🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/157147

• 📄 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.

✅ You can merge normally! (2 Unrelated Failures)

As of commit e60eb70 with merge base 8a88c6e ():

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

cc @kwen2501

good catch!

there are 2 usage of pin_memory(device=) in fully_shard. Is it possible to remove another as well?

note to myself:
at aten level, there is a major refactor to deprecate the device argument in pin_memory(device=) #126376

now it throws warnings if using device argument https://github.com/pytorch/pytorch/blame/20e40492b046b9287726d3ec656117e4dc38f0e2/aten/src/ATen/native/Memory.cpp#L45-L49

cc @jeejakp12 as we remove device from pin_memory

there is a chance of breaking 3rd party device. but according to #126376, even 3rd party device should avoid using pin_memory(device) as there are other ways in that PR

approved. but probably need to wait for CI (linter etc)

@pytorchmergebot merge

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