[llvm-dev] [cfe-dev] Buildbot Noise

[David Blaikie via llvm-dev]

On Fri, Oct 16, 2015 at 11:24 AM, Renato Golin <renato.golin at linaro.org>
wrote:

> On 16 October 2015 at 19:10, Robinson, Paul
> <Paul_Robinson at playstation.sony.com> wrote:
> > That part is cultural.
>
> It is... :)
>
>
> > Which ones are "will never work" and which are "ought to get fixed soon"?
>
> In theory, the ones that have bugs associated with it, which is not
> easy to find unless someone adds a comment with the XFAIL, which not
> many people do.
>
> I'm not saying it's better, I'm just saying how it is.
>
>
> > For the people who now get impatient with red bots, there is no barrier
> > to having them be impatient with orange bots.   And for the people who
> > complain about irrelevant noise, there would be less of it, which will
> > make them less irritating the rest of us. :-)
>
> I honestly don't see either way as problems, so I can't tell you how
> much better it would be. :)
>
> Maybe you should talk to David directly. I really don't mind whatever
> you guys choose.
>
>
> > My internal dashboard has 4 values that would be meaningful upstream:
> > pass, infra-failure, fail, no-new-fail. Buildbot seems to have the first
> > three already; one more does not seem like an excessive conceptual
> burden.
>
> We don't have the infra problem totally nailed yet, only if the bots
> crash/restart, not if it's SVN or server issue. But you're right, it's
> not binary.
>
> I personally don't treat orange and red bots differently, for me the
> only colour I want to see is green. But some people complain that if
> they have to distinguish between different shades of red, they just
> won't.
>
> I find it an odd behaviour in a community that wants to be diverse and
> "safe",


If you're drawing parallels to the CoC thread, I don't think that's really
relevant/appropriate. Safe in terms of personal/emotional/physical safety
is rather different than the issues being discussed here.


> to openly state you don't care about other people's problems
>

I've tried to clarify that it's not that I don't care about the technical
problems you're grappling with/are discovered on ARM, just that the unique
(not overlapping with other, faster bots) emails sent (& the red bots in
the dashboard, to a lesser degree) seem to not be actionable by most
contributors (possibly any other than you and a small group of persons
specifically interested in behavior on those platforms - as demonstrated by
multiple issues that you've said yourself you investigate mostly alone (&
that it takes a while)).


> and think that it's the other person that should "fix", but anyway.
>
> cheers,
> --renato
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20151019/9117ed27/attachment.html>