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

[fleet_executor] Complete compute interceptor #37485

Merged

Conversation

wangxicoding
Copy link
Contributor

PR types

Others

PR changes

Others

Describe

Compute interceptor 完成上下游消息依赖与消息发送。

@paddle-bot-old
Copy link

Thanks for your contribution!
Please wait for the result of CI firstly. See Paddle CI Manual for details.

Copy link
Contributor

@FeixLiu FeixLiu left a comment

Choose a reason for hiding this comment

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

LGTM

void Compute(const InterceptorMessage& msg);

private:
std::unordered_set<int64_t> upstream_deps_;
// FIXME(wangxi): if use step_ and max_steps_, how to restart step_ from 0
Copy link
Contributor

Choose a reason for hiding this comment

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

There is a message RESET, which was designed to reset the step_ to 0.

Copy link
Contributor Author

Choose a reason for hiding this comment

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

那感觉还得加一个start的interceptor,发送开始信息,然后到了指定micro_step后,再发reset信息。

@FeixLiu
Copy link
Contributor

FeixLiu commented Nov 24, 2021

现在这么设计是不是停不下来呀,需要等那个step_的开发?outbuffer和inready都可以根据RESET消息重置吧?

@wangxicoding
Copy link
Contributor Author

现在这么设计是不是停不下来呀,需要等那个step_的开发?outbuffer和inready都可以根据RESET消息重置吧?

是的,停不下来,reset不能把interceptor的线程停下来,只有stop可以。

@wangxicoding wangxicoding merged commit be3b774 into PaddlePaddle:develop Nov 24, 2021
@wangxicoding wangxicoding deleted the complete_compute_interceptor branch November 24, 2021 02:46
Zjq9409 pushed a commit to Zjq9409/Paddle that referenced this pull request Dec 10, 2021
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.

2 participants