Re: [PATCH] devtools: add Co-developed-by to commit log check

[Stephen Hemminger <stephen@networkplumber.org>]

On Thu, 12 Mar 2026 21:27:01 +0100
Thomas Monjalon <thomas@monjalon.net> wrote:

> 12/03/2026 21:10, Stephen Hemminger:
> > On Thu, 12 Mar 2026 19:56:25 +0100
> > Thomas Monjalon <thomas@monjalon.net> wrote:
> >   
> > > 12/03/2026 19:54, Stephen Hemminger:  
> > > > Add Co-developed-by to the recognized tag pattern and tag
> > > > sequence order in check-git-log.sh.    
> > > 
> > > Why using such tag? Signed-off-by is not enough?  
> > 
> > 
> > The tags were inherited from the kernel development process.
> > 
> > Signed-off-by is the DCO (Developer Certificate of Origin) attestation. 
> > It says "I certify that I have the right to submit this code under the project's license.
> > 
> > Co-developed-by explicitly marks someone as a co-author of the patch. 
> > Without it, there's only one author recorded in the git metadata.
> > 
> > I just wanted check-git-log.sh to be quiet if it was used.
> > 
> > The Co-developed-by is new, and probably came about because some people
> > need/want to have their contributions recorded in the git statistics.
> > Silly corporate overlords count contributions and it matters to them...
> > 
> > PS: If DPDK was exactly following the kernel process, every time a patch was merged into a sub-tree
> > it would get a Signed-off-by from a maintainer. Because the maintainer is validating that
> > the submitter had the correct rights.  
> 
> True, but in DPDK we use SoB as an author or co-author mark
> and it is well in line with the DCO.
> 
> I know that we had some exceptions asking for Co-developed-by
> because of a corporate ask, and I am on the side of being flexible.
> But making it a part of our official process could make things confused
> I think.

Agree. Want to allow developers to use Co-developed-by but not
suggest it or document it as required. Some people seem to want to use it.