As part of my various end-of-year benchmarks, recently I looked at the Linux LTS kernel performance on AMD EPYC 9005 over the past year, the AMD EPYC Milan-X performance over the past four years, and various other performance comparisons over time to look the evolution of the Linux software performance. Another run I had carried out was looking at the AMD EPYC 8004 "Siena" series since its launch just over two years ago. Here is a look at how an up-to-date Linux software stack can deliver some additional performance gains for these energy efficiency and cost-optimized server processors.

A patch queued up into tip/tip.git's x86/cpu Git branch ahead of the upcoming Linux 6.20~7.0 kernel cycle enables the Intel Transactional Synchronization Extensions (TSX) functionality by default on the mainline kernel for capable CPUs and those not affected by side-channel attacks due to TSX Async Abort (TAA) and similar vulnerabilities. For newer Intel CPUs with safe TSX support, this change can mean better performance with the kernel defaults...

If all goes well the upcoming Ubuntu 26.04 LTS release will further enhance the NVIDIA graphics performance under its default GNOME Wayland session. The improvements might be upstreamed to GNOME 50 in time but otherwise it's looking like Ubuntu 26.04 will carry its own patch(es) for improving the NVIDIA Wayland performance...

The Linux kernel patches talked about at the start of the year for more easily changing the boot logo of Tux are now queued into a "for-next" branch and thus expected to be submitted for the upcoming Linux 6.20~7.0 kernel cycle. Those wanting to replace the Tux icon with an alternative logo during the Linux kernel boot process could already patch the file manually but this new code allows for an easy replacement via Kconfig options...

For those looking for a speedy Basic Linear Algebra Subprograms "BLAS" library, OpenBLAS 0.3.31 is now available for this optimized open-source implementation...

One of the initiatives launched by Intel in 2025 was LLM-Scaler as part of Project Battlematrix. The open-source LLM Scaler is a Docker-based solution for helping to deploy Generative AI "GenAI" workloads on Intel Battlemage graphics cards with frameworks like vLLM, ComfyUI, SGLang, and more. There continues to be routine new feature releases of LLM Scaler for broadening the large language models supported and other improvements...

Wild 0.8 is now available as this speedy linker focused on iterative development, a goal of incremental linking, and written in the Rust programming language...

The GNOME 50 Alpha "50.alpha" release is now available for testing ahead of this open-source desktop's official release in March...

EndeavourOS 2026.01.12 "Ganymede Neo" is out as the first update of the year to this Arch Linux based distribution...

For those shopping for an AI-ready mobile workstation with NVIDIA RTX PRO Blackwell graphics, the Lenovo ThinkPad P1 Gen 8 offers a lot of potential for developers, AI researchers, content creators, and others. This Linux-friendly mobile workstation is well built and aligns with ThinkPad P-Series expectations while being ready to be tasked with demanding workloads.

The upcoming Linux 6.20~7.0 kernel cycle will provide expanded GPU temperature reporting capabilities for Intel graphics cards. Additional temperature sensors will now be exposed under Linux with the Intel Xe driver using the hardware monitoring (HWMON) interface for easy consumption by different Linux user-space software...

A significant update to Burn was released today, the MIT and Apache 2.0 licensed tensor library and deep learning framework written in the Rust programming language. Burn 0.20 brings some low-level changes as it continues to strive to deliver high performance AI across the diverse hardware ecosystem...

Sent out today was the latest DRM-Misc-Next pull request of new material ahead of the next kernel cycle either Linux 6.20 or 7.0 depending upon what Linus Torvalds decides to call it...

Whisper.cpp as the open-source high performance inference project built around OpenAI's Whisper and from the same developers as Llama.cpp / GGML is out with a big new release. Whisper.cpp 1.8.3 is capable of delivering a 12x performance boost for systems with integrated AMD and Intel graphics...

Started last year was D7VK as a project bringing Direct3D 7 implemented over the Vulkan API for enjoying better performance and support for legacy Windows games on Linux, akin to DXVK and VKD3D-Proton for newer versions of Direct3D over Vulkan that is used by Valve's Steam Play (Proton). Back in December D7VK added a Direct3D 6 front-end for allowing even older game titles to be accelerated using the modern Vulkan API. Today D7VK 1.2 is out for furthering the D3D6 support...

Libvirt 12.0 released today as this open-source virtualization API for management across different virtualization technologies/hypervisors. With libvirt 12.0, improving Bhyve as the FreeBSD hypervisor was a big focus...

DeviceTree patches worked on recently allow for the mainline Linux kernel to run on the ASUS "Kommando" IPMI Expansion Card. This is interesting for opening up new possibilities for this external IPMI/BMC expansion card but too bad that less than three years after launching it's difficult to find...

The oVirt 4.5.7 open-source virtualization management platform released this week after not seeing any new releases in two years. While Red Hat had started the oVirt open-source project for which their Red Hat Virtualization platform is based, since they shifted that to maintenance mode to focus on the Red Hat OpenShift platform and stopped contributing to oVirt, it's been up to the open-source community to keep it going...

In late 2024 the folks at Raspberry Pi announced the Raspberry Pi AI HAT+ as an AI accelerator capable of 26 TOPS and costing $110 for pairing with Raspberry Pi single board computers. Today they announced the much more capable Raspberry Pi AI HAT+ 2 that can begin to take on some generative AI "GenAI" models...

Separate from the Mesa merge request talked about earlier today for new RADV code that can deliver 10x faster ray-tracing pipeline compilation for this open-source Radeon Vulkan driver, another merge request landed today in Mesa 26.0 that was also carried out by Valve contractor Natalie Vock. That second merge request now in Mesa 26.0 delivers some additional gains for at least some ray-tracing games on RDNA3 and RDNA4 GPUs...

More than two years after the release of GRUB 2.12, GRUB 2.14 shipped today as the newest feature release of this widely-used bootloader on Linux systems and elsewhere...

There still are several months to go until the official Ubuntu 26.04 LTS release -- including one month until the feature freeze and the future Linux 6.20~7.0 kernel is expected to land too before the latter kernel freeze in early April. But for those curious how Ubuntu 26.04 is looking so far for servers, here are some very early benchmarks of it on AMD EPYC 9005 "Turin" in its present development state. The main motivation here for this early look was stemming from the recent rolling-release CachyOS benchmarks on AMD EPYC and wanting to see how it goes up against the current development state of Ubuntu Linux.

A new merge request opened today for Mesa's Radeon Vulkan driver "RADV" by Valve contractor Natalie Vock provides another significant boost for the Vulkan ray-tracing performance in multiple titles...

Last year GlobalFoundries acquired MIPS while an interesting new development announced today is that GlobalFoundries has acquired the ARC Processor IP and its solutions business from Synopsys. The Synopsys ARC Processor IP will be brought into the MIPS umbrella...