trunk/353bbb744f8d1c4b215c4e1d84205c8348a38354: [caffe2/veclib] Add reduce_max to NEON's f32 layer (#196716)
- PyTorch: 786 events in the last 90 days
- PyTorch: 776th Release in the last 90 days
- Previous: earlier the same day · trunk/31527a43dbf6adf4df2e6eb5c7b38094fec6b6f6: Widen FSDP param group stream type annotations to torch.Stream (#196563)
What happened
Summary: Add NEON intrinsic-based Vectorized::reduce_max(), otherwise implementation fallbacks to vec_base.h Test Plan: CI Differential Revision: D119614281 Pull Request resolved: #196716 Approved by: https://github.com/Skylion007 , https://github.com/fadara01
Summary assembled by rule from the sources below