[Intel-wired-lan] [PATCH net-next v4 3/3] ice: Enable switching default tx scheduler topology

Wilczynski, Michal michal.wilczynski at intel.com
Wed Jul 20 14:48:56 UTC 2022


Hi,

I created  a new version that includes Lukasz commit with devlink param 
implementation.
Thanks,
Michał

On 7/14/2022 10:31 PM, Tony Nguyen wrote:
>
>
> On 7/12/2022 12:51 AM, Wilczynski, Michal wrote:
>>
>>
>> On 7/12/2022 2:25 AM, Tony Nguyen wrote:
>>>
>>>
>>> On 7/4/2022 5:32 AM, Michal Wilczynski wrote:
>>>> Introduce support for tx scheduler topology change, based on user
>>>> selection, from default 9-layer to 5-layer. In order for switch to be
>>>> successful there is a new NVM(version 3.20 or older) and DDP 
>>>> package(OS
>>>> Package 1.3.29 or older).
>>>>
>>>> Enable 5-layer topology switch in init path of the driver. To 
>>>> accomplish
>>>> that upload of the DDP package needs to be delayed, until change in Tx
>>>> topology is finished. To trigger the Tx change user selection 
>>>> should be
>>>> changed in NVM using devlink. Then the platform should be rebooted.
>>>>
>>>> Example way of switching NVM settings in devlink:
>>>>
>>>> Show:
>>>> devlink dev param show pci/0000:4b:00.0
>>>>    name txbalancing type driver-specific
>>>>      values:
>>>>        cmode permanent value true
>>>>
>>>> Set:
>>>> devlink dev param set pci/0000:4b:00.0 name txbalancing value true 
>>>> cmode
>>>> permanent
>>>
>>> Did I miss the devlink implementation?
>>
>> It's not upstreamed yet, however I had to provide some example on how 
>> this might be
>> configured. There was a review comment about that. I tested it by 
>> using epct with OOT
>> driver first. Obviously I can't provide this method for upstream 
>> community.
>
> As there is a plan to send the devlink implementation soon, these 
> should all be put together into a series that has the functionality as 
> well as the ability to get/set it. If it is too large for a series, 
> provide some method to view the other portions of code such as RFC 
> patches or point to a repo that contains it.
>
> I don't believe functionality that would, currently, require 
> installing the OOT driver to change modes is acceptable.



More information about the Intel-wired-lan mailing list