Hello Pavel, thanks a lot for your very helpful list of all the things to (re)consider!
Your hint regarding "position independent relative jumps" helped me to find the flaw in my previous assumptions. I misunderstood the nature of the device created by the function `rp2040_flash_mtd_initialize` [1]. I assumed, it would be an MTD device encompassing the full size of the flash. But in fact it was just a small piece (8 kB) located at the end of the allocated flash. Thus, my understanding of the absolute positions of my mtd-partitions "within" this parent device, was completely off-track. Thanks a lot for pushing me toward the solution for this time-consuming puzzle in my head! With this weirdness out of the way, I will prepare a merge request for rp2040 implementation of `board_boot_image`. Cheers, Lars [1] https://github.com/apache/nuttx/blob/master/arch/arm/src/rp2040/rp2040_flash_mtd.c#L505
