* custom pytest-shard so I can control the verbosity (also index by 1 since it's confusing)
* normal runs (not keep-going) always rerun each failed test 9 times (3 per process, 3 processes). Previously it would only run the entire test file 3 times, so if a test before you segfaulted, you only got 2 tries
Example of quieter log https://github.com/pytorch/pytorch/actions/runs/7481334046/job/20363147497
"items in shard" only gets printed once at the beginning, and the reruns just say how many got skipped.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/117069
Approved by: https://github.com/huydhn
* custom pytest-shard so I can control the verbosity (also index by 1 since it's confusing)
* normal runs (not keep-going) always rerun each failed test 9 times (3 per process, 3 processes). Previously it would only run the entire test file 3 times, so if a test before you segfaulted, you only got 2 tries
Example of quieter log https://github.com/pytorch/pytorch/actions/runs/7481334046/job/20363147497
"items in shard" only gets printed once at the beginning, and the reruns just say how many got skipped.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/117069
Approved by: https://github.com/huydhn
New function for continue on error
Another solution might be to run the entire suite to the end and use last failed, but I'm worried about concurrent processes writing to the same last failed cache entry, it's a bit different than the usual test rerunning strategy we use especially regarding segfaults and other ways the test suite can suddenly end, and there are some cases where the entire test suite should immediately get rerun in a new process (ex cuda error that causes sync to fail).
Find example logs on commit 2f1510839727f6ef2631040d5f0edde26265015d
TODO: continue on error for --subprocess and test_distributed aren't working fully
Pull Request resolved: https://github.com/pytorch/pytorch/pull/112098
Approved by: https://github.com/huydhn
Fixes#113940. This vendors the relevant parts of [`packaging==23.2.0`]() to have access to `Version` and `InvalidVersion` without taking a runtime dependency on `setuptools` or `packaging`.
I didn't find any vendoring policy so I put it under `torch._vendor.packaging`. While I have only vendored the files we need, I have not touched or trimmed the files otherwise.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/114108
Approved by: https://github.com/malfet, https://github.com/albanD
Fixes a bug in TD metrics generation where it wouldn't be able to find the rank and relevance that a heuristic gave a test run if that heuristic had divided that test into multiple test runs.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/113789
Approved by: https://github.com/clee2000
Allow heuristics to actually downgrade the relevance of a test. Note that NONE/UNLIKELY tests will still get executed, but they will be ran at the end of the CI
The Relevance chosen affects the outcome when Heuristics offer conflicting predictions. A relevance higher up in this list means higher confidence in the declared relevance:
HIGH > NONE > PROBABLE > UNLIKELY > UNRANKED
Given that we assume ordering based on the list in init right now since the lists are appended, do a similar thing for UNLIKELY and NONE
ex HEURISTICS = [a, b, c, d]
currently all things in b.high and added after a.high
if b.none includes things in a.high, a.high trumps
if b.none includes things in a.probable, then b.none trumps since none is stronger than probable
if b.unlikely includes things from a.high/probable, a.high/probable trumps since unlikely and probable are at a higher strength
Pull Request resolved: https://github.com/pytorch/pytorch/pull/112671
Approved by: https://github.com/clee2000
Changes the heuristic framework to support multiple prioritizing individual classes within a test file.
Components of this included:
- Updating TestPrioritizations to accept individual test classes being prioritized. Previously, when a heuristic wanted to prioritize a test file it would pass in the test's name, now to prioritize a class within a test it uses the notation "test::classname"
- Changes are fully backwards compatible with existing heuristics
- Test sharding now supports sharding individual tests (for when they're prioritized)
- When a TestClass is prioritized, we pass the appropriate "-k" flags down to pytest
Pull Request resolved: https://github.com/pytorch/pytorch/pull/112161
Approved by: https://github.com/huydhn
`test_typing.py` was written to use `pytest` in https://github.com/pytorch/pytorch/pull/54234 which unfortunately rendered it incompatible with run_test.py, and therefore it was not running in CI all this time.
In this PR, same functionality is re-written using unittest framework, and `parametrize` from `torch.testing._internal._common_utils`.
Valid `test_typing.py` with ufmt
Disable `fail/bitwise_ops.py` and `pass/jit.py` as it regressed at some point as well as one of examples in `namedtuple.py` as `torch.linalg.qr` type is no longer revealed correctly.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/111428
Approved by: https://github.com/clee2000
Move print to the beginning instead because putting it at the end makes it so you have to scroll through when debugging, and nothing in that function indicates that it should be printing anything
Also the line for printing disabled issues out of the for loop
Pull Request resolved: https://github.com/pytorch/pytorch/pull/110621
Approved by: https://github.com/huydhn
To reduce the amount of logs
* for successes, only print the part that says what tests ran and don't print the rest. Zip the log into an artifact. The line listing al the test names is really long, but if you view source of the raw logs, it will not wrap so it will only be one line. The log classifier can also be configured to ignored this line. Gets rid of lines like `test_ops.py::TestCommonCPU::test_multiple_devices_round_cpu_int64 SKIPPED [0.0010s] (Only runs on cuda) [ 9%]`
* for failures/reruns, print logs. Do not zip.
Also
* change log artifact name
Examples of various logs:
a074db0f7f failures
1b439e24c4 failures
possibly controversial haha
should i include an option for always printing?
Pull Request resolved: https://github.com/pytorch/pytorch/pull/110033
Approved by: https://github.com/huydhn
Minor logging changes that just kind of annoyed me:
* prevent constant printing of `No CUDA runtime is found, using CUDA_HOME='/usr/local/cuda'` by moving import within the function (idk if this is ok)
* prevent constant printing of `Ignoring disabled issues: ['']` (no idea why it was not gated behind a function or main)
* change all prints in run_tests.py to be through stderr so theres no weird interleaving (although if everything goes through stderr, might as well just print everything through stdout...)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/110188
Approved by: https://github.com/huydhn, https://github.com/malfet, https://github.com/ZainRizvi
Extend metric library to allow setting global metrics on a process level which will always be emitted.
Current use case for them is to include shard information every time a metric is emitted by run_test.py
<!--
copilot:poem
-->
### <samp>🤖 Generated by Copilot at 0cae92c</samp>
> _`run_test` refactored_
> _Sharding metrics in Rockset_
> _Autumn of testing_
Pull Request resolved: https://github.com/pytorch/pytorch/pull/110035
Approved by: https://github.com/clee2000
When a test fails, we will now emit fine grained details about how accurately heuristics predicted the relevance of that test.
## Context
Why only look at failing tests? Our only signal that a PR is most likely relevant to a test is whether or not a test fails on it. Green tests don't tell us if the success was due to the code being good vs being irrelevant. This isn't a perfect measure, since it can miscategorize unstable and flaky failures as having been "missed" by the heuristics, but it's a reasonable approximation.
## What's measured?
The metrics this PR collects are designed to answer the following questions
### How comprehensive are the heuristics?
- What's the false negative rate, the % of failures that ideally should have been prioritized but weren't? (Both at an aggregate level and at a per heuristic level)
### How precise are the heuristics?
- What % of failed tests were prioritized by a given heuristic? What % was prioritized overall?
- How relevant was a failed test was considered to be? (Both a aggregate level and at a per heuristic level)
- What % of time was a given heuristic prioritizing a failing test higher than any other heuristic?
Pull Request resolved: https://github.com/pytorch/pytorch/pull/108192
Approved by: https://github.com/huydhn
ghstack dependencies: #108117
This PR looks big, but it's mostly just refactorings with a bit of dead code deletion. Exceptions are:
- Some metric emissions were changed to comply with the new TD format
- Some logging changes
- We now run tests in three batches (highly_relevant, probably_relevant, unranked_relevance) instead of the previous two (prioritized and general)
Refactorings done:
- Moves all test reordering code to the new TD framework
- Refactors run_test.py to cleanly support multiple levels of test priorities
- Deletes some dead code that was originally written for logging
Pull Request resolved: https://github.com/pytorch/pytorch/pull/107071
Approved by: https://github.com/clee2000, https://github.com/huydhn
Make it so that scripts can import and run the `emit_metrics` function even if they don't have boto3 installed, in which case it will still validate the inputs but skip the actual metric emission part.
It's purely a refactor without any real logic changes
Motivation: So that run_test.py and the target determination code can use this library easily without worrying about if it was imported or if it's dependencies are installed.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/107070
Approved by: https://github.com/huydhn
Update distutils.Version to packaging.version due to the deprecation warning.
```python
/root/Git.d/pytorch/pytorch/torch/testing/_internal/common_methods_invocations.py:17136: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
active_if=TEST_SCIPY and LooseVersion(scipy.__version__) < "1.4.0"),
/root/Git.d/pytorch/pytorch/torch/testing/_internal/common_methods_invocations.py:17138: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
active_if=TEST_SCIPY and LooseVersion(scipy.__version__) < "1.4.0"),
/root/Git.d/pytorch/pytorch/torch/testing/_internal/common_methods_invocations.py:17140: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
active_if=TEST_SCIPY and LooseVersion(scipy.__version__) < "1.4.0"),
```
Pull Request resolved: https://github.com/pytorch/pytorch/pull/107207
Approved by: https://github.com/soulitzer
Companion with https://github.com/pytorch/test-infra/pull/4424
Uses the file rating generated by the test infra PR to re order tests. For each test file, sum the file ratings from the changed files in the PR, and put the tests in order of sum.
A lot of tests are probably going to end up as "prioritized" since it takes anything with a rating > 0 right now.
Sharding is done twice, once on the prioritized tests, and once on the general/non prioritized tests. Prioritized tests have an order, so they should be sharded according to that order, while general tests don't have an order and are sharded by test time, which should result in more balanced shards.
I'll change the metric name before I merge, i want to quarantine my testing stuff from actual results
Pull Request resolved: https://github.com/pytorch/pytorch/pull/106347
Approved by: https://github.com/ZainRizvi
Fixes https://github.com/pytorch/pytorch/issues/106754
This PR:
- moves test/autograd/test_fallback.py to test_autograd_fallback.py and
removes it from test_autograd.py (necessary for the next step)
- adds test_autograd_fallback.py to parallel test blocklist.
- lintrunner really wanted to make changes to the files, but other than
that, it is a move.
The problem is that we set a global option (the autograd fallback mode)
during these tests which may cause the tests to interfere with each
other.
Test Plan:
- python test/run_test.py -i test_autograd_fallback
NOTE to diff train oncall:
- You'll also need to modify the test/autograd/test_fallback.py TARGET in
caffe2/test/TARGETS since we renamed the file.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/106866
Approved by: https://github.com/soulitzer
Companion with https://github.com/pytorch/test-infra/pull/4424
Uses the file rating generated by the test infra PR to re order tests. For each test file, sum the file ratings from the changed files in the PR, and put the tests in order of sum.
A lot of tests are probably going to end up as "prioritized" since it takes anything with a rating > 0 right now.
