Re: [dpdk-dev] [PATCH v2 1/2] net/pcap: physical interface MAC support

[Ferruh Yigit <ferruh.yigit@intel.com>]

On 4/18/2018 5:35 AM, Kuusisaari, Juhamatti wrote:
> Hello Ferruh,
> 
>> On 4/17/2018 1:53 PM, Juhamatti Kuusisaari wrote:
>>> Support for PCAP MAC address using physical interface MAC.
>>> Support for getting proper link status, speed and duplex.
>>>
>>> Signed-off-by: Juhamatti Kuusisaari <juhamatti.kuusisaari@coriant.com>
>>> ---
>>>  config/common_base              |  1 +
>>>  drivers/net/pcap/rte_eth_pcap.c | 52
>>> ++++++++++++++++++++++++++++++++++++++++-
>>>  2 files changed, 52 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/config/common_base b/config/common_base index
>>> c2b0d91..9804585 100644
>>> --- a/config/common_base
>>> +++ b/config/common_base
>>> @@ -410,6 +410,7 @@ CONFIG_RTE_LIBRTE_PMD_NULL=y  # Compile
>> software
>>> PMD backed by PCAP files  #  CONFIG_RTE_LIBRTE_PMD_PCAP=n
>>> +CONFIG_RTE_LIBRTE_PMD_PCAP_IF_MAC_SUPPORT=n
>>
>> Hi Juhamatti,
>>
>> Why a build time config option for this? Can we make it a runtime devarg?
> 
> Sure, we can make it a devarg. Or do we even need that? Are there a lot of test dependencies that would need to be fixed if we have it enabled by default?

Not test dependencies but this may be overkill for some usecases, I prefer
making this dynamically configurable, no strong opinion though.

> 
>> Overall we are trying to reduce config options already and this seems no
>> need to be build time option at all.
>>
>> btw, this is a little late in release cycle, so lets target this patch for next
>> release.
> 
> The patch is on top of net-next, this should be just fine.

Perhaps we should rename the sub-tree :) because this is not happening first
time. next-net is not for next release, as it has been Linux, it is for this
release but just a sub-tree for net PMDs.

> 
>> Thanks,
>> ferruh
> 
> Thanks,
> --
>  Juhamatti
>