Debian just added the kernel of FreeBSD to the unstable branch of its GNU/Linux distribution. The package contains pristine 5.3 kernel source together with patchset for conformance to Debian’s requirements. This is a first step towards inclussion of the Debian GNU/kFreeBSD port as a candidate for future Debian releases.
Debian adds kernel of FreeBSD
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
Did I miss somthing…..whats the point?
Debian is more than a Linux distro…it’s a distro that can be adapted to different kernels. Debian HURD is the example I know of.
FreeBSD propaganda
http://people.freebsd.org/~murray/bsd_flier.html
WHy does the Linux community hate the idea of FreeBSD. UFS 2 is far superior to ext3, Debian is the best linux flavor out there, nothing wrong in Debian incorporating FeeBSD. Kudos to the guys at debian
Their propoganda talks about the upcoming release of linux 2.4… It’s VERY outdated. At this point (2.6.11) I think Linux has pretty well caught up to BSD in most things…
How about old propaganda considering that comparison sheet talks about the 2.4 release of the linux kernel “will” … and also how RedHat used to turn on way too many services. Which was true about 5 years ago and was why RedHat out of the box installs would get owned as fast as a windows box.
So… wait a minute… is this trying to say I could have a working Linux system and just apt-get myself to FreeBSD?
I know Debian’s been incorporating other kernels for a while now. I imagine not all of them are binary-compatible, though. I suppose BSD’s have Linux emulation, but since *BSD has its own core-utils type stuff, wouldn’t that mean the GNU packages…? I think I’m just confused by the wording.
you should chose your words more thoughtfully. free is a bit too subjective. i think you should say “less restrictive”. the difference is what free is quantifying. anyhow, i think this is a great idea. at least a good exercise.
“Their propoganda talks about the upcoming release of linux 2.4… It’s VERY outdated. At this point (2.6.11) I think Linux has pretty well caught up to BSD in most things…”
I don’t think so. As Linus says, the linux kernel 2.6.x is slower than 2.4.x
http://news.com.com/Torvalds+Put+Linux+to+the+test/2100-7344_3-5646…
BTW, why the debian project is searching another kernel and making a new unstable branch, if linux is the best ?
You’re certainly not a lawyer ! Except that it’s good for Debian to have many kernels, I didn’t understand anything you wrote !
Suddenly calling it GNU/* makes sense. This is a *GNU* distro *without* Linux.
If by free you mean allowing 3rd parties to steal and profit from the hard work of volunteer developers without giving anything back, then yes, by all means, the BSDL is far more free than the GPL.
I’ll chose fairness over freedom anyday. Apparently from SF/Freshmeat license stats most people agree with me.
Hmmm Guess i’m lucky since im running on 2.6.10.
Debian is not searching for a kernel they are just adding another choice. But its a dead idea.. Probably very few users will be interested in BSD.. at least i dont know any that are.
BTW, why the debian project is searching another kernel and making a new unstable branch, if linux is the best ?
They are not in searching of another kernel… Debian already run on other kernels like Hurd and NetBSD far time ago, they started the use of freebsd kernel just because someone wanted, without any technical reason.
BTW, why the debian project is searching another kernel and making a new unstable branch, if linux is the best ?
They are not searching another kernel, they have added the FreeBSD core, like they are also maintaining a GNU/Hurd branch.
Debian provides a multitude of choice, which is a Good Thing ™.
If by free you mean allowing 3rd parties to steal and profit from the hard work of volunteer developers without giving anything back, then yes, by all means, the BSDL is far more free than the GPL.
Look… You guys just have to get over this. Those who contribute to BSD-licensed projects *know* how their work can be used. So, if they choose to volunteer, that’s because they are willing to have their work used in that way. Thus it’s not stealing, and there is nothing unfair about it. But it is unfair of you to judge based on presumptions that don’t apply to the situation.
I suspect that the FSF is quite happy with the idea of Debian GNU/kFreeBSD, because they’ve long felt that Linus gets too much credit. A system with a BSD kernel and GNU userland that casual users would find indistinguishable from a Debian GNU/Linux system would provide an interesting demonstration.
Also, back around ’91 the GNU people were starting to look at extending the Networking-2 partial releases coming out of Berkeley to full “GNU/BSD” systems, when two things intervened: the release of Linux and the AT&T/BSD lawsuit.
So the Debian folks are building the system that might have been.
I don’t think so. As Linus says, the linux kernel 2.6.x is slower than 2.4.x
http://www.redhat.com/software/rhel/3features/kernel/
BTW, why the debian project is searching another kernel and making a new unstable branch, if linux is the best ?
It is a shame that you haven’t even learned that open source is about choice.
Actually, in some situations 2.4 IS faster then 2.6.
If Debian releases a version of Free BSD with GNU userland as GNU/FreeBSD, wouldn’t that make the kernal in that release covered under the GNU license? Isn’t it viral? In this way the Debian folks would be “stealing” (it’s not stealing all right, get over it) the kernal from FreeBSD, and would continue to “use” the work by applying bug fixes from the FreeBSD coders.
Btw, they should be able to add any changes or fixes they come up with for the FreeBSD kernel back to FreeBSD. All they would have to do is give the code back to FreeBSD before they release it (and therefor change the new GNU license), so that it is covered under the FreeBSD license and then take it back for their needs.
If Debian releases a version of Free BSD with GNU userland as GNU/FreeBSD, wouldn’t that make the kernal [sic] in that release covered under the GNU license?
No, it wouldn’t. The GPL doesn’t put any requirements on the license of the OS on which you run GPL’d programs.
The biggest drawback on Debian seems, that the devs don’t seem to like user documentation.
This is a major advantage of the BSD systems, they are all well documented, only few Linux distros are to such extend.
The actual discussion in LKML is at: http://kerneltrap.org/node/4940
WHy does the Linux community hate the idea of FreeBSD. UFS 2 is far superior to ext3, Debian is the best linux flavor out there, nothing wrong in Debian incorporating FeeBSD.
I’m curious, what makes UFS 2 bettr than ext3?
How would UFS 2 measure up against other popular file systems like XFS, and ReiserFS?
This is a *GNU* distro *without* Linux.
Except for the fact that GNU didn’t make that distro. The only reason it’s called “GNU/kFreeBSD” is because Debian chose to call it that.
If running GPL licensed GNU userland bits on a non GPL kernel was an issue the GNU project would never have gotten off the ground. SunOS 4.x not Linux for a long time was the primary GNU userland development platform for many people.
I see this as very interesting, it gives great choice. It also shows how mature and portal the GNU userland needs to be for this to work.
I suspect that one day we might also see a Debian GNU/OpenSolaris done in a very similar way to the Debian GNU/*BSD releases. Briging a nice (for me anyway) historical twist to the GNU project.
There is a project to use the BSD kernel as core of Gentoo: http://www.gentoo.org/proj/en/gentoo-alt/bsd/
It’s only few months old.
What about going on the other direction, I mean, joining the Linux kernel with BSD (non-GPL) stuff? So we wouldn’t have to (according to Stallman) call it ‘GNU/Linux’ anymore…
I have used ReiserFS, ext3 and UFS 2. Compared to UFS ext3 felt slow, also I have had ext3 crash on me much more than UFS, could also be because I have used FreeBSD in my server for only a month. ReiserFS is pretty fast, I would personally choose REiserFS over ext3 but there aren’t many disrubtuions of Linux that allow ReiserFS install the ones I know are Ubuntu and XandrOS. Ubuntu was freaking fast on Reiserfs compared with Fedora using ext3.
According to Firefox, the last modification date for this page was January 5, 2002 19:32:27.
The propaganda was accurate… The propagandist should use more up-to-date information, though.
Add slackware to your list. they use ReiserFS by default. at least last time i check slackware 10 was using ReiserFS as its first option.
There is no point comparing BSDL and GPL they don’t have that much in common.
BSDL (for reasons that elude me) is just a small restriction short of being in the public domain.
GPL is an FSF tool to spread “freedom”.
How is it a thread about FreeBSD being added to the Debian Family turns into a war between licensing philosophies?
Where are all the threads on building this kernel and bootstrapping it to another partition and bootloader leaving one the ability to choose between the various kernel options underneath the Debian Hierarchy?
You can add Gentoo to the list. I even think there is a livecd for Reiser4 installs somewhere (not official though).
Ubuntu was freaking fast on Reiserfs compared with Fedora using ext3.
Very bad comparison because you compare two different distros with different format systems. It would make sense if you were compare Ubuntu with reiser and Ubuntu with default fs(ext3).
FYI, you can install ReiserFS on Fedora on the boot screen : linux reiser selinux=0.
I was under the impression that the thing about BSD was the development model which leads to an integrated OS with a clear direction.
I fail to see how copying their kernel could bring that to Debian…
FSF/GNU extremists (some of them) don’t like FreeBSD because the BSD license is more free than the GPL.
Or maybe they like FreeBSD just fine. But get tired of the BSD extremists missing the point of the GPL. We hear you talking about free this and free that, but you never define this free thing, which leads me to believe you don’t really understand what you’re talking about.
If it gives freedom to take away freedom, it results in less total freedom than one that takes away that freedom from you, requiring to preserve freedom.
