From ae72a46e495742863dba44fcf5fdc673c9d2afbc Mon Sep 17 00:00:00 2001 From: Gunes Bayir Date: Sun, 29 Jan 2023 13:24:24 +0000 Subject: =?UTF-8?q?Add=20new=20operator=20AddMulAdd=20for=20Neon=E2=84=A2?= =?UTF-8?q?=20backend=20for=20Float/Quantized=20types?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This is a fused operator that merges Add + Mul + Add [+ Relu-based-Activation] layers and have an intermediate output after the first Add. It's supported for FP16/32/QASYMM8/QASYMM8_SIGNED data types. The subsequent Add and Mul are intended for scaling and the coefficients only have one dimension (per channel). The inputs are - input1 : nD tensor [X, Y, Z, W, ..] - input2 : nD tensor [X, Y, Z, W, ..] - add_coef : 1D tensor [X] - mul_coef : 1D tensor [X] The outputs are - out1 : nD tensor (intermediate output) [X, Y, Z, W, ..] - out2 : nD tensor (final output) [X, Y, Z, W, ..] The operation can be summarized as follows: out1 <- input1 + input2 out2 <- Act(out1 * mul_coef + add_coef) The activation function can be Identity, Relu, Bounded Relu or Lower/Upper Bounded Relu. The intermediate output can be skipped by providing a nullptr. The reason of providing this operator is to be able to fuse in case of Residual network patterns and save computations by reducing memory back and forward. Resolves: COMPMID-5463 Signed-off-by: Gunes Bayir Change-Id: I8ef577aa623b036e9a9f655cc088493fd19a6109 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9055 Comments-Addressed: Arm Jenkins Reviewed-by: Jakub Sujak Reviewed-by: Viet-Hoa Do Tested-by: Arm Jenkins Benchmark: Arm Jenkins --- src/cpu/kernels/addmuladd/list.h | 48 ++++++++++++++++++++++++++++++++++++++++ 1 file changed, 48 insertions(+) create mode 100644 src/cpu/kernels/addmuladd/list.h (limited to 'src/cpu/kernels/addmuladd/list.h') diff --git a/src/cpu/kernels/addmuladd/list.h b/src/cpu/kernels/addmuladd/list.h new file mode 100644 index 0000000000..a7c22c06d8 --- /dev/null +++ b/src/cpu/kernels/addmuladd/list.h @@ -0,0 +1,48 @@ +/* + * Copyright (c) 2023 Arm Limited. + * + * SPDX-License-Identifier: MIT + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in all + * copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ +#ifndef SRC_CPU_KERNELS_ADDMULADD_LIST +#define SRC_CPU_KERNELS_ADDMULADD_LIST + +#include "arm_compute/core/ITensor.h" +#include "arm_compute/core/Types.h" +#include "arm_compute/core/Window.h" + +namespace arm_compute +{ +namespace cpu +{ +#define DECLARE_ADD_MUL_ADD_KERNEL(func_name) \ + void func_name(const ITensor *input1, const ITensor *input2, const ITensor *bn_mul, const ITensor *bn_add, \ + ITensor *add_output, ITensor *final_output, ConvertPolicy policy, const ActivationLayerInfo &act_info, const Window &window) + +DECLARE_ADD_MUL_ADD_KERNEL(add_mul_add_fp32_neon); +DECLARE_ADD_MUL_ADD_KERNEL(add_mul_add_fp16_neon); +DECLARE_ADD_MUL_ADD_KERNEL(add_mul_add_u8_neon); +DECLARE_ADD_MUL_ADD_KERNEL(add_mul_add_s8_neon); + +#undef DECLARE_ADD_MUL_ADD_KERNEL + +} // namespace cpu +} // namespace arm_compute +#endif /* SRC_CPU_KERNELS_ADDMULADD_LIST */ -- cgit v1.2.1