Re: [vfio-users] Hyper-v enlightments on recent NVIDIA drivers?

> I also read in a forum (https://lime-technology.com/forum/index.php?top > ic=43824.0) that Alex Williamson is working on a QEMU patch to allow > Windows guests to use Hyper-V enlightments without Nvidia beeing able > to detect is. Is this true?. Where could I follow the development / > test it?

Re: [vfio-users] Occasional freeze when 3 VMs with passthrough are running

> Another question: is kernel 4.2 broken with OVMF? I thought about upgrading to 15.10 to see if the issue is still there. Mostly yes. It freezes on the first stages of booting with 100% one core cpu load. I don't remember if there are any fixes available except using "lts" 4.1.13 or 4.3.x

Re: [vfio-users] Physical disk passthrough and OVMF

I wrote to this list as it appears that there were some folks with similar problems few months ago. Okay then, the bisect message will be on edk2 list when it's done. On Mar 29, 2016 6:25 PM, "Laszlo Ersek" <ler...@redhat.com> wrote: > Blank Field <ihatethisfield@...&

Re: [vfio-users] Guest does not boot OS with GPU passed through, stuck at TianoCore boot splash

100% one core CPU? What kind of disk device you have? Are you passing through a physical disk? What you can do - convert libvirt xml into a usual qemu command line, and add an ISA debug console and check OVMF debug output in it. On Apr 10, 2016 5:04 PM, "Stefan Seil" wrote:

Re: [vfio-users] Guest does not boot OS with GPU passed through, stuck at TianoCore boot splash

He's also should be able to attach just the .efi file or install graphics drivers via qxl. On Apr 11, 2016 6:48 PM, "Laszlo Ersek" <ler...@redhat.com> wrote: > On 04/10/16 16:46, Blank Field wrote: > > > What you can do - convert libvirt xml into a usual qemu comm

[vfio-users] Physical disk passthrough and OVMF

Hello again, fellow subscribers. Since 4.3.X kernels I am experiencing some strange problems with disk IO. I am passing through the whole /dev/sda(since i also dualboot into that system for some weird software or hardware) to my VM, and since recent versions of kernel the VM gets stuck in an

Re: [vfio-users] Physical disk passthrough and OVMF

OVMF is from git, upgrading QEMU ATM to 2.5.x. Apparently, fedora has 2.5 qemu only in beta 24. Thanks for the advice, didn't know about the new qemu release. On Mar 27, 2016 1:01 AM, "Dan Ziemba" <zman0...@gmail.com> wrote: > On Fri, 2016-03-25 at 01:10 +0300, Blank Field wro

[vfio-users] Translation project, info updates needed.

I've started a (yet)one-man-army translation project with the objective of bringing info and knowledge about VFIO to my native language. I've partially translated aw's blog post about iommu groups dated back to august 2014, and since that time a lot has changed. For example, the post says that

Re: [vfio-users] GPU HDMI audio passthrough weird behavior?

On Apr 24, 2016 10:54 AM, "Karsten Mettke" wrote:. > I am starting the kernel with > vfio_iommu_type1.allow_unsafe_interrupts=1 kvm.allow_unsafe_assigned_interrupts=1 kvm.ignore_msrs=1 > It was required for the GPU already so i dontknow if that would help with hdmi. Were you

Re: [vfio-users] BAR 0: can't reserve

> On Thu, Apr 28, 2016 at 2:41 PM, Blank Field <ihatethisfi...@gmail.com> > wrote: > >> Nevermind, the end user was dumb as usual. >> >> I've developed a habit of doing grub2-mkconfig -o /boot/grub2/grub.cfg >> which efi grub ignores, storing it's config on the

Re: [vfio-users] BAR 0: can't reserve

Nevermind, the end user was dumb as usual. I've developed a habit of doing grub2-mkconfig -o /boot/grub2/grub.cfg which efi grub ignores, storing it's config on the ESP instead. So you should do grub2-mkconfig -o /boot/efi/EFI/fedora/grub.cfg I wondered why this "video=efifb:off" wouldn't appear

[vfio-users] BAR 0: can't reserve

vfio-pci :01:00.0: BAR 0: can't reserve [mem 0xc000-0xcfff 64bit pref] cat /proc/iomem shows... b000- : PCI Bus :00 b000-bfff : PCI Bus :02 b000-bfff : :02:00.0 c000-cfff : PCI Bus :01 c000-cfff : :01:00.0

