Package Details: nvidia-full-beta 396.24-1

Git Clone URL: (read-only)
Package Base: nvidia-full-beta
Description: Full NVIDIA driver package for Arch's official 'linux' package (driver, utilities, and libraries) (beta version)
Upstream URL:
Keywords: lib32-opencl-driver lib32-vulkan-driver libglvnd nvidia-settings opencl-driver vulkan-driver
Licenses: custom:NVIDIA
Conflicts: nvidia, nvidia-173xx, nvidia-96xx
Provides: nvidia=396.24
Submitter: Det
Maintainer: Det
Last Packager: Det
Votes: 18
Popularity: 1.269096
First Submitted: 2013-06-27 12:48
Last Updated: 2018-05-02 20:58

Required by (16)

Sources (3)

Latest Comments

Det commented on 2018-04-29 15:56

Different operating system.

The Linux kernel breaks compatibility on the kernel level with each new kernel release, because otherwise you'd be bottlenecked on what you could change in the code.

You rebuild drivers yourself on 4.16 -> 4.17, or update the prebuilt ones provided by your distribution. commented on 2018-04-29 15:55

@wyre: It needs to be rebuild everytime I upgrade the kernel WTTFF.. This is bad. Any fix?

wyre commented on 2018-04-20 13:39

@Det, I'm ArchLinux user, I'm just learning about AUR usage ;-)

Det commented on 2018-04-20 13:14

Yeah. That is also standard AUR knowledge, which I won't be separately specifying for distributions not meant to use this repository.

wyre commented on 2018-04-20 12:40

Also I think you should specify this package need to be build every time kernel is upgraded.

@Det, may this package solve this question?

Det commented on 2018-04-19 17:42

How can you people not see which packages are getting installed, and which aren't? You know what /usr/lib32 is for, and can check the installed files of in the page, right?

You are all Arch users, not Manjaro/Antergos?

E: Steam also has lib32-libgl as dependency, so I don't understand, how can you not succeed in checking which of those then is installed?:

OlliC commented on 2018-04-19 17:05

Not building lib32 by default is probably bad. Needed a while to figure out why Steam does not start.

wyre commented on 2018-04-19 15:38

I think you should pin to top the @l0k1 comment. Thank you for your info about Steam working with this driver, mate.

Det commented on 2018-03-31 10:52

nvidia-full-beta 390.48-2: fix /usr/lib/libGLESv2* symlinks

Det commented on 2018-03-14 10:35

That's what the flag and the comment is there for, yes.

You only need to do it the first time, but I might flip the default.

l0k1 commented on 2018-03-14 10:29

This package has by default the lib32 parts not actually building, I had to manually flip the bit in the PKGBUILD to get it to build and install them. Without those lib32 libs Steam does not work.

Det commented on 2017-10-03 18:12

Real boring these things. They just need improvement.


hcorion commented on 2017-06-29 20:35

Ohhh, but you didn't do a rebuild. Silly me.

hcorion commented on 2017-06-29 20:24

Oh, I see you did, Hmmm.

If I run vulkanCapsViewer it outputs. Could not create Vulkan instance, Error: ERROR_EXTENSION_NOT_PRESENT.

I've got a NVIDIA GTX 1070

hcorion commented on 2017-06-29 20:19

Except Vulkan doesn't work, can you apply the same fix that you did to the non-full beta drivers to these?

hcorion commented on 2017-06-29 18:40

Yeah, thanks a lot!

Det commented on 2017-06-29 18:34

Works now?

hcorion commented on 2017-06-29 18:31

Hi, I got this error when I was trying to update this on Linux 4.11.7.

==> Starting package_nvidia-utils-full-beta()...
install: cannot stat 'nvidia_icd.json': No such file or directory

Det commented on 2017-05-01 07:40

Added the "illegal" 4.11 patch:

Det commented on 2017-02-13 09:24

-2 should be compatible with xorg-server 1.19.1-3:


..and in sync with nvidia-utils 375.26-3:


zacjor commented on 2016-12-02 22:38

The dependency lib32-libcl has been replaced with lib32-opencl-icd-loader.

Det commented on 2016-07-16 04:51


Note, you might not wanna clear your Pacman cache from .27:

dustball commented on 2016-04-07 18:47

Why exactly does this require "nvidia-utils-full-beta-all", which provides "libglvnd", and then conflicts with "libglvnd"?

This is pulled in as a requirement that can't ever be fulfilled!

