Repository navigation
ci(cache): write R2 from opted-out main lanes and fix the stats docs - #342
Conversation
The docs told readers to look for a Cache writes row, which sccache 0.17 does not print. The first main run against R2 showed what a writing run actually reports: Average cache write 0.463 s with Cache write errors at 0 across 4 misses, where the read-only pull request run had 4 write errors for the same 4 misses. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
Navigate logical layers of code changes, visualize relationships, and explore their blast radius. No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Advanced Run ID: 📒 Files selected for processing (1)
Included review availability: This review used your included allowance. Your plan provides up to 1 included review per hour; 0 remain after this review. 📝 WalkthroughWalkthroughThe build caching documentation updates the sccache 0.17 indicators for a writing run. It now specifies a nonzero ChangesBuild caching documentation
Priority: ⬇️ Low Estimated code review effort: 1 (Trivial) | ~3 minutes Change: Other Merge Risk: ⚪ Minimal · up to This documentation-only update accurately describes sccache 0.17 output and does not change build or cache behavior; no actionable merge risk remains. Architecture SummaryArchitecture risk: 🔵 Low · up to The change affects 1 system. Changed systems: Architecture concerns Review detailsSystems and components
Before / after behavior
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
Full details: Title checkExplanation The title correctly mentions the cache statistics documentation update, but it also claims that opted-out main lanes write to R2. The provided changeset contains only documentation changes and does not implement that CI behavior.
Comment |
The first main run against R2 showed E2E Build / CPU Smoke compiling 267 objects in READ_ONLY mode. The R2 write mode followed the same cacheWriter decision as the Actions cache, which honors save-if, and two lanes, python-generated and e2e-build-cpu, set save-if: "false" to keep duplicate target directories out of the Actions cache budget. R2 has no such budget, so those lanes recompiled on every run without ever filling the shared cache. R2 now writes from every trusted default-branch run regardless of save-if; pull requests and tags still read only. Those jobs already receive the read-write key on main through the sccache-writer environment, so no new run can write. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
3a9ae2c to
e01cafd
Compare
Two follow-ups from the first
mainrun against the R2 compiler cache (run 37057694286).🛠️ Opted-out lanes now write R2 on main
E2E Build / CPU Smoke compiled 267 objects on
maininREAD_ONLYmode. The R2 write mode followed the samecacheWriterdecision as the Actions cache, which honorssave-if, and two lanes (python-generated,e2e-build-cpu) setsave-if: "false"to keep duplicate target directories out of the Actions cache budget. R2 has no such budget, so those lanes recompiled every run without filling the shared cache. R2 now writes from every trusted default-branch run regardless ofsave-if; pull requests and tags still read only. Every cache job already receives the read-write key onmainthrough thesccache-writerenvironment, so this lets no new run write.📝 Name the right statistic
The docs told readers to check a
Cache writesrow, which sccache 0.17 does not print. A writing run shows a nonzeroAverage cache writewithCache write errors 0: Rust Test / Native App reported 0.463 s across 4 misses with no write errors, and UI wrote 423 objects with none.🧪 Validation
node --test .github/actions/rust-build-cache/*.test.mjswith sccache 0.17.0: 29 passed, including new cases that an opted-out lane writes onmainand stays read-only on a pull request.🤖 Generated with Claude Code