Deprecation warning for ucx_net_devices='auto' on UCX 1.11+ #681
Conversation
This file contains 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
pentschev
added
3 - Ready for Review
|
cc @charlesbluca this should prevent UCX from hanging even when specifying |
quasiben
approved these changes
Jul 23, 2021
|
@gpucibot merge |
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Outdated suggestions cannot be applied.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.
Add deprecation warning for
ucx_net_devices='auto'on UCX 1.11+, fallback to UCX's default when that's requested.Using
ucx_net_devices='auto'can cause hangs, this is because it would need to be updated to include both the IPoIB and MLX interfaces inUCX_NET_DEVICES. Given UCX can now successfully handle automatic CUDA<->IB mapping, -- including on DGX A100, which this option wasn't capable of handling -- there's no reason for us to support that anymore and deprecating it should be the best approach at this time.