This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

DP83826E: PHY discovers at different addresse

Part Number: DP83826E

Working with a TI DP83826E PHY on embedded linux.  The PHY address is (should be set) by strap resistors.  I have probed the board and confirmed the settings on the resistors make sense.

When the board boots after a power up the PHYs appear at address 7.

But if the board is rebooted, they appear at address 1.

I have stopped in u-boot in both power up and reboot use cases, and used mii to proble the MDIO bus and confirmed that is what mii detects as where the PHYs are.

Each PHY is on a separate MDIO bus.  The MAC controller for the buses is a Altera cyclone5 FPGA.

I do not understand how the PHYs could be appearing at different addresses, just based on the whether linux kernel is started after a power cycle, or on a reboot.  I have gone through driver code extensively and do not find any attempt to program the address, nor do I see in the data sheet any such feature.

The only theory I have is something in the Cyclone5 FPGA alters how the devices appear, driven by some initialization that the kernel does that persists in the FPGA through reboot, but not power cycles.

Am I missing something about the DP83826E part that would make it behave this way?