*head explodes*
Mandrake (the late) has had reiserFS as far back as 9.x. Maybe not at it’s 1st choice but it was there.
>It also shows how mature and portal the GNU userland needs >to be for this to work.
That’s why the initial idea of debian team – use bsd’s libc + kernel with gnu software failed misreably. Some parts of gnu userland turned out to be unportable, due to lack of development culture, and loads of linuxisms and glibcisms in the sources… And anyone who ever tried to compile auto*, configure, etc. oriented programs on platform with non-gnu userland/kernel/compiler/includes/libraries knows that it’s going to be a major PITA next time too… Because one needs “right” awk, “right” cc, “right” ld and “right” everything. Where “right” is NOT standards (I mean posix/sus) compatible, but is GNU. The destruction of a single positive result of unix wars (standards) is in progress, due to gnu/linux crowd with their “mature” and “portal” software.
Mandrake (the late) has had reiserFS as far back as 9.x. Maybe not at it’s 1st choice but it was there.
Also SuSE used ReiserFS as its first choice for a long time. But at that time, it was actually a bad choice. ReiserFS caused huge problems with RAID and SMP configurations, because everyone was testing it using IDE and on simple personal computers.
Because one needs “right” awk, “right” cc, “right” ld and “right” everything. Where “right” is NOT standards (I mean posix/sus) compatible
You forgot to mention: Neither is xBSD fully POSIX and e.g. UNIX98 complaint.
Did I tell anything good about BSDs?
BTW I use several linux distros (including debian) + fbsd + openbsd + solaris + windows, without special preference (though, windows tends to cause more troubles
). I’m not pro or against any system in particular, just against zealous praising of smth. based on false assumptions.
I don’t think so. As Linus says, the linux kernel 2.6.x is slower than 2.4.x
I think the benchmark you are referring to is comparing Red Hat’s heavily modfied 2.4 kernel, which contains 2.6 features.
Regardless 2.4 is still faster than Free/Net/Dragonfly/Open BSD.
ext3 has worked fine for me, and if you don’t like it there is jfs, xfs, reiserfs, and reiser4 which are all better than ufs2.
I have nothing against BSD, but it is not GNU/Linux; it is far behind in terms of important functionality, stability, performance, and software support.
So osnews readers are interested in OSes, and yet they don’t know that Debian is a F/OSS system that just happens to rely on L***x for a kernel, but likes the idea of being able to use FBSD, NetBSD, the HRD or whatever.
>Regardless 2.4 is still faster than Free/Net/Dragonfly/Open >BSD.
Why do the ghosts have short haircuts? (the question makes more sense than your statement). Faster doing what? On which platforms?
>ext3 has worked fine for me, and if you don’t like it there >is jfs, xfs, reiserfs, and reiser4 which are all better >than ufs2.
Better with regards to what?
>I have nothing against BSD, but it is not GNU/Linux;
And you are a GNU/Linux zealot
> it is far behind in terms of important functionality,
> stability, performance, and software support.
I’ve allready commented on perfomance.
I could (possibly) agree about software support and scalability. But stability? important functionality? Sounds very much like Torvald’s “Solaris has no advantages against linux” raving….
Why would i want to use the BSD kernel instead of the Linux kernel? are there big difference?
I assume you mean FreeBSD’s kernel, because various BSD’s use different kernels and differ in features.
They are both UNIX-like kernels. Particulary Linux strives for POSIX conformance, atleast as much as it can. Linux’s ability for threading and support for a large number of processors has much more maturity and ability than the FreeBSD’s. There are many other differences, that is the main one.
There is very little reason to use the FreeBSD kernel in my opinion.
…they simply express their freedom in different ways.
The BSDL and the GPL are both good licenses, they are both compatible, and both FreeBSD and Linux distributions include programs covered by each license.
Both licenses have their own strengths and weaknesses (though the GPL seems more popular, at least according to freshmeat). They are both equally, if differently, free. Anyone trying to argue that one is better than the other is not worth listening to. They are simply trying to start another sterile flamewar. Ignore them. Sometimes I suspect they really are MS astroturfers simply trying to sow disunity among F/OSS enthusiasts.
I like both licenses, just as I like both the idea of Debian Linux and Debian BSD. I really wish some people with anti-GPL agendas would stop trolling, and some pro-GPL posters would stop biting…
Why do the ghosts have short haircuts? (the question makes more sense than your statement). Faster doing what? On which platforms?
Generally GNU/Linux is faster. All the recent benchmarks I have seen point to this. All the benchmarks I have seen, usually bechmarking networking performance or a particular service, confirm Linux being faster on the x86 architecuture. However we will not know if FreeBSD is faster on pp64 or S/390 because it does not run on it.
