The FreeBSD Foundation exists to support the FreeBSD community and the FreeBSD project. Some of its projects are aimed at improving the experience of FreeBSD on specific hardware. There is an ongoing, and expanding, laptop experience project. To expand that project further, the foundation has provided Framework laptops to a bunch of developers working on the FreeBSD laptop and desktop experience. I’m one of those developers, and here are some initial notes on the process. The notes assume experience with FreeBSD.

Some disclaimers up front: the FreeBSD foundation is a lot like KDE e.V., which supports the KDE community and project. I wear a board hat for KDE e.V., but on the FreeBSD side I’m “just a ports developer”. Of course, the ports I try to work on are the KDE ones, so there’s a happy synergy here.

An anonymous donor sponsored these machines. While I am part of the FreeBSD donations@ team, I was not involved in the overall decision-making around this donation.

The machine I got is a Framework 13 with an AMD 7000 series CPU. That’s not the very-very latest one, which has a Ryzen 300 series in it, but it is at least 3 CPU generations newer than any other machine I have. For me in particular of interest is that it has the same GPU series, AMD Polaris 12, as my FreeBSD 14-STABLE desktop machine, so I can share experimentation with graphics drivers between them.

I picked the 2.8K display with rounded corners, because that’s potentially an interesting edge-case for the KDE Plasma 6 desktop; if there’s any funny-stuff needed for those corners, then we need to know about it.

Let’s Get Physical

Although it’s completely irrelevant for the long-term use of the laptop, I’ve got to hand it to the Framework folks: the packaging is really nice. Recyclable cardboard, well-laid-out, understandable boxes. I don’t often get a “huh, that’s clever” reaction when unpacking consumer electronics.

There’s a screwdriver included, cunningly hidden beneath the do-it-yourself-installation memory modules. That’s clever.

When it comes to putting the machine together, the installation guide with videos is both comprehensive and easy-to-follow. “Put DDR5 SO-DIMM modules in corresponding sockets” and “insert NVMe into socket” is straightforward, I do that all the time when (re)building desktop machines.

The bezel, on the other hand …

The bezel around the screen is just a thin bit of plastic. I got a red one, because FreeBSD (there is no KDE Blue option). It is essential to place it correctly, with all the screen-cables nicely aligned. I did not, and just clicked the bezel in place, pushed down on it and then closed the laptop, “per the instructions”. Except the bezel stuck out about 2mm, and on re-opening the laptop, it just about tore the bezel in half.

After 20 tricky minutes I could get the laptop open again and removed the bezel, repaired it, and tried again. I don’t really have a suggestion to improve the bezel installation except “try very carefully to close the laptop a bit, re-open, close a bit further, re-open, …” until it’s clear that the lid closes properly. Take some time to (re)route the cables to the screen so that they are as flat as possible.

Accessories

The little modules for the Framework laptop are pretty nifty. I’m already thinking I should have gotten an additional USB-C one. I selected one unusual module, RJ-45 wired ethernet, because my experience with FreeBSD and WiFi is not a good one. However, that’s what this whole laptop project is for. The FreeBSD Foundation has already funded work on laptop WiFi, so it’s probably over-cautiousness on my part.

With all the physical bits in place, the big question…

Will it run Doom?

Of course. Don’t be silly.

Will it run FreeBSD?

Yes, but that takes a little bit of effort. Download a FreeBSD 14.2 image and write it to a USB stick on some other machine. Leave it on your desk for now.

Boot the Framework laptop for the first time and let it do memory training and whatnot. Do not connect any devices and let it complain that there’s nothing to boot.

Reboot, still with nothing attached, and spam F2 during boot. You have to do this to get to the EFI shell / system configuration before it tries to boot anything. Disable secure boot. Linuxes have a signed GRUB shim nowadays, or other bits and pieces so they work with secure boot. FreeBSD 14.2 does not, yet.

Now insert the USB stick, reboot, and go through the installer process. It’s a text installer (still, as I still haven’t built FreeBSD support in Calamares) and gets you to a working system in about 5 minutes. Having the wired ethernet helps avoid any trouble here.

Reboot after installation and you can get a text console. All that technology for a late-80s user experience.

Will it run X11?

Yes, but the 14.2-RELEASE Errata point out that DRM kernel modules do not work if you grab the pre-built ones. This was true on March 12th 2025, so:

• Run pkg to install the package manager (initially it is a stub)
• Run pkg install git to install git (this pulls in a surprising amount of other stuff)
• Get the system sources (with git)
• Rebuild the world and install it
• Get the ports tree (with git)
• Build graphics/drm-61-kmod from ports (just make ; make install, and the port itself is a real quick build)
• Build graphics/gpu-firmware-amd-kmod from ports, remember FLAVOR=polaris12 for the GPU in this laptop (otherwise the default flavor is built)

After that, enable the amdgpu module in rc.conf, or load it by hand. Any old X11 stuff will do, but I suggest installing x11/kde and x11/sddm.

Will it run KDE Plasma 6 Wayland?

Hahaha. No. But yes.

KDE Plasma 6 on Wayland in general works. But on this specific machine, with this specific grapics card, Plasma starts, all the processes of a KDE Plasma desktop are running, and the screen displays a single white text-cursor in the upper-left corner.

It’s not this-specific-machine, either, since I have a desktop with Intel CPU and an AMD RX550 video card that behaves the same.

Last time I dug into KWin internals in an attempt to figure this out I ended up with “some part of the OpenGL stack is lying” and then gave up. Now with a fresh laptop that just cries out for a modern desktop, I’m going to try again.

Thank you everyone for keeping the lights on for a bit longer. KDE snaps have been restored. I also released 24.12.3! In addition, I have moved “most” snaps to core24. The remaining snaps need newer qt6/kf6, which is a WIP. “The Bad luck girl” has been hit once again with another loss, so with that, I will be reducing my hours on snaps while I consider my options for my future. I am still around, just a bit less.

Thanks again everyone, if you can get me through one more ( lingering broken arm ) surgery I would be forever grateful! https://gofund.me/d5d59582

Welcome to a new issue of "This Week in Plasma"! Every week we cover the highlights of what's happening in the world of KDE Plasma and its associated apps like Discover, System Monitor, and more.

This week, Plasma 6.4 began to take shape. A bunch of impactful features and UI improvements landed, not to mention some juicy technical changes in the form of a newly-implemented Wayland protocol and HDR energy efficiency improvements. Just a whole lot of good stuff! Check it out below:

Notable new Features

Plasma 6.4.0

After being punted from Plasma 6.3, per-virtual-desktop custom tile layouts are now implemented for 6.4! (Vlad Zahorodnii, link)

Clicking the "Details" button on a system notification showing file transfer progress will now reveal a graph showing the transfer speed over time! (Méven Car, link)

You can now fully disable System Tray icons provided by apps that lack an internal setting for this (looking at you, Discord). Note that this could potentially break apps as they won't know their tray icon isn't being shown, so only use this feature if you know what you're doing! A warning message explains this, too. (Nate Graham, link)

Notable UI Improvements

Plasma 6.3.4

Plasma's sidebar-style UI elements (e.g. the Activity Switcher sidebar) now overlap panels when shown outside of Edit Mode. This looks nicer and helps communicate focus better. (Niccolò Venerandi, link)

Plasma 6.4.0

Improved KRunner search result ordering by adding the power and session actions into the default set of favorite actions, ensuring they appear first when searched for. (Nate Graham, link)

Refined the heuristic for when a panel widget's popup will be displayed centered on the panel or the screen, so that it happens more often in cases where you obviously configured your panel with this in mind. (Niccolò Venerandi, link)

In the panel configuration dialog, the little wireframe visualizations for options now all visually reflect their panel's actual position on screen. (Niccolò Venerandi, link)

You can now configure which modifier keys plus a scroll trigger KWin's zoom effect. (Vlad Zahorodnii, link)

Improved keyboard navigation in KRunner's popup: now if the pointer happens to be hovering over an item, you can still use the arrow keys to move the selection highlight to a different item. (Christoph Wolk, link)

If you're being slowly driven mad by the system notification telling you how to regain control when an app like Input Leap is using the input devices, you can now disable it like you can any other notification. (David Redondo, link)

Plasma widgets in the System Tray that hide completely when they deem themselves not relevant no longer do this when placed in standalone form on the panel; we reasoned that in this case, if you put them there yourself, you probably always want to see them! (Niccolò Venerandi, link)

Widgets using the ExpandableListItem component — commonly seen in the System Tray — now display tooltips on hover for any list items with labels so long they've become elided. This was very uncommon, which is how we missed it until now! (Kai Uwe Broulik, link)

When you've configured the Kickoff Application Launcher to only show app names or only show app descriptions, you'll no longer see tooltips with the labels you said you didn't want. (Nate Graham, link)

Frameworks 6.13

Implemented touch scrolling in open/save dialogs. (Marco Martin, link)

Improved KRunner search result ordering in another way as well, by returning to the older style of strictly respecting the ordering that the user user configured. (Nate Graham, link)

Notable Bug Fixes

Plasma 6.3.3

Fixed a bug that could cause Discover to get stuck refreshing forever following flaky network connectivity. (Aleix Pol Gonzalez, link)

Fixed some layout glitches affecting the folder chooser dialog at certain window sizes. (Luke Horwell, link)

Plasma 6.3.4

Fixed a bug that would cause fit-content panels with Task Manager widgets on them to not immediately shrink as expected when apps or windows were closed. (Niccolò Venerandi, link)

Fixed a glitch in the Bluetooth wizard's scrollable device view that made scrolling using a touchscreen unreliable. (Marco Martin, link)

Plasma 6.4.0

A notorious Plasma 6 panel bug has been fixed: now when there are multiple panels sharing the same screen edge, they're all displayed properly, and reserve only as much space away from the screen edge as the thickest panel. (Niccolò Venerandi. link 1 and link 2)

Fixed a visual glitch involving the ruler for resizing custom-length panels in auto-hide mode. (Niccolò Venerandi, link)

Other bug information of note:

• 2 very high priority Plasma bugs (same as last week). Current list of bugs
• 23 15-minute Plasma bugs (down from 24 last week). Current list of bugs
• 98 KDE bugs of all kinds fixed over the past week. Full list of bugs

Notable in Performance & Technical

Plasma 6.3.4

Improved the pixel-perfection of various KWin effects, including Wobbly Windows. (Xaver Hugl, link)

Plasma 6.4.0

KWin's codebase has been formally split between an X11 version and a Wayland version, allowing the Wayland version to develop faster and the X11 version to avoid accumulating bugs due to changes in a shared base combined with a lack of testing (82% of users with telemetry turned on use Wayland now). This will continue until Plasma 7, at which point it's highly likely the dedicated X11 session will be removed. Note that the Wayland-only version will continue to run XWayland-using apps just as it can right now. More information about this can be found in Vlad's blog post on the topic!

Added support for P010 color-formatted videos, improving power efficiency for playing full-screen HDR video content. (Xaver Hugl, link)

Implemented support for the get_input_idle_notification Wayland Protocol. (Xaver Hugl, link)

Improved the reliability with which the Power & Battery widget is able to detect and display battery information for Bluetooth devices. (Kai Uwe Broulik, link)

It's no longer possible to create new Plasma Vaults using the EncFS encryption system, as it's been discontinued and has known security vulnerabilities. You can still use existing EncFS vaults; you just can't create new ones. (Nate Graham, link)

Continued to resolve binding loops and QML warnings throughout Plasma, succeeding at handling a large fraction of the ones seen in basic usage that are within KDE's power to resolve without Qt changes. (Christoph Wolk, link 1, link 2, link 3, link 4, link 5, link 6, link 7, link 8, link 9, link 10, link 11, link 12, link 13, link 14, link 15, and link 16)

How You Can Help

KDE has become important in the world, and your time and contributions have helped us get there. As we grow, we need your support to keep KDE sustainable.

You can help KDE by becoming an active community member and getting involved somehow. Each contributor makes a huge difference in KDE — you are not a number or a cog in a machine!

