CUDA 13.4 samples update - v13.4-public

Release 13.4 of the CUDA Samples supported by CUDA Toolkit 13.4.
See Changelog for more information.
This commit is contained in:
Dheemanth
2026-09-09 15:07:08 -07:00
committed by GitHub
parent b7c5481c55
commit 5443602d89
871 changed files with 6843 additions and 321849 deletions

View File

@@ -1,4 +1,4 @@
/* Copyright (c) 2022, NVIDIA CORPORATION. All rights reserved.
/* Copyright (c) 2026, NVIDIA CORPORATION. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
@@ -26,46 +26,54 @@
*/
/*
* This application demonstrates how to use the CUDA API to use multiple GPUs,
* with an emphasis on simple illustration of the techniques (not on
* performance).
* Sums a large vector across every GPU in the system.
*
* Note that in order to detect multiple GPUs in your system you have to disable
* SLI in the nvidia control panel. Otherwise only one GPU is visible to the
* application. On the other side, you can still extend your desktop to screens
* attached to both GPUs.
* The input is split into one slice per GPU, each with its own stream, so the H2D copy,
* reduction kernel, and D2H copy run concurrently across devices. cub::BlockReduce reduces
* within each thread block; the host then adds the per-block values into a per-GPU total and
* combines those into the final sum, checked against a CPU reference. CUDA events time the
* GPU phase.
*/
// System includes
#include <assert.h>
#include <math.h>
#include <stdio.h>
#include <stdlib.h>
#include <vector>
// CUDA runtime
#include <cuda_runtime.h>
// helper functions and utilities to work with CUDA
#include <helper_cuda.h>
#include <helper_functions.h>
// CCCL / CUB block-wide reduction
#include <cub/cub.cuh>
#ifndef MAX
#define MAX(a, b) (a > b ? a : b)
#endif
#include "simpleMultiGPU.h"
////////////////////////////////////////////////////////////////////////////////
// Data configuration
////////////////////////////////////////////////////////////////////////////////
const int MAX_GPU_COUNT = 32;
const int DATA_N = 1048576 * 32;
////////////////////////////////////////////////////////////////////////////////
// Simple reduction kernel.
// Refer to the 'reduction' CUDA Sample describing
// reduction optimization strategies
////////////////////////////////////////////////////////////////////////////////
__global__ static void reduceKernel(float *d_Result, float *d_Input, int N)
// Everything one GPU needs: its slice of the input, its buffers, and its stream
struct TGPUplan
{
int dataN; // elements assigned to this GPU
float *h_Data; // pinned input slice
float *d_Data; // the slice, on the device
float *d_Sum; // one partial sum per block
float *h_Sum_from_device; // those partial sums, copied back
cudaStream_t stream; // orders this GPU's copies and kernel
};
// Data configuration
constexpr int DATA_N = 1048576 * 32;
// Reduction launch configuration
constexpr int BLOCK_N = 32; // thread blocks launched per GPU
constexpr int THREAD_N = 256; // threads per block
// Per-block reduction kernel.
// Threads walk the input in a grid-stride loop, then cub::BlockReduce combines their totals.
// Thread 0 writes one partial sum per block (BLOCK_N values total); the host
// adds those up to obtain this GPU's final sum.
