iMX8DX SCU reset when booting with M4 powered up

cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

iMX8DX SCU reset when booting with M4 powered up

Jump to solution
1,032 Views
Sselas
Contributor I

Hi !

I am working on an application for the cortex M4 on my board. I'm dealing with a Colibri iMX8DX v1.0 E mounted on a Toradex Aster v1.1 B on which I flashed a Linux embedded v7.1 Minimal.

I rerouted the debug console of the Cortex M4 to the UARTB (or UART0) because the 144 and 146 SODIMM pins that would be connected to the tightly coupled UART of the M4 were not connected on the Aster Card. Following this, I was able to run some examples from the MCUXpresso SDK and see their result on my debug console.

Now, when I try to boot Linux after launching the m4boot_0 command in U-Boot, the kernel almost finishes its initialization before reset with "Reset cause: SCFW fault reset".

Just so I avoid any weird interference with badly controlled pins, the firmware I am currently attempting to make work alongside Linux is the looped Hello World demo from the MCUXpresso SDK.

The boot log is attached to this message. I have no idea what causes this error, any tips would be so welcome.

Tags (5)
0 Kudos
Reply
1 Solution
1,018 Views
Sselas
Contributor I

Thanks for your response @Manuel_Salas , I just found the problem, actually. I was getting desperate because this behavior kept happening despite putting a simpler program in, attempting multiple device trees and overlays...
The problem appears to be the BSP I was using, I read somewhere a few minutes ago that version 7.1.0 had trouble dealing with the m4 and flashing the version 6.8 instead fixed it all.

Best regards,

Tom

View solution in original post

2 Replies
1,021 Views
Manuel_Salas
NXP TechSupport
NXP TechSupport

Hello @Sselas 

I hope you are doing very well.

 

Sometimes this happens when core M4 is trying to use a peripheral that Cortex A is also trying to use with Linux, Please be sure that you are not using the same peripheral in both sides, or change your Linux Device tree to avoid it.

Best regards,

Salas.

0 Kudos
Reply
1,019 Views
Sselas
Contributor I

Thanks for your response @Manuel_Salas , I just found the problem, actually. I was getting desperate because this behavior kept happening despite putting a simpler program in, attempting multiple device trees and overlays...
The problem appears to be the BSP I was using, I read somewhere a few minutes ago that version 7.1.0 had trouble dealing with the m4 and flashing the version 6.8 instead fixed it all.

Best regards,

Tom