on CoreOS, /lib64/, /usr/lib64/ and co. all reside on a read-only filesystem. You might need to create a new directory elsewhere and its location listed in a file under /etc/ld.so.conf.d/

1. There is no /etc/ld.so.conf.d in 1298.5.0, only /etc/ld.so.conf is a symlink to /usr/lib/ld.so.conf which is mounted in readonly filesystem

There was a similar problem was with .bashrc ! the solution is to remove the link and replace it with the actual file)
I created a /opt/bin and /opt/lib64 but ldconfig complained about libEGL.so.1 which I made it s symlink

sudo ln -sfT libEGL.so.375.20 /opt/lib64/libEGL.so.1

2. What about the kernel modules?

    sudo mkdir -p /opt/lib64/modules/4.9.9-coreos-r1/kernel

But modprobe will not be able to find them, because it is non-standard path (insmod should work).
I wonder if it is simpler to re-mount the drive as rw and copy the kernel modules and all, rather than trying to install in /opt and work around issues.

creating device nodes under /dev/

Do you mean after modprobe or perhaps insmod?

From [here](https://groups.google.com/forum/#!topic/coreos-user/s9Sy_gBl94o):
> Kernel modules can be installed under `/opt/lib/modules/$(uname -r)/` as long as 
you use `--dirname /opt` when calling modprobe. 

So now I only need to use udev, otherwise will have to resort to [manually adding the nodes](https://gist.githubusercontent.com/tleyden/74f593a0beea300de08c/raw/95ed93c5751a989e58153db6f88c35515b7af120/nvidia_devices.sh) (,[same](http://askubuntu.com/a/748905/103961))

Im struggling to set up systemd-nspawn to work correctly.
Failed to register machine: Start job for unit machine-coreos_developer_container.bin.1409.5.0.scope failed with 'failed'
Failed to create directory /tmp/nspawn-root-Cs8uub/sys/fs/selinux: Read-only file system
Failed to create directory /tmp/nspawn-root-Cs8uub/sys/fs/selinux: Read-only file system

Im running the scripts in a privileged docker container. I have mapped /dev and /var/run into the container. Dockerfile as follows.
FROM ubuntu:xenial-20170510

RUN apt-get update && apt-get -y install sudo git curl bzip2 systemd-container grep rpm

CMD /bin/bash

How do I get through this and get the drivers to build?

If you specify a bad coreos channel-version combination, it creates the bz2 file anyway, causing the following situation (beta 1409.5.0 does not exist, but stable 1409.5.0 does):

core@ip ~/coreos-nvidia-master $ ./build.sh 384.59 beta 1409.5.0
Keeping container around after build: false
Additional flags: 
Downloading CoreOS beta developer image 1409.5.0
Decompressing
bunzip2: coreos_developer_container.bin.1409.5.0.bz2 is not a bzip2 file.
Caught error
Cleaning up
Cleaning up
core@ip ~/coreos-nvidia-master $ ./build.sh 384.59 stable 1409.5.0
Keeping container around after build: false
Additional flags: 
Downloading CoreOS stable developer image 1409.5.0
Decompressing
bunzip2: coreos_developer_container.bin.1409.5.0.bz2 is not a bzip2 file.
Caught error
Cleaning up
Cleaning up

I added this line to _export.sh

tar -C ${ARTIFACT_DIR}/tls -cvj $(basename -a ${ARTIFACT_DIR}/tls/*.so.*) > libraries-tls-${VERSION}.tar.bz2

It creates the right file, but also another empty file. Will work on a PR when I get time

There are 2 tls dlls and both should be packaged, but the one in tls directory has the newer ABI. The old ABI one is not passed to nvidia-docker. Some applications need the tls library to work.

$./build.sh 367.27 alpha 1097.0.0

I am following your instructions, but during the ./build.sh step I get the following message:

++ basename -a 'pkg/run_files/1097.0.0/NVIDIA-Linux-x86_64-367.27/kernel/*.ko'

• tar -C pkg/run_files/1097.0.0/NVIDIA-Linux-x86_64-367.27/kernel -cvj '*.ko'
  tar: *.ko: Cannot stat: No such file or directory
  tar: Exiting with failure status due to previous errors

Did I miss something?

I'v been embed 396.26 to coreos pxe image.
nvidia-smi has not supported setting the fan speed for a long time.
nvidia-settings need x11 server or fake x server .

Any suggestions install x11 || run nvidia-settings || 3rd tools on coreos ?