opencl

NVIDIA offers GPUDirect to reduce memory transfer overheads. I'm wondering if there is a similar concept for AMD/ATI? Specifically: 1) Do AMD GPUs avoid the second memory transfer when interfacing with network cards, as described here . In case the graphic is lost at some point, here is a description of the impact of GPUDirect on getting data from a GPU on one machine to be transferred across a network interface: With GPUDirect, GPU memory goes to Host memory then straight to the network interface card. Without GPUDirect, GPU memory goes to Host memory in one address space, then the CPU has to

I have read some papers talking about "persistent threads" for GPGPU, but I don't really understand it. Can any one give me an example or show me the use of this programming fashion? What I keep in my mind after reading and googling "persistent threads": Presistent Threads it's no more than a while loop that keep thread running and computing a lot of bunch of works. Is this correct? Thanks in advance Reference: http://www.idav.ucdavis.edu/publications/print_pub?pub_id=1089 http://developer.download.nvidia.com/GTC/PDF/GTC2012/PresentationPDF/S0157-GTC2012-Persistent-Threads-Computing.pdf CUDA

I'm new to C, C++ and OpenCL and doing my best to learn them at the moment. Here's a preexisting C++ function that I'm trying to figure out how to port to OpenCL using either the C or C++ bindings. #include <vector> using namespace std; class Test { private: double a; vector<double> b; vector<long> c; vector<vector<double> > d; public: double foo(long x, double y) { // mathematical operations // using x, y, a, b, c, d // and also b.size() // to calculate return value return 0.0; } }; Broadly my question is how to pass in all the class members that this function accesses into the binding and

Is it possible to define my own functions in OpenCL code, in order that the kernels could call them? It yes, where can I see some simple example? Kayhano Function used to create program is ... cl_program clCreateProgramWithSource ( cl_context context, cl_uint count, const char **strings, const size_t *lengths, cl_int *errcode_ret) You can place functions inside the strings parameter like this, float AddVector(float a, float b) { return a + b; } kernel void VectorAdd( global read_only float* a, global read_only float* b, global write_only float* c ) { int index = get_global_id(0); //c[index] =

Consider a simple example: vector addition. If I build a program for CL_DEVICE_TYPE_GPU, and I build the same program for CL_DEVICE_TYPE_CPU, what is the difference between them(except that "CPU program" is running on CPU, and "GPU program" is running on GPU)? Thanks for your help. There are a few differences between the device types. The simple answer to your vector question is: Use a gpu for large vectors, and cpu for smaller workloads. 1) Memory copying. GPUs rely on the data you are working on to be passed into them, and the results are later read back to the host. This is done over PCI-e,

As I know, GPUs switch between warps to hide the memory latency. But I wonder in which condition, a warp will be switched out? For example, if a warp perform a load, and the data is there in the cache already. So is the warp switched out or continue the next computation? What happens if there are two consecutive adds? Thanks First of all, once a thread block is launched on a multiprocessor (SM), all of its warps are resident until they all exit the kernel. Thus a block is not launched until there are sufficient registers for all warps of the block, and until there is enough free shared memory