Draft
Conversation
- Add CI profile with thin LTO and parallel codegen (~2x faster builds) - Add Rust dependency caching with Swatinem/rust-cache - Upgrade to modern GitHub Actions (dtolnay/rust-toolchain, actions/checkout@v4) - Replace slow snap zig install with mlugg/setup-zig - Skip redundant Rust pre-commit checks (already run in rust.yml) - Use CI profile for PR/main builds, release profile for tags Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
prek requires these tools to be available for the pre-commit hooks. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Remove separate cargo build step - clippy and test share the same dev profile, so clippy compiles once and test reuses those artifacts. The CI profile is kept for Python wheel builds where optimized output matters. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Replace vendored OpenSSL with rustls (pure Rust TLS) - Removes ~30-60s of OpenSSL compilation from source - rustls compiles much faster and has no C dependencies - Add mold linker for faster linking on Linux - mold is significantly faster than the default linker Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
nextest is a faster test runner with better parallelism, typically 2-3x faster than cargo test. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Remove mold linker (saves ~12s, linking isn't the bottleneck) - Only test library, skip binary tests (cuts test count from 184 to 92) Expected time: ~60s with warm cache Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Use separate cache key for native builds (datago-native) - Set manylinux: off for PRs to skip auditwheel - Reduce test matrix to min/max Python versions (3.10, 3.13) - Combine pip install commands - Use quiet pytest output Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Set lto = false in CI profile (faster linking) - Set CARGO_INCREMENTAL=1 for incremental builds - Use opt-level = 2 instead of 3 Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
MarcoForte
force-pushed
the
ci/optimize-build-times
branch
from
27f8033 to
5989ead
Compare
MarcoForte
force-pushed
the
ci/optimize-build-times
branch
from
e8a0b2a to
bb52c97
Compare
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Add required-features = ["cli"] to binary target - Pass --no-default-features to maturin build - CLI (clap + prettytable) not needed for Python bindings Build time: 41s → 30s Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
astral-sh's fork has performance improvements and better security defaults. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
MarcoForte
force-pushed
the
ci/optimize-build-times
branch
from
3d4b212 to
3b20968
Compare
Collaborator
|
I went a bit overboard in the rust build options for speed, but for real when using this from python I don't think that the bottleneck and we could relax them. I'm thinking lto specifically in Cargo.toml / profile.release, I think it adds a fair bit to the build time |
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
2 participants
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.
playing around with claude code, asking to loop over improving cicd times iteratively.
Maybe some optimisations are useful.