Adds a pre-commit hook (technically a pre-push hook) to the PyTorch repo.
**This is currently an opt-in feature**, which one can opt into by running `python scripts/setup_hooks.py` locally.
### Features
- **Run Lintrunner Before Push**: Before every `git push`, automatically runs lintrunner on your changes.
- Really need to skip the checks? Run `git push --no-verify`
- **Consistent, Isolated, Lintrunner Environment**: During pre-push, Lintrunner runs in it's own virtual en environment that contain all lintrunner dependencies in a consistent, isolated environment. No more lintrunner failures because you created a new .venv. (Did you know you needed to run `lintrunner init` every time you make a new .venv?)
- **Dependencies Automatically Updated**: If .lintrunner.toml is updated, this will automatically re-run `lintrunner init` to ensure you install the latest dependencies specified
### Installation
- Run `python scripts/setup_hooks.py`. Now every `git push` will first run lintrunner.
### Additional details
- The lintrunner used by the pre-push hook runs in a special per-repo virtual environment managed by the commit-hook tool located under `$USER/.cache/pre-commit`
- Does not affect your regularly used lintrunner
- Manual invocations of lintrunner will continue to depend on your local environment instead of the special pre-push one. If there's enough interest, we could explore consolidating them.
- Does not run `lintrunner -a` for you.
- You still need to manually run that (can be changed later though!)
- Have staged/unstaged changes? No worries
- This runs `git stash` before running the pre-commit hooks and pops back your changes afterwards, so only the changes actaully being pushed will be tested
### Downsides
- No streaming UI updates
- While you still get the same output from lintrunner that you're used to, the commit-hook framework doesn't show any output while lintrunner is actually running. Instead, it shows the entire output after linter has completed execution, which could be a few minutes (especially if it has to run `lintrunner init` first)
- `uv` installation is required to run the setup script. The setup script will ask users to install uv if it's not available.
- This is required to be able to install the pre-commit package in a safe way that's available no matter what .venv you are running in.
### Opting out
- Disable hook for a single push: Run `git push --no-verify`
- Disable hook permanently: If something goes wrong and you need to wipe your setup:
- Delete the `$USER/.cache/pre-commit` folder and the `.git/hooks/pre-push` file in your local repo.
- You can now rerun `python scripts/setup_hooks.py` to setup your git push hook again if you want.
### Potential Future Changes
Things that could be done to make this even better if folks like these ideas:
- Automatic setup
- Our `CONTRIBUTING.md` file tells devs to run `make setup-env`. That could be a good entry point to hook the installation into
- Fix the console output streaming
- Make every lintrunner invocation (including manual ones) use the same repo-specific venv that the commit-hook uses.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/158389
Approved by: https://github.com/seemethere
Triton XPU shares its version file with the community one. When the community updates Triton version, it will temporarily break the XPU CI/CD because they use different repositories and commits. To decouple Triton version bumps between the community and XPU, we propose splitting the version into two separate files.
Refer the latest community triton version bump PR #153117
Pull Request resolved: https://github.com/pytorch/pytorch/pull/155313
Approved by: https://github.com/etaf, https://github.com/EikanWang, https://github.com/atalman
Now that we have bc_lint in CI, this script is no longer needed (nor has it ever been conclusive). I've already updated the Runbook to not need this script.
Suppressing bc_lint as this script is not shipped as a part of torch--it is not user facing! For context, this script is (rarely) used by the release notes manager to ensure BC across releases. It had been broken for at least since 2.6.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/151453
Approved by: https://github.com/albanD, https://github.com/jbschlosser
Reference: https://docs.astral.sh/ruff/formatter/black/#assert-statements
> Unlike Black, Ruff prefers breaking the message over breaking the assertion, similar to how both Ruff and Black prefer breaking the assignment value over breaking the assignment target:
>
> ```python
> # Input
> assert (
> len(policy_types) >= priority + num_duplicates
> ), f"This tests needs at least {priority+num_duplicates} many types."
>
>
> # Black
> assert (
> len(policy_types) >= priority + num_duplicates
> ), f"This tests needs at least {priority+num_duplicates} many types."
>
> # Ruff
> assert len(policy_types) >= priority + num_duplicates, (
> f"This tests needs at least {priority + num_duplicates} many types."
> )
> ```
Pull Request resolved: https://github.com/pytorch/pytorch/pull/144546
Approved by: https://github.com/malfet
As of python 3.9 annotated lists can be written as `list[T]` and `List[T]` has been deprecated. However schema_check was converting `list[T]` to simply be `list`. This change teaches it to handle `list[T]` the same as `List[T]`.
