Lost Peripherals after Recent Update

synthsncats

I've also lost my ability to use USB peripherals. I had to manually fix it through the terminal by invoking fn + control + alt/option + F7 (luckily my macbook pro's keyboard still worked). I thought I had the previous kernel (but don't), so I installed LTS kernel instead. All appears to be working well.

noise617

Tried to install LTS kernel as described here... what could be causing the error?

user@pc ~ $ sudo clr-boot-manager set-timeout 5 && sudo clr-boot-manager update
Password: 
New timeout value is: 5

user@pc ~ $ sudo eopkg install linux-lts bbswitch broadcom-sta linux-lts-headers nvidia-340-glx-driver nvidia-390-glx-driver  nvidia-glx-driver razer-drivers  v4l2loopback vhba-module 
Safety switch forces the upgrade of following packages:
python-pycurl  tzdata
System error. Program terminated.
Repo item nvidia-340-glx-driver not found
Please use 'eopkg help' for general help.
Use --debug to see a traceback.

JoshStrobl

noise617 You don't need to install everything listed. You only need to install the kernel and whatever other modules you installed separately. If you never installed any or opened up the Hardware Drivers tool, chances are you aren't using any, and thus you only need to install linux-lts. Hence why it says "so if you use these kernel modules"

synthsncats

The nvidia-340-glx-driver is no longer in the repo. Though, it should install fine with the other nvidia drivers in the link you followed.

noise617

Thank you for the answers.

So, to whom it may concern, here's a brief report. I installed linux-lts by first running this command to always display the Solus startup menu:

sudo clr-boot-manager set-timeout 5 && sudo clr-boot-manager update

Then, just using Software Center, I checked what modules listed here I have installed. It turned out, none. Also, FYI, nvidia-340-glx-driver and razer-drivers are no longer in the repository at all.

Next, I installed the LTS kernel by just running:

sudo eopkg install linux-lts

I rebooted. USB devices do work under LTS kernel, but there's still the damn delay before they become active. Around 30 seconds, when the OS is already started. Not sure what to do with this.

Anyway, if you want to make the currently booted kernel your permanent choice, here's what you need to do:

After successfully booting into a kernel from the current or lts branches running sudo clr-boot-manager update will make the booted kernel branch the default boot option going forward.

tenroc

noise617 I have a similar issue, on every distro I tried, which I partially fixed.
What does journald says? run:
journalctl -p 3 -xb
My laptop was complaining about:
usb 1-2: hub failed to enable device, error -62
Which was holding up the initialization of usb devices until it timed out.
I identified the faulty usb device with lsusb, and I disabled it at boot with a systemd service.
I haven't found a way to disable the device before systemd init though. So I still have to wait for it to time out to decrypt my drive at boot.

noise617

tenroc I cannot type journalctl -p 3 -xb when the devices actually don't work because there's no keyboard at that time. But here's what it says when the devices activate, like I said about 30 sec after start.

-- Journal begins at Wed 2021-10-13 23:39:22, ends at Tue 2021-10-26 03:54:45. --
Oct 25 12:12:50 m700 spice-vdagent[1028]: Cannot access vdagent virtio channel /dev/virtio-ports/com.redhat.spice.0
Oct 25 12:12:51 m700 pulseaudio[1126]: [pulseaudio] pid.c: Daemon already running.
Oct 25 12:13:16 m700 pulseaudio[1025]: [pulseaudio] bluez5-util.c: GetManagedObjects() failed: org.freedesktop.DBus.Error.NoRe>

Also, I once pressed Ctrl + C during disk decryption password entry, in the
initial startup process when keyboard still worked, and this error was displayed. Not sure if it is related:

tenroc

noise617 what's Bus 1 - Device 9? (lsusb will tell you).
Also, the internet tells me :

USB error -110 means "Timeout expired before the transfer completed", which could be caused by anything. Most commonly, it's because power was exceeded;
(https://stackoverflow.com/questions/13653692/device-descriptor-read-64-error-110)

You could try to unplug your power cord, shutdown your computer, let it rest for a few seconds and replug it if you still face the error. Not sure if it is related to your issue about unresponsive peripherals though.
"device not accepting address xx" seems to cause the unresponsiveness on my laptop.

alexanderzhirov

Also faced a problem.

With the latest updates, I noticed problems working with USB, and on different computers.
That is, if I connect flash drives, they are not visible in the system at all.
If I turn on the same computer from under Windows, the flash drives are recognized. Then, without pulling them out of the computer, I reboot back into Linux - the flash drives become visible.
This situation is on two different computers with Solus, with the latest updates.

Wini-Buh

Same issue here... No mouse and no keyboard (both USB devices) available to log in. I need to reboot, then activate and exit the BIOS setup (without changing something there) after that the mouse and the keyboard are working.
This issue was definitively introduced with the latest update.

