GPU Programming with OpenCL Training Course

Introduction

• Fundamentals of OpenCL
• Comparison: OpenCL vs. CUDA vs. SYCL
• Overview of OpenCL features and architectural design
• Establishing the Development Environment

Initial Steps

• Creating a new OpenCL project in Visual Studio Code
• Examining the project structure and file organization
• Compiling and executing the program
• Displaying output results using printf and fprintf

OpenCL API

• Comprehending the function of the OpenCL API within the host program
• Querying device information and capabilities via the OpenCL API
• Instantiating contexts, command queues, buffers, kernels, and events using the OpenCL API
• Enqueuing commands such as read, write, copy, map, unmap, execute, and wait through the OpenCL API
• Handling errors and exceptions using the OpenCL API

OpenCL C

• Understanding the role of OpenCL C in device-side programming
• Writing OpenCL C kernels that execute on the device and manipulate data
• Using OpenCL C data types, qualifiers, operators, and expressions
• Leveraging OpenCL C built-in functions, including math, geometric, and relational operations
• Utilizing OpenCL C extensions and libraries, such as atomic, image, and cl_khr_fp16

OpenCL Memory Model

• Distinguishing between host and device memory models
• Working with OpenCL memory spaces, including global, local, constant, and private
• Managing OpenCL memory objects such as buffers, images, and pipes
• Applying OpenCL memory access modes like read-only, write-only, and read-write
• Implementing OpenCL memory consistency models and synchronization mechanisms

OpenCL Execution Model

• Differentiating between host and device execution models
• Defining parallelism using OpenCL work-items, work-groups, and ND-ranges
• Utilizing OpenCL work-item functions like get_global_id, get_local_id, and get_group_id
• Applying OpenCL work-group functions such as barrier, work_group_reduce, and work_group_scan
• Using OpenCL device functions including get_num_groups, get_global_size, and get_local_size

Debugging

• Identifying common errors and bugs in OpenCL applications
• Inspecting variables, breakpoints, and call stacks using the Visual Studio Code debugger
• Debugging and analyzing OpenCL programs on AMD devices with CodeXL
• Debugging and analyzing OpenCL programs on Intel devices with Intel VTune
• Debugging and analyzing OpenCL programs on NVIDIA devices with NVIDIA Nsight

Optimization

• Analyzing factors that influence OpenCL program performance
• Enhancing arithmetic throughput using OpenCL vector data types and vectorization techniques
• Reducing control overhead and improving locality via OpenCL loop unrolling and loop tiling
• Optimizing memory access and bandwidth using OpenCL local memory and local memory functions
• Measuring and improving execution time and resource utilization using OpenCL profiling and profiling tools