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
• Compile, using cocl:

$ cocl 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

Advanced usage

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)
• cuDNN API implementations for: convolutions (using im2col algorithm over Cedric Nugteren's CLBlast, pooling, ReLU, tanh, and sigmoid

How Coriander works

Kernel compilation proceeds in two steps:

Slides on the IWOCL website, here

Installation

Coriander development is carried out using the following platforms:

• Ubuntu 16.04, with:
  • NVIDIA K80 GPU
• Mac Sierra, with:
  • Intel HD Graphics 530
  • Radeon Pro 450

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.md

Testing

See testing.md

Simplifications made by Coriander

Coriander makes the following relaxations/simplifications:

• ints are generally assumed to be no longer than 32-bit, and truncated to 32-bit mostly
• floats are assumed to be singles. doubles in the original kernels are converted to floats in the OpenCL code
• buffer offsets are generally taken to be int32s for now. This might change in the future

Related projects

• Eigen-CL: Minimally-tweaked fork of Eigen, which can be compiled/run using Coriander, on an OpenCL device, https://bitbucket.org/hughperkins/eigen/commits/branch/eigen-cl
• Tensorflow-CL: Fork of Tensorflow, that can be built and run on an OpenCL-1.2 enabled GPU, using Coriander, https://github.com/hughperkins/tensorflow-cl
• CLBlast: Cedric Nugteren's excellent BLAS implementation for OpenCL CLBlast

License

Apache 2.0

News

• May 20:
  • renamed to Coriander
• May 18:
  • Presented Coriander at this year's IWOCL :-) Full IWOCL program here, and there is a link to my own slides
• May 5:
  • Eigen unit tests at https://bitbucket.org/hughperkins/eigen/src/75842846799e15f1c26ef6885565d64c3d0a67b2/unsupported/test/Coriander/?at=eigen-cl pass on Mac Pro 4th Generation with both:
    • Intel HD Graphics 530, and
    • Radeon Pro 450 (using env var CL_GPUOFFSET=1 to select)
  • I suspect this may have broken some other stuff, since one of the unit tests fails now, but I think it's a gentle step forward
• May 1:
  • dnn tests pass on Radeon Pro 450, on Mac Sierra now
  • fix crash bugs in pooling forward/backward, on Mac Sierra
  • thanks to my employer ASAPP giving me use of a nice Mac Book Pro 4th Generation, with Radeon Pro 450, unit tests now pass on said hardware :-)
• April 29:
  • Updated to latest EasyCL. This lets you use environment variable CL_GPUOFFSET to choose different gpus, eg set to 1 to use second gpu, to 2 to use third gpu, etc
• Older news