Re: [patch net-next v2 14/15] devlink: move port_del() to devlink_port_ops

[Jiri Pirko <jiri@resnulli.us>]

Mon, May 29, 2023 at 08:33:34AM CEST, kuba@kernel.org wrote:
>On Sat, 27 May 2023 09:42:45 +0200 Jiri Pirko wrote:
>> >I didn't think this thru last time, I thought port_new will move 
>> >in another patch, but that's impossible (obviously?).
>> >
>> >Isn't it kinda weird that the new callback is in one place and del
>> >callback is in another? Asymmetric ?  
>> 
>> Yeah, I don't know how to do it differently. port_new() has to be
>> devlink op, as it operates not on the port but on the device. However,
>> port_del() operates on device. I was thinking about changing the name of
>> port_del() to port_destructor() or something like that which would make
>> the symmetricity issue bit less visible. IDK, up to you. One way or
>> another, I think this could be easily done as a follow-up (I have 15
>> patches now already anyway).
>
>One could argue logically removing a port is also an operation of 
>the parent (i.e. the devlink instance). The fact that the port gets
>destroyed in the process is secondary. Ergo maybe we should skip 
>this patch?

Well, the port_del() could differ for different port flavours. The
embedding structure of struct devlink_port is also different.

Makes sense to me to skip the flavour switch and have one port_del() for
each port.