Bug 2227968

Summary: power_supply_uevent on USB leads to page_fault_oops
Product: [Fedora] Fedora Reporter: David Bowen <david>
Component: kernelAssignee: Kernel Maintainer List <kernel-maint>
Status: CLOSED DUPLICATE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: medium Docs Contact:
Priority: unspecified    
Version: 38CC: acaringi, adscvr, airlied, alciregi, bskeggs, hdegoede, hpa, jarodwilson, josef, kernel-maint, lgoncalv, linville, masami256, mchehab, ptalbert, stanley.king, steved
Target Milestone: ---Keywords: Regression
Target Release: ---   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: ---
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2023-08-20 21:37:35 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
The interesting section of the journal when the problem occurred
none
USB devices
none
USB device tree
none
Modules
none
oops.conf
none
abrt output none

Description David Bowen 2023-07-31 23:47:09 UTC 
Over the last month or so when I change the USB devices the system appears to hang.

What is actually happening is that the entire USB input system shuts down and so I can't login. I usually solve this problem by doing a hard restart. As you can imagine, this is starting to become tedious.

If I try to re-create this I can usually do it after about five times.

In my case I use a USB KVM switch which has been working fine for a couple of years. I replaced it with a new one - same problem. I used new cables - same problem. I used a new Logitech unifying dongle - same problem. All my HID devices are wireless so I didn't try replacing those, they aren't physically connected to the machine.

It looks as though this user has a similar problem:

https://www.spinics.net/linux/fedora/fedora-users/msg519556.html

Reproducible: Sometimes

Steps to Reproduce:
Repeatedly plug and unplug a Logitech unifying receiver.


Actual Results:  
kernel: BUG: unable to handle page fault for address: ffffabefc3981a50

Expected Results:  
Historically I've been able to add and remove devices or use the KVM switch with no problems.

$ uname -a
Linux somehost 6.4.6-200.fc38.x86_64 #1 SMP PREEMPT_DYNAMIC Mon Jul 24 20:51:12 UTC 2023 x86_64 GNU/Linux
$ rpm -qa | grep kernel
kernel-srpm-macros-1.0-19.fc38.noarch
libreport-plugin-kerneloops-2.17.11-1.fc38.x86_64
abrt-addon-kerneloops-2.17.1-1.fc38.x86_64
kernel-devel-6.3.12-200.fc38.x86_64
kernel-modules-core-6.3.12-200.fc38.x86_64
kernel-core-6.3.12-200.fc38.x86_64
kernel-modules-6.3.12-200.fc38.x86_64
kernel-6.3.12-200.fc38.x86_64
kernel-modules-extra-6.3.12-200.fc38.x86_64
kernel-modules-core-6.4.4-200.fc38.x86_64
kernel-core-6.4.4-200.fc38.x86_64
kernel-modules-6.4.4-200.fc38.x86_64
kernel-devel-6.4.4-200.fc38.x86_64
kernel-6.4.4-200.fc38.x86_64
kernel-modules-extra-6.4.4-200.fc38.x86_64
kernel-headers-6.4.4-200.fc38.x86_64
kernel-modules-core-6.4.6-200.fc38.x86_64
kernel-core-6.4.6-200.fc38.x86_64
kernel-modules-6.4.6-200.fc38.x86_64
kernel-devel-6.4.6-200.fc38.x86_64
kernel-devel-matched-6.4.6-200.fc38.x86_64
kernel-6.4.6-200.fc38.x86_64
kernel-modules-extra-6.4.6-200.fc38.x86_64
Comment 1 David Bowen 2023-07-31 23:51:39 UTC 
Created attachment 1980988 [details]
The interesting section of the journal when the problem occurred

Kernel trace leading from power_supply_uevent to page_fault_oops
Comment 2 David Bowen 2023-07-31 23:53:52 UTC 
Created attachment 1980989 [details]
USB devices

Log from running "lsusb"
Comment 3 David Bowen 2023-07-31 23:54:24 UTC 
Created attachment 1980990 [details]
USB device tree

Log from running "lsusb -t"
Comment 4 David Bowen 2023-07-31 23:56:51 UTC 
Created attachment 1980991 [details]
Modules

Result of running "lsmod"
Comment 5 David Bowen 2023-08-01 00:06:01 UTC 
Created attachment 1980992 [details]
oops.conf

I've added:

DropNotReportableOopses=no

to oops.conf
Comment 6 David Bowen 2023-08-01 00:27:17 UTC 
Created attachment 1981005 [details]
abrt output

It's not great because it's recorded as "not-reportable".
Comment 7 Stan King 2023-08-07 22:28:33 UTC 
(I'm adding this comment to bugzillae 2227968 and 2227221, which appear to be duplicates of each other.)

I'm having this same problem, with very similar dmesg output, triggered by the USB disconnect that occurs when switching ports on a KVM switch that implements a USB hub to provide the connection to the keyboard and mouse.

It started with kernel kernel-6.4.4-200.fc38.x86_64 and persists with 6.4.7.

Kernel kernel-6.3.12-200.fc38.x86_64 is OK.

I see this on three of my systems, the newest of which has an Intel i5-11400 processor.
Two of my systems seem immune to this, the oldest of which has an Intel i7-3770 processor.

I'd be happy to try a debug kernel on my affected i7-8550U system, but I'd have to be stepped through the generation process.
Comment 8 David Bowen 2023-08-20 21:37:35 UTC 

*** This bug has been marked as a duplicate of bug 2227221 ***