cancel
Showing results for 
Search instead for 
Did you mean: 

OpenCL

newbee37
Journeyman III

(python) pyopencl.enqueue_nd_range_kernel -> (native) clEnqueueNDRangeKernel mapping

Hi All!

Sorry in advance for a silly question but...

PyOpenCL signature:

pyopencl.enqueue_nd_range_kernel(
queue,
 
kernel,
 
global_work_size,
 
local_work_size,
 
global_work_offset=None,
 
wait_for=None,
 
g_times_l=False,
 
allow_empty_ndrange=False)

OpenCL signature:

cl_int clEnqueueNDRangeKernel(
    cl_command_queue command_queue,
    cl_kernel kernel,
    cl_uint work_dim,
    const size_t* global_work_offset,
    const size_t* global_work_size,
    const size_t* local_work_size,
    cl_uint num_events_in_wait_list,
    const cl_event* event_wait_list,
    cl_event* event);

How the pyopencl maps the parameters?

For instance, if I write something like kernel(queue, (5000,), None, args, res), where the 5000 goes? What will be value of the work_dim?

 

Thanks a lot!

 

 

0 Likes
0 Replies