Solved Kernel panic on RTX 5060 Ti (16GB)

B

bumpyhead

Hi folks,

Just got a 5060 Ti, plugged into the system, and installed the drivers, including
Code: 
nvidia-driver-580.105.08       NVidia graphics card binary drivers for hardware OpenGL rendering
nvidia-drm-61-kmod-580.105.08.1403000 NVIDIA DRM Kernel Module
nvidia-kmod-580.105.08.1403000 kmod part of NVidia graphics card binary drivers for hardware OpenGL rendering
In /boot/loader.conf, I have these lines
Code: 
nvidia_load="YES"
nvidia-modset_load="YES"
In /etc/rc.conf, I have these lines
Code: 
kld_list="cuse nvidia-modeset nvidia"
When the system boots, it gives a panic screen with "fatal trap 12: page fault while in kernel mode". See attached image.
Now I have to unplug the power of the graphic card and use the iGPU instead. Is there an easy fix? Thanks in advance!
 

Attachments

  • 08E04C17-C2B5-494C-8A3B-F11C65E9E076.jpeg
    08E04C17-C2B5-494C-8A3B-F11C65E9E076.jpeg
    258.3 KB · Views: 30
The panic is due to the kernel module being loaded in both /boot/loader.conf and /etc/rc.conf. I put these lines only in rc.conf, and now the system can boot, with nvidia modules loaded.

However I cannot start X. When I try to start X, the system gives a black screen with a tiny white rectangle on the top-left corner, and completely freezes.

Below are /usr/local/etc/X11/xorg.conf.d/10-nvidia.conf
Code: 
Section "Device"
  Identifier "NVIDIA Card"
  Driver "nvidia"
  BusID     "PCI:1:0:0"
  Option "AllowEmptyInitialConfiguration" "true"
  Option "PrimaryGPU" "yes"
EndSection

Section "Monitor"
  Identifier "Monitor0"
  Option "PreferredMode" "2560x1440"
EndSection

Section "Screen"
  Identifier "Screen0"
  Monitor    "Monitor0"
  Device     "NVIDIA Card"
  GPUDevice  "NVIDIA Card"
  SubSection "Display"
    Depth 24 
    Modes "2560x1440"
  EndSubSection
EndSection
and below is the contents of the Xorg.0.log
Code: 
[   361.525] 
X.Org X Server 1.21.1.18
X Protocol Version 11, Revision 0
[   361.526] Current Operating System: FreeBSD saturn 14.3-RELEASE-p5 FreeBSD 14.3-RELEASE-p5 releng/14.3-n271450-382f54740ee4 GENERIC amd64
[   361.526]  
[   361.526] Current version of pixman: 0.46.2
[   361.526]     Before reporting problems, check http://wiki.x.org
    to make sure that you have the latest version.
[   361.526] Markers: (--) probed, (**) from config file, (==) default setting,
    (++) from command line, (!!) notice, (II) informational,
    (WW) warning, (EE) error, (NI) not implemented, (??) unknown.
[   361.526] (==) Log file: "/var/log/Xorg.0.log", Time: Fri Nov 21 18:13:19 2025
[   361.528] (==) Using config file: "/etc/X11/xorg.conf"
[   361.528] (==) Using config directory: "/usr/local/etc/X11/xorg.conf.d"
[   361.528] (==) Using system config directory "/usr/local/share/X11/xorg.conf.d"
[   361.529] (==) No Layout section.  Using the first Screen section.
[   361.529] (**) |-->Screen "Screen0" (0)
[   361.529] (**) |   |-->Monitor "Monitor0"
[   361.530] (**) |   |-->Device "NVIDIA Card"
[   361.530] (**) |   |-->GPUDevice "NVIDIA Card"
[   361.530] (**) Allowing byte-swapped clients
[   361.530] (==) Automatically adding devices
[   361.530] (==) Automatically enabling devices
[   361.531] (==) Automatically adding GPU devices
[   361.531] (==) Automatically binding GPU devices
[   361.531] (==) Max clients allowed: 256, resource mask: 0x1fffff
[   361.535] (==) FontPath set to:
    /usr/local/share/fonts/misc/,
    /usr/local/share/fonts/TTF/,
    /usr/local/share/fonts/OTF/,
    /usr/local/share/fonts/Type1/,
    /usr/local/share/fonts/100dpi/,
    /usr/local/share/fonts/75dpi/,
    catalogue:/usr/local/etc/X11/fontpath.d
[   361.535] (==) ModulePath set to "/usr/local/lib/xorg/modules"
[   361.535] (II) The server relies on udev to provide the list of input devices.
    If no devices become available, reconfigure udev or disable AutoAddDevices.
[   361.535] (II) Module ABI versions:
[   361.535]     X.Org ANSI C Emulation: 0.4
[   361.535]     X.Org Video Driver: 25.2
[   361.535]     X.Org XInput driver : 24.4
[   361.535]     X.Org Server Extension : 10.0
[   361.545] (II) xfree86: Adding drm device (/dev/dri/card0)
[   361.545] (II) Platform probe for /dev/dri/card0
[   361.545] (**) OutputClass "nvidia" ModulePath extended to "/usr/local/lib/nvidia/xorg,/usr/local/lib/xorg/modules,/usr/local/lib/xorg/modules"
[   361.545] (**) OutputClass "nvidia" setting /dev/dri/card0 as PrimaryGPU
[   361.545] (--) PCI: (0@0:2:0) 8086:4682:1043:8694 rev 12, Mem @ 0x6404000000/16777216, 0x4000000000/268435456, I/O @ 0x00005000/64, BIOS @ 0x????????/65536
[   361.545] (--) PCI:*(1@0:0:0) 10de:2d04:1569:f328 rev 161, Mem @ 0x84000000/67108864, 0x6000000000/17179869184, 0x6400000000/33554432, I/O @ 0x00004000/128, BIOS @ 0x????????/65536
[   361.545] (II) LoadModule: "glx"
[   361.546] (II) Loading /usr/local/lib/xorg/modules/extensions/libglx.so
[   361.549] (II) Module glx: vendor="X.Org Foundation"
[   361.549]     compiled for 1.21.1.18, module version = 1.0.0
[   361.549]     ABI class: X.Org Server Extension, version 10.0
[   361.549] (II) LoadModule: "nvidia"
[   361.549] (II) Loading /usr/local/lib/xorg/modules/drivers/nvidia_drv.so
[   361.552] (II) Module nvidia: vendor="NVIDIA Corporation"
[   361.552]     compiled for 1.6.99.901, module version = 1.0.0
[   361.552]     Module class: X.Org Video Driver
[   361.552] (II) NVIDIA dlloader X Driver  580.105.08  Wed Oct 29 22:05:05 UTC 2025
[   361.552] (II) NVIDIA Unified Driver for all Supported NVIDIA GPUs
[   361.553] (--) Using syscons driver with X support (version 2.0)
[   361.553] (--) using VT number 4

[   361.553] (II) Loading sub module "fb"
[   361.553] (II) LoadModule: "fb"
[   361.553] (II) Module "fb" already built-in
[   361.553] (II) Loading sub module "wfb"
[   361.553] (II) LoadModule: "wfb"
[   361.554] (II) Loading /usr/local/lib/xorg/modules/libwfb.so
[   361.554] (II) Module wfb: vendor="X.Org Foundation"
[   361.554]     compiled for 1.21.1.18, module version = 1.0.0
[   361.554]     ABI class: X.Org ANSI C Emulation, version 0.4
[   361.555] (WW) VGA arbiter: cannot open kernel arbiter, no multi-card support
[   361.555] (==) NVIDIA(0): Depth 24, (==) framebuffer bpp 32
[   361.555] (==) NVIDIA(0): RGB weight 888
[   361.555] (==) NVIDIA(0): Default visual is TrueColor
[   361.555] (==) NVIDIA(0): Using gamma correction (1.0, 1.0, 1.0)
[   361.555] (II) Applying OutputClass "nvidia" options to /dev/dri/card0
[   361.555] (**) NVIDIA(0): Option "AllowEmptyInitialConfiguration" "true"
[   361.555] (**) NVIDIA(0): Enabling 2D acceleration
[   361.555] (II) Loading sub module "glxserver_nvidia"
[   361.555] (II) LoadModule: "glxserver_nvidia"
[   361.555] (II) Loading /usr/local/lib/xorg/modules/extensions/libglxserver_nvidia.so
[   361.730] (II) Module glxserver_nvidia: vendor="NVIDIA Corporation"
[   361.730]     compiled for 1.6.99.901, module version = 1.0.0
[   361.730]     Module class: X.Org Server Extension
[   361.730] (II) NVIDIA GLX Module  580.105.08  Wed Oct 29 22:09:45 UTC 2025
[   361.731] (II) NVIDIA: The X server supports PRIME Render Offload.
 
SirDice said:
You don't need to install graphics/nvidia-drm-61-kmod. You don't appear to have enabled it either.
Click to expand...
Thank you for your reply. I removed nvidia-drm-* but the issue remains. I am using 14.3-RELEASE-p5 generic kernel. If you could kindly help have a look, that would be great.

/boot/loader.conf
Code: 
geom_mirror_load="YES"
kern.geom.label.disk_ident.enable="0"
kern.geom.label.gptid.enable="0"
cryptodev_load="YES"
zfs_load="YES"
snd_uaudio_load="YES"
if_re_load="YES"
if_re_name="/boot/modules/if_re.ko"
tmpfs_load="YES"
hw.nvidiadrm.modset=1
vboxdrv_load="YES"
hw.re.max_rx_mbuf_sz="2048"
if_iwm_load="NO"
kern.msgbufsize=1146880
devmatch_blocklist="if_iwm"
coretemp_load="YES"
hw.usb.no_suspend=1

/etc/rc.conf
Code: 
kld_list="cuse nvidia-modeset nvidia"
devmatch_blocklist="if_iwlwifi ng_ubt"
ifconfig_re0="inet 10.0.0.99 netmask 255.255.255.0"
defaultrouter="10.0.0.1"
ip6addrctl_enable="NO"
route6d_enable="NO"
rc_startmsgs="NO"
sshd_enable="YES"
ntpd_enable="YES"
ntpd_sync_on_start="YES"
ntpd_oomprotect="YES"
moused_nondefault_enable="NO"
dumpdev="AUTO"
zfs_enable="YES"
devfs_system_ruleset="system"
unbound_enable="YES"
linux_enable="YES"
linux_mounts_enable="NO"
dbus_enable="YES"
webcamd_enable="YES"
cupsd_enable="YES"
sendmail_enable="NONE"
smtpd_enable="YES"
var_run_enable="YES"
syslog_ng_enable="YES"
syslogd_enable="NO"
virecover_enable="NO"
powerd_enable="YES"
vm_enable="YES"
vm_dir="zfs:zroot/vm"
redis_enable="YES"
zed_enable="YES"
daily_scrub_zfs_enable="YES"
daily_scrub_zfs_default_threshold="30"
powerd_flags="-a hiadaptive -b adaptive -n adaptive"
seatd_enable="YES"

Code: 
# pkg info | grep -i nvidia
libva-nvidia-driver-0.0.14_2   NVDEC-based backend for VAAPI
nvidia-driver-580.105.08       NVidia graphics card binary drivers for hardware OpenGL rendering
nvidia-kmod-580.105.08.1403000 kmod part of NVidia graphics card binary drivers for hardware OpenGL rendering
nvidia-settings-580.105.08     Display Control Panel for X NVidia driver
nvidia-xconfig-580.105.08      Tool to manipulate X configuration files for the NVidia driver
py311-pynvml-12.0.0            Python Bindings for the NVIDIA Management Library

