On 9/24/21 2:02 PM, Arınç ÜNAL wrote: >> 0x200 is not the address of the Realtek switch on the MDIO bus, 0x200 is >> the offset with mdio mux that needs to be toggled (bit 9). You still >> need to provide the Ethernet switch's address on the MDIO bus which >> appears to be 0. > > Oh, we flip the 9th bit. 2 to the power of 9 = 0x200. Got it! > > I tried 0 and 29 as the PHY ID. I'd assume the DSA realtek-smi driver > would start probing the switch, however, nothing happens. Full log in > attachments. > > [ 2.026772] bcm_iproc 18029200.spi: using bspi-mspi mode > [ 2.033467] libphy: Fixed MDIO Bus: probed > [ 2.038123] libphy: iProc MDIO bus: probed > [ 2.042331] iproc-mdio 18003000.mdio: Broadcom iProc MDIO bus registered > [ 2.049823] libphy: mdio_mux: probed > [ 2.054206] libphy: mdio_mux: probed > [ 2.058713] b53-srab-switch 18007000.ethernet-switch: found switch: > BCM53012, rev 0 > [ 2.066671] bgmac_bcma: Broadcom 47xx GBit MAC driver loaded > > Quoting Documentation/devicetree/bindings/net/dsa/realtek-smi.txt for > further reference. > >> Realtek SMI-based Switches >> ========================== >> >> The SMI "Simple Management Interface" is a two-wire protocol using >> bit-banged GPIO that while it reuses the MDIO lines MCK and MDIO does >> not use the MDIO protocol. This binding defines how to specify the >> SMI-based Realtek devices. Ah this is the key here, using the MDIO controller won't work sorry about misleading you. I suppose you will have to go back to the previous Device Tree representation you had, but change the dsa,member property and then you should be in business baring additional bugs/features. -- Florian
