Initial PR for VSINPU execution provider

[chenfeiyue-cfy]

Description

-It is an initial PR for VSINPU execution provider

Motivation and Context

• For support VeriSilicon hardware
• TIM-VX(Tensor Interface Module) (https://github.com/VeriSilicon/TIM-VX) is an integrated software solution by Verisilicon for our hardware(A311D/i.MX 8M Plus etc.) design, it is easy to use Verisilicon’s hardware by simply connecting onnxruntime with the TIM-VX API by this VSINPU execution provider.

[chenfeiyue-cfy]

@sunshinemyson @thezha Could you please confirm if we can sign this CLA？

[chenfeiyue-cfy]

@microsoft-github-policy-service agree company="VeriSilicon"

[jywu-msft]

/azp run Windows ARM64 QNN CI Pipeline,Windows x64 QNN CI Pipeline,Windows CPU CI Pipeline,Windows GPU CI Pipeline,Windows GPU TensorRT CI Pipeline,ONNX Runtime Web CI Pipeline,Linux CPU CI Pipeline,Linux CPU Minimal Build E2E CI Pipeline,Linux GPU CI Pipeline,Linux GPU TensorRT CI Pipeline

[jywu-msft]

/azp run Linux OpenVINO CI Pipeline,Linux QNN CI Pipeline,MacOS CI Pipeline,orttraining-amd-gpu-ci-pipeline,orttraining-linux-ci-pipeline,orttraining-linux-gpu-ci-pipeline,orttraining-ortmodule-distributed,onnxruntime-binary-size-checks-ci-pipeline,Big Models,Linux Android Emulator QNN CI Pipeline

[azure-pipelines]

Azure Pipelines successfully started running 10 pipeline(s).

[azure-pipelines]

Azure Pipelines successfully started running 10 pipeline(s).

[jywu-msft]

can you provide some details about the patch files under vsinpu/patches? why are they there?

[chenfeiyue-cfy]

can you provide some details about the patch files under vsinpu/patches? why are they there?

Sure! There are three patches under this directory.I will introduce the function of each patch file separately.

• AccuracyCorrection.patch: This patch is used to address the precision discrepancy (with a difference of 1) in the uint8 data type.
• local_testing_record_res.patch: It is a patch of workaround method. The purpose of this patch is to temporarily and informally record each output in order, making it easier to compare the correctness and similarity between our hardware results and the golden values using other scripts later.
• mlas_crosscompiling.patch: We need to temporary disable mlas_fp16 under our ARM64 HWs, because there exists conflict of compiliation flag when cross compiling the project. This patch is only needed on our some Specific HW.