Conversation
Every CUDA kernel family (persist-D / split-D / M4N2 across fp8 / fp4 / fp16) now reads NHD gmem natively -- packed-NHD and strided fused-QKV chunk views alike -- so the _keep_or_pack materialization fallback has no legitimate trigger left: tensors outside the NHD layout contract are contract violators and now fail loudly in the C++ layout gates instead of silently paying a copy. Also drops the is_nhd_zero_copy_input probe that only served the old-build materialization path. Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Every CUDA kernel family (persist-D / split-D / M4N2 across fp8 / fp4 / fp16) now reads NHD gmem natively -- packed-NHD and strided fused-QKV chunk views alike -- so the _keep_or_pack materialization fallback has no legitimate trigger left: tensors outside the NHD layout contract are contract violators and now fail loudly in the C++ layout gates instead of silently paying a copy. Also drops the is_nhd_zero_copy_input probe that only served the old-build materialization path.