FreeBSD 14 Released 38
Mononymous writes: FreeBSD 14 has been officially released. You can get it from FreeBSD.org, or via freebsd-update and source update methods for existing systems. Some highlights:
- OpenSSH version 9.5p1
- OpenSSL version 3.0.12, a major upgrade from OpenSSL 1.1.1t in FreeBSD 13.2
- OpenZFS release 2.2
- The bhyve hypervisor now supports TPM and GPU passthrough
This version will now create user home directories in /home by default, instead of the traditional /usr/home. More information on the release and changes can be found via the release announcement page.
- OpenSSH version 9.5p1
- OpenSSL version 3.0.12, a major upgrade from OpenSSL 1.1.1t in FreeBSD 13.2
- OpenZFS release 2.2
- The bhyve hypervisor now supports TPM and GPU passthrough
This version will now create user home directories in /home by default, instead of the traditional /usr/home. More information on the release and changes can be found via the release announcement page.
This is it (Score:4)
Finally, the year of FreeBSD!
Re: (Score:2)
Netcraft does not confirm it.
Re: (Score:2)
They are just now getting FIDO/U2F support for SSH? Wow. That was added in OpenSSH 8.2, released in February 2020, and FreeBSD 13 was released in April 2021.
They also updated zstd to version 1.5.2, which includes a data corruption bug that was fixed in zstd 1.5.5 from April. (Versions 1.5.4 and 1.5.5 both included some nice speed-ups as well.)
Good job staying current, FreeBSD!
Re: (Score:2)
They were much slower to ship security key support for SSH than the OpenBSD people -- it's not like I am holding them to Linux distro standards. And they managed to pull a version of zstd from January but not notice by November that a data-corruption bug was fixed in April.
Re: (Score:2)
It's almost like FreeBSD is slow to port software. And this is totally new information!
Did you not see the headline where they only now are getting GPU pass-through in the included hypervisor? That's been a thing on most hypervisors now for like 10+ years, since Intel launched their virtualization extensions for directed IO back in the Core 2 Duo days of 2008?
Re:This is it (Score:4, Funny)
Re: (Score:2)
Does it also not run the Weyland-Yutani window manager?
Re: (Score:2)
"Does it also not run the Weyland-Yutani window manager?"
It's a bug hunt, man!
Re: (Score:2)
Nuke it from the repos, it's the only way to be sure!
Series Question (Score:1, Flamebait)
How does FreeBSD handle the shift from Linux towards Wayland and SystemD both are in their codebase heavily intertwined with Linux (wayland itself is just a protocol so, but not the rest)
X11 is dying, and are they going to shift from there to Wayland as well. SystemD might be the next problem with the infrastructure and programs sitting on top becoming more and more intertwined with it.
Can anyone from the BSD cloud enlighten me on this?
Re:Series Question (Score:4, Informative)
"X11 is dying"
Says who, you? Seems pretty damn healthy on all the systems we use. We could certainly do with a decent update - maybe X11R7 or X12? - but Wayland isn't it, far too many compromises (google them).
Re: (Score:2, Insightful)
That sure does sound like the post of someone going through the phases [wikipedia.org].
Re: (Score:2)
Remote display of applications is more important to me than running a dual-head display with the monitors at 72 Hz and 75 Hz precisely.
Re: (Score:1)
That's fine. Hold on to that while it still works. The refresh rate issue is a symptom of the development path, not a missing feature. In all matters of IT, older software on life support diverges from the general use case.
Today it's a refresh rate.
Tomorrow it's a hardware bug not getting fixed.
Some time it's a security bug not getting fixed.
And then at some point it's an OS release that is unsupported.
Re: (Score:2)
Narrator: "He was."
Re: (Score:2)
ssh -X works great on wayland client
If you're running wayland on remote, then you have whole different set of options. (no clue why you would do this at this point)
Re: (Score:2)
What a bizarre reply. What are you smoking?
Re: (Score:2)
> Says who, you?
The X global event loop is a problem.
Run xev and then go open your password manager.
How much do you trust all your Firefox extensions? That just updated overnight.
Re: (Score:2)
"The X global event loop is a problem."
Only if you don't have a clue about how to code multithreaded or multiprocess applications.
Re: (Score:2)
The fact is that there are signs of at least Gnome ditching X11 support, and it wouldn't be a shock to see other environments eventually follow. To the point about a "decent update", the problem being that the folks that would have done that did Wayland.
So as it stands, you have a tradeoff of compromises, X11 has some hiccups, such as scaling and context menus blocking screen lockers.
The biggest issue with Wayland is that it reorganizes the scope of the job of various things X11 did. X11 dealt with all so
Re: (Score:2)
It's not that the stuff is hard, it's that it has sucked to have things lumped together. Seaparation of concerns.
X11 was designed around the scenario that applications will draw lines, arcs, and text, with some pixmaps to be used sparingly, rendering to then typical 800x600 displays.. Remoting is fundamental, however low latency and reliable local networking is assumed. Nowadays the X drawing primitives are hopelessly outdated and everything leans heavily ito pixmaps everywhere, even for basic shape and
Re: (Score:2)
" Remoting is fundamental, however low latency and reliable local networking is assumed"
X was designed back when unreliable10mb/s vampire tapped 10base2 ethernet was considered cutting edge. If it could work then it can certainly work now on gigabit+ twisted pair and fibre networks. When was the last time you worked somewhere where the LAN was flaky?
"Nowadays the X drawing primitives are hopelessly outdated"
Depends what you want to do. I've written a number of games just using raw Xlib including 3D stuff. T
Re: (Score:2)
Gnome ditching X11 support
Sounds like it's time to ditch Gnome. Oh wait, that was 15 years ago.
Re: (Score:2)
"X11 is dying"
Says who, you? Seems pretty damn healthy on all the systems we use.
Says the developers of X who effectively have relegated it to life support only. Says the developers of hardware and software that is required to support functionality going forward, functionality they have said they currently aren't planning on incorporating.
We could certainly do with a decent update - maybe X11R7 or X12? - but Wayland isn't it, far too many compromises (google them).
And that's why X11 is dying. Even while hiding from the uncomfortable truth you admit that it could do with an update, one that certainly at this point has no pathway to ever coming. You may not like Wayland, but that doesn't change the uncomfortable f
Re: (Score:2)
"Says the developers of X who effectively have relegated it to life support onl"
X.org were not the developers, they simply maintained it.
" that doesn't change the uncomfortable fact that X11 is very much dying"
Funny thing is , I remember people saying just that 20 years ago. Get back to me when it actually comes true.
"Just don't expect it to feature ${new_thing} anytime soon, or ever really"
Someone else who's never heard of extensions.
Re: (Score:2)
We could certainly do with a decent update - maybe X11R7 or X12? - but Wayland isn't it, far too many compromises (google them).
You realize that the people who would write such an update are probably the ones working on Wayland, right?
There is a reason why the at-large community has decided that X11 / X.org is done and started from scratch on a replacement. And yes, it takes time to create such a complex piece of software from the ground up.
I will never understand this idea that "we only need updates to $PROJECT!!" when all the people that worked on $PROJECT and have intimate details of the deficiencies of design and limitations wi
Re: (Score:2)
If the people behind Wayland heard about portability, Wayland would not be an issue with the BSDs. But Wayland is full of Linux specific items, I would not be surprised if one day Wayland depends upon systemd.
This is a case if "no good turn goes unpunished". In the very early days of Linux, many items were ported over from BSD for using in Linux. glib and tcp (IIRC) for 2 items, that is because the BSD people tried to make things portable. Many of us remember the move from glibc to the GNU version, caus
Re: (Score:2)
"X11 is dying"
Netcraft confirms it!
Re: (Score:2)
As for wayland, you can run wayland on FreeBSD. Or X11. Or both with xwayland. FreeBSD runs the Linux drm drivers which helps a lot, as does mesa and vulkan integration.
As for systemd, meh. We've dealt with it in mostly ad-hoc ways to date. It's mostly a system administration problem, not a deeply embedded into programs problem for the most part.
Forgive my ironic chuckle at x11 is dying, though. I've heard FreeBSD is dying since the 90s. I'll believe it when I see it... though recent X release rates do tend
Re: Series Question (Score:2)
"Nobody wants to see a profitable business selling a FreeBSD version of something like what Canonical would make."
Then why are they working on an OS whose license promotes that behavior?
Re: (Score:2)
https://cgit.freebsd.org/src/commit/?id=bbb2d2ce4220 [freebsd.org]
Here you go, including the removel of the 26-year-old comment as to why home was under usr!