[dpdk-ci] [PATCH v4] add script to decide best tree match for patches

[Jeremy Plsek]

On Fri, Apr 19, 2019 at 3:41 PM Thomas Monjalon <thomas at monjalon.net> wrote:
>
> 19/04/2019 20:06, Jeremy Plsek:
> > On Fri, Apr 19, 2019 at 1:55 PM Thomas Monjalon <thomas at monjalon.net> wrote:
> > > 19/04/2019 19:33, Jeremy Plsek:
> > > > I thinks so. Only more patchsets will tell, but that could be improved
> > > > after the fact. So I think it's fine to merge it in.
> > > >
> > > > At first glance, the only part that I don't think is implemented is
> > > > mentioned here: https://bugs.dpdk.org/show_bug.cgi?id=166#c35
> > > > > We must match the common prefix of the git trees.
> > > > > Examples:
> > > > > dpdk-next-net-intel + dpdk = dpdk
> > > > > dpdk-next-net-intel + dpdk-next-net-mlx = dpdk-next-net
> > > > Since some of the patches are being set to dpdk-next-net-mlx instead
> > > > of dpdk-next-net. But I'm fine with how it is right now and wouldn't
> > > > mind it getting changed to this later on.
> > >
> > > It is supposed to be fixed.
> > > Please could you give an example of a misbehaviour?
> >
> > The most recent example is series 4380. For me, that returned dpdk-next-net-mlx.
>
> The series 4380 is mlx only, so it fine to match dpdk-next-net-mlx.
> Why do you expect something else?
>
Ok, I guess I was misunderstanding the comment. I thought that
anything under dpdk-next-net-* would just be lumped into
dpdk-next-net. The scripts are good to go then.



--
Jeremy Plsek
UNH InterOperability Laboratory