switching NNC as default for TorchScript support

[jjsjann123]

Disable nvfuser by default in TorchScript
Add deprecation warning for nvfuser usage via TorchScript and PrimTorch

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/105185

• 📄 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.

✅ No Failures

As of commit 9d12d22:
💚 Looks good so far! There are no failures yet. 💚

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

[malfet]

LGTM (though not sure if NNC would be default in this case)

[davidberard98]

I think this looks like the right way to make NNC the default

[davidberard98]

although the failing test looks legitimate, may need to skip it?

[jjsjann123]

I'll try to take a look at the failing tests, but I also noticed some message from bot on that test: #103440

[jjsjann123]

Looks like the error is just coming from warning message I added 😆 skipped that part.

[jjsjann123]

failure coming from inductor and ROCM I'm not sure if they are relevant or not.:

2023-07-17T19:13:40.5706312Z Running inductor/test_cuda_repro 1/1 ... [2023-07-17 19:13:39.883309]
2023-07-17T19:13:40.5706447Z test_unary_ufuncs 1/1 failed! Received signal: SIGSEGV
2023-07-17T19:13:40.5707152Z Executing ['/opt/conda/envs/py_3.8/bin/python', '-bb', 'inductor/test_cuda_repro.py', '--shard-id=0', '--num-shards=1', '-v', '-vv', '-rfEX', '-p', 'no:xdist', '--use-pytest', '--sc=inductor/test_cuda_repro_0', '-x', '--reruns=2', '--import-slow-tests', '--import-disabled-tests'] ... [2023-07-17 19:13:39.883653]
2023-07-17T19:13:40.5707267Z Traceback (most recent call last):
2023-07-17T19:13:40.5707447Z   File "test/run_test.py", line 1683, in <module>
2023-07-17T19:13:40.5707565Z     main()
2023-07-17T19:13:40.5707692Z   File "test/run_test.py", line 1656, in main
2023-07-17T19:13:40.5707799Z     failure_messages += run_tests(
2023-07-17T19:13:40.5707927Z   File "test/run_test.py", line 1561, in run_tests
2023-07-17T19:13:40.5708017Z     raise RuntimeError(
2023-07-17T19:13:40.5708184Z RuntimeError: test_unary_ufuncs 1/1 failed! Received signal: SIGSEGV

[jjsjann123]

Looks like CI is clean now.

But there's an EasyCLA that I need to sign? is this something new or should I have pushed the branch to pytorch/pytorch and start the PR directly there?

cc'ing @davidberard98

[davidberard98]

/easycla

[davidberard98]

looks like that might have fixed it

[jjsjann123]

looks like that might have fixed it

Awesome~~ I guess this PR is good to go then.

[jjsjann123]

@pytorchbot merge

[pytorchmergebot]

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