OS News Archive

With the release of version 14.02, the Genode project has added two major features to the OS-construction framework: Using VirtualBox on top of Genode/NOVA, a wide range of unmodified guest OSes can now be integrated as components into Genode-based systems. The second feature is the addition of the file systems of the NetBSD kernel as rump kernels. The release documentation covers plenty of further improvements.

With storage and virtualization, the new release addresses two topics that are fundamental for using Genode as general-purpose OS, and both topics have been approached in a pretty holistic manner.

When it comes to storage, the project has significantly advanced over the past year but a few key pieces were still missing, namely mature file systems and a block cache. After having investigated FUSE-based file systems in the previous release, the project started exploring so-called rump kernels, which enable the execution of subsystems of the NetBSD kernel at user level. Originally, the rump kernel project was created to ease the development of drivers on NetBSD. The basic idea behind them is to link a driver to a stripped-down version of the NetBSD kernel that does not contain any privileged instruction. Hence, it can be executed in user mode. To interact with the outside world, a rump kernel uses a small so-called "hypercall" interface. By implementing this interface on top of the Genode API, rump kernels have become usable on Genode now. The immediate benefit is the availability of the time-tested file systems of the NetBSD kernel. But in the future, other NetBSD subsystem such as the TCP/IP stack or device drivers could be considered just as well.

At block level, the project took the chance to redesign the internal interfaces of the existing block-level components to support fully asynchronous operation. This step enables the effective use of modern disk-controller features such as native command queuing, and even the out-of-order processing of block requests. As the cherry on top of this line of work, there is a new block-cache component.

Over the past 5 years, virtualization has always played a role for the project. It started with running the paravirtualized OKLinux on top of the OKL4 microkernel. Later L4Linux was made available to Genode running on the Fiasco.OC kernel. Once the NOVA hypervisor found its way to the framework, the project embraced the use of the Vancouver virtual machine monitor, which enabled the use of unmodified Linux kernels. However, none of these solutions appealed well for a large user base, mainly because they were difficult to use or lacked features. By adding support for VirtualBox on top of NOVA, the project has finally found an answer to the question for product-grade virtualization on top of the framework. The integration of VirtualBox with Genode was no ordinary porting work but quite an engineering feat, which turns the architecture of VirtualBox pretty much upside down. In contrast to the host operating systems already supported by VirtualBox, Genode's version does not extend the host kernel in any way. VirtualBox leverages hardware-based virtualization (VT-x) but lives as a plain user-level program with no special privileges.

As with each new version, there are numerous smaller improvements and new features. For example, a new pseudo file system called trace-fs makes it possible to interactively use Genode's event tracing mechanism via Unix tools such as cat, grep, and echo.

All the changes are covered in detail in the release documentation of version 14.02.

NuttX saw its 100th release at the end of last month.

BareMetal OS now supports TCP/IP by way of a port of LwIP, originally by Adam Dunkels for embedded devices.

Jonathan Mahler, on 'long-form' articles:

Up until the moment I read this article, I had no idea there was a specific term for long(er) articles, let alone that some consider it a genre. I realised that virtually all of my reviews are apparently "long-form"; the Jolla review, for instance, was 9000 words long. I've done much crazier than that, though - the Palm article was 22000 words long.

However, in both of these cases, I never intended for the articles to become that long, or in fact, to achieve any specific length. When I start out, I just have a number of things that I want to discuss, and I won't stop writing until all of those things are in the article. I will make a distinction between things that get lots of attention (say, the gestures in Sailfish) and things that get a passing mention (e.g., the backplate), usually based on some sort of combination between what I personally find interesting and what you, the readers, might find interesting. Since the gestures in Sailfish are at the core of the user experience, it gets a lot of attention; because the backplate and its hardware potential offers little to no benefit right now, it gets a passing mention.

I also like to pick some sort of overarching red thread, like the whole The Last Resort thing in the Jolla/Sailfish review, to tie everything together and frame the article. This can be a dangerous thing, since it's usually very personal and can easily be misinterpreted as pretentious or have other unwelcome side-effects. Originally, I framed the Jolla/Sailfish article using Manifest Destiny, but I quickly realised that its pitch-black consequences were unacceptable in a mere technology article.

