[Bug 1227221] New: Blank Screen After Booting Kernel 6.4.0-150600.23.7.3 Post Leap Logo on MacBook Pro 5.1
https://bugzilla.suse.com/show_bug.cgi?id=1227221 Bug ID: 1227221 Summary: Blank Screen After Booting Kernel 6.4.0-150600.23.7.3 Post Leap Logo on MacBook Pro 5.1 Classification: openSUSE Product: openSUSE Distribution Version: Leap 15.6 Hardware: 64bit OS: Other Status: NEW Severity: Critical Priority: P5 - None Component: Kernel:Drivers Assignee: kernel-bugs@suse.de Reporter: jacopo.radice@outlook.com QA Contact: qa-bugs@suse.de Target Milestone: --- Found By: --- Blocker: --- User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:109.0) Gecko/20100101 Firefox/115.0 Build Identifier: After updating the kernel to version 6.4.0-150600.23.7.3 on my MacBook Pro 5.1, the screen goes blank after the openSUSE Leap logo during boot. The system does not reach the DE login page (XFCE). This issue was also present in the previous kernel version 6.4.0-150600.21.2. The problem persists across different attempts and reboots. If I boot from kernel 5.14.21-150500.298.1.g40e256a.x86_64 I have no issues. #### Reproducible: Always ### Steps to Reproduce: 1. Boot MacBook Pro 5.1 with kernel version 6.4.0-150600.23.7.3-default. 2. Observe the openSUSE Leap logo during boot. 3. Notice the screen goes blank and does not proceed to the DE login page (XFCE). ### Actual Results: The screen remains blank after the openSUSE Leap logo and does not reach the DE login page. ### Expected Results: The system should proceed to the DE login page (XFCE) after the openSUSE Leap logo. #### Additional Information: - The issue persists in both kernel versions 6.4.0-150600.21.2 and 6.4.0-150600.23.7.3. - The system works correctly with previous kernel versions (e.g., 5.14.21-150500.298.1). - The touchpad issue in kernel version 5.14.21-150500.55.49-default was resolved previously by Takashi Iwai, who might have insights into this new issue. ### Package Versions: rpm -q kernel-default kernel-default-extra kernel-default-optional kernel-default: kernel-default-5.14.21-150500.295.1.g28dcd75.x86_64 kernel-default-5.14.21-150500.298.1.g40e256a.x86_64 kernel-default-5.14.21-150500.293.1.ge008690.x86_64 kernel-default-5.14.21-150500.294.1.g51d8359.x86_64 kernel-default-6.4.0-150600.21.2.x86_64 kernel-default-6.4.0-150600.23.7.3.x86_64 kernel-default-extra: kernel-default-extra-6.4.0-150600.21.2.x86_64 kernel-default-extra-6.4.0-150600.23.7.3.x86_64 kernel-default-extra-5.14.21-150500.293.1.ge008690.x86_64 kernel-default-extra-5.14.21-150500.298.1.g40e256a.x86_64 kernel-default-extra-5.14.21-150500.295.1.g28dcd75.x86_64 kernel-default-extra-5.14.21-150500.294.1.g51d8359.x86_64 kernel-default-optional: kernel-default-optional-5.14.21-150500.295.1.g28dcd75.x86_64 kernel-default-optional-6.4.0-150600.21.2.x86_64 kernel-default-optional-6.4.0-150600.23.7.3.x86_64 kernel-default-optional-5.14.21-150500.298.1.g40e256a.x86_64 - kernel-default-optional-5.14.21-150500.294.1.g51d8359.x86_64 - kernel-default-optional-5.14.21-150500.293.1.ge008690.x86_64 Additional Context Previously, I encountered a similar issue with the touchpad on my MacBook Pro 5.1, which was resolved by Takashi Iwai. Given his expertise and familiarity with the hardware, I have added him to the CC list for this bug report. -- You are receiving this mail because: You are on the CC list for the bug.
https://bugzilla.suse.com/show_bug.cgi?id=1227221 Jacopo Radice <jacopo.radice@outlook.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |tiwai@suse.com -- You are receiving this mail because: You are on the CC list for the bug.
https://bugzilla.suse.com/show_bug.cgi?id=1227221 https://bugzilla.suse.com/show_bug.cgi?id=1227221#c2 --- Comment #2 from Jacopo Radice <jacopo.radice@outlook.com> --- (In reply to Takashi Iwai from comment #1)
Is it only about the graphics? That is, when you boot with "nomodeset" boot option, does it reach to the desktop (but with the EFI framebuffer)?
Also, please check with the recent kernel in OBS Kernel:stable:Backport repo and see whether the same problem is seen or not.
Hi Takashi, I have performed the tests as you suggested. Here are the results: Test with "nomodeset" Option: Booting with the nomodeset option on kernel version 6.4.0-150600.23.7.3 resulted in the same issue. The screen goes blank after the openSUSE Leap logo, and the system does not reach the desktop environment. Test with Kernel from OBS Kernel:stable:Backport Repo: I installed and tested the latest kernel from the Kernel:stable:Backport repository. The system still experiences the same issue with both the nomodeset option and without it. The screen goes blank after the openSUSE Leap logo, and the system does not reach the desktop environment. Additional Information: The issue persists across multiple kernel versions, including 6.4.0-150600.21.2, 6.4.0-150600.23.7.3, and the latest from the Kernel:stable:Backport repository. The system works correctly with older kernel versions (e.g., 5.14.21-150500.298.1). Thank you for your assistance. Best regards, Jacopo -- You are receiving this mail because: You are on the CC list for the bug.
https://bugzilla.suse.com/show_bug.cgi?id=1227221 https://bugzilla.suse.com/show_bug.cgi?id=1227221#c3 --- Comment #3 from Jacopo Radice <jacopo.radice@outlook.com> --- Created attachment 875797 --> https://bugzilla.suse.com/attachment.cgi?id=875797&action=edit dmeseg log -- You are receiving this mail because: You are on the CC list for the bug.
https://bugzilla.suse.com/show_bug.cgi?id=1227221 https://bugzilla.suse.com/show_bug.cgi?id=1227221#c4 --- Comment #4 from Jacopo Radice <jacopo.radice@outlook.com> --- Created attachment 875798 --> https://bugzilla.suse.com/attachment.cgi?id=875798&action=edit Xorg.log -- You are receiving this mail because: You are on the CC list for the bug.
https://bugzilla.suse.com/show_bug.cgi?id=1227221 https://bugzilla.suse.com/show_bug.cgi?id=1227221#c5 --- Comment #5 from Jacopo Radice <jacopo.radice@outlook.com> --- Created attachment 875799 --> https://bugzilla.suse.com/attachment.cgi?id=875799&action=edit journalctl.log -- You are receiving this mail because: You are on the CC list for the bug.
https://bugzilla.suse.com/show_bug.cgi?id=1227221 https://bugzilla.suse.com/show_bug.cgi?id=1227221#c7 --- Comment #7 from Jacopo Radice <jacopo.radice@outlook.com> --- Created attachment 875800 --> https://bugzilla.suse.com/attachment.cgi?id=875800&action=edit dmesg_nomodeset Hi Takashi, Following your instructions, I managed to boot into single-user mode with the nomodeset option and collect the dmesg output. Please find the attached log file for your review. Best regards, Jacopo -- You are receiving this mail because: You are on the CC list for the bug.
https://bugzilla.suse.com/show_bug.cgi?id=1227221 https://bugzilla.suse.com/show_bug.cgi?id=1227221#c9 --- Comment #9 from Jacopo Radice <jacopo.radice@outlook.com> --- (In reply to Takashi Iwai from comment #8)
Thanks. And after the single login with nomodeset, can you switch to runlevel 3, i.e. call "init 3" on the login shell? Does the system still work afterwards?
If yes, move to runlevel 5 via "init 5". This will switch to the GUI.