And you are a GNU/Linux zealot
Not really I ran FreeBSD for years and I currently have NetBSD installed and is to inevitably be a personal webserver. I am just pointing out the truth.
Linux does have important functionality such as the ability to run on quad processor servers reliabily and efficiently.
Stability is a matter of opinion, I would say.
>There are many other differences, that is the main one.
Why is that difference main? what about NFS perfomance and stability, for example? Code maturity (i.e. amount of childish ugly hacks and in kernel code)? Security? how many smp-boxes do you have on your desk? How many processors do you need on the router or firewall? There are systems with scalability that is by orders of magniture ahead of linux’s kernel (I mean solaris). So, does that make solaris a silver bullet?
If there is a 20% difference in kernel speed, I say, so what?
If kernel speed is so important to you that you need that final 20% boost, for example if you are doing ganing or high-performance computing, then you might look at the choices available. For everyone else, I would say that functionality, stability and reliability are more important than that last 20% speed boost.
By “everyone else”, I mean 95% of OSNews readers and posters to this forums. Nobody will be playing games on FreeBSD and few on Linux, so that leaves the high-performance computing crowd. Who’s in that?
There are systems with scalability that is by orders of magniture ahead of linux’s kernel (I mean solaris). So, does that make solaris a silver bullet?
I think you are quite mistaken. Scalability is mainly a platform issue. Linux scales on IA-64 upto 256 CPU and beyond per NUMA node. That IA-32 has very poor scalability is no secret fact. Otherwise, fell free to test Solaris 10 x86 on 32-way SMP.
I do not know any childish ugly hacks in Linux, if that is what you are implying, not present in different forms in other open-source systems. And when Solaris 10 source code is released to the masses I am sure there will be some there also.
There are systems with scalability that is by orders of magniture ahead of linux’s kernel
Right, that is why Linux is chosen to run on the majority of super-computers.
Security?
By your experience, I guesst you do not need somebody else telling you that Red Hat Enterprise Linux 4 also supports mandatory access control as like Trusted Solaris.
>I think you are quite mistaken. Scalability is mainly a >platform issue. Linux scales on IA-64 upto 256 CPU and >beyond per NUMA node. That IA-32 has very poor scalability >is no secret fact. Otherwise, fell free to test Solaris 10 >x86 on 32-way SMP.
I’m speaking about sparc n-head hardware. Where linux does not really run well
As for x86 – linux 2.6 and solaris 10 perfomance on SMP (tested myself with 2 and 4 processors) is on par. BUT Solaris does have special facilities that are unavailable or have inferior rivals on linux even on x86 and x86-64. Zones, dtrace, and (I am desperately waiting for it 
) zfs.
So Solaris has better scalibility than linux? That’s news to me, and several benchmarks I’ve seen that show the two are pretty much even.
ReiserFS should be fine with any distribution. The ability to use it is built into the kernel and has been there for quite some time. The only question is whether your distro gives you the choice when you are installing it. All except the newbie distros should give you a choice of several types.
Oh, and Linspire is a big backer of Reiser, so add them to the list which defaults to ReiserFS. (version 4, in this case)
>Right, that is why Linux is chosen to run on the majority of >super-computers.
🙂 So, windows is the superior platform, just for reason of being run on majority of desktops?
>By your experience, I guesst you do not need somebody else >telling you that Red Hat Enterprise Linux 4 also supports >mandatory access control as like Trusted Solaris.
I was thinking about openbsd and their understanding of security as the endless process of code auditing, not MAC/Trusted*
I think we both know that security of the system is not determined only by facilities present.
I’m speaking about sparc n-head hardware. Where linux does not really run well
Then specify that explicitly in your critism. SPARC makes a extremely small fraction of Linux hardware installation base. If you need SMP scalability, you definitively should try to run Linux on IA-64 or POWER5. And both platforms are definitively not inferior compared to SPARC.
BUT Solaris does have special facilities that are unavailable or have inferior rivals on linux even on x86 and x86-64. Zones, dtrace, and (I am desperately waiting for it
) zfs.
Dtrace is the typical example always popping up in the discussion – there seems to be no other “real” argument for Solaris.
Concerning zones/containers, why don’t you try to run a NetBSD kernel in it? This is one field where Xen is definitively more superior in term of technology.
Can you then give me a reason why AFS and GFS cannot serve your porpose?
>Dtrace is the typical example always popping up in the >discussion – there seems to be no other “real” argument for >Solaris.
Well, it is an argument
>Concerning zones/containers, why don’t you try to run a >NetBSD kernel in it? This is one field where Xen is >definitively more superior in term of technology.
True. Though, I can’t imagine a real-life situation with the need to run netbsd under solaris. W2k3 – probably. NetBSD – why?
