frozenleaves/pytorch
1
0
Fork 0
mirror of https://github.com/pytorch/pytorch.git synced 2025-10-20 12:54:11 +08:00
Code Issues Packages Projects Releases Wiki Activity
94,664 Commits 5,406 Branches 1,221 Tags
c6a8db0b9acbefc66f02e7ff46ad6bbedabd8b4b
Commit Graph

4478 Commits

Author SHA1 Message Date
drisspg
fe80f03726 Add B200 files to labeler and update codeowners (#165767) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/165767
Approved by: https://github.com/slayton58
 2025-10-17 23:24:17 +00:00
Nikita Shulga
6ece527fc5 [CI] Add aarch64 operator benchmark (#165585) 
Running on Graviton4
Skip ConvTranspose1d benchmarks if PyTorch is compiled with ACL, due to https://github.com/pytorch/pytorch/issues/165654
Pull Request resolved: https://github.com/pytorch/pytorch/pull/165585
Approved by: https://github.com/huydhn
 2025-10-17 14:42:14 +00:00
Yuanyuan Chen
e925dfcc6b Enable all SIM rules except disabled ones (#164645) 
`SIM` rules are useful for simplifying boolean expressions and enhances code readability.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/164645
Approved by: https://github.com/ezyang, https://github.com/mlazos
 2025-10-17 07:27:11 +00:00
Shangdi Yu
d82527b32a [Windows] Add AOTI cross-compilation CI (#165573) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/165573
Approved by: https://github.com/malfet
ghstack dependencies: #165560
 2025-10-17 01:05:35 +00:00
Wei Wang
d7e275d4b4 [CI][CUDA] Add periodic b200 distributed job (#159323) 
1. Run distributed job with B200 runner, periodically.
2. discovered generic distributed test issue that certain unit test hard-coded ranks, calling for require_exact_world_size(world_size) API instead of require_world_size(world_size).

Pull Request resolved: https://github.com/pytorch/pytorch/pull/159323
Approved by: https://github.com/eqy

Co-authored-by: Aidyn-A <aidyn.b.aitzhan@gmail.com>
 2025-10-16 21:54:04 +00:00
Jithun Nair
d5db3aee0d [CI] Use 1-GPU runners for rocm-mi355.yml (#165658) 
Should only need 1-GPU runners for rocm-mi355.yml since it runs `default` test config which only needs 1 GPU

Pull Request resolved: https://github.com/pytorch/pytorch/pull/165658
Approved by: https://github.com/jeffdaily
 2025-10-16 21:53:22 +00:00
Maggie Moss
d795fb225a [RFC] Add pyrefly to lintrunner (#165179) 
This will add pyrefly to lint runner as a warning only - and allow us to collect feedback about the tool before switching to pyrefly as the main type checker.

References the steps outlined here: : https://github.com/pytorch/pytorch/issues/163283:

test plan:
`lintrunner init`
`lintrunner`
confirm when pyrefly errors are present results look like: https://gist.github.com/maggiemoss/e6cb2d015dd1ded560ae1329098cf33f

Pull Request resolved: https://github.com/pytorch/pytorch/pull/165179
Approved by: https://github.com/ezyang
 2025-10-16 20:07:09 +00:00
Huy Do
6dedd34c31 [CD] Skip 12.9 build on Windows (#165665) 
Per title

Pull Request resolved: https://github.com/pytorch/pytorch/pull/165665
Approved by: https://github.com/Camyll, https://github.com/malfet
 2025-10-16 19:11:27 +00:00
Thanh Ha
85586d7efc Make c7i the default for _linux-build.yml (#164747) 
Use linux.c7i.2xlarge as the default runner for the _linux-build.yml workflow. In testing we found that switching from c5 - c7i grants a 15-20% faster build times despite c7i costing 5% more. This should reduce costs of jobs using _linux-build.yml.

