diff options
author | David Monahan <david.monahan@arm.com> | 2023-12-08 12:50:02 +0000 |
---|---|---|
committer | David Monahan <david.monahan@arm.com> | 2024-01-22 16:23:48 +0000 |
commit | bd738081b8fcea4599a06f01d5c07979f3b0fcb3 (patch) | |
tree | fa0c5ef9765e5f4de8551114752473c0a71747cd /src/backends/gpuFsa/layers/GpuFsaConvolution2d.hpp | |
parent | 5bc14146eb310465f3d7c59ac294aa32bc69984a (diff) | |
download | armnn-bd738081b8fcea4599a06f01d5c07979f3b0fcb3.tar.gz |
IVGCVSW-7165 - PreCompiledWorkload and Conv2d Integration work for GpuFsa
* Add PreCompiledWorkload implementation for GpuFsa
* Add ConstantWorkload implementation for GpuFsa
* Add Input/Output workloads for GpuFsa
* Added CopyMemGeneric workload for GpuFsa
* Separate creation and validation of sketch tensors into seperate functions
Signed-off-by: Kevin May <kevin.may@arm.com>
Signed-off-by: David Monahan <david.monahan@arm.com>
Change-Id: Ie7299a4c61073b5ca03d9f8681458869ef7ce743
Diffstat (limited to 'src/backends/gpuFsa/layers/GpuFsaConvolution2d.hpp')
-rw-r--r-- | src/backends/gpuFsa/layers/GpuFsaConvolution2d.hpp | 30 |
1 files changed, 30 insertions, 0 deletions
diff --git a/src/backends/gpuFsa/layers/GpuFsaConvolution2d.hpp b/src/backends/gpuFsa/layers/GpuFsaConvolution2d.hpp new file mode 100644 index 0000000000..3346dc1028 --- /dev/null +++ b/src/backends/gpuFsa/layers/GpuFsaConvolution2d.hpp @@ -0,0 +1,30 @@ +// +// Copyright © 2023-2024 Arm Ltd and Contributors. All rights reserved. +// SPDX-License-Identifier: MIT +// +#pragma once + +#include <armnn/Descriptors.hpp> +#include <armnn/Tensor.hpp> + +#include <arm_compute/core/Error.h> +#include <arm_compute/dynamic_fusion/sketch/gpu/GpuWorkloadSketch.h> +#include <gpuFsa/GpuFsaBackend.hpp> + +namespace armnn +{ + +using namespace arm_compute::experimental::dynamic_fusion; + +arm_compute::Status GpuFsaConvolution2dValidate(const TensorInfo& input, + const Convolution2dDescriptor& descriptor, + const TensorInfo& weights, + const Optional<TensorInfo>& biases); + +void GpuFsaConvolution2dCreateOp(GpuFsaPreCompiledBlob* blob, + const TensorInfo& input, + const Convolution2dDescriptor& descriptor, + const TensorInfo& weights, + const Optional<TensorInfo>& biases); + +} // namespace armnn |