Part Number: DP83822I
Hi team,
We have serval units that lost network after hours stability test. we use RMII mode
The findings so far:
1. when issue occurred , the TX has packet increment but the RX keeps no change after ping operation
root@**:~# ip -s link show eth0
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq state UP mode DEFAULT group default qlen 1000
link/ether e8:27:25:13:32:4b brd ff:ff:ff:ff:ff:ff
RX: bytes packets errors dropped missed mcast
2007598 18219 0 0 0 0
TX: bytes packets errors dropped carrier collsns
12060576 36165 0 0 0 0
2. we tried the MII loopback according to datasheet and https://docs.ampnuts.ru/ti.com.datasheet/DP83822I/Application_note_SNLA266.PDF
001F 8000 //software reset (clears register)
0000 6100 //programs DUT to 100BASE-TX mode and enables MII Loopback
001F 4000 //digital reset (doesn’t clear register)
Disable AUTO-MIDX(0x0019) : bit 15 set to 0
ethtool --change eth0 speed 100 duplex full autoneg off
~ # ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <NO-CARRIER,BROADCAST,MULTICAST,PROMISC,UP> mtu 1500 qdisc fq state DOWN group default qlen 1000
link/ether e8:27:25:1b:79:98 brd ff:ff:ff:ff:ff:ff
~ #
after above steps eth0 cannot up , but on normal unit it can up and have a address, and i also can see both RX/TX increments with ping operation
3.Another finding is ip link eth0 down && ip link eth0 up can recover the network communication
This is the 0112.log when network down including 0x0 to 0x1F, ethtool eth0 and ethtool -S eth0
Could you help check if i am missing any vital things on MII loop or anything wrong?
