cuda

CUDA

Overview

CUDA is NVIDIA's parallel computing platform and programming model for GPU-accelerated applications. It provides direct access to the GPU's virtual instruction set and parallel compute elements for executing kernels in C, C++, and Fortran.

cuda-samples version: v13.3 (CUDA Toolkit 13.3) CUDALibrarySamples: main (July 2026) Language: C/C++ (.cu files) Licenses: BSD-3-Clause (cuda-samples), Apache-2.0 (CUDALibrarySamples)

v13.3 layout note: cuda-samples renamed its top-level Samples/ directory to cpp/ (Python samples now live under python/). C++ samples are grouped under cpp/0_Introduction, cpp/4_CUDA_Libraries, cpp/6_Performance, plus new cpp/9_CUDA_Tile (CUDA Tile C++) and CCCL 3.3 feature samples (cubDeviceFind, cubDeviceSegmentedScan, cubDeviceTransform, libcuxxRandom, libcuxxMdspan) under cpp/4_CUDA_Libraries/.

Quick Start

// Minimal kernel + launch
__global__ void addVectors(float *a, float *b, float *c, int n) {
    int i = blockIdx.x * blockDim.x + threadIdx.x;
    if (i < n) c[i] = a[i] + b[i];
}

int main() {
    int n = 1 << 20;
    float *d_a, *d_b, *d_c;
    cudaMalloc(&d_a, n * sizeof(float));
    cudaMalloc(&d_b, n * sizeof(float));
    cudaMalloc(&d_c, n * sizeof(float));

    int threads = 256;
    int blocks = (n + threads - 1) / threads;
    addVectors<<<blocks, threads>>>(d_a, d_b, d_c, n);
    cudaDeviceSynchronize();

    cudaFree(d_a); cudaFree(d_b); cudaFree(d_c);
    return 0;
}

Core Concepts

• Kernel: __global__ function executed on GPU by many parallel threads
• Grid/Block/Thread: Launch hierarchy — <<<gridDim, blockDim>>> configures parallelism
• Device memory: Must be explicitly allocated with cudaMalloc and freed with cudaFree
• Streams: Async execution queues; default stream is synchronous with host
• Unified Memory (cudaMallocManaged): Automatically migrates data between CPU and GPU; check support with device_prop.managedMemory