changes in oVirt 4.3 and vGPU?

Hi, I just installed a Tesla T4 card, installed the nvidia's RPM, I see the mdev_type stuff etc. Following their instructions, I'm trying to set a Windows 10 VM to use the vGPU (the VM works without any vGPU), I get this error in the event... VM Win-10-test is down with error. Exit message: internal error: qemu unexpectedly closed the monitor: 2019-02-08T14:01:11.287955Z qemu-kvm: warning: All CPU(s) up to maxcpus should be described in NUMA config, ability to start up with partial NUMA mappings is obsoleted and will be removed in future 2019-02-08T14:01:11.313878Z qemu-kvm: -device vfio-pci,id=hostdev0,sysfsdev=/sys/bus/mdev/devices/486b48a3-01c7-4a67-9727-279813bae0e8,display=off,bus=pci.0,addr=0x8: vfio error: 486b48a3-01c7-4a67-9727-279813bae0e8: error getting device from group 0: Input/output error Verify all devices in group 0 are bound to vfio-<bus> or pci-stub and not already in use. Could someone explain to me what am I missing and what to do? I don't see any docs about it. Thanks

Exit message: internal error: qemu unexpectedly closed the monitor: 2019-02-08T14:01:11.287955Z qemu-kvm: warning: All CPU(s) up to maxcpus should be described in NUMA config, ability to start up with partial NUMA mappings is obsoleted and will be removed in future
I got that error on a fresh 4.3 yesterday while creating a plain boring CentOS VM, and I don't have any Nvidia stuff. Might not be related / could be a bug somewhere else. Anyone else seeing this? Greg On Fri, Feb 8, 2019 at 9:26 AM Hetz Ben Hamo <hetz@hetz.biz> wrote:
Hi,
I just installed a Tesla T4 card, installed the nvidia's RPM, I see the mdev_type stuff etc. Following their instructions, I'm trying to set a Windows 10 VM to use the vGPU (the VM works without any vGPU), I get this error in the event...
VM Win-10-test is down with error. Exit message: internal error: qemu unexpectedly closed the monitor: 2019-02-08T14:01:11.287955Z qemu-kvm: warning: All CPU(s) up to maxcpus should be described in NUMA config, ability to start up with partial NUMA mappings is obsoleted and will be removed in future 2019-02-08T14:01:11.313878Z qemu-kvm: -device vfio-pci,id=hostdev0,sysfsdev=/sys/bus/mdev/devices/486b48a3-01c7-4a67-9727-279813bae0e8,display=off,bus=pci.0,addr=0x8: vfio error: 486b48a3-01c7-4a67-9727-279813bae0e8: error getting device from group 0: Input/output error Verify all devices in group 0 are bound to vfio-<bus> or pci-stub and not already in use.
Could someone explain to me what am I missing and what to do? I don't see any docs about it.
Thanks _______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/X57H4C4SYRVPEG...
-- GREG SHEREMETA SENIOR SOFTWARE ENGINEER - TEAM LEAD - RHV UX Red Hat NA <https://www.redhat.com/> gshereme@redhat.com IRC: gshereme <https://red.ht/sig>

Greg Sheremeta <gshereme@redhat.com> writes:
Exit message: internal error: qemu unexpectedly closed the monitor: 2019-02-08T14:01:11.287955Z qemu-kvm: warning: All CPU(s) up to maxcpus should be described in NUMA config, ability to start up with partial NUMA mappings is obsoleted and will be removed in future
I got that error on a fresh 4.3 yesterday while creating a plain boring CentOS VM, and I don't have any Nvidia stuff. Might not be related / could be a bug somewhere else. Anyone else seeing this?
Yes, I also get that message. AFAIK it is harmless (although still confusing to users) and it probably means what it says: oVirt assigns to NUMA nodes only CPUs present at the VM start. See also https://bugzilla.redhat.com/1437559.
Greg
On Fri, Feb 8, 2019 at 9:26 AM Hetz Ben Hamo <hetz@hetz.biz> wrote:
Hi,
I just installed a Tesla T4 card, installed the nvidia's RPM, I see the mdev_type stuff etc. Following their instructions, I'm trying to set a Windows 10 VM to use the vGPU (the VM works without any vGPU), I get this error in the event...
VM Win-10-test is down with error. Exit message: internal error: qemu unexpectedly closed the monitor: 2019-02-08T14:01:11.287955Z qemu-kvm: warning: All CPU(s) up to maxcpus should be described in NUMA config, ability to start up with partial NUMA mappings is obsoleted and will be removed in future 2019-02-08T14:01:11.313878Z qemu-kvm: -device vfio-pci,id=hostdev0,sysfsdev=/sys/bus/mdev/devices/486b48a3-01c7-4a67-9727-279813bae0e8,display=off,bus=pci.0,addr=0x8: vfio error: 486b48a3-01c7-4a67-9727-279813bae0e8: error getting device from group 0: Input/output error Verify all devices in group 0 are bound to vfio-<bus> or pci-stub and not already in use.
Could someone explain to me what am I missing and what to do? I don't see any docs about it.
Thanks _______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/X57H4C4SYRVPEG...

Answering to myself: Disable ECC on Tesla/GRID cards if you want to use vGPU. Reboot, and it works ;)

Hetz, I'll be reaching out to you if I have vGPU problems!! I receive my T4 on Monday. On Feb 8 2019, at 12:16 pm, Hetz Ben Hamo <hetz@hetz.biz> wrote:
Answering to myself: Disable ECC on Tesla/GRID cards if you want to use vGPU. Reboot, and it works ;) _______________________________________________ Users mailing list -- users@ovirt.org To unsubscribe send an email to users-leave@ovirt.org Privacy Statement: https://www.ovirt.org/site/privacy-policy/ oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/TLL5VZO2VU5TWB...
participants (4)
-
femi adegoke
-
Greg Sheremeta
-
Hetz Ben Hamo
-
Milan Zamazal