A couple small drive-by changes I noticed as well:
- Path concatenation should use `os.path.join`, not `+`
- Spelling in error message
Pull Request resolved: https://github.com/pytorch/pytorch/pull/145154
Approved by: https://github.com/bobrenjc93
Changes by apply order:
1. Replace all `".."` and `os.pardir` usage with `os.path.dirname(...)`.
2. Replace nested `os.path.dirname(os.path.dirname(...))` call with `str(Path(...).parent.parent)`.
3. Reorder `.absolute()` ~/ `.resolve()`~ and `.parent`: always resolve the path first.
`.parent{...}.absolute()` -> `.absolute().parent{...}`
4. Replace chained `.parent x N` with `.parents[${N - 1}]`: the code is easier to read (see 5.)
`.parent.parent.parent.parent` -> `.parents[3]`
5. ~Replace `.parents[${N - 1}]` with `.parents[${N} - 1]`: the code is easier to read and does not introduce any runtime overhead.~
~`.parents[3]` -> `.parents[4 - 1]`~
6. ~Replace `.parents[2 - 1]` with `.parent.parent`: because the code is shorter and easier to read.~
Pull Request resolved: https://github.com/pytorch/pytorch/pull/129374
Approved by: https://github.com/justinchuby, https://github.com/malfet
Changes by apply order:
1. Replace all `".."` and `os.pardir` usage with `os.path.dirname(...)`.
2. Replace nested `os.path.dirname(os.path.dirname(...))` call with `str(Path(...).parent.parent)`.
3. Reorder `.absolute()` ~/ `.resolve()`~ and `.parent`: always resolve the path first.
`.parent{...}.absolute()` -> `.absolute().parent{...}`
4. Replace chained `.parent x N` with `.parents[${N - 1}]`: the code is easier to read (see 5.)
`.parent.parent.parent.parent` -> `.parents[3]`
5. ~Replace `.parents[${N - 1}]` with `.parents[${N} - 1]`: the code is easier to read and does not introduce any runtime overhead.~
~`.parents[3]` -> `.parents[4 - 1]`~
6. ~Replace `.parents[2 - 1]` with `.parent.parent`: because the code is shorter and easier to read.~
Pull Request resolved: https://github.com/pytorch/pytorch/pull/129374
Approved by: https://github.com/justinchuby, https://github.com/malfet
Summary:
In this diff we implement a way to ensure the internal thrift schema from cfgr (configerator/structs/caffe2/torch/export/schema.thrift) and the schema in OSS (torch/_export/serde/schema.thrift) are in sync, by adding a unittest to reflect on the type names and fields from each schema and compare them field by field.
When we detect new fields/types from torch/_export/serde/schema.thrift, there'll be a test failure on the trunk and the error message hints people to add the missing field/type to the thrift schema from cfgr, so that they are always in sync in practice.
Test Plan: buck test mode/opt caffe2/test:test_export -- -r test_thrift_schema_in_sync
Differential Revision: D66716834
Pull Request resolved: https://github.com/pytorch/pytorch/pull/141989
Approved by: https://github.com/yiming0416
Summary: To make sure schema.py and schema.thrift are kept in sync, we use the int keys from thrift and use Python Annotated type to associate fields between thrift and schema.py. Later we will use this association to build a single source of truth between the schemas.
Test Plan: CI
Differential Revision: D66253157
Pull Request resolved: https://github.com/pytorch/pytorch/pull/141611
Approved by: https://github.com/yiming0416
Differential Revision: D63206258
This diff introduces a mechanism to generate a json-compatible deserializer in cpp using nlohmann json (already being used by AOTI).
Why we need this? Because there will be a lot of cases where people don't want to use Python to load the graph (e.g. cpp runtime), and instead they can use this header to deserialize the JSON graph.
Every time we call update_schema.py to update the schema, the header will be auto generated and included into the source files.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/136398
Approved by: https://github.com/angelayi
Buck1 is no longer supported in favor of buck2. This CI tests the old buck1 flow, however it is difficult to maintain especially since buck1 doesn't support aarch64 mac.
I am suggesting that this CI be deprecated until a decision on buck2 is made, and buck2 support is added. As of now, there seems to be no push towards adding buck2 support.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/140067
Approved by: https://github.com/huydhn
Move the slow test json to be in the pytorch/pytorch repo and make a job that will update it weekly. The job uses the same environment as the commit hash. It uses similar code to the hash updates, but the hash update contains a lot of code that is specific to the hash update, so I chose to pick out the parts that are relevant
Remove references to the old file and set up testing to read from the new file instead
The old update cadence was every day, the new one is every week
The auto slow test infra + the lack of pinning between pytorch and test-infra makes it really hard to tell if a test started failing because of a change or because of the slow test json changing. While this can have benefits, like disable test issues being effective everywhere immediately, it can also be very confusing, especially since we don't have the same insight into slow tests like we do for disable issues.
Example PR made: https://github.com/pytorch/pytorch/pull/132383 (with all the changes from this PR because it was working on top of this)
We should just get rid of this at some point in favor of the slowTest decorator, but there are some tests that take 5+ minutes to run and I don't want to track them down right now
Pull Request resolved: https://github.com/pytorch/pytorch/pull/132379
Approved by: https://github.com/huydhn
