[Ksummit-2012-discuss] [ATTEND] stable kernel stuff and grumpy maintainers
jwboyer at redhat.com
Mon Jun 18 17:28:47 UTC 2012
On Mon, Jun 18, 2012 at 10:21:39AM -0700, Greg KH wrote:
> On Mon, Jun 18, 2012 at 01:15:53PM -0400, Josh Boyer wrote:
> > On Mon, Jun 18, 2012 at 10:12:24AM -0700, Greg KH wrote:
> > > Hi all,
> > >
> > > Despite being on the Program Committee, I figured it was only fair that
> > > I too submit my name as someone who would like to attend the Kernel
> > > Summit this year.
> > >
> > > I'd like to discuss:
> > > - The stable and longterm kernel release process, is it going well for
> > > everyone? Are there things we can do differently? How can I kick
> > > maintainers who don't mark patches for stable backports in ways that
> > > do not harm them too much? How can I convey decisions about the
> > > longterm kernel selection process in a better way so that it isn't
> > > surprising to people?
> > I'd like to discuss this as well. I think overall the stable trees are
> > going fairly smoothly, but there might be room for improvement. Perhaps
> > even resulting in a bit less burden on you.
> Hey, I'm all for less burden, but honestly, the stable releases aren't
> all that tough for me, it's a semi-automated process now, almost
> entirely scripted, with the exception of the patch review process, which
> needs to not be automated. I can give a demo of how it all works if
> people are interested.
I'd be interested in a demo, yes.
In the interest of progress, the one suggestion I'd have that
immediately comes to mind is changing the timing of the RC releases.
Right now they seem to be following a "RC Friday afternoon, release
Monday" kind of schedule. That probably works well for the maintainer
of the release, but it makes testing things out from a distro
perspective somewhat harder. Most of us are gone by Friday afternoon,
if not physically then certainly mentally. Switching to RC monday,
release thurs might be helpful.
> The toughest part for me right now is finding all the patches to apply,
> (i.e. digging through distro kernel trees) and dealing with random
> emails from people who don't read Documentation/stable_kernel_rules.txt.
I think we can improve the distro side of things, yes.
More information about the Ksummit-2012-discuss