A Student Developer Wants To Write A Software Implementation Of Vulkan For Mesa
Vulkan is going on one year old and while the hardware driver support has continued to advance, we haven't yet seen a software implementation of Vulkan for running on a CPU. Of course, not for expecting any performance miracle or the like, but as a vendor-neutral platform for being able to test Vulkan's behavior, certain fallback scenarios, and other use-cases like Mesa's LLVMpipe/swrast/Softpipe software rasterizers.
There is a student developer now wanting to write a Vulkan software implementation for Mesa either as part of Google Summer of Code (GSoC) or X.Org's Endless Vacation of Code (EVoC). But as the X.Org Foundation had recently tightened up the EVoC rules to only be relevant to existing X.Org contributors, it would have to be a GSoC project assuming it gets accepted and the student can find a mentor.
The student developer wrote on Mesa-dev, "I would like to write a software implementation of Vulkan for inclusion in mesa3d. I wanted to use a tiled renderer coupled with llvm and either write or use a whole-function-vectorization pass."
Let's hope that the student developer is genuine, capable, and that the work indeed will be able to come to fruition. Of course, longtime Phoronix readers will recall several years ago when another ambitious student wanting to write an OpenGL 4.1 state tracker from scratch in just one summer while that was quickly defeated... Seeing a software Vulkan implementation would certainly be interesting from a technical standpoint and the above mentioned reasons.
There is a student developer now wanting to write a Vulkan software implementation for Mesa either as part of Google Summer of Code (GSoC) or X.Org's Endless Vacation of Code (EVoC). But as the X.Org Foundation had recently tightened up the EVoC rules to only be relevant to existing X.Org contributors, it would have to be a GSoC project assuming it gets accepted and the student can find a mentor.
The student developer wrote on Mesa-dev, "I would like to write a software implementation of Vulkan for inclusion in mesa3d. I wanted to use a tiled renderer coupled with llvm and either write or use a whole-function-vectorization pass."
Let's hope that the student developer is genuine, capable, and that the work indeed will be able to come to fruition. Of course, longtime Phoronix readers will recall several years ago when another ambitious student wanting to write an OpenGL 4.1 state tracker from scratch in just one summer while that was quickly defeated... Seeing a software Vulkan implementation would certainly be interesting from a technical standpoint and the above mentioned reasons.
3 Comments