i.MX8QM MEK imx-scu scu: RPC send msg timeout

[stephanhoffmann]

Hello all,

I am working with a i.MX8QM MEK to evaluate multimedia performance.

After a while the system crashed. Unfortunately there is no console log available.

Since then the MEX can't boot anymore, neither IMX Linux LF5.15.71_2.2.0 nor my own buildroot based Linux image. Both of them worked before.

Error messages on the console log are:

[    4.162577] imx-scu scu: RPC send msg timeout
[    4.166639]  gpu0-pid0: failed to power up resource 144 ret -110
[    4.172642] imx-scu-clk: probe of gpu_core0_clk failed with error -110
[    7.234560] imx-scu scu: RPC send msg timeout
[    7.238593]  pwm0: failed to power off resource 191 ret -110
[   10.306558] imx-scu scu: RPC send msg timeout
[   10.310592]  pwm1: failed to power off resource 192 ret -110
and so on with many other resources

I assume something happened to the SCU. I attached the full console log from booting IMX Linux LF5.15.71_2.2.0

Does anybody have an idea what could have happened, how I can investigate further or how this could possibly be fixed?

Kind regards

Stephan

[stephanhoffmann]

The issue started when the board sat idle with nobody watching it. Maybe a GStreamer pipeline was running at that time, but I am not sure. Unfortunately there is no console log available. I tried to find the journal, but it seems to have been stored on a ramdisk and thus is lost, too.

Maybe it is a hardware defect, but I would like to be able to do further analysis.

[joanxie]

what operation or command do you made on the mek board? I mean do you playback a video all the time? for example you playback a video two days or many hours, then the board crashed, if yes, do you have any script? I need to reproduce this on my mek board, then I can tell you what's wrong with this board

[stephanhoffmann]

Hello joanxie,

thank you for your reply.

I cannot tell more than that the board stood idle during my break because I forgot to switch it off. When I came back it did not respond anymore and was not able to boot from any of the SD cards that worked before.

I do not think that you can reproduce this issue. Can you give any hint what I could do to analyze the problem further? Maybe I could measure some voltages or boot a special diagnostic image.

Kind Regards

Stephan

[joanxie]

only according to your logfile,  it seems scfw doesn't respond anymore, you can enable scfw console to check if scfw is alive. you also can measure power measurement to check if there is anything abnormal in power on/factory reset/software restart

[joanxie]

if your own board, I suggest that you can send your HW design to review, submit case from

https://www.nxp.com/support/support:SUPPORTHOME?tid=sbmenu

then nxp HW engineer can help you review your design to ensure your design is ok

[stephanhoffmann]

Good Morning Joanxie,

as I wrote in the title it is an original i.MX8 MEK.

[joanxie]

what operation did you do causing this issue? how could I reproduce this issue on imx8qm mek board with

LF5.15.71_2.2.0? pls share the test steps to double check

[stephanhoffmann]

The issue started when the board sat idle with nobody watching it. Maybe a GStreamer pipeline was running at that time, but I am not sure. Unfortunately there is no console log available. I tried to find the journal, but it seems to have been stored on a ramdisk and thus is lost, too.

Maybe it is a hardware defect, but I would like to be able to do further analysis.