linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Ioana Ciornei <ioana.ciornei@nxp.com>
To: Andrew Lunn <andrew@lunn.ch>
Cc: "gregkh@linuxfoundation.org" <gregkh@linuxfoundation.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"f.fainelli@gmail.com" <f.fainelli@gmail.com>
Subject: RE: [PATCH 06/12] staging: dpaa2-ethsw: add ACL entry to redirect STP to CPU
Date: Wed, 6 Nov 2019 13:47:47 +0000	[thread overview]
Message-ID: <DB6PR0402MB27899B298481E7A3460BB9BFE0790@DB6PR0402MB2789.eurprd04.prod.outlook.com> (raw)
In-Reply-To: <20191105155954.GE17620@lunn.ch>

> Subject: Re: [PATCH 06/12] staging: dpaa2-ethsw: add ACL entry to redirect
> STP to CPU
> 
> > The control queues do not form an actual interface in the sense that
> > the CPU does not receive unknown unicast, broadcast and multicast
> > frames by default.  For each frame that we want to direct to the CPU
> > we must add an ACL entry.
> 
> So this appears to be one of the dumbest switches so far :-(

To be fair, the actual hardware can do much more.
The problem here is that the firmware models the switch in a non-Linux style.

> 
> Can you add an ACL which is all L2 broadcast/multicast?  That would be a good
> first step.

I can add that but it would not be enough.
For example, unknown unicast would not be matched thus would not reach the CPU.

> 
> Does the ACL stop further processing of the frame? Ideally you want the
> switch to also flood broadcast/multicast out other ports, if they are in a
> bridge. If it cannot, you end up with the software bridge doing the flooding.

Yes, the ACL stops any further processing. 

> 
> So i also assume it does not perform learning on CPU frames? That probably
> means you need to connect up the fdb add/remove calls to add in ACLs. And
> you will need to implement ndo_set_rx_mode. Each unicast and multicast
> address needs to be turned into an ACL. What i don't know is if the network
> stack will automatically add the interfaces own MAC address. You might have
> to handle that special case.
> 
>     Andrew

Your assumption is true, learning, with the current implementation, is not possible for CPU frames.
In .ndo_start_xmit() we inject directly into the switch port's Tx queues, thus bypassing the entire learning process.

All in all, I do not see a way out just by using the ACL mechanism (because of unknown unicast frames).

I have to talk in detail with the firmware team, but from what I can understand if we make the following changes in firmware it would better fit the Linux framework:
 * make the CPU receive unknown unicast, broadcast and multicast frames by default (without any ACLs)
 * frames originated on the CPU should not bypass the learning process (it should have its own Tx queues that go through the same learning process)

Thanks,
Ioana



  reply	other threads:[~2019-11-06 13:47 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-11-05 12:34 [PATCH 00/12] staging: dpaa2-ethsw: add support for control interface traffic Ioana Ciornei
2019-11-05 12:34 ` [PATCH 01/12] staging: dpaa2-ethsw: get control interface attributes Ioana Ciornei
2019-11-05 12:34 ` [PATCH 02/12] staging: dpaa2-ethsw: setup buffer pool for control traffic Ioana Ciornei
2019-11-05 12:34 ` [PATCH 03/12] staging: dpaa2-ethsw: setup RX path rings Ioana Ciornei
2019-11-05 12:34 ` [PATCH 04/12] staging: dpaa2-ethsw: setup dpio Ioana Ciornei
2019-11-05 12:34 ` [PATCH 05/12] staging: dpaa2-ethsw: add ACL table at port probe Ioana Ciornei
2019-11-05 12:34 ` [PATCH 06/12] staging: dpaa2-ethsw: add ACL entry to redirect STP to CPU Ioana Ciornei
2019-11-05 14:22   ` Andrew Lunn
2019-11-05 14:31     ` Ioana Ciornei
2019-11-05 15:59       ` Andrew Lunn
2019-11-06 13:47         ` Ioana Ciornei [this message]
2019-11-06 14:50           ` Andrew Lunn
2019-11-06 15:22             ` Ioana Ciornei
2019-11-06 16:01               ` Andrew Lunn
2019-11-05 12:34 ` [PATCH 07/12] staging: dpaa2-ethsw: seed the buffer pool Ioana Ciornei
2019-11-05 12:34 ` [PATCH 08/12] staging: dpaa2-ethsw: handle Rx path on control interface Ioana Ciornei
2019-11-05 14:31   ` Andrew Lunn
2019-11-05 12:34 ` [PATCH 09/12] staging: dpaa2-ethsw: add .ndo_start_xmit() callback Ioana Ciornei
2019-11-05 12:34 ` [PATCH 10/12] staging: dpaa2-ethsw: enable the CTRL_IF based on the FW version Ioana Ciornei
2019-11-05 12:34 ` [PATCH 11/12] staging: dpaa2-ethsw: enable the control interface Ioana Ciornei
2019-11-05 12:34 ` [PATCH 12/12] staging: dpaa2-ethsw: remove control traffic from TODO file Ioana Ciornei
2019-11-05 13:24 ` [PATCH 00/12] staging: dpaa2-ethsw: add support for control interface traffic Greg KH
2019-11-05 13:49   ` Ioana Ciornei
2019-11-05 14:02   ` Andrew Lunn
2019-11-05 14:22     ` Ioana Ciornei
2019-11-05 14:44       ` Andrew Lunn
2019-11-05 15:32     ` Greg KH

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=DB6PR0402MB27899B298481E7A3460BB9BFE0790@DB6PR0402MB2789.eurprd04.prod.outlook.com \
    --to=ioana.ciornei@nxp.com \
    --cc=andrew@lunn.ch \
    --cc=f.fainelli@gmail.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=linux-kernel@vger.kernel.org \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).