Periodic fatal USB failure

Seacat17 · 2024-11-01 18:01:02

I have 3 USB slots in this laptop.

One USB 2.0
One USB 3.0
And one USB-C

It's been around a month since I started getting this weird issue. Out of nowhere my computer freezes for a few seconds and then USB 3.0 and USB-C slots become absolutely unusable. No matter what hardware I'll plug into them - it will not be recognized.

To fix this issue I need to reboot my computer but I'd like to get rid of it completely because it pisses me off at this point.

I was playing online right now and my SPU is connected to my USB 3.0 slot. Suddenly, it powered off so I knew the issue is there.

I immediately hopped into the terminal and did "sudo dmesg"

Here's the output of the error:

[    6.204576] amdgpu 0000:05:00.0: [drm] fb0: amdgpudrmfb frame buffer device
[    7.650256] usb 1-1: USB disconnect, device number 2
[    8.063890] usb 1-1: new high-speed USB device number 5 using xhci_hcd
[    8.210379] usb 1-1: New USB device found, idVendor=04e8, idProduct=6863, bcdDevice= 4.00
[    8.210391] usb 1-1: New USB device strings: Mfr=7, Product=8, SerialNumber=9
[    8.210397] usb 1-1: Product: SAMSUNG_Android
[    8.210402] usb 1-1: Manufacturer: SAMSUNG
[    8.210406] usb 1-1: SerialNumber: RFCT32RPZRN
[    8.310866] usbcore: registered new interface driver cdc_ether
[    8.317868] rndis_host 1-1:1.0 usb0: register 'rndis_host' at usb-0000:05:00.3-1, RNDIS device, 02:2e:14:13:0e:61
[    8.317898] usbcore: registered new interface driver rndis_host
[    8.337068] rndis_host 1-1:1.0 enp5s0f3u1: renamed from usb0
[   11.312008] block nvme0n1: No UUID available providing old NGUID
[   12.116199] Bluetooth: RFCOMM TTY layer initialized
[   12.116205] Bluetooth: RFCOMM socket layer initialized
[   12.116211] Bluetooth: RFCOMM ver 1.11
[   12.199868] warning: `mate-multiload-' uses wireless extensions which will stop working for Wi-Fi 7 hardware; use nl80211
[ 3002.286752] usb 1-2: timeout: still 12 active urbs on EP #86
[ 3006.413424] xhci_hcd 0000:05:00.3: xHCI host not responding to stop endpoint command
[ 3006.416692] xhci_hcd 0000:05:00.3: xHCI host controller not responding, assume dead
[ 3006.416692] xhci_hcd 0000:05:00.3: HC died; cleaning up
[ 3006.429357] usb 1-1: USB disconnect, device number 5
[ 3006.429399] rndis_host 1-1:1.0 enp5s0f3u1: unregister 'rndis_host' usb-0000:05:00.3-1, RNDIS device
[ 3006.429636] usb 1-2: Not enough bandwidth for altsetting 0
[ 3006.429640] usb 1-2: 2:0: usb_set_interface failed (-19)
[ 3006.443595] usb 1-2: Not enough bandwidth for altsetting 0
[ 3006.443602] usb 1-2: 2:0: usb_set_interface failed (-19)
[ 3006.443608] usb 1-2: 2:1: cannot set freq 48000 to ep 0x86
[ 3006.466960] usb 1-2: USB disconnect, device number 3
[ 3006.530330] usb 1-4: USB disconnect, device number 4

Does anyone know what's going on with my computer?

Last edited by Seacat17 (2024-11-02 19:17:57)

WorMzy · 2024-11-01 18:18:33

Mod note: not an Arch discussion, moving to Kernel and Hardware.

gromit · 2024-11-02 11:51:34

Which kernel are you currently using? Did you already try the linux-lts image?

seth · 2024-11-02 15:30:16

Please use [code][/code] tags, not "quote" tags. Edit your post in this regard.

If it's not the 6.11 kernel (timeframe would fit, though): https://wiki.archlinux.org/title/Power_ … utosuspend ?
Did you install sth. like TLP?
"usbcore.autosuspend=-1" will disable it but stuff like TLP can and will alter that value at runtime.

