Refactoring FSDP2 (_composable/fsdp) test cases to be device agnostic #149848

AnantGulati · 2025-03-24T09:36:03Z

The motivation for this PR is refactor existing test cases in the folder test/distributed/_composable/fsdp/ or fsdp2(as referred to in torch titan) to be device agnostic such that any accelerator type is supported (for eg. CUDA, HPU, XPU etc)

The changes are in line with previously merged changes for fsdp (present in the folder test/distributed/fsdp/ ) test cases: #139184

cc @H-Huang @awgu @wanchaol @fegin @fduwjj @wz337 @wconstab @d4l3k @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @chenyang78 @kadeng @chauhang @amjames @kwen2501 @c-p-i-o

pytorch-bot · 2025-03-24T09:36:06Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/149848

📄 Preview Python docs built from this PR
📄 Preview C++ docs built from this PR
❓ Need help or want to give feedback on the CI? Visit the bot commands wiki or our office hours

Note: Links to docs will display an error until the docs builds have been completed.

❌ 1 Cancelled Job

As of commit 36f1694 with merge base 7e16cb9 ():

CANCELLED JOB - The following job was cancelled. Please retry:

Apply lint suggestions (gh)

This comment was automatically generated by Dr. CI and updates every 15 minutes.

AnantGulati · 2025-04-02T09:26:15Z

@wconstab could you please help with this PR
Thanks

AnantGulati · 2025-04-07T13:14:56Z

@wconstab @kwen2501 Could you please help with this PR.
Thanks

AnantGulati · 2025-04-24T04:57:49Z

@wconstab Could you please help with this review.
Thanks

AnantGulati · 2025-04-28T05:23:11Z

@wconstab @H-Huang @wanchaol @fegin @kwen2501 @d4l3k
Could you please help with this review
Thanks

cyyever · 2025-04-28T05:29:33Z

@pytorchbot rebase

pytorchmergebot · 2025-04-28T05:31:06Z

@pytorchbot started a rebase job onto refs/remotes/origin/viable/strict. Check the current status here

pytorchmergebot · 2025-04-28T05:31:09Z

Rebase failed due to Command git -C /home/runner/work/pytorch/pytorch rebase refs/remotes/origin/viable/strict pull/149848/head returned non-zero exit code 1

Rebasing (1/9)
Auto-merging test/distributed/_composable/fsdp/test_fully_shard_grad_scaler.py
CONFLICT (content): Merge conflict in test/distributed/_composable/fsdp/test_fully_shard_grad_scaler.py
Auto-merging test/distributed/_composable/fsdp/test_fully_shard_init.py
error: could not apply 920e06d304d... adding fsdp files
hint: Resolve all conflicts manually, mark them as resolved with
hint: "git add/rm <conflicted_files>", then run "git rebase --continue".
hint: You can instead skip this commit: run "git rebase --skip".
hint: To abort and get back to the state before "git rebase", run "git rebase --abort".
hint: Disable this message with "git config set advice.mergeConflict false"
Could not apply 920e06d304d... adding fsdp files

Raised by https://github.com/pytorch/pytorch/actions/runs/14700946048

cyyever · 2025-04-28T05:31:44Z

@AnantGulati Please rebase and solve any conflict.

AnantGulati · 2025-04-28T06:10:09Z

@pytorchbot rebase

pytorchmergebot · 2025-04-28T06:11:50Z

@pytorchbot started a rebase job onto refs/remotes/origin/viable/strict. Check the current status here

pytorchmergebot · 2025-04-28T06:11:52Z

Rebase failed due to Command git -C /home/runner/work/pytorch/pytorch rebase refs/remotes/origin/viable/strict pull/149848/head returned non-zero exit code 1

Rebasing (1/13)
Auto-merging test/distributed/_composable/fsdp/test_fully_shard_grad_scaler.py
CONFLICT (content): Merge conflict in test/distributed/_composable/fsdp/test_fully_shard_grad_scaler.py
Auto-merging test/distributed/_composable/fsdp/test_fully_shard_init.py
error: could not apply 920e06d304d... adding fsdp files
hint: Resolve all conflicts manually, mark them as resolved with
hint: "git add/rm <conflicted_files>", then run "git rebase --continue".
hint: You can instead skip this commit: run "git rebase --skip".
hint: To abort and get back to the state before "git rebase", run "git rebase --abort".
hint: Disable this message with "git config set advice.mergeConflict false"
Could not apply 920e06d304d... adding fsdp files

Raised by https://github.com/pytorch/pytorch/actions/runs/14701438311

AnantGulati · 2025-04-28T06:26:11Z

@pytorchbot rebase

pytorchmergebot · 2025-04-28T06:27:45Z

@pytorchbot started a rebase job onto refs/remotes/origin/viable/strict. Check the current status here

pytorchmergebot · 2025-04-28T06:27:47Z

Rebase failed due to Command git -C /home/runner/work/pytorch/pytorch rebase refs/remotes/origin/viable/strict pull/149848/head returned non-zero exit code 1

Rebasing (1/16)
Auto-merging test/distributed/_composable/fsdp/test_fully_shard_grad_scaler.py
CONFLICT (content): Merge conflict in test/distributed/_composable/fsdp/test_fully_shard_grad_scaler.py
Auto-merging test/distributed/_composable/fsdp/test_fully_shard_init.py
error: could not apply 920e06d304d... adding fsdp files
hint: Resolve all conflicts manually, mark them as resolved with
hint: "git add/rm <conflicted_files>", then run "git rebase --continue".
hint: You can instead skip this commit: run "git rebase --skip".
hint: To abort and get back to the state before "git rebase", run "git rebase --abort".
hint: Disable this message with "git config set advice.mergeConflict false"
Could not apply 920e06d304d... adding fsdp files

Raised by https://github.com/pytorch/pytorch/actions/runs/14701672157

cyyever · 2025-04-29T01:37:38Z

@AnantGulati You have to manually rebase..

kwen2501 · 2025-04-29T06:37:52Z

cc @weifengpy @mori360. It is a big change but would be nice to have. Thank you!

kwen2501

LGTM! Thanks a lot for the contribution!

test/distributed/_composable/fsdp/test_fully_shard_comm.py

test/distributed/_composable/fsdp/test_fully_shard_overlap.py

AnantGulati · 2025-04-30T14:17:44Z

@cyyever @kwen2501 All change included in new PR.
Could you please approve the CI flow
Thanks

cyyever · 2025-04-30T15:11:19Z

@pytorchbot merge -i

AnantGulati · 2025-05-16T06:15:06Z

@pytorchbot rebase

pytorchmergebot · 2025-05-16T06:16:44Z

@pytorchbot started a rebase job onto refs/remotes/origin/viable/strict. Check the current status here

pytorchmergebot · 2025-05-16T06:16:46Z

Tried to rebase and push PR #149848, but it was already up to date. Try rebasing against main by issuing:
@pytorchbot rebase -b main

AnantGulati · 2025-05-16T09:18:01Z

@pytorchbot merge

pytorchmergebot · 2025-05-16T09:20:38Z

Merge started

Your change will be merged once all checks pass (ETA 0-4 Hours).

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging

Check the merge workflow status
here

pytorchmergebot · 2025-05-16T09:20:54Z

Merge failed

Reason: 1 jobs have failed, first few of them are: Apply lint suggestions

Details for Dev Infra team

Raised by workflow job

AnantGulati · 2025-05-19T04:15:33Z

@pytorchbot merge

pytorchmergebot · 2025-05-19T04:17:27Z

Merge started

Your change will be merged once all checks pass (ETA 0-4 Hours).

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging

Check the merge workflow status
here

pytorchmergebot · 2025-05-19T04:17:41Z

Merge failed

Reason: 1 jobs have failed, first few of them are: Apply lint suggestions

Details for Dev Infra team

Raised by workflow job

cyyever · 2025-05-19T05:44:55Z

@pytorchbot merge -f "unrelated lint failures"

pytorchmergebot · 2025-05-19T05:46:40Z

Merge started

Your change will be merged immediately since you used the force (-f) flag, bypassing any CI checks (ETA: 1-5 minutes). Please use -f as last resort and instead consider -i/--ignore-current to continue the merge ignoring current failures. This will allow currently pending tests to finish and report signal before the merge.

