diff options
| author | ramelg01 <ramy.elgammal@arm.com> | 2022-02-04 11:37:00 +0000 |
|---|---|---|
| committer | Ramy Elgammal <ramy.elgammal@arm.com> | 2022-02-08 22:51:40 +0000 |
| commit | ba8690b4f0f82dfb66e3da819361e4032e9fa4db (patch) | |
| tree | 1a882d2ddae128b883b4b3560a28fb53272f1707 /src/gpu/cl/kernels/ClWinogradFilterTransformKernel.h | |
| parent | d56d94dff3b6fb1cc1807df19a8ead08c7f0faae (diff) | |
| download | ComputeLibrary-ba8690b4f0f82dfb66e3da819361e4032e9fa4db.tar.gz | |
Improve start-up time for winograd_filter_transform_*_nhwc
- pass tensor's dimensions at runtime rather than compile time
- Add guard macro to compile only kernel of internest
Resolves: COMPMID-5118
Signed-off-by: Ramy Elgammal <ramy.elgammal@arm.com>
Change-Id: Ie42c3c07fdd817ce62e7cad354381bc22c6e9264
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/7058
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/gpu/cl/kernels/ClWinogradFilterTransformKernel.h')
| -rw-r--r-- | src/gpu/cl/kernels/ClWinogradFilterTransformKernel.h | 5 |
1 files changed, 4 insertions, 1 deletions
diff --git a/src/gpu/cl/kernels/ClWinogradFilterTransformKernel.h b/src/gpu/cl/kernels/ClWinogradFilterTransformKernel.h index fe0c3da174..b2130304e6 100644 --- a/src/gpu/cl/kernels/ClWinogradFilterTransformKernel.h +++ b/src/gpu/cl/kernels/ClWinogradFilterTransformKernel.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2018-2021 Arm Limited. + * Copyright (c) 2018-2022 Arm Limited. * * SPDX-License-Identifier: MIT * @@ -70,6 +70,9 @@ public: // Inherited methods overridden: void run_op(ITensorPack &tensors, const Window &window, cl::CommandQueue &queue) override; + +private: + int32_t _src_dim_z{ 0 }; }; } // namespace kernels } // namespace opencl |