Hi Takashi, I followed your instructions to switch runlevels with the nomodeset option. Here are the detailed results: Runlevel 3: After booting into single-user mode with nomodeset, I switched to runlevel 3 using init 3. The system functioned correctly in runlevel 3. I was able to log in and use the command-line interface without issues. I verified system functionality by running several basic commands (e.g., ls, pwd, top), and everything worked as expected. Runlevel 5: When I switched to runlevel 5 using init 5, the system attempted to start the graphical interface. However, the display went blank, indicating that the graphical user interface did not load as expected. There were no visible error messages on the screen; it simply went blank. Please let me know the next steps or if you need any additional information to troubleshoot this issue further. Thank you for your assistance. Best regards, Jacopo -- You are receiving this mail because: You are on the CC list for the bug.
https://bugzilla.suse.com/show_bug.cgi?id=1227221 https://bugzilla.suse.com/show_bug.cgi?id=1227221#c11 --- Comment #11 from Jacopo Radice <jacopo.radice@outlook.com> --- (In reply to Takashi Iwai from comment #10)
Hm, could you run like Xorg -retro on Linux console with runlevel 3? Does the X screen appear properly?
And, is your system with auto-login to a user account? If so, try to turn off (I guess modifying /etc/sysconfig/displaymanager should work), and check whether the display-manager itself works with nomodeset or not.
Hi Takashi, I followed your latest instructions to test the X server and check the auto-login settings. Here are the detailed results: X Server Test: After booting into single-user mode with nomodeset and switching to runlevel 3 (init 3), I logged in and ran the command Xorg -retro. The X server failed to start, and I encountered the following errors (full output attached in screenshots): livecodeserver Current version of pixman: 0.42.2 (EE) Fatal server error: (EE) no screens found(EE) (EE) (EE) Please consult the X.Org Foundation support at http://wiki.x.org for help. (EE) Please also check the log file at "/var/log/Xorg.0.log" for additional information. (EE) (EE) Server terminated with error (1). Closing log file. Auto-Login Check: I checked the /etc/sysconfig/displaymanager file for auto-login settings, and it appears auto-login is not enabled. Given that the X server fails to start with the no screens found error, it seems there might be an issue with the graphics configuration or drivers. Please let me know the next steps or if you need any additional information. Thank you for your continued assistance. Jacopo -- You are receiving this mail because: You are on the CC list for the bug.
https://bugzilla.suse.com/show_bug.cgi?id=1227221 https://bugzilla.suse.com/show_bug.cgi?id=1227221#c12 --- Comment #12 from Jacopo Radice <jacopo.radice@outlook.com> --- Created attachment 875807 --> https://bugzilla.suse.com/attachment.cgi?id=875807&action=edit Xorg0.log That's the full log that Xorg --retro gives. -- You are receiving this mail because: You are on the CC list for the bug.
https://bugzilla.suse.com/show_bug.cgi?id=1227221 https://bugzilla.suse.com/show_bug.cgi?id=1227221#c14 --- Comment #14 from Jacopo Radice <jacopo.radice@outlook.com> --- Created attachment 875810 --> https://bugzilla.suse.com/attachment.cgi?id=875810&action=edit Xorg0.log as sudo I've also run it by using sudo. I don't know why in the log run via sudo it says: "Current Operating System: Linux pc0 5.14.21-150500.298.g40e256a-default" While I've done the procedure via kernel 6.4.0-150600.23.7-default. In fact, on the terminal screenshot that I take with my phone it says: "Current Operating System: Linux pc0 5.14.21-150500.298.g40e256a-default #1 SMP PREEMPT_DYNAMIC Fri Jun 14 14:33:11 UTC 2024 (33f31da) x86_x64 -- You are receiving this mail because: You are on the CC list for the bug.
https://bugzilla.suse.com/show_bug.cgi?id=1227221 https://bugzilla.suse.com/show_bug.cgi?id=1227221#c16 --- Comment #16 from Jacopo Radice <jacopo.radice@outlook.com> --- (In reply to Takashi Iwai from comment #15)
You seem using nouveau X driver. Could you try to uninstall xf86-video-nouveau and retry?
Hi Takashi, Thank you for your response. I understand that you suggested uninstalling the xf86-video-nouveau driver. However, I am unable to do so because my laptop uses an NVIDIA GeForce 9400M graphics card, which is quite old and is no longer supported by the proprietary NVIDIA drivers. Given this situation, I am reliant on the nouveau driver for my graphics. Uninstalling it would leave me without any functional driver for my graphics card. Here is a summary of the current situation: Graphics Card: NVIDIA GeForce 9400M Driver: nouveau (required due to lack of support from proprietary NVIDIA drivers) Issue: X server fails to start with no screens found error when using the nomodeset option. Could you please advise on any alternative solutions or configurations that might help resolve this issue while still using the nouveau driver? Your guidance is much appreciated. Jacopo -- You are receiving this mail because: You are on the CC list for the bug.
