Linalg/XeGPU nanoGPT forward-pass example#196
Draft
nbpatel wants to merge 6 commits into
Draft
Conversation
The example was discovered by lit but had no RUN line, causing the
pre-commit suite to report UNRESOLVED ("Test has no 'RUN:' line").
Add a RUN line that dumps the xegpu-wg IR (1 layer, host-side, no GPU)
and FileChecks the module header, matching the sibling xegpu examples.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Contributor
|
Neat, thanks for contributing this! The file is rather long, maybe it'd be clearer if split into pieces, e.g., separate file for the main |
Contributor
Author
makes sense. Done |
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This PR adds a nanoGPT model script under examples/xegpu, a full GPT-2-style forward pass (6 layers, C=256, H=4, T=256) running end-to-end on the Intel GPU via XeGPU lowering path, with multi-head attention computed by a single fused flash-attention kernel per block (online-softmax). The model lowers to one MLIR module; verified against a numpy reference.
Architecture follows the GPT-2 block stack from Andrej Karpathy's nanoGPT, adapted here for testing XeGPU path
Assisted by Claude Code