diff options
| author | Usama Arif <usama.arif@arm.com> | 2019-03-26 13:44:01 +0000 |
|---|---|---|
| committer | Usama Arif <usama.arif@arm.com> | 2019-03-26 15:47:46 +0000 |
| commit | ae0001e4e9af43e03398a8aa36c038cf4fb2600a (patch) | |
| tree | c15543469a3fb30f936d29fbc545de32ecb44bb8 /src/core/CL/kernels/CLL2NormalizeLayerKernel.cpp | |
| parent | ec6997563a7cccf58431267cca39435ecd57cd32 (diff) | |
| download | ComputeLibrary-ae0001e4e9af43e03398a8aa36c038cf4fb2600a.tar.gz | |
COMPMID-2077 Optimise CLL2NormalizeLayerKernel
Change-Id: If3866d90f84d62578b7e26f524fa4d7757a5970c
Signed-off-by: Usama Arif <usama.arif@arm.com>
Reviewed-on: https://review.mlplatform.org/c/906
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Diffstat (limited to 'src/core/CL/kernels/CLL2NormalizeLayerKernel.cpp')
| -rw-r--r-- | src/core/CL/kernels/CLL2NormalizeLayerKernel.cpp | 14 |
1 files changed, 7 insertions, 7 deletions
diff --git a/src/core/CL/kernels/CLL2NormalizeLayerKernel.cpp b/src/core/CL/kernels/CLL2NormalizeLayerKernel.cpp index e33dab0f09..cb2e29449c 100644 --- a/src/core/CL/kernels/CLL2NormalizeLayerKernel.cpp +++ b/src/core/CL/kernels/CLL2NormalizeLayerKernel.cpp @@ -117,7 +117,7 @@ void CLL2NormalizeLayerKernel::configure(const ICLTensor *input, const ICLTensor { case 0: kernel_name = "x"; - idx = num_arguments_per_1D_tensor() * 3; + idx = num_arguments_per_2D_tensor() * 3; break; case 1: kernel_name = "y"; @@ -169,17 +169,17 @@ void CLL2NormalizeLayerKernel::run(const Window &window, cl::CommandQueue &queue case 0: { window_sum.set(Window::DimX, Window::Dimension(0, 0, 0)); - Window in_slice = window.first_slice_window_1D(); - Window sum_slice = window_sum.first_slice_window_1D(); + Window in_slice = window.first_slice_window_2D(); + Window sum_slice = window_sum.first_slice_window_2D(); do { unsigned int idx = 0; - add_1D_tensor_argument(idx, _input, in_slice); - add_1D_tensor_argument(idx, _sum, sum_slice); - add_1D_tensor_argument(idx, _output, in_slice); + add_2D_tensor_argument(idx, _input, in_slice); + add_2D_tensor_argument(idx, _sum, sum_slice); + add_2D_tensor_argument(idx, _output, in_slice); enqueue(queue, *this, in_slice); } - while(window.slide_window_slice_1D(in_slice) && window.slide_window_slice_1D(sum_slice)); + while(window.slide_window_slice_2D(in_slice) && window.slide_window_slice_2D(sum_slice)); } break; case 1: |