Ben,
Are you happy with this? I really see this as something we could have done at
the time we introduced datapath flow key masks (megaflows), but did not. The
basic change is that we allow datapath flow setup without requiring key fields
that are fully wildcarded to be present.
Jarno
On Sep 11, 2014, at 2:11 PM, Daniele Di Proietto <ddiproie...@vmware.com> wrote:
> Hi Jarno,
>
> Thanks for the patch.
>
> The problem was that after the recent changes to odp-util.c
> {ovs-,ovs-appctl }dpctl do not work anymore. Here¹s a line that used to
> work:
>
> ovs-appctl dpctl/add-flow dummy@br0
> "in_port(1),eth(src=00:00:00:00:00:00/00:00:00:00:00:00,dst=ff:ff:ff:ff:ff:
> ff/00:00:00:00:00:00),eth_type(0x0800),ipv4(src=10.0.0.1/0.0.0.0,dst=10.1.0
> .1/0.0.0.0,proto=17/0x00,tos=0x10/0,ttl=64/0,frag=no),udp(src=0/0,dst=0/0)"
> 2
>
> After this patch (and I suppose after patch 5818 for the kernel datapath)
> not only the above line works, but we¹re also able to write the flow more
> concisely, like this:
>
> ovs-appctl dpctl/add-flow dummy@br0
> "in_port(1),eth_type(0x0800),ipv4(frag=no)² 2
>
> Tested-by: Daniele Di Proietto <ddiproie...@vmware.com>
>
>
> On 9/11/14, 11:25 AM, "Jarno Rajahalme" <jrajaha...@nicira.com> wrote:
>
>>
>> On Sep 11, 2014, at 11:11 AM, Ben Pfaff <b...@nicira.com> wrote:
>>
>>> On Thu, Sep 11, 2014 at 11:12:29AM -0700, Jarno Rajahalme wrote:
>>>> This could fix using ovs-dpctl to add userspace datapath flows.
>>>
>>> It's a big patch, could you expand slightly on what it does before I
>>> start reading it?
>>
>> I planned to follow up with the note that Daniele said he faced problems
>> using ovs-dpctl to control DPDK datapath on the current master, after my
>> recent patches went in. I¹d hope Daniele will check if this patch
>> addresses his problem.
>>
>> This patch allows parsing Netlink keys that do not include all ³expected²
>> keys. For example, if eth_type is 0x0800, you could still leave out the
>> IPv4 key, with the result that it is fully wildcarded. If only a key
>> (without a mask) is parsed, all expected keys need to be present, like
>> before.
>>
>> The remaining patches from my ³masked set actions² series introduce the
>> same flexibility to the Linux kernel module.
>>
>> I could have sent this privately to Daniele first, but it was (marginally
>> :-) easier for me to send to the listŠ
>>
>> Jarno
>>
>> _______________________________________________
>> dev mailing list
>> dev@openvswitch.org
>> https://urldefense.proofpoint.com/v1/url?u=http://openvswitch.org/mailman/
>> listinfo/dev&k=oIvRg1%2BdGAgOoM1BIlLLqw%3D%3D%0A&r=MV9BdLjtFIdhBDBaw5z%2BU
>> 6SSA2gAfY4L%2F1HCy3VjlKU%3D%0A&m=VXLWm8e0av6r5eu0OEbLJwPvzsO%2B8qxcyBPguAn
>> oSss%3D%0A&s=83c77a3f4b5df749a990290a0b4481a56bd870372e28464e509411e77f24c
>> e84
>
_______________________________________________
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