Combine these things, and the article is done. Whether the resulting article turns out to be 2000 words or 10000 words is irrelevant to me; if it contains everything I want to convey, it's done. If it leaves things out just to be short and more digestible, it's a bad article. If it contains useless, irrelevant crap just to pad the word count, it's a bad article. Years ago, when both my best friend and I were writing our master's theses, we ended up with very, very different word counts - mine was 27000, hers was a mere 8000. Both contained all the required information; nothing more and nothing less. Both were graded positively. Word count is a measure of nothing.

By now, some of you might be wondering why the sales pitch for the Palm article did contain the word count - which seems to contradict the above. My reasoning there was simple: we were selling the Palm article. I figured that since I was asking people to pay money for an article that was freely available on that very same page, I should at least give them information about what they were spending their money on.

Long articles like the ones mentioned above are not for everyone. In fact, their potential audience is much, much smaller than, say, a three paragraph jab at software patents. While those jabs are fun - sort of - it's these long articles that are by far the most fulfilling to write. The Palm article alone took months and months of work - research, making notes, educating myself about low-level stuff, devising a structure, setting a tone, organising the six hundred different subjects I wanted to cover, the actual writing process, revising it all, while also doing my regular job, and so on - but it is by far the most rewarding experience I've ever had for OSNews.

I'll never forget getting emails from former Palm executives and engineers - big names - congratulating me on a job well done.

Writing articles like that is not easy, with my biggest enemy being a lack of time because OSNews is a hobby, not a full-time job (I wish it was!). A few weeks after publishing the Palm article, I started work on a similar article about Psion and Symbian, but due to work and personal life (which was rather tumultuous in 2013) sucking up a lot of time last year, I never found the time to continue work on it. With things having settled down since December, I'm making plans to dust off the Psion and Symbian material, possibly take a few weeks off work, and finish it.

That article could end up being 8000 words, or 50000 words. I don't know. The goal is not be long, but to be comprehensive, and this is my inherent problem with the term "long-form". This term puts the focus on length instead of content, which absolutely baffles me. A good article is not defined by its length - or lack thereof - but by its content.

We report regularly about Visopsys - one of the few hobby operating systems that survived where virtually all the others (SkyOS, Syllable, etc.) died out. They've got a new website, which seems like a nice occasion to give it some attention again.

ACM Queue posted an article titled "Unikernels: Rise of the Virtual Library Operating System", written by the main developers of MirageOS. The article is a good overview of the motivation and design of Mirage, which is a specialized lightweight hypervisor guest operating system for Xen. It's developed in OCaml and benefits from its features like memory safety, module system and compiler optimisations. It's comparable to OSv on many fronts.

More here.

The Genode project has released version 13.11 of their OS framework. This time, the focus lies on exploring new ways for bringing existing protocol stacks to the Genode world. FUSE-based file systems and the Linux TCP/IP stack have become available as user-level libraries, and the improved Qt5 port covers QML. Hardware-wise, the new version extends the support for Exynos-5, Raspberry Pi, and ARM TrustZone.

With the release cycle of version 13.11, the Genode developers took the chance to explore plenty of experimental features across the whole software stack.

I doubt this has any usefulness whatsoever, but that doesn't make it any less awesome. Great work (via Steven Troughton-Smith).

The Contiki operating system, known for its super-slim IPv6 stack, has been updated to version 2.7 with support for new Systems-on-a-Chip: single-chip devices with a microprocessor and an integrated 2.4 GHz radio. This new version has a much strengthened IPv6 wireless mesh networking setup that allows a bunch of chips to autonomously form wireless networks that can be connected directly to the Internet.

So, Tizen is now shipping... On a camera. On a related note, Samsung has let out some details of Tizen 3.0, planned for next year. It's going 64bit, will gain multiuser support, and will switch from X to Wayland.