compute

EMCompute

Function compute 

Source 
#[unsafe(no_mangle)]
pub extern "C" fn compute(
    kernel: *mut CKernel,
    data_for_gpu: *mut GroupOfBinders,
    gpu_data_len: usize,
) -> i32
Expand description

the simple and compact function for sending your computing task to the gpu side

kernel para = CKernel type which acts as Manifest for your gpu task data_for_gpu = pointer to array of GroupOfBinders which contains data which must be sent to gpu gpu_data_len = len of the array of the GroupOfBinders

unlike CUDA , you dont need to copy data to gpu manually , this function does it for you in the most performant possible way

if you find any bug or any problem , help us to fix it -> https://github.com/SkillfulElectro/EMCompute.git