From cfa2bba98169cb5ab1945462514be1b6badf7d98 Mon Sep 17 00:00:00 2001 From: Georgios Pinitas Date: Thu, 27 Jun 2019 17:00:52 +0100 Subject: COMPMID-2178: Update GEMM assembly code. Perform offset reduction and requantization within the assembly wrapper. Change-Id: I5d5b3e1f6f9ef4c71805362c57f88ff199c027a3 Signed-off-by: Georgios Pinitas Reviewed-on: https://review.mlplatform.org/c/1541 Comments-Addressed: Pablo Marquez Reviewed-by: Gian Marco Iodice Comments-Addressed: Arm Jenkins Tested-by: Arm Jenkins --- arm_compute/runtime/NEON/functions/NEGEMMLowpMatrixMultiplyCore.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'arm_compute/runtime/NEON/functions/NEGEMMLowpMatrixMultiplyCore.h') diff --git a/arm_compute/runtime/NEON/functions/NEGEMMLowpMatrixMultiplyCore.h b/arm_compute/runtime/NEON/functions/NEGEMMLowpMatrixMultiplyCore.h index d3b27e4faf..5b6a0dd943 100644 --- a/arm_compute/runtime/NEON/functions/NEGEMMLowpMatrixMultiplyCore.h +++ b/arm_compute/runtime/NEON/functions/NEGEMMLowpMatrixMultiplyCore.h @@ -122,7 +122,8 @@ private: int32_t _a_offset; int32_t _b_offset; bool _run_vector_matrix_multiplication; - bool _dot_product_path; + bool _assembly_path; + bool _fused_assembly_path; bool _reshape_b_only_on_first_run; bool _is_prepared; bool _fuse_output_stage; -- cgit v1.2.1