Disable DMA transpose on attention forward kernel #42
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
While we are stabilizing the DMA transpose
|
Please download and review the generated API doc at https://github.com/aws-neuron/nki-samples/actions/runs/12696429583 |
aws-qieqingy
reviewed
Jan 9, 2025
| @@ -44,20 +44,13 @@ class FlashConfig: | |||
| @nki.jit(mode='trace') | |||
| def transpose_p_local(p_local_transposed, p_local, LARGE_TILE_SZ): | |||
| for i in nl.affine_range(LARGE_TILE_SZ // 512): | |||
| if nisa.get_nc_version() == nisa.nc_version.gen3: | |||
There was a problem hiding this comment.
could we have a comment on why is the DMA transpose is disabled?
|
Internal testing for the PR passed. Ready to merge with another approval. |
aws-qieqingy
approved these changes
Jan 20, 2025
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Outdated suggestions cannot be applied.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.
Issue #, if available:
Description of changes:
Disable DMA transpose on attention forward kernel while we are stabilizing the DMA transpose in trn2
Testing:
Please see detailed unit test requirements in the CONTRIBUTING.md
nki.baremetalnki.benchmarkPull Request Checklist