-
Notifications
You must be signed in to change notification settings - Fork 3.6k
Fix CI failures due to Horovod and MPI dependencies #12330
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
Conversation
|
Seems to be parallel effort with #12318 |
|
@Borda How do you force the GPU CI to use the changed docker build? |
No, because it doesn't compile correctly yet... |
carmocca
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is how you force the Docker images to get rebuilt and pushed to dockerhub:
After they finish, you revert the above commit and CI should pull the image you just generated.
|
👀 |
This reverts commit f3a9440.
|
My idea of including missing mpi libs did not work. |
“You think I’ll weep?
No, I’ll not weep. Storm and tempest.” |
| HOROVOD_WITHOUT_TENSORFLOW=1 \ | ||
| HOROVOD_WITHOUT_MXNET=1 \ | ||
| HOROVOD_WITH_GLOO=1 \ | ||
| HOROVOD_WITHOUT_MPI=1 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Isn't this part of the problem?
What does this PR do?
Tries to fix #12314
Does your PR introduce any breaking changes? If yes, please list them.
Before submitting
PR review
Anyone in the community is welcome to review the PR.
Before you start reviewing make sure you have read Review guidelines. In short, see the following bullet-list:
Did you have fun?
Make sure you had fun coding 🙃