Re: xen PV on HVM and initial domain merge in linux-next

[Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>]

On Wed, Oct 20, 2010 at 04:16:17PM +0100, Stefano Stabellini wrote:
> On Wed, 20 Oct 2010, Stephen Rothwell wrote:
> > Hi Stefano,
> > 
> > [just casting the net a bit wider ...]
> > 
> > On Tue, 19 Oct 2010 18:51:47 +0100 Stefano Stabellini <stefano.stabellini@eu.citrix.com> wrote:
> > >
> > > I forgot to CC the LKML and linux-next...
> > > 
> > > On Tue, 19 Oct 2010, Stefano Stabellini wrote:
> > > > Stephen,
> > > > I have two patch series to merge in linux-next:
> > > > 
> > > > PV on HVM: receive interrupts as xen events
> > > > xen: initial domain support
> > > > 
> > > > they have all the acked-by needed and are both stable since several
> > > > weeks, however they depend on Konrad's xen-pcifront series and for this
> > > > reason I waited until now to ask for a merge in linux-next.
> > > > 
> > > > Could you please pull:
> > > > 
> > > > git://xenbits.xen.org/people/sstabellini/linux-pvhvm.git linux-next-initial-domain-v4
> > > > 
> > > > it contains both series rebased on Konrad's pcifront series merged on
> > > > linux-next (warning: it still contains the merge commit of
> > > > xen-pcifront-0.8.2 in linux-next).
> > > > Let me know if you have any conflicts or if you need me to change the
> > > > branch somehow.
> > 
> > Not following the Xen develpment at all, I would like to have a positive
> > reply from the listed Xen contacts, please,
> > 
> 
> Sure.
> Jeremy?

The patches touch the Xen PCI components, so:

Acked-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>

And yeah, we need to update the MAINTAINERS file. Let me spin one out
with the folks who are currently involved in this.

> 
> 
> > I do have concerns that this is turning up so late, but I realise that
> > that is mainly due to a misunderstanding on the part of some of the Xen
> > community.
> > 
> 
> Thank you very much for understanding!
> 
> 
> > Also, the above tree is based on next-20101019 which means that I cannot
> > use it as is.  All the trees merged into linux-next must be base on some
> > other stable tree (almost always Linus' tree).  linux-next is rebuilt
> > from scratch every day, so I cannot ever include a previous day's version.
> > 
> > Merging in other stable trees is OK (as long as the other maintainer is
> > aware of that and makes sure that their tree does not reabse).
> > 
> > Basically what you send to me should be what you intend to send to Linus
> > during the next merge window.
>  
> All right.
> I merged Jeremy's and Konrad's branches (the ones you just merged on
> linux-next) on top of linux 2.6.36 rc8, then I rebased my series on top
> of the result.
> Please checkout this branch:
> 
> git://xenbits.xen.org/people/sstabellini/linux-pvhvm.git 2.6.36-rc8-initial-domain-v5

Just as a note, you might want to create a branch titled

#linux-next in your git tree and link the reviewed/acked/tested branch to it. That way
Stephen does not have to change his entries in the linux-next tree every release.

> 
> and let me know if it is suitable, it shouldn't have any merge
> conflicts.

/me nods. Works nicely. Thank you for doing this on such a short-time frame.