Code: 
# kldstat
Id Refs Address                Size Name
 1  117 0xffffffff80200000  1f41500 kernel
 2    1 0xffffffff82143000    1b580 snd_uaudio.ko
 3    1 0xffffffff8215f000     36c8 coretemp.ko
 4    1 0xffffffff82163000     7808 cryptodev.ko
 5    1 0xffffffff8216b000    8f898 vboxdrv.ko
 7    1 0xffffffff82211000   11d718 if_re.ko
 8    1 0xffffffff8232f000   5e9328 zfs.ko
 9    1 0xffffffff82919000    21350 geom_mirror.ko
10    1 0xffffffff83510000     b0b0 tmpfs.ko
11    1 0xffffffff8351c000     6710 cuse.ko
12    1 0xffffffff83523000   1650f8 nvidia-modeset.ko
13    1 0xffffffff83800000  605d178 nvidia.ko
14    2 0xffffffff83689000    30b08 linux.ko
15    6 0xffffffff836ba000     c2a8 linux_common.ko
16    1 0xffffffff836c7000     3390 acpi_wmi.ko
17    1 0xffffffff836cb000     4250 ichsmb.ko
18    1 0xffffffff836d0000     2178 smbus.ko
19    1 0xffffffff836d3000     6020 ig4.ko
20    1 0xffffffff836da000    2de48 linux64.ko
21    1 0xffffffff83708000     2278 pty.ko
22    1 0xffffffff8370b000     3560 fdescfs.ko
23    1 0xffffffff8370f000     73c0 linprocfs.ko
24    1 0xffffffff83717000     440c linsysfs.ko
25    1 0xffffffff8371c000     3360 uhid.ko
26    1 0xffffffff83720000     4364 ums.ko
27    1 0xffffffff83725000     3360 wmt.ko
28    1 0xffffffff89a00000   33e438 vmm.ko
29    1 0xffffffff83729000     21dc nmdm.ko
30    1 0xffffffff8372c000     7798 if_bridge.ko
31    1 0xffffffff83734000     60e0 bridgestp.ko
32    1 0xffffffff8373b000     4850 nullfs.ko
33    1 0xffffffff83740000     2a80 mac_ntpd.ko

This is the latest Xorg.0.log (based on the above configuration)
Code: 
[    72.792] 
X.Org X Server 1.21.1.18
X Protocol Version 11, Revision 0
[    72.793] Current Operating System: FreeBSD saturn 14.3-RELEASE-p5 FreeBSD 14.3-RELEASE-p5 releng/14.3-n271450-382f54740ee4 GENERIC amd64
[    72.793]  
[    72.793] Current version of pixman: 0.46.2
[    72.793]     Before reporting problems, check http://wiki.x.org
    to make sure that you have the latest version.
[    72.793] Markers: (--) probed, (**) from config file, (==) default setting,
    (++) from command line, (!!) notice, (II) informational,
    (WW) warning, (EE) error, (NI) not implemented, (??) unknown.
[    72.793] (==) Log file: "/var/log/Xorg.0.log", Time: Fri Nov 21 20:28:12 2025
[    72.796] (==) Using config directory: "/usr/local/etc/X11/xorg.conf.d"
[    72.797] (==) Using system config directory "/usr/local/share/X11/xorg.conf.d"
[    72.798] (==) No Layout section.  Using the first Screen section.
[    72.798] (**) |-->Screen "Screen0" (0)
[    72.798] (**) |   |-->Monitor "Monitor0"
[    72.800] (**) |   |-->Device "NVIDIA Card"
[    72.800] (**) |   |-->GPUDevice "NVIDIA Card"
[    72.800] (**) Allowing byte-swapped clients
[    72.800] (==) Automatically adding devices
[    72.800] (==) Automatically enabling devices
[    72.800] (==) Automatically adding GPU devices
[    72.800] (==) Automatically binding GPU devices
[    72.801] (==) Max clients allowed: 256, resource mask: 0x1fffff
[    72.806] (==) FontPath set to:
    /usr/local/share/fonts/misc/,
    /usr/local/share/fonts/TTF/,
    /usr/local/share/fonts/OTF/,
    /usr/local/share/fonts/Type1/,
    /usr/local/share/fonts/100dpi/,
    /usr/local/share/fonts/75dpi/,
    catalogue:/usr/local/etc/X11/fontpath.d
[    72.806] (==) ModulePath set to "/usr/local/lib/xorg/modules"
[    72.806] (II) The server relies on udev to provide the list of input devices.
    If no devices become available, reconfigure udev or disable AutoAddDevices.
[    72.807] (II) Module ABI versions:
[    72.807]     X.Org ANSI C Emulation: 0.4
[    72.807]     X.Org Video Driver: 25.2
[    72.807]     X.Org XInput driver : 24.4
[    72.807]     X.Org Server Extension : 10.0
[    72.821] (!!) More than one possible primary device found
[    72.821] (--) PCI: (0@0:2:0) 8086:4682:1043:8694 rev 12, Mem @ 0x6404000000/16777216, 0x4000000000/268435456, I/O @ 0x00005000/64, BIOS @ 0x????????/65536
[    72.821] (--) PCI: (1@0:0:0) 10de:2d04:1569:f328 rev 161, Mem @ 0x84000000/67108864, 0x6000000000/17179869184, 0x6400000000/33554432, I/O @ 0x00004000/128, BIOS @ 0x????????/65536
[    72.821] (II) LoadModule: "glx"
[    72.821] (II) Loading /usr/local/lib/xorg/modules/extensions/libglx.so
[    72.825] (II) Module glx: vendor="X.Org Foundation"
[    72.825]     compiled for 1.21.1.18, module version = 1.0.0
[    72.825]     ABI class: X.Org Server Extension, version 10.0
[    72.825] (II) LoadModule: "nvidia"
[    72.825] (II) Loading /usr/local/lib/xorg/modules/drivers/nvidia_drv.so
[    72.827] (II) Module nvidia: vendor="NVIDIA Corporation"
[    72.827]     compiled for 1.6.99.901, module version = 1.0.0
[    72.827]     Module class: X.Org Video Driver
[    72.828] (II) NVIDIA dlloader X Driver  580.105.08  Wed Oct 29 22:05:05 UTC 2025
[    72.828] (II) NVIDIA Unified Driver for all Supported NVIDIA GPUs
[    72.828] (--) Using syscons driver with X support (version 2.0)
[    72.828] (--) using VT number 4

