I have a problem with the kernel-4.17-rc on Xfce4-gtk3 in VirtualBox . After I login, with any program that I load it could not be moved, resized. Others kernels works
Bottom line is 304.xx and 340.xx nvidia proprietary drivers don't support xorg 1.20, and all indications are new libglvnd support will not be backported to these.
One option is to roll forward with xorg 1.20 and either force these nvidia users to nouveau / prime nouveau, which as you point out will most likely end in severe anguish for the less technically inclined.
This transition is not straight forward and will require manual intervention and some knowledge to setup / troubleshoot.
Another possible option is to package an xorg 1.19 compat package group to allow the 304.xx and 3.40.xx drivers to continue to operate. Keep this xorg 1.19 compat package group around for six months or so and announce the phasing out of 304.xx / 340.xx drivers, giving users time to plan and test their transition to xorg 1.20 compatible video drivers.
All this is predicated on nvidia not backporting these legacy drivers to xorg 1.20 compliance.
Personally I am moving my primary laptop back to testing as even the 396.xx driver does not work with bumblebee and xorg 1.20. My other optimus laptop I'll keep on unstable to test.
2 Likes
Xorg 1.20 is in Arch Stable, what are legacy nvidia Arch users doing?
writing bugreports: https://bugs.archlinux.org/task/58660?project=1&string=nvidia-340xx
and let me guess what will happen... these drivers will go to !aur
3 Likes
Without xorg 1.20+ compatibility they are useless, whether in official repos or AUR.
Without nvidia backporting only way to use these drivers is with xorg <= 1.19.
Just another 
from nvidia.
1 Like
And another 
for the community. Because it isnt surprising that Nvidia would do this, but it's just surprising that nobody where prepared for it...
I mean we're moving on with all these stuff in webdad, gnome, firefox, KDE-Stuff and whatever BUT cant¹ get² our³ base⁴ system⁵ right⁶? For me it's so frustrating to see this.
1 Like
I am in the testing and today I have a vlc-nightly update and when doing this update I have this message:
attention: linux-firmware: local (20180518.86d39f8-1) is newer then core (20180518.2a9b2cf-1)
should I downgrade?
Normally, you can just ignore these messages if your system is working without problems.
1 Like
xorg 1.20 with GLVND, I hope this will help optimus users.
Current differences from testing
:: Different overlay package(s) in repository community x86_64
-------------------------------------------------------------------------------
PACKAGE testing unstable
-------------------------------------------------------------------------------
bumblebee 3.2.1-18 3.2.1-19
compiz-manjaro 0.9.13.1.r4143-1 0.9.13.1.r4143-2
deepin-session-ui 4.3.7.3-1 4.3.7.3-2
sddm-classic 0.15.0-2 0.15.0-3
skippy-xd 2015.12.21.r0.gd0a11f4-1 2016.10.20.r0.g72d1ddf-1
wallpaperd 0.2.1-0.1 -
xf86-video-qxl 0.1.5-5 -
xlogin 0.r11.a578cf2-1 0.r14.6e566b5-1
:: Different sync package(s) in repository community x86_64
-------------------------------------------------------------------------------
PACKAGE testing unstable
-------------------------------------------------------------------------------
xf86-video-qxl - 0.1.5-6
:: Different overlay package(s) in repository extra x86_64
-------------------------------------------------------------------------------
PACKAGE testing unstable
-------------------------------------------------------------------------------
libva-mesa-driver 18.0.4-0 18.1.0-0
mesa 18.0.4-0 18.1.0-0
mesa-vdpau 18.0.4-0 18.1.0-0
opencl-mesa 18.0.4-0 18.1.0-0
vulkan-intel 18.0.4-0 18.1.0-0
