geiccobs / denaro-cuda-miner

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Denaro CUDA miner (pool)

Usage

Download the latest files, you'll need just cuda.

Use ./cuda --help to see the full list of arguments.
Let's look at them:

  • --address - your wallet address, you can get it from https://t.me/DenaroCoinBot (learn about multi-address support)
  • --node - (def: https://node.denaro.is) node address to connect to
  • --pool - (def: https://pool.denaro.is) pool address to connect to
  • --silent - (def: false) don't print anything to stdout
  • --verbose - (def: false) don't clear stdout after each share (useful for debugging)
  • --device - (def: 0) GPU device ID, you can get it from nvidia-smi
  • --threads - (def: automatic) number of threads related to GPU
  • --blocks - (def: automatic) number of blocks related to GPU
  • --share - (def: 9) difficulty of shares, increase it if you see a lot of shares
  • --fee - (def: 5) dev fee, means that 1 every X blocks are mined by the dev (me <3)

Platforms

This miner is tested on both Linux and Windows (WSL 2).
Obviously it is working only on NVIDIA GPUs.

Installation

git clone https://github.com/geiccobs/denaro-cuda-miner
cd denaro-cuda-miner

Compiling by source

You can skip this if you want to use pre-built binary.

Install CUDA toolkit

sudo apt install build-essential libjson-c-dev libcurl4-openssl-dev cmake git # install dependencies
mkdir build && cd build # create a build directory
cmake .. && make # build the source code

Having problems with cmake version?

In this tutorial we'll be using the .sh installer.
Get the latest cmake installer from their download website.

sudo apt purge cmake # clear all the previous versions

chmod 700 cmake.sh # make the script executable
sudo mkdir /opt/cmake # create a directory for cmake
./cmake.sh --skip-license --prefix=/opt/cmake # install cmake
sudo ln -s /opt/cmake/bin/cmake /usr/bin/cmake # create a symlink
cmake --version # check the version

rm -f cmake.sh # remove the installer

Having problems with libjson-c?

git clone https://github.com/json-c/json-c.git # clone the source code
cd json-c # go to the source code directory
mkdir build && cd build # create a build directory
cmake .. && make # build the source code
sudo make install # install the library

cd ../.. # go back to the root directory
rm -rf json-c # remove the source code

Discussed topics

Dev fee

I've done this work basically for free, without having any idea about CUDA, nor an NVIDIA GPU.
Dev fee can be obviously turned off, just by setting --fee parameter to huge values or 0.

Multiple GPUs

I'll be working on it.
Currently, you can select which GPU to use by setting --device parameter, but you can't use multiple GPUs at the same time.

Avoid crashing

Seems to have finally addressed the majority of the crashing issues, if you want to be sure anyway, go ahead.

To avoid your miner stopping after a crash (because yes, it can always happen) you can start it using the following command in your terminal:

while true; do ./cuda --address YOUR_ADDRESS; sleep 1; done

There you go, your miner won't need to be manually restarted after any issue.

Multi-address

Multi-address support is available, you can use it by setting --address parameter to a comma-separated list of addresses.

Here's an example:

./cuda --address YOUR_ADDRESS,ADDRESS_2,ADDRESS_3

About


Languages

Language:Cuda 98.9%Language:CMake 1.1%