c++ - How to have a matrix as paramater when working with CUDA and Matlab -


I'm trying to explain that the matrix from matlux should be used to separate the C ++ queue file (* I want to process in different threads! I definitely do not want to divide the matrix in different threads! I want each thread to have the same matrix!

I Something tried to do this:

  _global_ zero radial average (int * image [] []) {...}   

However It does not work. I am getting the following error:

  kernel.cu (8): Error: In the compilation of "C: / User / Admin ~" in an array like this 1 error There are no elements 1 / AppData / Loca / Temp / tmpxft_00000c44_00000000-8_kernel.cpp1.ii ".   

Can you think of any way to do this? Or is it possible? < / P>

BTW: I am not using any libraries for C ++, only CUDA-AP

Matlab The correct way to define a kernel to pass the matrix is ​​clearly shown on the page (sometimes) linked to your question.

In summary, define such a kernel (matlab gpuArray is automatically passed as a device pointer in the kernel):

 < Code> __ global radial average (int * image) {...}   

After you retrieve the kernel by using the CUDAKernel toolbox, your Matlab Create an integer gpuArray from the matrix, and pass it to the kernel function.

Comments

Popular posts from this blog

c - Mpirun hangs when mpi send and recieve is put in a loop -

python - Apply coupon to a customer's subscription based on non-stripe related actions on the site -

java - Unable to get JDBC connection in Spring application to MySQL -