I wish small stuff like that would be sorted out long before release so it is all nice and well, but I run into problems like these often too. Also I have on several occasions landed back on L1 as a source for a solution as well. You get bonus points if you land in a thread you already answered in the past 
.
Or - in fact - created ā¦ 
2 Likes
Iām experimenting with running Nvidia and nouveau at the same time, for different cards. In my case, itās a 2080ti and a 1070 in bazzite.
If I run modprobe nouveau manually, everything basically works. I havenāt tested this too extensively because Iām not about doing things manually.
If I create a systemd service (after graphical.target, wantedby default.target) the desktop is very unstable and crashes from kde animations.
If I run an @reboot crontab, the nouveau card doesnāt always show the desktop correctly after waking from sleep. Logging out / in corrects the issue.
A bazzite dev I was talking to in discord explained that this cannot not be done as the drivers will conflict. Totally reasonable to not officially support.
That being said. . . anyone else experiment with this? Is there a less sketchy way to use two Nvidia graphics cards at once like a ānormalā desktop? Besides nouveau across the board preferably.
Everyone says this and their source is nearly always ātrust me, bro.ā They told me the same thing with an AMD dGPU + nVidia dGPU.
If manually running modprobe works then perhaps the systemd unit needs to start later ā Iām guessing after Plasma/Gnome/whatever has fully initialized. I reckon the problem is graphical.target is just the logon manager (sddm?) and not the full desktop (which isnāt handling twinVidia properly at startup).
Perhaps something like:
[Unit]
Description=Secondary GPU Module Loader
After=graphical.target
[Service]
Type=oneshot
RemainAfterExit=yes
TimeoutStartSec=10
ExecCondition=/bin/test -d /run/user/*
ExecStart=/sbin/modprobe <gpu module name goes here>
[Install]
WantedBy=multi-user.target
The idea is a folder appears under /run/user with the userās <uid> when they logon. At that point test -d /run/user/* becomes true, the service waits for TimeoutStartSec seconds then modprobes.
Note Iām dumb, Iām not an expert, and I havenāt tested this. At all. It might not work. It probably wonāt work. But maybe itās a nudge in the right direction.
As for waking from sleep thereās a number of possibilities for this but the brute force approach might be easiest. Something like automatically rmmod the module prior to system sleep.
1 Like
Being able to modprobe later in the boot process was exactly what I was looking for, thanks!
1 Like
In BTOP
OPTIONS for cpu, change the sensor input to TSIO or k10temp sensor from default. I see my cpu temp in the upper right of the cpu panel right underneath the clock frequency of the cpu.
1 Like
I swear, all of my Linux problems have to do with sleep. To be fair, my Windows problems did as wellā¦
Anyway, Iām hoping for some pointers in what direction to go research / troubleshoot, as Iām kinda out of ideas.
Problem 1:
On my desktop I have issues with graphical display glitches on my monitor after waking from sleep, but only if I have my TV plugged in (even if it isnāt enabled). Admittedly, TV is an HDMI 2.1 connection on an AMD graphics card (via one of the few active adapters that mostly-work), so Iām already in a weird state, but I just want to be able to sleep and wake my desktop without having to deal with display glitches from having a device plugged in Iām not even using half the time.
The display issues on my monitor (1440p/165 Hz Freesync via Displayport) are little hiccups that display a horizontal line like if there was an extreme screen tearing issue. My best guess is that freesync is glitching? They happen around a few times every five or so minutes, but seem partially dependent on what is displayed on screen.
The TV also has (different) display issues after resuming from sleep, but I know what are causing those and I have workarounds to them already. Plus Iām using a jank setup and expect jank in return. If the TV isnāt plugged in when the desktop goes to sleep, zero issues with the monitor.
R9 5900X, 7900XTX, EndeavourOS (KDE/Wayland)
Problem 2:
My laptop does not like waking from sleep in Linux. I swear this worked once upon a time and broke after a distro swap, but even reverting to my original distro it still doesnāt work. Which makes me think this is either driver-related issue, a firmware-related issue, or a āaetherspoonās memory doesnāt work rightā issue.
Basically, if the laptop sleeps for any reason (lid close, command to sleep, etc), it will hang on waking up. On EndeavourOS I get a blinking cursor in the top left and the keyboard initially responds, then stops responding. On other distros, the visual symptoms change (sometimes it will load some graphics and just hang at that point) but the end result (keyboard initially works then stops working, mouse/trackpad never works, machine locks up) remains.
