Re: [vfio-users] NVIDIA error: Failed to initialize DMA. Failed to allocate push buffer

Yes, this was it! Before making may more changes, I looked in /proc/iomem and found this: 1802100-20020ff : PCI Bus :80 200-2000fff 200-200120f : PCI Bus :81 200-2000fff : :81:00.0 200-2257fff : efifb HERE, wro

Re: [vfio-users] NVIDIA error: Failed to initialize DMA. Failed to allocate push buffer

I think here is the strongest hint; the host dmesg is floodeed with messages "BAR 1: can't reserve" 2021-12-30T10:01:59.456992+ gauss.lan.incorrekt.net kernel: vfio-pci :81:00.0: vfio_ecap_init: hiding ecap 0x1e@0x258 2021-12-30T10:01:59.457413+ gauss.lan.incorrekt.net kernel: vfio-

Re: [vfio-users] NVIDIA error: Failed to initialize DMA. Failed to allocate push buffer

Some more information: 1. driver seem to be loading fine in guest bronekk@euclid:~$ sudo dmesg | grep -E "nvidia|0d:00" [0.810066] pci :0d:00.0: [10de:1eb1] type 00 class 0x03 [0.814518] pci :0d:00.0: reg 0x10: [mem 0xc000-0xc0ff] [0.818518] pci :0d:00.0: reg 0

[vfio-users] NVIDIA error: Failed to initialize DMA. Failed to allocate push buffer

Hi Hoping someone solved this one before. My host if Epyc Milan, running on Asrock ROMED8-2T, GPU is NVIDIA Quadro RTX 4000, running on fresh Arch Linux install. The guest is Debian 11 , with NVIDIA-460 drivers . I can see the drivers are correctly loaded in the guest (with nvidia-smi), but Xo

Re: [vfio-users] NVIDIA has enabled GPU passthrough beta support for a Windows virtual machine on GeForce GPUs

I tried version 465.89 driver with Windows 10 20H2 guest on Ubuntu 20.04 with no luck. Using GTX1050Ti test this driver. The Whole guest gets frozen during the driver installation and BSOD loop after force reset. Interestingly, The driver is normal if let Windows to get the driver from WSUS automa

Re: [vfio-users] NVIDIA has enabled GPU passthrough beta support for a Windows virtual machine on GeForce GPUs

On 2021-03-31 11:38, Bronek Kozicki wrote: Some good news, straight from the source: https://nvidia.custhelp.com/app/answers/detail/a_id/5173/~/geforce-gpu-passthrough-for-windows-virtual-machine-%28beta%29 Now, if only I was able to buy a GeForce card ... Tested it, fortunately they did not

Re: [vfio-users] NVIDIA has enabled GPU passthrough beta support for a Windows virtual machine on GeForce GPUs

On Wed, 31 Mar 2021, at 12:51 PM, Patrick O'Callaghan wrote: > On Wed, 2021-03-31 at 10:38 +0100, Bronek Kozicki wrote: > > Some good news, straight from the source: > > > > https://nvidia.custhelp.com/app/answers/detail/a_id/5173/~/geforce-gpu-passthrough-for-windows-virtual-machine-%28beta%29 >

Re: [vfio-users] NVIDIA has enabled GPU passthrough beta support for a Windows virtual machine on GeForce GPUs

On Wed, 2021-03-31 at 10:38 +0100, Bronek Kozicki wrote: > Some good news, straight from the source: > > https://nvidia.custhelp.com/app/answers/detail/a_id/5173/~/geforce-gpu-passthrough-for-windows-virtual-machine-%28beta%29 How is this different from what we've been doing up to now? poc

[vfio-users] NVIDIA has enabled GPU passthrough beta support for a Windows virtual machine on GeForce GPUs

Some good news, straight from the source: https://nvidia.custhelp.com/app/answers/detail/a_id/5173/~/geforce-gpu-passthrough-for-windows-virtual-machine-%28beta%29 Now, if only I was able to buy a GeForce card ... -- Bronek Kozicki b...@incorrekt.com __

Re: [vfio-users] nvidia

providing a rom manually is the one thing i have not tried. ok, i will give it a shot. On 7/7/20 7:12 PM, Alex Williamson wrote: On Tue, Jul 7, 2020 at 4:33 PM Roger Lawhorn > wrote: Hello, I have an nvidia 980 ti oc 6gb card. I cannot use it with qemu as a pas

Re: [vfio-users] nvidia

On Tue, Jul 7, 2020 at 4:33 PM Roger Lawhorn wrote: > Hello, > I have an nvidia 980 ti oc 6gb card. > I cannot use it with qemu as a passthrough card. > I have had to passthrough my amd cards only. > I have read of nvidia making it impossible to use some of their cards in > virtual machines. > Is

[vfio-users] nvidia

Hello, I have an nvidia 980 ti oc 6gb card. I cannot use it with qemu as a passthrough card. I have had to passthrough my amd cards only. I have read of nvidia making it impossible to use some of their cards in virtual machines. Is this true? Is there a workaround? I had a laptop with a nvidia 8

Re: [vfio-users] Nvidia device not recognised / nouveau 0000:00:04.0: unknown chipset (ffffffff)

Any further info I can give or things I can try? If there a way in kvm vfio to set the chipset type? I'm not sure how to work out if I'm trying the impossible. Using Ubuntu 18.04 = qemu 2.11 to start with, but happy to follow whatever OS guide to help debug/troubleshoot. cheers, Woody On 20

[vfio-users] Nvidia device not recognised / nouveau 0000:00:04.0: unknown chipset (ffffffff)

Hi VFIOers! new to VFIO and trying a Metabox (clevo) P950ER. Laptop has Nvidia 1070M with Max-Q, no UEFI in the Bios. I have an older Metabox with a 970M (also no UEFI in the nvidia) to try things on to, but I'm using that for work until the P950ER is up and running. Following: https://hei

Re: [vfio-users] NVidia code 43 with Winddows 10 guest - does NVidia have another trick up their sleeve?

Hi, I managed to get it working this morning. The GPU ROM dumping procedure definitely had something to do with it. I took another shot at switching GPU slots, booting off my 740 and trying to dump the 1060's ROM. I was again unsuccessful, getting the input/output errors, both booting into the f

Re: [vfio-users] NVidia code 43 with Winddows 10 guest - does NVidia have another trick up their sleeve?

On Thu, 04 Jan 2018 17:29:28 + Bronek Kozicki wrote: > Perhaps disabling ROM and adding spice as "boot" video card will help. That's > how I use my Quadro M5000 , apparently the only supported way of doing it. > Turns out spice is useful for other purposes, even if its Windows display > ou

Re: [vfio-users] NVidia code 43 with Winddows 10 guest - does NVidia have another trick up their sleeve?

ping on a phone.   Original Message   From: alex.william...@redhat.com Sent: 4 January 2018 4:32 pm To: robert.j.dahlst...@gmail.com Cc: vfio-users@redhat.com Subject: Re: [vfio-users] NVidia code 43 with Winddows 10 guest - does NVidia have another trick up their sleeve? On Wed, 3 Jan 2018 17:28:

Re: [vfio-users] NVidia code 43 with Winddows 10 guest - does NVidia have another trick up their sleeve?

On Wed, 3 Jan 2018 17:28:39 -0600 Robert Dahlstrom wrote: > Hello, I have been trying to setup a Windows 10 guest on my Linux host, > passing through an nvidia 1060. I've been having a very difficult time, > getting the dreaded code 43. I haven't followed an individual guide, but > have taken bit

Re: [vfio-users] NVidia code 43 with Winddows 10 guest - does NVidia have another trick up their sleeve?

I bet you've already tried this but looking through your xml file for the guest I see a few inconsistencies when compared against the suggestions on the Arch Wiki [here](https://wiki.archlinux.org/index.php/PCI_passthrough_ via_OVMF#.22Error_43:_Driver_failed_to_load.22_on_Nvidia_ GPUs_passed_to_Wi

[vfio-users] NVidia code 43 with Winddows 10 guest - does NVidia have another trick up their sleeve?

Hello, I have been trying to setup a Windows 10 guest on my Linux host, passing through an nvidia 1060. I've been having a very difficult time, getting the dreaded code 43. I haven't followed an individual guide, but have taken bits from https://wiki.archlinux.org/ind ex.php/PCI_passthrough_via_OVM

Re: [vfio-users] Nvidia GTX970 overheat in game with vgapassthrough

e fan didn't solve the issue for me. I suspect that the actual issue is that they undervolted the GPU to make it run cooler. Date: Sat, 22 Jul 2017 11:40:33 + > From: Ivan Volosyuk > To: vfio-users > Subject: [vfio-users] Nvidia GTX970 overheat in game with > vgapa

Re: [vfio-users] Nvidia GTX970 overheat in game with vgapassthrough

Install msi afterburner and tell me what you see for the fan speed and temp readings. On 07/22/2017 07:40 AM, Ivan Volosyuk wrote: I was using KVM/QEMU vga-passthrough successfully for some time now and haven't seen much issues with my configuration.Yesterday, I installed PLAYERUNKNOWN'S BATTL

[vfio-users] Nvidia GTX970 overheat in game with vgapassthrough

I was using KVM/QEMU vga-passthrough successfully for some time now and haven't seen much issues with my configuration.Yesterday, I installed PLAYERUNKNOWN'S BATTLEGROUNDS and I noticed something which looks like GPU overheating issue: Initially the game works fine. I noticed that GPU fan starts ro

Re: [vfio-users] NVIDIA driver cause to black screen and can't launch desktop environment

> You are missing the PCIe Root Bridge, and attaching the GPU to it. Linux is > picky about that. > > Try with this: > > -device ioh3420,chassis=0,id=testbus \ > > -device > vfio-pci,host=01:00.0,multifunction=true,bus=testbus,addr=00.0,x-vga=on \ > -device vfio-pci,host=01:00.1,bus=testbus,addr

Re: [vfio-users] NVIDIA driver cause to black screen and can't launch desktop environment

>> I run QEMU with the following command: >> -device vfio-pci,host=01:00.0,x-vga=on -device vfio-pci,host=01:00.1 You are missing the PCIe Root Bridge, and attaching the GPU to it. Linux is picky about that. Try with this: -device ioh3420,chassis=0,id=testbus \ -device vfio-pci,host=01:00.0,

[vfio-users] NVIDIA driver cause to black screen and can't launch desktop environment

Hi all, I'm able to pass through a NVIDIA graphic card(GTX750 Ti) to RHEL 5.11 successfully, For performance reasons, I need to install NVIDIA's official driver, After installing NVIDIA's official driver, the guest system always encountered black screen and can't enter desktop environment. But I

Re: [vfio-users] nvidia gtx 780 - MSI-X?

On Fri, 17 Mar 2017 00:31:55 -0400 "taii...@gmx.com" wrote: > I have enabled MSI through the windows registry tweak but I am wondering > if I can bump that up to MSI-X somehow. > > I ask here and not on a windows forum because I have no idea if the card > supports this natively (and I have no

[vfio-users] nvidia gtx 780 - MSI-X?

I have enabled MSI through the windows registry tweak but I am wondering if I can bump that up to MSI-X somehow. I ask here and not on a windows forum because I have no idea if the card supports this natively (and I have no bare metal windows machines to test it on) - Thanks ___

Re: [vfio-users] Nvidia Code 43 on Fedora 24 host and Windows 10 guest.

On Thu, Oct 6, 2016 at 10:44 AM, Thomas Lindroth wrote: > The linux desktop have an annoying habit of moving all windows around > when the virtual desktop size change and keep the new position when > the size change back. I eventually created those scripts to move my > regular programs to hardcod

Re: [vfio-users] Nvidia Code 43 on Fedora 24 host and Windows 10 guest.

On 10/06/2016 05:53 PM, Rokas Kupstys wrote: > Hey could you share /home/cocobo/scripts/single_head and > /home/cocobo/scripts/dual_head scripts? They sound very interesting if > they are what i think they are. They are probably not as interesting as you think. single_head: #!/bin/bash # use xwin

Re: [vfio-users] Nvidia Code 43 on Fedora 24 host and Windows 10 guest.

Hey could you share /home/cocobo/scripts/single_head and /home/cocobo/scripts/dual_head scripts? They sound very interesting if they are what i think they are. Rokas Kupstys On 2016.10.05 17:35, Thomas Lindroth wrote: > On 10/05/2016 09:28 AM, Dawid Osuchowski wrote:> I've ran dd if=NV287MH.2K0

Re: [vfio-users] Nvidia Code 43 on Fedora 24 host and Windows 10 guest.

On Wed, 5 Oct 2016 16:35:08 +0200 Thomas Lindroth wrote: > On 10/05/2016 09:28 AM, Dawid Osuchowski wrote:> I've ran dd if=NV287MH.2K0 > of=strip bs=1 skip=1536, put the ROM file in > > /usr/share/qemu, run restorecon on it and it started working! I'm so > > thankful for both of you Alex and Tho

Re: [vfio-users] Nvidia Code 43 on Fedora 24 host and Windows 10 guest.

On 10/05/2016 09:28 AM, Dawid Osuchowski wrote:> I've ran dd if=NV287MH.2K0 of=strip bs=1 skip=1536, put the ROM file in > /usr/share/qemu, run restorecon on it and it started working! I'm so > thankful for both of you Alex and Thomas, i've been trying to get it to > work for at least 4-5 days and

Re: [vfio-users] Nvidia Code 43 on Fedora 24 host and Windows 10 guest.

I've ran dd if=NV287MH.2K0 of=strip bs=1 skip=1536, put the ROM file in /usr/share/qemu, run restorecon on it and it started working! I'm so thankful for both of you Alex and Thomas, i've been trying to get it to work for at least 4-5 days and you've helped me in just one day. While we are still he

Re: [vfio-users] Nvidia Code 43 on Fedora 24 host and Windows 10 guest.

On Wed, 5 Oct 2016 04:04:01 +0200 Thomas Lindroth wrote: > On 10/05/2016 03:46 AM, Alex Williamson wrote: > > If you want to send me your ROM file, I'll compare to mine and see > > whether it works with my card (not sure whether it should or not given > > different vendors). Thanks, > > This

Re: [vfio-users] Nvidia Code 43 on Fedora 24 host and Windows 10 guest.

On 10/05/2016 03:46 AM, Alex Williamson wrote: > If you want to send me your ROM file, I'll compare to mine and see > whether it works with my card (not sure whether it should or not given > different vendors). Thanks, This conversation from irc explains where that strip.rom file comes from. He

Re: [vfio-users] Nvidia Code 43 on Fedora 24 host and Windows 10 guest.

On Tue, 4 Oct 2016 19:19:09 -0600 Alex Williamson wrote: > On Tue, 4 Oct 2016 21:40:17 +0200 > Dawid Osuchowski wrote: > > > I've rebooted the machine, the lspci output was the same without > > launching anything prior to it. > > > > If by OVMF boot splash screen you mean the "TianoCore" scree

Re: [vfio-users] Nvidia Code 43 on Fedora 24 host and Windows 10 guest.

On Tue, 4 Oct 2016 21:40:17 +0200 Dawid Osuchowski wrote: > I've rebooted the machine, the lspci output was the same without > launching anything prior to it. > > If by OVMF boot splash screen you mean the "TianoCore" screen before the > OS is starting, then yes i'm able to see it. > > I've put

Re: [vfio-users] Nvidia Code 43 on Fedora 24 host and Windows 10 guest.

I've rebooted the machine, the lspci output was the same without launching anything prior to it. If by OVMF boot splash screen you mean the "TianoCore" screen before the OS is starting, then yes i'm able to see it. I've put the ROM file in the /usr/share/qemu directory and run restorecon on it (a

Re: [vfio-users] Nvidia Code 43 on Fedora 24 host and Windows 10 guest.

On Tue, 4 Oct 2016 19:43:58 +0200 Dawid Osuchowski wrote: > Hello! I'm trying to do a PCI passthrough of my GPU to Windows 10 guest > machine using libvirt and vfio-pci. I've followed Alex Williamson's > guide from vfio-blog > http://vfio.blogspot.com/2015/05/vfio-gpu-how-to-series-part-3-host.ht

[vfio-users] Nvidia Code 43 on Fedora 24 host and Windows 10 guest.

Hello! I'm trying to do a PCI passthrough of my GPU to Windows 10 guest machine using libvirt and vfio-pci. I've followed Alex Williamson's guide from vfio-blog http://vfio.blogspot.com/2015/05/vfio-gpu-how-to-series-part-3-host.html and it still doesn't work as i'm getting Code 43 on my MSI nVidia

Re: [vfio-users] Nvidia code 43 with Ubuntu 16.04 host

On Tue, Sep 6, 2016 at 12:00 PM, Alex Williamson < alex.l.william...@gmail.com> wrote: > It means his version of libvirt is too old to recognize the tag and > generates an error parsing the xml because it's unknown. The referenced > xml doesn't include anything that would trigger hyper-v afaict,

Re: [vfio-users] Nvidia code 43 with Ubuntu 16.04 host

Solving the libvirt version issue is easy enough. Add ppa:jacob/virtualisation and upgrade libvirt. He's got 2.1.0 in there, along with updated versions of qemu and virt-manager. On Tue, Sep 6, 2016 at 1:00 PM, Alex Williamson wrote: > On Tue, Sep 6, 2016 at 11:47 AM, Will Marler wrote: > >>

Re: [vfio-users] Nvidia code 43 with Ubuntu 16.04 host

On Tue, Sep 6, 2016 at 11:47 AM, Will Marler wrote: > Simon, are you able to boot your VM with GPU passthrough and no Code 43 > errors without enabling any hyperv features? Do this first, then move on to > enabling Hyper V. > > "I cannot place the vendor_id flag in the xml because it is validated

Re: [vfio-users] Nvidia code 43 with Ubuntu 16.04 host

Simon, are you able to boot your VM with GPU passthrough and no Code 43 errors without enabling any hyperv features? Do this first, then move on to enabling Hyper V. "I cannot place the vendor_id flag in the xml because it is validated in the Ubuntu package" -- what does this mean? Is there someth

Re: [vfio-users] Nvidia code 43 with Ubuntu 16.04 host

On Tue, Sep 6, 2016 at 8:33 AM, Rokas Kupstys wrote: > While nothing wrong at the same time it is fragile. Script is replacing > some parameters with something else. If command line will change enough for > search string to no longer be present in there it will stop working. Maybe > this is what

Re: [vfio-users] Nvidia code 43 with Ubuntu 16.04 host

While nothing wrong at the same time it is fragile. Script is replacing some parameters with something else. If command line will change enough for search string to no longer be present in there it will stop working. Maybe this is what happened? Rokas Kupstys On 2016.09.06 13:59, Alex Williamson

Re: [vfio-users] Nvidia code 43 with Ubuntu 16.04 host

On Sep 6, 2016 1:23 AM, "Rokas Kupstys" wrote: > > Why on earth are you using wrapper script? If you use shellscript to start qemu then add these flags there. If you are using libvirt - libvirt also supports these features. This mess is looking for a trouble and i believe you found it. libvirt di

Re: [vfio-users] Nvidia code 43 with Ubuntu 16.04 host

Why on earth are you using wrapper script? If you use shellscript to start qemu then add these flags there. If you are using libvirt - libvirt also supports these features. This mess is looking for a trouble and i believe you found it. Rokas Kupstys On 2016.09.06 00:28, Simon Ye wrote: > I'm tryi

[vfio-users] Nvidia code 43 with Ubuntu 16.04 host

I'm trying to use a Windows 10 Enterprise guest but keep getting error 43 with my Nvidia 750Ti card. I have set the kvm hidden state flag and well as overriden the vendor_id flag via a wrapper script as suggested by other users on this mailing list: #!/bin/sh exec /usr/bin/qemu-system-x86_64 \ `

Re: [vfio-users] NVIDIA GPU Passthrough to Win10 - Driver Disabled (Code 43)

Success! The cookie goes to Philip for the suggestion that the card wasn't working in UEFI mode due to firmware. I flashed a newer version of the firmware on the GPU. Repeating all steps in my original post resulted in a successful installation and the GPU is now successfully functioning on the G

Re: [vfio-users] NVIDIA GPU Passthrough to Win10 - Driver Disabled (Code 43)

On Mon, 25 Jul 2016 21:39:56 -0400 Steven Bell wrote: > Hey Alex, > > I really appreciate all your help and your quick responses (and everyone > else too, thanks!). > > I have found a Utility from MSI that will detect if there is an updated > version of the my card's firmware, but it won't run

Re: [vfio-users] NVIDIA GPU Passthrough to Win10 - Driver Disabled (Code 43)

Hey Alex, I really appreciate all your help and your quick responses (and everyone else too, thanks!). I have found a Utility from MSI that will detect if there is an updated version of the my card's firmware, but it won't run on the VM (I assume it's looking for some version bit for Win10_64bit

Re: [vfio-users] NVIDIA GPU Passthrough to Win10 - Driver Disabled (Code 43)

On Mon, 25 Jul 2016 20:15:15 -0400 Steven Bell wrote: > So I tried a BIOS VM, freshly created from scratch. Reinstalled Windows 10 > and was able to get the NIC and USB passthrough working fine. When I > removed the VNC Graphic device and Video Display, then add the NVIDIA card > (graphics and au

Re: [vfio-users] NVIDIA GPU Passthrough to Win10 - Driver Disabled (Code 43)

So I tried a BIOS VM, freshly created from scratch. Reinstalled Windows 10 and was able to get the NIC and USB passthrough working fine. When I removed the VNC Graphic device and Video Display, then add the NVIDIA card (graphics and audio) the machine doesn't boot and needs to be forced off. Even r

Re: [vfio-users] NVIDIA GPU Passthrough to Win10 - Driver Disabled (Code 43)

On 07/25/2016 04:29 PM, Steven Bell wrote: On Mon, Jul 25, 2016 at 5:23 PM, Philip Abernethy wrote: Did you make sure that the card actually supports UEFI mode? Because my GTX 660 didn't. But the OS (Arch in my case) would happily boot with the UEFI set to UEFI-only mode and the card would silen

Re: [vfio-users] NVIDIA GPU Passthrough to Win10 - Driver Disabled (Code 43)

Interesting. I'll give the nvflash a try and post back my results. On Mon, Jul 25, 2016 at 5:23 PM, Philip Abernethy wrote: > Did you make sure that the card actually supports UEFI mode? Because my > GTX 660 didn't. But the OS (Arch in my case) would happily boot with the > UEFI set to UEFI-only

Re: [vfio-users] NVIDIA GPU Passthrough to Win10 - Driver Disabled (Code 43)

Did you make sure that the card actually supports UEFI mode? Because my GTX 660 didn't. But the OS (Arch in my case) would happily boot with the UEFI set to UEFI-only mode and the card would silently run in legacy mode. I had to update the card's ROM. I used nvflash to check the card's UEFI capabil

Re: [vfio-users] NVIDIA GPU Passthrough to Win10 - Driver Disabled (Code 43)

On Mon, 25 Jul 2016 17:11:13 -0400 Steven Bell wrote: > Hi Alex, > > My original XML config created by virt-manager did not have any hyperv > related options. There was no tag, nor was there a clock for > hyperv. > > Why would these be missing? Since they aren't there and I still have a > prob

Re: [vfio-users] NVIDIA GPU Passthrough to Win10 - Driver Disabled (Code 43)

Hi Alex, My original XML config created by virt-manager did not have any hyperv related options. There was no tag, nor was there a clock for hyperv. Why would these be missing? Since they aren't there and I still have a problem, should I try adding them and setting the vendor_id value to "Nvidia

Re: [vfio-users] NVIDIA GPU Passthrough to Win10 - Driver Disabled (Code 43)

On Mon, 25 Jul 2016 15:50:51 -0500 Jayme Howard wrote: > I don't have an XML example handy, but you're missing the Nvidia workaround > flags. In the CLI version, they look like the following: > > OPTS="$OPTS -cpu > host,kvm=off,hv_time,hv_relaxed,hv_vapic,hv_spinlocks=0x1fff,hv_vendor_id=Nvidia

Re: [vfio-users] NVIDIA GPU Passthrough to Win10 - Driver Disabled (Code 43)

Hi Jayme, Thanks for the reply! My QEMU version is 2.6.0-5.fc23. I'm pointed to the Fedora virt-preview repo. I'd imagine that's why I've got 2.6 not 2.5. I'll look into the hv_flags and see if I can't find the XML equivalent to enable them. Will post my results. On Mon, Jul 25, 2016 at 4:50 PM,

Re: [vfio-users] NVIDIA GPU Passthrough to Win10 - Driver Disabled (Code 43)

I don't have an XML example handy, but you're missing the Nvidia workaround flags. In the CLI version, they look like the following: OPTS="$OPTS -cpu host,kvm=off,hv_time,hv_relaxed,hv_vapic,hv_spinlocks=0x1fff,hv_vendor_id=Nvidia43FIX" All those hv_ flags are necessary to get around Nvidia's vi

[vfio-users] NVIDIA GPU Passthrough to Win10 - Driver Disabled (Code 43)

Hello, I am currently trying to setup a Windows 10 VM on a Fedora 23 host with QEMU-KVM passing through a NIC, a USB controller, and a NVIDIA GPU (GTX 670). With my current setup, the NIC and USB Controller are both passed through and function without issue. The GPU driver gives the message "Windo

[vfio-users] Nvidia drivers (host) and KMS: fail load vfio

Hi. i've tested today the active the KMS in the nvidia drivers on my arch https://wiki.archlinux.org/index.php/kernel_mode_setting https://wiki.archlinux.org/index.php/NVIDIA#DRM_kernel_mode_setting but found problems when active my VFIO installation: ul 23 23:25:15 sL1pKn07 systemd-udevd[1006]

Re: [vfio-users] Nvidia glitches on Windows 10 guest

I forgot to explain something which is maybe important for my issue. I have this problem ONLY when i use both NVIDIA cards. My 960 card is for Linux and the 970 for Win10 Guest. If i use only the 970 for the Guest and the IGP Intel for the Host (and disable the 960), i can allocate without problem

Re: [vfio-users] Nvidia glitches on Windows 10 guest

I did a bootable memtest and i have not errors with my RAM ... I install actually Win7 to see if this issue appears with it too. 2016-06-22 7:47 GMT+02:00 Milos Kaurin : > Would you be able to run a Bootable memtest just to make sure you dort > have faulty ram? > On Jun 21, 2016 8:15 PM, "Berilli

Re: [vfio-users] Nvidia glitches on Windows 10 guest

Would you be able to run a Bootable memtest just to make sure you dort have faulty ram? On Jun 21, 2016 8:15 PM, "Berillions" wrote: > New try with a direct live for an Euro 2016's match : > > 4GB VM : corruption, glitches and freeze > 2GB VM : nothing ... > > > > 2016-06-21 20:35 GMT+02:00 Beril

Re: [vfio-users] Nvidia glitches on Windows 10 guest

New try with a direct live for an Euro 2016's match : 4GB VM : corruption, glitches and freeze 2GB VM : nothing ... 2016-06-21 20:35 GMT+02:00 Berillions : > Test if i allocate 4GB to the VM. > > - VM not launched : > >> AnonHugePages: 0 kB >> HugePages_Total:2050 >> HugePages_Free

Re: [vfio-users] Nvidia glitches on Windows 10 guest

Test if i allocate 4GB to the VM. - VM not launched : > AnonHugePages: 0 kB > HugePages_Total:2050 > HugePages_Free: 2050 > HugePages_Rsvd:0 > HugePages_Surp:0 > Hugepagesize: 2048 kB > - VM launched : > AnonHugePages: 10240 kB > HugePages_Total:205

Re: [vfio-users] Nvidia glitches on Windows 10 guest

On Tue, Jun 21, 2016 at 12:28 PM, Berillions wrote: > To enable the Hugepages and to allocate 4096, i only follow the Debian > Wiki like i said. And concerning the VM, i only add in the .xml : > " > > " > And when the VM is running does HugePages_Free in /proc/meminfo decrease from Hu

Re: [vfio-users] Nvidia glitches on Windows 10 guest

To enable the Hugepages and to allocate 4096, i only follow the Debian Wiki like i said. And concerning the VM, i only add in the .xml : " " 2016-06-21 20:13 GMT+02:00 Alex Williamson : > On Tue, Jun 21, 2016 at 12:05 PM, Berillions wrote: > >> You're right, I badly understood the use

