Update DirectML 1.5.1 to 1.8.0 for ORT1.10 #9765
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
MaskRCNN failed when `Cast` tried to execute `Xor` with emptiness (zero in dimensions). This is perfectly legal and should be treated as a nop. Ultimately DML itself should treat this case as a nop, just like how C's `memcpy` treats 0 count as a nop, but I'm just addressing it in ORT now, as enabling it in DML would impact more operators to be consistent (probably should incrementally add a flag to tensor validation so operators can be opted in gradually). Corresponding WindowsAI PR: https://microsoft.visualstudio.com/WindowsAI/_git/WindowsAI/pullrequest/5195850 Related work items: #27469839, #28761382
When used in ORT, a common method shouldn't copy and return initializer data Related work items: #29514403
Tensors that contain 0-sized dimensions were being broadcasted to higher dimensions, which would remove the possibility to remove them from the graph. 0-sized dimensions represent empty tensors, so whatever operator needs to broadcast it shouldn't try to call into DML.
This extends a workaround needed to match node inputs with Tensors to the EP code handling constant input upload. This was causing issues in a couple of models, including EfficientDet, although that model still fails due to this bug: https://microsoft.visualstudio.com/OS/_workitems/edit/29970551 Related work items: #29706035
GPU timeouts have already been disabled in command queues created by Winml, but not the ones created by the DML EP within the ORT API
…ut size New validation [here](https://microsoft.visualstudio.com/DefaultCollection/WindowsAI/_git/WindowsAI/pullrequest/5354070?_a=files&path=%2Fdml%2FSharedValidation%2FDmlBatchNormalizationOperatorValidator.h) causes some BatchNorm cases to fail (e.g. OnnxConformanceTestsTaef::BatchNormalization (BatchNormalization_2x2x2)). I'm unsure how long this bug existed, but based on Nick's investigation, it apparently still worked anyway. Related work items: #27678610
Update 8D BatchNorm Related work items: #27678610
0 is valid in Tile in "repeats" parameter. The CPU kernel handles it fine. So should the DML EP. Related work items: #29970551
…r into DmlDev I'm just doing this manually this once because some commits were squashed, and I want to be sure they merge cleanly for the nightly automatic merge. "git merge upstream/master"
Add opset 13 for ops which are unchanged. The culled list, removing any operators that changed signature (beyond just data type) or any that we don't support: Abs-13 Add-13 ArgMax-13 ArgMin-13 Cast-13 Ceil-13 Clip-13 Concat-13 Constant-13 DepthToSpace-13 Div-13 Equal-13 Erf-13 Exp-13 Expand-13 Flatten-13 Floor-13 Gather-13 GatherElements-13 GatherND-13 Gemm-13 Greater-13 Identity-13 IsNaN-13 LRN-13 Less-13 Log-13 MatMul-13 Max-13 Mean-13 MeanVarianceNormalization-13 Min-13 Mod-13 Mul-13 Neg-13 Pad-13 Pow-13 Reciprocal-13 ReduceL1-13 ReduceL2-13 ReduceLogSum-13 ReduceLogSumExp-13 ReduceMax-13 ReduceMean-13 ReduceMin-13 ReduceProd-13 ReduceSumSquare-13 Relu-13 Reshape-13 Scatter-13 ScatterElements-13 ScatterND-13 Sigmoid-13 Sign-13 Slice-13 SpaceToDepth-13 Sqrt-13 Sub-13 Sum-13 Tanh-13 Tile-13 Transpose-13 Upsample-13 Related work items: #36946821
https://microsoft.visualstudio.com/WindowsAI/_build/results?buildId=41740400&view=results  
fdwr
force-pushed
the
user/dwayner/DML1.8forORT1.10
branch
from
90720cb
to
e0ffc30
Compare
fdwr
force-pushed
the
user/dwayner/DML1.8forORT1.10
branch
from
22f650b
to
289b1bd
Compare
fdwr
changed the title
jeffbloo
previously approved these changes
Nov 19, 2021
jeffbloo
approved these changes
Nov 20, 2021
|
Azure Pipelines successfully started running 1 pipeline(s). |
|
/azp run Linux CPU CI Pipeline, Linux CPU Minimal Build E2E CI Pipeline, Linux GPU CI Pipeline, Linux GPU TensorRT CI Pipeline, Linux Nuphar CI Pipeline, Linux OpenVINO CI Pipeline, MacOS CI Pipeline, ONNX Runtime Web CI Pipeline, Windows CPU CI Pipeline |
|
You have several pipelines (over 10) configured to build pull requests in this repository. Specify which pipelines you would like to run by using /azp run [pipelines] command. You can specify multiple pipelines using a comma separated list. |
|
Azure Pipelines successfully started running 9 pipeline(s). |
|
/azp run Windows GPU TensorRT CI Pipeline, onnxruntime-python-checks-ci-pipeline, orttraining-amd-gpu-ci-pipeline, orttraining-linux-ci-pipeline, orttraining-linux-gpu-ci-pipeline, orttraining-ortmodule-distributed |
|
Azure Pipelines successfully started running 6 pipeline(s). |
This was referenced Dec 8, 2021
Closed
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description:
Motivation and Context