Hello, I am a novice, I use orin Nano development board, I want to add other drivers in the kernel, and I need to modify the device tree, how do I need to operate, how do I need to compile and package after operation. Is there a detailed tutorial?
Refer the below link:
https://docs.nvidia.com/jetson/archives/r36.3/DeveloperGuide/SD/Kernel.html
Thank you for your help, I have carefully read this information, which only has the kernel compilation tutorial, not the relevant configuration of the kernel. CORSS_COMPILE=~/gcc/aarch64āglibcāstable-2022.08-1/bin/ aarch64-buildroot-linux-gnu-menuconfig Go to the kernel configuration screen, check imx412, After compiling with make ARM=arm64 CORSS_COMPILE=~/gcc/aarch64āglibcāstable-2022.08-1/bin/aarch64-buildroot-linux-gnu ā j8, After compiling I could not find the imx412.ko file produced by imx412 in the kernel. What else do I need to do?
I am compiling on PC Ubuntu20.04
1 Like
After compilation, were you able to flash it on the Jetson unit?
Refer below link for flashing:
https://docs.nvidia.com/jetson/archives/r36.3/DeveloperGuide/IN/QuickStart.html
I have not yet brushed the machine, but after compiling the kernel, I can not find the driver file after configuring the kernel in the compilation file. Excuse me, am I operating the kernel correctly? Hereās how I manipulate the kernel configuration:
$ tar xf public_sources.tbz ~/nvidia/ ~/nvidia/
$ cd ~/nvidia/Linux_for_Tegra/source
$ tar xf kernel_src.tbz2
$ tar xf kernel_oot_modules_src.tbz2
$ tar xf nvidia_kernel_display_driver_source.tbz2
cd ~/nvidia//Linux_for_Tegra/source
$ export CROSS_COMPILE=<toolchain-path>/bin/aarch64-buildroot-linux-gnu-
$ make -C kernel
$cd ~/nvidia/Linux_for_Tegra/source/kernel/kernel-jammy-src
$make ARCH=arm64 CROSS_COMPILE=~/nvidia/gcc/bin/aarch64-buildroot-linux-gnu- menuconfig
$make ARCH=arm64 CROSS_COMPILE=~/nvidia/gcc/bin/aarch64-buildroot-linux-gnu- -j8
Which version of linux for tegra you are flashing?
Please provide the link you are following to use the steps mentioned by you.
development environment(PC)ļ¼Ubuntu 20.04
Version in use: Jetson Linux 36.3
From where did you get these steps?
please provide the linkā¦
Maybe the steps are wrong, this is how I compile and prune kernels for other development boards, just replacing the compiler. Could you please provide the steps of tailoring orin Nano core?
I had already provided the below link in earlier post:
You need to follow these steps for Linux for Tegra 36.3
https://docs.nvidia.com/jetson/archives/r36.3/DeveloperGuide/SD/Kernel.html
Also,
**You can modify your imx412 to āyā in the tegra_defconfig file from the below link **
** kernel-5.10/arch/arm64/configs/tegra_defconfig**
to enable the configs and then run nvbuild.sh
Refer the below for more details:
Please make ARCH=arn64 menuconfig where to execute, I this $/Linux_for_Tegra/source error
zynq@ubuntu:~/nvidia/Linux_for_Tegra/source$ make ARCH=arn64 menuconfig
make: *** No rule to make target 'menuconfig'. Stop.
zynq@ubuntu:~/nvidia/Linux_for_Tegra/source$ cd ./kernel/
zynq@ubuntu:~/nvidia/Linux_for_Tegra/source/kernel$ ls
kernel-jammy-src Makefile
zynq@ubuntu:~/nvidia/Linux_for_Tegra/source/kernel$ make ARCH=arn64 menuconfig
make: *** No rule to make target 'menuconfig'. Stop.
zynq@ubuntu:~/nvidia/Linux_for_Tegra/source/kernel$ cd kernel-jammy-src/
zynq@ubuntu:~/nvidia/Linux_for_Tegra/source/kernel/kernel-jammy-src$ make ARCH=arn64 menuconfig
Makefile:623: arch/arn64/Makefile: No such file or directory
make: *** No rule to make target 'arch/arn64/Makefile'. Stop.
zynq@ubuntu:~/nvidia/Linux_for_Tegra/source/kernel/kernel-jammy-src$
Ask make ARCH=arn64 menuconfig where to execute, no matter where to execute will give an errorć
I see some spelling mistake here.
It should be arm64 instead of arn64
zynq@ubuntu:~/nvidia/Linux_for_Tegra/source$ cd ./kernel/kernel-jammy-src/
zynq@ubuntu:~/nvidia/Linux_for_Tegra/source/kernel/kernel-jammy-src$ make ARCH=arm64 menuconfig
zynq@ubuntu:~/nvidia/Linux_for_Tegra/source/kernel/kernel-jammy-src$ cd ..
zynq@ubuntu:~/nvidia/Linux_for_Tegra/source/kernel$ cd ..
zynq@ubuntu:~/nvidia/Linux_for_Tegra/source$ ./nvbuild.sh -o ~/nvidia/out
Then I searched the ~/nvidia/out file and couldnāt find the imx412.ko file.
For this, I think you need to manually go and add entry for inside tegra_defconfig file( in the path kernel-5.10/arch/arm64/configs/tegra_defconfig )
example:
CONFIG_VIDEO_IMX412=y
, before compiling using nvbuild.sh
Refer this link:
Thanks! imx412.ko was successfully compiled by modifying tegra_defconfig. But why is it not possible to tailor the kernel by āmake ARCH=arm64 menuconfigā?
1 Like
You should ask NVIdia for that !
Please mark my post as solution so it will be helpful for others.
What if I want to do some kernel tailoring by making ARCH=arm64 menuconfig?
@KevinFFF could you please answer this query
There are 4 steps to update kernel config.
Step 1. Generate base kernel config (.config)
$ cd <Linux_for_Tegra>/sources/kernel/kernel-5.10
$ make ARCH=arm64 defconfig
Step 2. Configure the kernel config in menu and update .config
$ make ARCH=arm64 menuconfig
Step 3. Generate defconfig
$ make ARCH=arm64 savedefconfig
Step 4. Update defconfig
$ cp defconfig arch/arm64/configs/defconfig
You can also add the specific kernel config in arch/arm64/configs/defconfig directly.
After boot up, please run the following command to check if it is applied correctly.
$ zcat /proc/config.gz | grep CONFIG_VIDEO_IMX412
2 Likes