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
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
As FindPythonInterp and FindPythonLibs has been deprecated since cmake-3.12
Replace `PYTHON_EXECUTABLE` with `Python_EXECUTABLE` everywhere (CMake variable names are case-sensitive)
This makes PyTorch buildable with python3 binary shipped with XCode on MacOS
TODO: Get rid of `FindNumpy` as its part of Python package
Pull Request resolved: https://github.com/pytorch/pytorch/pull/124613
Approved by: https://github.com/cyyever, https://github.com/Skylion007
The `usort` config in `pyproject.toml` has no effect due to a typo. Fixing the typo make `usort` do more and generate the changes in the PR. Except `pyproject.toml`, all changes are generated by `lintrunner -a --take UFMT --all-files`.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/127126
Approved by: https://github.com/kit1980
The `usort` config in `pyproject.toml` has no effect due to a typo. Fixing the typo make `usort` do more and generate the changes in the PR. Except `pyproject.toml`, all changes are generated by `lintrunner -a --take UFMT --all-files`.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/127126
Approved by: https://github.com/kit1980
ghstack dependencies: #127122, #127123, #127124, #127125
The `usort` config in `pyproject.toml` has no effect due to a typo. Fixing the typo make `usort` do more and generate the changes in the PR. Except `pyproject.toml`, all changes are generated by `lintrunner -a --take UFMT --all-files`.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/127123
Approved by: https://github.com/Skylion007
ghstack dependencies: #127122
Adds a ruff lint rule to ban raising raw exceptions. Most of these should at the very least be runtime exception, value errors, type errors or some other errors. There are hundreds of instance of these bad exception types already in the codebase, so I have noqa'd most of them. Hopefully this error code will get commiters to rethink what exception type they should raise when they submit a PR.
I also encourage people to gradually go and fix all the existing noqas that have been added so they can be removed overtime and our exception typing can be improved.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/124570
Approved by: https://github.com/ezyang
Update ruff to 0.4.1 .
This version fixes a lot false negatives/false positives, is 20-40% faster, and has various other bug fixes.
Below is a before and after table showing the execution time of ruff lint and ruff format in milliseconds courtesy of https://astral.sh/blog/ruff-v0.4.0
| Repository | Linter (v0.3) | Linter (v0.4) | Formatter (v0.3) | Formatter (v0.4) |
|----------------------------------------------------|---------------|---------------|------------------|------------------|
| [pytorch/pytorch](https://github.com/pytorch/pytorch) | 328.7 | 251.8 | 351.1 | 274.9 |
Pull Request resolved: https://github.com/pytorch/pytorch/pull/124549
Approved by: https://github.com/ezyang
Refactor release only changes to two step execution.
1. Step ``tag-docker-images.sh`` . Tags latest docker images for current release. This step takes about 30min to complete. This step may fail due to space issues on the local host or http connection when pulling image. Hence should be rerun if failed.
2. Apply release only changes ``apply-release-changes.sh`` prepares a PR with release only changes.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/121728
Approved by: https://github.com/jeanschmidt
Fixes#120044
Should fix build from source instructions on release branch here: https://github.com/pytorch/pytorch#from-source
Please note we are using /test/ channel for release here to make sure it works, before actual release is completed.
Test main:
```
make triton
pip3 uninstall -y triton
WARNING: Skipping triton as it is not installed.
Looking in indexes: https://download.pytorch.org/whl/nightly/
Collecting pytorch-triton==3.0.0+a9bc1a3647
Downloading https://download.pytorch.org/whl/nightly/pytorch_triton-3.0.0%2Ba9bc1a3647-cp310-cp310-linux_x86_64.whl (239.0 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 239.0/239.0 MB 8.7 MB/s eta 0:00:00
Requirement already satisfied: filelock in /home/atalman/miniconda3/envs/py310/lib/python3.10/site-packages (from pytorch-triton==3.0.0+a9bc1a3647) (3.13.1)
Installing collected packages: pytorch-triton
Attempting uninstall: pytorch-triton
Found existing installation: pytorch-triton 2.2.0
Uninstalling pytorch-triton-2.2.0:
Successfully uninstalled pytorch-triton-2.2.0
Successfully installed pytorch-triton-3.0.0+a9bc1a3647
```
Test release/2.2:
```
make triton
pip3 uninstall -y triton
WARNING: Skipping triton as it is not installed.
