Bug 429412

Summary: System OOPS on soundserver crash
Product: [Fedora] Fedora Reporter: milady <kami_amaterasu>
Component: kernelAssignee: Kernel Maintainer List <kernel-maint>
Status: CLOSED CURRENTRELEASE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: urgent Docs Contact:
Priority: low    
Version: 8CC: jks, ma, than
Target Milestone: ---   
Target Release: ---   
Hardware: i686   
OS: Linux   
Whiteboard:
Fixed In Version: 2.6.23.15-137.fc8 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2008-02-11 22:39:14 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Attachments:
Description Flags
Soundserver crash window
none
Soundserver backtrace
none
kernel OOPS trace none

Description milady 2008-01-19 15:13:09 UTC
Description of problem:
The arts soundserver crashes at the beginning of the KDE session. The moment I
click OK to close the window informing the crash, the system freezes (I think
it's a kernel oops) and it stops responding even to keyboard commands, I am
forced to shut it down using the power off button of the computer.


Version-Release number of selected component (if applicable):
kernel-2.6.23.9-85.fc8

How reproducible:
Always

Steps to Reproduce:
1.Start KDE session
2.
3.
  
Actual results:
System freeze

Expected results:
Working KDE session

Additional info:

Comment 1 milady 2008-01-19 15:13:09 UTC
Created attachment 292258 [details]
Soundserver crash window

Comment 2 milady 2008-01-19 15:13:53 UTC
Created attachment 292259 [details]
Soundserver backtrace

Comment 3 milady 2008-01-19 15:14:59 UTC
Created attachment 292260 [details]
kernel OOPS trace

If I leave the soundserver crash window alone, I can use the system just fine.
As soon as I try to close it or look at the backtrace, the kernel freezes.

Comment 4 milady 2008-01-21 15:57:17 UTC
Same as bug 391971, except mine triggers a kernel OOPS.

Comment 5 Chuck Ebbert 2008-01-21 20:00:34 UTC
Page protection violation.
Writing to a read-only page.

00000000 <.text>:
   0:   00 57 56                add    %dl,0x56(%edi)
   3:   89 d6                   mov    %edx,%esi
   5:   53                      push   %ebx
   6:   89 c3                   mov    %eax,%ebx
   8:   89 e0                   mov    %esp,%eax
   a:   25 00 f0 ff ff          and    $0xfffff000,%eax
   f:   ff 40 14                incl   0x14(%eax)
  12:   89 da                   mov    %ebx,%edx
  14:   83 c2 04                add    $0x4,%edx
  17:   19 ff                   sbb    %edi,%edi
  19:   39 50 18                cmp    %edx,0x18(%eax)
  1c:   83 df 00                sbb    $0x0,%edi
  1f:   85 ff                   test   %edi,%edi
  21:   ba f2 ff ff ff          mov    $0xfffffff2,%edx
  26:   75 08                   jne    0x30
  28:   89 f0                   mov    %esi,%eax
  2a:   90                      nop
===>
   0:   0f b1 0b                cmpxchg %ecx,(%ebx)
   3:   89 c2                   mov    %eax,%edx
   5:   89 e0                   mov    %esp,%eax
   7:   25 00 f0 ff ff          and    $0xfffff000,%eax
   c:   ff 48 14                decl   0x14(%eax)
   f:   5b                      pop    %ebx
  10:   89 d0                   mov    %edx,%eax
  12:   5e                      pop    %esi
  13:   5f                      pop    %edi
  14:   c3                      ret


Comment 6 Chuck Ebbert 2008-01-21 20:01:03 UTC
(In reply to comment #4)
> Same as bug 391971, except mine triggers a kernel OOPS.

Can you confirm this is a single-processor system?


Comment 7 Chuck Ebbert 2008-01-21 20:09:37 UTC
If you boot with the kernel option "noreplace-smp" does it prevent the kernel
OOPS when this happens?

Comment 8 milady 2008-01-21 20:16:31 UTC
My computer is a single-processor system (2 GHz Intel Pentium M).

Comment 9 milady 2008-01-21 20:21:25 UTC
(In reply to comment #7)
> If you boot with the kernel option "noreplace-smp" does it prevent the kernel
> OOPS when this happens?

No more kernel OOPS with this option. The soundserver still crashes though.

Comment 10 Chuck Ebbert 2008-01-23 00:25:58 UTC
*** Bug 429415 has been marked as a duplicate of this bug. ***

Comment 11 Chuck Ebbert 2008-01-23 00:32:24 UTC
Patch went in kernel 2.6.23.14-118.

Comment 12 Martin J├╝rgens 2008-01-23 14:53:45 UTC
> And can you confirm this happened on a single-processor machine?

Yes, AMD Sempron 3000+ with Cool 'n' Quiet activated on Asrock mainboard.

Comment 13 milady 2008-01-25 19:22:19 UTC
I think I know what may have caused the bug on my system: I'm using a webcam for
which the driver is not yet included in the kernel modules. I need the
OmniVision 519 driver and I compiled the version provided on www.rastageeks.org

Not having the webcam connected at startup solves the soundserver problem and
therefore the kernel oops...

Hope this helps

Comment 14 Chuck Ebbert 2008-02-08 00:19:39 UTC
*** Bug 431823 has been marked as a duplicate of this bug. ***

Comment 15 Fedora Update System 2008-02-11 03:34:02 UTC
kernel-2.6.23.15-137.fc8 has been submitted as an update for Fedora 8

Comment 16 Fedora Update System 2008-02-11 22:38:47 UTC
kernel-2.6.23.15-137.fc8 has been pushed to the Fedora 8 stable repository.  If problems still persist, please make note of it in this bug report.