Relates to pytorch/test-infra#7175.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/164747
Approved by: https://github.com/atalman
 2025-10-16 17:37:51 +00:00
Nikita Shulga
23fb7e9f4b [CI] Add arch prefix in front of op benchmark results (#165584) 
To be able to run x86 and aarch64 benchmarks later on
Pull Request resolved: https://github.com/pytorch/pytorch/pull/165584
Approved by: https://github.com/huydhn
ghstack dependencies: #165583
 2025-10-16 01:50:52 +00:00
Huy Do
c2bd41ac9f Build vLLM nightly wheels for CUDA 13.0 (#163239) 
Now that https://github.com/vllm-project/vllm/pull/24599 has been merged
Pull Request resolved: https://github.com/pytorch/pytorch/pull/163239
Approved by: https://github.com/malfet, https://github.com/atalman
 2025-10-16 01:03:26 +00:00
Nikita Shulga
7e6721fb0a [BE] Remove confusing opbenchmark-on-demand-build (#165583) 
As it doesn't have a test shard, so what's the point or running the build? Was added in https://github.com/pytorch/pytorch/pull/143733 and looks like test shard never existed for it

Moreover, allow one to specify benchmark size as argument, so one
technically can do a workflow dispatch with different opbenchmark sizes
Pull Request resolved: https://github.com/pytorch/pytorch/pull/165583
Approved by: https://github.com/huydhn
 2025-10-15 23:48:28 +00:00
PyTorch UpdateBot
59d30d1b75 [vision hash update] update the pinned vision hash (#165496) 
This PR is auto-generated nightly by [this action](https://github.com/pytorch/pytorch/blob/main/.github/workflows/nightly.yml).
Update the pinned vision hash.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/165496
Approved by: https://github.com/pytorchbot
 2025-10-15 04:35:50 +00:00
PyTorch UpdateBot
3915898c22 [audio hash update] update the pinned audio hash (#165495) 
This PR is auto-generated nightly by [this action](https://github.com/pytorch/pytorch/blob/main/.github/workflows/nightly.yml).
Update the pinned audio hash.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/165495
Approved by: https://github.com/pytorchbot
 2025-10-15 04:32:49 +00:00
Jean Schmidt
1ec0755a7e [ISSUES] Update ci:sev template to include a note about ci: disable-autorevert label (#165459) 
We noticed that disabling autorevert in any and all ci:sevs is too impactful, as ci: sevs are sometimes created just to communicate an action or a impactful change. But sometimes durring a SEV we might not want to disable autorevert anyways, a example is a ci: sev impacting jobs we don't use as basis for autorevert.

So, a note is added reminding the ci:sev author to optionally add this tag to disable auto-revert

Note: using this opportunity to fix the ci: disable-autorevert issues. As it is best for the title to be simple and the displayed message in the GitHub interface to be decorated with emoji :)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/165459
Approved by: https://github.com/malfet
 2025-10-14 20:32:46 +00:00
Jeff Daily
d7e3f493d9 [ROCm][CI] add mi355 to inductor perf test nightly (#165326) 
Fixes #ISSUE_NUMBER

Pull Request resolved: https://github.com/pytorch/pytorch/pull/165326
Approved by: https://github.com/jeffdaily

Co-authored-by: Jeff Daily <jeff.daily@amd.com>
 2025-10-14 20:03:21 +00:00
Shangdi Yu
09a4187b8e Update windows cuda build to use 12.8 (#165345) 
As title

Motivation: The rest of the pytorch and inductor build is using 12.8 and we're deprecating cuda 12.6 builds soon per https://github.com/pytorch/pytorch/issues/165111

Pull Request resolved: https://github.com/pytorch/pytorch/pull/165345
Approved by: https://github.com/atalman, https://github.com/malfet
 2025-10-14 13:58:20 +00:00
PyTorch MergeBot
c5972ebdfb Revert "Update windows cuda build to use 12.8 (#165345)" 
This reverts commit ca96c675001fa87b9d9c648972415ab8b1591f11.

Reverted https://github.com/pytorch/pytorch/pull/165345 on behalf of https://github.com/pytorch-auto-revert due to Reverted automatically by pytorch's autorevert, to avoid this behaviour add the tag autorevert: disable ([comment](https://github.com/pytorch/pytorch/pull/165345#issuecomment-3400344079))
 2025-10-14 06:46:33 +00:00
Shangdi Yu
ca96c67500 Update windows cuda build to use 12.8 (#165345) 
As title

Motivation: The rest of the pytorch and inductor build is using 12.8 and we're deprecating cuda 12.6 builds soon per https://github.com/pytorch/pytorch/issues/165111

Pull Request resolved: https://github.com/pytorch/pytorch/pull/165345
Approved by: https://github.com/atalman
 2025-10-14 02:33:44 +00:00
Catherine Lee
64699b8042 [trymerge] Do not check for rules when reverting (#165342) 
Why do we need to check for merge rules when reverting?
Pull Request resolved: https://github.com/pytorch/pytorch/pull/165342
Approved by: https://github.com/malfet
 2025-10-13 19:07:00 +00:00
Catherine Lee
684df93975 [CI] Default keep-going true for tags of form ciflow/something/commitsha (#165180) 
Tags of the form `ciflow/something/commitsha` are usually created by running the workflow from HUD

