Slashdot Mirror

bizzaro CISC instruction set piece of shite

I guess you didn't get the memo. Turns out RISC wasn't the good idea everyone thought it would be in the 1990's.

Indeed. Even Linus thinks the x86 is an ugly duckling.

I never understood why I couldn't get a fast terminal prompt and have the remainder of the daemons start up in the background, all reniced to low priority during the initialization process, or maybe slowly started up to avoid disk contention. I personally amortize the bootup time by buying a bunch of ram, and dd'ing all the files in /usr/bin, /usr/lib/*.so, /lib/*.so, /etc and maybe /usr/share/apps, depending on how much ram I have. This pulls everything into the buffer cache and improves KDE startup significantly.

I'm sure I could optimize this by running some kind of kernel auditor that tracked every file that was run (executable) or loaded (.so), wrote out a list at shutdown, and reused that list to precache on reboot.

A few interesting links: http://kerneltrap.org/node/2157 , http://www.uwsg.iu.edu/hypermail/linux/kernel/0609 .2/2180/boot_linux_faster.pdf , http://initscripts-ng.alioth.debian.org/soc2006-bo otsystem/deliverable3.html , http://preload.sf.net/ , http://packages.ubuntu.com/dapper-backports/admin/ readahead .

I2 from a connectivity standpoint really isn't anything different from I1. It's still an IP routed network and all your normal IP routed toys (www, ftp, home brew app) still work as advertised. The term Abilene is actually the name of the I2 backbone network. It was spawned as a second generation IP network (and yes it can route IPV6 natively) to connect research institutions. Allowing them to utilize the network for research and high speed data transfers. What kind of research? Well anything really, hear about those doctors doing remote robotic surgery? That data probably was traversing the Abilene backbone. Grid computing in acedemia? Probably connected via Abilene .

Just like the I1 backbone, Abilene, being a backbone network, peers and eventually splits off to regional controlled networks. The one I am familiar with is OARNet (Ohio Acedemic Research Network). OARNet provides high speed connectivity to Ohio campuses and peers with the Abilene I2 backbone in Indianapolis (from OARNet's POP in Cleveland).

From an I2 connected campus (meaning you have access to the Abilene backbone at somepoint), there really isn't anything special to connect over the Abilene core. The network gurus had the IPV4 routing setup in such a way that if you connected to an IP address that was available via Abilene, the data would go that path. Otherwise, it would route out over the standard I1 connection. Most of the time when I would have to download some big ISO images, I would specifically look for an Abilene (I2) connected peer. Downloads over 10MBit weren't uncommon (mostly limited by the load on at the server on the other end). Pretty cool really. I'm sure others around are using it for more important stuff other than downloading ISOs :).

I didn't even know there was a DVD version, as I don't see it for 6.06 or for 6.10. Nevertheless, the CD install doesn't give you any of those options. Note that my complaint wasn't that things weren't installed (kcontrol is still there for example, but you have to run it directly), but that they weren't discoverable (ie, in the K menu) unless you already knew about them, the fact that the DVD version is so well hidden merely highlights my point. Perhaps the DVD version is only available through the mail?

There is no such thing as autorun on OS X

Actually, there is, but only if you run classic in OS X. It's called Autostart in QuickTime. If you have Classic installed *and* running, it will still work. This page mentions it, and there used to be a test exploit page located at http://www.u-struct.com/diary/img/20020131_OSissue _E/ but it seems that link is no longer active. It's an exploit that has been known about for years, but it's very low risk now. You're only at risk if you run the Classic environment, and then it can be disabled in your classic QuickTime preferences. More information about disabling QuickTime's autostart can be found here.

That is a good question. I hope they released the sourcecode somewhere. I don't have an Xbox, but I would like to have that program--I have had two strokes.

Maybe it works on Linux in general, and they wanted to make it more widespread? The glove appears to work on PCs to. No Mac drivers apparently, but according to this page, they eventually made linux ones.

