enricobuehler
658c6daa42
windows-client / client (arm64, --no-default-features, aarch64-pc-windows-msvc, C:\t-a64) (pull_request) Successful in 2m45s
apple / swift (pull_request) Successful in 2m5s
apple / distribute (pull_request) Skipped
apple / screenshots (pull_request) Skipped
macos-host / check (pull_request) Successful in 1m13s
windows-client / client (x64, , x86_64-pc-windows-msvc, C:\t) (pull_request) Successful in 7m10s
ci / web (pull_request) Successful in 2m3s
ci / docs-site (pull_request) Successful in 1m57s
ci / bun-nix (pull_request) Successful in 57s
secret-scan / gitleaks (pull_request) Successful in 27s
ci / docs-drift (pull_request) Failing after 44s
ci / rust-arm64 (pull_request) Successful in 5m26s
setup-binary / build (x86_64) (pull_request) Successful in 4m40s
setup-binary / build (aarch64) (pull_request) Successful in 4m45s
setup-binary / probe the published channel (pull_request) Skipped
setup-binary / publish to the generic registry (pull_request) Skipped
android / android (pull_request) Successful in 8m40s
nix / flake (pull_request) Successful in 9m55s
ci / rust (pull_request) Successful in 34m14s
The first both-vendor matrix run caught the newest-wins scan consuming an older, still-published slot after a newer one under --chaos. A stream must never show that frame: drop it (CAS PUBLISHED -> FREE) and count it, so WP7's reader rule is pinned by the spike that found it.