Guide CUDA et Programmation GPU

VérifiéSûr

Bonnes pratiques, motifs de conception et garde-fous pour la programmation CUDA et GPU. Couvre la hiérarchie mémoire, l'optimisation d'occupation et les patterns de kernels pour le calcul haute performance.

Spar Skills Guide Bot
DeveloppementIntermédiaire
4002/06/2026
Claude CodeCursorWindsurfCopilotCodex
#cuda#gpu-programming#kernel-design#memory-hierarchy

Recommandé pour

Notre avis

Cette compétence fournit des garde-fous complets, des modèles et des bonnes pratiques pour la programmation CUDA/GPU, en mettant l'accent sur la hiérarchie mémoire, la conception de noyaux et la vérification des erreurs.

Points forts

  • Met l'accent sur l'accès mémoire coalescé pour optimiser la bande passante
  • Fournit des macros de vérification d'erreurs réutilisables et robustes
  • Couvre à la fois les bonnes pratiques et les pièges courants en parallélisme GPU

Limites

  • Peut être verbeux pour des noyaux simples
  • Nécessite CUDA 11+ et ne couvre pas les versions antérieures
  • Ne traite pas des bibliothèques spécifiques comme cuBLAS ou cuDNN
Quand l'utiliser

Utilisez cette compétence lors du développement ou de la révision de noyaux CUDA pour des applications GPU critiques en performance.

Quand l'éviter

Évitez pour un prototypage rapide sans souci de performance optimale ou pour des projets ne ciblant pas les GPU NVIDIA.

Analyse de sécurité

Sûr
Score qualité90/100

The skill provides CUDA programming guidelines, macros, and best practices with no executable commands, destructive actions, or data exfiltration risks. It is purely instructional.

Aucun point d'attention détecté

Exemples

Add error checking to CUDA kernel
Add CUDA error checking to this kernel launch and memory operations. Use the recommended macro pattern from the CUDA guide.
Optimize memory access pattern
Analyze the memory access pattern in this CUDA kernel and suggest changes to ensure coalesced global memory access. Follow the memory hierarchy guidelines.
Kernel design review
Review this CUDA kernel for occupancy and grid-stride loops. Explain how to set block dimensions and grid size following best practices.

name: cuda-guide description: | CUDA/GPU computing guardrails, patterns, and best practices for AI-assisted development. Use when working with CUDA files (.cu, .cuh), or when the user mentions CUDA/GPU programming. Provides kernel design patterns, memory hierarchy guidelines, and occupancy optimization specific to this project's coding standards. license: MIT metadata: author: samuel version: "1.0" category: language language: cuda extensions: ".cu,.cuh"

CUDA Guide

Applies to: CUDA 11+, GPU Computing, Deep Learning, Scientific Computing, HPC

Core Principles

  1. Parallelism First: Design algorithms for thousands of concurrent threads; serial thinking is the primary enemy of GPU performance
  2. Memory Hierarchy Awareness: Global memory is 100x slower than shared memory and 1000x slower than registers; every kernel design starts with memory access planning
  3. Coalesced Access: Adjacent threads must access adjacent memory addresses; a single misaligned access pattern can reduce bandwidth by 32x
  4. Occupancy Over Cleverness: Maximize active warps per SM by managing register count, shared memory usage, and block dimensions together
  5. Minimize Host-Device Transfers: PCIe bandwidth is the bottleneck; overlap transfers with computation using streams and pinned memory

Guardrails

Error Checking

  • ALWAYS check CUDA API return values with a macro wrapper
  • ALWAYS call cudaGetLastError() after every kernel launch
  • ALWAYS call cudaDeviceSynchronize() before reading kernel results on the host
  • Use compute-sanitizer (successor to cuda-memcheck) in development builds
  • Handle cudaErrorMemoryAllocation gracefully; never assume GPU memory is infinite
#define CUDA_CHECK(call)                                                    \
    do {                                                                    \
        cudaError_t err = call;                                            \
        if (err != cudaSuccess) {                                          \
            fprintf(stderr, "CUDA error at %s:%d: %s\n",                   \
                    __FILE__, __LINE__, cudaGetErrorString(err));           \
            exit(EXIT_FAILURE);                                            \
        }                                                                   \
    } while (0)