>Can you then give me a reason why AFS and GFS cannot serve >your porpose?
Builtin volume management on filesystem level, copy-on-write, fine-grained I/O priority/scheduling, transactions (rollbacks!!!). AFS and GFS are network-distributed centric, I’m interested in a huge fs on a single server. If zfs is really as good as sun tells, I can see no rival for it. Though – I agree, it’s vapourware for now.
You do know that you can do this right?
You can take the FreeBSD kernal and re-release it under the GPL.. its perfectly legit. Thats part of the power the BSD license gives you. You just can go the other direction.
I was thinking just the other day of taking some orginal BSD source code and turning it into a GPL application. So, I did quite a bit of research on this to double check it.
True. Though, I can’t imagine a real-life situation with the need to run netbsd under solaris. W2k3 – probably. NetBSD – why?
This is just an example. There are however many good reasons why people are interested to run a strictly separated virtualization rather than just zones/jails type “soft” spearation, for security reasons, to begin with.
That you can run an entirely different kernel is just a demonstration of the strict virtualization.
AFS and GFS are network-distributed centric, I’m interested in a huge fs on a single server.
AFS and GFS are just two examples. There are also GPFS, SANFS, and many others. If you look around in the fields with extreme I/O demands, these are the typical FS used. I do not see the point why anyone should have a single server centric solution, or emphasize this.
The stability & hardiness of FreeBSD with the beauty that is apt..
Ahh.
FreeBSD is great, just not as a “distro”. As an Operating System – yes. A sweet-n-easy joyride it is not.
A DebianBSD would be!!!
“I really think that 90% of GNU Hippies dont’ care about the licenses, they just dont’ want people making money off of their code (such as Microsoft). And yes Microsoft using BSDL code is a win-win situation for a billion reasons.
I think that 90% of BSD Hippies wish they could get a handful of those billion reasons Microsoft put in the bank”
You’ve missed the point, BSD is about excellence. What BSD devs care about is writing good code and have the community use it. There is sactisfaction in the fact that your work is thanked and appreciated by millions of people.. hope one day you can realize it’s not all about money.
“Regardless 2.4 is still faster than Free/Net/Dragonfly/Open BSD.”
> “Linux > BSD!!”
ext3 has worked fine for me, and if you don’t like it there is jfs, xfs, reiserfs, and reiser4 which are all better than ufs2.
> “Linux > BSD!!”
I have nothing against BSD, but it is not GNU/Linux; it is far behind in terms of important functionality, stability, performance, and software support.
> “Linux > BSD!!”
Come on, where is any of the BSDs beyond linux in stability? Try to stress-test linux and then a FreeBSD 4.10 or a OpenBSD box and then tell me something.. You’re talking about linux, the OS that still has problems with fork() bombing.. FreeBSD won’t crash with a fully loaded desktop (GNOME) and fork()ing it up to 8000 processes.. linux crashed on my as soon as I suspended the fork bomb and then executed it again. Just an example out of dozens I could think off.. Kernel security is important too, try to search for linux kernel exploits vs *bsd kernel exploits, then come back saying linux is more stable.
Performance ? Not that big of a difference, and I think you’ll be amazed when the work on FreeBSD’s 5.X branch is over.. And it’s not about performance only, what good is performance if the system isn’t stable ?
Software support ? Ports system + the ability to run 99% of Linux software pretty much even the situation, wouldn’t you agree ?
About filesystems.. I’ve had ext3 dying after an unexpected power loss.. As for UFS, I’ve never, ever lost a filesystem due to ANY REASON in 6 years of FreeBSD. Be it UFS 1 or 2. Can’t comment on the other filesystems as I’ve never tried them, but UFS seems pretty stable and mature to me – as it should be, it’s many, many years old.
“I do not know any childish ugly hacks in Linux, if that is what you are implying, not present in different forms in other open-source systems. And when Solaris 10 source code is released to the masses I am sure there will be some there also.
There are systems with scalability that is by orders of magniture ahead of linux’s kernel
Right, that is why Linux is chosen to run on the majority of super-computers.”
lol, and you deny you’re a linux zealot.
Last time I checked, it was Solaris 8, HP-UX and AIX running on super computers, not Linux.
“What BSD devs care about is writing good code…”
oh please if you were talking about NetBSD’s code then i would agree but if you say that FreeBSD devs give a fuck about clean code and not implementing hacks then i have completely different opinion.
maybe you have something better to do and not only troll around here ?
Isn’t FreeBSD a registered trademark, now owned by The FreeBSD Foundation? I don’t see why The FreeBSD Foundation should have given permission to use the trademark in a project that is going to be in competition with FreeBSD, since FreeBSD markets the integration of the entire system and not just the kernel as one of its strengths.
Yeh, they create the operating system’s Frankenstein
Thanks guys!
p.s.This guys have the “monster”, but in the other hand don’t ever had gnome2.10 in the packages!!!
LOL
wang burglar sausage!
”
Last time I checked, it was Solaris 8, HP-UX and AIX running on super computers, not Linux.
”
From the Top500 supercomputers list
NUMBER 1
IBM’s BlueGene/L Supercomputer — runs Linux
NUMBER 2
SGI’s Columbia Supercomputer — runs Linux
NUMBER 3
Earth Simulator — runs Enhanced NEC Unix
NUMBER 4
IBM’s MareNostrum — runs Linux
NUMBER 5
IBM’s (I believe) Thunder — runs Linux
So, out of the top 5 spots, Linux holds 4…but wait, there’s more. It was recently estimated that Linux holds 300 spots out of the top 500 for Supercomputers. So, if you combine AIX, Solaris and HP-UX, then they almost compare to Linux, but not quite.
See how much more believable you sound when you do a little research instead of saying “last time I checked”.
From http://www.forbes.com/home/enterprisetech/2005/03/15/cz_dl_0315linu…
Meuer reckons Linux powers 301 of the 500 top machines, compared to 189 on Unix, two on FreeBSD, a Unix variant, and one on Microsoft’s (nasdaq: MSFT – news – people ) Windows. (Seven machines are categorized as “other.”)
Moreover, Seager says Linux outguns popular Unix operating systems like AIX and Solaris from Sun Microsystems (nasdaq: SUNW – news – people ) because those systems contain features that make them great for commercial users but add a lot of system overhead that ends up limiting overall performance. One example: a “virtualization” feature in AIX lets many applications share the same processor but “just hammers performance,” Seager says.
Well, guess what:
Some people like:
1) Linux Kernel
2) FreeBSD Kernel
3) FreeBSD Utilities
4) GNU Utilities
I can honelsty say, I like the FreeBSD kernel and do like the GNU Utilities. The FreeBSD kernel is easy to update, via CVS and easy to rebuild, by simply editing a text file.
PS: I like the run levels in *BSD vs the Sys V. or in Linux. And yes, there are some Linux Distro’s that use the *BSD run levels. I am wondering how the run levels will work in the Debian Release.
This does scratch an itch for some folks. Forget the all the talking of smack. Use what you like and prefer.
Enjoy
Long live the depenguinator.
Last time I checked, it was Solaris 8, HP-UX and AIX running on super computers, not Linux.
The blatant mismatch with the reality is already commented by other. But what surprises and wonders me is actually that – sorry – some of your claim appears just ridiculous:
– For AIX, you should know that IBM is slowly transiting from AIX to Linux, especially in term of HPC. IBM pSeries HPC solutions running all AIX was mainly on POWER4+. You should update your information.
– The high of Solaris based HPC solutions were 2000 (5 years ago!), and they still ranked typically below top 40. With CPU like POWER5 or Itanium 2 emerging in recent years, Solaris based HPC is just not competitive anymore.
– HP/UX? How well do you actually know Hewlett-Packard in the first place? PA-RISC and HP/UX was never the HPC platform of choice for HP. The AlphaServers raking at place 2 and 3 on Top 500 list in 2002/2003 was all running Tru64.
it is nice to have a variety of seed. the farmer then has a better chance of adapting to changes. we all know this. does this not apply to distros. we need variety for who knows what will be required in the future. we need to keep the airwaves free!
That’s what I ‘d like to see, apt-get handling binary packages on FreeBSD (kernel+userland). Ports are nice and all, but binaries aren’t their strength.
FreeBSD kernel + GNU… not that interesting IMHO
What’s wrong with packages? Alternatively, what’s better about apt-get?
I think this is dependent on how you look at the term “free software.” The way I see it, it’s not meant to imply that the user is free to do what they will with the software, but that the software itself is free. As if it were a person — the software is emancipated so to speak, no one can “control” it in any way.
This is a lot different than saying anyone is free to do what they want with the software. By looking at the software itself as having been emancipated, you can see why the term “free software” adheres more properly to GPL than the BSD Licesne. If corporations are free to grab hold of software and exploit it so to speak, then the software itself has lost it’s free nature.
This will make a lot more sense with AI programs and the idea of conscious data — to the future!
And before you all ask… yes, I’m a communist.
No wonder Debian Linux 3.1 is taking so long to get release
the Debian team are doing other things.
Last time I checked, it was Solaris 8, HP-UX and AIX running on super computers, not Linux.
Sorry, but you’re wrong. Linux dominates the supercomputer field. Blue Gene, the SGI supercomputer and other top entries all run on Linux.
http://news.com.com/Blue+Gene,+Linux+top+supercomputing+list/2100-7…
You’re talking about linux, the OS that still has problems with fork() bombing.. FreeBSD won’t crash with a fully loaded desktop (GNOME) and fork()ing it up to 8000 processes..
This is true for some distros’ default settings, however it’s trivial to fix this by limiting processes. On some distros (Debian, I think) the default is quite safe and the system is immune to such fork bombs.
What BSD devs care about is writing good code and have the community use it. There is sactisfaction in the fact that your work is thanked and appreciated by millions of people.. hope one day you can realize it’s not all about money.
That’s pretty much the same motivation that’s behind Linux devs as well. And the GPL license is certainly not all about money! Can’t we just stop these pissing contests and agree that both OSes are good (and work very well together, might I add)?
“No wonder Debian Linux 3.1 is taking so long to get release
the Debian team are doing other things.”
Buh-dum-bum..
But they just dropped SPARC, ARM, MIPS, PA-RISC and S/390, and I think maybe M68K from stable.
I just thought of something…
This HURD, FreeBSD, $KERNEL release thing is the equivalent of maintaining multiple architectures in this world of approaching singularity…
It only makes sense to sharpen the ‘architecture’ mentality once the ‘kernel’ on ‘narrower assortment of architectures’ evolution kicks in..
>>>Direct comment link re: what’s wrong with FreeBSD
By Larry (IP: —.de) – Posted on 2005-04-08 20:42:41
FSF/GNU extremists (some of them) don’t like FreeBSD because the BSD license is more free than the GPL.<<<
Sounds like another LINSPIRE fanboy. Hrm… “GNU extremists”. Freedom is such a terrible thing.
>>>Direct comment link re: what’s wrong with FreeBSD
By Larry (IP: —.de) – Posted on 2005-04-08 20:42:41
FSF/GNU extremists (some of them) don’t like FreeBSD because the BSD license is more free than the GPL.<<<
Sounds like another LINSPIRE fanboy. Hrm… “GNU extremists”. Freedom is such a terrible thing.
It’s like this:
BSDBush: These people are extremists. They’re jealous of our freedom.
Free is free. Free is me being able to download a distro and apps for free, and not have to bend over for a licensing agreement. If you can’t understand what freedom is, then you’ve been duped in this corporate culture of software patents and licenses.
Does this mean I’m not going to give anyone any money for their hardwork? By no means! I’m going to invest in the Free Software Foundation and the distro of my choice. Screw the companies that want to act like the rest of the proprietary herd.
Quit bringing Shrub into this. A total strawman on your part.
Gent: I think this is dependent on how you look at the term “free software.” The way I see it, it’s not meant to imply that the user is free to do what they will with the software, but that the software itself is free. As if it were a person — the software is emancipated so to speak, no one can “control” it in any way.
Problem… The GPL only really affects the distribution of the software.
For example: I could “create my own version” of gcc and use it as I will and never return so much as a single line of source code to the community. If I gave the software to someone else however, I’m obligated to provide source code. However, the community still doesn’t have it. As a result the “freedom” from the GPL really isn’t that much “greater” than BSD.
You’re talking about linux, the OS that still has problems with fork() bombing.. FreeBSD won’t crash with a fully loaded desktop (GNOME) and fork()ing it up to 8000 processes.
That argument was debunked by Alan Cox, Red Hat kernel maintainer on this post:
http://fcp.homelinux.org/modules/newbb/viewtopic.php?topic_id=12794…
h: You’re talking about linux, the OS that still has problems with fork() bombing.. FreeBSD won’t crash with a fully loaded desktop (GNOME) and fork()ing it up to 8000 processes..
This depends significantly on the kind of fork bomb. And maybe you should shed a light on the type you are talking about. A typical bash process, for example, has a 300 kB large (non-shared) memory footprint. 8000 processes means a bit larger than 2.3 GB. Even today, this is not kind of typical hardware configuration you can find. And requiring a system with less physical RAM to handle 8000 bash processes well is just a physical impossibility.
This is true for some distros’ default settings, however it’s trivial to fix this by limiting processes. On some distros (Debian, I think) the default is quite safe and the system is immune to such fork bombs.
The default is always a trouble since the ulimit/limits.conf depends significantly on the hardware and application it should run.
For example: I could “create my own version” of gcc and use it as I will and never return so much as a single line of source code to the community. If I gave the software to someone else however, I’m obligated to provide source code. However, the community still doesn’t have it. As a result the “freedom” from the GPL really isn’t that much “greater” than BSD.
I understand. Buth again, I think you’re looking at it from an aspect of what the user can do. The GPL is written in a way to protect the software itself from control. It effectively gives the software “rights.”
