12/02/2024 09:37, Oleksandr Kolomeiets:
> From: Dariusz Sosnowski <dsosnow...@nvidia.com>
> > From: Thomas Monjalon <tho...@monjalon.net>
> > > 01/02/2024 10:59, Oleksandr Kolomeiets:
> > > > "flow actions_update" updates a flow rule specified by a rule ID with
> > > >
> > > > a new action list by making a call to "rte_flow_actions_update()":
> > > > flow actions_update {port_id} {rule_id}
> > > >
> > > > actions {action} [/ {action} [...]] / end [user_id]
> > > >
> > > > Creating, updating and destroying a flow rule:
> > > > testpmd> flow create 0 group 1 pattern eth / end actions drop /
> > > > end
> > > > Flow rule #0 created
> > > > testpmd> flow actions_update 0 0 actions queue index 1 / end
> > > > Flow rule #0 updated with new actions
> > > > testpmd> flow destroy 0 rule 0
> > > > Flow rule #0 destroyed
> > >
> > > Why not a simple "flow update" command name?
> >
> > +1. This would also make it consistent with async version of this command
> > - "flow queue {port_id} update ...".
>
> Indeed, shortening the command from "flow actions_update" to "flow update"
> seems more natural.
> However, note that the command updates only the actions of a flow rule and
> leaves all other parameters unchanged.
> My concern is that in the future there can be some "flow pattern_update"
> command, thus making "flow update" command ambiguous.
> Also, the name is consistent with the underlying rte_flow_actions_update()
> function.
> With that in mind, please clarify if the name should still be changed.
If a function is added for pattern update, we could still implement it
with the same command prefix "flow update" and call functions as appropriate.
So yes I still think the command should be "flow update".
