[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <bb73ca0f-145c-4900-b077-e4d654dc1a0c@lunn.ch>
Date: Sat, 10 Aug 2024 17:46:14 +0200
From: Andrew Lunn <andrew@...n.ch>
To: Oleksij Rempel <o.rempel@...gutronix.de>
Cc: Heiner Kallweit <hkallweit1@...il.com>,
Russell King <linux@...linux.org.uk>,
"David S. Miller" <davem@...emloft.net>,
Eric Dumazet <edumazet@...gle.com>,
Jakub Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>,
kernel@...gutronix.de, linux-kernel@...r.kernel.org,
netdev@...r.kernel.org
Subject: Re: [PATCH net-next v2 2/3] phy: Add Open Alliance helpers for the
PHY framework
> > > 6.4 Polarity Detection and Correction (POL)
> > > 6.4.1 Polarity Detection (DET)
> > > 6.4.2 Polarity Correction (COR)
> >
> > Could these be mapped to ETH_TP_MDI* ?
>
> Yes, but it will look confusing in the user space. To make better
> representation in ethtool we will probably need a new port type. For
> example instead of PORT_TP it will be PORT_STP (single twiste pair) or
> PORT_SPE (single pair ethernet). What do you think?
Thinking about this some more....
The Marvell PHYs can indicate per pair if the polarity is wrong and
has been corrected, for 10BaseT and 1000BaseT. The datasheet indicates
that for 100BaseTX polarity does not matter.
So i agree, MDI crossover is the wrong way to indicate this. We should
add polarity indication, for up to 4 pairs.
Andrew
Powered by blists - more mailing lists