Add cpu_thread setting logic to xnn_executor_runner#8902
Conversation
|
Stack from ghstack (oldest at bottom): |
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/8902
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit ddabcf7 with merge base 09ad20a ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
mcr229
left a comment
There was a problem hiding this comment.
this seems fine, but honestly coming back to this after year+ since our first release, do we need to have an xnn_executor_runner which is separate from the executor_runner? Should our CMake just be if BUILD_XNNPACK=ON, we link XNNPACK to executor_runner, if not then we dont?
sounds like a great idea to me! this change is still an incremental improvement though so I will land it. On another note, CI is green, rebasing to make sure extra changes from previous commit go away and then merging without waiting for CI to run again |
No description provided.