Bug 1980331

Description of problem:
Currently, when using manage vGPU dialog to add multiple mdev devices to VM the maximum available mdev devices is calculated by the number of mdev_bus that are being used by the vGPU card instead of the multiply of vGPU mdev_bus by mdev_type max_instances value. 

for example, when using host with 2xTesla M60 we have 4 mdev_bus in use:
[root@lion01 ~]# ls -l /sys/class/mdev_bus/
total 0
lrwxrwxrwx. 1 root root 0 Jul  7 01:11 0000:84:00.0 -> ../../devices/pci0000:80/0000:80:02.0/0000:82:00.0/0000:83:08.0/0000:84:00.0
lrwxrwxrwx. 1 root root 0 Jul  7 01:11 0000:85:00.0 -> ../../devices/pci0000:80/0000:80:02.0/0000:82:00.0/0000:83:10.0/0000:85:00.0
lrwxrwxrwx. 1 root root 0 Jul  7 01:11 0000:8b:00.0 -> ../../devices/pci0000:80/0000:80:03.0/0000:86:00.0/0000:87:10.0/0000:89:00.0/0000:8a:08.0/0000:8b:00.0
lrwxrwxrwx. 1 root root 0 Jul  7 01:11 0000:8c:00.0 -> ../../devices/pci0000:80/0000:80:03.0/0000:86:00.0/0000:87:10.0/0000:89:00.0/0000:8a:10.0/0000:8c:00.0

And if we want to use nvidia-18 mdev_type, the max_instance per bus is 4:
mdev_type: nvidia-18 --- description: num_heads=4, frl_config=60, framebuffer=2048M, max_resolution=5120x2880, max_instance=4 --- name: GRID M60-2Q

So the maximum available mdev instances to use in that case should be 16 instead of 4.


Also, in case we are using more than one host with vGPU under the same engine, the upper limit should be the one of the host with the most available vGPU instances (if for example we are using 1st host with 1 vGPU card and the 2nd host is with 4 vGPU card, so the upper limit should be according to the 2nd host) 

This bug is relevant to Bug 1860646 - [RFE] Manage vGPU dialog, add option for assigning more than one vGPU instance to VM

A workaround for this issue is to use edit VM dialog -> custom properties -> mdev_type dropbox.


Version-Release number of selected component (if applicable):
ovirt-engine-4.4.7.6-0.11.el8ev
vdsm-4.40.70.6-1.el8ev.x86_64
libvirt-daemon-7.0.0-14.1.module+el8.4.0+11095+d46acebf.x86_64
qemu-kvm-5.2.0-16.module+el8.4.0+11536+725e25d9.2.x86_64
NVIDIA-vGPU-rhel-8.4-460.73.02.x86_64

How reproducible:
100%

Steps to Reproduce:
Assuming the host is with 2xTesla M60 (which is using 4 mdev_bus).
1. Browse WebAdmin -> VM -> host devices tab, open manage vGPU dialog.
2. try to add more than 1 nvidia-22 instance (max_instance=1).
3.

Actual results:
It is possible to add only 1 nvidia-22 instance to VM.

Expected results:
the maximum available instances to use should be 4 (max_instance=1)x(mdev_bus=4).

Additional info:
manage vGPU screenshot attached.