Sharding is done twice, once on the prioritized tests, and once on the general/non prioritized tests. Prioritized tests have an order, so they should be sharded according to that order, while general tests don't have an order and are sharded by test time, which should result in more balanced shards.
I'll change the metric name before I merge, i want to quarantine my testing stuff from actual results
Pull Request resolved: https://github.com/pytorch/pytorch/pull/106347
Approved by: https://github.com/ZainRizvi
This PR turns translation validation on by default for tests and accuracy benchmark
runs. It also installs Z3 on CI.
The main changes are:
- Add `--no-translation-validation` as an option in _test/run_tests.py_
- Set `PYTORCH_TEST_WITH_TV` environment variable
- Add `TEST_WITH_TV` variable in _torch/testing/_internal/common_utils.py_
- Turn translation validation on for accuracy benchmarks in _benchmarks/dynamo/common.py_
- Add Z3 installation on CI scripts
Pull Request resolved: https://github.com/pytorch/pytorch/pull/103611
Approved by: https://github.com/ezyang
Prevents following cryptic error if one attempts to use `run_tests.py` on system that also has torchaudio installed in dev mode (as `tools` from https://github.com/pytorch/audio might take precedence, but this is not how script should behave):
```
Unable to import test_selections from tools/testing. Running without test selection stats.... Reason: No module named 'tools.stats'
Traceback (most recent call last):
File "/Users/nshulga/git/pytorch/pytorch/test/run_test.py", line 1673, in <module>
main()
File "/Users/nshulga/git/pytorch/pytorch/test/run_test.py", line 1604, in main
selected_tests = get_selected_tests(options)
File "/Users/nshulga/git/pytorch/pytorch/test/run_test.py", line 1418, in get_selected_tests
path = os.path.join(str(REPO_ROOT), TEST_TIMES_FILE)
NameError: name 'TEST_TIMES_FILE' is not defined
```
But make sure to remove it in the end, otherwise it will not work if torch is installed from wheel, but tests are running from clean repo checkout.
<!--
copilot:poem
-->
### <samp>🤖 Generated by Copilot at dd52521</samp>
> _Sing, O Muse, of the cunning code review_
> _That fixed the tests of the `tools` module_
> _By adding and removing the root path_
> _As a shepherd guides his flock to and fro._
Pull Request resolved: https://github.com/pytorch/pytorch/pull/104214
Approved by: https://github.com/kit1980
There is a `HAVE_TEST_SELECTION_TOOLS` conditional, but turns out it does not really work, so fix it by defining all missing prototypes and make it work as single-shard instance
Add lint rule to test stat it would succeed for runnign only test_cuda with released version of PyTorch
Pull Request resolved: https://github.com/pytorch/pytorch/pull/104111
Approved by: https://github.com/clee2000, https://github.com/ZainRizvi
Because we always run tests with pytest now.
Marking it as `bc-breaking` as there could technically be some scripts depending on it somewhere...
<!--
copilot:poem
-->
### <samp>🤖 Generated by Copilot at 1760568</samp>
> _`pytest` option gone_
> _simpler test runner script_
> _autumn leaves fall fast_
Pull Request resolved: https://github.com/pytorch/pytorch/pull/104125
Approved by: https://github.com/seemethere
Added a feature to upload test statistics to DynamoDB and Rockset using a new function `emit_metric` in `tools/stats/upload_stats_lib.py`.
Added metrics to measure test reordering effectiveness in `tools/testing/test_selections.py`.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/102691
Approved by: https://github.com/malfet
sharding on rocm is broken, i cant replicate on dummy PRs even though it seems to happen pretty often on main, so adding this to increase my sample size. Hopefully this is enough print statements...
Pull Request resolved: https://github.com/pytorch/pytorch/pull/102713
Approved by: https://github.com/huydhn
Currently file level reruns + stepcurrent are incompatible and it's making PRs green when they are actually red, so turn off stepcurrent + file level reruns when keep-going is used until I figure out a better way to do this.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/102569
Approved by: https://github.com/huydhn
The console log blows up to much when running in rerun disabled tests mode (x50) e132f09e88. Each log is around 1GB and the whole uncompressed logs is ~50GB. After compression, it will be around 1GB, still too big. The increase comes mainly from the multiple SKIPPED message for non-disabled tests, which is expected due to how SkipTest and pytest-flakyfinder currently work.
I update `test/conftest.py` to completely ignore skipped tests when rerunning disabled test instead of collecting then skipping 50 tests each. The benefit of doing is is much more than I originally expect:
* Rerun disabled tests jobs now finish in less than half an hour as they should be
* Fix OOM runner crash because of too many collected tests
* Fix verbosity issue as now only disabled tests are run x50 times. There are only few hundreds of them atm
* Fix timed out issue when rerunning disabled distributed and ASAN tests. They are just too slow when running at x50
### Testing
When rerunning disabled tests https://github.com/pytorch/pytorch/actions/runs/5084508614, only disabled tests on the platform are run, for example `test_ops_jit` on https://ossci-raw-job-status.s3.amazonaws.com/log/13770164954 only ran 100 tests (`test_variant_consistency_jit_linalg_lu_cuda_float32` + `test_variant_consistency_jit_linalg_lu_factor_cuda_complex64`) x50.
```
Executing ['/opt/conda/envs/py_3.10/bin/python', '-bb', 'test_ops_jit.py', '--shard-id=1', '--num-shards=2', '-v', '-vv', '-rfEX', '-p', 'no:xdist', '--use-pytest', '--sc=test_ops_jit_1', '--flake-finder', '--flake-runs=50', '--import-slow-tests', '--import-disabled-tests', '--rerun-disabled-tests'] ... [2023-05-25 21:32:49.763856]
Expand the folded group to see the log file of test_ops_jit 2/2
##[group]PRINTING LOG FILE of test_ops_jit 2/2 (/var/lib/jenkins/workspace/test/test-reports/test_ops_jit_h2wr_t2c.log)
Test results will be stored in test-reports/python-pytest/test_ops_jit/test_ops_jit-51a83bd44549074e.xml
============================= test session starts ==============================
platform linux -- Python 3.10.11, pytest-7.3.1, pluggy-1.0.0 -- /opt/conda/envs/py_3.10/bin/python
cachedir: .pytest_cache
hypothesis profile 'pytorch_ci' -> database=None, max_examples=50, derandomize=True, suppress_health_check=[HealthCheck.too_slow]
rootdir: /var/lib/jenkins/workspace
configfile: pytest.ini
plugins: hypothesis-5.35.1, cpp-2.3.0, flakefinder-1.1.0, rerunfailures-11.1.2, shard-0.1.2, xdist-3.3.0, xdoctest-1.1.0
collecting ... collected 1084 items
Running 100 items in this shard: test/test_ops_jit.py::TestJitCUDA::test_variant_consistency_jit_linalg_lu_cuda_float32 (x50), test/test_ops_jit.py::TestJitCUDA::test_variant_consistency_jit_linalg_lu_factor_cuda_complex64 (x50)
stepcurrent: Cannot find last run test, not skipping
test_ops_jit.py::TestJitCUDA::test_variant_consistency_jit_linalg_lu_cuda_float32 PASSED [2.1876s] [ 1%]
test_ops_jit.py::TestJitCUDA::test_variant_consistency_jit_linalg_lu_factor_cuda_complex64 PASSED [4.5615s] [ 2%]
```
* [pull](https://github.com/pytorch/pytorch/actions/runs/5093566864)
* [trunk](https://github.com/pytorch/pytorch/actions/runs/5095364311)
* [periodic](https://github.com/pytorch/pytorch/actions/runs/5095378850)
* [slow](https://github.com/pytorch/pytorch/actions/runs/5095390285)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/102107
Approved by: https://github.com/clee2000, https://github.com/malfet
Per title. I extract this part out of the draft PR that I'm working on https://github.com/pytorch/pytorch/pull/102107 because
the remaining issues with rerun disabled tests: log size and unexpected runner failures requires some further investigations while this one is clearing breaking in trunk atm.
Until we can support disable C++ tests, there is no need to run them in rerun disabled tests mode.
### Testing
Coming from https://github.com/pytorch/pytorch/pull/102107, for example https://github.com/pytorch/pytorch/actions/runs/5062224659/jobs/9087747981
```
2023-05-23T22:46:50.1953318Z Running cpp/basic 1/1 ... [2023-05-23 22:46:50.195077]
2023-05-23T22:46:50.1953847Z Skipping C++ tests when running under RERUN_DISABLED_TESTS mode
2023-05-23T22:46:50.2066032Z Running cpp/atest 1/1 ... [2023-05-23 22:46:50.206348]
2023-05-23T22:46:50.2066435Z Skipping C++ tests when running under RERUN_DISABLED_TESTS mode
2023-05-23T22:46:52.2666743Z No CUDA runtime is found, using CUDA_HOME='/usr/local/cuda'
2023-05-23T22:46:52.2691817Z Ignoring disabled issues: []
...
```
Pull Request resolved: https://github.com/pytorch/pytorch/pull/102132
Approved by: https://github.com/clee2000
not very elegant
checked on separate conda env that doesnt have the usual ci dependencies
the two pytest extensions at fault are pytest-rerunfailures and pytest-shard, also included pytest-flakefinder just incase
no idea if this is a good way to do this
could also check individually and add flags based on that, but was told that needing to requiring all the ci dependencies to be downloaded was also ok
Pull Request resolved: https://github.com/pytorch/pytorch/pull/100916
Approved by: https://github.com/huydhn
After an investigation, running C++ tests with https://github.com/pytest-dev/pytest-cpp is just slower than running them directly, plain and simple. I'm curious on the exact root cause, but that's a story for another day.
`time build/bin/test_lazy` takes half a minute to run 610 tests on `linux-bionic-cuda11.8-py3.10-gcc7 / test (default, 2, 5, linux.4xlarge.nvidia.gpu)` while `time pytest /var/lib/jenkins/workspace/build/bin/test_lazy -v` takes 20+ minutes on the same runner. This is a very costly price to pay.
The saving grace here is that https://github.com/pytest-dev/pytest-cpp supports pytest-xdist to run tests in parallel with `-n auto`, so `time pytest /var/lib/jenkins/workspace/build/bin/test_lazy -v -n auto` takes only 3 minutes. This is still not as fast as running C++ tests directly, but it's order of magnitude faster than running them sequentially.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/101440
Approved by: https://github.com/clee2000
After https://github.com/pytorch/pytorch/pull/99559, we can now run C++ test with `run_test.py`. Although advance features such as `--import-slow-tests` and `--import-disabled-tests` won't work for now, there will still be a gain in reliability and performance as C++ can now be retried and run in parallel.
This covers all C++ tests in the CI including aten, libtorch, and Vulkan C++ tests across all platforms Linux, Windows, MacOS.
