gp-make-hot — make([]byte) each iter vs reuse / single slice
Claim
Measured on this machine: primary hot case make_each_iter ratio before/after ~160.26× (faster after). Default severity stays off. Do not escalate to error from this fill alone — ratio is strong but policy/default and G3 OK-idiom history keep severity unchanged.
How measured
- Harness:
packages/goperf/examples/bench/evidence_gp.go - Protocol: warm=3; median of 9 rounds; GC/collect where applicable
- Raw lines:
gp-make-hot.jsonl
Runtime (this evidence file)
| Field | Value |
|---|---|
| When | 2026-10-23 Asia/Shanghai (machine clock; JSONL ts is UTC) |
| Toolchain | go1.27.3 |
Median after (reuse_buf) | 0.057417 ms (ops=200000, ratio=1.000000) |
Median before (make_each_iter) | 9.201458 ms (ops=200000, ratio=160.256684) |
Median after (single_slice_fill) | 0.362708 ms (ops=50000, ratio=1.000000) |
Median before (make_append_slices) | 2.214292 ms (ops=50000, ratio=6.104889) |
| Primary ratio before/after | ~160.26× (make_each_iter) |
Direction and absolute ms are N/machine/opt-sensitive. Re-run the harness after toolchain or machine changes.
Lint mapping
goperf-lint gp-make-hot default stays off. Escalate recommended: no.
Limits
- Microbench; not an app profile.
- Small-N rows may sit near timer noise; prefer hot (non-small) cases for ratio claims.
- I/O rules (print/echo/puts/writeln) discard to
/dev/null— env-dependent.