https://bugzilla.suse.com/show_bug.cgi?id=1227221 https://bugzilla.suse.com/show_bug.cgi?id=1227221#c18 --- Comment #18 from Jacopo Radice <jacopo.radice@outlook.com> --- (In reply to Takashi Iwai from comment #17)
xf86-video-nouveau is a user-space driver, and it isn't needed at all for running with the proper DRM device. And the recent kernel provides a DRM device even for the simple framebuffer including EFI fb. So everything should work (even better) without xf86-video-nouveau stuff.
Just give it a try.
Hi Takashi, Thank you for your continued assistance. I followed your instructions to uninstall the xf86-video-nouveau driver. Here is a detailed account of the steps I took and the results: Steps Taken: Booted into runlevel 3. Uninstalled the xf86-video-nouveau driver using: sudo zypper rm xf86-video-nouveau Rebooted the system. Tested the X server with Xorg -retro. Results: Unfortunately, nothing has changed. The X server still fails to start with the same no screens found error. Upon rebooting, the system was stuck in runlevel 3 and could not load any GUI even with the 5.x kernel. This behavior persisted even after multiple attempts to start the X server. Rollback to Previous State: Luckily, I had created a snapshot with snapper before removing the xf86-video-nouveau driver. I performed a rollback to that snapshot by booting it in read-only mode. The system is now back to its previous state, and the old kernel works as usual with the GUI. Given these outcomes, it seems that removing the xf86-video-nouveau driver did not resolve the issue and instead led to additional complications. I am currently back to the original state with the nouveau driver reinstalled. Could you please provide further guidance on how to proceed? Is there an alternative configuration or driver that might help resolve this issue? Your expertise is greatly appreciated. Best regards, Jacopo -- You are receiving this mail because: You are on the CC list for the bug.
https://bugzilla.suse.com/show_bug.cgi?id=1227221 https://bugzilla.suse.com/show_bug.cgi?id=1227221#c20 --- Comment #20 from Jacopo Radice <jacopo.radice@outlook.com> --- Created attachment 875814 --> https://bugzilla.suse.com/attachment.cgi?id=875814&action=edit Xorg0.log no-nouveau Yes I confirm. This time I directly do the procedure by using the root user but the outcome is the same. That's the log -- You are receiving this mail because: You are on the CC list for the bug.
https://bugzilla.suse.com/show_bug.cgi?id=1227221 https://bugzilla.suse.com/show_bug.cgi?id=1227221#c22 --- Comment #22 from Jacopo Radice <jacopo.radice@outlook.com> --- (In reply to Takashi Iwai from comment #21)
The log shows:
[ 17.985] (II) LoadModule: "nouveau" [ 17.986] (WW) Warning, couldn't open module nouveau [ 17.986] (EE) Failed to load module "nouveau" (module does not exist, 0)
It means that you likely have some fixed configuration to use nouveau driver, instead of the automatic selection.
Check /etc/X11/xorg.conf.d/*.conf. If any of the file doesn't belong to a package (you can check via rpm -qf /etc/X11/xorg.conf.d/*), it's a non-standard config.
Hi Takashi, I followed your instructions and identified that the file /etc/X11/xorg.conf.d/20-nouveau.conf was not owned by any package. Here are the steps I took: Inspected the contents of /etc/X11/xorg.conf.d/20-nouveau.conf and found the following configuration: Section "Device" Identifier "Nvidia Card" Driver "nouveau" EndSection Edited the file to comment out the lines forcing the use of the nouveau driver: # Section "Device" # Identifier "Nvidia Card" # Driver "nouveau" # EndSection Uninstalled the xf86-video-nouveau driver again: sudo zypper rm xf86-video-nouveau Rebooted the system using the kernel 6.4.0-150600.23.7-default without any additional settings from GRUB. Results: The display now works, and I can log into XFCE successfully. Everything seems to be functioning correctly so far. Is there anything else I should check or any additional logs you need? Thank you for your guidance and support. Best regards, Jacopo -- You are receiving this mail because: You are on the CC list for the bug.
participants (1)
-
bugzilla_noreply＠suse.com