Inside this week's LWN.net Weekly Edition:

• Front: Finding concurrency bugs with sched_ext; Rust abstractions; 6.14 Merge window; Sealed system mappings; OpenSUSE board; Julia; Site tour.
• Briefs: Binutils 2.44; Firefox 135.0; Freedesktop GitLab; GNU C Library 2.41; GTK; Servo; Thunderbird updates; Sanctions; Quotes; ...
• Announcements: Newsletters, conferences, security updates, patches, and more.

The Servo Rust-based rendering engine project has published an article summarizing its progress in 2024, and plans for the future:

Servo main dependencies (SpiderMonkey, Stylo and WebRender) have been upgraded, the new layout engine has kept evolving adding support for floats, tables, flexbox, fonts, etc. By the end of 2024 Servo passes 1,515,229 WPT subtests (79%). Many other new features have been under active development: WebGPU, Shadow DOM, ReadableStream, WebXR, ... Servo now supports two new platforms: Android and OpenHarmony. And we have got the first experiments of applications using Servo as a web engine (like Tauri, Blitz, QtWebView, Cuervo, Verso and Moto).

Over the past year or so, LWN has added a number of useful new features for our subscribers to enhance the experience of reading and commenting on our content. Those features are of little use, however, to readers who do not know about them. It has been more than a decade since we last provided a tour of the site—it seems that another is in order. Walk this way for a look at the LWN kernel source database (KSDB), enhanced commenting features, EPUB downloads, and more.

Jake Hillion gave a presentation at FOSDEM about using sched_ext, the BPF scheduling framework that was introduced in kernel version 6.12, to help find elusive concurrency problems. In collaboration with Johannes Bechberger, he has built a scheduler that can reveal theoretically possible but unobserved concurrency bugs in test code in a few minutes. Since their scheduler only relies on mainline kernel features, it can theoretically be applied to any application that runs on Linux — although there are a number of caveats since the project is still in its early days.

Security updates have been issued by Debian (firefox-esr), Fedora (fastd, ovn, and yq), Mageia (libreoffice), Slackware (mozilla), SUSE (google-osconfig-agent, grafana, helm, and rime-schema-all), and Ubuntu (linux-azure, linux-azure-5.4, linux-lowlatency, openjdk-17, openjdk-21, openjdk-23, openjdk-8, and openjdk-lts).

Jeff Xu has been working on a patch set that makes certain mappings in a process's address space impossible to change, sealing them against tampering. This has some potential security benefits — mainly, making sure that someone cannot relocate the vsyscall and vDSO mappings — but some kernel developers haven't been impressed with the patches. While the core functionality (sealing the mappings) is sound, some of the supporting code for enabling and disabling the new feature caused concern by going against the normal design for such things. Reviewers also questioned how this feature would interact with checkpointing and with sandboxing.

Version 135.0 of the Firefox web browser has been released. Changes include more languages for the translations feature, increasing roll-out of the credit-card autofill and AI chatbot features, and (perhaps most welcome):

Firefox now includes safeguards to prevent sites from abusing the history API by generating excessive history entries, which can make navigating with the back and forward buttons difficult by cluttering the history. This intervention ensures that such entries, unless interacted with by the user, are skipped when using the back and forward buttons.

Security updates have been issued by Debian (openjdk-17), Fedora (chromium, fastd, ovn, and yq), Mageia (libxml2 and redis), Oracle (gstreamer1-plugins-base, gstreamer1-plugins-good), Red Hat (buildah, bzip2, galera, mariadb, grafana, keepalived, libsoup, mariadb:10.11, mariadb:10.5, mingw-glib2, podman, python-jinja2, and rsync), SUSE (bind, ignition, java-11-openjdk, java-17-openjdk, krb5, libxml2, openssl-1_1, orc, python-asteval, rsync, and xrdp), and Ubuntu (harfbuzz, libndp, libvpx, and opencv).

By the time that Linus Torvalds released 6.14-rc1 and closed the merge window for this development cycle, some 9,307 non-merge changesets had been pulled into the mainline repository — the lowest level of merge-window activity seen in years. There were, nonetheless, a number of interesting changes in the 5,000 commits pulled since the first-half merge-window summary was written.

Matthias Clasen has written a short update on a GTK hackfest that took place at FOSDEM and what's coming in GTK 4.18. This includes fixes for pointer sizes in Wayland when fractional scaling is enabled, removal of the old GL renderer in favor of the GL renderer introduced in GTK 4.13.6, and deprecation of X11 and Broadway backends with intent to remove them in GTK 5.

The deprecated backends will remain available until then, and no action is required by developers at this time, Clasen wrote: "There is no need to act on deprecations until you are actively porting your app to the next major version of GTK, which is not on the horizon yet".

