mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Andrew Lunn <andrew@lunn.ch>
To: Jonas Jelonek <jonas@jonasjelonek.de>,
	Christian Marangi <ansuelsmth@gmail.com>
Cc: Andrew Lunn <andrew+netdev@lunn.ch>,
	Heiner Kallweit <hkallweit1@gmail.com>,
	Russell King <linux@armlinux.org.uk>,
	"David S . Miller" <davem@davemloft.net>,
	Eric Dumazet <edumazet@kernel.org>,
	Jakub Kicinski <kuba@kernel.org>, Paolo Abeni <pabeni@redhat.com>,
	Simon Horman <horms@kernel.org>,
	netdev@vger.kernel.org, Rob Herring <robh@kernel.org>,
	Krzysztof Kozlowski <krzk+dt@kernel.org>,
	Conor Dooley <conor+dt@kernel.org>,
	Jonathan Corbet <corbet@lwn.net>,
	Shuah Khan <skhan@linuxfoundation.org>,
	Randy Dunlap <rdunlap@infradead.org>,
	linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org,
	devicetree@vger.kernel.org
Subject: Re: [RFC PATCH net-next 0/2] Add Realtek XSGMII interface mode
Date: Sun, 22 Nov 2026 22:24:24 +0100	[thread overview]
Message-ID: <df35e0d4-9f1f-4882-bc6b-257add4e9fb1@lunn.ch> (raw)
In-Reply-To: <20261117084748.2387647-1-jonas@jonasjelonek.de>

On Tue, Nov 17, 2026 at 08:47:46AM +0000, Jonas Jelonek wrote:
> Hi,
> 
> This RFC adds an explicit PHY interface mode for Realtek's proprietary
> XSGMII transport. It carries eight 10/100/1000 Mbps ports over one
> 10.3125 Gbaud SerDes lane. Each logical port remains limited to 1 Gbps;
> the aggregate serial rate is not a per-port Ethernet speed.

We should be careful with naming. We have already have QSGMII and
PSGMII, for Quad and Pent. X is too vague. So we should somehow
indicate 8, Oct.

But...

> Protocol identity and naming
> ----------------------------
> 
> XSGMII is the name Realtek uses for this interface. Calling it simply
> "octal SGMII" would describe its capacity without identifying its
> protocol. Existing modes such as QSGMII and PSGMII already distinguish
> multiplexed MAC-PHY transports; the PSGMII introduction is an example of
> adding the generic mode definitions separately from a driver user [1].
> 
> Cisco USGMII is a relevant comparison, but equivalent port capacity does
> not establish interoperability. Cisco defines Q-USGMII for four ports
> over 5.0 Gb/s and O-USGMII for eight ports over 10.0 Gb/s, using 8b/10b
> coding and a packet-control header [2]. Linux already represents the
> quad variant as "qusgmii", with its own control-word decoding [3].
> 
> Realtek documents XSGMII and O-USGMII as separate RTL8218E interfaces:
> XSGMII runs at 10.3125 Gb/s and O-USGMII at 10.0 Gb/s, with different
> hardware strap selections [4]. The SDK also distinguishes them in
> phy_8218e_macIntfSerdesMode_get() and applies separate initialization
> tables. Realtek's O-USGMII is therefore the candidate for comparison
> with Cisco's octal USGMII; it is not evidence that XSGMII can be renamed
> to USGMII. Compatibility of those O-USGMII implementations would need
> separate confirmation.
> 
> This proposal retains the vendor's XSGMII name rather than assigning an
> existing protocol name to a distinct hardware configuration. It does
> not assume a particular XSGMII line encoding merely from the line rate.

https://deepwiki.com/lotusmomo/airoha_sdk/3.1-sgmiixsgmii-drivers-and-apis

Airoha also seems to have an XSGMII. Is it the same, or different?

Adding Christian.

I hope we are not going to have vendors introducing there own
variants, and we need

PHY_INTERFACE_MODE_REALTEK_OSGMII & PHY_INTERFACE_MODE_AIROHA_OSGMII,
etc.

       Andrew

      parent reply	other threads:[~2026-11-22 21:24 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-17  8:47 Jonas Jelonek
2026-11-17  8:47 ` [RFC PATCH net-next 1/2] dt-bindings: net: add " Jonas Jelonek
2026-11-17  8:47 ` [RFC PATCH net-next 2/2] net: phy: " Jonas Jelonek
2026-11-22 21:24 ` Andrew Lunn [this message]

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=df35e0d4-9f1f-4882-bc6b-257add4e9fb1@lunn.ch \
    --to=andrew@lunn.ch \
    --cc=andrew+netdev@lunn.ch \
    --cc=ansuelsmth@gmail.com \
    --cc=conor+dt@kernel.org \
    --cc=corbet@lwn.net \
    --cc=davem@davemloft.net \
    --cc=devicetree@vger.kernel.org \
    --cc=edumazet@kernel.org \
    --cc=hkallweit1@gmail.com \
    --cc=horms@kernel.org \
    --cc=jonas@jonasjelonek.de \
    --cc=krzk+dt@kernel.org \
    --cc=kuba@kernel.org \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux@armlinux.org.uk \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=rdunlap@infradead.org \
    --cc=robh@kernel.org \
    --cc=skhan@linuxfoundation.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

all inboxes | Powered by JetHome®