Bug 830432 - Fedora 17 fails to boot due to a kernel BUG at mm/slub.c:3471! on ASUS P8Z68-V PRO GEN3. kernel 3.4.0-1.fc17 x86_64
Summary: Fedora 17 fails to boot due to a kernel BUG at mm/slub.c:3471! on ASUS P8Z68...
Keywords:
Status: CLOSED DUPLICATE of bug 830359
Alias: None
Product: Fedora
Classification: Fedora
Component: kernel
Version: 17
Hardware: x86_64
OS: Linux
unspecified
urgent
Target Milestone: ---
Assignee: Kernel Maintainer List
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
: 835388 (view as bug list)
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2012-06-09 14:14 UTC by Andrey Arapov
Modified: 2012-07-09 12:48 UTC (History)
12 users (show)

Fixed In Version:
Clone Of:
Environment:
Last Closed: 2012-07-09 12:48:10 UTC
Type: Bug
Embargoed:


Attachments (Terms of Use)
photo/screenshot 1 (3.64 MB, image/jpeg)
2012-06-09 14:14 UTC, Andrey Arapov
no flags Details
photo/screenshot 2 (3.60 MB, image/jpeg)
2012-06-09 14:15 UTC, Andrey Arapov
no flags Details
photo/screenshot 3 (4.04 MB, image/jpeg)
2012-06-15 07:55 UTC, Andrey Arapov
no flags Details
Screenshoot from Boot Variant 1/2 (911.70 KB, image/jpeg)
2012-07-03 19:17 UTC, Florian Sievert
no flags Details
Screenshoot from Boot Variant 2/2 (814.09 KB, image/jpeg)
2012-07-03 19:17 UTC, Florian Sievert
no flags Details

Description Andrey Arapov 2012-06-09 14:14:09 UTC
Created attachment 590613 [details]
photo/screenshot 1

Description of problem:
Fedora 17 fails to boot due to a kernel BUG at mm/slub.c:3471!  on ASUS P8Z68-V PRO GEN3.

Please see attached photos (screenshots) of the problem.


PC Base Board Information
	Manufacturer: ASUSTeK Computer INC.
	Product Name: P8Z68-V PRO GEN3

Version-Release number of selected component (if applicable):
PC ---> DMI: System manufacturer System Product Name/P8Z68-V PRO GEN3, BIOS 3202 02/17/2012
kernel ---> kernel.x86_64                         3.4.0-1.fc17

How reproducible:
Just start booting Linux 3.4.0-1.fc17 (x86_64)

Steps to Reproduce:
1. Start booting Linux 3.4.0-1.fc17 (x86_64)
2. Get a hang (I think even before initrd loads)
  
Actual results:
Linux fails to boot. Was quite hard to get some output on a screen to make a photo (screenshot).

Expected results:
Linux should boot normally.

Additional info:
Please request if any extra details needed about my configuration (HW/SW).

Comment 1 Andrey Arapov 2012-06-09 14:15:40 UTC
Created attachment 590614 [details]
photo/screenshot 2

Comment 2 Eivind Olsen 2012-06-14 21:36:32 UTC
I was about to report this bug as well, but guess it's more helpful if I add my information here..?

I see the same bug with the same kernel, but with a slightly different mainboard - I use the ASUS P8P67 DELUXE Rev 1.xx (BIOS version 2302).
Here's what I found on the screen in manually typed format, I skipped all the registers but have them on a photo if they're really needed.

kernel BUG at mm/slub.c:3471!
invalid opcode: 0000 [#1] SMP
CPU 2
Modules linked in: nouveau mxm_wmi wmi video i2c_algo_bit drm_kms_helper ttm drm i2...

Pid: 259, comm: udevd Not tainted 3.4.0-1.fc17.x86_64 #1 System manufacturer System Product Name...
RIP: 0010:[<ffffffff81169345>]  [<ffffffff81169345>] kfree+0x115/0x120
RSP: 0018:ffff880403bf1e88  EFLAGS: 00010246
RAX, RDX etc...
Process udevd (pid: 259, threadinfo ffff880403bf0000, task ffff880403f90000)
Stack:
...
Call Trace:
 [<ffffffff811ebe4b>] sysfs_release+0x8b/0xb0
 [<ffffffff8117f098>] fput+0x118/0x260
 [<ffffffff8117b2b6>] filp_close+0x66/0xa0
 [<ffffffff8117b57c>] sys_close+0x7c/0xc0
 [<ffffffff815f8ae9>] system_call_fastpath+0x16/0xc0
Code: 89 d9 48 89 da 4c 89 d6 e8 f6 fd 47 00 eb a4 49 f7 02 00 c0 00 00 74 13 4c 89 d7 e8 9...

udevd[185]: worker [259] terminated by signal 11 (Segmentation fault)

udevd[185]: worker [259] failed while handling '/devices/LNXSYSTM:00/LNXCPU:01'

Comment 3 Andrey Arapov 2012-06-15 07:55:29 UTC
Created attachment 592064 [details]
photo/screenshot 3

Adding one more screenshot, system hanged in the same way, right after grub.