[    72.829] (II) Loading sub module "fb"
[    72.829] (II) LoadModule: "fb"
[    72.829] (II) Module "fb" already built-in
[    72.829] (II) Loading sub module "wfb"
[    72.829] (II) LoadModule: "wfb"
[    72.829] (II) Loading /usr/local/lib/xorg/modules/libwfb.so
[    72.830] (II) Module wfb: vendor="X.Org Foundation"
[    72.830]     compiled for 1.21.1.18, module version = 1.0.0
[    72.830]     ABI class: X.Org ANSI C Emulation, version 0.4
[    72.830] (WW) VGA arbiter: cannot open kernel arbiter, no multi-card support
[    72.830] (==) NVIDIA(0): Depth 24, (==) framebuffer bpp 32
[    72.830] (==) NVIDIA(0): RGB weight 888
[    72.830] (==) NVIDIA(0): Default visual is TrueColor
[    72.830] (==) NVIDIA(0): Using gamma correction (1.0, 1.0, 1.0)
[    72.830] (**) NVIDIA(0): Option "AllowEmptyInitialConfiguration" "true"
[    72.830] (**) NVIDIA(0): Enabling 2D acceleration
[    72.830] (II) Loading sub module "glxserver_nvidia"
[    72.830] (II) LoadModule: "glxserver_nvidia"
[    72.830] (II) Loading /usr/local/lib/xorg/modules/extensions/libglxserver_nvidia.so
[    72.938] (II) Module glxserver_nvidia: vendor="NVIDIA Corporation"
[    72.938]     compiled for 1.6.99.901, module version = 1.0.0
[    72.938]     Module class: X.Org Server Extension
[    72.938] (II) NVIDIA GLX Module  580.105.08  Wed Oct 29 22:09:45 UTC 2025
[    72.939] (II) NVIDIA: The X server supports PRIME Render Offload.
 
Nothing out of the ordinary here. Except missing the monitor identification.

bumpyhead said:
When I try to start X, the system gives a black screen with a tiny white rectangle on the top-left corner, and completely freezes.
Click to expand...
Might be a silly question, but did you actually hook up the monitor to the graphics card's output? Because it sounds like it's still connected to the onboard graphics.
 
