 16.04) Build CUDA_TROMP:

• Open terminal and run the following commands:
  • Ubuntu 14.04:
    • wget http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1404/x86_64/cuda-repo-ubuntu1404_8.0.44-1_amd64.deb
    • sudo dpkg -i cuda-repo-ubuntu1404_8.0.44-1_amd64.deb
  • Ubuntu 16.04:
    • wget http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64/cuda-repo-ubuntu1604_8.0.44-1_amd64.deb
    • sudo dpkg -i cuda-repo-ubuntu1604_8.0.44-1_amd64.deb
  • sudo apt-get update
  • sudo apt-get install cuda
  • sudo apt-get install cuda-toolkit-8-0
  • sudo apt-get install cmake build-essential libboost-all-dev
  • git clone -b Linux https://github.com/nicehash/nheqminer.git
  • cd nheqminer/Linux_cmake/nheqminer_cuda_tromp && cmake . && make -j $(nproc)
  • or specify your compute version for example 50 like so cd nheqminer/Linux_cmake/nheqminer_cuda_tromp && cmake COMPUTE=50 . && make

Linux (16.04) Build OCL_XMP, OCL_SILENTARMY:

• Open terminal and run the following commands:
  • AMD APP SDK
  • and make sure you have the AMD drivers installed
  • install them to the default paths
  • sudo apt-get install mesa-common-dev
  • sudo apt-get install cmake build-essential libboost-all-dev
  • git clone -b Linux https://github.com/nicehash/nheqminer.git
  • cd nheqminer/Linux_cmake/nheqminer_AMD && cmake . -DOPENCL_LIBRARY=/usr/lib/x86_64-linux-gnu/libOpenCL.so -DOPENCL_INCLUDE_DIRECTORY=/opt/AMDAPPSDK-3.0/include && make -j $(nproc)
  • cp ../../3rdparty/amd_bins_linux/* -r .
  • cp ../../3rdparty/amd_silentarmy_kernels/* -r .

Run instructions:

Parameters: -h Print this help and quit -l [location] Stratum server:port -u [username] Username (bitcoinaddress) -a [port] Local API port (default: 0 = do not bind) -d [level] Debug print level (0 = print all, 5 = fatal only, default: 2) -b [hashes] Run in benchmark mode (default: 200 iterations)

CPU settings -t [num_thrds] Number of CPU threads -e [ext] Force CPU ext (0 = SSE2, 1 = AVX, 2 = AVX2)

NVIDIA CUDA settings -ci CUDA info -cd [devices] Enable CUDA mining on spec. devices -cb [blocks] Number of blocks -ct [tpb] Number of threads per block Example: -cd 0 2 -cb 12 16 -ct 64 128

If run without parameters, miner will start mining with 75% of available logical CPU cores. Use parameter -h to learn about available parameters:

Example to run benchmark on your CPU:

    nheqminer -b

Example to mine on your CPU with your own BTC address and worker1 on NiceHash USA server:

    nheqminer -l equihash.usa.nicehash.com:3357 -u YOUR_BTC_ADDRESS_HERE.worker1

Example to mine on your CPU with your own BTC address and worker1 on EU server, using 6 threads:

    nheqminer -l equihash.eu.nicehash.com:3357 -u YOUR_BTC_ADDRESS_HERE.worker1 -t 6

Note: if you have a 4-core CPU with hyper threading enabled (total 8 threads) it is best to run with only 6 threads (experimental benchmarks shows that best results are achieved with 75% threads utilized)

Example to mine on your CPU as well on your CUDA GPUs with your own BTC address and worker1 on EU server, using 6 CPU threads and 2 CUDA GPUs:

    nheqminer -l equihash.eu.nicehash.com:3357 -u YOUR_BTC_ADDRESS_HERE.worker1 -t 6 -cd 0 1

nheqminer-djezo's Issues

Building on windows (NV)

k so made it work by

Changing all references in project files from $(AMDSDK) to $(CUDA_PATH)

ie \include\

and path for OpenCL.lib from \x86_64\ to \x64

removed unused Linux reference in cl_ext.hpp (CL_DEVICE_TOPOLOGY_AMD)

fixed linux ref/typo in Zcashstratum.cpp ( sleep(1) -> Sleep(1) )

think I have dun it rite
https://github.com/blaklite/nheqminer-djezo

nheqminer problem when compile in Ubuntu 17.10 + Cuda 9.0/9.1

Environmeent: Ubuntu 17.10 64bit + Cuda 9.0/9.1
nheqminer code version: ba3edc3
Error message:

Building NVCC (Device) object CMakeFiles/nheqminer_cuda_tromp.dir/__/__/cuda_tromp/nheqminer_cuda_tromp_generated_equi_miner.cu.o
nvcc fatal   : Unsupported gpu architecture 'compute_20'
CMake Error at nheqminer_cuda_tromp_generated_equi_miner.cu.o.cmake:222 (message)

Mac possibly?

Any thoughts an getting this working an a mac w/ CUDA

Made sure it's using Apple Command Line tools 8.2 to resolve compiler issues.

Initial attempts

cmake . && make -j $(sysctl -n hw.ncpu)

-- The C compiler identification is AppleClang 8.0.0.8000042
-- The CXX compiler identification is AppleClang 8.0.0.8000042
-- Check for working C compiler: /Library/Developer/CommandLineTools/usr/bin/cc
-- Check for working C compiler: /Library/Developer/CommandLineTools/usr/bin/cc -- works
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Detecting C compile features
-- Detecting C compile features - done
-- Check for working CXX compiler: /Library/Developer/CommandLineTools/usr/bin/c++
-- Check for working CXX compiler: /Library/Developer/CommandLineTools/usr/bin/c++ -- works
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- Looking for pthread.h
-- Looking for pthread.h - found
-- Looking for pthread_create
-- Looking for pthread_create - found
-- Found Threads: TRUE
-- Found CUDA: /usr/local/cuda (found version "8.0")
-- Boost version: 1.64.0
-- Found the following Boost libraries:
--   system
--   log_setup
--   log
--   date_time
--   filesystem
--   thread
--   regex
--   chrono
--   atomic
CUDA FOUND
-- Configuring done
-- Generating done
-- Build files have been written to: /Users/USER/nheqminer/Linux_cmake/nheqminer_cuda_tromp
[  5%] Building NVCC (Device) object CMakeFiles/nheqminer_cuda_tromp.dir/__/__/cuda_tromp/nheqminer_cuda_tromp_generated_equi_miner.cu.o
nvcc warning : The 'compute_20', 'sm_20', and 'sm_21' architectures are deprecated, and may be removed in a future release (Use -Wno-deprecated-gpu-targets to suppress warning).
In file included from /Users/USER/nheqminer/Linux_cmake/nheqminer_cuda_tromp/../../cuda_tromp/equi_miner.cu:9:
/Users/USER/nheqminer/Linux_cmake/nheqminer_cuda_tromp/../../cuda_tromp/../cpu_tromp/equi.h:7:10: fatal error: 'osx_barrier.h' file not found
#include "osx_barrier.h"
         ^
1 error generated.
CMake Error at nheqminer_cuda_tromp_generated_equi_miner.cu.o.cmake:222 (message):
  Error generating
  /Users/USER/nheqminer/Linux_cmake/nheqminer_cuda_tromp/CMakeFiles/nheqminer_cuda_tromp.dir/__/__/cuda_tromp/./nheqminer_cuda_tromp_generated_equi_miner.cu.o


make[2]: *** [CMakeFiles/nheqminer_cuda_tromp.dir/__/__/cuda_tromp/nheqminer_cuda_tromp_generated_equi_miner.cu.o] Error 1
make[1]: *** [CMakeFiles/nheqminer_cuda_tromp.dir/all] Error 2
make: *** [all] Error 2

error in CUDA9?

I am using cuda9 but following errors occurred:
nvcc fatal : Unsupported gpu architecture 'compute_20'
CMake Error at nheqminer_cuda_tromp_generated_equi_miner.cu.o.cmake:207 (message):

Is there any suggestion?

nheqminer_AMD unable to locate OpenCL device

Seems odd to being able to find OpenCL devices.

./nheqminer_AMD -oi

    ==================== www.nicehash.com ====================
            Equihash CPU&GPU Miner for NiceHash v0.4b
    Thanks to Zcash developers for providing base of the code.
           Special thanks to tromp, xenoncat, mbevand
                 and eXtremal-ik7 for providing
          optimized CPU, CUDA and AMD equihash solvers.
    ==================== www.nicehash.com ====================

Number of OpenCL devices found: 0
Platform: AMD Accelerated Parallel Processing | PlatformNum: 0
Devices:

clinfo | egrep -i 'Platform Version|Device Type|Board name'

Platform Version: OpenCL 2.0 AMD-APP (2348.3)
Device Type: CL_DEVICE_TYPE_GPU
Board name: Radeon RX 560 Series
Device Type: CL_DEVICE_TYPE_GPU
Board name: Radeon RX 560 Series

Compile Flags :-
cmake . -DOPENCL_LIBRARY=/opt/AMDAPPSDK-3.0/lib/x86_64/sdk/libOpenCL.so.1 -DOPENCL_INCLUDE_DIRECTORY=/opt/AMDAPPSDK-3.0/include

Anything I'm missing out ?

CUDA error

"CUDA error 'an illegal memory access was encountered' in func 'eq_cuda_context<RB, SM, SSM, THREADS, PACKER>::solve' line 2095" given when running the windows release.