pavo

Members
  • Posts

    22
  • Joined

  • Last visited

Everything posted by pavo

  1. I have seen some RX 6xxx series GPUs have issues booting when booting with verbose enabled. Try removing verbose booting. Running RX 6600 XT on my Intel Server.
  2. MacOS VMs should be using virtio for ethernet type. It gives you the best compatibility and the highest throughput.
  3. Yeah pretty much the same results but it is 100% an issue with Unraid not a SMBIOs issue. Something changed in the way vfio or the kernel handles passing through the wifi pci card to the VM. It's crashes the VM on any OS if you are using any version of Unraid higher than 6.10.3. Test this on both AMD and Intel systems with the same issue.
  4. So I have confirmed an issue with any version of Unraid above 6.10.3 and passing through a Fenvi T919 pci card to a macOS VM. It will show inside the macOS VM but as soon as you connect the Wifi to any network, it will crash macOS. This also happens on any other VM OS as in Windows and Linux VMs. There has been something changed in Unraid higher than 6.10.3 that doesn't allow this device to be passed through to a VM correctly. If you need anything from me so you guys can figure this issue out, please let me know and I will post it.
  5. Not sure why, I have 1x HDMI and 2x DP monitors working without WEG perfectly fine. The last time I needed WEG was back in the R9 290X days, went to a Sapphire Nitro+ OC RX 480, Sapphire Nitro+ Vega 64, Sapphire Radeon VII and now Sapphire Nitro+ RX 6900 XT all without needing WEG using HDMI and multiple DP ports.
  6. Yeah macOS doesn't like passive adapters. You would need an active DP to VGA adapter.
  7. @ghost82 agdpmod=pikera is only needed if you use WhateverGreen, which the RX 6xxx XT series does not need. All display ports work without WEG.
  8. Would be awesome to have the ability to create new Libvirt networks that use Linux bridges in the Network Settings section of the webGUI. Having to make a new xml and manually define and start is such a pita..... Also add a checkbox to make it persistent or not. Similar to this....
  9. Good to know... thank you. Will have to test this out because I have 10Gb transfer rate from external NAS without any issues. Completely agree... Forgot to fix the picture Will have to test this, I use SSDTs to create ACPI devices for all my passed-through devices which makes them show as internal. Have been using these settings for both BigSur and Monterey and csrutil works perfectly fine in recovery. Having control over SIP doesn't mean you should. With using kext injection within OpenCore and the varies plugins you can use for Lilu, I haven't ran into any reason to make changes too SIP. Plus it makes it more secure operationally. Correct using `cpu=host` does mean host passthrough, but in order to make a AMD cpu have a vendor-id of `GenuineIntel` You must provide the cpu type before it, hence the reason for cpu=host. Using it this way doesn't require any AMD kernel patches and passes through ALL host CPU features that is supported by the OS, including 7leaf cpu features. As you can see I am also using a Intel CPUID in the config.plist for OpenCore to emulate an Intel CPU.
  10. Let me know what you guys think.... https://github.com/Pavo-IM/Hackinabox
  11. @ich777 First of all.... very impressive container. Does this also update Qemu versions? The latest Qemu version is 6.0.0 rc4 but I am only seeing Qemu 5.1.
  12. Good news, no more AMD reset bug with RX 6xxx XT series GPUs in macOS 11.4 beta 1 on UnRAID.
  13. Nested virtualization that uses macOS's native hypervisor will not work with AMD cpus, because macOS doesn't support the AMD-V (SVM) cpu feature that AMD cpus have. MacOS's native hypervisor only support Intel's vmx CPU feature.
  14. Because virtualbox provides its own hypervisor cpu instruction set. I was talking about macOS's native hypervisor. Until someone writes a kernel extension to add AMD-V (SVM) support to the kernel and the hypervisor framework for macOS, anything that uses the native macOS hypervisor will not work.
  15. This will never work because macOS kernel nor the hypervisor for macOS support AMD-V (SVM) cpu feature. Basically put, anything that uses macOS's hypervisor requires Intel's vmx cpu feature.
  16. Anyway to get this added to Unraid? https://github.com/gnif/vendor-reset/
  17. Disable SIP for Big Sur is FF0F0000 value and you can boot recovery if you enable JumpstartHotPlug under UEFI > APFS, but only use it to boot the Recovery, not normal boot.
  18. Why use a qcow? Why not just a USB drive like normal and either passthrough the USB Controller or pass it through as a USB device.
  19. Well anyways here is the xml I used for Unraid https://github.com/Pavo-IM/Hackinabox
  20. grrr for some reason the code block is taking out what needs to be posted.
  21. You can use the following addition to the xml and have no need for FakeSMC or VirtualSMC. <qemu:commandline> <qemu:arg value='-device'/> <qemu:arg value='isa-applesmc,osk="************************"'/> </qemu:commandline>