Mousius commented on pull request #9733: URL: https://github.com/apache/tvm/pull/9733#issuecomment-1006442787
hi @driazati / @areusch, I can't see the output in the other Jenkins as I'm not authorised but based on the Jenkinsfile this seems like a good thing to just try :smile_cat: CI already gets super bogged down with many builds so I don't think there'll be a real difference to that situation by implementing this, but the happy path for when CI has capacity will be vastly improved leading to potentially less builds building up anyway. The later stages will be using more GPU instances which are currently never used unless we get that far in the pipeline, so this should be fairly safe and there's always the revert button if it really performs poorly. I would definitely advocate for #9129 as well to maximize free capacity but that can be a future addition once we've seen how this works in practice. P.S. It'd be nice to see these all as GitHub checks in parallel as well :smile_cat: -- 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. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
