Defer pg agent listener thread until contexts are initialized #28013
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
ProcessGroupAgent currently kicks off the listener thread in its constructor. However, serving requests requires contexts to be initialized, e.g., RRefContext and agent_ global var in api.py, which might not be done yet when the first request arrives. ProcessGroupAgent does not know what would be the appropriate time to start the listener thread, hence exposing an API for higher layer code to explicitly start listeners. [ghstack-poisoned]
|
@pytorchbot retest this please |
xush6528
reviewed
Oct 15, 2019
| @@ -103,7 +103,8 @@ PyObject* rpc_init(PyObject* /* unused */) { | |||
| &ProcessGroupAgent::sync, | |||
| py::call_guard<py::gil_scoped_release>()); | |||
|
|
|||
| module.def("_init_rpc_agent", [](std::shared_ptr<RpcAgent> agent) { | |||
| module.def("_start_rpc_agent", [](std::shared_ptr<RpcAgent> agent) { | |||
| agent->start(); | |||
There was a problem hiding this comment.
Shouldn't we reverse this order?
|
LGTM ! |
…zed" ProcessGroupAgent currently kicks off the listener thread in its constructor. However, serving requests requires contexts to be initialized, e.g., RRefContext and agent_ global var in api.py, which might not be done yet when the first request arrives. ProcessGroupAgent does not know what would be the appropriate time to start the listener thread, hence exposing an API for higher layer code to explicitly start listeners. Differential Revision: [D17932271](https://our.internmc.facebook.com/intern/diff/D17932271) [ghstack-poisoned]
…zed" ProcessGroupAgent currently kicks off the listener thread in its constructor. However, serving requests requires contexts to be initialized, e.g., RRefContext and agent_ global var in api.py, which might not be done yet when the first request arrives. ProcessGroupAgent does not know what would be the appropriate time to start the listener thread, hence exposing an API for higher layer code to explicitly start listeners. Differential Revision: [D17932271](https://our.internmc.facebook.com/intern/diff/D17932271) [ghstack-poisoned]
mrshenli
added a commit
that referenced
this pull request
Oct 15, 2019
ProcessGroupAgent currently kicks off the listener thread in its constructor. However, serving requests requires contexts to be initialized, e.g., RRefContext and agent_ global var in api.py, which might not be done yet when the first request arrives. ProcessGroupAgent does not know what would be the appropriate time to start the listener thread, hence exposing an API for higher layer code to explicitly start listeners. ghstack-source-id: 86f4c71 Pull Request resolved: #28013
thiagocrepaldi
pushed a commit
to thiagocrepaldi/pytorch
that referenced
this pull request
Feb 4, 2020
…h#28013) Summary: Pull Request resolved: pytorch#28013 ProcessGroupAgent currently kicks off the listener thread in its constructor. However, serving requests requires contexts to be initialized, e.g., RRefContext and agent_ global var in api.py, which might not be done yet when the first request arrives. ProcessGroupAgent does not know what would be the appropriate time to start the listener thread, hence exposing an API for higher layer code to explicitly start listeners. Test Plan: Imported from OSS Differential Revision: D17932271 Pulled By: mrshenli fbshipit-source-id: 3b408477594d4d19319e7cd08dd6f383a7ed7670
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Stack from ghstack:
ProcessGroupAgent currently kicks off the listener thread in its
constructor. However, serving requests requires contexts to be
initialized, e.g., RRefContext and agent_ global var in api.py,
which might not be done yet when the first request arrives.
ProcessGroupAgent does not know what would be the appropriate time
to start the listener thread, hence exposing an API for higher
layer code to explicitly start listeners.
Differential Revision: D17932271