diff options
| author | Michele Di Giorgio <michele.digiorgio@arm.com> | 2020-06-04 15:05:38 +0100 |
|---|---|---|
| committer | Michele Di Giorgio <michele.digiorgio@arm.com> | 2020-06-15 13:59:04 +0000 |
| commit | 4a61653202afb018f4f259d3c144a735d73f0a20 (patch) | |
| tree | 082fd42e91cc0914dcacc0746bbe3e117d74210c /arm_compute/core/NEON | |
| parent | ccd94966cc58ef5148577e71ba1a4ff5aae1f3bb (diff) | |
| download | ComputeLibrary-4a61653202afb018f4f259d3c144a735d73f0a20.tar.gz | |
COMPMID-3480: Perform in-place computations in NEArithmeticAdditionKernel
Change-Id: I0089657dd95d7c7b8592984def8e8de1d7e6d085
Signed-off-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3308
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'arm_compute/core/NEON')
| -rw-r--r-- | arm_compute/core/NEON/kernels/NEArithmeticAdditionKernel.h | 20 |
1 files changed, 10 insertions, 10 deletions
diff --git a/arm_compute/core/NEON/kernels/NEArithmeticAdditionKernel.h b/arm_compute/core/NEON/kernels/NEArithmeticAdditionKernel.h index bff34dfda2..50c90b521a 100644 --- a/arm_compute/core/NEON/kernels/NEArithmeticAdditionKernel.h +++ b/arm_compute/core/NEON/kernels/NEArithmeticAdditionKernel.h @@ -68,12 +68,12 @@ public: * - (QASYMM8_SIGNED,QASYMM8_SIGNED) -> QASYMM8_SIGNED * - (QSYMM16,QSYMM16) -> QSYMM16 * - * @param[in] input1 First input tensor. Data types supported: U8/QASYMM8/QASYMM8_SIGNED/S16/QSYMM16/F16/S32/F32 - * @param[in] input2 Second input tensor. Data types supported: U8/QASYMM8/QASYMM8_SIGNED/S16/QSYMM16/F16/S32/F32 - * @param[out] output The output tensor. Data types supported: U8/QASYMM8/QASYMM8_SIGNED/S16/QSYMM16/F16/S32/F32. - * @param[in] policy Overflow policy. + * @param[in, out] input1 First input tensor. Data types supported: U8/QASYMM8/QASYMM8_SIGNED/S16/QSYMM16/F16/S32/F32 + * @param[in] input2 Second input tensor. Data types supported: U8/QASYMM8/QASYMM8_SIGNED/S16/QSYMM16/F16/S32/F32 + * @param[out] output (Optional) The output tensor. Can be nullptr. Data types supported: U8/QASYMM8/QASYMM8_SIGNED/S16/QSYMM16/F16/S32/F32. + * @param[in] policy Overflow policy. */ - void configure(const ITensor *input1, const ITensor *input2, ITensor *output, ConvertPolicy policy); + void configure(ITensor *input1, const ITensor *input2, ITensor *output, ConvertPolicy policy); /** Static function to check if given info will lead to a valid configuration of @ref NEArithmeticAdditionKernel * * @param[in] input1 First input tensor info. Data types supported: U8/QASYMM8/QASYMM8_SIGNED/S16/QSYMM16/F16/S32/F32 @@ -91,11 +91,11 @@ public: private: /** Common signature for all the specialised add functions * - * @param[in] input1 First input tensor. Data types supported: U8/QASYMM8/S16/QSYMM16/F16/S32/F32 - * @param[in] input2 Second input tensor. Data types supported: U8/QASYMM8/S16/QSYMM16/F16/S32/F32 - * @param[out] output The output tensor. Data types supported: U8/QASYMM8/S16/QSYMM16/F16/S32/F32. - * @param[in] policy Overflow policy. - * @param[in] window Region on which to execute the kernel. + * @param[in, out] input1 First input tensor. Data types supported: U8/QASYMM8/S16/QSYMM16/F16/S32/F32 + * @param[in] input2 Second input tensor. Data types supported: U8/QASYMM8/S16/QSYMM16/F16/S32/F32 + * @param[out] output (Optional) The output tensor. Can be nullptr. Data types supported: U8/QASYMM8/S16/QSYMM16/F16/S32/F32. + * @param[in] policy Overflow policy. + * @param[in] window Region on which to execute the kernel. */ using AddFunction = void(const ITensor *input1, const ITensor *input2, ITensor *output, ConvertPolicy policy, const Window &window); /** Add function to use for the particular tensor types passed to configure() */ |