From 4b90865ab985d571f70c60583cdfb8c7a65f1670 Mon Sep 17 00:00:00 2001 From: Gian Marco Iodice Date: Thu, 18 Oct 2018 10:21:02 +0100 Subject: COMPMID-1413 - Improve the performance of GEMMLowp with 8 bit dot product on OpenCL COMPMID-1424 - Add dot product support for CLDepthwise QASYMM8 3x3 NHWC non-unit stride With this patch we are able to improve the performance of MobileNet v1-qasymm8 by 37 % Tried to use the dot product instruction in CLDepthwise QASYMM8 3x3 NHWC non-unit stride but I have not seen any benefit (maybe because we have few arithemtic operation and we do not have more load instructions). However Depthwise convolution has been improved by 30% Change-Id: Id768a99c2e53a04276707e427af5d0ec93419ada Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/155082 Tested-by: bsgcomp Reviewed-by: Georgios Pinitas --- src/core/CL/kernels/CLGEMMInterleave4x4Kernel.cpp | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'src/core/CL/kernels/CLGEMMInterleave4x4Kernel.cpp') diff --git a/src/core/CL/kernels/CLGEMMInterleave4x4Kernel.cpp b/src/core/CL/kernels/CLGEMMInterleave4x4Kernel.cpp index ae54e77972..f333c1bff3 100644 --- a/src/core/CL/kernels/CLGEMMInterleave4x4Kernel.cpp +++ b/src/core/CL/kernels/CLGEMMInterleave4x4Kernel.cpp @@ -115,7 +115,7 @@ CLGEMMInterleave4x4Kernel::CLGEMMInterleave4x4Kernel() { } -void CLGEMMInterleave4x4Kernel::configure(const ICLTensor *input, ICLTensor *output, int mult_interleave4x4_height, bool reinterpret_input_as_3d) +void CLGEMMInterleave4x4Kernel::configure(const ICLTensor *input, ICLTensor *output, int mult_interleave4x4_height, bool reinterpret_input_as_3d, bool unroll_block) { ARM_COMPUTE_ERROR_ON_NULLPTR(input, output); @@ -132,6 +132,7 @@ void CLGEMMInterleave4x4Kernel::configure(const ICLTensor *input, ICLTensor *out // Create build options CLBuildOptions build_opts; build_opts.add_option("-DMULT_INTERLEAVE4X4_HEIGHT=" + support::cpp11::to_string(mult_interleave4x4_height)); + build_opts.add_option_if(unroll_block, "-DUNROLL_BLOCK"); build_opts.add_option_if(_reinterpret_input_as_3d, "-DREINTERPRET_INPUT_AS_3D"); build_opts.add_option_if(_reinterpret_input_as_3d, "-DHEIGHT_GEMM3D=" + support::cpp11::to_string(input->info()->dimension(1))); build_opts.add_option_if(_reinterpret_input_as_3d, "-DDEPTH_GEMM3D=" + support::cpp11::to_string(input->info()->dimension(2))); -- cgit v1.2.1