Here are some interesting links for the glove: Linux patch to blacklist the P5 from HID (apparently it reports itself as HID device, but does not comply with standard.) Library to use the P5 in Linux geocities page with lots of links to P5 info Page with lots of technical details about the glove

One question I have: I tried Essential Reality's site, and there is another company listed there. Did they get bought out or go out of business? Is the glove still being made?

After looking at all this, I want one of these. This would be a kick-ass input device even for regular computing. You could probably use it as a mouse and keyboard (one handed--no pr0n jokes please ;-)...and more!

Have you ever read an email by Mr Schilling? Try this thread on lkml, and tell me who is being the most annoying. He drags himself through the mud by alienating people with his attitude.

I've actually taken this a step farther. I've pieced together a RAID 1 array, using an old computer, stuffed with a high powered power supply and lots of 400 gig harddrives. (They are cheap. Yes, I know there are bigger drives, but I can afford $150 out of a paycheck easier than $400+.) I have about 2 terabytes so far, but I figure that with PCI expansion cards, I can get up to almost 3 terabytes, with fairly secure data, easily. If you want to ommit the redundency, you can hit 6 terrabytes+.

Alternate (amd64)

Alternate (i386)

Alternate (powerpc)

Desktop (amd64)

Desktop (i386)

Desktop (powerpc)

Server (amd64)

Server (i386)

Server (powerpc)

Alternate (amd64)

Alternate (i386)

Alternate (powerpc)

Desktop (amd64)

Desktop (i386)

Desktop (powerpc)

Server (amd64)

Server (i386)

Server (powerpc)

Alternate (amd64)

Alternate (i386)

Alternate (powerpc)

Desktop (amd64)

Desktop (i386)

Desktop (powerpc)

Server (amd64)

Server (i386)

Server (powerpc)

Alternate (amd64)

Alternate (i386)

Alternate (powerpc)

Desktop (amd64)

Desktop (i386)

Desktop (powerpc)

Server (amd64)

Server (i386)

Server (powerpc)

Alternate (amd64)

Alternate (i386)

Alternate (powerpc)

Desktop (amd64)

Desktop (i386)

Desktop (powerpc)

Server (amd64)

Server (i386)

Server (powerpc)

Alternate (amd64)

Alternate (i386)

Alternate (powerpc)

Desktop (amd64)

Desktop (i386)

Desktop (powerpc)

Server (amd64)

Server (i386)

Server (powerpc)

Alternate (amd64)

Alternate (i386)

Alternate (powerpc)

Desktop (amd64)

Desktop (i386)

Desktop (powerpc)

Server (amd64)

Server (i386)

Server (powerpc)

Alternate (amd64)

Alternate (i386)

Alternate (powerpc)

Desktop (amd64)

Desktop (i386)

Desktop (powerpc)

Server (amd64)

Server (i386)

Server (powerpc)

Alternate (amd64)

Alternate (i386)

Alternate (powerpc)

Desktop (amd64)

Desktop (i386)

Desktop (powerpc)

Server (amd64)

Server (i386)

Server (powerpc)

On one side we have the 'use it if you can' camp, and on the other, the 'it's not free so screw that' camp. We need both, though we could do without some of the scaremongering so favoured by the latter.

Remindes me painfully of this: http://www.ussg.iu.edu/hypermail/linux/kernel/0512 .0/0972.html The only breakthrough here is for hardware companies who don't want to publish specs for their hardware. Whoever at Novell designed this driver deal doesn't truly understand what Free Software is all about (or Open Source for that matter) and why it's important to have documented hardware.

http://www.ussg.iu.edu/hypermail/linux/kernel/9906 .0/0746.html

He is, of course, referring to all the research in the '80s and '90s on microkernels and IPC-based operating systems.

http://www.ussg.iu.edu/hypermail/linux/kernel/9906 .0/0746.html

He is, of course, referring to all the research in the '80s and '90s on microkernels and IPC-based operating systems.

Is this a new attempt at including Digital Restrictions Management in Linux? I've read posts on here that have discussed the DRM inclusions in GNOME's GStreamer. Could KDE be next? Real, Fluendo, and other copyright trolls are a threat to watch out for, similar to the looming threat of binary kernel modules. We as Linux users must be on guard against the copyright mafia's incessant attempts to neuter our computers.

