Bug 1340616 - vioscsi: bsod after windows10 install
Summary: vioscsi: bsod after windows10 install
Keywords:
Status: CLOSED NOTABUG
Alias: None
Product: Virtualization Tools
Classification: Community
Component: virtio-win
Version: unspecified
Hardware: x86_64
OS: Linux
unspecified
urgent
Target Milestone: ---
Assignee: Vadim Rozenfeld
QA Contact: Virtualization Bugs
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2016-05-28 22:49 UTC by Michał Zegan
Modified: 2017-08-17 07:26 UTC (History)
12 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2017-08-17 07:26:47 UTC
Embargoed:


Attachments (Terms of Use)

Description Michał Zegan 2016-05-28 22:49:45 UTC
I tried to install windows10 pro on a virtual machine with virtio-scsi device.
However, I seem to have some problems. The installation is always successful. then the system starts, and there is the oob screen. When you wait a while on the oob screen, or, if you managed to go past oob, if you wait on any other screen, the system will crash. The bsod errors I've seen are: "irql not less or equal (tcpipsys)", "unhandled system thread exception (tcpipsys)" and "page fault". It seems that is caused by virtio scsi driver as when I change to virtio blk, those errors do not show up anymore. Usually they happened shortly after showing first OOB screen. There were, however, times when I managed to go through most of oob, and the bsod probably showed up in the middle of the last oob step (after creation of the admin user account). after reboot there was a login screen, I managed to log in, and there was another bsod after few minutes.

Comment 1 Vadim Rozenfeld 2016-05-29 07:49:44 UTC
(In reply to Michał Zegan from comment #0)
> I tried to install windows10 pro on a virtual machine with virtio-scsi
> device.
> However, I seem to have some problems. The installation is always
> successful. then the system starts, and there is the oob screen. When you
> wait a while on the oob screen, or, if you managed to go past oob, if you
> wait on any other screen, the system will crash. The bsod errors I've seen
> are: "irql not less or equal (tcpipsys)", "unhandled system thread exception
> (tcpipsys)" and "page fault". It seems that is caused by virtio scsi driver
> as when I change to virtio blk, those errors do not show up anymore. Usually
> they happened shortly after showing first OOB screen. There were, however,
> times when I managed to go through most of oob, and the bsod probably showed
> up in the middle of the last oob step (after creation of the admin user
> account). after reboot there was a login screen, I managed to log in, and
> there was another bsod after few minutes.

Hi Michał,

Could you please specify kvm and qemu versions as well as qemu command line and virtio-win drivers version (build).

Thanks,
Vadim.

Comment 2 Michał Zegan 2016-05-29 10:13:44 UTC
Well. One update before I provide info: not all bsods I was experiencing were due to vioscsi. those tcpipsys bugs going after oob were not related to it, the bug that was probably vioscsi happens periodically when the system is idle.
info:
qemu version: 2.5
virtio win version: virtio-win-0.1.117
linux kernel version 4.5
qemu command line

/bin/qemu-system-x86_64 \
-name vm -D vm.log \
-spice tls-port=25000,x509-cacert-file=../ca/cacert.pem,x509-cert-file=cert.pem,x509-key-file=key.pem,x509-dh-key-file=dh.pem,password=mypass \
-machine q35,accel=kvm,iommu=on -nodefaults -cpu host \
-chardev stdio,id=interface,mux=on,signal=off \
-chardev socket,id=control,server,nowait,path=control.sock \
-mon chardev=interface,mode=readline -mon chardev=control,mode=control \
-smp cores=4 -m size=2048M,maxmem=4096M,slots=2 \
-device ib700 -device pvpanic -device isa-serial,chardev=interface \
-device virtio-balloon-pci,disable-modern=false \
-device virtio-rng-pci,disable-modern=false \
-device virtio-serial-pci,disable-modern=false \
-chardev socket,id=agent,server,nowait,path=agent.sock \
-device virtserialport,chardev=agent,name=org.qemu.guest_agent.0 \
-chardev spicevmc,name=vdagent,id=vdagent \
-device virtserialport,chardev=vdagent,name=com.redhat.spice.0 \
-device qxl-vga \
-device ich9-intel-hda -device hda-micro \
-netdev bridge,br=vmnet,id=net \
-device virtio-net-pci,netdev=net \
-device nec-usb-xhci -device nec-usb-xhci \
-device virtio-scsi-pci,disable-modern=false \
-chardev spicevmc,name=usbredir,id=usbredir1 \
-device usb-redir,chardev=usbredir1 \
-chardev spicevmc,name=usbredir,id=usbredir2 \
-device usb-redir,chardev=usbredir2 \
-device usb-tablet \
-drive id=cd1,media=cdrom,if=none \
-device ide-cd,drive=cd1,bus=ide.0 \
-drive id=cd2,if=none,media=cdrom \
-device ide-cd,drive=cd2,bus=ide.1 \
-drive id=disk,if=none,media=disk,format=qcow2,file=disk.img \
-device scsi-hd,drive=disk \
-drive id=fwcode,if=pflash,format=raw,file=firmware.bin \
-drive id=fwvars,if=pflash,format=raw,file=vars.bin \
-watchdog-action reset -rtc base=localtime \
-boot menu=on \
$@

