Skip to content

fix(build): suppress MLX folding warning after flags - #24

Merged
mudler merged 1 commit into
mudler:mainfrom
localai-org-maint-bot:cron/fix-mlx-folding-root
Aug 2, 2026
Merged

fix(build): suppress MLX folding warning after flags#24
mudler merged 1 commit into
mudler:mainfrom
localai-org-maint-bot:cron/fix-mlx-folding-root

Conversation

@localai-org-maint-bot

Copy link
Copy Markdown
Contributor

Summary

  • move the narrow AppleClang -Wgnu-folding-constant suppression into the MLX-only Objective-C++ translation unit
  • remove the ineffective CMake source option that target-level -Werror re-promotes
  • keep every other warning fatal and record the Darwin build-verification checkpoint

This follows up #23 after LocalAI PR mudler/LocalAI#11137 still reported two MLX-header folding diagnostics in Darwin CI.

Verification

  • git diff upstream/main...HEAD --check
  • python3 scripts/check-doc-checkpoint.py --base upstream/main --head HEAD (OK)
  • source invariant: exactly one translation-unit pragma; no folding-warning exception remains in CMake

AppleClang and CMake are unavailable on this Linux cron host, so the Darwin CI lane is the binding compiler verification.

Move the narrow AppleClang diagnostic suppression into the MLX-only translation unit so target-level -Werror cannot re-promote it. Keep all other warnings fatal.

FOLLOWING_AGENTS_PROTOCOL

Assisted-by: Codex:gpt-5 [systematic-debugging]
@mudler
mudler merged commit 3f32c84 into mudler:main Aug 2, 2026
7 of 10 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants