Re: [PATCH net-next 0/8] net/mlx5: Prepare eswitch infrastructure for satellite PF support

[Paolo Abeni]

On 5/14/26 11:53 AM, Paolo Abeni wrote:
> On 5/14/26 9:56 AM, Moshe Shemesh wrote:
>> On 5/14/2026 5:25 AM, Jakub Kicinski wrote:
>>> External email: Use caution opening links or attachments
>>>
>>>
>>> On Sun, 10 May 2026 08:34:40 +0300 Tariq Toukan wrote:
>>>> This series prepares the mlx5 eswitch command interface and vport
>>>> infrastructure for satellite PF support.
>>>
>>> Could you perhaps start by explaining what "satellite PF" is?
>>> And how it differs from "socket direct"
>>
>> Satellite PF is another type of Physical Function, its role and 
>> privileges are similar to the host PF, but unlike host PF the Satellite 
>> PF is on the DPU and not on another host. So it's kind of "Satellite" 
>> for the ECPF which is also on the DPU.
>>
>> The next patchset will introduce the Satellite PF, while this patchset 
>> only does some preparations. Small changes to prepare the eswitch to 
>> manage another PF.
>>
>> While Socket Direct is a hardware PCIe topology, that adds another PCIe 
>> link to the NIC, the Satellite PF is just logical entity, by firmware 
>> configuration, no hardware change.
> 
> My understanding is that Jakub wants you to capture the above info in
> the cover letter.

I take my own words back. Actually I'm not sure about the above and
given the constant ML flood I think we prefer to avoid a repost if
possible. The series LGTM, please wait a bit for Jakub eventual explicit
repost request.

Thanks,

Paolo