Notes:
* To support C++ test discovery, the env variable `CPP_TESTS_DIR` can be set to where the C++ test binaries is located
* Support pytest -k argument via run_test as this is used by pytest-cpp to replace `--gtest-filter`
* The XML output is in pytest format, but it's ok now because we don't have slow test or flaky test support for C++ test yet
* ~~I need to figure out why conftest.py doesn't work when I invoke pytest directly for C++ test, so `--sc` is not available for C++ tests at the moment. Proper pytest plugin like stepwise works fine though. I'll investigate and fix it in a separate PR~~ Found the cause, `conftest.py` is per directory and needs to be in any arbitrary directory that holds C++ test
* Two tests `test_api` and `test_tensorexpr` timed out on ASAN, I suspect that ASAN is now used on top of the python executable, which is slower than running native C++ code. IMO, it's ok to run these tests as before on ASAN for now
Pull Request resolved: https://github.com/pytorch/pytorch/pull/99956
Approved by: https://github.com/clee2000, https://github.com/ZainRizvi
Today, we prioritize running test files that were edited in the user's PR, with the idea being to run them before we run any other test.
Except, if the modified test is supposed to run serially, then we still end up running it after all the parallelized tests have finished running.
This PR fixes that to _always_ run the prioritized tests before the regular tests, regardless of if the test is supposed to run serially or in parallel
Pull Request resolved: https://github.com/pytorch/pytorch/pull/100748
Approved by: https://github.com/huydhn
i think get_reordered_tests broken since master -> main switch
add typing for some functions
checked for `prioritized` in the logs
limited testing because I only care about one very small part of the log thats near the beginning
Pull Request resolved: https://github.com/pytorch/pytorch/pull/100752
Approved by: https://github.com/huydhn
Fixes #ISSUE_NUMBER
* change hook so that test still gets saved in --sc when fails in test setup (caused an off by 1 error due to setup being called before the logreport hook)
* allow reruns for all tests now that --sc is used
* increase number of reruns now that --sc is used
Pull Request resolved: https://github.com/pytorch/pytorch/pull/100200
Approved by: https://github.com/huydhn
* add stepcurrent flag (--sc) based off the stepwise flag that saves the currently running test so that test running can resume from the last successful test after segfaults, takes in an argument for a key so that different test runs dont overwrite each other
* send sigint to process when timeout so that xml can be made
* add currently unused stepcurrent skip flag (--scs) based off stepwise skip flag that skips the failing test, was going to use if for the keep-going label but having trouble with CI
Pull Request resolved: https://github.com/pytorch/pytorch/pull/98035
Approved by: https://github.com/huydhn
This depends on [pytest-cpp](https://github.com/pytest-dev/pytest-cpp) to discover and run C++ tests with pytest. C++ tests are built under `${WORKSPACE}/build/bin` directory and copied to the test job under the same path.
* To expose them to `run_test`, I choose to use the mock path prefix `cpp`, for example `build/bin/c10_Array_test` would be named as `cpp/c10_Array_test` and the `python test/run_test.py --cpp -i cpp/c10_Array_test` would run the test in the same way as other Python tests. I could copy them from `build/bin` to `test/cpp`, but it will be mixed with the source code and CMake file. So this looks easier
* Some executable under `build/bin` are not C++ tests, and they are exclude, for example `build/bin/torch_shm_manager`
* C++ tests need to run with pytest directly as python command doesn't understand it
* The change is gated by the new `--cpp` argument to `run_test.py`, for example `python test/run_test.py --cpp` will run all available C++ tests
* The tests can be run in parallel
* Failing tests can be retried with `--reruns=2` and `--sw`
```
============================= test session starts ==============================
platform darwin -- Python 3.9.15, pytest-7.2.0, pluggy-1.0.0 -- /Users/huydo/miniconda3/envs/py3.9/bin/python3
cachedir: .pytest_cache
hypothesis profile 'default' -> database=DirectoryBasedExampleDatabase('/Users/huydo/Storage/mine/pytorch/test/.hypothesis/examples')
rootdir: /Users/huydo/Storage/mine/pytorch, configfile: pytest.ini
plugins: xdoctest-1.1.0, cpp-2.3.0, rerunfailures-10.3, shard-0.1.2, flakefinder-1.1.0, hypothesis-6.56.4, xdist-3.0.2, repeat-0.9.1
collecting ... collected 3 items / 2 deselected / 1 selected
Running 1 items in this shard: build/bin/scalar_tensor_test::TestScalarTensor.TestScalarTensorMPS
stepwise: skipping 2 already passed items.
../build/bin/scalar_tensor_test::TestScalarTensor::TestScalarTensorMPS RERUN [100%]
../build/bin/scalar_tensor_test::TestScalarTensor::TestScalarTensorMPS RERUN [100%]
../build/bin/scalar_tensor_test::TestScalarTensor::TestScalarTensorMPS FAILED [100%]
```
* `--import-slow-tests` and `--import-disabled-tests` won't work for now and that's ok to have it as a future task.
I also add `pytest-cpp==2.3.0` to Linux Docker, MacOS, and Windows.
### Testing
Build PyTorch and run `python test/run_test.py --cpp` on my laptop. CI change would come later in a separate PR. Also running `python test/run_test.py --help` now shows all C++ test discovered under `build/bin`
Pull Request resolved: https://github.com/pytorch/pytorch/pull/99559
Approved by: https://github.com/clee2000
Sharing code between the code that handles test results in parallel vs serial mode.
