Add trailing dims memoization to improve performance of permute_copy_out (#5246)
Summary: Pull Request resolved: https://github.com/pytorch/executorch/pull/5246 Otherwise we recompute trailing dims every time we call coordinateToIndex. ghstack-source-id: 242067817 exported-using-ghexport Reviewed By: mergennachin Differential Revision: D62154216 fbshipit-source-id: a690230892b6eca8dcd14dfd859c1195b16d231c
S
Scott Wolchok committed
17cf7826207915d3309ca729961454d0cfca220c
Parent: 5e7efe6
Committed by Facebook GitHub Bot <facebook-github-bot@users.noreply.github.com>
on 9/12/2024, 3:07:16 AM