#define CUDA_CHECK_KERNEL()                                                 \
    do {                                                                    \
        cudaError_t err = cudaGetLastError();                              \
        if (err != cudaSuccess) {                                          \
            fprintf(stderr, "Kernel launch error at %s:%d: %s\n",          \
                    __FILE__, __LINE__, cudaGetErrorString(err));           \
            exit(EXIT_FAILURE);                                            \
        }                                                                   \
    } while (0)

Memory Management

  • Pair every cudaMalloc with a cudaFree; prefer RAII wrappers in C++ host code
  • Use cudaMallocManaged (Unified Memory) for prototyping; switch to explicit transfers for production
  • Use cudaMallocHost (pinned memory) when streaming data to the GPU; pageable memory cannot overlap with compute
  • Prefer cudaMemcpyAsync with streams over synchronous cudaMemcpy
  • Never access device pointers from host code or host pointers from device code (except Unified Memory)
  • Call cudaMemset or cudaMemsetAsync to zero-initialize device buffers

Kernel Design

  • Block size must be a multiple of warp size (32); prefer 128, 256, or 512
  • Calculate grid size as (n + block_size - 1) / block_size
  • Always include bounds checking: if (idx < n) at the top of every kernel
  • Use grid-stride loops for kernels that must handle arbitrary data sizes
  • Document thread mapping: which dimension maps to which data axis
  • Mark device-only helpers as __device__, host+device as __host__ __device__
// Grid-stride loop: works with any grid size, any data size
__global__ void saxpy(float a, const float* x, float* y, int n) {
    for (int i = blockIdx.x * blockDim.x + threadIdx.x;
         i < n;
         i += blockDim.x * gridDim.x) {
        y[i] = a * x[i] + y[i];
    }
}

Synchronization

  • Use __syncthreads() after every shared memory write before any thread reads another thread's value
  • Never place __syncthreads() inside a conditional branch that not all threads in a block will reach (deadlock)
  • Use __syncwarp() (CUDA 9+) for warp-level synchronization instead of relying on implicit warp-synchronous execution
  • Use cudaDeviceSynchronize() sparingly in production; prefer stream synchronization with cudaStreamSynchronize()
  • Use CUDA events (cudaEventRecord / cudaEventSynchronize) for fine-grained inter-stream ordering

Performance

  • Profile before optimizing: use Nsight Compute for kernel analysis, Nsight Systems for system-level view
  • Target >50% theoretical occupancy; use the CUDA Occupancy Calculator to tune block dimensions
  • Aim for >60% of peak memory bandwidth in memory-bound kernels
  • Avoid warp divergence: ensure threads within a warp take the same branch when possible
  • Prefer float over double on consumer GPUs (2x throughput difference)
  • Minimize atomic operations on global memory; use shared memory atomics with a final reduction

Memory Hierarchy

Understanding the memory hierarchy is the single most important factor in CUDA performance.

| Memory Type | Scope | Latency (cycles) | Size | Cached | Read/Write | |-------------|-------|-------------------|------|--------|------------| | Registers | Thread | 1 | ~255 per thread | N/A | R/W | | Shared | Block | ~5 | 48-164 KB per SM | N/A | R/W | | L1 Cache | SM | ~28 | 48-192 KB per SM | Auto | R | | L2 Cache | Device | ~200 | 4-40 MB | Auto | R/W | | Global | Device | ~400-600 | 4-80 GB (HBM/GDDR) | Yes | R/W | | Constant | Device | ~5 (cached) | 64 KB | Yes (broadcast) | R | | Texture | Device | ~400 (cached) | Global pool | Yes (spatial) | R |

Decision guide:

