diff options
author | Michele Di Giorgio <michele.digiorgio@arm.com> | 2021-01-25 11:49:03 +0000 |
---|---|---|
committer | Michele Di Giorgio <michele.digiorgio@arm.com> | 2021-01-26 18:23:41 +0000 |
commit | 4cfab1886549d7582e3867c51278132e5d37681b (patch) | |
tree | e3d8e502a360581c422787c7452acad77fa1824f /src/runtime/gpu/cl/operators/ClAdd.h | |
parent | 1cf1a7f046f3ff9518cdad21b1f0c164e46345f7 (diff) | |
download | ComputeLibrary-4cfab1886549d7582e3867c51278132e5d37681b.tar.gz |
Make CLArithmeticSubtraction kernel and function state-less
Resolves COMPMID-4008
Change-Id: Ic5f40610e771f31e6d301dfae976c81e9c79fa8b
Signed-off-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4917
Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/runtime/gpu/cl/operators/ClAdd.h')
-rw-r--r-- | src/runtime/gpu/cl/operators/ClAdd.h | 6 |
1 files changed, 3 insertions, 3 deletions
diff --git a/src/runtime/gpu/cl/operators/ClAdd.h b/src/runtime/gpu/cl/operators/ClAdd.h index 2854c16180..f751d8dc83 100644 --- a/src/runtime/gpu/cl/operators/ClAdd.h +++ b/src/runtime/gpu/cl/operators/ClAdd.h @@ -58,11 +58,11 @@ public: * - (QSYMM16,QSYMM16) -> QSYMM16 * * @param[in] compile_context The compile context to be used. - * @param[in, out] src1 First source tensor. Data types supported: U8/QASYMM8/QASYMM8_SIGNED/S16/QSYMM16/S32/F16/F32. + * @param[in, out] src1 First source tensor info. Data types supported: U8/QASYMM8/QASYMM8_SIGNED/S16/QSYMM16/S32/F16/F32. * The source tensor is [in, out] because its TensorInfo might be modified inside the kernel in case of broadcasting of dimension 0. - * @param[in, out] src2 Second source tensor. Data types supported: U8/QASYMM8/QASYMM8_SIGNED/S16/QSYMM16/S32/F16/F32. + * @param[in, out] src2 Second source tensor info. Data types supported: U8/QASYMM8/QASYMM8_SIGNED/S16/QSYMM16/S32/F16/F32. * The source tensor is [in, out] because its TensorInfo might be modified inside the kernel in case of broadcasting of dimension 0. - * @param[out] dst Destination tensor. Data types supported: U8/QASYMM8/QASYMM8_SIGNED/S16/QSYMM16/S32/F16/F32. + * @param[out] dst Destination tensor info. Data types supported: U8/QASYMM8/QASYMM8_SIGNED/S16/QSYMM16/S32/F16/F32. * @param[in] policy Policy to use to handle overflow. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ |