Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/hughperkins/coriander
Build NVIDIA® CUDA™ code for OpenCL™ 1.2 devices
https://github.com/hughperkins/coriander
amd coriander gpu intel mac nvidia opencl radeon ubuntu
Last synced: about 2 months ago
JSON representation
Build NVIDIA® CUDA™ code for OpenCL™ 1.2 devices
- Host: GitHub
- URL: https://github.com/hughperkins/coriander
- Owner: hughperkins
- License: apache-2.0
- Created: 2016-09-28T01:28:05.000Z (almost 8 years ago)
- Default Branch: master
- Last Pushed: 2024-02-14T05:37:58.000Z (7 months ago)
- Last Synced: 2024-04-10T05:25:25.231Z (6 months ago)
- Topics: amd, coriander, gpu, intel, mac, nvidia, opencl, radeon, ubuntu
- Language: LLVM
- Homepage:
- Size: 7.77 MB
- Stars: 831
- Watchers: 48
- Forks: 87
- Open Issues: 40
-
Metadata Files:
- Readme: README.md
- Funding: .github/FUNDING.yml
- License: LICENSE
Awesome Lists containing this project
README
# Coriander
Build applications written in NVIDIA® CUDA™ code for OpenCL™ 1.2 devices.
## Concept
- leave applications in NVIDIA® CUDA™
- compile into OpenCL 1.2
- run on any OpenCL 1.2 GPU
## How to use
- Write an NVIDIA® CUDA™ sourcecode file, or find an existing one
- Let's use [cuda_sample.cu](https://github.com/hughperkins/Coriander/blob/76a849d9510276bc67167c9a7676d64ff04c3e4a/test/cuda_sample.cu)
- Compile, using `cocl`:
```
$ cocl_py cuda_sample.cu
...
... (bunch of compily stuff) ...
...
./cuda_sample.cu compiled into ./cuda_sample
```
Run:
```
$ ./cuda_sample
Using Intel , OpenCL platform: Intel Gen OCL Driver
Using OpenCL device: Intel(R) HD Graphics 5500 BroadWell U-Processor GT2
hostFloats[2] 123
hostFloats[2] 222
hostFloats[2] 444
```
[Options](doc/options.md)
## What Coriander provides
- compiler for host-side code, including memory allocation, copy, streams, kernel launches
- compiler for device-side code, handling templated C++ code, converting it into bog-standard OpenCL 1.2 code
- cuBLAS API implementations for GEMM, GEMV, SCAL, SAXPY (using Cedric Nugteren's [CLBlast](https://github.com/cnugteren/CLBlast))
- cuDNN API implementations for: convolutions (using `im2col` algorithm over Cedric Nugteren's [CLBlast](https://github.com/cnugteren/CLBlast), pooling, ReLU, tanh, and sigmoid
## How Coriander works
Kernel compilation proceeds in two steps:
Slides on the [IWOCL](http://iwocl.org) website, [here](http://www.iwocl.org/wp-content/uploads/iwocl2017-hugh-perkins-cuda-cl.pdf)
## Installation
Coriander development is carried out using the following platforms:
- Ubuntu 16.04, with:
- NVIDIA K80 GPU and/or NVIDIA K520 GPU (via aws)
- Mac Book Pro 4th generation (thank you [ASAPP](http://asapp.com) :-) ), with:
- Intel HD Graphics 530
- Radeon Pro 450
- Sierra OS
Other systems should work too, ideally. You will need at a minimum at least one OpenCL-enabled GPU,
and appropriate OpenCL drivers installed, for the GPU. Both linux and Mac systems stand a reasonable chance of working ok.
For installation, please see [installation](doc/installation.md)
## Plugins
You can install the following plugins:
- [Coriander-clblast](https://github.com/hughperkins/coriander-clblast): just do `cocl_plugins.py install --repo-url https://github.com/hughperkins/coriander-clblast`
- [Coriander-dnn](https://github.com/hughperkins/coriander-dnn): just do `cocl_plugins.py install --repo-url https://github.com/hughperkins/coriander-dnn`
- Your plugin here?
[How to create a plugin](doc/plugins.md)
## Add to your own `cmake` project
- use `cocl_add_executable` and `cocl_add_library`
- see [cmake usage](doc/cmake_usage.md)
## Testing
See [testing](doc/testing.md)
## Assumptions/relaxations made by Coriander
See [assumptions](doc/assumptions.md)
## Roadmap
## Libraries
Coriander uses the following libraries:
- [clang/llvm](http://llvm.org/): c/c++ parser/compiler; many contributors
- [thrust](https://github.com/thrust/thrust): parallel GPU library, from NVIDIA®
- [yaml-cpp](https://github.com/jbeder/yaml-cpp): yaml for c++, by Jesse Beder
- [EasyCL](https://github.com/hughperkins/EasyCL): wrapper for OpenCL 1.2 boilerplate
- [argparsecpp](https://github.com/hughperkins/argparsecpp): command-line parser for c++
- [gtest](https://github.com/google/googletest): unit tests for c++, from Google
## Related projects
- [Eigen-CL](https://bitbucket.org/hughperkins/eigen/commits/branch/eigen-cl): Minimally-tweaked fork of Eigen, for OpenCL 1.2
- [tf-coriander](https://github.com/hughperkins/tf-coriander): Tensorflow for OpenCL-1.2
## How to Cite
Please cite: [CUDA-on-CL: a compiler and runtime for running NVIDIA® CUDA™ C++11 applications on OpenCL™ 1.2 Devices](http://dl.acm.org/citation.cfm?id=3078156)
## License
[Apache 2.0](LICENSE)
## News
- June 23:
- factorized CLBlast implementation of NVIDIA® CUDA™ cuBLAS API, into new plugin [coriander-clblast](https://github.com/hughperkins/coriander-clblast)
- June 21:
- created a new release [v6.0.0](https://github.com/hughperkins/coriander/releases/tag/v6.0.0), that marks a bunch of changes:
- incorporates of course the earlier changes:
- took some big steps towards portability and Windows compilation, ie using python 2.7 scripts, rather than bash scripts, and fixing many Windows-related compilation issues
- the plugin architecture
- factorizing the partial NVIDIA® CUDA™ cuDNN API implementation into a new plugin [coriander-dnn](https://github.com/hughperkins/coriander-dnn)
- moved the default installation directory from `/usr/local` to `~/coriander`
- this means that plugins can be installed without `sudo`
- it also makes it relatively easy to wipe and reinstall, for more effective jenkins testing
- `install_distro.py` is now considerably more tested than a few days ago, and handles downloading `llvm-4.0` automatically
- [Older news](doc/news.md)