Edit: x-ref, https://bbs.archlinux.org/viewtopic.php?id=300719

Last edited by seth (2024-11-02 15:32:01)

Seacat17 · 2024-11-02 19:21:25

seth wrote:

Please use [code][/code] tags, not "quote" tags. Edit your post in this regard.
If it's not the 6.11 kernel (timeframe would fit, though): https://wiki.archlinux.org/title/Power_ … utosuspend ?
Did you install sth. like TLP?
"usbcore.autosuspend=-1" will disable it but stuff like TLP can and will alter that value at runtime.
Edit: x-ref, https://bbs.archlinux.org/viewtopic.php?id=300719

Kernel: 6.11.5-arch1-1

USB autosuspend is definitely not the problem because the USB devices are literally in use when this happens. Also, this shouldn't suspend the USB slots.

No, I did not install TLP or anything like that.

Seacat17 · 2024-11-02 19:23:07

gromit wrote:

Which kernel are you currently using? Did you already try the linux-lts image?

No, I did not try LTS. How do I?

seth · 2024-11-02 20:53:54

Install linux-lts and add it to your bootloader. Then boot it.
Howerver from the linked thread, the regression might have been backported into the LTS kernel, so an alternative approach would be to install 6.10.x instead of 6.11.x from either your cache or the ALA as suggested over there.

Seacat17 · 2024-11-03 01:53:36

seth wrote:

Install linux-lts and add it to your bootloader. Then boot it.
Howerver from the linked thread, the regression might have been backported into the LTS kernel, so an alternative approach would be to install 6.10.x instead of 6.11.x from either your cache or the ALA as suggested over there.

Is there any chance that switching to Zen kernel may fix the problem? I'm gonna try LTS and 6.10.x though, I've looked into how to do that.

(I know this is a bit off-topic but... Gosh, I love Linux so much, genuinely, you can switch a goddamn KERNEL on-the-fly, it's so amazing)

seth · 2024-11-03 07:15:47

Is there any chance that switching to Zen kernel may fix the problem?

That's highly unlikely. The zen kernel mostly tunes some performance/scheduler settings but doesn't add any kind of special driver patches.

Seacat17 · 2024-11-03 16:28:04

The error happened on LTS kernel as well.

[   20.811149] usb 1-1: USB disconnect, device number 2
[   21.230296] usb 1-1: new high-speed USB device number 5 using xhci_hcd
[   21.374361] usb 1-1: New USB device found, idVendor=04e8, idProduct=6863, bcdDevice= 4.00
[   21.374367] usb 1-1: New USB device strings: Mfr=7, Product=8, SerialNumber=9
[   21.374370] usb 1-1: Product: SAMSUNG_Android
[   21.374372] usb 1-1: Manufacturer: SAMSUNG
[   21.374373] usb 1-1: SerialNumber: RFCT32RPZRN
[   21.446821] usbcore: registered new interface driver cdc_ether
[   21.454534] rndis_host 1-1:1.0 eth0: register 'rndis_host' at usb-0000:05:00.3-1, RNDIS device, 02:2e:14:13:0e:61
[   21.454580] usbcore: registered new interface driver rndis_host
[   21.476896] rndis_host 1-1:1.0 enp5s0f3u1: renamed from eth0
[   39.104560] nvidia 0000:01:00.0: Using 47-bit DMA addresses
[ 1213.020175] usb 1-2: timeout: still 12 active urbs on EP #86
[ 1217.183555] xhci_hcd 0000:05:00.3: xHCI host not responding to stop endpoint command
[ 1217.186836] xhci_hcd 0000:05:00.3: xHCI host controller not responding, assume dead
[ 1217.186836] xhci_hcd 0000:05:00.3: HC died; cleaning up
[ 1217.199543] usb 1-1: USB disconnect, device number 5
[ 1217.199586] rndis_host 1-1:1.0 enp5s0f3u1: unregister 'rndis_host' usb-0000:05:00.3-1, RNDIS device
[ 1217.199691] usb 1-2: Not enough bandwidth for altsetting 0
[ 1217.199695] usb 1-2: 2:0: usb_set_interface failed (-19)
[ 1217.199766] usb 1-2: Not enough bandwidth for altsetting 1
[ 1217.199769] usb 1-2: 1:1: usb_set_interface failed (-19)
[ 1217.201390] usb 1-2: Not enough bandwidth for altsetting 1
[ 1217.201393] usb 1-2: 1:1: usb_set_interface failed (-19)
[ 1217.212923] usb 1-2: Not enough bandwidth for altsetting 1
[ 1217.212927] usb 1-2: 1:1: usb_set_interface failed (-19)
[ 1217.213838] usb 1-2: Not enough bandwidth for altsetting 1
[ 1217.213840] usb 1-2: 1:1: usb_set_interface failed (-19)
[ 1217.214668] usb 1-2: Not enough bandwidth for altsetting 1
[ 1217.214671] usb 1-2: 1:1: usb_set_interface failed (-19)
[ 1217.215477] usb 1-2: Not enough bandwidth for altsetting 1
[ 1217.215479] usb 1-2: 1:1: usb_set_interface failed (-19)
[ 1217.243889] usb 1-2: USB disconnect, device number 3
[ 1217.433854] usb 1-4: USB disconnect, device number 4

Gonna try Zen kernel now. Why? For science.

beroal · 2024-11-03 16:41:08

May it be a hardware failure?

