Bug 1223109

Summary: Booting issue after upgrade to kernel-default-6.8.6-1.1
Product: [openSUSE] openSUSE Tumbleweed Reporter: Bartosz Kaczyński <bkaczynski>
Component: KernelAssignee: openSUSE Kernel Bugs <kernel-bugs>
Status: RESOLVED FIXED QA Contact: E-mail List <qa-bugs>
Severity: Major    
Priority: P5 - None CC: BjAlvestad, bkaczynski, jslaby, jweberhofer
Version: CurrentFlags: jslaby: needinfo? (bkaczynski)
Target Milestone: ---   
Hardware: x86-64   
OS: openSUSE Tumbleweed   
Whiteboard:
Found By: --- Services Priority:
Business Priority: Blocker: ---
Marketing QA Status: --- IT Deployment: ---
Attachments: hwinfo
photo-of-system-stalls

Description Bartosz Kaczyński 2024-04-19 07:05:11 UTC 
Created attachment 874370 [details]
hwinfo

Disclaimer: This is my first bug report on Bugzilla and I may not have the necessary technical knowledge to address all concerns. The response on the forum led me here [1]

I am experiencing a problem with booting the system after updating the kernel to the kernel-default-6.8.6-1.1 version. I am attaching a picture of the moment when the system hangs.

I am also including results of the `inxi -Faz` command[2] from after restoring to the previous version of the system.

[1] https://forums.opensuse.org/t/boot-issue-following-recent-kernel-update/174306
[2] https://paste.opensuse.org/pastes/2db537ab61eb
Comment 1 Bartosz Kaczyński 2024-04-19 07:06:09 UTC 
Created attachment 874371 [details]
photo-of-system-stalls
Comment 2 Jiri Slaby 2024-04-19 08:47:30 UTC 
Great report if this is your first one.

It's e1000e's fault. It takes a spin lock in e1000_watchdog_task():
  spin_lock(&adapter->stats64_lock);
  e1000e_update_stats(adapter);
  -> e1000e_update_phy_stats()
     -> e1000e_read_phy_reg_mdic()
        -> usleep_range() ----> Boom.

Caused by 6.8.5 commit:
commit 0a4e3c2d976aa4dd38951afd6267f74ef3fade0e
Author: Vitaly Lifshits <vitaly.lifshits@intel.com>
Date:   Thu Jan 4 16:16:52 2024 +0200

    e1000e: Workaround for sporadic MDI error on Meteor Lake systems
    
    commit 6dbdd4de0362c37e54e8b049781402e5a409e7d0 upstream.

Reported upstream:
https://lore.kernel.org/all/809b5785-e65f-47f4-b52b-f9d2af0a3484@kernel.org/
Comment 3 Takashi Iwai 2024-04-19 13:53:11 UTC 
*** Bug 1222945 has been marked as a duplicate of this bug. ***
Comment 4 Jiri Slaby 2024-04-30 07:36:02 UTC 
The fix is at:
https://lore.kernel.org/intel-wired-lan/20240417190320.3159360-1-vitaly.lifshits@intel.com/T/#u

Now pushing. Could you test Kernel:stable once it builds?
Comment 5 Bartosz Kaczyński 2024-05-01 06:14:03 UTC 
I've added Kernel:stable repo and upgraded with --allow-vendor-change flag and after 

I added the Kernel:stable repository and updated the system (zypper dup --allow-vendor-change).
After restarting, it booted up correctly.
# uname -r
6.8.8-2.ge7fcfd8-default
Comment 6 Jiri Slaby 2024-05-02 10:40:10 UTC 
*** Bug 1223691 has been marked as a duplicate of this bug. ***
Comment 7 Bartosz Kaczyński 2024-05-16 15:09:19 UTC 
No issues with booting after recent kernel update

~> uname -r
6.8.9-1-default

I mark this bug as resolved-fixed.

Thank you.