Fix typing errors in torch.distributed.distributed_c10d.*

[xuzhao9]

Stack from ghstack:

• Fix typing errors in torch.distributed.*, close issue #42967. #47534 Fix typing errors in torch.distributed.*, close issue Enable torch.distributed typechecks during CI #42967.
• Fix typing errors in torch.distributed.nn.* directory. #47533 Fix typing errors in torch.distributed.nn.* directory.
• Fix typing errors in torch.distributed.distributed_c10d.* #47532 Fix typing errors in torch.distributed.distributed_c10d.*
• Fix type annotation errors in torch.distributed.* directory #47531 Fix type annotation errors in torch.distributed.* directory

Differential Revision: D24952501

[dr-ci]

💊 CI failures summary and remediations

As of commit 23127ef (more details on the Dr. CI page):

---

• 2/2 failures introduced in this PR

---

🕵️ 2 new failures recognized by patterns

The following CI failures do not appear to be due to upstream breakages:

docker-pytorch-linux-bionic-py3.8-gcc9 (1/2)

Step: "Check if image should be built" (full log | diagnosis details | 🔁 rerun)

ERROR: Something has gone wrong and the previous image isn't available for the merge-base of your branch

+ docker manifest inspect 308535385114.dkr.ecr.us-east-1.amazonaws.com/pytorch/pytorch-linux-bionic-py3.8-gcc9:9cc3a74f0e401cccf5d075d1c2835af60ce8c310 
 102 115 46 100 105 102 102 46 116 97 114 46 103 122 105 112 34 44 10 32 32 32 32 32 32 32 32 32 34 115 105 122 101 34 58 32 53 52 49 52 53 50 51 56 44 10 32 32 32 32 32 32 32 32 32 34 100 105 103 101 115 116 34 58 32 34 115 104 97 50 53 54 58 54 52 55 49 54 54 100 55 53 53 98 99 51 97 101 98 98 48 97 52 56 99 99 54 98 56 53 50 102 52 57 100 100 97 50 52 100 50 102 53 57 49 52 53 48 54 55 55 56 56 53 55 101 100 100 102 55 101 56 51 97 97 55 97 34 10 32 32 32 32 32 32 125 10 32 32 32 93 10 125]} 
++ git merge-base HEAD 2981ef28c139b416449642419401d7ae6f3f9a8a 
+ git rev-parse 2981ef28c139b416449642419401d7ae6f3f9a8a:.circleci/docker 
9cc3a74f0e401cccf5d075d1c2835af60ce8c310 
+++ git merge-base HEAD 2981ef28c139b416449642419401d7ae6f3f9a8a 
++ git rev-parse 2981ef28c139b416449642419401d7ae6f3f9a8a:.circleci/docker 
+ PREVIOUS_DOCKER_TAG=9cc3a74f0e401cccf5d075d1c2835af60ce8c310 
+ [[ 9cc3a74f0e401cccf5d075d1c2835af60ce8c310 = \9\c\c\3\a\7\4\f\0\e\4\0\1\c\c\c\f\5\d\0\7\5\d\1\c\2\8\3\5\a\f\6\0\c\e\8\c\3\1\0 ]] 
+ echo 'ERROR: Something has gone wrong and the previous image isn'\''t available for the merge-base of your branch' 
ERROR: Something has gone wrong and the previous image isn't available for the merge-base of your branch 
+ echo '       contact the PyTorch team to restore the original images' 
       contact the PyTorch team to restore the original images 
+ exit 1 

pytorch_linux_bionic_py3_6_clang9_test (2/2)

Step: "Run tests" (full log | diagnosis details | 🔁 rerun)

Nov 16 20:56:34 [E request_callback_no_python.cpp:592] Received error while processing request type 2: RuntimeError: Can not pickle torch.futures.Future

Nov 16 20:56:34 At: 
Nov 16 20:56:34   /opt/conda/lib/python3.6/site-packages/torch/distributed/rpc/internal.py(98): serialize 
Nov 16 20:56:34   /opt/conda/lib/python3.6/site-packages/torch/distributed/rpc/internal.py(150): serialize 
Nov 16 20:56:34  
Nov 16 20:56:34 [E request_callback_no_python.cpp:592] Received error while processing request type 2: RuntimeError: Can not pickle torch.futures.Future 
Nov 16 20:56:34  
Nov 16 20:56:34 At: 
Nov 16 20:56:34   /opt/conda/lib/python3.6/site-packages/torch/distributed/rpc/internal.py(98): serialize 
Nov 16 20:56:34   /opt/conda/lib/python3.6/site-packages/torch/distributed/rpc/internal.py(150): serialize 
Nov 16 20:56:34  
Nov 16 20:56:34 [E request_callback_no_python.cpp:592] Received error while processing request type 2: RuntimeError: Can not pickle torch.futures.Future 
Nov 16 20:56:34  
Nov 16 20:56:34 At: 
Nov 16 20:56:34   /opt/conda/lib/python3.6/site-packages/torch/distributed/rpc/internal.py(98): serialize 
Nov 16 20:56:34   /opt/conda/lib/python3.6/site-packages/torch/distributed/rpc/internal.py(150): serialize 
Nov 16 20:56:34  
Nov 16 20:56:35 ok (1.647s) 
Nov 16 20:56:36   test_return_future_remote (__main__.ProcessGroupRpcTestWithSpawn) ... RPC was initialized with the PROCESS_GROUP backend which is deprecated and slated to be removed and superseded by the TENSORPIPE backend. It is recommended to migrate to the TENSORPIPE backend. 
Nov 16 20:56:36 RPC was initialized with the PROCESS_GROUP backend which is deprecated and slated to be removed and superseded by the TENSORPIPE backend. It is recommended to migrate to the TENSORPIPE backend. 
Nov 16 20:56:36 RPC was initialized with the PROCESS_GROUP backend which is deprecated and slated to be removed and superseded by the TENSORPIPE backend. It is recommended to migrate to the TENSORPIPE backend. 
Nov 16 20:56:36 RPC was initialized with the PROCESS_GROUP backend which is deprecated and slated to be removed and superseded by the TENSORPIPE backend. It is recommended to migrate to the TENSORPIPE backend. 

---

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 or post in the (internal) Dr. CI Users group.

See how this bot performed.

This comment has been revised 48 times.

[xuzhao9]

Neither mypy nor I can find from_buffer() function in torch.ByteStorage.

[xuzhao9]

torch.cuda.current_device() returns int, and torch.device() returns torch.device.

[xuzhao9]

If is_nccl_backend is set, current_device is an int, which will not be accepted by torch.zeros().

[xuzhao9]

If is_nccl_backend is set, current_device is an int, which will not be accepted by torch.empty().

[xuzhao9]

If is_nccl_backend is set, current_device is an int, which will not be accepted by torch.Tensor.to().

[facebook-github-bot]

@xuzhao9 merged this pull request in 915050e.