netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
To: Florian Fainelli <f.fainelli@gmail.com>,
	Gavin Shan <gwshan@linux.vnet.ibm.com>,
	netdev@vger.kernel.org
Cc: davem@davemloft.net, joel@jms.id.au, weixue@trustnetic.com
Subject: Re: [PATCH net-next 00/10] NCSI Support
Date: Fri, 08 Jul 2016 08:05:39 +1000	[thread overview]
Message-ID: <1467929139.27157.57.camel@kernel.crashing.org> (raw)
In-Reply-To: <577E921C.5050904@gmail.com>

On Thu, 2016-07-07 at 10:32 -0700, Florian Fainelli wrote:
> 
> I know nothing about NCSI, pretty much like Jon Snow, but from a cursory
> look at your patches, is not there a way to make the NCSCI capable
> network devices strictly adhere to the net_device APIs and calling
> conventions?

They already do.

Basically imagine NC-SI being a replacement for the PHY of the device.

Instead of having a PHY, your MII is directly connected to another NIC
(the main system NIC) which conveniently provide that sideband MII
interface for you.

You then use the NC-SI protocol which uses a specific ethernet packet
type to communicate with that other NIC as you would otherwise communicate
with a PHY using MDIO. At least conceptually. For config. Normal traffic
is still normal ethernet.

The protocol is more complex than MDIO though, you can be connected to
multiple NICs, you have to provide your MAC address to the other guy so
he can route some incoming traffic to you etc...

So in the current implementation by Gavin, all that is needed for an
existing NIC driver to support NC-SI is a few lines of code to register
with the stack, which will under the hood create the ncsi_dev associated
with the net_dev and manage the protocol.

>From a "user" point of view you still have a normal netdev doing normal
ethernet, the NC-SI protocol is transparent and strictly used to maintain
the link active between you and the other NIC thorugh which your traffic
goes.

Cheers,
Ben.

  reply	other threads:[~2016-07-07 22:05 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-07-03  5:32 [PATCH net-next 00/10] NCSI Support Gavin Shan
2016-07-03  5:32 ` [PATCH net-next 01/10] net/ncsi: Resource management Gavin Shan
2016-07-03  5:32 ` [PATCH net-next 02/10] net/ncsi: NCSI command packet handler Gavin Shan
2016-07-03  5:32 ` [PATCH net-next 03/10] net/ncsi: NCSI response " Gavin Shan
2016-07-03  5:32 ` [PATCH net-next 04/10] net/ncsi: Package and channel management Gavin Shan
2016-07-03  5:32 ` [PATCH net-next 05/10] net/ncsi: NCSI AEN packet handler Gavin Shan
2016-07-03  5:32 ` [PATCH net-next 06/10] net/farady: Helper functions to create or destroy MDIO interface Gavin Shan
2016-07-03  5:32 ` [PATCH net-next 07/10] net/farady: Read MAC address from chip Gavin Shan
2016-07-03  5:32 ` [PATCH net-next 08/10] net/farady: Support NCSI mode Gavin Shan
2016-07-03  5:32 ` [PATCH net-next 09/10] net/farady: Match driver according to compatible property Gavin Shan
2016-07-03  5:32 ` [PATCH net-next 10/10] net/farady: Mask PHY interrupt with NCSI mode Gavin Shan
2016-07-03 22:03 ` [PATCH net-next 00/10] NCSI Support Or Gerlitz
2016-07-03 22:49   ` Benjamin Herrenschmidt
2016-07-04  0:24     ` Gavin Shan
2016-07-05 17:44   ` Alexei Starovoitov
2016-07-05 21:42     ` Benjamin Herrenschmidt
2016-07-06  2:07       ` Alexei Starovoitov
2016-07-06  2:14         ` Benjamin Herrenschmidt
2016-07-07  9:12     ` Or Gerlitz
2016-07-07  9:17       ` Benjamin Herrenschmidt
2016-07-07  9:18         ` Benjamin Herrenschmidt
2016-07-07 13:05         ` Gavin Shan
2016-07-07 13:44         ` Or Gerlitz
2016-07-07 16:34           ` Gavin Shan
2016-07-07 17:32 ` Florian Fainelli
2016-07-07 22:05   ` Benjamin Herrenschmidt [this message]
2016-07-08  1:10   ` Gavin Shan

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=1467929139.27157.57.camel@kernel.crashing.org \
    --to=benh@kernel.crashing.org \
    --cc=davem@davemloft.net \
    --cc=f.fainelli@gmail.com \
    --cc=gwshan@linux.vnet.ibm.com \
    --cc=joel@jms.id.au \
    --cc=netdev@vger.kernel.org \
    --cc=weixue@trustnetic.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 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).