Passing a struct into a Kernel

Accelerated Computing CUDA CUDA Programming and Performance
1

Hi,

I am relatively new to both CUDA and C memory management. I have been reading over the programmer’s guide to CUDA and searching forums but I cannot seem to figure out if what I am trying to do is possible.

I want to have a struct, define an array of that type and copy it to the device so I can use the values in the struct to do computations in the Kernel. I have been able to do this with just float arrays and such but the struct one gives me an error “error: expression must be a pointer to a complete object type”

Am I breaking some commandment trying to do this or am I just going about it the wrong way?

Thanks in advance.

2

I figured it out. Was time to RTM if you know what I mean.