Repository navigation
chore: bump cubecl and cubek to main - #6221
Merged
Merged
Conversation
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #6221 +/- ##
==========================================
- Coverage 66.16% 66.16% -0.01%
==========================================
Files 878 878
Lines 127748 127748
==========================================
- Hits 84528 84526 -2
- Misses 43220 43222 +2 ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Checklist
cargo run-checkshas been executed.changelog policy in CONTRIBUTING.md.
Related Issues/PRs
Changes
Pins cubecl to
9f8a9cc6and cubek tod28d5532(current mains). Only the cubecl and cubek source lines change inCargo.lock.e2m1) casts lower on every LLVM target; a CUDA or HIP device without a nativee2m1x2conversion reports it as emulated, so NVFP4 codes decode asf16pairs, and the mma fragment loads accept that read.k, instead of panicking when it binds.Testing
cargo check --locked --workspace --all-targets: passes.cargo run-checks. The cubecl and cubek changes were tested in their own PRs (ROCm gfx1151 and CUDA L4).AI Usage
Claude Code carried out the bump autonomously (the pins, the lockfile edit), ran the check above, and wrote this description and created the PR.
🤖 Generated with Claude Code