diff options
Diffstat (limited to 'src/backends/gpuFsa/layers/GpuFsaResize.cpp')
| -rw-r--r-- | src/backends/gpuFsa/layers/GpuFsaResize.cpp | 79 |
1 files changed, 79 insertions, 0 deletions
diff --git a/src/backends/gpuFsa/layers/GpuFsaResize.cpp b/src/backends/gpuFsa/layers/GpuFsaResize.cpp new file mode 100644 index 0000000000..b46b601bad --- /dev/null +++ b/src/backends/gpuFsa/layers/GpuFsaResize.cpp @@ -0,0 +1,79 @@ +// +// Copyright © 2024 Arm Ltd and Contributors. All rights reserved. +// SPDX-License-Identifier: MIT +// + +#include "GpuFsaResize.hpp" +#include "UtilsGpuFsa.hpp" + +#include <aclCommon/ArmComputeTensorUtils.hpp> + +#include <arm_compute/dynamic_fusion/sketch/gpu/GpuWorkloadContext.h> +#include <arm_compute/dynamic_fusion/sketch/gpu/GpuWorkloadSketch.h> +#include <arm_compute/dynamic_fusion/sketch/gpu/operators/GpuResize.h> +#include <arm_compute/dynamic_fusion/sketch/gpu/operators/GpuOutput.h> + +using namespace arm_compute::experimental::dynamic_fusion; +using namespace armnn::armcomputetensorutils; + +namespace armnn +{ + +arm_compute::Status GpuFsaResizeValidate(const TensorInfo& input, + const ResizeDescriptor& descriptor) +{ + // Create a new workload sketch, for validation purposes + auto compileCtx = arm_compute::CLKernelLibrary::get().get_compile_context(); + auto workloadContext = GpuWorkloadContext(&compileCtx); + GpuWorkloadSketch sketch{ &workloadContext }; + + arm_compute::TensorInfo aclInputInfo = BuildArmComputeTensorInfo(input, descriptor.m_DataLayout); + aclInputInfo.set_are_values_constant(input.IsConstant()); + arm_compute::ITensorInfo* inputInfo = workloadContext.create_tensor_info(aclInputInfo); + + ResizeAttributes resizeAttributes = CreateResizeAttributes(descriptor); + + return GpuResize::validate_op(sketch, inputInfo, resizeAttributes); +} + +void GpuFsaResizeCreateOp(GpuFsaPreCompiledBlob* blob, + const TensorInfo& input, + const ResizeDescriptor& descriptor) +{ + GpuWorkloadSketch* sketch = blob->sketch.get(); + GpuWorkloadContext* workloadContext = blob->workloadContext.get(); + std::vector<arm_compute::ITensorInfo*> inputTensorInfos = {}; + std::vector<arm_compute::ITensorInfo*> outputTensorInfos = {}; + + arm_compute::TensorInfo aclInputInfo = BuildArmComputeTensorInfo(input, descriptor.m_DataLayout); + aclInputInfo.set_are_values_constant(input.IsConstant()); + + inputTensorInfos.emplace_back(workloadContext->create_tensor_info(aclInputInfo)); + + ResizeAttributes resizeAttributes = CreateResizeAttributes(descriptor); + + // Validate operator, check status and update reasonIfUnsupported + arm_compute::Status aclStatus = GpuResize::validate_op(*sketch, + inputTensorInfos[0], + resizeAttributes); + + const bool supported = aclStatus.error_code() == arm_compute::ErrorCode::OK; + if (!supported) + { + throw BackendCapabilityException("\"GpuFsa\" backend failed during resize validation"); + } + + arm_compute::ITensorInfo* addOutputInfo = GpuResize::create_op(*sketch, + inputTensorInfos[0], + resizeAttributes); + + // Temporary fix until fusing attempt is make for GpuFsa backend and Output layer workload is created. + outputTensorInfos.emplace_back(workloadContext->create_tensor_info()); + GpuOutput::create_op(*sketch, addOutputInfo, outputTensorInfos[0]); + + // Store the TensorInfos within the blob as unique_ptrs to be used later + blob->inputTensorInfos = std::make_unique<std::vector<arm_compute::ITensorInfo*>>(inputTensorInfos); + blob->outputTensorInfos = std::make_unique<std::vector<arm_compute::ITensorInfo*>>(outputTensorInfos); +} + +} // namespace armnn
\ No newline at end of file |