Skip to content

Explore Venus + MoltenVK for GPU acceleration #4551

Description

@osy

Currently we use VirGL + ANGLE to translate GL (guest) to Metal (host). This works decently (on Linux) but the downside is that it’s buggy (crashes) and more modern Linux applications and games are moving to Vulkan.

Venus translate guest Vulkan calls to host Vulkan calls.

MoltenVK translates host Vulkan calls to Metal calls.

It is worth exploring this pairing to see if it’s a) more stable and b) more performant.

Note that neither solution currently has Windows guest support so that will have to be developed separately.

Metadata

Metadata

Assignees

No one assigned

    Labels

    enhancementNew feature or request

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions