On Fri, 9 Sep 2011, Ian Campbell wrote:
> On Fri, 2011-09-09 at 06:19 -0400, Stefano Stabellini wrote:
> > On Thu, 8 Sep 2011, Ian Campbell wrote:
> > > On Thu, 2011-09-08 at 13:47 -0400, stefano.stabellini@xxxxxxxxxxxxx
> > > wrote:
> > > > Signed-off-by: Stefano Stabellini <stefano.stabellini@xxxxxxxxxxxxx>
> > > >
> > > > diff -r ef27b472d4f3 Config.mk
> > > > --- a/Config.mk Thu Sep 08 17:19:12 2011 +0000
> > > > +++ b/Config.mk Thu Sep 08 17:29:50 2011 +0000
> > > > @@ -195,6 +195,8 @@ endif
> > > > # Only available through the git protocol at the moment
> > > > QEMU_UPSTREAM_URL ?=
> > > > git://xenbits.xen.org/people/sstabellini/qemu-dm.git
> > > > QEMU_UPSTREAM_TAG ?= origin/xen-stable-0.15
> > > > +SEABIOS_UPSTREAM_URL=git://git.qemu.org/seabios.git
> > > > +SEABIOS_UPSTREAM_TAG ?= 7fc039e9c262b4199fab497f3e12f4e425c37560
> > >
> > > I guess we should have a default tree on xenbits for this?
> >
> > I am not sure it is a good idea, after all we don't plan to fork it,
> > right?
>
> Planning not to fork is not the same as not forking.
>
> For example, imagine a bug (or security issue) is discovered in seabios
> and we want to backport it to our (by then) xen-4.2-testing branch. Do
> we take all the other updates to seabios trunk into our stable branch to
> get the one we want or do we ask them to push it to a branch just for
> us? Or do we now need to scrabble to setup trees, update the links and
> retest etc while trying to get an (perhaps critical) update out? Far
> better to just start off hosting a tree on xenbits, even if it doesn't
> diverge at all.
>
Seabios already provides two stable branches: if a security issue is
discovered, the Seabios guys will backport the fix to their stable
branches and we just need to update the TAG (or set the TAG to
0.6.1-stable to begin with).
The only problem is that no stable branches have xen support yet.
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
|
Home