Messages in this thread | | | Date | Fri, 13 Jan 2023 13:14:01 +0200 | From | Vladimir Oltean <> | Subject | Re: [PATCH v4 1/3] dsa: marvell: Provide per device information about max frame size |
| |
On Fri, Jan 13, 2023 at 12:02:19PM +0100, Lukasz Majewski wrote: > Hi Vladimir, > > > On Fri, Jan 13, 2023 at 11:39:08AM +0100, Lukasz Majewski wrote: > > > Are there any more comments, or is this patch set eligible for > > > pulling into net-next tree? > > > > How about responding to the comment that was already posted first? > > Could you be more specific? > > > On the beginning (first posted version) the patch included 9 patches > (which included work for ADDR4 for some mv88e6020 setup). > > But after the discussion, I've decided to split this patch set to > smaller pieces; > > First to add the set_max_frame size with basic definition for mv88e6020 > and mv88e6071 and then follow with more complicated changes (for which > there is no agreement on how to tackle them). > > For the 'set_max_frame' feature Alexander Dyuck had some comments > regarding defensive programming approach, but finally he agreed with > Andrew's approach. > > As of now - the v4 has been Acked by Andrew, so it looks like at least > this "part" of the work is eligible for upstreaming. > > > Or there are any more issues about which I've forgotten ?
Do you agree that for the chip families which neither implement port_set_jumbo_size() nor set_max_frame_size(), a max MTU of 1492 will be returned, which currently produces warnings at probe time and should be fixed first, prior to refactoring the code? https://patchwork.kernel.org/project/netdevbpf/patch/20230106101651.1137755-1-lukma@denx.de/#25149891
| |