Seacat17 · 2024-11-03 16:58:32

beroal wrote:

May it be a hardware failure?

Cannot stop thinking about this one, tho I really hope it's a software failure because this machine is barely 3 years old.

beroal · 2024-11-03 17:13:43

Seacat17 wrote:

beroal wrote:
May it be a hardware failure?
Cannot stop thinking about this one, tho I really hope it's a software failure because this machine is barely 3 years old.

Is the issue present with an OS that is not Linux?

Seacat17 · 2024-11-03 18:07:17

beroal wrote:

Seacat17 wrote:
beroal wrote:
May it be a hardware failure?
Cannot stop thinking about this one, tho I really hope it's a software failure because this machine is barely 3 years old.
Is the issue present with an OS that is not Linux?

I do not have any other OSes... Also for this issue to occur I need to wait for quite some time. It occurs absolutely randomly.

seth · 2024-11-03 18:32:19

https://bbs.archlinux.org/viewtopic.php?id=300719 would have a strikignly similar HW issue.
Science away, but a betting person might bet on testin 6.10

Seacat17 · 2024-11-03 22:01:04

Zen kernel has the same problem...

[11131.269794] perf: interrupt took too long (2525 > 2500), lowering kernel.perf_event_max_sample_rate to 79000
[18342.733861] usb 1-2: timeout: still 12 active urbs on EP #86
[18347.094624] xhci_hcd 0000:05:00.3: xHCI host not responding to stop endpoint command
[18347.098615] xhci_hcd 0000:05:00.3: xHCI host controller not responding, assume dead
[18347.098615] xhci_hcd 0000:05:00.3: HC died; cleaning up
[18347.110507] usb 1-2: Not enough bandwidth for altsetting 0
[18347.110511] usb 1-2: 2:0: usb_set_interface failed (-19)
[18347.110545] usb 1-1: USB disconnect, device number 11
[18347.110619] rndis_host 1-1:1.0 enp5s0f3u1: unregister 'rndis_host' usb-0000:05:00.3-1, RNDIS device
[18347.127943] usb 1-2: USB disconnect, device number 3
[18347.149046] usb 1-4: USB disconnect, device number 4

I'm sorry it takes so long for me to post results. It takes really long time to catch the error.

6.10 it is.

Is there any other log I can provide other than dmesg?

Last edited by Seacat17 (2024-11-03 22:02:10)

seth · 2024-11-04 07:50:44

Is there any other log I can provide other than dmesg?

system journal for userspace context to gage the cause/trigger.

Did you already test 6.10.x?

seth · 2024-11-07 15:05:29

x-ref, https://bbs.archlinux.org/viewtopic.php … 2#p2207082

Seacat17 · 2024-11-11 22:45:06

I am deeply sorry for the lack of communication. I was busy with offline duties.

I installed linux-6.10.10.arch1-1-x86_64 but I can't get my RTX 3050 to work with this kernel version. I have already removed nvidia-dkms and installed nvidia but the GPU is still not recognized.

EDIT: trying to install older Nvidia driver version gives me this:

[seacat17@seacat17-laptop ~]$ sudo pacman -U file:///nvidia-550.78-7-x86_64.pkg.tar.zst file:///opencl-nvidia-550.78-1-x86_64.pkg.tar.zst file:///nvidia-utils-550.78-1-x86_64.pkg.tar.zst file:///libxnvctrl-550.78-1-x86_64.pkg.tar.zst file:///lib32-nvidia-utils-550.78-2-x86_64.pkg.tar.zst
[sudo] password for seacat17: 
loading packages...
warning: downgrading package nvidia (565.57.01-2 => 550.78-7)
warning: downgrading package opencl-nvidia (565.57.01-1 => 550.78-1)
warning: downgrading package nvidia-utils (565.57.01-1 => 550.78-1)
warning: downgrading package libxnvctrl (565.57.01-1 => 550.78-1)
warning: downgrading package lib32-nvidia-utils (565.57.01-1 => 550.78-2)
resolving dependencies...
looking for conflicting packages...

