Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

python: trim registration and loading of dialects and passes #1084

Merged
merged 1 commit into from
Jul 21, 2022
Merged

python: trim registration and loading of dialects and passes #1084

merged 1 commit into from
Jul 21, 2022

Conversation

ashay
Copy link
Collaborator

@ashay ashay commented Jul 19, 2022

In the interest of merging upstream LLVM quickly, a previous patch
(7f08169) updated the torch-mlir build to register all dialects and
passes through Python bindings. This patch limits the dialects and
passes to only those that are used in torch-mlir.

Key to this change are the removal of
MLIRPythonExtension.RegisterEverything and the introduction of a new
Python module (_mlir_libs/_site_initialize_0.py), where we register
the dialects and passes used by torch-mlir.

@ashay ashay marked this pull request as draft July 19, 2022 19:41
@silvasean
Copy link
Contributor

Nice :)

@ashay ashay marked this pull request as ready for review July 19, 2022 21:20
@ashay ashay marked this pull request as draft July 19, 2022 21:26
@ashay
Copy link
Collaborator Author

ashay commented Jul 20, 2022

I just now found out that the build error is due to the use of the default linker (ld?). I am using lld on my local machine, which doesn't produce the error. Still investigating.

Copy link
Collaborator

@stellaraccident stellaraccident left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks. I left some notes. I suspect that the C/C++ cross linking thing is actually what is tripping up ld, since what you are doing is introducing a kind of ODR violation.

I don't have time tonight, but if you struggle with this, let me know and I can straighten it out.

python/CMakeLists.txt Outdated Show resolved Hide resolved
python/CMakeLists.txt Show resolved Hide resolved
python/TorchMLIRModule.cpp Outdated Show resolved Hide resolved
python/TorchMLIRModule.cpp Outdated Show resolved Hide resolved
python/torch_mlir/_mlir_libs/_mlirRegisterEverything.py Outdated Show resolved Hide resolved
Copy link
Collaborator

@stellaraccident stellaraccident left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Much better - thanks!

@ashay
Copy link
Collaborator Author

ashay commented Jul 20, 2022

I suspect that the C/C++ cross linking thing is actually what is tripping up ld, since what you are doing is introducing a kind of ODR violation.

Thanks so much, that was spot on! I believe I have separated the C and C++ APIs now, but please let me know if you think more changes are necessary. Locally, using ld now works, so hopefully CI will pass as well.

@ashay ashay marked this pull request as ready for review July 20, 2022 23:29
In the interest of merging upstream LLVM quickly, a previous patch
(7f08169) updated the torch-mlir build to register all dialects and
passes through Python bindings.  This patch limits the dialects and
passes to only those that are used in torch-mlir.

Key to this change are the removal of
`MLIRPythonExtension.RegisterEverything` and the introduction of a new
Python module (`_mlir_libs/_site_initialize_0.py`), where we register
the dialects and passes used by torch-mlir.
@ashay ashay merged commit ad283c1 into llvm:main Jul 21, 2022
@ashay ashay deleted the ashay/trim-python-linking branch July 21, 2022 01:34
@stellaraccident
Copy link
Collaborator

I suspect that the C/C++ cross linking thing is actually what is tripping up ld, since what you are doing is introducing a kind of ODR violation.

Thanks so much, that was spot on! I believe I have separated the C and C++ APIs now, but please let me know if you think more changes are necessary. Locally, using ld now works, so hopefully CI will pass as well.

I wish it didn't rely on such esoteric knowledge to spot such things. But glad it helped regardless :) Thanks for doing the adaptations! Much appreciated.

@powderluv
Copy link
Collaborator

@ashay could this be causing the nightly failure and also the

 ImportError: dlopen(/Users/kkiningh/Workspace/pytorch-mlir/torch-mlir/build/tools/torch-mlir/python_packages/torch_mlir/torch_mlir/_mlir_libs/_mlirRegisterEverything.cpython-39-darwin.so, 0x0002): symbol not found 
in flat namespace '_mlirRegisterAllDialects'  

reported on discord ?

gpetters94 pushed a commit to gpetters94/mlir-npcomp that referenced this pull request Jul 27, 2022
In the interest of merging upstream LLVM quickly, a previous patch
(7f08169) updated the torch-mlir build to register all dialects and
passes through Python bindings.  This patch limits the dialects and
passes to only those that are used in torch-mlir.

Key to this change are the removal of
`MLIRPythonExtension.RegisterEverything` and the introduction of a new
Python module (`_mlir_libs/_site_initialize_0.py`), where we register
the dialects and passes used by torch-mlir.
qedawkins pushed a commit to nod-ai/torch-mlir that referenced this pull request Oct 3, 2022
* initial docker documentation

Signed-off-by: Alexandre Eichenberger <[email protected]>

* split README with no redundant place for info

Signed-off-by: Alexandre Eichenberger <[email protected]>

* update

Signed-off-by: Alexandre Eichenberger <[email protected]>

* update

Signed-off-by: Alexandre Eichenberger <[email protected]>

* update

Signed-off-by: Alexandre Eichenberger <[email protected]>

* update

Signed-off-by: Alexandre Eichenberger <[email protected]>

* update

Signed-off-by: Alexandre Eichenberger <[email protected]>

* respond to suggestions

Signed-off-by: Alexandre Eichenberger <[email protected]>

* specify that onnx-mlir.py script generates only code suitable to be exec in Linux and/or Docker env

Signed-off-by: Alexandre Eichenberger <[email protected]>

* fix checkdocs

Signed-off-by: Alexandre Eichenberger <[email protected]>

* responded to review suggestion on onnx-mlir --help

Signed-off-by: Alexandre Eichenberger <[email protected]>

* use ONNX-MLIR everywhere

Signed-off-by: Alexandre Eichenberger <[email protected]>
qedawkins pushed a commit to nod-ai/torch-mlir that referenced this pull request Oct 3, 2022
* Add check-onnx-backend to Mac CI. (llvm#1069)

* Add check-onnx-backend to Mac CI.

Signed-off-by: Ettore Tiotto <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* Additional Docker help and split README for easier reading (llvm#1084)

* initial docker documentation

Signed-off-by: Alexandre Eichenberger <[email protected]>

* split README with no redundant place for info

Signed-off-by: Alexandre Eichenberger <[email protected]>

* update

Signed-off-by: Alexandre Eichenberger <[email protected]>

* update

Signed-off-by: Alexandre Eichenberger <[email protected]>

* update

Signed-off-by: Alexandre Eichenberger <[email protected]>

* update

Signed-off-by: Alexandre Eichenberger <[email protected]>

* update

Signed-off-by: Alexandre Eichenberger <[email protected]>

* respond to suggestions

Signed-off-by: Alexandre Eichenberger <[email protected]>

* specify that onnx-mlir.py script generates only code suitable to be exec in Linux and/or Docker env

Signed-off-by: Alexandre Eichenberger <[email protected]>

* fix checkdocs

Signed-off-by: Alexandre Eichenberger <[email protected]>

* responded to review suggestion on onnx-mlir --help

Signed-off-by: Alexandre Eichenberger <[email protected]>

* use ONNX-MLIR everywhere

Signed-off-by: Alexandre Eichenberger <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* add verify for concat

Signed-off-by: Ethan Wang <[email protected]>

* check all inputs

Signed-off-by: Ethan Wang <[email protected]>

* Support filtering out lit tests based on targets (llvm#1087)

Currently we ignore what targets llvm was built for in the lit tests, but recent changes to onnx-mlir explicitly initialize the available targets.
This makes the corresponding change to the lit configuration, so that we can filter out the lit tests based on the available targets.

Signed-off-by: Stella Stamenova <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* Switch URLs to use main instead of master (llvm#1094)

Signed-off-by: Charles Volzka <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* Fix MacOS build badge (llvm#1092)

Signed-off-by: Gong Su <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* onnx-mlir.py warning about binary output (.so and .jar) (llvm#1090)

not directly usable if host is not Linux

Signed-off-by: Gong Su <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* Make the doc example obey ONNX_MLIR_BUILD_TESTS (llvm#1083)

* Make the doc example obey ONNX_MLIR_BUILD_TESTS

Currently, ONNX_MLIR_BUILD_TESTS controls EXCLUDE_FROM_ALL, however, the targets added through add_executable will always build. We follow the llvm pattern and explicitly set EXCLUDE_FROM_ALL in the add_onnx_mlir_executable function if it is set for the directory, so that add_executable targets don't always build.

Signed-off-by: Stella Stamenova <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* Explicitly install into lib on all systems (llvm#1088)

Signed-off-by: Gong Su <[email protected]>

Co-authored-by: Alexandre Eichenberger <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* add check (llvm#1098)

Signed-off-by: Tong Chen <[email protected]>

Co-authored-by: Alexandre Eichenberger <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* fix typos and add ssh-client to dockerfile (llvm#1096)

* fix typos and add ssh-client to dockerfile

Signed-off-by: Ethan Wang <[email protected]>

* sync doc and script

Signed-off-by: Ethan Wang <[email protected]>

Co-authored-by: Alexandre Eichenberger <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* Emit print statement only when the verbose option is in effect. (llvm#1097)

Signed-off-by: Ettore Tiotto <[email protected]>

Co-authored-by: Alexandre Eichenberger <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* format & refine code by request

Signed-off-by: Ethan Wang <[email protected]>

* Support older versions 6, 11, 12 for Clip Op (llvm#1100)

Signed-off-by: Tung D. Le <[email protected]>

Co-authored-by: Alexandre Eichenberger <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* using front to get first input

Signed-off-by: Ethan Wang <[email protected]>

* add 3 lit test for concat  verifier

Signed-off-by: Ethan Wang <[email protected]>

* add newline

Signed-off-by: Ethan Wang <[email protected]>

* Add check-onnx-backend to Mac CI. (llvm#1069)

* Add check-onnx-backend to Mac CI.

Signed-off-by: Ettore Tiotto <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* Additional Docker help and split README for easier reading (llvm#1084)

* initial docker documentation

Signed-off-by: Alexandre Eichenberger <[email protected]>

* split README with no redundant place for info

Signed-off-by: Alexandre Eichenberger <[email protected]>

* update

Signed-off-by: Alexandre Eichenberger <[email protected]>

* update

Signed-off-by: Alexandre Eichenberger <[email protected]>

* update

Signed-off-by: Alexandre Eichenberger <[email protected]>

* update

Signed-off-by: Alexandre Eichenberger <[email protected]>

* update

Signed-off-by: Alexandre Eichenberger <[email protected]>

* respond to suggestions

Signed-off-by: Alexandre Eichenberger <[email protected]>

* specify that onnx-mlir.py script generates only code suitable to be exec in Linux and/or Docker env

Signed-off-by: Alexandre Eichenberger <[email protected]>

* fix checkdocs

Signed-off-by: Alexandre Eichenberger <[email protected]>

* responded to review suggestion on onnx-mlir --help

Signed-off-by: Alexandre Eichenberger <[email protected]>

* use ONNX-MLIR everywhere

Signed-off-by: Alexandre Eichenberger <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* Switch URLs to use main instead of master (llvm#1094)

Signed-off-by: Charles Volzka <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* Fix MacOS build badge (llvm#1092)

Signed-off-by: Gong Su <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* fix typos and add ssh-client to dockerfile (llvm#1096)

* fix typos and add ssh-client to dockerfile

Signed-off-by: Ethan Wang <[email protected]>

* sync doc and script

Signed-off-by: Ethan Wang <[email protected]>

Co-authored-by: Alexandre Eichenberger <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* Update document (llvm#1077)

* create

Signed-off-by: Tong Chen <[email protected]>

* delete HowTOAddAnOperation.md

Signed-off-by: Tong Chen <[email protected]>

* modify testing

Signed-off-by: Tong Chen <[email protected]>

* create

Signed-off-by: Tong Chen <[email protected]>

* delete HowTOAddAnOperation.md

Signed-off-by: Tong Chen <[email protected]>

* modify testing

Signed-off-by: Tong Chen <[email protected]>

* fix

Signed-off-by: Tong Chen <[email protected]>

* create

Signed-off-by: Tong Chen <[email protected]>

* add comment

Signed-off-by: Tong Chen <[email protected]>

* delete HowTOAddAnOperation.md

Signed-off-by: Tong Chen <[email protected]>

* modify testing

Signed-off-by: Tong Chen <[email protected]>

* fix

Signed-off-by: Tong Chen <[email protected]>

* create

Signed-off-by: Tong Chen <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* Update LLVM level (llvm#1095)

* Update LLVM level to 700997a

Signed-off-by: Ettore Tiotto <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* Pass a type converter to all ONNX operations. (llvm#1102)

Signed-off-by: Ettore Tiotto <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* Nuke KrnlDummyCastOp now that we use MLIR's UnrealizedConversionCastOp (llvm#1103)

* Nuke KrnlDummyCastOp now that we use MLIR's UnrealizedConversionCastOp

Signed-off-by: Ettore Tiotto <[email protected]>

* Remove a dependency in src/Dialect/Krnl/CMakeList.txt.  Regenerate docs via 'ninja onnx-mlir-docs'.

Signed-off-by: Ettore Tiotto <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* Add --emitObj option to onnx-mlir (llvm#1104)

Signed-off-by: Ettore Tiotto <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* fix warnings (llvm#1093)

Signed-off-by: Ian Bearman <[email protected]>

Co-authored-by: Stella Stamenova <[email protected]>
Co-authored-by: Alexandre Eichenberger <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* Add -march option to onnx-mlir (llvm#1107)

Signed-off-by: Ettore Tiotto <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* Fix Doc spelling and broken links, removed warnings about using main (llvm#1106)

* removed warning about main vs master in CONTRIBUTING, fixed links and spelling mistakes

Signed-off-by: Alexandre Eichenberger <[email protected]>
Signed-off-by: Ethan Wang <[email protected]>

* Update BuildONNX.md

Signed-off-by: Ethan Wang <[email protected]>

Co-authored-by: Ettore Tiotto <[email protected]>
Co-authored-by: Alexandre Eichenberger <[email protected]>
Co-authored-by: Stella Stamenova <[email protected]>
Co-authored-by: Charles Volzka <[email protected]>
Co-authored-by: gongsu832 <[email protected]>
Co-authored-by: chentong319 <[email protected]>
Co-authored-by: Tung D. Le <[email protected]>
Co-authored-by: Ian Bearman <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants