trunk/8376f7cabcc19198d1a827b2eb0c4a9e57137381: [inductor] Hold GPU lock across multi-kernel comparison (#192876)
- PyTorch 近 90 天出现 185 次
- PyTorch 近 90 天第 182 次发版
- 上一次：同一天稍早 · trunk/76588348eb3e6d543e06f04751268a19f57bca8d: Fix post-local SGD checkpoint reload race (#194614)
发生了什么
Summary Benchmark all candidates in a runtime multi-kernel comparison under one registered GPU benchmark lock. This prevents concurrent compile workers from interleaving candidate measurements on the same GPU. Existing per-candidate benchmark locking remains nested through the reentrant lock contract. Candidate ordering, forced and cached choices, and benchmark failure behavior are unchanged. Test plan python -m pyt…
摘要按规则整理自下方来源原文