[Intel-wired-lan] [PATCH v1 1/1] igc: Complete to commit Add support for TSO

[Neftin, Sasha]

On 2/13/2020 00:47, Alexander Duyck wrote:
> On Wed, Feb 12, 2020 at 2:39 PM Brown, Aaron F <aaron.f.brown at intel.com> wrote:
>>
>>> From: Intel-wired-lan <intel-wired-lan-bounces at osuosl.org> On Behalf Of
>>> Sasha Neftin
>>> Sent: Wednesday, February 5, 2020 4:31 AM
>>> To: intel-wired-lan at lists.osuosl.org
>>> Subject: [Intel-wired-lan] [PATCH v1 1/1] igc: Complete to commit Add
>>> support for TSO
>>>
>>> commit f38b782dccab ("igc: Add support for TSO")
>>> Add option to setting transmit command (TUCMD) of the context
>>> descriptor based on skb_shinfo gso_type and SKB_GSO_UDP_L4 flag.
>>>
>>> Signed-off-by: Sasha Neftin <sasha.neftin at intel.com>
>>> ---
>>>   drivers/net/ethernet/intel/igc/igc_defines.h | 1 +
>>>   drivers/net/ethernet/intel/igc/igc_main.c    | 3 ++-
>>>   2 files changed, 3 insertions(+), 1 deletion(-)
>>>
>> Aside from possibly including the responses to Jesse's questions on this...
>> Tested-by: Aaron Brown <aaron.f.brown at intel.com>
> 
> I'm not sure the patch makes any sense. Does the driver support UDP
> GSO? I don't see the feature flag (NETIF_F_GSO_UDP_L4) anywhere that
> enables it.
> 
Yes, it is supported. The MAC is similar to igb from segmentation 
perspectives. I missed add it. Please, let me finish some testing and I 
will add NETIF_F_GSO_UDP_L4 flag to _features_check and _probe methods. 
I will process another patch with complements.
> If it doesn't enable it then it doesn't make much sense to update the
> code to add this one bit until it does since you can't actually test
> it as the stack will not ask you to segment UDP frames.
>