Comment 4 Andrey Arapov 2012-06-24 11:20:53 UTC
Absolutely the same behavior with newest kernels also:
kernel-3.4.2-4.fc17.x86_64
kernel-3.4.3-1.fc17.x86_64

System hangs on boot or automatically reboots (in 3.4.3)
Tried "noacpi" kernel argument, didn't help.

The only kernel that can boot on my P8Z68 is kernel-3.3.7-1.fc17.x86_64 currently... (Haven't tried older kernel's)

Also the latest BIOS update didn't help:
[    0.000000] DMI: System manufacturer System Product Name/P8Z68-V PRO GEN3, BIOS 3402 05/07/2012

Comment 5 Sean McKinley 2012-06-26 18:54:57 UTC
*** Bug 835388 has been marked as a duplicate of this bug. ***

Comment 6 Sean McKinley 2012-06-26 19:07:39 UTC
I believe I have the same problem since I came across the same slub bug after I filed my bug. I noticed that we have the same chipset (Intel Z68) and we are both using the nouveau driver (I have a GTX 560 Ti). After I installed the proprietary nvidia driver and the kernel modules for 3.3.4-5 (the only properly working kernel I have installed) and 3.4.3-1 from RPM fusion, the system has been behaving properly. 

It's not a solution, but if that works for you then maybe that will help suggest where the problem lies.

Comment 7 Florian Sievert 2012-07-03 19:15:53 UTC
I also believe, I am having two system being affected by this issue. 

My main system was running F16 and F17 for a while quite stable. Suddenly the system just loading the kernel and seems to die right away resulting in a black screen being completely dead. The issue does also occur after installting F17 again (while the installation itself was fine). For my suprise, the issue also happening with a F16. I am currently running Gentoo without any issues or something indicating a hardware problem. Mainboard is a P8Z68, graphic card is a GTX 560 (Asus, nvidia). I don't get any screen nor are logs created. The issue occurs also when booting the recovery mode.

The second system is at a friend I installed the system about a week ago. After the installation, we booted multiple times, but no issue was observed. Some days laters, my friend got a message similiar to one reported by the others here. (screenshots will be attached). After occuring it does also occur when booting the old kernel. The system is a different mainboard being a Asus P5B, but also having a GTX 560 (Gigabyte, nouveau only).

So reading the other comments, the GTX 560 seems to be a common base. Also the chip seems to be same generation, except the older one from friends computer.

Comment 8 Florian Sievert 2012-07-03 19:17:34 UTC
Created attachment 596077 [details]
Screenshoot from Boot Variant 1/2

Comment 9 Florian Sievert 2012-07-03 19:17:59 UTC
Created attachment 596078 [details]
Screenshoot from Boot Variant 2/2

Comment 10 Andrey Arapov 2012-07-06 06:50:04 UTC
Newest kernel (3.4.4-3.fc17.x86_64) still has the same behavior...
As the workaround to boot your P8Z68 - simply stop using your GTX 560 Ti... In UEFI (formerly BIOS) you can set Primary Display to IGFX (instead of Auto or PCI/PCIE) - this solves the problem.. but GTX 560 Ti will be Not in use :/

Or you can physically remove GTX 560 Ti...

Comment 11 Josh Boyer 2012-07-06 13:32:04 UTC
You might give this kernel a try.  It has one fix for GTX 560 Ti issues:

https://admin.fedoraproject.org/updates/kernel-3.4.4-5.fc17

Comment 12 Andrey Arapov 2012-07-08 22:06:30 UTC
Just updated to kernel-3.4.4-5.fc17 - it works ! This bug can be closed.

Solution:
[root@P8Z68 ~]# yum update -y
[root@P8Z68 ~]# reboot
[root@P8Z68 ~]# rpm -q --changelog kernel-3.4.4-5.fc17 |grep nouveau |head -1
- Fix breakage in nouveau with nv_two_heads (rhbz 830359)

Thank you!

Comment 13 Josh Boyer 2012-07-09 12:48:10 UTC

*** This bug has been marked as a duplicate of bug 830359 ***


Note You need to log in before you can comment on or make changes to this bug.