Re: Missing signoff in the arm-soc tree

[Mark Brown <broonie@kernel.org>]

[-- Attachment #1: Type: text/plain, Size: 1597 bytes --]

On Mon, Sep 14, 2026 at 09:53:00PM +0200, Arnd Bergmann wrote:
> On Mon, Sep 14, 2026, at 18:25, Mark Brown wrote:

> > These issues are *still* present today.  It's making noise whenever I
> > check -next for ongoing issues, it'd be good if the branch could be
> > rebuilt.

> I assume you hit this on the branch I pushed out earlier today, which
> started fresh from 7.3 with just a single merge. I've amended that
> commit with an S-o-b now, but I'm not sure if that is what you are
> asking.

It is in whatever was there when I fetched the trees this morning which
was some time before I finished the merge and ran the summary stuff so
you may have updated it later.

> When you write "and also isn't showing up as a merge", do you mean
> your script has an exception for merge commits to not need S-o-b
> but fails to detect this one as a merge? I see that about half the
> merges in linux-next (including your daily merges) are not signed off,
> so I'm not sure if we are all doing this wrong the same way, or if
> there is a different problem with the soc tree.

As indicated in the original mail and repeated followups the relevant
commit is 3edcea7f0413e84863300998ad3838eb87e6fca7 which looks from the
changelog like it was originally a merge but is not described as such in
the git metadata, it is just an apparently empty commit with a single
parent with a suspiciously similar changelog.  My theory was that
something had gone wrong with a rebase or scripting.  I only check for
signoff problems on non-merge commits, merge commits are ignored since
they are not normally signed off.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]