All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jamal Hadi Salim <jhs@mojatatu.com>
To: Jiri Pirko <jiri@resnulli.us>, Scott Feldman <sfeldma@gmail.com>
Cc: John Fastabend <john.r.fastabend@intel.com>,
	Netdev <netdev@vger.kernel.org>,
	"David S. Miller" <davem@davemloft.net>,
	"nhorman@tuxdriver.com" <nhorman@tuxdriver.com>,
	Andy Gospodarek <andy@greyhouse.net>, Thomas Graf <tgraf@suug.ch>,
	"dborkman@redhat.com" <dborkman@redhat.com>,
	"ogerlitz@mellanox.com" <ogerlitz@mellanox.com>,
	"jesse@nicira.com" <jesse@nicira.com>,
	"pshelar@nicira.com" <pshelar@nicira.com>,
	"azhou@nicira.com" <azhou@nicira.com>,
	"ben@decadent.org.uk" <ben@decadent.org.uk>,
	"stephen@networkplumber.org" <stephen@networkplumber.org>,
	"Kirsher, Jeffrey T" <jeffrey.t.kirsher@intel.com>,
	"vyasevic@redhat.com" <vyasevic@redhat.com>,
	Cong Wang <xiyou.wangcong@gmail.com>,
	Eric Dumazet <edumazet@google.com>,
	Florian Fainelli <f.fainelli@gmail.com>,
	Roopa Prabhu <roopa@cumulusnetworks.com>,
	John Linville <linville@tuxdriver.com>,
	"jasowang@redhat.com" <jasow
Subject: Re: [patch net-next v3 02/17] net: make vid as a parameter for ndo_fdb_add/ndo_fdb_del
Date: Fri, 28 Nov 2014 07:57:00 -0500	[thread overview]
Message-ID: <5478711C.8080700@mojatatu.com> (raw)
In-Reply-To: <20141127215529.GA17784@nanopsycho.orion>

On 11/27/14 16:55, Jiri Pirko wrote:
> Thu, Nov 27, 2014 at 09:59:37PM CET, sfeldma@gmail.com wrote:
>> Ya right now the driver just doesn't call br_fdb_external_learn_add()
>> if LEARNING_SYNC is not set.  It's a port driver setting so it seems
>> fine to handle it in the port driver.  You could move the check up to
>> br_fdb_external_learn_add(), but then you have an extra call every 1s
>> for each fdb entry being refreshed.  (1s or whatever the refresh
>> frequency is).  Easier to avoid this overhead and make the decision at
>> the source.
>
> I have been thinking about moving the check into bridge code, it to make
> it there as well as in drivers. This is easily changeable on demenad
> later though, so I left this for now.
>

It seems more comfortable to move to the core if you are doing polling
with timers... i.e you kick it per-offload via some timer. The arming
being done by the setting of LEARNING_SYNC
There are cases where all this is done via interrupts. i.e the hardware
will issue an interrupt only then do you poll..

Maybe Scott's approach is the correct one. I am indifferent to be
honest, these are some of those things that can be easily refactored
as more hardware shows up.

cheers,
jamal

  reply	other threads:[~2014-11-28 12:57 UTC|newest]

