Arch Linux User Repository

The PKGBUILD supports customization via environment variables. Here's a list of all of them:
_makemenuconfig: Invokes make menuconfig (default configuration menu) before compilation starts
_makenconfig: Invokes make nconfig (nicer configuration menu) before compilation starts
_makexconfig: Invokes make xconfig (GUI configuration menu) before compilation starts
_use_current: Will use the configuration of the running kernel, if the running kernel is compiled with IKCONFIG_PROC
_copyfinalconfig: Copies the final kernel configuration into the repository root as kconfig-new before compilation starts
_localmodcfg: Only compiles modules found in modprobed-db's database (which decreases compilation time and kernel size)
_use_llvm_lto: Compiles the kernel with LLVM instead of GCC. Should work, if not write a comment
_subarch: Specifies the sub architecture to compile for (see the PKGBUILD file for a list of all subarches). Default is 40 (Generic-x86-64), which is \compatible with all amd64 processors. Must be a number
_debug: Force enables debug options when set to y, force disables debug options when set to n or uses the config defaults when unset

All of these variables just need to be set for them to apply, except for _subarch and _debug.

If you want a dead simple way to compile this package, you can try out my build tool. It includes a few no nonsense changes to the kernel configuration and allows you to configure the PKGBUILD in a simple manner and decrease build time. And if you don't want that, using makepkg as-is works fine too.

Kernel has been updated and is now maintained again.

Thank you metak for creating this package in the first place.

I am trying to compile Nvidis-340xx-DKMS and am having the following error. It compiles normal on CK, can you help me?
ERROR: modpost: "agp_backend_release" [/home/luiz/git-Download/nvidia-340xx/src/NVIDIA-Linux-x86_64-340.108-no-compat32/kernel/nvidia.ko] undefined!
ERROR: modpost: "agp_copy_info" [/home/luiz/git-Download/nvidia-340xx/src/NVIDIA-Linux-x86_64-340.108-no-compat32/kernel/nvidia.ko] undefined!
ERROR: modpost: "agp_unbind_memory" [/home/luiz/git-Download/nvidia-340xx/src/NVIDIA-Linux-x86_64-340.108-no-compat32/kernel/nvidia.ko] undefined!
ERROR: modpost: "agp_free_memory" [/home/luiz/git-Download/nvidia-340xx/src/NVIDIA-Linux-x86_64-340.108-no-compat32/kernel/nvidia.ko] undefined!
ERROR: modpost: "agp_backend_acquire" [/home/luiz/git-Download/nvidia-340xx/src/NVIDIA-Linux-x86_64-340.108-no-compat32/kernel/nvidia.ko] undefined!
ERROR: modpost: "agp_find_bridge" [/home/luiz/git-Download/nvidia-340xx/src/NVIDIA-Linux-x86_64-340.108-no-compat32/kernel/nvidia.ko] undefined!
make[2]: *** [scripts/Makefile.modpost:111: /home/luiz/git-Download/nvidia-340xx/src/NVIDIA-Linux-x86_64-340.108-no-compat32/kernel/Module.symvers]


@adsoncicilioti

glad i helped you!

GREAT @fidasx!!

That fixes! nvidia_uvm loaded and apps that uses CUDA working! Thanks!

@adsoncicilioti

download the patch and save it inside the linux-clear with name nvidia-5.9-uvm-fix.patch then open the PKGBUILD and add it under the other patches

like

Add nvidia-uvm.patch

echo "Applying nvidia-uvm.patch ..."
patch -Np1 -i ../nvidia-5.9-uvm-fix.patch

Thanks @fidasx! But, i'm so sorry.. I don't know how make this patch.

Like i says on commentary below, i've use the compiled package from @metak repo. When i try compiles with yay, using modprobed-db to made local .conf modules, my touchpad don't work.

Do you can give me an clarification about how do this patch?

@adsoncicilioti you need to patch the kernel for nvidia_uvm to work

use this if you want

https://paste.gg/p/anonymous/8ea12096fac14ebbb6bd4f5d2af8ec50

Hi @metak! nice work with Clear Kernel. I'm using the 5.9.6 from your repo, and all boot parms requireds by Clear, including module.sig_unenforce.

With standard and Zen Kernel, all nvidia modules are loaded correctly - nvidia, nvidia_drm, nvidia_modset, nvidia_uvm. On clear nvidia_uvm module not is loaded, and apps like blender, and other that uses CUDA cores, not recognize the nvidia (cuda device).

I've try with # modprobe nvidia_uvm command but the error is returned: modprobe: ERROR: could not insert 'nvidia_uvm': Unknown symbol in module, or unknown parameter (see dmesg).

Running dmesg i got:

  184.986139] nvidia_uvm: module uses symbols from proprietary module nvidia, inheriting taint.
[  184.986268] nvidia_uvm: Unknown symbol radix_tree_preloads (err -2)

I've don't understand why the standard and zen Kernels loads and Clear not. Any idea?

@krathalan I can change that with next update.

@metak some users (like myself) may have their ~/.config/modprobed.db file actually be a symlink to somewhere else that is shared with other systems. You should change the -f on line 215 from if [ -f $HOME/.config/modprobed.db ]; then to -e, so it reads if [ -e $HOME/.config/modprobed.db ]; then.

From man test, -f checks that: FILE exists and is a regular file (emphasis mine). -e only checks that: FILE exists. -f returns false if the file exists but is a symlink.

So once I've built the kernel with my customized config options from nconfig, I see the config file is copied into the main directory. How do I use this when I want to recompile again, or when the kernel updates and there are more config options (like make oldconfig would show you)?