You are not logged in.
Hi all ,
i have been struggling with this for two years now and no progress ,
installing the nvidia driver gives me a black screen no matter what i did , i followed the wiki for the letter , tried building the kernel from git , tried every nvidia driver available (beta , dkms)
on gnome i can get X and the DE to work until i suspend and i get black screen again ,
am on Plasma now & sddm ,
the hardware is Asus zenbook ux303ub using nvidia 940m in an optimus configuration ,
and here is the X logs :
Offline
Can you post dmesg and lspci logs? That way we know a bit more about your hardware and what is happening.
Offline
Can you post dmesg and lspci logs? That way we know a bit more about your hardware and what is happening.
thanks for the reply ,
lspci :
dmesg :
Offline
You have an optimus system, you will want to look at https://wiki.archlinux.org/index.php/NVIDIA_Optimus for setup information
Offline
You have an optimus system, you will want to look at https://wiki.archlinux.org/index.php/NVIDIA_Optimus for setup information
i tried that already with no luck
Offline
That is neither an error message, nor does it contain exact information on what you did. These instructions are known to work for many people, given that they are followed correctly.
Read: https://bbs.archlinux.org/viewtopic.php?id=57855
What doesn't work? Post all the files you adjusted verbatim, post an updated xorg log/dmesg. If you did follow the instructions for SDDM, have you made sure that the file is execute-/readable to non root users?
Mod note: Moving to NC
Edit: Went through your post history, we literally did already have this discussion 2 years ago. Which showed that these instructions did work, though with the caveat that suspend might be broken, which is a known risk with this setup. Although from what I've heard people have been suspending and resuming fine with recent setups. So depending on how long ago your last attempt was, doing that now might lead to better results. If they don't you are likely to have to use the bumblebee or nvidia-xrun methods to only use the nvidia card on demand. You might also want to update your UEFI, that also has a chance of leading to better results as suspension resume faults are often related there.
Last edited by V1del (2018-04-20 02:30:26)
Offline
That is neither an error message, nor does it contain exact information on what you did. These instructions are known to work for many people, given that they are followed correctly.
Read: https://bbs.archlinux.org/viewtopic.php?id=57855
What doesn't work? Post all the files you adjusted verbatim, post an updated xorg log/dmesg. If you did follow the instructions for SDDM, have you made sure that the file is execute-/readable to non root users?
Mod note: Moving to NC
Edit: Went through your post history, we literally did already have this discussion 2 years ago. Which showed that these instructions did work, though with the caveat that suspend might be broken, which is a known risk with this setup. Although from what I've heard people have been suspending and resuming fine with recent setups. So depending on how long ago your last attempt was, doing that now might lead to better results. If they don't you are likely to have to use the bumblebee or nvidia-xrun methods to only use the nvidia card on demand. You might also want to update your UEFI, that also has a chance of leading to better results as suspension resume faults are often related there.
The setup didn't work previously , i simply gave up and continued using intel , i even opened a bug report to Nvidia , regarding the logs i will supply them in a few hours , that's the problem with arch users always presuming every one is a noop but that's ok . And for the record i have been using arch for 8 years .
Last edited by step-2 (2018-04-20 03:41:59)
Offline
We only know what you provide, it's quite easy to quell that assumption by showing what you did. If you really did do that you might want to post a link to the bug report so we know what has already been tried and what hasn't. If we take your first post on its own, there's absolutely no indication on what attempts at the setup have already been taken and in what way they failed.
In the worst case you just picked the short straw and your particular system really simply doesn't work satisfactorily with this setup, we haven't ruled out all the straws yet by a long shot though.
Offline
We only know what you provide, it's quite easy to quell that assumption by showing what you did. If you really did do that you might want to post a link to the bug report so we know what has already been tried and what hasn't. If we take your first post on its own, there's absolutely no indication on what attempts at the setup have already been taken and in what way they failed.
In the worst case you just picked the short straw and your particular system really simply doesn't work satisfactorily with this setup, we haven't ruled out all the straws yet by a long shot though.
Offline
So that doesn't really contain anything. Back to square one. Set up Optimus according to the article (and as I have a hunch that your suspension problems likely result from that, make sure that you enable modesetting on the nvidia driver, i.e. add a config file at /etc/modprobe.d/nvidia-drm.conf
options nvidia-drm modeset=1) after having done the setup and a reboot and if something still doesn't work post the logs and all files you've adjusted as requested earlier. Should you have to force a reboot due to a freeze post after the successful next boot
journalctl -b-1in addition to these other outputs.
Offline
So that doesn't really contain anything. Back to square one. Set up Optimus according to the article (and as I have a hunch that your suspension problems likely result from that, make sure that you enable modesetting on the nvidia driver, i.e. add a config file at /etc/modprobe.d/nvidia-drm.conf
options nvidia-drm modeset=1) after having done the setup and a reboot and if something still doesn't work post the logs and all files you've adjusted as requested earlier. Should you have to force a reboot due to a freeze post after the successful next boot
journalctl -b-1in addition to these other outputs.
journalctl -b-1 :
xorg :
Offline
There's no indication to an attempt at loading the nvidia driver in the xorg log. Do you have bumblebee installed? Remove it, also remove xf86-video-intel it can only interfere, you still haven't posted the config files...
Offline
I have xf86-video-intel installed , i will remove it , with nvidia-beta i can see the cursor and i can move it but i can't see sddm , i will supply the logs after work .
Offline
There's no indication to an attempt at loading the nvidia driver in the xorg log. Do you have bumblebee installed? Remove it, also remove xf86-video-intel it can only interfere, you still haven't posted the config files...
journalctl -b-1 :
i force rebooted but no X log .
Last edited by step-2 (2018-04-21 01:15:39)
Offline
Offline
[ 3.659] (II) NVIDIA(0): Detected PCI Express Link width: 4X
[ 3.659] (II) NVIDIA(0): Validated MetaModes:
[ 3.659] (II) NVIDIA(0): "NULL"
[ 3.659] (II) NVIDIA(0): Virtual screen size determined to be 640 x 480
[ 3.659] (WW) NVIDIA(0): Unable to get display device for DPI computation.
[ 3.659] (==) NVIDIA(0): DPI set to (75, 75); computed from built-in defaulthttps://wiki.archlinux.org/index.php/NV … timus#SDDM
Are you really trying to exclusively run on the nvidia chip or do you want to drive this in hybrid mode?
Offline
[ 3.659] (II) NVIDIA(0): Detected PCI Express Link width: 4X [ 3.659] (II) NVIDIA(0): Validated MetaModes: [ 3.659] (II) NVIDIA(0): "NULL" [ 3.659] (II) NVIDIA(0): Virtual screen size determined to be 640 x 480 [ 3.659] (WW) NVIDIA(0): Unable to get display device for DPI computation. [ 3.659] (==) NVIDIA(0): DPI set to (75, 75); computed from built-in defaulthttps://wiki.archlinux.org/index.php/NV … timus#SDDM
Are you really trying to exclusively run on the nvidia chip or do you want to drive this in hybrid mode?
Thanks for the reply ,
I followed these instructions with the same results ,
Yes i want to run on Nvidia cause the hardware run cooler according to gnome , in gnome there is the issue of suspending .
Offline
I followed these instructions with the same results ,
What was the contents of /etc/X11/xorg.conf and /usr/share/sddm/scripts/Xsetup after you followed the instructions?
Offline
step-2 wrote:I followed these instructions with the same results ,
What was the contents of /etc/X11/xorg.conf and /usr/share/sddm/scripts/Xsetup after you followed the instructions?
/etc/X11/xorg.conf :
Section "Module"
Load "modesetting"
EndSection
Section "Device"
Identifier "nvidia"
Driver "nvidia"
BusID "1:0:0"
Option "AllowEmptyInitialConfiguration" EndSection
/usr/share/sddm/scripts/Xsetup :
xrandr --setprovideroutputsource modesetting NVIDIA-0
xrandr --auto
Offline
Sidenote: Please use code tags.
The Device section in your post isn't closed; please post the xorg log for this setup.
Offline
Sidenote: Please use code tags.
The Device section in your post isn't closed; please post the xorg log for this setup.
X log :
Offline
Did sddm display correctly before suspend? If not please stop/disable sddm and try startx.
Remember to configure xinitrc including the xrandr lines before trying startx.
Offline
Did sddm display correctly before suspend? If not please stop/disable sddm and try startx.
Remember to configure xinitrc including the xrandr lines before trying startx.
am using gdm at the moment , with the current configuration it's the only display manager that allow me to start plasma ,
and when i suspend i get the black screen ,
starting x manually gives me a black screen .
Offline
If you start just "Xorg", you get a black screen and a cursor - that's completely normal.
The problem w/ GDM might be the invocation of wayland on the intel chip, see https://wiki.archlinux.org/index.php/GD … rg_backend
Why, [edit: "ie. in how far"] do other DMs not "allow [you] to start plasma"?
Last edited by seth (2018-04-22 06:23:09)
Offline
If you start just "Xorg", you get a black screen and a cursor - that's completely normal.
The problem w/ GDM might be the invocation of wayland on the intel chip, see https://wiki.archlinux.org/index.php/GD … rg_backend
Why, [edit: "ie. in how far"] do other DMs not "allow [you] to start plasma"?
starting X results in a black screen with no backlight ,
and yes gdm is running on X backend & xf86-video-intel is uninstalled ,
i have tried xdm , lightdm and sddm all with the same result ,
lightdm used to show but can't launch plasma , some of the configuration files must have broke it ,
also when waking from suspending i get black screen but i can see the backlight and the machine freeze and i have to hard reboot .
Last edited by step-2 (2018-04-22 12:47:03)
Offline