Bug 1119171

Summary:	The value of steal time in RHEL7.1 guest becomes "0.0 st" after S3
Product:	Red Hat Enterprise Linux 7	Reporter:	huiqingding <huding>
Component:	qemu-kvm	Assignee:	Amit Shah <amit.shah>
Status:	CLOSED WONTFIX	QA Contact:	Virtualization Bugs <virt-bugs>
Severity:	medium	Docs Contact:
Priority:	medium
Version:	7.1	CC:	hhuang, huding, juzhang, rbalakri, virt-maint, xfu
Target Milestone:	rc
Target Release:	---
Hardware:	x86_64
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2015-03-04 06:04:55 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:
Bug Depends On:
Bug Blocks:	923626

Description huiqingding 2014-07-14 08:16:05 UTC

Description of problem:
Boot two RHEL7.1 guests with "-smp 1", bonding cpu of these two guests to one same host cpu. Add 100% cpu load in guest, use top command to check steal time in two guests, found the steal time is about "50%". Do S3 inside guest, after resume, the steal time becomes "0.0"

Version-Release number of selected component (if applicable):
Host side:
kernel-3.10.0-133.el7.x86_64
qemu-kvm-1.5.3-66.el7.x86_64
Guest side:
kernel-3.10.0-133.el7.x86_64

How reproducible:
100%

Steps to Reproduce:
1.Boot two guests using the following commands:

# taskset -c 1 /usr/libexec/qemu-kvm -cpu SandyBridge \
-enable-kvm  -m 2G  -smp 1,maxcpus=160 \
-monitor stdio \
-M pc \
-global PIIX4_PM.disable_s3=0 \
-global PIIX4_PM.disable_s4=0 \
-drive file=/home/rhel7-64-ga.qcow2,if=none,id=drive-scsi-disk,format=qcow2,cache=none,werror=stop,rerror=stop \
-device virtio-scsi-pci,id=scsi0,addr=0x4 \
-device scsi-hd,drive=drive-scsi-disk,bus=scsi0.0,id=data-disk2,bootindex=0 \
-netdev tap,id=hostnet0,vhost=on,id=hostnet0,script=/etc/qemu-ifup \
-device virtio-net-pci,netdev=hostnet0,id=virtio-net-pci0,mac=c2:9a:2f:9c:de:10,bus=pci.0,addr=0x5 \
-qmp tcp:0:4445,server,nowait \
-spice port=5901,password=redhat-vga,disable-ticketing -vga qxl -global qxl-vga.vram_size=33554432

2. Add 100% cpu load in two guests
# for ((;;)) do x=1; done

3. Inside two guests, run top command to check steal time value.

4. Inside two guests, do S3
# pm-suspend

5. After resuming the guest, run top command to check steal time value.

Actual results:
after step3, the steal time of two guests are about 50%
after step5, the steal time of two guests are 0.0

Expected results:
after step5, the steal time of two guests should be about 50%

Additional info:

Comment 1 huiqingding 2014-07-14 08:30:12 UTC

I also test the following version, also hit this problem:
Host side:
kernel-3.10.0-125.el7.x86_64
qemu-kvm-1.5.3-66.el7.x86_64
Guest side:
kernel-3.10.0-125.el7.x86_64

and
Host side:
kernel-3.10.0-124.el7.x86_64
qemu-kvm-1.5.3-66.el7.x86_64
Guest side:
kernel-3.10.0-124.el7.x86_64

Best regards
Huiqing

Comment 2 huiqingding 2014-07-14 08:34:37 UTC

Please note:

I also tried rhel7.0.z build as well, also hit this problem:
Host side:
kernel-3.10.0-123.6.2.el7.x86_64
qemu-kvm-1.5.3-66.el7.x86_64
Guest side:
kernel-3.10.0-123.6.2.el7.x86_64


Best regards
Huiqing