Re: [PATCH 2/3] net: phy: bcm5481x: Implement MII-Lite mode

I wouldn't consider MII-Lite a separate mode (like eg. RMII), only a 
special case of MII with just those four signals not connected. As far I 
understand it, there is no need to configure the MAC for MII-Lite if the 
MAC input signals (RXER, CRS, COL) stay inactive. Because if missing COL 
(Collision), half duplex cannot be supported. The clock is limited to 25 
MHz, thus no gigabit. Besides 100Mbps, also 10Mbps, full duplex is 
supported with 2.5 MHz MII clock.
At least in the case of Broadcom PHYs, only the PHY must be explicitly 
told to switch to MII-Lite. No problem with the impossibility of half 
duplex, because all BroadR-Reach modes are full duplex only.
In turn, the RMII has less data lines (only two) and the MAC needs to be 
configured differently so it is clearly a another mode.

   Kamil

On 6/20/25 17:19, Andrew Lunn wrote:

On Fri, Jun 20, 2025 at 03:44:28PM +0200, Kamil Horák (2N) wrote:

quoted

The Broadcom bcm54810 and bcm54811 PHYs are capable to operate in
simplified MII mode, without TXER, RXER, CRS and COL signals as defined
for the MII. While the PHY can be strapped for MII mode, the selection
between MII and MII-Lite must be done by software.

Please could you say more about what mii-lite is. Rather than adding a
bool DT property, i'm asking myself should we add interface mode for
it?

Is it a mode of its own? MII normally means Fast Ethernet, 100Mbps. Is
that what MII-Lite supports? How does it differ from RMII? Should we
be calling this PHY_INTERFACE_MODE_LMII?

	Andrew