at-alloc-hot — new / alloc in loop (Node proxy)
Status: measured (node-proxy)
Claim
Hot per-iteration new Uint8Array is slower than reusing one buffer on Node/V8 (proxy for ArkTS alloc-in-loop smell).
How measured
-
Harness:
examples/bench/evidence_at001_at003.js -
runtime: node-proxy — not ArkTS VM / device SoftAP
-
warm=3; median of 9; N=200000
-
Raw lines:
at-alloc-hot.jsonl
Runtime (this evidence file)
| Field | Value |
|---|---|
| When | 2026-10-22 Asia/Shanghai (JSONL ts is UTC) |
| runtime | node-proxy |
| node | v20.19.2 |
| v8 | 11.3.244.8-node.38 |
Median before (new_in_loop) | ~4.768 ms |
Median after (reuse_buffer) | ~0.1461 ms |
| ratio before/after | ~32.6× |
Lint mapping
arktsperf-lint at-alloc-hot: severity stays package default (see evidence-policy). JSONL is Node proxy, not device.
Limits
- Proxy only; ArkTS/ArkRuntime may differ.
- Default stays off (G3 OK-idiom); no escalate.