On Wed, Apr 25, 2012 at 2:40 PM, Ryosuke Niwa <[email protected]> wrote: > On Wed, Apr 25, 2012 at 2:34 PM, Adrienne Walker <[email protected]> wrote: >> >> Also, when you say "pay attention," what level of involvement should >> we expect for one committer making the bots go red on a port that they >> are not primarily working on? Should we expect test expectation >> rebaselining? Patch reverting? Adding "= FAIL" test expectations and >> filing a bug? A friendly ping in irc? An apology and cupcakes? > > > FWIW, all contributors are expected NOT to regress ANY bots, and should put > reasonable efforts into fixing build & test failures after landing their > patches. i.e. if any build bot fails due to someone's patch, they're > expected to fix it. >
Given that some people often land changes that break baselines on every bot, and that we clearly don't expect the patch submitter to fix every bot in this situation, how do you square that with what you just wrote? What's your definition of "reasonable efforts"? We did discuss this a fair amount at the committers meeting - see http://trac.webkit.org/wiki/April%202012%20Keeping%20the%20bots%20green . Given that some bots have very long cycle times, and that it's not yet practical to ensure every bot is green for your change prior to landing (we don't have full bot coverage in EWS nor a way to manage updating broken baselines), I don't think such a blanket statement is realistic. -- Dirk _______________________________________________ webkit-dev mailing list [email protected] http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev
