Re: [RFC PATCH net-next 0/4] switchdev: avoid duplicate packet forwarding

2015-06-17 Thread Jamal Hadi Salim
On 06/16/15 17:11, Jiri Pirko wrote: Tue, Jun 16, 2015 at 06:47:47PM CEST, sfel...@gmail.com wrote: On Mon, Jun 15, 2015 at 11:04 PM, Jiri Pirko wrote: So what we need is a unique mark for device ports within a fwding group, such as a bridge. Yep, have a group of netdevs, pick one of them

Re: [RFC PATCH net-next 0/4] switchdev: avoid duplicate packet forwarding

2015-06-17 Thread Scott Feldman
On Tue, Jun 16, 2015 at 11:30 PM, Jiri Pirko wrote: > Scott, again, I'm not talking about br0 ifindex! I'm talking about > ifindexes of the port netdevs. > > br0 > sw1p1 ifindex=1 > sw1p2 ifindex=2 > sw2p1 ifindex=3 > sw2p2 ifindex=4 > > > now we have 2 groups, one in each switch. So

Re: [RFC PATCH net-next 0/4] switchdev: avoid duplicate packet forwarding

2015-06-16 Thread Jiri Pirko
Wed, Jun 17, 2015 at 01:53:56AM CEST, sfel...@gmail.com wrote: >On Tue, Jun 16, 2015 at 2:11 PM, Jiri Pirko wrote: >> Tue, Jun 16, 2015 at 06:47:47PM CEST, sfel...@gmail.com wrote: >>>On Mon, Jun 15, 2015 at 11:04 PM, Jiri Pirko wrote: Tue, Jun 16, 2015 at 01:25:51AM CEST, da...@davemloft.ne

Re: [RFC PATCH net-next 0/4] switchdev: avoid duplicate packet forwarding

2015-06-16 Thread Scott Feldman
On Tue, Jun 16, 2015 at 2:11 PM, Jiri Pirko wrote: > Tue, Jun 16, 2015 at 06:47:47PM CEST, sfel...@gmail.com wrote: >>On Mon, Jun 15, 2015 at 11:04 PM, Jiri Pirko wrote: >>> Tue, Jun 16, 2015 at 01:25:51AM CEST, da...@davemloft.net wrote: From: sfel...@gmail.com Date: Sat, 13 Jun 2015 11:

Re: [RFC PATCH net-next 0/4] switchdev: avoid duplicate packet forwarding

2015-06-16 Thread Jiri Pirko
Tue, Jun 16, 2015 at 06:47:47PM CEST, sfel...@gmail.com wrote: >On Mon, Jun 15, 2015 at 11:04 PM, Jiri Pirko wrote: >> Tue, Jun 16, 2015 at 01:25:51AM CEST, da...@davemloft.net wrote: >>>From: sfel...@gmail.com >>>Date: Sat, 13 Jun 2015 11:04:26 -0700 >>> The switchdev port driver must do two

Re: [RFC PATCH net-next 0/4] switchdev: avoid duplicate packet forwarding

2015-06-16 Thread Scott Feldman
On Mon, Jun 15, 2015 at 11:04 PM, Jiri Pirko wrote: > Tue, Jun 16, 2015 at 01:25:51AM CEST, da...@davemloft.net wrote: >>From: sfel...@gmail.com >>Date: Sat, 13 Jun 2015 11:04:26 -0700 >> >>> The switchdev port driver must do two things: >>> >>> 1) Generate a fwd_mark for each switch port, using s

Re: [RFC PATCH net-next 0/4] switchdev: avoid duplicate packet forwarding

2015-06-15 Thread Jiri Pirko
Tue, Jun 16, 2015 at 01:25:51AM CEST, da...@davemloft.net wrote: >From: sfel...@gmail.com >Date: Sat, 13 Jun 2015 11:04:26 -0700 > >> The switchdev port driver must do two things: >> >> 1) Generate a fwd_mark for each switch port, using some unique key of the >>switch device (and optionally po

Re: [RFC PATCH net-next 0/4] switchdev: avoid duplicate packet forwarding

2015-06-15 Thread David Miller
From: sfel...@gmail.com Date: Sat, 13 Jun 2015 11:04:26 -0700 > The switchdev port driver must do two things: > > 1) Generate a fwd_mark for each switch port, using some unique key of the >switch device (and optionally port). This is a one-time operation done >when port's netdev is setup

Re: [RFC PATCH net-next 0/4] switchdev: avoid duplicate packet forwarding

2015-06-15 Thread roopa
On 6/13/15, 11:04 AM, sfel...@gmail.com wrote: From: Scott Feldman (RFC because we're at rc7+ now) With switchdev support for offloading L2/L3 forwarding data path to a switch device, we have a general problem where both the device and the kernel may forward the packet, resulting in duplicate

Re: [RFC PATCH net-next 0/4] switchdev: avoid duplicate packet forwarding

2015-06-15 Thread roopa
On 6/13/15, 11:04 AM, sfel...@gmail.com wrote: From: Scott Feldman (RFC because we're at rc7+ now) With switchdev support for offloading L2/L3 forwarding data path to a switch device, we have a general problem where both the device and the kernel may forward the packet, resulting in duplicate

[RFC PATCH net-next 0/4] switchdev: avoid duplicate packet forwarding

2015-06-13 Thread sfeldma
From: Scott Feldman (RFC because we're at rc7+ now) With switchdev support for offloading L2/L3 forwarding data path to a switch device, we have a general problem where both the device and the kernel may forward the packet, resulting in duplicate packets on the wire. Anytime a packet is forwarde