Kit de desenvolvimento Intel® para aplicativos OpenCL™

Writing to a Shared Resource

According to the OpenCL™ specification, you need to ensure that the commands that change the content of a shared memory object, complete in the previous command queue before the memory object is used by commands, executed in another command-queue. One way to achieve this is using events:

Use Row-Wise Data Accesses

OpenCL™ enables you to submit kernels on one-, two- or three-dimensional index space. Consider using one-dimensional ranges for cache locality and to save index computations.

If a two- or three-dimensional range naturally fits your data dimensions, try to keep work-items scanning along rows, not columns. For example:

OpenCL™ Standard

OpenCL™ (Open Computing Language) standard provides a uniform programming environment for software developers to write portable general-purpose parallel code for high-performance computing servers, client computer systems, and other computing systems. OpenCL is developed by multiple companies through the Khronos* OpenCL committee, and Intel is a key contributor to the OpenCL standard since its inception.

Intel OpenCL implementation targets Intel® Xeon® processors and Intel Xeon Phi™ coprocessors.

Assine o Kit de desenvolvimento Intel® para aplicativos OpenCL™