[profiler] add enabled API to autograd profiler

[rohan-varma]

Stack from ghstack:

• [profiler] add enabled API to autograd profiler #31380 [profiler] add enabled API to autograd profiler

For being able to profile async RPCs, we attach a RecordFunction object to the future that is created during the RPC to persist it across the lifetime of the RPC (this is implemented in the next PR:#31381 ). Since we'd only like to do this when profiling is enabled, this PR adds an enabled API to the autograd profiler.

Differential Revision: D19050391

[pritamdamania87]

Don't we already check if the profiler is enabled in #31381 and attach the RecordFunction only in those cases? Otherwise the rf_ pointer is nullptr?

[rohan-varma]

@pritamdamania87 : That check relies on this change, in particular, it uses torch.autograd._profiler_enabled() here: https://github.com/pytorch/pytorch/pull/31381/files#diff-5e438ccfe0d61a53077ec02243a79f07R277

[kostmo]

💊 CircleCI build failures summary and remediations

As of commit 9b40a7e:

• 1/2 broken upstream at merge base 700109e (see grid view)

  You may want to rebase on the viable/strict branch (expand for instructions)

  If your commit is newer than viable/strict, you can try basing on an older, stable commit:

  git fetch origin viable/strict
  git rebase --onto viable/strict $(git merge-base origin/master HEAD)
  

  If your commit is older than viable/strict:

  git fetch origin viable/strict
  git rebase viable/strict
  

  Check out the recency history of this "viable master" tracking branch.

• 1/2 failures introduced in this PR

Detailed failure analysis

One may explore the probable reasons each build failed interactively on the Dr. CI website.

🕵️ 1 new failure recognized by patterns

The following build failures don't appear to be due to upstream breakage:

caffe2_onnx_py2_gcc5_ubuntu16_04_test (1/1)

Step: "Test" (full log | pattern match details)

Dec 21 11:16:48 E RuntimeError: required keyword attribute 'Y_scale' is undefined

Dec 21 11:16:48 k = 'Y_scale' 
Dec 21 11:16:48  
Dec 21 11:16:48     def _node_getitem(self, k): 
Dec 21 11:16:48         r""" 
Dec 21 11:16:48         Accessor for attributes of a node which is polymorphic over 
Dec 21 11:16:48         return type. 
Dec 21 11:16:48      
Dec 21 11:16:48         NB: This is monkey-patched onto Node. 
Dec 21 11:16:48         """ 
Dec 21 11:16:48 >       sel = self.kindOf(k) 
Dec 21 11:16:48 E       RuntimeError: required keyword attribute 'Y_scale' is undefined 
Dec 21 11:16:48  
Dec 21 11:16:48 ../.local/lib/python2.7/site-packages/torch/onnx/utils.py:836: RuntimeError 
Dec 21 11:16:48 =============================== warnings summary =============================== 
Dec 21 11:16:48 /usr/local/lib/python2.7/dist-packages/scipy/_lib/_numpy_compat.py:10 
Dec 21 11:16:48   /usr/local/lib/python2.7/dist-packages/scipy/_lib/_numpy_compat.py:10: DeprecationWarning: Importing from numpy.testing.nosetester is deprecated since 1.15.0, import from numpy.testing instead. 
Dec 21 11:16:48     from numpy.testing.nosetester import import_nose 
Dec 21 11:16:48  
Dec 21 11:16:48 /usr/local/lib/python2.7/dist-packages/scipy/stats/morestats.py:12 
Dec 21 11:16:48   /usr/local/lib/python2.7/dist-packages/scipy/stats/morestats.py:12: DeprecationWarning: Importing from numpy.testing.decorators is deprecated since numpy 1.15.0, import from numpy.testing instead. 
Dec 21 11:16:48     from numpy.testing.decorators import setastest 

🚧 1 upstream failure recognized by patterns:

These builds matched patterns, but were probably caused by upstream breakages:

• pytorch_linux_xenial_py3_clang5_asan_test

---

This comment was automatically generated by Dr. CI (expand for details).

Follow this link to opt-out of these comments for your Pull Requests.

Please report bugs/suggestions on the GitHub issue tracker.

This comment has been revised 7 times.

[rohan-varma]

Test failures are unrelated:

Dec 21 11:16:48 self = <[MemoryError("std::bad_alloc") raised in repr()] SafeRepr object at 0x7f634539b5f0>
Dec 21 11:16:48 k = 'Y_scale'
Dec 21 11:16:48 
Dec 21 11:16:48     def _node_getitem(self, k):
Dec 21 11:16:48         r"""
Dec 21 11:16:48         Accessor for attributes of a node which is polymorphic over
Dec 21 11:16:48         return type.
Dec 21 11:16:48     
Dec 21 11:16:48         NB: This is monkey-patched onto Node.
Dec 21 11:16:48         """
Dec 21 11:16:48 >       sel = self.kindOf(k)
Dec 21 11:16:48 E       RuntimeError: required keyword attribute 'Y_scale' is undefined
Dec 21 11:16:48 
Dec 21 11:16:48 ../.local/lib/python2.7/site-packages/torch/onnx/utils.py:836: RuntimeError

and

Dec 21 11:36:57 SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior /var/lib/jenkins/workspace/third_party/fbgemm/src/QuantUtilsAvx2.cc:98:14 in 
Dec 21 11:36:57 Traceback (most recent call last):
Dec 21 11:36:57   File "test/run_test.py", line 456, in <module>
Dec 21 11:36:57     main()
Dec 21 11:36:57   File "test/run_test.py", line 449, in main
Dec 21 11:36:57     raise RuntimeError(message)
Dec 21 11:36:57 RuntimeError: test_quantization failed!
Dec 21 11:36:57 + cleanup
Dec 21 11:36:57 + retcode=1
Dec 21 11:36:57 + set +x

[facebook-github-bot]

This pull request has been merged in cc2d5ca.