You can acquire a copy of that software, do what you will, use it as you will etc, but there is no way for you, under the GPL to exploit the software in and of itself for the generation of capital. At least not without giving back. This effectively abolishes the aspect of private property in the form of expropriated labor within the realm of GPLed “free software.”
I’ve heard a lot of people argue how free software is not a communist ideal, and how the ability to charge proves it. But the software itself is actually quite fitting to what I’m talking about, as communism does not particularly seek to end simple money transactions (although it does), but more importantly seeks to change a social condition.
The GPL upholds this changed social condition where the BSD License does not. The software is “free software” in the same way we look at “free laborers/free workers.”
This is what I’m saying. It’s the difference between the outlook of capitalism, which is a freedom and right to private property, and the outlook of communism, which realizes the infringement on the freedom of the worker DUE to private property.
In this case you replace the worker with the software itself, so that the capital it generates may very well benefit a company or a person materially speaking, however, the labor itself is not exploited as in order to do this they will have to return that value to the community at large.
I don’t want to threadjack this into a political discussion (as hard as that may be to believe). So let me just say plainly, that this is my outlook, most people who I have met willing to argue the similarities between the free software movement and communist ideals are very uneducated on the issue itself. So if you don’t understand where I’m coming from, it could just be because of our completely differing points of view. If you do understand where I’m coming from, but simply disagree… it’s your right to do so.
The default is always a trouble since the ulimit/limits.conf depends significantly on the hardware and application it should run.
True, which is the argument given as a reason to not limit processes by some distros. Personally, I think a simple question in the installation process (i.e. is this going to be a Desktop/Workstation) could be used to provide better defaults depending on intended use.
Supercomputing on Linux:
http://news.bbc.co.uk/1/hi/technology/3496754.stm
http://news.bbc.co.uk/1/hi/technology/3532706.stm
http://news.bbc.co.uk/1/hi/technology/3983131.stm
http://news.bbc.co.uk/1/hi/technology/2940422.stm
http://news.bbc.co.uk/1/hi/sci/tech/1490272.stm
Debian’s Stunning New Release Strategy i.e. Vancouver Proposals:
http://xrl.us/ffg4
So, you are telling me that you better know the linux kernel than Linus … OK
I think you must stop to be a linux zealot and accept that the linux kernel has some problems.
“Regardless 2.4 is still faster than Free/Net/Dragonfly/Open BSD.
I have nothing against BSD, but it is not GNU/Linux; it is far behind in terms of important functionality, stability, performance, and software support.”
I think you haven’t seen (or forgotten) the benchmark between BSD and linux done by the linux user fefe:
http://bulk.fefe.de/scalability/
In fact, FreeBSD and NetBSD scale better than linux. BTW, why at the beginning of this bench, when linux was winning, every linux users point to that site. But now that BSD community teach fefe how to configure BSD systems and that now BSD scales better, every linux users have forgotten this link …
In fact, you have something against BSD, and telling that you are not won’t make you an “open minded” linux user …
linux is making the same that windows done in 80/90, it is good enough to be used and less expensive than UNIX (but it is changing), that’s all …
Just tell me which distribution on linux this all computers are running. I am curious.
You haven’t in mind that the linux running on this super computer has some proprietary patches that you will never get the hand on ?
Cool something to keep an eye on.
“Just tell me which distribution on linux this all computers are running. I am curious.
You haven’t in mind that the linux running on this super computer has some proprietary patches that you will never get the hand on ?”
Not now but soon…SOON…Muwuahahahaha!!!!
http://linux.slashdot.org/article.pl?sid=05/04/08/2148207&tid=98&ti…
I forgot to mention that I had tried ext3 and reiserf son Ubuntu and found ReiserFS to be more faster, could be because Ubuntu has it that way. I am not a know it all in this matter but one of my collegues was arguing that ReiserFS is nothing but ext3, had to show it to him. People still prefer ext3, is more acceptable for some reason.
You know, enterprise doesn’t care about your freedom, but they care about their money and the control of their IT. If the GPL 3.0 is coming soon, they will:
1) fork linux and use a version that come with GPL 2.0 and proprietary patch
and/or
2) choose another system (BSD system for example or invest in their own unix system if they own so)
“You know, enterprise doesn’t care about your freedom, but they care about their money and the control of their IT. If the GPL 3.0 is coming soon, they will:
1) fork linux and use a version that come with GPL 2.0 and proprietary patch
and/or
2) choose another system (BSD system for example or invest in their own unix system if they own so)”
And you Sir obviously don´t care much for having fun:) At least your nick suggests you have a sense of humor. Perhaps ComicStoreGuy would be more appropriate. “Worst post ever!”