以上回答指出了块大小如何影响性能，并建议基于占用率最大化的常见启发式方法来选择块大小。并不想提供选择块大小的标准，值得一提的是，CUDA 6.5（现在是发布候选版本）包括几个新的运行时函数来帮助计算占用率和启动配置，请参见：
CUDA专业提示：占用API简化启动配置
其中一个有用的函数是
cudaOccupancyMaxPotentialBlockSize，它启发式地计算实现最大占用率的块大小。该函数提供的值可以作为启动参数手动优化的起点。下面是一个小例子。
#include <stdio.h>
__global__ void MyKernel(int *a, int *b, int *c, int N)
{
int idx = threadIdx.x + blockIdx.x * blockDim.x;
if (idx < N) { c[idx] = a[idx] + b[idx]; }
}
void main()
{
const int N = 1000000;
int blockSize;
int minGridSize;
int gridSize;
int* h_vec1 = (int*) malloc(N*sizeof(int));
int* h_vec2 = (int*) malloc(N*sizeof(int));
int* h_vec3 = (int*) malloc(N*sizeof(int));
int* h_vec4 = (int*) malloc(N*sizeof(int));
int* d_vec1; cudaMalloc((void**)&d_vec1, N*sizeof(int));
int* d_vec2; cudaMalloc((void**)&d_vec2, N*sizeof(int));
int* d_vec3; cudaMalloc((void**)&d_vec3, N*sizeof(int));
for (int i=0; i<N; i++) {
h_vec1[i] = 10;
h_vec2[i] = 20;
h_vec4[i] = h_vec1[i] + h_vec2[i];
}
cudaMemcpy(d_vec1, h_vec1, N*sizeof(int), cudaMemcpyHostToDevice);
cudaMemcpy(d_vec2, h_vec2, N*sizeof(int), cudaMemcpyHostToDevice);
float time;
cudaEvent_t start, stop;
cudaEventCreate(&start);
cudaEventCreate(&stop);
cudaEventRecord(start, 0);
cudaOccupancyMaxPotentialBlockSize(&minGridSize, &blockSize, MyKernel, 0, N);
gridSize = (N + blockSize - 1) / blockSize;
cudaEventRecord(stop, 0);
cudaEventSynchronize(stop);
cudaEventElapsedTime(&time, start, stop);
printf("Occupancy calculator elapsed time: %3.3f ms \n", time);
cudaEventRecord(start, 0);
MyKernel<<<gridSize, blockSize>>>(d_vec1, d_vec2, d_vec3, N);
cudaEventRecord(stop, 0);
cudaEventSynchronize(stop);
cudaEventElapsedTime(&time, start, stop);
printf("Kernel elapsed time: %3.3f ms \n", time);
printf("Blocksize %i\n", blockSize);
cudaMemcpy(h_vec3, d_vec3, N*sizeof(int), cudaMemcpyDeviceToHost);
for (int i=0; i<N; i++) {
if (h_vec3[i] != h_vec4[i]) { printf("Error at i = %i! Host = %i; Device = %i\n", i, h_vec4[i], h_vec3[i]); return; };
}
printf("Test passed\n");
}
编辑
cudaOccupancyMaxPotentialBlockSize 定义在 cuda_runtime.h 文件中，定义如下：
template<class T>
__inline__ __host__ CUDART_DEVICE cudaError_t cudaOccupancyMaxPotentialBlockSize(
int *minGridSize,
int *blockSize,
T func,
size_t dynamicSMemSize = 0,
int blockSizeLimit = 0)
{
return cudaOccupancyMaxPotentialBlockSizeVariableSMem(minGridSize, blockSize, func, __cudaOccupancyB2DHelper(dynamicSMemSize), blockSizeLimit);
}
参数的含义如下。
minGridSize = Suggested min grid size to achieve a full machine launch.
blockSize = Suggested block size to achieve maximum occupancy.
func = Kernel function.
dynamicSMemSize = Size of dynamically allocated shared memory. Of course, it is known at runtime before any kernel launch. The size of the statically allocated shared memory is not needed as it is inferred by the properties of func.
blockSizeLimit = Maximum size for each block. In the case of 1D kernels, it can coincide with the number of input elements.
请注意，从CUDA 6.5开始，需要根据API建议的1D块大小计算自己的2D/3D块尺寸。
另请注意，CUDA驱动程序API包含功能等效的API，用于占用率计算，因此可以在驱动程序API代码中使用cuOccupancyMaxPotentialBlockSize，方法与上面示例中运行时API相同。
