dinsdag, december 20, 2005

A quick glance at E17

Those interested in having a look at E17 but not willing to actually start using it, can use the following command to see E17 in a nice little window:
Xnest -ac :1 | enlightenment -display :1

You need to have the xnest package installed or you can install the xserver-xephyr package and use:
Xephyr -ac :1 | enlightenment -display :1


Totally OT: Qemu 0.8.0 is available now! Lots of new features! :-)

MPlayer and accompanying browser plugin

I've added MPlayer and Mplayerplug-in Debian packages to my Ubuntu Breezy repository. They are current snapshots of the CVS repositories. MPlayer can play nearly all audio and video files you can find, and the accompanying plugin makes those playable from within your browser.

apt-get install mplayerplugin

vrijdag, december 16, 2005

Libjingle and E17

Google introduced a new project called Jingle earlier today. I just added Ubuntu Breezy packages of this new stuff to my repository.

Furthermore, I added E17 CVS packages. These should be built automatically using a cronjob.

All goodies available here:
deb http://lumumba.uhasselt.be/takis/breezy ./

woensdag, december 07, 2005

More Debian packages

I've added more packages to my Ubuntu Breezy repository. Besides snapshots of the standard GIT tools and Cogito, I've added qgit, JRTPLib, UIML.Net, Cassowary.Net, XMLRPC for C#, and the WxWidgets .Net bindings.

GIT world domination

The Wine project has just started using GIT! This is the second large/huge project using Linus' Torvalds version control system.

What's incredibly neat is that they've put up the entire project history into the repository starting from 1993! Have a look over here at the webinterface.


The Wine people have put up a Wine GIT Wiki to get people started using GIT. For those interested in using GIT, Jeff Garzik's GIT HOWTO makes an interesting read.


In related GIT news: Artem Khodush wrote a webbased GIT browser.

maandag, december 05, 2005

GIT and Cogito snapshots

I've made some GIT and Cogito snapshots available for Ubunty Breezy. The packages are created from a script invoked by cron, daily.

deb http://lumumba.uhasselt.be/takis/breezy ./

woensdag, november 09, 2005


The 28th of October me and my girlfriend went for a walk in the woods. There are lots of markings in the woods, of different colours representing different routes. The idea was to take a route which would be about 5km and we should complete it in an hour or so. So, we took the orange route, which was 5.5km. Three and a half hours later we finally got out of the forrest... :-) Some kids had apparantly moved some of the markings...

dinsdag, november 08, 2005

CVS to GIT and back

The first step is to import a CVS tree into GIT:

mkdir foobar.git
cd foobar.git
git-cvsimport -p x -v -d :ext:takis@cvs.sourceforge.net:/cvsroot/someproject foobar


We will not make any modifications to the project in that directory, this directory is purely intended to be a GIT conversion of the CVS repository. Instead, we'll create a clone of that archive and do our hackery in there:


git clone -l foobar.git/ foobar-pi.git/


The "l" flag makes sure that git uses hard links instead of copying files, if possible.

Next, we'll make some modifications to the project and commit them using git-commit or cg-commit. As said earlier, we'll make those modifications in the foobar-pi.git directory:

cd foobar-pi.git
... do stuff ...
cg-commit


If new changes are committed in the CVS repository, we need to fetch those in our foobar.git repository, which is kind of a GIT mirror of the online CVS repository.


cd foobar.git
git-cvsimport -p x -v -d :ext:takis@cvs.sourceforge.net:/cvsroot/someproject foobar



Now, we'll pull in the CVS-commits to the clone directory.

cd foobar-pi.git
git pull



This results in a way to have a nice GIT repository of a project being hosted using CVS. Your GIT repository makes it easy to view changes in the project as you've got the full history locally on your system.

To prepare you modifications to be send as a patch, do the following:

git-format-patch origin..HEAD


This will give you a set of files, containing patches, which you can post to a mailinglist or maintainer.

If you've got write access to the CVS-repository, you probably want to commit these patches back into the CVS yourself. To do this, checkout the CVS repository somewhere on your filesystem:

export CVS_RSH=ssh
...


Get into the CVS working directory, and set the environment variable GIT_DIR to point to the .git directory within your git directory:

export GIT_DIR=/usr/local/src/foobar-pi.git/.git/


View the GIT-logs and select which commits you want to push into the CVS-repository:

git-log


Copy and paste the commit-identifier (being an SHA1-hash) and use it as a parameter to git-cvsexportcommit:

git-cvsexportcommit 4a20cbafdf25a141b31a8333284a332d1a4d6072
unset GIT_DIR


Do not forget to unset the GIT_DIR environment variable!

This will output:

$VAR1 = '4a20cbafdf25a141b31a8333284a332d1a4d6072';
...
Ready for you to commit, just run:

cvs commit -F .msg Uiml/uiml.net.build


Repeat the command shown above:

cvs commit -F .msg Uiml/uiml.net.build


The patch which you wrote and committed in your clone GIT tree, is now committed back into the CVS repository.

Now, you can resync your GIT-CVS mirror tree:

cd foobar.git
git-cvsimport -p x -v -d :ext:takis@cvs.sourceforge.net:/cvsroot/someproject foobar


Now, your patches which you committed in the CVS repository, will be pulled back into the GIT-tree which acts as a mirror of the CVS-repository.

Now, you'll want to merge this back into your GIT-clone tree, which is where you do your actual work:

cd foobar-pi.git
git pull


Now, you've come full circle. You patches which were initially committed in foobar-pi.git, were committed to the CVS tree, and pulled back into the mirror GIT tree (foobar.git) and finally back into your clone/working GIT tree (foobar-pi.git). Ofcourse your patches were already in foobar-pi.git and GIT will notice this.

Oneway CVS to GIT

The first step is to import a CVS tree into GIT:

mkdir foobar.git
cd foobar.git
git-cvsimport -p x -v -d :ext:takis@cvs.sourceforge.net:/cvsroot/someproject foobar


We will not make any modifications to the project in that directory, this directory is purely intended to be a GIT conversion of the CVS repository. Instead, we'll create a clone of that archive and do our hackery in there:


git clone -l foobar.git/ foobar-pi.git/


The "l" flag makes sure that git uses hard links instead of copying files, if possible.

Next, we'll make some modifications to the project and commit them using git-commit or cg-commit. As said earlier, we'll make those modifications in the foobar-pi.git directory:

cd foobar-pi.git
... do stuff ...
cg-commit


If new changes are committed in the CVS repository, we need to fetch those in our foobar.git repository, which is kind of a GIT mirror of the online CVS repository.


cd foobar.git
git-cvsimport -p x -v -d :ext:takis@cvs.sourceforge.net:/cvsroot/someproject foobar



Now, we'll pull in the CVS-commits to the clone directory.

cd foobar-pi.git
git pull



This results in a way to have a nice GIT repository of a project being hosted using CVS. Your GIT repository makes it easy to view changes in the project as you've got the full history locally on your system.

To prepare you modifications to be send as a patch, do the following:

git-format-patch origin..HEAD


This will give you a set of files, containing patches, which you can post to a mailinglist or maintainer.

maandag, oktober 31, 2005

Patch author tracking

Some time after the SCO accusations that the Linux kernel would contain code owned by SCO, the Linux kernel people decided to setup a system for tracking the origin of patches. Unfortunately, this system isn't exactly faultproof. I've only submitted about 10 patches out of which three have been wrongly accredited. And because of GIT's use of hashes as identifiers, these mistakes do not get rectified.

Some time ago two patches of mine ended up in 2.6.11.5 stating someone else being the author. Now, I just noticed another patch of mine ended up somewhere in 2.6.x wrongly accrediting someone else.

The White Stripes in Brussels

Saw The White Stripes in Brussels tonight! It was awesome! :)





dinsdag, oktober 18, 2005

Thales


Thales
Originally uploaded by t4k1s.
Just got back from an interesting interview at Thales.

donderdag, oktober 13, 2005

picorb IDL string

picorb now supports the IDL string type.

picorb recursive structs

Just finished adding recursive structs to picorb. For in, inout, out parameters and return types. For the moment I have only tested omniORB's interoperability with the new recursive structs code as I need to modify the testing code to do so. Next up will be the Java code, but that's for tomorrow.

What's next on my schedule are: adding support for sequences, user-defined exceptions, variable length types and arrays.

picorb floppy

I've just created a bootable floppy of which 400K is used, which can be used to demonstrate picorb. GRUB by itself takes 152K on the FAT-formatted disk. eCos and the demo CORBA server implementation take 255K of which eCos takes about 240K. For now, I've taken a basic configuration of eCos, which I intend to strip down. The demo server implements a CORBA object supporting an interface containing 21 methods.

woensdag, oktober 12, 2005

My ORB on PowerPC

My tiny ORB is now also working on my MPC823E embedded board. I've named it "picorb".

dinsdag, oktober 11, 2005

CORBA for eCos

I've just succeeded in getting my new tiny ORB running on eCos within the Qemu emulator! My testsuite of TAO, omniORB and J2SE ORB calls all succeed without a problem. Next up is testing on my PowerPC 823E embedded board to verify the correctness of the endianness related code.

eCos on Qemu

Enabling some tracing output regarding DHCP, BOOTP and in general networking showed that no DHCP requests where being answered by the internal Qemu DHCP server. So, I disabled DHCP for now, and switched to a hardcoded network configuration. This seems to work, the network card gets nicely configured. My ORB runs and binds to the specified interface. But... through the slirp networking code, it doesn't seem to work. Switching to TUN/TAP hasn't given any positive results either.

eCos

I've been playing with eCos lately. As I haven't got enough spare PCs -and because it allows me to work more efficiently- I'm using eCos on Qemu. The network card emulated by Qemu, is the widely support good old NE2000 (both ISA and PCI versions are emulated). Unfortunately, eCos doesn't support that network card, which in one way kinda surprised me, as it was a very widely supported card on desktop OSs. In fact, eCos just supports a few network cards on PC systems. Ofcourse, in another way, this isn't so surprising, x86 and it's related hardware aren't exactly the target of eCos.

Luckily, Antony T Curtis wrote a patch for Qemu to support another network card, the AMD PC-Net II. But, unfortunately, the latest release of eCos doesn't support that card either. Now, I've got three versions of eCos on my system: the latest release from the webpage, the Debian package and the latest CVS tree. After some playing around, I noticed that Qemu kept segfaulting when using the PC-Net card. Some googling turned up a few patches - which BTW have made it into the Xen repository (didn't know Xen was using Qemu) but haven't made it to the Qemu-patches page.

Now, the networkcard is nicely recognised by eCos, but the CVS branch of eCos has some other issues. Ofcourse, this is to be expected, it's the unstable CVS branch where developers push in their latest changes, but still it is kind of annoying. There hasn't been an eCos release in two years or so, so hopefully a new release will follow soon.

For the moment I'm stuck with some annoying issues: When booting Redboot, it just hangs for half a minute. The 2.0 release and also the Debian package with the same configuration on the same system, don't show this behavior. And after this, there's some problem with configuring the network card. I'm not sure whether the networkcard implementation in Qemu is to blame or the eCos driver. And I can't revert to the older eCos release to find out, as it doesn't support the card.
I'm pretty sure my code isn't to blame, as the official test mentioned in the eCos manual (ftp_test.c) doesn't work either. A mail to the eCos mailinglist could help.

maandag, oktober 10, 2005

Tiny CORBA implementation

The 28th of September I started writing a new CORBA implementation. The major goal being minimal footprint while being GIOP compatible. The current footprint of a small Hello World style applications implementing one object containing 21 methods is 16024 bytes as an ELF Linux executable. The textsegment is 13008 bytes for the moment.

Currently, I have the base types working and structs, and I've tested interoperability with omniORB, TAO and Sun's Java ORB of the J2SE SDK. It works on both big-endian and little-endian systems.

I reused the omniORB IDL compiler frontend and implemented a new backend for my ORB, which reliefs me from the pain of writing skeletons manually, which I had to do the first days of the implementation.

I've started porting it to eCos but that isn't quiet working yet.

donderdag, september 01, 2005

My diet


My diet
Originally uploaded by t4k1s.
Im curious whether my current diet will indeed make me lose some weight ;-)

dinsdag, april 12, 2005

Filesystems

I wanted to give GIT a go on the Linux kernel sourcetree, and decided to have a quick look at the performance of the various filesystems available for Linux today. I just tested the ones which were actually usable and drop-dead easy to setup: EXT2, EXT3, ReiserFS, SGI's XFS and IBM's JFS.

The only reason I'm posting this on my blog, is that I wanted to deleted the file I kept this info in and wanted to show it to a few friends who might find it interesting. So, I do not and is not a true benchmark. I just tested one thing, without isolating the to be tested process at all (I was running GNOME at the same time and various other apps). Again, it was just supposed to give me a quick and very rough idea about which filesystem would be fastest for doing lost of extracting of Linus kernel tarballs.

EXT2

pissaris@nyx:/extra/pissaris$ time tar xjf /home/shared/workshop/linux-2.6.7.tar.bz2
real 0m39.594s
user 0m35.705s
sys 0m2.498s

REISERFS

pissaris@nyx:/extra/pissaris$ time tar xjf /home/shared/workshop/linux-2.6.7.tar.bz2
real 0m48.444s
user 0m36.386s
sys 0m5.674s

XFS

pissaris@nyx:/extra/pissaris$ time tar xjf /home/shared/workshop/linux-2.6.7.tar.bz2
real 0m51.189s
user 0m36.353s
sys 0m3.387s


EXT3

pissaris@nyx:/extra/pissaris$ time tar xjf /home/shared/workshop/linux-2.6.7.tar.bz2
real 0m53.162s
user 0m36.205s
sys 0m2.965s



JFS
pissaris@nyx:/extra/pissaris$ time tar xjf /home/shared/workshop/linux-2.6.7.tar.bz2
real 1m20.934s
user 0m36.022s
sys 0m2.652s

vrijdag, maart 18, 2005

DVD-Video authoring

The fastest way I am aware of creating a DVD-Video from whatever moviefile you have lying around is using ffmpeg, dvdauthor and growisofs in the following way:

ffmpeg -i mymovie.avi -target pal-dvd - | dvdauthor -o dvd -
dvdauthor -o dvd -T
cd dvd
growisofs -dvd-compat -Z /dev/dvdrw -dvd-video .


Now, dvdauthor may complain about the aspect ratio, in which case you can explicitly set the aspect ratio using the "-aspect" parameter of ffmpeg.

zaterdag, maart 12, 2005

Another one misunderstanding "NO WARRANTY"

Here's another guy who doesn't seem to get it. Thom Holwerda posted on eXpert Zone:

OSS developers should learn to live with the fact that their software is no longer exclusively used by geeks and enthusiasts -- and if they cannot cope with this, then clearly state that your software is a hobby project

No, no, no. Most GNOME software -of which he's talking- uses the (L)GPL license. The GPL clearly states that there is *NO* warranty.

And furthermore, even if it was no hobby project and someone working in a company got the permission to release the developed software using the GPL license, then again, the users have no right to demand the implementation of the features they want. It's a gift. Accept it or ignore it.

OSNews' Eugenia on Gnome

I was rather unpleasantly surprised by Eugenia's article on OSNews today.


"A feature will be implemented if and only if there is a developer who wants to implement it" (it was later mentioned in the discussion that this is if the developer actually has such a need himself for the feature).


Ofcourse! OSS is software by enthousiasts *for* enthousiasts. OSS developers are not a heap of programming slaves waiting to be commanded by demanding users.


I do not like or agree with the above statement, because I see software as simple tools. Only good tools can ultimately succeed. And I want Open Source to succeed.

The key to the success is more users contributing. Not complaining to developers to work harder for free.


For example, the No1 Gnome feature request, a menu editor that actually works properly, is still not realized after so many years.

So, what are you waiting for? During so many years, I'm pretty sure you could have implemented it yourself.

... users had to wait 6 months to get a checkbox on the preference panel ...

Oh please, it's a darn shame that none of them got to implement it themselves.




Open source devs generally only code whatever they personally need, and that's a huge difference from a commercial application where the "customer" is being asked repeatedly what features he/she needs in the application.

Sure. The customer pays, he gets his satisfaction, the programmer gets his by getting his wage. That's called a job. The other thing is called a hobby, which people do just for the fun of it. To please themselves. A hobby is not about pleasing others. It might please others, but that's not what makes it a hobby.

You may argue that in the second case you pay real money to get such support, but in my book, engineering is engineering. ... So, no matter if something is developed for OSS or for commercial reasons, the principle of engineering remains the same, because in both cases, the software is released out there to be consumed by [innocent] people.

NO. The software is written so satisfy the developers needs and as a free bonus he releases it. If someone might be able to use it, then it's a free gift. If they don't, well, then they are free to ignore it.




So in both cases, there is some responsibility on what the user would expect out of a given application and in the case of Gnome, there are a few millions of users that developers should take into account.

No, the OSS developer who is a hobbist programmer has _no_ responsibility.

Here's a nice quote from the GPL:


NO WARRANTY

11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY
FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN
OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES
PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED
OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS
TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE
PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING,
REPAIR OR CORRECTION.




If these developers really don't want user feedback, they should close down their bugzilla, stop offering their software freely (only use it for their own needs), stop sending press releases out and stop asking for donations on their front page. It's as simple as that.

No, it's not as simple as that. There's users like the one quoted above, who nag and complain about OSS not being good enough, and there's users who are happy to be able to use and modify software for free. If you don't like OSS, you should stop visiting bugzilla, stop donating money and stop reading press releases. I happen to like all the aforementioned a lot.


If the "plain user" entity in the OSS world is such a taboo why the hell would I want to use OSS software?

You don't have to! Nobody is forcing you!


Just because the code is open?

Well, that's what defines OSS, doesn't it?

I don't personally have any real use for the source code (and most normal users don't either). I don't do C anymore (in the case of Gnome)

Well, if you don't like the software and are not prepared to enhance it, well, then that's pity. I suggest you go and find software that'll better fit your needs.


and I can't possibly pay $100-200 per hour to a consultant to add features for me.

If you can't spend $200 for getting a needed feature implemented, you certainly can't spend more on getting a WinPC full of software.

The point of OSS is that you _can_ add the feature yourself or ask someone to do it for you. If you can, good for you, if someone is kind enough to do it for you for free *be grateful*, otherwise pay someone.

Try getting your favorite feature into MS Visual Studio or any other closed source software package. Those companies will _truly_ ignore you. They might do market research to please to commonly asked features, but they certainly won't satisfy you specific needs.

All that "the source code is open, send me patches" it's all looney-baloney for the vast majority of users.

Please reread the "NO WARRANTY" section of the GPL.

You don't seem to get the nice thing about OSS, that is, if I write software to satisfy my needs, there might be other people who've got the same needs. And it will cost me little effort to just make it available. If they need different features, they can add them, and I can enjoy them as well.


However, I wouldn't mind at all paying about $30 per year to a big project like Gnome, but get assured that users aren't cut off from its development process (the current donation scheme does not take care of this).

They are *not* cut off. They are free to contribute. They are just not entitled to command any open source programmer and that seems rather reasonable to me.



They can only quickly describe what they need, and that's that.

Sure, I like to quickly describe what I need too, but apparantly the real world doesn't work like that. I like a Porche 911 for example. And, I don't like paying for it either, nor would I like building it myself. So, please you get me one. For free.



Normal users don't use bugzilla. Only power users & developers do. Besides, no one likes spending time to register.

Ah! So, you don't like devoting 5 minutes of your precious time to register at bugzilla, but you expect other people to devote many hourse of their *precious* time to implement your favorite software for free for you!


