[ovs-dev] [PATCH 1/2] Add support for tun_key to OVS datapath

Kyle Mestery (kmestery) kmestery at cisco.com
Tue Oct 9 01:34:07 UTC 2012

On Oct 8, 2012, at 4:49 PM, Jesse Gross <jesse at nicira.com> wrote:
> On Fri, Oct 5, 2012 at 1:07 PM, Kyle Mestery (kmestery)
> <kmestery at cisco.com> wrote:
>> On Oct 5, 2012, at 12:47 PM, Jesse Gross wrote:
>>> On Mon, Oct 1, 2012 at 8:52 AM, Kyle Mestery <kmestery at cisco.com> wrote:
>>>> This is a first pass at providing a tun_key which can be
>>>> used as the basis for flow-based tunnelling. The
>>>> tun_key includes and replaces the tun_id in both struct
>>>> ovs_skb_cb and struct sw_tun_key.
>>>> This patch allows all existing tun_id behaviour to still work. Existing
>>>> users of tun_id are redirected to tun_key->tun_id to retain compatibility.
>>>> However, when the userspace code is updated to make use of the new tun_key,
>>>> the old behaviour will be deprecated and removed.
>>>> NOTE: With these changes, the tunneling code no longer assumes input and
>>>> output keys are symmetric.  If they are not, PMTUD needs to be disabled
>>>> for tunneling to work.
>>>> Signed-off-by: Kyle Mestery <kmestery at cisco.com>
>>>> Cc: Simon Horman <horms at verge.net.au>
>>>> Cc: Jesse Gross <jesse at nicira.com>
>>> One thought that I had was to try to get the kernel portions in one
>>> release and the userspace portions that actually use them in the next.
>>> That way we would give people a window to smoothly upgrade because
>>> the kernel module would support both mechanisms for that release.
>>> We're currently planning to branch for the next release in two weeks
>>> (although I'm going to be traveling for the second one and unable to
>>> review code).  I think we're getting pretty close here, so do you
>>> think my plan is doable?
>> I think that makes sense. In essence, my first patch is mostly the kernel pieces.
>> I am also traveling a week from Sunday for a week, and will likely not have the
>> cycles to do much during that week. I'll see to addressing your comments on the
>> first patch, and try to resubmit again early next week.
> OK.  Pravin said that he could help out by doing some of the polishing
> work so we can get everything in before the release branch.  If you've
> made any changes since the last patchset can you send it out so he can
> take a look at the latest version?

Yes, I'm planning to do that tomorrow. I spent today going through and fixing a few more
of your comment items. I'll post the latest patches tomorrow morning. It would be great
to get some feedback again when you guys get in, so I can iterate again tomorrow. I
have a lot of time to work on this tomorrow specifically this week.


More information about the dev mailing list