Description of problem: When a Windows 7 64-bit VM is created via virt-manager, and started, whether from virtual CDROM or Disk, the clock ticks at a very high rate. In one second, about 10 to 15 second speed by, roughly. I've filed this initially as a virt-manager bug, because I'm not sure where the bug actually resides, so starting at the top of the chain makes sense. My OS is Arch Linux, and I'm using the latest package versions from the main repos. Version-Release number of selected component (if applicable): virt-manager 2.1.0 libvirt 5.2.0 qemu 4.0.0 How reproducible: Always. Steps to Reproduce: 1. Create new VM for Windows 7 64-bit. Use default settings. 2. Start VM. 3. Observe absurd clock speeds, visible by the fact the everything happens absurdly quickly. 4. Groan when setting change confirmation dialogs almost instantly time-out... Actual results: 1:~10 host / guest clock speed, affecting everything the guest does ~ everything sped-up. Expected results: 1:1 clock speed observed for host / guest.
Created attachment 1559974 [details] Guest config
Created attachment 1559975 [details] Guest log
I hope that this is sufficient info. :)
Thanks for the report. Can you do 'sudo virsh edit $vmname' and change the hypervclock line to: <timer name='hypervclock' present='yes'/> Then fully power off the VM and restart it, see if it makes a difference
Um... if you look at the config file, the line already looks exactly like that. On that note... I had randomly tried Windows 10, and it was 1:1, even with the exact same setting. I compared the configs, and there was nothing interesting.
Sorry that should have have been s/yes/no/ like: <timer name='hypervclock' present='no'/> Can you try that?
No difference. :(
Did this ever work in the past, or is this the first time you are attempting a win7 install? A few things to try: * Delete all the <timer> elements from guest config, poweroff and restart the VM * Try creating another win7 VM, but in 'Customize before install' select chipset PC instead of the default Q35, see if that makes any difference
Windows 7 used to work in the past, with the previous qemu major release. I'll try out your suggestions, though.
Changing to i440FX resolved it. :) When did Q35 become the default?
We started using q35 by default for modern guest OS in virt-manager 2.0. I'll see if I can reproduce
Had the same issue. Same Windows 7 guest had no issues under Fedora 30 but starting experiencing this bug when I migrated it to a host running virt-manager-2.1.0-4.1 and qemu-4.0.0-2.1 on OpenSUSE Tumbleweed 20190517. Changing it to i440FX fixed it for me as well.
Changing the machine type to `pc-q35-3.1' also fixes it. https://bbs.archlinux.org/viewtopic.php?id=246835
Ok, it's the same qemu-4.0 bug that also affects VFIO Nvidia passthrough: https://bugs.launchpad.net/qemu/+bug/1826422 As per AW advice, adding <ioapic driver='kvm'/> to the xml also makes it work for this Win 7 case. A fix has been queued for qemu-4.0.1 / qemu-4.1 https://lists.gnu.org/archive/html/qemu-devel/2019-06/msg00589.html
Thanks for figuring out the culprit and the qemu-devel patch! Moving this to the Fedora tracker to track the backport
This only affects rawhide/virt-preview. I looked into backporting the patch but I'm going to instead wait for qemu 4.1 or 4.0.1 stable release. This adds a new machine type, so if I backport the patch and for some reason the change is reverted later, fedora is in a weird position having to maintain that machine type on its own.
This bug appears to have been reported against 'rawhide' during the Fedora 31 development cycle. Changing version to '31'.
This bug appears to have been reported against 'rawhide' during the Fedora 31 development cycle. Changing version to 31.
This message is a reminder that Fedora 31 is nearing its end of life. Fedora will stop maintaining and issuing updates for Fedora 31 on 2020-11-24. It is Fedora's policy to close all bug reports from releases that are no longer maintained. At that time this bug will be closed as EOL if it remains open with a Fedora 'version' of '31'. Package Maintainer: If you wish for this bug to remain open because you plan to fix it in a currently maintained version, simply change the 'version' to a later Fedora version. Thank you for reporting this issue and we are sorry that we were not able to fix it before Fedora 31 is end of life. If you would still like to see this bug fixed and are able to reproduce it against a later version of Fedora, you are encouraged change the 'version' to a later Fedora version prior this bug is closed as described in the policy above. Although we aim to fix as many bugs as possible during every release's lifetime, sometimes those efforts are overtaken by events. Often a more recent Fedora release includes newer upstream software that fixes bugs or makes them obsolete.
Fedora 31 changed to end-of-life (EOL) status on 2020-11-24. Fedora 31 is no longer maintained, which means that it will not receive any further security or bug fix updates. As a result we are closing this bug. If you can reproduce this bug against a currently maintained version of Fedora please feel free to reopen this bug against that version. If you are unable to reopen this bug, please file a new report against the current release. If you experience problems, please add a comment to this bug. Thank you for reporting this bug and we are sorry it could not be fixed.