Greetings, I have a server that run a vm with several nic pcie cards as pass-though. the board has an intel gpu and the hdmi output is not used. I thought of booting up another vm and pass the gpu into it. I found several discussions on the subject stating that it isn't possible but all of them are several years old. I wonder if that has changed? is there a way to pt the vga to a vm? Thanks, Dagg.
On Sat, Aug 08, 2020 at 06:44:26PM +0200, daggs wrote:
Greetings,
I have a server that run a vm with several nic pcie cards as pass-though. the board has an intel gpu and the hdmi output is not used. I thought of booting up another vm and pass the gpu into it. I found several discussions on the subject stating that it isn't possible but all of them are several years old. I wonder if that has changed? is there a way to pt the vga to a vm?
Thanks,
Dagg.
Hi, libvirt has supported host device assignment for quite some time, so from that perspective, you can follow [1] and see if your card works in the VM. The GPU assignment involves the userspace VFIO driver and there are a few factors in play which may render the assignment either not work at all or work with limitations. In any case, this question would be better asked at vfio-users@redhat.com mailing list should you experience any issues with [1]. Regards, Erik [1] https://libvirt.org/formatdomain.html#host-device-assignment PS: You can even try assigning the GPU from within virt-manager which produces the right XML bits for libvirt. I've never tried assigning an integrated Intel GPU on my laptop (for obvious reasons), so I can't give you the kind of answer guaranteeing this would work 100%.
Greetings Erik,
Sent: Tuesday, August 11, 2020 at 10:11 AM From: "Erik Skultety" <eskultet@redhat.com> To: "daggs" <daggs@gmx.com> Cc: libvirt-users@redhat.com Subject: Re: pass-though Intel gpu int o a vm
On Sat, Aug 08, 2020 at 06:44:26PM +0200, daggs wrote:
Greetings,
I have a server that run a vm with several nic pcie cards as pass-though. the board has an intel gpu and the hdmi output is not used. I thought of booting up another vm and pass the gpu into it. I found several discussions on the subject stating that it isn't possible but all of them are several years old. I wonder if that has changed? is there a way to pt the vga to a vm?
Thanks,
Dagg.
Hi, libvirt has supported host device assignment for quite some time, so from that perspective, you can follow [1] and see if your card works in the VM. The GPU assignment involves the userspace VFIO driver and there are a few factors in play which may render the assignment either not work at all or work with limitations. In any case, this question would be better asked at vfio-users@redhat.com mailing list should you experience any issues with [1]. I'll try ask them, thanks.
Regards, Erik
[1] https://libvirt.org/formatdomain.html#host-device-assignment
PS: You can even try assigning the GPU from within virt-manager which produces the right XML bits for libvirt. I've never tried assigning an integrated Intel GPU on my laptop (for obvious reasons), so I can't give you the kind of answer guaranteeing this would work 100%.
the system in question is a headless server so virt-manager is nto an option. thanks for the suggestion thought. Dagg.
PS: You can even try assigning the GPU from within virt-manager which produces the right XML bits for libvirt. I've never tried assigning an integrated Intel GPU on my laptop (for obvious reasons), so I can't give you the kind of answer guaranteeing this would work 100%.
the system in question is a headless server so virt-manager is nto an option. thanks for the suggestion thought.
virt-manager can connect to libvirtd running on the headless server machine remotely, so I still think it could be of help to you ;). Erik
what is the proper uri to use?
Sent: Wednesday, August 12, 2020 at 7:23 PM From: "Erik Skultety" <eskultet@redhat.com> To: "daggs" <daggs@gmx.com> Cc: libvirt-users@redhat.com Subject: Re: pass-though Intel gpu int o a vm
PS: You can even try assigning the GPU from within virt-manager which produces the right XML bits for libvirt. I've never tried assigning an integrated Intel GPU on my laptop (for obvious reasons), so I can't give you the kind of answer guaranteeing this would work 100%.
the system in question is a headless server so virt-manager is nto an option. thanks for the suggestion thought.
virt-manager can connect to libvirtd running on the headless server machine remotely, so I still think it could be of help to you ;).
Erik
On 8/12/20 12:46 PM, daggs wrote:
what is the proper uri to use?
If you run virt-manager on some other machine, then select "File / New Connection", you will get a dialog box that will prompt you for the info needed, and create the proper URI for you. Utually it is "qemu+ssh://root@$remotehost/system".
Sent: Wednesday, August 12, 2020 at 7:23 PM From: "Erik Skultety" <eskultet@redhat.com> To: "daggs" <daggs@gmx.com> Cc: libvirt-users@redhat.com Subject: Re: pass-though Intel gpu int o a vm
PS: You can even try assigning the GPU from within virt-manager which produces the right XML bits for libvirt. I've never tried assigning an integrated Intel GPU on my laptop (for obvious reasons), so I can't give you the kind of answer guaranteeing this would work 100%.
the system in question is a headless server so virt-manager is nto an option. thanks for the suggestion thought.
virt-manager can connect to libvirtd running on the headless server machine remotely, so I still think it could be of help to you ;).
Erik
participants (3)
-
daggs -
Erik Skultety -
Laine Stump