The grepc(1) program is something that originated in this project, as it helped me find code quickly in glibc and the Linux kernel. However, I've found it incredibly useful outside of this project. I'll take some space to announce it, as it's much more than just a tool for writing manual pages, and I expect it to be useful to most --if not all-- C programmers.

It is a command-line tool that finds C source code (for example, a function definition) in arbitrary projects. It doesn't use any indexing mechanism (unlike ctags and similar tools). This means that it can be used right after cloning some repository, without having to first generate an index.

Git is ubiquitous; in the last two decades, the version-control system has truly achieved world domination. Almost every developer uses it and the vast majority of open-source projects are hosted in Git repositories. That does not mean, however, that it is perfect. Patrick Steinhardt used his main-track session at FOSDEM 2026 to discuss some of its shortcomings and how they are being addressed to prepare Git for the next decade.

The postmarketOS project has published a recap from FOSDEM 2026, including the FOSS on Mobile devroom, and a summary of its post-FOSDEM hackathon. This includes decisions on governance and the project's AI policy:

AI policy: our current AI policy does not state that we forbid the use of generative AI in postmarketOS, so far this document just lists why we think it is a bad idea and misaligned with the project values. We discussed this and will soon change it (via merge request) to clearly state that we don't want generative AI to be used in the project. It was also noted that currently the policy is too long, it would make sense to split it into the actual policy and still keep, but separate the reasoning from it.

[...] Power delegation and teams: in over two hours we discussed how to move forward with [postmarketOS change request] PMCR 0008 to organize ourselves better, and how it fits with soon having a legal entity. We figured that we need to rename "The Board" (which is currently for financial oversight) to "Financial Team", as we will soon have a new board for the legal entity. In the end our idea was to have the new board refer to an "assembly" for all important decisions, and this "assembly" would just be all Trusted Contributors in postmarketOS. The Core Contributors team would be dissolved in favor of having several topic-specific teams (a lot of which we already have, such as the infra team). This way we would have a very flat decision structure. The PMCR will be updated soon and discussed further there. Casey also asked on fedi for further feedback and got a lot of input.

Other topics include reaching out to resellers to sell phones with postmarketOS preinstalled, security, and more.

Greg Kroah-Hartman has unleashed six new stable kernels: 6.18.10, 6.6.124, 6.12.70, 6.1.163, 5.15.200, and 5.10.250. Each one contains important fixes throughout the tree; users are advised to upgrade.

His professional accomplishments and impact are almost endless, but often captured by one moniker: "grandfather of the Internet," acknowledging the foundational contributions made by his many students at the University of California, Irvine; the University of Delaware; the University of Pennsylvania; and Carnegie Mellon University.

See also: this announcement by Manny Farber on Farber's "Interesting People" list.

Matthias Clasen has published a short summary of the GTK hackfest held prior to FOSDEM 2026. Topics include discussions on unstable APIs, a decision to bump the C runtime requirement to C11 in the next development cycle, limiting changes in GTK3 to crash and build fixes, as well as the state of accessibility:

On the accessibility side, we are somewhat worried about the state of AccessKit. The code upstream is maintained, but we haven't seen movement in the GTK implementation. We still default to the AT-SPI backend on Linux, but AccessKit is used on Windows and macOS (and possibly Android in the future); it would be nice to have consumers of the accessibility stack looking at the code and issues.

On the AT-SPI side we are still missing proper feature negotiation in the protocol; interfaces are now versioned on D-Bus, but there's no mechanism to negotiate the supported set of roles or events between toolkits, compositors, and assistive technologies, which makes running newer applications on older OS versions harder.

Michiel Leenaars, director of strategy at the NLnet Foundation, used his keynote at FOSDEM to sound warnings for the community for free and open-source (FOSS) software; in particular, he talked about the threats posed by geopolitical politics, dangerous allies, and large language models (LLMs). His talk was a mix of observations and suggestions that pertain to FOSS in general and to Europe in particular as geopolitical tensions have mounted in recent months.

Version 3.0 of the Offpunk offline-first, command-line web, Gemini, and Gopher browser has been released. Notable changes in this release include integration of the unmerdify library to "remove cruft" from web sites, the xkcdpunk standalone tool for viewing xkcd comics in the terminal, and a cookies command to enable browsing web sites (such as LWN.net) while being logged in.

Something wonderful happened on the road leading to 3.0: Offpunk became a true cooperative effort. Offpunk 3.0 is probably the first release that contains code I didn't review line-by-line. Unmerdify (by Vincent Jousse), all the translation infrastructure (by the always-present JMCS), and the community packaging effort are areas for which I barely touched the code.

