diff options
author | Sheri Zhang <sheri.zhang@arm.com> | 2021-02-10 15:32:38 +0000 |
---|---|---|
committer | Michele Di Giorgio <michele.digiorgio@arm.com> | 2021-02-22 11:12:23 +0000 |
commit | ac6499a474503dfcf587d95b85823aeb5861103f (patch) | |
tree | fc0bdc870d564ef45468a5033f954b10505ec956 /src/core/NEON/kernels/NEDirectConvolutionLayerKernel.cpp | |
parent | 669d17d9055089be1a20f6984630b95019fe2e78 (diff) | |
download | ComputeLibrary-ac6499a474503dfcf587d95b85823aeb5861103f.tar.gz |
Comply with Trademark rules for use of Neon, Arm and Mali
Full trademarks available in README.md
Resolves: COMPMID-4257
Signed-off-by: Sheri Zhang <sheri.zhang@arm.com>
Change-Id: Ibfba2adf2eef3449433f467464ebd87d7198474d
Signed-off-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/5116
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/core/NEON/kernels/NEDirectConvolutionLayerKernel.cpp')
-rw-r--r-- | src/core/NEON/kernels/NEDirectConvolutionLayerKernel.cpp | 6 |
1 files changed, 3 insertions, 3 deletions
diff --git a/src/core/NEON/kernels/NEDirectConvolutionLayerKernel.cpp b/src/core/NEON/kernels/NEDirectConvolutionLayerKernel.cpp index 87b9fb1bf1..77742163fa 100644 --- a/src/core/NEON/kernels/NEDirectConvolutionLayerKernel.cpp +++ b/src/core/NEON/kernels/NEDirectConvolutionLayerKernel.cpp @@ -1,5 +1,5 @@ /* - * Copyright (c) 2017-2020 Arm Limited. + * Copyright (c) 2017-2021 Arm Limited. * * SPDX-License-Identifier: MIT * @@ -149,7 +149,7 @@ inline bool run_optim_small_tensor(const ITensor *t) // Optimized convolver for 1x1 kernels used only where input width and height are both <= 8 // For big Z as in Input=7x7x832, this implementation is faster than the general code becuase it doesn't need to -// store intermidiate results in memory. Temporary results are stored in NEON registers directly and then written to the output buffer. +// store intermidiate results in memory. Temporary results are stored in Neon registers directly and then written to the output buffer. template <unsigned int stridex> class convolver_w1x1_i8x8_f32 { @@ -1381,4 +1381,4 @@ void NEDirectConvolutionLayerKernel::run(const Window &window, const ThreadInfo } } } -} // namespace arm_compute
\ No newline at end of file +} // namespace arm_compute |