OS corruption on the ReServer J44012

[JaocbSumosuke]

We are having issues with a reoccurring problem with the OS corrupting. After some time in use, the J4012 will stop booting up to Linux and fail at this point in the boot process:

bpmp: base binary md5 is fcc6f4f897ab64a32c3de470c48f0036 bpmp: combined binary md5 is a5fd653d3815177549cc44e087b1bd9e bpmp: firmware tag is a5fd653d3815177549cc-fcc6f4f897a Control register check for CBB_CENTRAL_CBB_FIREWALL_MCB_BLF failed, expected value: 80030000 != actual value: 00000000 address: 13a159d0 Write control check for CBB_CENTRAL_CBB_FIREWALL_MCB_BLF failed, expected value: 0f18000b != actual value: 00000000 address: 13a159c8 Control register check for CBB_CENTRAL_CBB_FIREWALL_MCB_BLF failed, expected value: 80030000 != actual value: 00000000 address: 13a159d0 Write control check for CBB_CENTRAL_CBB_FIREWALL_PWM7_BLF failed, expected value: 00100008 != actual value: 0010000a address:

It seems like the QSPI boot stack is getting corrupted somehow and is only recoverable via reflash.

This is happening on both freshly imaged J4012’s using reserver’s JetPack v. 5.1.1 Image from here: https://wiki.seeedstudio.com/reServer_Industrial_Getting_Started/#flash-jetpack

And we’re also seeing it on devices straight from the box (no re-image).

Any assistance would be appreciated.

[ackPeng]

What operations were performed during the process? We will use this information to reproduce and resolve the issue.

[JaocbSumosuke]

Hello, what kind of information were you looking for? AS in what is running on the device and peripherals are being used?

We would see this error after the device powers on after some kind of failure.
Thanks.

[ackPeng]

Yes, for example, what type of program was running? Were there any disk operations? Log information about any device malfunctions, etc., can all be shared.