>> > I think I will just start marking patches as changes-requested when I see that >> > they break tests without replying and without reviewing. >> > Please respect reviewer's time. >> >> That is completely fine if the tests are working in the first place. And >> even when they're not (like in this case), pointing it out is fine, and >> I'll obviously go investigate. But please at least reply to the email, >> not all of us watch patchwork regularly. > > Please see Documentation/bpf/bpf_devel_QA.rst. > patchwork status is the way we communicate the intent. > If the patch is not in the queue it won't be acted upon. I do realise that you guys use patchwork as the status tracker, but from a submitter PoV, in practice a change there is coupled with an email either requesting something change, or notifying of merge. Which is fine, and I'm not asking you to do anything differently. I'm just suggesting that if you start silently marking patches as 'changes requested' without emailing the submitter explaining why, that will just going to end up creating confusion, and you'll get questions and/or identical resubmissions. So it won't actually solve anything... (And to be clear, I'm not saying this because I plan to deliberately submit patches with broken selftests in the future!) -Toke
