diff options
| author | Gian Marco Iodice <gianmarco.iodice@arm.com> | 2019-06-13 15:58:32 +0100 |
|---|---|---|
| committer | Gian Marco Iodice <gianmarco.iodice@arm.com> | 2019-06-14 09:51:57 +0000 |
| commit | 2b52add96d7eced9eec36d526b800e2bc269984f (patch) | |
| tree | b230b87c8722d4bfb4f844562d0948b95d364b7b | |
| parent | 2732cca12bac29e1515cee1db5005c73893c61b4 (diff) | |
| download | ComputeLibrary-2b52add96d7eced9eec36d526b800e2bc269984f.tar.gz | |
COMPMID-2405: Add lws_hint to CLGEMMLowpMatrixAReductionKernel
Change-Id: Ie194e6df5db07d37b15d735e15856bf02d441360
Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Reviewed-on: https://review.mlplatform.org/c/1338
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
| -rw-r--r-- | src/core/CL/kernels/CLGEMMLowpReductionKernel.cpp | 12 |
1 files changed, 10 insertions, 2 deletions
diff --git a/src/core/CL/kernels/CLGEMMLowpReductionKernel.cpp b/src/core/CL/kernels/CLGEMMLowpReductionKernel.cpp index 225c358b20..e878dbcdcd 100644 --- a/src/core/CL/kernels/CLGEMMLowpReductionKernel.cpp +++ b/src/core/CL/kernels/CLGEMMLowpReductionKernel.cpp @@ -1,5 +1,5 @@ /* - * Copyright (c) 2017-2018 ARM Limited. + * Copyright (c) 2017-2019 ARM Limited. * * SPDX-License-Identifier: MIT * @@ -127,6 +127,14 @@ void CLGEMMLowpMatrixAReductionKernel::configure(const ICLTensor *mtx_a, ICLTens auto win_config = validate_and_configure_window_matrix_a_reduction(_input->info(), _output->info()); ARM_COMPUTE_ERROR_THROW_ON(win_config.first); ICLKernel::configure_internal(win_config.second); + + _config_id = kernel_name; + _config_id += "_"; + _config_id += support::cpp11::to_string(_input->info()->dimension(0)); + _config_id += "_"; + _config_id += support::cpp11::to_string(_input->info()->dimension(1)); + _config_id += "_"; + _config_id += support::cpp11::to_string(_input->info()->dimension(2)); } Status CLGEMMLowpMatrixAReductionKernel::validate(const ITensorInfo *mtx_a, const ITensorInfo *vector_sum_row) @@ -156,7 +164,7 @@ void CLGEMMLowpMatrixAReductionKernel::run(const Window &window, cl::CommandQueu unsigned int idx = 0; add_3D_tensor_argument(idx, _input, slice_in); add_2D_tensor_argument(idx, _output, slice_out); - enqueue(queue, *this, slice_out); + enqueue(queue, *this, slice_out, lws_hint()); } while(collapsed.slide_window_slice_2D(slice_out)); } |