[spp] [spp 03539] Re: [PATCH 0/6] Replace deprecated APIs

[Hideyuki Yamashita]

Hello Yasufumi-san,

Please see inline.

> > Hello Yasufumi-san,
> >
> > Thanks for your comments.
> > Please see inline.
> >
> >>> Hello Yasufumi-san,
> >>>
> >>> Thanks so much for your comments.
> >>>
> >>> About your first point, I think the follwing naming may be fit with DPDK
> >>> and SPP naming conventions, what do you think?
> >>>
> >>> attach -> spp_eth_dev_attach
> >>> detach -> spp_eth_dev_detach
> >> I think you should not use prefix `spp` without if it defines a behaviour of SPP itself. This function is more essential to attach device.
> >>
> >> For `spp_eth_dev_attach`, You might think similar name from original `rte_eth_dev_attach`, but I think we do not keep the name of deprecated APIs. More simply, how about `dev_attach_by_devargs` instead of?
> > I have no problem about naming `dev_attach_by_devargs`.
> >
> >> For `spp_eth_dev_detach`, I think `dev_detach_by_port_id` is more prefer. Other than the name of the function, I am curious why it takes second argument `name` because it is not used in the function.
> > I have no problem about namign  `dev_detach_by_port_id`.
> > As you say, second argument is NOT used within the function and I will
> > remove it from revised version.
> >
> >> By the way, I am not clear what is the difference between your functions and `rte_eth_dev_attach` defined in `dpdk/lib/librte_ethdev/rte_ethdev.c` exactly. Could you explain about the difference shortly for helping my understanding?
> > Basically both of those functions uses `rte_eal_hotplug_add` to hotplug
> > device.
> Thank you for considering updates.
> >
> > Difference is described as following:
> > `rte_eth_dev_attach` uses global variable named
> > eth_dev_last_created_port when retrieving newly created dpdk port while
> > my new function uses `rte_eth_dev_get_port_by_name` becauase this function is defined outside
> > of dpdk library and can not retrieve global variable within dpdk library.
> I've got understand your point. Could you confirm why you avoid to use global variable?
rte_eth_dev_attach is function inside dpdk library so it is somehow
natural to use global variable when retrieve newly created dpdk port.
As I said earlier, this time new function is located in application side.
It is unnatural to use global variable of library from application
rather I prefer to use existing function for retrieving allocated dpdk
from appliction(rte_eth_dev_get_port_by_name).

I hope above explanation answers to your question.
BTW, I am preparing revising patch set including cover letter.

BR,
Hideyuki Yamashita
NTT TechnoCross

> Thanks
> > BR,
> > Hideyuki Yamashita
> > NTT TechnoCross
> >  >> Thanks
> >>> About your second point, I appologize about my mistake.
> >>>
> >>> Once I get ack for above new naming of two interfaces from you,
> >>> I will revise my patch set including cover-letter and will send
> >>> those to mailing list.
> >>>
> >>> Thanks for your co-opearation!
> >>>
> >>> BR,
> >>> Hideyuki Yamashita
> >>> NTT TechnoCorss
> >>>
> >>>> On 2018/11/07 14:07, x-fn-spp at sl.ntt-tx.co.jp wrote:
> >>>>> From: Hideyuki Yamashita <yamashita.hideyuki at po.ntt-tx.co.jp>
> >>>>>
> >>>>> >From DPDK-18.08, the follwing APIs become deprecated and
> >>>>> will be deleted in DPDK18.08.
> >>>>> - rte_eth_dev_attach()
> >>>>> - rte_eth_dev_detach()
> >>>>>
> >>>>> For rte_eth_dev_attach(), use of rte_eal_hotplug_add() is recommended.
> >>>>> For rte_eth_dev_detach(), use of rte_eal_hotplug_remove() is recommended.
> >>>>>
> >>>>> To follow the above changes, this patch set provides replacement of
> >>>>> those APIs.
> >>>> Hideyuki,
> >>>>
> >>>> Thank you for suggesting to update to v18.08!
> >>>>
> >>>> Could you re-consider the name of function you added by referring conventions of DPDK and SPP? It has almost no means if just `attach` or `detach`. It is preferable to be self explanatory for how your function works.
> >>>>
> >>>> Commit messages are also required to be revised. Update for the change of function name and modify invalid descriptions for documentation guidelines.
> >>>>
> >>>> Thanks
> >>>>> Signed-off-by: Hideyuki Yamashita <yamashita.hideyuki at po.ntt-tx.co.jp>
> >>>>> Signed-off-by: Naoki Takada <takada.naoki at lab.ntt.co.jp>
> >>>>>
> >>>>> Hideyuki Yamashita (6):
> >>>>>      shared: addition of attach()
> >>>>>      spp_nfv: replacement of rte_eth_dev_attach()
> >>>>>      spp_vf:replacement of rte_eth_dev_attach()
> >>>>>      shared: addition of detach()
> >>>>>      spp_nfv: replacement of rte_eth_dev_detach()
> >>>>>      spp_vm: replacement of rte_eth_dev_detach().
> >>>>>
> >>>>>     src/nfv/nfv.c       | 12 ++++-----
> >>>>>     src/shared/common.c | 64 +++++++++++++++++++++++++++++++++++++++++++++
> >>>>>     src/shared/common.h | 32 +++++++++++++++++++++++
> >>>>>     src/vf/spp_vf.c     |  4 +--
> >>>>>     src/vm/main.c       |  2 +-
> >>>>>     5 files changed, 105 insertions(+), 9 deletions(-)
> >>>>>
> >>>>
> >>>>
> >
> >
> >
> > 
> 
> -- Yasufumi Ogawa
> NTT Network Service Systems Labs