As my first attempt at a wireless lan I researched what hardware was compatible and found that the DWL-650 was the Prizm chipset and ordered a pair. Guess what? The brain-dead reseller I should have known better than to order from shiped me two DWL-650+ cards. They wanted me to pay money to further do business with them but I sold the unopened one (I didn't read the box, I just assumed and tore the first one open and inserted it in my computer) on Half.com.
Now that I still have the other one on the shelf I am rather tickled to find this project.
I suppose I should finish this by saying that DragonFly is a BSD with changes that make it operate more like a microkernel. System calls and device drivers communicate with messages.
Matt's trying to remove the MP lock (aka giant lock) bottleneck.
This page explains what FreeBSD does when you run top on an MP system and you see processes waiting for *Giant:
Approach 1: global kernel lock (aka Giant Lock)
- only one processor executing in kernel at any time
- acquire lock when entering kernel
- release lock when exiting kernel
- need to prevent interrupts on all CPUs
- disable locally, otherwise deadlock
- interrupts on other CPUs acquire lock
- simple first implementation of SMP kernel
- limits kernel concurrency
- limits performance for larger numbers of CPUs
Mach was the true kernel with BSD running on it like a server. Mach doesn't have fork() or open()/close() or filesystems, that was done in the BSD server.
The GNU Hurd also using Mach (but that's being replaced) as the kernel but instead of only having one monolithic server to do all the UNIX-y stuff it has separate servers for each filesystem type and process management etc.
It's easy to disassociate the nameserver from the OS using the ports tree so it's easy to upgrade the system and more secure since you can run in a change-root.
AT&T filed suit against Berkeley Software, claiming that proprietary code agreements had been haphazardly violated. In the same year, BSD filed countersuit, reciprocating bad intentions and fueling internal rivalry
But Solaris drivers are dynamically loaded. What's the difference between the kernel loading a driver and exec()ing a program? A GPLed dynamically loaded should be able to be legally run from an non-GPLed kernel.
No, that's the (misguided) point of the open-source movement's worst figurehead.
Do you really think that this movement could have been started by a pacifist? When RMS announced a project to develop a free UNIX-replacement everybody demonstrated that they believed it could not be done. In 1984 things were different than in 1991 because the world was changing due to his actions.
It was started as an experimental project, and participated in as a hobby.
Define "start". To me things that come first are the start.
The point should be to have a goal of creating something, and letting that stand on its own merit.
Creation and destruction are two aspects of the same process. Trying to separate them is meaningless. Every time you improve Linux you approach closer to the point where MicroSoft is obsolete.
Without distinguishing features like Apt, the argument for going with Debian is diminished. Sure, there are still arguments to be made, but they are less obvious. Here's an exercise for you: imagine you are standing at the water cooler chatting with workmates, and a non-technical colleague just said they are thinking of trying Linux at home and were going to install Red Hat but they heard Debian is really good, but has a tricky installer. They think they'll just try Red Hat because that's what they've heard of other people using, but are interested in your opinion because you're "in computers". You've got exactly 15 seconds to succinctly explain why Debian may be better for them than Red Hat.
> Autor: Dr. Stefan Hildemann > Datum: 10.06.03 18:33 > > Ich hatte heute die MÃglichkeit, mir die belastenden Code-Abschnitte > anzuschauen.
Today I had the possibility of seeing the purloined Linux code in SCO.
> Durch einen Irrtum seitens des vertretenden Anwaltsbüros mussten mein > Kollege und ich, im Unterschied zu den 7 anderen Beauftragten, die > heute EInsicht erhielten, keine VerschwiegenheitserklÃrung > unterschreiben.
By a mistake on the part of the representing law office my colleague and I, in contrast to the other seven assigned ones, did not have to sign the SCO NDA.
> Ganz im Gegensatz zu den Prüfern der Firma Microsoft, > die offenbar sogar gegenüber ihren eigenen Vorgesetzten Stillschweigen > wahren müssen, und nur der firmeninternen Revisionsabteilung gegenüber > Bericht erstatten dürfen.
My situation is completely unlike that of those Microsoft's examiners of the SCO code, who must protect secrecy even away from their own superiors, and only report to their internal audit department.
> Nun, zum Code selbst:
Now, concerning the code:
> Vorgelegt wurden unter notarieller Aufsicht 46 Seiten a jeweils eine > HÃlfte Code aus Linux (grÃÃYtenteils ausgedruckte Posts direkt aus der > Linux-Kernel-Mailingliste) und die andere HÃlfte Listings von SCO.
Under notarial supervision 46 pages each showing on one half the code from Linux (to a large extent printed out E-mail messages directly from the Linux Kernel mailing list) and the other half listings of SCO code were demonstrated.
