Compare commits

..

56 Commits

Author SHA1 Message Date
7782f2866c [dynamo] fix 3.11+ refleak for 2.3.1 release (#126235) 2024-05-15 13:21:19 -04:00
be9a4076f0 Do not import transformers when import torch._dynamo (#124634) (#125755)
Fixes https://github.com/pytorch/pytorch/issues/123954

Pull Request resolved: https://github.com/pytorch/pytorch/pull/124634
Approved by: https://github.com/thiagocrepaldi, https://github.com/Chillee
ghstack dependencies: #124343
2024-05-15 13:15:44 -04:00
d114e0488c [Release only] Advance triton version to 2.3.1 (#126204)
* Advance tition version to 2.3.1

* update_commit_pin

* pin

* pin rocm

* lint
2024-05-15 11:13:52 -04:00
194698a4ac Revert "[Release only] Release 2.3 start using triton package from pypi" (#126202)
Revert "[Release only] Release 2.3 start using triton package from pypi (#123…"

This reverts commit 97ff6cfd9c86c5c09d7ce775ab64ec5c99230f5d.
2024-05-14 17:01:07 -04:00
1d6a938090 [cherry-pick][device_mesh] add back the private init backend option (#124780) (#126147)
[device_mesh] add a private init backend option (#124780)

This PR adds a private init backend option, to tackle the issues sub
mesh creation:

in device mesh slicing we don't want to create process groups again,
so explicitly turn the group creation off it's useful

Also I think there might be more submesh creation functionality so
having this flag would ensure that there's no new group created

Differential Revision: [D56497780](https://our.internmc.facebook.com/intern/diff/D56497780)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/124780
Approved by: https://github.com/awgu
2024-05-14 13:38:55 -04:00
4f0b3ad855 Fix ref leak in dtype.to_complex()/to_real() (#126101)
* Fix ref leak in `dtype.to_complex()`/`to_real()` (#125154)

By using `Py_NewRef`

Also, wrap `THPDtype_to_real`/`THPDtype_to_complex` calls with `HANDLE_TH_ERRORS`

Add regression test for the above issues, by calling to_complex for integral dtypes, that raises an exception and by preserving reference count to the same to_complex/to_real call to detect if leak is happeneing.

Replace
```cpp
auto dtype = (PyObject*)torch::getTHPDtype(current_dtype);
Py_INCREF(dtype);
return dtype;
```
with a more compact/streamlined equivalent
```cpp
return Py_NewRef(torch::getTHPDtype(current_dtype));
```

Fixes https://github.com/pytorch/pytorch/issues/124868

Pull Request resolved: https://github.com/pytorch/pytorch/pull/125154
Approved by: https://github.com/Skylion007, https://github.com/albanD

(cherry picked from commit 744f341aa4eaa4f2e7068e5f83fa6fccb0a02ccc)

* Revert "Fix ref leak in `dtype.to_complex()`/`to_real()` (#125154)"

This reverts commit a1b04d8832c64d6112b432ce1e7725f02761cf21.

* Fix ref leak in `dtype.to_complex()`/`to_real()` (#125154)

By using `Py_NewRef`

Also, wrap `THPDtype_to_real`/`THPDtype_to_complex` calls with `HANDLE_TH_ERRORS`

Add regression test for the above issues, by calling to_complex for integral dtypes, that raises an exception and by preserving reference count to the same to_complex/to_real call to detect if leak is happeneing.

Replace
```cpp
auto dtype = (PyObject*)torch::getTHPDtype(current_dtype);
Py_INCREF(dtype);
return dtype;
```
with a more compact/streamlined equivalent
```cpp
return Py_NewRef(torch::getTHPDtype(current_dtype));
```

Fixes https://github.com/pytorch/pytorch/issues/124868

Pull Request resolved: https://github.com/pytorch/pytorch/pull/125154
Approved by: https://github.com/Skylion007, https://github.com/albanD

(cherry picked from commit 744f341aa4eaa4f2e7068e5f83fa6fccb0a02ccc)

* Revert "Fix ref leak in `dtype.to_complex()`/`to_real()` (#125154)"

This reverts commit 5a28bad418ad8187036c64b3fe487109dfc4ff5d.

* Refactor autocast C++ APIs to be device-agnostic (#124359)

# Motivation
This PR aims to refactor autocast **C++** APIs to be device-agnostic and deprecate the device-specific autocast  **C++** APIs.
In C++ side,
- `is_enabled()` -> `is_enabled(device_type)`.
- `set_enabled(new_enabled)` -> `set_enabled(device_type, new_enabled)`.
- `get_autocast_dtype()` -> `get_autocast_dtype(device_type)`
- `set_autocast_dtype(dtype)` -> `set_autocast_dtype(device_type, dtype)`

These following C++ APIs are deprecated and should be removed in PyTorch 2.5
- `is_cpu_enabled`
- `set_cpu_enabled`
- `get_autocast_cpu_dtype`
- `set_autocast_cpu_dtype`
- `is_xpu_enabled`
- `set_xpu_enabled`
- `get_autocast_xpu_dtype`
- `set_autocast_xpu_dtype`
- `is_ipu_enabled`
- `set_ipu_enabled`
- `get_autocast_ipu_dtype`
- `set_autocast_ipu_dtype`
- `is_hpu_enabled`
- `set_hpu_enabled`
- `get_autocast_hpu_dtype`
- `set_autocast_hpu_dtype`
- `is_xla_enabled`
- `set_xla_enabled`
- `get_autocast_xla_dtype`
- `set_autocast_xla_dtype`
- `is_privateuseone_enabled`
- `set_privateuseone_enabled`
- `get_autocast_privateuseone_dtype`
- `set_autocast_privateuseone_dtype`

In Python side,
provide 4 generic autocast APIs:
- `torch.is_autocast_enabled(device_type)`
- `torch.set_autocast_enabled(device_type, new_enabled)`
- `torch.get_autocast_dtype(device_type)`
- `torch.set_autocast_dtype(device_type, dtype)`

# Additional Context
We will submit another PR to refactor autocast **Python** APIs based on this PR.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/124359
Approved by: https://github.com/jgong5, https://github.com/albanD

* refactor autocast python APIs (#124479)

Refactor autocast usage scenario in `torch/amp/autocast_mode.py` and `torch/utils/checkpoint.py` to fix the bug - convention conflict between `torch.xxx.get_autocast_xxx_dtype` defined in `autocast_mode.py` and `torch.xxx.get_autocast_dtype` defined in `checkpoint.py`.

Use device-agnostic APIs like `torch.get_autocast_dtype`, ..., instead.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/124479
Approved by: https://github.com/jgong5, https://github.com/gujinghui, https://github.com/EikanWang, https://github.com/albanD
ghstack dependencies: #124359

* Fix ref leak in `dtype.to_complex()`/`to_real()` (#125154)

By using `Py_NewRef`

Also, wrap `THPDtype_to_real`/`THPDtype_to_complex` calls with `HANDLE_TH_ERRORS`

Add regression test for the above issues, by calling to_complex for integral dtypes, that raises an exception and by preserving reference count to the same to_complex/to_real call to detect if leak is happeneing.

Replace
```cpp
auto dtype = (PyObject*)torch::getTHPDtype(current_dtype);
Py_INCREF(dtype);
return dtype;
```
with a more compact/streamlined equivalent
```cpp
return Py_NewRef(torch::getTHPDtype(current_dtype));
```

Fixes https://github.com/pytorch/pytorch/issues/124868

Pull Request resolved: https://github.com/pytorch/pytorch/pull/125154
Approved by: https://github.com/Skylion007, https://github.com/albanD

* Revert "refactor autocast python APIs (#124479)"

This reverts commit 495b0c9aec07472d82b9fa5e5cdaab35ec16898d.

* Revert "Refactor autocast C++ APIs to be device-agnostic (#124359)"

This reverts commit 83106b7c4f7d6cc84ccc8beb0949eaa94c830d8d.

---------

Co-authored-by: Nikita Shulga <2453524+malfet@users.noreply.github.com>
Co-authored-by: Huy Do <huydhn@gmail.com>
Co-authored-by: Yu, Guangye <guangye.yu@intel.com>
2024-05-14 13:29:45 -04:00
bf1b3a056a Lint: Update older-python test to 3.6 (#126108)
Lint: Update older-python test to 3.6 (#125843)

As python-3.5 can no longer connect to pypi after today's cert update
Fixes https://github.com/pytorch/pytorch/issues/125841

(cherry picked from commit 7e86a7c0155295539996e0cf422883571126073e)

Co-authored-by: Nikita Shulga <2453524+malfet@users.noreply.github.com>
2024-05-13 14:45:40 -07:00
c365674171 Fixes format utils executable (#123482)
Fixes format utils executable (#123407)

Fixes an issue with the format utils executable, which was causing it to run as a no-op. :(

Pull Request resolved: https://github.com/pytorch/pytorch/pull/123407
Approved by: https://github.com/wz337, https://github.com/fegin

(cherry picked from commit 18c9d460682293bae23386e9bc40932eee1d8add)

Co-authored-by: Lucas Pasqualin <lpasqualin@meta.com>
2024-05-13 14:44:35 -07:00
768e4b9420 [MPS] Fix large copy (#126104)
[MPS] Fix large copy (#124635)

By slicing `copyFromBuffer:sourceOffset:toBuffer:destinationOffset:size:` into 2Gb chunks

Add regression test, but limit it to machines with 12Gb of RAM or more, and MacOS 14+, as on MacOS 13 attempt to alloc 4Gb tensor fails with:
```
/AppleInternal/Library/BuildRoots/c651a45f-806e-11ed-a221-7ef33c48bc85/Library/Caches/com.apple.xbs/Sources/MetalPerformanceShaders/MPSCore/Types/MPSNDArray.mm:724: failed assertion `[MPSNDArray initWithDevice:descriptor:] Error: total bytes of NDArray > 2**32'
```

Fixes https://github.com/pytorch/pytorch/issues/124335

Pull Request resolved: https://github.com/pytorch/pytorch/pull/124635
Approved by: https://github.com/kulinseth

(cherry picked from commit abf3f90781263ee45e2e79cf7f80102ffa7f1b14)

Co-authored-by: Nikita Shulga <nshulga@meta.com>
2024-05-13 14:42:29 -07:00
e25474c05d [MPS] Native nonzero implementation (#126100)
[MPS] Native nonzero implementation (#125355)

Fixes https://github.com/pytorch/pytorch/issues/124850

Replace previous MPSGraph nonzero construction with native nonzero op. For older OSes, fallback to CPU (previous implementation was not reliable and was comparable to CPU in speed).

Pull Request resolved: https://github.com/pytorch/pytorch/pull/125355
Approved by: https://github.com/kulinseth

(cherry picked from commit a40d6df448de1acb263ed8f6ff9e7d26f5a1a161)

Co-authored-by: Denis Vieriu <dvieriu@apple.com>
2024-05-13 14:39:39 -07:00
d8b35dac22 Publish PyTorch docs to pytorch/cpp repo (#126102)
Publish PyTorch docs to pytorch/cpp repo (#122895)

Updating the documents push to go to https://github.com/pytorch/docs repo instead of https://github.com/pytorch/pytorch.github.io as part of updating the PyTorch docs set up.
Co-authored-by: Nikita Shulga <2453524+malfet@users.noreply.github.com>
Co-authored-by: Svetlana Karslioglu <svekars@meta.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/122895
Approved by: https://github.com/malfet

(cherry picked from commit 535a84c12553a6a423bb3f7f7b175ee7bf8d5464)

Co-authored-by: sekyondaMeta <127536312+sekyondaMeta@users.noreply.github.com>
2024-05-13 14:35:46 -07:00
bbb838654c Separate arm64 and amd64 docker builds (#126099)
Separate arm64 and amd64 docker builds (#125617)

Fixes https://github.com/pytorch/pytorch/issues/125094

Please note: Docker CUDa 12.4 failure is existing issue, related to docker image not being available on gitlab:
```
docker.io/nvidia/cuda:12.4.0-cudnn8-devel-ubuntu22.04: docker.io/nvidia/cuda:12.4.0-cudnn8-devel-ubuntu22.04: not found
```
 https://github.com/pytorch/pytorch/actions/runs/8974959068/job/24648540236?pr=125617

Here is the reference issue: https://gitlab.com/nvidia/container-images/cuda/-/issues/225

Tracked on our side: https://github.com/pytorch/builder/issues/1811
Pull Request resolved: https://github.com/pytorch/pytorch/pull/125617
Approved by: https://github.com/huydhn, https://github.com/malfet

(cherry picked from commit b29d77b54f5cee9e786f8a72329837043f36a349)

Co-authored-by: atalman <atalman@fb.com>
2024-05-13 14:33:28 -07:00
d983cb78e2 [MPS] Fix abs for complex types (#126096)
[MPS] Fix `abs` for complex types (#125662)

By calling `realPartOfTensor:` if input type is complex on Sonoma and fall back to `at::view_as_real` trick on Ventura.

Split `unary_op` template into `unary_op` and `unary_op_noresize`, which skips resize and empty checks

Marked `abs`, `isclose` and `nn.functional.softsign` OpInfo tests as supported by complex types

Fixes https://github.com/pytorch/pytorch/issues/125135

Pull Request resolved: https://github.com/pytorch/pytorch/pull/125662
Approved by: https://github.com/kulinseth

(cherry picked from commit 0fd1fc17c3a53cb4cede1992d431d5384b2813f3)

Co-authored-by: Nikita Shulga <nikita.shulga@gmail.com>
2024-05-13 14:32:05 -07:00
75e01e7df0 Add userbase library dir to windows dll search path (#126095)
Add userbase library dir to windows dll search path (#125684)

Fixes https://github.com/pytorch/pytorch/issues/125109 which is a regression introduced by https://github.com/pytorch/builder/pull/1467 that adds dynamic dependency to mkl, which if installed in the user-dir is placed into `sysconfig.sysconfig.get_config_var("userbase") / "Library" / "bin"`

Fix this one, but adding `userbase` folder to the DLL search path

Testing before this fix:
```
Python 3.12.3 (tags/v3.12.3:f6650f9, Apr  9 2024, 14:05:25) [MSC v.1938 64 bit (AMD64)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import torch
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "C:\Users\Administrator\AppData\Roaming\Python\Python312\site-packages\torch\__init__.py", line 141, in <module>
    raise err
OSError: [WinError 126] The specified module could not be found. Error loading "C:\Users\Administrator\AppData\Roaming\Python\Python312\site-packages\torch\lib\shm.dll" or one of its dependencies.
>>> exit()
```

After:
```
c:\Program Files\Python312>python
Python 3.12.3 (tags/v3.12.3:f6650f9, Apr  9 2024, 14:05:25) [MSC v.1938 64 bit (AMD64)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import torch
>>> exit()
```
Co-authored-by: Nikita Shulga <2453524+malfet@users.noreply.github.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/125684
Approved by: https://github.com/malfet

(cherry picked from commit fdfef759a676ee7a853872e347537bc1e4b51390)

Co-authored-by: atalman <atalman@fb.com>
2024-05-13 14:30:59 -07:00
a696b3b7f6 Restore DILL_AVAILABLE for backwards compat with torchdata (#126094)
Restore DILL_AVAILABLE for backwards compat with torchdata (#122616)

Signed-off-by: Edward Z. Yang <ezyang@meta.com>

Pull Request resolved: https://github.com/pytorch/pytorch/pull/122616
Approved by: https://github.com/peterbell10

(cherry picked from commit f42818321b5694c5be76f16173c7ac9223ec8ae3)

Co-authored-by: Edward Z. Yang <ezyang@meta.com>
2024-05-13 14:30:14 -07:00
eqy
2e165ec9c2 Revert "Include torch warn in each error in cudnn/Conv_v8.cpp (#120719)" (#125790)
This reverts commit 5fd7f5c4e336c2c3041e10529990c620cc8cf9a5.

Reverted https://github.com/pytorch/pytorch/pull/120719 on behalf of https://github.com/janeyx99 due to sorry but am reverting as this prints unwanted warnings even when an exception is not thrown  ([comment](https://github.com/pytorch/pytorch/pull/120719#issuecomment-1994491826))

Co-authored-by: PyTorch MergeBot <pytorchmergebot@users.noreply.github.com>
2024-05-13 15:48:42 -04:00
1199df476e [Doc] Update docstrings for torch/random.py (#125443)
[Doc] Update docstrings for torch/random.py (#125265)

Updates the docstrings for torch/random.py to clarify what device / RNG each function operates on.

While trying to understand the difference between
```
state = torch.random.get_rng_state()
some_code
torch.random.set_rng_state(state)
```
and
```
with torch.random.fork_rng():
    some_code
```
I found out that there was a note about this in the docstring that wasn't being rendered on the website. I fixed that note and added additional clarifications on other functions in this file.

Test Plan:
Built the docs and verified that everything renders correctly.

<img width="911" alt="Screenshot 2024-04-30 at 2 22 08 PM" src="https://github.com/pytorch/pytorch/assets/9263852/f219bc35-89bd-4f5b-ba60-255b089499a4">

<img width="901" alt="Screenshot 2024-04-30 at 2 22 13 PM" src="https://github.com/pytorch/pytorch/assets/9263852/c141e7fa-afc9-4c66-b460-96668ce35606">

Pull Request resolved: https://github.com/pytorch/pytorch/pull/125265
Approved by: https://github.com/Balandat, https://github.com/lezcano
2024-05-13 15:46:47 -04:00
97ff6cfd9c [Release only] Release 2.3 start using triton package from pypi (#123580) 2024-04-08 16:27:33 -04:00
fb38ab7881 Fix for MPS regression in #122016 and #123178 (#123385)
Fixes #122016 and #123178. This regression is related to an OS side change that requires a slight adjustment from us on PyTorch side to restore the previous behavior. Additionally we cleared out pre-MacOS13 related workarounds.

Before the fix on MacOS 14.4:

```
python -c "import torch;x=torch.zeros(3, device='mps');x[1] = 1; x[2] = 3; print(x)"
tensor([0., 3., 3.], device='mps:0')
```

After the fix:
```
python -c "import torch;x=torch.zeros(3, device='mps');x[1] = 1; x[2] = 3; print(x)"
tensor([0., 1., 3.], device='mps:0')
```

This also fixes complex number initialization and as such makes `nn.functional.rms_norm` pass on MacOS-14+

Co-authored-by: Nikita Shulga <2453524+malfet@users.noreply.github.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/123234
Approved by: https://github.com/malfet, https://github.com/kulinseth

(cherry picked from commit 05289a278c3eaca271061649982f38c435b50674)

Co-authored-by: Joona Havukainen <jhavukainen@apple.com>
2024-04-05 18:46:31 -04:00
23961cef85 [Release/2.3] Set py3.x build-environment name consistently (#123446)
https://github.com/pytorch/pytorch/pull/122157 checks for the Python version using `"$BUILD_ENVIRONMENT" != *py3.8*`, but some build environment uses a different style with `py3_8` instead causing numpy 2.x to be installed there wrongly, i.e. 03b987fe3f
Pull Request resolved: https://github.com/pytorch/pytorch/pull/122247
Approved by: https://github.com/malfet

(cherry picked from commit 6fefc52a2b4f814c5bc85f4087a92ad7f6ee3abe)

Co-authored-by: Huy Do <huydhn@gmail.com>
2024-04-05 09:01:19 -07:00
634cf5069a [Wheel] Change libtorch_cpu OpenMP search path (#123417) (#123442)
To prevent delocate from double-packing it, which makes Torch wheels
unusable with torch.compile out of the box

Fixes https://github.com/pytorch/pytorch/issues/122705

Pull Request resolved: https://github.com/pytorch/pytorch/pull/123417
Approved by: https://github.com/atalman

Co-authored-by: Nikita Shulga <nikita.shulga@gmail.com>
2024-04-05 10:22:39 -04:00
12d0e693d0 update submodule onnx==1.16.0 (#123387)
Fixes #121258

CC @malfet @atalman
Pull Request resolved: https://github.com/pytorch/pytorch/pull/123125
Approved by: https://github.com/malfet

(cherry picked from commit 19c2ed15c099c7ed9f96074584af6ab9da206f92)

Co-authored-by: pbialecki <piotr.bialecki@hotmail.de>
2024-04-04 20:47:38 -04:00
38acd812ab [MPS] Fwd-fix for clamp regression (#122148) (#123383)
Forward fix for regressions introduced by https://github.com/pytorch/pytorch/pull/121381 as we failed to run MPS CI twice on it

- Do not call `minimumWithNaNPropagationWithPrimaryTensor` for integral tensors as it will crash with
  ```
    /AppleInternal/Library/BuildRoots/ce725a5f-c761-11ee-a4ec-b6ef2fd8d87b/Library/Caches/com.apple.xbs/Sources/MetalPerformanceShaders/MPSCore/Utility/MPSKernelDAG.mm:805: failed assertion `Error getting visible function: (null) Function isNaN_i16_i8 was not found in the library'
   ```
- Change the order of max and min call as it's apparently important for
  consistency, as `min(max(a, b), c)` might not equal to `max(min(a, c), b)` if `c` is not always less or equal than `b`

Pull Request resolved: https://github.com/pytorch/pytorch/pull/122148
Approved by: https://github.com/huydhn

Co-authored-by: Nikita Shulga <nikita.shulga@gmail.com>
2024-04-04 16:29:42 -07:00
b197f540bc Use numpy 2.0.0rc1 in CI (#123356)
Bump numpy version to 2.0.0rc1 in CI

Related to: https://github.com/pytorch/pytorch/issues/107302
Pull Request resolved: https://github.com/pytorch/pytorch/pull/123286
Approved by: https://github.com/huydhn, https://github.com/kit1980, https://github.com/ZainRizvi

(cherry picked from commit 26b4ccf9d171a4abb3b25d9f88fc594ea5aca1ce)

Co-authored-by: atalman <atalman@fb.com>
2024-04-04 19:02:49 -04:00
dc81d19aac [CI] Test that NumPy-2.X builds are backward compatible with 1.X (#123354)
By compiling PyTorch against 2.x RC, but running all the tests with Numpy-1.X

This has no affects on binary builds
Pull Request resolved: https://github.com/pytorch/pytorch/pull/122157
Approved by: https://github.com/atalman

(cherry picked from commit 03b987fe3fa93f398c0af5b40e512950c39a7cb6)

Co-authored-by: Nikita Shulga <2453524+malfet@users.noreply.github.com>
2024-04-04 19:00:35 -04:00
108305e47b Upgrade submodule pybind to 2.12.0 (#123355)
To fix https://github.com/pytorch/pytorch/issues/122056

Building with NP 2.0 allows me to run locally with both NP 2.0 and 1.26.
Any other test we should run @rgommers  ?

FYI @Skylion007 @atalman
Pull Request resolved: https://github.com/pytorch/pytorch/pull/122899
Approved by: https://github.com/Skylion007

(cherry picked from commit 6c2f36c9845f310db8ece23c0d2e4ad6f702bc57)

Co-authored-by: albanD <desmaison.alban@gmail.com>
2024-04-04 18:07:42 -04:00
a8b009185d Make PyTorch compilable against upcoming Numpy-2.0 (#121880) (#123380)
Test plan:
```
% python -c "import torch;import numpy;print(numpy.__version__, torch.tensor(numpy.arange(3, 10)))"
2.1.0.dev0+git20240312.9de8a80 tensor([3, 4, 5, 6, 7, 8, 9])
% python -c "import torch;print(torch.rand(3, 3).numpy())"
[[0.0931946  0.44874293 0.8480404 ]
 [0.93877375 0.10188377 0.67375803]
 [0.02520031 0.89019287 0.5691561 ]]

```
Fixes https://github.com/pytorch/pytorch/issues/121798

Pull Request resolved: https://github.com/pytorch/pytorch/pull/121880
Approved by: https://github.com/albanD

(cherry picked from commit 38d9bb5abcc31ba97927a5399b88afe2cf60bf64)

Co-authored-by: Nikita Shulga <2453524+malfet@users.noreply.github.com>
2024-04-04 14:22:26 -07:00
b67b277268 Fix torch.clamp in MPS to handle NaN correctly (#121381) (#122785)
Fixes #120899

So this is interesting. There are methods that specifically propagate NaN instead of clamping to real numbers.
https://developer.apple.com/documentation/metalperformanceshadersgraph/mpsgraph/3857573-maximumwithnanpropagationwithpri

Co-authored-by: Nikita Shulga <2453524+malfet@users.noreply.github.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/121381
Approved by: https://github.com/malfet

(cherry picked from commit 40acc84aafa82f00a5b3966302638f344bef07bd)

Co-authored-by: Roger Lam <mrlamroger@gmail.com>
2024-04-04 13:26:29 -07:00
a8f93a5c71 [ONNX] beartype to emit warning instead of error by default (#123363)
Making exporter more "robust" to advances in beartype tool.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/123205
Approved by: https://github.com/justinchuby, https://github.com/thiagocrepaldi
2024-04-04 16:13:58 -04:00
fa07dc5132 [MPS] Fix naive matmul for BFloat16 (#123289)
Will only work on MacOS14 or newer, so compile the shader with `MTLLanguageVersion_3_1` when appropriate

Fixes https://github.com/pytorch/pytorch/issues/121583
Pull Request resolved: https://github.com/pytorch/pytorch/pull/121731
Approved by: https://github.com/albanD

(cherry picked from commit 5498804ec2ac9aa62ba3bbf20149118142567d9b)

Co-authored-by: Nikita Shulga <nikita.shulga@gmail.com>
2024-04-04 16:04:46 -04:00
2a82d31f78 fix breaking changes for ONNX Runtime Training (#123271)
Fixes breaking changes for ONNX Runtime Training.

PR https://github.com/pytorch/pytorch/pull/121102 introduced incompatibility with ORT training because of change in parameter type. Creating a PR to add previous parameter types and verified that it works with ORT training.

Error with current scenario:

```
site-packages/onnxruntime/training/ortmodule/torch_cpp_extensions/cpu/aten_op_executor/aten_op_executor.cc:60:40: error: invalid conversion from ‘const DLManagedTensor*’ to ‘DLManagedTensor*’ [-fpermissive]
at::Tensor tensor = at::fromDLPack(dlpack);

site-packages/torch/include/ATen/DLConvertor.h:15:46: note:   initializing argument 1 of ‘at::Tensor at::fromDLPack(DLManagedTensor*)’
TORCH_API Tensor fromDLPack(DLManagedTensor* src);
```
Co-authored-by: Nikita Shulga <2453524+malfet@users.noreply.github.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/122000
Approved by: https://github.com/malfet

(cherry picked from commit 765c3fc138fda4b49978403ee1394040221957cc)

Co-authored-by: Abhishek Jindal <abjindal@microsoft.com>
2024-04-03 18:52:06 -04:00
4bb5cb51e6 Fix swap_tensors path in _apply for modules that inherit from RNNBase (RNN, GRU, LSTM) (#122800) (#123116)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/122800
Approved by: https://github.com/albanD

(cherry picked from commit cc12668053ad847ff4a430e99eeebf99c136f3cd)
2024-04-02 16:16:37 -07:00
ef38d0572e nn.Module: use swap_tensors for Tensor subclasses (#122755) (#123106)
This fixes a bug when casting a module that has DTensor parameters. The old behavior will swap the .data field of the Tensor subclass which is incorrect behavior when dealing with tensor subclasses that may have multiple child tensors.

This uses the `swap_tensors` method to swap all of the tensors not just the .data field.

Test plan:

```
pytest test/distributed/_tensor/test_api.py -k 'test_distribute_module_casting'
python test/distributed/fsdp/test_wrap.py -k test_auto_wrap_smoke_test_cuda_init_mode1_cpu_offload0_use_device_id_True
```

Pull Request resolved: https://github.com/pytorch/pytorch/pull/122755
Approved by: https://github.com/wanchaol, https://github.com/mikaylagawarecki

(cherry picked from commit e6ee8322d767ab241ce1651e7c178f539e8e3199)

Co-authored-by: Tristan Rice <rice@fn.lc>
2024-04-02 16:16:16 -07:00
5a53185e65 Remove cuda dependencies when building AOTriton (#122982) (#123179)
Downloading CUDA sometimes failed and breaks the build process, but
AOTriton does not need these packages. This commit comments out the
related downloading scripts.
2024-04-02 19:08:22 -04:00
bc9e23abb5 Fix performance regression and memory storage handling of Flash Attention on ROCM (#122857) (#122967)
This PR fixes the two major issues that was discovered after the initial merge of PR #121561
1. The Flash Attention support added by has severe performance regressions on regular shapes (power of two head dimensions and sequence lengths) compared with PR #115981. Its performance is worse than the math backend and only has numerical stability advantages. This PR fixes this problem.
2. There is a flaw of memory storage handling in PR #121561 which does not copy the gradients back to the designated output tensor. This PR removes the deprecated `TensorStorageSanitizer` class which is unnecessary due to the more flexible backward kernel shipped by PR #121561

Pull Request resolved: https://github.com/pytorch/pytorch/pull/122857
Approved by: https://github.com/jeffdaily, https://github.com/drisspg
2024-04-02 18:53:19 -04:00
8194fae625 Pin protobuf to 3.20.2 on macOS (#123197)
The newer protobuf 5.26.0 releasing on March 13rd is causing failures with `test_hparams_*` from `test_tensorboard` in which the stringify metadata is wrong when escaping double quote. For example, 3bc2bb6781.  This looks like an upstream issue from Tensorboard where it doesn't work with this brand new protobuf version https://github.com/tensorflow/tensorboard/blob/master/tensorboard/pip_package/requirements.txt#L29

The package has been pinned on Docker https://github.com/pytorch/pytorch/blob/main/.ci/docker/requirements-ci.txt#L155, so it should be pinned on macOS too.  We want to eventually just have one requirements.txt file.

Fixes https://github.com/pytorch/pytorch/issues/122008
Fixes https://github.com/pytorch/pytorch/issues/121927
Fixes https://github.com/pytorch/pytorch/issues/121946
Pull Request resolved: https://github.com/pytorch/pytorch/pull/121918
Approved by: https://github.com/kit1980
2024-04-02 15:08:09 -04:00
12acd4c9b3 [Cherrypick][DeviceMesh] Cache and reuse sliced result (#122975) (#123073)
Fixes #118849

Add a map for parent_to_child_mappings in _mesh_resources so we can cache and reuse submesh slicing result so that we can avoid recreating submesh and the underlying sub pg repeatedly, which could lead to funky behaviors.

We will follow up with reusing pg from the parent_mesh during submesh creation.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/122975
Approved by: https://github.com/wanchaol
2024-04-02 15:05:07 -04:00
857797d148 [CherryPick] Inductor cpp wrapper: fix dtype of ShapeAsConstantBuffer (#122297) (#123064)
For `at::scalar_tensor` the default dtype will be `float` ([link to scalar_tensor](0d8e960f74/aten/src/ATen/native/TensorFactories.cpp (L856)), [link to default dtype](0d8e960f74/c10/core/TensorOptions.h (L551))) if we don't set the `dtype` value. However, the input scalar value is not necessarily a `float` value. With `torch::tensor(x)`, the dtype of the tensor will be decided according to the dtype of the scalar.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/122297
Approved by: https://github.com/jgong5, https://github.com/desertfire
2024-04-02 15:03:25 -04:00
233dfe4d6a Proper view support for jagged layout NestedTensor (#122854)
* Proper view support for jagged layout NestedTensor (#113279)

This PR:
* Introduces an ATen op for creating true jagged views from a dense values buffer
    * `_nested_view_from_jagged(values, offsets, lengths, ragged_idx, dummy)`
    * This ops is implemented on the Python side using torch.library so we can return a subclass instance
    * `jagged_from_list()` now uses this instead of the old autograd.Function `NestedViewFromBuffer`
    * The latter op is used for non-contiguous JTs returned via `torch.nested.narrow()`
    * `dummy` is an awful hack to ensure that `NestedTensor.__torch_dispatch__()` is invoked for our view
* Introduces an ATen op for accessing the `values` component of an NT via a view
    * `_nested_get_values(nt)`
* **Removes** the autograd.Functions `ViewNestedFromBuffer` and `ViewBufferFromNested` in favor of `nested_from_values_offsets()` / `nested_from_values_offsets_lengths()` and `nt.values()`, respectively.
* Changes test code to prefer `as_nested_tensor()` over `jagged_from_list()` directly
    * Similarly, avoid `buffer_from_jagged()`, preferring `values()`
* Depends on general subclass view fake-ification on the PT2 side (handled solely in previous PRs in the stack)

With these changes, the semantics of jagged layout NTs are such that they are considered a true view of the underlying `values` buffer. This means views of jagged NTs are views of the underlying buffer as well, simplifying some handling.

Differential Revision: [D54269922](https://our.internmc.facebook.com/intern/diff/D54269922)
Co-authored-by: voznesenskym <voznesenskym@gmail.com>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/113279
Approved by: https://github.com/ezyang

(cherry picked from commit cd6bfc7965fc5ae20720bae0994e332e56f819c0)

* Update executorch.txt

* Update executorch.txt

* Fix linter error

---------

Co-authored-by: Joel Schlosser <jbschlosser@meta.com>
Co-authored-by: Guang Yang <42389959+guangy10@users.noreply.github.com>
2024-04-02 11:46:53 -07:00
e22b534b10 Upgrade submodule oneDNN to v3.3.6 for release/2.3 (#122164) (#122930)
As the title. Including issue fixes for aarch64:
- https://github.com/oneapi-src/oneDNN/pull/1831
- https://github.com/oneapi-src/oneDNN/pull/1834

---

## Validation results
(on Intel CPU + Linux)
**Static quantization with Inductor on CV models**

Quant method | Geomean throughput ratio (v3.3.6/baseline)
-- | --
ptq | 0.982937
ptq (cpp wrapper) | 0.978384
qat | 0.978828

**Torchbench cpu userbenchmark with Inductor**

Items | Perf Geomean Ratio (v3.3.6/baseline)
-- | --
eager_throughtput_bf16_infer | 1.00x
eager_throughtput_fp32_infer | 1.00x
jit_llga_throughtput_amp_bf16 | 1.01x
jit_llga_throughtput_fp32 | 1.00x
eager_throughtput_fx_int8 | 1.00x
eager_throughtput_bf16_train | 1.46x
eager_throughtput_fp32_train | 1.41x

**Dynamo benchmarks tests**
Precision | Shape | Wrapper | Thread | Eager old/new GEOMEAN | Inductor old/new GEOMEAN
-- | -- | -- | -- | -- | --
Float32 | Static | Default | Multiple | 1.003836812 | 1.003425
Float32 | Static | Default | Single | 1.000181451 | 0.999611
Float32 | Dynamic | Default | Multiple | 1.003980183 | 1.006563
Float32 | Dynamic | Default | Single | 1.000076939 | 0.999969
AMP | Static | Default | Multiple | 0.996824772 | 0.998715
AMP | Static | Default | Single | 0.996402574 | 1.001483
AMP | Dynamic | Default | Multiple | 0.994919866 | 1.000467
AMP | Dynamic | Default | Single | 0.9962054 | 1.000767

(on Aarch64)
https://github.com/pytorch/pytorch/pull/122164#issuecomment-2007912919

---

Pull Request resolved: https://github.com/pytorch/pytorch/pull/122164
Approved by: https://github.com/snadampal, https://github.com/malfet, https://github.com/atalman
2024-04-02 12:57:11 -04:00
8602990e3f [CherryPick] Back out "[DeviceMesh] Add support for nD slicing (#119752)" (#121763) (#122495)
Summary:
Original commit changeset: e52b8809c8d8

Original Phabricator Diff: D54778906

We have to backout this diff.
D54778906 seems to be causing test failures for APF blocking trunk health and hence release. Just starting to look at the issue. T182209248

Test Plan: Sandcastle

Reviewed By: satgera

Differential Revision: D54825114

Pull Request resolved: https://github.com/pytorch/pytorch/pull/121763
Approved by: https://github.com/osalpekar

(cherry picked from commit e99fa0042cd3dcd2eded24585d59c53f2da9d9f5)
2024-03-28 14:25:08 -07:00
685cc955df [ROCm] Update triton rocm branch to release/2.3.x (#122493)
* Update triton rocm branch to release/2.3.x

* Remove ROCM_TRITION_VERSION and update to 2.3.0

* Remove unnecessary ROCm conditionalisation

* Skip failing UT
2024-03-28 14:18:37 -07:00
b1c2430fbd remove torchao dependency (#122635)
* remove torchao dependency (#122524)

Test Plan:
CI

```
buck2 run mode/dev-nosan mode/inplace executorch/examples/models/llama2:export_llama -- -c ~/llama/ultra_new_checkpoint.pt -p ~/llama/params.json -kv -E 8,8 -d fp32 --pt2e_quantize "xnnpack_dynamic" -2
```

```
buck run //executorch/backends/xnnpack/test:test_xnnpack_ops -- executorch.backends.xnnpack.test.ops.linear.TestLinear.test_qd8_fp32_per_token_weight_per_channel_group_int4
```

Differential Revision: D55263008

Pull Request resolved: https://github.com/pytorch/pytorch/pull/122524
Approved by: https://github.com/jerryzh168

(cherry picked from commit c677221798d8ce87c97aac1bd9ae34af0767c383)

* Update executorch.txt

* Update _decomposed.py

* Update executorch.txt

* Update executorch.txt

* Update executorch.txt

* Update executorch.txt

* Update executorch.txt

---------

Co-authored-by: Guang Yang <guangyang@meta.com>
Co-authored-by: Guang Yang <42389959+guangy10@users.noreply.github.com>
2024-03-28 12:25:12 -07:00
3002eb2556 [export] hack skip index_put_ in dce (#122683) (#122721)
Summary: Ideally we should do whats in the todo. Just doing this for now to unblock llama capture

Test Plan: capturing llama and using pt2e to quantize it

Differential Revision: D55354487

Pull Request resolved: https://github.com/pytorch/pytorch/pull/122683
Approved by: https://github.com/kimishpatel

(cherry picked from commit 41d24df08f72e059c4eebdde4315e63a9918406f)

Co-authored-by: Jacob Szwejbka <jakeszwe@meta.com>
2024-03-27 21:29:53 -07:00
e1a846d6b8 Fix auto_functionalize (#121990) (#122654)
Differential Revision: D54964130

When we re-export, auto_functionalize HOP will be in the graph. Therefore, we need to implement proper functionalization rule for it. Since the content inside auto_functionalize is guaranteed be functional, it is ok to just fall through it.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/121990
Approved by: https://github.com/ydwu4, https://github.com/zou3519

(cherry picked from commit 0d845f7b0781f091452a5fd31de14e1c2117f3d4)

Co-authored-by: Tugsbayasgalan (Tugsuu) Manlaibaatar <tmanlaibaatar@meta.com>
2024-03-27 21:28:56 -07:00
4a9a8c606d [export] add pass to remove auto functionalized hop (#122246) (#122655)
Summary: Adds a pass that blindly removes the functionalize hop without consideration on if its safe. Useful for ExecuTorch today and other usecases that have additional logic that can reason about when this pass is safe to use

Test Plan: added unit test

Differential Revision: D55103867

Pull Request resolved: https://github.com/pytorch/pytorch/pull/122246
Approved by: https://github.com/angelayi

(cherry picked from commit c84f81b395fff969bbd2f784efad8ab1a8aa52de)

Co-authored-by: Jacob Szwejbka <jakeszwe@meta.com>
2024-03-27 21:05:15 -07:00
d3201f48b1 Revert "Revert "CI: Specify libc and libstdcxx versions in conda environments"" (#122523)
This reverts commit 74832f12fae2e1bc51bf1f9971dcd12c90a971f5.
2024-03-22 17:41:42 -04:00
74832f12fa Revert "CI: Specify libc and libstdcxx versions in conda environments" (#122497)
This reverts commit b4f90aae1b375bfe06d3c4a099240e06f93c81c4.
2024-03-22 11:27:50 -04:00
02cdb400d7 Use temporary name for triton package, fix lint (#122438)
* Use temporary name for triton package

* Fix lint
2024-03-21 17:30:38 -04:00
37257774c6 Triton wheel build using 2.3.x branch (#122403)
* Triton build 2.3.x

* Revert "[Release Only] Build triton using pinned version rather branch (#121765)"

This reverts commit d69c4219127e2cf5d9637b0daacc0a24e65f8133.

* Triton wheel change

* release
2024-03-21 12:52:21 -04:00
c4e5434423 necessary change to make torch2.3 work with triton2.2 (#122139) 2024-03-21 08:24:53 -04:00
b4f90aae1b CI: Specify libc and libstdcxx versions in conda environments (#121929)
Without this we get mismatches between the GLIBC and GLIBCXX ABI used
by conda packages vs pytorch.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/121556
Approved by: https://github.com/isuruf, https://github.com/malfet

(cherry picked from commit 7a53dedb07ed72b85d1e083ce38c43c7810fc5f1)

Co-authored-by: Peter Bell <peterbell10@live.co.uk>
2024-03-14 17:56:46 -04:00
94d6463255 [RELEASE ONLY CHANGES] Increase timeout for linux binary jobs, fix workflow lint (#121851)
* [release only] Increase timeout job for linux binary builds by 30min

* fix lint
2024-03-13 19:50:57 -04:00
6a89a753b1 [RELEASE ONLY CHANGES] Apply release only changes Release 2.3 (#121813)
* [Release only changes] Release only changes #2

* common+lint
2024-03-13 11:03:48 -04:00
d69c421912 [Release Only] Build triton using pinned version rather branch (#121765) 2024-03-12 19:05:23 -04:00
6725db07ae [RELEASE ONLY CHANGES] Apply release only changes Release 2.3 (#121726)
* Apply release only changes

* temp changes

* tweak

* fix

* Revert "tweak"

This reverts commit 38edcac21448829ac114c73423c84614628e2598.
2024-03-12 18:14:35 -04:00
2607 changed files with 57126 additions and 150872 deletions

View File

@ -1,4 +1,3 @@
# We do not use this library in our Bazel build. It contains an
# infinitely recursing symlink that makes Bazel very unhappy.
third_party/ittapi/
third_party/opentelemetry-cpp

View File

@ -229,7 +229,6 @@ case "$image" in
BASEKIT_VERSION=2024.0.0-49522
NINJA_VERSION=1.9.0
CONDA_CMAKE=yes
TRITON=yes
;;
pytorch-linux-jammy-py3.8-gcc11-inductor-benchmarks)
ANACONDA_PYTHON_VERSION=3.8

View File

@ -1 +1 @@
d4b3e5cc607e97afdba79dc90f8ef968142f347c
e2a8f9548aecb62a68e264607174a7d207ed2929

View File

@ -1 +1 @@
bbe6246e37d8aa791c67daaf9d9d61b26c9ccfdc
d08e16b738ab550c3af51305df624d5c823dc445

View File

@ -1 +0,0 @@
b8c64f64c18d8cac598b3adb355c21e7439c21de

View File

@ -1 +1 @@
45fff310c891f5a92d55445adf8cc9d29df5841e
958fccea74da58e7e0595ab88ae6cd3f6795a173

View File

@ -33,12 +33,12 @@ pip_install coloredlogs packaging
pip_install onnxruntime==1.17.0
pip_install onnx==1.15.0
# pip_install "onnxscript@git+https://github.com/microsoft/onnxscript@3e869ef8ccf19b5ebd21c10d3e9c267c9a9fa729" --no-deps
pip_install onnxscript==0.1.0.dev20240315 --no-deps
pip_install onnxscript==0.1.0.dev20240301 --no-deps
# Cache the transformers model to be used later by ONNX tests. We need to run the transformers
# package to download the model. By default, the model is cached at ~/.cache/huggingface/hub/
IMPORT_SCRIPT_FILENAME="/tmp/onnx_import_script.py"
as_jenkins echo 'import transformers; transformers.AutoModel.from_pretrained("sshleifer/tiny-gpt2"); transformers.AutoTokenizer.from_pretrained("sshleifer/tiny-gpt2"); transformers.AutoModelForSpeechSeq2Seq.from_pretrained("openai/whisper-large-v3");' > "${IMPORT_SCRIPT_FILENAME}"
as_jenkins echo 'import transformers; transformers.AutoModel.from_pretrained("sshleifer/tiny-gpt2"); transformers.AutoTokenizer.from_pretrained("sshleifer/tiny-gpt2");' > "${IMPORT_SCRIPT_FILENAME}"
# Need a PyTorch version for transformers to work
pip_install --pre torch --index-url https://download.pytorch.org/whl/nightly/cpu

View File

@ -11,8 +11,7 @@ mkdir -p $pb_dir
ln -s /usr/lib64 "$pb_dir/lib64"
curl -LO "https://github.com/protocolbuffers/protobuf/releases/download/v3.17.3/protobuf-all-3.17.3.tar.gz" --retry 3
tar -xvz --no-same-owner -C "$pb_dir" --strip-components 1 -f protobuf-all-3.17.3.tar.gz
tar -xvz -C "$pb_dir" --strip-components 1 -f protobuf-all-3.17.3.tar.gz
NPROC=$[$(nproc) - 2]
pushd "$pb_dir" && ./configure && make -j${NPROC} && make -j${NPROC} check && sudo make -j${NRPOC} install && sudo ldconfig
popd

View File

@ -13,11 +13,8 @@ conda_reinstall() {
}
if [ -n "${ROCM_VERSION}" ]; then
TRITON_REPO="https://github.com/openai/triton"
TRITON_REPO="https://github.com/ROCmSoftwarePlatform/triton"
TRITON_TEXT_FILE="triton-rocm"
elif [ -n "${BASEKIT_VERSION}" ]; then
TRITON_REPO="https://github.com/intel/intel-xpu-backend-for-triton"
TRITON_TEXT_FILE="triton-xpu"
else
TRITON_REPO="https://github.com/openai/triton"
TRITON_TEXT_FILE="triton"

View File

@ -3,7 +3,7 @@ set -xe
# Intel® software for general purpose GPU capabilities.
# Refer to https://dgpu-docs.intel.com/releases/LTS_803.29_20240131.html
# Refer to https://dgpu-docs.intel.com/releases/stable_647_21_20230714.html
# Intel® oneAPI Base Toolkit (version 2024.0.0) has been updated to include functional and security updates.
# Refer to https://www.intel.com/content/www/us/en/developer/tools/oneapi/base-toolkit-download.html
@ -21,7 +21,7 @@ function install_ubuntu() {
| gpg --dearmor | tee /usr/share/keyrings/oneapi-archive-keyring.gpg > /dev/null
# Add the signed entry to APT sources and configure the APT client to use the Intel repository
echo "deb [arch=amd64 signed-by=/usr/share/keyrings/intel-graphics.gpg] https://repositories.intel.com/gpu/ubuntu jammy/lts/2350 unified" \
echo "deb [arch=amd64 signed-by=/usr/share/keyrings/intel-graphics.gpg] https://repositories.intel.com/gpu/ubuntu jammy/production/2328 unified" \
| tee /etc/apt/sources.list.d/intel-gpu-jammy.list
echo "deb [signed-by=/usr/share/keyrings/oneapi-archive-keyring.gpg] https://apt.repos.intel.com/oneapi all main" \
| tee /etc/apt/sources.list.d/oneAPI.list

View File

@ -85,10 +85,10 @@ librosa>=0.6.2 ; python_version < "3.11"
#Pinned versions:
#test that import:
mypy==1.9.0
mypy==1.8.0
# Pin MyPy version because new errors are likely to appear with each release
#Description: linter
#Pinned versions: 1.9.0
#Pinned versions: 1.8.0
#test that import: test_typing.py, test_type_hints.py
networkx==2.8.8
@ -134,9 +134,9 @@ opt-einsum==3.3
#Pinned versions: 3.3
#test that import: test_linalg.py
optree==0.11.0
optree==0.9.1
#Description: A library for tree manipulation
#Pinned versions: 0.11.0
#Pinned versions: 0.9.1
#test that import: test_vmap.py, test_aotdispatch.py, test_dynamic_shapes.py,
#test_pytree.py, test_ops.py, test_control_flow.py, test_modules.py,
#common_utils.py, test_eager_transforms.py, test_python_dispatch.py,
@ -147,9 +147,9 @@ optree==0.11.0
#test_pointwise_ops.py, test_dtensor_ops.py, test_torchinductor.py, test_fx.py,
#test_fake_tensor.py, test_mps.py
pillow==10.3.0
pillow==10.2.0
#Description: Python Imaging Library fork
#Pinned versions: 10.3.0
#Pinned versions: 10.2.0
#test that import:
protobuf==3.20.2
@ -228,11 +228,12 @@ scikit-image==0.20.0 ; python_version >= "3.10"
#Pinned versions: 0.20.3
#test that import:
scipy==1.10.1 ; python_version <= "3.11"
scipy==1.12.0 ; python_version == "3.12"
scipy==1.6.3 ; python_version < "3.10"
scipy==1.8.1 ; python_version == "3.10"
scipy==1.10.1 ; python_version == "3.11"
# Pin SciPy because of failing distribution tests (see #60347)
#Description: scientific python
#Pinned versions: 1.10.1
#Pinned versions: 1.6.3
#test that import: test_unary_ufuncs.py, test_torch.py,test_tensor_creation_ops.py
#test_spectral_ops.py, test_sparse_csr.py, test_reductions.py,test_nn.py
#test_linalg.py, test_binary_ufuncs.py

View File

@ -1 +1 @@
3.0.0
2.3.1

View File

@ -61,20 +61,15 @@ COPY ci_commit_pins/timm.txt timm.txt
RUN if [ -n "${INDUCTOR_BENCHMARKS}" ]; then bash ./install_inductor_benchmark_deps.sh; fi
RUN rm install_inductor_benchmark_deps.sh common_utils.sh timm.txt huggingface.txt
# Install XPU Dependencies
ARG BASEKIT_VERSION
COPY ./common/install_xpu.sh install_xpu.sh
RUN bash ./install_xpu.sh && rm install_xpu.sh
ARG TRITON
# Install triton, this needs to be done before sccache because the latter will
# try to reach out to S3, which docker build runners don't have access
COPY ./common/install_triton.sh install_triton.sh
COPY ./common/common_utils.sh common_utils.sh
COPY ci_commit_pins/triton-xpu.txt triton-xpu.txt
COPY triton_version.txt triton_version.txt
# TODO: will add triton xpu commit
COPY ci_commit_pins/triton.txt triton.txt
RUN if [ -n "${TRITON}" ]; then bash ./install_triton.sh; fi
RUN rm install_triton.sh common_utils.sh triton-xpu.txt triton_version.txt
RUN rm install_triton.sh common_utils.sh triton.txt
# (optional) Install database packages like LMDB and LevelDB
ARG DB
@ -90,6 +85,11 @@ RUN if [ -n "${VISION}" ]; then bash ./install_vision.sh; fi
RUN rm install_vision.sh cache_vision_models.sh common_utils.sh
ENV INSTALLED_VISION ${VISION}
# Install XPU Dependencies
ARG BASEKIT_VERSION
COPY ./common/install_xpu.sh install_xpu.sh
RUN bash ./install_xpu.sh && rm install_xpu.sh
# (optional) Install non-default CMake version
ARG CMAKE_VERSION
COPY ./common/install_cmake.sh install_cmake.sh

View File

@ -1,9 +1,5 @@
#!/bin/bash
set -ex
source "$(dirname "${BASH_SOURCE[0]}")/../pytorch/common_utils.sh"
LOCAL_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
ROOT_DIR=$(cd "$LOCAL_DIR"/../.. && pwd)
TEST_DIR="$ROOT_DIR/test"

View File

@ -3,20 +3,6 @@
# shellcheck source=./common.sh
source "$(dirname "${BASH_SOURCE[0]}")/common.sh"
# Workaround for dind-rootless userid mapping (https://github.com/pytorch/ci-infra/issues/96)
WORKSPACE_ORIGINAL_OWNER_ID=$(stat -c '%u' "/var/lib/jenkins/workspace")
cleanup_workspace() {
echo "sudo may print the following warning message that can be ignored. The chown command will still run."
echo " sudo: setrlimit(RLIMIT_STACK): Operation not permitted"
echo "For more details refer to https://github.com/sudo-project/sudo/issues/42"
sudo chown -R "$WORKSPACE_ORIGINAL_OWNER_ID" /var/lib/jenkins/workspace
}
# Disable shellcheck SC2064 as we want to parse the original owner immediately.
# shellcheck disable=SC2064
trap_add cleanup_workspace EXIT
sudo chown -R jenkins /var/lib/jenkins/workspace
git config --global --add safe.directory /var/lib/jenkins/workspace
if [[ "$BUILD_ENVIRONMENT" == *onnx* ]]; then
# TODO: This can be removed later once vision is also part of the Docker image
pip install -q --user --no-use-pep517 "git+https://github.com/pytorch/vision.git@$(cat .github/ci_commit_pins/vision.txt)"

View File

@ -223,24 +223,6 @@ if [[ "${BUILD_ENVIRONMENT}" != *android* && "${BUILD_ENVIRONMENT}" != *cuda* ]]
export BUILD_STATIC_RUNTIME_BENCHMARK=ON
fi
# Do not change workspace permissions for ROCm CI jobs
# as it can leave workspace with bad permissions for cancelled jobs
if [[ "$BUILD_ENVIRONMENT" != *rocm* ]]; then
# Workaround for dind-rootless userid mapping (https://github.com/pytorch/ci-infra/issues/96)
WORKSPACE_ORIGINAL_OWNER_ID=$(stat -c '%u' "/var/lib/jenkins/workspace")
cleanup_workspace() {
echo "sudo may print the following warning message that can be ignored. The chown command will still run."
echo " sudo: setrlimit(RLIMIT_STACK): Operation not permitted"
echo "For more details refer to https://github.com/sudo-project/sudo/issues/42"
sudo chown -R "$WORKSPACE_ORIGINAL_OWNER_ID" /var/lib/jenkins/workspace
}
# Disable shellcheck SC2064 as we want to parse the original owner immediately.
# shellcheck disable=SC2064
trap_add cleanup_workspace EXIT
sudo chown -R jenkins /var/lib/jenkins/workspace
git config --global --add safe.directory /var/lib/jenkins/workspace
fi
if [[ "$BUILD_ENVIRONMENT" == *-bazel-* ]]; then
set -e
@ -266,6 +248,7 @@ else
( ! get_exit_code python setup.py clean bad_argument )
if [[ "$BUILD_ENVIRONMENT" != *libtorch* ]]; then
# rocm builds fail when WERROR=1
# XLA test build fails when WERROR=1
# set only when building other architectures

View File

@ -159,7 +159,7 @@ function install_torchvision() {
}
function install_tlparse() {
pip_install --user "tlparse==0.3.7"
pip_install --user "tlparse==0.3.5"
PATH="$(python -m site --user-base)/bin:$PATH"
}
@ -178,7 +178,7 @@ function install_torchrec_and_fbgemm() {
function clone_pytorch_xla() {
if [[ ! -d ./xla ]]; then
git clone --recursive --quiet https://github.com/pytorch/xla.git
git clone --recursive -b r2.3 https://github.com/pytorch/xla.git
pushd xla
# pin the xla hash so that we don't get broken by changes to xla
git checkout "$(cat ../.github/ci_commit_pins/xla.txt)"

View File

@ -45,7 +45,6 @@ time python test/run_test.py --verbose -i distributed/test_device_mesh
time python test/run_test.py --verbose -i distributed/tensor/parallel/test_ddp_2d_parallel
time python test/run_test.py --verbose -i distributed/tensor/parallel/test_fsdp_2d_parallel
time python test/run_test.py --verbose -i distributed/tensor/parallel/test_tp_examples
time python test/run_test.py --verbose -i distributed/tensor/parallel/test_tp_random_state.py
# Other tests
time python test/run_test.py --verbose -i test_cuda_primary_ctx

View File

@ -59,16 +59,16 @@ print("sample mean: ", sample_mean)
print("sample sigma: ", sample_sigma)
if math.isnan(sample_mean):
raise Exception("""Error: sample mean is NaN""") # noqa: TRY002
raise Exception("""Error: sample mean is NaN""")
elif math.isnan(sample_sigma):
raise Exception("""Error: sample sigma is NaN""") # noqa: TRY002
raise Exception("""Error: sample sigma is NaN""")
z_value = (sample_mean - mean) / sigma
print("z-value: ", z_value)
if z_value >= 3:
raise Exception( # noqa: TRY002
raise Exception(
f"""\n
z-value >= 3, there is high chance of perf regression.\n
To reproduce this regression, run

View File

@ -105,7 +105,6 @@ if [ "$is_main_doc" = true ]; then
echo undocumented objects found:
cat build/coverage/python.txt
echo "Make sure you've updated relevant .rsts in docs/source!"
echo "You can reproduce locally by running 'cd docs && make coverage && cat build/coverage/python.txt'"
exit 1
fi
else

View File

@ -6,27 +6,6 @@
set -ex
# shellcheck source=./common.sh
source "$(dirname "${BASH_SOURCE[0]}")/common.sh"
# Do not change workspace permissions for ROCm CI jobs
# as it can leave workspace with bad permissions for cancelled jobs
if [[ "$BUILD_ENVIRONMENT" != *rocm* ]]; then
# Workaround for dind-rootless userid mapping (https://github.com/pytorch/ci-infra/issues/96)
WORKSPACE_ORIGINAL_OWNER_ID=$(stat -c '%u' "/var/lib/jenkins/workspace")
cleanup_workspace() {
echo "sudo may print the following warning message that can be ignored. The chown command will still run."
echo " sudo: setrlimit(RLIMIT_STACK): Operation not permitted"
echo "For more details refer to https://github.com/sudo-project/sudo/issues/42"
sudo chown -R "$WORKSPACE_ORIGINAL_OWNER_ID" /var/lib/jenkins/workspace
}
# Disable shellcheck SC2064 as we want to parse the original owner immediately.
# shellcheck disable=SC2064
trap_add cleanup_workspace EXIT
sudo chown -R jenkins /var/lib/jenkins/workspace
git config --global --add safe.directory /var/lib/jenkins/workspace
fi
echo "Environment variables:"
env
@ -111,6 +90,9 @@ if [[ -n $TESTS_TO_INCLUDE ]]; then
INCLUDE_CLAUSE="--include $TESTS_TO_INCLUDE"
fi
# shellcheck source=./common.sh
source "$(dirname "${BASH_SOURCE[0]}")/common.sh"
echo "Environment variables"
env
@ -229,6 +211,8 @@ if [[ "$BUILD_ENVIRONMENT" == *asan* ]]; then
export LD_PRELOAD=/usr/lib/llvm-15/lib/clang/15.0.7/lib/linux/libclang_rt.asan-x86_64.so
# Disable valgrind for asan
export VALGRIND=OFF
# Increase stack size, because ASAN red zones use more stack
ulimit -s 81920
(cd test && python -c "import torch; print(torch.__version__, torch.version.git_version)")
echo "The next four invocations are expected to crash; if they don't that means ASAN/UBSAN is misconfigured"
@ -315,12 +299,9 @@ test_inductor_distributed() {
pytest test/distributed/_composable/fsdp/test_fully_shard_training.py -k test_train_parity_multi_group
pytest test/distributed/_composable/fsdp/test_fully_shard_training.py -k test_train_parity_with_activation_checkpointing
pytest test/distributed/_composable/fsdp/test_fully_shard_training.py -k test_train_parity_2d_mlp
pytest test/distributed/_composable/fsdp/test_fully_shard_training.py -k test_train_parity_hsdp
pytest test/distributed/_composable/fsdp/test_fully_shard_training.py -k test_train_parity_2d_transformer_checkpoint_resume
pytest test/distributed/_composable/fsdp/test_fully_shard_frozen.py
pytest test/distributed/_composable/fsdp/test_fully_shard_mixed_precision.py -k test_compute_dtype
pytest test/distributed/_composable/fsdp/test_fully_shard_mixed_precision.py -k test_reduce_dtype
pytest test/distributed/fsdp/test_fsdp_tp_integration.py -k test_fsdp_tp_integration
# this runs on both single-gpu and multi-gpu instance. It should be smart about skipping tests that aren't supported
# with if required # gpus aren't available
@ -332,13 +313,13 @@ test_inductor() {
python tools/dynamo/verify_dynamo.py
python test/run_test.py --inductor --include test_modules test_ops test_ops_gradients test_torch --verbose
# Do not add --inductor for the following inductor unit tests, otherwise we will fail because of nested dynamo state
python test/run_test.py --include inductor/test_torchinductor inductor/test_torchinductor_opinfo inductor/test_aot_inductor --verbose
python test/run_test.py --include inductor/test_torchinductor inductor/test_torchinductor_opinfo --verbose
# docker build uses bdist_wheel which does not work with test_aot_inductor
# TODO: need a faster way to build
if [[ "$BUILD_ENVIRONMENT" != *rocm* ]]; then
BUILD_AOT_INDUCTOR_TEST=1 python setup.py develop
CPP_TESTS_DIR="${BUILD_BIN_DIR}" LD_LIBRARY_PATH="${TORCH_LIB_DIR}" python test/run_test.py --cpp --verbose -i cpp/test_aoti_abi_check cpp/test_aoti_inference
CPP_TESTS_DIR="${BUILD_BIN_DIR}" LD_LIBRARY_PATH="${TORCH_LIB_DIR}" python test/run_test.py --cpp --verbose -i cpp/test_aot_inductor
fi
}
@ -451,17 +432,6 @@ test_perf_for_dashboard() {
"${target_flag[@]}" --"$mode" --"$dtype" --backend "$backend" "$@" \
--output "$TEST_REPORTS_DIR/${backend}_max_autotune_${suite}_${dtype}_${mode}_cuda_${target}.csv"
fi
if [[ "$DASHBOARD_TAG" == *cudagraphs_low_precision-true* ]] && [[ "$mode" == "inference" ]]; then
# TODO: This has a new dtype called quant and the benchmarks script needs to be updated to support this.
# The tentative command is as follows. It doesn't work now, but it's ok because we only need mock data
# to fill the dashboard.
python "benchmarks/dynamo/$suite.py" \
"${target_flag[@]}" --"$mode" --quant --backend "$backend" "$@" \
--output "$TEST_REPORTS_DIR/${backend}_cudagraphs_low_precision_${suite}_quant_${mode}_cuda_${target}.csv" || true
# Copy cudagraph results as mock data, easiest choice?
cp "$TEST_REPORTS_DIR/${backend}_with_cudagraphs_${suite}_${dtype}_${mode}_cuda_${target}.csv" \
"$TEST_REPORTS_DIR/${backend}_cudagraphs_low_precision_${suite}_quant_${mode}_cuda_${target}.csv"
fi
done
done
}
@ -623,12 +593,6 @@ test_inductor_torchbench_cpu_smoketest_perf(){
done
}
test_torchbench_gcp_smoketest(){
pushd "${TORCHBENCHPATH}"
python test.py -v
popd
}
test_python_gloo_with_tls() {
source "$(dirname "${BASH_SOURCE[0]}")/run_glootls_test.sh"
assert_git_not_dirty
@ -1208,9 +1172,6 @@ elif [[ "${TEST_CONFIG}" == *torchbench* ]]; then
llama_v2_7b_16h resnet50 timm_efficientnet mobilenet_v3_large timm_resnest \
shufflenet_v2_x1_0 hf_GPT2
PYTHONPATH=$(pwd)/torchbench test_inductor_torchbench_cpu_smoketest_perf
elif [[ "${TEST_CONFIG}" == *torchbench_gcp_smoketest* ]]; then
checkout_install_torchbench
TORCHBENCHPATH=$(pwd)/torchbench test_torchbench_gcp_smoketest
else
checkout_install_torchbench
# Do this after checkout_install_torchbench to ensure we clobber any

View File

@ -78,7 +78,7 @@ TRITON_VERSION=$(cat $PYTORCH_ROOT/.ci/docker/triton_version.txt)
if [[ "$PACKAGE_TYPE" =~ .*wheel.* && -n "${PYTORCH_EXTRA_INSTALL_REQUIREMENTS:-}" ]]; then
# Only linux Python < 3.12 are supported wheels for triton
TRITON_CONSTRAINT="platform_system == 'Linux' and platform_machine == 'x86_64' and python_version < '3.12'"
TRITON_REQUIREMENT="triton==${TRITON_VERSION}; ${TRITON_CONSTRAINT}"
TRITON_REQUIREMENT="pytorch-triton==${TRITON_VERSION}; ${TRITON_CONSTRAINT}"
if [[ -n "$PYTORCH_BUILD_VERSION" && "$PYTORCH_BUILD_VERSION" =~ .*dev.* ]]; then
TRITON_SHORTHASH=$(cut -c1-10 $PYTORCH_ROOT/.ci/docker/ci_commit_pins/triton.txt)
TRITON_REQUIREMENT="pytorch-triton==${TRITON_VERSION}+${TRITON_SHORTHASH}; ${TRITON_CONSTRAINT}"

View File

@ -36,7 +36,6 @@ hicpp-exception-baseclass,
hicpp-avoid-goto,
misc-*,
-misc-const-correctness,
-misc-include-cleaner,
-misc-use-anonymous-namespace,
-misc-unused-parameters,
-misc-no-recursion,

1
.gitattributes vendored
View File

@ -4,4 +4,3 @@
.github/generated-* linguist-generated=true
.github/scripts/gql_mocks.json linguist-generated=true
third_party/LICENSES_BUNDLED.txt linguist-generated=true
tools/build/bazel/requirements.txt linguist-generated=true

View File

@ -8,18 +8,7 @@ body:
value: >
#### Before submitting a bug, please make sure the issue hasn't been already addressed by searching through [the
existing and past issues](https://github.com/pytorch/pytorch/issues)
It's likely that your bug will be resolved by checking our FAQ or troubleshooting guide [documentation](https://pytorch.org/docs/main/dynamo/index.html)
Note: if you're submitting an issue that you generated from a fuzzer. Please do the following:
- Ensure rtol/atol are at default tolerances
- Dont compare indices of max/min etc, because that avoids the above requirement
- If comparing eager and torch.compile at fp16/bf16, you should use fp32 as baseline
If the above requirements are met, add the label "topic: fuzzer" to your issue.
It's likely that your bug will be resolved by checking our FAQ or troubleshooting guide [documentation](https://pytorch.org/docs/master/dynamo/index.html)
- type: textarea
attributes:
label: 🐛 Describe the bug
@ -44,7 +33,7 @@ body:
label: Minified repro
description: |
Please run the minifier on your example and paste the minified code below
Learn more here https://pytorch.org/docs/main/torch.compiler_troubleshooting.html
Learn more here https://pytorch.org/docs/master/compile/troubleshooting.html
placeholder: |
env TORCHDYNAMO_REPRO_AFTER="aot" python your_model.py
or

View File

@ -9,10 +9,6 @@ inputs:
use-gha:
description: If set to any value, use GHA to download the artifact. Otherwise use s3.
required: false
s3-bucket:
description: S3 bucket to download builds
required: false
default: "gha-artifacts"
runs:
using: composite
@ -22,10 +18,9 @@ runs:
uses: seemethere/download-artifact-s3@v4
with:
name: ${{ inputs.name }}
s3-bucket: ${{ inputs.s3-bucket }}
- name: Download PyTorch Build Artifacts from GHA
if: ${{ inputs.use-gha }}
if: inputs.use-gha
uses: actions/download-artifact@v3
with:
name: ${{ inputs.name }}
@ -34,10 +29,6 @@ runs:
shell: bash
run: unzip -o artifacts.zip
- name: Remove artifacts.zip
shell: bash
run: rm artifacts.zip
- name: Output disk space left
shell: bash
run: df -H

View File

@ -13,13 +13,6 @@ inputs:
required: true
type: string
description: JSON description of what test configs to run.
selected-test-configs:
required: false
type: string
description: |
A comma-separated list of test configurations from the test matrix to keep,
The empty list means we are going to keep every configurations by defaults
default: ""
job-name:
type: string
required: false
@ -47,9 +40,6 @@ outputs:
ci-no-td:
description: True if ci-no-td label was on PR or [ci-no-td] in PR body.
value: ${{ steps.filter.outputs.ci-no-td }}
ci-td-distributed:
description: True if ci-td-distributed label was on PR or [ci-td-distributed] in PR body.
value: ${{ steps.filter.outputs.ci-td-distributed }}
runs:
using: composite
@ -133,7 +123,6 @@ runs:
--workflow "${GITHUB_WORKFLOW}" \
--job-name "${JOB_NAME}" \
--test-matrix "${{ inputs.test-matrix }}" \
--selected-test-configs "${{ inputs.selected-test-configs }}" \
--pr-number "${PR_NUMBER}" \
--tag "${TAG}" \
--event-name "${EVENT_NAME}" \

View File

@ -1,207 +0,0 @@
name: linux-build
inputs:
build-environment:
required: true
description: Top-level label for what's being built/tested.
docker-image-name:
required: true
description: Name of the base docker image to build with.
build-generates-artifacts:
required: false
default: "true"
description: If set, upload generated build artifacts.
build-with-debug:
required: false
default: "false"
description: If set, build in debug mode.
sync-tag:
required: false
default: ""
description: |
If this is set, our linter will use this to make sure that every other
job with the same `sync-tag` is identical.
cuda-arch-list:
required: false
default: "5.2"
description: Runner label to select worker type
runner:
required: false
default: "linux.2xlarge"
description: |
List of CUDA architectures CI build should target.
test-matrix:
required: false
type: string
description: |
An option JSON description of what test configs to run later on. This
is moved here from the Linux test workflow so that we can apply filter
logic using test-config labels earlier and skip unnecessary builds
s3-bucket:
description: S3 bucket to download artifact
required: false
default: "gha-artifacts"
aws-role-to-assume:
description: role to assume for downloading artifacts
required: false
default: ""
GITHUB_TOKEN:
description: GitHub token
required: true
HUGGING_FACE_HUB_TOKEN:
description: Hugging Face Hub token
required: false
default: ""
outputs:
docker-image:
value: ${{ steps.calculate-docker-image.outputs.docker-image }}
description: The docker image containing the built PyTorch.
test-matrix:
value: ${{ steps.filter.outputs.test-matrix }}
description: An optional JSON description of what test configs to run later on.
runs:
using: composite
steps:
- name: Setup Linux
uses: ./.github/actions/setup-linux
- name: configure aws credentials
uses: aws-actions/configure-aws-credentials@v3
if: ${{ inputs.aws-role-to-assume != '' }}
with:
role-to-assume: ${{ inputs.aws-role-to-assume }}
role-session-name: gha-linux-build
role-duration-seconds: 10800
aws-region: us-east-1
- name: Calculate docker image
id: calculate-docker-image
uses: pytorch/test-infra/.github/actions/calculate-docker-image@main
with:
docker-image-name: ${{ inputs.docker-image-name }}
- name: Use following to pull public copy of the image
id: print-ghcr-mirror
env:
ECR_DOCKER_IMAGE: ${{ steps.calculate-docker-image.outputs.docker-image }}
shell: bash
run: |
tag=${ECR_DOCKER_IMAGE##*/}
echo "docker pull ghcr.io/pytorch/ci-image:${tag/:/-}"
- name: Pull docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
with:
docker-image: ${{ steps.calculate-docker-image.outputs.docker-image }}
- name: Parse ref
id: parse-ref
shell: bash
run: .github/scripts/parse_ref.py
- name: Get workflow job id
id: get-job-id
uses: ./.github/actions/get-workflow-job-id
if: always()
with:
github-token: ${{ inputs.GITHUB_TOKEN }}
# Apply the filter logic to the build step too if the test-config label is already there
- name: Select all requested test configurations (if the test matrix is available)
id: filter
uses: ./.github/actions/filter-test-configs
with:
github-token: ${{ inputs.GITHUB_TOKEN }}
test-matrix: ${{ inputs.test-matrix }}
job-name: ${{ steps.get-job-id.outputs.job-name }}
- name: Download pytest cache
uses: ./.github/actions/pytest-cache-download
continue-on-error: true
with:
cache_dir: .pytest_cache
job_identifier: ${{ github.workflow }}_${{ inputs.build-environment }}
s3_bucket: ${{ inputs.s3-bucket }}
- name: Build
if: steps.filter.outputs.is-test-matrix-empty == 'False' || inputs.test-matrix == ''
id: build
env:
BUILD_ENVIRONMENT: ${{ inputs.build-environment }}
BRANCH: ${{ steps.parse-ref.outputs.branch }}
# TODO duplicated
AWS_DEFAULT_REGION: us-east-1
PR_NUMBER: ${{ github.event.pull_request.number }}
SHA1: ${{ github.event.pull_request.head.sha || github.sha }}
SCCACHE_BUCKET: ossci-compiler-cache-circleci-v2
SCCACHE_S3_KEY_PREFIX: ${{ github.workflow }}
XLA_CLANG_CACHE_S3_BUCKET_NAME: ossci-compiler-clang-cache-circleci-xla
PR_LABELS: ${{ toJson(github.event.pull_request.labels.*.name) }}
TORCH_CUDA_ARCH_LIST: ${{ inputs.cuda-arch-list }}
DOCKER_IMAGE: ${{ steps.calculate-docker-image.outputs.docker-image }}
XLA_CUDA: ${{ contains(inputs.build-environment, 'xla') && '0' || '' }}
DEBUG: ${{ inputs.build-with-debug == 'true' && '1' || '0' }}
OUR_GITHUB_JOB_ID: ${{ steps.get-job-id.outputs.job-id }}
HUGGING_FACE_HUB_TOKEN: ${{ inputs.HUGGING_FACE_HUB_TOKEN }}
shell: bash
run: |
# detached container should get cleaned up by teardown_ec2_linux
container_name=$(docker run \
-e BUILD_ENVIRONMENT \
-e MAX_JOBS="$(nproc --ignore=2)" \
-e AWS_DEFAULT_REGION \
-e PR_NUMBER \
-e SHA1 \
-e BRANCH \
-e SCCACHE_BUCKET \
-e SCCACHE_S3_KEY_PREFIX \
-e XLA_CUDA \
-e XLA_CLANG_CACHE_S3_BUCKET_NAME \
-e SKIP_SCCACHE_INITIALIZATION=1 \
-e TORCH_CUDA_ARCH_LIST \
-e PR_LABELS \
-e OUR_GITHUB_JOB_ID \
-e HUGGING_FACE_HUB_TOKEN \
--env-file="/tmp/github_env_${GITHUB_RUN_ID}" \
--security-opt seccomp=unconfined \
--cap-add=SYS_PTRACE \
--tty \
--detach \
--user jenkins \
-v "${GITHUB_WORKSPACE}:/var/lib/jenkins/workspace" \
-w /var/lib/jenkins/workspace \
"${DOCKER_IMAGE}"
)
docker exec -t "${container_name}" sh -c '.ci/pytorch/build.sh'
- name: Archive artifacts into zip
if: inputs.build-generates-artifacts == 'true' && steps.build.outcome != 'skipped'
shell: bash
run: |
zip -1 -r artifacts.zip dist/ build/custom_test_artifacts build/lib build/bin .additional_ci_files
- name: Store PyTorch Build Artifacts on S3
uses: seemethere/upload-artifact-s3@v5
if: inputs.build-generates-artifacts == 'true' && steps.build.outcome != 'skipped'
with:
name: ${{ inputs.build-environment }}
retention-days: 14
if-no-files-found: error
path: artifacts.zip
s3-bucket: ${{ inputs.s3-bucket }}
- name: Upload sccache stats
if: steps.build.outcome != 'skipped'
uses: seemethere/upload-artifact-s3@v5
with:
s3-prefix: |
${{ github.repository }}/${{ github.run_id }}/${{ github.run_attempt }}/artifact
retention-days: 365
if-no-files-found: warn
path: sccache-stats-*.json
s3-bucket: ${{ inputs.s3-bucket }}
- name: Teardown Linux
uses: pytorch/test-infra/.github/actions/teardown-linux@main
if: always()

View File

@ -1,384 +0,0 @@
name: linux-test
inputs:
build-environment:
required: true
type: string
description: Top-level label for what's being built/tested.
test-matrix:
required: true
type: string
description: JSON description of what test configs to run.
docker-image:
required: true
type: string
description: Docker image to run in.
sync-tag:
required: false
type: string
default: ""
description: |
If this is set, our linter will use this to make sure that every other
job with the same `sync-tag` is identical.
use-gha:
required: false
type: string
default: ""
description: If set to any value, upload to GHA. Otherwise upload to S3.
dashboard-tag:
required: false
type: string
default: ""
s3-bucket:
description: S3 bucket to download artifact
required: false
type: string
default: "gha-artifacts"
aws-role-to-assume:
description: role to assume for downloading artifacts
required: false
type: string
default: ""
HUGGING_FACE_HUB_TOKEN:
description: |
HF Auth token to avoid rate limits when downloading models or datasets from hub
required: false
default: ""
GITHUB_TOKEN:
description: GitHub token
required: true
#env:
# GIT_DEFAULT_BRANCH: ${{ inputs.default_branch }}
runs:
using: composite
steps:
- name: Setup Linux
uses: ./.github/actions/setup-linux
- name: configure aws credentials
if : ${{ inputs.aws-role-to-assume != '' }}
uses: aws-actions/configure-aws-credentials@v3
with:
role-to-assume: ${{ inputs.aws-role-to-assume }}
role-session-name: gha-linux-test
aws-region: us-east-1
- name: Calculate docker image
id: calculate-docker-image
uses: pytorch/test-infra/.github/actions/calculate-docker-image@main
with:
docker-image-name: ${{ inputs.docker-image }}
- name: Use following to pull public copy of the image
id: print-ghcr-mirror
env:
ECR_DOCKER_IMAGE: ${{ steps.calculate-docker-image.outputs.docker-image }}
shell: bash
run: |
tag=${ECR_DOCKER_IMAGE##*/}
echo "docker pull ghcr.io/pytorch/ci-image:${tag/:/-}"
- name: Pull docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
with:
docker-image: ${{ steps.calculate-docker-image.outputs.docker-image }}
- name: Check if in a ARC runner
shell: bash
id: check_arc_runner
run: echo "IN_ARC_RUNNER=$([ -f /.inarc ] && echo true || echo false)" >> "$GITHUB_OUTPUT"
- name: Install nvidia driver, nvidia-docker runtime, set GPU_FLAG
id: install-nvidia-driver
uses: pytorch/test-infra/.github/actions/setup-nvidia@main
if: ${{ contains(inputs.build-environment, 'cuda') && !contains(matrix.config, 'nogpu') && steps.check_arc_runner.outputs.IN_ARC_RUNNER == 'false' }}
- name: Lock NVIDIA A100 40GB Frequency
shell: bash
run: |
sudo nvidia-smi -pm 1
sudo nvidia-smi -ac 1215,1410
nvidia-smi
if: contains(matrix.runner, 'a100')
- name: Start monitoring script
id: monitor-script
shell: bash
continue-on-error: true
run: |
python3 -m pip install psutil==5.9.1 nvidia-ml-py==11.525.84
python3 -m tools.stats.monitor > usage_log.txt 2>&1 &
echo "monitor-script-pid=${!}" >> "${GITHUB_OUTPUT}"
- name: Download build artifacts
uses: ./.github/actions/download-build-artifacts
with:
name: ${{ inputs.build-environment }}
s3-bucket: ${{ inputs.s3-bucket }}
- name: Download TD artifacts
continue-on-error: true
uses: ./.github/actions/download-td-artifacts
- name: Parse ref
id: parse-ref
shell: bash
run: .github/scripts/parse_ref.py
- name: Get workflow job id
id: get-job-id
uses: ./.github/actions/get-workflow-job-id
if: always()
with:
github-token: ${{ inputs.GITHUB_TOKEN }}
- name: Check for keep-going label and re-enabled test issues
# This uses the filter-test-configs action because it conviniently
# checks for labels and re-enabled test issues. It does not actually do
# any filtering. All filtering is done in the build step.
id: keep-going
uses: ./.github/actions/filter-test-configs
with:
github-token: ${{ inputs.GITHUB_TOKEN }}
test-matrix: ${{ inputs.test-matrix }}
job-name: ${{ steps.get-job-id.outputs.job-name }}
- name: Test
id: test
env:
BUILD_ENVIRONMENT: ${{ inputs.build-environment }}
PR_NUMBER: ${{ github.event.pull_request.number }}
GITHUB_REPOSITORY: ${{ github.repository }}
GITHUB_WORKFLOW: ${{ github.workflow }}
GITHUB_JOB: ${{ github.job }}
GITHUB_RUN_ID: ${{ github.run_id }}
GITHUB_RUN_NUMBER: ${{ github.run_number }}
GITHUB_RUN_ATTEMPT: ${{ github.run_attempt }}
JOB_ID: ${{ steps.get-job-id.outputs.job-id }}
JOB_NAME: ${{ steps.get-job-id.outputs.job-name }}
BRANCH: ${{ steps.parse-ref.outputs.branch }}
SHA1: ${{ github.event.pull_request.head.sha || github.sha }}
BASE_SHA: ${{ github.event.pull_request.base.sha || github.sha }}
TEST_CONFIG: ${{ matrix.config }}
SHARD_NUMBER: ${{ matrix.shard }}
NUM_TEST_SHARDS: ${{ matrix.num_shards }}
REENABLED_ISSUES: ${{ steps.keep-going.outputs.reenabled-issues }}
CONTINUE_THROUGH_ERROR: ${{ steps.keep-going.outputs.keep-going }}
VERBOSE_TEST_LOGS: ${{ steps.keep-going.outputs.ci-verbose-test-logs }}
NO_TEST_TIMEOUT: ${{ steps.keep-going.outputs.ci-no-test-timeout }}
NO_TD: ${{ steps.keep-going.outputs.ci-no-td }}
TD_DISTRIBUTED: ${{ steps.keep-going.outputs.ci-td-distributed }}
SCCACHE_BUCKET: ossci-compiler-cache-circleci-v2
SCCACHE_S3_KEY_PREFIX: ${{ github.workflow }}
SHM_SIZE: ${{ contains(inputs.build-environment, 'cuda') && '2g' || '1g' }}
DOCKER_IMAGE: ${{ inputs.docker-image }}
XLA_CUDA: ${{ contains(inputs.build-environment, 'xla') && '0' || '' }}
XLA_CLANG_CACHE_S3_BUCKET_NAME: ossci-compiler-clang-cache-circleci-xla
PYTORCH_TEST_CUDA_MEM_LEAK_CHECK: ${{ matrix.mem_leak_check && '1' || '0' }}
PYTORCH_TEST_RERUN_DISABLED_TESTS: ${{ matrix.rerun_disabled_tests && '1' || '0' }}
DASHBOARD_TAG: ${{ inputs.dashboard-tag }}
HUGGING_FACE_HUB_TOKEN: ${{ inputs.HUGGING_FACE_HUB_TOKEN }}
shell: bash
run: |
set -x
if [[ $TEST_CONFIG == 'multigpu' ]]; then
TEST_COMMAND=.ci/pytorch/multigpu-test.sh
elif [[ $BUILD_ENVIRONMENT == *onnx* ]]; then
TEST_COMMAND=.ci/onnx/test.sh
else
TEST_COMMAND=.ci/pytorch/test.sh
fi
# detached container should get cleaned up by teardown_ec2_linux
# TODO: Stop building test binaries as part of the build phase
# Used for GPU_FLAG since that doesn't play nice
# shellcheck disable=SC2086,SC2090
container_name=$(docker run \
${GPU_FLAG:-} \
-e BUILD_ENVIRONMENT \
-e PR_NUMBER \
-e GITHUB_ACTIONS \
-e GITHUB_REPOSITORY \
-e GITHUB_WORKFLOW \
-e GITHUB_JOB \
-e GITHUB_RUN_ID \
-e GITHUB_RUN_NUMBER \
-e GITHUB_RUN_ATTEMPT \
-e JOB_ID \
-e JOB_NAME \
-e BASE_SHA \
-e BRANCH \
-e SHA1 \
-e AWS_DEFAULT_REGION \
-e IN_WHEEL_TEST \
-e SHARD_NUMBER \
-e TEST_CONFIG \
-e NUM_TEST_SHARDS \
-e REENABLED_ISSUES \
-e CONTINUE_THROUGH_ERROR \
-e VERBOSE_TEST_LOGS \
-e NO_TEST_TIMEOUT \
-e NO_TD \
-e TD_DISTRIBUTED \
-e PR_LABELS \
-e MAX_JOBS="$(nproc --ignore=2)" \
-e SCCACHE_BUCKET \
-e SCCACHE_S3_KEY_PREFIX \
-e XLA_CUDA \
-e XLA_CLANG_CACHE_S3_BUCKET_NAME \
-e PYTORCH_TEST_CUDA_MEM_LEAK_CHECK \
-e PYTORCH_TEST_RERUN_DISABLED_TESTS \
-e SKIP_SCCACHE_INITIALIZATION=1 \
-e HUGGING_FACE_HUB_TOKEN \
-e DASHBOARD_TAG \
--env-file="/tmp/github_env_${GITHUB_RUN_ID}" \
--security-opt seccomp=unconfined \
--cap-add=SYS_PTRACE \
--ipc=host \
--shm-size="${SHM_SIZE}" \
--tty \
--detach \
--name="${container_name}" \
--user jenkins \
-v "${GITHUB_WORKSPACE}:/var/lib/jenkins/workspace" \
-w /var/lib/jenkins/workspace \
"${DOCKER_IMAGE}"
)
# Propagate download.pytorch.org IP to container
grep download.pytorch.org /etc/hosts | docker exec -i "${container_name}" sudo bash -c "/bin/cat >> /etc/hosts"
echo "DOCKER_CONTAINER_ID=${container_name}" >> "${GITHUB_ENV}"
docker exec -t "${container_name}" sh -c "pip install $(echo dist/*.whl)[opt-einsum] && ${TEST_COMMAND}"
- name: Upload pytest cache if tests failed
uses: ./.github/actions/pytest-cache-upload
continue-on-error: true
if: failure() && steps.test.conclusion && steps.test.conclusion == 'failure'
with:
cache_dir: .pytest_cache
shard: ${{ matrix.shard }}
sha: ${{ github.event.pull_request.head.sha || github.sha }}
test_config: ${{ matrix.config }}
job_identifier: ${{ github.workflow }}_${{ inputs.build-environment }}
- name: Print remaining test logs
shell: bash
if: always() && steps.test.conclusion
run: |
cat test/**/*_toprint.log || true
- name: Stop monitoring script
if: always() && steps.monitor-script.outputs.monitor-script-pid
shell: bash
continue-on-error: true
env:
MONITOR_SCRIPT_PID: ${{ steps.monitor-script.outputs.monitor-script-pid }}
run: |
kill "$MONITOR_SCRIPT_PID"
- name: Upload test artifacts
uses: ./.github/actions/upload-test-artifacts
if: always() && steps.test.conclusion && steps.test.conclusion != 'skipped'
with:
file-suffix: ${{ github.job }}-${{ matrix.config }}-${{ matrix.shard }}-${{ matrix.num_shards }}-${{ matrix.runner }}_${{ steps.get-job-id.outputs.job-id }}
use-gha: ${{ inputs.use-gha }}
s3-bucket: ${{ inputs.s3-bucket }}
- name: Collect backtraces from coredumps (if any)
if: always()
shell: bash
run: |
# shellcheck disable=SC2156
find . -iname "core.[1-9]*" -exec docker exec "${DOCKER_CONTAINER_ID}" sh -c "gdb python {} -ex 'bt' -ex 'q'" \;
- name: Store Core dumps on S3
uses: seemethere/upload-artifact-s3@v5
if: failure()
with:
name: coredumps-${{ matrix.config }}-${{ matrix.shard }}-${{ matrix.num_shards }}-${{ matrix.runner }}
retention-days: 14
if-no-files-found: ignore
path: ./**/core.[1-9]*
- name: Teardown Linux
uses: pytorch/test-infra/.github/actions/teardown-linux@main
if: always()
# NB: We are currently having an intermittent GPU-related issue on G5 runners with
# A10G GPU. Once this happens, trying to reset the GPU as done in setup-nvidia does
# not seem to help. Here are some symptoms:
# * Calling nvidia-smi timeouts after 60 second
# * Fail to run nvidia-smi with an unable to determine the device handle for GPU
# unknown error
# * Test fails with a missing CUDA GPU error when initializing CUDA in PyTorch
# * Run docker --gpus all fails with error response from daemon
#
# As both the root cause and recovery path are unclear, let's take the runner out of
# service so that it doesn't get any more jobs
- name: Check NVIDIA driver installation step
if: failure() && steps.install-nvidia-driver.outcome && steps.install-nvidia-driver.outcome != 'skipped'
shell: bash
env:
RUNNER_WORKSPACE: ${{ runner.workspace }}
run: |
set +e
set -x
nvidia-smi
# NB: Surprisingly, nvidia-smi command returns successfully with return code 0 even in
# the case where the driver has already crashed as it still can get the driver version
# and some basic information like the bus ID. However, the rest of the information
# would be missing (ERR!), for example:
#
# +-----------------------------------------------------------------------------+
# | NVIDIA-SMI 525.89.02 Driver Version: 525.89.02 CUDA Version: 12.0 |
# |-------------------------------+----------------------+----------------------+
# | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
# | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
# | | | MIG M. |
# |===============================+======================+======================|
# | 0 ERR! Off | 00000000:00:1E.0 Off | ERR! |
# |ERR! ERR! ERR! ERR! / ERR! | 4184MiB / 23028MiB | ERR! Default |
# | | | ERR! |
# +-------------------------------+----------------------+----------------------+
#
# +-----------------------------------------------------------------------------+
# | Processes: |
# | GPU GI CI PID Type Process name GPU Memory |
# | ID ID Usage |
# |=============================================================================|
# +-----------------------------------------------------------------------------+
#
# This should be reported as a failure instead as it will guarantee to fail when
# Docker tries to run with --gpus all
#
# So, the correct check here is to query one of the missing piece of info like
# GPU name, so that the command can fail accordingly
nvidia-smi --query-gpu=gpu_name --format=csv,noheader --id=0
NVIDIA_SMI_STATUS=$?
# These are acceptable return code from nvidia-smi as copied from setup-nvidia GitHub action
if [ "$NVIDIA_SMI_STATUS" -ne 0 ] && [ "$NVIDIA_SMI_STATUS" -ne 14 ]; then
echo "NVIDIA driver installation has failed, shutting down the runner..."
.github/scripts/stop_runner_service.sh
fi
# For runner with multiple GPUs, we also want to confirm that the number of GPUs are the
# power of 2, i.e. 1, 2, 4, or 8. This is to avoid flaky test issue when one GPU fails
# https://github.com/pytorch/test-infra/issues/4000
GPU_COUNT=$(nvidia-smi --list-gpus | wc -l)
NVIDIA_SMI_STATUS=$?
# These are acceptable return code from nvidia-smi as copied from setup-nvidia GitHub action
if [ "$NVIDIA_SMI_STATUS" -ne 0 ] && [ "$NVIDIA_SMI_STATUS" -ne 14 ]; then
echo "NVIDIA driver installation has failed, shutting down the runner..."
.github/scripts/stop_runner_service.sh
fi
# Check the GPU count to be a power of 2
if [ "$GPU_COUNT" -le 8 ] && [ "$GPU_COUNT" -ne 1 ] && [ "$GPU_COUNT" -ne 2 ] && [ "$GPU_COUNT" -ne 4 ] && [ "$GPU_COUNT" -ne 8 ]; then
echo "NVIDIA driver detects $GPU_COUNT GPUs. The runner has a broken GPU, shutting it down..."
.github/scripts/stop_runner_service.sh
fi