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging

Check the merge workflow status
here

pytorch-bot bot added oncall: distributed Add this issue/PR to distributed oncall triage queue topic: not user facing topic category labels Mar 24, 2025

pytorchbot added the open source label Mar 24, 2025

colesbury requested a review from wconstab March 24, 2025 18:32

colesbury added the triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module label Mar 24, 2025

AnantGulati changed the title ~~Refactoring FSDP2 (_composable/fsdp) test cases to be device agnostic (1/n)~~ Refactoring FSDP2 (_composable/fsdp) test cases to be device agnostic Apr 14, 2025

kwen2501 requested review from weifengpy and mori360 April 29, 2025 06:36

kwen2501 approved these changes Apr 29, 2025

View reviewed changes

test/distributed/_composable/fsdp/test_fully_shard_comm.py Outdated Show resolved Hide resolved

test/distributed/_composable/fsdp/test_fully_shard_comm.py Outdated Show resolved Hide resolved

test/distributed/_composable/fsdp/test_fully_shard_overlap.py Outdated Show resolved Hide resolved

pytorch-bot bot added the ciflow/trunk Trigger trunk jobs on your pull request label Apr 30, 2025

pytorchmergebot added the merging label Apr 30, 2025

AnantGulati added 8 commits May 16, 2025 09:14

Adding remaining files

cc42868

Adding bug fixes

9635fb8

adding fsdp files

2657b8f

Adding explicit device call in grad scalar

e35ae11

adding rebase and lint fix

967ecd7

adding suggestions from comment and bug fixes

b486137

adding fix for test overlap

9da8217

fixing ci error

36f1694

AnantGulati force-pushed the AnantGulati_fsdp2_test_cases branch from baffbdd to 36f1694 Compare May 16, 2025 06:14

pytorch-bot bot added the ciflow/trunk Trigger trunk jobs on your pull request label May 16, 2025

pytorchmergebot added the merging label May 16, 2025

pytorchmergebot removed the merging label May 16, 2025

pytorchmergebot added the merging label May 19, 2025

pytorchmergebot removed the merging label May 19, 2025

pytorchmergebot added the merging label May 19, 2025

pytorchmergebot closed this in 5506baa May 19, 2025

pytorchmergebot added Merged and removed merging labels May 19, 2025

Refactoring FSDP2 (_composable/fsdp) test cases to be device agnostic #149848

Refactoring FSDP2 (_composable/fsdp) test cases to be device agnostic #149848

Uh oh!

Conversation

AnantGulati commented Mar 24, 2025 • edited by pytorch-bot bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

pytorch-bot bot commented Mar 24, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/149848

❌ 1 Cancelled Job

Uh oh!

AnantGulati commented Apr 2, 2025

Uh oh!

AnantGulati commented Apr 7, 2025

Uh oh!

AnantGulati commented Apr 24, 2025

Uh oh!

AnantGulati commented Apr 28, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

cyyever commented Apr 28, 2025

Uh oh!

pytorchmergebot commented Apr 28, 2025

Uh oh!

pytorchmergebot commented Apr 28, 2025

Uh oh!

cyyever commented Apr 28, 2025

Uh oh!

AnantGulati commented Apr 28, 2025

Uh oh!

pytorchmergebot commented Apr 28, 2025

Uh oh!

pytorchmergebot commented Apr 28, 2025

Uh oh!

AnantGulati commented Apr 28, 2025

Uh oh!

pytorchmergebot commented Apr 28, 2025

Uh oh!

pytorchmergebot commented Apr 28, 2025

Uh oh!

cyyever commented Apr 29, 2025

Uh oh!

kwen2501 commented Apr 29, 2025

Uh oh!

kwen2501 left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

AnantGulati commented Apr 30, 2025

Uh oh!

cyyever commented Apr 30, 2025

Uh oh!

AnantGulati commented May 16, 2025

Uh oh!

pytorchmergebot commented May 16, 2025

Uh oh!

pytorchmergebot commented May 16, 2025

Uh oh!

AnantGulati commented May 16, 2025

Uh oh!

pytorchmergebot commented May 16, 2025

Merge started

Uh oh!

pytorchmergebot commented May 16, 2025

Merge failed

Uh oh!

AnantGulati commented May 19, 2025

Uh oh!

pytorchmergebot commented May 19, 2025

Merge started

Uh oh!

pytorchmergebot commented May 19, 2025

Merge failed

Uh oh!

cyyever commented May 19, 2025

Uh oh!

AnantGulati commented Mar 24, 2025 •

edited by pytorch-bot bot

Loading

pytorch-bot bot commented Mar 24, 2025 •

edited

Loading

AnantGulati commented Apr 28, 2025 •

edited

Loading