The ACO "Amd COmpiler" started by Valve for the Mesa RADV Vulkan driver has shown it can do wonders for Linux gaming performance and reducing game load times compared to AMD's official AMDGPU LLVM shader compiler back-end. Recently thanks to the work of Qiang Yu there has been much work hitting upstream Mesa for beginning to enable using the ACO compiler by the RadeonSI Gallium3D driver...
With Chrome 114 is the start of Google beginning to roll-out Maglev as their new mid-tier compiler for further enhancing the JavaScript browser performance...
A new set of patches this weekend begin laying the groundwork for Rust abstractions for Linux network device drivers so that Rust code can be used for constructing new network device drivers. The patches also include a dummy Rust network driver...
One of the most interesting aspects of the new AMD Ryzen 7040 series laptop processors is the new "Ryzen AI" capabilities with the new XDNA AI engine capabilities built into the SoC, leveraging IP from their Xilinx acquisition. Linux support details remain scarce but at least one of their (Windows) demos for showcasing Ryzen AI is open-source...
On Friday a big set of patches affecting the AMDGPU/Radeon/AMDKFD kernel drivers were submitted for DRM-Next to queue until the Linux 6.5 kernel merge window opens in the coming weeks. A lot of new feature code is part of this pull for benefiting new hardware, continuing to refine AMD GPU power management under Linux, and more...
With Intel TDX and AMD SEV-SNP for better securing virtual machines on the mainline Linux kernel, memory is accepted/initialized immediately at boot time by the VMs although the capability exists to have "unaccepted memory" where that memory is only dealt with by the VMs later on or on an as-needed basis. For two years now Intel engineers have been working on this unaccepted memory support and this week posted their thirteenth iteration of these fundamental Linux kernel patches...
The Portable Computing Language "PoCL" began as an open-source CPU-based OpenCL implementation that has become quite a comprehensive implementation over the years. But in leveraging the LLVM/Clang compiler stack, over time PoCL has grown beyond just a CPU implementation to also support OpenCL execution on NVIDIA GPUs, AMD HSA-capable GPUs, and more. The latest now coming with PoCL 4.0 is support for Intel Level Zero execution for running this OpenCL implementation over Intel Arc Graphics GPUs...
Pages