View File

@ -9,10 +9,6 @@ inputs:
job_identifier:
description: Text that uniquely identifies a given job type within a workflow. All shards of a job should share the same job identifier.
required: true
s3_bucket:
description: S3 bucket to download PyTest cache
required: false
default: "gha-artifacts"
runs:
using: composite
@ -34,7 +30,6 @@ runs:
CACHE_DIR: ${{ inputs.cache_dir }}
JOB_IDENTIFIER: ${{ inputs.job_identifier }}
REPO: ${{ github.repository }}
BUCKET: ${{ inputs.s3_bucket }}
run: |
python3 .github/scripts/pytest_cache.py \
--download \
@ -43,4 +38,3 @@ runs:
--job_identifier $JOB_IDENTIFIER \
--temp_dir $RUNNER_TEMP \
--repo $REPO \
--bucket $BUCKET \

View File

@ -15,12 +15,10 @@ runs:
category=$1
# If it is GCP runner (runner name contains gcp), do not run this
runner_name_str=${{ runner.name }}
if [[ -f /.inarc ]]; then
echo "ARC Runner, no info on ec2 metadata"
elif [[ $runner_name_str == *"gcp"* ]]; then
echo "Runner is from Google Cloud Platform, No info on ec2 metadata"
else
if [[ $runner_name_str != *"gcp"* ]]; then
curl -fsSL "http://169.254.169.254/latest/meta-data/${category}"
else
echo "Runner is from Google Cloud Platform, No info on ec2 metadata"
fi
}
echo "ami-id: $(get_ec2_metadata ami-id)"
@ -28,14 +26,8 @@ runs:
echo "instance-type: $(get_ec2_metadata instance-type)"
echo "system info $(uname -a)"
- name: Check if in a ARC runner
shell: bash
id: check_arc_runner
run: echo "IN_ARC_RUNNER=$([ -f /.inarc ] && echo true || echo false)" >> $GITHUB_OUTPUT
- name: Start docker if docker deamon is not running
shell: bash
if: ${{ steps.check_arc_runner.outputs.IN_ARC_RUNNER == 'false' }}
run: |
if systemctl is-active --quiet docker; then
echo "Docker daemon is running...";
@ -66,7 +58,6 @@ runs:
env | grep '^CI' >> "/tmp/github_env_${GITHUB_RUN_ID}"
- name: Kill any existing containers, clean up images
if: ${{ steps.check_arc_runner.outputs.IN_ARC_RUNNER == 'false' }}
shell: bash
run: |
# ignore expansion of "docker ps -q" since it could be empty
@ -105,28 +96,3 @@ runs:
echo "${RESOLVED_IP} ${PT_DOMAIN}" | sudo tee -a /etc/hosts
cat /etc/hosts
- name: Check that the docker daemon is running
shell: bash
continue-on-error: true
if: ${{ steps.check_arc_runner.outputs.IN_ARC_RUNNER == 'true' }}
run: |
set +x
max_attempts=30
delay=10
attempt=1
for attempt in $(seq 1 $max_attempts); do
echo "Attempt $attempt of $max_attempts: Checking if Docker daemon is running..."
if docker info > /dev/null 2>&1; then
echo "Docker is running. Proceeding with the next steps"
exit 0
else
echo "Docker is not running yet."
echo "Retrying in $delay seconds..."
sleep $delay
fi
done
echo "Reached maximum attempts to connect to Docker. Exiting."
exit 1

View File

@ -11,10 +11,6 @@ inputs:
Suffix to add to the filename of the artifacts. This should include the
workflow job id, see [Job id in artifacts].
required: true
s3-bucket:
description: S3 bucket to download builds
required: false
default: "gha-artifacts"
runs:
using: composite
@ -46,7 +42,7 @@ runs:
env:
FILE_SUFFIX: ${{ inputs.file-suffix }}
run: |
# Remove any previous usage logs if they exist
# Remove any previous test reports if they exist
rm -f logs-*.zip
# this workflow is also run in bazel build test, but we dont generate usage reports for it
# so check to see if the file exists first
@ -57,18 +53,6 @@ runs:
zip -r "logs-${FILE_SUFFIX}.zip" test -i '*.log'
fi
- name: Zip debugging artifacts for upload
if: runner.os != 'Windows' && !inputs.use-gha
shell: bash
env:
FILE_SUFFIX: ${{ inputs.file-suffix }}
run: |
# Remove any previous debugging artifacts if they exist
rm -f debug-*.zip
if [ -d 'test/debug' ]; then
zip -r "debug-${FILE_SUFFIX}.zip" test/debug
fi
# Windows zip
- name: Zip JSONs for upload
if: runner.os == 'Windows' && !inputs.use-gha
@ -103,7 +87,6 @@ runs:
uses: seemethere/upload-artifact-s3@v5
if: ${{ !inputs.use-gha }}
with:
s3-bucket: ${{ inputs.s3-bucket }}
s3-prefix: |
${{ github.repository }}/${{ github.run_id }}/${{ github.run_attempt }}/artifact
retention-days: 14
@ -114,7 +97,6 @@ runs:
uses: seemethere/upload-artifact-s3@v5
if: ${{ !inputs.use-gha }}
with:
s3-bucket: ${{ inputs.s3-bucket }}
s3-prefix: |
${{ github.repository }}/${{ github.run_id }}/${{ github.run_attempt }}/artifact
retention-days: 14
@ -126,25 +108,12 @@ runs:
if: ${{ !inputs.use-gha }}
continue-on-error: true
with:
s3-bucket: ${{ inputs.s3-bucket }}
s3-prefix: |
${{ github.repository }}/${{ github.run_id }}/${{ github.run_attempt }}/artifact
retention-days: 14
if-no-files-found: ignore
path: logs-*.zip
- name: Store Debug Artifacts on S3
uses: seemethere/upload-artifact-s3@v5
if: ${{ !inputs.use-gha }}
continue-on-error: true
with:
s3-bucket: ${{ inputs.s3-bucket }}
s3-prefix: |
${{ github.repository }}/${{ github.run_id }}/${{ github.run_attempt }}/artifact
retention-days: 14
if-no-files-found: ignore
path: debug-*.zip
# GHA upload
- name: Store Test Downloaded JSONs on Github
uses: actions/upload-artifact@v3

View File

@ -1 +1 @@
ea437b31ce316ea3d66fe73768c0dcb94edb79ad
87aeb554d3e2f7855b7abe5120c282f59648ed7a

View File

@ -1 +1 @@
2c4665ffbb64f03f5d18016d3398af4ac4da5f03
2c127da8b5e2e8f44b50994c6cb931bcca267cfe

View File

@ -1 +1 @@
58a412cb271a3f98ae2e01fd1d24bdbb66645d4e
r2.3

View File

@ -1,13 +0,0 @@
# Use this to auto apply labels based on other labels. Applies to both PRs and
# issues. Currently only supports any and all
- any:
- "module: custom operators"
- "module: aotdispatch"
then:
- "module: pt2-dispatcher"
- any:
- "module: dynamo"
- "module: pt2-dispatcher"
- "module: inductor"
then:
- "oncall: pt2"

3
.github/labeler.yml vendored
View File

@ -35,9 +35,6 @@
- test/distributed/tensor/parallel/test_fsdp_2d_parallel.py
- torch/distributed/_tensor/**
- torch/distributed/fsdp/**
- torch/csrc/inductor/**
- test/cpp/aoti_abi_check/**
- test/cpp/aoti_inference/**
"module: cpu":
- aten/src/ATen/cpu/**

View File

@ -28,6 +28,7 @@
- caffe2/python/onnx/**
approved_by:
- BowenBao
- abock
- justinchuby
- shubhambhokare1
- thiagocrepaldi
@ -235,23 +236,6 @@
- Lint
- pull
- name: XPU ATen
patterns:
- aten/src/ATen/xpu/**
- c10/xpu/**
- torch/csrc/xpu/**
- torch/xpu/**
- test/xpu/**
- third_party/xpu.txt
approved_by:
- EikanWang
- jgong5
- gujinghui
mandatory_checks_name:
- EasyCLA
- Lint
- pull
- name: Distributions
patterns:
- torch/distributions/**
@ -373,14 +357,12 @@
- name: CPU inductor
patterns:
- torch/_inductor/mkldnn_lowerings.py
- torch/_inductor/fx_passes/mkldnn_fusion.py
- torch/_inductor/fx_passes/quantization.py
- torch/_inductor/codegen/cpp.py
- test/inductor/test_mkldnn_pattern_matcher.py
- test/inductor/test_cpu_repo.py
- test/inductor/test_cpu_cpp_wrapper.py
- aten/src/ATen/cpu/**
- aten/src/ATen/native/quantized/cpu/**
- test/quantization/core/test_quantized_op.py
- torch/ao/quantization/quantizer/x86_inductor_quantizer.py

View File

@ -1,6 +1,5 @@
tracking_issue: 24422
ciflow_tracking_issue: 64124
TD_rollout_issue: 123120
ciflow_push_tags:
- ciflow/binaries
- ciflow/binaries_conda
@ -16,12 +15,9 @@ ciflow_push_tags:
- ciflow/trunk
- ciflow/unstable
- ciflow/xpu
- ciflow/torchbench
retryable_workflows:
- lint
- pull
- trunk
- linux-binary
- windows-binary
labeler_config: labeler.yml
label_to_label_config: label_to_label.yml

View File

@ -5,7 +5,7 @@
# functorch/docs/requirements.txt
# .ci/docker/requirements-ci.txt
boto3==1.19.12
jinja2==3.1.3
jinja2==3.0.1
lintrunner==0.10.7
ninja==1.10.0.post1
nvidia-ml-py==11.525.84

View File

@ -1,4 +1,4 @@
# iOS simulator requirements
coremltools==5.0b5
protobuf==3.20.2
optree==0.11.0
optree==0.9.1

View File

@ -26,7 +26,7 @@ pytest-cpp==2.3.0
rockset==1.0.3
z3-solver==4.12.2.0
tensorboard==2.13.0
optree==0.11.0
optree==0.9.1
# NB: test_hparams_* from test_tensorboard is failing with protobuf 5.26.0 in
# which the stringify metadata is wrong when escaping double quote
protobuf==3.20.2

View File

@ -1,99 +0,0 @@
set -ex
# Set ROCM_HOME isn't available, use ROCM_PATH if set or /opt/rocm
ROCM_HOME="${ROCM_HOME:-${ROCM_PATH:-/opt/rocm}}"
# Find rocm_version.h header file for ROCm version extract
rocm_version_h="${ROCM_HOME}/include/rocm-core/rocm_version.h"
if [ ! -f "$rocm_version_h" ]; then
rocm_version_h="${ROCM_HOME}/include/rocm_version.h"
fi
# Error out if rocm_version.h not found
if [ ! -f "$rocm_version_h" ]; then
echo "Error: rocm_version.h not found in expected locations." >&2
exit 1
fi
# Extract major, minor and patch ROCm version numbers
MAJOR_VERSION=$(grep 'ROCM_VERSION_MAJOR' "$rocm_version_h" | awk '{print $3}')
MINOR_VERSION=$(grep 'ROCM_VERSION_MINOR' "$rocm_version_h" | awk '{print $3}')
PATCH_VERSION=$(grep 'ROCM_VERSION_PATCH' "$rocm_version_h" | awk '{print $3}')
ROCM_INT=$(($MAJOR_VERSION * 10000 + $MINOR_VERSION * 100 + $PATCH_VERSION))
echo "ROCm version: $ROCM_INT"
# Check TRITON_ROCM_DIR is set
if [[ -z "${TRITON_ROCM_DIR}" ]]; then
export TRITON_ROCM_DIR=third_party/amd/backend
fi
# Remove packaged libs and headers
rm -rf $TRITON_ROCM_DIR/include/*
LIBTINFO_PATH="/usr/lib64/libtinfo.so.5"
LIBNUMA_PATH="/usr/lib64/libnuma.so.1"
LIBELF_PATH="/usr/lib64/libelf.so.1"
OS_SO_PATHS=(
$LIBELF_PATH
$LIBNUMA_PATH
$LIBTINFO_PATH
)
for lib in "${OS_SO_PATHS[@]}"
do
cp $lib $TRITON_ROCM_DIR/lib/
done
# Required ROCm libraries
if [[ "${MAJOR_VERSION}" == "6" ]]; then
libamdhip="libamdhip64.so.6"
else
libamdhip="libamdhip64.so.5"
fi
# Required ROCm libraries - ROCm 6.0
ROCM_SO=(
"${libamdhip}"
"libhsa-runtime64.so.1"
"libamd_comgr.so.2"
"libdrm.so.2"
"libdrm_amdgpu.so.1"
)
if [[ $ROCM_INT -ge 60100 ]]; then
ROCM_SO+=("librocprofiler-register.so.0")
fi
for lib in "${ROCM_SO[@]}"
do
file_path=($(find $ROCM_HOME/lib/ -name "$lib")) # First search in lib
if [[ -z $file_path ]]; then
if [ -d "$ROCM_HOME/lib64/" ]; then
file_path=($(find $ROCM_HOME/lib64/ -name "$lib")) # Then search in lib64
fi
fi
if [[ -z $file_path ]]; then
file_path=($(find $ROCM_HOME/ -name "$lib")) # Then search in ROCM_HOME
fi
if [[ -z $file_path ]]; then
file_path=($(find /opt/ -name "$lib")) # Then search in /opt
fi
if [[ -z $file_path ]]; then
echo "Error: Library file $lib is not found." >&2
exit 1
fi
cp $file_path $TRITON_ROCM_DIR/lib
# When running locally, and not building a wheel, we need to satisfy shared objects requests that don't look for versions
LINKNAME=$(echo $lib | sed -e 's/\.so.*/.so/g')
ln -sf $lib $TRITON_ROCM_DIR/lib/$LINKNAME
done
# Copy Include Files
cp -r $ROCM_HOME/include/hip $TRITON_ROCM_DIR/include
# Copy linker
mkdir -p $TRITON_ROCM_DIR/llvm/bin
cp $ROCM_HOME/llvm/bin/ld.lld $TRITON_ROCM_DIR/llvm/bin/

View File

@ -1,99 +0,0 @@
#!/bin/bash
set -x
WHEELHOUSE_DIR=/artifacts
PATCHELF_BIN=patchelf
ROCM_LIB=backends/amd/lib
ROCM_LD=backends/amd/llvm/bin
PREFIX=triton
fname_without_so_number() {
LINKNAME=$(echo $1 | sed -e 's/\.so.*/.so/g')
echo "$LINKNAME"
}
replace_needed_sofiles() {
find $1 -name '*.so*' -o -name 'ld.lld' | while read sofile; do
origname=$2
patchedname=$3
if [[ "$origname" != "$patchedname" ]]; then
set +e
origname=$($PATCHELF_BIN --print-needed $sofile | grep "$origname.*")
ERRCODE=$?
set -e
if [ "$ERRCODE" -eq "0" ]; then
echo "patching $sofile entry $origname to $patchedname"
$PATCHELF_BIN --replace-needed $origname $patchedname $sofile
fi
fi
done
}
mkdir -p "/tmp_dir"
pushd /tmp_dir
for pkg in /$WHEELHOUSE_DIR/*triton*.whl; do
echo "Modifying $pkg"
rm -rf tmp
mkdir -p tmp
cd tmp
cp $pkg .
unzip -q $(basename $pkg)
rm -f $(basename $pkg)
$PATCHELF_BIN --set-rpath ${LD_SO_RPATH:-'$ORIGIN:$ORIGIN/../../lib'} $PREFIX/$ROCM_LD/ld.lld
$PATCHELF_BIN --print-rpath $PREFIX/$ROCM_LD/ld.lld
# Modify libtriton.so as it sits in _C directory apart from its dependencies
find $PREFIX/_C -type f -name "*.so*" | while read sofile; do
echo "Setting rpath of $sofile"
$PATCHELF_BIN --set-rpath ${C_SO_RPATH:-'$ORIGIN:$ORIGIN/'../$ROCM_LIB} ${FORCE_RPATH:-} $sofile
$PATCHELF_BIN --print-rpath $sofile
done
# All included dependencies are included in a single lib directory
deps=()
deps_soname=()
while read sofile; do
echo "Setting rpath of $sofile to ${LIB_SO_RPATH:-'$ORIGIN'}"
$PATCHELF_BIN --set-rpath ${LIB_SO_RPATH:-'$ORIGIN'} ${FORCE_RPATH:-} $sofile
$PATCHELF_BIN --print-rpath $sofile
deps+=("$sofile")
deps_soname+=("$(basename $sofile)")
done < <(find $PREFIX/$ROCM_LIB -type f -name "*.so*")
patched=()
for filepath in "${deps[@]}"; do
filename=$(basename $filepath)
destpath=$PREFIX/$ROCM_LIB/$filename
if [[ "$filepath" != "$destpath" ]]; then
cp $filepath $destpath
fi
patchedpath=$(fname_without_so_number $destpath)
patchedname=$(basename $patchedpath)
if [[ "$destpath" != "$patchedpath" ]]; then
mv $destpath $patchedpath
fi
patched+=("$patchedname")
echo "Copied $filepath to $patchedpath"
done
# Go through all required shared objects and see if any of our other objects are dependants. If so, replace so.ver wth so
for ((i=0;i<${#deps[@]};++i)); do
echo "replacing "${deps_soname[i]} ${patched[i]}
replace_needed_sofiles $PREFIX/$ROCM_LIB ${deps_soname[i]} ${patched[i]}
replace_needed_sofiles $PREFIX/_C ${deps_soname[i]} ${patched[i]}
replace_needed_sofiles $PREFIX/$ROCM_LD ${deps_soname[i]} ${patched[i]}
done
# Re-bundle whl with so adjustments
zip -rqy $(basename $pkg) *
if [[ -z "${MANYLINUX_VERSION}" ]]; then
newpkg=$pkg
else
newpkg=$(echo $pkg | sed -e "s/\linux_x86_64/${MANYLINUX_VERSION}/g")
fi
# Remove original whl
rm -f $pkg
# Move rebuilt whl to original location with new name.
mv $(basename $pkg) $newpkg
done

View File

@ -10,6 +10,9 @@ from typing import Optional
SCRIPT_DIR = Path(__file__).parent
REPO_DIR = SCRIPT_DIR.parent.parent
# TODO: Remove me once Triton version is again in sync for vanilla and ROCm
ROCM_TRITION_VERSION = "2.3.0"
def read_triton_pin(rocm_hash: bool = False) -> str:
triton_file = "triton.txt" if not rocm_hash else "triton-rocm.txt"
@ -29,6 +32,27 @@ def check_and_replace(inp: str, src: str, dst: str) -> str:
return inp.replace(src, dst)
def patch_setup_py(
path: Path,
*,
version: str,
name: str = "triton",
expected_version: Optional[str] = None,
) -> None:
with open(path) as f:
orig = f.read()
# Replace name
orig = check_and_replace(orig, 'name="triton",', f'name="{name}",')
# Replace version
if not expected_version:
expected_version = read_triton_version()
orig = check_and_replace(
orig, f'version="{expected_version}",', f'version="{version}",'
)
with open(path, "w") as f:
f.write(orig)
def patch_init_py(
path: Path, *, version: str, expected_version: Optional[str] = None
) -> None:
@ -68,17 +92,21 @@ def build_triton(
with TemporaryDirectory() as tmpdir:
triton_basedir = Path(tmpdir) / "triton"
triton_pythondir = triton_basedir / "python"
triton_repo = "https://github.com/openai/triton"
if build_rocm:
triton_repo = "https://github.com/ROCmSoftwarePlatform/triton"
triton_pkg_name = "pytorch-triton-rocm"
else:
triton_repo = "https://github.com/openai/triton"
triton_pkg_name = "pytorch-triton"
check_call(["git", "clone", triton_repo], cwd=tmpdir)
if release:
ver, rev, patch = version.split(".")
check_call(
["git", "checkout", f"release/{ver}.{rev}.x"], cwd=triton_basedir
)
if build_rocm:
check_call(["git", "checkout", "release/2.3.x"], cwd=triton_basedir)
else:
check_call(
["git", "checkout", f"release/{ver}.{rev}.x"], cwd=triton_basedir
)
else:
check_call(["git", "checkout", commit_hash], cwd=triton_basedir)
@ -91,7 +119,7 @@ def build_triton(
print("source:\n path: .\n", file=meta)
print(
"build:\n string: py{{py}}\n number: 1\n script: cd python; "
"python setup.py install --record=record.txt\n",
"python setup.py install --single-version-externally-managed --record=record.txt\n",
" script_env:\n - MAX_JOBS\n",
file=meta,
)
@ -137,15 +165,18 @@ def build_triton(
patch_init_py(
triton_pythondir / "triton" / "__init__.py",
version=f"{version}",
expected_version=None,
expected_version=ROCM_TRITION_VERSION if build_rocm else None,
)
if build_rocm:
check_call(
[f"{SCRIPT_DIR}/amd/package_triton_wheel.sh"],
cwd=triton_basedir,
shell=True,
# TODO: Remove me when ROCM triton is updated
patch_setup_py(
triton_pythondir / "setup.py",
name=triton_pkg_name,
version=f"{version}",
expected_version=ROCM_TRITION_VERSION,
)
check_call("scripts/amd/setup_rocm_libs.sh", cwd=triton_basedir, shell=True)
print("ROCm libraries setup for triton installation...")
check_call(
@ -156,11 +187,8 @@ def build_triton(
shutil.copy(whl_path, Path.cwd())
if build_rocm:
check_call(
[f"{SCRIPT_DIR}/amd/patch_triton_wheel.sh"],
cwd=triton_basedir,
shell=True,
)
check_call("scripts/amd/fix_so.sh", cwd=triton_basedir, shell=True)
return Path.cwd() / whl_path.name

View File

@ -29,7 +29,7 @@ def parse_args() -> Any:
"--onto-branch", type=str, required=True, help="the target release branch"
)
parser.add_argument(
"--github-actor", type=str, required=True, help="all the world's a stage"
"--github-actor", type=str, required=True, help="all the worlds a stage"
)
parser.add_argument(
"--classification",

View File

@ -23,10 +23,8 @@ def main() -> None:
job_link = f"[job]({run_url})" if run_url is not None else "job"
msg = (
f"The {args.action} {job_link} was canceled or timed out. This most often happen if two merge requests were issued"
+ " for the same PR, or if merge job was waiting for more than 6 hours for tests to finish."
+ " In later case, please do not hesitate to reissue the merge command\n"
+ f" For more information see [pytorch-bot wiki]({BOT_COMMANDS_WIKI})."
f"The {args.action} {job_link} was canceled. If you believe this is a mistake,"
+ f" then you can re trigger it through [pytorch-bot]({BOT_COMMANDS_WIKI})."
)
gh_post_pr_comment(org, project, args.pr_num, msg)

View File

@ -18,7 +18,7 @@ ESTIMATED_TOKENS = [0]
TOKEN = os.environ["GITHUB_TOKEN"]
if not TOKEN:
raise Exception("GITHUB_TOKEN is not set") # noqa: TRY002
raise Exception("GITHUB_TOKEN is not set")
REPO_ROOT = Path(__file__).parent.parent.parent

Binary file not shown.