Packages (5) lib32-nvidia-utils-550.78-2  libxnvctrl-550.78-1  nvidia-550.78-7
             nvidia-utils-550.78-1  opencl-nvidia-550.78-1

Total Installed Size:   858.99 MiB
Net Upgrade Size:      -150.90 MiB

:: Proceed with installation? [Y/n] 
(5/5) checking keys in keyring                     [######################] 100%
(5/5) checking package integrity                   [######################] 100%
(5/5) loading package files                        [######################] 100%
(5/5) checking for file conflicts                  [######################] 100%
error: failed to commit transaction (conflicting files)
nvidia-utils: /usr/lib/libnvidia-egl-gbm.so exists in filesystem (owned by egl-gbm)
nvidia-utils: /usr/lib/libnvidia-egl-gbm.so.1 exists in filesystem (owned by egl-gbm)
nvidia-utils: /usr/share/egl/egl_external_platform.d/15_nvidia_gbm.json exists in filesystem (owned by egl-gbm)
Errors occurred, no packages were upgraded.
[seacat17@seacat17-laptop ~]$

Last edited by Seacat17 (2024-11-11 22:46:44)

seth · 2024-11-11 23:11:45

egl-gbm was split from nvidia-utils, you'll have to remove that.
For dkms you'll need the linux-headers package matching your version and if you want to use a precompiled nvidia package, it has to be the one that was compiled for (in this case) 6.10.10 which would be https://gitlab.archlinux.org/archlinux/ … 2fecc97de4

Seacat17 · 2024-11-11 23:24:55

seth wrote:

egl-gbm was split from nvidia-utils, you'll have to remove that.
For dkms you'll need the linux-headers package matching your version and if you want to use a precompiled nvidia package, it has to be the one that was compiled for (in this case) 6.10.10 which would be https://gitlab.archlinux.org/archlinux/ … 2fecc97de4

Linux Headers 6.10.10 is not in my Pacman cache... What can I do about that?

seth · 2024-11-12 07:22:08

https://wiki.archlinux.org/title/Arch_Linux_Archive

Seacat17 · 2024-11-12 16:27:55

seth wrote:

https://wiki.archlinux.org/title/Arch_Linux_Archive

Latest is linux-headers-6.9.arch1-1-x86_64

EDIT: NVM I am an idiot. They were in the middle of the list.

Last edited by Seacat17 (2024-11-12 16:28:54)

Seacat17 · 2024-11-12 16:33:08

Linux 6.10.10 is ready for the battle.

I'm gonna test it as hard as I can to make the error appear (or to confirm that it does not appear in 6.10.10).

seth · 2024-11-12 16:33:33

https://en.wikipedia.org/wiki/Collation#Alphabetical

Arch Linux

#1 2024-11-01 18:01:02

Periodic fatal USB failure

#2 2024-11-01 18:18:33

Re: Periodic fatal USB failure

#3 2024-11-02 11:51:34

Re: Periodic fatal USB failure

#4 2024-11-02 15:30:16

Re: Periodic fatal USB failure

#5 2024-11-02 19:21:25

Re: Periodic fatal USB failure

#6 2024-11-02 19:23:07

Re: Periodic fatal USB failure

#7 2024-11-02 20:53:54

Re: Periodic fatal USB failure

#8 2024-11-03 01:53:36

Re: Periodic fatal USB failure

#9 2024-11-03 07:15:47

Re: Periodic fatal USB failure

#10 2024-11-03 16:28:04

Re: Periodic fatal USB failure

#11 2024-11-03 16:41:08

Re: Periodic fatal USB failure

#12 2024-11-03 16:58:32

Re: Periodic fatal USB failure

#13 2024-11-03 17:13:43

Re: Periodic fatal USB failure

#14 2024-11-03 18:07:17

Re: Periodic fatal USB failure

#15 2024-11-03 18:32:19

Re: Periodic fatal USB failure

#16 2024-11-03 22:01:04

Re: Periodic fatal USB failure

#17 2024-11-04 07:50:44

Re: Periodic fatal USB failure

#18 2024-11-07 15:05:29

Re: Periodic fatal USB failure

#19 2024-11-11 22:45:06

Re: Periodic fatal USB failure

#20 2024-11-11 23:11:45

Re: Periodic fatal USB failure

#21 2024-11-11 23:24:55

Re: Periodic fatal USB failure

#22 2024-11-12 07:22:08

Re: Periodic fatal USB failure

#23 2024-11-12 16:27:55

Re: Periodic fatal USB failure

#24 2024-11-12 16:33:08

Re: Periodic fatal USB failure

#25 2024-11-12 16:33:33

Re: Periodic fatal USB failure

Board footer