Re: [Qemu-devel] [PATCH] 9pfs: include <linux/limits.h> for XATTR_SIZE_MAX

["Daniel P. Berrange" <berrange@redhat.com>]

On Mon, Jul 31, 2017 at 10:23:08PM +0800, Fam Zheng wrote:
> On Sun, 07/30 19:23, Peter Maydell wrote:
> > On 30 July 2017 at 17:51, Kamil Rytarowski <n54@gmx.com> wrote:
> > > On 29.07.2017 21:34, Peter Maydell wrote:
> > >> On 29 July 2017 at 14:50, Patrick Steinhardt <ps@pks.im> wrote:
> > >>> On Fri, Jul 28, 2017 at 02:20:49PM -0300, Philippe Mathieu-Daudé wrote:
> > >>>> This is likely to break on BSD, but now than patchew has a NetBSD job
> > >>>> you can trigger a build RESENDing this patch.
> > >>
> > >
> > > I just checked patchew, and there is FreeBSD job. How far are we from
> > > adding more BSDs?
> > 
> > I now test OpenBSD and NetBSD as well in my pre-merge
> > test setup. Patchew could add them as well if desired.
> > (vm setup instructions at http://wiki.qemu.org/Hosts/BSD)
> 
> No objection to adding more BSDs to patchew as long as I can find a few more
> gigabytes RAM to run the VM (BTW I'm also thinking about converting long running
> VMs to boot/shutdown on demand, to support more types of guests). But still want
> to ask this: how likely it is for a patch to compile on one BSD flavor but fail
> on the other?

While they share common ancestry, they are largely independant projects,
so each has its own quirks & potentially differing features. IOW it isn't
like Linux distros, where there's a common kernel & userspace in every
distros, and you're largely just dealing with software version differences.
So if you have the resources, I think it'd be worth running patchew across
the different BSDs that QEMU claims support for. I'd far rather see the
failures upfront, than when Peter tries to merge my pull request.

Regards,
Daniel
-- 
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|