GNU Hurd Is "Almost There" with x86_64, SMP and ~75% of Debian Packages Building | Hacker News Reader

GNU Hurd Is "Almost There" with x86_64, SMP and ~75% of Debian Packages Building

phoronix.com

57 pointssergiogdr23 comments

This headline was pretty much true 5 years ago, 10 years ago, 15 years ago...

Don't get me wrong, I think Hurd is interesting, but I seriously doubt it's going to have a big impact on anything as it reflects the software engineering philosophies of the 1980s.

The "75% of Debian archive builds" claim is exactly the same 7 years ago. In fact, look at this slide from the 2019 presentation: https://archive.fosdem.org/2019/schedule/event/roadmap_for_t... (page 8)

It is barely distinguishable from the first slide featured in the Phoronix article from today: https://www.phoronix.net/image.php?id=2026&image=gnu_hurd_1 It seems like there has been progress on other fronts, so I'm not sure why Phoronix ran a headline focused on very old news.

Interestingly, the 2018 version of the slide claims "80% of Debian archive builds"; I wonder what caused the regression. https://archive.fosdem.org/2018/schedule/event/microkernel_h... (page 26)

> so I'm not sure why Phoronix ran a headline focused on very old news.

It's just coverage of FOSDEM 2026 and I guess they assumed that the FOSDEM slides would show notable changes rather than the state of play.

Another example of if llms are so good. Why isn't a gap like this closed very quickly?

GNU projects and LLM contribs mix like water and oil.

> reflects the software engineering philosophies of the 1980s.

It has a microkernel architecture. That's already an improvement over the "modern" monolithic kernels we are stuck with today. Given Big Tech's interest in hardening security and sandboxing you'd think this would get more attention.

True but it's not exactly new. I remember Andrew Tanenbaum and Linus Torvald's heated discussions in the early 90s :) Minix featured a microkernel before linux existed.

Yeah, but we are still far off making it mainstream beyond some key use cases, QNX, INTEGRITY, language runtimes on top of type 1 hypervisors, all kernel extension points being pushed into userspace across Apple,Google,Microsoft offerings, Nintendo Switch,....

Given the tectonic shift in priorities for Linux kernel development over the past decade, I'm willing to bet that many key developers would be open to a microkernel architecture now than ~25+ years ago. CPUs now have hardware features that reduce the overhead of MMU context changes which gets rid of a significant part of the cost of having isolated address spaces to contain code. The Meltdown and Spectre attacks really forced the security issue to the point where major performance costs to improve security became acceptable in a way that was not the case in the '90s or '00s.

Link to the actual presentation:

https://fosdem.org/2026/schedule/event/7FZXHF-updates_on_gnu...

While I laughed at the headline, it also fondly reminds me of reading Slashdot in the late 90s-early 00s, back before the Internet and programming and computers had all gone to shit.

Good luck guys. At least working on this for decades is less damaging to the world than anything people do at Google and Facebook.

I always liked reading about it in uni in the mid-late 00s. It made me feel smart in my OS tutorials when I could rattle off all the design choices and how they differed from Linux and Windows

Hurd is long past ever being anything but a pet project of RMS and his familiars.

Please let me know what it is "there". Then I'll start caring. Until then ...

Hah! It's been almost there for over a decade...

I remember talking about waiting for it to be ready in 1991.

Make that plural: decades

X11 ?

"Hurd Isn't Soup"

Hurd Uses Repurposed Debian

To be fair, GNU/Linux distros like Debian lean very heavily on the GNU part. They owe a lot to the GSF and its work is highly praised.

Just their kernel somehow seems to be stuck in vaporware status. Probably because a lot of developers would think "why work on this when we already have Linux" which is a fair point too.

I wonder if microkernels more relevant now than ever given their reduced attack surface, and also the recent availability of more cores.

One big criticism from decades ago was the loss in efficiency. But what's changed since microkernels were conceived is how many processor cores are available to offload userspace drivers from the kernel.

Is this a valid viewpoint? Is it time for microkernels to overtake monolithic kernels?

They already did. There are more microkernels around than monolithic. All big CPU's use them internally, all phones use them.

Reply on news.ycombinator.com