Pull Request resolved: https://github.com/pytorch/pytorch/pull/165180
Approved by: https://github.com/huydhn
 2025-10-13 16:12:37 +00:00
PyTorch UpdateBot
4874cce52f [xla hash update] update the pinned xla hash (#165302) 
This PR is auto-generated nightly by [this action](https://github.com/pytorch/pytorch/blob/main/.github/workflows/nightly.yml).
Update the pinned xla hash.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/165302
Approved by: https://github.com/pytorchbot
 2025-10-13 12:36:29 +00:00
PyTorch UpdateBot
957b0e9793 [vision hash update] update the pinned vision hash (#165017) 
This PR is auto-generated nightly by [this action](https://github.com/pytorch/pytorch/blob/main/.github/workflows/nightly.yml).
Update the pinned vision hash.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/165017
Approved by: https://github.com/pytorchbot
 2025-10-13 04:35:52 +00:00
PyTorch UpdateBot
b04def139e [audio hash update] update the pinned audio hash (#165113) 
This PR is auto-generated nightly by [this action](https://github.com/pytorch/pytorch/blob/main/.github/workflows/nightly.yml).
Update the pinned audio hash.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/165113
Approved by: https://github.com/pytorchbot
 2025-10-13 04:35:36 +00:00
PyTorch UpdateBot
a2601630cd [vllm hash update] update the pinned vllm hash (#164628) 
This PR is auto-generated nightly by [this action](https://github.com/pytorch/pytorch/blob/main/.github/workflows/nightly.yml).
Update the pinned vllm hash.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/164628
Approved by: https://github.com/pytorchbot

Co-authored-by: Huy Do <huydhn@gmail.com>
 2025-10-12 18:26:07 +00:00
Huy Do
5ad7611b52 Reland vision pinned commit hash update (#164492) 
Redo https://github.com/pytorch/pytorch/pull/154694

Pull Request resolved: https://github.com/pytorch/pytorch/pull/164492
Approved by: https://github.com/yangw-dev
 2025-10-12 04:53:27 +00:00
Thanh Ha
79a33e2db2 Switch docs build from c5 to c7i (#165082) 
Switch docs build from c5 to c7i which should increase build
performance by roughly 15-20% while reducing costs by 10-15%.

Signed-off-by: Thanh Ha <thanh.ha@linuxfoundation.org>
 2025-10-11 10:59:18 -04:00
Huy Do
4400c5d31e Continue to build nightly CUDA 12.9 for internal (#163029) 
Revert part of https://github.com/pytorch/pytorch/pull/161916 to continue building CUDA 12.9 nightly

Pull Request resolved: https://github.com/pytorch/pytorch/pull/163029
Approved by: https://github.com/malfet
 2025-10-11 08:26:47 +00:00
Yang Wang
cafca357fb Fix h100 daily inductor running dispatch (#165185) 
casued by merged pr: e7ed1a00eb

the if condition should also updated

Pull Request resolved: https://github.com/pytorch/pytorch/pull/165185
Approved by: https://github.com/malfet, https://github.com/huydhn
 2025-10-10 21:28:58 +00:00
Jithun Nair
0ec0120b19 Move aws OIDC credentials steps into setup-rocm.yml (#164769) 
The AWS ECR login step needs `id-token: write` permissions. We move the steps to get OIDC-based credentials from `_rocm-test.yml` to `setup-rocm.yml`. This lays the groundwork to enable access to AWS ECR in workflows in other repos such as torchtitan that use [linux_job_v2.yml](https://github.com/pytorch/test-infra/blob/main/.github/workflows/linux_job_v2.yml), which also uses [setup-rocm.yml](335f4f80a0/.github/workflows/linux_job_v2.yml (L168)).

Any caller workflows that eventually execute `setup-rocm` action will thus need to provide the `id-token: write` permission.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/164769
Approved by: https://github.com/huydhn
 2025-10-10 21:24:29 +00:00
Catherine Lee
370b1c12d2 [CI] Put the no gpu tests on machines that don't have gpus (#165183) 
I think this is just a copy paste error?