Thread overview: 149+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-11-25 10:28 [patch net-next v3 00/17] introduce rocker switch driver with hardware accelerated datapath api - phase 1: bridge fdb offload Jiri Pirko
2014-11-25 10:28 ` [patch net-next v3 01/17] bridge: rename fdb_*_hw to fdb_*_hw_addr to avoid confusion Jiri Pirko
2014-11-25 15:03   ` Andy Gospodarek
2014-11-25 15:26   ` Jamal Hadi Salim
2014-11-25 22:11   ` Thomas Graf
2014-11-25 22:33   ` Florian Fainelli
2014-11-25 10:28 ` [patch net-next v3 02/17] net: make vid as a parameter for ndo_fdb_add/ndo_fdb_del Jiri Pirko
2014-11-25 15:13   ` Andy Gospodarek
2014-11-25 15:18     ` Jiri Pirko
2014-11-25 15:43       ` John Fastabend
2014-11-25 16:38         ` Jiri Pirko
2014-12-09 11:57         ` Or Gerlitz
2014-11-25 15:38   ` Jamal Hadi Salim
2014-11-25 16:01     ` John Fastabend
2014-11-25 16:18       ` Jamal Hadi Salim
2014-11-25 16:30         ` John Fastabend
2014-11-25 16:50           ` Jamal Hadi Salim
2014-11-26  1:44             ` Simon Horman
2014-11-26  2:36             ` Scott Feldman
2014-11-26  3:19               ` Jamal Hadi Salim
2014-11-26  3:59                 ` Scott Feldman
2014-11-26 11:28                   ` Jamal Hadi Salim
2014-11-26 11:40                     ` Jiri Pirko
2014-11-26 11:54                       ` Jamal Hadi Salim
2014-11-26 12:06                         ` Jamal Hadi Salim
2014-11-27  6:50                     ` Scott Feldman
2014-11-27 12:14                       ` Jamal Hadi Salim
2014-11-27 20:59                         ` Scott Feldman
2014-11-27 21:55                           ` Jiri Pirko
2014-11-28 12:57                             ` Jamal Hadi Salim [this message]
2014-11-28 10:14               ` Roopa Prabhu
2014-11-28 10:33                 ` Scott Feldman
2014-11-25 16:19     ` Roopa Prabhu
2014-11-25 16:33       ` John Fastabend
2014-11-25 16:57         ` Jamal Hadi Salim
2014-11-25 16:43       ` Jiri Pirko
2014-11-25 18:53   ` Samudrala, Sridhar
2014-11-25 20:40     ` Jiri Pirko
2014-11-25 22:14   ` Thomas Graf
2014-11-25 22:39     ` Florian Fainelli
2014-11-25 23:11       ` Thomas Graf
2014-11-26  7:54     ` Jiri Pirko
2014-11-25 10:28 ` [patch net-next v3 03/17] net: rename netdev_phys_port_id to more generic name Jiri Pirko
2014-11-25 15:15   ` Andy Gospodarek
2014-11-25 15:44   ` Jamal Hadi Salim
2014-11-25 10:28 ` [patch net-next v3 04/17] net: introduce generic switch devices support Jiri Pirko
2014-11-25 15:02   ` Andy Gospodarek
2014-11-25 15:51   ` Jamal Hadi Salim
2014-11-25 16:49     ` Jiri Pirko
2014-11-25 17:08       ` Jamal Hadi Salim
2014-11-25 21:54         ` Thomas Graf
2014-11-26  3:33           ` Jamal Hadi Salim
2014-11-26  4:18             ` Scott Feldman
2014-11-26 11:36               ` Jamal Hadi Salim
2014-11-26 16:08                 ` Thomas Graf
2014-11-26 17:09                   ` Jamal Hadi Salim
2014-11-26 17:59                     ` Jiri Pirko
2014-11-26 21:50                       ` Thomas Graf
2014-11-26 23:32                         ` Jamal Hadi Salim
2014-11-27 13:03                           ` Thomas Graf
2014-11-27 13:32                             ` Jamal Hadi Salim
2014-11-27 13:50                               ` Jiri Pirko
2014-11-28 13:13                                 ` Jamal Hadi Salim
2014-11-27  5:58                 ` Scott Feldman
2014-11-27 12:46                   ` Jamal Hadi Salim
2014-11-27  3:13             ` Simon Horman
2014-11-27 12:35               ` Jamal Hadi Salim
2014-11-25 16:07   ` Roopa Prabhu
2014-11-25 16:50     ` Jiri Pirko
2014-11-25 10:28 ` [patch net-next v3 05/17] rtnl: expose physical switch id for particular device Jiri Pirko
2014-11-25 15:16   ` Andy Gospodarek
2014-11-25 10:28 ` [patch net-next v3 06/17] net-sysfs: " Jiri Pirko
2014-11-25 15:28   ` Andy Gospodarek
2014-11-25 15:53   ` Jamal Hadi Salim
2014-11-25 22:27   ` Florian Fainelli
2014-11-25 10:28 ` [patch net-next v3 07/17] rocker: introduce rocker switch driver Jiri Pirko
2014-11-25 15:57   ` Jamal Hadi Salim
2014-11-25 16:57     ` Jiri Pirko
2014-11-25 17:10     ` David Miller
2014-11-25 17:15       ` Jiri Pirko
2014-11-25 17:15         ` Jamal Hadi Salim
2014-11-25 17:51         ` Andy Gospodarek