Indeed, looking at the vmsplice patch you can clearly see that it supports nonblocking I/O (with the SPLICE_F_NBLOCK flag)
http://www.ussg.iu.edu/hypermail/linux/kernel/0604 .2/1399.html

http://news.com.com/2102-7344_3-6061491.html?tag=s t.util.print
Linux in a binary world... a doomsday scenario
http://www.ussg.iu.edu/hypermail/linux/kernel/0512 .0/0972.html
http://www.ussg.iu.edu/hypermail/linux/kernel/0602 .1/0274.html
http://www.ussg.iu.edu/hypermail/linux/kernel/0602 .0/1051.html

http://news.com.com/2102-7344_3-6061491.html?tag=s t.util.print
Linux in a binary world... a doomsday scenario
http://www.ussg.iu.edu/hypermail/linux/kernel/0512 .0/0972.html
http://www.ussg.iu.edu/hypermail/linux/kernel/0602 .1/0274.html
http://www.ussg.iu.edu/hypermail/linux/kernel/0602 .0/1051.html

http://news.com.com/2102-7344_3-6061491.html?tag=s t.util.print
Linux in a binary world... a doomsday scenario
http://www.ussg.iu.edu/hypermail/linux/kernel/0512 .0/0972.html
http://www.ussg.iu.edu/hypermail/linux/kernel/0602 .1/0274.html
http://www.ussg.iu.edu/hypermail/linux/kernel/0602 .0/1051.html

I'd never heard of this new "cars/quarter" unit (invented by the same guy who gave us the LoC unit, presumably), so I had to look it up to see that this glue can hold around 10,000 psi (70,000 kPa).

Just use this link instead.

There's a lot of historical decisions of dubious validity in retrospect, but there's also an excellent technical defense that can be made for Mach. In short, Mach is really cool. Mach IPC makes signals, sockets, pipes, shared memory, SysV IPC, etc. look positively clumsy. What's FreeBSD's answer to the Mach Inteface Generator? CORBA?

So OS X gets a lot of mileage out of Mach messaging - AppleEvents, distributed notifications, run loops, etc. If OS X processes seem good at talking to one another - think VoiceOver, Spotlight, the window server, iLife's media sharing, even copy and paste - it's due in part to the fast, flexible IPC mechanisms enabled by Mach.

The 4/4 memory split only applies to 32 bit environments. Haven't the G3/G4/G5 been 64 bit?

In principle, yes; in practice, OS X has a 32 bit kernel even on 64 bit machines, not least of all for driver binary compatibility. You want to know the win here - take a look at the binary compatibility driver story on 64 bit Linux or 64 bit Windows. Apple allows 64 bit processes on Tiger without breaking everyone's hardware.

(Incidentally, only the G5 is 64 bit.)

Are you suggesting that FreeBSD, Linux, Windows, or any other modern operating system doesn't use dynamic libraries?

Yes. Benchmarks typically compare statically linked libraries on Linux (because they're faster) to dynamically linked binaries on OS X (because that's all Apple ships).

Yes, there is an advantage to not using the buffer cache in some cases, something you can do in linux with the O_DIRECT filedescriptor flag

Thanks, I wasn't familiar with that flag on Linux. From googling, it looks like it does somewhat different things, in particular, not speeding up sequential file access.

In any case, I'll certainly agree that there Linux-specific filesystem optimizations; I was just commenting on a technique I found to give a substantial boost to OS X programs with sequential access patterns.

I was waiting for someone to bring this one up.

Kernel configuration. It's not just a job, it's an adventure!

....he makes some polite, reasonable replies to the answers to his post. Nice to see.

Yea, interestingly enough Indiana University's Webmail (new beta) does some work to combat this. They make it harder for the individual user to blindly click a redirect link. You can check out the feature list here.

As 2600 says... if you see something, say something.

... he really doesn't care about binary drivers like nVidia

I'm afraid you could not be more misinformed. You'd be wise to follow the LKML to find out what Linus actually thinks and says.

