https://github.com/thilinarmtb/gemv
Wrapper for HIP/CUDA gemv using vendor provided libraries.
https://github.com/thilinarmtb/gemv
Last synced: 3 months ago
JSON representation
Wrapper for HIP/CUDA gemv using vendor provided libraries.
- Host: GitHub
- URL: https://github.com/thilinarmtb/gemv
- Owner: thilinarmtb
- Created: 2023-12-01T16:57:57.000Z (almost 2 years ago)
- Default Branch: main
- Last Pushed: 2024-04-12T19:54:14.000Z (over 1 year ago)
- Last Synced: 2025-02-01T09:13:23.368Z (8 months ago)
- Language: C
- Homepage:
- Size: 88.9 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# GEMV
`GEMV` is a library for performing platform independent dense matrix-vector
products on the GPUs.
## Build Instructions
This project uses conda to manage dependencies (CMake, clang-format, clang-tidy
and other dependencies for documentation). Dependencies can be installed by
executing following commands after installing
[conda](https://docs.conda.io/en/latest/miniconda.html).
```sh
conda env create -f environment-dev.yml
conda activate gemv
```
Then simply run `gemv.sh` script to build and install the library.
```sh
./gemv.sh --install
```
You can format the source code with `clang-format` using the option `--format`.
```
./gemv.sh --format
```
Use `--help` to see all the options supported by `gemv.sh` script.
```
./gemv.sh --help
```