How can I initialize a __device__ variable via the CUDAKernel interface?

I have this global _device___ variable in my original C/Cuda kernel code, that I'm now porting to a kernel compatible to the CUDAKernel interface, how can I initialize it? I used cudaMemcpyToSymbol in C, can I simply use a gpuArray variable declaration in its place?

回答 (0 件)

カテゴリ

ヘルプ センター および File ExchangeGPU Computing についてさらに検索

質問済み:

2013 年 5 月 22 日

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by