Awesome_Donkey commented on 2016-03-27 20:15


I've ran into an issue when trying to install this (nvidia-full-beta nvidia-libgl-full-beta nvidia-utils-full-beta lib32-nvidia-libgl-full-beta lib32-opencl-nvidia-full-beta). The installation will fail due to conflicting files;

error: failed to commit transaction (conflicting files)
nvidia-libgl-full-beta: /usr/lib/ exists in filesystem
nvidia-libgl-full-beta: /usr/lib/ exists in filesystem
nvidia-utils-full-beta: /usr/lib/ exists in filesystem
nvidia-utils-full-beta: /usr/lib/ exists in filesystem
nvidia-utils-full-beta: /usr/lib/ exists in filesystem
nvidia-utils-full-beta: /usr/lib/ exists in filesystem
nvidia-utils-full-beta: /usr/lib/nvidia/ exists in filesystem
nvidia-utils-full-beta: /usr/lib/nvidia/ exists in filesystem
nvidia-utils-full-beta: /usr/lib/nvidia/ exists in filesystem
nvidia-utils-full-beta: /usr/lib/nvidia/ exists in filesystem
nvidia-utils-full-beta: /usr/lib/nvidia/ exists in filesystem
nvidia-utils-full-beta: /usr/lib/nvidia/ exists in filesystem
Errors occurred, no packages were upgraded.

I've ran pacman -Qo on every single conflict, and they all are owned by libglvnd How would I proceed to fix this conflict? I'm at a loss here - I'm running nvidia lib32-nvidia-libgl lib32-nvidia-utils lib32-opencl-nvidia nvidia-libgl nvidia-utils


EDIT: Fixed it by running yaourt -Rdd libglvnd lib32-libglvnd and trying again. However I've run into another issue;

error: missing 'lib32-libgl' dependency for 'lib32-glu'
error: missing 'lib32-libgl' dependency for 'steam'

I've tried reinstalling lib32-nvidia-libgl-full-beta multiple times.

EDIT 2: Got it. Had to remove everything and start from scratch. Nonetheless there's a conflict with libglvnd somewhere. Anyways all's well now. :)

Athemis commented on 2015-08-03 18:28

Works like a charm. Thanks! :)

Det commented on 2015-08-03 18:24

Does it work now?

Athemis commented on 2015-08-03 17:36

Does not work with 355.06:

* nvidia_uvm.ko is built by default now, the ouput path of the module has changed (put directly in 'kernel' instead of 'kernel/uvm').

* Versioning/filename of$pkgver changed to

* was added

Det commented on 2015-07-29 06:23

352.30 — Changelog:

✔ Added support for the following GPU:
   ✓ Tesla K80
   ✓ GeForce 910M

✔ Fixed a bug that caused poor video post-processing performance in VDPAU when operating on a large number of video streams simultaneously.
✔ Updated nvidia-installer to use modprobe(8) when leaving the NVIDIA kernel module loaded after installation, instead of insmod(8) or libkmod. This allows the kernel module to honor any configuration directives that apply to it in /etc/modprobe.d when it is loaded.
✔ Fixed a bug that allowed console messages from the Linux kernel to be drawn over the user interface of nvidia-installer.

Det commented on 2015-04-14 13:37


Det commented on 2015-03-25 18:14


Det commented on 2015-03-05 13:52

346.47-2 - Added patch for 4.0 RC:

Det commented on 2015-02-26 09:30

346.47 — Changelog:

• Added support for the following GPUs:
   Quadro K620M
   Quadro K2200M
   GeForce GTX 965M
• Fixed a bug that could cause rendering corruption in GLX clients using PBOs and/or VBOs when using GLX indirect rendering.
• Fixed a bug that caused Xinerama layouts which included X screens with 'Option "UseDisplayDevice" "none"' to be represented incorrectly in the nvidia-settings control panel.
• Fixed a bug that could cause glXSwapBuffer() to block for longer than necessary in multi-threaded GLX applications using the GLX_NV_delay_before_swap extension.
• Fixed a bug that caused OpenGL applications using the NV_path_rendering extension to crash after a modeswitch event.
• Fixed a bug that caused DisplayPort audio to stop working after monitors are hotplugged.

Det commented on 2015-01-16 10:34

Added patches for kernel...:

- 3.18 (Optimus):
- 3.19:

Det commented on 2014-12-08 20:42

346.22 - Changelog:

• Added support for X.Org xserver ABI 19 (xorg-server 1.17).
• Improved compatibility with recent Linux kernels.
• Fixed a bug that prevented internal 4K panels on some laptops from being driven at a sufficient bandwidth to support their native resolutions.
• Fixed a regression that prevented the NVIDIA kernel module from loading in some virtualized environments such as Amazon Web Services.
• Fixed a regression that caused displays to be detected incorrectly on some notebook systems.
• Fixed a bug that could cause X to freeze when using Base Mosaic.
• Fixed a regression that prevented the NVIDIA X driver from recognizing Base Mosaic layouts generated by the nvidia-settings control panel.

Commander commented on 2014-09-19 03:19

343.22 - Changelog:

Det commented on 2014-08-06 21:15

343.13 - Changelog:

** WARNING: This version drops support for pre-GeForce 400 "Fermi" GPUs: **

This warning is included in a couple of places in the package to (try to) make sure people don't miss it.

