Re: [PATCH v3 0/3] enhance TX checksum command and csum forwarding engine

[Olivier MATZ <olivier.matz-pdR9zngts4EAvxtiuMwx3w@public.gmane.org>]

Hi Konstantin,

On 01/21/2015 05:28 PM, Ananyev, Konstantin wrote:
>> I added the support of Ether over GRE, IP over GRE and IP over IP
>> tunnels in csumonly to do the test. I ask the csum forward engine
>> to calculate inner IP+TCP checksums, and outer IP (case 6 in [1]).
>> Here are the results:
>>
>> 1/ When I use I40E_TXD_CTX_UDP_TUNNELING:
>> - vxlan: all checksums ok
>> - eth over gre: all checksums ok
>> - ip over gre: not transmitted by hw
>> - ip over ip: all checksums wrong (set to 0 by hw)
>>
>> 2/ When I use I40E_TXD_CTX_GRE_TUNNELING:
>> - vxlan: checksums ok
>> - eth over gre: all checksums ok
>> - ip over gre: all checksums ok
>> - ip over ip: all checksums wrong (set to 0 by hw)
>>
>> 3/ When I use 00b:
>> - vxlan: all checksums ok
>> - eth over gre: all checksums ok
>> - ip over gre: all checksums ok
>> - ip over ip: checksums wrong (set to 0 by hw)
>
> Wow, so there is absolutely no difference in results for L4TUNT=2(GRE) and L4TUNT=0, right?
> And IP over IP doesn't work at all?

Right. I probably missed something in i40e driver. The application seems
to fill the mbuf properly.

> I suppose you set L4TUNLEN as described in spec for each case, right?

I think so.

> That looks really weird to me and as I can see completely contradicts with what spec.
> I suppose we'll need to reproduce all that tests on our HW too.
> Could you send to us a patch with your changes, so we can try same thing?
> Or just a dump of TDD and TCD values for each case.

Sure, I'm going to send all my code and tests in a RFC patchset in
a few minutes. By the way, I'm off tomorrow, I won't be able to
answer.

Regards,
Olivier