2012373 – [RHEL8.6.0][qemu6.1] Boot win2022 guest very slowly with 'qemu64' model with big memory on Milan host

RHEL Engineering is moving the tracking of its product development work on RHEL 6 through RHEL 9 to Red Hat Jira (issues.redhat.com). If you're a Red Hat customer, please continue to file support cases via the Red Hat customer portal. If you're not, please head to the "RHEL project" in Red Hat Jira and file new tickets here. Individual Bugzilla bugs in the statuses "NEW", "ASSIGNED", and "POST" are being migrated throughout September 2023. Bugs of Red Hat partners with an assigned Engineering Partner Manager (EPM) are migrated in late September as per pre-agreed dates. Bugs against components "kernel", "kernel-rt", and "kpatch" are only migrated if still in "NEW" or "ASSIGNED". If you cannot log in to RH Jira, please consult article #7032570. That failing, please send an e-mail to the RH Jira admins at rh-issues@redhat.com to troubleshoot your issue as a user management inquiry. The email creates a ServiceNow ticket with Red Hat. Individual Bugzilla bugs that are migrated will be moved to status "CLOSED", resolution "MIGRATED", and set with "MigratedToJIRA" in "Keywords". The link to the successor Jira issue will be found under "Links", have a little "two-footprint" icon next to it, and direct you to the "RHEL project" in Red Hat Jira (issue links are of type "https://issues.redhat.com/browse/RHEL-XXXX", where "X" is a digit). This same link will be available in a blue banner at the top of the page informing you that that bug has been migrated.

Bug 2012373 - [RHEL8.6.0][qemu6.1] Boot win2022 guest very slowly with 'qemu64' model with big memory on Milan host

Summary: [RHEL8.6.0][qemu6.1] Boot win2022 guest very slowly with 'qemu64' model with ...

Keywords:
Status:	CLOSED WONTFIX
Alias:	None
Product:	Red Hat Enterprise Linux 8
Classification:	Red Hat
Component:	qemu-kvm
Sub Component:
Version:	8.6
Hardware:	x86_64
OS:	Linux
Priority:	unspecified
Severity:	low
Target Milestone:	rc
Target Release:	---
Assignee:	Virtualization Maintenance
QA Contact:	liunana
Docs Contact:	Jiri Herrmann
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+	depends on / blocked

Reported:	2021-10-09 02:11 UTC by liunana
Modified:	2023-10-26 02:17 UTC (History)
CC List:	7 users (show)
Fixed In Version:
Doc Type:	Known Issue
Doc Text:	.Windows Server 2022 guests in some cases boot very slowly on AMD Milan Virtual machines (VMs) that use the Windows Server 2022 guest operating system and the `qemu64` CPU model currently take a very long time to boot on hosts with an AMD EPYC 7003 series processor (also known as `AMD Milan`). To work work around the problem, do not use `qemu64` as the CPU model, because it is an unsupported setting for VMs in RHEL 8. For example, use the `host-model` CPU instead.
Clone Of:
Environment:
Last Closed:	2023-04-09 07:28:00 UTC
Type:	---
Target Upstream Version:
Embargoed:
Dependent Products:
Flags:	nanliu: needinfo- nanliu: needinfo-

Attachments	(Terms of Use)

Links
System	ID	Private	Priority	Status	Summary	Last Updated
Red Hat Issue Tracker	RHELPLAN-99349	0	None	None	None	2021-10-09 02:15:32 UTC

Description liunana 2021-10-09 02:11:03 UTC

Description of problem:
Booting Win2022 guest will take a long time (about 15 minutes) with configurations "'-cpu qemu64' + -m 124928(half memory of host)" on Milan host.


Version-Release number of selected component (if applicable):
    amd-milan-09.khw1.lab.eng.bos.redhat.com
    4.18.0-345.1.el8.x86_64
    qemu-kvm-6.1.0-2.module+el8.6.0+12861+13975d62.x86_64

How reproducible:
6/6


Steps to Reproduce:
1. boot win2022 guest with -cpu qemu64 + -m 124928(half memory of host) on Milan host, full command [1]
2.
3.

