Bug 924500

Summary: kvm f19 guest reports "BUG: bad unlock balance detected" (backtrace) when using vnc/cirrus on f17 host
Product: [Fedora] Fedora Reporter: Reartes Guillermo <rtguille>
Component: xorg-x11-drv-cirrusAssignee: X/OpenGL Maintenance List <xgl-maint>
Status: CLOSED WORKSFORME QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: medium Docs Contact:
Priority: unspecified    
Version: 19CC: awilliam, robatino, xgl-maint
Target Milestone: ---   
Target Release: ---   
Hardware: x86_64   
OS: Linux   
Whiteboard: RejectedBlocker
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2013-05-15 03:58:02 UTC Type: Bug
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
syslog file
none
X.org log file
none
anaconda.log
none
program.log
none
screenshot none

Description Reartes Guillermo 2013-03-21 22:15:47 UTC
Created attachment 714124 [details]
syslog file

Description of problem:

A back-trace is generated during the boot process. The generated back-trace does not break the boot process. The welcome screen is reached.

Version-Release number of selected component (if applicable):
F19a TC1 (anaconda 19.12)

How reproducible:
always

Steps to Reproduce:
1. Boot Fedora-19-Alpha-TC1-x86_64-DVD.iso
2. the backtrace is generated, boot process continues
3. welcome screen is reached.
  
Actual results:
A backtrace, i can reach the welcome screen.

Expected results:
no back-traces

Additional info:

22:00:57,157 INFO kernel:[    8.872536] Console: switching to colour frame buffer device 160x64
22:00:57,157 INFO kernel:[    9.038882] cirrus 0000:00:02.0: fb0: cirrusdrmfb frame buffer device
22:00:57,157 INFO kernel:[    9.038884] cirrus 0000:00:02.0: registered panic notifier
22:00:57,157 INFO kernel:[    9.044168] [drm] Initialized cirrus 1.0.0 20110418 for 0000:00:02.0 on minor 0
22:00:57,157 WARNING kernel:[    9.052222] 
22:00:57,157 WARNING kernel:[    9.052316] =====================================
22:00:57,157 WARNING kernel:[    9.052542] [ BUG: bad unlock balance detected! ]
22:00:57,157 WARNING kernel:[    9.052779] 3.9.0-0.rc3.git0.4.fc19.x86_64 #1 Not tainted
22:00:57,157 WARNING kernel:[    9.053022] -------------------------------------
22:00:57,157 WARNING kernel:[    9.053022] plymouthd/283 is trying to release lock (&crtc->mutex) at:
22:00:57,157 WARNING kernel:[    9.053022] [<ffffffff81719c8e>] mutex_unlock+0xe/0x10
22:00:57,157 WARNING kernel:[    9.053022] but there are no more locks to release!
22:00:57,157 WARNING kernel:[    9.053022] 
22:00:57,157 WARNING kernel:[    9.053022] other info that might help us debug this:
22:00:57,157 WARNING kernel:[    9.053022] no locks held by plymouthd/283.
22:00:57,157 WARNING kernel:[    9.053022] 
22:00:57,157 WARNING kernel:[    9.053022] stack backtrace:
22:00:57,157 WARNING kernel:[    9.053022] Pid: 283, comm: plymouthd Not tainted 3.9.0-0.rc3.git0.4.fc19.x86_64 #1
22:00:57,157 WARNING kernel:[    9.053022] Call Trace:
22:00:57,157 WARNING kernel:[    9.053022]  [<ffffffff81719c8e>] ? mutex_unlock+0xe/0x10
22:00:57,157 WARNING kernel:[    9.053022]  [<ffffffff810d6b5c>] print_unlock_imbalance_bug+0xec/0x100
22:00:57,157 WARNING kernel:[    9.053022]  [<ffffffff81719c8e>] ? mutex_unlock+0xe/0x10
22:00:57,157 WARNING kernel:[    9.053022]  [<ffffffff810db58e>] lock_release+0x25e/0x310
22:00:57,157 WARNING kernel:[    9.053022]  [<ffffffff81719bab>] __mutex_unlock_slowpath+0x9b/0x170
22:00:57,157 WARNING kernel:[    9.053022]  [<ffffffff81719c8e>] mutex_unlock+0xe/0x10
22:00:57,157 WARNING kernel:[    9.053022]  [<ffffffffa01af579>] drm_modeset_unlock_all+0x39/0x60 [drm]
22:00:57,157 WARNING kernel:[    9.053022]  [<ffffffffa01b3c64>] drm_mode_addfb+0x1a4/0x1b0 [drm]
22:00:57,157 WARNING kernel:[    9.053022]  [<ffffffffa01a3421>] drm_ioctl+0x501/0x5c0 [drm]
22:00:57,157 WARNING kernel:[    9.053022]  [<ffffffffa01b3ac0>] ? drm_mode_cursor_ioctl+0x130/0x130 [drm]
22:00:57,157 WARNING kernel:[    9.053022]  [<ffffffff812f25f6>] ? inode_has_perm.isra.48.constprop.64+0x56/0x80
22:00:57,157 WARNING kernel:[    9.053022]  [<ffffffff811edd35>] do_vfs_ioctl+0x305/0x530
22:00:57,157 WARNING kernel:[    9.053022]  [<ffffffff812f29db>] ? selinux_file_ioctl+0x5b/0x110
22:00:57,157 WARNING kernel:[    9.053022]  [<ffffffff811edfe1>] sys_ioctl+0x81/0xa0
22:00:57,157 WARNING kernel:[    9.053022]  [<ffffffff81726a19>] system_call_fastpath+0x16/0x1b
22:00:57,157 INFO kernel:[    9.185094]  vdb: unknown partition table

Comment 1 Reartes Guillermo 2013-03-21 22:16:16 UTC
Created attachment 714126 [details]
X.org log file

Comment 2 Reartes Guillermo 2013-03-21 22:16:42 UTC
Created attachment 714127 [details]
anaconda.log

Comment 3 Reartes Guillermo 2013-03-21 22:17:07 UTC
Created attachment 714128 [details]
program.log

Comment 4 Reartes Guillermo 2013-03-21 22:20:35 UTC
Created attachment 714129 [details]
screenshot

Comment 5 Adam Williamson (Red Hat non-Fedora) 2013-03-23 01:53:54 UTC
The criteria only require virtualization to work at Beta, and then only N on N and N on N-1, not N on N-2. For me, both 19 on 19 and 19 on 18 work at present. So I'm -1 to this.

Comment 6 Adam Williamson (Red Hat non-Fedora) 2013-03-27 17:45:21 UTC
Discussed at 2013-03-27 blocker review meeting: http://meetbot.fedoraproject.org/fedora-blocker-review/2013-03-27/f19alpha-blocker-review-3.2013-03-27-16.01.log.txt . Rejected as a blocker: virt is not required for Alpha, qxl is preferred to cirrus, and N-on-N-2 is not a required config anyway...

Comment 7 Adam Williamson (Red Hat non-Fedora) 2013-05-14 19:13:53 UTC
Assigning to correct component. FWIW, airlied/ajax say we *really* want to get people off using Cirrus for VMs, qxl is much better supported; please switch to qxl/Spice if you can.

Comment 8 Reartes Guillermo 2013-05-15 02:58:58 UTC
> FWIW, airlied/ajax say we *really* want to get people off using
> Cirrus for VMs, qxl is much better supported; please switch to 
> qxl/Spice if you can.

@Adam

I am aware of it, all my stuff uses spice/qlx since quite some time, and yes, it is much better.

I was just testing cirrus.

I tested it again with F19b TC4 and i did not see it.
It appears this can be closed.

Comment 9 Adam Williamson (Red Hat non-Fedora) 2013-05-15 03:58:02 UTC
Great, thanks for testing.