Breakwater

Backtest / #1118

perf: Cache Version hash value in dedicated slot

pypa/packaging#1118 · merged 2026-03-13 · base f2bbd4f578 → head 1676a10cda · ran on Docker

PASS

No behaviour divergence found by any probe.

Known regressions in this PR

Performance · Version

No realistic input family measured slower.

Measured in backtest packaging-v2 (same commits, run bt-1118-e98cb3, image bw-pair:dcd905da4078ada1). 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×input size (characters)F2F2: 1.01× at ~9 charsF2: 1.01× at ~9 charsF2: 1.01× at ~9 charsF2: 1.01× at ~10 charsF2: 1.01× at ~10 charsF2: 1.01× at ~10 charsF2: 1.01× at ~10 chars
  • F2
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
F2repeated_hash_same_versionsclaimed speedup · plan1.01× – 1.01×0.00 → 0.00neutral

Receipt

TargetBackendImageRun AuditAdjudication
packaging.version:Versionlocal Dockerbw-pair:4ccf3cc09b8d136bbt-1118-0b0b01$0.044$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.