Intel today formally announced the Optimization Zone as a new initiative at the company that began last October and is building up a centralized repository for maximizing performance and software optimizations around Intel hardware.

The Intel Optimization Zone is their new centralized hub for documenting software performance optimizations for running the best on Intel hardware from BIOS tunables for given workloads to recommended application/server software settings for achieving peak performance on Intel hardware.

The Intel Optimization Zone will provide tuning guides, optimization recipes and best-known methods, performance analysis guidance, and optimal

Archinstall 4.0 is out today and just in time for the April 1 monthly refresh to the Arch Linux installer. With Archinstall 4.0, this Arch Linux OS installer is now using the Textual TUI library rather than the Curses library.

Archinstall 4.0 is now using Textual rather than Curses to provide a more modern text-based user interface. THe developers feel this provides a better looking terminal-based installer for Arch Linux rather than the mature curses library.

For those wanting to make use of Linux GPU compute software under Windows 11 by way of Windows Subsystem for Linux (WSL2), AMD's ROCDXG "librocdxg" library is now deemed production-ready for delivering open-source ROCm compatibility with WSL.

Last week's release of ROCm 7.2.1 paired with the new Adrenalin 26.2.2 Windows 11 driver now is ready to go for a nice ROCm experience under WSL. The ROCDXG library enables ROCm functionality inside WSL,

Hitting the mailing list today are patches providing initial support for AMD's next-gen NPU "AIE4" platform, complete with SR-IOV support.

AMD is working on bringing up their next-gen NPU IP under Linux. The patches provide initial support for the AMD NPU AIE4 platform atop the AMDXDNA accelerator driver in the Linux kernel. AMD's Ryzen AI NPU leverages the AI Engine "AIE" processors that originally was based on work by Xilinx. Each AI Engine consists of a vector processor,

Coreboot 26.03 was christened today as the newest quarterly feature release for this open-source system firmware implementation that strives to replace proprietary BIOS/firmware. Most notable with Coreboot 26.03 is full support for Intel Core Ultra Series 3 Panther Lake SoCs.

Coreboot 26.03 brings full support for Intel Panther Lake SoCs. Intel engineers continue working on Coreboot for client processors primarily for meeting Google Chrome OS design requirements. Occasionally we also see such ports used in turn by the likes of

The open-source RadeonSI Gallium3D driver with Rusticl for modern Rust-based OpenCL is nearing formal OpenCL 3.0 conformance with all necessary OpenCL test cases passing. Making this all the more interesting is that this is the first modern AMD graphics hardware in a decade likely to see formal recognition for OpenCL conformance with AMD having not submitted any of their own OpenCL conformance results since 2015.

Rusticl lead developer Karol Herbst of Red Hat posted on Mastodon this weekend that he has fixed all the remaining Conformance Test

Last week I ran benchmarks quantify the performance benefit to Intel FRED for Flexible Return and Event Delivery initially found with the new Core Ultra Series 3 "Panther Lake" processors and also for upcoming Nova Lake and Diamond Rapids CPUs. The FRED performance impact was very beneficial across a variety of workloads but rather strangely was not enabled by default. Mere hours after publishing that article, an Intel engineer posted a patch to enable FRED by default . Now this week that patch appears all-set for merging with the upcoming Linux 7.

A proposal was submitted today for launching the RISC-V Compiler Collaboration "RVCC" as an LLVM Incubator project to focus on compiler optimizations for better performance on RISC-V. But before getting too excited, there is already some opposition to the proposal.

RVCC aims to be a staging area where RISC-V performance optimization patches for LLVM/Clang can be collected and iterate quicker than going into LLVM proper. RVV aims to lower the time / review costs on patches while still being developed/tested, the project would