Re: [PATCH v3] git-publish: add profile

[Stefan Hajnoczi <stefanha@redhat.com>]

[-- Attachment #1: Type: text/plain, Size: 3941 bytes --]

On Thu, Mar 06, 2025 at 05:13:20PM +0100, Albert Esteve wrote:
> On Thu, Mar 6, 2025 at 11:51 AM Stefano Garzarella <sgarzare@redhat.com> wrote:
> >
> > On Thu, 6 Mar 2025 at 11:44, Albert Esteve <aesteve@redhat.com> wrote:
> > >
> > > On Thu, Mar 6, 2025 at 10:58 AM Stefano Garzarella <sgarzare@redhat.com> wrote:
> > > >
> > > > On Thu, 6 Mar 2025 at 06:24, Stefan Hajnoczi <stefanha@redhat.com> wrote:
> > > > >
> > > > > On Wed, Mar 05, 2025 at 05:45:46PM +0100, Matias Ezequiel Vara Larsen wrote:
> > > > > > Add git-publish profile and document how to use it.
> > > > > >
> > > > > > Signed-off-by: Matias Ezequiel Vara Larsen <mvaralar@redhat.com>
> > > > > > ---
> > > > > > v2 -> v3:
> > > > > >    * remove uppercase and the line split and use italic
> > > > > > v1 -> v2:
> > > > > >    * add link to git-publish
> > > > > >    * document that the project contains a profile already
> > > > > > ---
> > > > > >  .gitpublish |  3 +++
> > > > > >  README.md   | 15 +++++++++++++++
> > > > > >  2 files changed, 18 insertions(+)
> > > > > >  create mode 100644 .gitpublish
> > > > > >
> > > > > > diff --git a/.gitpublish b/.gitpublish
> > > > > > new file mode 100644
> > > > > > index 0000000..d412896
> > > > > > --- /dev/null
> > > > > > +++ b/.gitpublish
> > > > > > @@ -0,0 +1,3 @@
> > > > > > +[gitpublishprofile "default"]
> > > > >
> > > > > Great, this will make it easier to send spec changes!
> > > > >
> > > > > > +base = master
> > > > > > +to = virtio-comment@lists.linux.dev
> > > > >
> > > > > From https://github.com/oasis-tcs/virtio-spec?tab=readme-ov-file:
> > > > >
> > > > >   When not to use the virtio-comment mailing list:
> > > > >     questions and change proposals for Virtio drivers and devices
> > > > >     implementing the specification. (please use the virtio-dev mailing
> > > > >     list for this).
> > > >
> > > > That statement talks about the driver and device "implementing" the
> > > > spec, not the spec itself, indeed there is another statement:
> > > >
> > > >   When to use the virtio-comment mailing list:
> > > >      questions and change proposals for the Virtio specification,
> > > >      including the specification of basic functionality, transports and
> > > >      devices.
> > > >
> > > > So I guess this patch is right, since here we are talking about spec
> > > > changes.
> > > >
> > > > BTW I'm honest, I'm always confused about virtio-comment and virtio-dev
> > > > xD, so another "Thank you" to Matias for this patch!
> > >
> > > I'm in that club too :D I had to fail once to get it right.
> >
> > :D
> >
> > >
> > > The problem with hardcoding the gitpublishto setting to the default
> > > profile is that users need to be aware that if they need to send it to
> > > virtio-dev, they need to manually remove virtio-comment and add
> > > virtio-dev to the command line. I mean, the patch is good because in
> > > most cases it will be the right mailing list to use. But perhaps it
> > > would be nice to mention it in the README (at least, mention what is
> > > the default), instead of relying on them going to .gitpublish file and
> > > check?
> >
> > I see, but do we really have a case where changes to the spec need to
> > be sent to virtio-dev?
> 
> Judging by the README I'd say probably not? But looking at the
> messages in the list, I found one example of a spec change in
> virtio-dev [1]. Although that is not a good reason per se, as it may
> just be using the wrong list.
> 
> [1] https://lore.kernel.org/virtio-dev/20241107105103.19152-1-mgurtovoy@nvidia.com/T/#u
> 
> >
> > If not, I'd avoid to mention to increase confusion.
> 
> Yeah, let's do that then.

I read the README.md more carefully and I think Stefano is right:
patches should be sent to virtio-comment.

There's no need to mention virtio-dev.

I was just doing it wrong all along :D.

Stefan

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