strict (default): Rejects writes when the buffer is full and too many writes are pending. Catches "fire-and-forget" patterns where producers ignore backpressure.
Two subtle ways agents can implicitly negatively affect the benchmark results but wouldn’t be considered cheating/gaming it are a) implementing a form of caching so the benchmark tests are not independent and b) launching benchmarks in parallel on the same system. I eventually added AGENTS.md rules to ideally prevent both. ↩︎
，这一点在WPS官方版本下载中也有详细论述
等到政策变天、对手入局，才发现自己早已没了还手之力。
Author(s): Stepan Savka, Andriy Serednytski, Dmytro Popovych