Lenovo Legion 5 Slim, R7 7840HS, RTX 4070, EndeavourOS (KDE/Wayland) but also other distros that are still using X.
Lately Iāve been experiencing really slow boots. If I hit escape during bootup I see boot hanging with slow output of several lines like usb 3-7: device descriptor read/64, error -110 before I think the kernel gives up and moves on. dmesg shows me this:
dmesg
[ 23.291836] usb 3-7: device descriptor read/64, error -110
[ 23.573800] usb 3-7: new high-speed USB device number 5 using xhci_hcd
[ 28.922841] usb 3-7: device descriptor read/64, error -110
[ 44.794957] usb 3-7: device descriptor read/64, error -110
[ 44.904248] usb usb3-port7: attempt power cycle
[ 45.344780] usb 3-7: new high-speed USB device number 6 using xhci_hcd
[ 50.144773] usb 3-7: Device not responding to setup address.
[ 55.149884] usb 3-7: Device not responding to setup address.
[ 55.357747] usb 3-7: device not accepting address 6, error -71
[ 55.736773] usb 3-7: new high-speed USB device number 7 using xhci_hcd
[ 60.537319] usb 3-7: Device not responding to setup address.
[ 65.542422] usb 3-7: Device not responding to setup address.
[ 65.749773] usb 3-7: device not accepting address 7, error -71
[ 65.752559] usb usb3-port7: unable to enumerate USB device
How do I identify which device corresponds to usb 3-7? I thought lsusb -t would tell me, but it doesnāt show anything useful:
lsusb
/: Bus 001.Port 001: Dev 001, Class=root_hub, Driver=xhci_hcd/12p, 480M
|__ Port 001: Dev 002, If 0, Class=Human Interface Device, Driver=usbhid, 12M
|__ Port 002: Dev 003, If 0, Class=Hub, Driver=hub/4p, 480M
|__ Port 003: Dev 005, If 0, Class=Human Interface Device, Driver=usbhid, 12M
|__ Port 003: Dev 004, If 0, Class=Hub, Driver=hub/4p, 480M
|__ Port 001: Dev 006, If 0, Class=Hub, Driver=hub/4p, 480M
|__ Port 004: Dev 007, If 0, Class=Human Interface Device, Driver=usbhid, 12M
|__ Port 004: Dev 007, If 1, Class=Human Interface Device, Driver=usbhid, 12M
|__ Port 004: Dev 007, If 2, Class=Human Interface Device, Driver=usbhid, 12M
/: Bus 002.Port 001: Dev 001, Class=root_hub, Driver=xhci_hcd/6p, 10000M
|__ Port 003: Dev 002, If 0, Class=Hub, Driver=hub/4p, 5000M
|__ Port 001: Dev 003, If 0, Class=Hub, Driver=hub/4p, 5000M
/: Bus 003.Port 001: Dev 001, Class=root_hub, Driver=xhci_hcd/12p, 480M
|__ Port 005: Dev 002, If 0, Class=Human Interface Device, Driver=usbhid, 12M
|__ Port 005: Dev 002, If 1, Class=Human Interface Device, Driver=usbhid, 12M
|__ Port 006: Dev 003, If 0, Class=Audio, Driver=snd-usb-audio, 480M
|__ Port 006: Dev 003, If 1, Class=Audio, Driver=snd-usb-audio, 480M
|__ Port 006: Dev 003, If 2, Class=Audio, Driver=snd-usb-audio, 480M
|__ Port 006: Dev 003, If 3, Class=Vendor Specific Class, Driver=[none], 480M
|__ Port 008: Dev 008, If 0, Class=Audio, Driver=snd-usb-audio, 480M
