Breakwater

Backtest / #1115

perf: skip hash property

pypa/packaging#1115 · merged 2026-03-11 · base e839d73030 → head baa9ea8c7f · ran on Docker

PASS

No behaviour divergence found by any probe.

Performance · Version

No realistic input family measured slower.

Measured in backtest packaging-v2 (same commits, run bt-1115-cc92fd, image bw-pair:c5e7d6dc99b58cc8). The semantic results above are from a later run; perf doesn't use the semantic planner, so it was not re-measured.

head slowerhead faster1×1001,00010,000input size (characters)F5F5: 1.00× at ~36 charsF5: 1.00× at ~54 charsF5: 1.00× at ~176 charsF5: 1.00× at ~704 charsF5: 1.00× at ~3,130 charsF5: 1.00× at ~13,164 charsF5: 1.00× at ~59,244 chars
  • F5
Head/base cachegrind instructions per call, by input size (log-log). Each line is one input family Nemotron proposed; points above 1× are where the "faster" code does more work.
FamilyInputsRatio range Slope base → headVerdict
F5many distinct complex version strings (epoch, many release segments, pre, post, dev, long local) hashed once eachclaimed speedup · plan1.00× – 1.00×0.86 → 0.86neutral

Receipt

TargetBackendImageRun AuditAdjudication
packaging.version:Versionlocal Dockerbw-pair:b75c39f11bbb1fbfbt-1115-bf552c$0.041$0.000

Nemotron cost on Nebius Token Factory. Backend: where the untrusted code ran. local Docker (an image cited by its content-addressed tag). Every image is content-addressed (repo at each commit + harness), so every card re-runs from the image and seed 0. Raw data: audit.