arbal/ethminer repository overview

⁠ethminer

This container is a CUDA enabled build of ethereum-mining/ethminer. It tracks the head of the master branch at build time. I used the guidance provided in the build.md⁠ of the repository to put this image together.

⁠Tip Jar

If you're using this container image, tips in eth are appreciated! :bowtie: 0xAe26Cd1573745236807E67cE3be34Ca4226f4cF3

⁠Overview

To minimize the complexity of pulling in the required dependencies and minimizing the size of the image, I use the nvidia/cuda devel image to build the ethminer binary. In the second stage, I copy the build artifacts to a new image dependent on nvidia/cuda runtime.

The entrypoint for the final image is /opt/ethminer/ethminer.

This provides the ability to pass arguments at container execution time. For example:

$ docker run --gpus all -it kriation/ethminer:latest --list-devices


ethminer 0.19.0+commit.47ae149e
Build: linux/release/gnu

 Id Pci Id    Type Name                          CUDA SM   Total Memory
 --- --------- ---- ----------------------------- ---- ---  ------------
   0 01:00.0   Gpu  GeForce GTX 1660 SUPER        Yes  7.5       1.52 GB

The prerequisite to using this image is that the NVIDIA container toolkit is configured properly on the container host. The documentation⁠ from NVIDIA on setting the toolkit up is straightforward.

⁠Build

For those interested in building a custom image that isn't available in Docker Hub, you can pass a build argument of CUDA_VERSION set to any of the tags listed in the CUDA Supported Tags⁠ documentation.

An example is below:

$ docker build --build-arg CUDA_VERSION=10.0 -t kriation/ethminer:10.0 .

The resulting image will be based off of the CUDA 10.0 base.