Re: [vfio-users] Nvidia glitches on Windows 10 guest

On Tue, Jun 21, 2016 at 12:05 PM, Berillions wrote: > You're right, I badly understood the use of Hugepage. > > Like you say, 4096 is enough to use 4GB in my VM but unfortunatly with > this, i still have freeze, corruption in the VM ... > > In resume : > - 4096 Hugepages + 2GB VM = good > - 4096

Re: [vfio-users] Nvidia glitches on Windows 10 guest

You're right, I badly understood the use of Hugepage. Like you say, 4096 is enough to use 4GB in my VM but unfortunatly with this, i still have freeze, corruption in the VM ... In resume : - 4096 Hugepages + 2GB VM = good - 4096 Hugepages + 4GB VM = bad 2016-06-21 19:58 GMT+02:00 Alex Williamson

Re: [vfio-users] Nvidia glitches on Windows 10 guest

On Tue, Jun 21, 2016 at 11:53 AM, Berillions wrote: > When i said in my message "real 8Gb Memory Ram", it means that I have 2 > slots each one which contain 4Gb Memory Ram. > "real 12Gb Memory Ram", it means that I have 3 slots each one which > contain 4Gb Memory Ram. > > And yes, i created Hugep

Re: [vfio-users] Nvidia glitches on Windows 10 guest

When i said in my message "real 8Gb Memory Ram", it means that I have 2 slots each one which contain 4Gb Memory Ram. "real 12Gb Memory Ram", it means that I have 3 slots each one which contain 4Gb Memory Ram. And yes, i created Hugepages for the VM with the Debian Wiki ( https://wiki.debian.org/Hu

Re: [vfio-users] Nvidia glitches on Windows 10 guest

Are you using huge pages? On June 21, 2016 10:31:22 AM PDT, Berillions wrote: >Hi guys, > >I found where come from the issue. I have really in my PC 12Gb Memory >RAM >(but the issue exist with real 8Gb Memory Ram). > >If i give 4Gb or more to the VM, i have the glitches, corruption and >freeze.

Re: [vfio-users] Nvidia glitches on Windows 10 guest

On Tue, Jun 21, 2016 at 11:31 AM, Berillions wrote: > Hi guys, > > I found where come from the issue. I have really in my PC 12Gb Memory RAM > (but the issue exist with real 8Gb Memory Ram). > > If i give 4Gb or more to the VM, i have the glitches, corruption and > freeze. But, if i give only 2Gb

Re: [vfio-users] Nvidia glitches on Windows 10 guest

Hi guys, I found where come from the issue. I have really in my PC 12Gb Memory RAM (but the issue exist with real 8Gb Memory Ram). If i give 4Gb or more to the VM, i have the glitches, corruption and freeze. But, if i give only 2Gb for the guest the issue does not exist and i can use Win10 correc

[vfio-users] Nvidia glitches on Windows 10 guest

- Which is the DPC ? - How to see correctly the dps latency in the guest ? - for the cpu command line, i have only -cpu host, kvm=off - You can see the script that i use to launch the vm here: http://forum.hardware.fr/hfr/OSAlternatifs/Logiciels-2/unique-passthrough-linux-sujet_74047_11.htm#t139105

Re: [vfio-users] Nvidia glitches on Windows 10 guest

Quick question, are you posting a new email everytime? Because i got like 2 or 3 of those on the same request, very annoying. Never heard about this kind of issue though, virtual net device load should be insignificant. 2016-06-16 12:17 GMT+02:00 Quentin Deldycke : > Possibly because managing a v

Re: [vfio-users] Nvidia glitches on Windows 10 guest

Possibly because managing a virtual device is expensive in cpu time? Possibly because these cpus who are managing these interrupts or shared with the vm? Possibly because your dpc is bad when having a network interface? Much possibilities, no information, much things already answered in this maili

[vfio-users] Nvidia glitches on Windows 10 guest

Nobody to help me how to resolv this issue ? Or where i must to search to give your more informations. I have my linux system on a ssd and the qemu image file on a HDD. I don't understand why Network driver reduce and finally freeze the system. Uninstall it and the Win10 works great. I need you

[vfio-users] Nvidia glitches on Windows 10 guest

Hi, I found where come from the problem and it's the Network option in Qemu which freeze, slow my Win10 system. I tried more Network options : - Virtio option : Very bad performance when installed from virtio CD. Uninstall the driver and Win10 works like a charm but no Internet sadly. - rtl8139 o

[vfio-users] Nvidia card passed through but unable to install drivers in Windows guest

