[Pipeline] Use dedicated simplification pipeline for TorchDynamo frontend #3376
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
sjain-stanford
changed the title
sjain-stanford
force-pushed
the
sambhav/fx_pipeline_fixes
branch
from
7e89083 to
2cb35fc
Compare
penguin-wwy
approved these changes
May 22, 2024
|
Can we get a post review from @rsuderman ? While probably ok, I wish we had left time for a design change like this to get reviewed by all parties vs an author and merge in 2 hours. |
|
Appreciate the note @stellaraccident , I will be on standby for @rsuderman 's review and send a follow-on to address/revise/revert as appropriate. |
BaneTrifa
pushed a commit
to BaneTrifa/torch-mlir
that referenced
this pull request
May 24, 2024
…tend (llvm#3376) Discord Thread: https://discord.com/channels/636084430946959380/1238330633328005243 ## Context: [This](https://github.com/llvm/torch-mlir/blob/main/python/torch_mlir/fx.py#L61) was updated to support e2e tests for the TorchDynamo frontend in Torch-MLIR, where we run FX decompositions and import the FX IR to generate Torch dialect, followed by `torch-function-to-torch-backend-pipeline`, skipping only the shape/type refinement for now. However, we should be able to skip many of the torch simplification passes, as depicted in the [frontend roadmap](https://github.com/llvm/torch-mlir/blob/main/docs/images/roadmap_frontend.png). Based on IREE's TorchDynamo [pipeline](https://github.com/iree-org/iree/blob/main/compiler/plugins/input/Torch/InputConversion/Passes.cpp#L29), the only two passes we seem to require are: `ReduceOpVariantsPass` and `DecomposeComplexOpsPass`. This is inline with our findings as well based on initial exploration. This PR creates a dedicated frontend simplification pipeline for TorchDynamo / FX Importer which calls only `ReduceOpVariantsPass` and `DecomposeComplexOpsPass`. We rely on the e2e fx_importer tests to ensure we're not regressing by removing many of the passes that were historically needed for TorchScript. One notable change here is that we do not call the `LowerToBackendContractPass` anymore, which used to call `TorchSimplificationPipeline` iteratively until VerifyBackendContract was clean. Some of this was required for the shape/type refinement to converge, which seems a non-issue for Dynamo frontend. Do we anticipate this (the iterative invocation of TorchSimplificationPipeline followed by VerifyBackendContract) to be worth retaining in the Dynamo frontend pipeline? If so, I can make those changes, PLMK.
sjarus
pushed a commit
to sjarus/torch-mlir
that referenced
this pull request
Jun 6, 2024
…tend (llvm#3376) Discord Thread: https://discord.com/channels/636084430946959380/1238330633328005243 ## Context: [This](https://github.com/llvm/torch-mlir/blob/main/python/torch_mlir/fx.py#L61) was updated to support e2e tests for the TorchDynamo frontend in Torch-MLIR, where we run FX decompositions and import the FX IR to generate Torch dialect, followed by `torch-function-to-torch-backend-pipeline`, skipping only the shape/type refinement for now. However, we should be able to skip many of the torch simplification passes, as depicted in the [frontend roadmap](https://github.com/llvm/torch-mlir/blob/main/docs/images/roadmap_frontend.png). Based on IREE's TorchDynamo [pipeline](https://github.com/iree-org/iree/blob/main/compiler/plugins/input/Torch/InputConversion/Passes.cpp#L29), the only two passes we seem to require are: `ReduceOpVariantsPass` and `DecomposeComplexOpsPass`. This is inline with our findings as well based on initial exploration. This PR creates a dedicated frontend simplification pipeline for TorchDynamo / FX Importer which calls only `ReduceOpVariantsPass` and `DecomposeComplexOpsPass`. We rely on the e2e fx_importer tests to ensure we're not regressing by removing many of the passes that were historically needed for TorchScript. One notable change here is that we do not call the `LowerToBackendContractPass` anymore, which used to call `TorchSimplificationPipeline` iteratively until VerifyBackendContract was clean. Some of this was required for the shape/type refinement to converge, which seems a non-issue for Dynamo frontend. Do we anticipate this (the iterative invocation of TorchSimplificationPipeline followed by VerifyBackendContract) to be worth retaining in the Dynamo frontend pipeline? If so, I can make those changes, PLMK.
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.
Discord Thread: https://discord.com/channels/636084430946959380/1238330633328005243
Context:
This was updated to support e2e tests for the TorchDynamo frontend in Torch-MLIR, where we run FX decompositions and import the FX IR to generate Torch dialect, followed by
torch-function-to-torch-backend-pipeline, skipping only the shape/type refinement for now. However, we should be able to skip many of the torch simplification passes, as depicted in the frontend roadmap.Based on IREE's TorchDynamo pipeline, the only two passes we seem to require are:
ReduceOpVariantsPassandDecomposeComplexOpsPass. This is inline with our findings as well based on initial exploration.This PR creates a dedicated frontend simplification pipeline for TorchDynamo / FX Importer which calls only
ReduceOpVariantsPassandDecomposeComplexOpsPass. We rely on the e2e fx_importer tests to ensure we're not regressing by removing many of the passes that were historically needed for TorchScript.One notable change here is that we do not call the
LowerToBackendContractPassanymore, which used to callTorchSimplificationPipelineiteratively until VerifyBackendContract was clean. Some of this was required for the shape/type refinement to converge, which seems a non-issue for Dynamo frontend. Do we anticipate this (the iterative invocation of TorchSimplificationPipeline followed by VerifyBackendContract) to be worth retaining in the Dynamo frontend pipeline? If so, I can make those changes, PLMK.