Gentoo Removed Chromium, a Kernel Maintainer Is Stepping Back From Some Review, and an Arch Developer's Resignation Left Packages Needing Owners
Gentoo removed its Chromium package on 5 November, a kernel maintainer is stepping back from some patch review, and an Arch resignation left packages needing maintainers; the Herald compares the three responses.
Overview
Three Linux projects lost or rationed maintainer time between 1 August and 13 November 2026, and each answered differently. Gentoo removed its source-built Chromium package: the removal commit in Gentoo’s ebuild repository carries the UTC date 2026-11-05. Lorenzo Stoakes, who says he holds four kernel maintainerships, told The Register that he is keeping them but stepping back from some patch review. And after the Arch Linux developer Morten Linderud resigned on 1 August, Phoronix listed packages that needed new maintainers; in the Herald’s own check on 14 November, described below, nine of the ten packages Phoronix names list a maintainer. The Herald’s reading, labelled as such under Implications, is that what gave way differed in each case: ownership of packages at Arch, a whole package at Gentoo, and the breadth of one maintainer’s review in the kernel.
Background
Gentoo and Chromium. LWN.net, in an article by Joe Brockmeier dated 6 October, explains that Gentoo users build software from source with Portage and can choose compile-time options, and that Chromium has a complex build system, bundles many of its dependencies and releases often. LWN quotes the package maintainer Sam James: “Chromium is horrible to maintain as a source package. It is easily the package in Gentoo that burns through maintainers the most.” It quotes Matt Jolly calling the work “a commitment to volunteer at least a business day’s worth of your free time each week”, and reports that Chromium’s stable channel gets a new major branch roughly every two weeks and a new version every week. Jolly adds that “we’ve had the release cycle shorten twice since I started maintaining the package in 2023”.
Review volume in the kernel. The Register’s Liam Proven reports that Stoakes, a principal software engineer at Arm, wrote to the kernel mailing list: “Given the current throughput in the kernel resulting from AI-generated series, I unfortunately lack the bandwidth to generally be able to offer review except in those areas I maintain.” The same article quotes Greg Kroah-Hartman, from an “Ask the Expert” session at the Open Source Summit that The Register’s FOSS desk attended and that the article of 13 November places in the previous month: “All of the maintainers are burnt out, and we have been for ten years.” Both are the speakers’ own characterisations, and neither Register article cited here gives a measured figure for review workload.
An Arch resignation. Phoronix, whose byline gives 1 August 2026 at 07:47 AM EDT (US Eastern), reports that Linderud, an Arch developer, security team member and package maintainer for a decade, announced he was resigning, and quotes him that “it’s a good time to let go”. Phoronix says packages needing a new maintainer include mkinitcpio, Bolt, nvme-cli, Go-Tools, Pacman, arch-install-scripts, archlinux-keyring and archlinux-repro, and that he was the sole maintainer of WPA_Supplicant and fsverity-utils. The article gives no workload reason for the departure and restates his resignation email and blog post; none of the other cited sources covers it.
Current situation
Gentoo: from last rites to removal. According to LWN, James declared “last rites” for the package on 24 September, after a bug report in which a user complained on 9 September that the stable package was outdated; last rites are Gentoo’s notice that a package will be unmaintained and open to a new maintainer. LWN says the period ran to 24 October. In the bug thread, Jolly said that restoring the package would need at least three people: “History has shown that one maintainer isn’t working”. Linuxiac reported on 18 October that removal was scheduled for 24 October unless a volunteer took over and that the package was already masked; it summarises Gentoo’s mailing-list announcement and bug tracker and adds a list of alternative browsers. The 5 November commit, titled “www-client/chromium: treeclean” and citing the same bug number LWN links, deletes 17 files and 14,945 lines. By the Herald’s arithmetic that is 12 days after 24 October, and none of the cited sources explains the gap or says whether anyone volunteered.
Two pieces of context change how this reads. LWN checked other distributions: Fedora’s Chromium package and the one in Debian’s security repository were each one version behind the 154.0.8037.97 upstream release of 1 October, and LWN concludes that “Keeping up with Chromium is not an easy task.” And Gentoo users keep options: Linuxiac says Google Chrome stays available as a precompiled package, while LWN cautions that the Flathub Chromium Flatpak is built by a third party. One figure differs in wording: Linuxiac describes six hours a week as the current maintainer’s personal time, while LWN quotes Jolly as counting only compilation time on a fast PC; the Herald follows LWN.
