Again, likely a uboot issue,,,, there are 2 'strains' of uboot,,,, bsp and mainline
mainline has no trust (ATF, andriod trusted firmware), the uboot is u-boot.itb (which contains trust)
vrs bsp,, uboot.img,,, this needs trust.img
Maybe debian does it differently
Without a README you really have no idea which version is which
The SPI forms are a bit different, the idbloader has some 'extra' blank sectors inserted into it
Also starts at zero,,, these are ONLY for writing to SPI
---
I would check if the dtb is referenced in extlinux.conf and that it exists in the right place
(because of the 2 partition, there should be NO leading /boot in path)
So, I would try a different uboot package, for reasons I prefer an updated mrfixit uboot, a bsp uboot (for s3 sleep)
Whatever you choose, copy mbr, 1st 16M (somewhere, usb stick?),,,, blank (/dev/zero) 1st 16M,,,, restore mbr
Then write choosen uboot package
Do check that it is mbr and not GPT partitioned, I don't know what debian does (GPT=34 sectors)
mainline has no trust (ATF, andriod trusted firmware), the uboot is u-boot.itb (which contains trust)
vrs bsp,, uboot.img,,, this needs trust.img
Maybe debian does it differently
Without a README you really have no idea which version is which
The SPI forms are a bit different, the idbloader has some 'extra' blank sectors inserted into it
Also starts at zero,,, these are ONLY for writing to SPI
---
I would check if the dtb is referenced in extlinux.conf and that it exists in the right place
(because of the 2 partition, there should be NO leading /boot in path)
So, I would try a different uboot package, for reasons I prefer an updated mrfixit uboot, a bsp uboot (for s3 sleep)
Whatever you choose, copy mbr, 1st 16M (somewhere, usb stick?),,,, blank (/dev/zero) 1st 16M,,,, restore mbr
Then write choosen uboot package
Do check that it is mbr and not GPT partitioned, I don't know what debian does (GPT=34 sectors)