dongjoon-hyun commented on PR #58054: URL: https://github.com/apache/spark/pull/58054#issuecomment-5361511683
Thank you, @viirya , for the LGTM and for the independent sweep of the zero-slot consumers -- confirming that barrier and pipelined gang admission are the only terminal readers of that state is exactly the assurance the scope decision needed. Noted the non-blocking coverage remark; I'll keep the end-to-end pipelined-hold rejection test in mind as a follow-up. And thank you, @peter-toth , for the many meticulous rounds on this PR -- the review shaped most of what is good about the final design, from the backend-owned requirement authority to the scope decision itself. The latest config-off caveat (finding 38) is addressed in 44baf1d, and the stale replies on the findings-29/35 threads are corrected to reflect the pipelined scope exclusion. From my side all known threads are now addressed at head. -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
