What does the following command give you?
for d in $(ls /sys/kernel/iommu_groups/15/devices/); do lspci -nnvs
"$d"; done
I've put the output from that command in the attachment. In that
output the device indicated as a Mediatek device is the one I tried
to add to the VM which is producing the error.
That doesn't look promising. You have a lot of things in that iommu
group including both of your ethernet devices.
That is nothing I have done explicitly as far as I am aware, from my
perspective Fedora has done that itself.
Without that device the VM boots quite happily, and from what
George was showing, the second version of the 6.18 kernel doesn't
appear to have support for the chipset that I read was supposed to
be being provided in the 6.18 kernel. It may be too early in the
kernel release.
I've checked the host system and the device that I think is the
wifi device, as shown by INXI has bus-ID 08:00.0 which is the bus
address shown in the message above, so I believe I have selected
the correct device. The only thing I can think of is QEMU is
expecting the host to have provided a driver for that device which
in the current kernels doesn't exist, and, if that is the case that
defeats the purpose of the test I'm trying to do which is check if
the 6.18 kernel is going to provide a driver. Am I looking at
things incorrectly?
The driver might not be available yet:
https://marc.info/?l=linux-wireless&m=176721477530840&w=2
I've looked at that thread and that person is using a later version
of the kernel to what I see available and as the thread says the
driver isn't available yet. What I also found interesting is the ID
of Steve's device is 14c3:7927, whereas my device ID is 14c3:6639, so
maybe my device is not using the MT7927 chipset but an MT6639 chipset
which also seems to be not supported in the current kernels?
I don't see any indication that either of those are close to being
supported.
Yeah that is what I have read as well, which I've also read is
surprising as Mediatek have the reputation of being very swift with
providing drivers for their chipset for Linux, to the extent of often
having chipset drivers available before the chipsets get wide usage. The
motherboard wifi device also wasn't usable under Windows (Windows
couldn't see the device) until I did a Bios update, which then enable
wifi to work, so I was expecting the same under Linux.
I've tried going back to my usb dongle which uses the mt7925 chipset but
that doesn't work properly, if I unplug the device and plug it back in
again the mt7925u driver loads and wifi is usable (I've seen other
people express the same requirements for the device to be usable. I've
also seen a review on github of the usb device and that tester said it
was a very reliable device and users should have no issues with using it
under Linux because he didn't), but if I boot the system the auto load
of the mt7925u driver fails with error 110, which from what I've read
potentially indicates an incompatibility with Windows Fast Start, but I
have always had that disabled in Windows and the Bios, but even with
that load failure if I unplug the device and plug it back in again the
driver loads quite happily, so I'm struggling to understand what the
issue is.
regards,
BEGIN:VCARD
VERSION:4.0
N:Morris;Stephen;;;
FN:Stephen Morris
EMAIL;PREF=1;TYPE=home:[email protected]
END:VCARD
--
_______________________________________________
users mailing list -- [email protected]
To unsubscribe send an email to [email protected]
Fedora Code of Conduct:
https://docs.fedoraproject.org/en-US/project/code-of-conduct/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives:
https://lists.fedoraproject.org/archives/list/[email protected]
Do not reply to spam, report it:
https://pagure.io/fedora-infrastructure/new_issue
