Imagination and our commitment to open source
As you may have heard, Imagination is now delving into the world of open source and an initial merge request for an early Vulkan® driver and compiler for some of our PowerVR graphics IP has been added to the Mesa 3D graphics library. While over the years we have made contributions to the Linux kernel for embedded CPUs containing our IP, this marks the first time we have made significant contributions to the open-source community for GPUs. The Vulkan driver and compiler can be found in the public Mesa repo and we also offer a corresponding Linux driver and a firmware in public repos. The Linux kernel and firmware will be upstreamed later.
For this initial release, we are supporting some of our established IP and also some of our recent IP that will shortly be hitting shipping silicon. The older GPU is our PowerVR GX6250 from 2014, with a Chromebook Acer R13 laptop featuring a MediaTek MT8173 SoC acting as our target development device. The other two GPUs are from our recent A-Series and B-Series: namely the IMG AXE 1-16M and IMG BXS-4-64. These two GPUs will be featured in industrial and IoT devices where the products will have a long lifespan, and as such, long-term support is critical. By providing an open-source driver option, we are enabling the community to ensure long-term support, free from any corporate concerns or restraints. While we have started with the Vulkan API, for those interested in OpenGL® ES we will be providing support for this via Zink or ANGLE.
To read the full article, click here
Related Semiconductor IP
- nQrux® Root of Trust IP
- AXI to UCIe Bridge IP
- UCIe 2.x Controller IP
- SWI3S (SoundWire I3S Interface) Peripheral Controller Core IP
- OpenTitan-based RISC-V Secure Element
Related Blogs
- Running OpenCode with llama.cpp and Qwen 3.5 on E-Series
- Jeff Bier's Impulse Response - Open Source Digital Signal Processing?
- Using OSVVM for DVB-S2 IP Core Validation
- Beware licensing issues when using open source
Latest Blogs
- From Bug Hunting to Engineering Methodology: Lessons from AI-Driven Development
- Manufacturing Intelligence: Turning EDA Data into Trusted Action
- Execute-in-Place: Getting More from Embedded NVM
- Rethinking RTL flows with AI-driven hybrid formal verification
- Arm and NVIDIA: Building the trusted compute foundation for the agentic AI era