You don’t have to be a programmer, either. Many other opportunities exist:

• Triage and confirm bug reports, maybe even identify their root cause
• Contribute designs for wallpapers, icons, and app interfaces
• Design and maintain websites
• Translate user interface text items into your own language
• Promote KDE in your local community
• …And a ton more things!

You can also help us by making a donation! Any monetary contribution — however small — will help us cover operational costs, salaries, travel expenses for contributors, and in general just keep KDE bringing Free Software to the world.

To get a new Plasma feature or a bugfix mentioned here, feel free to push a commit to the relevant merge request on invent.kde.org.

Dear digiKam fans and users,

After four months of active maintenance and many weeks triaging bugs, the digiKam team is proud to present version 8.6.0 of its open source digital photo manager.

• New Features and Major Changes
  • Face Management Updates and Improvements
  • Auto-Tags Management
  • Image Quality Management
  • Red Eye Correction
• Internal Components Update
• Notable Bug Fixes
• Generalities
• Future Plans
• Final Words

The digiKam team has continued to work on a better Artificial Intelligence integration in digiKam, and many parts have been improved with the 8.6.0 release.

This is a recipe post. I loathe and despise recipe sites, but this is one I regularly need to look up. In Canadian measures, which is what I still do my baking in even after 30 years in Europe.

Mix well and let stand while collecting the rest:

• 1½ cups rye bran
• 1 c. milk
• ½ tsp. salt

Ready:

• ⅓ c. oil
• 1 egg

Ready:

• ⅔ c. brown sugar
• 1 c. flour
• 2 tsp. baking powder
• ½ c. raisins

Mix the wets. Stir in the drys. Fill muffin pan. Bake 25 minutes at 180℃. Makes 11 muffins because my muffin pan has one broken cup. Uses rye bran because the local windmill has that “left over” as animal feed after milling rye – apparently very few people in the Netherlands use bran anyway, and rye bran even less so. Instead of flour and baking powder use zelfrijzend bakmeel.

Edit: reduce sugar to ½ cup and add 1 tbsp. of molasses to improve flavor and color and reduce sweetness.

Let’s go for my web review for the week 2025-11.

Kill your Feeds - Stop letting algorithms dictate how you think

Tags: tech, social-media, criticism

Simple steps to escape the algorithmic social media circus.

https://usher.dev/posts/2025-03-08-kill-your-feeds/

Music labels will regret coming for the Internet Archive, sound historian says

Tags: tech, culture, archive, history, law, copyright

Once again the music labels can’t understand the cultural value of building archives. Let’s hope they loose the lawsuit.

https://arstechnica.com/tech-policy/2025/03/music-labels-will-regret-coming-for-the-internet-archive-sound-historian-says/

What does “PhD-level” AI mean? OpenAI’s rumored $20,000 agent plan explained

Tags: tech, ai, machine-learning, gpt, marketing, business

Here we go for a brand new marketing stunt from OpenAI. You can also tell the pressure is rising since all of this is still operating at a massive loss.

https://arstechnica.com/ai/2025/03/what-does-phd-level-ai-mean-openais-rumored-20000-agent-plan-explained/

Microsoft is reportedly plotting a future without OpenAI

Tags: tech, ai, machine-learning, gpt, microsoft, business

Are we surprised? Not really… This kind of struggle was an obvious outcome from the heavy dependencies between both companies.

https://techstartups.com/2025/03/07/microsoft-is-plotting-a-future-without-openai/

Cognitive Behaviors that Enable Self-Improving Reasoners, or, Four Habits of Highly Effective STaRs

Tags: tech, ai, machine-learning, gpt, cognition, research

I like this kind of research as it also says something about our own cognition. The results comparing two models and improving them are fascinating.

https://arxiv.org/abs/2503.01307

Google’s Gemma 3 is an open source, single-GPU AI with a 128K context window

Tags: tech, ai, machine-learning, gpt

More smaller footprint models are becoming available. This is becoming interesting (even though the open source claim is overrated here).

https://arstechnica.com/gadgets/2025/03/googles-new-gemma-3-ai-model-is-optimized-to-run-on-a-single-gpu/