Re: [vfio-users] Windows 7 successfully migrated from BIOS to OVMF/UEFI

On May 16, 2016 12:27 PM, "Leonhard Preis" wrote: > > My dear fellow vfio users, > > today I want to share my success[1] with you. In case you have Windows 7 and currently use BIOS but would like to switch to UEFI or are about to set up your first Windows VM with passthrough and

Re: [vfio-users] cannot boot windows guest after update

Are you sure your grub cfg is alright? On May 5, 2016 1:33 AM, "thibaut noah" wrote: > By modifying my xml i managed to go into repair mode, trying to just run > windows 10 get me a blue screen with inaccessible boot device code. > Looking into that. > > 2016-05-05 0:15

Re: [vfio-users] input-linux first impressions

I've the exact same problems when i used qemu's built in SDL window. On Apr 14, 2016 5:13 PM, "Ivan Volosyuk" wrote: > First of all it is pretty cool to be able to skip the synergy and > short-circuit multiple layers on the way of keyboard and mouse events. It > was

Re: [vfio-users] UEFI & virt-manager on Arch

I'll try to predict: A config file somewhere in /etc/ should be redacted. On Apr 15, 2016 11:33 AM, "Garland Key" wrote: > 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

Re: [vfio-users] Looking for performance tips, qemu machine type and pci/pcie-root questionning?

On Apr 15, 2016 9:01 AM, "thibaut noah" wrote: > Speaking of drives, it seems from what i read that it is possible for qemu/kvm to read a native (non virtualize) install of windows from a passthrough drive. > If so is there something special to do? Might go to the hassle

Re: [vfio-users] Fwd: UEFI & virt-manager on Arch

That article is being rewritten in a neighbor thread, beware. On Apr 15, 2016 3:36 PM, "Garland Key" wrote: > 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,

Re: [vfio-users] KMS and vfio modules: vfio fail if active with nvidia

On Jul 24, 2016 2:17 AM, "Blank Field" <ihatethisfi...@gmail.com> wrote: > Search for kernel parameter options. Also, if nvidia-drm is a module, you can do modinfo to find out all parameters. Yeah, doing modinfo on nvidia, nvidia_modeset, nvidia_uvm and nvidia_drm should show

Re: [vfio-users] KMS and vfio modules: vfio fail if active with nvidia

On Jul 24, 2016 2:10 AM, "sL1pKn07 SpinFlo" wrote: > linux /boot/vmlinuz-linux > root=UUID=1b76a96e-0c70-49c5-a90b-c5308f969a2f rw intel_iommu=on > nvidia-drm.modeset=1 > > how select one specific card in kms? Search for kernel parameter options. Also, if nvidia-drm is a

Re: [vfio-users] KMS and vfio modules: vfio fail if active with nvidia

On Jul 24, 2016 1:33 AM, "sL1pKn07 SpinFlo" wrote: > > Hi. i found problems when active the KMS suport for my linux > > https://wiki.archlinux.org/index.php/NVIDIA#DRM_kernel_mode_setting > My phone almost hung up when i pressed "respond inline" button. Use pastebin or fpaste.

Re: [vfio-users] KMS and vfio modules: vfio fail if active with nvidia

On Jul 24, 2016 2:32 AM, "sL1pKn07 SpinFlo" wrote: > i'm not sure what you mean with: > > > Also keeping the first gpu busy before KMS shoots out may help > > how make this? Doing vfio-pci.ids=pci-ids-of-your-gpu. I doubt that would help. Read my previous message about modinfo

[vfio-users] New presentation from KVM Forum

Hey all! I've just noticed a new blogpost on vfio.blogspot.com with a link to this years' presentation from AW. http://awilliam.github.io/presentations/KVM-Forum-2016 You can clearly see the progress that has been made since previous presentation. Cheers for VFIO.

[vfio-users] Occasional kernel Oops happened.

Alright, i know my machine has a lot of dodgy constructions, but that was something new. The VM just froze, libvirtd just froze and became . Had to kill -9 libvirtd after killing qemu, and then the VM has started again successfully. dmesg(last): https://paste.fedoraproject.org/511253/44033414/