On Tue, 2020-05-12 at 08:38 +0200, David Hildenbrand wrote:
> On 12.05.20 07:21, Lukas Bulwahn wrote:
> > Hi David,
> >
> > with your commit 6d6b93b9afd8 ("MAINTAINERS: Add myself as virtio-balloon
> > co-maintainer"), visible on next-20200508, ./scripts/checkpatch.pl -f
> > MAINTAINERS complains:
> >
> > WARNING: Misordered MAINTAINERS entry - list file patterns in alphabetic order
> > #17982: FILE: MAINTAINERS:17982:
> > +F: include/uapi/linux/virtio_balloon.h
> > +F: include/linux/balloon_compaction.h
> >
> > This is due to wrong ordering of the entries in your submission. If you
> > would like me to send you a patch fixing that, please just let me know.
> >
> > It is a recent addition to checkpatch.pl to report ordering problems in
> > MAINTAINERS, so you might have not seen that at submission time.
>
> Thanks for the notification Lukas,
>
> b962ee8622d0 ("checkpatch: additional MAINTAINER section entry ordering
> checks") is not in Linus' tree yet AFAIKS.
>
> I can see that 3b50142d8528 ("MAINTAINERS: sort field names for all
> entries") is upstream. I do wonder if we should just do another batch
> update after the checkpatch patch is upstream instead, I guess more will
> pile up?
>
> @mst, joe, what do you prefer?
>
> 1. I can resend the original patch.
> 2. Lukas can send a fixup that we might want to squash.
> 3. We wait until the checkpatch change goes upstream and to a final
> batch update.
A fixup patch would work.
I think if Linus every once in awhile just before an -rc1 runs
scripts/parse-maintainers like:
commit 3b50142d8528 ("MAINTAINERS: sort field names for all entries")
then these sorts of individual patches would not matter much.
This first time the script was run, I think there was just 1 patch
conflict from -next to Linus' tree, and that scripted change was
fairly large.
As the changes will generally be smaller in the future, it's unlikely
there will be a significant number of conflicts.