View File

@ -1,7 +1,6 @@
#!/usr/bin/env python3
import json
import logging
import os
import re
import subprocess
@ -9,7 +8,6 @@ import sys
import warnings
from enum import Enum
from functools import lru_cache
from logging import info
from typing import Any, Callable, Dict, List, Optional, Set
from urllib.request import Request, urlopen
@ -19,7 +17,33 @@ REENABLE_TEST_REGEX = "(?i)(Close(d|s)?|Resolve(d|s)?|Fix(ed|es)?) (#|https://gi
PREFIX = "test-config/"
logging.basicConfig(level=logging.INFO)
# Same as shard names
VALID_TEST_CONFIG_LABELS = {
f"{PREFIX}{label}"
for label in {
"backwards_compat",
"crossref",
"default",
"deploy",
"distributed",
"docs_tests",
"dynamo",
"force_on_cpu",
"functorch",
"inductor",
"inductor_distributed",
"inductor_huggingface",
"inductor_timm",
"inductor_torchbench",
"jit_legacy",
"multigpu",
"nogpu_AVX512",
"nogpu_NO_AVX2",
"slow",
"tsan",
"xla",
}
}
def is_cuda_or_rocm_job(job_name: Optional[str]) -> bool:
@ -38,9 +62,9 @@ SUPPORTED_PERIODICAL_MODES: Dict[str, Callable[[Optional[str]], bool]] = {
}
# The link to the published list of disabled jobs
DISABLED_JOBS_URL = "https://ossci-metrics.s3.amazonaws.com/disabled-jobs.json"
DISABLED_JOBS_URL = "https://ossci-metrics.s3.amazonaws.com/disabled-jobs.json?versionId=qO7aEr.Og33PtLXfNq0j0yj.bbLC7SzR"
# and unstable jobs
UNSTABLE_JOBS_URL = "https://ossci-metrics.s3.amazonaws.com/unstable-jobs.json"
UNSTABLE_JOBS_URL = "https://ossci-metrics.s3.amazonaws.com/unstable-jobs.json?versionId=7NhgpqKTtGXVUnL1C79KboTW_5qQx8y5"
# Some constants used to handle disabled and unstable jobs
JOB_NAME_SEP = "/"
@ -66,12 +90,6 @@ def parse_args() -> Any:
parser.add_argument(
"--test-matrix", type=str, required=True, help="the original test matrix"
)
parser.add_argument(
"--selected-test-configs",
type=str,
default="",
help="a comma-separated list of test configurations from the test matrix to keep",
)
parser.add_argument(
"--workflow", type=str, help="the name of the current workflow, i.e. pull"
)
@ -137,25 +155,19 @@ def get_labels(pr_number: int) -> Set[str]:
}
def filter_labels(labels: Set[str], label_regex: Any) -> Set[str]:
"""
Return the list of matching labels
"""
return {l for l in labels if re.match(label_regex, l)}
def filter(test_matrix: Dict[str, List[Any]], labels: Set[str]) -> Dict[str, List[Any]]:
"""
Select the list of test config to run from the test matrix. The logic works
as follows:
If the PR has one or more test-config labels as specified, only these test configs
will be selected. This also works with ciflow labels, for example, if a PR has both
ciflow/trunk and test-config/functorch, only trunk functorch builds and tests will
be run.
If the PR has one or more labels as specified in the VALID_TEST_CONFIG_LABELS set, only
these test configs will be selected. This also works with ciflow labels, for example,
if a PR has both ciflow/trunk and test-config/functorch, only trunk functorch builds
and tests will be run
If the PR has none of the test-config label, all tests are run as usual.
"""
filtered_test_matrix: Dict[str, List[Any]] = {"include": []}
for entry in test_matrix.get("include", []):
@ -165,46 +177,23 @@ def filter(test_matrix: Dict[str, List[Any]], labels: Set[str]) -> Dict[str, Lis
label = f"{PREFIX}{config_name.strip()}"
if label in labels:
msg = f"Select {config_name} because label {label} is present in the pull request by the time the test starts"
info(msg)
print(
f"Select {config_name} because label {label} is presented in the pull request by the time the test starts"
)
filtered_test_matrix["include"].append(entry)
test_config_labels = filter_labels(labels, re.compile(f"{PREFIX}.+"))
if not filtered_test_matrix["include"] and not test_config_labels:
info("Found no test-config label on the PR, so all test configs are included")
# Found no test-config label and the filtered test matrix is empty, return the same
valid_test_config_labels = labels.intersection(VALID_TEST_CONFIG_LABELS)
if not filtered_test_matrix["include"] and not valid_test_config_labels:
# Found no valid label and the filtered test matrix is empty, return the same
# test matrix as before so that all tests can be run normally
return test_matrix
else:
msg = f"Found {test_config_labels} on the PR so only these test configs are run"
info(msg)
# When the filter test matrix contain matches or if a valid test config label
# is found in the PR, return the filtered test matrix
return filtered_test_matrix
def filter_selected_test_configs(
test_matrix: Dict[str, List[Any]], selected_test_configs: Set[str]
) -> Dict[str, List[Any]]:
"""
Keep only the selected configs if the list if not empty. Otherwise, keep all test configs.
This filter is used when the workflow is dispatched manually.
"""
if not selected_test_configs:
return test_matrix
filtered_test_matrix: Dict[str, List[Any]] = {"include": []}
for entry in test_matrix.get("include", []):
config_name = entry.get("config", "")
if not config_name:
continue
if config_name in selected_test_configs:
filtered_test_matrix["include"].append(entry)
return filtered_test_matrix
def set_periodic_modes(
test_matrix: Dict[str, List[Any]], job_name: Optional[str]
) -> Dict[str, List[Any]]:
@ -385,33 +374,30 @@ def process_jobs(
# - If the target record has the job (config) name, only that test config
# will be skipped or marked as unstable
if not target_job_cfg:
msg = (
print(
f"Issue {target_url} created by {author} has {issue_type.value} "
+ f"all CI jobs for {workflow} / {job_name}"
)
info(msg)
return _filter_jobs(
test_matrix=test_matrix,
issue_type=issue_type,
)
if target_job_cfg == BUILD_JOB_NAME:
msg = (
print(
f"Issue {target_url} created by {author} has {issue_type.value} "
+ f"the build job for {workflow} / {job_name}"
)
info(msg)
return _filter_jobs(
test_matrix=test_matrix,
issue_type=issue_type,
)
if target_job_cfg in (TEST_JOB_NAME, BUILD_AND_TEST_JOB_NAME):
msg = (
print(
f"Issue {target_url} created by {author} has {issue_type.value} "
+ f"all the test jobs for {workflow} / {job_name}"
)
info(msg)
return _filter_jobs(
test_matrix=test_matrix,
issue_type=issue_type,
@ -477,7 +463,7 @@ def parse_reenabled_issues(s: Optional[str]) -> List[str]:
def get_reenabled_issues(pr_body: str = "") -> List[str]:
default_branch = f"origin/{os.environ.get('GIT_DEFAULT_BRANCH', 'main')}"
default_branch = os.getenv("GIT_DEFAULT_BRANCH", "main")
try:
commit_messages = subprocess.check_output(
f"git cherry -v {default_branch}".split(" ")
@ -508,15 +494,10 @@ def perform_misc_tasks(
"ci-no-test-timeout", check_for_setting(labels, pr_body, "ci-no-test-timeout")
)
set_output("ci-no-td", check_for_setting(labels, pr_body, "ci-no-td"))
# Only relevant for the one linux distributed cuda job, delete this when TD
# is rolled out completely
set_output(
"ci-td-distributed", check_for_setting(labels, pr_body, "ci-td-distributed")
)
# Obviously, if the job name includes unstable, then this is an unstable job
is_unstable = job_name and IssueType.UNSTABLE.value in job_name
if not is_unstable and test_matrix and test_matrix.get("include"):
if not is_unstable and test_matrix:
# Even when the job name doesn't mention unstable, we will also mark it as
# unstable when the test matrix only includes unstable jobs. Basically, this
# logic allows build or build-and-test jobs to be marked as unstable too.
@ -586,16 +567,6 @@ def main() -> None:
# No PR number, no tag, we can just return the test matrix as it is
filtered_test_matrix = test_matrix
if args.selected_test_configs:
selected_test_configs = {
v.strip().lower()
for v in args.selected_test_configs.split(",")
if v.strip()
}
filtered_test_matrix = filter_selected_test_configs(
filtered_test_matrix, selected_test_configs
)
if args.event_name == "schedule" and args.schedule == "29 8 * * *":
# we don't want to run the mem leak check or disabled tests on normal
# periodically scheduled jobs, only the ones at this time

View File

@ -378,9 +378,7 @@ def main() -> None:
for template, workflows in template_and_workflows:
# added Iterable check to appease the mypy gods
if not isinstance(workflows, Iterable):
raise Exception( # noqa: TRY002
f"How is workflows not iterable? {workflows}"
) # noqa: TRY002
raise Exception(f"How is workflows not iterable? {workflows}")
for workflow in workflows:
workflow.generate_workflow_file(workflow_template=template)

View File

@ -21,6 +21,8 @@ DOCKER_IMAGE_TYPES = ["runtime", "devel"]
def generate_docker_matrix() -> Dict[str, List[Dict[str, str]]]:
ret: List[Dict[str, str]] = []
# CUDA amd64 Docker images are available as both runtime and devel while
# CPU arm64 image is only available as runtime.
for cuda, version in generate_binary_build_matrix.CUDA_ARCHES_FULL_VERSION.items():
for image in DOCKER_IMAGE_TYPES:
ret.append(
@ -31,9 +33,19 @@ def generate_docker_matrix() -> Dict[str, List[Dict[str, str]]]:
cuda
],
"image_type": image,
"platform": "linux/arm64,linux/amd64",
"platform": "linux/amd64",
}
)
ret.append(
{
"cuda": "cpu",
"cuda_full_version": "",
"cudnn_version": "",
"image_type": "runtime",
"platform": "linux/arm64",
}
)
return {"include": ret}

View File

@ -4,7 +4,6 @@
import argparse
import json
import operator
import os
import re
import sys
@ -127,7 +126,7 @@ def find_job_id_name(args: Any) -> Tuple[str, str]:
# Sort the jobs list by start time, in descending order. We want to get the most
# recently scheduled job on the runner.
jobs.sort(key=operator.itemgetter("started_at"), reverse=True)
jobs.sort(key=lambda job: job["started_at"], reverse=True)
for job in jobs:
if job["runner_name"] == args.runner_name:

Binary file not shown.

View File

@ -6,9 +6,6 @@ CONDA_ENV=$(conda env list --json | jq -r ".envs | .[-1]")
eval "$(command conda 'shell.bash' 'hook' 2> /dev/null)"
conda activate "${CONDA_ENV}"
# Use uv to speed up lintrunner init
python3 -m pip install uv
CACHE_DIRECTORY="/tmp/.lintbin"
# Try to recover the cached binaries
if [[ -d "${CACHE_DIRECTORY}" ]]; then

View File

@ -1,20 +0,0 @@
#!/bin/bash
set -euxo pipefail
# Download requirements
cd llm-target-determinator
pip install -q -r requirements.txt
cd ../codellama
pip install -e .
# Run indexer
cd ../llm-target-determinator
torchrun \
--standalone \
--nnodes=1 \
--nproc-per-node=1 \
indexer.py \
--experiment-name indexer-files \
--granularity FILE

View File

@ -9,7 +9,6 @@ from unittest import main, mock, TestCase
import yaml
from filter_test_configs import (
filter,
filter_selected_test_configs,
get_labels,
mark_unstable_jobs,
parse_reenabled_issues,
@ -18,6 +17,7 @@ from filter_test_configs import (
remove_disabled_jobs,
set_periodic_modes,
SUPPORTED_PERIODICAL_MODES,
VALID_TEST_CONFIG_LABELS,
)
@ -273,13 +273,13 @@ class TestConfigFilter(TestCase):
testcases = [
{
"test_matrix": '{include: [{config: "default", runner: "linux"}]}',
"expected": '{"include": []}',
"description": "Request test-config/cfg but the test matrix doesn't have it",
"expected": '{"include": [{"config": "default", "runner": "linux"}]}',
"description": "No match, keep the same test matrix",
},
{
"test_matrix": '{include: [{config: "default", runner: "linux"}, {config: "plain-cfg"}]}',
"expected": '{"include": []}',
"description": "A valid test config label needs to start with test-config/",
"expected": '{"include": [{"config": "default", "runner": "linux"}, {"config": "plain-cfg"}]}',
"description": "No match because there is no prefix or suffix, keep the same test matrix",
},
{
"test_matrix": '{include: [{config: "default", runner: "linux"}, {config: "cfg", shard: 1}]}',
@ -294,8 +294,9 @@ class TestConfigFilter(TestCase):
)
self.assertEqual(case["expected"], json.dumps(filtered_test_matrix))
def test_filter_with_test_config_label(self) -> None:
def test_filter_with_valid_label(self) -> None:
mocked_labels = {f"{PREFIX}cfg", "ciflow/trunk"}
VALID_TEST_CONFIG_LABELS.add(f"{PREFIX}cfg")
testcases = [
{
@ -316,51 +317,6 @@ class TestConfigFilter(TestCase):
)
self.assertEqual(case["expected"], json.dumps(filtered_test_matrix))
def test_filter_selected_test_configs(self) -> None:
testcases = [
{
"test_matrix": '{include: [{config: "default"}]}',
"selected_test_configs": "",
"expected": '{"include": [{"config": "default"}]}',
"description": "No selected test configs",
},
{
"test_matrix": '{include: [{config: "default"}]}',
"selected_test_configs": "foo",
"expected": '{"include": []}',
"description": "A different test config is selected",
},
{
"test_matrix": '{include: [{config: "default"}]}',
"selected_test_configs": "foo, bar",
"expected": '{"include": []}',
"description": "A different set of test configs is selected",
},
{
"test_matrix": '{include: [{config: "default"}]}',
"selected_test_configs": "foo, bar,default",
"expected": '{"include": [{"config": "default"}]}',
"description": "One of the test config is selected",
},
{
"test_matrix": '{include: [{config: "default"}, {config: "bar"}]}',
"selected_test_configs": "foo, bar,Default",
"expected": '{"include": [{"config": "default"}, {"config": "bar"}]}',
"description": "Several test configs are selected",
},
]
for case in testcases:
selected_test_configs = {
v.strip().lower()
for v in case["selected_test_configs"].split(",")
if v.strip()
}
filtered_test_matrix = filter_selected_test_configs(
yaml.safe_load(case["test_matrix"]), selected_test_configs
)
self.assertEqual(case["expected"], json.dumps(filtered_test_matrix))
def test_set_periodic_modes(self) -> None:
testcases: List[Dict[str, str]] = [
{
@ -685,7 +641,6 @@ class TestConfigFilter(TestCase):
ci_verbose_test_logs: bool = False,
ci_no_test_timeout: bool = False,
ci_no_td: bool = False,
ci_td_distributed: bool = False,
is_unstable: bool = False,
reenabled_issues: str = "",
) -> str:
@ -694,7 +649,6 @@ class TestConfigFilter(TestCase):
f"ci-verbose-test-logs={ci_verbose_test_logs}\n"
f"ci-no-test-timeout={ci_no_test_timeout}\n"
f"ci-no-td={ci_no_td}\n"
f"ci-td-distributed={ci_td_distributed}\n"
f"is-unstable={is_unstable}\n"
f"reenabled-issues={reenabled_issues}\n"
)

View File

@ -205,6 +205,7 @@ def mocked_read_merge_rules(repo: Any, org: str, project: str) -> List[MergeRule
approved_by=["pytorch/metamates", "ngimel"],
mandatory_checks_name=[
"Lint",
"Facebook CLA Check",
"pull / linux-xenial-cuda11.3-py3.7-gcc7 / build",
],
ignore_flaky_failures=True,
@ -397,7 +398,7 @@ class TestTryMerge(TestCase):
def test_gql_retrieve_checksuites(self, *args: Any) -> None:
"Fetch comments and conclusions for PR with 60 commits"
pr = GitHubPR("pytorch", "pytorch", 94787)
self.assertEqual(len(pr.get_checkrun_conclusions()), 182)
self.assertEqual(len(pr.get_checkrun_conclusions()), 183)
def test_team_members(self, *args: Any) -> None:
"Test fetching team members works"
@ -741,30 +742,6 @@ class TestBypassFailures(TestCase):
self.assertTrue(len(failed) == 0)
self.assertTrue(len(ignorable["UNSTABLE"]) == 1)
# Add another test case where there is no unstable keyword in the job name, but
# the job has already been marked as unstable
pr = GitHubPR("pytorch", "executorch", 3318)
checks = pr.get_checkrun_conclusions()
checks = get_classifications(
pr.pr_num,
pr.project,
checks,
[],
)
print(checks)
workflow_name = "test-llama-app"
job_name = "mobile-job (android)"
self.assertTrue(
checks[f"Android / {workflow_name} / {job_name}"].classification
== "UNSTABLE"
)
pending, failed, ignorable = categorize_checks(
checks, list(checks.keys()), ok_failed_checks_threshold=1
)
self.assertTrue(len(pending) == 0)
self.assertTrue(len(failed) == 0)
self.assertTrue(len(ignorable["UNSTABLE"]) == 1)
def test_get_classifications_broken_trunk(self, *args: Any) -> None:
# The mock merge base is the actual value returned by gh_fetch_merge_base
test_cases = [
@ -856,41 +833,6 @@ class TestBypassFailures(TestCase):
self.assertTrue(len(ignorable["FLAKY"]) == 4)
self.assertTrue(len(ignorable["BROKEN_TRUNK"]) == 2)
def test_get_classifications_wrong_workflow_name(self, *args: Any) -> None:
pr = GitHubPR("pytorch", "pytorch", 123104)
checks = pr.get_checkrun_conclusions()
check_name = "linux-binary-conda / conda-py3_8-cuda11_8-build / build"
check_name_workflow_path = ".github/workflows/generated-linux-binary-conda-nightly.yml / conda-py3_8-cuda11_8-build / build"
# Mock a check where the workflow name uses the full path
checks[check_name_workflow_path] = JobCheckState(
check_name_workflow_path,
checks[check_name].url,
checks[check_name].status,
checks[check_name].classification,
checks[check_name].job_id,
checks[check_name].title,
checks[check_name].summary,
)
del checks[check_name]
checks = get_classifications(
pr.pr_num,
pr.project,
checks,
[],
)
pending, failed, ignorable = categorize_checks(
checks,
list(checks.keys()),
)
self.assertTrue(len(pending) == 0)
self.assertTrue(len(failed) == 0)
self.assertTrue(len(ignorable["FLAKY"]) == 1)
self.assertTrue(len(ignorable["BROKEN_TRUNK"]) == 0)
@mock.patch("trymerge.read_merge_rules", side_effect=xla_merge_rules)
def test_dont_ignore_flaky_failures(self, *args: Any) -> None:
"""

View File

@ -123,7 +123,6 @@ fragment PRCheckSuites on CheckSuiteConnection {
workflow {
name
}
databaseId
url
}
checkRuns(first: 50) {
@ -1399,10 +1398,7 @@ def find_matching_merge_rule(
)
required_checks = list(
filter(
lambda x: ("EasyCLA" in x)
or ("Facebook CLA Check" in x)
or not skip_mandatory_checks,
mandatory_checks,
lambda x: "EasyCLA" in x or not skip_mandatory_checks, mandatory_checks
)
)
pending_checks, failed_checks, _ = categorize_checks(
@ -1413,13 +1409,6 @@ def find_matching_merge_rule(
else 0,
)
# categorize_checks assumes all tests are required if required_checks is empty.
# this is a workaround as we want to keep that behavior for categorize_checks
# generally.
if not required_checks:
pending_checks = []
failed_checks = []
hud_link = f"https://hud.pytorch.org/{pr.org}/{pr.project}/commit/{pr.last_commit()['oid']}"
if len(failed_checks) > 0:
if reject_reason_score < 30000:
@ -1619,59 +1608,28 @@ def remove_job_name_suffix(name: str, replacement: str = ")") -> str:
def is_broken_trunk(
check: JobCheckState,
name: str,
drci_classifications: Any,
) -> bool:
if not check or not drci_classifications:
if not name or not drci_classifications:
return False
name = check.name
job_id = check.job_id
# Consult the list of broken trunk failures from Dr.CI
return any(
(name == broken_trunk["name"]) or (job_id and job_id == broken_trunk["id"])
name == broken_trunk["name"]
for broken_trunk in drci_classifications.get("BROKEN_TRUNK", [])
)
def is_unstable(
check: JobCheckState,
drci_classifications: Any,
) -> bool:
if not check or not drci_classifications:
return False
name = check.name
job_id = check.job_id
# The job name has the unstable keyword. This is the original way to mark a job
# as unstable on HUD, Dr.CI, and trymerge
if "unstable" in name:
return True
# Consult the list of unstable failures from Dr.CI
return any(
(name == unstable["name"] or (job_id and job_id == unstable["id"]))
for unstable in drci_classifications.get("UNSTABLE", [])
)
def is_flaky(
check: JobCheckState,
name: str,
drci_classifications: Any,
) -> bool:
if not check or not drci_classifications:
if not name or not drci_classifications:
return False
name = check.name
job_id = check.job_id
# Consult the list of flaky failures from Dr.CI
return any(
(name == flaky["name"] or (job_id and job_id == flaky["id"]))
for flaky in drci_classifications.get("FLAKY", [])
)
return any(name == flaky["name"] for flaky in drci_classifications.get("FLAKY", []))
def is_invalid_cancel(
@ -1744,7 +1702,7 @@ def get_classifications(
if check.status == "SUCCESS" or check.status == "NEUTRAL":
continue
if is_unstable(check, drci_classifications):
if "unstable" in name:
checks_with_classifications[name] = JobCheckState(
check.name,
check.url,
@ -1758,7 +1716,7 @@ def get_classifications(
# NB: It's important to note that when it comes to ghstack and broken trunk classification,
# Dr.CI uses the base of the whole stack
if is_broken_trunk(check, drci_classifications):
if is_broken_trunk(name, drci_classifications):
checks_with_classifications[name] = JobCheckState(
check.name,
check.url,
@ -1770,7 +1728,7 @@ def get_classifications(
)
continue
elif is_flaky(check, drci_classifications):
elif is_flaky(name, drci_classifications):
checks_with_classifications[name] = JobCheckState(
check.name,
check.url,

View File

@ -60,7 +60,7 @@ def rebase_onto(
repo._run_git("rebase", onto_branch, branch)
if repo.rev_parse(branch) == repo.rev_parse(onto_branch):
raise Exception(SAME_SHA_ERROR) # noqa: TRY002
raise Exception(SAME_SHA_ERROR)
if dry_run:
push_result = repo._run_git("push", "--dry-run", "-f", remote_url, refspec)
@ -100,7 +100,7 @@ def rebase_ghstack_onto(
repo._run_git("rebase", onto_branch, orig_ref)
if repo.rev_parse(orig_ref) == repo.rev_parse(onto_branch):
raise Exception(SAME_SHA_ERROR) # noqa: TRY002
raise Exception(SAME_SHA_ERROR)
# steal the identity of the committer of the commit on the orig branch
email = repo._run_git("log", orig_ref, "--pretty=format:%ae", "-1")
@ -126,7 +126,7 @@ def rebase_ghstack_onto(
print(push_result)
if ghstack_result.returncode != 0:
print(ghstack_result.stderr.decode("utf-8"))
raise Exception(f"\n```{push_result}```") # noqa: TRY002
raise Exception(f"\n```{push_result}```")
# The contents of a successful push result should look like:
# Summary of changes (ghstack 0.6.0)

View File

@ -8,7 +8,7 @@
# NOTE: If testing pytorch/builder changes you can change this variable to change what pytorch/builder reference
# the binary builds will check out
{%- set builder_repo = "pytorch/builder" -%}
{%- set builder_branch = "main" -%}
{%- set builder_branch = "release/2.3" -%}
{%- macro concurrency(build_environment) -%}
concurrency:

View File

@ -100,8 +100,8 @@ jobs:
with:
name: !{{ config["build_name"] }}
path: "${{ runner.temp }}/artifacts/"
!{{ common.checkout(deep_clone=False, directory="pytorch") }}
!{{ common.checkout(deep_clone=False, directory="builder", repository=common.builder_repo, branch=common.builder_branch) }}
!{{ common.checkout(deep_clone=False, directory="pytorch", checkout_pr_head=False) }}
!{{ common.checkout(deep_clone=False, directory="builder", repository=common.builder_repo, branch=common.builder_branch, checkout_pr_head=False) }}
- name: ROCm set GPU_FLAG
run: |
echo "GPU_FLAG=--device=/dev/mem --device=/dev/kfd --device=/dev/dri --group-add video --group-add daemon" >> "${GITHUB_ENV}"

View File

@ -81,8 +81,8 @@ jobs:
elif [ -d "/Applications/Xcode_13.3.1.app" ]; then
echo "DEVELOPER_DIR=/Applications/Xcode_13.3.1.app/Contents/Developer" >> "${GITHUB_ENV}"
fi
!{{ common.checkout(deep_clone=False, directory="pytorch") }}
!{{ common.checkout(deep_clone=False, directory="builder", repository=common.builder_repo, branch=common.builder_branch) }}
!{{ common.checkout(deep_clone=False, directory="pytorch", checkout_pr_head=False) }}
!{{ common.checkout(deep_clone=False, directory="builder", repository=common.builder_repo, branch=common.builder_branch, checkout_pr_head=False) }}
- name: Install sccache (only for non-forked PRs, and pushes to trunk)
uses: nick-fields/retry@v2.8.2
if: ${{ github.event_name == 'push' || github.event.pull_request.head.repo.full_name == github.repository }}

View File

@ -65,8 +65,8 @@ jobs:
steps:
!{{ common.setup_ec2_windows() }}
!{{ set_runner_specific_vars() }}
!{{ common.checkout(deep_clone=False, directory="pytorch") }}
!{{ common.checkout(deep_clone=False, directory="builder", repository=common.builder_repo, branch=common.builder_branch) }}
!{{ common.checkout(deep_clone=False, directory="pytorch", checkout_pr_head=False) }}
!{{ common.checkout(deep_clone=False, directory="builder", repository=common.builder_repo, branch=common.builder_branch, checkout_pr_head=False) }}
- name: Populate binary env
shell: bash
run: |
@ -105,8 +105,8 @@ jobs:
with:
name: !{{ config["build_name"] }}
path: "${{ env.PYTORCH_FINAL_PACKAGE_DIR }}"
!{{ common.checkout(deep_clone=False, directory="pytorch") }}
!{{ common.checkout(deep_clone=False, directory="builder", repository=common.builder_repo, branch=common.builder_branch) }}
!{{ common.checkout(deep_clone=False, directory="pytorch", checkout_pr_head=False) }}
!{{ common.checkout(deep_clone=False, directory="builder", repository=common.builder_repo, branch=common.builder_branch, checkout_pr_head=False) }}
- name: Populate binary env
shell: bash
run: |

View File

@ -37,7 +37,7 @@ jobs:
keep-going: ${{ steps.filter.outputs.keep-going }}
steps:
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
with:
fetch-depth: 1
submodules: false
@ -59,25 +59,25 @@ jobs:
runs-on: ${{ matrix.runner }}
steps:
- name: Setup SSH (Click me for login details)
uses: pytorch/test-infra/.github/actions/setup-ssh@main
uses: pytorch/test-infra/.github/actions/setup-ssh@release/2.3
with:
github-secret: ${{ secrets.GITHUB_TOKEN }}
# [see note: pytorch repo ref]
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
- name: Setup Linux
uses: ./.github/actions/setup-linux
- name: Calculate docker image
id: calculate-docker-image
uses: pytorch/test-infra/.github/actions/calculate-docker-image@main
uses: pytorch/test-infra/.github/actions/calculate-docker-image@release/2.3
with:
docker-image-name: ${{ inputs.docker-image-name }}
- name: Pull docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
uses: pytorch/test-infra/.github/actions/pull-docker-image@release/2.3
with:
docker-image: ${{ steps.calculate-docker-image.outputs.docker-image }}
@ -141,5 +141,5 @@ jobs:
if: always()
- name: Teardown Linux
uses: pytorch/test-infra/.github/actions/teardown-linux@main
uses: pytorch/test-infra/.github/actions/teardown-linux@release/2.3
if: always()

View File

@ -37,7 +37,7 @@ jobs:
keep-going: ${{ steps.filter.outputs.keep-going }}
steps:
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
with:
fetch-depth: 1
submodules: false
@ -59,25 +59,25 @@ jobs:
runs-on: ${{ matrix.runner }}
steps:
- name: Setup SSH (Click me for login details)
uses: pytorch/test-infra/.github/actions/setup-ssh@main
uses: pytorch/test-infra/.github/actions/setup-ssh@release/2.3
with:
github-secret: ${{ secrets.GITHUB_TOKEN }}
# [see note: pytorch repo ref]
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
- name: Setup Linux
uses: ./.github/actions/setup-linux
- name: Calculate docker image
id: calculate-docker-image
uses: pytorch/test-infra/.github/actions/calculate-docker-image@main
uses: pytorch/test-infra/.github/actions/calculate-docker-image@release/2.3
with:
docker-image-name: ${{ inputs.docker-image-name }}
- name: Pull docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
uses: pytorch/test-infra/.github/actions/pull-docker-image@release/2.3
with:
docker-image: ${{ steps.calculate-docker-image.outputs.docker-image }}
@ -186,5 +186,5 @@ jobs:
if: always()
- name: Teardown Linux
uses: pytorch/test-infra/.github/actions/teardown-linux@main
uses: pytorch/test-infra/.github/actions/teardown-linux@release/2.3
if: always()

View File

@ -42,7 +42,7 @@ jobs:
reenabled-issues: ${{ steps.filter.outputs.reenabled-issues }}
steps:
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
with:
fetch-depth: 1
submodules: false
@ -64,36 +64,31 @@ jobs:
runs-on: ${{ matrix.runner }}
steps:
- name: Setup SSH (Click me for login details)
uses: pytorch/test-infra/.github/actions/setup-ssh@main
uses: pytorch/test-infra/.github/actions/setup-ssh@release/2.3
with:
github-secret: ${{ secrets.GITHUB_TOKEN }}
# [see note: pytorch repo ref]
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
- name: Setup Linux
uses: ./.github/actions/setup-linux
- name: Calculate docker image
id: calculate-docker-image
uses: pytorch/test-infra/.github/actions/calculate-docker-image@main
uses: pytorch/test-infra/.github/actions/calculate-docker-image@release/2.3
with:
docker-image-name: ${{ inputs.docker-image-name }}
- name: Pull docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
uses: pytorch/test-infra/.github/actions/pull-docker-image@release/2.3
with:
docker-image: ${{ steps.calculate-docker-image.outputs.docker-image }}
- name: Check if in a ARC runner
shell: bash
id: check_arc_runner
run: echo "IN_ARC_RUNNER=$([ -f /.inarc ] && echo true || echo false)" >> "$GITHUB_OUTPUT"
- name: Install nvidia driver, nvidia-docker runtime, set GPU_FLAG
uses: pytorch/test-infra/.github/actions/setup-nvidia@main
if: ${{ inputs.cuda-version != 'cpu' && steps.check_arc_runner.outputs.IN_ARC_RUNNER == 'false' }}
uses: pytorch/test-infra/.github/actions/setup-nvidia@release/2.3
if: ${{ inputs.cuda-version != 'cpu' }}
- name: Output disk space left
run: |
@ -201,5 +196,5 @@ jobs:
file-suffix: bazel-${{ github.job }}_${{ steps.get-job-id.outputs.job-id }}
- name: Teardown Linux
uses: pytorch/test-infra/.github/actions/teardown-linux@main
uses: pytorch/test-infra/.github/actions/teardown-linux@release/2.3
if: always()

View File

@ -139,13 +139,13 @@ jobs:
run: env
- name: "[FB EMPLOYEES] Enable SSH (Click me for login details)"
uses: pytorch/test-infra/.github/actions/setup-ssh@main
uses: pytorch/test-infra/.github/actions/setup-ssh@release/2.3
continue-on-error: true
with:
github-secret: ${{ secrets.github-token }}
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
with:
no-sudo: ${{ inputs.build_environment == 'linux-aarch64-binary-manywheel' }}
@ -173,7 +173,6 @@ jobs:
- name: Checkout PyTorch to pytorch dir
uses: malfet/checkout@silent-checkout
with:
ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
submodules: recursive
path: pytorch
quiet-checkout: true
@ -187,7 +186,7 @@ jobs:
- name: Checkout pytorch/builder to builder dir
uses: malfet/checkout@silent-checkout
with:
ref: main
ref: release/2.3
submodules: recursive
repository: pytorch/builder
path: builder
@ -213,7 +212,7 @@ jobs:
- name: Pull Docker image
if: ${{ steps.filter.outputs.is-test-matrix-empty == 'False' }}
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
uses: pytorch/test-infra/.github/actions/pull-docker-image@release/2.3
with:
docker-image: ${{ inputs.DOCKER_IMAGE }}
@ -270,7 +269,7 @@ jobs:
- name: Teardown Linux
if: always()
uses: pytorch/test-infra/.github/actions/teardown-linux@main
uses: pytorch/test-infra/.github/actions/teardown-linux@release/2.3
- name: Chown workspace
if: always()

