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.

Sign up for GitHub

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

Jump to bottom

common: delay ps initialization for better RDMA compatibility #91

Merged
merged 5 commits into from
Sep 23, 2019
Merged

common: delay ps initialization for better RDMA compatibility #91

merged 5 commits into from
Sep 23, 2019

Conversation

bobzhuyb
Copy link
Member

No description provided.

@bobzhuyb bobzhuyb requested a review from ymjiang August 30, 2019 12:34
@bobzhuyb bobzhuyb changed the title delay ps initialization for better RDMA compatibility common: delay ps initialization for better RDMA compatibility Aug 30, 2019
ymjiang
ymjiang approved these changes Sep 23, 2019
View reviewed changes
Copy link
Member

@ymjiang ymjiang left a comment •
edited
Loading

Choose a reason for hiding this comment

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

Tested. With this PR, RDMA training can run normally even if adding os.fork() after bps.init() (baseline will crash).

@ymjiang ymjiang merged commit f18f1a8 into master Sep 23, 2019
@ymjiang ymjiang deleted the delay_ps_init branch September 23, 2019 07:51
@DeruiLiu DeruiLiu mentioned this pull request Aug 6, 2020
Closed
This pull request was closed.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@ymjiang ymjiang ymjiang approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@bobzhuyb @ymjiang