CCMiner: how to compile on Linux (Ubuntu / Fedora) and macOS

CCMiner - How to install and use? (Download and Configure for Windows)


CCMiner: how to compile on Linux (Ubuntu/Fedora) and macOS

In this article, we will look at building methods for CCMiner for Linux and macOS.

You can also read our setup article CCminer for Windows.

you can use ./ to configure and build with default settings.

Recommended to run ./ before ./configure (autoconf and automake for to work must be installed on your system)

./configure has a parameter called –with-cuda that allows you to specify where your CUDA 6.5 toolkit is installed (usually / usr / local / cuda, but some distributions may have a different default location)

How to compile CCMiner on Ubuntu (16.04 LTS+)

Step 1. Installing Cuda and Nvidia driver

First install the Cuda toolkit and nVidia driver and type `nvidia-smi` to check if your card is detected.

Step 2. Installing the dependencies

sudo apt-get install libcurl4-openssl-dev libssl-dev libjansson-dev automake autotools-dev build-essential 

Step 3: gcc/g ++ 5

Ubuntu now ships with gcc 6 or 7, so install gcc / g ++ 5 and make it the default (required by cuda toolkit)

sudo apt-get install gcc-5 g ++ - 5

Then use the ./ helper in the ccminer source folder, edit the file and the file if necessary.

  • ./
  • ./ccminer –version

How to compile CCMiner on Fedora 25+

Note: you can find an alternative method via rpms: see and

Step 1: gcc and dependencies

1. dnf install gcc gcc-c++ autoconf automake 
2. dnf install jansson-devel openssl-devel libcurl-devel zlib-deve  

Step 2: nvidia drivers

1. dnf install kernel-devel       
2. dnf install$(rpm -E %fedora).noarch.rpm      
3. dnf check-update       
4. dnf install xorg-x11-drv-nvidia-cuda kmod-nvidia 
5. ln -s /usr/lib64/ 

Step 3: CUDA SDK (Download the .run shared archive from nvidia)

1. --override is required to ignore "too recent" gcc 6.3       
2. --silent is required to install only the toolkit (no kmod)       
3. ./ --toolkit --silent --override       
4. nvcc --version 
  • Add nvcc binary path to system
ln -s /usr/local/cuda-8.0 / usr / local / cuda # (if not already done)
echo 'export PATH = $ PATH: / usr / local / cuda / bin'> /etc/profile.d/
  • Add cudart library path to the system
echo / usr / local / cuda / lib64> /etc/

Step 4: Fix Toolkit Incompatibility with gcc 6

You need to build an older version of GCC / G ++, I recommend 5.4 see

Note: this manual method will override gcc by default, it might be better to use a custom tool chain prefix

dnf install libmpc-devel mpfr-devel gmp-devel
./configure --prefix = / usr / local --enable-languages = c, c ++, lto --disable-multilib
make -j 8 && make install
(while at this point you have time to cook something: p)

# or, for previous versions of fedora, edit the file /usr/local/cuda/include/host_config.h
# and comment / delete line 121: #error is an unsupported version of GNU! Versions of gcc over 5 are not supported!
  • ./
  • ./ccminer -n

How to compile CCMiner on macOS

Step 1: download and install CUDA Toolkit 8 or later

Step 2: Install Homebrew

ruby -e "$ (curl -fsSL"

Step 3: Dependencies

brew install pkg-config autoconf automake curl openssl llvm
  • ./
  • ./ccminer -n

Recommended Related Articles:

This post is also available in: Русский

Leave a Reply

Your email address will not be published. Required fields are marked *