On 3/28/2018 11:15 AM, Zhang, Qi Z wrote: > Hi Ferruh: > >> -----Original Message----- >> From: Yigit, Ferruh >> Sent: Wednesday, March 28, 2018 6:01 PM >> To: Zhang, Qi Z <qi.z.zh...@intel.com>; Olivier Matz >> <olivier.m...@6wind.com>; dev@dpdk.org; Xing, Beilei >> <beilei.x...@intel.com>; Lu, Wenzhuo <wenzhuo...@intel.com> >> Subject: Re: [dpdk-dev] [PATCH] net/i40evf: add multicast MAC address >> filtering >> >> On 3/27/2018 2:51 PM, Zhang, Qi Z wrote: >>> >>> >>>> -----Original Message----- >>>> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Olivier Matz >>>> Sent: Thursday, January 25, 2018 10:36 PM >>>> To: dev@dpdk.org; Xing, Beilei <beilei.x...@intel.com>; Zhang, Qi Z >>>> <qi.z.zh...@intel.com>; Lu, Wenzhuo <wenzhuo...@intel.com> >>>> Subject: [dpdk-dev] [PATCH] net/i40evf: add multicast MAC address >>>> filtering >>>> >>>> Add support the set_mc_addr_list device operation in the i40evf PMD. >>>> >>>> The configured addresses are stored in the device private area, so >>>> they can be flushed before adding new ones. >>>> >>>> Signed-off-by: Olivier Matz <olivier.m...@6wind.com> >>> >>> Acked-by: Qi Zhang <qi.z.zh...@intel.com> >>> >>> Thanks for enable this! >> >> Hi Qi, >> >> This feature was already documented as supported in i40e_vf.ini which >> seems wrong but now it is supported. >> >> i40e also documents this feature as supported, can you please check if the >> feature is supported there and update i40e.ini if required? > > Do you mean > > " Multicast MAC filter = Y" ? > > My understanding is, though i40e PF/VF doesn't support set_mc_addr_list, but > we still can use mac_addr_add to add a multi cast address, so we still have > that flag on in i40e.ini.
Right, current "mac_addr_add" dev_ops lets setting multicast addresses. In this logic if HW supports setting multicast address PMD may claim this support. Or this may explicitly mean implementing set_mc_addr_list dev_ops in driver. These feature list is to help user / application developer about what to expect and what to not from drivers. A user may expect set_mc_addr_list dev_ops is implemented when this feature documented and it may be confusing when it fails. Overall which information, hw supports setting mc mad address or set_mc_addr_list implemented, adds more value from user point of view? What do you think? > > Regards > Qi > >> >> Thanks, >> ferruh
