ciflow/inductor/200700: [inductor] Run very large short reductions outside combo kernels
- PyTorch 近 90 天出现 2241 次
- PyTorch 近 90 天第 2212 次发版
- 上一次:同一天稍早 · ciflow/torchtitan/200591
发生了什么
Summary: The combo kernel partitioner already gives a reduction its own kernel when numel * rnumel exceeds LARGE_NUMELS , but only for long reductions (rnumel > 2048). A large short reduction still shares a combo kernel with small ones and inherits their config. For an all-persistent combo, select_combo_heuristics takes the configs of the sub-kernel with the largest rnumel, which is usually a small one. In a product…
摘要按规则整理自下方来源原文