Blocks
- 早期的卡,最多支持65535个blocks, 后面的卡最多可以支持 2 31 − 1 = 2147483647 2^{31}-1=2147483647 231−1=2147483647
可以理解为blocks可以随便造,主要是要把threads的大小选好
Threads
- 一般来讲每一个block不能超过512(Compute Capability 1.x)/ 1024(2.x和之后版本)个线程
你电脑不旧的话,一般都是不超过1024
- 每个块的线程数应该是warp size大小的整数倍,在所有当前的硬件上,warp size大小是32。
比如我的电脑:
Device : “GeForce RTX 2080 Ti”
driverVersion : 10010
runtimeVersion : 10000
CUDA Driver Version / Runtime Version 10.1 / 10.0
CUDA Capability Major/Minor version number : 7.5
Total amount of global memory : 10.73 GBytes (11523260416 bytes)
GPU Clock rate : 1545 MHz(1.54 GHz)
Memory Clock rate : 7000 Mhz
Memory Bus Width : 352-bit
L2 Cache Size: 5767168 bytes
Total amount of constant memory: 65536 bytes
Total amount of shared memory per block: 49152 bytes
Total number of registers available per block: 65536
Warp Size: 32
Maximum number of threads per multiprocessor: 1024
Maximum number of thread per block: 1024
Maximum sizes of each dimension of a block: 1024 x 1024 x 64
Maximum sizes of each dimension of a grid: 2147483647 x 65535 x 65535
一般来讲都是32,你也可以那个用一下代码测试一下:
#include <stdio.h>
int main(void) {
cudaDeviceProp deviceProp;
if (cudaSuccess != cudaGetDeviceProperties(&deviceProp, 0)) {
printf("Get device properties failed.\n");
return 1;
} else {
printf("The warp size is %d.\n", deviceProp.warpSize);
return 0;
}
}
- threads太小(32)或者太大(512)都不好,经验来讲,最好选择128 threads/256 threads 为好