From 19835e591cb0b66a0f5000ae1505bf299e50337d Mon Sep 17 00:00:00 2001 From: Gian Marco Date: Tue, 30 Jan 2018 13:35:54 +0000 Subject: COMPMID-882 - Optimizing GEMMLowp on OpenCL reshaping matrices This new optimization allows to achieve 36.3 % of MAC utilisation on Mate 9 @ 1GHz. The performance have been reported here https://confluence.arm.com/display/MLENG/GEMMLowp+performance%3A+ACL+18.02 Change-Id: I71b6a217068763dfdc11bbf3574ee0eb94f93679 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/118531 Reviewed-by: Anthony Barbier Tested-by: Jenkins --- src/core/CL/CLKernelLibrary.cpp | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'src/core/CL/CLKernelLibrary.cpp') diff --git a/src/core/CL/CLKernelLibrary.cpp b/src/core/CL/CLKernelLibrary.cpp index 0847612d21..5452b8a1be 100644 --- a/src/core/CL/CLKernelLibrary.cpp +++ b/src/core/CL/CLKernelLibrary.cpp @@ -237,7 +237,8 @@ const std::map CLKernelLibrary::_kernel_program_map = { "gemmlowp_matrix_b_reduction", "gemmlowp.cl" }, { "gemmlowp_mm_bifrost", "gemmlowp.cl" }, { "gemmlowp_mm_midgard", "gemmlowp.cl" }, - { "gemmlowp_mm_interleaved_transposed", "gemmlowp.cl" }, + { "gemmlowp_mm_interleaved_transposed_bifrost", "gemmlowp.cl" }, + { "gemmlowp_mm_interleaved_transposed_midgard", "gemmlowp.cl" }, { "gemmlowp_offset_contribution", "gemmlowp.cl" }, { "gemmlowp_output_stage_quantize_down", "gemmlowp.cl" }, { "gemmlowp_output_stage_quantize_down_fixedpoint", "gemmlowp.cl" }, -- cgit v1.2.1