libosinfo 1.0 seems to be causing regressions with virt-manager on Fedora 25.
Just run the virt-manager test suite and check the failures. Not sure if this
is an upstream bug or a packaging issue though. Simple reproducer is:
$ cat test.py
import gi
gi.require_version('Libosinfo', '1.0')
from gi.repository import Libosinfo as libosinfo
loader = libosinfo.Loader()
loader.process_default_path()
db = loader.get_db()
f23 = db.get_os("http://fedoraproject.org/fedora/23";)
devs = f23.get_all_devices(None)
for idx in range(devs.get_length()):
dev = devs.get_nth(idx)
print dev.get_id(), dev.get_class(), dev.get_bus_type()
$ python test.py
http://pcisig.com/pci/1af4/1041 None pci
http://pcisig.com/pci/1af4/1042 None pci
http://pcisig.com/pci/1af4/1043 None pci
http://pcisig.com/pci/1af4/1044 None pci
http://pcisig.com/pci/1af4/1045 None pci
http://pcisig.com/pci/1af4/1048 None pci
http://pcisig.com/pci/1af4/1049 None pci
http://pcisig.com/pci/1af4/1052 None pci
http://pcisig.com/pci/1af4/1004 None pci
http://usb.org/usb/80ee/0021 None usb
http://pcisig.com/pci/1af4/1001 None pci
http://pcisig.com/pci/1af4/1005 None pci
http://pcisig.com/pci/1af4/1000 None pci
http://pcisig.com/pci/1af4/1002 None pci
http://pcisig.com/pci/1af4/1003 None pci
http://pcisig.com/pci/1af4/1009 None pci
http://pcisig.com/pci/1b36/0100 None pci
http://pcisig.com/pci/8086/2415 None pci
Notice that get_class() returns None. virt-manager does all the device support
matching by filtering on device class, and it seems class=None is breaking
that. get_name() returns empty results as well which will also break
virt-manager. I haven't dug into the 'why' though
Thanks,
Cole
_______________________________________________
virt-tools-list mailing list
virt-tools-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/virt-tools-list
