yeah it asks for a password upon launch but still comes up with that error
you do have libvirt, installed?
yes and running
i may have setup libvirt wrong but i cant find any documentation telling me how so idk
What distro you running?
manjaro
only because I have never attempted to install arch but i have used manjaro and they just released kernel 4.1 their defualt kernel so i though that would be the perfect distro
apparently i just need to give up and use arch or preferably some other distro or do i?
Would it be possible to do this with a gaming laptop?
i have a 5700HQ and a GTX970M 16gigs of RaM
I dont usually Use Linux
as long as the ufei on your laptop supports iommu then yeah i don't see why not
the cpu supports intel VT-d which is necessary so thats good and you also have 2 gpus but you will need a dedicated output for the gpu vs the cpu so idk its worth trying
Then again I'm out of ideas, based on what you have learned and accomplished (hopefully you kept notes) I'd start from scratch with a fresh install of the OS and build out from there, if you have the same results then it may not be what your doing but a deficiency in the build your using, I had this issue with openSuse 13.2 the distro was fine but it just didn't like something in my hardware or the hardware configuration and I could never get a stable install to build off of even though everyone else I talked to didn't have the same issues, switched to Fedora 22 and had no problems with stability at all.
It would seem to me to be one of the features and downfalls of Linux, but at least there are always other distros and options unlike the cookie-cutter nature of Windows where your install is the exact same as mine as long as the version/build number is the same, you have made great progress and learned a lot, but sometimes in doing a project like this you make changes that are irreversible and a wipe and reinstall is necessary to have a virgin environment to try again.
I know how to use terminal, it's just the methods are scattered and not sure which will be viable to OpenSUSE since wendell only tested this on Arch. Having the video that was displayed before would of been nice to watch to get a foothold on what needs to be done, since I've been playing around in linux for a long time now and haven't switched over yet since there are other applications (including games) that are not available in linux even if I emulated them. So not going to do a full switch but I do dust off my linux usage every once a while.
Although I do hope that someone out there will wiki up some guides on different distro's to reference it all.
I have been following this and other resources for passing through hardware to a VM. However so far I have been unable to get my system to boot without error after blacklisting and attempting to pass my graphics card to vfio-pci - "Your screen, graphics card and input device settings could not be detected correctly." I have the system booting off Intel's iGPU but, except when I add the PCI ID to "/etc/modprobe.d/vfio.conf" as suggested here (at which point I can't actually boot properly because of the error), the dedicated graphics card still ends up seeing usage (even though the cable is plugged into the motherboard's HDMI connector) and lspci doesn't indicate vfio has hold of it (though for all I know, even with the addition of the vfio.conf line vfio may not have hold of the card and it's just erroring because that was not something linux was happy with me doing).
On a side note, the ports directly on the dedicated GPU are inoperable at least.
Any recommendations on where I can go from here?
better than me i suppose everytime i fire up my shell script i hang my system with a blank screen
I wish I had the answers for you but I don't best advice I could give you is to start fresh following this guide...
When using vifo-pci there are many steps that must be completed as I'd guess you already know but judging from your post I'd guess your problem is at this point in the guide I linked to above.
"If you're using kernel v4.1 or newer, the vfio-pci driver supports the same ids option so you can directly attach devices to vfio-pci and skip pci-stub. vfio-pci is not generally built statically into the kernel, so we need to force it to be loaded early. To do this on Fedora we need to setup the module options we want to use with modprobe.d. I typically use a file named /etc/modprobe.d/local.conf for local, ie. system specific, configuration. "
Notice the text that I highlighted boldly. Not saying it is your issue but is a reasonable place to start to make sure things are happening and at the right sequence so the card is isolated from the host OS and available to the guest to control uninhibited.
@wendell hey i watch your videos all the time, i love the level of detail you include in your videos and you inspired me to update my computer to the new skylake processor with the z170 deluxe but im unable to boot the vm properly this is the thread im trying to fix my issue in
https://forum.teksyndicate.com/t/pci-stub-not-claming-secondary-graphics-card/88728/9
do you think you can lend me a hand? im using ubuntu because its the only linux os im familiar with. any help would be amazing.
http://vfio.blogspot.com/
try vfio instead of pci-stub? remove all traces of pci-stub plz
are you able to get to uefi? does the card initalize? video card is new enough to have a uefi bios?
1 Like
i cant get to the uefi, the card will not initialize and i got the gtx 980 because i was under the impression that was a known working card for playing games in vm windows.
the website you linked me to doesnt have enough details for me to understand how to remove pci-stub and replace it with vfio. as far as i know i thought i was using stub to hold the card until i lunch the vm where vfio will take over and hand it to the guest os.
sudo /usr/vm1
[sudo] password for spawn:
WARNING: Image format was not specified for '/home/spawn/Downloads/Windows.iso' and probing guessed raw.
Automatically detecting the format is dangerous for raw images, write operations on block 0 will be restricted.
Specify the 'raw' format explicitly to remove the restrictions.
qemu-system-x86_64: -device vfio-pci,host=02:00.0,bus=root.1,addr=00.0,multifunction=on,x-vga=on: vfio: error, group 1 is not viable, please ensure all devices within the iommu_group are bound to their vfio bus driver.
qemu-system-x86_64: -device vfio-pci,host=02:00.0,bus=root.1,addr=00.0,multifunction=on,x-vga=on: vfio: failed to get group 1
qemu-system-x86_64: -device vfio-pci,host=02:00.0,bus=root.1,addr=00.0,multifunction=on,x-vga=on: Device initialization failed
qemu-system-x86_64: -device vfio-pci,host=02:00.0,bus=root.1,addr=00.0,multifunction=on,x-vga=on: Device 'vfio-pci' could not be initializedSo if I dedicate 6 cpu cores to the virtual machine, can I use all eight when i'm not running the vm? same with ram, if i dedicate half, will i be able to use all of it when the virtual machine is shut down?
Nice job if you want even better performance delete kvm=off in your script you only need it for passing through nvidia cards and use hyper-v enlightenment so Windows knows its in a virtual machine and optimizes itself. I've heard that the i915 patch is needed but only if your emulating a legacy bios. If your using ovmf its not needed. Vcpu punning is also a really good way of improving your VM.
What does your script look like pci stub shouldn't be the issue
1 Like