<div dir="ltr">For me, I usually delete this line in <device> tags<div> </div><div> <address type='pci' domain='0x0000' bus='0x00' slot='0x08' function='0x0'/> </div><div> </div><div>Don't delete the <address> tag inside the <source>.</div><div>I usually delete this line, this line means the virtual address that pass-throughed device will set in VM.</div><div>And it will not get error if you save XML file, it will generate automatically while VM started.</div><div>So that u don't have to worry about virtual address conflicts.</div></div><div class="gmail_extra"> <div class="gmail_quote">2016-02-23 4:58 GMT+08:00 Eric Griffith <<a href="mailto:egriffith92@gmail.com" target="_blank">egriffith92@gmail.com</a>>: <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr"> <div class="gmail_extra"><div><div class="h5"><div class="gmail_quote">On Thu, Jan 28, 2016 at 8:47 PM, Stewart Adam <<a href="mailto:maillist@diffingo.com" target="_blank">maillist@diffingo.com</a>> wrote: <blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">On 2016-01-21 12:32 PM, Eric Griffith wrote: <blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"> On Jan 17, 2016 16:10, "Stewart Adam" <<a href="mailto:maillist@diffingo.com" target="_blank">maillist@diffingo.com</a> <mailto:<a href="mailto:maillist@diffingo.com" target="_blank">maillist@diffingo.com</a>>> wrote: > > On 2016-01-17 1:27 PM, Eric Griffith wrote: >> >> On Jan 17, 2016 13:15, "Stewart Adam" <<a href="mailto:maillist@diffingo.com" target="_blank">maillist@diffingo.com</a> <mailto:<a href="mailto:maillist@diffingo.com" target="_blank">maillist@diffingo.com</a>> <div><div> >> <mailto:<a href="mailto:maillist@diffingo.com" target="_blank">maillist@diffingo.com</a> <mailto:<a href="mailto:maillist@diffingo.com" target="_blank">maillist@diffingo.com</a>>>> wrote: >> > >> > On 2016-01-07 11:30 AM, Eric Griffith wrote: >> >> >> >> My problem is.. The Win7 (and Win8.1) guests are perfectly fine and stable >> >> until I try to install the amd crimson drivers. Then Windows BSODs, yelling >> >> about an unhandled exception. If I unpack the install and install things >> >> piecemeal, then I can install the HDMI audio driver, and everything else >> >> okay...except for the display driver. Installing the display driver causes >> >> the BSOD. >> > >> > How are you booting the guest (e.g. manually executing QEMU, or manually >> created libvirt XML or virt-manager+libvirt)? >> > >> >> Virt-manager+libvirt >> >> > I had a very similar symptom on my guest, in the end it was the guest's >> machine flag "vmport=off" that virt-manager had added that resulted in the >> hang/BSOD upon installation of the driver. >> >> I tried removing the vmport=off flag but still BSOD'ed :( >> >> > After attempting a driver install, does the guest boot correctly but hang >> when attempting to show the login screen (i.e. when the 3D driver initializes)? >> >> Sometimes yes, other times itd just do the boot as if I had never tried to >> install the driver. > > > Sounds like the same issue I was experiencing... Have you also added a ioh3420 device edited your configuration so that the GPU is attached to it instead the default PCI/PCIE buses? If you're not sure, please attach the output of "virsh dumpxml your-vm-name". I did not do that, I just passed through the card and the card's audio. Are there any available instructions on how to do so? Also I appreciate the pointer, Stewart :) </div></div></blockquote> Sorry for the delay - here are those instructions: You'll want to execute 'virsh edit domname' as root to edit your domain's config file, and then add this XML after the 'pci-bridge' controller block: <controller type='pci' index='3' model='pcie-root-port'> <model name='ioh3420'/> <target chassis='1' port='0x1'/> <alias name='pci.3'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x1c' function='0x0' multifunction='on'/> </controller> Then scroll down to the <hostdev> block for your GPU, and alter it to use the controller (bus 3) we just created: <hostdev mode='subsystem' type='pci' managed='yes'> <driver name='vfio'/> <source> <address domain='0x0000' bus='0x01' slot='0x00' function='0x0'/> </source> <alias name='hostdev0'/> <address type='pci' domain='0x0000' bus='0x03' slot='0x00' function='0x0' multifunction='on'/> </hostdev> Regards, Stewart </blockquote></div> </div></div><div><div><div>Hey Stewart, long time no talk. Sorry about the delaying in getting back to you on this topic, classes really kicking into high gear and life became kind of crazy. </div>Copy pasted below is the dumpxml of my Win8.1 VM. I tried to add in the two code blocks that you linked above, but I got errors when I did so. I've bolded and marked in red where I added the first code block, the pci definition, and then did the same for the two devices that make up my GPU. </div><div>Edit: just incase anyone's mail client doesn't like the bold / text color change, I've also off-set the area by adding a blank line. </div><div> </div>XML WALL OF TEXT: <domain type='kvm'> <name>win8.1</name> <uuid>(UUID)</uuid> <memory unit='KiB'>4194304</memory> <currentMemory unit='KiB'>4194304</currentMemory> <vcpu placement='static'>2</vcpu> <os> <type arch='x86_64' machine='pc-i440fx-2.5'>hvm</type> <loader readonly='yes' type='pflash'>/usr/share/edk2.git/ovmf-x64/OVMF_CODE-pure-efi.fd</loader> <nvram>/var/lib/libvirt/qemu/nvram/win8.1_VARS.fd</nvram> </os> <features> <acpi/> <apic/> <hyperv> <relaxed state='on'/> <vapic state='on'/> <spinlocks state='on' retries='8191'/> </hyperv> </features> <cpu mode='custom' match='exact'> <model fallback='allow'>core2duo</model> </cpu> <clock offset='localtime'> <timer name='rtc' tickpolicy='catchup'/> <timer name='pit' tickpolicy='delay'/> <timer name='hpet' present='no'/> <timer name='hypervclock' present='yes'/> </clock> <on_poweroff>destroy</on_poweroff> <on_reboot>restart</on_reboot> <on_crash>restart</on_crash> <pm> <suspend-to-mem enabled='no'/> <suspend-to-disk enabled='no'/> </pm> <devices> <emulator>/usr/bin/qemu-kvm</emulator> <disk type='file' device='cdrom'> <driver name='qemu' type='raw'/> <source file='/usr/share/virtio-win/virtio-win-0.1.102.iso'/> <target dev='sda' bus='sata'/> <readonly/> <boot order='3'/> <address type='drive' controller='0' bus='0' target='0' unit='0'/> </disk> <disk type='file' device='disk'> <driver name='qemu' type='qcow2'/> <source file='/var/lib/libvirt/images/win8.1.qcow2'/> <target dev='sdb' bus='sata'/> <boot order='1'/> <address type='drive' controller='0' bus='0' target='0' unit='1'/> </disk> <disk type='file' device='cdrom'> <driver name='qemu' type='raw'/> <source file='/var/lib/libvirt/images/Win8.1_English_x64.iso'/> <target dev='sdc' bus='sata'/> <readonly/> <boot order='2'/> <address type='drive' controller='0' bus='0' target='0' unit='2'/> </disk> <controller type='usb' index='0' model='ich9-ehci1'> <address type='pci' domain='0x0000' bus='0x00' slot='0x05' function='0x7'/> </controller> <controller type='usb' index='0' model='ich9-uhci1'> <master startport='0'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x05' function='0x0' multifunction='on'/> </controller> <controller type='usb' index='0' model='ich9-uhci2'> <master startport='2'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x05' function='0x1'/> </controller> <controller type='usb' index='0' model='ich9-uhci3'> <master startport='4'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x05' function='0x2'/> </controller> </div> <div> <controller type='pci' index='0' model='pci-root'/> <controller type='sata' index='0'> <address type='pci' domain='0x0000' bus='0x00' slot='0x04' function='0x0'/> </controller> <controller type='virtio-serial' index='0'> <address type='pci' domain='0x0000' bus='0x00' slot='0x06' function='0x0'/> </controller> <interface type='network'> <mac address='52:54:00:7d:c6:85'/> <source network='default'/> <model type='virtio'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x02' function='0x0'/> </interface> <serial type='pty'> <target port='0'/> </serial> <console type='pty'> <target type='serial' port='0'/> </console> <channel type='spicevmc'> <target type='virtio' name='com.redhat.spice.0'/> <address type='virtio-serial' controller='0' bus='0' port='1'/> </channel> <input type='tablet' bus='usb'/> <sound model='ich9'> <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/> </sound> <hostdev mode='subsystem' type='pci' managed='yes'> <source> <address domain='0x0000' bus='0x01' slot='0x00' function='0x0'/> </source> <address type='pci' domain='0x0000' bus='0x00' slot='0x07' function='0x0'/> </hostdev> <hostdev mode='subsystem' type='pci' managed='yes'> <source> <address domain='0x0000' bus='0x01' slot='0x00' function='0x1'/> </source> <address type='pci' domain='0x0000' bus='0x00' slot='0x08' function='0x0'/> </hostdev> <hostdev mode='subsystem' type='usb' managed='yes'> <source> <vendor id='0x24f0'/> <product id='0x0140'/> </source> </hostdev> <hostdev mode='subsystem' type='usb' managed='yes'> <source> <vendor id='0x046d'/> <product id='0xc52b'/> </source> </hostdev> <hostdev mode='subsystem' type='usb' managed='yes'> <source> <vendor id='0x0a5c'/> <product id='0x21e8'/> </source> </hostdev> <redirdev bus='usb' type='spicevmc'> </redirdev> <redirdev bus='usb' type='spicevmc'> </redirdev> <memballoon model='virtio'> <address type='pci' domain='0x0000' bus='0x00' slot='0x09' function='0x0'/> </memballoon> </devices> </domain> </div><div>Adding them in was no problem in and of itself, it was saving it that caused a problem. Virsh didn't like those two definitions. Editing the "address type=pci" lines of both, and changing THOSE bus= definitions to say 0x03, instead of 0x00 gives me this error: error: XML error: Invalid PCI address 0000:03:07.0. slot must be <= 0 </div><div>Editing the one above it, the "address domain=" line, and changing the bus from 0x01 to 0x03 gives me this error: </div><div> error: XML error: The device information for 0000:00:1c.0 has no PCI connection types listed </div>Any thoughts? </div><div class="gmail_extra">Half tempted to say 'screw it' and just hold off on doing a VFIO setup until I snag an Nvidia Polaris card later this year >< </div></div> _______________________________________________ vfio-users mailing list <a href="mailto:vfio-users@redhat.com">vfio-users@redhat.com</a> <a href="https://www.redhat.com/mailman/listinfo/vfio-users" rel="noreferrer" target="_blank">https://www.redhat.com/mailman/listinfo/vfio-users</a> </blockquote></div> </div>