Bug 769300

Summary: Strange xen kernel code when installing a vm
Product: [Fedora] Fedora Reporter: Volnei <vcputtini>
Component: kernelAssignee: Kernel Maintainer List <kernel-maint>
Status: CLOSED INSUFFICIENT_DATA QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: medium Docs Contact:
Priority: unspecified    
Version: 16CC: gansalmon, itamar, jonathan, kernel-maint, ketuzsezr, leiwang, lersek, madhu.chinakonda, qguan, qwan
Target Milestone: ---   
Target Release: ---   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2012-05-14 17:09:14 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
dmesg
none
configuration file of vm
none
dmesg results
none
OS dmesg
none
xm-dmesg
none
kernel-dmesg
none
xm-dmesg
none
SO dmesg log file
none
xen dmesg none

Description Volnei 2011-12-20 13:27:16 UTC
Description of problem:

I was installing a window$ vm and the system was quite slow.
as 100% of CPU is being used. For about 30 seconds.
When the system normalized kernel was the message on the console.

What could cause this situation?

Message from syslogd@barney at Dec 20 11:15:09 ...
 kernel:[ 1672.067672] Code: cc 51 41 53 b8 10 00 00 00 0f 05 41 5b 59 c3 cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc 51 41 53 b8 11 00 00 00 0f 05 


Thank you

Version-Release number of selected component (if applicable):


How reproducible:
Is not constant, it happens sometimes

Steps to Reproduce:
1.
2.
3.
  
Actual results:
Message from syslogd@barney at Dec 20 11:15:09 ...
 kernel:[ 1672.067672] Code: cc 51 41 53 b8 10 00 00 00 0f 05 41 5b 59 c3 cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc 51 41 53 b8 11 00 00 00 0f 05 

Expected results:


Additional info:

kernel-3.1.5-6.fc16.x86_64
xen-runtime-4.1.2-2.fc16.x86_64
netxen-firmware-4.0.534-4.fc15.noarch
xen-hypervisor-4.1.2-2.fc16.x86_64
xen-libs-4.1.2-2.fc16.x86_64
xen-4.1.2-2.fc16.x86_64
xen-licenses-4.1.2-2.fc16.x86_64

Comment 1 Konrad Rzeszutek Wilk 2011-12-20 15:37:14 UTC
That implies a bug. Can you attach the full dmesg output, the detailed steps of how you installed it (and the version of Windows). Thanks.

Comment 2 Volnei 2011-12-20 15:47:46 UTC
Created attachment 548871 [details]
dmesg

Comment 3 Volnei 2011-12-20 15:48:32 UTC
Created attachment 548872 [details]
configuration file of vm

Comment 4 Volnei 2011-12-20 15:52:23 UTC
I'm not doing anything special.
I created a LVM volume and installed windows just that.

Windows XP SP2 32bits

I'll now install the W7-32bit to see if the same problem occurs.

Comment 5 Volnei 2011-12-20 15:57:36 UTC
I forgot a few details:
Sometimes it happened when I was installing a program under Windows, but not always.

Thanks

Comment 6 Volnei 2011-12-20 18:02:41 UTC
Created attachment 548907 [details]
dmesg results

Comment 7 Volnei 2011-12-20 18:08:24 UTC
Hi,

This bug occurs whenever I make a massive processing.
For example was installing an SDK and the bug occurred several times.

This test done using the kernel-3.1.5-2.fc16.x86_64
and the problem is the same as in kernel-3.1.5-6.fc16.x86_64



thank you

Comment 8 Konrad Rzeszutek Wilk 2011-12-20 20:33:04 UTC
Volnei,

Thanks for uploading all those logs. Looking at the kernel stack it looks as if the kernel detects processes being stuck for a very lengthy time doing something. Most of that something is with a hypercall, meaning that the xen hypervisor is doing something for a very long time - which it should not do.

Can you on the xen hypervisor line add 'loglvl=all guest_loglvl=all' and when see those errors again, run 'xm dmesg' and attach the output here? Also run 'xm debug-keys \*' and that should print tons of detailed information.

The other thing that I wonder if you could is try installing the PV drivers. There are GPL version of them: http://www.meadowcourt.org/downloads/ [and here is a good link in case you run in trouble:http://www.jolokianet.net/70Knowledge/Virtualization/Open_Source_Windows_Paravirtualization_Drivers_for_Xen]

Thanks.

Comment 9 Volnei 2011-12-21 10:02:21 UTC
Ok.

Two questions:

1) Where I put the parameters (loglvl=all guest_loglvl=all')? Already that I use the kernel = "/usr/lib/xen/boot/hvmloader"?

2) In case, I'm using the XP w/SP3. Wich driver I should install?
   I install gplpv_xp_0.11.0.308.msi is this the right?


Thanks

Comment 10 Volnei 2011-12-21 19:06:11 UTC
Created attachment 549070 [details]
OS dmesg

Comment 11 Volnei 2011-12-21 19:06:43 UTC
Created attachment 549072 [details]
xm-dmesg

Comment 12 Volnei 2011-12-21 19:13:59 UTC
Hi,

I was compiling two programs simultaneously in VM XP, and the problem
has occurred.The VM was using only 1 vcpu.

By curiosity I reconfigured the VM to use 2 vcpus and redid the compilations and the problem didn't happen.

I already had installed the gplpv_XP_0.11..0.308.msi

I hope that this information can help you

Thanks

Comment 13 Volnei 2012-01-04 12:56:18 UTC
Created attachment 550675 [details]
kernel-dmesg

Comment 14 Volnei 2012-01-04 12:56:48 UTC
Created attachment 550676 [details]
xm-dmesg

Comment 15 Volnei 2012-01-04 13:01:38 UTC
Hi,

New bug messages when I ran a compilation on vm windows XP 32bits SP3.
Depends on of the charge of system, this error back to happens.

Please see new attachments.

Thank you

Comment 16 Volnei 2012-02-10 11:50:37 UTC
Created attachment 560880 [details]
SO dmesg log file

Comment 17 Volnei 2012-02-10 11:51:09 UTC
Created attachment 560881 [details]
xen dmesg

Comment 18 Volnei 2012-02-10 11:54:37 UTC
The problem persists. 

What are the possible causes of this bug?

Attached new log files

Thank you


netxen-firmware-4.0.534-4.fc15.noarch
xen-licenses-4.1.2-4.fc16.x86_64
xen-libs-4.1.2-4.fc16.x86_64
xen-hypervisor-4.1.2-4.fc16.x86_64
xen-4.1.2-4.fc16.x86_64
xen-runtime-4.1.2-4.fc16.x86_64
kernel-3.2.3-2.fc16.x86_64

Comment 19 Dave Jones 2012-03-22 17:12:48 UTC
[mass update]
kernel-3.3.0-4.fc16 has been pushed to the Fedora 16 stable repository.
Please retest with this update.

Comment 20 Dave Jones 2012-03-22 17:15:03 UTC
[mass update]
kernel-3.3.0-4.fc16 has been pushed to the Fedora 16 stable repository.
Please retest with this update.

Comment 21 Dave Jones 2012-03-22 17:24:14 UTC
[mass update]
kernel-3.3.0-4.fc16 has been pushed to the Fedora 16 stable repository.
Please retest with this update.