> * blocking-b2g:tef+ is for bugs that we've got agreement with partners about 
> needing as part of v1.0.0.0

This was my original understanding of this flag, it's not the
information that I was told in the triaging sessions we've had the
last couple of days. We have been marking many bugs as tef+ even when
no partner was attending the meeting, which obviously means that we
don't know if it's a partner requirement.

So how should we be using this flag?

If we are to only mark bugs as tef+ once we've gotten confirmed that
it's a partner requirement, should we add some additional flag for
things that we want to propose to partners as might needing to be a
requirement. I.e. bugs that are passing our initial screening of
nominations?

> * status-b2g18 represents the fix status on the v1.* branches
>
> Now for landing.
>
> * v1.0.0.0 Landing
> ** Gecko: mozilla-b2g18
> ** Gaia: master branch
> ** Landing requirements: blocking-b2g:tef+ (representing agreement with 
> partners), and approval-gaia-master:+ or approval-mozilla-b2g18:+

I'm not sure that it makes sense to require both tef+ *and* an
approval flag. So far we've not had the approval flag requirement and
I think that has enabled us to get patches landed quickly. We do have
a lot of regressions, but the types of regressions that I've seen
hasn't been once that I think we would have caught by going through an
approval step.

Requiring this flag *does* have a cost in that it *will* delay the
landing of patches. We do have close to 100 bugs to fix and 10 days to
do it, so any slow down is putting a lot of risk in our ability to
meet that goal.

As a data point I'll note that we were able to keep an extremely high
rate of bug fixing at the work week, and one of the reasons for this
was that we were able to keep very short patch-review-land cycles.

/ Jonas
_______________________________________________
dev-b2g mailing list
[email protected]
https://lists.mozilla.org/listinfo/dev-b2g

Reply via email to