pyopencl

import pyopencl as cl import numpy as np def run_opencl_kernel(): # 1. Create a context try: ctx = cl.create_some_context() except cl.LogicError as e: print(f"Error creating OpenCL context: {e}") print("Please ensure you have OpenCL drivers/ICD installed.") return # 2. Create a command queue queue = cl.CommandQueue(ctx) # 3. Prepare host data a = np.random.rand(50000).astype(np.float32) # 4. Create device buffers and transfer data mf = cl.mem_flags a_buf = cl.Buffer(ctx, mf.READ_WRITE | mf.COPY_HOST_PTR, hostbuf=a) # 5. Define and build the OpenCL kernel program prg = cl.Program(ctx, """ __kernel void twice(__global float *a) { int gid = get_global_id(0); a[gid] = 2*a[gid]; } """).build() # 6. Execute the kernel # global_size is the total number of work-items # local_size (None) lets the OpenCL driver decide prg.twice(queue, a.shape, None, a_buf) # 7. Create output array and transfer data back to host result = np.empty_like(a) cl.enqueue_copy(queue, result, a_buf).wait() # 8. Verify the result print("Original first 5 elements:", a[:5]) print("Resulting first 5 elements:", result[:5]) assert np.allclose(result, 2*a) print("Kernel executed successfully and results verified!") if __name__ == '__main__': run_opencl_kernel()