Actual results:
Booting will take a long time and is very slow

Expected results:
Win2022 can boot successfully quickly

Additional info:

[1] full command
/usr/libexec/qemu-kvm \
    -name 'avocado-vt-vm1'  \
    -sandbox on  \
    -machine pc,memory-backend=mem-machine_mem  \
    -nodefaults \
    -device VGA,bus=pci.0,addr=0x2 \
    -m 124928 \
    -object memory-backend-ram,size=124928M,id=mem-machine_mem  \
    -smp 128,maxcpus=128,cores=64,threads=1,dies=1,sockets=2  \
    -cpu 'qemu64',enforce,-mpx,hv_stimer,hv_synic,hv_vpindex,hv_relaxed,hv_spinlocks=0x1fff,hv_vapic,hv_time,hv_frequencies,hv_runtime,hv_tlbflush,hv_reenlightenment,hv_stimer_direct,hv_ipi,+kvm_pv_unhalt \
    -chardev socket,id=chardev_serial0,path=/tmp/serial-serial0,wait=off,server=on \
    -device isa-serial,id=serial0,chardev=chardev_serial0  \
    -device qemu-xhci,id=usb1,bus=pci.0,addr=0x3 \
    -device usb-tablet,id=usb-tablet1,bus=usb1.0,port=1 \
    -device virtio-scsi-pci,id=virtio_scsi_pci0,bus=pci.0,addr=0x4 \
    -blockdev node-name=file_image1,driver=file,auto-read-only=on,discard=unmap,aio=threads,filename=/home/kvm_autotest_root/images/win2022-64-virtio-scsi.qcow2,cache.direct=on,cache.no-flush=off \
    -blockdev node-name=drive_image1,driver=qcow2,read-only=off,cache.direct=on,cache.no-flush=off,file=file_image1 \
    -device scsi-hd,id=image1,drive=drive_image1,write-cache=on \
    -device virtio-net-pci,mac=9a:42:98:3e:f5:d6,id=idou98LU,netdev=id4pr52t,bus=pci.0,addr=0x5  \
    -netdev tap,id=id4pr52t,vhost=on \
    -vnc :0  \
    -rtc base=localtime,clock=host,driftfix=slew  \
    -boot menu=off,order=cdn,once=c,strict=off \
    -enable-kvm \
    -monitor stdio \

Comment 1 liunana 2021-10-09 02:15:20 UTC

Additional info:
[2]
Host mem:
# free -m
              total        used        free      shared  buff/cache   available
Mem:         257488       81960      131269          50       44258      173674


[3]
I can't reproduce this bug with follows combations:
    -cpu qemu64 + -m 8G            boot win2022 quickly
    -cpu EPYC-Milan + -m 124928    boot win2022 quickly
    -cpu host + -m 124928          boot win2022 quickly

Besides, I can't reproduce this bug on Intel SPR host(intel-eaglestream-spr-07.khw1.lab.eng.bos.redhat.com).

[4]
Can't reproduce this issue with 'qemu-kvm-6.0.0-29.module+el8.6.0+12490+ec3e565c.x86_64'.

Comment 2 John Ferlan 2021-10-18 12:51:16 UTC

Assigned to Amnon for initial triage per bz process and age of bug created or assigned to virt-maint without triage.

Comment 3 Dr. David Alan Gilbert 2021-10-25 15:03:01 UTC

I'm not sure qemu64 is a supported cpu type, but still, an interesting bug.
Can you please test this on a Rome or Naples to see if it really is Milan specific?

Comment 4 liunana 2021-10-26 09:48:29 UTC

(In reply to Dr. David Alan Gilbert from comment #3)
> I'm not sure qemu64 is a supported cpu type, but still, an interesting bug.
> Can you please test this on a Rome or Naples to see if it really is Milan
> specific?

It works well on a Rome machine even I boot Win2022 with almost all host memory, Win2022 guest can boot up quickly.
I will check Naples later once I get one.

