Where is the nano's boot screen pic located?

We need to hide the implement platform info for confidential, so boot screen has to be erased. Would experts tell me how? Thanks a lot.

1 Like

Not sure what you mean by ‘boot screen has to be erased’. There’s no platform-implementation info in the BMP splash, if that’s what you are referring to - it’s just a NVIDIA logo BMP. You can disable it by removing the BMP partition from your flash*.xml file, and removing the bmp.blob from Linux_for_Tegra/bootloader.

If you are wanting to dispense with any boot spew to the UART, there are topics on how to do that that on this board. You can disable some of the bootloader spew, but not all w/o source. And you can also disable kernel spew with the ‘quiet’ cmd-line arg IIRC.

HTH,

Tom

Thanks for reply. I can’t find bmp.blob and no Linux_for_Tegra/bootloader path existed on jetson nano, whose system is flashed in SD card. I’m confused about you said. Maybe you mean TX2 or Xavier?

No, what twarren was talking about your host. Not Jetson Nano device.

I think you directly put the image to sdcard so you don’t know what is Linux_for_Tegra, right?
Please try to use sdkmanager and you will see.

I use sdkmanager to flash nano, I remove bmp.blob and every files that contain “bmp.blob”:
./JetPack_4.2.1_Linux_GA_P3448/Linux_for_Tegra/create-jetson-nano-sd-card-image.sh: ‘part_num=12;part_name=BMP;part_size=81920;part_file=bmp.blob’
./JetPack_4.2.1_Linux_GA_P3448/Linux_for_Tegra/bootloader/l4t_bup_gen.func: ‘part_name=BMP; part_type=bl,porg_sd; part_spec=${_common_spec}; part_file=bmp.blob’
./JetPack_4.2.1_Linux_GA_P3448/Linux_for_Tegra/bootloader/t210ref/cfg/flash_l4t_t214_emmc.xml: bmp.blob
./JetPack_4.2.1_Linux_GA_P3448/Linux_for_Tegra/bootloader/t210ref/cfg/flash_l4t_t210_spi_sd_p3448.xml: bmp.blob
./JetPack_4.2.1_Linux_GA_P3448/Linux_for_Tegra/bootloader/t210ref/cfg/gnu_linux_tegraboot_emmc_full.xml: bmp.blob
./JetPack_4.2.1_Linux_GA_P3448/Linux_for_Tegra/bootloader/t210ref/cfg/flash_l4t_t214_spi.xml: bmp.blob
./JetPack_4.2.1_Linux_GA_P3448/Linux_for_Tegra/bootloader/t210ref/cfg/flash_l4t_t210_emmc_p3448.xml: bmp.blob

But NVIDIA logo still shows on boot screen, I’m crazy…

You only need remove the BMP partition from flash_l4t_t210_spi_sd_p3448.xml for Nano. Remove the entire section, starting with 'partition name=“BMP” and ending with /partition:

    <partition name="BMP" id="21" type="data">
        <allocation_policy> sequential </allocation_policy>
        <filesystem_type> basic </filesystem_type>
        <size> 81920 </size>
        <file_system_attribute> 0 </file_system_attribute>
        <allocation_attribute> 0x8 </allocation_attribute>
        <percent_reserved> 0 </percent_reserved>
        <filename> bmp.blob </filename>
    </partition>

You can also delete the bmp.blob binary from your Linux_for_tegra/bootloader subdir, but that isn’t strictly necessary.

Then reflash your Nano (put it into recovery mode first) via:
sudo ./flash.sh jetson-nano-qspi-sd mmcblk0p1

Theoretically, you should also be able to reflash w/SDK manager, but I don’t use that on a daily basis, so I’d prefer if you use the above command. And note that the above is all done on your host system, not on the Nano itself.

Tom

Thanks for reply. Unfortunately, it doesn’t work, logo is still here. My steps are:
1, run sdkmanager, if I’ve put nano into recovery mode, flash will run automatically, scripts can’t be modified, so I disconnect nano, then flash is pending.
2, Modify flash_l4t_t210_spi_sd_p3448.xml as you said, bmp.blob is removed also, then put nano into recovery mode and perform sudo ./flash.sh jetson-nano-qspi-sd mmcblk0p1.
I’m nearby the edge of crazy, please help me…

I think I should rebuild cboot.bin, but I don’t know how. Would anyone help me?

Could any experts help me? Thanks in advanced.

2, Modify flash_l4t_t210_spi_sd_p3448.xml as you said, bmp.blob is removed also, then put nano into recovery mode and perform sudo ./flash.sh jetson-nano-qspi-sd mmcblk0p1.

This should work. Could you confirm there is no bmp.blob under bootloader and the partition does not exist?

1, run sdkmanager, if I’ve put nano into recovery mode, flash will run automatically, scripts can’t be modified, so I disconnect nano, then flash is pending.

Please do not use sdkmanager at this moment. Otherwise it may reinstall the missing file.

Thanks for reply. It can’t work. I can confirm that actions needed were done:

nvidia@nvidia-Precision-Tower-3620:~/nvidia/nvidia_sdk/JetPack_4.2.1_Linux_GA_P3448$ sudo find . -name "bmp.blob"
nvidia@nvidia-Precision-Tower-3620:~/nvidia/nvidia_sdk/JetPack_4.2.1_Linux_GA_P3448$ sudo find . -name "flash_l4t_t210_spi_sd_p3448.xml" | grep "bmp"
nvidia@nvidia-Precision-Tower-3620:~/nvidia/nvidia_sdk/JetPack_4.2.1_Linux_GA_P3448$ cd Linux_for_Tegra/
nvidia@nvidia-Precision-Tower-3620:~/nvidia/nvidia_sdk/JetPack_4.2.1_Linux_GA_P3448/Linux_for_Tegra$ sudo ./flash.sh jetson-nano-qspi-sd mmcblk0p1
###############################################################################
# L4T BSP Information:
# R32 , REVISION: 2.0
###############################################################################
# Target Board Information:
# Name: jetson-nano-qspi-sd, Board Family: t210ref, SoC: Tegra 210, 
# OpMode: production, Boot Authentication: , 
###############################################################################
./tegraflash.py --chip 0x21 --applet "/home/nvidia/nvidia/nvidia_sdk/JetPack_4.2.1_Linux_GA_P3448/Linux_for_Tegra/bootloader/nvtboot_recovery.bin" --skipuid --cmd "dump eeprom boardinfo cvm.bin" 
Welcome to Tegra Flash
version 1.0.0
Type ? or help for help and q or quit to exit
Use ! to execute system commands
 
[   0.0045 ] Generating RCM messages
[   0.0085 ] tegrarcm --listrcm rcm_list.xml --chip 0x21 0 --download rcm /home/nvidia/nvidia/nvidia_sdk/JetPack_4.2.1_Linux_GA_P3448/Linux_for_Tegra/bootloader/nvtboot_recovery.bin 0 0
[   0.0113 ] RCM 0 is saved as rcm_0.rcm
......

What confuse me most is when I flash nano at the first time, there is no directory “/JetPack_4.2.1_Linux_GA_P3448/Linux_for_Tegra”, I have to use sdkmanager, then such directory will be create, and flash.sh. So I can’t understand:

1.txt (19.8 KB)

I’ve tried with my nano and remove the bmp.blob. The nv logo is gone during boot up.

Yes, you need sdkmanager to install the Linux_for_Tegra for first time. Afterwards, you don’t need to always use it to help flash. You could directly use flash.sh to flash the board. sdkmanager is also using this script to flash.

To prevent sdkmanager reinstall bmp.blob back to your bootloader, I told you please not use sdkmanager during the test.

Thanks for reply. I do use flash.sh after confirming as command thread:

nvidia@nvidia-Precision-Tower-3620:~/nvidia/nvidia_sdk/JetPack_4.2.1_Linux_GA_P3448$ sudo find . -name "bmp.blob"
nvidia@nvidia-Precision-Tower-3620:~/nvidia/nvidia_sdk/JetPack_4.2.1_Linux_GA_P3448$ sudo find . -name "flash_l4t_t210_spi_sd_p3448.xml" | grep "bmp"
nvidia@nvidia-Precision-Tower-3620:~/nvidia/nvidia_sdk/JetPack_4.2.1_Linux_GA_P3448$ cd Linux_for_Tegra/
nvidia@nvidia-Precision-Tower-3620:~/nvidia/nvidia_sdk/JetPack_4.2.1_Linux_GA_P3448/Linux_for_Tegra$ sudo ./flash.sh jetson-nano-qspi-sd mmcblk0p1
......

Would you tell me what’s wrong in my command thread? I’ve attached flash record as 1.txt, partition BMP is not be flashed. But why green NVIDIA logo is still exist?

Could you share the uart log for us?

Excuse me. Where or how to get uart log?

FYI

Thanks. I will attach logs later for no uart cable. I have two nanos, both can’t work after flash. I also reinstall sdkmanager, but nvidia logo both exist.

Hi, uart logs please refer to uart.txt attached. Thanks.
uart.txt (93.6 KB)

I do find these logs:

[0005.220] bmp blob is not loaded and initialized
[0005.225] Failed to display boot-logo

But I swear the logo is still exist.