View File

@ -127,14 +127,14 @@ jobs:
} >> "${GITHUB_ENV} }}"
- name: "[FB EMPLOYEES] Enable SSH (Click me for login details)"
uses: pytorch/test-infra/.github/actions/setup-ssh@main
uses: pytorch/test-infra/.github/actions/setup-ssh@release/2.3
continue-on-error: true
with:
github-secret: ${{ secrets.github-token }}
# Setup the environment
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
with:
no-sudo: ${{ inputs.build_environment == 'linux-aarch64-binary-manywheel' }}
@ -155,7 +155,6 @@ jobs:
- name: Checkout PyTorch to pytorch dir
uses: malfet/checkout@silent-checkout
with:
ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
submodules: recursive
path: pytorch
@ -168,7 +167,7 @@ jobs:
- name: Checkout pytorch/builder to builder dir
uses: malfet/checkout@silent-checkout
with:
ref: main
ref: release/2.3
submodules: recursive
repository: pytorch/builder
path: builder
@ -199,12 +198,12 @@ jobs:
path: "${{ runner.temp }}/artifacts/"
- name: Install nvidia driver, nvidia-docker runtime, set GPU_FLAG
uses: pytorch/test-infra/.github/actions/setup-nvidia@main
uses: pytorch/test-infra/.github/actions/setup-nvidia@release/2.3
if: ${{ inputs.GPU_ARCH_TYPE == 'cuda' && steps.filter.outputs.is-test-matrix-empty == 'False' }}
- name: Pull Docker image
if: ${{ steps.filter.outputs.is-test-matrix-empty == 'False' }}
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
uses: pytorch/test-infra/.github/actions/pull-docker-image@release/2.3
with:
docker-image: ${{ inputs.DOCKER_IMAGE }}
@ -214,7 +213,7 @@ jobs:
- name: Teardown Linux
if: always()
uses: pytorch/test-infra/.github/actions/teardown-linux@main
uses: pytorch/test-infra/.github/actions/teardown-linux@release/2.3
- name: Chown workspace
if: always()

View File

@ -95,7 +95,7 @@ jobs:
SHA1: ${{ github.event.pull_request.head.sha || github.sha }}
steps:
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
with:
no-sudo: true

View File

@ -23,7 +23,7 @@ jobs:
keep-going: ${{ steps.filter.outputs.keep-going }}
steps:
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
with:
fetch-depth: 1
submodules: false
@ -44,7 +44,7 @@ jobs:
runs-on: ${{ matrix.runner }}
steps:
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
- name: Set up JDK 8
uses: actions/setup-java@v3
@ -53,7 +53,7 @@ jobs:
distribution: 'temurin'
- name: Setup miniconda
uses: pytorch/test-infra/.github/actions/setup-miniconda@main
uses: pytorch/test-infra/.github/actions/setup-miniconda@release/2.3
with:
python-version: 3.8
environment-file: .github/requirements/conda-env-${{ runner.os }}-${{ runner.arch }}

View File

@ -28,21 +28,7 @@ on:
description: |
If this is set, our linter will use this to make sure that every other
job with the same `sync-tag` is identical.
s3-bucket:
description: S3 bucket to download artifact
required: false
type: string
default: "gha-artifacts"
aws-role-to-assume:
description: role to assume for downloading artifacts
required: false
type: string
default: ""
upload-aws-role-to-assume:
description: role to assume for downloading artifacts
required: false
type: string
default: ""
secrets:
GH_PYTORCHBOT_TOKEN:
required: false
@ -80,7 +66,7 @@ jobs:
name: build-docs-${{ matrix.docs_type }}-${{ inputs.push }}
steps:
- name: Setup SSH (Click me for login details)
uses: pytorch/test-infra/.github/actions/setup-ssh@main
uses: pytorch/test-infra/.github/actions/setup-ssh@release/2.3
with:
github-secret: ${{ secrets.GITHUB_TOKEN }}
instructions: |
@ -91,27 +77,19 @@ jobs:
# [see note: pytorch repo ref]
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
- name: Setup Linux
uses: ./.github/actions/setup-linux
- name: configure aws credentials
if : ${{ inputs.aws-role-to-assume != '' }}
uses: aws-actions/configure-aws-credentials@v3
with:
role-to-assume: ${{ inputs.aws-role-to-assume }}
role-session-name: gha-linux-test
aws-region: us-east-1
- name: Calculate docker image
id: calculate-docker-image
uses: pytorch/test-infra/.github/actions/calculate-docker-image@main
uses: pytorch/test-infra/.github/actions/calculate-docker-image@release/2.3
with:
docker-image-name: ${{ inputs.docker-image }}
- name: Pull docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
uses: pytorch/test-infra/.github/actions/pull-docker-image@release/2.3
with:
docker-image: ${{ steps.calculate-docker-image.outputs.docker-image }}
@ -119,7 +97,6 @@ jobs:
uses: ./.github/actions/download-build-artifacts
with:
name: ${{ inputs.build-environment }}
s3-bucket: ${{ inputs.s3-bucket }}
- name: Generate netrc (only for docs-push)
if: inputs.push
@ -179,14 +156,6 @@ jobs:
uses: ./.github/actions/chown-workspace
if: always()
- name: configure aws credentials
if : ${{ inputs.upload-aws-role-to-assume != '' }}
uses: aws-actions/configure-aws-credentials@v3
with:
role-to-assume: ${{ inputs.upload-aws-role-to-assume }}
role-session-name: gha-linux-test
aws-region: us-east-1
- name: Upload Python Docs Preview
uses: seemethere/upload-artifact-s3@v5
if: ${{ github.event_name == 'pull_request' && matrix.docs_type == 'python' && steps.build-docs.outcome == 'success' }}
@ -218,5 +187,5 @@ jobs:
s3-prefix: pytorch/pytorch/${{ github.event.pull_request.number }}/functorchdocs
- name: Teardown Linux
uses: pytorch/test-infra/.github/actions/teardown-linux@main
uses: pytorch/test-infra/.github/actions/teardown-linux@release/2.3
if: always()

View File

@ -46,7 +46,7 @@ jobs:
keep-going: ${{ steps.filter.outputs.keep-going }}
steps:
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
with:
fetch-depth: 1
submodules: false
@ -80,7 +80,7 @@ jobs:
steps:
# [see note: pytorch repo ref]
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
- name: Populate CI build options
shell: bash
@ -102,7 +102,7 @@ jobs:
brew install libtool
- name: Setup miniconda for iOS
uses: pytorch/test-infra/.github/actions/setup-miniconda@main
uses: pytorch/test-infra/.github/actions/setup-miniconda@release/2.3
with:
python-version: "3.9"
environment-file: .github/requirements/conda-env-iOS.txt

View File

@ -1,109 +0,0 @@
name: linux-build
on:
workflow_call:
inputs:
build-environment:
required: true
type: string
description: Top-level label for what's being built/tested.
docker-image-name:
required: true
type: string
description: Name of the base docker image to build with.
build-generates-artifacts:
required: false
type: boolean
default: true
description: If set, upload generated build artifacts.
build-with-debug:
required: false
type: boolean
default: false
description: If set, build in debug mode.
sync-tag:
required: false
type: string
default: ""
description: |
If this is set, our linter will use this to make sure that every other
job with the same `sync-tag` is identical.
cuda-arch-list:
required: false
type: string
default: "5.2"
description: Runner label to select worker type
runner:
required: false
type: string
default: "linux.2xlarge"
description: |
List of CUDA architectures CI build should target.
test-matrix:
required: false
type: string
description: |
An option JSON description of what test configs to run later on. This
is moved here from the Linux test workflow so that we can apply filter
logic using test-config labels earlier and skip unnecessary builds
s3-bucket:
description: S3 bucket to download artifact
required: false
type: string
default: "gha-artifacts"
aws-role-to-assume:
description: role to assume for downloading artifacts
required: false
type: string
default: ""
secrets:
HUGGING_FACE_HUB_TOKEN:
required: false
description: |
HF Auth token to avoid rate limits when downloading models or datasets from hub
outputs:
docker-image:
value: ${{ jobs.build.outputs.docker-image }}
description: The docker image containing the built PyTorch.
test-matrix:
value: ${{ jobs.build.outputs.test-matrix }}
description: An optional JSON description of what test configs to run later on.
jobs:
build:
# Don't run on forked repos
if: github.repository_owner == 'pytorch'
runs-on: ${{ inputs.runner }}
timeout-minutes: 240
outputs:
docker-image: ${{ steps.linux-build.outputs.docker-image }}
test-matrix: ${{ steps.linux-build.outputs.test-matrix }}
steps:
- name: Setup SSH (Click me for login details)
uses: pytorch/test-infra/.github/actions/setup-ssh@main
with:
github-secret: ${{ secrets.GITHUB_TOKEN }}
# [pytorch repo ref]
# Use a pytorch/pytorch reference instead of a reference to the local
# checkout because when we run this action we don't *have* a local
# checkout. In other cases you should prefer a local checkout.
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
- name: Linux Build
id: linux-build
uses: ./.github/actions/linux-build
with:
build-environment: ${{ inputs.build-environment }}
docker-image-name: ${{ inputs.docker-image-name }}
build-generates-artifacts: ${{ inputs.build-generates-artifacts }}
build-with-debug: ${{ inputs.build-with-debug }}
sync-tag: ${{ inputs.sync-tag }}
cuda-arch-list: ${{ inputs.cuda-arch-list }}
test-matrix: ${{ inputs.test-matrix }}
s3-bucket: ${{ inputs.s3-bucket }}
aws-role-to-assume: ${{ inputs.aws-role-to-assume }}
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
HUGGING_FACE_HUB_TOKEN: ${{ secrets.HUGGING_FACE_HUB_TOKEN }}

View File

@ -1,105 +0,0 @@
name: linux-build-rg
on:
workflow_call:
inputs:
build-environment:
required: true
type: string
description: Top-level label for what's being built/tested.
docker-image-name:
required: true
type: string
description: Name of the base docker image to build with.
build-generates-artifacts:
required: false
type: boolean
default: true
description: If set, upload generated build artifacts.
build-with-debug:
required: false
type: boolean
default: false
description: If set, build in debug mode.
sync-tag:
required: false
type: string
default: ""
description: |
If this is set, our linter will use this to make sure that every other
job with the same `sync-tag` is identical.
cuda-arch-list:
required: false
type: string
default: "5.2"
description: |
List of CUDA architectures CI build should target.
runner-group:
required: false
type: string
default: "arc-lf-linux.2xlarge"
description: Runner group to select group type
test-matrix:
required: false
type: string
description: |
An option JSON description of what test configs to run later on. This
is moved here from the Linux test workflow so that we can apply filter
logic using test-config labels earlier and skip unnecessary builds
s3-bucket:
description: S3 bucket to download artifact
required: false
type: string
default: "gha-artifacts"
aws-role-to-assume:
description: role to assume for downloading artifacts
required: false
type: string
default: ""
secrets:
HUGGING_FACE_HUB_TOKEN:
required: false
description: |
HF Auth token to avoid rate limits when downloading models or datasets from hub
outputs:
docker-image:
value: ${{ jobs.build.outputs.docker-image }}
description: The docker image containing the built PyTorch.
test-matrix:
value: ${{ jobs.build.outputs.test-matrix }}
description: An optional JSON description of what test configs to run later on.
jobs:
build:
# Don't run on forked repos
if: github.repository_owner == 'pytorch'
runs-on:
group: ${{ inputs.runner-group }}
timeout-minutes: 240
outputs:
docker-image: ${{ steps.linux-build.outputs.docker-image }}
test-matrix: ${{ steps.linux-build.outputs.test-matrix }}
steps:
# [pytorch repo ref]
# Use a pytorch/pytorch reference instead of a reference to the local
# checkout because when we run this action we don't *have* a local
# checkout. In other cases you should prefer a local checkout.
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
- name: Linux Build
id: linux-build
uses: ./.github/actions/linux-build
with:
build-environment: ${{ inputs.build-environment }}
docker-image-name: ${{ inputs.docker-image-name }}
build-generates-artifacts: ${{ inputs.build-generates-artifacts }}
build-with-debug: ${{ inputs.build-with-debug }}
sync-tag: ${{ inputs.sync-tag }}
cuda-arch-list: ${{ inputs.cuda-arch-list }}
test-matrix: ${{ inputs.test-matrix }}
s3-bucket: ${{ inputs.s3-bucket }}
aws-role-to-assume: ${{ inputs.aws-role-to-assume }}
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
HUGGING_FACE_HUB_TOKEN: ${{ secrets.HUGGING_FACE_HUB_TOKEN }}

View File

@ -47,23 +47,6 @@ on:
An option JSON description of what test configs to run later on. This
is moved here from the Linux test workflow so that we can apply filter
logic using test-config labels earlier and skip unnecessary builds
selected-test-configs:
description: |
A comma-separated list of test configurations from the test matrix to keep,
The empty list means we are going to keep every configurations by defaults
required: false
type: string
default: ""
s3-bucket:
description: S3 bucket to download artifact
required: false
type: string
default: "gha-artifacts"
aws-role-to-assume:
description: Role to assume for downloading artifacts
required: false
type: string
default: ""
secrets:
HUGGING_FACE_HUB_TOKEN:
required: false
@ -90,7 +73,7 @@ jobs:
test-matrix: ${{ steps.filter.outputs.test-matrix }}
steps:
- name: Setup SSH (Click me for login details)
uses: pytorch/test-infra/.github/actions/setup-ssh@main
uses: pytorch/test-infra/.github/actions/setup-ssh@release/2.3
with:
github-secret: ${{ secrets.GITHUB_TOKEN }}
@ -99,22 +82,14 @@ jobs:
# checkout because when we run this action we don't *have* a local
# checkout. In other cases you should prefer a local checkout.
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
- name: Setup Linux
uses: ./.github/actions/setup-linux
- name: configure aws credentials
uses: aws-actions/configure-aws-credentials@v3
if: ${{ inputs.aws-role-to-assume != '' }}
with:
role-to-assume: ${{ inputs.aws-role-to-assume }}
role-session-name: gha-linux-build
aws-region: us-east-1
- name: Calculate docker image
id: calculate-docker-image
uses: pytorch/test-infra/.github/actions/calculate-docker-image@main
uses: pytorch/test-infra/.github/actions/calculate-docker-image@release/2.3
with:
docker-image-name: ${{ inputs.docker-image-name }}
@ -128,7 +103,7 @@ jobs:
echo "docker pull ghcr.io/pytorch/ci-image:${tag/:/-}"
- name: Pull docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
uses: pytorch/test-infra/.github/actions/pull-docker-image@release/2.3
with:
docker-image: ${{ steps.calculate-docker-image.outputs.docker-image }}
@ -150,7 +125,6 @@ jobs:
with:
github-token: ${{ secrets.GITHUB_TOKEN }}
test-matrix: ${{ inputs.test-matrix }}
selected-test-configs: ${{ inputs.selected-test-configs }}
job-name: ${{ steps.get-job-id.outputs.job-name }}
- name: Download pytest cache
@ -159,7 +133,6 @@ jobs:
with:
cache_dir: .pytest_cache
job_identifier: ${{ github.workflow }}_${{ inputs.build-environment }}
s3_bucket: ${{ inputs.s3-bucket }}
- name: Build
if: steps.filter.outputs.is-test-matrix-empty == 'False' || inputs.test-matrix == ''
@ -224,7 +197,6 @@ jobs:
retention-days: 14
if-no-files-found: error
path: artifacts.zip
s3-bucket: ${{ inputs.s3-bucket }}
- name: Upload sccache stats
if: steps.build.outcome != 'skipped'
@ -235,8 +207,7 @@ jobs:
retention-days: 365
if-no-files-found: warn
path: sccache-stats-*.json
s3-bucket: ${{ inputs.s3-bucket }}
- name: Teardown Linux
uses: pytorch/test-infra/.github/actions/teardown-linux@main
uses: pytorch/test-infra/.github/actions/teardown-linux@release/2.3
if: always()

View File

@ -1,85 +0,0 @@
name: linux-test-rg
on:
workflow_call:
inputs:
build-environment:
required: true
type: string
description: Top-level label for what's being built/tested.
test-matrix:
required: true
type: string
description: JSON description of what test configs to run.
docker-image:
required: true
type: string
description: Docker image to run in.
sync-tag:
required: false
type: string
default: ""
description: |
If this is set, our linter will use this to make sure that every other
job with the same `sync-tag` is identical.
timeout-minutes:
required: false
type: number
default: 240
description: |
Set the maximum (in minutes) how long the workflow should take to finish
use-gha:
required: false
type: string
default: ""
description: If set to any value, upload to GHA. Otherwise upload to S3.
dashboard-tag:
required: false
type: string
default: ""
s3-bucket:
description: S3 bucket to download artifact
required: false
type: string
default: "gha-artifacts"
aws-role-to-assume:
description: role to assume for downloading artifacts
required: false
type: string
default: ""
secrets:
HUGGING_FACE_HUB_TOKEN:
required: false
description: |
HF Auth token to avoid rate limits when downloading models or datasets from hub
env:
GIT_DEFAULT_BRANCH: ${{ github.event.repository.default_branch }}
jobs:
test:
# Don't run on forked repos or empty test matrix
if: github.repository_owner == 'pytorch' && toJSON(fromJSON(inputs.test-matrix).include) != '[]'
strategy:
matrix: ${{ fromJSON(inputs.test-matrix) }}
fail-fast: false
runs-on: ${{ matrix.runner }}
timeout-minutes: ${{ matrix.mem_leak_check == 'mem_leak_check' && 600 || inputs.timeout-minutes }}
steps:
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
- name: Linux Test
id: linux-test
uses: ./.github/actions/linux-test
with:
build-environment: ${{ inputs.build-environment }}
test-matrix: ${{ inputs.test-matrix }}
docker-image: ${{ inputs.docker-image }}
sync-tag: ${{ inputs.sync-tag }}
use-gha: ${{ inputs.use-gha }}
dashboard-tag: ${{ inputs.dashboard-tag }}
s3-bucket: ${{ inputs.s3-bucket }}
aws-role-to-assume: ${{ inputs.aws-role-to-assume }}
HUGGING_FACE_HUB_TOKEN: ${{ secrets.HUGGING_FACE_HUB_TOKEN }}
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

View File

@ -1,86 +0,0 @@
name: linux-test-label
on:
workflow_call:
inputs:
build-environment:
required: true
type: string
description: Top-level label for what's being built/tested.
test-matrix:
required: true
type: string
description: JSON description of what test configs to run.
docker-image:
required: true
type: string
description: Docker image to run in.
sync-tag:
required: false
type: string
default: ""
description: |
If this is set, our linter will use this to make sure that every other
job with the same `sync-tag` is identical.
timeout-minutes:
required: false
type: number
default: 240
description: |
Set the maximum (in minutes) how long the workflow should take to finish
use-gha:
required: false
type: string
default: ""
description: If set to any value, upload to GHA. Otherwise upload to S3.
dashboard-tag:
required: false
type: string
default: ""
s3-bucket:
description: S3 bucket to download artifact
required: false
type: string
default: "gha-artifacts"
aws-role-to-assume:
description: role to assume for downloading artifacts
required: false
type: string
default: ""
secrets:
HUGGING_FACE_HUB_TOKEN:
required: false
description: |
HF Auth token to avoid rate limits when downloading models or datasets from hub
env:
GIT_DEFAULT_BRANCH: ${{ github.event.repository.default_branch }}
jobs:
test:
# Don't run on forked repos or empty test matrix
if: github.repository_owner == 'pytorch' && toJSON(fromJSON(inputs.test-matrix).include) != '[]'
strategy:
matrix: ${{ fromJSON(inputs.test-matrix) }}
fail-fast: false
runs-on:
group: ${{ matrix.runner }}
timeout-minutes: ${{ matrix.mem_leak_check == 'mem_leak_check' && 600 || inputs.timeout-minutes }}
steps:
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
- name: Linux Test
id: linux-test
uses: ./.github/actions/linux-test
with:
build-environment: ${{ inputs.build-environment }}
test-matrix: ${{ inputs.test-matrix }}
docker-image: ${{ inputs.docker-image }}
sync-tag: ${{ inputs.sync-tag }}
use-gha: ${{ inputs.use-gha }}
dashboard-tag: ${{ inputs.dashboard-tag }}
s3-bucket: ${{ inputs.s3-bucket }}
aws-role-to-assume: ${{ inputs.aws-role-to-assume }}
HUGGING_FACE_HUB_TOKEN: ${{ secrets.HUGGING_FACE_HUB_TOKEN }}
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

View File

@ -37,16 +37,6 @@ on:
required: false
type: string
default: ""
s3-bucket:
description: S3 bucket to download artifact
required: false
type: string
default: "gha-artifacts"
aws-role-to-assume:
description: role to assume for downloading artifacts
required: false
type: string
default: ""
secrets:
HUGGING_FACE_HUB_TOKEN:
required: false
@ -67,7 +57,7 @@ jobs:
timeout-minutes: ${{ matrix.mem_leak_check == 'mem_leak_check' && 600 || inputs.timeout-minutes }}
steps:
- name: Setup SSH (Click me for login details)
uses: pytorch/test-infra/.github/actions/setup-ssh@main
uses: pytorch/test-infra/.github/actions/setup-ssh@release/2.3
if: ${{ !contains(matrix.runner, 'gcp.a100') }}
with:
github-secret: ${{ secrets.GITHUB_TOKEN }}
@ -76,22 +66,14 @@ jobs:
docker exec -it $(docker container ps --format '{{.ID}}') bash
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
- name: Setup Linux
uses: ./.github/actions/setup-linux
- name: configure aws credentials
if : ${{ inputs.aws-role-to-assume != '' }}
uses: aws-actions/configure-aws-credentials@v3
with:
role-to-assume: ${{ inputs.aws-role-to-assume }}
role-session-name: gha-linux-test
aws-region: us-east-1
- name: Calculate docker image
id: calculate-docker-image
uses: pytorch/test-infra/.github/actions/calculate-docker-image@main
uses: pytorch/test-infra/.github/actions/calculate-docker-image@release/2.3
with:
docker-image-name: ${{ inputs.docker-image }}
@ -105,19 +87,14 @@ jobs:
echo "docker pull ghcr.io/pytorch/ci-image:${tag/:/-}"
- name: Pull docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
uses: pytorch/test-infra/.github/actions/pull-docker-image@release/2.3
with:
docker-image: ${{ steps.calculate-docker-image.outputs.docker-image }}
- name: Check if in a ARC runner
shell: bash
id: check_arc_runner
run: echo "IN_ARC_RUNNER=$([ -f /.inarc ] && echo true || echo false)" >> "$GITHUB_OUTPUT"
- name: Install nvidia driver, nvidia-docker runtime, set GPU_FLAG
id: install-nvidia-driver
uses: pytorch/test-infra/.github/actions/setup-nvidia@main
if: ${{ contains(inputs.build-environment, 'cuda') && !contains(matrix.config, 'nogpu') && steps.check_arc_runner.outputs.IN_ARC_RUNNER == 'false' }}
uses: pytorch/test-infra/.github/actions/setup-nvidia@release/2.3
if: contains(inputs.build-environment, 'cuda') && !contains(matrix.config, 'nogpu')
- name: Lock NVIDIA A100 40GB Frequency
run: |
@ -139,7 +116,6 @@ jobs:
uses: ./.github/actions/download-build-artifacts
with:
name: ${{ inputs.build-environment }}
s3-bucket: ${{ inputs.s3-bucket }}
- name: Download TD artifacts
continue-on-error: true
@ -200,7 +176,6 @@ jobs:
VERBOSE_TEST_LOGS: ${{ steps.keep-going.outputs.ci-verbose-test-logs }}
NO_TEST_TIMEOUT: ${{ steps.keep-going.outputs.ci-no-test-timeout }}
NO_TD: ${{ steps.keep-going.outputs.ci-no-td }}
TD_DISTRIBUTED: ${{ steps.keep-going.outputs.ci-td-distributed }}
SCCACHE_BUCKET: ossci-compiler-cache-circleci-v2
SCCACHE_S3_KEY_PREFIX: ${{ github.workflow }}
SHM_SIZE: ${{ contains(inputs.build-environment, 'cuda') && '2g' || '1g' }}
@ -253,7 +228,6 @@ jobs:
-e VERBOSE_TEST_LOGS \
-e NO_TEST_TIMEOUT \
-e NO_TD \
-e TD_DISTRIBUTED \
-e PR_LABELS \
-e MAX_JOBS="$(nproc --ignore=2)" \
-e SCCACHE_BUCKET \
@ -266,6 +240,7 @@ jobs:
-e HUGGING_FACE_HUB_TOKEN \
-e DASHBOARD_TAG \
--env-file="/tmp/github_env_${GITHUB_RUN_ID}" \
--ulimit stack=10485760:83886080 \
--security-opt seccomp=unconfined \
--cap-add=SYS_PTRACE \
--ipc=host \
@ -315,7 +290,6 @@ jobs:
with:
file-suffix: ${{ github.job }}-${{ matrix.config }}-${{ matrix.shard }}-${{ matrix.num_shards }}-${{ matrix.runner }}_${{ steps.get-job-id.outputs.job-id }}
use-gha: ${{ inputs.use-gha }}
s3-bucket: ${{ inputs.s3-bucket }}
- name: Collect backtraces from coredumps (if any)
if: always()
@ -333,7 +307,7 @@ jobs:
path: ./**/core.[1-9]*
- name: Teardown Linux
uses: pytorch/test-infra/.github/actions/teardown-linux@main
uses: pytorch/test-infra/.github/actions/teardown-linux@release/2.3
if: always()
# NB: We are currently having an intermittent GPU-related issue on G5 runners with

View File

@ -71,11 +71,11 @@ jobs:
test-matrix: ${{ steps.filter.outputs.test-matrix }}
steps:
- name: Clean up disk space before running MacOS workflow
uses: pytorch/test-infra/.github/actions/check-disk-space@main
uses: pytorch/test-infra/.github/actions/check-disk-space@release/2.3
# [see note: pytorch repo ref]
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
- name: Set xcode version
env:
@ -87,7 +87,7 @@ jobs:
- name: Setup miniconda
if: inputs.environment-file == ''
uses: pytorch/test-infra/.github/actions/setup-miniconda@main
uses: pytorch/test-infra/.github/actions/setup-miniconda@release/2.3
with:
python-version: ${{ inputs.python-version }}
environment-file: .github/requirements/conda-env-${{ runner.os }}-${{ runner.arch }}
@ -97,7 +97,7 @@ jobs:
# environment even though the arch is x86-64
- name: Setup miniconda using the provided environment file
if: inputs.environment-file != ''
uses: pytorch/test-infra/.github/actions/setup-miniconda@main
uses: pytorch/test-infra/.github/actions/setup-miniconda@release/2.3
with:
python-version: ${{ inputs.python-version }}
environment-file: ${{ inputs.environment-file }}
@ -207,4 +207,4 @@ jobs:
- name: Clean up disk space
if: always()
continue-on-error: true
uses: pytorch/test-infra/.github/actions/check-disk-space@main
uses: pytorch/test-infra/.github/actions/check-disk-space@release/2.3

View File

@ -40,7 +40,7 @@ jobs:
reenabled-issues: ${{ steps.filter.outputs.reenabled-issues }}
steps:
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
with:
submodules: false
@ -81,7 +81,7 @@ jobs:
use-gha: true
- name: Setup miniconda
uses: pytorch/test-infra/.github/actions/setup-miniconda@main
uses: pytorch/test-infra/.github/actions/setup-miniconda@release/2.3
with:
python-version: ${{ inputs.python-version }}
environment-file: .github/requirements/conda-env-${{ runner.os }}-${{ runner.arch }}
@ -159,4 +159,4 @@ jobs:
- name: Clean up disk space
if: always()
continue-on-error: true
uses: pytorch/test-infra/.github/actions/check-disk-space@main
uses: pytorch/test-infra/.github/actions/check-disk-space@release/2.3

View File

@ -79,11 +79,11 @@ jobs:
done
- name: Clean up disk space before running MacOS workflow
uses: pytorch/test-infra/.github/actions/check-disk-space@main
uses: pytorch/test-infra/.github/actions/check-disk-space@release/2.3
# [see note: pytorch repo ref]
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
- name: Download build artifacts
uses: ./.github/actions/download-build-artifacts
@ -98,7 +98,7 @@ jobs:
use-gha: true
- name: Setup miniconda
uses: pytorch/test-infra/.github/actions/setup-miniconda@main
uses: pytorch/test-infra/.github/actions/setup-miniconda@release/2.3
with:
python-version: ${{ inputs.python-version }}
environment-file: .github/requirements/conda-env-${{ runner.os }}-${{ runner.arch }}
@ -227,4 +227,4 @@ jobs:
- name: Clean up disk space
if: always()
continue-on-error: true
uses: pytorch/test-infra/.github/actions/check-disk-space@main
uses: pytorch/test-infra/.github/actions/check-disk-space@release/2.3

View File

@ -58,7 +58,7 @@ jobs:
steps:
# [see note: pytorch repo ref]
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
with:
no-sudo: true
@ -80,12 +80,12 @@ jobs:
- name: Calculate docker image
id: calculate-docker-image
uses: pytorch/test-infra/.github/actions/calculate-docker-image@main
uses: pytorch/test-infra/.github/actions/calculate-docker-image@release/2.3
with:
docker-image-name: ${{ inputs.docker-image }}
- name: Pull docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
uses: pytorch/test-infra/.github/actions/pull-docker-image@release/2.3
with:
docker-image: ${{ steps.calculate-docker-image.outputs.docker-image }}

View File

@ -23,7 +23,7 @@ jobs:
keep-going: ${{ steps.filter.outputs.keep-going }}
steps:
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
with:
fetch-depth: 1
submodules: false
@ -54,10 +54,10 @@ jobs:
SUPPORT_ABI: '${{ matrix.support_abi }}'
steps:
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
- name: Setup miniconda
uses: pytorch/test-infra/.github/actions/setup-miniconda@main
uses: pytorch/test-infra/.github/actions/setup-miniconda@release/2.3
with:
python-version: 3.8
environment-file: .github/requirements/conda-env-${{ runner.os }}-${{ runner.arch }}.txt

View File

@ -60,10 +60,10 @@ jobs:
git config --global core.fsmonitor false
- name: Clean up leftover processes on non-ephemeral Windows runner
uses: pytorch/test-infra/.github/actions/cleanup-runner@main
uses: pytorch/test-infra/.github/actions/cleanup-runner@release/2.3
- name: Setup SSH (Click me for login details)
uses: pytorch/test-infra/.github/actions/setup-ssh@main
uses: pytorch/test-infra/.github/actions/setup-ssh@release/2.3
with:
github-secret: ${{ secrets.GITHUB_TOKEN }}
instructions: |
@ -78,7 +78,7 @@ jobs:
# [see note: pytorch repo ref]
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
with:
no-sudo: true

View File

@ -54,10 +54,10 @@ jobs:
git config --global core.fsmonitor false
- name: Clean up leftover processes on non-ephemeral Windows runner
uses: pytorch/test-infra/.github/actions/cleanup-runner@main
uses: pytorch/test-infra/.github/actions/cleanup-runner@release/2.3
- name: Setup SSH (Click me for login details)
uses: pytorch/test-infra/.github/actions/setup-ssh@main
uses: pytorch/test-infra/.github/actions/setup-ssh@release/2.3
with:
github-secret: ${{ secrets.GITHUB_TOKEN }}
instructions: |
@ -73,7 +73,7 @@ jobs:
# [see note: pytorch repo ref]
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
with:
no-sudo: true
@ -92,7 +92,7 @@ jobs:
retry_wait_seconds: 30
command: |
set -eu
python3 -m pip install rockset==1.0.3 'xdoctest>=1.1.0'
python3 -m pip install rockset==1.0.3
- name: Start monitoring script
id: monitor-script

View File

@ -54,7 +54,7 @@ jobs:
steps:
# [see note: pytorch repo ref]
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
- name: Setup XPU
uses: ./.github/actions/setup-xpu
@ -72,12 +72,12 @@ jobs:
- name: Calculate docker image
id: calculate-docker-image
uses: pytorch/test-infra/.github/actions/calculate-docker-image@main
uses: pytorch/test-infra/.github/actions/calculate-docker-image@release/2.3
with:
docker-image-name: ${{ inputs.docker-image }}
- name: Pull docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
uses: pytorch/test-infra/.github/actions/pull-docker-image@release/2.3
with:
docker-image: ${{ steps.calculate-docker-image.outputs.docker-image }}

View File

@ -3,7 +3,7 @@ name: Build Triton wheels
on:
push:
branches:
- main
- release/2.3
tags:
# NOTE: Binary build pipelines should only get triggered on release candidate builds
# Release candidate tags look like: v1.11.0-rc1
@ -47,12 +47,12 @@ jobs:
BUILD_DEVICE: ${{ matrix.device }}
steps:
- name: Setup SSH (Click me for login details)
uses: pytorch/test-infra/.github/actions/setup-ssh@main
uses: pytorch/test-infra/.github/actions/setup-ssh@release/2.3
with:
github-secret: ${{ secrets.GITHUB_TOKEN }}
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
with:
submodules: false
@ -60,7 +60,7 @@ jobs:
uses: ./.github/actions/setup-linux
- name: Pull Docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
uses: pytorch/test-infra/.github/actions/pull-docker-image@release/2.3
with:
docker-image: ${{ env.DOCKER_IMAGE }}
@ -119,12 +119,13 @@ jobs:
- uses: actions/upload-artifact@v3
with:
name: pytorch-triton-wheel-${{ matrix.py_vers }}-${{ matrix.device }}
# NB: Use the same name here and all wheels can be downloaded by referring to the same artifact
name: pytorch-triton-wheel
if-no-files-found: error
path: ${{ runner.temp }}/artifacts/*
- name: Teardown Linux
uses: pytorch/test-infra/.github/actions/teardown-linux@main
uses: pytorch/test-infra/.github/actions/teardown-linux@release/2.3
if: always()
upload-wheel:
@ -156,15 +157,8 @@ jobs:
- name: Download Build Artifacts
uses: actions/download-artifact@v3
with:
# Download all available artifacts
path: ${{ runner.temp }}/artifacts-all
- name: Select Wheel Artifacts
shell: bash
run: |
set -x
mkdir -p "${RUNNER_TEMP}/artifacts/"
mv "${RUNNER_TEMP}"/artifacts-all/pytorch-triton-wheel-*/* "${RUNNER_TEMP}/artifacts/"
name: pytorch-triton-wheel
path: ${{ runner.temp }}/artifacts/
- name: Set DRY_RUN (only for tagged pushes)
if: ${{ github.event_name == 'push' && (github.event.ref == 'refs/heads/main' || startsWith(github.event.ref, 'refs/tags/v')) }}
@ -209,12 +203,12 @@ jobs:
PY_VERS: ${{ matrix.py_vers }}
steps:
- name: Setup SSH (Click me for login details)
uses: pytorch/test-infra/.github/actions/setup-ssh@main
uses: pytorch/test-infra/.github/actions/setup-ssh@release/2.3
with:
github-secret: ${{ secrets.GITHUB_TOKEN }}
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
with:
submodules: false
@ -222,7 +216,7 @@ jobs:
uses: ./.github/actions/setup-linux
- name: Pull Docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
uses: pytorch/test-infra/.github/actions/pull-docker-image@release/2.3
with:
docker-image: ${{ env.DOCKER_IMAGE }}
@ -252,12 +246,13 @@ jobs:
- uses: actions/upload-artifact@v3
with:
name: pytorch-triton-conda-${{ matrix.py_vers }}
# NB: Use the same name here and all wheels can be downloaded by referring to the same artifact
name: pytorch-triton-conda
if-no-files-found: error
path: ${{ runner.temp }}/artifacts/*
- name: Teardown Linux
uses: pytorch/test-infra/.github/actions/teardown-linux@main
uses: pytorch/test-infra/.github/actions/teardown-linux@release/2.3
if: always()
upload-conda:
@ -272,15 +267,8 @@ jobs:
- name: Download Build Artifacts
uses: actions/download-artifact@v3
with:
# Download all available artifacts
path: ${{ runner.temp }}/artifacts-all
- name: Select Conda Artifacts
shell: bash
run: |
set -x
mkdir -p "${RUNNER_TEMP}/artifacts/"
mv "${RUNNER_TEMP}"/artifacts-all/pytorch-triton-conda-*/* "${RUNNER_TEMP}/artifacts/"
name: pytorch-triton-conda
path: ${{ runner.temp }}/artifacts/
- name: Set DRY_RUN (only for tagged pushes)
if: ${{ github.event_name == 'push' && (github.event.ref == 'refs/heads/main' || startsWith(github.event.ref, 'refs/tags/v')) }}

View File

@ -31,7 +31,7 @@ jobs:
runs-on: linux.20_04.4x
steps:
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
with:
submodules: false
fetch-depth: 1

View File

@ -11,7 +11,7 @@ jobs:
runs-on: ubuntu-latest
steps:
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
- name: Run close_nonexistent_disable_issues.py
env:

View File

@ -74,21 +74,21 @@ jobs:
# [see note: pytorch repo ref]
# deep clone (fetch-depth 0) required for git merge-base
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
- name: Setup Linux
uses: ./.github/actions/setup-linux
- name: Build docker image
id: build-docker-image
uses: pytorch/test-infra/.github/actions/calculate-docker-image@main
uses: pytorch/test-infra/.github/actions/calculate-docker-image@release/2.3
with:
docker-image-name: ${{ matrix.docker-image-name }}
always-rebuild: true
push: true
- name: Pull docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
uses: pytorch/test-infra/.github/actions/pull-docker-image@release/2.3
with:
docker-image: ${{ steps.build-docker-image.outputs.docker-image }}
@ -120,5 +120,5 @@ jobs:
if: always()
- name: Teardown Linux
uses: pytorch/test-infra/.github/actions/teardown-linux@main
uses: pytorch/test-infra/.github/actions/teardown-linux@release/2.3
if: always()

View File

@ -7,6 +7,7 @@ on:
- Dockerfile
- docker.Makefile
- .github/workflows/docker-release.yml
- .github/scripts/generate_docker_release_matrix.py
push:
branches:
- nightly
@ -40,7 +41,7 @@ jobs:
matrix: ${{ steps.generate-matrix.outputs.matrix }}
steps:
- name: Checkout PyTorch
uses: pytorch/pytorch/.github/actions/checkout-pytorch@main
uses: pytorch/pytorch/.github/actions/checkout-pytorch@release/2.3
with:
fetch-depth: 1
submodules: true
@ -68,7 +69,7 @@ jobs:
CUDNN_VERSION: ${{ matrix.cudnn_version }}
steps:
- name: Setup SSH (Click me for login details)
uses: pytorch/test-infra/.github/actions/setup-ssh@main
uses: pytorch/test-infra/.github/actions/setup-ssh@release/2.3
with:
github-secret: ${{ secrets.GITHUB_TOKEN }}
# [see note: pytorch repo ref]
@ -129,17 +130,27 @@ jobs:
if: ${{ github.event.ref == 'refs/heads/nightly' && matrix.image_type == 'runtime' }}
run: |
PYTORCH_DOCKER_TAG="${PYTORCH_VERSION}-cuda${CUDA_VERSION_SHORT}-cudnn${CUDNN_VERSION}-runtime"
CUDA_SUFFIX="-cu${CUDA_VERSION}"
if [[ ${CUDA_VERSION_SHORT} == "cpu" ]]; then
PYTORCH_DOCKER_TAG="${PYTORCH_VERSION}-runtime"
CUDA_SUFFIX=""
fi
PYTORCH_NIGHTLY_COMMIT=$(docker run ghcr.io/pytorch/pytorch-nightly:"${PYTORCH_DOCKER_TAG}" \
python -c 'import torch; print(torch.version.git_version[:7],end="")')
docker tag ghcr.io/pytorch/pytorch-nightly:"${PYTORCH_DOCKER_TAG}" \
ghcr.io/pytorch/pytorch-nightly:"${PYTORCH_NIGHTLY_COMMIT}-cu${CUDA_VERSION}"
docker push ghcr.io/pytorch/pytorch-nightly:"${PYTORCH_NIGHTLY_COMMIT}-cu${CUDA_VERSION}"
ghcr.io/pytorch/pytorch-nightly:"${PYTORCH_NIGHTLY_COMMIT}${CUDA_SUFFIX}"
docker push ghcr.io/pytorch/pytorch-nightly:"${PYTORCH_NIGHTLY_COMMIT}${CUDA_SUFFIX}"
# Please note, here we ned to pin specific verison of CUDA as with latest label
if [[ ${CUDA_VERSION_SHORT} == "12.1" ]]; then
docker tag ghcr.io/pytorch/pytorch-nightly:"${PYTORCH_NIGHTLY_COMMIT}${CUDA_SUFFIX}" \
ghcr.io/pytorch/pytorch-nightly:latest
docker push ghcr.io/pytorch/pytorch-nightly:latest
fi
docker tag ghcr.io/pytorch/pytorch-nightly:"${PYTORCH_NIGHTLY_COMMIT}-cu${CUDA_VERSION}" \
ghcr.io/pytorch/pytorch-nightly:latest
docker push ghcr.io/pytorch/pytorch-nightly:latest
- name: Teardown Linux
uses: pytorch/test-infra/.github/actions/teardown-linux@main
uses: pytorch/test-infra/.github/actions/teardown-linux@release/2.3
if: always()

View File

@ -48,7 +48,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu-aarch64
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-main
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-2.3
DESIRED_PYTHON: "3.8"
runs_on: linux.arm64.2xlarge
ALPINE_IMAGE: "arm64v8/alpine"
@ -69,7 +69,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu-aarch64
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-main
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-2.3
DESIRED_PYTHON: "3.8"
build_name: manywheel-py3_8-cpu-aarch64
build_environment: linux-aarch64-binary-manywheel
@ -91,7 +91,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu-aarch64
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-main
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-2.3
DESIRED_PYTHON: "3.8"
build_name: manywheel-py3_8-cpu-aarch64
secrets:
@ -111,7 +111,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu-aarch64
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-main
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-2.3
DESIRED_PYTHON: "3.9"
runs_on: linux.arm64.2xlarge
ALPINE_IMAGE: "arm64v8/alpine"
@ -132,7 +132,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu-aarch64
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-main
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-2.3
DESIRED_PYTHON: "3.9"
build_name: manywheel-py3_9-cpu-aarch64
build_environment: linux-aarch64-binary-manywheel
@ -154,7 +154,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu-aarch64
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-main
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-2.3
DESIRED_PYTHON: "3.9"
build_name: manywheel-py3_9-cpu-aarch64
secrets:
@ -174,7 +174,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu-aarch64
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-main
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-2.3
DESIRED_PYTHON: "3.10"
runs_on: linux.arm64.2xlarge
ALPINE_IMAGE: "arm64v8/alpine"
@ -195,7 +195,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu-aarch64
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-main
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-2.3
DESIRED_PYTHON: "3.10"
build_name: manywheel-py3_10-cpu-aarch64
build_environment: linux-aarch64-binary-manywheel
@ -217,7 +217,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu-aarch64
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-main
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-2.3
DESIRED_PYTHON: "3.10"
build_name: manywheel-py3_10-cpu-aarch64
secrets:
@ -237,7 +237,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu-aarch64
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-main
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-2.3
DESIRED_PYTHON: "3.11"
runs_on: linux.arm64.2xlarge
ALPINE_IMAGE: "arm64v8/alpine"
@ -258,7 +258,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu-aarch64
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-main
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-2.3
DESIRED_PYTHON: "3.11"
build_name: manywheel-py3_11-cpu-aarch64
build_environment: linux-aarch64-binary-manywheel
@ -280,7 +280,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu-aarch64
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-main
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-2.3
DESIRED_PYTHON: "3.11"
build_name: manywheel-py3_11-cpu-aarch64
secrets:
@ -300,7 +300,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu-aarch64
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-main
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-2.3
DESIRED_PYTHON: "3.12"
runs_on: linux.arm64.2xlarge
ALPINE_IMAGE: "arm64v8/alpine"
@ -321,7 +321,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu-aarch64
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-main
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-2.3
DESIRED_PYTHON: "3.12"
build_name: manywheel-py3_12-cpu-aarch64
build_environment: linux-aarch64-binary-manywheel
@ -343,7 +343,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu-aarch64
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-main
DOCKER_IMAGE: pytorch/manylinuxaarch64-builder:cpu-aarch64-2.3
DESIRED_PYTHON: "3.12"
build_name: manywheel-py3_12-cpu-aarch64
secrets:

View File

@ -48,7 +48,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/conda-builder:cpu-main
DOCKER_IMAGE: pytorch/conda-builder:cpu-2.3
DESIRED_PYTHON: "3.8"
build_name: conda-py3_8-cpu
build_environment: linux-binary-conda
@ -66,7 +66,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/conda-builder:cpu-main
DOCKER_IMAGE: pytorch/conda-builder:cpu-2.3
DESIRED_PYTHON: "3.8"
build_name: conda-py3_8-cpu
build_environment: linux-binary-conda
@ -87,7 +87,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/conda-builder:cpu-main
DOCKER_IMAGE: pytorch/conda-builder:cpu-2.3
DESIRED_PYTHON: "3.8"
build_name: conda-py3_8-cpu
secrets:
@ -108,7 +108,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.8"
runs_on: linux.24xlarge
build_name: conda-py3_8-cuda11_8
@ -128,7 +128,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.8"
build_name: conda-py3_8-cuda11_8
build_environment: linux-binary-conda
@ -150,7 +150,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.8"
build_name: conda-py3_8-cuda11_8
secrets:
@ -171,7 +171,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.8"
runs_on: linux.24xlarge
build_name: conda-py3_8-cuda12_1
@ -191,7 +191,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.8"
build_name: conda-py3_8-cuda12_1
build_environment: linux-binary-conda
@ -213,7 +213,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.8"
build_name: conda-py3_8-cuda12_1
secrets:
@ -233,7 +233,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/conda-builder:cpu-main
DOCKER_IMAGE: pytorch/conda-builder:cpu-2.3
DESIRED_PYTHON: "3.9"
build_name: conda-py3_9-cpu
build_environment: linux-binary-conda
@ -251,7 +251,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/conda-builder:cpu-main
DOCKER_IMAGE: pytorch/conda-builder:cpu-2.3
DESIRED_PYTHON: "3.9"
build_name: conda-py3_9-cpu
build_environment: linux-binary-conda
@ -272,7 +272,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/conda-builder:cpu-main
DOCKER_IMAGE: pytorch/conda-builder:cpu-2.3
DESIRED_PYTHON: "3.9"
build_name: conda-py3_9-cpu
secrets:
@ -293,7 +293,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.9"
runs_on: linux.24xlarge
build_name: conda-py3_9-cuda11_8
@ -313,7 +313,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.9"
build_name: conda-py3_9-cuda11_8
build_environment: linux-binary-conda
@ -335,7 +335,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.9"
build_name: conda-py3_9-cuda11_8
secrets:
@ -356,7 +356,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.9"
runs_on: linux.24xlarge
build_name: conda-py3_9-cuda12_1
@ -376,7 +376,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.9"
build_name: conda-py3_9-cuda12_1
build_environment: linux-binary-conda
@ -398,7 +398,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.9"
build_name: conda-py3_9-cuda12_1
secrets:
@ -418,7 +418,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/conda-builder:cpu-main
DOCKER_IMAGE: pytorch/conda-builder:cpu-2.3
DESIRED_PYTHON: "3.10"
build_name: conda-py3_10-cpu
build_environment: linux-binary-conda
@ -436,7 +436,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/conda-builder:cpu-main
DOCKER_IMAGE: pytorch/conda-builder:cpu-2.3
DESIRED_PYTHON: "3.10"
build_name: conda-py3_10-cpu
build_environment: linux-binary-conda
@ -457,7 +457,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/conda-builder:cpu-main
DOCKER_IMAGE: pytorch/conda-builder:cpu-2.3
DESIRED_PYTHON: "3.10"
build_name: conda-py3_10-cpu
secrets:
@ -478,7 +478,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.10"
runs_on: linux.24xlarge
build_name: conda-py3_10-cuda11_8
@ -498,7 +498,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.10"
build_name: conda-py3_10-cuda11_8
build_environment: linux-binary-conda
@ -520,7 +520,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.10"
build_name: conda-py3_10-cuda11_8
secrets:
@ -541,7 +541,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.10"
runs_on: linux.24xlarge
build_name: conda-py3_10-cuda12_1
@ -561,7 +561,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.10"
build_name: conda-py3_10-cuda12_1
build_environment: linux-binary-conda
@ -583,7 +583,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.10"
build_name: conda-py3_10-cuda12_1
secrets:
@ -603,7 +603,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/conda-builder:cpu-main
DOCKER_IMAGE: pytorch/conda-builder:cpu-2.3
DESIRED_PYTHON: "3.11"
build_name: conda-py3_11-cpu
build_environment: linux-binary-conda
@ -621,7 +621,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/conda-builder:cpu-main
DOCKER_IMAGE: pytorch/conda-builder:cpu-2.3
DESIRED_PYTHON: "3.11"
build_name: conda-py3_11-cpu
build_environment: linux-binary-conda
@ -642,7 +642,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/conda-builder:cpu-main
DOCKER_IMAGE: pytorch/conda-builder:cpu-2.3
DESIRED_PYTHON: "3.11"
build_name: conda-py3_11-cpu
secrets:
@ -663,7 +663,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.11"
runs_on: linux.24xlarge
build_name: conda-py3_11-cuda11_8
@ -683,7 +683,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.11"
build_name: conda-py3_11-cuda11_8
build_environment: linux-binary-conda
@ -705,7 +705,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.11"
build_name: conda-py3_11-cuda11_8
secrets:
@ -726,7 +726,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.11"
runs_on: linux.24xlarge
build_name: conda-py3_11-cuda12_1
@ -746,7 +746,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.11"
build_name: conda-py3_11-cuda12_1
build_environment: linux-binary-conda
@ -768,7 +768,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.11"
build_name: conda-py3_11-cuda12_1
secrets:
@ -788,7 +788,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/conda-builder:cpu-main
DOCKER_IMAGE: pytorch/conda-builder:cpu-2.3
DESIRED_PYTHON: "3.12"
build_name: conda-py3_12-cpu
build_environment: linux-binary-conda
@ -806,7 +806,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/conda-builder:cpu-main
DOCKER_IMAGE: pytorch/conda-builder:cpu-2.3
DESIRED_PYTHON: "3.12"
build_name: conda-py3_12-cpu
build_environment: linux-binary-conda
@ -827,7 +827,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/conda-builder:cpu-main
DOCKER_IMAGE: pytorch/conda-builder:cpu-2.3
DESIRED_PYTHON: "3.12"
build_name: conda-py3_12-cpu
secrets:
@ -848,7 +848,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.12"
runs_on: linux.24xlarge
build_name: conda-py3_12-cuda11_8
@ -868,7 +868,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.12"
build_name: conda-py3_12-cuda11_8
build_environment: linux-binary-conda
@ -890,7 +890,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/conda-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.12"
build_name: conda-py3_12-cuda11_8
secrets:
@ -911,7 +911,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.12"
runs_on: linux.24xlarge
build_name: conda-py3_12-cuda12_1
@ -931,7 +931,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.12"
build_name: conda-py3_12-cuda12_1
build_environment: linux-binary-conda
@ -953,7 +953,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/conda-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.12"
build_name: conda-py3_12-cuda12_1
secrets:

View File

@ -43,7 +43,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:cpu-main
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:cpu-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: cxx11-abi
build_name: libtorch-cpu-shared-with-deps-cxx11-abi
@ -62,7 +62,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:cpu-main
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:cpu-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: cxx11-abi
build_name: libtorch-cpu-shared-with-deps-cxx11-abi

View File

@ -48,7 +48,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:cpu-main
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:cpu-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: cxx11-abi
build_name: libtorch-cpu-shared-with-deps-cxx11-abi
@ -67,7 +67,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:cpu-main
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:cpu-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: cxx11-abi
build_name: libtorch-cpu-shared-with-deps-cxx11-abi
@ -89,7 +89,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:cpu-main
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:cpu-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: cxx11-abi
build_name: libtorch-cpu-shared-with-deps-cxx11-abi
@ -111,7 +111,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:cuda11.8-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: cxx11-abi
build_name: libtorch-cuda11_8-shared-with-deps-cxx11-abi
@ -131,7 +131,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:cuda11.8-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: cxx11-abi
build_name: libtorch-cuda11_8-shared-with-deps-cxx11-abi
@ -154,7 +154,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:cuda11.8-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: cxx11-abi
build_name: libtorch-cuda11_8-shared-with-deps-cxx11-abi
@ -176,7 +176,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:cuda12.1-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: cxx11-abi
build_name: libtorch-cuda12_1-shared-with-deps-cxx11-abi
@ -196,7 +196,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:cuda12.1-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: cxx11-abi
build_name: libtorch-cuda12_1-shared-with-deps-cxx11-abi
@ -219,7 +219,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:cuda12.1-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: cxx11-abi
build_name: libtorch-cuda12_1-shared-with-deps-cxx11-abi
@ -241,7 +241,7 @@ jobs:
DESIRED_CUDA: rocm5.7
GPU_ARCH_VERSION: 5.7
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:rocm5.7-main
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:rocm5.7-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: cxx11-abi
build_name: libtorch-rocm5_7-shared-with-deps-cxx11-abi
@ -263,7 +263,7 @@ jobs:
GPU_ARCH_VERSION: 5.7
GPU_ARCH_TYPE: rocm
SKIP_ALL_TESTS: 1
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:rocm5.7-main
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:rocm5.7-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: cxx11-abi
steps:
@ -277,7 +277,6 @@ jobs:
- name: Checkout PyTorch
uses: malfet/checkout@silent-checkout
with:
ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
submodules: recursive
path: pytorch
quiet-checkout: true
@ -289,7 +288,7 @@ jobs:
- name: Checkout pytorch/builder
uses: malfet/checkout@silent-checkout
with:
ref: main
ref: release/2.3
submodules: recursive
repository: pytorch/builder
path: builder
@ -305,7 +304,7 @@ jobs:
- name: Pull Docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
with:
docker-image: pytorch/libtorch-cxx11-builder:rocm5.7-main
docker-image: pytorch/libtorch-cxx11-builder:rocm5.7-2.3
- name: Test Pytorch binary
uses: ./pytorch/.github/actions/test-pytorch-binary
- name: Teardown ROCm
@ -325,7 +324,7 @@ jobs:
DESIRED_CUDA: rocm5.7
GPU_ARCH_VERSION: 5.7
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:rocm5.7-main
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:rocm5.7-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: cxx11-abi
build_name: libtorch-rocm5_7-shared-with-deps-cxx11-abi
@ -347,7 +346,7 @@ jobs:
DESIRED_CUDA: rocm6.0
GPU_ARCH_VERSION: 6.0
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:rocm6.0-main
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:rocm6.0-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: cxx11-abi
build_name: libtorch-rocm6_0-shared-with-deps-cxx11-abi
@ -369,7 +368,7 @@ jobs:
GPU_ARCH_VERSION: 6.0
GPU_ARCH_TYPE: rocm
SKIP_ALL_TESTS: 1
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:rocm6.0-main
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:rocm6.0-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: cxx11-abi
steps:
@ -383,7 +382,6 @@ jobs:
- name: Checkout PyTorch
uses: malfet/checkout@silent-checkout
with:
ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
submodules: recursive
path: pytorch
quiet-checkout: true
@ -395,7 +393,7 @@ jobs:
- name: Checkout pytorch/builder
uses: malfet/checkout@silent-checkout
with:
ref: main
ref: release/2.3
submodules: recursive
repository: pytorch/builder
path: builder
@ -411,7 +409,7 @@ jobs:
- name: Pull Docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
with:
docker-image: pytorch/libtorch-cxx11-builder:rocm6.0-main
docker-image: pytorch/libtorch-cxx11-builder:rocm6.0-2.3
- name: Test Pytorch binary
uses: ./pytorch/.github/actions/test-pytorch-binary
- name: Teardown ROCm
@ -431,7 +429,7 @@ jobs:
DESIRED_CUDA: rocm6.0
GPU_ARCH_VERSION: 6.0
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:rocm6.0-main
DOCKER_IMAGE: pytorch/libtorch-cxx11-builder:rocm6.0-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: cxx11-abi
build_name: libtorch-rocm6_0-shared-with-deps-cxx11-abi

View File

@ -43,7 +43,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-main
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: pre-cxx11
build_name: libtorch-cpu-shared-with-deps-pre-cxx11
@ -62,7 +62,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-main
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: pre-cxx11
build_name: libtorch-cpu-shared-with-deps-pre-cxx11

View File

@ -48,7 +48,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-main
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: pre-cxx11
build_name: libtorch-cpu-shared-with-deps-pre-cxx11
@ -67,7 +67,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-main
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: pre-cxx11
build_name: libtorch-cpu-shared-with-deps-pre-cxx11
@ -89,7 +89,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-main
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: pre-cxx11
build_name: libtorch-cpu-shared-with-deps-pre-cxx11
@ -111,7 +111,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: pre-cxx11
build_name: libtorch-cuda11_8-shared-with-deps-pre-cxx11
@ -131,7 +131,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: pre-cxx11
build_name: libtorch-cuda11_8-shared-with-deps-pre-cxx11
@ -154,7 +154,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: pre-cxx11
build_name: libtorch-cuda11_8-shared-with-deps-pre-cxx11
@ -176,7 +176,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: pre-cxx11
build_name: libtorch-cuda12_1-shared-with-deps-pre-cxx11
@ -196,7 +196,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: pre-cxx11
build_name: libtorch-cuda12_1-shared-with-deps-pre-cxx11
@ -219,7 +219,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: pre-cxx11
build_name: libtorch-cuda12_1-shared-with-deps-pre-cxx11
@ -241,7 +241,7 @@ jobs:
DESIRED_CUDA: rocm5.7
GPU_ARCH_VERSION: 5.7
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: pre-cxx11
build_name: libtorch-rocm5_7-shared-with-deps-pre-cxx11
@ -263,7 +263,7 @@ jobs:
GPU_ARCH_VERSION: 5.7
GPU_ARCH_TYPE: rocm
SKIP_ALL_TESTS: 1
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: pre-cxx11
steps:
@ -277,7 +277,6 @@ jobs:
- name: Checkout PyTorch
uses: malfet/checkout@silent-checkout
with:
ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
submodules: recursive
path: pytorch
quiet-checkout: true
@ -289,7 +288,7 @@ jobs:
- name: Checkout pytorch/builder
uses: malfet/checkout@silent-checkout
with:
ref: main
ref: release/2.3
submodules: recursive
repository: pytorch/builder
path: builder
@ -305,7 +304,7 @@ jobs:
- name: Pull Docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
with:
docker-image: pytorch/manylinux-builder:rocm5.7-main
docker-image: pytorch/manylinux-builder:rocm5.7-2.3
- name: Test Pytorch binary
uses: ./pytorch/.github/actions/test-pytorch-binary
- name: Teardown ROCm
@ -325,7 +324,7 @@ jobs:
DESIRED_CUDA: rocm5.7
GPU_ARCH_VERSION: 5.7
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: pre-cxx11
build_name: libtorch-rocm5_7-shared-with-deps-pre-cxx11
@ -347,7 +346,7 @@ jobs:
DESIRED_CUDA: rocm6.0
GPU_ARCH_VERSION: 6.0
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: pre-cxx11
build_name: libtorch-rocm6_0-shared-with-deps-pre-cxx11
@ -369,7 +368,7 @@ jobs:
GPU_ARCH_VERSION: 6.0
GPU_ARCH_TYPE: rocm
SKIP_ALL_TESTS: 1
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: pre-cxx11
steps:
@ -383,7 +382,6 @@ jobs:
- name: Checkout PyTorch
uses: malfet/checkout@silent-checkout
with:
ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
submodules: recursive
path: pytorch
quiet-checkout: true
@ -395,7 +393,7 @@ jobs:
- name: Checkout pytorch/builder
uses: malfet/checkout@silent-checkout
with:
ref: main
ref: release/2.3
submodules: recursive
repository: pytorch/builder
path: builder
@ -411,7 +409,7 @@ jobs:
- name: Pull Docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
with:
docker-image: pytorch/manylinux-builder:rocm6.0-main
docker-image: pytorch/manylinux-builder:rocm6.0-2.3
- name: Test Pytorch binary
uses: ./pytorch/.github/actions/test-pytorch-binary
- name: Teardown ROCm
@ -431,7 +429,7 @@ jobs:
DESIRED_CUDA: rocm6.0
GPU_ARCH_VERSION: 6.0
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-2.3
LIBTORCH_VARIANT: shared-with-deps
DESIRED_DEVTOOLSET: pre-cxx11
build_name: libtorch-rocm6_0-shared-with-deps-pre-cxx11

View File

@ -44,7 +44,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.8"
build_name: manywheel-py3_8-cuda11_8
build_environment: linux-binary-manywheel
@ -64,7 +64,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.8"
build_name: manywheel-py3_8-cuda11_8
build_environment: linux-binary-manywheel
@ -84,7 +84,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.8"
build_name: manywheel-py3_8-cuda12_1
build_environment: linux-binary-manywheel
@ -104,7 +104,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.8"
build_name: manywheel-py3_8-cuda12_1
build_environment: linux-binary-manywheel

View File

@ -48,7 +48,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-main
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-2.3
DESIRED_PYTHON: "3.8"
build_name: manywheel-py3_8-cpu
build_environment: linux-binary-manywheel
@ -66,7 +66,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-main
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-2.3
DESIRED_PYTHON: "3.8"
build_name: manywheel-py3_8-cpu
build_environment: linux-binary-manywheel
@ -87,7 +87,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-main
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-2.3
DESIRED_PYTHON: "3.8"
build_name: manywheel-py3_8-cpu
secrets:
@ -107,7 +107,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu-cxx11-abi
GPU_ARCH_TYPE: cpu-cxx11-abi
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-main
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-2.3
DESIRED_DEVTOOLSET: cxx11-abi
DESIRED_PYTHON: "3.8"
build_name: manywheel-py3_8-cpu-cxx11-abi
@ -126,7 +126,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu-cxx11-abi
GPU_ARCH_TYPE: cpu-cxx11-abi
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-main
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-2.3
DESIRED_DEVTOOLSET: cxx11-abi
DESIRED_PYTHON: "3.8"
build_name: manywheel-py3_8-cpu-cxx11-abi
@ -148,7 +148,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu-cxx11-abi
GPU_ARCH_TYPE: cpu-cxx11-abi
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-main
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-2.3
DESIRED_DEVTOOLSET: cxx11-abi
DESIRED_PYTHON: "3.8"
build_name: manywheel-py3_8-cpu-cxx11-abi
@ -170,7 +170,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.8"
build_name: manywheel-py3_8-cuda11_8
build_environment: linux-binary-manywheel
@ -190,7 +190,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.8"
build_name: manywheel-py3_8-cuda11_8
build_environment: linux-binary-manywheel
@ -212,7 +212,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.8"
build_name: manywheel-py3_8-cuda11_8
secrets:
@ -233,7 +233,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.8"
build_name: manywheel-py3_8-cuda12_1
build_environment: linux-binary-manywheel
@ -253,7 +253,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.8"
build_name: manywheel-py3_8-cuda12_1
build_environment: linux-binary-manywheel
@ -275,7 +275,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.8"
build_name: manywheel-py3_8-cuda12_1
secrets:
@ -296,7 +296,7 @@ jobs:
DESIRED_CUDA: rocm5.7
GPU_ARCH_VERSION: 5.7
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-2.3
DESIRED_PYTHON: "3.8"
build_name: manywheel-py3_8-rocm5_7
build_environment: linux-binary-manywheel
@ -317,7 +317,7 @@ jobs:
GPU_ARCH_VERSION: 5.7
GPU_ARCH_TYPE: rocm
SKIP_ALL_TESTS: 1
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-2.3
DESIRED_PYTHON: "3.8"
steps:
- name: Setup ROCm
@ -330,7 +330,6 @@ jobs:
- name: Checkout PyTorch
uses: malfet/checkout@silent-checkout
with:
ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
submodules: recursive
path: pytorch
quiet-checkout: true
@ -342,7 +341,7 @@ jobs:
- name: Checkout pytorch/builder
uses: malfet/checkout@silent-checkout
with:
ref: main
ref: release/2.3
submodules: recursive
repository: pytorch/builder
path: builder
@ -358,7 +357,7 @@ jobs:
- name: Pull Docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
with:
docker-image: pytorch/manylinux-builder:rocm5.7-main
docker-image: pytorch/manylinux-builder:rocm5.7-2.3
- name: Test Pytorch binary
uses: ./pytorch/.github/actions/test-pytorch-binary
- name: Teardown ROCm
@ -378,7 +377,7 @@ jobs:
DESIRED_CUDA: rocm5.7
GPU_ARCH_VERSION: 5.7
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-2.3
DESIRED_PYTHON: "3.8"
build_name: manywheel-py3_8-rocm5_7
secrets:
@ -399,7 +398,7 @@ jobs:
DESIRED_CUDA: rocm6.0
GPU_ARCH_VERSION: 6.0
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-2.3
DESIRED_PYTHON: "3.8"
build_name: manywheel-py3_8-rocm6_0
build_environment: linux-binary-manywheel
@ -420,7 +419,7 @@ jobs:
GPU_ARCH_VERSION: 6.0
GPU_ARCH_TYPE: rocm
SKIP_ALL_TESTS: 1
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-2.3
DESIRED_PYTHON: "3.8"
steps:
- name: Setup ROCm
@ -433,7 +432,6 @@ jobs:
- name: Checkout PyTorch
uses: malfet/checkout@silent-checkout
with:
ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
submodules: recursive
path: pytorch
quiet-checkout: true
@ -445,7 +443,7 @@ jobs:
- name: Checkout pytorch/builder
uses: malfet/checkout@silent-checkout
with:
ref: main
ref: release/2.3
submodules: recursive
repository: pytorch/builder
path: builder
@ -461,7 +459,7 @@ jobs:
- name: Pull Docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
with:
docker-image: pytorch/manylinux-builder:rocm6.0-main
docker-image: pytorch/manylinux-builder:rocm6.0-2.3
- name: Test Pytorch binary
uses: ./pytorch/.github/actions/test-pytorch-binary
- name: Teardown ROCm
@ -481,7 +479,7 @@ jobs:
DESIRED_CUDA: rocm6.0
GPU_ARCH_VERSION: 6.0
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-2.3
DESIRED_PYTHON: "3.8"
build_name: manywheel-py3_8-rocm6_0
secrets:
@ -501,7 +499,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-main
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-2.3
DESIRED_PYTHON: "3.9"
build_name: manywheel-py3_9-cpu
build_environment: linux-binary-manywheel
@ -519,7 +517,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-main
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-2.3
DESIRED_PYTHON: "3.9"
build_name: manywheel-py3_9-cpu
build_environment: linux-binary-manywheel
@ -540,7 +538,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-main
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-2.3
DESIRED_PYTHON: "3.9"
build_name: manywheel-py3_9-cpu
secrets:
@ -560,7 +558,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu-cxx11-abi
GPU_ARCH_TYPE: cpu-cxx11-abi
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-main
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-2.3
DESIRED_DEVTOOLSET: cxx11-abi
DESIRED_PYTHON: "3.9"
build_name: manywheel-py3_9-cpu-cxx11-abi
@ -579,7 +577,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu-cxx11-abi
GPU_ARCH_TYPE: cpu-cxx11-abi
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-main
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-2.3
DESIRED_DEVTOOLSET: cxx11-abi
DESIRED_PYTHON: "3.9"
build_name: manywheel-py3_9-cpu-cxx11-abi
@ -601,7 +599,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu-cxx11-abi
GPU_ARCH_TYPE: cpu-cxx11-abi
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-main
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-2.3
DESIRED_DEVTOOLSET: cxx11-abi
DESIRED_PYTHON: "3.9"
build_name: manywheel-py3_9-cpu-cxx11-abi
@ -623,7 +621,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.9"
build_name: manywheel-py3_9-cuda11_8
build_environment: linux-binary-manywheel
@ -643,7 +641,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.9"
build_name: manywheel-py3_9-cuda11_8
build_environment: linux-binary-manywheel
@ -665,7 +663,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.9"
build_name: manywheel-py3_9-cuda11_8
secrets:
@ -686,7 +684,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.9"
build_name: manywheel-py3_9-cuda12_1
build_environment: linux-binary-manywheel
@ -706,7 +704,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.9"
build_name: manywheel-py3_9-cuda12_1
build_environment: linux-binary-manywheel
@ -728,7 +726,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.9"
build_name: manywheel-py3_9-cuda12_1
secrets:
@ -749,7 +747,7 @@ jobs:
DESIRED_CUDA: rocm5.7
GPU_ARCH_VERSION: 5.7
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-2.3
DESIRED_PYTHON: "3.9"
build_name: manywheel-py3_9-rocm5_7
build_environment: linux-binary-manywheel
@ -770,7 +768,7 @@ jobs:
GPU_ARCH_VERSION: 5.7
GPU_ARCH_TYPE: rocm
SKIP_ALL_TESTS: 1
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-2.3
DESIRED_PYTHON: "3.9"
steps:
- name: Setup ROCm
@ -783,7 +781,6 @@ jobs:
- name: Checkout PyTorch
uses: malfet/checkout@silent-checkout
with:
ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
submodules: recursive
path: pytorch
quiet-checkout: true
@ -795,7 +792,7 @@ jobs:
- name: Checkout pytorch/builder
uses: malfet/checkout@silent-checkout
with:
ref: main
ref: release/2.3
submodules: recursive
repository: pytorch/builder
path: builder
@ -811,7 +808,7 @@ jobs:
- name: Pull Docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
with:
docker-image: pytorch/manylinux-builder:rocm5.7-main
docker-image: pytorch/manylinux-builder:rocm5.7-2.3
- name: Test Pytorch binary
uses: ./pytorch/.github/actions/test-pytorch-binary
- name: Teardown ROCm
@ -831,7 +828,7 @@ jobs:
DESIRED_CUDA: rocm5.7
GPU_ARCH_VERSION: 5.7
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-2.3
DESIRED_PYTHON: "3.9"
build_name: manywheel-py3_9-rocm5_7
secrets:
@ -852,7 +849,7 @@ jobs:
DESIRED_CUDA: rocm6.0
GPU_ARCH_VERSION: 6.0
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-2.3
DESIRED_PYTHON: "3.9"
build_name: manywheel-py3_9-rocm6_0
build_environment: linux-binary-manywheel
@ -873,7 +870,7 @@ jobs:
GPU_ARCH_VERSION: 6.0
GPU_ARCH_TYPE: rocm
SKIP_ALL_TESTS: 1
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-2.3
DESIRED_PYTHON: "3.9"
steps:
- name: Setup ROCm
@ -886,7 +883,6 @@ jobs:
- name: Checkout PyTorch
uses: malfet/checkout@silent-checkout
with:
ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
submodules: recursive
path: pytorch
quiet-checkout: true
@ -898,7 +894,7 @@ jobs:
- name: Checkout pytorch/builder
uses: malfet/checkout@silent-checkout
with:
ref: main
ref: release/2.3
submodules: recursive
repository: pytorch/builder
path: builder
@ -914,7 +910,7 @@ jobs:
- name: Pull Docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
with:
docker-image: pytorch/manylinux-builder:rocm6.0-main
docker-image: pytorch/manylinux-builder:rocm6.0-2.3
- name: Test Pytorch binary
uses: ./pytorch/.github/actions/test-pytorch-binary
- name: Teardown ROCm
@ -934,7 +930,7 @@ jobs:
DESIRED_CUDA: rocm6.0
GPU_ARCH_VERSION: 6.0
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-2.3
DESIRED_PYTHON: "3.9"
build_name: manywheel-py3_9-rocm6_0
secrets:
@ -954,7 +950,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-main
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-2.3
DESIRED_PYTHON: "3.10"
build_name: manywheel-py3_10-cpu
build_environment: linux-binary-manywheel
@ -972,7 +968,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-main
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-2.3
DESIRED_PYTHON: "3.10"
build_name: manywheel-py3_10-cpu
build_environment: linux-binary-manywheel
@ -993,7 +989,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-main
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-2.3
DESIRED_PYTHON: "3.10"
build_name: manywheel-py3_10-cpu
secrets:
@ -1013,7 +1009,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu-cxx11-abi
GPU_ARCH_TYPE: cpu-cxx11-abi
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-main
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-2.3
DESIRED_DEVTOOLSET: cxx11-abi
DESIRED_PYTHON: "3.10"
build_name: manywheel-py3_10-cpu-cxx11-abi
@ -1032,7 +1028,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu-cxx11-abi
GPU_ARCH_TYPE: cpu-cxx11-abi
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-main
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-2.3
DESIRED_DEVTOOLSET: cxx11-abi
DESIRED_PYTHON: "3.10"
build_name: manywheel-py3_10-cpu-cxx11-abi
@ -1054,7 +1050,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu-cxx11-abi
GPU_ARCH_TYPE: cpu-cxx11-abi
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-main
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-2.3
DESIRED_DEVTOOLSET: cxx11-abi
DESIRED_PYTHON: "3.10"
build_name: manywheel-py3_10-cpu-cxx11-abi
@ -1076,7 +1072,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.10"
build_name: manywheel-py3_10-cuda11_8
build_environment: linux-binary-manywheel
@ -1096,7 +1092,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.10"
build_name: manywheel-py3_10-cuda11_8
build_environment: linux-binary-manywheel
@ -1118,7 +1114,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.10"
build_name: manywheel-py3_10-cuda11_8
secrets:
@ -1139,7 +1135,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.10"
build_name: manywheel-py3_10-cuda12_1
build_environment: linux-binary-manywheel
@ -1159,7 +1155,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.10"
build_name: manywheel-py3_10-cuda12_1
build_environment: linux-binary-manywheel
@ -1181,7 +1177,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.10"
build_name: manywheel-py3_10-cuda12_1
secrets:
@ -1202,7 +1198,7 @@ jobs:
DESIRED_CUDA: rocm5.7
GPU_ARCH_VERSION: 5.7
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-2.3
DESIRED_PYTHON: "3.10"
build_name: manywheel-py3_10-rocm5_7
build_environment: linux-binary-manywheel
@ -1223,7 +1219,7 @@ jobs:
GPU_ARCH_VERSION: 5.7
GPU_ARCH_TYPE: rocm
SKIP_ALL_TESTS: 1
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-2.3
DESIRED_PYTHON: "3.10"
steps:
- name: Setup ROCm
@ -1236,7 +1232,6 @@ jobs:
- name: Checkout PyTorch
uses: malfet/checkout@silent-checkout
with:
ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
submodules: recursive
path: pytorch
quiet-checkout: true
@ -1248,7 +1243,7 @@ jobs:
- name: Checkout pytorch/builder
uses: malfet/checkout@silent-checkout
with:
ref: main
ref: release/2.3
submodules: recursive
repository: pytorch/builder
path: builder
@ -1264,7 +1259,7 @@ jobs:
- name: Pull Docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
with:
docker-image: pytorch/manylinux-builder:rocm5.7-main
docker-image: pytorch/manylinux-builder:rocm5.7-2.3
- name: Test Pytorch binary
uses: ./pytorch/.github/actions/test-pytorch-binary
- name: Teardown ROCm
@ -1284,7 +1279,7 @@ jobs:
DESIRED_CUDA: rocm5.7
GPU_ARCH_VERSION: 5.7
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-2.3
DESIRED_PYTHON: "3.10"
build_name: manywheel-py3_10-rocm5_7
secrets:
@ -1305,7 +1300,7 @@ jobs:
DESIRED_CUDA: rocm6.0
GPU_ARCH_VERSION: 6.0
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-2.3
DESIRED_PYTHON: "3.10"
build_name: manywheel-py3_10-rocm6_0
build_environment: linux-binary-manywheel
@ -1326,7 +1321,7 @@ jobs:
GPU_ARCH_VERSION: 6.0
GPU_ARCH_TYPE: rocm
SKIP_ALL_TESTS: 1
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-2.3
DESIRED_PYTHON: "3.10"
steps:
- name: Setup ROCm
@ -1339,7 +1334,6 @@ jobs:
- name: Checkout PyTorch
uses: malfet/checkout@silent-checkout
with:
ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
submodules: recursive
path: pytorch
quiet-checkout: true
@ -1351,7 +1345,7 @@ jobs:
- name: Checkout pytorch/builder
uses: malfet/checkout@silent-checkout
with:
ref: main
ref: release/2.3
submodules: recursive
repository: pytorch/builder
path: builder
@ -1367,7 +1361,7 @@ jobs:
- name: Pull Docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
with:
docker-image: pytorch/manylinux-builder:rocm6.0-main
docker-image: pytorch/manylinux-builder:rocm6.0-2.3
- name: Test Pytorch binary
uses: ./pytorch/.github/actions/test-pytorch-binary
- name: Teardown ROCm
@ -1387,7 +1381,7 @@ jobs:
DESIRED_CUDA: rocm6.0
GPU_ARCH_VERSION: 6.0
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-2.3
DESIRED_PYTHON: "3.10"
build_name: manywheel-py3_10-rocm6_0
secrets:
@ -1407,7 +1401,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-main
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-2.3
DESIRED_PYTHON: "3.11"
build_name: manywheel-py3_11-cpu
build_environment: linux-binary-manywheel
@ -1425,7 +1419,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-main
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-2.3
DESIRED_PYTHON: "3.11"
build_name: manywheel-py3_11-cpu
build_environment: linux-binary-manywheel
@ -1446,7 +1440,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-main
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-2.3
DESIRED_PYTHON: "3.11"
build_name: manywheel-py3_11-cpu
secrets:
@ -1466,7 +1460,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu-cxx11-abi
GPU_ARCH_TYPE: cpu-cxx11-abi
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-main
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-2.3
DESIRED_DEVTOOLSET: cxx11-abi
DESIRED_PYTHON: "3.11"
build_name: manywheel-py3_11-cpu-cxx11-abi
@ -1485,7 +1479,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu-cxx11-abi
GPU_ARCH_TYPE: cpu-cxx11-abi
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-main
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-2.3
DESIRED_DEVTOOLSET: cxx11-abi
DESIRED_PYTHON: "3.11"
build_name: manywheel-py3_11-cpu-cxx11-abi
@ -1507,7 +1501,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu-cxx11-abi
GPU_ARCH_TYPE: cpu-cxx11-abi
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-main
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-2.3
DESIRED_DEVTOOLSET: cxx11-abi
DESIRED_PYTHON: "3.11"
build_name: manywheel-py3_11-cpu-cxx11-abi
@ -1529,7 +1523,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.11"
build_name: manywheel-py3_11-cuda11_8
build_environment: linux-binary-manywheel
@ -1549,7 +1543,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.11"
build_name: manywheel-py3_11-cuda11_8
build_environment: linux-binary-manywheel
@ -1571,7 +1565,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.11"
build_name: manywheel-py3_11-cuda11_8
secrets:
@ -1592,7 +1586,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.11"
build_name: manywheel-py3_11-cuda12_1
build_environment: linux-binary-manywheel
@ -1612,7 +1606,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.11"
build_name: manywheel-py3_11-cuda12_1
build_environment: linux-binary-manywheel
@ -1634,7 +1628,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.11"
build_name: manywheel-py3_11-cuda12_1
secrets:
@ -1655,7 +1649,7 @@ jobs:
DESIRED_CUDA: rocm5.7
GPU_ARCH_VERSION: 5.7
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-2.3
DESIRED_PYTHON: "3.11"
build_name: manywheel-py3_11-rocm5_7
build_environment: linux-binary-manywheel
@ -1676,7 +1670,7 @@ jobs:
GPU_ARCH_VERSION: 5.7
GPU_ARCH_TYPE: rocm
SKIP_ALL_TESTS: 1
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-2.3
DESIRED_PYTHON: "3.11"
steps:
- name: Setup ROCm
@ -1689,7 +1683,6 @@ jobs:
- name: Checkout PyTorch
uses: malfet/checkout@silent-checkout
with:
ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
submodules: recursive
path: pytorch
quiet-checkout: true
@ -1701,7 +1694,7 @@ jobs:
- name: Checkout pytorch/builder
uses: malfet/checkout@silent-checkout
with:
ref: main
ref: release/2.3
submodules: recursive
repository: pytorch/builder
path: builder
@ -1717,7 +1710,7 @@ jobs:
- name: Pull Docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
with:
docker-image: pytorch/manylinux-builder:rocm5.7-main
docker-image: pytorch/manylinux-builder:rocm5.7-2.3
- name: Test Pytorch binary
uses: ./pytorch/.github/actions/test-pytorch-binary
- name: Teardown ROCm
@ -1737,7 +1730,7 @@ jobs:
DESIRED_CUDA: rocm5.7
GPU_ARCH_VERSION: 5.7
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-2.3
DESIRED_PYTHON: "3.11"
build_name: manywheel-py3_11-rocm5_7
secrets:
@ -1758,7 +1751,7 @@ jobs:
DESIRED_CUDA: rocm6.0
GPU_ARCH_VERSION: 6.0
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-2.3
DESIRED_PYTHON: "3.11"
build_name: manywheel-py3_11-rocm6_0
build_environment: linux-binary-manywheel
@ -1779,7 +1772,7 @@ jobs:
GPU_ARCH_VERSION: 6.0
GPU_ARCH_TYPE: rocm
SKIP_ALL_TESTS: 1
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-2.3
DESIRED_PYTHON: "3.11"
steps:
- name: Setup ROCm
@ -1792,7 +1785,6 @@ jobs:
- name: Checkout PyTorch
uses: malfet/checkout@silent-checkout
with:
ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
submodules: recursive
path: pytorch
quiet-checkout: true
@ -1804,7 +1796,7 @@ jobs:
- name: Checkout pytorch/builder
uses: malfet/checkout@silent-checkout
with:
ref: main
ref: release/2.3
submodules: recursive
repository: pytorch/builder
path: builder
@ -1820,7 +1812,7 @@ jobs:
- name: Pull Docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
with:
docker-image: pytorch/manylinux-builder:rocm6.0-main
docker-image: pytorch/manylinux-builder:rocm6.0-2.3
- name: Test Pytorch binary
uses: ./pytorch/.github/actions/test-pytorch-binary
- name: Teardown ROCm
@ -1840,7 +1832,7 @@ jobs:
DESIRED_CUDA: rocm6.0
GPU_ARCH_VERSION: 6.0
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-2.3
DESIRED_PYTHON: "3.11"
build_name: manywheel-py3_11-rocm6_0
secrets:
@ -1860,7 +1852,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-main
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-2.3
DESIRED_PYTHON: "3.12"
build_name: manywheel-py3_12-cpu
build_environment: linux-binary-manywheel
@ -1878,7 +1870,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-main
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-2.3
DESIRED_PYTHON: "3.12"
build_name: manywheel-py3_12-cpu
build_environment: linux-binary-manywheel
@ -1899,7 +1891,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu
GPU_ARCH_TYPE: cpu
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-main
DOCKER_IMAGE: pytorch/manylinux-builder:cpu-2.3
DESIRED_PYTHON: "3.12"
build_name: manywheel-py3_12-cpu
secrets:
@ -1919,7 +1911,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu-cxx11-abi
GPU_ARCH_TYPE: cpu-cxx11-abi
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-main
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-2.3
DESIRED_DEVTOOLSET: cxx11-abi
DESIRED_PYTHON: "3.12"
build_name: manywheel-py3_12-cpu-cxx11-abi
@ -1938,7 +1930,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu-cxx11-abi
GPU_ARCH_TYPE: cpu-cxx11-abi
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-main
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-2.3
DESIRED_DEVTOOLSET: cxx11-abi
DESIRED_PYTHON: "3.12"
build_name: manywheel-py3_12-cpu-cxx11-abi
@ -1960,7 +1952,7 @@ jobs:
# favor of GPU_ARCH_VERSION
DESIRED_CUDA: cpu-cxx11-abi
GPU_ARCH_TYPE: cpu-cxx11-abi
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-main
DOCKER_IMAGE: pytorch/manylinuxcxx11-abi-builder:cpu-cxx11-abi-2.3
DESIRED_DEVTOOLSET: cxx11-abi
DESIRED_PYTHON: "3.12"
build_name: manywheel-py3_12-cpu-cxx11-abi
@ -1982,7 +1974,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.12"
build_name: manywheel-py3_12-cuda11_8
build_environment: linux-binary-manywheel
@ -2002,7 +1994,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.12"
build_name: manywheel-py3_12-cuda11_8
build_environment: linux-binary-manywheel
@ -2024,7 +2016,7 @@ jobs:
DESIRED_CUDA: cu118
GPU_ARCH_VERSION: 11.8
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda11.8-2.3
DESIRED_PYTHON: "3.12"
build_name: manywheel-py3_12-cuda11_8
secrets:
@ -2045,7 +2037,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.12"
build_name: manywheel-py3_12-cuda12_1
build_environment: linux-binary-manywheel
@ -2065,7 +2057,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.12"
build_name: manywheel-py3_12-cuda12_1
build_environment: linux-binary-manywheel
@ -2087,7 +2079,7 @@ jobs:
DESIRED_CUDA: cu121
GPU_ARCH_VERSION: 12.1
GPU_ARCH_TYPE: cuda
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-main
DOCKER_IMAGE: pytorch/manylinux-builder:cuda12.1-2.3
DESIRED_PYTHON: "3.12"
build_name: manywheel-py3_12-cuda12_1
secrets:
@ -2108,7 +2100,7 @@ jobs:
DESIRED_CUDA: rocm5.7
GPU_ARCH_VERSION: 5.7
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-2.3
DESIRED_PYTHON: "3.12"
build_name: manywheel-py3_12-rocm5_7
build_environment: linux-binary-manywheel
@ -2129,7 +2121,7 @@ jobs:
GPU_ARCH_VERSION: 5.7
GPU_ARCH_TYPE: rocm
SKIP_ALL_TESTS: 1
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-2.3
DESIRED_PYTHON: "3.12"
steps:
- name: Setup ROCm
@ -2142,7 +2134,6 @@ jobs:
- name: Checkout PyTorch
uses: malfet/checkout@silent-checkout
with:
ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
submodules: recursive
path: pytorch
quiet-checkout: true
@ -2154,7 +2145,7 @@ jobs:
- name: Checkout pytorch/builder
uses: malfet/checkout@silent-checkout
with:
ref: main
ref: release/2.3
submodules: recursive
repository: pytorch/builder
path: builder
@ -2170,7 +2161,7 @@ jobs:
- name: Pull Docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
with:
docker-image: pytorch/manylinux-builder:rocm5.7-main
docker-image: pytorch/manylinux-builder:rocm5.7-2.3
- name: Test Pytorch binary
uses: ./pytorch/.github/actions/test-pytorch-binary
- name: Teardown ROCm
@ -2190,7 +2181,7 @@ jobs:
DESIRED_CUDA: rocm5.7
GPU_ARCH_VERSION: 5.7
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm5.7-2.3
DESIRED_PYTHON: "3.12"
build_name: manywheel-py3_12-rocm5_7
secrets:
@ -2211,7 +2202,7 @@ jobs:
DESIRED_CUDA: rocm6.0
GPU_ARCH_VERSION: 6.0
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-2.3
DESIRED_PYTHON: "3.12"
build_name: manywheel-py3_12-rocm6_0
build_environment: linux-binary-manywheel
@ -2232,7 +2223,7 @@ jobs:
GPU_ARCH_VERSION: 6.0
GPU_ARCH_TYPE: rocm
SKIP_ALL_TESTS: 1
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-2.3
DESIRED_PYTHON: "3.12"
steps:
- name: Setup ROCm
@ -2245,7 +2236,6 @@ jobs:
- name: Checkout PyTorch
uses: malfet/checkout@silent-checkout
with:
ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
submodules: recursive
path: pytorch
quiet-checkout: true
@ -2257,7 +2247,7 @@ jobs:
- name: Checkout pytorch/builder
uses: malfet/checkout@silent-checkout
with:
ref: main
ref: release/2.3
submodules: recursive
repository: pytorch/builder
path: builder
@ -2273,7 +2263,7 @@ jobs:
- name: Pull Docker image
uses: pytorch/test-infra/.github/actions/pull-docker-image@main
with:
docker-image: pytorch/manylinux-builder:rocm6.0-main
docker-image: pytorch/manylinux-builder:rocm6.0-2.3
- name: Test Pytorch binary
uses: ./pytorch/.github/actions/test-pytorch-binary
- name: Teardown ROCm
@ -2293,7 +2283,7 @@ jobs:
DESIRED_CUDA: rocm6.0
GPU_ARCH_VERSION: 6.0
GPU_ARCH_TYPE: rocm
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-main
DOCKER_IMAGE: pytorch/manylinux-builder:rocm6.0-2.3
DESIRED_PYTHON: "3.12"
build_name: manywheel-py3_12-rocm6_0
secrets:

Some files were not shown because too many files have changed in this diff Show More