ComputeLibrary.git -

Age	Commit message (Collapse)	Author
2021-06-18	Integrate improved CPU depthwise convolution kernels	Michele Di Giorgio
	* Replace assembly kernels for depthwise convolution with more optimized ones. * Add int8 assembly kernels. * Fix implicit padding on optimized kernels Resolves: COMPMID-3867, COMPMID-4361 Change-Id: I0b0867e05f61be4f368f62190d55e14d0ab3ebf2 Signed-off-by: Michele Di Giorgio <michele.digiorgio@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/5622 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
2021-03-08	Fix cl_gemmlowp_reshaped_rhs_only_fused_output_stage_fixedpoint example	SiCongLi
	CLGEMMLowpReshapedOnlyRHSKernel only accepts 1D bias. Resolves COMPMID-4295 Change-Id: Ib41c42ded3a6e6e7009a238b2deeffb90aa9ea41 Signed-off-by: SiCongLi <sicong.li@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/5218 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2021-03-04	Add tuner mode support in GeMM benchmark	Gian Marco Iodice
	Change-Id: Ie1ad9880d22daa2a8ee4f239b5b23d3a7ba4cb1b Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/5210 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Reviewed-by: SiCong Li <sicong.li@arm.com>
2021-02-09	Fix Floating Point Exception for cl_gemm_reshaped	Manuel Bottini
	Return proper error in the examples when such cases occur Resolves: COMPMID-4237 Change-Id: I864c6f73a629a961774c7ae1cb62c28edcbed379 Signed-off-by: Manuel Bottini <manuel.bottini@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/5028 Reviewed-by: Giorgio Arena <giorgio.arena@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2020-12-08	Fix invalid arguments error for cl_gemmlowp* examples without arguments	SiCong Li
	* Pass QASYMM8 as default argument for cl_gemmlowp examples * Fix the (arbitrary) quantization info to be the same across tensors to avoid quantization information mismatch issue Resolves COMPMID-4044 Signed-off-by: SiCong Li <sicong.li@arm.com> Change-Id: I14aa6648bbebdcd9b0bb77c47b46c27aa79d6253 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4650 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-12-01	Update default C++ standard to C++14	Georgios Pinitas
	(3RDPARTY_UPDATE) Resolves: COMPMID-3849 Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com> Change-Id: I6369f112337310140e2d6c8e79630cd11138dfa0 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4544 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-11-30	COMPMID-3982 Add QASYMM8 support to gemm tuner (reshaped and reshaped_only_rhs)	SiCongLi
	Signed-off-by: SiCongLi <sicong.li@arm.com> Change-Id: Ic6492cfd2701374d837ac53d51b0ddc07e6b1fd1 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4550 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>