Random reboot on the i.mx8mq

cancel
Showing results for 
Search instead for 
Did you mean: 

Random reboot on the i.mx8mq

462 Views
Contributor IV

We have a custom board that's running an i.MX8M processor.  The board is running a variation of P9.0.0_v2.0.0_GA modified to work with our custom hardware.  When running the board will reboot randomly.  The board will be up from anywhere between 2 hours to a couple of weeks before the reboot occurs.  The reboot appears to be caused by the software not being able to service the watchdog as when we disable the watchdog the board hangs instead of rebooting.  Therefore, it appears that something is causing the processor to lockup.

Has anyone seen this type of issue?  On our past boards this type of issue has been DRAM related.  However, as this is a off-the-shelf SoM we don't have much control of the DRAM configuration settings. Has anyone seen this type of issue in their custom i.MX8M boards or on the NXP provided developer boards. 

If anyone has seen this issue what steps did they take to fix it?

Thank you,

Sheldon 

Tags (2)
0 Kudos
6 Replies

54 Views
Contributor II

Thank you, for your response Sheldon. we will try those out. Your help is much appreciated!

- Omkar

0 Kudos

54 Views
NXP TechSupport
NXP TechSupport

Hi Sheldon

this may be caused by derating settings and one can try attached patch and solutions

suggested on

I.MX8M would automatically crashes under do not do anything 

Best regards
igor
-----------------------------------------------------------------------------------------------------------------------
Note: If this post answers your question, please click the Correct Answer button. Thank you!
-----------------------------------------------------------------------------------------------------------------------

0 Kudos

54 Views
Contributor IV

Hello Igor,

We applied nearly identical patches to the arm trusted firmware project last week.  The patches are attached for reference. 

The difference is that we applied one additional patch "atf-0002-add-save-restore-support-for-ddrphy-phymstr.patch." However, even after applying the patches we still get the random reboot issue.  

Do you know if applying the extra atf-0002-add-save-restore-support-for-ddrphy-phymstr.patch patch could be causing an issue?

Thank you,

Sheldon

0 Kudos

54 Views
NXP TechSupport
NXP TechSupport

Hi Sheldon

P9.0.0_v2.0.0_GA uses L4.14.98-2.0.0, for that version one can try apply

latest patches from Linux 4.14.98_2.3.1-patch Documentation

Best regards
igor

0 Kudos

54 Views
Contributor II

Hello sheldonrucker and  igorpadykov,

I am responding to this thread, because the problem you describe sounds eerily similar to the one we are seeing on one of our custom boards based around i.MX8M and LPDDR4. The device stays up for 2-10 days and then randomly reboots, apparently because the software/kernel hangs.

Did you ever get to the bottom of this issue? If yes, which fix ultimately helped you get over this? I would really appreciate if you can share some additional information regarding this issue, what options did you try, what worked/did not work etc.

Thank you,

Omkar

0 Kudos

54 Views
Contributor IV

Hello Omkar,

With the attached patches we were able to get the unit to stay up without rebooting as long as the unit wasn't thrashing around in memory.  For example, we run a test where we switch between streaming 1080p and UHD content stored on the device via the Android Gallery application every five seconds.  When running this test the unit will reboot randomly.  However if we just run a UHD video on loop via the Galley player the unit will stay up without rebooting.  We've run the same test on the NXP reference board and found that it does not reboot.

Our board uses the Variscite DART-MX8M SoM so one of the U-Boot patches only applies to their board  (u-boot-0001-lpddr4-timing-no-dqs2qd.patch).  So if you apply the patches as is that patch won't be applied to your board.  You'll need to make the edit by hand to the u-boot LPDDR timing file for your board.  That change is important as without it we were unable to get a unit that didn't reboot randomly.  

Note that we don't consider the attached patches a final fix.  They help but they don't lead to a fully stable system.  They help in that before the patches the unit would randomly reboot while sitting in an Android application and doing nothing.  Not it only seems to reboot when thrashing about in memory.


Either way I hope this helps!


Thank you,

Sheldon

0 Kudos