[Piglit] Require Signed-off-by for patches?

[Jordan Justen]

On Wed, Nov 13, 2013 at 5:12 PM, Matt Turner <mattst88 at gmail.com> wrote:
> On Wed, Nov 13, 2013 at 12:06 PM, Jordan Justen <jljusten at gmail.com> wrote:
>> What are the arguments against just following the kernel's
>> Signed-off-by practice?
>
> What are the arguments for it?
>
> The kernel's submitting patches documentation says that
>
>  - Signed-off-by: this is a developer's certification that he or she has
>    the right to submit the patch for inclusion into the kernel.  It is an
>    agreement to the Developer's Certificate of Origin, the full text of
>    which can be found in Documentation/SubmittingPatches.  Code without a
>    proper signoff cannot be merged into the mainline.
>
> and
>
> The Signed-off-by: tag indicates that the signer was involved in the
> development of the patch, or that he/she was in the patch's delivery path.

That text contains one of the arguments. Signed-of-by is a way for the
contributor to assert that they are confident the open source project
can use the code they've submitted. In other words, they didn't just
copy it off some web-site, or out of some closed source code.

Yes, this is usually implied by submitting the patch, but with
Signed-of-by, they are specifically calling it out. I assume the
kernel folks thought about this a fair amount before developing the
signature process, so it seems like a good idea to just leverage that.

Another argument would be that developers are somewhat likely to
already be familiar with the process from the kernel and other open
source projects. Frank's feedback on Paul's patch is an example of
this. When I started contributing to mesa/piglit I also assumed the
same thing since we used all the other parts of the kernel's *-by
process.

Adding -s to git commit just doesn't seem to be much of burden, nor
much of change in process.

-Jordan