Test Env:
  4.18.0-348.4.el8.x86_64
  qemu-kvm-6.1.0-2.module+el8.6.0+12861+13975d62.x86_64

# free -m
              total        used        free      shared  buff/cache   available
Mem:          63806       26495       36530          13         781       36653
Swap:         32067           0       32067


QEMU cmlline:
    -machine pc,memory-backend=mem-machine_mem  \
    -m 62832 \
    -object memory-backend-ram,size=62832M,id=mem-machine_mem  \



Please help to check this, thanks.


Best regards
Liu Nana

Comment 5 Dr. David Alan Gilbert 2021-10-26 10:36:31 UTC

But you're running at 62832 Memory there rather than nearly twice that on the milan test; usign 62832 on a milan is that fast or slow?

Comment 6 liunana 2021-10-26 15:14:28 UTC

(In reply to Dr. David Alan Gilbert from comment #5)
> But you're running at 62832 Memory there rather than nearly twice that on
> the milan test; 

Yes, My local Rome host has no big memory, and I can't boot a guest when I overcommit too much memory.
I will find another Rome machine with huge memory to try this.


>usign 62832 on a milan is that fast or slow?

It's fast on Milan.

Comment 21 RHEL Program Management 2023-04-09 07:28:00 UTC

After evaluating this issue, there are no plans to address it further or fix it in an upcoming release.  Therefore, it is being closed.  If plans change such that this issue will be fixed in an upcoming release, then the bug can be reopened.

Comment 22 liunana 2023-07-06 11:59:36 UTC

I still can reproduce this issue on latest RHEL.8.9.
Host info:
    4.18.0-500.el8.x86_64
    qemu-kvm-6.2.0-35.module+el8.9.0+19166+e262ca96.x86_64
    amd-genoa-02.khw1.lab.eng.bos.redhat.com
    Model name:          AMD EPYC 9754 128-Core Processor
Guest: Win2022

# free -m
              total        used        free      shared  buff/cache   available
Mem:         127929       11493      115932          19         503      115575
Swap:          4095           0        4095


