[webkit-dev] Rolling out a patch requires a justification beyond a test failure in downstream projects

[Elliott Sprehn]

Do you have an example of when this has occurred? It's good to have
examples if we want to prevent this in the future.


On Tue, Dec 11, 2012 at 1:44 PM, Ryosuke Niwa <rniwa at webkit.org> wrote:

> Hi,
>
> I've encountered a couple of incidences where people roll out patches
> saying that test X is failing on some downstream project Y without giving
> any details as to how those tests are failing and why that's a real WebKit
> regression we should care about.
>
> First off, I don't think we should be rolling out patches based solely on
> a downstream test unless there is a clear evidence that the failure is a
> real regression in WebKit that affects more than just the said downstream
> project. You may talk to the author and he or she might be nice enough to
> agree to roll out the patch, but I don't think we should be rolling out
> patches right away regardless.
>
> Second, if there is a clear WebKit regression, then you should communicate
> the following information at minimum:
>
>    1. The exact location of the test that failed - URL, etc...
>    2. The nature of the failure - assertion failure, feature it's
>    testing, etc...
>    3. The output (before and) after the failure started happening.
>    4. Instructions to run the tests locally
>
> Without this, the author is left with no clue whatsoever
> to diagnose and/or fix the problem.
>
> - R. Niwa
>
>
> _______________________________________________
> webkit-dev mailing list
> webkit-dev at lists.webkit.org
> http://lists.webkit.org/mailman/listinfo/webkit-dev
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-dev/attachments/20121211/fc65afff/attachment.html>