With the Windows Subsystem for Linux (WSL), Microsoft makes it easy to run Linux software on its Windows system. Previously, this was limited to command line tools or servers that are available via HTTP, for example. In future, the WSL will also support graphic applications such as Microsoft announced at its in-house fair Build which takes place purely virtually this year.
So far, GUI applications could already be used in WSL – but only via a few detours, such as applications created specifically for Windows, such as an X server from external sources. According to Microsoft, these kinds of handicrafts will no longer apply. In order to be able to display the graphic applications, Microsoft uses according to the detailed blog post to a Wayland server in the WSL. This in turn communicates with an RDP client on the actual Windows system.
The Gnome file manager runs in the WSL next to a Windows application. (Image: Microsoft)
The WSL also supports the use of graphics cards and GPUs. This enables the use of hardware acceleration for your own developments in the WSL environment. Microsoft refers here to the general parallelization of compute tasks and specifically to the use for training and working with machine learning models. It is not yet known whether direct access to the hardware is possible here. Since the WSL itself now uses Hyper-V, it can be assumed that this also applies to the GPU support. Microsoft plans to publish details on this later.
Microsoft also wants to significantly simplify the installation of the WSL itself wsl.exe -install suffice. Thanks to the integration of various distributions in the Windows Store, using the WSL is already comparatively easy.