Note that the original version of this code had an inconsistency between the two versions where it would execute `print_to_stderr(err_message)` on every test that ran in parallel, but for serial tests it would only invoke `print_to_stderr(err_message)` if `continue_on_error` was also specified. By sharing code, this PR changes that behavior to be consistent between the two modes.
Also adding some comments.
<!--
copilot:poem
-->
### <samp>🤖 Generated by Copilot at 029342c</samp>
> _Sing, O Muse, of the skillful coder who refined_
> _The PyTorch testing script, `run_test.py`, and shined_
> _A light on its obscure logic, with docstrings and comments_
> _And made it run more smoothly, with better error contents_
Pull Request resolved: https://github.com/pytorch/pytorch/pull/99467
Approved by: https://github.com/huydhn, https://github.com/malfet
Common advice we give for handling memory fragmentation issues is to
allocate a big block upfront to reserve memory which will get split up later.
For programs with changing tensor sizes this can be especially helpful to
avoid OOMs that happen the first time we see a new largest input and would
otherwise have to allocate new segments.
However the issue with allocating a block upfront is that is nearly impossible
to correctly estimate the size of that block. If too small, space in the block
will run out and the allocator will allocate separate blocks anyway. Too large,
and other non-PyTorch libraries might stop working because they cannot allocate
any memory.
This patch provides the same benefits as using a pre-allocating block but
without having to choose its size upfront. Using the cuMemMap-style APIs,
it adds the ability to expand the last block in a segment when more memory is
needed.
Compared to universally using cudaMallocAsync to avoid fragmentation,
this patch can fix this common fragmentation issue while preserving most
of the existing allocator behavior. This behavior can be enabled and disabled dynamically.
This should allow users to, for instance, allocate long-lived parameters and state in individual buffers,
and put temporary state into the large expandable blocks, further reducing
fragmentation.
See inline comments for information about the implementation and its limitations.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/96995
Approved by: https://github.com/eellison
In C++ we have TORCH_LIBRARY_FRAGMENT. This PR adds the same
functionality to the Python torch.library API.
The motivation for this is: for the simple custom op API, we don't want
users to need to deal with Library objects. One way to hide this from
users is to create library fragments.
Test Plan:
- tests that you can create multiple fragments and def+impl operators on each.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/98439
Approved by: https://github.com/ezyang, https://github.com/bdhirsh
This has been bugging me for a while as I'm working on these Python scripts and they are not tracked by ufmt linter. So I add these script into that linter.
```
[[linter]]
code = 'UFMT'
include_patterns = [
'.github/**/*.py',
'test/run_test.py',
```
This change should just work and not break anything as ufmt (black + usort) linter is very safe to use for standalone util scripts.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/97588
Approved by: https://github.com/kit1980
Fixes#96347
This PR:
- Makes the functorch tests run as a part of the "default" shards
- Delete the functorch CI shard from all CI job configurations (if it exists)
- Increase the "default" shard count by 1 for each job, unless it was
previously set to 1, to accommodate the new functorch tests and not
regress time-to-signal.
- Adds a bunch of skips for ROCM and torchdynamo configurations. We can
investigate them later.
NB: I might go through some more iterations to figure out what other
skips need to be added, but this iteration of the PR seems to pass most CI.
suite.
Test Plan:
- wait for CI
Pull Request resolved: https://github.com/pytorch/pytorch/pull/96464
Approved by: https://github.com/huydhn
Enables the last few files under pytest.
xdist was causing problems with `profiler/test_profiler` `test_source_multithreaded` due to creating extra threads. Luckily we don't use it so we can disable it with `-p no:xdist`, but this is incompatible with pytest-rerunfailures==10.2, so upgrade to 10.3. I'd update the windows ami but idk how.
`dynamo/test_optimizers` and `dynamo/test_repros` both had tests that used skip_if_pytest. https://github.com/pytorch/pytorch/pull/93251/files suggests that it is due to pytest assertion rewriting, so I added `PYTEST_DONT_REWRITE` to their module docstrings to prevent pytest from rewriting assertions.
Disable test by issue in `dynamo/test_dynamic_shapes` seems sane.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/96698
Approved by: https://github.com/huydhn, https://github.com/malfet
Enable pytest for a few unique files. pytest runs tests in a different order than unittest (but still a consistent ordering with respect to itself) and some tests change global state, causing other tests to fail.
`test_transpose_non_contiguous` in `test_torchinductor.py` gets impacted from some other test but I'm not sure which one, so my solution is to reset the metrics before the rest of the test is run.
`test_register_patterns` in `test_quantize_fx.py` adds extra keys to global variables, so remove them when the test is done via unittest's `addCleanUp` which also works on pytest.
pytest doesn't really have an equivalent for `load_tests` so change it to be like `test_jit` that imports all the classes. I also attempted to dynamically import them, but I failed.
`test_public_api_surface` in `test_fx.py` checks for a backwards compatibility classification. There is a different test in test_fx that results in `fuser_utils` being imported. pytest runs this test before `test_public_api_surface` while unittest runs it after, so pytest sees `fuser_utils` when crawling through the modules.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/96397
Approved by: https://github.com/huydhn
Set environment variable
```
PYTORCH_TEST_DO_NOT_USE_PYTEST=1
```
to not use pytest in pytorch unit testing.
