diff options
author | Georgios Pinitas <georgios.pinitas@arm.com> | 2021-04-22 21:13:21 +0100 |
---|---|---|
committer | Georgios Pinitas <georgios.pinitas@arm.com> | 2021-05-18 14:48:39 +0000 |
commit | 856f66e6c61b77d03f754cd0fa8439891f0e4aca (patch) | |
tree | f9379cd0853ac407109e54c3d53b385ceee066c2 /arm_compute/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.h | |
parent | 37f4b2ef1ea225a90ccb563fcb2c08f8fb0fb5d5 (diff) | |
download | ComputeLibrary-856f66e6c61b77d03f754cd0fa8439891f0e4aca.tar.gz |
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 <georgios.pinitas@arm.com>
Change-Id: I7055435d831b05b749b26302082e4ac45f26dfb0
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/5498
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'arm_compute/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.h')
-rw-r--r-- | arm_compute/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.h | 10 |
1 files changed, 8 insertions, 2 deletions
diff --git a/arm_compute/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.h b/arm_compute/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.h index e7f4cb9d01..e5de45c34f 100644 --- a/arm_compute/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.h +++ b/arm_compute/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.h @@ -41,7 +41,13 @@ class CLGEMMLowpOffsetContributionKernel; class CLGEMMLowpOffsetContributionOutputStageKernel; class CLGEMMLowpMatrixAReductionKernel; class CLGEMMLowpMatrixBReductionKernel; -class CLGEMMReshapeRHSMatrixKernel; +namespace opencl +{ +namespace kernels +{ +class ClGemmReshapeRhsMatrixKernel; +} // namespace kernels +} // namespace opencl /** Basic function to execute GEMMLowpMatrixMultiplyCore on OpenCL. */ class CLGEMMLowpMatrixMultiplyCore : public IFunction @@ -140,7 +146,7 @@ private: std::unique_ptr<CLDepthConvertLayerKernel> _weights_to_qasymm8; std::unique_ptr<CLGEMMLowpMatrixMultiplyNativeKernel> _mm_native_kernel; std::unique_ptr<CLGEMMLowpMatrixMultiplyReshapedOnlyRHSKernel> _mm_reshaped_only_rhs_kernel; - std::unique_ptr<CLGEMMReshapeRHSMatrixKernel> _mtx_b_reshape_kernel; + std::unique_ptr<opencl::kernels::ClGemmReshapeRhsMatrixKernel> _mtx_b_reshape_kernel; std::unique_ptr<CLGEMMLowpMatrixAReductionKernel> _mtx_a_reduction_kernel; std::unique_ptr<CLGEMMLowpMatrixBReductionKernel> _mtx_b_reduction_kernel; std::unique_ptr<CLGEMMLowpOffsetContributionKernel> _offset_contribution_kernel; |