Domain: tu-bs.de
Stories and comments across the archive that link to tu-bs.de.
Comments · 31
-
Re:The uses are endless
All you need is a web came and a laser pointer
http://www.cs.tu-bs.de/rob/david.html -
Augment this with DAVID 3D laser scanning at home
Very neat, for the backend.
For the frontend, this technique for 3D laser scanning you can do at home using a cheap webcam would be a great match!:
http://www.cs.tu-bs.de/rob/david.html
As a plastic scale model kit builder I can think of a number of parts I could scan and replicate! -
Is the -rt patchset hard-real-time?
ADEOS - the microkernel used in RTAI - is "hard real-time", as is VxWorks. TimeSys' Linux patches are soft real-time.
Small correction: if by "TimeSys' Linux patches" you mean the -rt patchset that i'm maintaining (and to which Thomas is a major contributor), and in particular if you mean the CONFIG_PREEMPT_RT kernel feature, then the answer is a clear "no": it's not "soft real-time", it's intended to be "hard real-time" in the same sense as ADEOS/RTAI is.
The -rt patch-set implements a fully preemptible Linux kernel, which allows a higher-prio event to preempt any lower-prio processing: it can preempt device driver interrupt processing or other "irqs off" critical sections or other normally non-preemptible (for example spin-locked) code within the kernel, immediately. (it does all the necessary hard-real-time things one would expect: it pushes interrupt processing into special kernel threads, it implements priority inheritance for all Linux locking primitives to guarantee processing and to get out of priority inversion scenarios, etc.)
See more about the technology behind the -rt patchset in Paul McKenney's article on LWN.net, and on Kernel.org's RT Wiki. You can also try out an -rt kernel based Linux distribution yourself, grab a Knoppix-based PREEMPT_RT-kernel live-CD from: here.
-
Performance overhead of the -rt patch-set
RT has a pretty big speed penalty.
I can definitely say that unlike some other approaches, the -rt Linux kernel does not introduce a "big speed penalty".
Under normal desktop loads the overhead is very low. You can try it out yourself, grab a Knoppix-based PREEMPT_RT-kernel live-CD from here: http://debian.tu-bs.de/project/tb10alj/osadl-knop
p ix.iso.From early on, one of our major goals with the -rt patchset (which includes the CONFIG_PREEMPT_RT kernel feature that makes the Linux kernel "completely preemptible") was to make the cost to non-RT tasks as small as possible.
One year ago, a competing real-time kernel project (RTAI/ipipe - which adds a real-time microkernel to 'above' Linux) has done a number of performance tests to compare PREEMPT_RT (which has a different, "integrated" real-time design that makes the Linux kernel itself hard-real-time capable) to the RTAI kernel and to the vanilla kernel - to figure out the kind of overhead various real-time kernel design approaches introduce.
(Please keep in mind that these tests were done by a "competing" project, with the goal to uncover the worst-case overhead of real-time kernels like PREEMPT_RT. So it included highly kernel-intensive workloads that run lmbench while the box is also flood-pinged, has heavy block-IO interrupt traffic, etc. It did not include "easy" workloads like mostly userspace processing, which would have shown no runtime overhead at all. Other than the choice of the "battle terrain" the tests were conducted in a completely fair manner, and the tests were conducted with review and feedback from me and other -rt developers.)
The results were:
LMbench running times:
| Kernel............ | plain | IRQ.. | ping-f| IRQ+p | IRQ+hd|
| Vanilla-2.6.12-rc6 | 175 s | 176 s | 185 s | 187 s | 246 s |
| with RT-V0.7.48-25 | 182 s | 184 s | 201 s | 202 s | 278 s |(Smaller is better. The full test results can be found in this lkml posting.)
I.e. the overhead of PREEMPT_RT, for highly kernel-intensive lmbench workloads, was 4.0%. [this has been a year ago, we further reduced this overhead since then.] In fact, for some lmbench sub-benchmarks such as mmap() and fork(), PREEMPT_RT was faster.
