[ovs-dev] [PATCH 01/12] xenserver: Fix comment in interface-reconfigure.

Ian Campbell Ian.Campbell at citrix.com
Tue Feb 23 10:04:41 PST 2010

On Tue, 2010-02-23 at 18:02 +0000, Ben Pfaff wrote:
> On Tue, Feb 23, 2010 at 10:20:10AM +0000, Ian Campbell wrote:
> > I made some minor comments on individual patches but generally all looks
> > fine to me (take with a pinch of salt given my lack of knowledge of the
> > next branch).
> I only saw one comment, on patch 3/4.  Were there others?

I think I got ahead of myself, there was only one comment in the end.

> Anyway, I pushed these out, so I'll have to fix them up.
> > The unit test stuff looks nice, if only I could steal it without
> > learning m4 ;-) I've attached the unit test I use plus a script which
> > will setup a XenServer host appropriately. They might be a little
> > Ian-specific and are completely devoid of comments but feel free to ask
> > if you are interested in using or cribbing from them and something isn't
> > clear.
> Thanks.  Are these the scripts that are (were?) in the XAPI tree?  I
> remember seeing a few of them related to testing interface-reconfigure,
> but I never got around to trying them out.

They are still there AFAIK. The version in the xapi tree doesn't
understand vswitch so I have a local patch to make that work which was
applied to the version I attached here.

I need to find the time to rework the changes in such a way as the test
suite can run in either bridge or vswitch mode.

> I haven't decided yet whether m4 is a godsend or a horrible curse, and
> I've been working with it for a long time now.
> > I assume these will these end up with a Signed-off-by etc when they
> > committed? At some point in the future, as part of moving XCP and
> > XenServer over to the next branch, I'll want to cherry-pick these
> > changes and upstream them to the XCP project which will require a
> > signed-off-by.
> Thanks for pointing out the need for "Signed-off-by" lines.  We have not
> been using them for OVS (I guess our lawyers have not told us that we
> need them?), but I'm perfectly willing to add them anyway, and so I will
> stick them on all of these commits before I push them.


More information about the dev mailing list