qemu commandline:
/usr/libexec/qemu-kvm \
     -name 'avocado-vt-vm1'  \
     -sandbox on  \
     -blockdev node-name=file_ovmf_code,driver=file,filename=/usr/share/OVMF/OVMF_CODE.secboot.fd,auto-read-only=on,discard=unmap \
     -blockdev node-name=drive_ovmf_code,driver=raw,read-only=on,file=file_ovmf_code \
     -blockdev node-name=file_ovmf_vars,driver=file,filename=/home/kar/workspace/root/avocado/data/avocado-vt/avocado-vt-vm1_win2022-64-virtio-scsi-ovmf_qcow2_filesystem_VARS.raw,auto-read-only=on,discard=unmap \
     -blockdev node-name=drive_ovmf_vars,driver=raw,read-only=off,file=file_ovmf_vars \
     -machine q35,pflash0=drive_ovmf_code,pflash1=drive_ovmf_vars,memory-backend=mem-machine_mem,kernel-irqchip=split \
     -device pcie-root-port,id=pcie-root-port-0,multifunction=on,bus=pcie.0,addr=0x1,chassis=1 \
     -device pcie-pci-bridge,id=pcie-pci-bridge-0,addr=0x0,bus=pcie-root-port-0  \
     -nodefaults \
     -device VGA,bus=pcie.0,addr=0x2 \
     -m 64G \
     -object memory-backend-ram,size=64G,id=mem-machine_mem  \
     -smp 224,maxcpus=224,cores=112,threads=1,dies=1,sockets=2  \
     -cpu 'qemu64',enforce,hv_stimer,hv_synic,hv_vpindex,hv_relaxed,hv_spinlocks=0x1fff,hv_vapic,hv_time,hv_frequencies,hv_runtime,hv_tlbflush,hv_reenlightenment,hv_stimer_direct,hv_ipi,kvm_pv_unhalt=on \
     -device pvpanic,ioport=0x505,id=idOR0s3n \
     -chardev socket,server=on,path=/tmp/serial-serial0,wait=off,id=chardev_serial0 \
     -device isa-serial,id=serial0,chardev=chardev_serial0  \
     -device pcie-root-port,id=pcie-root-port-1,port=0x1,addr=0x1.0x1,bus=pcie.0,chassis=2 \
     -device qemu-xhci,id=usb1,bus=pcie-root-port-1,addr=0x0 \
     -device usb-tablet,id=usb-tablet1,bus=usb1.0,port=1 \
     -device pcie-root-port,id=pcie-root-port-2,port=0x2,addr=0x1.0x2,bus=pcie.0,chassis=3 \
     -device virtio-scsi-pci,id=virtio_scsi_pci0,bus=pcie-root-port-2,addr=0x0 \
     -blockdev '{"node-name": "file_image1", "driver": "file", "auto-read-only": true, "discard": "unmap", "aio": "threads", "filename": "/home/kvm_autotest_root/images/win2022-64-virtio-scsi-ovmf.qcow2", "cache": {"direct": true, "no-flush": false}}' \
     -blockdev '{"node-name": "drive_image1", "driver": "qcow2", "read-only": false, "cache": {"direct": true, "no-flush": false}, "file": "file_image1"}' \
     -device scsi-hd,id=image1,drive=drive_image1,write-cache=on \
     -device pcie-root-port,id=pcie-root-port-3,port=0x3,addr=0x1.0x3,bus=pcie.0,chassis=4 \
     -device virtio-net-pci,mac=9a:42:d4:3c:2d:6c,id=idgJ8Q1I,netdev=idw3KDQu,bus=pcie-root-port-3,addr=0x0  \
     -netdev tap,id=idw3KDQu,vhost=on \
     -vnc :0  \
     -rtc base=localtime,clock=host,driftfix=slew  \
     -boot menu=off,order=cdn,once=c,strict=off \
     -enable-kvm \
     -monitor stdio \
     -global mch.extended-tseg-mbytes=48 \
     -chardev file,id=firmware,path=/tmp/edk2.log \
     -device isa-debugcon,iobase=0x402,chardev=firmware \

Hi Bandan,

Would you please help to check this issue? 'qemu64' is still supported in qemu-kvm-6.2.0.
Do we have any plan for this fix? If so I will re-open this bug, thanks.


