[Mesa-dev] Thoughts after hitting 100 merge requests?

[Rob Clark]

On Mon, Jan 28, 2019 at 2:29 PM Ilia Mirkin <imirkin at alum.mit.edu> wrote:
>
> A few thoughts. Given past experience, I don't really expect these to
> have any serious impact on the direction taken, but I also don't want
> to just sit brooding in silence. Please note that full time/paid
> contributors probably have a different view of things than volunteer
> contributors. There's not a ton of the latter left.

There is a desire to make it easier for drive-by contributors (which
are generally taken to mean volunteer contributors but that shouldn't
mean to make it harder for long term volunteer contributors).  Ideally
it is not an either/or but something that could be fixed in the tools.
(At least that is my rough opinion.. I haven't contributed any patches
to gitlab itself and probably wouldn't know where to start.)

>
> 1. Emailed patches allow me to quickly and easily keep track of what's
> going on, and provide feedback where necessary. In under 5 seconds, I
> can evaluate whether a patch is of interest to me (usually much faster
> than that), and it ends up in my email which I already look at. Going
> to gitlab to look at this stuff is just not going to happen for me. If
> reviews all move to gitlab, I just won't do reviews, and will know a
> lot less about what's going on across the repository.

this seems like a solvable problem

> 2. I've seen a bunch of things land where I would have had comments
> beforehand. Once the patch is in, I don't really have an easy way to
> provide feedback. In the past if such a thing would happen, I just
> take the subject of the patch, pop it into the search in gmail, and
> reply to the email. It's unclear what I should do now -- for most
> things thus far, just pinging on IRC has worked out, but there's
> nowhere to either see the discussion that led to the change nor to
> provide post-commit feedback. In most places where I've done "tracked"
> reviews, there's linkage in the commit message to be able to find the
> reviews in the review system.

it would be nice to see gitlab injecting some tags (not just r-b, but
also a link back to the MR) into the git commit msg

> 3. Given the present state, where not everyone looks at gitlab,
> patches going through it are getting less review than they would on
> the ML. If entire subsystems agree to exclusively go through gitlab --
> fine. However at this point, core mesa, st/mesa, and gallium patches
> are sneaking through without being seen pre-commit by the people who
> normally review such things.

I think some way to automagically apply path based labels to gitlab
MRs would solve at least some of this..  I was told that it should be
possible to do something like this on top of the gitlab API but
haven't had time to translate sugestion into patch.

Or maybe more useful would be a way to subscribe to path based
notifications without requiring a label..

It would be at least nice to get cover-letters CC'd to list so at
least mesa-dev has a sort of executive summary of what is going on.

BR,
-R


>
> Cheers,
>
>   -ilia
>
> On Fri, Jan 11, 2019 at 11:57 AM Jason Ekstrand <jason at jlekstrand.net> wrote:
> >
> > All,
> >
> > The mesa project has now hit 100 merge requests (36 are still open).  I (and I'm sure others) would be curious to hear people's initial thoughts on the process.  What's working well?  What's not working?  Is it total fail and should we go back to mailing lists?
> >
> > --Jason
> > _______________________________________________
> > mesa-dev mailing list
> > mesa-dev at lists.freedesktop.org
> > https://lists.freedesktop.org/mailman/listinfo/mesa-dev
> _______________________________________________
> mesa-dev mailing list
> mesa-dev at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/mesa-dev