Debian FreeBSD Kernel Option Ready For Squeeze

[phoronix]

Phoronix: Debian FreeBSD Kernel Option Ready For Squeeze

Previously it was shared that the Debian team was working on bringing the FreeBSD kernel to this important Linux distribution as an option along side the Linux kernel. Today the Debian release team has announced that the FreeBSD kernel in FreeBSD (kFreeBSD) for x86 and x86_64 systems is ready and will be "handled equal with the other release ports" beginning with Debian Squeeze...

http://www.phoronix.com/vr.php?view=NzU4Nw

[ethana2]

FreeBSD based Ubuntu Server with ZFS coming soon?

[yoshi314]

FreeBSD based Ubuntu Server with ZFS coming soon?

i wish both linux and *bsd supported the same amount of filesystems. right now i'm most likely locked to linux, because of my setup involving ext4 on top of lvm on top of luks. i'm afraid that none of those technologies is supported on bsd :/

that is the major blocker for me against testing bsd on my box - i probably wouldn't be able to access any of my partitions from it.

[oneman]

wow what an unbelievable waste of time, I can't wait to run it!

[yogi_berra]

wow what an unbelievable waste of time, I can't wait to run it!

You should try Gentoo's port of it: http://www.gentoo.org/proj/en/gentoo-alt/bsd/fbsd/

[Ex-Cyber]

I don't really see the appeal of running a Linux distro's userland on FreeBSD, apart from the simple novelty of seeing it running. I do, however, see the appeal of using a FreeBSD kernel port as a testbed to develop the kind of general portability experience and infrastructure necessary to move to another kernel. We have no guarantee that Linux will continue to be the foremost FOSS kernel indefinitely. Even if it stays strong, we might all be running some kind of backwards-incompatible "Linux 3" (paravirtualized 2.x VMs notwithstanding) ten years from now.

[nanonyme]

Yeah, undoubtedly Hurd will develop far enough by then that we can use it instead. Wait, what? *chuckle* I don't really see any opensource alternatives to Linux very promising due to Linux simply having pretty much the best driver support. If you really want to see some problem in Linux, it's that it's GPL and therefore can't ever support ZFS.

[benmoran]

GPL is it's biggest strength.

[drag]

I don't really see the appeal of running a Linux distro's userland on FreeBSD, apart from the simple novelty of seeing it running. I do, however, see the appeal of using a FreeBSD kernel port as a testbed to develop the kind of general portability experience and infrastructure necessary to move to another kernel. We have no guarantee that Linux will continue to be the foremost FOSS kernel indefinitely. Even if it stays strong, we might all be running some kind of backwards-incompatible "Linux 3" (paravirtualized 2.x VMs notwithstanding) ten years from now.

Portability is a important quality to have in software. It is a nice test of correctness, for example. For chasing out hidden bugs or bad design choices.

There may not be any really apparent reason behind the madness, but it exists.

This sort of reason is why it took Linux distros months of work to properly support AMD64, but it took Microsoft years to release a OS.

As you should know a OS is made up of layers of abstraction. By keeping things compartmentalized then you can swap out parts when better things come along and it is much easier to maintain and improve the software. Changing the kernel out from underneath your OS is a good way to test things for correctnees, right?

This sort of thing is why bug fixes and security fixes can be made for enterprise Linux systems in a few days, while it can take weeks for Microsoft to do the same for Windows.

If your high level applications broke when they get ported to FreeBSD it means that it is likely that the Linux kenrel is not following specifications correctly and that changes in the kernel can break your applications.

That sort of thing.


IT may not seem to be like a big deal, but you have to realize that this sort of thing is a end result of decades of people following this sort of design approach... so that the GNU/kFreeBSD folks can create a stable, full featured, operating system with probably only a couple dozen or so people working on it in their spare time as a hobby.

This sort of flexibility and easy customizability is a huge asset.

[kraftman]

GPL is it's biggest strength.

Of course.

@Nanonyme

There are many other Linux advantages then just driver support. There's a chance Linux will have ZFS :>

@Drag

If your high level applications broke when they get ported to FreeBSD it means that it is likely that the Linux kenrel is not following specifications correctly and that changes in the kernel can break your applications.

Or opposite.


Btw. no more bashing Phoronix benchmarks :P You/we can test two different kernels in exactly the same environment :>