SirDice said:
Might be a silly question, but did you actually hook up the monitor to the graphics card's output? Because it sounds like it's still connected to the onboard graphics.
Click to expand...
Yes, I did! the monitor is hooked into the graphic card's output. Neither HDMI or DP works. I also tried single/dual monitors. None of these works.

When startx, it just shows a black screen with a keyboard cursor on the top-left, and have to hard reboot.
 
bumpyhead said:
Yes, I did! the monitor is hooked into the graphic card's output. Neither HDMI or DP works.
Click to expand...
Odd, because your Xorg log doesn't seem to detect anything. It should look something like this (this is an old system, with an old NVidia driver, but the output will look similar):
Code: 
[    29.474] (II) NVIDIA(0): Creating default Display subsection in Screen section
        "Default Screen Section" for depth/fbbpp 24/32
[    29.474] (==) NVIDIA(0): Depth 24, (==) framebuffer bpp 32
[    29.474] (==) NVIDIA(0): RGB weight 888
[    29.474] (==) NVIDIA(0): Default visual is TrueColor
[    29.474] (==) NVIDIA(0): Using gamma correction (1.0, 1.0, 1.0)
[    29.475] (**) NVIDIA(0): Enabling 2D acceleration
[    30.820] (--) NVIDIA(0): Valid display device(s) on GPU-0 at PCI:2:0:0
[    30.820] (--) NVIDIA(0):     CRT-0
[    30.820] (--) NVIDIA(0):     DFP-0
[    30.821] (--) NVIDIA(0):     DFP-1 (boot)
[    30.825] (II) NVIDIA(0): NVIDIA GPU GeForce GT 520M (GF119) at PCI:2:0:0 (GPU-0)
[    30.825] (--) NVIDIA(0): Memory: 524288 kBytes
[    30.825] (--) NVIDIA(0): VideoBIOS: 75.19.0a.00.09
[    30.825] (II) NVIDIA(0): Detected PCI Express Link width: 16X
[    30.841] (--) NVIDIA(GPU-0): CRT-0: disconnected
[    30.841] (--) NVIDIA(GPU-0): CRT-0: 400.0 MHz maximum pixel clock
[    30.841] (--) NVIDIA(GPU-0):
[    30.842] (--) NVIDIA(GPU-0): DFP-0: disconnected
[    30.842] (--) NVIDIA(GPU-0): DFP-0: Internal TMDS
[    30.842] (--) NVIDIA(GPU-0): DFP-0: 165.0 MHz maximum pixel clock
[    30.842] (--) NVIDIA(GPU-0):
[    30.873] (--) NVIDIA(GPU-0): LG Electronics LG FULL HD (DFP-1): connected
[    30.874] (--) NVIDIA(GPU-0): LG Electronics LG FULL HD (DFP-1): Internal TMDS
[    30.874] (--) NVIDIA(GPU-0): LG Electronics LG FULL HD (DFP-1): 330.0 MHz maximum pixel clock
[    30.874] (--) NVIDIA(GPU-0):


bumpyhead said:
and have to hard reboot.
Click to expand...
Do you have another system? Laptop perhaps? You can use it to remotely login and shutdown/restart the machine cleanly. Or check your BIOS/UEFI settings regarding the power button, make sure it's set to "delayed" instead of "instant" off. Then you can simply tap the power button and the system will cleanly shutdown.
 
What happenes if you add hw.nvidia.registry.EnableGpuFirmware=1 in your /boot/loader.conf and restart your computer?

At least some of RTX5xxx GPUs are known not to work without GSP enabled and the above is the configuration to forcibly enable it.

This is documented in pkg-message.in of x11/nvidia-driver which should be displayed when it is installed.

Why disabled by default is because enabling GSP is reported to results in failure on suspend/resume, thus, disbled by default and should work fine for pre-RTX5xxx GPUs.

And why needed to set in /boot/loader.conf is because it MUST be already set when driver kmods are loaded. Setting in /etc/sysctl.conf could result in race conditions.
 
