From 6777359cd5b7aa322723d881e9ada426668491fa Mon Sep 17 00:00:00 2001 From: Pablo Marquez Tello Date: Fri, 6 Oct 2023 13:49:44 +0100 Subject: CpuSubKernel changes to enable fp16 in armv8a multi_isa builds * FP16 kernels must be instantiated in fp16.cpp. * Partially resolves MLCE-1102 Change-Id: I497fe0ba6e84493a5072c3e80bbba7ecd5de8095 Signed-off-by: Pablo Marquez Tello Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10448 Tested-by: Arm Jenkins Reviewed-by: Viet-Hoa Do Comments-Addressed: Arm Jenkins Benchmark: Arm Jenkins --- src/cpu/kernels/CpuSubKernel.cpp | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'src/cpu/kernels/CpuSubKernel.cpp') diff --git a/src/cpu/kernels/CpuSubKernel.cpp b/src/cpu/kernels/CpuSubKernel.cpp index 2b2c6f2e92..c8706ff651 100644 --- a/src/cpu/kernels/CpuSubKernel.cpp +++ b/src/cpu/kernels/CpuSubKernel.cpp @@ -1,5 +1,5 @@ /* - * Copyright (c) 2021-2022 Arm Limited. + * Copyright (c) 2021-2023 Arm Limited. * * SPDX-License-Identifier: MIT * @@ -31,6 +31,7 @@ #include "src/core/helpers/AutoConfiguration.h" #include "src/core/helpers/WindowHelpers.h" #include "src/cpu/kernels/add/generic/neon/impl.h" +#include "src/cpu/kernels/sub/neon/impl.h" #include "src/cpu/kernels/sub/neon/list.h" #if defined(ENABLE_FP32_KERNELS) @@ -57,7 +58,7 @@ static const std::vector available_kernels = { REGISTER_FP32_NEON(arm_compute::cpu::sub_same_neon)}, {"neon_fp16_sub", [](const CpuSubKernelDataTypeISASelectorData &data) { return (data.dt == DataType::F16) && data.isa.fp16; }, - REGISTER_FP16_NEON(arm_compute::cpu::sub_same_neon)}, + REGISTER_FP16_NEON(arm_compute::cpu::sub_same_neon_fp16)}, {"neon_u8_sub", [](const CpuSubKernelDataTypeISASelectorData &data) { return (data.dt == DataType::U8); }, REGISTER_INTEGER_NEON(arm_compute::cpu::sub_same_neon)}, {"neon_s16_sub", [](const CpuSubKernelDataTypeISASelectorData &data) { return (data.dt == DataType::S16); }, -- cgit v1.2.1