|__ Port 008: Dev 008, If 1, Class=Audio, Driver=snd-usb-audio, 480M
|__ Port 008: Dev 008, If 2, Class=Audio, Driver=snd-usb-audio, 480M
|__ Port 008: Dev 008, If 3, Class=Audio, Driver=snd-usb-audio, 480M
|__ Port 008: Dev 008, If 4, Class=Audio, Driver=snd-usb-audio, 480M
|__ Port 008: Dev 008, If 5, Class=Audio, Driver=snd-usb-audio, 480M
|__ Port 008: Dev 008, If 6, Class=Human Interface Device, Driver=usbhid, 480M
|__ Port 009: Dev 009, If 0, Class=Video, Driver=uvcvideo, 480M
|__ Port 009: Dev 009, If 1, Class=Video, Driver=uvcvideo, 480M
|__ Port 009: Dev 009, If 2, Class=Audio, Driver=snd-usb-audio, 480M
|__ Port 009: Dev 009, If 3, Class=Audio, Driver=snd-usb-audio, 480M
/: Bus 004.Port 001: Dev 001, Class=root_hub, Driver=xhci_hcd/5p, 20000M/x2
/: Bus 005.Port 001: Dev 001, Class=root_hub, Driver=xhci_hcd/2p, 480M
/: Bus 006.Port 001: Dev 001, Class=root_hub, Driver=xhci_hcd/2p, 20000M/x2
/: Bus 007.Port 001: Dev 001, Class=root_hub, Driver=xhci_hcd/2p, 480M
/: Bus 008.Port 001: Dev 001, Class=root_hub, Driver=xhci_hcd/2p, 10000M
/: Bus 009.Port 001: Dev 001, Class=root_hub, Driver=xhci_hcd/2p, 480M
/: Bus 010.Port 001: Dev 001, Class=root_hub, Driver=xhci_hcd/2p, 10000M
/: Bus 011.Port 001: Dev 001, Class=root_hub, Driver=xhci_hcd/1p, 480M
|__ Port 001: Dev 002, If 0, Class=Human Interface Device, Driver=usbhid, 12M
/: Bus 012.Port 001: Dev 001, Class=root_hub, Driver=xhci_hcd/0p, 5000M
Iāve tried unplugging everything connected to USB ports, but that doesnāt have any effect.
System
System:
Kernel: 6.15.2-300.vanilla.fc42.x86_64 arch: x86_64 bits: 64
Desktop: GNOME v: 48.2 Distro: Fedora Linux 42 (Workstation Edition)
Machine:
Type: Desktop System: ASRock product: X870E Taichi Lite v: N/A
serial: <superuser required>
Mobo: ASRock model: X870E Taichi Lite serial: <superuser required>
UEFI: American Megatrends LLC. v: 3.25 date: 05/14/2025
CPU:
Info: 8-core model: AMD Ryzen 7 9800X3D bits: 64 type: MT MCP cache:
L2: 8 MiB
Speed (MHz): avg: 3594 min/max: 603/5272 cores: 1: 3594 2: 3594 3: 3594
4: 3594 5: 3594 6: 3594 7: 3594 8: 3594 9: 3594 10: 3594 11: 3594 12: 3594
13: 3594 14: 3594 15: 3594 16: 3594
Graphics:
Device-1: Advanced Micro Devices [AMD/ATI] Navi 31 [Radeon RX 7900 XT/7900
XTX/7900 GRE/7900M] driver: amdgpu v: kernel
Device-2: Advanced Micro Devices [AMD/ATI] Granite Ridge [Radeon Graphics]
driver: amdgpu v: kernel
Device-3: Realtek Streamplify CAM driver: snd-usb-audio,uvcvideo type: USB
Display: wayland server: X.Org v: 24.1.6 with: Xwayland v: 24.1.6
compositor: gnome-shell driver: dri: radeonsi gpu: amdgpu resolution:
1: 2560x1440~120Hz 2: 2560x1440~60Hz
API: OpenGL v: 4.6 vendor: amd mesa v: 25.0.7 renderer: AMD Radeon RX
7900 XT (radeonsi navi31 LLVM 20.1.5 DRM 3.63
6.15.2-300.vanilla.fc42.x86_64)
API: EGL Message: EGL data requires eglinfo. Check --recommends.