Looking in indexes: https://download.pytorch.org/whl/test/
Collecting pytorch-triton==2.2.0
Using cached https://download.pytorch.org/whl/test/pytorch_triton-2.2.0-cp310-cp310-linux_x86_64.whl (183.1 MB)
Requirement already satisfied: filelock in /home/atalman/miniconda3/envs/py310/lib/python3.10/site-packages (from pytorch-triton==2.2.0) (3.13.1)
Installing collected packages: pytorch-triton
Attempting uninstall: pytorch-triton
Found existing installation: pytorch-triton 3.0.0+a9bc1a3647
Uninstalling pytorch-triton-3.0.0+a9bc1a3647:
Successfully uninstalled pytorch-triton-3.0.0+a9bc1a3647
Successfully installed pytorch-triton-2.2.0
```
Pull Request resolved: https://github.com/pytorch/pytorch/pull/121169
Approved by: https://github.com/seemethere
It's a bit annoying to have to read through the test name in verbose mode just to see what the test's sentinel file is actually called when encountering an unexpected success. Now that we have sentinel files, we can directly list the file path from root in the error message.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/120766
Approved by: https://github.com/Skylion007
Previously we were downloading all of (eager311, dynamo38, dynamo311).
Now we just download what's necessary. This is useful for
update_failures.py because the dynamo tests finish much faster than the
eager tests and it only needs the result from the dynamo tests.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/119027
Approved by: https://github.com/jamesjwu
ghstack dependencies: #118874, #118882, #118931
Previously, you could run update_failures.py (with a commit hash) and it
would add new expected failures and skips for newly failing tests and
remove expected failures for newly passing tests.
This PR teaches update_failures.py to also remove skips for tests that
are now passing without them.
The way we do this is:
- dynamo_test_failures.py doesn't actually skip tests -- it runs the
test and then suppresses the signal.
- if the test actually passed, then the test gets skipped with a special
skip message
- we teach update_failures.py to look for the presence of that skip
message.
Test Plan:
- Used this to generate https://github.com/pytorch/pytorch/pull/118928
Pull Request resolved: https://github.com/pytorch/pytorch/pull/118931
Approved by: https://github.com/yanboliang
ghstack dependencies: #118874, #118882
```
def f():
def g():
return ()
print(g.__name__)
f()
```
The following script should print `g` (with or without torch.compile),
but prints `f.<locals>.g` with torch.compile.
The problem looks like we use the co_qualname when reconstructing the
NestedUserFunctionVariable. I switched this over to use the co_name.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/118768
Approved by: https://github.com/yanboliang, https://github.com/jansel
Changelog:
- Don't count running PYTORCH_TEST_WITH_DYNAMO=1 on dynamo/ tests in the pass
rate. This was a bug (we were counting all of these as failing, but in
reality, most of these pass). The net effect is that the passrate is (artifically)
6% higher.
- Have the histogram script filter out skips based on the passrate metric.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/118752
Approved by: https://github.com/jamesjwu
Summary:
This diff implements a mechanism for safely update torch.export serialization schema, aka schema.py, which is the API surface having the strongest compatibility guarantee.
The diff is consist of 3 changes:
- Added a script to "build" or "materialize" schema.py into a platform neutral format (yaml), which serves as the committed form of the seialization schema.
- Added unittest to compare against schema.py and schema.yaml, so that it forces developers to execute the updater script when there is mismatch between two files.
- Added a checker inside the updater script, so that all the compatible change will result in a minor version bump, and all the incompatible changes will result in a major version bump.
torch.export's serialization BC/FC policy is (tentatively) documented here: https://docs.google.com/document/d/1EN7JrHbOPDhbpLDtiYG4_BPUs7PttpXlbZ27FuwKhxg/edit#heading=h.pup7ir8rqjhx , we will update the
As noted in the code doc, people should be able to run the following command to update schema properly from now on:
```
python scripts/export/update_schema.py --prefix <path_to_torch_development_diretory>
or
buck run caffe2:export_update_schema -- --prefix /data/users/$USER/fbsource/fbcode/caffe2/
```
Test Plan:
buck test mode/opt caffe2/test:test_export -- -r test_schema
buck run caffe2:update_export_schema -- --prefix /data/users/$USER/fbsource/fbcode/caffe2/
Differential Revision: D52971020
Pull Request resolved: https://github.com/pytorch/pytorch/pull/118424
Approved by: https://github.com/angelayi