Kernel version: 5.14.14-202.current

Here are some information from the journal:

Okt 26 07:46:56 winipc kernel: xhci_hcd 0000:00:14.0: Abort failed to stop command ring: -110
Okt 26 07:46:56 winipc kernel: xhci_hcd 0000:00:14.0: xHCI host controller not responding, assume dead
Okt 26 07:46:56 winipc kernel: xhci_hcd 0000:00:14.0: HC died; cleaning up
Okt 26 07:46:57 winipc kernel: usb 4-3: device not accepting address 2, error -108
Okt 26 07:46:57 winipc kernel: usb usb4-port3: couldn't allocate usb_device
Okt 26 07:46:57 winipc kernel: usb usb3-port1: couldn't allocate usb_device

alexanderzhirov

Has anyone figured out what the problem might be? I take it the problem is not in the kernel? Still USB are visible, but they do not turn on automatically.

noise617

tenroc

Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 001 Device 004: ID 045e:075d Microsoft Corp. LifeCam Cinema
Bus 001 Device 003: ID feed:3060 foostan Corne Keyboard (crkbd)
Bus 001 Device 017: ID 2717:ff40 Xiaomi Inc. Mi/Redmi series (MTP)
Bus 001 Device 002: ID 1ea7:0064 SHARKOON Technologies GmbH 2.4GHz Wireless rechargeable vertical mouse [More&Better]
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub

I cannot imagine that it is related to power: it is a stationary mini PC, I had never had any such problems before these recent updates.

By the way, I wonder if it is possible to track down the reason for the malfunction during the kernel boot-up process. Like, by entering a command in Budgie Desktop Settings / Autostart.

Axios

I had that problem both gnome and budgie with the update before last was trying get some files between machines plug usb in would not show up have not tried it with last fridays update tho will try it see if it still has problems later.
Usb keyboard mouse never had any problems (no-Encryption or Nvidia)
I have updated the budgie machine but never got around to the gnome update.

tenroc

noise617 Could still be related to power, even with a minipc. But yes, I agree, it seems that it's not the issue here. You don't even have anything currently plugged in as device 11 on bus 1. dmesg (and by extent journald) do store logs at kernel boot time. So if you don't see any relevant error with dmesg, I'm out of ideas. I guess we'll see if a kernel update fixes the issues at some point.

Axios

Seems my Budgie Computer is working fine now with usb thumbs
Maybe you need check your packages for broken? and see what it does with the next update
When I was having problems was update before last did current seem not to have problems.

noise617

tenroc Do you know if the problem can be diagnosticezed using this idea? Like, loading the PC with the kernel which completely disables USB devices and seeing what the errors are?

By the way, I wonder if it is possible to track down the reason for the malfunction during the kernel boot-up process. Like, by entering a command in Budgie Desktop Settings / Autostart.

tenroc

noise617 That's what I meant, dmesg and journald already log persistently accross reboots, so if you try to boot with the bugged kernel, and then with the previous one, you should get usable logs.
sudo journalctl | grep -E "(xhci|\: usb)" should give you errors related specifically with xhci (the usb controller) and usb devices.

I had some free time so I tried to compare boot logs with the bugged kernel, and with the last working one, and the issue seems to be situated here:

oct. 30 11:20:08 solus kernel: xhci_hcd 0000:00:14.0: Timeout while waiting for setup device command oct. 30 11:20:14 tenroc-laptop kernel: xhci_hcd 0000:00:14.0: Timeout while waiting for setup device command oct. 30 11:20:14 tenroc-laptop kernel: usb 2-3: device not accepting address 4, error -62 oct. 30 11:20:15 tenroc-laptop kernel: usbcore: registered new device driver apple-mfi-fastcharge oct. 30 11:20:15 tenroc-laptop kernel: usbcore: registered new interface driver bcm5974 oct. 30 11:20:19 tenroc-laptop kernel: xhci_hcd 0000:00:14.0: Timeout while waiting for setup device command oct. 30 11:20:25 tenroc-laptop kernel: xhci_hcd 0000:00:14.0: xHCI host not responding to stop endpoint command. oct. 30 11:20:25 tenroc-laptop kernel: xhci_hcd 0000:00:14.0: xHCI host controller not responding, assume dead oct. 30 11:20:25 tenroc-laptop kernel: xhci_hcd 0000:00:14.0: HC died; cleaning up oct. 30 11:20:25 tenroc-laptop kernel: xhci_hcd 0000:00:14.0: Timeout while waiting for setup device command oct. 30 11:20:25 tenroc-laptop kernel: usb 1-3: USB disconnect, device number 2 oct. 30 11:20:25 tenroc-laptop kernel: usb 2-3: device not accepting address 5, error -62 oct. 30 11:20:25 tenroc-laptop kernel: usb usb2-port3: unable to enumerate USB device oct. 30 11:20:25 tenroc-laptop kernel: usb 1-3-port2: couldn't allocate usb_device oct. 30 11:20:25 tenroc-laptop kernel: usb 1-3.1: USB disconnect, device number 4 oct. 30 11:20:25 tenroc-laptop kernel: usb 1-5: USB disconnect, device number 3 oct. 30 11:20:25 tenroc-laptop kernel: xhci_hcd 0000:00:14.0: dma_pool_free buffer-512, 0000000080da38b5/0x00000000fff8b600 (bad dma)

The difference between the two bootsequences is that with the bugged kernel, xhci times out at some point, dies, and then proceed to unload every usb devices. In my case, usb 1-5 (the second to last line in the log) is my internal trackpad / keyboard, and therefore I get stuck on the decryption prompt. The last line (bad dma) also only appears on bugged kernel.

The last kernel update also didn't fix the issue btw, if anyone is wondering.

lf-araujo

This is the first time I found a bug on this distro after three years using continuously. Still pretty good run. Yeah, last update disabled all sound in my machine, will try to revert back.

synthsncats

lf-araujo Same here. I was amazed at how Solus just works, without me intervening into its innards like in other distros.

synthsncats

The latest kernel update still "disables" my USB peripherals. I'm staying put in LTS, until someone knows what exactly is causing this and how to fix it.

lf-araujo

synthsncats A solution for my problem was posted to reddit:

« Previous Page Next Page »