Currently when a patch fails to build, we un-apply the patch, and build without patch. However, we do not analyze the results. We should add a step to analyze these results. Also when ToT fails to build, instead of marking the patch as failed, we should keep trying indefinitely (like old EWS) until ToT is green again.
Created attachment 372485 [details] Proposed patch Sample run: Without failure on ToT: https://ews-build.webkit-uat.org/#/builders/33/builds/4178 With failure on ToT: https://ews-build.webkit-uat.org/#/builders/33/builds/4182
Example of similar behavior on old EWS: https://webkit-queues.webkit.org/patch/369760/mac-wk2-ews
Comment on attachment 372485 [details] Proposed patch Clearing flags on attachment: 372485 Committed r246609: <https://trac.webkit.org/changeset/246609>
All reviewed patches have been landed. Closing bug.
<rdar://problem/51911524>