https://github.com/cogciprocate/ocl
OpenCL for Rust
https://github.com/cogciprocate/ocl
amd async gpgpu intel nvidia ocl opencl-api opencl-library rust scientific-computing
Last synced: 1 day ago
JSON representation
OpenCL for Rust
- Host: GitHub
- URL: https://github.com/cogciprocate/ocl
- Owner: cogciprocate
- License: other
- Created: 2015-10-31T02:03:56.000Z (over 9 years ago)
- Default Branch: master
- Last Pushed: 2024-04-05T21:11:41.000Z (about 1 year ago)
- Last Synced: 2025-04-06T14:07:56.387Z (8 days ago)
- Topics: amd, async, gpgpu, intel, nvidia, ocl, opencl-api, opencl-library, rust, scientific-computing
- Language: Rust
- Size: 34.1 MB
- Stars: 755
- Watchers: 14
- Forks: 74
- Open Issues: 34
-
Metadata Files:
- Readme: README.md
- License: LICENSE-APACHE
Awesome Lists containing this project
- awesome-cuda-and-hpc - ocl
- awesome-cuda-and-hpc - ocl
- awesome-rust-list - ocl
- awesome-rust-list - ocl
README
ocl
===
#### [Documentation](https://docs.rs/ocl) | [Change Log](https://github.com/cogciprocate/ocl/blob/master/RELEASES.md)
[](https://crates.io/crates/ocl) [](https://docs.rs/ocl)
[](https://en.wikipedia.org/wiki/Cross-platform)
[](https://travis-ci.org/cogciprocate/ocl)
Pure OpenCL™ bindings and interfaces for
[Rust](https://www.rust-lang.org/).
## Goals
To provide:
- A simple and intuitive interface to OpenCL devices
- The full functionality and power of the OpenCL API
- An absolute minimum of boilerplate
- Zero or virtually zero performance overhead
- Thread-safe and automatic management of API pointers and resources
## Usage
Ensure that an OpenCL library is installed for your platform and that `clinfo`
or some other diagnostic command will run. Add the following to your project's
`Cargo.toml`:
```toml
[dependencies]
ocl = "0.19"
```
And add the following to your crate root (lib.rs or main.rs):
```rust
extern crate ocl;
```
## Example
From [`examples/trivial.rs`]:
```rust
extern crate ocl;
use ocl::ProQue;
fn trivial() -> ocl::Result<()> {
let src = r#"
__kernel void add(__global float* buffer, float scalar) {
buffer[get_global_id(0)] += scalar;
}
"#;
let pro_que = ProQue::builder()
.src(src)
.dims(1 << 20)
.build()?;
let buffer = pro_que.create_buffer::()?;
let kernel = pro_que.kernel_builder("add")
.arg(&buffer)
.arg(10.0f32)
.build()?;
unsafe { kernel.enq()?; }
let mut vec = vec![0.0f32; buffer.len()];
buffer.read(&mut vec).enq()?;
println!("The value at index [{}] is now '{}'!", 200007, vec[200007]);
Ok(())
}
```
See the the remainder of [`examples/trivial.rs`] for more information about
how this library leverages Rust's zero-cost abstractions to provide the full
power and performance of the C API in a simple package.
## Recent Changes
* 0.18.0: Creating a
[`Kernel`](https://docs.rs/ocl/0.18.0/ocl/struct.Kernel.html) now requires
the use of the new
[`KernelBuilder`](https://docs.rs/ocl/0.18.0/ocl/struct.KernelBuilder.html).
See the [change
log](https://github.com/cogciprocate/ocl/blob/master/RELEASES.md) for more
information.
##### Introduction to OpenCL
For a quick but thorough primer on the basics of OpenCL, please see [Matthew
Scarpino's excellent article, 'A Gentle Introduction to OpenCL' at
drdobbs.com](http://www.drdobbs.com/parallel/a-gentle-introduction-to-opencl/231002854)
(his
[book](https://www.amazon.com/OpenCL-Action-Accelerate-Graphics-Computations/dp/1617290173/ref=sr_1_2?ie=UTF8&qid=1500745843&sr=8-2&keywords=opencl)
is great too).
##### Diving Deeper
Already familiar with the standard OpenCL core API? See the [`ocl-core`] crate
for access to the complete feature set in the conventional API style with
Rust's safety and convenience.
##### Version Support
OpenCL versions 1.1 and above are supported. OpenCL version 1.0 is **not**
supported due to its inherent thread unsafety.
##### Vulkan™ and the Future
The OpenCL API already posesses all of the new attributes of the Vulkan API
such as low-overhead, high performance, and unfettered hardware access. For all
practical purposes, Vulkan is simply a graphics-focused superset of OpenCL's
features (sorta kinda). OpenCL 2.1+ and Vulkan kernels/shaders now both
compile into SPIR-V making the device side of things the same. I wouldn't be
suprised if most driver vendors implement the two host APIs identically.
In the future it's possible the two may completely merge (or that Vulkan will
absorb OpenCL). Whatever happens, nothing will change as far as the front end
of this library is concerned. This library will maintain its focus on the
compute side of things.
##### License
Licensed under either of:
* Apache License, Version 2.0, ([LICENSE-APACHE](LICENSE-APACHE) or http://www.apache.org/licenses/LICENSE-2.0)
* MIT license ([LICENSE-MIT](LICENSE-MIT) or http://opensource.org/licenses/MIT)
at your option.
##### Contribution
Unless you explicitly state otherwise, any contribution intentionally submitted
for inclusion in the work by you, as defined in the Apache-2.0 license, shall be dual licensed as above, without any
additional terms or conditions.
*“OpenCL and the OpenCL logo are trademarks of Apple Inc. used by
permission by Khronos.”*
*“Vulkan and the Vulkan logo are trademarks of
the Khronos Group Inc.”*
[OpenCL libraries for your CPU]: https://software.intel.com/en-us/intel-opencl/download
[AMD]: https://software.intel.com/en-us/intel-opencl/download
[`ocl-core`]: https://github.com/cogciprocate/ocl/tree/master/ocl-core
[issue]: https://github.com/cogciprocate/ocl_rust/issues
[provide feedback]: https://github.com/cogciprocate/ocl_rust/issues
[`examples`]: https://github.com/cogciprocate/ocl/tree/master/ocl/examples
[`examples/trivial.rs`]: https://github.com/cogciprocate/ocl/blob/master/ocl/examples/trivial.rs
[voodoo]: https://github.com/cogciprocate/voodoo
[intel-win64]: https://software.intel.com/en-us/articles/opencl-drivers#win64
[intel-linux64-redhat-suse]: https://software.intel.com/en-us/articles/opencl-drivers#lin64
[intel-linux64-ubuntu]: https://software.intel.com/en-us/articles/opencl-drivers#ubuntu64
[amd-app-sdk]: http://developer.amd.com/tools-and-sdks/opencl-zone/amd-accelerated-parallel-processing-app-sdk/