CentOS

I'm running in to issues trying to install 346.46-0.x86_64 nvidia driver after updating to CentOS 7.4 .
I'm using same packages / process as was used on CentOS 7.3 ( blacklist nouveau ) and install Cuda 7.5 with 346.46 driver. I'm getting ` Error: Driver 'nvidia' is already registered, aborting..` errors

Anyone else running in to similar issues with 7.4 ( 3.10.0-693.2.2.el7.x86_64 kernel )

```
Sep 13 21:01:44 sars-image-gpure-1505335543 kernel: NVRM: request_mem_region failed for 16M @ 0xfd000000. This can#012NVRM: occur when a driver such as rivatv is loaded and claims#012NVRM: ownership of the device's registers.
Sep 13 21:01:44 sars-image-gpure-1505335543 kernel: nvidia: probe of 0000:00:04.0 failed with error -1
Sep 13 21:01:44 sars-image-gpure-1505335543 kernel: Error: Driver 'nvlink' is already registered, aborting...
Sep 13 21:01:44 sars-image-gpure-1505335543 kernel: Error: Driver 'nvidia' is already registered, aborting...
Sep 13 21:01:44 sars-image-gpure-1505335543 kernel: NVRM: DRM init failed
Sep 13 21:01:44 sars-image-gpure-1505335543 kernel: NVRM: NVIDIA init module failed!
Sep 13 21:01:04 sars-image-gpure-1505335543 kernel: nvidia: loading out-of-tree module taints kernel.
Sep 13 21:01:04 sars-image-gpure-1505335543 kernel: nvidia: module license 'NVIDIA' taints kernel.
Sep 13 21:01:04 sars-image-gpure-1505335543 kernel: Disabling lock debugging due to kernel taint
Sep 13 21:01:04 sars-image-gpure-1505335543 kernel: nvidia: module verification failed: signature and/or required key missing - tainting kernel
Sep 13 21:01:04 sars-image-gpure-1505335543 kernel: Error: Driver 'nvidia' is already registered, aborting...
Sep 13 21:01:04 sars-image-gpure-1505335543 kernel: NVRM: DRM init failed
Sep 13 21:01:04 sars-image-gpure-1505335543 kernel: NVRM: NVIDIA init module failed!
```

Hello
this is mostly guess work here but

Sep 13 21:01:04 sars-image-gpure-1505335543 kernel: nvidia: module verification failed: signature and/or required key missing - tainting kernel

would suggest the driver is not gpl licensed, if you're trying to install with yum you cold try or import the missing gpg key.
Hope this helps

Have you tried using the elrepo version of NVidia? It does all the work for you.

https://elrepo.org/tiki/kmod-nvidia

No need to manually blacklist, it takes care of all that.

I tried using the Nvidia RPM's and Their run files, both appear to be very broken in CentOS 7.4 . From my research this appears to a more widespread issue between nvidia and latest kernel:
https://devtalk.nvidia.com/default/topi ... -kernels-/