I'm running Windows 10 with PCI passthrough on top of Arch with the latest
kernel and I've had no problems since I got it running. If you have
questions about the process, you've come to the right place. I'd follow
the blogspot postings on how to do this and then modify it for Arch use.
If you ge
Hi all,
I've successfully passed through my GTX 970 and its audio device (see
below).
02:00.0 VGA compatible controller: NVIDIA Corporation GM204 [GeForce
GTX 970] (rev a1) (prog-if 00 [VGA controller])
Kernel driver in use: vfio-pci
02:00.1 Audio device: NVIDIA Corporation GM204 High
is part of VM definition, a sibling to hostdev nodes etc. You
> can create this part in virt-manager GUI, no need to edit xml by hand -
> just make sure to select the bridge you have created.
>
>
>
> B.
>
>
>
> *From: *Garland Key
> *Sent: *Monday, 18 April 2016 2
#x27;t possible on Arch. That said, I can't find any documentation
> on how to convince qemu to use an existing network bridge. If you're
> willing, please add this information as well. If you already know how, any
> pointers woul
y, manually crafted bridge, explicitly
> used in VM definitions.
>
>
> B.
>
> *From: *Garland Key
> *Sent: *Monday, 18 April 2016 22:21
> *To: *Nicolas Roy-Renaud; vfio-users
> *Subject: *Re: [vfio-users] [FEEDBACK NEEDED] Rewriting the Arch wiki
> article
>
> I'
all - none defined (undefined the
> default one) and none started. Just my, manually crafted bridge, explicitly
> used in VM definitions.
>
>
> B.
>
> *From: *Garland Key
> *Sent: *Monday, 18 April 2016 22:21
> *To: *Nicolas Roy-Renaud; vfio-users
> *Subject: *Re:
t. For the
sake of being thorough, you would need to add *MODULES="vfio vfio-pci"*
On Mon, Apr 18, 2016 at 5:27 PM, Alex Williamson <
alex.l.william...@gmail.com> wrote:
> On Mon, Apr 18, 2016 at 3:14 PM, Garland Key
> wrote:
>
>> Please add what to do if you
find any documentation
on how to convince qemu to use an existing network bridge. If you're
willing, please add this information as well. If you already know how, any
pointers would be greatly appreciated.
Best,
Garland
On Mon, Apr 18, 2016 at 5:14 PM, Garland Key
wrote:
> Plea
Please add what to do if you have two identical GPUs. Here is exactly what
is needed to make it work.
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- -
*/etc/modprobe.d/vfio.conf*
install vfio-pci /sbin/vfio-pci-override-vga.sh
options vfio-pci disable_vga=
lting XML for me:
>
>
> default
> 278badf4-d33e-493e-b604-083b3dcc2804
>
>
>
>
> Garland Key schrieb am Sa., 16. Apr. 2016
> um 14:30 Uhr:
>
>> I'm running Arch with Plasma 5 and Network Manager. AFAIK, the
>> virt-manager package on Arch h
I'm running Arch with Plasma 5 and Network Manager. AFAIK, the
virt-manager package on Arch has broken networking. I've successfully
created a bridge using Network Manager, but I can't seem to add the bridge
in virt-manager or via virsh.
Is this possibe? I need it for Synergy. My mobo has two
Arch doesn't provide these files. In the wiki, you're asked to extract the
files from the rpm and copy them to the appropriate directly. From there,
you modify the config file to point to the bin files so that virt-manager
can see it.
On Fri, Apr 15, 2016 at 7:09 AM, Laszlo Ersek wrote:
> On 0
I always appreciate when someone with the knowledge takes the time to
explain why a specific solution works - thank you, Laszlo.
On Fri, Apr 15, 2016 at 6:08 AM, Laszlo Ersek wrote:
> On 04/15/16 10:45, Philip Abernethy wrote:
> > I'm using extra/ovmf on my machine and just had to add
> >
> > nv
I'm 100% sure that I've added the latest edk2.git UEFI files to the correct
place, but virt-manager doesn't see it. Under Firmware: it says UEFI not
found.
Is this a known issue? Is it possible that the edk2 files need specific
permissions? If so, what should the permissions be?
___
obviously it was quicker, good luck !
>
> 2016-04-15 5:47 GMT+02:00 Garland Key :
>>
>> Thanks for the help and offer for future help. I'll tinker with it to
>> see if I can get nvidia to ignore the second gpu. I kept my image
>> file for the win10 install, so
at 10:15 PM, thibaut noah wrote:
> Btw. Your last mail was sent to me only. You should update the one send to
> the mailing list
>
> On Apr 14, 2016 6:37 PM, "Garland Key" wrote:
>>
>> I've figured everything out except for how to run the script.
lso a guy who made a video and a huge post on a forum :
>
> http://www.se7ensins.com/forums/threads/how-to-setup-a-gaming-virtual-machine-with-gpu-passthrough-qemu-kvm-libvirt-and-vfio.1371980/
>
> Modules are indeed in mkinit, read the wiki ;)
>
> 2016-04-14 5:00 GMT+02:00 Ga
ems+="/sbin/vfio-pci-override-vga.sh /usr/bin/find
/usr/bin/dirname" in: /etc/dracut.conf.d/local.conf.
Should I instead add install_items+="/sbin/vfio-pci-override-vga.sh
/usr/bin/find /usr/bin/dirname" to /etc/modprobe.d/vfio.conf?
An
Please add what to do if both of your gpus are identical. This is an issue
that was a huge headache for me. I actually gave up and went to fedora
over it. I missed Arch after two weeks and went back. I was able to get
it to work in fedora on the first go. I haven't given it another go just
yet
I'm having this same issue but the 100% cpu spike isn't degrading gameplay
somehow, so I don't care for now. Hoping it will be fixed in next kernel
update.
On Mon, Mar 21, 2016 at 1:13 PM Jon Panozzo
wrote:
> Can confirm 4.2.8 kernel doesn’t exhibit this behavior with the same QEMU
> package an
Is it possible to pass or share the host temperatures and fan speeds to a
Windows 10 guest using virt-manager?
___
vfio-users mailing list
vfio-users@redhat.com
https://www.redhat.com/mailman/listinfo/vfio-users
mal management access would both
> use the same methods to control fan speed. Unless you have a specific
> reason to use, say, speedfan, you'd probably better off just configuring
> lm_sensors or ACPI thermal control correctly.
>
>
> On 2016-03-16 10:24, Garland Key wrote:
>
By going through the process in Fedora, I found out that there was actually
no problem with your method for arch - it actually worked. It turns out
that windows iso was corrupt. Once I started using Fedora, I decided that
I'd stick with it.
On Tue, Mar 15, 2016, 10:33 AM sL1pKn07 SpinFlo wrote:
in 5 years
(Fedora 23). I had forgotten how nice it is to have everything curated for
you with secure configurations by default. I followed the instructions in
the blog and it worked on the first try.
Best,
David
On Tue, Mar 1, 2016 at 11:51 AM Garland Key
wrote:
> Thank you! lspci -nnk
I'm using Fedora 23 with the virt-preview repo. I'd like to create a
bridge with eno1 (host ethernet) in virt-manager so that I'll be able to
use Synergy. Unfortunately, eno1 isn't listed and I don't see anyway to
add it in the gui. I've googled the problem and didn't find that any
solutions. T
want to bridging VM network
> into eth. Unless eth network isn't DHCP.
> Even so, you can type IP address in guest OS.
>
> Anyway, it's better to check problems if you can paste XML information.
>
> 2016-03-09 8:55 GMT+08:00 Garland Key :
>
>> 1. I'm using t
time, or it will skip CD-ROM option.
> You can try EFI Shell command to boot up Windows 10 installation.
>
> 3. Try edit network to Bridge by using virsh.
>
>
>
>
>
>function='0x0'/>
>
>
> 2016-03-09 4:00 GMT+08:00 Gar
*Specs*OS: Arch Linux x86_64
Mobo: ASUS Sabertooth X99
CPU: i7-5930K
RAM: 32G DDR4
GPU: Gigabyte Nvidia GTX 970
I've successfully installed qemu, libvirt, edk2.git-ovmf-x64 and
vfio-linux. I have successfully dedicated one gpu to vfio and the other to
the nvidia driver. Each of my gpus are liste
Thank you! lspci -nnk shows that my second gpu is bound to vfio-pci! I'll
move on to wrapping this up.
On Tue, Mar 1, 2016 at 11:46 AM sL1pKn07 SpinFlo wrote:
> 2016-03-01 10:00 GMT+01:00 Garland Key :
> > @sL1pKn07 SpinFlo
> >
> > I've started working on this a
icate that
anything is bound to it.
You said that there were zero configurations involving modprobe. Does this
include
loading the following modules in initramfs? vfio vfio_iommu_type1 vfio_pci
vfio_virqfd
On Thu, Feb 25, 2016 at 10:10 AM Garland Key
wrote:
> @sL1pKn07 SpinFlo I worked on the
@sL1pKn07 SpinFlo I worked on the solution you provided and it seems to be
working
so far. When I pull up nvidia-settings, only the first card is recognized
by the driver. I'm
going to move forward and setup the vm now. Thanks so much for the
information! I'll
document this on the Arch Wiki and
I just started using google inbox and it will automatically group all
emails with the same subject line together. It's very tidy and I didn't
think I'd like it until I realized it did this. This is now my daily email
driver and I switch to gmail if I need to write a more complicated email.
On Tu
@sL1pKn07 SpinFlo
This is a jackpot of information - THANK YOU! I'll let you know if/when
I'm able to make this work :D
On Tue, Feb 23, 2016 at 2:03 PM sL1pKn07 SpinFlo wrote:
> double identical gpu, i mean
>
> greetings
>
> 2016-02-23 20:00 GMT+01:00 sL1pKn07 SpinFlo :
> > my arch script:
> >
Once both cards are bound to pci-stub, do I need to unbind them before
binding them to nvidia and vfio-pci? I just realized my script never
unbinds the gpus from the kernel, so hopefully it will work this time.
On Tue, Feb 23, 2016 at 1:47 PM Garland Key
wrote:
> Okay - I'll try it
#2 I wouldn't expect that to work. I'm pretty bad with the boot process
> so I'll leave that for someone else.
> On Feb 23, 2016 1:33 PM, "Garland Key"
> wrote:
>
>> @Alex & @Ben J
>>
>> I tried binding both cards to pci-stub in boot opti
Alex. I'll
> let you know if and when the advice you and Ben have shared works.
>
> On Sun, Feb 21, 2016 at 12:40 PM Alex Williamson <
> alex.william...@redhat.com> wrote:
>
>> On Sun, Feb 21, 2016 at 10:01 AM, Garland Key <
>> david.garland@gmail.com
hared works.
On Sun, Feb 21, 2016 at 12:40 PM Alex Williamson
wrote:
> On Sun, Feb 21, 2016 at 10:01 AM, Garland Key > wrote:
>
>> It seems xorg crashed because the nvidia kernel module failed. Here's
>> the xorg log file:
>>
>
> Did the intended device
I currently just have a script that I launch
> manually. My X server fails on startup with that same error but as soon as
> I bind the nvidia card the server starts successfully.
> On Feb 21, 2016 12:20 PM, "Garland Key"
> wrote:
>
>> Just to be sure I understand you
> nvidia driver. If you bind one of the cards to nvidia, X will probably
> launch.
> On Feb 21, 2016 12:01 PM, "Garland Key"
> wrote:
>
>> It seems xorg crashed because the nvidia kernel module failed. Here's
>> the xorg log file:
>>
>> [1
On Sun, Feb 21, 2016 at 12:01 PM Garland Key
wrote:
> It seems xorg crashed because the nvidia kernel module failed. Here's the
> xorg log file:
>
> [10.474] (--) Log file renamed from "/var/log/Xorg.pid-1427.log" to
> "/var/log/Xorg.0.log"
> [1
7;s kernel log for additional error
messages and
[10.489] (EE) NVIDIA: consult the NVIDIA README for details.
[10.489] (EE) No devices detected.
[10.489] (EE)
Fatal server error:
[10.489] (EE) no screens found(EE)
[10.489] (EE)
Please consult the The X.Org Foundation suppor
Thanks Alex,
I consider myself an intermediate power user. I know enough to understand
what the line that you provided does, but not enough to write a solution of
my own. I inserted the line that you provided above in
/modprobe.d/vfio.conf and it caused X to fail. I had to alt+f2 to comment
it out
:00.0) and GPU audio (02:00.1) are here:
> */sys/devices/pci:00/:00:02.0/:02:00.0/* &
> */sys/devices/pci:00/:00:02.0/:02:00.1/*
>
>
>
>
>
>
>
> *#!/bin/shDEVS=":02:00.0 0000:0
ver_override done
modprobe -i vfio-pci*
Hope to hear from you soon!
Best,
Garland Key
___
vfio-users mailing list
vfio-users@redhat.com
https://www.redhat.com/mailman/listinfo/vfio-users
44 matches
Mail list logo
