Slashdot Mirror
OpenBSD SMP In The Works
Cajal writes "Four students at the University of Waterloo are working to add SMP support to OpenBSD as part of the Spinlocks project. More information is available in a story
at the OpenBSD Journal's site. They expect to have an initial working MP kernel in January."
..is that I can never decide to buy a CD set because everytime I think wait one release, the next one will have new feature xxx included!! (Where xxx is some new pf feature, or systrace, or SMP, or....
In related news, the Egyptians are on the cusp of discovering Construction, which will allow them to build Aqueduct and Coliseum. However, this is not expected to improve the odds of their feared Chariot against invading Mechanized Infantry.
But most other OS have had it for a while now. Why has BSD taken so long?
And it makes for a good research project as well.
But I ask here, as an honest interested person, why one would wait until SMP is correctly and efficiently implemented into OpenBSD when they could simply use any old recent version of Windows or Linux on SMP hardware to get symmetric multiprocessor support for a high-load server?
I understand that Research -> Products -> Corporate $$$, but is this perhaps too little too late for OpenBSD?
If you celebrate Xmas, befriend me (538
They can increase their userbase to (hopefully) include some of the larger companies and corporations that opt to use FreeBSD because of it's SMP support and greater performance than OpenBSD. Hopefully, this modification will also include some performance modifications so it can also compete speedwise with FreeBSD.
While security takes precedence over performance in my book, there are definitely some things that need the performance of FreeBSD.
This is one feature i've been looking forward to playing with (not NEEDING) for a while, i can't wait to try it when it's available somewhere.
--Fuzz
No, I would have assumed *BSD all had SMP support too, after all, Linux and Windows support SMP... strange
Morphing Software
Sweet now I can have more than one fish in the box.
The human condition is to not accept the human condition.
aww come on man, if you're gonna say it at least say it right, It's Soviet Russia
The last time I spoke to Theo in person, he wasn't too keen on SMP. That wasn't too long ago.
SMP tanks roll over YOU!
The University of Waterloo, eh? Well, knowing them, the versions of OpenBSD with SMP support will require a Windows XP activation key...
Or maybe they figured out a way to port OpenBSd to Windows. Or something. Waterloo?
Lack of eloquence does not denote lack of intelligence, though they often coincide.
We want the option of SMP on SUN kit as well. Seriously SMP is well needed seeing as most OpenBSD box's are yesterdays coperate kit. Also given intels SMP on a chip play along with Power4 dual cores on the market now that by the time they make it into your average geeks home SMP should be stable.
You've apparently never asked Theo about this. I have several times, going back years ago, and his reply has always been either "nobody wants it", or "SMP machines are too expensive and nobody has them". And OBTW, he wasn't interested in loaner or donated hardware to do the work on.
I guess somebody cares now....
The problem is, I don't know how much of OpenBSD's kernel really relies on the assumption that it'll only ever run on one CPU and it may take some time before OpenBSD becomes as stable and secure as it needs to be.
OpenBSD is designed for an "edge server" environment, where scalability isn't as important as security.
If I'm not mistaken, OpenBSD has had very strong IPv6 support for years.
Microsoft has been raiding the University of Waterloo for programmers for years now.
-------
Warning: Slashdot may contain traces of nuts.
Those who don't use Linux are doomed to reimplement it... again.
and for a limited time the ekrout combo is available for just $3.99 and may be super-sized for a meager 39 cents extra. standard meal includes medium size fountain soda and large fries.
He's a fag, too.
God hates fags.
However, the crack about ipv6 is stupid. OpenBSD has had a working ipv6 stack for a long time (they were first OS to ship with an IPsec stack at all). Get your facts straight.
Yes... I suppose it IS a troll...
They have ipv6. OpenBSD ships with ipv6 active and operating. They've been looking at/working on SMP for some time, but they (read: Theo) wants to make sure it meets the standards of the rest of the OS. SMP adds quite a few (theoretically at least) security problems to deal with, and they want to be sure those problems are fully addressed.
'Sensible' is a curse word.
I've been using OpenBSD in several mission critical networking roles for 3 years now, and I can safely say that I haven't needed SMP.
The conventional wisdom that an operating system should be judged according to it's bells and whistles is what's wrong with the software industry. An OS should be judged by two things: Does it do the job I require of it, and does it do it well?
