FEC MDIO WRITE error

[adarsh1997]

Hi
I am using imx6dl custom design board .Some Boards are rebooted sometimes without doing anything.

From syslog i found that after thes logs came board is rebooted :

fec 2188000.ethernet eth0: MDIO write timeout

Nov 29 07:31:26 imx6solosabresd kernel: [ 1873.439097] ------------[ cut here ]------------

Nov 29 07:31:29 imx6solosabresd kernel: [ 1875.214625] WARNING: CPU: 1 PID: 91 at drivers/net/phy/phy.c:716 phy_error+0x24/0x5c

Anyone have idea why mdio write timeout error  is happening and creating reboot issue.

In dmesg also sometimes this error came:

fec 2188000.ethernet eth0: MDIO write timeout

Nov 29 05:08:30 imx6solosabresd kernel: [ 261.512429] ------------[ cut here ]------------

Nov 29 05:08:33 imx6solosabresd kernel: [ 263.290457] WARNING: CPU: 1 PID: 91 at drivers/net/phy/phy.c:716 phy_error+0x24/0x5c

Nov 29 05:08:33 imx6solosabresd kernel: [ 266.260619] Modules linked in: imx_vdoa

Nov 29 05:08:37 imx6solosabresd kernel: [ 268.104484] CPU: 1 PID: 91 Comm: kworker/1:1 Not tainted 5.4.24-2.1.0 #1

Nov 29 05:08:37 imx6solosabresd kernel: [ 270.673943] Hardware name: Freescale i.MX6 Quad/DualLite (Device Tree)

Nov 29 05:08:37 imx6solosabresd kernel: [ 270.680496] Workqueue: events_power_efficient phy_state_machine

Nov 29 05:08:37 imx6solosabresd kernel: [ 270.680503] sysrq: This sysrq operation is disabled.

Nov 29 05:08:37 imx6solosabresd kernel: [ 270.680531] [<8010fe30>] (unwind_backtrace) from [<8010b670>] (show_stack+0x10/0x14)

Nov 29 05:08:37 imx6solosabresd kernel: [ 270.699213] [<8010b670>] (show_stack) from [<80ac974c>] (dump_stack+0x90/0xa4)

Nov 29 05:08:37 imx6solosabresd kernel: [ 270.706458] [<80ac974c>] (dump_stack) from [<8012f9f8>] (__warn+0xbc/0xd8)

Nov 29 05:08:37 imx6solosabresd kernel: [ 270.713346] [<8012f9f8>] (__warn) from [<8012fa78>] (warn_slowpath_fmt+0x64/0xc4)

Nov 29 05:08:37 imx6solosabresd kernel: [ 270.720842] [<8012fa78>] (warn_slowpath_fmt) from [<806652ec>] (phy_error+0x24/0x5c)

Nov 29 05:08:37 imx6solosabresd kernel: [ 270.728597] [<806652ec>] (phy_error) from [<806664e4>] (phy_state_machine+0xc8/0x14c)

Nov 29 05:08:37 imx6solosabresd kernel: [ 270.736443] [<806664e4>] (phy_state_machine) from [<801494fc>] (process_one_work+0x1d4/0x41c)

Nov 29 05:08:37 imx6solosabresd kernel: [ 270.744980] [<801494fc>] (process_one_work) from [<80149790>] (worker_thread+0x4c/0x528)

Nov 29 05:08:37 imx6solosabresd kernel: [ 270.753083] [<80149790>] (worker_thread) from [<8014ed8c>] (kthread+0x124/0x150)

Nov 29 05:08:37 imx6solosabresd kernel: [ 270.760491] [<8014ed8c>] (kthread) from [<801010e8>] (ret_from_fork+0x14/0x2c)

Nov 29 05:08:37 imx6solosabresd kernel: [ 270.767716] Exception stack(0xa8211fb0 to 0xa8211ff8)

Nov 29 05:08:37 imx6solosabresd kernel: [ 270.772774] 1fa0: 00000000 00000000 00000000 00000000

Nov 29 05:08:37 imx6solosabresd kernel: [ 270.780959] 1fc0: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000

Nov 29 05:08:37 imx6solosabresd kernel: [ 270.789140] 1fe0: 00000000 00000000 00000000 00000000 00000013 00000000

Nov 29 05:08:37 imx6solosabresd kernel: [ 270.859850] ---[ end trace 8c3180956b158c55 ]---

NOTE:we are using marvell mv88e6097 ethernet switch also.

[igorpadykov]

Hi Adarsh

>Some Boards are rebooted sometimes without doing anything.

one can consider memory errors and run ddr test

https://community.nxp.com/t5/i-MX-Processors-Knowledge-Base/i-MX-6-7-DDR-Stress-Test-Tool/ta-p/11082...

Best regards
igor

[adarsh1997]

Hi @igorpadykov 
We found that sometimes mdio and phy error came then only board rebooted:
fec 2188000.ethernet eth0: MDIO write timeout

Do you have any idea why MDIO write timeout is coming?

I am using kernel version 5.4