AI Architecture & Systems

Kernel Programming Models and Execution Fundamentals

2.1.1CPU Thread, GPU Thread, Block, and Grid#

2.1.2Warp/Wavefront Cooperation and Cooperative Groups#

2.1.3CUDA/HIP Memory Spaces and Synchronization Primitives#

2.1.4Kernel Launch, Stream, Event, and Asynchronous Execution#

2.1.5Device Memory, Pinned Memory, and Unified Memory#

2.1.6Race Conditions, Deadlock, Out-of-Bounds Access, and Memory Consistency#