(Note that the comparison of PREEMPT_RT vs. I-pipe/RTAI is apples to oranges in terms of design approach and feature-set: PREEMPT_RT is Linux extended with hard-realtime capabilities (i.e. normal Linux tasks get real-time capabilities and guarantees, so it's an "integrated" approach), while ipipe is a 'layered' design with a completely separate real-time-OS domain "ontop" of Linux - which special, isolated domain has to be programmed via special non-Linux APIs. The "integrated" real-time design approach that we took with -rt is alot more complex and it is alot harder to achieve.)
See more about the technology behind the -rt patchset in Paul McKenney's article on LWN.net, and on Kernel.org's RT Wiki.
-
Re:Ha, MS Office ...
And for those that do not believe me, I have:
Text processor
Spreadsheet
Presentation creator
Oh, and a great bibliography management system (nor MSOffice or OOo get close enough to this)
Oh, and it also has a sql mode to communicate with SQL databases, and lets not talk about its scripting capabilities! (VBA ppppffft). -
Anyone got freenx working under breezy yet?
I compiled the sources from http://debian.tu-bs.de/project/kanotix/unstable/ to create the
.deb packages. After installation and configuration (nx user was added, local user was added to nx server allowed list, there are no ssh authentication problems), any attempts to connect via the nx client ultimately fail with no indication as to what went wrong.
The indentical setup works on my Mandriva box so this is really making me scratch my head as to what could be going on. -
In the meantime ...
If you have a suitable machine you can grab kanotix 64 and try out a full 64 bit OS from a cd, and install it if you want to! Unlimited support incidents come from the simple link to the Kanotix irc channel from the desktop or a quick trip to the forum. Donations go by Paypal to "donate the_@_here kanotix.com" so know you know what to do with your $12!
-
Re:MacOS X: use partimage in knoppix
A little more googling might give you a more recent knoppix version (as of 2day it's 3.7 and 3.8 out soon), or there might be another bootable CD linux distro for PPC that contains partimage.
http://debian.tu-bs.de/knoppix/powerPC/
Once downloaded boot, wait for cute GUI, open a root shell, if it's not there then just open a terminal and run
su
partimage
If you are a little more hard core, use dd; this is how I did it:
http://linux.slashdot.org/comments.pl?sid=135938&c id=11352968 -
Re:Analysis
There are several sites dedicated to critical readings of Watchmen, because it is so dense.
These are all dripping spoilers, so care should be taken in following these links. Having Watchmen spoiled is something I wouldn't wish on anyone.
Watching the detectives, a Hypertext guide to Watchmen.
Taking Off the Mask, a bacheolor's thesis by Samuel Asher Effron, class 1996.
-
Re:How about drivers for the current crop of hardw
Thanks for the info. I'll try it out. I'm downlowding KANOTIX, which supposedly has WiFi enabled using ndiswrapper on a bootable CD. I'm going to test that out, too.
-
Knoppix
How come everyone thinks debian is so hard?
Here da go --> knoppix for ppc. Burn CD, boot, run knx-hdinstall, boot from hdd, edit /etc/apt/sources.list to point to your favorite debian mirror and then 'apt-get update && apt-get upgrade' and you're done. Or you can just stick with knoppix. It doesn't get much easier than that. -
Re:Gentoo for PPCif there isn't a Knoppix live cd for PPC
You mean this.
-
Re:Ingenious...
Ob. Futurama...
Leela: The jig's up, Nixon. We'll trade you the tape for the body.
Nixon: Oh, expletive deleted. You've got a deal.
A Head in the Polls
A Head in the Polls -
Re:PPC?
-
Re: Dreamcast instead?However, to connect to your network (at more than 56kbps) and play back anything from there, a Dreamcast (unlike the Xbox) will need an optional Broadband or LAN Adapter, still at several times the price of the console itself (if you can point out a cheap source, please do):
Sega.com has discontinued Broadband Adapter distribution. You can buy used ones on eBay, or you can go to Lik-Sang and purchase one for $120.
Moreover, the "Divx" you are referring to seems to be a "Pocket DivX", specifically transcoded to run on less powerful hardware (that could not cope with the full-featured format):Why does my video and audio skip and or freeze?
A: Again play with the encoding settings... try reducing the video bitrate, decimating by 2 or reduces the audio bitrate down to 22khz mono to avoid these skips. If you have found a magical setting that works GREAT share you findings (...)
The current decoder CORE cannot handle resolutions larger than 496x496. But a bigger issue is that anything higher then that the Dreamcast hardware can not handle (A/V skips). So do not look for any larger resolution anytime soon. -
Re:Another Postfix book is coming soon
(Replying to my own message. Ugh.) Ralf's postfix page is here.
-
Knoppix for PPC
Does anyone know where I can find Knoppix for PPC? I'd love to play around with it on some old iMacs I have lying around. However I only found an old release. I would think that linux-live cds should work even better om macs than on pcs, since there is so much less hardware to configure. The knoppix homepage states that the minimum requirements include a x86 processor.
-
Clickable Links
http://kano.mipooh.net/kanotix/
It is made by a german Knoppix hacker named Kano, who has a big page of patches for Knoppix here:
http://www.kano.mipooh.net/
It comes with kernel 2.4.23 patched with forcedeth and XFS.
It uses grub, Xfree86 4.3, is based on Debian/sid.
ACPI and DMA enabled by default (can be disabled with acpi=off respectively nodma)
The forum (german and english):
http://kanotix.mipooh.net/index.php
Download:
http://debian.tu-bs.de/knoppix/kanotix/
Torrent:
http://kano.mipooh.net/kanotix/KANOTIX-X-MAS-2003- PREVIEW.iso.torrent
-
Re:Setting up postfix to do this?Man did I munge the name!
It's Ralf Hildebrandt, and his most useful homepage can be found here.
My apologies, Ralf.
-
Re:Serious question: Quantum computing resources
I found this article to be a very useful. A nice starting block.
-
Dear John Carmack: HA HA HA, I have Doom III
I have your leaked copy of doom_iii.rar off of quakenet on IRC!
If you're reading this right now, I am raping your precious baby... ooh yeah!
http://www-public.tu-bs.de:8080/~y0019284/Bilder/1 .jpg
http://www-public.tu-bs.de:8080/~y0019284/Bilder/2 .jpg
http://users.pandora.be/kalakov/doomiii.JPG
http://www.iebeta.net/comments.php?catid=1&id= 164
-SexyKellyOsbourne -
Dear John Carmack: HA HA HA, I have Doom III
I have your leaked copy of doom_iii.rar off of quakenet on IRC!
If you're reading this right now, I am raping your precious baby... ooh yeah!
http://www-public.tu-bs.de:8080/~y0019284/Bilder/1 .jpg
http://www-public.tu-bs.de:8080/~y0019284/Bilder/2 .jpg
http://users.pandora.be/kalakov/doomiii.JPG
http://www.iebeta.net/comments.php?catid=1&id= 164
-SexyKellyOsbourne -
Re:I think he's right in a way> They would have been paid for with money seized
> from taxpayers, so if you have a job, you're paying
> for the software anyway, whether you want to use it
> or not.This already happens. Using stuff I know about directly as evidence, look at the fermilab tools homepage. Now granted this stuff wasn't developed spontaneously, every last bit of it is an internal tool that was made freely available to the public, but the point still stands. Fermilab is operated by universities across the country, but is owned by the DOE.
> think a U.S. Department of Software Development
> would result in better software?Not by default, nor every time, but it's definetly capable of it. Again I'll use fermi as an example. Nedit is a really well done GUI text editor, I call it a "second tier" editor because it's not directly a vi or emacs clone, but it's very own beast. These days it's a highly developed, well maintained editor.
So while I'd agree to an extent that the government shouldn't have a Department of Open Source that leeches taxpayer money to create free software, I *do* have to butt in and inform you that many (if not all) government owned laboratories across the US already produce and release free software.
;)So it *can* work.
-
Re:Emacs is all you want aparently.
In addition to ftp, there is ssh/scp support with the tramp package at ftp://ls6-ftp.cs.uni-d ort mund.de/pub/src/emacs/tramp.tar.gz.
The following links may be useful for writing php code in emacs:
-
Whacked things to do with a Palm Robot
Strap a webcam to the top, add an accelerometer to it and program the Palm with a fuzzy algorithm to produce the most violent jerky movement while streaming video to a 72" TV being watched by a batch of too-drunk hackers?
