jtb20 wrote: > Could you please contain your stacking enthusiasm a bit? Your PR stack here > is basically exhausting our entire CI capacity any time you rebase it. Please > keep the number of simultaneously open PRs low, or use a workflow that does > not result in mass rebases. > > See > https://llvm.grafana.net/public-dashboards/21c6e0a7cdd14651a90e118df46be4cc. > I assume that big spike in queuing is due to this stack getting updated.
Please excuse my naivety, I'm not terribly used to this workflow! Believe it or not, I'm not particularly a fan of stacked PRs. I'm not quite sure how to do it better though. A few huge patches would seem to be much harder to review. I guess quite a few of these are incremental-ish fixes on top of a baseline, so can be folded back -- but it's going to be hard to keep track of context and the interdependencies between components if I do that. Open to suggestions, or I can think some more about different ways to slice this work up? https://github.com/llvm/llvm-project/pull/188765 _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
