VM Guests Hang While Rebooting Due to KVM Clock Bug

[nutznboltz]

SRU Justification:

Impact: Various Qemu/KVM guests hang while attempting to reboot due to a missing kvm clock kernel patch.

Testcase: Some guests are more susceptible to this issue. A 32-bit RHEL-5 (or compatible) 2-CPU VM guest hangs 100% of the time (fully reproducible.) I have personally seen Ubuntu VM guests hang too but this is much harder to reproduce.

Fix: A one line patch from 2.6.39

diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 980317a..0556e05 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -2100,8 +2100,8 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
   if (check_tsc_unstable()) {
    kvm_x86_ops->adjust_tsc_offset(vcpu, -tsc_delta);
    vcpu->arch.tsc_catchup = 1;
- kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu);
   }
+ kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu);
   if (vcpu->cpu != cpu)
    kvm_migrate_timers(vcpu);
   vcpu->cpu = cpu;

This is a resubmission of LP: #795717

See also http://patchwork.ozlabs.org/patch/109436/

See also LP: #882579

ProblemType: Bug
DistroRelease: Ubuntu 11.04
Package: linux-image-2.6.38-12-generic 2.6.38-12.51
ProcVersionSignature: Ubuntu 2.6.38-12.51-generic 2.6.38.8
Uname: Linux 2.6.38-12-generic x86_64
AlsaVersion: Advanced Linux Sound Architecture Driver Version 1.0.23.
Architecture: amd64
ArecordDevices:
 **** List of CAPTURE Hardware Devices ****
 card 0: SB [HDA ATI SB], device 0: ALC892 Analog [ALC892 Analog]
   Subdevices: 1/1
   Subdevice #0: subdevice #0
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/controlC0: ksta 2106 F.... pulseaudio
CRDA: Error: [Errno 2] No such file or directory
Card0.Amixer.info:
 Card hw:0 'SB'/'HDA ATI SB at 0xfe4f4000 irq 16'
   Mixer name : 'Realtek ALC892'
   Components : 'HDA:10ec0892,104383c0,00100302'
   Controls : 32
   Simple ctrls : 18
Card1.Amixer.info:
 Card hw:1 'HDMI'/'HDA ATI HDMI at 0xfe6e8000 irq 19'
   Mixer name : 'ATI RS690/780 HDMI'
   Components : 'HDA:1002791a,00791a00,00100000'
   Controls : 4
   Simple ctrls : 1
Card1.Amixer.values:
 Simple mixer control 'IEC958',0
   Capabilities: pswitch pswitch-joined penum
   Playback channels: Mono
   Mono: Playback [on]
Date: Wed Nov 2 07:54:16 2011
HibernationDevice: RESUME=UUID=d12bbb7b-b9a4-425c-86a7-c50665bf7833
InstallationMedia: Ubuntu 10.10 "Maverick Meerkat" - Release amd64 (20101007)
IwConfig:
 lo no wireless extensions.

 eth0 no wireless extensions.

 virbr0 no wireless extensions.
MachineType: System manufacturer System Product Name
ProcEnviron:
 LANGUAGE=en_US:en
 PATH=(custom, user)
 LANG=en_US.UTF-8
 SHELL=/bin/bash
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-2.6.38-12-generic root=/dev/mapper/vg0-lv0 ro quiet splash vt.handoff=7
RelatedPackageVersions:
 linux-restricted-modules-2.6.38-12-generic N/A
 linux-backports-modules-2.6.38-12-generic N/A
 linux-firmware 1.52
RfKill:

SourcePackage: linux
UpgradeStatus: Upgraded to natty on 2011-05-26 (159 days ago)
dmi.bios.date: 08/24/2010
dmi.bios.vendor: American Megatrends Inc.
dmi.bios.version: 1606
dmi.board.asset.tag: To Be Filled By O.E.M.
dmi.board.name: M4A89GTD-PRO/USB3
dmi.board.vendor: ASUSTeK Computer INC.
dmi.board.version: Rev 1.xx
dmi.chassis.asset.tag: Asset-1234567890
dmi.chassis.type: 3
dmi.chassis.vendor: Chassis Manufacture
dmi.chassis.version: Chassis Version
dmi.modalias: dmi:bvnAmericanMegatrendsInc.:bvr1606:bd08/24/2010:svnSystemmanufacturer:pnSystemProductName:pvrSystemVersion:rvnASUSTeKComputerINC.:rnM4A89GTD-PRO/USB3:rvrRev1.xx:cvnChassisManufacture:ct3:cvrChassisVersion:
dmi.product.name: System Product Name
dmi.product.version: System Version
dmi.sys.vendor: System manufacturer

[Herton R. Krzesinski (herton)]

@nutznboltz: I'm including the patch again into natty, and probably we will have a new kernel for you to verify next week. I'm marking this bug as duplicate of the original one (bug 795717). Make sure you follow that bug, a new request for verification will be made there. Once the kernel with the fix is in -proposed, I'll send again a message on that bug requesting the kernel to be verified (tested) to fix the problem. If it's not tested, the patch will be reverted again.

[nutznboltz (nutznboltz-deactivatedaccount)]

@herton, thanks for including the patch into Natty! :)