__global__ static void reduceKernel(float *d_Result, const float *d_Input, int N)
{
using BlockReduce = cub::BlockReduce<float, THREAD_N>;
__shared__ typename BlockReduce::TempStorage temp_storage;
const int tid = blockIdx.x * blockDim.x + threadIdx.x;
const int threadN = gridDim.x * blockDim.x;
float sum = 0;
@@ -73,160 +81,145 @@ __global__ static void reduceKernel(float *d_Result, float *d_Input, int N)
for (int pos = tid; pos < N; pos += threadN)
sum += d_Input[pos];
d_Result[tid] = sum;
// Reduce the per-thread partial sums across the block (result valid in thread 0)
sum = BlockReduce(temp_storage).Sum(sum);
if (threadIdx.x == 0)
d_Result[blockIdx.x] = sum;
}
////////////////////////////////////////////////////////////////////////////////
// Program main
////////////////////////////////////////////////////////////////////////////////
int main(int argc, char **argv)
int main()
{
// Solver config
TGPUplan plan[MAX_GPU_COUNT];
// GPU reduction results
float h_SumGPU[MAX_GPU_COUNT];
float sumGPU;
double sumCPU, diff;
int i, j, gpuBase, GPU_N;
const int BLOCK_N = 32;
const int THREAD_N = 256;
const int ACCUM_N = BLOCK_N * THREAD_N;
printf("Starting simpleMultiGPU\n");
checkCudaErrors(cudaGetDeviceCount(&GPU_N));
if (GPU_N > MAX_GPU_COUNT) {
GPU_N = MAX_GPU_COUNT;
int deviceCount = 0;
cudaGetDeviceCount(&deviceCount);
printf("CUDA-capable device count: %i\n", deviceCount);
// This sample needs at least two GPUs to show work running on several devices at once
if (deviceCount < 2) {
printf("Two or more CUDA-capable devices are required. Exiting.\n");
return EXIT_SUCCESS;
}
printf("CUDA-capable device count: %i\n", GPU_N);
printf("Generating input data...\n\n");
std::vector<TGPUplan> plan(deviceCount);
std::vector<float> h_SumGPU(deviceCount); // one result per GPU
// Subdividing input data across GPUs
// Get data sizes for each GPU
for (i = 0; i < GPU_N; i++) {
plan[i].dataN = DATA_N / GPU_N;
for (int i = 0; i < deviceCount; i++) {
plan[i].dataN = DATA_N / deviceCount;
}
// Take into account "odd" data sizes
for (i = 0; i < DATA_N % GPU_N; i++) {
for (int i = 0; i < DATA_N % deviceCount; i++) {
plan[i].dataN++;
}
// Assign data ranges to GPUs
gpuBase = 0;
for (i = 0; i < GPU_N; i++) {
plan[i].h_Sum = h_SumGPU + i;
gpuBase += plan[i].dataN;
}
// Create streams for issuing GPU command asynchronously and allocate memory
// (GPU and System page-locked)
for (i = 0; i < GPU_N; i++) {
checkCudaErrors(cudaSetDevice(i));
checkCudaErrors(cudaStreamCreate(&plan[i].stream));
// Allocate memory
checkCudaErrors(cudaMalloc((void **)&plan[i].d_Data, plan[i].dataN * sizeof(float)));
checkCudaErrors(cudaMalloc((void **)&plan[i].d_Sum, ACCUM_N * sizeof(float)));
checkCudaErrors(cudaMallocHost((void **)&plan[i].h_Sum_from_device, ACCUM_N * sizeof(float)));
checkCudaErrors(cudaMallocHost((void **)&plan[i].h_Data, plan[i].dataN * sizeof(float)));
for (int i = 0; i < deviceCount; i++) {
cudaSetDevice(i);
cudaStreamCreate(&plan[i].stream);
for (j = 0; j < plan[i].dataN; j++) {
cudaMalloc(&plan[i].d_Data, plan[i].dataN * sizeof(float));
cudaMalloc(&plan[i].d_Sum, BLOCK_N * sizeof(float));
cudaMallocHost(&plan[i].h_Sum_from_device, BLOCK_N * sizeof(float));
cudaMallocHost(&plan[i].h_Data, plan[i].dataN * sizeof(float));
for (int j = 0; j < plan[i].dataN; j++) {
plan[i].h_Data[j] = (float)rand() / (float)RAND_MAX;
}
}
// Start timing and compute on GPU(s)
printf("Computing with %d GPUs...\n", GPU_N);
// create and start timer
StopWatchInterface *timer = NULL;
sdkCreateTimer(&timer);
printf("Computing with %d GPUs...\n", deviceCount);
// Time the multi-GPU work with CUDA events (recorded on device 0)
