GPU Programming

3.

Fundamentals of CUDA Programming

3.1.

CUDA Ecosystem and Setup

3.1.1.

CUDA Toolkit Components

3.1.1.1.

3.1.1.2.

Runtime Libraries

3.1.1.3.

Development Tools

3.1.2.

Development Environment Setup

3.1.2.1.

Driver Installation

3.1.2.2.

Toolkit Installation

3.1.2.3.

IDE Integration

3.1.2.4.

Verification and Testing

3.2.

CUDA Programming Model

3.2.1.

Host and Device Concepts

3.2.1.1.

Host Code (CPU)

3.2.1.2.

Device Code (GPU)

3.2.1.3.

Heterogeneous Computing Model

3.2.2.

Kernels and Functions

3.2.2.1.

Kernel Declaration

3.2.2.2.

Device Functions

3.2.2.3.

3.2.2.4.

Function Qualifiers

3.2.3.

Thread Hierarchy

3.2.3.1.

3.2.3.2.

3.2.3.3.

3.2.3.4.

Thread Indexing

3.2.4.

Built-in Variables

3.2.4.1.

3.2.4.2.

3.2.4.3.

3.2.4.4.

3.2.4.5.

3.3.

First CUDA Programs

3.3.1.

Hello World on GPU

3.3.1.1.

Kernel Definition

3.3.1.2.

3.3.1.3.

Compilation Process

3.3.2.

Vector Addition Example

3.3.2.1.

Memory Allocation

3.3.2.2.

3.3.2.3.

Kernel Implementation

3.3.2.4.

Result Verification

3.3.3.

Error Checking Fundamentals

3.3.3.1.

CUDA Error Codes

3.3.3.2.

Error Handling Macros

3.3.3.3.

Debugging Basics

3.4.

Memory Management

3.4.1.

Memory Spaces Overview

3.4.1.1.

3.4.1.2.

3.4.1.3.

Memory Hierarchy

3.4.2.

Basic Memory Operations

3.4.2.1.

3.4.2.2.

3.4.2.3.

3.4.2.4.

Memory Transfer Directions

3.4.3.

3.4.3.1.

cudaMallocManaged

3.4.3.2.

Automatic Data Migration

3.4.3.3.

Page Faulting Mechanism

3.4.3.4.

Performance Considerations

Previous

2. GPU Programming Models and APIs

Go to top

Next

4. Intermediate CUDA Programming