[dpdk-dev] [PATCH 2/2] ethdev: allow unknown link speed

[Ferruh Yigit]

On 4/13/2020 3:26 PM, Andrew Rybchenko wrote:
> On 4/8/20 1:26 AM, Thomas Monjalon wrote:
>> When querying the link informations, the link status is
>> a mandatory major information.
>> Other boolean values are supposed to be accurate:
>> 	- duplex mode (half/full)
>> 	- negotiation (auto/fixed)
>>
>> This API update is making explicit that the link speed information
>> is optional.
>> The value ETH_SPEED_NUM_NONE (0) was already part of the API.
>> The value ETH_SPEED_NUM_UNKNOWN (infinite) is added to cover
>> two different cases:
>> 	- speed is not known by the driver
>> 	- device is virtual
>>
>> Suggested-by: Morten Brørup <mb at smartsharesystems.com>
>> Suggested-by: Benoit Ganne <bganne at cisco.com>
>> Signed-off-by: Thomas Monjalon <thomas at monjalon.net>
> 
> I'm afraid it requires more efforts to keep things consistent,
> e.g.:
>  - printing of link speed in proc-info
>  - comparison of rate vs link_speed in testpmd
>  - printing of link speed in testpmd as %u Mbps
>  - division by 1000 in test-pipeline
>  - link speed printing in doc/guides/sample_app_ug/link_status_intr.rst
>  - link speed printing in many-many examples
> 
> IMHO, it is not nice to print UING32_MAX and let use
> guess what it means.
> 

+1

Hi Thomas,

The two patches in the set seems unrelated, and first one looks OK, do you want
to separate them so that this doesn't block that one?