From 318c9f40770b2d1c06f8c0fe3f7929812503733e Mon Sep 17 00:00:00 2001 From: David Mansell Date: Wed, 8 Jul 2020 13:28:45 +0100 Subject: COMPMID-3578: Update FP32/int8 kernel selection. Upgrade the current 'is_preferred()' mechanism with a new framework, where kernels instead provide an estimated cycle count figure. Compatibility with old mechanism is achieved via a wrapper which replaces a "true" result with an estimate of 0, and a "false" result with UINT64_MAX. This mechanism is then used to select between 'interleaved' and 'hybrid' FP32 NEON kernels. This uses a simple system based on counting MACs performed and bytes of data transferred (for rearrange/merge operations) and dividing by fixed performance figures, which are provided for A53, A55, A73 and 'default' figures (based on A76). Separately, a new route for performing int8 GEMMs by using the int16 kernel is provided. This performs significantly (for uint8) or slightly (for int8) better on A53 than the existing int8 route. Optimized 8-to-16 bit transforms are also included. Change-Id: I53b2e59eb9368793c78c2081e17d2445361bcc47 Signed-off-by: David Mansell Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/c/VisualCompute/ComputeLibrary/+/250120 Tested-by: bsgcomp Reviewed-by: Gian Marco Iodice Comments-Addressed: bsgcomp Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3609 Tested-by: Arm Jenkins Reviewed-by: Georgios Pinitas --- .../kernels/arm_gemm/performance_parameters.hpp | 37 ++++++++++++++++++++++ 1 file changed, 37 insertions(+) create mode 100644 src/core/NEON/kernels/arm_gemm/performance_parameters.hpp (limited to 'src/core/NEON/kernels/arm_gemm/performance_parameters.hpp') diff --git a/src/core/NEON/kernels/arm_gemm/performance_parameters.hpp b/src/core/NEON/kernels/arm_gemm/performance_parameters.hpp new file mode 100644 index 0000000000..059ab5f7df --- /dev/null +++ b/src/core/NEON/kernels/arm_gemm/performance_parameters.hpp @@ -0,0 +1,37 @@ +/* + * Copyright (c) 2020 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. + */ +#pragma once + +namespace arm_gemm { + +struct PerformanceParameters { + float kernel_macs_cycle; + float prepare_bytes_cycle = 0.0f; + float merge_bytes_cycle = 0.0f; + + PerformanceParameters(float k) : kernel_macs_cycle(k) { } + PerformanceParameters(float k, float p, float m) : kernel_macs_cycle(k), prepare_bytes_cycle(p), merge_bytes_cycle(m) { } +}; + +} // namespace arm_gemm -- cgit v1.2.1