Slashdot Mirror
FreeBSD 5.0-RC1 Now Available
Dan writes "Murray Stokely of FreeBSD release engineering team announces that they're one milestone closer with the immediate availability of FreeBSD first release candidate for the i386, alpha, sparc64, and ia64 platforms. ISO images and FTP installation directories are available now from the FreeBSD FTP site."
Look it moves... It must be alive!
To spoil your joke: yes, it runs Linux. (kldload linux ; ./sick-linux-binary ; rejoice! )
great, I just installed 4.7 last night...
...because those Linux folks started to get to culty--kinda like the Mac people.
Here's a direct link to the pertinent section. It details kernel, userland, and security updates that have gone into the 5.0 tree of FreeBSD.
If you celebrate Xmas, befriend me (538
BSD is dead?
BSD sux, Linux rocks!
"I switched to FreeBSD and was amazed..."
"I tried FreeBSD and it sucked..."
"Not to troll, but why should I use FreeBSD instead of Linux"
FreeBSD and SMP sucks!
"In Soviet Russia, the RC1 releases YOU!"
"BSD != DEAD"
So you se my friends, no need to post further! Thank you, come again.
Why do you think they call it a "release candidate"?
Programming can be fun again. Film at 11.
I don't know whether you're just a troll or unclear on basic concepts, but you're wrong on multiple counts. 802.11b cards are well-supported, and journalling is a band-aid for a filesystem. FreeBSD's filesystem is well-designed and doesn't need that band-aid. As for token ring, ehh...
Isn't it great how people can release things for hardware you wouldn't even know how to buy if you wanted to. I've often wondered how elements like the FreeBSD team and Linux get people interested in doing these things. Its not like an "itch you need to scratch" because you don't even have the body part to have the itch on!
An Eye for an Eye will make the whole world blind - Gandhi
For more information about the Linux ABI, see this manpage.
Beware, Nugget is watching... See?
SMPng: The next generation support for SMP machines (work in progress). There is now partial support for multiple processors to be running in the kernel at the same time.
KSE: Kernel Scheduled Entities allow a single process to have multiple kernel-level threads, similar to Scheduler Activations.
New architectures: Support for the sparc64 and ia64 architectures, in addition to the i386, pc98, and alpha.
GCC: The compiler toolchain is now based on GCC 3. X , rather than GCC 2.95. X .
MAC: Support for extensible, loadable Mandatory Access Control policies.
GEOM: A flexible framework for transformations of disk I/O requests. An experimental disk encryption facility has been developed based on GEOM.
FFS: The FFS filesystem now supports background fsck (8) operations (for faster crash recovery) and filesystem snapshots.
UFS2: A new UFS2 on-disk format has been added, which supports extended per-file attributes and larger file sizes.
Cardbus: Support for Cardbus devices.
Beware, Nugget is watching... See?
That is the funny part. VMWare runs courtesy of the Linux ABI. So you'd be running a Linux OS on a FreeBSD system using a Linux binary.
Beware, Nugget is watching... See?
Definitely not. Please stick with 4.x.x line for a while yet, and your satisfaction (from the POV of production use) will be greater.
Due to massive changes ans some binary incompatibilities, you should wait with using 5.x line for any serious production until x >= 1.
However, if you want to test it in a light and inconsequential scenario, by all means you should do it, to reduce the shock later on, and to play with some really cool stuff... Things like native kernel threads, GEOM disk abstraction layer, background fsck and a lot of others make 5.x line a distinct flavor...
Don't forget to brandelf it too, if said sick-linux-binary happens to have a broken elf header which says it's a Solaris binary or whatever, i.e:Since while FreeBSD will use the elf header to make things like Linux emulation work, Linux just ignores it, meaning a lot of tools like to produce incorrect headers. Tsk
A major improvement in FreeBSD 5.x over 4.x is the new modular init. Instead of one monolithic script (classical BSD) or several scripts in a symlink farm with manual sorting and dependency resolution (SysV / Debian, RedHat, SuSE...), it uses an internal automatic sorting and dependency resolution comparable to apt-get or modprobe on GNU/Linux. I would like to see mainstream adoption of this in the GNU/Linux world of this. To date, Gentoo Linux is the only distribution offering and supporting this excellent feature.
gopher://cramer.plaintext.cc http://cramer.plaintext.cc:70
Ofcourse I have 'kern.fallback_elf_brand=3' before even starting to think about running sick-linux-binary without brandelf ;)
02-12-09 14:33 BSD: FreeBSD 5.0-RC1 Now Available
My monitor sometimes thinks it is a crystal ball; using it I can predict future /. headlines. Here goes:
03-01-06 9:25 BSD: FreeBSD 5.0-RC2 Now Available
03-01-14 9:25 BSD: FreeBSD 5.0-RC3 Now Available
03-01-25 9:25 BSD: FreeBSD 5.0-RC4 Now Available
03-02-02 9:25 BSD: FreeBSD 5.0-RC5 Now Available
03-02-17 9:25 BSD: FreeBSD 5.0 Released
03-02-19 9:25 BSD: FreeBSD 5.0.1 Released
Funny thing though, apart from the different version numbers the discussion is always exactly the same...
Wenn ist das Nunstueck git und Slotermeyer? Ja! Beiherhund das Oder die Flipperwaldt gersput.
Vinum has not been mantained at a high level for some time and I have heard that there is a replacement in 5.0 that emulates the IBM AIX volume manager (which kicks ass in my opinion).
Any word on this?
Either give it away or get top dollar, but never sell yourself cheap.
As a note to the curious, the Linux compatibility improved "greatly" in the latter half of the 4.x series. It's more likely that one or two simple functions were fixed (I haven't followed closely, and it's been a while), but this means that previously tempermental software- the Amiga/Elate SDK, for one example- now runs flawlessly as of 4.5 or so.
;)
Loki games should be no problem, not that they were before.
I was going to say...not necessarily support I'd be looking for. Unless of course you're going to build a DNS server - you know, one server to rule them all, one server to BIND them...
I want to delete my account but Slashdot doesn't allow it.
I understand fully, and also agree. Getting VMware to work fully on FreeBSd is a kluge, especially if you want to do full screen mode. It just doesn't work. I've found that vmware1 works better than vmware2. Getting it networked has also proved difficult, though I've seen it done. One of the nasty parts of VMware is its use of that Linux /proc filesystem, which contains way more than just process information...
Overall, I think VMware is too expensive now. I'd rather see a VirtualPC for FreeBSD.
Beware, Nugget is watching... See?
I tried to get DP-2 working in Virtual PC but networking seemed to get stuck. Anyone have luck with getting FreeBSD to work in Virtual PC? Are there patches that need to be applied?
The next time I buy a computer, it will have a 64-bit processor. Since ia64 is what FreeBSD seems to be supporting, and I have never had a problem with Intel, that is probably what I will get. There doesn't seem to be any real wok done on AMD's 64-bit chip for FreeBSD. You can actually buy an ia64 currently, whereas AMD's is coming "real soon now." The last time I purchased an AMD (about ten years ago) it ran about half as fast as the supposedly equivalent Intel chip. I honestly don't know why so many people have such a love of AMD. I am glad that Intel has competition, though. The main reason why ia64 is having troubles is because not much supports it yet. Once FreeBSD gets a usable ia64 port, it will just be a matter of time for any other OS that wants it.
Best Slashdot comment ever
You can help by getting off your rear and writing to your congressman or senator. Tell them FreeBSD is important to you. Tell them that without FreeBSD, you would have to find less managable and intelligently designed alternatives. Let them know that this is an issue that effects YOU directly, that YOU vote, and that your vote will be influenced, indeed dependent, on his or her policy on FreeBSD.
You CAN make a difference. Don't treat voting as a right, treat it as a duty. Keep informed, keep your political representatives informed on how you feel. And, most importantly of all, vote.
KMSMA (WWBD?)
FreeBSD has grown larger and larger -- back in the 3.x days I could run it easily. The 4.x series have consumed much more memory, even when the kernel is compiled to use the same features. I had heard that one of the 5.x trees goals was to regain some of that "thin" nature which IMHO is one of FreeBSD's biggest draws. Anyone know how that is coming along?
MORTAR COMBAT!
Well, FreeBSD can run linux binaries. Observe:
/compat/linux
bash# uname -a
FreeBSD abox.some.dom 4.7-STABLE FreeBSD 4.7-STABLE: Sun Dec. 8 19:28:39 EDT 2002
root@abox.some.dom:/usr/obj/usr/src/sys/abox i386
bash-2.05b# chroot
%uname -a
Linux linuxbox.some.dom 2.4.2 FreeBSD 4.7-STABLE: Sun Dec. 8 19:28:39 EDT 2002
user@abox i386 unknown
So, does it run linux? Of course. It can also run SCO unix. "Does linux run FreeBSD or SCO?" is a better question. There's an effort to provide similar capabilities in linux, but it looks like they've just started. Help 'em out, ok?
Is there a release roadmap, guessing when 5.0 final and 5.1 will come out? I haven't seen it on the FreeBSD site.
You're quite confused, but I don't blame you.
4.4BSD was the last full release from the Computer
Science Research Group at UC Berkeley. I think it
was in 1994. FreeBSD and NetBSD were based in large
part on this code. (This is an oversimplification
but it's good enough.)
Mac OS X is based on NeXTStep, which includes BSD
code from 4.3BSD, which came before 4.4BSD. Mac OS
X was updated using FreeBSD 3.4 as a reference.
There was no wholesale integration of FreeBSD 3.4.
Mac OS X 10.2 was updated using FreeBSD 4.3 as a
reference, I believe. Again, no wholesale
integration. The same will be the case with
FreeBSD 5.
Ben "You have your mind on computers, it seems."
Somebody told a biologist that "BSD is stable" and
they drew the wrong conclusion.
Ben "You have your mind on computers, it seems."
I think, between our two comments, lies the truth.
Apple does use 4.4BSD subsystem elements for 10.2--it says so in their documentation and its man pages are peppered with this reference. The rest of your information does clarify any oversimplifications or other inaccuracies...thanks.
Vos teneo officium eram periculosus ut vos recipero is.
someone mod the parent up. An AC who is not smoking crack!
If you had nuts on your chin, would they be chin nuts?
Actually that means its not fault tolerant... :) You are the unstable one who ejects things without unmounting them.
The kernel does panic if you try to mount a CD that wasn't fixated though... or at least it did back in 4.6. That is the same kind of fault intolerance as the floppy eject thing I bet.
The arguments of a person with such a "flexible" version of the English language tend not to change my mind.
Best Slashdot comment ever
The base install should allow for boot scripts, minimal drivers, and system binaries. Try using the FreeBSD installer to install a system onto a 64 MB flash IDE drive, for instance -- you can't, because the base system takes up too much space. RAM really isn't the issue, although "whopping" 16 MB is fairly whopping considering the target audience.
MORTAR COMBAT!
Actually, I've had no problems with x.0-RELEASEs. We installed 3.0-RELEASE on our machines the day it was released. We were waiting for it because we needed support for our SCSI card. This was before I knew about -SNAPSHOTs. Anyway, we installed it and ended up running it for like two years without a reboot. I remember a few security issues that could be patched while the machine was running but I don't remember any showstopper stability issues or system corruption issues. In fact, in all the releases that I've installed since 2.something-really-low, I don't think I've ever seen an unstable or dangerous -RELEASE.
Darwin uses a monolithic kernel based on FreeBSD 4.4 and the OSF/mk Mach 3, combining BSD's POSIX support with the fine-grained multithreading and real-time performance of Mach.
The previous was correct in his post.
Now that FreeBSD has cardbus support can we expect Apple to grab some code and improve their cardbus support. While Apple's CardBus support does the basics there are many drivers that I have heard could not be written simply because the API support did not exist. Is this something Apple can grab. I'm not funny up to date on what Apple grabs from which various BSD projects. -Tim
Tim Smith - Ramblings from Nerd Land
The KSE facility will not, in all probability, be production-ready in time for the 5.0-RELEASE. See FreeBSD KSE Project page. For SMP, see FreeBSD SMP Project.
-- Sig down
This is true, but it does increase the probability greatly. Generally, whenever I have a problem with a company, I make a point not to buy from them again. I had an AMD 486DX4 100MHz and an Intel 486DX 66MHz, and the Intel chip was noticably faster, in practally identical systems otherwise. It isn't the slowness that annoyed me (it did cost less), but the fact that they claimed otherwise.
Best Slashdot comment ever
Much progress on SPARC64 since then.
The next chips out of Intel IA2 (or whatever) will be largely based on the Alpha chip from DEC-now-Compaq-now-HP. The Alpha is pretty good product victimized by absurdly stupid management/marketing/pick-what-you-like. So it's a worthy question. Why support a POS architecture when the successor, due out in another year (GA), is vastly different? It may be a lot of work that won't translate over well.
Computer Science is Applied Philosophy
> Real OS's shouldn't need cartoon character mascots.
Right! They should have pudgy guys dressed up in blue butterfly outfits!
Chris Mattern
The next chips out of Intel IA2 (or whatever) will be largely based on the Alpha chip. . .
Do you have any links to back that up? That would be the coolest thing in the world, but from what I've read recently HP is burying Alpha and PA-RISC in favor of Intel's Itanium, and that Intel has some of the technology from Alpha but apparently can't just take off with a new line of Alphas by themselves.
...and both of which implement the IA-64 architecture. The FreeBSD port isn't to Itanium I, it's to IA-64, so it should be able to work on Itanium II as well (although there may be work needed if Itanium I and II different in any ways not covered by the IA-64 architecture spec that matter to the OS).
They may use similar implementation techniques to ones used in various Alphas (there's no such thing as "the Alpha chip", there are multiple Alpha chipsinstruction set architecture will be dropped in favor of an ISA similar to Alpha.
I have seen nothing to indicate that the successor will be "vastly different" in its instruction-set architecture, so that work done to port to Itanium I-based machines "won't translate over very well" for Itanium II-based machines.
FreeBSD 4.5 works fine with both my mouse and keyboard so I know its a bug.
http://saveie6.com/
VMware uses /proc (or rather, /usr/compat/linux/proc ;-)
/proc.
/proc from tape, rendering it totally useless for me as a file system. But then again, I'm not a plan 9 fan either, which probably makes me a heretic in many UNIX users eyes.
This is Evil, I quite agree. But from what little research I've done, even a getppid() call on Linux seems to involve opening
I only wished mount had an option to make a file system visible under emulation only.
Last I checked, I was unable to restore
Native VMware support for FreeBSD is when I unzip my purse again. I have a hard time believing it'll take more than a day or two for a VMware engineer to fix up the fallout from a "make World" on FreeBSD. Oh well. I think too many FreeBSD users overestimate the engineer/marketer ratio at VMware, and I believe they'll have a hard time getting an engineer off his proverbial to do such a port, and train the support staff ("look, when you tell the user to type "uname -a" and he mentiones FreeBSD, go to page 5 of your cheat sheet"). I'm only half joking there; educating the support staff is an important job, and while I feel VMware support is less than stellar, I challenge any commercial operation to do a better job (or Plex86 to come up with a better Open Source equivalent, FWIW, and I sure lack the time to assist there).
Bert Driehuis -- All I asked was a friggin' rotatin' chair. Throw me a bone here, people.
Right now there's no reason at all for me to switch.
Hey, if Linux does the job for you, why switch? One mans elitist stance is another mans sign of quality. And vice versa. I think that if you look around in the Linux world, you'll find that Debian is much closer to FreeBSD, than Redhat is. If you look around in the BSD world, you'll find that FreeBSD is much closer to Redhat than NetBSD is.
All are excellent OSes. If it were anything near practical, I'd be multibooting Linux for productivity apps, FreeBSD for server development, NetBSD for kernel development, Debian for server deployment, OpenBSD for security critical stuff, Win98 for games and Win2k for Windows support. All of them tasks I perform at times. Stuck with limited disk space and the annoyance of reboots, I use FreeBSD for work and Win98 for games. And I payed the Microsoft tax for the games. So sue me.
As Opus so eloquently put it, "to each his dentifrice".
Bert Driehuis -- All I asked was a friggin' rotatin' chair. Throw me a bone here, people.
(I'm not aware of any pthreads implementation being of the "npw" variety, so I changed the subject to match what I think the original poster intended -- but then again, the _np suffix has bit me more than once in the past, so there very well might be a Non Portable Windows standard by now, pardon my ignorance).
Or you asking about kernel threads?
