The latest NVIDIA drivers failed with Kernel 6.19-rc1, as detailed in an earlier post, and after some more research I was able to work out a hack to fix this, and in the process discovered a bit of a surprise..

I managed to get the NVIDIA-proprietary code for 580.119.02 to compile successfully, but then:
…………………………
ERROR: modpost: GPL-incompatible module nvidia.ko uses GPL-only symbol '__vma_start_write'
ERROR: modpost: GPL-incompatible module nvidia-drm.ko uses GPL-only symbol '__vma_start_write'
make[4]: *** [/usr/src/linux-6.19-rc1/scripts/Makefile.modpost:147: Module.symvers] Error 1
make[3]: *** [/usr/src/linux-6.19-rc1/Makefile:2004: modpost] Error 2
make[2]: *** [/usr/src/linux-6.19-rc1/Makefile:248: __sub-make] Error 2
make[2]: Leaving directory '/home/rgadsdon/NVIDIA-Linux-x86_64-580.119.02/kernel'
make[1]: *** [Makefile:248: __sub-make] Error 2
make[1]: Leaving directory '/usr/src/linux-6.19-rc1'
make: *** [Makefile:138: modules] Error 2

I was testing this on a laptop with a GTX980M, which is still supported with this driver..

So, I decided to try hacking the MIT/GPL code in the kernel-open directory..
After getting this compile successfully, I got a ‘no device found‘ error after # modprobe nvidia, and assumed I had got something wrong.. After more testing, I found that the MIT/GPL (‘kernel-open‘) code would not work at all with this GPU, although the NVIDIA (‘kernel‘) code for the same driver version did! It seems the ‘kernel-open‘ code has already dropped support for ‘older’ GTX GPUs, and only worked with newer RTX models..
……………….
[ 1029.103879] nvidia-nvlink: Nvlink Core is being initialized, major device number 237
[ 1029.103887] NVRM: The NVIDIA GPU 0000:01:00.0 (PCI ID: 10de:13d7)
NVRM: installed in this system is not supported by open
NVRM: nvidia.ko because it does not include the required GPU
NVRM: System Processor (GSP).
NVRM: Please see the 'Open Linux Kernel Modules' and 'GSP
NVRM: Firmware' sections in the driver README, available on
NVRM: the Linux graphics driver download page at
NVRM: www.nvidia.com.
[ 1029.107031] nvidia 0000:01:00.0: probe with driver nvidia failed with error -1
[ 1029.107118] NVRM: The NVIDIA probe routine failed for 1 device(s).
[ 1029.107120] NVRM: None of the NVIDIA devices were initialized.
[ 1029.107666] nvidia-nvlink: Unregistered Nvlink Core, major device number 237

Switching to a system with an RTX 3050 GPU, I was able to get the hacked driver to load and run successfully..
……………………..
CC [M] nvidia.mod.o
CC [M] .module-common.o
LD [M] nvidia.ko
CC [M] nvidia-uvm.mod.o
LD [M] nvidia-uvm.ko
CC [M] nvidia-modeset.mod.o
LD [M] nvidia-modeset.ko
CC [M] nvidia-drm.mod.o
LD [M] nvidia-drm.ko
CC [M] nvidia-peermem.mod.o
LD [M] nvidia-peermem.ko
make[2]: Leaving directory '/home/rgadsdon/NVIDIA-Linux-x86_64-580.119.02/kernel-open'
make[1]: Leaving directory '/usr/src/linux-6.19-rc1'

So far, this seems to be working correctly (tested with KDE/Plasma, 4K video, and VMware with Win10 Guest).       I will publish the ‘hack’ details – and appropriate warnings – in a separate article..

Robert Gadsdon. December 18th 2025..