This change is related to some recent changes, e.g. #96210, #96016, #95844, #95659, that enabled the use of pytest in many test modules. Those test modules were testing normally before, but failed immediately after pytest is used. Sample stacktraces are:
```python
root@8e3168a83ee2:/opt/pytorch/pytorch# python test/run_test.py -v -i test_optim -- -v --save-xml
Ignoring disabled issues: []
/opt/pytorch/pytorch/test/run_test.py:1225: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
if torch.version.cuda is not None and LooseVersion(torch.version.cuda) >= "11.6":
Selected tests:
test_optim
parallel (file granularity) tests:
test_optim
serial (file granularity) tests:
Ignoring disabled issues: []
Ignoring disabled issues: []
Running test_optim ... [2023-03-09 12:51:59.358110]
Executing ['/usr/local/bin/python', '-bb', 'test_optim.py', '-v', '--save-xml', '-v', '--use-pytest', '-vv', '-rfEX', '-x', '--reruns=2'] ... [2023-03-09 12:51:59.358810]
Test results will be stored in test-reports/python-pytest/test_optim/test_optim-5e41643c8bac8ace.xml
Traceback (most recent call last):
File "/opt/pytorch/pytorch/test/test_optim.py", line 4581, in <module>
run_tests()
File "/opt/pytorch/pytorch/torch/testing/_internal/common_utils.py", line 796, in run_tests
exit_code = pytest.main(args=pytest_args)
File "/usr/local/lib/python3.10/site-packages/_pytest/config/__init__.py", line 148, in main
config = _prepareconfig(args, plugins)
File "/usr/local/lib/python3.10/site-packages/_pytest/config/__init__.py", line 329, in _prepareconfig
config = pluginmanager.hook.pytest_cmdline_parse(
File "/usr/local/lib/python3.10/site-packages/pluggy/_hooks.py", line 265, in __call__
return self._hookexec(self.name, self.get_hookimpls(), kwargs, firstresult)
File "/usr/local/lib/python3.10/site-packages/pluggy/_manager.py", line 80, in _hookexec
return self._inner_hookexec(hook_name, methods, kwargs, firstresult)
File "/usr/local/lib/python3.10/site-packages/pluggy/_callers.py", line 55, in _multicall
gen.send(outcome)
File "/usr/local/lib/python3.10/site-packages/_pytest/helpconfig.py", line 103, in pytest_cmdline_parse
config: Config = outcome.get_result()
File "/usr/local/lib/python3.10/site-packages/pluggy/_result.py", line 60, in get_result
raise ex[1].with_traceback(ex[2])
File "/usr/local/lib/python3.10/site-packages/pluggy/_callers.py", line 39, in _multicall
res = hook_impl.function(*args)
File "/usr/local/lib/python3.10/site-packages/_pytest/config/__init__.py", line 1060, in pytest_cmdline_parse
self.parse(args)
File "/usr/local/lib/python3.10/site-packages/_pytest/config/__init__.py", line 1348, in parse
self._preparse(args, addopts=addopts)
File "/usr/local/lib/python3.10/site-packages/_pytest/config/__init__.py", line 1231, in _preparse
self.pluginmanager.load_setuptools_entrypoints("pytest11")
File "/usr/local/lib/python3.10/site-packages/pluggy/_manager.py", line 287, in load_setuptools_entrypoints
plugin = ep.load()
File "/usr/local/lib/python3.10/importlib/metadata/__init__.py", line 171, in load
module = import_module(match.group('module'))
File "/usr/local/lib/python3.10/importlib/__init__.py", line 126, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
File "<frozen importlib._bootstrap>", line 1050, in _gcd_import
File "<frozen importlib._bootstrap>", line 1027, in _find_and_load
File "<frozen importlib._bootstrap>", line 1006, in _find_and_load_unlocked
File "<frozen importlib._bootstrap>", line 688, in _load_unlocked
File "/usr/local/lib/python3.10/site-packages/_pytest/assertion/rewrite.py", line 168, in exec_module
exec(co, module.__dict__)
File "/usr/local/lib/python3.10/site-packages/xdist/looponfail.py", line 16, in <module>
import execnet
File "/usr/local/lib/python3.10/site-packages/execnet/__init__.py", line 14, in <module>
from .gateway_base import DataFormatError
File "/usr/local/lib/python3.10/site-packages/execnet/gateway_base.py", line 1138, in <module>
FLOAT_FORMAT_SIZE = struct.calcsize(FLOAT_FORMAT)
BytesWarning: Comparison between bytes and string
FINISHED PRINTING LOG FILE of test_optim (/opt/pytorch/pytorch/test/test-reports/test_optim_1pnlesrz.log)
test_optim failed!
Traceback (most recent call last):
File "/opt/pytorch/pytorch/test/run_test.py", line 1428, in <module>
main()
File "/opt/pytorch/pytorch/test/run_test.py", line 1386, in main
raise RuntimeError(
RuntimeError: test_optim failed!
Tip: You can keep running tests even on failure by passing --keep-going to run_test.py.
If running on CI, add the 'keep-going' label to your PR and rerun your jobs.
```
I'd like to propose this option that allows users to use the good old python unit test way instead of pytest to run their testing in CI.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/96444
Approved by: https://github.com/malfet
Run more tests through pytest.
Use a block list for tests that shouldn't run through pytest. As far as I can tell, the number of tests run, skipped, and xfailed for those not on the blocklist are the same.
Regarding the main module:
Usually tests are run in CI, we call `python <test file>`, which causes the file to be imported under the module name `__main__`. However, pytest searches for the module to be imported under the file name, so the file will be reimported. This can cause issues for tests that run module level code and change global state, like test_nn, which modifies lists imported from another file, or tests in test/lazy, which initialize a backend that cannot coexist with a second copy of itself.