cudaEvent_t start, stop;
cudaSetDevice(0);
cudaEventCreate(&start);
cudaEventCreate(&stop);
cudaEventRecord(start);
// start the timer
sdkStartTimer(&timer);
// Copy data to GPU, launch the kernel and copy data back. All asynchronously
for (i = 0; i < GPU_N; i++) {
// Set device
checkCudaErrors(cudaSetDevice(i));
// Queue every GPU's copy-in, kernel and copy-out before waiting on any of them,
// so the devices run concurrently instead of one after another
for (int i = 0; i < deviceCount; i++) {
TGPUplan &gpu = plan[i];
cudaSetDevice(i);
// Copy input data from CPU
checkCudaErrors(cudaMemcpyAsync(
plan[i].d_Data, plan[i].h_Data, plan[i].dataN * sizeof(float), cudaMemcpyHostToDevice, plan[i].stream));
cudaMemcpyAsync(gpu.d_Data, gpu.h_Data, gpu.dataN * sizeof(float), cudaMemcpyHostToDevice, gpu.stream);
// Perform GPU computations
reduceKernel<<<BLOCK_N, THREAD_N, 0, plan[i].stream>>>(plan[i].d_Sum, plan[i].d_Data, plan[i].dataN);
getLastCudaError("reduceKernel() execution failed.\n");
// Launch Kernel
reduceKernel<<<BLOCK_N, THREAD_N, 0, gpu.stream>>>(gpu.d_Sum, gpu.d_Data, gpu.dataN);
// Read back GPU results
checkCudaErrors(cudaMemcpyAsync(
plan[i].h_Sum_from_device, plan[i].d_Sum, ACCUM_N * sizeof(float), cudaMemcpyDeviceToHost, plan[i].stream));
cudaMemcpyAsync(gpu.h_Sum_from_device, gpu.d_Sum, BLOCK_N * sizeof(float), cudaMemcpyDeviceToHost, gpu.stream);
}
// Process GPU results
for (i = 0; i < GPU_N; i++) {
float sum;
// Set device
checkCudaErrors(cudaSetDevice(i));
for (int i = 0; i < deviceCount; i++) {
cudaSetDevice(i);
// Wait for all operations to finish
cudaStreamSynchronize(plan[i].stream);
// Finalize GPU reduction for current subvector
sum = 0;
float sum = 0;
for (j = 0; j < ACCUM_N; j++) {
for (int j = 0; j < BLOCK_N; j++) {
sum += plan[i].h_Sum_from_device[j];
}
*(plan[i].h_Sum) = (float)sum;
h_SumGPU[i] = sum;
// Shut down this GPU
checkCudaErrors(cudaFreeHost(plan[i].h_Sum_from_device));
checkCudaErrors(cudaFree(plan[i].d_Sum));
checkCudaErrors(cudaFree(plan[i].d_Data));
checkCudaErrors(cudaStreamDestroy(plan[i].stream));
// Free up this GPU's resources. h_Data stays alive for the CPU check below.
cudaFreeHost(plan[i].h_Sum_from_device);
cudaFree(plan[i].d_Sum);
cudaFree(plan[i].d_Data);
cudaStreamDestroy(plan[i].stream);
}
sumGPU = 0;
float sumGPU = 0;
for (i = 0; i < GPU_N; i++) {
for (int i = 0; i < deviceCount; i++) {
sumGPU += h_SumGPU[i];
}
sdkStopTimer(&timer);
printf(" GPU Processing time: %f (ms)\n\n", sdkGetTimerValue(&timer));
sdkDeleteTimer(&timer);
cudaSetDevice(0);
cudaEventRecord(stop);
cudaEventSynchronize(stop);
float gpuTime = 0;
cudaEventElapsedTime(&gpuTime, start, stop);
printf(" GPU Processing time: %f (ms)\n\n", gpuTime);
cudaEventDestroy(start);
cudaEventDestroy(stop);
// Compute on Host CPU
printf("Computing with Host CPU...\n\n");
sumCPU = 0;
double sumCPU = 0;
for (i = 0; i < GPU_N; i++) {
for (j = 0; j < plan[i].dataN; j++) {
for (int i = 0; i < deviceCount; i++) {
for (int j = 0; j < plan[i].dataN; j++) {
sumCPU += plan[i].h_Data[j];
}
}
// Compare GPU and CPU results
printf("Comparing GPU and Host CPU results...\n");
diff = fabs(sumCPU - sumGPU) / fabs(sumCPU);
const double diff = fabs(sumCPU - sumGPU) / fabs(sumCPU);
printf(" GPU sum: %f\n CPU sum: %f\n", sumGPU, sumCPU);
printf(" Relative difference: %E \n\n", diff);
// Cleanup and shutdown
for (i = 0; i < GPU_N; i++) {
checkCudaErrors(cudaSetDevice(i));
checkCudaErrors(cudaFreeHost(plan[i].h_Data));
for (int i = 0; i < deviceCount; i++) {
cudaFreeHost(plan[i].h_Data);
}
exit((diff < 1e-5) ? EXIT_SUCCESS : EXIT_FAILURE);
return (diff < 1e-5) ? EXIT_SUCCESS : EXIT_FAILURE;
}