← 返回事件
持续讨论AI发版

trunk/ff30875f25530b355ab830bbde00b7a5363bafa4: [inductor] Extend partitioned scatter to additional ops (#194078)

图:PyTorch Releases

发生了什么

The partitioned scatter pass only matched aten.index_put / aten.index_put_ , so it missed ops that emit the exact same contended atomic_add store. This extends the pass to cover them and tightens the partition cap so the rewrite cannot cost more memory traffic than the scatter it replaces. Adds support for: _unsafe_masked_index_put_accumulate lowering aten.scatter_add , aten.scatter_reduce.two , aten.scatter.reduce…

摘要按规则整理自下方来源原文

为什么在扩散

来源