There are many many jobs that do not require SMP. There are many many jobs being done on SMP boxes that do not require SMP. As the price of processors has diminished, SMP is just a cool thing to buy. I'd be willing to put money down saying that 75% of the SMP boxes out there aren't needed (if that was measurable).
So, if you want to judge your OS based on features you don't need, then go for it. I use OpenBSD because it is the best choice for that particular need. If you want to assume that one OS is the Uber-OS because of the back panel of the box, then go for it. I'll assume a particular OS is best for the task at hand, and go with that.
I'm not part of the OpenBSD project (nor do I play one on TV), but one of the central points behind it is that they don't put in things unless they are needed. So far it doesn't seem like SMP has been justified in the great scheme of things (no surprise given the actual need in the wild). I'd much rather have them working on things I'm going to be using instead of evaluating other products based on things I won't.
God loves fags
Man who spin locks safe cracker.
Still doesn't. Most BSD's don't.
You mean Civilization.
Suck it, Trebek!
Did anyone else notice that these four students are using PowerBooks (I assume running OS X). Check out this picture. You also have to love the reference to the cult movie Hackers.
Actually OpenBSD has had ipv6 for some time now. Kind of lame to have ipv6 (something that has yet to be implemented at any reasonable amount) before SMP (something required for servers doing just about anything but firewalling in a corp environment).
Can I get an eye poke?
Dog House Forum
This should help OpenBSD to die scalably.
Um.... OpenBSD has had ipv6 since May 19, 1999: http://www.openbsd.org/plus25.html
And even without SMP support, you have always been able to run OpenBSD on a dual-cpu x86. I've been running it on my dual p2-233 for two years now. Sure it only uses one processor, but for the time I want to put in to configuring it, I figure a secure box on one proc is better than an unsecure box on two.
And yes, I have checked out the other BSD projects. After trying linux for a year, I got sick of the bloat, and wanted to get as lean as possible. In my opinion, at the time OpenBSD (2.9, I think) was the cleanest.
"Dying tickles!" -- Ralph Wiggum
I see tons of negativity from the clueless as usual. *sigh*
:P)
Hey, people? Do you know how many people *don't* need SMP, as opposed to those who do? Did you ever think that, given the number of SMP-supporting *nixes out there, OpenBSD felt like concentrating on more important things (like security) first?
Yeah.
Congrats, OpenBSD! (Still the only operating system to pit Daemonettes against Catgirls in pits of pudding! Hehehe. Get it? Daemonettes? Catgirls? Script Kitties? HEHEHEHEHE. Sorry.
I think its great as a school project, but that seems to be all that it is. It would be great if others gave a helping hand, and made sure the project gets "finished"(mostly working anyways) "These guys have an interesting 4th year project for CE" would be more better, or at least repel more Trolls
Smp don't have a lot of meaning because almost nobody has many processors. But people are always interested in more features. I guess if everyone buys a hyper threaded pentium 4, then smp support is essential.
Troll.
OpenBSD has had IPv6 since version 2.7 out in June 2000.
And for the record, FreeBSD has had IPv6 since March, 2000, version 4.0
And let's not forget who brought you OpenSSH.
SMP isn't the top priority. Giving up stability and security for the latest and greatest features are not what everyone wants. A friend of mine complained about FreeBSD not having good SMP support, I asked him if he had an SMP machine, he said "No." I hope that is enough to illustrate my point.
Sorry to go off on this, but mod the parent down if you mod me down please. People always trounce on any of the BSD's while praising Linux here.
Hear that? That's my karma in the toilet. Flush.
"Weapons should be hardy rather than decorative" - Miyamoto Musashi
I think that goes for OS's too
Well, don't they? Forwards and backwards that means the same thing essentially. :-)
Did you expect something more from a dying OS?
OpenBSD is a very promising OS, and SMP support will finally let it play with the big boys in the free *nix playground :)
"two things: Does it do the job I require of it, and does it do it well?" In every engineering project there's cost. You can choose two of the three: Better, Faster, Cheaper. You may have forgotten this, but the accountants won't
http://en.wikipedia.org/wiki/Signature_bloc
Yes, one(1!) remote hole in the default install, in more than 7 years.
If Microsoft could boast with this they would spend the equivalent of a small country's GNP on letting you damned well know about it.
If Linux could boast with this then Microsoft wouldn't have that small country's GNP to spend, since everybody would be running Linux instead.
If you by "always behind" mean that Doom 3 can't be run on OBSD then yes, you are correct on that too.
OpenBSD was designed to be a secure server, even with default install. Adding SMP-support have never been a priority in part, I think, because of the inherent security issues that arise from such a implementation. I'm sure there are other reasons as well, none of them beeing that they never got around to implementing it.
So are we agreed then, OpenBSD is always behind, and unsecure?
I thought so. =)
"Education is a system of induced ignorance" / Noam Chomsky
Which of the popular applications and uses of OpenBSD would benefit from SMP support?
--about time. Before I just didn't understand. I keep reading about open bsd being a server distro, very secure. Well, duh, servers are where you see multiple processors more than cheap(er) desktops. and if all it is is to be some sort of minimum home brewed gateway router thing, they can stop now, it's "done", go on to some other project.
All in all, though, I'd say adding multiple processor support is a good thing. I wish them well, and perhaps I'll try it someday once this is more stable.
Hi.
Is there any reason you're posting in monospace, other than to be an eyesore?
you are a moron. just how egotistical are you? stick to linux - unix for tools.
It's strange how things like this end up changing would would have been. Do it right the first time, because if it gets adopted, and it wasn't done right, efforts will be diluted.
I'm glad to see it's happening though. At least somebody's throwing some brainpower at it rather than waiting around for Theo & friends. (no fault to Theo, I know SMP is "in the works" - OpenBSD is secure, first and foremost. That's what I, and many others, care about most. Kudos to you and your team on this! You have a highly-regarded, ultra secure OS that has kept many cracker-types and script-kiddies at bay for many years. You have saved many people many thousands or millions of dollars with the protection your software project has provided. You have given nothing to the headache medicine providers of the IT industry.)
One more processor for my dual-capable Sun SS20 and I'll have a grand-ole time playing with this. Just too bad it comes with only a single 10-speed ethernet port. Anybody know about S-bus fast ethernet cards?
To these brave deveopers: Way to go! Thanks for getting the ball rolling and best of luck with your project (and dealing with the publicity! :)
Be careful what you say there. Most BSDs? What are you classifying as BSD? Most commercial unix systems like solaris are based on BSD.
I assume of course you are talking about the free BSD operating systems. Considering there are only three, I guess two of three can constitute most. But FreeBSD has had SMP for a while already, although until 5.x it hasn't been very good.
On the comercial side of things, most BSDs *do* have SMP. And very good SMP as well.
Why isn't this posted in the BeOS section?
I'll bite, because it has nothing to do with BeOS.
Oops.
Sure its cool and isnt a bad thing, but for the target market that OBSD has, is SMP really *that* important?
---- Booth was a patriot ----
The Kernel spinlocks you!
i would also like to add that if you are in any way technically inclined you can build a Linux From Scratch ( google for that ) and make it as small as you wanted. :) hell, even layout the directories as you like, GNU/Linux is kewl like that.
he wasn't interested in loaner or donated hardware to do the work on
Hmm. now why didn't he say "Because SMP is insanely hard to do right and I'm not motivated enough or smart enough to get it done without making the system even more unstable. And besides we get more publicity if we encrypt swap"?
The OpenBSD project seems filled with people who need to prove their masculinity and grab attention rather than people who want to make sound design decisions. And in addition to making technical decisions, they also follow suit politically. For example, oh, such as holding an OpenSSH vulnerability hostage to promote adoption of their privilege separation patch.
OpenBSD has plenty of stability problems. See Dan Bernstein's downtime reports. And Bernstein is someone who wants to believe in OpenBSD but is losing patience.
OpenBSD receives plenty of attention thanks to its loudmouthed egomaniacs. Fortunately, FreeBSD is a better, more rationally designed system. Of course the FreeBSD developers are a little resentful of Linux's attention and the GPL and have made system changes out of spite, but it's nothing as bad as what OpenBSD does on a regular basis.
I really did want to like OpenBSD. Don't believe the hype.
Ok, it's oversimplification time. As I understand it, the three main flavors of BSD and their foci are:
Free - Well-rounded BSD for popular architectures.
Open - Ultrasecure BSD for many architectures. Lots of code auditing, but always just behind Free in some area or another.
Net - Runs everywhere. Won't be done until it runs on toasters and wristwatches.
I have much less experience with the BSDs than I do with the various GNU/Linux distros, so I hope someone will answer my question rather than flame in response. I've long been an advocate of reducing the number of different and nearly equal (in functionality) ways to do the same thing, regarding what I would call redundant software projects. In my view, the necessity of competition in a market for physical goods (the need to keep costs down and quality up by preventing monopolies) does not exist with OSS. If a company producing OSS decides to raise prices or slow development or include unpopular features, anyone else is free to keep using older versions at the very least or fork the project and continue development with positive goals at best. With the necessity for competition removed, a market containing multiple, redundant, competing (for mind share) OSS projects is inferior to a market containing a single, popular OSS project that satisfies a specific need. Incompatibilites crop up. The support base is divided. Developer time is divided. Skills must be learned twice. I would greatly prefer a single desktop environment (and widget set and cut 'n' paste mechanism) over the current situation. In Erpo's-the-emperor land, there is only one gnu/linux distro.
Remember, these are just my opinions.
My question is this: while there are a number of idealogical, license-, or ego-related reasons why maintainers of gnu/linux distros (or desktop environments, or whatever) would resist a merging of sofware or elimination of obviously inferior options (obviously inferior in the "there's 1001 gnutella clients and 99% of them suck" sense), is there the same kind of resistance in the BSD community to merging all three main flavors?
But.. OpenBSD is more secure! Just try cracking
my OpenBSD based webserver at 198.247.175.96
Very true, I could do this.
So why not? Because I don't actively run that computer. I used to muck around with it for C++-on-UNIX development, php development, and in general learning UNIX. Now I just use it to encode CDs to ogg (and this one time to build a library for a friend).
My main desktop computer is a WinXP box. I haven't found a UNIX-based OS that suits my needs as a desktop OS (I'd purchase a PowerBook today if I had an extra $3088).
Why wasn't Linux From Scratch on my list of things to try out when I was tyring things out? Simply because I didn't know about it. I got turned on to OpenBSD by a friend, tried the others, but stuck with OpenBSD. I ran 2.9 for a long time, upgraded to 3.0 for some feature that 2.9 lacked, and have stuck with it.
I do everything in monotype. You are not the first to complain. I find it easier to read. My only wish is that it be sans-serif.
"Dying tickles!" -- Ralph Wiggum
In addition to just making sure the kernel has it right, you've got to make sure that it's not going to open up any holes in the userspace too.
And 640k ought to be enough for anybody.
I want to delete my account but Slashdot doesn't allow it.
Please take note that this announcement didn't come from OpenBSD.org. The guys doing it as just doing it as their own project (which is neat). One of the main reasons people enjoy OpenBSD is that it's code has been audited by Theo and his folks and prepared to be the most security oriented distro around.
http://www.remix.net/
Hmmmm....last I heard it was something like "Only multiple remote kernel holes in the default install, in more than 7 years!"
You're still posting in monospace.
It still sucks.
Since there is no modifier to allow me to have monospaced posts marked -2, fuckwit, I'm asking you nicely, please stop.
And, as always, the experience of one person is merely anecdotal (definition:Based on casual observations or indications rather than rigorous or scientific analysis).
I thought BSD was dead? :/
I find it funny that all of these wonderful features are being added to a system that won't boot beyond 8G.
I've been having my own personal hell getting grub or lilo to boot this system on a second HD. My only success so far is to use the entire disk and then chainload it. Suggestions are welcome.
This would be an extension of that... one version of unix reimplementing something that is already in linux.
But linux has already done both of those! Get a new goal!
-- 'The' Lord and Master Bitman On High, Master Of All
And thus you put your foot in your mouth as you explain the exact reasons why IPv6 came first while at the same time question why SMP is late--OBSD has a different focus than your typical lame-brained OS.
Hmm ... two years up here, never seen a crash, and I run OBSD on 3 different architectures. Only rebooted for things like kernel patches. He probably has some subtly incompatible or buggy hardware -- such is life. OpenBSD is pretty much just as stable as Linux or other BSDs, but hardware support isn't always as expansive or broadly tested as, say, Linux (exactly as you would expect with only a fraction of the users).
