MCPcopy Create free account
hub / github.com/ARM-software/ComputeLibrary / run_op

Method run_op

src/gpu/cl/kernels/ClWidthConcatenate2TensorsKernel.cpp:149–172  ·  view source on GitHub ↗

Source from the content-addressed store, hash-verified

147}
148
149void ClWidthConcatenate2TensorsKernel::run_op(ITensorPack &tensors, const Window &window, ::cl::CommandQueue &queue)
150{
151 ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL(this);
152 ARM_COMPUTE_ERROR_ON_INVALID_SUBWINDOW(ICLKernel::window(), window);
153
154 Window slice = window.first_slice_window_4D();
155
156 const auto src0 =
157 utils::cast::polymorphic_downcast<const ICLTensor *>(tensors.get_const_tensor(TensorType::ACL_SRC_VEC));
158 const auto src1 =
159 utils::cast::polymorphic_downcast<const ICLTensor *>(tensors.get_const_tensor(TensorType::ACL_SRC_VEC + 1));
160 auto dst = utils::cast::polymorphic_downcast<ICLTensor *>(tensors.get_tensor(TensorType::ACL_DST));
161
162 do
163 {
164 unsigned int idx = 0;
165 add_4D_tensor_argument(idx, src0, slice);
166 add_4D_tensor_argument(idx, src1, slice);
167 add_4D_tensor_argument(idx, dst, slice);
168 _kernel.setArg<cl_int>(idx++, _depth);
169 _kernel.setArg<cl_int>(idx++, _input1_width);
170 enqueue(queue, *this, window, lws_hint());
171 } while (window.slide_window_slice_4D(slice));
172}
173} // namespace kernels
174} // namespace opencl
175} // namespace arm_compute

Callers

nothing calls this directly

Calls 4

first_slice_window_4DMethod · 0.80
get_const_tensorMethod · 0.80
slide_window_slice_4DMethod · 0.80
get_tensorMethod · 0.45

Tested by

no test coverage detected