Public bug reported:
In addition to bug #1916775, my Win10 Home guest freezes randomly and
infrequently. Unlike bug #1916775, this is unrecoverable and I see on
the host (Debian 4.19.171-2) via iotop that all disk IO has stopped. My
only recourse is a hard reset of the guest.
My setup uses PCI-pass-through graphics (GTX 1650), host cpu (Ryzen 7
3800XT). It seems to occur more frequently when I plug in 3 monitors
rather than 2 into the pass-through graphics card. It occurs whether or
not I use the qcow disk drive.
qemu-system-x86_64
-cpu
host,kvm=on,l3-cache=on,hv_relaxed,hv_vapic,hv_time,hv_spinlocks=0x1fff,hv_vendor_id=hv_dummy
-smp 8
-rtc clock=host,base=localtime
-machine type=q35,accel=kvm,kernel_irqchip=on
-enable-kvm
-drive if=pflash,format=raw,readonly,file=/usr/share/OVMF/OVMF_CODE.fd
-drive if=pflash,format=raw,file=/tmp/OVMF_VARS.fd
-m 32G
-usb
-device usb-tablet
-vga none
-serial none
-parallel none
-boot cd
-nographic
-device usb-host,vendorid=0x045e,productid=0x00db
-device usb-host,vendorid=0x1bcf,productid=0x0005
-drive
id=disk0,index=0,format=qcow2,if=virtio,cache=off,file=./win10_boot_priv.qcow2
-drive
id=disk2,index=2,aio=native,cache.direct=on,if=virtio,cache=off,format=raw,discard=unmap,detect-zeroes=unmap,file=/dev/vg0/win10_hdpriv
-device vfio-pci,host=09:00.0,addr=0x02.0x0,multifunction=on
-device vfio-pci,host=09:00.1,addr=0x02.0x1
-device vfio-pci,host=09:00.2,addr=0x02.0x2
-device vfio-pci,host=09:00.3,addr=0x02.0x3
-netdev tap,id=netid,ifname=taplan,script=no,downscript=no
-device e1000,netdev=netid,mac=52:54:00:01:02:03
** Affects: qemu
Importance: Undecided
Status: New
** Description changed:
- In addition to bug #1916775, my Win10 Home guest freezes randomly and
infrequently. Unlike bug
- #1916775, this is unrecoverable and I see on the host (Debian 4.19.171-2) via
iotop that all disk IO has stopped. My only recourse is a hard reset of the
guest.
+ In addition to bug #1916775, my Win10 Home guest freezes randomly and
+ infrequently. Unlike bug #1916775, this is unrecoverable and I see on
+ the host (Debian 4.19.171-2) via iotop that all disk IO has stopped. My
+ only recourse is a hard reset of the guest.
My setup uses PCI-pass-through graphics (GTX 1650), host cpu (Ryzen 7
3800XT). It seems to occur more frequently when I plug in 3 monitors
rather than 2 into the pass-through graphics card. It occurs whether or
not I use the qcow disk drive.
- qemu-system-x86_64
- -cpu
host,kvm=on,l3-cache=on,hv_relaxed,hv_vapic,hv_time,hv_spinlocks=0x1fff,hv_vendor_id=hv_dummy
- -smp 8
- -rtc clock=host,base=localtime
- -machine type=q35,accel=kvm,kernel_irqchip=on
- -enable-kvm
- -drive if=pflash,format=raw,readonly,file=/usr/share/OVMF/OVMF_CODE.fd
- -drive if=pflash,format=raw,file=/tmp/OVMF_VARS.fd
- -m 32G
- -usb
- -device usb-tablet
- -vga none
- -serial none
- -parallel none
- -boot cd
- -nographic
- -device usb-host,vendorid=0x045e,productid=0x00db
- -device usb-host,vendorid=0x1bcf,productid=0x0005
- -drive
id=disk0,index=0,format=qcow2,if=virtio,cache=off,file=./win10_boot_priv.qcow2
- -drive
id=disk2,index=2,aio=native,cache.direct=on,if=virtio,cache=off,format=raw,discard=unmap,detect-zeroes=unmap,file=/dev/vg0/win10_hdpriv
- -device vfio-pci,host=09:00.0,addr=0x02.0x0,multifunction=on
- -device vfio-pci,host=09:00.1,addr=0x02.0x1
- -device vfio-pci,host=09:00.2,addr=0x02.0x2
- -device vfio-pci,host=09:00.3,addr=0x02.0x3
- -netdev tap,id=netid,ifname=taplan,script=no,downscript=no
- -device e1000,netdev=netid,mac=52:54:00:01:02:03
+ qemu-system-x86_64
+ -cpu
host,kvm=on,l3-cache=on,hv_relaxed,hv_vapic,hv_time,hv_spinlocks=0x1fff,hv_vendor_id=hv_dummy
+ -smp 8
+ -rtc clock=host,base=localtime
+ -machine type=q35,accel=kvm,kernel_irqchip=on
+ -enable-kvm
+ -drive if=pflash,format=raw,readonly,file=/usr/share/OVMF/OVMF_CODE.fd
+ -drive if=pflash,format=raw,file=/tmp/OVMF_VARS.fd
+ -m 32G
+ -usb
+ -device usb-tablet
+ -vga none
+ -serial none
+ -parallel none
+ -boot cd
+ -nographic
+ -device usb-host,vendorid=0x045e,productid=0x00db
+ -device usb-host,vendorid=0x1bcf,productid=0x0005
+ -drive
id=disk0,index=0,format=qcow2,if=virtio,cache=off,file=./win10_boot_priv.qcow2
+ -drive
id=disk2,index=2,aio=native,cache.direct=on,if=virtio,cache=off,format=raw,discard=unmap,detect-zeroes=unmap,file=/dev/vg0/win10_hdpriv
+ -device vfio-pci,host=09:00.0,addr=0x02.0x0,multifunction=on
+ -device vfio-pci,host=09:00.1,addr=0x02.0x1
+ -device vfio-pci,host=09:00.2,addr=0x02.0x2
+ -device vfio-pci,host=09:00.3,addr=0x02.0x3
+ -netdev tap,id=netid,ifname=taplan,script=no,downscript=no
+ -device e1000,netdev=netid,mac=52:54:00:01:02:03
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1925966
Title:
Win10 guest freezes randomly
Status in QEMU:
New
Bug description:
In addition to bug #1916775, my Win10 Home guest freezes randomly and
infrequently. Unlike bug #1916775, this is unrecoverable and I see on
the host (Debian 4.19.171-2) via iotop that all disk IO has stopped.
My only recourse is a hard reset of the guest.
My setup uses PCI-pass-through graphics (GTX 1650), host cpu (Ryzen 7
3800XT). It seems to occur more frequently when I plug in 3 monitors
rather than 2 into the pass-through graphics card. It occurs whether
or not I use the qcow disk drive.
qemu-system-x86_64
-cpu
host,kvm=on,l3-cache=on,hv_relaxed,hv_vapic,hv_time,hv_spinlocks=0x1fff,hv_vendor_id=hv_dummy
-smp 8
-rtc clock=host,base=localtime
-machine type=q35,accel=kvm,kernel_irqchip=on
-enable-kvm
-drive if=pflash,format=raw,readonly,file=/usr/share/OVMF/OVMF_CODE.fd
-drive if=pflash,format=raw,file=/tmp/OVMF_VARS.fd
-m 32G
-usb
-device usb-tablet
-vga none
-serial none
-parallel none
-boot cd
-nographic
-device usb-host,vendorid=0x045e,productid=0x00db
-device usb-host,vendorid=0x1bcf,productid=0x0005
-drive
id=disk0,index=0,format=qcow2,if=virtio,cache=off,file=./win10_boot_priv.qcow2
-drive
id=disk2,index=2,aio=native,cache.direct=on,if=virtio,cache=off,format=raw,discard=unmap,detect-zeroes=unmap,file=/dev/vg0/win10_hdpriv
-device vfio-pci,host=09:00.0,addr=0x02.0x0,multifunction=on
-device vfio-pci,host=09:00.1,addr=0x02.0x1
-device vfio-pci,host=09:00.2,addr=0x02.0x2
-device vfio-pci,host=09:00.3,addr=0x02.0x3
-netdev tap,id=netid,ifname=taplan,script=no,downscript=no
-device e1000,netdev=netid,mac=52:54:00:01:02:03
To manage notifications about this bug go to:
https://bugs.launchpad.net/qemu/+bug/1925966/+subscriptions