... he really doesn't care about binary drivers like nVidia

I'm afraid you could not be more misinformed. You'd be wise to follow the LKML to find out what Linus actually thinks and says.

Anyhow, I think .NET suffers from the same mentality. Sure, I've played with it a bit. It was very simple to jump into C# from Java, they had a fairly rich set of core libraries. Microsoft keeps pitching it as 'easy' and I suspect there are too many folks krufting out C# apps rather than crafting them, thus my perception is this new framework does not have a high enough barrier of entry. My assumption is the money will follow the same pattern. That, and the .NET framework is also a bit young. I've worked with a lot of companies, and those few who are making the jump to .NET started in 2005/2004. No 'bleeding edge' bonus money over other platforms/frameworks. Why would I move to it?

I know a lot of people who have the same frusteration over a lowered barrier to entry. Making something outright *easier* is just better (no sense in having to blow lots of time making people learn something unnecessary), but there is a problem when it's easy to enter a field but hard to perform in the field competently.

I've known a few people (who I consider to be overreacting) who actually dislike use of debuggers, because they feel that it makes people less likely to actually properly think through what they are developing. It lowers the barrier to entry a lot, and that the cost of that loss of a filter causes more harm than the overall benefits. One of these people is Linus Torvalds. Another is my current boss. While I think that they both go too far, there is a sizeable kernel of truth.

I think that the single largest irritation I have with some of the newer platforms is that they greatly lower the barrier to entry with threads (see my .sig). Traditionally, it was a pain in the ass to write a simple preemptively-multithreaded program. Java and .NET make it *much* easier to write that hello world thread program.

The problem is that the syntax you must use and some trivial knowledge about what is safe to do and what isn't in a language or on a platform is the tiniest part of what must be known to correctly multithread a program. You *must understand how to correctly do multithreaded design* from the standpoint of designing a program. This is not easy to do correctly, and it's easy to take "shortcuts" that then come back and bite one in the ass. I would say that I have known maybe three or four people that I would trust to correctly write preemptively multithreaded systems that I'd be willing to trust my life to -- and in every single one of their cases, I've found nasty design-level bugs in their threaded or distributed designs that probably would not have cropped up in a simple single-threaded design.

Multithreading a a system has exactly two legitimate uses:

(1) Increasing performance. Unless you have a very specialized system designed for parallel processing, you are probably talking about possibly a factor of two speedup on a typical multiprocessor system. The problem is that a factor of two is *nothing*. You have a limited amount of development time T in which to make your application perform as quickly as possible. It's almost always possible to redesign such a system's higher-level logic (precompute, cache to avoid recomputation, better algorithms, whatever) to get at *least* a factor of two speed improvement on a CPU-bound application, and doing so is almost always cheaper in terms of development time than multithreading an application. Multithreading is damned expensive in terms of development time and debugging time. It makes profiling much more difficult. It is *not* immediately clear to a reader of your code what assumptions you make about the threading model; we developed data hiding for defining how data must be touched, but we have no good way at the language level for defining guarantees about how threads must interact. Result: you waste a lot of time for minimal performance gain.

(2) Ea

So huge you didn't even notice it, did you? :-)

See http://kb.iu.edu/data/aliv.html

The meaning of words can change. The law should not change with the meaning of the word. The law should only change with the changing of the law by legistators.

Absolutely right. When construing a statute, judges do indeed consider what its words meant at the time the statutes were written. Their overriding objective is to determine the intent of the legislators who wrote the statute.

The meaning of the word "make" used by the appellate court has been in common use since the Middle Ages.

So the GPLd code could be "attached" to the hardware in such a way that removing one from the other becomes impossible (or at least not trivial). It's easy to see how this free software/hardware combination could be used for DRM.

Here's an interview with Richard Stallman discussing Linus's decision to include DRM in the linux kernel.

And here's a post from linus on the kernel mailing list (thread "flame linus to a crisp") talking about DRM in the linux kernel.

So there you go GPLed DRM.

Not true.

http://www.kerneltraffic.org/kernel-traffic/kt2002 0708_174.html#5
http://www.ussg.iu.edu/hypermail/linux/kernel/0405 .2/1273.html

There are also some ABI implications, which is why most distros avoid 386 support. Can't be bothered to do any more googling.