  • Data reused within a thread -> registers (automatic via local variables)
  • Data shared across threads in a block -> __shared__ memory
  • Read-only data broadcast to all threads -> __constant__ memory
  • Large read-only data with spatial locality -> texture memory
  • Everything else -> global memory with coalesced access patterns

Key Patterns

Kernel Launch Configuration

// Query device for optimal configuration
void launch_optimized(const float* input, float* output, int n) {
    int block_size;
    int min_grid_size;

    // Let the runtime suggest optimal block size for maximum occupancy
    cudaOccupancyMaxPotentialBlockSize(
        &min_grid_size, &block_size, my_kernel, 0, n);

    int grid_size = (n + block_size - 1) / block_size;
    my_kernel<<<grid_size, block_size>>>(input, output, n);
    CUDA_CHECK_KERNEL();
}

Coalesced Memory Access

// BAD: Strided access -- adjacent threads access non-adjacent memory
// Each warp issues 32 separate memory transactions
__global__ void transpose_naive(const float* in, float* out, int W, int H) {
    int x = blockIdx.x * blockDim.x + threadIdx.x;
    int y = blockIdx.y * blockDim.y + threadIdx.y;
    if (x < W && y < H) {
        out[x * H + y] = in[y * W + x];  // Write is strided
    }
}

// GOOD: Use shared memory to coalesce both reads and writes
__global__ void transpose_coalesced(
    const float* in, float* out, int W, int H
) {
    __shared__ float tile[32][33]; // +1 padding avoids bank conflicts

    int x = blockIdx.x * 32 + threadIdx.x;
    int y = blockIdx.y * 32 + threadIdx.y;

    if (x < W && y < H) {
        tile[threadIdx.y][threadIdx.x] = in[y * W + x]; // Coalesced read
    }
    __syncthreads();

    x = blockIdx.y * 32 + threadIdx.x;
    y = blockIdx.x * 32 + threadIdx.y;

    if (x < H && y < W) {
        out[y * H + x] = tile[threadIdx.x][threadIdx.y]; // Coalesced write
    }
}

Shared Memory Tiling

// Dot product of two vectors using shared memory reduction
__global__ void dot_product(
    const float* a, const float* b, float* result, int n
) {
    __shared__ float cache[256];

    int tid = threadIdx.x;
    int idx = blockIdx.x * blockDim.x + threadIdx.x;

    // Each thread computes its partial sum via grid-stride
    float partial = 0.0f;
    for (int i = idx; i < n; i += blockDim.x * gridDim.x) {
        partial += a[i] * b[i];
    }
    cache[tid] = partial;
    __syncthreads();

    // Tree reduction in shared memory
    for (int s = blockDim.x / 2; s > 0; s >>= 1) {
        if (tid < s) {
            cache[tid] += cache[tid + s];
        }
        __syncthreads();
    }

    if (tid == 0) {
        atomicAdd(result, cache[0]);
    }
}

Warp-Level Primitives (CUDA 9+)

// Warp-level reduction using shuffle instructions -- no shared memory needed
__device__ float warp_reduce_sum(float val) {
    for (int offset = warpSize / 2; offset > 0; offset /= 2) {
        val += __shfl_down_sync(0xFFFFFFFF, val, offset);
    }
    return val;
}

// Block-level reduction combining warp shuffles and shared memory
__device__ float block_reduce_sum(float val) {
    __shared__ float warp_sums[32]; // One slot per warp (max 32 warps/block)

    int lane = threadIdx.x % warpSize;
    int warp_id = threadIdx.x / warpSize;

    val = warp_reduce_sum(val);

    if (lane == 0) {
        warp_sums[warp_id] = val;
    }
    __syncthreads();

    // First warp reduces the warp sums
    int num_warps = (blockDim.x + warpSize - 1) / warpSize;
    val = (threadIdx.x < num_warps) ? warp_sums[threadIdx.x] : 0.0f;
    if (warp_id == 0) {
        val = warp_reduce_sum(val);
    }

    return val;
}

Performance

Occupancy Calculator

// Query occupancy at compile time for tuning
void report_occupancy() {
    int block_size = 256;
    int num_blocks;

    cudaOccupancyMaxActiveBlocksPerMultiprocessor(
        &num_blocks, my_kernel, block_size, 0);

    cudaDeviceProp prop;
    cudaGetDeviceProperties(&prop, 0);

    int active_warps = num_blocks * (block_size / prop.warpSize);
    int max_warps = prop.maxThreadsPerMultiProcessor / prop.warpSize;
    float occupancy = (float)active_warps / max_warps;

    printf("Occupancy: %.1f%% (%d/%d warps)\n",
           occupancy * 100, active_warps, max_warps);
}

Nsight Profiling Workflow

# System-level trace: find CPU/GPU idle gaps, stream concurrency
nsys profile -o trace ./program
nsys stats trace.nsys-rep

# Kernel-level analysis: roofline, memory throughput, occupancy
ncu --set full -o kernel_report ./program
ncu -i kernel_report.ncu-rep    # Open in Nsight Compute GUI

# Quick single-metric check
ncu --metrics sm__throughput.avg.pct_of_peak_sustained_elapsed ./program

Memory Bandwidth Measurement

// Measure effective bandwidth of a kernel
void measure_bandwidth(int n) {
    size_t bytes = 2 * n * sizeof(float); // Read A + Write B

    cudaEvent_t start, stop;
    CUDA_CHECK(cudaEventCreate(&start));
    CUDA_CHECK(cudaEventCreate(&stop));

    CUDA_CHECK(cudaEventRecord(start));
    copy_kernel<<<grid, block>>>(d_in, d_out, n);
    CUDA_CHECK(cudaEventRecord(stop));
    CUDA_CHECK(cudaEventSynchronize(stop));

    float ms = 0;
    CUDA_CHECK(cudaEventElapsedTime(&ms, start, stop));

    float gb_per_sec = bytes / (ms * 1e6);
    printf("Effective bandwidth: %.2f GB/s\n", gb_per_sec);

    CUDA_CHECK(cudaEventDestroy(start));
    CUDA_CHECK(cudaEventDestroy(stop));
}

Tooling

Essential Commands

# Compile CUDA code
nvcc -arch=sm_80 -O3 -o program main.cu         # Single file
nvcc -arch=native -lineinfo -o program main.cu   # With debug line info

# CMake build
cmake -B build -DCMAKE_CUDA_ARCHITECTURES="70;80;86"
cmake --build build -j$(nproc)

# Runtime debugging
compute-sanitizer ./program                      # Memory errors (replaces cuda-memcheck)
compute-sanitizer --tool racecheck ./program     # Shared memory race conditions
compute-sanitizer --tool initcheck ./program     # Uninitialized device memory reads
compute-sanitizer --tool synccheck ./program     # Synchronization errors

# Profiling
nsys profile ./program                           # System-level timeline
ncu ./program                                    # Kernel-level metrics
ncu --kernel-name my_kernel --launch-skip 2 --launch-count 1 ./program

# Device info
nvidia-smi                                       # GPU status and memory usage
nvcc --version                                   # CUDA compiler version

CMakeLists.txt Template

cmake_minimum_required(VERSION 3.18)
project(myproject LANGUAGES CXX CUDA)

set(CMAKE_CXX_STANDARD 17)
set(CMAKE_CUDA_STANDARD 17)
set(CMAKE_CUDA_ARCHITECTURES 70 80 86)
set(CMAKE_CUDA_SEPARABLE_COMPILATION ON)

find_package(CUDAToolkit REQUIRED)

add_library(kernels src/kernels.cu)
target_include_directories(kernels PUBLIC include)

add_executable(main src/main.cpp)
target_link_libraries(main kernels CUDA::cudart)

enable_testing()
add_executable(tests tests/test_kernels.cu)
target_link_libraries(tests kernels CUDA::cudart)
add_test(NAME gpu_tests COMMAND tests)

References

For detailed patterns and examples, see:

External References

Skills similaires