Revert "[CUDA] Enable XQA by default for FP16/BF16 GQA (#29046)"#29217
Closed
hariharans29 wants to merge 1 commit into
Closed
Revert "[CUDA] Enable XQA by default for FP16/BF16 GQA (#29046)"#29217hariharans29 wants to merge 1 commit into
hariharans29 wants to merge 1 commit into
Conversation
This reverts commit 6be94de.
Member
Author
|
Closing as the root cause for main branch CI failure is NOT enabling XQA by default. It seems like the tfailing test is using fp32 inputs which is falling back to CPU and exposing the bug that it being fixes by #29069 |
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Description
This reverts commit 6be94de.
It is currently breaking a test in Linux x64 CUDA CI:
This test itself was introduced in PR #29002 which was merged last night but its CI ran 5 days ago and that test passed on CI. When the PR #29046 was in CI - that particular test did not exist and so #29046 also "passed" CI. But these 2 PRs in combination now break main branch and block other PR merges.
Sample PRs where it fails the Linux CUDA CIs:
#29200
#29207
Motivation and Context
Unblock PR merges into main as Linux CUDA CI is red in main branch now