pytorch
4f2d8690 - Fix distributed issue by including distributed files (#87615)
Commit
1 year agoFix distributed issue by including distributed files (#87615)
This fixes regression in distributed headers installation.
Caused by following PR: https://github.com/pytorch/pytorch/pull/85953
which removed the inclusions
Fixes #87173
Test plan from wheel build by this CI: https://github.com/pytorch/pytorch/actions/runs/3314742519
```
[ec2-user@ip-10-0-9-132 c10d]$ pwd
/home/ec2-user/actions-runner/_work/_temp/artifacts/torch/include/torch/csrc/distributed/c10d
[ec2-user@ip-10-0-9-132 c10d]$ ls -las
total 300
4 drwxr-xr-x 2 ec2-user ec2-user 4096 Oct 24 19:12 .
0 drwxr-xr-x 4 ec2-user ec2-user 29 Oct 24 19:12 ..
12 -rw-r--r-- 1 ec2-user ec2-user 9051 Oct 24 17:28 Backend.hpp
4 -rw-r--r-- 1 ec2-user ec2-user 216 Oct 24 17:28 c10d.h
4 -rw-r--r-- 1 ec2-user ec2-user 3880 Oct 24 17:28 comm.hpp
4 -rw-r--r-- 1 ec2-user ec2-user 604 Oct 24 17:28 debug.h
4 -rw-r--r-- 1 ec2-user ec2-user 1717 Oct 24 17:28 default_comm_hooks.hpp
4 -rw-r--r-- 1 ec2-user ec2-user 1316 Oct 24 17:28 error.h
4 -rw-r--r-- 1 ec2-user ec2-user 962 Oct 24 17:28 exception.h
4 -rw-r--r-- 1 ec2-user ec2-user 1461 Oct 24 17:28 FileStore.hpp
4 -rw-r--r-- 1 ec2-user ec2-user 771 Oct 24 17:28 GlooDeviceFactory.hpp
4 -rw-r--r-- 1 ec2-user ec2-user 1154 Oct 24 17:28 HashStore.hpp
4 -rw-r--r-- 1 ec2-user ec2-user 4058 Oct 24 17:28 logger.hpp
4 -rw-r--r-- 1 ec2-user ec2-user 2059 Oct 24 17:28 logging.h
8 -rw-r--r-- 1 ec2-user ec2-user 7979 Oct 24 17:28 NCCLUtils.hpp
4 -rw-r--r-- 1 ec2-user ec2-user 2756 Oct 24 17:28 Ops.hpp
4 -rw-r--r-- 1 ec2-user ec2-user 1814 Oct 24 17:28 ParamCommsUtils.hpp
4 -rw-r--r-- 1 ec2-user ec2-user 1478 Oct 24 17:28 PrefixStore.hpp
16 -rw-r--r-- 1 ec2-user ec2-user 13235 Oct 24 17:28 ProcessGroupGloo.hpp
12 -rw-r--r-- 1 ec2-user ec2-user 11298 Oct 24 17:28 ProcessGroup.hpp
12 -rw-r--r-- 1 ec2-user ec2-user 8645 Oct 24 17:28 ProcessGroupMPI.hpp
28 -rw-r--r-- 1 ec2-user ec2-user 26526 Oct 24 17:28 ProcessGroupNCCL.hpp
4 -rw-r--r-- 1 ec2-user ec2-user 3805 Oct 24 17:28 ProcessGroupRoundRobin.hpp
12 -rw-r--r-- 1 ec2-user ec2-user 10361 Oct 24 17:28 ProcessGroupUCC.hpp
8 -rw-r--r-- 1 ec2-user ec2-user 5062 Oct 24 17:28 ProcessGroupWrapper.hpp
8 -rw-r--r-- 1 ec2-user ec2-user 4201 Oct 24 17:28 PyProcessGroup.hpp
4 -rw-r--r-- 1 ec2-user ec2-user 1072 Oct 24 17:28 python_comm_hook.h
24 -rw-r--r-- 1 ec2-user ec2-user 23859 Oct 24 17:28 reducer.hpp
4 -rw-r--r-- 1 ec2-user ec2-user 2330 Oct 24 17:28 reducer_timer.hpp
4 -rw-r--r-- 1 ec2-user ec2-user 1683 Oct 24 17:28 sequence_num.hpp
4 -rw-r--r-- 1 ec2-user ec2-user 2108 Oct 24 17:28 socket.h
4 -rw-r--r-- 1 ec2-user ec2-user 2589 Oct 24 17:28 Store.hpp
4 -rw-r--r-- 1 ec2-user ec2-user 3264 Oct 24 17:28 TCPStore.hpp
8 -rw-r--r-- 1 ec2-user ec2-user 6944 Oct 24 17:28 TraceUtils.h
8 -rw-r--r-- 1 ec2-user ec2-user 4539 Oct 24 17:28 Types.hpp
4 -rw-r--r-- 1 ec2-user ec2-user 580 Oct 24 17:28 UCCForNCCL.hpp
4 -rw-r--r-- 1 ec2-user ec2-user 2301 Oct 24 17:28 UCCTracing.hpp
8 -rw-r--r-- 1 ec2-user ec2-user 4933 Oct 24 17:28 UCCUtils.hpp
4 -rw-r--r-- 1 ec2-user ec2-user 584 Oct 24 17:28 UnixSockUtils.hpp
24 -rw-r--r-- 1 ec2-user ec2-user 20796 Oct 24 17:28 Utils.hpp
4 -rw-r--r-- 1 ec2-user ec2-user 575 Oct 24 17:28 WinSockUtils.hpp
8 -rw-r--r-- 1 ec2-user ec2-user 4259 Oct 24 17:28 Work.hpp
```
Pull Request resolved: https://github.com/pytorch/pytorch/pull/87615
Approved by: https://github.com/malfet
Author
Committer
Parents
Loading