Best regards
Nana

Comment 23 Bandan Das 2023-07-21 02:41:45 UTC

(In reply to liunana from comment #22)
> I still can reproduce this issue on latest RHEL.8.9.
> Host info:
>     4.18.0-500.el8.x86_64
>     qemu-kvm-6.2.0-35.module+el8.9.0+19166+e262ca96.x86_64
>     amd-genoa-02.khw1.lab.eng.bos.redhat.com
>     Model name:          AMD EPYC 9754 128-Core Processor
> Guest: Win2022
> 
> # free -m
>               total        used        free      shared  buff/cache  
> available
> Mem:         127929       11493      115932          19         503     
> 115575
> Swap:          4095           0        4095
> 
> 
> qemu commandline:
> /usr/libexec/qemu-kvm \
>      -name 'avocado-vt-vm1'  \
>      -sandbox on  \
>      -blockdev
> node-name=file_ovmf_code,driver=file,filename=/usr/share/OVMF/OVMF_CODE.
> secboot.fd,auto-read-only=on,discard=unmap \
>      -blockdev
> node-name=drive_ovmf_code,driver=raw,read-only=on,file=file_ovmf_code \
>      -blockdev
> node-name=file_ovmf_vars,driver=file,filename=/home/kar/workspace/root/
> avocado/data/avocado-vt/avocado-vt-vm1_win2022-64-virtio-scsi-
> ovmf_qcow2_filesystem_VARS.raw,auto-read-only=on,discard=unmap \
>      -blockdev
> node-name=drive_ovmf_vars,driver=raw,read-only=off,file=file_ovmf_vars \
>      -machine
> q35,pflash0=drive_ovmf_code,pflash1=drive_ovmf_vars,memory-backend=mem-
> machine_mem,kernel-irqchip=split \
>      -device
> pcie-root-port,id=pcie-root-port-0,multifunction=on,bus=pcie.0,addr=0x1,
> chassis=1 \
>      -device
> pcie-pci-bridge,id=pcie-pci-bridge-0,addr=0x0,bus=pcie-root-port-0  \
>      -nodefaults \
>      -device VGA,bus=pcie.0,addr=0x2 \
>      -m 64G \
>      -object memory-backend-ram,size=64G,id=mem-machine_mem  \
>      -smp 224,maxcpus=224,cores=112,threads=1,dies=1,sockets=2  \
>      -cpu
> 'qemu64',enforce,hv_stimer,hv_synic,hv_vpindex,hv_relaxed,
> hv_spinlocks=0x1fff,hv_vapic,hv_time,hv_frequencies,hv_runtime,hv_tlbflush,
> hv_reenlightenment,hv_stimer_direct,hv_ipi,kvm_pv_unhalt=on \
>      -device pvpanic,ioport=0x505,id=idOR0s3n \
>      -chardev
> socket,server=on,path=/tmp/serial-serial0,wait=off,id=chardev_serial0 \
>      -device isa-serial,id=serial0,chardev=chardev_serial0  \
>      -device
> pcie-root-port,id=pcie-root-port-1,port=0x1,addr=0x1.0x1,bus=pcie.0,
> chassis=2 \
>      -device qemu-xhci,id=usb1,bus=pcie-root-port-1,addr=0x0 \
>      -device usb-tablet,id=usb-tablet1,bus=usb1.0,port=1 \
>      -device
> pcie-root-port,id=pcie-root-port-2,port=0x2,addr=0x1.0x2,bus=pcie.0,
> chassis=3 \
>      -device
> virtio-scsi-pci,id=virtio_scsi_pci0,bus=pcie-root-port-2,addr=0x0 \
>      -blockdev '{"node-name": "file_image1", "driver": "file",
> "auto-read-only": true, "discard": "unmap", "aio": "threads", "filename":
> "/home/kvm_autotest_root/images/win2022-64-virtio-scsi-ovmf.qcow2", "cache":
> {"direct": true, "no-flush": false}}' \
>      -blockdev '{"node-name": "drive_image1", "driver": "qcow2",
> "read-only": false, "cache": {"direct": true, "no-flush": false}, "file":
> "file_image1"}' \
>      -device scsi-hd,id=image1,drive=drive_image1,write-cache=on \
>      -device
> pcie-root-port,id=pcie-root-port-3,port=0x3,addr=0x1.0x3,bus=pcie.0,
> chassis=4 \
>      -device
> virtio-net-pci,mac=9a:42:d4:3c:2d:6c,id=idgJ8Q1I,netdev=idw3KDQu,bus=pcie-
> root-port-3,addr=0x0  \
>      -netdev tap,id=idw3KDQu,vhost=on \
>      -vnc :0  \
>      -rtc base=localtime,clock=host,driftfix=slew  \
>      -boot menu=off,order=cdn,once=c,strict=off \
>      -enable-kvm \
>      -monitor stdio \
>      -global mch.extended-tseg-mbytes=48 \
>      -chardev file,id=firmware,path=/tmp/edk2.log \
>      -device isa-debugcon,iobase=0x402,chardev=firmware \
> 
> Hi Bandan,
> 
> Would you please help to check this issue? 'qemu64' is still supported in
> qemu-kvm-6.2.0.
> Do we have any plan for this fix? If so I will re-open this bug, thanks.
> 
> 
> Best regards
> Nana

I am not sure whether this is worth fixing.
Can we not document this as a known issue for older qemu releases ?

Comment 26 liunana 2023-07-24 02:23:40 UTC

Hi Jiri,

Would you please help check Bandan's question in Comment 25? Thanks.


Best regards
Nana

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