ROCm for Windows Training Course

Introduction

• Understanding ROCm
• Introduction to HIP
• Comparing ROCm, CUDA, and OpenCL
• Overview of ROCm and HIP features and architecture
• Differences between ROCm for Windows and ROCm for Linux

Installation

• Installing ROCm on Windows
• Verifying the installation and checking device compatibility
• Updating or uninstalling ROCm on Windows
• Troubleshooting common installation issues

Getting Started

• Creating a new ROCm project using Visual Studio Code on Windows
• Exploring the project structure and files
• Compiling and running the program
• Displaying output using printf and fprintf

ROCm API

• Using the ROCm API in the host program
• Querying device information and capabilities
• Allocating and deallocating device memory
• Copying data between host and device
• Launching kernels and synchronizing threads
• Handling errors and exceptions

HIP Language

• Using HIP in the device program
• Writing kernels to execute on the GPU and manipulate data
• Utilizing data types, qualifiers, operators, and expressions
• Applying built-in functions, variables, and libraries

ROCm and HIP Memory Model

• Utilizing different memory spaces, such as global, shared, constant, and local
• Working with different memory objects, such as pointers, arrays, textures, and surfaces
• Applying different memory access modes, such as read-only, write-only, and read-write
• Using the memory consistency model and synchronization mechanisms

ROCm and HIP Execution Model

• Utilizing different execution models, such as threads, blocks, and grids
• Using thread functions, such as hipThreadIdx_x, hipBlockIdx_x, and hipBlockDim_x
• Applying block functions, such as __syncthreads and __threadfence_block
• Using grid functions, such as hipGridDim_x and hipGridSync, along with cooperative groups

Debugging

• Debugging ROCm and HIP programs on Windows
• Using the Visual Studio Code debugger to inspect variables, breakpoints, and the call stack
• Using the ROCm Debugger to debug programs on AMD devices
• Using the ROCm Profiler to analyze programs on AMD devices

Optimization

• Optimizing ROCm and HIP programs on Windows
• Using coalescing techniques to improve memory throughput
• Using caching and prefetching techniques to reduce memory latency
• Using shared memory and local memory techniques to optimize memory access and bandwidth
• Using profiling tools to measure and improve execution time and resource utilization

Summary and Next Steps