Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix sendrecv ut #6983

Merged
merged 7 commits into from
Dec 26, 2017
Merged

Conversation

typhoonzero
Copy link
Contributor

fix #6978
fix #6893

Copy link
Contributor

@Yancey1989 Yancey1989 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I haven't seen the code about how to deal with SIGINT in the send side(trainer), or maybe you have any other idea?

@@ -57,7 +59,10 @@ class RecvOp : public framework::OperatorBase {
}
}

virtual ~RecvOp() {
void Stop() override {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Maybe we also need the destructor function, and call Stop() in it.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

No. recv_op's Run may run in another thread (in unit test), Stop must be called out side dtor

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

So we also need to add the interface to pybind , maybe we coud do that in another PR.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yep. We can add it later when running in different clusters.

Yancey1989
Yancey1989 previously approved these changes Dec 26, 2017
Copy link
Contributor

@Yancey1989 Yancey1989 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

Copy link
Contributor

@Yancey1989 Yancey1989 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM++

@typhoonzero typhoonzero merged commit 4cade60 into PaddlePaddle:develop Dec 26, 2017
@typhoonzero typhoonzero deleted the fix_sendrecv_ut branch December 27, 2017 02:03
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

send recv unit test failed at develop branch recv_op need graceful shutdown
2 participants