Hi, thank you for your answer.
Maybe I was not clear, pretty new to this, so sorry.
We need to make a lot of the Jetson Orin NX ready to run.
So we want use an image for OS drive, and flash onyl QSPI (boot + initrd + kernel, right?).
Problem is:
If we run the flash command as explained in the thread, everything is fine.
./flash.sh -c bootloader/t186ref/cfg/flash_t234_qspi.xml jetson-orin-nano-devkit mmcblk0p1
if we run the same command a second time on a system that was already flashed (changes, bugs, …), it does not work any more:
[ 21.4101 ] Start flashing
[ 21.4137 ] tegradevflash_v2 --pt flash.xml.bin --create
[ 21.4170 ] Bootloader version 01.00.0000
[ 21.4500 ] Erasing spi: 0 ......... [Done]
[ 22.4567 ] Writing partition secondary_gpt with gpt_secondary_3_0.bin [ 16896 bytes ]
[ 22.4577 ] [................................................] 100%
[ 22.4607 ] 000000004d4d2c01: E> NV3P_SERVER: Failed to initialize partition table from GPT.
[ 22.6810 ]
[ 22.6810 ]
If we usel4t_ initrd_flash directly, it does not work, too, but for othe reasons, I guess:
./l4t_initrd_flash.sh -c ../../bootloader/t186ref/cfg/flash_t234_qspi.xml jetson-orin-nano-devkit mmcblk0p1
…
******************************************
* *
* Step 2: Generate rcm boot commandline *
* *
******************************************
ROOTFS_AB= ROOTFS_ENC= /home/localadmin/nvidia/nvidia_sdk/JetPack_5.1.2_Linux_JETSON_ORIN_NX_TARGETS/Linux_for_Tegra/flash.sh --no-flash --rcm-boot jetson-orin-nano-devkit mmcblk0p1
Error: Invalid target board - jetson-orin-nano-devkit.
If we use the SDKmanager GUI and run a “full flash” including os drive setup, it works…
So I may just oversea something.
Thank you for all the help
BR