The Linux 7.1 kernel is bringing performance improvements for Sheaves , the per-CPU caching layer introduced several kernel cycles ago (Linux 6.18) for better efficiency on today's high core count hardware. Sheaves began as an opt-in feature but since Linux 7.0 is now being used for all caches .

With the expanded use of Sheaves, regression reports have turned up in some scenarios and so for Linux 7.1 some regressions are hopefully now addressed. A recent focus was on improving

The SDL library widely used by cross-platform games and part of the Steam Runtime has now established a policy to block code contributions made using AI / Large Language Models (LLMs).

SDL is the latest open-source project to say no to code generated by AI/LLMs. After a ticket was opened last week questioning if the Simple DirectMedia Layer has a policy around AI/LLM contributions, stakeholders took to coming up with such a policy. This policy follows GitHub Copilot used on some recent SDL reviews and a

The workqueue changes merged today for the Linux 7.1 kernel are significant for today's modern high-end processors where there can be many CPU cores per last level cache (LLC / L3 cache). The new WQ_AFFN_CACHE_SHARD affinity scope can reduce some contention on such systems and help achieve greater performance.

Linux engineer Breno Leitao with Meta worked on the set of patches for introducing the WQ_AFFN_CACHE_SHARD affinity scope to address observed bottlenecks where there are many CPU cores

Archinstall 4.2 is now available as the latest update to this very convenient, text-based Arch Linux OS installer.

Significant with Archinstall 4.2 is a security fix to avoid writing encryption keyfiles as plain text to unencrypted root partitions. There was a mistake in Archinstall until now that if encryptiong partitions like /home but not the root partition, the key files would be written to the unencrypted root in plain text. With Archinstall 4,2 key files are only generated when the root is also encrypted. This

Last month I ran benchmarks showing the very positive performance impact FRED has on Intel's new Panther Lake processors while wondering why Flexible Return and Event Deliver wasn't enabled by default yet on Linux. Hours after that story was published, an Intel engineer posted the patch to enable FRED by default with the rationale they were waiting for hardware to be publicly released in order to evaluate the performance benefit. Days after that the FRED-by-default patch hit tip/tip.git and now as of yesterday that patch

Eric Engestrom stepped up again to serve as Mesa release manager for this quarter's Mesa 26.1 feature release. Mesa 26.1-rc1 was just released in kicking off the weekly release candidate dance until Mesa 26.1 stable is ready for debut in May.

Mesa 26.1 is bringing many new Vulkan driver extensions, various performance optimizations, and other improvements for gaming especially on Intel and AMD RadeonSI/RADV hardware as well as some nice improvements to NVK and the smaller drivers

A change proposal has been filed to build x86_64-v3 micro-architecture feature level packages alongside the existing x86_64-v1 packages for Fedora Linux.

The change proposal is for Fedora 45 due to be released in October and would be adding x86_64-v3 as a primary architecture alongside the existing x86_64 (v1) baseline packages. This change proposal in current form isn't about ending existing support

Linux developer Qais Yousef with Google has announced the alpha release of Sched QoS as a new initiative for user-space assisted scheduling. The scheduling model in turn is based in part on Apple's quality of service classes used by iOS for classifying software as user interactive, user initiative, utility, or background tasks.

Due to the very diverse hardware platforms where Linux runs and the range of workloads encountered, Qais Yousef has been pursuing Sched QoS for user-space assisted scheduling. Sched QoS relies on configuration-based hinting