diff options
author | Gian Marco Iodice <gianmarco.iodice@arm.com> | 2023-04-26 14:55:02 +0100 |
---|---|---|
committer | Gian Marco Iodice <gianmarco.iodice@arm.com> | 2023-05-02 09:27:49 +0000 |
commit | 7a0f1bdaf74cde263b2919c7d1652b0cb87a94f3 (patch) | |
tree | 62886dac919eb95811efd76d907960dfddef0b61 /src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyValhall.h | |
parent | a62129a02397ba87171ebf4477795f628dcec0f6 (diff) | |
download | ComputeLibrary-7a0f1bdaf74cde263b2919c7d1652b0cb87a94f3.tar.gz |
Add fp16 GeMM heuristic for Arm® Mali™-G710
- Performance improvements on various networks between 5-20%
Resolves COMPMID-6030
Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Change-Id: Idcf7de57e6f5a94a6a94ec78229dd53c24de44f4
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/c/VisualCompute/ComputeLibrary/+/514481
Tested-by: Viet-Hoa Do <viet-hoa.do@arm.com>
Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com>
Comments-Addressed: bsgcomp <bsgcomp@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9524
Benchmark: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyValhall.h')
-rw-r--r-- | src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyValhall.h | 3 |
1 files changed, 2 insertions, 1 deletions
diff --git a/src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyValhall.h b/src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyValhall.h index 0ec068fffd..f2952a3d30 100644 --- a/src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyValhall.h +++ b/src/gpu/cl/kernels/gemm/reshaped_only_rhs/ClGemmDefaultConfigReshapedRhsOnlyValhall.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2020-2022 Arm Limited. + * Copyright (c) 2020-2023 Arm Limited. * * SPDX-License-Identifier: MIT * @@ -53,6 +53,7 @@ private: std::pair<GEMMLHSMatrixInfo, GEMMRHSMatrixInfo> configure_G78_f32(unsigned int m, unsigned int n, unsigned int k, unsigned int b); std::pair<GEMMLHSMatrixInfo, GEMMRHSMatrixInfo> configure_G78_f16(unsigned int m, unsigned int n, unsigned int k, unsigned int b); std::pair<GEMMLHSMatrixInfo, GEMMRHSMatrixInfo> configure_G77_u8(unsigned int m, unsigned int n, unsigned int k, unsigned int b); + std::pair<GEMMLHSMatrixInfo, GEMMRHSMatrixInfo> configure_G710_f16(unsigned int m, unsigned int n, unsigned int k, unsigned int b); std::pair<GEMMLHSMatrixInfo, GEMMRHSMatrixInfo> configure_G715_f32(unsigned int m, unsigned int n, unsigned int k, unsigned int b); std::pair<GEMMLHSMatrixInfo, GEMMRHSMatrixInfo> configure_G715_f16(unsigned int m, unsigned int n, unsigned int k, unsigned int b); }; |