[Draft]: Add TileLang torch.compile backend - #2763
Conversation
|
Important Review skippedDraft detected. Please check the settings in the CodeRabbit UI or the ⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Pro Plus Run ID: You can disable this status message by setting the Use the checkbox below for a quick retry:
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
|
👋 Hi! Thank you for contributing to the TileLang project. Please remember to run We appreciate you taking this step! Our team will review your contribution, and we look forward to your awesome work! 🚀 |
1020e04 to
fc67141
Compare
Rename the TileLang Relax pass to LowerPrimitiveFunctionsToTIR and update its API, FFI, tests, and graph pipeline references. Relocate the schedule template design document and regression runner alongside the related examples and maintenance scripts.
Rename ReserveRootBlock to CanonicalizeScheduledTIR across the C++ pass, FFI and Python APIs, graph scheduling helpers, tests, and documentation.
Move graph-owned lowering and schedule-template checks into testing/python/graph, update their internal paths, and remove the redundant transformed-reduction issue test.
Rename the graph workload directory to tests/inductor, move the softmax workloads into its root, and update the local regression runner paths.
Remove the schedule-specific non-fragment accumulator fallbacks from MMA and WGMMA lowering, restoring all three CUDA GEMM files to their mainline behavior.
2 participants
No description provided.