DP83640: IMX7D MII Device Tree

Hi James,

To assist with debugging, can you probe RX_CLK and CLKOUT and tell me their frequencies? Can you also read registers 0x0, 0x1, 0x10, 0x11 and tell me their values?

I see RGMII written in your code. Are you expecting to use the RGMII interface, because the DP83640 does not support this.

Thanks,

Lucas

All information in this correspondence and in any related correspondence is provided “as is” and “with all faults”, and it is subject to TI’s Important Notice (http://www.ti.com/corp/docs/legal/important-notice.shtml).

Hi Lucas,

Thanks very much for the reply.

Screen grab of RX_CLK,

Screen grab of CLKOUT

Register values are as follows,

0x0 -> 0x0000

0x1 -> 0x0000

0x10 -> 0x0000 

0x11 -> 0x0000

The RGMII references are the default pin labels for the IMX7d, the MAC is configured into MII mode and these pins are all RGMII/RMII and MII capable.

In UBOOT i see the same thing on the scope but get very diffrent register values,

0x0 -> 0x3100

0x1 -> 0x78ed

0x10 -> 0x0715

0x11 -> 0x0000

Thanks,

James

Hi James,

I don't immediately see any issues from you clock outputs or register values. Can you provide a schematic for review?

Also for your information, you can use this usb-2-mdio tool as another option to read registers: https://www.ti.com/tool/USB-2-MDIO

Thanks,

Lucas

All information in this correspondence and in any related correspondence is provided “as is” and “with all faults”, and it is subject to TI’s Important Notice (http://www.ti.com/corp/docs/legal/important-notice.shtml).

Hi Lucas,

Thanks for your help, my gut at the moment is that its an issue on the CPU side and that the PHY is behaving / configured correctley.

Here are the relevant sections of the schematic, 

Many Thanks,

James

Hi James,

I'm reviewing your schematic and will provide feedback by Wednesday.

Thanks,

Lucas

Hi James,

While I didn't see anything in the schematic that could directly cause the issue, I do have some notes:

• I noticed GPIO1 is tied to 3.3V. How do you plan to use this IO? You may want to consider using a PU resistor as opposed to directly tying to 3.3V.
• I recommend adding a 0.1uF decoupling cap on each supply.

Can you share the circuitry on the reset pin and your magnetics/RJ45?

You can also try verifying the RX paths. Can you connect the link partner on the cable side to send packets and try reading these packets on the RX lines?

Thanks,

Lucas

All information in this correspondence and in any related correspondence is provided “as is” and “with all faults”, and it is subject to TI’s Important Notice (http://www.ti.com/corp/docs/legal/important-notice.shtml).

Hi Lucas,

Thanks for the reply. the GPIO1 is labelled as 3V3 but is actually a 3V3 1PPS signal from a GNSS receiver which has a pull up (not shown on the snippet i sent through)

I've also attached a larger snippet of the schematic that shows the magnetics a bit clearer.

I've also attached a link partner and can see RX Data on the RXD0 line on a scope (screen grab attached) but not seeing any data in uboot.

Thanks

James

Hi James,

Can you try an external mac side loopback test? This would be the same setup as verifying the RX paths, but send the data from the RX lines back to the TX lines and check the RX error count. This can be found in register 0x15[7:0].

Thanks,

Lucas

All information in this correspondence and in any related correspondence is provided “as is” and “with all faults”, and it is subject to TI’s Important Notice (http://www.ti.com/corp/docs/legal/important-notice.shtml).