I don't see how you came to that conclusion from my post. I was stating that things should be implemented in an order that makes the kernel most useful. I mean shit, even NT4 had SMP, which came out 7 years ago. OpenBSD is a great OS, but there are some seriously weird decisions being made on how things are implemented.
I mean ipv6 will be in use probably never now that nat has caught on. Now there are ugly hacks in the works for dns and nat to allow for two way communication.
What's the point of having the most secure OS out there if you can't use it for things that need to be secure because you are limited to one proc. Anyone know of any high traffic dynamic web sites that use OpenBSD? Ever wonder why?
Can I get an eye poke?
Dog House Forum
>On the comercial side of things, most BSDs *do* have SMP. And very good SMP as well.
Actually, that's crap. Most commercial Unixes are based on SVR4, not BSD. BSDi is (was?) probably the only close-to-mainstream commercial BSD.
It starts off sane, then wanders into lunacy. Brilliant joke or just a troll... I can't decide.
Thank you for your valuable opinion Adolf.
But.. OpenBSD is more secure! Just try cracking my OpenBSD based webserver at 198.247.175.96 try cracking mine at 127.0.0.1
am I the only one that's surprised that it didn't have this before?
Don't be suprised. OpenBSD's goals and claims to fame are stability, security and correctness. I donated a dual i386 mobo some time ago to them for the SMP project. I knew SMP on OpenBSD would be a long time coming, but also knew that when it came it would be well tested and solid.
If you want support for the latest whiz-bang crud to work on your desktop (at any cost) then install the Linux kernel du jour and whatever leet distro is hot this month.
Trolling is a art,
That's one example, but his are better documented than mine which is why I didn't share mine.
Informally, I've run OpenBSD, FreeBSD, and Linux on the same exact machine and had OpenBSD crash for no reason whereas FreeBSD and Linux run fine. *shrug*
I understand OpenBSD may have less eyeballs but that still doesn't make it acceptable to me. Especially not so when I consider all of the other reasons plus the alternative systems which are just as good if not better in terms of stability, support, etc.
Isnt OpenBSD already sponsored in part by DARPA and AFRL (air force research labratory)?
OpenBSD was designed to be a secure server, even with default install.
There's no 'even' about it being secure in the default install. There are other OSes that are pretty secure in their default install as well. That state, in most modern freenixes, is that most services aren't running, most ports aren't open by default.
So it's MORE likely that boxes with the 'default install' are secure. When the person who did the install starts adding services, configuring it, etc. it's likely it will become by definition somewhat less secure.
I'm probably going to get modded down for this, but it's got to be said (and this is from experience; it's not mindless cheerleading) --
This isn't the best use of OpenBSD developer time, if what they want is a better OpenBSD. Despite its good (but not perfect) security track record, this is an operating system that is riddled with mysterious problems when it comes to Unix compatibility.
I've got a bunch of source code that builds and runs fine on FreeBSD, Solaris, Linux, and a bunch of other Unices, but OpenBSD introduces reliability problems and some serious performance problems. (The same computer processes the same data with the same program at about 20% of the speed of Linux.)
SMP is a nice thing to have, but before OpenBSD can really be seriously considered for production use on any but the most trivial tasks, the reliability and performance issues need to be ironed out first.
Tired of FB/Google censorship? Visit UNCENSORED!
Date: Tue, 10 Dec 2002 00:11:57 -0500 (EST)
From: Richy Kim
To: Will Ellis-Adams
Cc: smp@openbsd.org
Subject: Re: spinlocks
ay papi,
I can speak for most of the group that our intentions were never as lofty as they appear on deadly or slashdot.
This is a 4th year design project. It was an opportunity to couple our academics requirements and our personal interests. But the goal, first and foremost, is to graduate -- on time.
Getting our code merged back is a trifling priority, if even.
And unfortunately, as academic projects go, we're restricted in contributors outside our group.
But thanks for encouragement, we will keep in touch with the list as we progress.
-richy kim
ha. i don't think you get it. sure one can install an OS with nothing running on it and nothing listening, but that is plain useless. the idea is to make the box useful, which means that services and applications can be enabled or turned on, knowing that that "default" code base has been audited. the services and applications that are included by default have gone through this audit process.