Info: Tools: api: glxinfo x11: xdriinfo, xdpyinfo, xprop, xrandr
Audio:
Device-1: Advanced Micro Devices [AMD/ATI] Navi 31 HDMI/DP Audio
driver: snd_hda_intel
Device-2: Advanced Micro Devices [AMD/ATI] Radeon High Definition Audio
[Rembrandt/Strix] driver: snd_hda_intel
Device-3: Advanced Micro Devices [AMD] Family 17h/19h/1ah HD Audio
driver: N/A
Device-4: Focusrite-Novation Scarlett Solo 4th Gen driver: snd-usb-audio
type: USB
Device-5: Generic USB Audio driver: hid-generic,snd-usb-audio,usbhid
type: USB
Device-6: Realtek Streamplify CAM driver: snd-usb-audio,uvcvideo type: USB
API: ALSA v: k6.15.2-300.vanilla.fc42.x86_64 status: kernel-api
Server-1: PipeWire v: 1.4.5 status: active
Network:
Device-1: Realtek RTL8126 5GbE driver: r8169
IF: enp10s0 state: up speed: 1000 Mbps duplex: full mac: <filter>
Device-2: MEDIATEK MT7925 Wi-Fi 7 160MHz driver: mt7925e
IF: wlp11s0 state: down mac: <filter>
Drives:
Local Storage: total: 3.64 TiB used: 3.23 TiB (88.8%)
ID-1: /dev/nvme0n1 vendor: Samsung model: SSD 990 PRO 4TB size: 3.64 TiB
Partition:
ID-1: / size: 3.64 TiB used: 3.23 TiB (88.8%) fs: btrfs dev: /dev/dm-0
ID-2: /boot size: 973.4 MiB used: 437.3 MiB (44.9%) fs: ext4
dev: /dev/nvme0n1p2
ID-3: /boot/efi size: 598.8 MiB used: 19.3 MiB (3.2%) fs: vfat
dev: /dev/nvme0n1p1
ID-4: /home size: 3.64 TiB used: 3.23 TiB (88.8%) fs: btrfs dev: /dev/dm-0
Swap:
ID-1: swap-1 type: zram size: 8 GiB used: 0 KiB (0.0%) dev: /dev/zram0
Sensors:
System Temperatures: cpu: 39.6 C mobo: N/A
Fan Speeds (rpm): N/A
GPU: device: amdgpu temp: 37.0 C fan: 2 device: amdgpu temp: 36.0 C
Info:
Memory: total: 48 GiB note: est. available: 46.13 GiB used: 3.68 GiB (8.0%)
Processes: 479 Uptime: 11m Shell: Zsh inxi: 3.3.38
Iāve had some issues with GPU power management that may be caused by the mobo, so a part of me wonders whether this is just more of that. I donāt know how to pinpoint issues to the motherboard though.
At one point this solved sleep not waking for me (on 7640U, a rolling distro though not on an arch base), your description reads like I experienced, and applied:
https://bbs.archlinux.org/viewtopic.php?id=296954
(see post #9 for a unit file - but revert it if you donāt see any difference)
Several minor kernel versions later of 6.14, something else caused it to fail, differently (display buffer of the time the screen closed, specifically) so more likely uefi/fw packages/kernel not talking to each other.
There are some sleep info/debug tools here I was meaning to look into further -
https://pypi.org/project/amd-debug-tools/
Though honestly, I gave up on this for now and just start on an older kernel (6.12 in my case).
If itās always the same āportā 3-7 at issue you might be able to either, disable in bios or otherwise to disable it with e.g. usbguard though I donāt know that latter would occur on boot.
Whatever it is doesnāt seem to complete registration for lsusb to report (aside 1 - lsusb.py may exist in your distribution & is arguably more readable, also try lsusb -tv). The āusbā device may actually be soldered to the board or even be e.g. a shorted or bad front panel connector itself (i.e. not necessarily the back panel ports). Iād plug in something else until I found which set of physical ports from the board is on that usb2.0 part of the tree, and start to disconnect anything in it.
aside 2 - systemd-analyze blame can also help identify services delaying startup.
Hm - yeah, this definitely sounds like what mine is going through. Sadly, the initial tips didnāt work for me; Iāll poke around further with the debug tools, I really donāt want to revert kernel versions.
Background
A week ago I was seeing really poor performance in looking glass (45 FPS, linux guest GPU maxed out). Eventually narrowed it down to a driver problem that I solved by reverting to 470xx via aur (gpu is a 1070) and using X11 instead of Wayland. Reading through the aur entry it was applying some patches to the driver for the latest kernel to get it to compile.
At this point, looking glass worked perfectly but the desktop ran pretty badly. Additionally, any time I tried to launch a game through steam I would get an error about not having a D3D11 compatible card.
So I was reading the docs from the more recent 570 driver and saw something about Ubuntu 24.04 being āofficiallyā supported. Went to find out what was so special about them and learned about the work they put into their HWE kernel.
Extended hardware support sounded like exactly what I needed, so I downloaded the 6.14-hwe kernel source from archive.ubuntu.org, followed the Kernel page in the arch wiki, learned what DKMS is for, installed the 570 drivers, and now I have a better (not perfect) functioning LG, a smooth desktop AND I can launch games with the linux guest gpu. I still have to use X11 instead of Wayland, but that seems OK to me.
Question
My question is: now that I am running a manually compiled kernel, does that mean Iām committed to compiling ubuntuās HWE kernel for the rest of time? Or is there a way to start using the ubuntu binary kernel in arch and receive automatic updates?