Some of these may help, but isn’t in any particular order…
About module naming:
https://forums.developer.nvidia.com/t/jetson-nano-board-setup/163668/6
Beware that the command “uname -r” is used by a kernel to find its modules, and that the result of this is from a combination of the base kernel version plus the CONFIG_LOCALVERSION (see the above URL). If you start with the original kernel version and configuration, then set CONFIG_LOCALVERSION to match the old kernel’s “uname -r”, then the original modules and module location will keep working. The location is:
/lib/modules/$(uname -r)/kernel
…and this is often forgotten and so a new kernel fails to find modules.
A partial extension of the above URL, but aimed more at configuring:
https://forums.developer.nvidia.com/t/attempts-to-set-up-ecryptfs-and-fscrypt-failed-and-failed-and-failed/119083/2
A URL regarding kernel build, but with more device tree and kernel argument focus:
https://forums.developer.nvidia.com/t/about-kernel/77995/18
Just a recipe when compiling natively from the Jetson (there would be more environment setup for cross compile):
# --- Setting Up: -------------------------------------------------------
# DO NOT BUILD AS ROOT/SUDO!!! You might need to install source code as root/sudo.
mkdir -p "${HOME}/build/kernel"
mkdir -p "${HOME}/build/modules"
mkdir -p "${HOME}/build/firmware"
export TOP="/usr/src/sources/kernel/kernel-4.9"
export TEGRA_KERNEL_OUT="${HOME}/build/kernel"
export TEGRA_MODULES_OUT="${HOME}/build/modules"
export TEGRA_FIRMWARE_OUT="${HOME}/build/firmware"
export TEGRA_BUILD="${HOME}/build"
# --- Notes: ------------------------------------------------------------
# It is assumed kernel source is at "/usr/src/sources/kernel/kernel-4.9".
# Check if you have 6 CPU cores, e.g., via "htop".
# If you are missing cores, then experiment with "sudo nvpmodel -m 0, -m 1, and -m 2".
# Perhaps use "htop" to see core counts.
# Using "-j 6" in hints below because of assumption of 6 cores.
# -----------------------------------------------------------------------
# Compile commands start in $TOP, thus:
cd $TOP
# Do not forget to provide a starting configuration. Probably copy of "/proc/config.gz",
# to $TEGRA_KERNEL_OUT, but also perhaps via:
make O=$TEGRA_KERNEL_OUT nconfig
# If building the kernel Image:
make -j 6 O=$TEGRA_KERNEL_OUT Image
# If you did not build Image, but are building modules:
make -j 6 O=$TEGRA_KERNEL_OUT modules_prepare
# To build modules:
make -j 6 O=$TEGRA_KERNEL_OUT modules
# To build device tree content:
make -j 6 O=$TEGRA_KERNEL_OUT dtbs
# To put modules in "$TEGRA_MODULES_OUT":
make -j 6 O=$TEGRA_KERNEL_OUT INSTALL_MOD_PATH=$TEGRA_MODULES_OUT
# To put firmware and device trees in "$TEGRA_FIRMWARE_OUT":
make -j 6 O=$TEGRA_KERNEL_OUT INSTALL_FW_PATH=$TEGRA_FIRMWARE_OUT
In the above, when run on a Jetson, it assumes source exists at “TOP”, and in this case “/usr/src/sources/kernel/kernel-4.9” was chosen. You could change this if source is somewhere else. Also, several targets which you probably won’t use are shown, e.g., a lot of people won’t build firmware, and if reusing the old modules, then probably won’t rebuild all modules. More typical is to build a single module.
Official documents are for cross compile and are a variation on the above “recipe”.
Advice: If you start with the original kernel config, and only add config, then you probably don’t need to rebuild modules. However, there are times when you remove features which were non-module format, and this could be a “more invasive” change (e.g., adding or removing the ability to use swap is invasive), and you will want to rebuild everything…kernel and modules, but not firmware.
Advice: You will see flashing as a method of installing a new kernel. This is usually unnecessary (assuming you have not burned security fuses), and the kernel in “/boot” (the “Image” file) only needs to change if you’ve changed an integrated feature. Adding or removing a module allows you to not bother changing the “Image” file when the “uname -r” remains constant. You can leave the original “Image” file in place, add a new one with a modified name (e.g., “Image-custom1”), and add a new entry to “extlinux.conf” and pick the new kernel via serial console…if it were to fail, you simply use the original entry; if it succeeds, then you can modify “extlinux.conf” to make this the default while still leaving the “safe original” in place for picking with serial console should the need arise.