CUDA-Utilities
所属分类：GPU/显卡
开发工具：Cuda
文件大小：32KB
下载次数：0
上传日期：2019-05-17 11:58:40
上 传 者：
sh-1993
说明： CUDA编程实用程序
(Utilities for CUDA programming)
文件列表:
Bessel.cu (3450, 2019-05-17)
Bessel.cuh (106, 2019-05-17)
InputOutput.cpp (2164, 2019-05-17)
InputOutput.cu (4770, 2019-05-17)
InputOutput.cuh (523, 2019-05-17)
InputOutput.h (354, 2019-05-17)
License (0, 2019-05-17)
License\Gpl License.txt (35147, 2019-05-17)
License\Lgpl License.txt (7651, 2019-05-17)
Matlab_like.cu (6014, 2019-05-17)
Matlab_like.cuh (974, 2019-05-17)
Polynomials.cu (14702, 2019-05-17)
Polynomials.cuh (1881, 2019-05-17)
cfloat3.cu (5250, 2019-05-17)
cfloat3.cuh (3980, 2019-05-17)
cuFFT_auxiliary.cu (640, 2019-05-17)
cuFFT_auxiliary.cuh (215, 2019-05-17)
cublasWrappers.cu (1974, 2019-05-17)
cublasWrappers.cuh (1244, 2019-05-17)
cusparseWrappers.cu (878, 2019-05-17)
cusparseWrappers.cuh (493, 2019-05-17)
float2Overloads.cu (3343, 2019-05-17)
float2Overloads.cuh (1880, 2019-05-17)
float3Overloads.cu (2292, 2019-05-17)
float3Overloads.cuh (1760, 2019-05-17)
# CUDA Utilities
Utilities for CUDA programming
Wrap all the CUDA API calls within ```gpuErrchk```. For example:
```
gpuErrchk(cudaMemcpy(d_x, x, N * sizeof(float), cudaMemcpyHostToDevice));
```
Use ```gpuErrchk``` to check for errors in your kernels. For example:
```
#ifdef DEBUG
gpuErrchk(cudaPeekAtLastError());
gpuErrchk(cudaDeviceSynchronize());
#endif
```
If ```DEBUG``` is defined, then the CUDA error check of the kernels will be executed.
Use ```iDivUp``` to find the number of blocks to be launched by a CUDA kernel. For example:
```
BLOCKSIZE = 256;
N = 10000; // Number of elements to be processed
kernel<<>>(...);
```
近期下载者：
相关文件：
收藏者：