2014-11-25 16:13   ` David Laight
2014-11-25 19:19     ` Scott Feldman
2014-11-25 21:57       ` tgraf
2014-11-25 10:28 ` [patch net-next v3 08/17] bridge: call netdev_sw_port_stp_update when bridge port STP status changes Jiri Pirko
2014-11-25 15:58   ` Andy Gospodarek
2014-11-25 16:53     ` Jiri Pirko
2014-11-25 22:18       ` Thomas Graf
2014-11-25 22:20     ` Florian Fainelli
2014-11-25 22:48   ` Roopa Prabhu
2014-11-26  1:35     ` Scott Feldman
2014-11-28 10:05       ` Roopa Prabhu
2014-11-28 10:51         ` Scott Feldman
2014-11-28 13:17           ` Jamal Hadi Salim
2014-11-28 13:27             ` Jiri Pirko
2014-11-28 16:23             ` Roopa Prabhu
2014-11-25 10:28 ` [patch net-next v3 09/17] bridge: add API to notify bridge driver of learned FBD on offloaded device Jiri Pirko
2014-11-25 16:01   ` Jamal Hadi Salim
2014-11-25 16:38   ` Andy Gospodarek
2014-11-25 22:36     ` Thomas Graf
2014-11-26  1:48       ` Scott Feldman
2014-11-26 10:26         ` Jiri Pirko
2014-11-25 22:44   ` Florian Fainelli
2014-11-26  2:03     ` Scott Feldman
2014-11-26  2:34       ` Florian Fainelli
2014-11-26  2:40         ` Scott Feldman
2014-11-26  8:16           ` Jiri Pirko
2014-11-26  3:22     ` Jamal Hadi Salim
2014-11-25 10:28 ` [patch net-next v3 10/17] rocker: implement rocker ofdpa flow table manipulation Jiri Pirko
2014-11-25 16:03   ` Jamal Hadi Salim
2014-11-25 10:28 ` [patch net-next v3 11/17] rocker: implement L2 bridge offloading Jiri Pirko
2014-11-25 16:04   ` Jamal Hadi Salim
2014-11-25 17:11     ` David Miller
2014-11-25 10:28 ` [patch net-next v3 12/17] rocker: implement ndo_fdb_dump Jiri Pirko
2014-11-25 16:04   ` Jamal Hadi Salim
2014-11-25 10:28 ` [patch net-next v3 13/17] bridge: move private brport flags to if_bridge.h so port drivers can use flags Jiri Pirko
2014-11-25 16:01   ` Andy Gospodarek
2014-11-25 16:05   ` Jamal Hadi Salim
2014-11-25 22:20   ` Florian Fainelli
2014-11-25 22:48   ` Thomas Graf
2014-11-26 12:12     ` Jiri Pirko
2014-11-25 10:28 ` [patch net-next v3 14/17] bridge: add new brport flag LEARNING_SYNC Jiri Pirko
2014-11-25 16:05   ` Roopa Prabhu
2014-11-25 16:06   ` Jamal Hadi Salim
2014-11-25 16:40   ` Andy Gospodarek
2014-11-25 10:28 ` [patch net-next v3 15/17] bridge: add new hwmode swdev Jiri Pirko
2014-11-25 16:03   ` Roopa Prabhu
2014-11-25 16:17     ` Andy Gospodarek
2014-11-25 22:56       ` Thomas Graf
2014-11-25 16:07   ` Jamal Hadi Salim
2014-11-25 23:00     ` Thomas Graf
2014-11-25 10:28 ` [patch net-next v3 16/17] bridge: add brport flags to dflt bridge_getlink Jiri Pirko
2014-11-25 16:08   ` Jamal Hadi Salim
2014-11-25 17:02     ` Scott Feldman
2014-11-25 17:48   ` Andy Gospodarek
2014-11-25 22:07   ` Thomas Graf
2014-11-26  9:25     ` Jiri Pirko
2014-11-26 10:48       ` Thomas Graf
2014-11-25 10:28 ` [patch net-next v3 17/17] rocker: add ndo_bridge_setlnk/getlink support for learning policy Jiri Pirko
2014-11-25 16:09   ` Jamal Hadi Salim
2014-11-25 18:55     ` Scott Feldman
2014-11-25 19:00       ` Jamal Hadi Salim
2014-11-25 20:42         ` Jiri Pirko
2014-11-26 11:07   ` Thomas Graf
2014-11-26 11:27     ` Jiri Pirko
2014-11-26 11:30       ` Thomas Graf
2014-11-26 11:42         ` Jiri Pirko
  -- strict thread matches above, loose matches on Subject: below --
2014-11-26  7:35 [patch net-next v3 02/17] net: make vid as a parameter for ndo_fdb_add/ndo_fdb_del Arad, Ronen

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=5478711C.8080700@mojatatu.com \
    --to=jhs@mojatatu.com \
    --cc=andy@greyhouse.net \
    --cc=azhou@nicira.com \
    --cc=ben@decadent.org.uk \
    --cc=davem@davemloft.net \
    --cc=dborkman@redhat.com \
    --cc=edumazet@google.com \
    --cc=f.fainelli@gmail.com \
    --cc=jeffrey.t.kirsher@intel.com \
    --cc=jesse@nicira.com \
    --cc=jiri@resnulli.us \
    --cc=john.r.fastabend@intel.com \
    --cc=linville@tuxdriver.com \
    --cc=netdev@vger.kernel.org \
    --cc=nhorman@tuxdriver.com \
    --cc=ogerlitz@mellanox.com \
    --cc=pshelar@nicira.com \
    --cc=roopa@cumulusnetworks.com \
    --cc=sfeldma@gmail.com \
    --cc=stephen@networkplumber.org \
    --cc=tgraf@suug.ch \
    --cc=vyasevic@redhat.com \
    --cc=xiyou.wangcong@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.