Bug 31618

Summary:	VM session LXQt freezes abruptly without apparent reason
Product:	Mageia	Reporter:	papoteur <yvesbrungard>
Component:	RPM Packages	Assignee:	Kernel and Drivers maintainers <kernel>
Status:	NEW ---	QA Contact:
Severity:	normal
Priority:	Normal	CC:	cyril.levet0780
Version:	Cauldron
Target Milestone:	---
Hardware:	All
OS:	Linux
Whiteboard:
Source RPM:		CVE:
Status comment:
Attachments:	From journal

Description papoteur 2023-03-02 18:05:31 CET

Description of problem:
The session stops. It stays displayed, but doesn't react to anythink. I have to kill the Virtualbox VM.

This occurred several times.

The Virtual machine is configured with VMSVGA and with 3D acceleration.
Graphic card is Intel:
lspcidrake -v |grep Card
Card:Intel 810 and later: Intel Corporation|HD Graphics 620 [DISPLAY_VGA] (vendor:8086 device:5916 subv:1043 subd:16e0) (rev: 02)

The last session ended with these lines in journal. I tried first to send the close signal, without success:


mars 02 17:01:01 CauldronMV CROND[16647]: (root) CMD (nice -n 19 run-parts --report /etc/cron.hourly)
mars 02 17:01:02 CauldronMV CROND[16646]: (root) CMDEND (nice -n 19 run-parts --report /etc/cron.hourly)
mars 02 17:41:00 CauldronMV kernel: clocksource: timekeeping watchdog on CPU1: acpi_pm wd-wd read-back delay of 71517ns
mars 02 17:41:00 CauldronMV kernel: clocksource: wd-tsc-wd read-back delay of 129904ns, clock-skew test skipped!
mars 02 17:45:01 CauldronMV kernel: vmwgfx 0000:00:02.0: [drm] *ERROR* flip_done timed out
mars 02 17:45:01 CauldronMV kernel: vmwgfx 0000:00:02.0: [drm] *ERROR* [CRTC:40:crtc-0] commit wait timed out
mars 02 17:45:11 CauldronMV kernel: vmwgfx 0000:00:02.0: [drm] *ERROR* flip_done timed out
mars 02 17:45:11 CauldronMV kernel: vmwgfx 0000:00:02.0: [drm] *ERROR* [PLANE:34:plane-0] commit wait timed out
mars 02 17:45:21 CauldronMV systemd-logind[761]: Power key pressed short.
mars 02 17:45:21 CauldronMV kernel: vmwgfx 0000:00:02.0: [drm] *ERROR* flip_done timed out
mars 02 17:45:21 CauldronMV kernel: vmwgfx 0000:00:02.0: [drm] *ERROR* [CRTC:40:crtc-0] commit wait timed out
mars 02 17:45:21 CauldronMV polkitd[738]: Unregistered Authentication Agent for unix-session:1 (system bus name :1.66, object path /org/lxqt/PolicyKit1/AuthenticationAgent, locale fr_FR.UTF>
mars 02 17:45:22 CauldronMV su[3469]: pam_unix(su-l:session): session closed for user root
mars 02 17:45:25 CauldronMV lxqt-session[2866]: lxqt-session: Window Manager won't terminate ... killing.
mars 02 17:45:25 CauldronMV systemd-logind[761]: The system will power off now!
mars 02 17:45:25 CauldronMV systemd-logind[761]: System is powering down.

Comment 1 Lewis Smith 2023-03-02 19:42:40 CET

I use 'real' LxQt often without seeing this, so suspect a VB problem.

Can you say whether this happened just once, or repeatedly?
And give the kernel & VirtualBox versions.

CC: (none) => lewyssmith

Comment 2 papoteur 2023-03-03 08:52:28 CET

Hi Lewis,
For the host:
uname -a
Linux YZenbook.home 5.15.88-desktop-1.mga8 #1 SMP Sat Jan 14 15:00:41 UTC 2023 x86_64 x86_64 x86_64 GNU/Linux
rpm -q virtualbox
virtualbox-7.0.6-1.mga8


For the guest:

uname -a
Linux CauldronMV 6.1.14-desktop-1.mga9 #1 SMP PREEMPT_DYNAMIC Mon Feb 27 20:47:10 UTC 2023 x86_64 GNU/Linux
rpm -q virtualbox-guest-additions
virtualbox-guest-additions-7.0.6-2.mga9

Comment 3 papoteur 2023-03-03 09:07:08 CET

Created attachment 13725 [details]
From journal

I did an extract from the journal when the keyword "flip_done" occured.
It shows that the problem occurred at least 4 times followed by completed restart.
I don't use the VM every day.
I think this is a bug in the Virtualbox graphical driver.

Comment 4 Lewis Smith 2023-03-03 20:40:19 CET

Thanks for the extra info.
In the guest VM which does not work, can you post the output of:
 $ inxi -MSGxx
which should give all the relevant guest system details.

Comment 5 Cyril Levet 2023-03-19 18:24:29 CET

Same problem here with Host Mageia 8 (KDE/Plasma) and guest Mageia Cauldron (KDE/Plasma). I have an Nvidia graphic card using proprietary drivers. Some times the Virtual Machine becomes irresponsive at all.
inxi -MSGxx
System:
  Host: localhost Kernel: 6.2.6-desktop-1.mga9 arch: x86_64 bits: 64
    compiler: gcc v: 12.2.1 Desktop: KDE Plasma v: 5.27.3 tk: Qt v: 5.15.7
    wm: kwin_x11 dm: SDDM Distro: Mageia 9
Machine:
  Type: Virtualbox System: innotek GmbH product: VirtualBox v: 1.2
    serial: <superuser required> Chassis: Oracle Corporation type: 1
    serial: <superuser required>
  Mobo: Oracle model: VirtualBox v: 1.2 serial: <superuser required>
    BIOS: innotek GmbH v: VirtualBox date: 12/01/2006
Graphics:
  Device-1: VMware SVGA II Adapter driver: vmwgfx v: 2.20.0.0 ports:
    active: Virtual-1 empty: Virtual-2, Virtual-3, Virtual-4, Virtual-5,
    Virtual-6, Virtual-7, Virtual-8 bus-ID: 00:02.0 chip-ID: 15ad:0405
  Display: x11 server: X.org v: 1.21.1.7 with: Xwayland v: 22.1.8
    compositor: kwin_x11 driver: X: loaded: v4l,vmware dri: vmwgfx gpu: vmwgfx
    display-ID: :0 screens: 1
  Screen-1: 0 s-res: 1920x954 s-dpi: 96
  Monitor-1: Virtual-1 mapped: Virtual1 res: 1920x954 size: N/A
  API: OpenGL v: 3.3 Mesa 23.0.0 renderer: SVGA3D; build: RELEASE; LLVM;
    direct-render: Yes

CC: (none) => cyril.levet0780

Comment 6 Lewis Smith 2023-03-21 08:57:31 CET

Thank you Cyril for the extra information.

(In reply to papoteur from comment #3)
> I think this is a bug in the Virtualbox graphical driver.
So passing this to kernel/driver who deal with this sort of problem.

Assignee: bugsquad => kernel
CC: lewyssmith => (none)