I've yet to see a real-world business application with even a 25% speedup with the addition of a second processor. SMP is expensive and overrated for all but certain "embarassingly parallel" computational tasks.
I worked for an "internet incubator" that had huge farms of 8 and 4 way Compaq servers running MS-SQL Server 7. One CPU would be pegged at near 100%, while the other processors on the machine would be near idle. With Oracle, the situation was some better, but again less than 25% speedup for 2nd processor, and less than 25% of 25% for 3rd, etc.
Maybe YOU need to grow up, and actually test the benefits of SMP rather than coo and goo-goo over it like some shiny toy.
Someone please tell me where it says on the goals page that SMP is a high priority? .. there can be no question as to the obvious reality for the official distribution in the forseeable future;-)
When you do security as a priority, fancy features that support threaded (read complex and untrustable from a security perspective) applications just don't quite get a front seat.
I do wish those dudes the best of successes, perhaps it will get merged in after the posix realtime extensions from rtmx, in the best case.
I sympathize with many who think they want SMP, but when the choice is security, stability, SMP, pick two
Todd Fries
Fine, but why do you pay for two processors if you only use one? Of course, you can run OpenBSD on an SMP machine, but if I'd like to run a non-SMP-enabled OS, I would certainly not buy an SMP machine.
That is a very good and valid point.
I purchased the machine off of a friend, equiped it with a few spare hard drives, ram, and 2nd processor, and initially instaled linux 2.4 on it. Later, after concluding that there must be something better than linux/XFree86 became an OpenBSD box.
Of course, it still runs XFree86 (it really is a shame what happened to Be), but it also runs about 90% less stuff than linux.
I've considered helping out the OpenBSD team by contributing to the SMP code, but I just don't have the background, time, energy or interest right now.
"Dying tickles!" -- Ralph Wiggum
Pick ONE.
Different approaches, different world views. It's hard to tell which is better (and why) *after* it's been done, much less *before* it's been done.
I don't think the duplication of effort is all that wasteful. Image the state of Linux and FreeBSD security if OpenBSD did not exist.
Most commercial unix systems like solaris are based on BSD.
Wrong. SunOS 4.x (renamed posthumously to Solaris 1.x I believe, by the Sun Marketroids) was a BSD based UNIX. Sun bit the bullet and shifted to SysV for Solaris 2.x, hence the renaming and major version increment (though IIRC SunOS 4.x was never called Solaris 1.x until Solaris 2.x came out - go figure). The Sun marketroids consequently spat on everything BSD based explaining how the way ahead was SysV. Before that, of course, they spat on SysV explaining how superior BSD was. Marketroids, you see. The other big players (AIX, HP-UX, and, I believe, Tru64 or whatever they call it now) are also SysV based. This was no doubt one of the compelling reasons for Sun's switch, at the time. Which brings me to my point - please name come commercial BSD implementations that have "very good" [sic] SMP.
Try NetBSD... safe,straightforward,useful.
I broke in and deleted /usr
.....
hahahahahaha
no wait
There are places where the networks are not touching,and there are places where they are-Boeing's Lori Gunter
I agree about the IP6 thing. Its far too complex compared to IP4 which makes programmers and network admins lives a total pain plus as you say NAT solves the main issue that IP6 was created to address (excuse the pun), ie running out of IP addresses. Having had the misfortune to try and set up IP6 I can safely say it should be back in the lab being sorted out , it sure isn't ready for prime time yet unless you have a Phd in network engineering.
these work good if you imagine rick moranis as Dark helmet reading them.
Rubbish. If your Compaq servers really behaved that way it's because something else was starving performance (RAM? I/O?). Indeed throwing a dozen CPUs at an I/O bound app will show little or no difference, but throwing CPUs at a CPU bound database improves performance dramatically.
OpenBFD.
PegQuin--I've got a sneakin' suspicion
Compaq Tru64 UNIX was once called Digital UNIX which was once called DEC OSF/1. The Open Software Foundation (OSF) was created, in some significant part, to counter AT&T's control of the UNIX market: the OSF wanted to prevent an AT&T/Sun hegemony. The OSF code base is very much rooted in BSD (and also includes Mach microkernel code [also used in Apple's Mac OS X]).
/bin/ps understanding both BSD and Sys V options.
