steelbrain

@azurita_a @luizfrombrazil Testa o reims, da pra usar uma vm do mac os em qualquer pc com qualquer gpu steelbrain/reims-vgpu

Introducing vphone-linux: running a real iOS 26 virtual iPhone on Linux (x86_64) I've been porting vphone-cli to Linux which runs a virtual iPhone using Apple's private "vresearch101" research VM platform, normally only possible on Apple Silicon Macs through Virtualization.framework. Instead, this boots the same iOS 26.6.2 VM on an ordinary x86 Linux PC: - CPU: a custom QEMU machine model of Apple's VM platform, running ARM64e (pointer authentication and all) through QEMU's TCG emulation, with SEP emulation good enough for iOS to boot. - GPU: iOS uses Apple's paravirtual GPU driver. On the host side, a fork of reims-vgpu plus metal2vulkan translate its Metal command streams and shaders to Vulkan. - Input: a virtual USB touchscreen and keyboard, so it's usable with a mouse. - Host control: a vsock channel to a guest daemon for scripting (install apps, read logs, take screenshots). Current state: - It boots to the home screen with the wallpaper, widgets, Liquid Glass UI, touch input and apps (Settings, Maps, Calendar, Photos and more) working. - It's still slower than a real device, since everything is emulated, but it's getting usable. Main Goal: Full iOS app development environment outside macs. I'll publish it on GitHub with setup instructions once it's cleaned up. You'll still need to supply the firmware yourself: no Apple binaries will be distributed.

@makrotasuss 看起来metal还没实现，所以没有壁纸…？不知道有没有可能在ai帮助下搓一个metal驱动，就像nvidia跑metal那个项目 或者这个项目有帮助吗 steelbrain/metal2vulkan 总之期待一下

Mac OS VMs under Fedora Asahi Linux I run a cluster of Apple Mac M1/M2 Minis and Studios for development use. They all run Fedora Asahi Linux and they work really well. I mainly use them as headless servers for software dev work. I've recently been working on some iOS projects, which naturally need Mac OS/XCode to do the builds. CI runners on Github and hosted Mac OS machines in general are very expensive - a used Mac Mini is now dirt cheap. Mac OS, however, is not a great headless server OS and running CI jobs should ideally be done with ephemeral storage to keep things clean and predictable. I therefore really wanted a way of running my Mac OS CI jobs under Linux on my existing M-series Apple hardware inside VMs. I guided a combination AI agents into investigating the shortcomings of steelbrain's alpha/experimental work. Things like partial/slow GPU support, VM boot failures (required a virtual clock), memory ballooning, etc. and ultimately reverse engineered, improved efficiency of and production-readied steelbrain's work. This involved a lot of testing, a lot of benchmarking, a few kernel panics. Mac OS VMs now run efficiently and with GPU acceleration under Fedora Asahi Linux. You can find the work here: aquarat/apple-macos-arm64-vm-guest I've been using them as runners for my Forgejo CI system, but this work should be easily extendable to Github Actions and beyond. Initial image generation still needs a Mac OS machine. There are graphs 😆

后者 metal2vulkan 主要是 steelbrain/reims-vgpu 这个实现。和我之前提到 switch 模拟器用到的 nvn2vulkan 原理都是指令翻译。 性能方面苹果 gpu 走的是 tbdr 大缓存换带宽路线，直接 vulkan 翻译损失比较大，不过问题不大可以大力出奇迹。

reims-vgpu working on Windows (WSL) with nVidia :) Am I the first? reims-vgpu#32

@lin72h Already pushed to the development branch! github.com/steelbrain/reims-vgpu/commit/871e09316…

@JoshuaSWarren @trymirai @ashxhart Hello! This is somewhat similar to something I’m doing. Would you like to collaborate? steelbrain/metal2vulkan