Security updates have been issued by AlmaLinux (git-lfs, libsoup, and unbound), Debian (dcmtk, ffmpeg, openjdk-11, pam-u2f, and python-aiohttp), Fedora (buku, chromium, jpegxl, nodejs18, nodejs20, and rust-routinator), Mageia (clamav, kernel, kmod-virtualbox, kmod-xtables-addons & dwarves, and kernel-linus), SUSE (apptainer, bind, buildah, chromedriver, clamav, dovecot24, ignition, kubelogin, libjxl, libQt5Bluetooth5-32bit, orc, owasp-modsecurity-crs, python-pydantic, python311-ipython, and stb), and Ubuntu (linux-azure and netdata).

Linus has released 6.14-rc1 and closed the merge window for this release.

This is actually a _tiny_ merge window, and that's ok. The holidays clearly meant that people did less development than during a normal cycle, and that then shows up as a much smaller-than-average release. I really felt like this year we got the whole holiday season release timing right, and this is just another sign of that.

Version 2.44 of the GNU Binutils package has been released. Perhaps the most significant change is the absence of the "gold" linker, which is deprecated and about to disappear entirely. Gold appeared in 2008 with some fanfare as a faster linker, but it has suffered from a lack of maintenance in recent years. This release also includes some architecture-specific assembler improvements, and some (non-gold) linker enhancements.

The 6.13.1, 6.12.12, 6.6.75, 6.1.128, 5.15.178, 5.10.234, and 5.4.290 stable kernel updates have all been released; each contains another set of important fixes.

Julia, a free, general-purpose programming language aimed at science, engineering, and related arenas of technical computing, has steadily improved and widened its scope of application since its initial public release in 2012. As part of its 1.11 release from late 2024, Julia made several inroads into areas outside of its traditional focus, provided its users with advances in tooling, and has seen several improvements in performance and programmer convenience. These recent developments in and around Julia go a long way to answer several longstanding complaints from both new and experienced users. We last looked in on the language one year ago, for its previous major release, Julia 1.10.

The election to replace outgoing openSUSE board members is underway, with four candidates vying for three seats. The election was initially scheduled to be completed in December, but the timeline was extended due to too few candidates standing for the seats. Voting closes on February 2 and the results are expected to be announced on February 3.

The Linux Foundation has published its long-awaited article on international sanctions and open-source development. This is the reasoning that went into the removal of a group of Russian kernel maintainers in October.

It is disappointing that the open source community cannot operate independently of international sanctions programs, but these sanctions are the law of each country and are not optional. Many developers work on open source projects in their spare time, or for fun. Dealing with U.S. and international sanctions was unlikely on the list of things that most (or very likely any) open source developers thought they were signing up for. We hope that in time relevant authorities will clarify that open source and standards activities may continue unabated. Until that time, however, with the direct and indirect sponsorship of developers by companies, the intersection of sanctions on corporate entities leaves us in a place where we cannot ignore the potential risks.

Security updates have been issued by AlmaLinux (libsoup), Debian (debian-security-support and redis), Fedora (expat, java-21-openjdk, lemonldap-ng, and phpMyAdmin), Mageia (chromium-browser-stable and git-lfs), Oracle (bzip2, git-lfs, libsoup, mariadb:10.11, mariadb:10.5, python-jinja2, redis, and unbound), Red Hat (git-lfs, libsoup, python-jinja2, rsync, and unbound), SUSE (buildah, chromium, google-osconfig-agent, govulncheck-vulndb, hauler, ignition, krb5, libxml2, python311-pydantic, SDL2_sound, and trivy), and Ubuntu (jquery, linux-azure, linux-azure-4.15, linux-azure-5.15, linux-hwe-5.4, linux-oracle, and mysql-8.0).

While the path toward the ability to write device drivers in Rust has been anything but smooth, steady progress has been made and that goal is close to being achieved — for some types of drivers at least. Device drivers need to be able to set up memory areas for direct memory access (DMA) transfers, though; that means Rust drivers will need a set of abstractions to interface with the kernel's DMA-mapping subsystem. Those abstractions have run into resistance that has the potential to block progress on the Rust-for-Linux project as a whole.

Visitors to the freedesktop.org GitLab instance are currently being greeted with a message noting that the company who has been hosting it for free for nearly five years, Equinix, has asked that it be moved (or start being paid for) by the end of April. The issue ticket opened by Benjamin Tissoires in order to track the planning of a move is clear that the project is grateful for the gift: "First, I'd like to thank Equinix Metal for the years of support they gave us. They were very kind and generous with us and even if it's a shame we have to move out on a short notice, all things come to an end."

The current cost for the services, much of which is for 50TB of bandwidth data transfer per month and a half-dozen beefy servers for running continuous-integration (CI) jobs, comes to around $24,000 per month. Tissoires believes that the project should start paying for service somewhere, in order to avoid upheaval of this sort, sometimes on short or no notice. "I personally think we better have fd.o pay for its own servers, and then have sponsors chip in. This way, when a sponsor goes away, it's technically much simpler to just replace the money than change datacenter." Various options are being discussed there, but any move is likely to disrupt normal services for a week or more.