OpenCL

From Wikipedia, the free encyclopedia

OpenCL (Open Computing Language) is a framework for writing programs that execute across heterogeneous platforms consisting of CPUs, GPUs, and other processors. OpenCL includes a language (based on C99) for writing kernels (functions that execute on OpenCL devices), plus APIs that are used to define and then control the platforms. OpenCL provides parallel computing using task-based and data-based parallelism.

OpenCL is analogous to the open industry standards OpenGL and OpenAL, for 3D graphics and computer audio, respectively. OpenCL extends the power of the GPU beyond graphics (GPGPU). OpenCL is managed by the non-profit technology consortium Khronos Group.

[edit] History

OpenCL was initially developed by Apple Inc., which holds trademark rights, and refined into an initial proposal in collaboration with technical teams at AMD, Intel and Nvidia. Apple submitted this initial proposal to the Khronos Group. On June 16, 2008, the Khronos Compute Working Group was formed^[1] with representatives from CPU, GPU, embedded-processor, and software companies. This group worked for five months to finish the technical details of the specification for OpenCL 1.0 by November 18, 2008.^[2] This technical specification was reviewed by the Khronos members and approved for public release on December 8, 2008.^[3]

OpenCL 1.0 is scheduled to be introduced in Mac OS X v10.6 ("Snow Leopard"). According to an Apple press release:^[4]

Snow Leopard further extends support for modern hardware with Open Computing Language (OpenCL), which lets any application tap into the vast gigaflops of GPU computing power previously available only to graphics applications. OpenCL is based on the C programming language and has been proposed as an open standard.

AMD has decided to support OpenCL (and DirectX 11) instead of the now deprecated Close to Metal in its Stream framework.^[5][6] RapidMind announced their adoption of OpenCL underneath their development platform, in order to support GPUs from multiple vendors with one interface.^[7] On December 9, 2008, Nvidia announced its intention to add full support for the OpenCL 1.0 specification to its GPU Computing Toolkit.^[8]

The OpenCL specification is under development at Khronos, which is open to any interested company to join.

[edit] Implementation

On December 10, 2008, AMD and Nvidia held the first public OpenCL demonstration, a 75-minute presentation at Siggraph Asia 2008. AMD showed a CPU-accelerated OpenCL demo explaining the scalability of OpenCL on one or more cores while Nvidia showed a GPU-accelerated demo.^[9][10]

On March 26, 2009, at GDC 2009, AMD and Havok demonstrated the first working implementation for OpenCL accelerating Havok Cloth on AMD Radeon HD 4000 series GPU.^[11]

On April 20, 2009, Nvidia announced the release of its OpenCL driver and SDK to developers participating in its OpenCL Early Access Program.^[12]

In September 2009, Apple plans to release the Snow Leopard version of Mac OS X, which contains an implementation of OpenCL.^[13] OpenCL in Snow Leopard will initially be supported on the NVIDIA Geforce 8600M GT, GeForce 8800 GT, GeForce 8800 GTS, Geforce 9400M, GeForce 9600M GT, GeForce GT 120, GeForce GT 130, ATI Radeon 4850, and Radeon 4870.^[14]

[edit] Example

This example will compute a Fast Fourier Transformation (FFT): ^[15]

// create a compute context with GPU device
context = clCreateContextFromType(0, CL_DEVICE_TYPE_GPU, NULL, NULL, NULL);
 
// create a work-queue
queue = clCreateWorkQueue(context, NULL, NULL, 0);
 
// allocate the buffer memory objects
memobjs[0] = clCreateBuffer(context, CL_MEM_READ_ONLY | CL_MEM_COPY_HOST_PTR, sizeof(float)*2*num_entries, srcA);
memobjs[1] = clCreateBuffer(context, CL_MEM_READ_WRITE, sizeof(float)*2*num_entries, NULL);
 
// create the compute program
program = clCreateProgramFromSource(context, 1, &fft1D_1024_kernel_src, NULL);
 
// build the compute program executable
clBuildProgramExecutable(program, false, NULL, NULL);
 
// create the compute kernel
kernel = clCreateKernel(program, "fft1D_1024");
 
// create N-D range object with work-item dimensions
global_work_size[0] = n;
local_work_size[0] = 64;
range = clCreateNDRangeContainer(context, 0, 1, global_work_size, local_work_size);
 
// set the args values
clSetKernelArg(kernel, 0, (void *)&memobjs[0], sizeof(cl_mem), NULL);
clSetKernelArg(kernel, 1, (void *)&memobjs[1], sizeof(cl_mem), NULL);
clSetKernelArg(kernel, 2, NULL, sizeof(float)*(local_work_size[0]+1)*16, NULL);
clSetKernelArg(kernel, 3, NULL, sizeof(float)*(local_work_size[0]+1)*16, NULL);
 
// execute kernel
clExecuteKernel(queue, kernel, NULL, range, NULL, 0, NULL);

The actual calculation: (Based on Fitting FFT onto the G80 Architecture)^[16]

// This kernel computes FFT of length 1024. The 1024 length FFT is decomposed into 
// calls to a radix 16 function, another radix 16 function and then a radix 4 function 
 
__kernel void fft1D_1024 (__global float2 *in, __global float2 *out, 
                          __local float *sMemx, __local float *sMemy) { 
  int tid = get_local_id(0); 
  int blockIdx = get_group_id(0) * 1024 + tid; 
  float2 data[16]; 
 
  // starting index of data to/from global memory 
  in = in + blockIdx;  out = out + blockIdx; 
 
  globalLoads(data, in, 64); // coalesced global reads 
  fftRadix16Pass(data);      // in-place radix-16 pass 
  twiddleFactorMul(data, tid, 1024, 0); 
 
  // local shuffle using local memory 
  localShuffle(data, sMemx, sMemy, tid, (((tid & 15) * 65) + (tid >> 4))); 
  fftRadix16Pass(data);               // in-place radix-16 pass 
  twiddleFactorMul(data, tid, 64, 4); // twiddle factor multiplication 
 
  localShuffle(data, sMemx, sMemy, tid, (((tid >> 4) * 64) + (tid & 15))); 
 
  // four radix-4 function calls 
  fftRadix4Pass(data);
  fftRadix4Pass(data + 4); 
  fftRadix4Pass(data + 8);
  fftRadix4Pass(data + 12); 
 
  // coalesced global writes 
  globalStores(data, out, 64); 
}

[edit] See also

• CUDA
• BrookGPU
• Lib Sh
• Intel Larrabee
• Grand Central Dispatch

[edit] References

[edit] External links

• Official site
• OpenCL header files and specification
• OpenCL overview
• OpenCL World
• GPU Modeling and Development in OpenCL
• OpenCL for NVIDIA
• The first public demonstration of OpenCL running on a GPU was done by NVIDIA on December 12, 2008 at Siggraph Asia
• AMD's OpenCL parallel computing demo from Siggraph Asia 2008
• OpenCL: What you need to know – article published in Macworld, August 2008
• HPCWire: OpenCL on the Fast Track
• The Khronos Group Releases OpenCL 1.0 Specification
• Java bindings for OpenCL
• OpenCL-Z: OpenCL Information utility