qemu VS kvm-guest-drivers-windows

https://gitlab.com/qemu-project/qemu/-/blob/master/hw/scsi/s...

QEMU's IOThreads allow the user to configure the threads and get something similar to thread per core architecture. But if 1 thread becomes a bottleneck, then some form of thread synchronization is needed again even with thread per core architecture. Some problems can be parallelized and they work well with thread per core.

Add `ENV ERL_FLAGS="+JPperf true"` to your Dockerfile and it will build just fine cross platform. The flag just changes some things during build time and won’t affect runtime performance.

[1] https://gitlab.com/qemu-project/qemu/-/issues/1034

> I don't know how rdcycle works on qemu.

That's a good question! I had to look it up myself ...

Obviously qemu TCG isn't a cycle-accurate emulation. Using RDCYCLE / reading the corresponding CSR eventually calls https://gitlab.com/qemu-project/qemu/-/blob/69680740eafa1838... which calls cpu_get_host_ticks is basically an arch-independent wrapper around RDTSC.

So it just measures the time taken to run using RDTSC. Which I guess is what you would want (maybe?). It would measure the time taken to emulate the vector instruction in host instructions.

> This benchmark is more meant for developers to figure out how to vectorize algorithms effectively, as in which instructions to choose.

Absolutely, I'm not saying the qemu results would say anything very deep, but they're kind of interesting from the point of view of either optimizing qemu or if you have to use qemu because the hardware you want isn't available / isn't cheap enough.

I see the commits that started switch support from RVV 071 to 100 start here, https://gitlab.com/qemu-project/qemu/-/commit/9ec6622db30df1c00d863c1ffc33341f9e0a534d

>> https://gitlab.com/qemu-project/qemu/-/issues/1696 ]

> Can I please just get the detail in mail instead of having to go look at random websites?

Maybe it's me but if I did boot boot linux 292.612 times to find a bug, you might as well click a link to a repository of a major open source project on a major git hosting service.

Is it really that weird to ask people online to check a website? Maybe I don't know the etiquette of these mail lists so this is a geniune question.

Capstone is used[1] by QEMU as disassembly engine in debug logs and in monitor mode debugger, by the way, so it's in the scope of the RISE effort.

[1] https://gitlab.com/qemu-project/qemu/-/blob/master/disas/cap...

A user on the qemu bugtracker found a way to get the Intel Arc working across resets without crashing the host: Just don't passthrough the audio device of the GPU and everything works!

[1]https://gitlab.com/qemu-project/qemu/-/commit/3412d8ec9810b819f8b79e8e0c6b87217c876e32 [2]https://gitlab.com/alex.williamson/qemu/-/commit/9a6d1822a2bd55f5dee1aec1b6529ae57949d5ba.patch

There's mature VirtIO drivers for just about everything already, under the virtio-win umbrella: https://github.com/virtio-win/kvm-guest-drivers-windows

My desktop PC is using libvirt+qemu (on an Arch host. I use Arch, btw) to PCI passthru my RTX 4090 GPU to a Windows guest. I installed the guest initially with emulated SATA for the main drive. Once Windows was up and running, I installed virtio-win and the guest is now using virtIO accelerated drivers for the network interface, main disk. I'm also sharing some filesystems using virtio-fs.

So, I started configuring the PC with ubuntu desktop (since I need a GUI for my use case). Then I setup a windows KVM with this. In my use case, I want the windows VM to be always connected to a VPN (Mullvad) that has the toggle of LAN connections turned on. This is because I want to use sunshine on the KVM to stream my desktop to a firestick. This is how I have been configuring one PC till now for my family to be able to watch TV. The issue is that I cannot use sunshine or parsec cos there are no encoders in the VM, thus the performance sucks. (I could use software encoders, but I would use too much CPU for windows VM, which would make me unable to run a LLM (this was my plan) )

Note also: it just got an experimental Windows driver!

[1]https://github.com/virtio-win/kvm-guest-drivers-windows/pull...

virtio-GPU for Windows for now is just a basic display driver, without any kind of acceleration, so there shouldn't be any difference between using it and QXL, which is fully emulated and in it, if necessary, you can increase the VRAM to be able to use high resolutions, as explained here. Note: this link is just from a quick search, I tried this for myself some time ago but my intention was not to increase the resolution but just to find out if it was possible to have some kind of 3D acceleration, as it happens in VirtualBox, but is not possible yet.

The only sorta-responsive display tech I've found is Virt with GL (3D) enabled. And if your physical box has 4k video you can forget about going anywhere near full screen with that guest VM.

I suggest you open a new issue in the repository describing this, or search if there is one open reporting the same. Also, keep in mind that as an early stage it might be a good idea to avoid copying important files using this, you could end up losing data unless you do an integrity check (recently I tried installing a game on the VM in a host directory and it didn't work. I'm thinking of an alternative or going back to using VirtualBox, since the shared folders worked very well for this purpose).

The problem was really bad with latest virtio drivers, so I downgraded to version 204, as per: https://github.com/virtio-win/kvm-guest-drivers-windows/issues/623