My workaround for this is to run tests from the `__main__` module. However, this results in pytest being unable to rewrite assertions (and possibly other things but I don't know what other things pytest does right now). A better solution might be to call `pytest <test file>` directly and move all the code in run_tests(argv) to be module level code or put it in a hook in conftest.py.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/95844
Approved by: https://github.com/huydhn
Summary: Currently running PyTorch tests with dynamo and inductor is
controlled by environment variables, and CI sets them based on test
config name matching. Change them to use options of run_test.py.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/94539
Approved by: https://github.com/huydhn
Part of my effort to move everything to pytest and decrease the number of testrunner frameworks in ci
Gives xmls but they might look a weird b/c module level tests vs tests in classes.
Doesn't give skip/disable test infra because those are tied to classes. (for future ref, could either put tests in classes or move the check_if_enable stuff into a pytest hook)
Tested in CI and checked that the same number of tests are run
Pull Request resolved: https://github.com/pytorch/pytorch/pull/95659
Approved by: https://github.com/huydhn
Time comparison between using MultithreadedTestCase and MultiProcessTestCase on op db tests is amazing!
using MultiThreadTestCase on a AWS dev node:
```
time pytest test/distributed/_tensor/test_dtensor_ops.py
============= 175 passed, 42 skipped, 397 xfailed in 80.30s (0:01:20) =======
real 1m22.330s
user 1m38.782s
sys 0m18.762s
```
MultiProcessTestCase spends from 40mins to more than 1h, even if using pytest parallel testing tools.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/92198
Approved by: https://github.com/XilunWu
Fixes#90940. This PR revamps how tests are run in parallel as well as device visibility at the docker container and within the run_test.py test runner.
First, running multiple test modules concurrently on the same GPU was causing instability for ROCm runners manifesting as timeouts. ROCm runners have at least 1 GPU each, but often 2 or more. This PR allows NUM_PROCS to be set equal to the number of devices available, but also takes care to set HIP_VISIBLE_DEVICES to avoid oversubscribing any GPU.
Second, we had introduced env vars `-e ROCR_VISIBLE_DEVICES` (#91031) to prepare for two GHA runners per CI node, to split up the GPU visibility at the docker level between the two runners. This effort wasn't fully realized; to date, we haven't had more than one runner per CI host. We abandon this effort in favor of all GPUs being visible to a single runner and managing GPU resources as stated above.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/91137
Approved by: https://github.com/kit1980, https://github.com/huydhn, https://github.com/pruthvistony
Rerun all disabled test to gather their latest result so that we can close disabled tickets automatically. When running under this mode (RERUN_DISABLED_TESTS=true), only disabled tests are run while the rest are skipped `<skipped message="Test is enabled but --rerun-disabled-tests verification mode is set, so only disabled tests are run" type="skip"/>`
The logic is roughly as follows, the test runs multiple times (n=50)
* If the disabled test passes, and it's flaky, do nothing because it's still flaky. In the test report, we'll see the test passes with the following skipped message:
```
<testcase classname="TestMultiprocessing" file="test_multiprocessing.py" line="357" name="test_fs" time="0.000" timestamp="0001-01-01T00:00:00">
<skipped message="{"flaky": True, "num_red": 4, "num_green": 0, "max_num_retries": 3, "rerun_disabled_test": true}" type="skip"/>
</testcase>
```
* If the disabled test passes every single time, and it is not flaky anymore, mark it so that it can be closed later. We will see the test runs and passes, i.e.
```
<testcase classname="TestCommonCUDA" name="test_out_warning_linalg_lu_factor_cuda" time="0.170" file="test_ops.py" />
```
* If the disabled test fails after all retries, this is also expected. So only report this but don't fail the job (because we don't care about red signals here), we'll see the test is skipped (without the `flaky` field), i.e.
```
<testcase classname="TestMultiprocessing" file="test_multiprocessing.py" line="357" name="test_fs" time="0.000" timestamp="0001-01-01T00:00:00">
<skipped message="{"num_red": 4, "num_green": 0, "max_num_retries": 3, "rerun_disabled_test": true}" type="skip"/>
</testcase>
```
This runs at the same schedule as `mem_leak_check` (daily). The change to update test stats, and (potentially) grouping on HUD will come in separated PRs.
### Testing
* pull https://github.com/pytorch/pytorch/actions/runs/3447434434
* trunk https://github.com/pytorch/pytorch/actions/runs/3447434928
Pull Request resolved: https://github.com/pytorch/pytorch/pull/88646
Approved by: https://github.com/clee2000
Fixes: https://github.com/pytorch/pytorch/issues/88010
This PR does a couple things to stop slow gradcheck from timing out:
- Splits out test_ops_fwd_gradients from test_ops_gradients, and factors out TestFwdGradients and TestBwdGradients which both inherit from TestGradients, now situated in common_utils (maybe there is a better place?)
- Skips CompositeCompliance (and several other test files) for slow gradcheck CI since they do not use gradcheck
- because test times for test_ops_fwd_gradients and test_ops_gradients are either unknown or wrong, we hardcode them for now to prevent them from being put together. We can undo the hack after we see actual test times are updated. ("def calculate_shards" randomly divides tests with unknown test times in a round-robin fashion.)
- Updates references to test_ops_gradients and TestGradients
- Test files that are skipped for slow gradcheck CI are now centrally located in in run_tests.py, this reduces how fine-grained we can be with the skips, so for some skips (one so far) we still use the old skipping mechanism, e.g. for test_mps
Pull Request resolved: https://github.com/pytorch/pytorch/pull/88216
Approved by: https://github.com/albanD