The kernel: narrower review, same maintainerships. The Register says Stoakes “isn’t quitting” kernel development. He told it: “I am retaining my maintainerships and have no plans on giving those up, I just want my inbox to be less horrendous.” He adds that AI is part of the reason and that “in general, it’s gone from ‘once a week’ to ‘daily’ to ‘half of what I see in review’”, and that the wave is hitting memory management “across the board”. That is his account; the article does not verify the proportion. It says his email came “just the day after” Linus Torvalds saw signs that the tide was turning. The Register’s earlier report on Torvalds’s Linux 7.4-rc2 post explains the apparent tension: Torvalds relayed a claim that on the bpf side fixes now arrive faster than new reports, and added “I think most other areas haven’t reached that point yet”. The two accounts do not conflict, since Torvalds spoke of bpf and Stoakes of memory management.
Support windows: six branches. In the same Register article Kroah-Hartman says: “I want to go down to two Long Term Support releases – four, maximum. Six is too many.” kernel.org’s release page, read on 14 November, lists six longterm branches (6.18, 6.12, 6.6, 6.1, 5.15 and 5.10), with projected end-of-life months of December 2028 for 6.18 and 6.12, December 2027 for 6.6 and 6.1, and December 2026 for 5.15 and 5.10. FOSS Force reported on 2 March that Kroah-Hartman had moved the end of life of 6.12 from December 2026 to December 2028 and of 6.18 from December 2027 to December 2028, after “lots of discussions with different companies and groups and the other stable kernel maintainer”. That article restates his announcement and adds a table.
Arch: owners on nine of ten. The Herald queried the JSON behind Arch’s public package search on 14 November (16:24 UTC) for the ten names in the Phoronix report. All eight packages Phoronix lists as needing a new maintainer show at least one maintainer: mkinitcpio, Bolt, nvme-cli, archlinux-keyring and archlinux-repro show two each, and Go-Tools, Pacman and arch-install-scripts one. Of the two named as solely his, WPA_Supplicant shows one maintainer and fsverity-utils shows none. This is the Herald’s own count. The search shows current maintainers only, so it cannot say which are new, and it covers official repositories, not the AUR; Phoronix says he also maintained “many other packages”, which the Herald did not check.
Implications
The Herald’s reading. Across the three cases, what gave way was not the same thing. At Arch, the sources show a person leaving and, by the Herald’s check, most of the named packages now listing maintainers; nothing cited says the departure was about workload. At Gentoo, the project dropped the package entirely, and Jolly says restoring it would take at least three people where one had not worked. In the kernel, a maintainer cut the breadth of his review while keeping his maintainerships. In the two cases where sources describe a cause, they point to rising workload: Chromium’s release cycle shortened, according to Jolly, and the volume of AI-generated patch series rose, according to Stoakes’s email. Torvalds’s caveat is that in bpf, at least, fixes may now be arriving faster than reports.
Support windows pull in two directions. By the Herald’s arithmetic from kernel.org’s dates, each of the six longterm branches was released in a different year from 2020 to 2025, and their projected lifetimes run from about three years (6.18) to about six (5.10). If the December 2026 dates hold, the list falls to four branches, the maximum Kroah-Hartman named; the sources do not say that is the plan. On the pattern in that table, longer lifetimes with a new branch each year mean more branches overlap, which works against a smaller count. The sources do not say how the two are to be reconciled.
What the sources leave open. None of the cited sources says whether any volunteer has come forward for Chromium, how Arch assigned the packages, or whether Stoakes’s change will spread to other maintainers. For users, the practical consequence in the Gentoo case is that source-built Chromium is gone from the repository, with Chrome binaries and a third-party Flatpak as the routes Linuxiac and LWN name.
Sourcing. The Register (kernel) and LWN (Gentoo) did original reporting; LWN’s page also carries comments from Gentoo developers, who are not independent of the project. Linuxiac, Phoronix and FOSS Force restate project announcements, as described above. The Gentoo commit and kernel.org are the projects’ own records. Earlier Herald coverage is in the report on Gentoo’s planned Chromium removal and the report on the Arch resignation.