You are not logged in.
I updated my kernel to 6.9.7-arch1-1 alongside nvidia-dkms today and after rebooting my system was unable to boot past the userspace. I was unable to enter tty via the function keys so my only resort is to chroot
I’m lost on what the issue could be, let me know if you require any logs (something like journalctl im not sure how to gather from the chroot environment to here)
Last edited by fasdf344 (2024-06-29 17:35:42)
Offline
Same problem here, RTX 3070, AMD Ryzen 5 3600 with X11 and i3-wm. Upgraded the kernel (to 6.9.7.arch1-1) along with nvidia (to 555.58-2) today and after rebooting got a black screen saying that the screen is not receiving signal, changing tty isn’t possible. Also tried adding modeset=1 on the kernel parameters as suggested by some users but it didn’t help (tried different combinations of modeset=1 with/without fbdev=1, kernel parameters/modprobe.conf, both, and adding the modules to the mkinitcpio.conf file).
I ended up reverting to the previous kernel and drivers for now but I have seen plenty of people reporting the same issue today. If you want a short term solution do that in the meantime.
Last edited by dimaheve (2024-06-29 22:33:26)
Offline
Also tried adding modeset=1 on the kernel parameters as suggested by
[edit: realizing that you were likely just casually paraphrasing]
The proper parameter would be "nvidia_drm.modeset=1", do you have the system journal for the bad boot?
Last edited by seth (2024-06-30 08:11:45)
Offline
Same problem here, RTX 3070, AMD Ryzen 5 3600 with X11 and i3-wm. Upgraded the kernel (to 6.9.7.arch1-1) along with nvidia (to 555.58-2) today and after rebooting got a black screen saying that the screen is not receiving signal, changing tty isn’t possible. Also tried adding modeset=1 on the kernel parameters as suggested by some users but it didn’t help (tried different combinations of modeset=1 with/without fbdev=1, kernel parameters/modprobe.conf, both, and adding the modules to the mkinitcpio.conf file).
I ended up reverting to the previous kernel and drivers for now but I have seen plenty of people reporting the same issue today. If you want a short term solution do that in the meantime.
May I ask what packages did you downgrade? I downgraded my kernel to 6.9.4 alongside my drivers which matched the kernel version and I still had the same issue
Offline
If your graphical.target insta-restarts switching the VT might fail despite there being no issue with the framebuffer itself, can you boot the multi-user.target (2nd link below)?
Offline
If your graphical.target insta-restarts switching the VT might fail despite there being no issue with the framebuffer itself, can you boot the multi-user.target (2nd link below)?
I changed the symlink using the link below
# systemctl get-default
multi-user.target
However it still shows a black screen when rebooting, heres the journal of the last boot
Offline
There's no obvious problem, try w/o nvidia_drm.modeset=1 to allow the simpledrm device - can you boot the multi-user.target this way?
Does nvidia-535xx-dkms from the AUR still work?
Offline
There's no obvious problem, try w/o nvidia_drm.modeset=1 to allow the simpledrm device - can you boot the multi-user.target this way?
Does nvidia-535xx-dkms from the AUR still work?
Downgrading to the 550 drivers I had in my pacman cache allowed me to boot up into tty, changed my symlink back to boot up into my desktop environment
I’m just confused on why when I downgraded earlier it still refused to boot?
Offline
Check the journal of the boot with the failing, supposedly downgraded, system - it's impossible to guess what might have gone wrong.
Offline
I have the exact same problem, and the only way I could get back to graphics mode was to downgrade to the state of June 28th , ie, before the kernel and nvidia drivers were updated:
https://archive.archlinux.org/repos/2024/06/28/$repo/os/$arch
Puzzling my head about this one...
Same problem here, RTX 3070, AMD Ryzen 5 3600 with X11 and i3-wm. Upgraded the kernel (to 6.9.7.arch1-1) along with nvidia (to 555.58-2) today and after rebooting got a black screen saying that the screen is not receiving signal, changing tty isn’t possible. Also tried adding modeset=1 on the kernel parameters as suggested by some users but it didn’t help (tried different combinations of modeset=1 with/without fbdev=1, kernel parameters/modprobe.conf, both, and adding the modules to the mkinitcpio.conf file).
I ended up reverting to the previous kernel and drivers for now but I have seen plenty of people reporting the same issue today. If you want a short term solution do that in the meantime.
Offline
Hi got a similar problem after updating to nvidia-driver 555.58 reverting to previous driver solve the issue
using a NVIDIA GeForce GTX 1050
2 similar monitor, had no problem for years
now one is not working
I found out in my case the display problem seems to be related to the display output from nvidia.
HDMI OK
DVI not working as it should (display dmesg fine up to x or wayland)
what output are you using?
Offline
also note (not my case):
If using a Turing/Ampere+ GPU (aka GTX 1650+), note that >=nvidia-drivers-555
enables the use of the GSP firmware by default. *If* experience regres
sions,
please see '/etc/modprobe.d/nvidia.conf' to optionally disable.
Offline
Hi got a similar problem after updating to nvidia-driver 555.58 reverting to previous driver solve the issue
using a NVIDIA GeForce GTX 1050
2 similar monitor, had no problem for years
now one is not working
I found out in my case the display problem seems to be related to the display output from nvidia.
HDMI OK
DVI not working as it should (display dmesg fine up to x or wayland)
what output are you using?
I encountered a similar issue with the DVI port after updating the NVIDIA driver to version 555. Has anyone else experienced this problem? Please share any solutions.
Offline
jmsgentoo wrote:Hi got a similar problem after updating to nvidia-driver 555.58 reverting to previous driver solve the issue
using a NVIDIA GeForce GTX 1050
2 similar monitor, had no problem for years
now one is not working
I found out in my case the display problem seems to be related to the display output from nvidia.
HDMI OK
DVI not working as it should (display dmesg fine up to x or wayland)
what output are you using?I encountered a similar issue with the DVI port after updating the NVIDIA driver to version 555. Has anyone else experienced this problem? Please share any solutions.
it seems not to be an isolated case with the DVI port ...
https://bbs.archlinux.org/viewtopic.php … 5#p2181025
Offline
I have the same problem (jmsgentoo linked it in prev post) - it is on the DVI port.
Last edited by catcattycat (2024-07-01 04:37:41)
Offline
I am having the same problem. I have two monitors connected -- one HDMI, another one DVI. It was fine until yesterday, but after update and reboot, DVI is showing only black screen.
It seems DVI is the problem. It also showed nothing when only DVI is connected. HDMI works fine when connected, whether the DVI is connected or not.
As gauthma posted, downgrading to the date before update (in my case June 20th) solved the problem, at least temporarily.
I have the exact same problem, and the only way I could get back to graphics mode was to downgrade to the state of June 28th , ie, before the kernel and nvidia drivers were updated:
https://archive.archlinux.org/repos/2024/06/28/$repo/os/$arch
edit:
Downgrading only nvidia and linux also worked. The "downgrade" AUR package was helpful. My current yay output:
:: Synchronizing package databases...
core is up to date
extra is up to date
multilib is up to date
:: Searching AUR for updates...
:: Searching databases for updates...
-> lib32-nvidia-utils: ignoring package upgrade (550.90.07-1 => 555.58-1)
-> linux: ignoring package upgrade (6.9.6.arch1-1 => 6.9.7.arch1-1)
-> nvidia: ignoring package upgrade (550.90.07-4 => 555.58-2)
-> nvidia-utils: ignoring package upgrade (550.90.07-3 => 555.58-2)
there is nothing to do
Last edited by freleefty (2024-07-01 12:31:36)
Offline
Hello !
I've got 3 screens, the DP and HDMI work, but exactly the same problem on the DVI port ...
Since the last nvidia driver's update on June 28th !
Last edited by bigfoot92 (2024-07-01 12:35:01)
Offline
'Fixed' my issue just by switching from the DVI port to the DP port. : )
Offline
Same problem, but DVI isn't in the picture.
4070 TI going to a single Dell 38" monitor across HDMI.
Screen goes blank (edit: "blank" here means no signal being sent to the monitor) before even getting to the luksOpen prompt during the boot process. Same behavior on linux and LTS, as well as nvidia-open-dkms.
Workaround was to downgrade to 550
Last edited by jason98893 (2024-07-02 22:20:36)
Offline
@jason98893 since most of this mandates modules in early boot, is that set up, and if so, did you rebuild your initramfs when the nvidia driver updated?
Offline
Does anyone know if this issue still happens after the latest upgrade (kernel 6.9.9.arch1-1 and nvidia 555.58.02-3)?
Offline
Almost certainly so. Whenever just the nvidia package updates without an actual version change or an accompanying nvidia-utils release, the only thing that changes is the kernel module getting rebuilt against the newer kernel version. If the bug is in the driver, then the code hasn't changed and the issue wouldn't be fixed.
Offline
@gauthma yes it does, I'm on the latest kernel (6.9.9-arch1-1) and there is no difference in behavior, my second (DVI) monitor is blank. The same happens with nvidia-dkms version 550
Offline
I have the same issue. Downgraded to:
core/linux 6.8.7.arch1-1 6.10.arch1-2 3.21 MiB
extra/nvidia 550.76-1 555.58.02-8 1.52 MiB
extra/nvidia-settings 550.78-1 555.58.02-1 0.00 MiB
extra/nvidia-utils 550.76-3 555.58.02-1 22.80 MiB
as workaround.
Connected via hdmi to kvm switch. Running a 1070.
This is the second time in 3-4 weeks that I've attempted latest levels. Same problem.
When it's borked I can't even get to vt's. Fortunately ssh is up so I can get in to repair without booting from a usb key.
Cheers,
Arkay.
Offline
I had this error yesterday
[2024-07-23T23:05:39+0200] [ALPM-SCRIPTLET] -> Running build hook: [fsck]
[2024-07-23T23:05:39+0200] [ALPM-SCRIPTLET] ==> ERROR: module not found: 'nvidia'
[2024-07-23T23:05:39+0200] [ALPM-SCRIPTLET] ==> Generating module dependencies
[2024-07-23T23:05:39+0200] [ALPM-SCRIPTLET] ==> Creating zstd-compressed initcpio image: '/boot/initramfs-linux-zen.img'
and I knew it was going cause problems so I tried resinstalling the Nvidia package but it did not help.
What did work is
[2024-07-23T23:13:26+0200] [ALPM] removed nvidia-535xx-dkms (535.183.01-1)
[2024-07-23T23:13:26+0200] [ALPM] removed nvidia-535xx-utils (535.183.01-1)
[2024-07-23T23:13:27+0200] [ALPM] installed nvidia-utils (555.58.02-1)
[2024-07-23T23:13:27+0200] [ALPM-SCRIPTLET] If you run into trouble with CUDA not being available, run nvidia-modprobe first.
[2024-07-23T23:13:27+0200] [ALPM-SCRIPTLET] If you use GDM on Wayland, you might have to run systemctl enable --now nvidia-resume.service
[2024-07-23T23:13:27+0200] [ALPM] installed nvidia-open-dkms (555.58.02-8)
[2024-07-23T23:13:28+0200] [ALPM] transaction completed
[2024-07-23T23:13:28+0200] [ALPM] transaction completed
[2024-07-23T23:13:28+0200] [ALPM] running '20-systemd-sysusers.hook'...
[2024-07-23T23:13:28+0200] [ALPM] running '30-systemd-daemon-reload-system.hook'...
[2024-07-23T23:13:28+0200] [ALPM] running '30-systemd-udev-reload.hook'...
[2024-07-23T23:13:28+0200] [ALPM] running '30-systemd-update.hook'...
[2024-07-23T23:13:28+0200] [ALPM] running '70-dkms-install.hook'...
[2024-07-23T23:13:28+0200] [ALPM-SCRIPTLET] ==> dkms install --no-depmod nvidia/555.58.02 -k 6.10.0-zen1-2-zen
[2024-07-23T23:14:59+0200] [ALPM-SCRIPTLET] ==> depmod 6.10.0-zen1-2-zen
[2024-07-23T23:15:01+0200] [ALPM] running '90-mkinitcpio-install.hook'...
[2024-07-23T23:15:01+0200] [ALPM-SCRIPTLET] ==> Building image from preset: /etc/mkinitcpio.d/linux-zen.preset: 'default'
[2024-07-23T23:15:01+0200] [ALPM-SCRIPTLET] ==> Using default configuration file: '/etc/mkinitcpio.conf'
[2024-07-23T23:15:01+0200] [ALPM-SCRIPTLET] -> -k /boot/vmlinuz-linux-zen -g /boot/initramfs-linux-zen.img
[2024-07-23T23:15:01+0200] [ALPM-SCRIPTLET] ==> Starting build: '6.10.0-zen1-2-zen'
[2024-07-23T23:15:01+0200] [ALPM-SCRIPTLET] -> Running build hook: [base]
[2024-07-23T23:15:01+0200] [ALPM-SCRIPTLET] -> Running build hook: [udev]
[2024-07-23T23:15:02+0200] [ALPM-SCRIPTLET] -> Running build hook: [autodetect]
[2024-07-23T23:15:02+0200] [ALPM-SCRIPTLET] -> Running build hook: [microcode]
[2024-07-23T23:15:02+0200] [ALPM-SCRIPTLET] -> Running build hook: [modconf]
[2024-07-23T23:15:02+0200] [ALPM-SCRIPTLET] -> Running build hook: [block]
[2024-07-23T23:15:03+0200] [ALPM-SCRIPTLET] -> Running build hook: [plymouth]
[2024-07-23T23:15:04+0200] [ALPM-SCRIPTLET] -> Running build hook: [encrypt]
[2024-07-23T23:15:05+0200] [ALPM-SCRIPTLET] ==> WARNING: Possibly missing firmware for module: 'qat_420xx'
[2024-07-23T23:15:05+0200] [ALPM-SCRIPTLET] -> Running build hook: [lvm2]
[2024-07-23T23:15:06+0200] [ALPM-SCRIPTLET] -> Running build hook: [filesystems]
[2024-07-23T23:15:06+0200] [ALPM-SCRIPTLET] -> Running build hook: [keyboard]
[2024-07-23T23:15:06+0200] [ALPM-SCRIPTLET] -> Running build hook: [keymap]
[2024-07-23T23:15:06+0200] [ALPM-SCRIPTLET] -> Running build hook: [fsck]
[2024-07-23T23:15:07+0200] [ALPM-SCRIPTLET] ==> Generating module dependencies
[2024-07-23T23:15:07+0200] [ALPM-SCRIPTLET] ==> Creating zstd-compressed initcpio image: '/boot/initramfs-linux-zen.img'
[2024-07-23T23:15:07+0200] [ALPM-SCRIPTLET] -> Early uncompressed CPIO image generation successful
[2024-07-23T23:15:08+0200] [ALPM-SCRIPTLET] ==> Initcpio image generation successful
I hope this is related and it helps
$ pacman -Ss nvidia | grep 'installed'
extra/egl-wayland 3:1.1.13-1 [installed]
extra/libvdpau 1.5-3 [installed]
extra/libxnvctrl 555.58.02-1 [installed]
extra/nvidia-open-dkms 555.58.02-8 [installed]
extra/nvidia-utils 555.58.02-1 [installed]
multilib/lib32-libvdpau 1.5-2 [installed]
Rlu: 222126
Offline