On Wed, Nov 18, 2026 at 12:00:20PM +0100, Vincent Jardin via B4 Relay wrote: > From: Vincent Jardin <[email protected]> > > IIC7 and IIC8, the i2c6 and i2c7 nodes, exist only on the SDHC2_DAT[7:4] > pads, and one RCW field, SDHC2_DAT74_PMUX, muxes all four of them: > - 0b010 gives both buses > - 0b001 gives GPIO2[18:15] > > Both nodes point at the same two pin states of that field. A board that > enables one of the two buses gets a working bus recovery from them, so > they stay in the SoC dtsi. > > A board that enables both cannot use them: the pinctrl core gives the > field to the first controller and the second one fails to probe, and > the bus recovery of one would switch the pads of the other one to GPIO > too. Say so above each node. > > Suggested-by: Vladimir Oltean <[email protected]> > Link: https://lore.kernel.org/all/[email protected]/ > Signed-off-by: Vincent Jardin <[email protected]> > ---
Reviewed-by: Vladimir Oltean <[email protected]>
