[NPU]: Add NPU support for the mrope operator #992
Merged
+293
−2
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Outdated suggestions cannot be applied.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.
Summary
Mainly to complete the adaptation of the _triton_qwen2vl_mrope operator on the NPU:
1、Solving the operator ub overflow problem
2、Resolves issues with some concat operations not supported by ascend.
3、Fixed legacy issues with the rope operator
Testing Done
Verified on Ascend NPU 910B4:
MROPE forward and backward pass tests
make testto ensure correctnessmake checkstyleto ensure code stylemake test-convergenceto ensure convergence