zp-json-hot — measured, still hint
Status
hint. Wall-clock JSONL: zp-json-hot.jsonl. Not corpus-triaged. The 2026-10-22 corpus verdict does not cover this slug.
Measurement
json.parseFromSlice of one small object, 2000 times, versus parsing once and reading the field. ReleaseFast. The result is kept so the after loop is not deleted.
| role | case | median_ms | ops | ratio |
|---|---|---|---|---|
| after | parse_once | 0.003 | 2000 | 1 |
| before | parse_each_iter | 7.043 | 2000 | 2683.175 |
A payload that changes every iteration cannot be hoisted. Default stays hint. This ratio is not a general speedup.
Unblock
Do not set the default to error from this file alone. A changing input, a cold path, or another ABI can erase the difference. Corpus triage is still required before any default change.