def test_variable_sized_fields():
src_field = Field.create_generic('src', spatial_dimensions=2)
dst_field = Field.create_generic('dst', spatial_dimensions=2)
update_rule = Assignment(dst_field[0, 0],
(src_field[0, 1] + src_field[0, -1] +
src_field[1, 0] + src_field[-1, 0]) / 4)
ast = create_cuda_kernel(sympy_cse_on_assignment_list([update_rule]))
kernel = make_python_function(ast)
size = (3, 3)
src_arr = np.random.rand(*size)
src_arr = add_ghost_layers(src_arr)
dst_arr = np.zeros_like(src_arr)
gpu_src_arr = gpuarray.to_gpu(src_arr)
gpu_dst_arr = gpuarray.to_gpu(dst_arr)
kernel(src=gpu_src_arr, dst=gpu_dst_arr)
gpu_dst_arr.get(dst_arr)
stencil = np.array([[0, 1, 0], [1, 0, 1], [0, 1, 0]]) / 4.0
reference = convolve(remove_ghost_layers(src_arr),
stencil,
mode='constant',
cval=0.0)
reference = add_ghost_layers(reference)
np.testing.assert_almost_equal(reference, dst_arr)
def _generate_fields(dt=np.uint64, num_directions=1, layout='numpy'):
field_sizes = FIELD_SIZES
if num_directions > 1:
field_sizes = [s + (num_directions, ) for s in field_sizes]
fields = []
for size in field_sizes:
field_layout = layout_string_to_tuple(layout, len(size))
src_arr = create_numpy_array_with_layout(size, field_layout, dtype=dt)
array_data = np.reshape(np.arange(1, int(np.prod(size) + 1)), size)
# Use flat iterator to input data into the array
src_arr.flat = add_ghost_layers(
array_data,
index_dimensions=1 if num_directions > 1 else 0).astype(dt).flat
dst_arr = np.zeros(src_arr.shape, dtype=dt)
buffer_arr = np.zeros(np.prod(src_arr.shape), dtype=dt)
fields.append((src_arr, dst_arr, buffer_arr))
return fields
def _generate_fields(dt=np.uint8, stencil_directions=1, layout='numpy'):
pytest.importorskip('pycuda')
field_sizes = FIELD_SIZES
if stencil_directions > 1:
field_sizes = [s + (stencil_directions, ) for s in field_sizes]
fields = []
for size in field_sizes:
field_layout = layout_string_to_tuple(layout, len(size))
src_arr = create_numpy_array_with_layout(size, field_layout).astype(dt)
array_data = np.reshape(np.arange(1, int(np.prod(size) + 1)), size)
# Use flat iterator to input data into the array
src_arr.flat = add_ghost_layers(
array_data, index_dimensions=1
if stencil_directions > 1 else 0).astype(dt).flat
gpu_src_arr = gpuarray.to_gpu(src_arr)
gpu_dst_arr = gpuarray.empty_like(gpu_src_arr)
size = int(np.prod(src_arr.shape))
gpu_buffer_arr = gpuarray.zeros(size, dtype=dt)
fields.append((src_arr, gpu_src_arr, gpu_dst_arr, gpu_buffer_arr))
return fields