[Maverick][GIT PULL] replacement of hacks with Yama

[Chase Douglas]

On Thu, 2010-07-01 at 06:51 -0600, Tim Gardner wrote:
> On 06/30/2010 03:30 PM, Leann Ogasawara wrote:
> > On Wed, 2010-06-30 at 13:45 -0700, Kees Cook wrote:
> >> Hi Tim,
> >>
> >> On Wed, Jun 30, 2010 at 08:12:24AM -0600, Tim Gardner wrote:
> >>> These commits all have '(cherry picked from commit *)' comments in
> >>> the commit log, but as the objects are not from Linus' tree, they
> >>> are not relevant. Can you amend these commit log messages?
> >>
> >> Ah, sorry, they're from security-testing-2.6#next:
> >> http://git.kernel.org/?p=linux/kernel/git/jmorris/security-testing-2.6.git;a=shortlog;h=refs/heads/next
> >>
> >> What is the best way to reference these kinds of cherry picks?
> >
> > I don't think we've standardized on a syntax but something like the
> > following would be fine with me:
> >
> > (cherry picked from commit<insert sha1>
> > from git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/security-testing-2.6.git next)
> >
> 
> This is a bit of a nit, but even that syntax is superfluous because 
> James Morris tree is not immutable. Like all of the 1st and subsequent 
> level maintainers he must occasionally rebase his tree which renders all 
> SHA1 references to his tree invalid.
> 
> The only trees that are never rebased are Linus' and the post release 
> Ubuntu repos.

I think it's useful just as we submit patches. If a patch comes from
someone we don't know already, we can double check it before we apply it
to our tree.

However, I don't see the usefulness once it has been applied to our tree
because upstream may rebase. Maybe we should strip them before we apply?

-- Chase