Repository navigation
fix: layer split across cuda - #2262
Conversation
|
No actionable comments were generated in the recent review.
|
| Layer / File(s) | Summary |
|---|---|
Guard node assignments src/core/layer_split_partition.cpp |
A shared helper skips invalid indices and null nodes. It assigns the current segment’s backend only when the node has no existing assignment. |
Estimated code review effort: 2 (Simple) | ~10 minutes
Change: Bug fix
Possibly related issues
- leejet/stable-diffusion.cpp#2225 — The issue describes illegal CUDA memory access during multi-CUDA layer splitting, which is related to the partition assignment behavior changed here.
Suggested reviewers: leejet
Merge Risk: 
Low · up to 5b4f4
-
Some valid multi-CUDA graphs may incur extra cross-device transfers and work, but execution remains correct and the impact is limited to shared-producer graph shapes.
Pre-merge checks: 
4 · 
1
-
Failed checks: 1 warning Passed checks: 4 passed-
Full details: Description check
-
Explanation
The description includes the required Summary, Related Issue / Discussion, Additional Information, and Checklist sections. It explains the goal, links the related issue, and confirms the contribution guidelines checklist.
-
Full details: Docstring Coverage
-
Explanation
Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff with cyclomatic complexity of at least 2. Documentation-only edits and deletions within surviving functions are included; unchanged diff context and fully deleted functions are excluded. Analyzed 1 functions across 1 files.
Resolution
Write docstrings for the functions missing them to satisfy the coverage threshold.
-
Finishing Touches
-
Generate unit tests (beta)
-
- Create a new PR
-
- Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts
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.
|
Thank you for your contribution. |
Summary
Allow split backend over multiple cuda gpus
Related Issue / Discussion
#2225
Additional Information
Checklist
Summary by CodeRabbit