So, Compaq Tru64 UNIX is a very good SMP BSD implementation.
Tru64 does a good pretty good job of being bilingual with basic apps like
Since noone else has mentioned it: BTW, Bill Joy, a founder of Sun Microsystems, is also the founder of BSD. He packaged the first Berkeley Software Distribution.
Hehe, your right about Solaris. I was afriad of that but was too lazy to do research. Shall we start
:) Anyways, it's a pointless argument, I'm not an SMP guru and next time I'll research more before posting.
BSD/OS has SMP. SunOS *did* have SMP in the later years. But admittedly it seemed to suck as much as Linux 2.0's SMP. SCO had SMP as well, granted it's also now defunkt. Another BSD based system was DEC's Ultrix, which also did SMP. Again, whether it was *good* was questionable. I never used it and don't know enough to comment.
Anyways, the point is, there were a lot of BSD based systems that did SMP. Also keep in mind SysV didn't do SMP even in 1988. If your talking about modern BSD based systems, and not BSD/Mach systems (where it's reallly Mach doing the SMP work I think) then we must limit ourselves to FreeBSD (SMP, soon to be very good SMP), OpenBSD (no SMP, but it seems will evenutally have SMP), NetBSD (Has SMP as of next major release, in other words it's complete), and BSD/OS (has SMP).
So currently most BSD actually have SMP I guess.
yes! you're exactly right! In most real world *business* applications, there are other bottlenecks that keep SMP from giving near 100% increase in performance for each additional processor (it was I/O and cache thrashing in this case). I *have* seen scientifc and engineering applications where that wasn't the case, but those 4 and 8 ways are usually a waste of money in view of cost/benefit.
You could mark him as a foe and add a -5 to all his posts.
IPv6 support is not really any more complicated, at an application level, than IPv4 support. In fact the two can be unified using a common interface which is even defined in POSIX.1g. NAT does *not* "solve" the problem of diminished IP space, it simply makes it less immediately burdensome. NAT brings with it a set of unique new problems (like accepting inbound connections). Maybe you've had trouble with IPv6, but that does not mean that it is "inherently broken" or anything like that. IPv6 support on some systems no doubt lacks documentation, but the core technologies are for the most part sound, stable, and ready for prime time.
:)
The key factor, I think, to lack of IPv6 rollout so far is missing support in routers and Windows. That will come, though.
Of course, IHBT perhaps.
"question = (to) ? be : !be;" --Shakespeare
I don't think *you* get it.
/etc/inetd.conf as installed from the installation media, it's not the 'default configuration.'
If it's not enabled in
What good is an httpd that points to some 'default' directory? Once you start loading in content, forms, etc. it's not 'audited'.
This is picking over nits. But I 'get it'.
Like I care about my karma, being maxed out for the last YEAR. Go mod something UP for chrissake. What a waste of mod points.
For your modding pleasure, this is offtopic, flamebait, trollish, overrated and quite sad that I even feel the need to mention this.
I'm not saying its broken , its just far too complex. I'm sure it would work very well if set up probably , but getting it set up probably is a monumental task. Being a network admin is hard enough without having to worry about all the new issues that IP6 brings to the table which for 99% of companies and home users are irrelevant anyway. And yes there is a common posix interface for ip4 and 6 but the standard C API is not everything. I've recently had to write a network protocol and part of this required sending orginating user IP addresses in our packets. This meant reserving 16 bytes of space in packets in case we ever upgrade from just ip4 which only requires 4 bytes. Plus ip6 addresses have to be stored in an array , not a simple unsigned int making life a lot harder on low level coders like myself.
A simple unsigned int on a SPARC, or a simple unsigned int on an x86?
SMP is not a server requirement, at least not for file/web/data servers. There the primary requirement is fast I/O (disk, network). Compute servers would be an exception.
I submit you're more likely to find SMP in the workstation of someone doing heavy duty crunching (image editing, PCB layout, simulations).
As OpenBSD is already the most secure OS available, it's no wonder that we have some of the top scientists in Canada working on bringing in SMP support. We would have taken a ton of code from FreeBSD, but we believe that their upcoming implementation is inferior to what we could come up with on our own. Besides, there's no telling what kind of glaring security holes there could be in FreeBSD's SMP implementation.
--
Theo DeRaadt
Founder, OpenBSD project.