So, before anything else, I want to thank all the people involved for sharing their energy and motivation. I'm very grateful for every contribution the project received. I'm also really happy to see "old names" replying from time to time on the mailing list. It makes me feel like there's an emerging Offpunk community where everybody can contribute at their own pace.

There were a lot of changes between 2.8 and 3.0, which probably means some new bugs and some regressions. We count on you, yes, you!, to report them and make 3.1 a lot more stable. It's as easy at typing "bugreport" in offpunk!

See the "Installing Offpunk" page to get started.

Sean Whitton has announced that Debian's tag2upload service is now out of beta and ready for use by Debian developers and maintainers.

During the beta we encountered only a few significant bugs. Now that we've fixed those, our rate of successful uploads is hovering around 95%. Failures are almost always due to packaging inconsistencies that older workflows don't detect, and therefore only need fixing once per package.

We don't think you need explicit approval from your co-maintainers anymore. Your upload workflows can be different to your teammates. They can be using dput, dgit or tag2upload.

LWN covered tag2upload in July 2024.

The most significant changes in 6.19 include initial support for Intel's linear address-space separation feature, support for Arm Memory system resource Partitioning And Monitoring, the listns() system call, a reworked restartable-sequences implementation, support for large block sizes in the ext4 filesystem, some networking changes for improved memory safety, the live update orchestrator, and much more. See the LWN merge-window summaries (part 1, part 2) and the KernelNewbies 6.19 page for details.

What is the goal of using ML models in Linux kernel? The main goal is to employ ML models for elaboration of a logic of particular Linux kernel subsystem based on processing data or/and an efficient subsystem configuration based on internal state of subsystem. As a result, it needs: (1) collect data for training, (2) execute ML model training phase, (3) test trained ML model, (4) use ML model for executing the inference phase. The ML model inference can be used for recommendation of Linux kernel subsystem configuration or/and for injecting a synthesized subsystem logic into kernel space (for example, eBPF logic).

It is rigorously undocumented and there are no real users, so it's not entirely clear what the purpose is, but there are undoubtedly interesting things that could be done with it.

Greg Kroah-Hartman has released the 6.18.9, 6.12.69, 6.6.123, 6.1.162, 5.15.199, and 5.10.249 stable kernels. As always, each contains important fixes throughout the tree; users are advised to upgrade.

This is a major release for the project, seeing several substantive new features that users have asked for over a long period of time. Region FX, clip recording, a touch-sensitive GUI, pianoroll windows, clip editing and more, not to mention dozens of bug fixes, new MIDI binding maps, improved GUI performance on macOS (for most) ...

We expect to get feedback on some of the major new features in this release, and plan to take that into account as we improve and refine them and the rest of Ardour going forward. We have no doubt that there will be both delight and disappointment with certain things - rather than assume that we don't know what we're doing, please leave us feedback on the forums so that Ardour gets better over time. Those of you new to our clip launching implementation might care to read up on the differences with Ableton Live.

In the coming weeks, we'll begin to sketch out what we have planned next for Ardour, in addition to responding to the feedback we get on this 9.0 release.

Control-flow integrity (CFI) is a set of techniques that make it more difficult for attackers to hijack indirect jumps to exploit a system. The Linux kernel has supported forward-edge CFI (which protects indirect function calls) since 2020, with the most recent implementation of the feature introduced in 2022. That version avoids the overhead introduced by the earlier approach by using a compiler flag (-fsanitize=kcfi) that is present in Clang but not in GCC. Now, Kees Cook has a patch set adding that support to GCC that looks likely to land in GCC 17.

The Linux From Scratch (LFS) project provides step-by-step instructions on building a customized Linux system entirely from source. Historically, the project has provided separate System V and systemd editions, which gave users a choice of init systems. Bruce Dubbs has announced the project will no longer produce the System V version:

There are two reasons for this decision. The first reason is workload. No one working on LFS is paid. We rely completely on volunteers. In LFS there are 88 packages. In BLFS there are over 1000. The volume of changes from upstream is overwhelming the editors. In this release cycle that started on the 1st of September until now, there have been 70 commits to LFS and 1155 commits to BLFS (and counting). When making package updates, many packages need to be checked for both System V and systemd. When preparing for release, all packages need to be checked for each init system.

The second reason for dropping System V is that packages like GNOME and soon KDE's Plasma are building in requirements that require capabilities in systemd that are not in System V. This could potentially be worked around with another init system like OpenRC, but beyond the transition process it still does not address the ongoing workload problem.

[...] As a personal note, I do not like this decision. To me LFS is about learning how a system works. Understanding the boot process is a big part of that. systemd is about 1678 "C" files plus many data files. System V is "22" C files plus about 50 short bash scripts and data files. Yes, systemd provides a lot of capabilities, but we will be losing some things I consider important.

The next version, 13.0, is expected in March and will only focus on systemd.