Comment 3 Vadim Rozenfeld 2016-05-30 01:33:41 UTC
(In reply to Michał Zegan from comment #2)
> Well. One update before I provide info: not all bsods I was experiencing
> were due to vioscsi. those tcpipsys bugs going after oob were not related to
> it, the bug that was probably vioscsi happens periodically when the system
> is idle.
> info:
> qemu version: 2.5
> virtio win version: virtio-win-0.1.117
> linux kernel version 4.5
> qemu command line
> 
> /bin/qemu-system-x86_64 \
> -name vm -D vm.log \
> -spice
> tls-port=25000,x509-cacert-file=../ca/cacert.pem,x509-cert-file=cert.pem,
> x509-key-file=key.pem,x509-dh-key-file=dh.pem,password=mypass \
> -machine q35,accel=kvm,iommu=on -nodefaults -cpu host \
> -chardev stdio,id=interface,mux=on,signal=off \
> -chardev socket,id=control,server,nowait,path=control.sock \
> -mon chardev=interface,mode=readline -mon chardev=control,mode=control \
> -smp cores=4 -m size=2048M,maxmem=4096M,slots=2 \
> -device ib700 -device pvpanic -device isa-serial,chardev=interface \
> -device virtio-balloon-pci,disable-modern=false \
> -device virtio-rng-pci,disable-modern=false \
> -device virtio-serial-pci,disable-modern=false \
> -chardev socket,id=agent,server,nowait,path=agent.sock \
> -device virtserialport,chardev=agent,name=org.qemu.guest_agent.0 \
> -chardev spicevmc,name=vdagent,id=vdagent \
> -device virtserialport,chardev=vdagent,name=com.redhat.spice.0 \
> -device qxl-vga \
> -device ich9-intel-hda -device hda-micro \
> -netdev bridge,br=vmnet,id=net \
> -device virtio-net-pci,netdev=net \
> -device nec-usb-xhci -device nec-usb-xhci \
> -device virtio-scsi-pci,disable-modern=false \
> -chardev spicevmc,name=usbredir,id=usbredir1 \
> -device usb-redir,chardev=usbredir1 \
> -chardev spicevmc,name=usbredir,id=usbredir2 \
> -device usb-redir,chardev=usbredir2 \
> -device usb-tablet \
> -drive id=cd1,media=cdrom,if=none \
> -device ide-cd,drive=cd1,bus=ide.0 \
> -drive id=cd2,if=none,media=cdrom \
> -device ide-cd,drive=cd2,bus=ide.1 \
> -drive id=disk,if=none,media=disk,format=qcow2,file=disk.img \
> -device scsi-hd,drive=disk \
> -drive id=fwcode,if=pflash,format=raw,file=firmware.bin \
> -drive id=fwvars,if=pflash,format=raw,file=vars.bin \
> -watchdog-action reset -rtc base=localtime \
> -boot menu=on \
> $@

Thanks. 
One more thing, I forgot to ask you - what is the Windows installation build and bitness? We need all this information for reproducing the problem on our local setups.

Vadim.

Comment 4 Michał Zegan 2016-05-30 06:58:39 UTC
Well, it is a 64 bit system: windows 10 build 1511 if I read correctly from the iso file name, for reference it is pl_windows_10_n_multiple_editions_version_1511_updated_feb_2016_x64_dvd_8380190.
Also because I have many very weird problems with my vm it is actually possible vioscsi is only a trigger for the bug, not the cause.

Comment 5 Vadim Rozenfeld 2016-05-30 23:17:18 UTC
Any chance that you can share dump files for the future analysis?
Another question, if you experience the same problem when booting in safe mode?
Do you you have any 3rd party drivers (especially anti-viruses) installed on your
system? If yes, can you try disabling all of them and see if it makes any changes.

Thanks,
Vadim.

Comment 6 juzhang 2016-05-31 01:15:50 UTC
Hi Vadim,

I guess you raised needifo to Michał Zegan.

Hi Wyu,

Could you have a look this bz as well?

Best Regards,
Junyi

Comment 7 Vadim Rozenfeld 2016-05-31 02:05:12 UTC
(In reply to juzhang from comment #6)
> Hi Vadim,
> 
> I guess you raised needifo to Michał Zegan.
> 
> Hi Wyu,
> 
> Could you have a look this bz as well?
> 
> Best Regards,
> Junyi

Thanks Junyi,

It will be really helpful if QE will spend some time on reproducing this issue.

Best regards,
Vadim.

Comment 8 Michał Zegan 2016-05-31 08:40:40 UTC
Unfortunately I currently cannot provide any dump as I deleted the disk immediately after experiencing this problem.
Also, it is always a fresh install and I just have no drivers/programs/whatever.

Comment 9 Michał Zegan 2016-06-14 10:00:21 UTC
Okay. After much testing, it seems to me that this bug is not an virtio scsi driver bug. It is caused, as well as some other blue screens, by some currently unknown problem in ovmf firmware or something that the firmware triggers. There is a workaround for this firmware triggered bug and when applied, virtio scsi works properly.

Comment 10 Vadim Rozenfeld 2016-06-15 01:23:59 UTC
(In reply to Michał Zegan from comment #9)
> Okay. After much testing, it seems to me that this bug is not an virtio scsi
> driver bug. It is caused, as well as some other blue screens, by some
> currently unknown problem in ovmf firmware or something that the firmware
> triggers. There is a workaround for this firmware triggered bug and when
> applied, virtio scsi works properly.

Thank a lot for your update.
Do you mind me closing this issue, or would you like to keep it open for a while?

Best regards,
Vadim.

Comment 11 Michał Zegan 2016-06-15 06:51:20 UTC
Closing is ok. I may be wrong, but I won't know in the mean time.

Comment 12 lijin 2017-08-17 07:26:47 UTC
close the bug according to comment#10 and comment#11


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