> Ob es sich dabei tatsÃchlich um SysV-Quellen handelt, ist so nicht > nachzuvollziehen, da sie aus dem Zusammenhang gerissen sind.
It is not really possible for one to determine the authenticity of the SCO code since it is completely removed from contextural clues.
> Interessant ist auch, das sÃmtliche Datumsangaben aus beiden entfernt > worden sind, selbst aus den Kommentaren.
Notibly all the dates were removed from the demonstrated source code and its comments.
> Die Kommenare selbst sind stellenweise tatsÃchlich identisch, selbst > einige Witze sind auf beiden Seiten gleich.
The comments are actually identical in some parts, even some jokes are alike on both sides.
> AuffÃllig ist aber, das an > den am meisten übereinstimmenden Stellen der vor den Kommenaren > stehende Quellcode doch recht unterschiedlich ist.
Remarkable is however, which is nevertheless quite different to to most agreeing places the source code standing before the comments.
However, it is remarkable that the source code which is standing before the comments is quite different nevertheless.
> Die grundlegende > Aufbau der beanstandeten Funktionen ist zwar Ãhnlich, die konkrete > Implementation aber doch recht verschieden.
The fundamental structure of the purloined functions is similar, the concrete implementation however nevertheless quite different.
> Variablen und > Funktionsnamen sind anders, Schlefien sind unterschiedlich > strukturiert, Bedingungen laufen über Kettenabfragen oder Bitmuster.
Variable and function names are different, Schlefien (?) are structured differently, conditions run over Kettenabfragen ("conditional branches"?) or Bitmuster ("Boolean logic"?).
> Alles in allem lÃsst sich nur eines sicher sagen: Die von den > jeweiligen Code-Abschnitten bereitgestellten Funktionen sind oftmals > gleich, was aber auch von vorn herein zu erwarten war.
All in all only one thing can be surely said: The functions made available by the respective code sections are often alike, which was to be expected considering the allegations.
"Reference to the SCO web site shows that they own certain intellectual property and that they correctly attribute the trademark to The Open Group. SCO has never owned "UNIX". SCO is licensed to use the registered trademark UNIX "on and in connection" with their products that have been certified by The Open Group, as are all other licensees," the statement said.
"These are the only circumstances in which a licensee may use the trademark UNIX on and in connection with its products. Statements that SCO 'owns the UNIX operating system', has 'licensed UNIX to XYZ' are clearly inaccurate and misleading."
Q: What is SVR6? A: SVR6 is the code name for the next-generation operating platform designed to take advantage of Web services and is the foundation of our SCOx strategy. As the owners of the UNIX operating system, it is incumbent upon SCO to advance the UNIX kernel for both 32-bit and 64-bit architectures. This will be accomplished through the support of key industry partners who will also contribute to this next-generation platform. SVR6 will be formally announced at our upcoming SCO Forum event to be held in Las Vegas, Nevada on August 17-19 at the MGM Grand Hotel.
Slashdot Mirror
As my first attempt at a wireless lan I researched what hardware was compatible and found that the DWL-650 was the Prizm chipset and ordered a pair. Guess what? The brain-dead reseller I should have known better than to order from shiped me two DWL-650+ cards. They wanted me to pay money to further do business with them but I sold the unopened one (I didn't read the box, I just assumed and tore the first one open and inserted it in my computer) on Half.com.
Now that I still have the other one on the shelf I am rather tickled to find this project.
I suppose I should finish this by saying that DragonFly is a BSD with changes that make it operate more like a microkernel. System calls and device drivers communicate with messages.
Matt's trying to remove the MP lock (aka giant lock) bottleneck.
This page explains what FreeBSD does when you run top on an MP system and you see processes waiting for *Giant:
Approach 1: global kernel lock (aka Giant Lock)
- only one processor executing in kernel at any time
- acquire lock when entering kernel
- release lock when exiting kernel
- need to prevent interrupts on all CPUs
- disable locally, otherwise deadlock
- interrupts on other CPUs acquire lock
- simple first implementation of SMP kernel
- limits kernel concurrency
- limits performance for larger numbers of CPUs
The GNU Hurd also using Mach (but that's being replaced) as the kernel but instead of only having one monolithic server to do all the UNIX-y stuff it has separate servers for each filesystem type and process management etc.
free clue
debunking denialists
It's easy to disassociate the nameserver from the OS using the ports tree so it's easy to upgrade the system and more secure since you can run in a change-root.
See this USENET post.
Is it just me or does the *BSD is Dying troll seem like he has real problems
get help kid
But Solaris drivers are dynamically loaded. What's the difference between the kernel loading a driver and exec()ing a program? A GPLed dynamically loaded should be able to be legally run from an non-GPLed kernel.
