Skip to content

Conversation

@mayeut
Copy link
Contributor

@mayeut mayeut commented Aug 25, 2025

This allows a speed-up of all QEMU targets by using a statically built clang that runs with the native architecture of the runner rather than calling gcc through QEMU.

CI should finish in roughly 30 minutes compared to 2h30 minutes in master.

This PR is mostly used as a PoC for this trick and see wether it would see adoption, what are the limitations and how it could be improved.

Clang used in this PR is built & published in the following repo: https://github.com/mayeut/static-clang-images

xref scikit-build/ninja-python-distributions#320

@henryiii henryiii merged commit 3df09e7 into scikit-build:main Nov 19, 2025
28 checks passed
@mayeut mayeut deleted the use-static-clang branch November 19, 2025 22:42
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