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

Add a CI test to validate shapelib #1317

Closed
powderluv opened this issue Aug 30, 2022 · 6 comments
Closed

Add a CI test to validate shapelib #1317

powderluv opened this issue Aug 30, 2022 · 6 comments
Assignees

Comments

@powderluv
Copy link
Collaborator

Currently CI has a test to

    - name: Ensure generated files are up to date	
      if: ${{ matrix.os-arch == 'ubuntu-x86_64' && matrix.llvm-build == 'in-tree' }}	
      run: |	
        ./build_tools/update_torch_ods.sh	
        ./build_tools/update_shape_lib.sh	
        if ! git diff --quiet; then	
          echo "#######################################################"	
          echo "Generated files are not up to date (see diff below)"	
          echo ">>> Please run ./build_tools/update_torch_ods.sh and ./build_tools/update_shape_lib.sh <<<"	
          echo "#######################################################"	
          git diff --color=always	
          exit 1	
        fi
        ```

Add the equivalent in the Docker CI image. 
@silvasean
Copy link
Contributor

Is there a way that we can avoid duplicating the tests in build_linux_packages.sh and buildAndTest.yml? It seems redundant to specify in both.

@powderluv
Copy link
Collaborator Author

There should be no tests in buildAndTest.yml (for linux).

@powderluv
Copy link
Collaborator Author

So I would like to understand what is the expected behaviour of the test so we slot it in correctly:

When do you expect there to be a valid difference to fail the build ? If latest PyTorch was updated we fail the local CI and ask the developer to update shape / torch ?

@silvasean
Copy link
Contributor

Yes. For now, it will be like any breaking PyTorch change -- it will cause local builds to fail. The solution is to have a deliberate weekly PyTorch bump.

@powderluv
Copy link
Collaborator Author

#1328 to track the periodic PyTorch bump

qedawkins pushed a commit to nod-ai/torch-mlir that referenced this issue Oct 3, 2022
…l loops (llvm#1317)

[LRN][Pooling][Loop] Use builder based interface to generate Krnl loops (Only partially done)

Signed-off-by: Whitney Tsang <[email protected]>
qedawkins pushed a commit to nod-ai/torch-mlir that referenced this issue Oct 3, 2022
The modification to use the builder based interface to generate Krnl loop is completed (llvm#1250, llvm#1283, llvm#1285, llvm#1292, llvm#1293, llvm#1303, llvm#1308, llvm#1314, llvm#1316, llvm#1317, llvm#1326, llvm#1403), and BuildKrnlLoop is no longer needed.

Signed-off-by: Whitney Tsang [email protected]
@silvasean
Copy link
Contributor

We centralized this all to build_linux_packages.sh

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

No branches or pull requests

2 participants