NS: Introduced by https://github.com/pytorch/pytorch/pull/161013

Not sure where it got copied from though, the other set of no gpu tests for the other cuda version already have cpu runners
Pull Request resolved: https://github.com/pytorch/pytorch/pull/165183
Approved by: https://github.com/malfet
 2025-10-10 20:59:09 +00:00
Catherine Lee
6fd1ca28e1 [lint] Run full lint on ciflow/trunk (#165169) 
Add some naming stuff to differentiate between full + partial

If we find that partial always == full, then we can get rid of it

https://github.com/pytorch/pytorch/issues/165168
Pull Request resolved: https://github.com/pytorch/pytorch/pull/165169
Approved by: https://github.com/Skylion007, https://github.com/malfet
 2025-10-10 20:38:51 +00:00
Nikita Shulga
7cddda1234 Update asan in slow to linux.2xlarge.memory 
Followup after f2ae7084eb
 2025-10-10 12:02:29 -07:00
Nikita Shulga
f2ae7084eb [BE] Use linux.2xlarge.memory for ASAN builds (#165164) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/165164
Approved by: https://github.com/janeyx99
 2025-10-10 18:13:42 +00:00
Thanh Ha
10a9fb641b Switch build jobs from linux.4xlarge to c7i (#165057) 
Switch build jobs that use linux.4xlarge which uses c5 instance types to c7i variant. This should improve performance by ~15-20% while cutting costs by ~10-15%.

Relates to pytorch/test-infra#7175
Pull Request resolved: https://github.com/pytorch/pytorch/pull/165057
Approved by: https://github.com/huydhn
 2025-10-10 15:13:40 +00:00
Nikita Shulga
44b1ff54e9 [CD] Do not propagate download.pytorch.org IP into container (#165075) 
Followup after https://github.com/pytorch/pytorch/pull/164969

Should fix binary build test failures
Pull Request resolved: https://github.com/pytorch/pytorch/pull/165075
Approved by: https://github.com/seemethere, https://github.com/huydhn
ghstack dependencies: #164968, #164969
 2025-10-10 04:27:29 +00:00
PyTorch MergeBot
daea35df5c Revert "[CD] Do not propagate download.pytorch.org IP into container (#165075)" 
This reverts commit 6d27a8e5093ee2a21d44dceeeffcb272e6e0f655.

Reverted https://github.com/pytorch/pytorch/pull/165075 on behalf of https://github.com/pytorch-auto-revert due to Reverted automatically by pytorch's autorevert, to avoid this behaviour add the tag autorevert: disable ([comment](https://github.com/pytorch/pytorch/pull/165075#issuecomment-3388228013))
 2025-10-10 04:20:51 +00:00
Nikita Shulga
6d27a8e509 [CD] Do not propagate download.pytorch.org IP into container (#165075) 
Followup after https://github.com/pytorch/pytorch/pull/164969

Should fix binary build test failures
Pull Request resolved: https://github.com/pytorch/pytorch/pull/165075
Approved by: https://github.com/seemethere, https://github.com/huydhn
ghstack dependencies: #164968, #164969
 2025-10-09 21:59:31 +00:00
Nikita Shulga
e7fd296930 [CI] Add full debug build to trunk (#164974) 
But not test, just import torch, as regression test for https://github.com/pytorch/pytorch/issues/164297

Test plan: Re-apply #164974 on top of this change and observer the failure in the workflows: https://github.com/pytorch/pytorch/actions/runs/18383302153/job/52375282838
Pull Request resolved: https://github.com/pytorch/pytorch/pull/164974
Approved by: https://github.com/seemethere, https://github.com/clee2000, https://github.com/atalman
ghstack dependencies: #164968, #164969
 2025-10-09 20:12:16 +00:00
Jithun Nair
ee6a1ecb0a [ROCm] Enable MI355 CI on PRs, and run full set of UTs on PRs (#160215) 
Useful to have PR testing for PRs such as https://github.com/pytorch/pytorch/pull/151360

Pull Request resolved: https://github.com/pytorch/pytorch/pull/160215
Approved by: https://github.com/malfet, https://github.com/atalman

Co-authored-by: Jeff Daily <jeff.daily@amd.com>
 2025-10-09 18:03:12 +00:00
Thanh Ha
b28b24a9fc Switch build jobs that use linux.12xlarge to c7i (#164941) 
This PR updates build jobs that currently use linux.12xlarge to the
c7i varient which should increase build times by 15% - 20% depending
on the job and reduce costs of these jobs by 10% - 15%.

