From 856f66e6c61b77d03f754cd0fa8439891f0e4aca Mon Sep 17 00:00:00 2001 From: Georgios Pinitas Date: Thu, 22 Apr 2021 21:13:21 +0100 Subject: Port CLGEMM to memory injecting interface Moves the following kernels: - CLGEMMMatrixMultiplyKernel - CLGEMMMatrixMultiplyNativeKernel - CLGEMMMatrixMultipluReshapedKernel - CLGEMMMatrixMultiplyReshapedOnlyRHSKernel Moves the following functions - CLGEMM Introduces facilities to easy handling of auxiliary temporary buffers under then new run interface. Such are: - CLAuxTensorHandler: That allows wrapping of workspace buffers memory to CLBuffer objects - Ability to inject TensorInfo to allocator without transferring ownership. This reduce the copy overhead if needed. Resolves: COMPMID-4188 Signed-off-by: Georgios Pinitas Change-Id: I7055435d831b05b749b26302082e4ac45f26dfb0 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/5498 Tested-by: Arm Jenkins Reviewed-by: Michalis Spyrou Comments-Addressed: Arm Jenkins --- arm_compute/core/Types.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'arm_compute/core/Types.h') diff --git a/arm_compute/core/Types.h b/arm_compute/core/Types.h index 9e054f26dd..ec9c419dbc 100644 --- a/arm_compute/core/Types.h +++ b/arm_compute/core/Types.h @@ -1753,11 +1753,11 @@ private: /** GEMM reshape information class. This class stores the necessary information about matrix A and matrix B reshape. * - * The matrix A can only be reshaped through @ref CLGEMMReshapeLHSMatrixKernel or @ref NEGEMMInterleave4x4Kernel - * Note: Optionally just for @ref CLGEMMReshapeLHSMatrixKernel is it possible to set mult_interleave4x4_height, the multiplication factor for the height of the 4x4 interleaved block + * The matrix A can only be reshaped through @ref opencl::kernels::ClGemmReshapeLhsMatrixKernel or @ref NEGEMMInterleave4x4Kernel + * Note: Optionally just for @ref opencl::kernels::ClGemmReshapeLhsMatrixKernel is it possible to set mult_interleave4x4_height, the multiplication factor for the height of the 4x4 interleaved block * - * The matrix B can only be reshaped through @ref CLGEMMReshapeRHSMatrixKernel or @ref NEGEMMTranspose1xWKernel - * Note: Optionally just for @ref CLGEMMReshapeRHSMatrixKernel is it possible to set mult_transpose1xW_width, the multiplication factor for the width of the 1xW transposed block + * The matrix B can only be reshaped through @ref opencl::kernels::ClGemmReshapeRhsMatrixKernel or @ref NEGEMMTranspose1xWKernel + * Note: Optionally just for @ref opencl::kernels::ClGemmReshapeRhsMatrixKernel is it possible to set mult_transpose1xW_width, the multiplication factor for the width of the 1xW transposed block * */ class GEMMReshapeInfo final -- cgit v1.2.1