From 94672fb2af6535adc6ea7fe8b8498580ad8cf3f4 Mon Sep 17 00:00:00 2001 From: Georgios Pinitas Date: Wed, 22 Jan 2020 18:36:27 +0000 Subject: COMPMID-3003: Integrate assembly kernels utilizing MMLA instruction. MMLA is a matrix-multiply instruction introduced on armv8.6-A Signed-off-by: Georgios Pinitas Change-Id: I572a54981d48f5a1e0e9e51102cb7ae28ad87806 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/2663 Tested-by: Arm Jenkins Reviewed-by: Michalis Spyrou Comments-Addressed: Arm Jenkins --- .../kernels/a64_interleaved_u8u32_mmla_12x8.hpp | 72 ++++++++++++++++++++++ 1 file changed, 72 insertions(+) create mode 100644 src/core/NEON/kernels/arm_gemm/kernels/a64_interleaved_u8u32_mmla_12x8.hpp (limited to 'src/core/NEON/kernels/arm_gemm/kernels/a64_interleaved_u8u32_mmla_12x8.hpp') diff --git a/src/core/NEON/kernels/arm_gemm/kernels/a64_interleaved_u8u32_mmla_12x8.hpp b/src/core/NEON/kernels/arm_gemm/kernels/a64_interleaved_u8u32_mmla_12x8.hpp new file mode 100644 index 0000000000..d66edd832a --- /dev/null +++ b/src/core/NEON/kernels/arm_gemm/kernels/a64_interleaved_u8u32_mmla_12x8.hpp @@ -0,0 +1,72 @@ +/* + * Copyright (c) 2019 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 + +#ifdef __aarch64__ + +#include +#include "../std_transforms_fixed.hpp" + +namespace arm_gemm { + +// Actual kernel implementations +void a64_interleaved_u8u32_mmla_12x8(const uint8_t *, const uint8_t *, uint32_t *, int, int, int); + +class interleaved_u8u32_mmla_12x8 { +public: + typedef uint8_t operand_type; + typedef uint32_t result_type; + + typedef void (*kern_type)(const uint8_t *, const uint8_t *, uint32_t *, int, int, int); + + /* Kernel blocking parameters */ + static unsigned int out_width() + { + return 12; + } + + static unsigned int out_height() + { + return 8; + } + + static unsigned int k_unroll() + { + return 8; + } + + // Use the standard fixed size transforms. + StdTransformsFixed transforms = {}; + + kern_type kernel=a64_interleaved_u8u32_mmla_12x8; + + interleaved_u8u32_mmla_12x8(const CPUInfo *ci) + { + UNUSED(ci); + } +}; + +} // namespace arm_gemm + +#endif // __aarch64__ -- cgit v1.2.1