Then explain this
Plus this: http://www.bsd-laptop.org/
I am the only one who read "menthol battery"? :)
Define "start". To me things that come first are the start.
Creation and destruction are two aspects of the same process. Trying to separate them is meaningless. Every time you improve Linux you approach closer to the point where MicroSoft is obsolete.
Yes, with Knoppix being based on Debian any fool run a Debian in minutes. Just boot the CD-ROM and there you are running Debian.
Installation is not that much more complicated. Tuning your system is...well can't have everything.
Day one.
15, Knoppix 14, Knoppix! 13, KNOPPIX!
Suppose you found that your terminal brain cancer (MS) was attacking your terminal lung cancer (SPAM)?
> Autor: Dr. Stefan Hildemann
> Datum: 10.06.03 18:33
>
> Ich hatte heute die MÃglichkeit, mir die belastenden Code-Abschnitte
> anzuschauen.
Today I had the possibility of seeing the purloined Linux code in SCO.
> Durch einen Irrtum seitens des vertretenden Anwaltsbüros mussten mein
> Kollege und ich, im Unterschied zu den 7 anderen Beauftragten, die
> heute EInsicht erhielten, keine VerschwiegenheitserklÃrung
> unterschreiben.
By a mistake on the part of the representing law office my colleague and I, in contrast to the other seven assigned ones, did not have to sign the SCO NDA.
> Ganz im Gegensatz zu den Prüfern der Firma Microsoft,
> die offenbar sogar gegenüber ihren eigenen Vorgesetzten Stillschweigen
> wahren müssen, und nur der firmeninternen Revisionsabteilung gegenüber
> Bericht erstatten dürfen.
My situation is completely unlike that of those Microsoft's examiners of the SCO code, who must protect secrecy even away from their own superiors, and only report to their internal audit department.
> Nun, zum Code selbst:
Now, concerning the code:
> Vorgelegt wurden unter notarieller Aufsicht 46 Seiten a jeweils eine
> HÃlfte Code aus Linux (grÃÃYtenteils ausgedruckte Posts direkt aus der
> Linux-Kernel-Mailingliste) und die andere HÃlfte Listings von SCO.
Under notarial supervision 46 pages each showing on one half the code from Linux (to a large extent printed out E-mail messages directly from the Linux Kernel mailing list) and the other half listings of SCO code were demonstrated.
> Ob es sich dabei tatsÃchlich um SysV-Quellen handelt, ist so nicht
> nachzuvollziehen, da sie aus dem Zusammenhang gerissen sind.
It is not really possible for one to determine the authenticity of the SCO code since it is completely removed from contextural clues.
> Interessant ist auch, das sÃmtliche Datumsangaben aus beiden entfernt
> worden sind, selbst aus den Kommentaren.
Notibly all the dates were removed from the demonstrated source code and its comments.
> Die Kommenare selbst sind stellenweise tatsÃchlich identisch, selbst
> einige Witze sind auf beiden Seiten gleich.
The comments are actually identical in some parts, even some jokes are alike on both sides.
> AuffÃllig ist aber, das an
> den am meisten übereinstimmenden Stellen der vor den Kommenaren
> stehende Quellcode doch recht unterschiedlich ist.
Remarkable is however, which is nevertheless quite different to to most agreeing places the source code standing before the comments.
However, it is remarkable that the source code which is standing before the comments is quite different nevertheless.
> Die grundlegende
> Aufbau der beanstandeten Funktionen ist zwar Ãhnlich, die konkrete
> Implementation aber doch recht verschieden.
The fundamental structure of the purloined functions is similar, the concrete implementation however nevertheless quite different.
> Variablen und
> Funktionsnamen sind anders, Schlefien sind unterschiedlich
> strukturiert, Bedingungen laufen über Kettenabfragen oder Bitmuster.
Variable and function names are different, Schlefien (?) are structured differently, conditions run over Kettenabfragen ("conditional branches"?) or Bitmuster ("Boolean logic"?).
> Alles in allem lÃsst sich nur eines sicher sagen: Die von den
> jeweiligen Code-Abschnitten bereitgestellten Funktionen sind oftmals
> gleich, was aber auch von vorn herein zu erwarten war.
All in all only one thing can be surely said: The functions made available by the respective code sections are often alike, which was to be expected considering the allegations.
> In der konkrete
Well the quotes may have been used in an inappropriate fashion but read this book and you will see that they are what RMS thinks (and rightly so.)
more like this
A current violation:
http://www.sco.com/scosource/linuxqanda.html
I gave them up after I found:
http://www.resellerratings.com/seller2121.html