Lunderberg commented on pull request #8196: URL: https://github.com/apache/tvm/pull/8196#issuecomment-857142889
On second thought, it looks like that difference was because `pytest-benchmark` does not include warmup runs by default. That run just happened to be the first that required initialization, which messed up some of the timing. With warmup runs, the performance is matched before/after the refactor. ![image](https://user-images.githubusercontent.com/3888575/121256884-e425d680-c861-11eb-9833-fe85c87ca30b.png) -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org