edrevo commented on pull request #9523: URL: https://github.com/apache/arrow/pull/9523#issuecomment-786911328
@seddonm1, thanks for the info around the tests you have made with tokio::spawn. I would be interested in a repro case for the behavior you are seeing where the last thread isn't scheduled and there's no usage of channels. Out of curiosity, to better understand tokio's internals. In my case, I was able to reproduce the hang but after changing to tokio's mspc the hang wasn't there anymore. Maybe I just got lucky, or maybe that change minimized the chance of a hang but doesn't eliminate it completely. You never know with these threading issues... Adding an explicit yield call sounds fine to me, tbh, so from my point of view, leaving that call in the code is fine. ---------------------------------------------------------------- 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