Why extracting data from PDFs is still a nightmare for data experts

Tags: tech, ai, machine-learning, gpt, ocr, computer-vision

So much data trapped in PDFs indeed… Unfortunately VLM are still not reliable enough to be unleashed without tight validation of the output.

https://arstechnica.com/ai/2025/03/why-extracting-data-from-pdfs-is-still-a-nightmare-for-data-experts/

A Quick Journey Into the Linux Kernel

Tags: tech, linux, kernel, programming

Nice exploration of the important areas in the kernel.

https://www.lucavall.in/blog/a-quick-journey-into-the-linux-kernel

Standards for ANSI escape codes

Tags: tech, unix, command-line, terminal, standard

A nice glimpse into the maze of the escape codes on the terminal.

https://jvns.ca/blog/2025/03/07/escape-code-standards/

Tech Notes: Rust trait object layout

Tags: tech, rust, type-systems, memory

OK more questions than answers I guess… That said, it shows interesting differences in design choices with C++ to support the traits system.

https://neugierig.org/software/blog/2025/03/trait-object-layout.html

When are Rust’s const fns executed?

Tags: tech, rust, compiler, optimization

A quick primer about compile time evaluations in Rust.

https://felixwrt.dev/posts/const-fn/

Why do I find Rust inadequate for codecs?

Tags: tech, rust, c, codec, performance, safety

Interesting reasons to let go of Rust, some spaces indeed can have a safety vs performance tradeoff which would justify using good old C.

https://palaiologos.rocks/posts/rust-codecs/

The Art of Formatting Code

Tags: tech, programming, tools, unicode

Ever wondered how to make a code formatter? This post does a good job showing the main problems you might encounter. The impact of Unicode is especially funny. Very interesting stuff.

https://mcyoung.xyz/2025/03/11/formatters/

Idiomatic Go

Tags: tech, programming, go

An evolving list of how to write idiomatic Go.

https://dmitri.shuralyov.com/idiomatic-go

Building Websites With Lots of Little HTML pages

Tags: tech, web, frontend, css

You can really do a lot with CSS transitions nowadays.

https://blog.jim-nielsen.com/2025/lots-of-little-html-pages/

Post-Processing Shaders as a Creative Medium - Maxime Heckel’s Blog

Tags: tech, gpu, shader, web

Nice set of tricks for post-processing effects all centering around pixelated patterns. Really neat.

https://blog.maximeheckel.com/posts/post-processing-as-a-creative-medium/

What Makes Code Hard To Read: Visual Patterns of Complexity

Tags: tech, programming, maintenance

The whole field is unfortunately a bit fuzzy. That said, this article gives interesting ideas about what to pay attention to when writing code to ease the readability.

https://seeinglogic.com/posts/visual-readability-patterns/

How Long Should Functions Be? - by Kent Beck

Tags: tech, engineering, craftsmanship, quality, tdd, programming

Nice post. Explains well why the answer is not a number to target. You want to impact the distribution.

https://tidyfirst.substack.com/p/how-long-should-functions-be

How to Write Useful Commit Messages

Tags: tech, version-control, codereview

Might be going a bit far if you use everything listed here. That said, it gives lots of good ideas so you might want to decide on what you should adopt on your project.

https://refactoringenglish.com/chapters/commit-messages/

CSS Relative Colors

Tags: tech, web, css, colors

Nice new tricks to specify colours in CSS.

https://ishadeed.com/article/css-relative-colors/

Stewardship over ownership

Tags: tech, engineering, maintenance, leadership

I like this. Sometimes a simple word can make all the difference in the way we behave. Code stewardship is indeed a better word.

https://ntietz.com/blog/stewardship-over-ownership/

The Staff+ Performance Cliff

Tags: tech, leadership, management, learning

It’s indeed not easy to go from individual contributions, to team level leadership, to organisation level leadership. Many things need to be learned or relearned at each step.

https://sylormiller.com/posts/2025/staff-plus-cliff/

There is No Automatic Reset for Engineering

Tags: tech, quality, project-management, product-management

Interesting thinking around a portfolio of activities. You can prioritise differently within it to manage quality vs speed of delivery over time.

https://agileotter.blogspot.com/2025/03/there-is-no-automatic-reset-in.html?m=1

Reducing Power Gradients - Psych Safety

Tags: organization, management, power, psychology

Interesting tips to reduce the power dynamics in organisations.

https://psychsafety.com/reducing-power-gradients/

Bye for now!