Hi, You must to disable kvm for the CPU to use Nvidia card. I don't use libvirt but with qemu, it's something like : qemu-system-x86_64 \ -enable-kvm \ -m 4096 \ -k fr \ -cpu host,kvm=off \ ___ vfio-users mailing list vfio-users@redhat.com https://www.r

[vfio-users] Nvidia card passed through but unable to install drivers in Windows guest

I am able to passthrough a nvidia card (tried both 630 and 950) to a windows 10 guest, but once I install the nvidia driver in windows I start getting a blue screen and the vm reboots constantly saying system thread exception not handled (nvlddmkm.sys). I have also tried windows 7, but I'm stil

[vfio-users] Nvidia glitches on Windows 10 guest

Gmail Google+ Agenda Web plus Messages envoyés VGA Passthrough issue with Nvidia cards

Re: [vfio-users] Nvidia Error 43 and vendor_id state='on'

Thank you for the Reply Nicolas. Thanks for the info on the USB controllers, i may in the future play with that and try to bind them only when the VM is running. As for my XML config, at the time I copied it, there was an issue with the Nvidia card giving an error 43 in device manager after boot.

Re: [vfio-users] Nvidia Error 43 and vendor_id state='on'

On 2016-06-11 18:25, David wrote: OK, here we go. Understand my XML is a little in pieces right now, as i am trying to re-add the settings and the one under . In order to get the VM to boot i had to remove the Nvidia card. I first tried putting back all 3 hyperv settings and the clock setting

Re: [vfio-users] Nvidia Error 43 and vendor_id state='on'

An update on my progress. I powered off the PC and reseated the video card and re-hooked the power cables. After another try with all the settings below out of the XML, i was able to get it to boot ok. No idea if the error 43 will return, but my last 2 test boots have gone OK. I have tried putt

Re: [vfio-users] Nvidia Error 43 and vendor_id state='on'

OK, here we go. Understand my XML is a little in pieces right now, as i am trying to re-add the settings and the one under . In order to get the VM to boot i had to remove the Nvidia card. I first tried putting back all 3 hyperv settings and the clock setting, but that didnt work. So then i to

Re: [vfio-users] Nvidia Error 43 and vendor_id state='on'

> > > ... > Hmm.. This is the first time I saw this. My "Nvidia High Definition Audio" device no longer gives out the "error 47", so thanks. I never cared to fix it before because I am not using the HDMI sound. Is there an updated list of recommended settings?

Re: [vfio-users] Nvidia Error 43 and vendor_id state='on'

Now would probably be a good time to share your hardware details and exact XML you're using. Nobody wants to waste their time helping someone debug a laptop with Optimus. ___ vfio-users mailing list vfio-users@redhat.com https://www.redhat.com/mailman/li

Re: [vfio-users] Nvidia Error 43 and vendor_id state='on'

On 2016-06-11 21:57, David wrote: I have continued to work on this. I had and in my xml, and that didnt restore boot. I have the card out of the system right now. With this new setup i believe i should be able to re-add the 3 settings and the one under . But i have found that when put tho

Re: [vfio-users] Nvidia Error 43 and vendor_id state='on'

I have continued to work on this. I had and in my xml, and that didnt restore boot. I have the card out of the system right now. With this new setup i believe i should be able to re-add the 3 settings and the one under . But i have found that when put those lines back in (copying them from

Re: [vfio-users] Nvidia Error 43 and vendor_id state='on'

Required: ... ... ... ... ___ vfio-users mailing list vfio-users@redhat.com https://www.redhat.com/mailman/listinfo/vfio-users

Re: [vfio-users] Nvidia Error 43 and vendor_id state='on'

When I get error 43, I generally just delete the video card (and the virtual one) from the device manager a few times, rebooting in between. Eventually it comes back just fine. It's definitely not ideal, but it's worked for me so far. On Sat, Jun 11, 2016 at 8:53 AM, David wrote: > I am getting

[vfio-users] Nvidia Error 43 and vendor_id state='on'

I am getting the error 43 for my Nvidia card in a Win 7 guest with UEFI. Host is Fedora 23 fully updated. Is this document correct = (https://wiki.archlinux.org/index.php/PCI_passthrough_via_OVMF) (Under the Troubleshooting section near the bottom) that with new versions of QEMU and Libvert I shoul

Re: [vfio-users] Nvidia GTX 970 sound is choppy

Have you enabled MSI? http://vfio.blogspot.com.au/2014/09/vfio-interrupts-and-how-to-coax-windows.html Another idea, not sure if it will help, but you can change: to Physically the audio card lives on the same slot as graphic card. I have GTX 970, but I don't use it audio and passthrough the so

Re: [vfio-users] Nvidia GTX 970 sound is choppy

Alternately, if that doesn't help you, Razer's got a virtual surround app that generates a virtual sound card. I use that as my default sound card in the VM. That doesn't solve the problem if you're using the sound directly on the host, though. I play through in-home streaming, and Steam picks u

  1   2   >