Det commented on 2014-07-23 00:24

(direct link: I'm not sure how is Pacaur able to affect Makepkg to prevent it from producing the correct pkgname=() array (Pacaur itself doesn't have to see the correct packages), but neither one of those solutions is really all that convenient, so if he (Spyhawk) doesn't want to set pkgname by sourcing the PKGBUILD itself, you'd probably just need to keep on manually building with makepkg, or picking another AUR tool.

LadFromWales commented on 2014-07-22 21:19

I posted post #450 regarding this package no longer functioning correctly with pacaur.
Response in post #452 states that this is an issue with the package as the packages installed aren't on the AUR or RPC.

Could this be looked into?

Det commented on 2014-07-09 11:58

340.24 Changelog:

I switched the source to the HTTP mirror, which gives me pretty much the full speed of ~10MB/s, while the FTP one is seemingly capped at 886 kB/s (many times much less). The web interface will still show the FTP mirror (overridden in the .AURINFO), as it gives you a much easier access to all the driver downloads.

Det commented on 2014-07-09 10:15


z1lt0id commented on 2014-07-09 01:01

It was due to me missing linux-headers, which should be in the deps of this PKGBuild.

Also 340.24 has just been released.

Det commented on 2014-07-01 17:12

No idea. Can't reproduce. Maybe try rebuilding manually.

z1lt0id commented on 2014-06-29 23:52

Won't build for me.
==> Starting build()...
-> Building Nvidia module for 3.15.2-1-ARCH...
NVIDIA: calling KBUILD...
make[1]: Entering directory '/usr/lib/modules/3.15.2-1-ARCH/build'
make[1]: *** No rule to make target 'modules'. Stop.
make[1]: Leaving directory '/usr/lib/modules/3.15.2-1-ARCH/build'
nvidia.ko failed to build!
Makefile:176: recipe for target 'nvidia.ko' failed
make: *** [nvidia.ko] Error 1
==> ERROR: A failure occurred in build().
==> ERROR: Makepkg was unable to build nvidia-full-beta.
==> Restart building nvidia-full-beta ? [y/N]

Det commented on 2014-04-08 15:53

337.12 Changelog:

Det commented on 2014-04-02 15:03

334.21-3: Fixed:
- FS#39203:
- FS#39525:

Det commented on 2014-03-09 18:50

When receiving the complaint about the existing files, just pass the '-f' flag to Pacman:

(lib32-)nvidia-libgl-full-beta: /usr/lib(32)/ exists in filesystem
(lib32-)nvidia-libgl-full-beta: /usr/lib(32)/ exists in filesystem
(lib32-)nvidia-libgl-full-beta: /usr/lib(32)/ exists in filesystem

Det commented on 2014-03-09 18:43

- Moved the EGL/GLES symlinks to (lib32-)nvidia-libgl-full-beta.[1]
- Added the Unified Memory module (nvidia-uvm.ko).[2]

[1] =
[2] =

Det commented on 2014-03-09 18:41

- Moved the EGL/GLES symlinks to (lib32-)nvidia-libgl-full-beta.[1]
- Added the Unified Memory module (nvidia-uvm.ko).[2]

[1] =
[2] =

Det commented on 2014-02-23 13:56

Added the EGL/OpenGL ES libraries: The necessary symlinks only replace those of (lib32-)mesa in post_install/upgrade, for now.

There's also a bunch of other stuff included, most of which are mentioned in FS#38604 (

Det commented on 2013-11-02 08:40

@thgravo, it does (it's a make dependency - the driver doesn't need the sources once it's compiled).

Det commented on 2013-11-02 08:38

@thgravo, it does.

Det commented on 2013-11-02 08:36


* Use a better patch:

thgravo commented on 2013-10-31 08:51

Shouldn't this depend on linux-headers?

Det commented on 2013-10-25 18:15

Never even thought of that.

TheRealSoup commented on 2013-10-23 10:59

Love what you've done with this package.

I do have a small concern with _major= using uname though, I can see 3 possible issues:
* having to reboot into the new kernel to be able to build nvidia for it
* building inside install media that has an older kernel
* building inside chroots where the host isn't Arch

Maybe something like this?
_major=$(pacman -Q linux|awk -F '[ .]' '{print $2"."$3}')

Det commented on 2013-06-27 13:04

So, I'll be copy-pasting the same thing here, which is that this package (along with 'nvidia-full-beta-all'[1]) is for people who are tired of the update conflicts and downloading the same source twice for the two different halves of the nvidia driver.

64-bit systems can also set the option "_lib32=1", which will pull in the 32-bit compatibility sources for all parties involved (again, obviating from the unnecessity of downloading multiple sources) and use them to build the 'lib32-nvidia-utils-full-beta' group.

This only needs to be set once and will remain until the lib32-* packages are removed.

[1] =