From a046e164b96a8441b2fa14ef578f7db46a0e97da Mon Sep 17 00:00:00 2001 From: Michele Di Giorgio Date: Tue, 8 Oct 2019 09:36:26 +0100 Subject: COMPMID-2600: Implement a new and generic depthwise convolution for CL QASYMM8 NHWC The NCHW case is supported at function level by permuting the inputs/outputs to NHWC. This patch also removes CLDirectConvolutionLayerOutputStageKernel which is deprecated and some kernels which were only used in the generic case of depthwise convolution. Change-Id: I91e0f02d0a2f4a4a352e08c248e648944137fe68 Signed-off-by: Michele Di Giorgio Reviewed-on: https://review.mlplatform.org/c/2056 Reviewed-by: Giorgio Arena Comments-Addressed: Arm Jenkins Tested-by: Arm Jenkins Reviewed-by: Gian Marco Iodice --- src/core/CL/cl_kernels/helpers_asymm.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'src/core/CL/cl_kernels/helpers_asymm.h') diff --git a/src/core/CL/cl_kernels/helpers_asymm.h b/src/core/CL/cl_kernels/helpers_asymm.h index 53e6719cd7..57ecccc2b2 100644 --- a/src/core/CL/cl_kernels/helpers_asymm.h +++ b/src/core/CL/cl_kernels/helpers_asymm.h @@ -381,11 +381,13 @@ DEQUANTIZE_IMPL(uchar, 4) DEQUANTIZE_IMPL(ushort, 4) DEQUANTIZE_IMPL(short, 4) +ASYMM_ROUNDING_DIVIDE_BY_POW2_IMPL(1) ASYMM_ROUNDING_DIVIDE_BY_POW2_IMPL(2) ASYMM_ROUNDING_DIVIDE_BY_POW2_IMPL(4) ASYMM_ROUNDING_DIVIDE_BY_POW2_IMPL(8) ASYMM_ROUNDING_DIVIDE_BY_POW2_IMPL(16) +ASYMM_MULT_IMPL(1) ASYMM_MULT_IMPL(2) ASYMM_MULT_IMPL(4) ASYMM_MULT_IMPL(8) -- cgit v1.2.1