Re: [dpdk-dev] [PATCH] doc: announce change in ETH item struct

[Andrew Rybchenko <arybchenko@solarflare.com>]

On 8/5/20 4:31 PM, Dekel Peled wrote:
> Thanks, PSB.
> 
>> -----Original Message-----
>> From: Andrew Rybchenko <arybchenko@solarflare.com>
>> Sent: Wednesday, August 5, 2020 1:49 PM
>> To: Dekel Peled <dekelp@mellanox.com>; dev@dpdk.org
>> Cc: jerinjacobk@gmail.com; stephen@networkplumber.org;
>> ajit.khaparde@broadcom.com; maxime.coquelin@redhat.com;
>> olivier.matz@6wind.com; david.marchand@redhat.com;
>> ferruh.yigit@intel.com; Asaf Penso <asafp@mellanox.com>
>> Subject: Re: [dpdk-dev] [PATCH] doc: announce change in ETH item struct
>>
>> On 8/4/20 7:01 PM, Dekel Peled wrote:
>>> Struct rte_flow_item_eth will be modified to include additional
>>> values, indicating existence or absence of VLAN headers following the
>>> ETH header, as proposed in RFC
>>>
>> https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmails.
>> dpdk.org%2Farchives%2Fdev%2F2020-
>> August%2F177349.html&amp;data=02%7C01%7Cdekelp%40mellanox.com%7
>> C7d12c0524d434c10f35a08d8392d3f53%7Ca652971c7d2e4d9ba6a4d149256f46
>> 1b%7C0%7C0%7C637322213783925660&amp;sdata=4rEXCifnCHd2%2FA6AU4
>> F3vjBcD7CfoTpT0traJ2z1fBk%3D&amp;reserved=0.
>>> Because of ABI break this change is proposed for 20.11.
>>>
>>> Signed-off-by: Dekel Peled <dekelp@mellanox.com>
>>> ---
>>>  doc/guides/rel_notes/deprecation.rst | 5 +++++
>>>  1 file changed, 5 insertions(+)
>>>
>>> diff --git a/doc/guides/rel_notes/deprecation.rst
>>> b/doc/guides/rel_notes/deprecation.rst
>>> index 5201142..6241709 100644
>>> --- a/doc/guides/rel_notes/deprecation.rst
>>> +++ b/doc/guides/rel_notes/deprecation.rst
>>> @@ -115,6 +115,11 @@ Deprecation Notices
>>>    following the IPv6 header, as proposed in RFC
>>>
>> https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmails.
>> dpdk.org%2Farchives%2Fdev%2F2020-
>> August%2F177257.html&amp;data=02%7C01%7Cdekelp%40mellanox.com%7
>> C7d12c0524d434c10f35a08d8392d3f53%7Ca652971c7d2e4d9ba6a4d149256f46
>> 1b%7C0%7C0%7C637322213783925660&amp;sdata=Ix4Y2vEXMoBek%2BkXw
>> QazY11a9mkc3aiZRJDX9cbUiZk%3D&amp;reserved=0.
>>>
>>> +* ethdev: The ``struct rte_flow_item_eth`` struct will be modified to
>>> +include
>>> +  additional values, indicating existence or absence of VLAN headers
>>> +  following the ETH header, as proposed in RFC
>>> +
>> https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmails.
>> dpdk.org%2Farchives%2Fdev%2F2020-
>> August%2F177349.html&amp;data=02%7C01%7Cdekelp%40mellanox.com%7
>> C7d12c0524d434c10f35a08d8392d3f53%7Ca652971c7d2e4d9ba6a4d149256f46
>> 1b%7C0%7C0%7C637322213783925660&amp;sdata=4rEXCifnCHd2%2FA6AU4
>> F3vjBcD7CfoTpT0traJ2z1fBk%3D&amp;reserved=0.
>>
>> It is unclear how it will coexist with VLAN items in a pattern.
> Same as the existing proto field coexist with following VLAN items.
> 
>> Are you going to add consistency checks on ethdev-layer?
> Not planned currently.

IMHO, it is a must requirement if you introduce interface which
has inter-dependencies and requires generic consistency check.
It must be a part of ethdev, not every PMD.

> 
>>
>> Also it is unclear why both bit fields and a number are required.
> I agree it is redundancy, added for flexibility, but can leave num_of_vlans only.
> 
>>
>> Referenced RFC lacks definition of S-VLAN anc C-VLAN in the context. Exact
>> definition to avoid ambiguity.
> These are well defined terms, I will add reference to spec.

Basically, I agree that changes in the ETH item may be
required and OK to acknowledge the deprecation notice
without deep technical details. The only requirement is
API backward compatibility (i.e. just recompiled code
to take ABI changes into account must work).

> 
>>
>> So, it looks required to modify the structure, but I'd not stick to referenced
>> RFC, since the result could differ a lot. May be reference it as just an
>> example.
> Thank you.
> 
>