Please could you point me to some documentation for Linux's transactional filesystem support? The last time I searched for information on the topic, I found out that it would be tremendously complicated, without any really important benefits.

ATI and NVidia are famous for being the two worst companies to buy from. Some links:

• The Linux Doomsday scenario
• The "Everbody loves Eric Raymond" cartoon strip it inspired ;-) - featuring ATI
• Stallman makes an example of NVidia in the GNU 20th anniversary essay

Actually their binary drivers are really crashy historically. There was this bug in all of the 3xxx releases, and most releases through the 4xxx could lock up when X was trying to do certain things (large copies from offscreen seemed to do the trick). I don't even know if the latter was ever fixed because I switched from my old tnt2 to an ATI r200-based card with open-source drivers.

The whole point of using C for benchmarks like SPEC is to remove the influence of any and all VM, interpreters and the like. It is as close to programming on the bare metal as reasonably possible these days.

Now FORTRAN is indeed used a great deal on supercomputers and is easier to optimize aggressively, this is true, but I've yet to see a modern O/S written in FORTRAN, although I have little doubt that this is possible. I'm sure you could link FORTRAN code into the Linux kernel, but I doubt it would be accepted in the main tree...

Now on the FORTRAN vs. C question if you read the classic "real programmers don't eat quiche" it seems that although the author doesn't think much about Unix he says good things about C. That ought to settle the debate ?

If not C99 is easier to parallelize than C89, and some swear C++ can match or exceed FORTRAN.

Normally this means the server is sending the file as some binary format, file extensions don't matter. Try this Ubuntu torrent which works for me.

Have your developers read this fantastic set of course pages:
http://osl.iu.edu/~lums/swc/

The course is sponsored by the python foundation, so there's a lot of python embedded. Still, its one of the best and most complete collection of best practices and resources I've come across.

If the gamer cannot make it to the CAVE http://www.avl.iu.edu/technology/cave/, Philips is bringing the (functional equivalent of) holodeck to the gamer.

Before we all go saying, Naaah or Nya nya it won't work, I say heck if the technology is there, let's see how it pans out. If it doesn't, then it will go the natural course of evolution in technology: wind up in the pr0n alleyways on the 'net.

The possibilities make me wanna say, Philips I wanna have your bab...er, I mean, Bring It On!

Good Call: http://www.microsoft.com/technet/archive/msdos/04_ dblsp.mspx

while it appears drivespace was the MS 6.20 fix. http://kb.iu.edu/data/abid.html

For fun, Don also maintains the Extremes of Conventional Music Notation where he records the extremes found in written music. Some interesting excerpted tidbits:

• softest pppppppp (8 p's) in Ligeti's Etudes for Piano, 1st Book
• loudest ffffffff (8 f's) in Ligeti: Etudes for Piano, 2nd Book, (the 1812 overture only reaches ffff)
• Instruments to be played by one performer in a piece - *Mahler: Symphony no. 5 calls for one clarinetist playing six different instruments.
• Most repeated notes in a melody - 32 in Prokofieff: Toccata, Op. 11 (1912)

There are many others, quite interesting.

For fun, Don also maintains the Extremes of Conventional Music Notation where he records the extremes found in written music. Some interesting excerpted tidbits:

• softest pppppppp (8 p's) in Ligeti's Etudes for Piano, 1st Book
• loudest ffffffff (8 f's) in Ligeti: Etudes for Piano, 2nd Book, (the 1812 overture only reaches ffff)
• Instruments to be played by one performer in a piece - *Mahler: Symphony no. 5 calls for one clarinetist playing six different instruments.
• Most repeated notes in a melody - 32 in Prokofieff: Toccata, Op. 11 (1912)

There are many others, quite interesting.

Basically, if an ISP can make good money off providing broadband service, why will it let a community body run such projects for free or cheap? The student loan system in USA is a vaguely related, yet relevant example. Why do we have 500+ billion for war but not for education or medicare? And free Internet?