Signed-off-by: Thanh Ha <thanh.ha@linuxfoundation.org>
 2025-10-09 09:58:52 -04:00
PyTorch MergeBot
5b8174bc28 Revert "[vllm hash update] update the pinned vllm hash (#164628)" 
This reverts commit 7b691546d2949790ffc8f6bd3c674faa6a46ff7c.

Reverted https://github.com/pytorch/pytorch/pull/164628 on behalf of https://github.com/huydhn due to There are some broken vLLM tests ([comment](https://github.com/pytorch/pytorch/pull/164628#issuecomment-3384560957))
 2025-10-09 07:43:02 +00:00
PyTorch MergeBot
a753ffa9af Revert "Use runner with more memory for ASAN builds (#165000)" 
This reverts commit f5fd18f7e24378bd9eb91404f697f1c81a8187d5.

Reverted https://github.com/pytorch/pytorch/pull/165000 on behalf of https://github.com/izaitsevfb due to not sure how, but this broke lint ([comment](https://github.com/pytorch/pytorch/pull/165000#issuecomment-3384286412))
 2025-10-09 06:22:28 +00:00
PyTorch UpdateBot
7b691546d2 [vllm hash update] update the pinned vllm hash (#164628) 
This PR is auto-generated nightly by [this action](https://github.com/pytorch/pytorch/blob/main/.github/workflows/nightly.yml).
Update the pinned vllm hash.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/164628
Approved by: https://github.com/pytorchbot
 2025-10-09 04:35:36 +00:00
Ivan Zaitsev
f5fd18f7e2 Use runner with more memory for ASAN builds (#165000) 
An attempt to [address OOM here](aed5ed1076/1).
Pull Request resolved: https://github.com/pytorch/pytorch/pull/165000
Approved by: https://github.com/seemethere, https://github.com/malfet, https://github.com/huydhn
 2025-10-09 01:09:28 +00:00
Nikita Shulga
f1229b6db9 [BE] Remove manual IP address resolution (#164969) 
As https://github.com/pytorch/pytorch/issues/100400 has been closed a while back
Pull Request resolved: https://github.com/pytorch/pytorch/pull/164969
Approved by: https://github.com/seemethere
ghstack dependencies: #164968
 2025-10-08 21:22:34 +00:00
Nikita Shulga
15800888b6 [CI] Print GPU info during setup linux (#164968) 
I.e. run `nvidia-smi` if present

Helps detecting what driver version this runner is on, which would have helped debugging some of the issues recently
Pull Request resolved: https://github.com/pytorch/pytorch/pull/164968
Approved by: https://github.com/ngimel
 2025-10-08 20:58:33 +00:00
Catherine Lee
e7ed1a00eb Run inductor-perf-test-nightly-h100 once per day (#164967) 
To reduce inductor costs, though I'm not sure how much this one matters specifically since h100s are reserved

Pull Request resolved: https://github.com/pytorch/pytorch/pull/164967
Approved by: https://github.com/BoyuanFeng
 2025-10-08 20:58:19 +00:00
Nikita Shulga
90c0825e2d [GHF] Allow reverts from pytorch-auto-revert app (#164911) 
This is a bit weird, but author_login is not a unique field, but author_url is.

Explicitly allow https://github.com/apps/pytorch-auto-revert to issue revert commands

Update mocks by running
```
sed -i -e s/8e262b0495bd934d39dda198d4c09144311c5ddd6cca6a227194bd48dbfe7201/47860a8f57a214a426d1150c29893cbc2aa49507f12b731483b1a1254bca3428/ gql_mocks.json
```

Test plan: Run
```python
from trymerge import GitHubPR
pr=GitHubPR("pytorch", "pytorch", 164660)
print(pr.get_last_comment().author_url, pr.get_comment_by_id(3375785595).author_url)
```
that should produce
```
https://github.com/pytorch-auto-revert https://github.com/apps/pytorch-auto-revert
```
Plus added a regression test that checks two particular comments for revert validity

`pytorch-auto-revert` user is my alter ego :)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/164911
Approved by: https://github.com/jeanschmidt
 2025-10-08 15:15:45 +00:00
PyTorch MergeBot
1927783aa3 Revert "Reland vision pinned commit hash update (#164492)" 
This reverts commit 6861a270624b44954826688f8dad668eb0154452.

Reverted https://github.com/pytorch/pytorch/pull/164492 on behalf of https://github.com/izaitsevfb due to see autorevert msg above, inductor breakage is legit ([comment](https://github.com/pytorch/pytorch/pull/164492#issuecomment-3379537888))
 2025-10-08 04:38:26 +00:00