One of the biggest behind-the-scenes changes in the upcoming Plasma 6.4 release is the split of kwin_x11 and kwin_wayland codebases. With this blog post, I would like to delve in what led us to making such a decision and what it means for the future of kwin_x11.

Background

KWin started as an X11 window manager almost two and a half decades ago. Over the course of the years, it transformed drastically. It gained support for compositing on X, and it became a Wayland compositor.

Sharing the same codebase was critical in the early days of kwin_wayland. We already had working window management abstractions, which had been tested for many years, so we could reuse them on Wayland instead of writing new from scratch. Also, if kwin_x11 gained a new feature, then kwin_wayland would likely gain it for free too.

As time went by, kwin_wayland outgrew kwin_x11. They still shared code but they became quite distinct projects with different mental models how things operate, e.g. how pixels get on the screen or how input works. It also didn’t help that many Plasma developers jumped the X11 ship and turned to the Wayland side as part of the “eating your own dog food” practice, which eventually led to the feature freeze in KWin/X11 back in 2018 due to the lack of sufficient testing and various breakages.

Some time around 2020, we started taking a more bold and aggressive approach to Wayland session development because we saw that Plasma Wayland was trailing behind other desktop environments and something had to be changed in order to catch up. Such a policy produced great results, and Plasma is now one of the leading Wayland desktop environments. Unfortunately, it also greatly contributed to the number of regressions in the X11 session.

Another issue was that there were some features that we couldn’t make work as expected on Wayland so we had to drop them for everyone, which understandably made X11 users unhappy.

Goals

A few years ago, we started contemplating the idea of splitting the X11 and Wayland codebases because of the growing list of regressions affecting the X11 session, and architecture restrictions imposed on KWin/Wayland by the way KWin/X11 works.

That would allows us to keep KWin/X11 working as is without it breaking too often and freely change KWin/Wayland in ways that we think are best suited to make the Plasma Wayland session even better. Of course, it is not a silver bullet solution: we replace one problem with another problem (mainly related to maintenance and ensuring interface compatibility between two projects).

Details

After various discussions online and at Akademy and also seeing (impressive) Plasma Wayland usage statistics, we decided that it’s the right time to do such a split. The main kwin repository is going to host KWin/Wayland, while the kwin-x11 repository is going to host KWin/X11.

KWin/X11 and KWin/Wayland are co-installable so users can freely switch between the X11 and Wayland sessions back and forth and also make sure that updating to 6.4 is not a big hassle for distributions. You’ll be able to have only KWin/X11 or only KWin/Wayland on your computer, or both.

The codebase split doesn’t affect Xwayland support in KWin/Wayland. In other words, X11 applications will continue running on Plasma Wayland.

Extensions

Like any other Plasma component, KWin’s functionality can be extended using plugins. There’s good and bad news. The good news is that extensions written in JavaScript and QML (for example, fancy effects that are available at the KDE Store) will continue working both with kwin_x11 and kwin_wayland as expected, so extension developers don’t need to do anything about it. The bad news is that C++ extensions should be specifically targeted for kwin_x11 and kwin_wayland because neither provides API and ABI compatibility guarantees for its C++ API.

As Wayland progress moves forward, it is likely that the scripting API of KWin/Wayland will be further extended.

Future of KWin/X11

KWin/X11 will be still maintained for the foreseeable future. But that maintenance work will boil down to fixing build errors, adapting to new KDE Frameworks and Plasma APIs, and backporting window-related fixes from KWin/Wayland. There are no plans to drop KWin/X11 in the Plasma 6 lifecycle, although it’s highly possible that it will happen in Plasma 7.

KWin/X11 won’t receive new features anymore; until recently, it received new features that had been developed against KWin/Wayland passively (because both lived in the same repository). However, it might be actually a good thing because the X11 session doesn’t receive that much testing nowadays.

Qt Contributor’s Summit 2025 is taking place in Munich in May. Unfortunately, I won’t be able to make it this year, so let’s talk about some of my recent contributions to our favorite cross-platform UI toolkit.

A couple of years ago I added a context menu to the line number bar in KWrite and Kate. Rather than having to go to the menu to toggle automatic line breaks or bookmarks, those options were now accessible from the context menu. I’ve always wanted to do the same to the scrollbar and have a mini-map option there. However, it wasn’t possible to extend the default context menu and I didn’t want to re-implement all of its Up/Down/Pg Up/Pg Down/etc logic. I therefore added a QScrollBar::createStandardContextMenu method similar to what QLineEdit has. As the name implies, it builds the regular context menu and hands it to you, so you can add your own actions to it.

Speaking of menus, I uplifted SH_Menu_SelectionWrap that decides whether menu selection wraps around when using the arrow keys from QStyle to QStyleHints. This allows Qt Quick Controls Menu to behave the same as QMenu. This behavior is off by default on macOS, for instance, and our own Qt Quick Controls 2 Desktop Style currently handles keyboard input itself to implement wrapping.

I’ve talked about it before that I sometimes just fire up a profiler and look at application startup performance. That’s while working on KWrite I noticed that it was spending a good amount of time parsing the shared-mime-info database. It’s basically a huge XML file containing information about all file types there are and how to detect them, either by file extension or by magic bytes within. Now, why would it even load that database when opening a blank new document?

There’s several places where KWrite and the underlying KTextEditor Framework need to know the type of a viewed file: to load the correct syntax highlighter, display the correct file icon in the tab bar and recent documents list, and so on. When the file type isn’t already known it tries to guess from its contents. Qt caches the database of course but the first one to create a QMimeDatabase triggers a load. However, none of this really matters for an empty document so it now assumes an empty document be of type text/plain. The correct type is actually application/x-zerosize but that’s really not what you want here.

My trusty ThinkPad has a “keyboard key” (Fn+F11) which under Windows supposedly opens some keyboard settings page. For the longest time, I thought it just wasn’t supported under Linux because I couldn’t create a global shortcut with it. Recently I noticed in KWin’s Debug Console that it did in fact recognize the key. Turns out Qt just didn’t have a corresponding Qt::Key. Starting from Qt 6.10 there will be a new Qt::Key_Keyboard value.

Qt Wayland Client

Since last time I talked about Qt Wayland, I have done a few more optimizations to its SHM (Shared Memory) backing store. It’s the canvas that Qt provides to software-rendered applications for drawing their user interface.

I finally merged support for scrolling the backing store. This lets an application, such as a text editor or terminal emulator, optimize scrolling through a view by merely moving the pixels and filling in only the small gap that’s now left. Unfortunately, this optimization cannot be provided when using fractional scaling because there’s no integer number of pixels we can just move up or down.

Furthermore, Qt Wayland no longer uses a backing store with alpha channel if the window didn’t request one. Most toplevel application windows are opaque after all. It doesn’t really matter from the compositor’s POV since a wl_surface can carry an “opaque region” that indicates what part of the window is actually opaque and Qt sets it accordingly. ARGB32 and RGB32 are the same in-memory size, too. Nevertheless, not using an alpha channel lets Qt skip certain operations, such as clearing the paint area before starting to draw.

While investigating a performance problem in animations under Wayland, I noticed that StackView’s new pushItem methods didn’t use any animation by default. As part of the effort to give QML tooling, such as qmlcachegen and qmllint, more context, many generic methods were supplemented by proper overloads. For example, StackView.push(var) takes either an Item, a Component, or a URL but you could call it with anything and that’s why it has to be invoked dynamically. StackView.pushItem on the other hand is declared three times with an explicit type which allows to generate more efficient code.

It didn’t make much sense that they would behave differently, of course. Indeed online documentation clearly said “If no operation is provided, PushTransition will be used.” We concluded it was a copy-paste mistake and I changed this for Qt 6.8.3, even though it’s technically a behavior change.

I’ll be jealous of your Weißwurstfrühstück in Munich and am looking forward to attending Qt Contributor’s Summit again in 2026!

Last week I decided to clean up a bit of digital cruft. That is, I moved a few of my websites onto a single VPS, saving quite a bit of monthly server hosting costs.

What I did was that I moved VPSes from Linode (Akamai) to DigitalOcean, but also migrated a full web hotel from One to DigitalOcean (converting email accounts to email forwards).

As this is something that I do very rarely, I decided to document the process here so that I don’t have to look everything up again next time around.

The grunt work was about migrating a number of L*MP services to a LEMP server. There are a couple of tasks involved here, mainly migration of databases and getting WordPress running in a subdirectory using Nginx. The rest of the exercise had to do with the moving of nameservers and waiting for DNS propagation to get certbot to provide certificates for the new location.

Migration of MySQL databases

The migration of a database between machines can be broken down into three stages:

1. Dumping the old database
2. Creating a new database and user
3. Sourcing the database contents into the new database

I choose to do it in these three stages, as I’d like to keep the old database dump as an additional backup. The other option would be to transfer the database contents in a single step, merging steps 1 and 3 into one

Nevertheless, I use mysqldump to dump the database contents, and then bzip2 to reduce the size of the dump. This is efficient since and SQL dump is quite verbose.

mysqldump -u username -p --databases databasename | grep -vE \"^(USE|CREATE DATABASE)\" | bzip2 -c - > dumpname.sql.bz2

This is derived from the answer by Anuboiz over at stack overflow. The resulting file is then transferred to the new server using scp together with the actual website.

The next step is to create a new database and a new database user. Here, I assume MariaDB (using the mysql commands), as my main target is WordPress. For other database engines, e.g. Postgresql, please check the docs for exact grammar, but the SQL commands should be very similar.

sudo mysql
mysql> CREATE DATABASE databasename;
mysql> USE databasename;
mysql> CREATE USER 'username'@'localhost' identified by 'password';
mysql> GRANT CREATE, ALTER, DROP, INSERT, UPDATE, DELETE, SELECT, REFERENCES, RELOAD on databasename.* TO 'username'@'localhost' WITH GRANT OPTION;
mysql> EXIT

Check out this digital ocean tutorial for details on the above commands.

The next step is to read the database contents into the new database. For this, we need to unzip the sql dump, e.g. bunzip2 dumpname.sql.bz2, which will result in a file called dumpname.sql. Please notice that bunzip2 unzips the file and removes the original, zipped, file. If you want to keep the original, use the -k option.

Once you have the dumpname.sql file available, you can read it into the database with the newly created user using the source command as shown below.

mysql -u username -p
enter the password here
mysql> USE databasename;
mysql> SOURCE dumpname.sql;
mysql> EXIT

Now you should have a new database with the old database contents on the new server, with an associated database user. For WordPress sites, make sure that you reflect any changes in the associated wp-config.php file.

WordPress in a subdirectory using Nginx

The other piece of the puzzle that was new to me was to run WordPress from a subdirectory, e.g. example.com/blog/, rather than from the root level, e.g. example.com/.

Removing most of the nginx server configuration, the following parts does the magic:

server {
        root /var/www/thelins.se;
        index index.php index.html;

        server_name thelins.se www.thelins.se;

        ...

        # For root
        location / {
                try_files $uri $uri/ /index.php?$args;
        }

        # For subdirectory
        location /johan/blog/ {
                try_files $uri $uri/ /johan/blog/index.php$args;
        }

        location ~ \.php$ {
                fastcgi_split_path_info ^(.+\.php)(/.+)$;
                fastcgi_pass unix:/run/php/php7.4-fpm.sock;
                fastcgi_index index.php;
                include fastcgi.conf;
        }

        ...
}

The trick was to ensure that the subdirectory try_files statement refer to the correct index.php. Notice that this has to be done for each WordPress instance, if you happen to have multiple WordPress installations in various subdirectories on the same domain.

Conclusions

Its a bit of hassle to migrate a lot of web sites at once, but the monetary saving from moving the low traffic sites onto a single VPS, and the simplification of the management and monitoring by moving all VPSes to a single provider makes it worth it.