I still see pthreads as a programming convenience, and as such, FreeBSD pthreads has served me very well.
Once you get to serious pthreads programming, all but a few commercial implementations fall flat on their face. Needless to say, to support those Serious Programming efforts, those commercial implementations generally do not rank highly on performance, as all that multi-CPU stuff more often than not eats CPU time in spinlocks, and most apps that on the surface could do with multiple CPU's turn out to be disk bound in the first place.
It is so rare that I see apps that actually would benefit from multiple CPU's that I'm consistently stunned to see this issue receiving attention from folks who are not doing fluid dynamics or some other highly parallelizable task.
Bert Driehuis -- All I asked was a friggin' rotatin' chair. Throw me a bone here, people.
If you're happy with Linux, why not stick to it?
I wouldn't dream of running my FreeBSD boxes with unproven drivers. I hate being stuck with a DVI Geforce4 card driving a DVI LCD (even though I have reason to believe the Xfree86 support on Linux would suck as badly). If weird device support were important to me, I'd be running a flavor of Linux by now.
And I hate all of the parochialism as much as you do. Trust me, I shed many a tear over the unavailability of a native FreeBSD port of VMware. But I cope with it because FreeBSD serves me better in other respects (and most of them immaterial -- my main dislike for Linux stems from the default colorized ls in Linux 0.something, back in the days when the whole install fit on te floppies).
Bert Driehuis -- All I asked was a friggin' rotatin' chair. Throw me a bone here, people.
Can you point me to an example?
Ben "You have your mind on computers, it seems."
FreeBSD 5 will have to show its mettle before Apple officially integrates the 5.0BSD subsystem in a later OS X update.
Are you confusing the "4.4BSD" that gets thrown around a lot, with FreeBSD 4.4?
Because 4.4BSD does not mean FreeBSD 4.4, it usually refers to the last official release of the Berkeley Software Distribution.
From this: The UNIX system family tree: Research and BSD
we can see that Darwin is made up in parts from 4.4BSD Lite2 for Rhapsody, FreeBSD 3.2 and NetBSD 1.4 for Darwin/Mac OSX 10.0, then updated with FreeBSD 4.4 for OSX 10.2.
Looking at the family tree, you can see that the current versions of FreeBSD, OpenBSD, NetBSD, BSDI and Darwin are all 4.4BSD Lite2 based.
There will probably never be a 5.0BSD, unless Berkeley picks BSD up again (and does a lot with it), since the legacy of 4.4BSD is mostly a foundation now.
War crimes, torture, lies, illegal spying... Would someone give Bush a blowjob, already, so he can be impeached?
BTW, there are some pretty excellent PDF files here that have a nice representation of UNIX history.
War crimes, torture, lies, illegal spying... Would someone give Bush a blowjob, already, so he can be impeached?
Hard to use? Jesus. The reason I use FreeBSD is because it is EASY to use. Linux drives me fucking batty every time I try to use it. Get some sort of bloody standard for system upgrades and I might use Linux. Get some sort of standard package system and I might use Linux. Get some sort of standard startup scripts and I might use Linux.
Frankly, if you had a hard time using FreeBSD, then noone showed you how to use the system and that is a shame. If you spent more than a few days with a working FreeBSD system and understood how it worked, I doubt you would ever go back to Linux.
-sirket
It might be a personal dislike
It is a personal dislike, and not even fully rational. I just don't like it.
nicely tuned colourisation in ls does help you navigate and perform your tasks easier
I never saw a nicely tuned color config. As a matter of fact, the default ls color scheme, and the default VIM color scheme hurt my eyes to the point of distracting me from my work.
Color is overused and underutilized. IMHO.
Bert Driehuis -- All I asked was a friggin' rotatin' chair. Throw me a bone here, people.
You got me. This was told to me by HP/Compaq sales folks at a presentation. They're trying to get into the company where I work and we're talking about their migration plans. I'll look and see if I have any soft docs...
Computer Science is Applied Philosophy
It was made to sound that the only real similarity will be in name. This says somethign drastic to me, I could be mistaken
Computer Science is Applied Philosophy
And there's probably not much (if anything) left of the P6 core (Pentium Pro, Pentium II, Pentium III, many Celerons) in the Pentium IV, and not much left of the P5 core (Pentium, Pentium MMX) in the P6 core - but they all implement the x86^H^H^HIA-32 instruction set, albeit with various additions over time (MMX, SSE, and assorted other stuff such as conditional moves and a 64-bit compare and exchange).
The Itanium 2 implementation of the IA-64 instruction set might not share much with the original Itanium implementation IA-64 instruction set, but that doesn't mean that they don't implement the same core instruction set.
In fact, the Intel FAQ on Itanium 2 explicitly says:
(emphasis mine).
FreeBSD is not just a kernel. This isn't some tarball you get from bsdkernel.org, compile and reboot with. It's a complete OS that includes the kernel and the userland together. The userland components are developed with and for the kernel by one group of people. That userland has thus evolved alongside the changing kernel in the -CURRENT CVS trunk, diverging greatly from the time that 4.0 was tagged -STABLE. When you upgrade, you get it all.
I'm sure you meant no harm, but unless corrected, the misunderstanding of the nature of the BSDs, and how they differ from Linux, will continue to be propogated.
--
My comments and opinions completely reflect those of anyone and anything I am remotely associated with.
A complete operating system based on 4.4BSD.
FreeBSD's distinguished roots derive from the latest BSD software releases from the Computer Systems Research Group at the University of California, Berkeley. The book The Design and Implementation of 4.4BSD Operating System, written by the 4.4BSD system architects, thus describes much of FreeBSD's core functionality in detail.
Drawing on the skills and experience of a diverse and world-wide group of volunteer developers, the FreeBSD Project has worked to extend the feature set of the 4.4BSD operating system in many ways, striving constantly to make each new release of the OS more stable, faster and containing new functionality driven by user requests.
MacOSX's BSD subsystem was never based on 3.3BSD, but on FreeBSD 3.3 (an important distinction). OSX 10.2's BSD subsystem was upgraded to a FreeBSD 4.4 base.
Therefore, to say that 'Apple does use 4.4BSD subsystem elements for 10.2' is a true statement, but this has not changed from 10.1 or 10.0 (or the Public Beta, etc), since all versions have been based on FreeBSD, and all versions of FreeBSD are still a "4.4BSD based" system.
If you reread your initial post, this was not what you claimed, and thus the propogation of misinformation continues...
--
My comments and opinions completely reflect those of anyone and anything I am remotely associated with.
No shit. I was referrering to the guy who corrected the guy who thought OS X had large bits of 3.3BSD. SIgh, anonymous cowards....
I am not exactly sure what your problem is with mergemaster, but it is a pretty simple tool. If you do not like it, though, you can write your own merge utility like I, and may other people, have done. I can not remember the last time I ran mergemaster.
I like having the entire source code for my system on the box. If I want to work on any part of my system, the code is right there and can be built en masse, or piece-meal as I see fit. If BIND breaks (as usual) I can simply jump into the BIND directory and make install the new code without rebuilding the rest of the system.
I also prefer the ports tree to apt. That is a personal preference. The ports tree is a simple and logically laid out method for installing software. The make files are easy to read and understand, and you can figure out what they are doing. apt, well, just is not as simple in my opinion.
I ran Debian on all of my computers for about a year and a half back in the 2.0.x days (before apt really caught on) and it was just too much work. I try to go back to it about once a year or so and just can not do it. Nothing in the system is laid out the way I expect it to be laid out, and to this day, kernel configuration and compiles are a pain in the ass (again a personal opinion).
When all is said and done, Linux can't offer me a good reason to switch back yet. Saying it is "as good" as FreeBSD is hardly a compelling reason. My FreeBSD boxes are so stable I forget they are running sometimes. A Redhat box that I installed about a month ago kept crashing due to a problem with journald (usually at about 4am). I installed FreeBSD on the box, went home, and slept very soundly.
"make buildworld && make installworld && make buildkernel KERNCONF=kernel && make installkernel KERNCONF=kernel" is just too easy.
-sirket
As far as your other complaints about file structure, it's a matter of preference, I know where the program configs are (/etc/someserver/) and the documentation (/usr/share/doc/someprogram/).
And yes, sometimes I forget about my Debian servers running at work.
"I keep looking in the want-ads under 'revolutionary' but there don't seem to be any listings.. "