cconvey commented on code in PR #12204: URL: https://github.com/apache/tvm/pull/12204#discussion_r949341528
########## src/runtime/hexagon/ops/conv2d_hvx.cc: ########## @@ -0,0 +1,473 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ + +#include <HAP_compute_res.h> +#include <hexagon_types.h> +#include <hvx_hexagon_protos.h> +#include <tvm/runtime/c_runtime_api.h> +#include <tvm/runtime/device_api.h> + +#include <algorithm> +#include <cassert> +#include <cinttypes> + +#include "tvm/runtime/hexagon/ops/conv2d.h" + +// Current limitations: +// - N in NHWC must be 1 +// - dilated convolutions are not supported +// - Bias is not accepted +// - Optional "relu" is not performed + +// Packed arguments: +// 0: DLTensor activations (NHWC) +// 1: DLTensor weights (HWIO) +// 2: int offset_top +// 3: int offset_left +// 4: int stride_h +// 5: int stride_w +// 6: DLTensor output (NHWC) +extern "C" int conv2d_packed(TVMValue* args, int* type_codes, int num_args, TVMValue* out_val, + int out_code, void* res_handle); + +namespace tvm { +namespace runtime { +namespace hexagon { + +inline uint16_t* getElementPtr(int block_out_y, int block_out_x, int block_out_c, int yi, int xio, Review Comment: If `uint16_t` is a stand-in for qfloat16, is there a typename supplied by the Hexagon SDK that might be more descriptive? This might not be a valid example, but I'm thinking of something _like_ `qhl_q0_t`. Or if the SDK provides no such type name, could it make sense to just define one here? E.g., `using qfloat16_ptr = uint16_t*;` -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: commits-unsubscr...@tvm.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
