Bug 1216575 - RPi2 TW images not start into desktop environment
Summary: RPi2 TW images not start into desktop environment
Status: CONFIRMED
Alias: None
Product: openSUSE Tumbleweed
Classification: openSUSE
Component: Kernel:Drivers (show other bugs)
Version: Current
Hardware: armv7 openSUSE Tumbleweed
: P5 - None : Normal (vote)
Target Milestone: ---
Assignee: Kernel Bugs
QA Contact: E-mail List
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2023-10-25 17:29 UTC by Kasimir _
Modified: 2024-02-29 17:38 UTC (History)
8 users (show)

See Also:
Found By: ---
Services Priority:
Business Priority:
Blocker: ---
Marketing QA Status: ---
IT Deployment: ---


Attachments
dmesg from Snapshot20231114 (33.69 KB, text/plain)
2023-11-16 10:39 UTC, Kasimir _
Details
dmesg from Snapshot20231127 (30.37 KB, text/plain)
2023-11-29 20:01 UTC, Kasimir _
Details
dmesg from Snapshot20240111 (34.74 KB, text/plain)
2024-01-13 11:54 UTC, Kasimir _
Details
journalctl from Snapshot20240111 (135.04 KB, text/plain)
2024-01-13 11:55 UTC, Kasimir _
Details
dmesg from Snapshot20240128 with boot options (753.00 KB, text/plain)
2024-01-30 17:30 UTC, Kasimir _
Details
dmesg from Snapshot20240131 with boot options on Dell screen (715.19 KB, text/plain)
2024-02-07 19:45 UTC, Kasimir _
Details
dmesg from Snapshot20240131 with boot options on HP screen (718.38 KB, text/plain)
2024-02-07 19:45 UTC, Kasimir _
Details

Note You need to log in before you can comment on or make changes to this bug.
Comment 2 Kasimir _ 2023-11-14 17:44:52 UTC
If I add the follow to the kernel command line, it boots up to the desktop environment:
modprobe.blacklist=vc4
Comment 4 Ivan Ivanov 2023-11-24 13:20:12 UTC
I can't see anything wrong in logs... Lets add some DRM experts.
Comment 5 Kasimir _ 2023-11-29 20:01:10 UTC
Created attachment 871065 [details]
dmesg from Snapshot20231127

No change with Snapshot20231127
Comment 6 David Walker 2023-12-07 03:26:28 UTC
FYI, I'm also seeing this on a RPI4b running Tumbleweed 20231129 and Gnome. modprobe.blacklist=vc4 fixes the black screen (although breaks other things, I assume). I'm not sure when it started, as I skipped multiple releases while out of town.
Comment 7 Ivan Ivanov 2023-12-07 08:10:50 UTC
Issues with RPi 4b are differnt, see bsc#1217512.
Comment 8 Kasimir _ 2024-01-13 11:54:37 UTC
Created attachment 871853 [details]
dmesg from Snapshot20240111

Are there any news?

The problems still exist with the Snapshot20240111.
Comment 9 Kasimir _ 2024-01-13 11:55:23 UTC
Created attachment 871854 [details]
journalctl from Snapshot20240111
Comment 10 Ivan Ivanov 2024-01-23 07:22:00 UTC
Hm, I can't see any errors, display manager reports that it is 
runing fine, but still nothing on the HDMI output??

# systemctl status display-manager.service
● display-manager.service - X Display Manager
     Loaded: loaded (/usr/lib/systemd/system/display-manager.service; enabled; preset: enabled)
     Active: active (running) since Tue 2024-01-23 07:17:18 UTC; 3min 23s ago
    Process: 2024 ExecStart=/usr/lib/X11/display-manager start (code=exited, status=0/SUCCESS)
   Main PID: 2032 (lightdm)
      Tasks: 7 (limit: 1957)
        CPU: 3.744s
     CGroup: /system.slice/display-manager.service
             ├─2032 /usr/sbin/lightdm
             ├─2039 /usr/bin/Xorg.bin :0 -seat seat0 -auth /run/lightdm/root/:0 -nolisten tcp vt7 -novtswitch
             └─2082 lightdm --session-child 13 20

Jan 23 07:17:17 2b11 systemd[1]: Starting X Display Manager...
Jan 23 07:17:17 2b11 display-manager[2028]: /etc/vconsole.conf available
Jan 23 07:17:17 2b11 display-manager[2028]: XKBLAYOUT: us
Jan 23 07:17:17 2b11 display-manager[2028]: XKBLAYOUT is already set. Better don't touch X11 keyboard configuration.
Jan 23 07:17:18 2b11 systemd[1]: Started X Display Manager.
Comment 11 Stefan Dirsch 2024-01-23 09:12:15 UTC
Did this OpenSource Driver stack ever work on RPi2? I think things didn't improve before RPi4. :-( Reassigning and adding our DRM experts ...
Comment 12 Ivan Ivanov 2024-01-23 11:26:23 UTC
It is working fine with RPi3's
Comment 13 Takashi Iwai 2024-01-29 15:03:13 UTC
Could you try to pass boot options "drm.debug=0x1e log_buf_len=16M" and get the dmesg?  And we can compare with the working case on Ivan's RPi3, for example.

Also, when you boot with nomodeset boot option, it'll keep working?
Comment 14 Kasimir _ 2024-01-30 17:30:59 UTC
Created attachment 872320 [details]
dmesg from Snapshot20240128 with boot options

Attached is the output of dmesg with the boot options "drm.debug=0x1e log_buf_len=16M".

Yes, with nomodeset boot option it works.
Comment 15 Takashi Iwai 2024-02-06 13:37:10 UTC
Thanks.  Through a quick glance, I see no obvious errors; the system switches from the simple fb to vc4 around 60 secs after boot, and it probed the HDMI monitor, set up 1920x1080.

Can you try with another monitor?  I'm not sure whether it's a problem of the driver or the combination of the hardware.
Comment 16 Ivan Ivanov 2024-02-07 18:42:18 UTC
This happens to me too. Regular Samsung monitor. I was thinking if this
is some kind of crash in the login manager... but I haven't had time to
look into this further. Note: this is 32 bit system.
Comment 17 Kasimir _ 2024-02-07 19:45:27 UTC
Created attachment 872552 [details]
dmesg from Snapshot20240131 with boot options on Dell screen

The same thing happens on a Dell (P2213f) and an HP (E223) monitor.
Comment 18 Kasimir _ 2024-02-07 19:45:58 UTC
Created attachment 872553 [details]
dmesg from Snapshot20240131 with boot options on HP screen
Comment 19 Ivan Ivanov 2024-02-16 15:03:18 UTC
After 2 weeks uptime I see this, could it be somehow related?

...
[drm:vc4_bo_create [vc4]] *ERROR* Failed to allocate from GEM DMA helper:
vc4-drm soc:gpu: [drm]                            V3D:  33420kb BOs (35)
vc4-drm soc:gpu: [drm]                     V3D shader:    128kb BOs (31)
vc4-drm soc:gpu: [drm]                           dumb:   4116kb BOs (5)
vc4-drm soc:gpu: [drm]                         binner:  16384kb BOs (1)
vc4-drm soc:gpu: [drm]                total purged BO:  27352kb BOs (56)
...
Comment 20 Kasimir _ 2024-02-29 17:38:02 UTC
With the current image [1] the problem still exists, the screen ultimately remains black.

[1] https://download.opensuse.org/ports/armv7hl/tumbleweed/appliances/openSUSE-Tumbleweed-ARM-X11-raspberrypi2.armv7l-2024.01.16-Snapshot20240225.raw.xz