TODO: Use threading lock to protect unsafe tensor moves in multi-gpu #1778
Conversation
This file contains hidden or 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
|
@LRL-ModelCloud Pleasea also test this PR asap and I will revert if there are issues. |
|
@Qubitium it works OK |
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Outdated suggestions cannot be applied.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.
@avtc Please test this code which should prevent the cuda async assert errors in 10 gpu + moe + mock quant. Pretty sure there is a limit within cuda that only allows 1 sync operation per device.
Oof.. I merged this PR by mistake. Please test and if ok, I will leave it be and if not, I will revert.