Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

https://fakenmc.github.io/cf4ocl

C Framework for OpenCL
https://fakenmc.github.io/cf4ocl

c framework gpgpu gpu hpc opencl opencl-api opencl-objects scientific-computing wrapper

Last synced: about 2 months ago
JSON representation

C Framework for OpenCL

Lists

README 

        
[![ci](https://github.com/nunofachada/cf4ocl/actions/workflows/main.yml/badge.svg)](https://github.com/nunofachada/cf4ocl/actions/workflows/main.yml)

[![codecov](https://codecov.io/gh/nunofachada/cf4ocl/branch/master/graph/badge.svg)](https://codecov.io/gh/nunofachada/cf4ocl)

[![Coverity Scan Build Status](https://scan.coverity.com/projects/11475/badge.svg)](https://scan.coverity.com/projects/fakenmc-cf4ocl)

[![Latest release](https://img.shields.io/github/release/nunofachada/cf4ocl.svg)](https://github.com/nunofachada/cf4ocl/releases)

[![LGPL Licence](https://img.shields.io/badge/license-LGPLv3-yellowgreen.svg)](https://opensource.org/licenses/LGPL-3.0/)

[![Supported platforms](https://img.shields.io/badge/platform-windows%20%7C%20macos%20%7C%20linux%20%7C%20bsd-orange.svg)](https://en.wikipedia.org/wiki/Cross-platform)



News

====



*6 Jan. 2024*



- This repository is now archived and will not be further developed by the original author. It's hard to make a case for using OpenCL in pure C nowadays, given the existence of excellent wrappers for other programming languages. If someone wants to pick up where I left off, let me know.



*4 July 2016*



- Version 2.1.0 is available for download in the [releases][] page.



Summary

=======



The C Framework for OpenCL, _cf4ocl_, is a cross-platform pure C object-oriented

framework for developing and benchmarking [OpenCL][] projects. It aims to:



1. Promote the rapid development of OpenCL host programs in C (with support for

C++) and avoid the tedious and error-prone boilerplate code usually required.

1. Assist in the benchmarking of OpenCL events, such as kernel execution and

data transfers. Profiling comes for **free** with _cf4ocl_.

1. Simplify the analysis of the OpenCL environment and of kernel requirements.

2. Allow for all levels of integration with existing OpenCL code: use as much or

as few of _cf4ocl_ required for your project, with full access to the underlying

OpenCL objects and functions at all times.



Features

========



* Object-oriented interface to the OpenCL API

  * New/destroy functions, no direct memory alloc/free

  * Easy (and extensible) device selection

  * Simple event dependency mechanism

  * User-friendly error management

* OpenCL version and platform independent

* Integrated profiling

* Advanced device query utility

* Offline kernel compiler and linker



Documentation

=============



* [User guide and API](http://www.fakenmc.com/cf4ocl/docs/latest/)

* [Tutorial](http://www.fakenmc.com/cf4ocl/docs/latest/tut.html)

* [Wiki](https://github.com/nunofachada/cf4ocl/wiki)

* [Utilities](http://www.fakenmc.com/cf4ocl/docs/latest/utils.html)

* [Function list](http://www.fakenmc.com/cf4ocl/docs/latest/funlist.html)

* [Library modules](http://www.fakenmc.com/cf4ocl/docs/latest/modules.html)

* [Examples](http://www.fakenmc.com/cf4ocl/docs/latest/examples.html)



Feedback and collaboration

==========================



Download or clone _cf4ocl_,

[build and install it](https://github.com/nunofachada/cf4ocl/wiki/Build-and-install-from-source),

and [code](https://github.com/nunofachada/cf4ocl/wiki/Using-cf4ocl-in-a-new-project)

a small example, such as the one below, which shows a clean and fast way to

create an OpenCL context with a user-selected device:



```c

#include 

int main() {



    /* Variables. */

    CCLContext * ctx = NULL;



    /* Code. */

    ctx = ccl_context_new_from_menu(NULL);

    if (ctx == NULL) exit(-1);



    /* Destroy context wrapper. */

    ccl_context_destroy(ctx);



    return 0;

}

```



If you like this project and want to contribute, take a look at the existing

[issues][]. We also need help with [binary packaging][] for different OSes.

Other improvements or suggestions are of course, welcome. We appreciate any

feedback.



Not yet integrated

==================



A few OpenCL API calls, most of which introduced with OpenCL 2.1, are

[not yet integrated][enhancements] with _cf4ocl_. However, this functionality is

still available to client code, because _cf4ocl_ can be used simultaneously with

raw OpenCL objects and functions.



Reference

=========



* Fachada, N., Lopes, V.V., Martins, R.C., Rosa, A.C., (2017)

cf4ocl: A C framework for OpenCL,

[Science of Computer Programming](https://www.journals.elsevier.com/science-of-computer-programming),

143:9–19, http://www.sciencedirect.com/science/article/pii/S0167642317300540

([arXiv preprint](https://arxiv.org/abs/1609.01257))



License

=======



Library code is licensed under [LGPLv3][], while the remaining code

(utilities, examples and tests) is licensed under [GPLv3][].



Other useful C frameworks/utilities for OpenCL

==============================================



If _cf4ocl_ does not meet your requirements, take a look at the following

projects:



* [Simple OpenCL][]

* [The OpenCL utility library][]

* [OCL-MLA][]

* [oclkit][]

* [hiCL][]

* [ocl-ke][]



[releases]: https://github.com/nunofachada/cf4ocl/releases

[OpenCL]: http://www.khronos.org/opencl/ "OpenCL"

[LGPLv3]: http://www.gnu.org/licenses/lgpl.html "LGPLv3"

[GPLv3]: http://www.gnu.org/licenses/gpl.html "GPLv3"

[Wiki]: https://github.com/nunofachada/cf4ocl/wiki

[enhancements]: https://github.com/nunofachada/cf4ocl/issues?q=is%3Aissue+is%3Aopen+label%3Aenhancement

[issues]: https://github.com/nunofachada/cf4ocl/issues "issues"

[binary packaging]: https://github.com/nunofachada/cf4ocl/wiki/Install-the-binaries

[Simple OpenCL]: https://github.com/morousg/simple-opencl "Simple OpenCL"

[The OpenCL utility library]: https://github.com/Oblomov/CLU "The OpenCL utility library"

[OCL-MLA]: http://tuxfan.github.io/ocl-mla/ "OCL-MLA"

[oclkit]: https://github.com/matze/oclkit "oclkit"

[hiCL]: https://github.com/issamsaid/hiCL "high level OpenCL abstraction layer for scientific computing"

[ocl-ke]: https://github.com/anyc/ocl-ke "OpenCL kernel extractor"