New: Starting a virtual machine ater suspend causes the host system hangup

https://bugzilla.kernel.org/show_bug.cgi?id=60629

            Bug ID: 60629
           Summary: Starting a virtual machine ater suspend causes the
                    host system hangup
           Product: Virtualization
           Version: unspecified
    Kernel Version: 2.6.35-32 - 3.9.9-1
          Hardware: x86-64
                OS: Linux
              Tree: Mainline
            Status: NEW
          Severity: high
          Priority: P1
         Component: kvm
          Assignee: [email protected]
          Reporter: [email protected]
        Regression: No

HW:
MB: Gigabyte EP43UD3l
CPU: CdQ Q8400
RAM: 4Gb

uname -a
Linux hostname 3.9.9-1-ARCH #1 SMP PREEMPT Wed Jul 3 22:45:16 CEST 2013 x86_64 GNU/Linux

I have the same problem as in
http://www.linux.org.ru/forum/linux-hardware/7455179
so just google-translate:

To get started: Ubuntu 10.10 64-bit, kernel 2.6.35-32 64-bit, Intel (R) Core (TM) 2 Quad CPU [email protected], motherboard gigabyte ep43-ds3l So this processor supports virtualization, with 3 - 3,5 Gb ram everything works correctly with no problems and in 32 and 64 bit. The problems begin when put 4Gb and more virtualization-enabled in the BIOS, there is no problem if virtualization-disabled. The problem: When the virtualization and 4gb of RAM or higher, after the computer from sleep (suspend-to-ram), and run Virtualbox or Vmware, comes complete frieze, or via ssh or in any other way, sysrq reaction too no. An educated bet discovered that before sleep:

cat /proc/cpuinfo | grep 'model name'
model name : Intel(R) Core(TM)2 Quad CPU Q8300 @ 2.50GHz
model name : Intel(R) Core(TM)2 Quad CPU Q8300 @ 2.50GHz
model name : Intel(R) Core(TM)2 Quad CPU Q8300 @ 2.50GHz
model name : Intel(R) Core(TM)2 Quad CPU Q8300 @ 2.50GHz

After sleep:

cat /proc/cpuinfo | grep 'model name'
model name : Intel(R) Core(TM)2 Quad CPU Q8300 @ 2.50GHz
model name : 06/17
model name : 06/17
model name : 06/17

I.e. the model name changed to something strange, turning off virtualization (RAM 4gb and above) it does not, and does not occur at 3Gb RAM and enabled, and off virtualization. Part of the problem was solved with the help of a script placed in / etc / pm / sleep.d /

#!/bin/sh
case $1 in
hibernate|suspend)
echo 0 > /sys/devices/system/cpu/cpu1/online
echo 0 > /sys/devices/system/cpu/cpu2/online
echo 0 > /sys/devices/system/cpu/cpu3/online
;;
thaw|resume)
echo 0 > /sys/devices/system/cpu/cpu1/online
echo 0 > /sys/devices/system/cpu/cpu2/online
echo 0 > /sys/devices/system/cpu/cpu3/online
echo 1 > /sys/devices/system/cpu/cpu1/online
echo 1 > /sys/devices/system/cpu/cpu2/online
echo 1 > /sys/devices/system/cpu/cpu3/online
/etc/init.d/microcode.ctl
echo 0 > /sys/devices/system/cpu/cpu1/online
echo 0 > /sys/devices/system/cpu/cpu2/online
echo 0 > /sys/devices/system/cpu/cpu3/online
echo 1 > /sys/devices/system/cpu/cpu1/online
echo 1 > /sys/devices/system/cpu/cpu2/online
echo 1 > /sys/devices/system/cpu/cpu3/online
;;
esac

(But, model anyway '06/17')

But there was a nasty thing if Vmware or Virtualbox left running, then after waking up again frieze, the script does not have time to work.

You are receiving this mail because:
You are watching the assignee of the bug.

https://bugzilla.kernel.org/show_bug.cgi?id=60629

Alan <[email protected]> changed:

What |Removed |Added
----------------------------------------------------------------------------
CC| |[email protected]

--- Comment #1 from Alan <[email protected]> ---
I'm a bit confused why this bug is filed here, not with virtualbox and vmware

Does it also occur with kvm ?

You are receiving this mail because:
You are watching the assignee of the bug.

https://bugzilla.kernel.org/show_bug.cgi?id=60629

[email protected] changed:

What |Removed |Added
----------------------------------------------------------------------------

             Status|NEW                         |RESOLVED
         Resolution|---                         |INVALID

--- Comment #2 from [email protected] ---
Problem solved by updating CPU microcode in BIOS ROM. Somewhat updating microcode by kernel not solves teh problem.
Now after suspend i have all my CPU cores have right model names. Sorry for that "bug".

You are receiving this mail because:
You are watching the assignee of the bug.