T-Aoki said:
What happenes if you add hw.nvidia.registry.EnableGpuFirmware=1 in your /boot/loader.conf and restart your computer?
Click to expand...
Yes, it works! I put these two lines in /boot/loader.conf
Code: 
hw.nvidia.registry.EnableGpuFirmware=1
hw.nvidiadrm.modset=1
Now I can startx and use my dual monitor set-up. I have marked the problem as "solved".

It is weird, however, that the second command seems useless.
Code: 
# sysctl -a | grep -i hw.nvidia | grep modeset
hw.nvidiadrm.modeset: 0

Remaining issue for me is hyprland stopped work (seems driver lacks KMS support). Below are log-files. Hope I could figure it out on my way or upgrading to 15.0 would help.
Code: 
[LOG] [AQ] drm: Enumerated device /dev/dri/card0
[LOG] [AQ] libseat: Device /dev/dri/card0 does not support kms
[ERR] [AQ] drm: Skipping device /dev/dri/card0, not a KMS device
[ERR] [AQ] drm: Found no gpus to use, cannot continue
[ERR] [AQ] DRM Backend failed
[LOG] [AQ] Starting the Aquamarine backend!
[LOG] [AQ] Starting the Wayland backend!
[ERR] [AQ] Wayland backend cannot start: wl_display_connect failed (is a wayland compositor running?)
[ERR] [AQ] Requested backend (wayland) could not start, enabling fallbacks
[ERR] [AQ] Implementation wayland failed, erasing.
[CRITICAL] [AQ] Cannot open backend: no allocator available
[CRITICAL] m_pAqBackend couldn't start! This usually means aquamarine could not find a GPU or encountered some issues. Make sure you're  running either on a tty or on a Wayland session, NOT an X11 one.
[CRITICAL] Critical error thrown: CBackend::create() failed!
 
bumpyhead said:
It is weird, however, that the second command seems useless.
# sysctl -a | grep -i hw.nvidia | grep modeset hw.nvidiadrm.modeset: 0
Click to expand...
The second one is only required when you need graphics/nvidia-drm-*-kmod[-devel].

bumpyhead said:
Remaining issue for me is hyprland stopped work (seems driver lacks KMS support).
Click to expand...
Any of graphics/nvidia-drm-*-kmod[-devel] and corresponding graphics/drm-*-kmod are needed to expose DRM/KMS support.
Install graphics/nvidia-drm-61-kmod (66 and latest doesn't work on 14.3 because 14.3 lacks some LinuxKPI functionalities they need) again.
It would pull in corresponding graphics/drm-61-kmod as its runtime dependency.

You also need to switch "nvidia-modeset nvidia" in your kld_list to nvidia-drm.
nvidia-drm would auto-load required kmods. nvidia-modeset would be auto-loaded, and nvidia-modeset auto-loads nvidia.

And you'll no longer need xorg.conf stuff, as modesetting driver should auto-configure it (this is why graphics/nvidia-drm-*-kmod[-devel] require corresponding graphics/drm-*-kmod to build / run). Remove (delete or rename not to be picked) your xorg configs you "manually" created if it hurts starting up xorg.
 
T-Aoki said:
You also need to switch "nvidia-modeset nvidia" in your kld_list to nvidia-drm.
nvidia-drm would auto-load required kmods. nvidia-modeset would be auto-loaded, and nvidia-modeset auto-loads nvidia.
Click to expand...
It has been a few days. The system is usable, but not stable.

Every time I boot the machine (with nvida-modeset & nvidia loaded), startx, then the system halts and I need to hard reset.

After the reset, I can boot into the machine, and startx normally. Even if I quit X, and restart X multiple times, it is fine. It seems the system/driver feels refreshed after a hard-reset.

Then, if I reboot normally, and startx will crash again.

I tried to enable/disable nvidia-drm, and it won't help. I also tried to load nvida-modeset through /boot/loader.conf, then the kernel just panic at startup.

Well, I am accepting it as of now, because I don't need to reboot a lot. Hopefully, upgrading to 15.0 would fix this.
 
You must log in or register to reply here.
Back
Top