RGMII Enabling Problem

[mehmetdeniz]

Hello,

We need to use the RGMII interface on our custom carrier board (designed for AGX Xavier).

We tried the following solutions from other topics. But we have not fixed this.
-Jetson ORIN: RGMII can't send data - #6 by kangqi236
-Jetson AGX Orin RGMII to use - #6 by WayneWWW
-Does Orin support 88E1512 PHY chip? - #18 by WayneWWW

I’m attaching the .dts (exported from our DTB file), debug logs & configured pinmux files.
tegra234-p3701-0000-p3737-0000.dts (488.3 KB)

agx_orin_dmesg.txt (96.3 KB)

tegra234-mb1-bct-pinmux-p3701-0000.dtsi (62.5 KB)

Also, the following picture describes our ethernet implementation from the kernel source:

Module: AGX Orin Dev. Kit module
SW: JetPack-5.0 GA Early Access

These are our all flashing steps:
1- Copy the attached files in the specified paths:

• tegra234-p3701-0000-p3737-0000.dtb → Linux_for_Tegra/kernel/dtb/tegra234-p3701-0000-p3737-0000.dtb
• tegra234-mb1-bct-pinmux-p3701-0000.dtsi → Linux_for_Tegra/bootloader/t186ref/BCT/tegra234-mb1-bct-pinmux-p3701-0000.dtsi

2- Apply the binaries, apply the EEPROM configurations (we’ve not used any EEPROM) & change the ODMDATA for 1Gb Ethernet connection.

$ cd ~/nvidia/nvidia_sdk/JetPack_5.0_AG_ORIN/Linux_for_Tegra/
$ sudo ./apply_binaries.sh
$ sed -i "s/cvm_eeprom_read_size = <0x100>;/cvm_eeprom_read_size = <0x0>;/g" bootloader/tegra234-mb2-bct-common.dtsi
$ sed -i "s/cvb_eeprom_read_size = <0x100>;/cvb_eeprom_read_size = <0x0>;/g" bootloader/tegra234-mb2-bct-common.dtsi
$ sed -i "s/ODMDATA=\"gbe-uphy-config-22,hsstp-lane-map-3,nvhs-uphy-config-0,hsio-uphy-config-0,gbe0-enable-10g\";/ODMDATA=\"gbe-uphy-config-0,nvhs-uphy-config-0,hsstp-lane-map-3,hsio-uphy-config-16\";/g" p3701.conf.common

3- Create a user to skip the Ubuntu configuration wizard.

$ sudo tools/l4t_create_default_user.sh -u nvidia -p nvidia -a -n nvidia-agx-orin --accept-license 

4- Then, flash it

$ sudo ./flash.sh --no-flash jetson-agx-orin-devkit mmcblk0p1
$ lsusb | grep 0955:7023
$ sudo ./flash.sh -r jetson-agx-orin-devkit mmcblk0p1

Thanks

[WayneWWW]

Hi,

You only need to set cvb eeprom to 0. Do not touch anything for cvm eeprom.

[WayneWWW]

BTW, why do you remove interrupts from your mdio DT?

[mehmetdeniz]

Hi @WayneWWW
I guess the TEGRA_ETHERNETPHY_DT_VERSION is not less than DT_VERSION_2.

Shall I remove the if-check statement (also only set the cvb_read_size) & try it again?

Thanks

[WayneWWW]

Hi,

The default TEGRA_ETHERNETPHY_DT_VERSION is 2 so we intend the interrupt should be added if you check the adaptation guide.

[mehmetdeniz]

Hi @WayneWWW
Thank you. I checked the adaptation guide & removed the if-check side on the dts file from kernel source.
The following screenshot is from it.

The full device-tree file:
tegra234-p3701-0000-p3737-0000.dts (488.4 KB)

The debug logs:
agx_orin_dmesg2.txt (97.9 KB)

Now, the ethernet interface established. Here are the ifconfig & ethtool command results below:

nvidia@nvidia-agx-orin:~$ ifconfig 
eth0: flags=4099<UP,BROADCAST,MULTICAST>  mtu 1466
        ether 48:b0:2d:69:33:d4  txqueuelen 1000  (Ethernet)
        RX packets 132  bytes 10209 (10.2 KB)
        RX errors 0  dropped 71  overruns 0  frame 0
        TX packets 0  bytes 0 (0.0 B)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

lo: flags=73<UP,LOOPBACK,RUNNING>  mtu 65536
        inet 127.0.0.1  netmask 255.0.0.0
        inet6 ::1  prefixlen 128  scopeid 0x10<host>
        loop  txqueuelen 1000  (Local Loopback)
        RX packets 278  bytes 19194 (19.1 KB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 278  bytes 19194 (19.1 KB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

rndis0: flags=4099<UP,BROADCAST,MULTICAST>  mtu 1500
        ether a2:83:40:ce:9b:09  txqueuelen 1000  (Ethernet)
        RX packets 0  bytes 0 (0.0 B)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 0  bytes 0 (0.0 B)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

usb0: flags=4099<UP,BROADCAST,MULTICAST>  mtu 1500
        ether a2:83:40:ce:9b:0b  txqueuelen 1000  (Ethernet)
        RX packets 0  bytes 0 (0.0 B)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 0  bytes 0 (0.0 B)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

nvidia@nvidia-agx-orin:~$  sudo ethtool eth0
[sudo] password for nvidia: 
Settings for eth0:
	Supported ports: [ TP MII FIBRE ]
	Supported link modes:   10baseT/Half 10baseT/Full 
	                        100baseT/Half 100baseT/Full 
	                        1000baseT/Half 1000baseT/Full 
	Supported pause frame use: Symmetric Receive-only
	Supports auto-negotiation: Yes
	Supported FEC modes: Not reported
	Advertised link modes:  10baseT/Half 10baseT/Full 
	                        100baseT/Half 100baseT/Full 
	                        1000baseT/Half 1000baseT/Full 
	Advertised pause frame use: No
	Advertised auto-negotiation: Yes
	Advertised FEC modes: Not reported
	Speed: Unknown!
	Duplex: Unknown! (255)
	Port: Twisted Pair
	PHYAD: 0
	Transceiver: internal
	Auto-negotiation: on
	MDI-X: Unknown
	Supports Wake-on: g
	Wake-on: d
	Link detected: no
nvidia@nvidia-agx-orin:~$

Thanks

[WayneWWW]

What PHY is in use here and what is current situation?

Is it working now?

[WayneWWW]

If the TX does not work, please make sure you put every file that is generated from the pinmux spreadsheet to the BCT. Not only the pinmux dtsi.

There should be a gpio dtsi too.

[mehmetdeniz]

Hello @WayneWWW,

Thanks a lot. Now, the RGMII is working.

The updated GPIO dtsi file: tegra234-mb1-bct-gpio-p3701-0000.dtsi (3.9 KB)

The latest dmesg: agx_orin_dmesg3.txt (94.1 KB)

Kind Regards,
Mehmet