vulkan-radeon 18.0.4-0 18.1.0-0
xf86-video-amdgpu 18.0.1-1 -
xf86-video-ati 1_7.9.0-1 -
xf86-video-dummy 0.3.8-2 -
xf86-video-fbdev 0.4.4+12+ge0bce0d-1 -
xf86-video-intel 1_2.99.917+829+gd7dfab62-1 -
xf86-video-nouveau 1.0.15-2 -
xf86-video-openchrome 0.6.0-2 -
xf86-video-sisusb 0.9.7-1 -
xf86-video-vesa 2.4.0-1 -
xf86-video-vmware 13.2.1-3 -
xf86-video-voodoo 1.2.5-9 -
xorg-server 1.19.6+24+g56547b196-1 -
xorg-server-common 1.19.6+24+g56547b196-1 -
xorg-server-devel 1.19.6+24+g56547b196-1 -
xorg-server-xdmx 1.19.6+24+g56547b196-1 -
xorg-server-xephyr 1.19.6+24+g56547b196-1 -
xorg-server-xnest 1.19.6+24+g56547b196-1 -
xorg-server-xvfb 1.19.6+24+g56547b196-1 -
xorg-server-xwayland 1.19.6+24+g56547b196-1 -
gdm - 3.28.1-1.1
libgdm - 3.28.1-1.1
:: Different sync package(s) in repository extra x86_64
-------------------------------------------------------------------------------
PACKAGE testing unstable
-------------------------------------------------------------------------------
egl-wayland - 0+23-1
eglexternalplatform - 0+3-1
xf86-video-amdgpu - 18.0.1-2
xf86-video-ati - 1:18.0.1-2
xf86-video-dummy - 0.3.8-3
xf86-video-fbdev - 0.4.4+13+g9af7f81-1
xf86-video-intel - 1:2.99.917+831+ge7bfc906-1
xf86-video-nouveau - 1.0.15-3
xf86-video-openchrome - 0.6.0-3
xf86-video-sisusb - 0.9.7-2
xf86-video-vesa - 2.4.0-2
xf86-video-vmware - 13.3.0-1
xf86-video-voodoo - 1.2.5-10
xorg-server - 1.20.0-2
xorg-server-common - 1.20.0-2
xorg-server-devel - 1.20.0-2
xorg-server-xdmx - 1.20.0-2
xorg-server-xephyr - 1.20.0-2
xorg-server-xnest - 1.20.0-2
xorg-server-xvfb - 1.20.0-2
xorg-server-xwayland - 1.20.0-2
:: Different overlay package(s) in repository multilib x86_64
-------------------------------------------------------------------------------
PACKAGE testing unstable
-------------------------------------------------------------------------------
lib32-libva-mesa-driver 18.0.4-0 18.1.0-0
lib32-mesa 18.0.4-0.1 18.1.0-0.1
lib32-mesa-vdpau 18.0.4-0 18.1.0-0
lib32-vulkan-intel 18.0.4-0 18.1.0-0
lib32-vulkan-radeon 18.0.4-0 18.1.0-0
1 Like
lib32-wayland required deleting
-
/usr/lib32/libwayland-egl.so , libwayland-egl.so.1 , libwayland-egl.so.1.0.0
-
/usr/lib32/pkgconfig/wayland-egl.pc
ansible 2.5.3-2 solved my problem 
1 Like
You mean after Xorg 1.20 broke our nvidia 304 and 340 series?
We downgraded Xorg 1.20 to the previous one.
Certainly not solved for us and not likely to be from the devtalk thread.
I suppose 1.20 will be breaking my Manjaro install shortly,
cheers
You may switch to testing. Only in unstable we have xorg-server 1.20 series for now, which is currently tagged as blocker package.
I'm using stable. I just posted in this thread because of the trauma of a non booting Arch yesterday and then discovering that nvidia will not do anything for legacy users.
Thanks anyway,
Hugh
Update: There were a few very minor hiccups, but I made the transition to Nouveau from the proprietary 304xx, and have been using several days. I've been pleasantly surprised.
To get kwin usable, I had to switch compositor rendering backend to XRender from the default OpenGL 2.0. I had to tweak a few files, but it was less difficult, and more successful than any previous attempt to run Nouveau.
So far so good
3 Likes
Would you be so kind as to explain the steps you took?
PM if you don't want to post in this thread but it would be useful.
Maybe you could start another thread somewhere.
I think there are going to be more people in our situation.
thanks in advance
2 Likes
If you have issues with bumblebee, you may remove xf86-input-mouse package and see if that helps you out.
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.
