virtio-net-osx

AFAIK the only way to achieve this is to pass-through a GPU via VT-d. This config does not appear to support this though. If you search github for this line, you will find other configs that support doing this- https://github.com/kholia/OSX-KVM/blob/master/macOS-libvirt.... Edit: It's relatively easy- blacklist (or uninstall) GPU drivers from host machine, and pass the PCI address to qemu. Performance is not perceptibly different from native. Other things you probably want to do are enable hugepages, pin virtual cpus to physical cores, and use vfio for network- https://github.com/pmj/virtio-net-osx. It's also possible to run boot OSX using UEFI mode, which removes the need for the binary blobs in this repo.

I think it's a requirement of PCIe for devices to support DMA to memory with 64-bit physical addresses.[1] In any case, the hardware clearly supports it in the case - the bug was that a 32-bit I/O vector was being passed off to the hardware claiming to be a 64-bit one, and subsequently interpreted as such. The situation was certainly bad with classic PCI - I have an nForce5-based motherboard whose PCI bus silently truncates 64-bit DMA buffer addresses to 32 bits. The onboard ethernet chip does in principle support 64-bit DMA, so I was getting all sorts of weird behaviour after upgrading my RAM from 2 to 6 GiB. The southbridge chip in question is now blacklisted for 64-bit DMA in the Linux kernel. The 32-bit DMA cutoff isn't the only such constraint on buffer physical addresses of course; historically, ISA devices could only DMA to 20-bit