Re: [Qemu-devel] [PATCH for-1.7] configure: Explicitly set ARFLAGS so we can build with GNU Make 4.0

[Michael Roth]

Quoting Peter Maydell (2013-11-05 13:36:57)
> On 5 November 2013 19:31, Andreas Färber <address@hidden> wrote:
> > Am 05.11.2013 20:09, schrieb Peter Maydell:
> >> I don't think there's any need to respin a patch
> >> just because somebody thinks it might be worth
> >> applying to stable.
> >
> > Point is, if you are ping'ing Anthony to apply this with his
> > fully-automated patches tool, then he won't add the Cc and in turn it
> > will unlikely be backported to 1.6.2.
> >
> > The response I got at QEMU Summit was that it is both the contributor's
> > and the submaintainer's responsibility to add the Cc to the commit
> > message. But maybe you're going to tell me that I misunderstood that as
> > well...
> 
> I don't remember what people said particularly, but I don't
> think our process for stable should rely on patch submitters
> having to add cc tags and resend stuff because that's a
> lot of faff for something most people don't care about.
> Whoever is maintaining stable ought IMHO to be making their
> own judgements about whether a fix is worth backporting
> based on their criteria for what kind of fix to put in
> that branch.

I think the way to look at it is getting stuff applied is first priority,
but if it's missing Cc: qemu-stable all bets are off on whether or not it
gets pulled in for a stable release. Whether a contributor/maintainer cares
enough to ensure this prior to commit is their call.

That said, if it's CC'd *to* qemu-stable it'll generally get picked up.
Maybe in the future if we find ourselves needing to automate things more
that may not be the case though, so it's a good practice to adopt.

In any case, ping? :)

> 
> -- PMM