-
Re:Guess this means Borland wins the IDEAll of this talk of IDE and no one has mentioned Code Forge? I bought this ide about a year or so ago, and find it very handy. It has support and syntax coloring for a bunch of Languages(such as c/c++, java, php, html, perl, etc...). It is fairly simple to use, comes with built in revision control. There is also a free version available. It integrates very well with ddd which IMHO is a very nice debugger.
For little scripts and the such I still use vi, or sometimes emacs (depends on my mood), but for big projects, I tend to use CodeForge. Just my $.02
:) -
Re:Visual debugger
Have you tried DDD? I don't know how well it does Java, but I like it for C and C++, which is what I've used it for so far, and it does have support for jdb.
-
Re:finally, a good IDE for Linux?!?!
ctags stopped cutting it when I moved from C to C++, and it's useless for Java
Exuberant ctags is far advanced over old ctags programs. The C support is much improved and C++ and Java support have been added and work like a charm. Works especially well in combination with Vim. If you don't like the way it handles Java, you can give JTags a try, but it's nowhere near as stable.
If you're doing Java development, you'll probably also want to use Jikes, as it integrates very nicely with QuickFix mode in Vim and make mode in Emacs. There's also a Jikes Debugger java debugger, but I've not used it.
a decent code beautifier, since indent doesn't work right for Java
jsbeautifier is one of many -- a search I did a few months ago turned up 10 or 12 beautifiers for Java, and even more for other languages.
Of course, if you want a good graphical debugger then ddd is the way to go -- it lets you get to the gdb command line if need be.
Sumner
-
Not if you're a true nerd ;)
The true nerd's PDA of choice is, of course, the Palm IIIx. Shall we count the ways... It has 4MB of RAM, so you can load it up with countless useless things (like IRPing to fool around with IR printers), flashable OS means you can run Linux on it if you are so inclined...
Let's not be fooled, dudes - a nerd doesn't use a PDA because it's useful, as in managing your shopping list (you order on-line from your local mart anyway) or keeping track of your appointments. We have them because they're way cool - in other words, we have an intuition that they represent a new paradigm, one where bloatware is refreshingly unwelcome.
The IIIe is obviously a ploy to scrounge a bit more market share - but I say, more power to 3Com if it succeeds. Meanwhile, if you're a self-respecting nerd, get a IIIx. ;) -
HPC software original sources...Some pieces of Sun's HPC software are derivations of freely available code. Their MPI implementation is (or rather was, the last time I looked) based on mpich from ANL. The linear algebra packages are based on ScaLAPACK and crew. Sun may be giving out some tuning implementation, but nothing that can't be found automatically (see the PHiPAC and ATLAS projects). PETsc and PVM are straight builds of older code, bugs and all.
Some of the more interesting pieces, like LSF, are only licensed by Sun, thus will not be included in this `deal.' (For a free improvement over LSF, check out GNU Queue. If it doesn't do something you want, you can support the community and extend it.) If you read the announcement carefully, you'll see that the only new codes to which it applies are the parallel file system (the Sun CTO thinks distributed file systems are dead, anyways), the Prism debugger, and the parallel run-time environment.
Of those, the only with no available substitute is the debugger. The ROMIO library is a good place to start for the MPI file I/O stuff (a good database would be a better place, imho). I already mentioned queue management software. The Ptools Consortium and the Globus Project have links to other HPC cluster tools.
Many of the pieces for debugging are available (combine ddd and gnuplot), but some notable ones are missing. The ability to control multiple GDBs easily from one processes and the visualization of parallel execution are needed, and quite difficult to implement. There seems to be interest in making GDB easier to use from other processes, which is a good start towards solving the larger problem of general, distributed debugging. And both the mpich and LAM MPI implementation have some profiling information, but few tools to dig through it.
To be fair, Sun has contributed (and supported contributions) to the original packages. Why they are releasing the rest under their Exploit the Community license is beyond me.
Jason, ejr@cs.berkeley.edu
-
DDD in itself is stable, but...
See the DDD FAQ for a discussion of DDD's stability. Basically, DDD in itself is very stable, but relies on a large number of (possibly unstable) components. LessTif's stability is at a tolerable level (and increasing); if your mileage varies, feel free to use a DDD binary linked with OSF/Motif.