[Mesa-dev] Gitlab migration

[Daniel Vetter]

On Thu, May 24, 2018 at 3:23 AM, Jason Ekstrand <jason at jlekstrand.net> wrote:
> On Wed, May 23, 2018 at 5:49 PM, Jordan Justen <jordan.l.justen at intel.com>
> wrote:
>>
>> On 2018-05-23 15:16:58,  wrote:
>> > On Wed, May 23, 2018 at 5:48 PM, Jordan Justen
>> > <jordan.l.justen at intel.com> wrote:
>>
>> > And I'm not strongly tied to bugzilla vs gitlab's issue tracking.
>>
>> Another project I'm involved with had a contingency that swore
>> github's "issues" were completely inadequate compared to bugzilla,
>> which is to say that I don't think there is consensus on this point.
>> :)
>>
>> For that (smaller) project, I thought github's issues were fine
>> compared to bugzilla. For (the larger) Mesa project, I'm not so sure.
>
>
> Agreed.  The gitlab/github issue system has some scaling problems.  The
> tagging system isn't as good as bugzilla's categorization.  Also, the
> ability to move bugs back and forth between mesa and kernel is a pretty
> killer feature.  Then again, the ability to tag a bug as "blocks 18.3" might
> be a nicer than tracker bugs.

So for political reasons I expect the kernel has to sail in the
shadows of mesa and xserver here. We need a convenient scape goat :-)

But if we e.g. have clear consensus among all the intel teams that we
want to move to gitlab with our issues pile, I'd very much help
support that on the kernel side. I think as long as we have all the
parts of the driver stack for one vendor in the same place, we're 90%
covered with reassignment needs. Anything across team needs to be
discusson on mailing lists and irc anyway, for it'll fall through all
the cracks otherwise.

And see my other reply, there's a metric pile of things I'd really
like to play around with gitlab CI integration on the kernel side,
because scripting it in a portable fashion client side for each
committer is a royal pain. It could very well be that we only start
with that at the drm tree integration level, and not at subtrees where
contributors push their patches. That kind of mutli-repo integration
flow is also something I need to figure out with gitlab first.
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch