xmrMiner supports all NVIDIA GPUs with a compute capability >=2.0 You can check the compute capability on this side.
-
NVIDIA CUDA >=6.0
- Debian/Ubuntu:
sudo apt-get install nvidia-cuda-dev nvidia-cuda-toolkit
- Debian/Ubuntu:
-
host compiler
- clang >=3.9 (support compile of the host and device code)
- gcc >=4.6 (depends on your current CUDA version)
- by default CUDA8 is not supporting gcc 6.x, never the less to compile with the newest gcc version you need follow this steps
- open the file
$CUDAINSTALLDIR/include/host_config.h
and change(remove) the following line
to#if __GNUC__ > 5 #error -- unsupported GNU version! gcc versions later than 5 are not supported! #endif /* __GNUC__ > 5 */
#if __GNUC__ > 5 #endif /* __GNUC__ > 5 */
-
SSL support
- Debian/Ubuntu:
sudo apt-get install libssl-dev
- Debian/Ubuntu:
-
CMake >=3.3.0
- Debian/Ubuntu:
sudo apt-get install cmake cmake-curses-gui
- Debian/Ubuntu:
-
Curl
- Debian/Ubuntu:
sudo apt-get install libcurl4-gnutls-dev
- Debian/Ubuntu:
-
git
- Debian/Ubuntu:
sudo apt-get install git
- Debian/Ubuntu:
-
Jansson
- Debian/Ubuntu:
sudo apt-get install libjansson-dev
- From Source:
git clone https://github.com/akheron/jansson.git
cd jansson
mkdir build
cd build
cmake .. -DCMAKE_INSTALL_PREFIX=$HOME/usr/jansson
make install
- set the path to jansson before you compile xmrMiner
export CMAKE_PREFIX_PATH=$CMAKE_PREFIX_PATH:$HOME/usr/jansson
- Debian/Ubuntu:
If you have compiled a dependency by hand please add the path to the install folder to CMAKE_PREFIX_PATH
e.g.,
export CMAKE_PREFIX_PATH=$CMAKE_PREFIX_PATH:$JANSSON_ROOT
- create and enter the project folder:
mkdir -p xmrMinerProject
cd xmrMinerProject
- download the
xmrMiner
source code
git clone ....
- create a temporary build folder and enter
mkdir -p build
cd build
- configure
xmrMiner
(search for all dependencies) and add the install path (creates a folderxmrMiner
within the home)
cmake -DCMAKE_INSTALL_PREFIX=$HOME/xmrMiner ../xmrMiner
- optional you can change all compile time options with a ncurses gui
ccmake .
inside the build folder- after a option in ccmake is changed you need to end ccmake with the key
c
and thang
- compile
make -j install
- optional to speedup the compile you can change the CMake option
CUDA_ARCH
to the compute capability of your NVIDIA GPUccmake
orcmake -DCUDA_ARCH=61
+ options from step4
(for Pascal)
- start xmrMiner
cd $HOME/xmrMiner
./xmrMiner --help