What I like to see, is some market research. Approach all kinds of users, put their gripes in line, make a note of their features they really need and evaluate them. Then, create a project plan and distribute the tasks that need to be done to your developers and make sure they deliver what they must deliver.

If you want to force people to work for free for you, that's called slavery.

People will say "that's not how OSS works", but as a user, I don't really care how OSS works.

Yes you do. You are imposing a working method on other people; telling them what to do. For free. For you.

I care about using software that's been properly developed taking users into account rather than purely developers' needs.

The features you need are not there, you are not intending to add them yourself. Okay. But you should not be annoyed that you can't force others to add them for you.


Be careful: I am not asking the OSS developers to implement every little thing that's asked out there, I am asking them to simply take users into account ...

Which does mean implementing every little thing that's asked out there.



I personally find it "deteriorating" for any user to use Open Source software made from such 'lone' developers and not by a company which specifically asks for feature requests or does market research. It would be like the user does not respect him/herself by using a tool that does not do all it could do or all things the user needs it to do

I think your statements show a lack of respect for the people out there who devote their free time to write software and then decide to give it away for free and release the full source code.


To me, software is a tool, nothing more. I am as practical as it goes when it comes to computers. I don't idolize them and I don't have a political ideology about software or hardware (and in fact, I personally take pity to anyone who does -- there's more important things in this world than to be political over bits and bytes).

But you do want a horde a programmers to work for free for you.



If something is open source, that's cool, it's a meta-feature that the closed source apps don't have, and I welcome it and I embrace it.

No you don't. You stated before that you don't care about the source code at all.



But when it comes to software written by hobbyists who don't want to implement anything apart from what they personally need, and systematically ignore their users

Again. They are hobbyists. They enjoy themselves writing software. Like you might enjoy taking a biketrip or whatever. They have every right to ignore their users. Just like you have the right to ignore my problems in my software while you are taking your biketrip.

(apart for their crashing bug reports, which is the only thing that they seem to need them for),

You should be grateful for that. They are trying to fix the problems *you* are having for free for you. Please try to send a bugreport to MS and see what response you'll get.


I refuse to use that software, because I respect myself and my choices.

Good for you. You surely seem to respect yourself. But you surely lack respect for the huge amount of work a huge amount of people made available for you for free.

While this might not be a huge problem for small time applications, it is a big problem when a project that's used by millions has the same attitude towards its users. It's disappointing, to say the least.

Yes. Your selfishness is disappointing.



Oh yes, I'm planning to build a house. I'll quickly describe how I want it, as I don't like to take time to explain clearly on how I want my house to be. But, once you've completed it, it should perfectly fit my needs, otherwise, I'll come to nag and complain about all the things you did wrong. And of course, I don't like paying for it. Oh, and as I'm not such a handy guy myself, I don't want to help out. So, you'll find me in Spain somewhere at the beach.




Update:
After reading Eugenia's article, I noticed this on her profile page:
... so if you have a problem with my spelling and grammar either:
a) do not come back (spare us and save your time too) b) send me a proofread version of the article in question.


This is truly hilarious. So, while she's saying that OSS developers should do everything to please the consumers of their software, she's saying to her readers -the consumers of her articles- that if they don't like her writings the should not read it or fix it themselves.

:-D

dinsdag, maart 08, 2005

2.6.11-bk3

A fix of mine for AMD8111e just made it into Linus Torvalds Bitkeeper tree as can be seen in the Linux 2.6.11-bk3 ChangeLog.

I've got to find a better way to keep track of which patches got into which trees. I have no idea for the moment which ones will need resending...

maandag, maart 07, 2005

My dear pet


poes
Originally uploaded by t4k1s.
Here's a picture of my dear pet. She didn't reach her first birthday.

Grave

I've just finished burying my dear pet.

My pet

During the night of Saterday on Sunday, while I was having a good time at dancecafe La Fete, my pet got hit by a car and died. It is my third pet dying in two years. This one was less then one year old <:-( It was on my way back home somewhere around 04:00 that I saw my pet lying just beside the street.

vrijdag, maart 04, 2005

More patches!

A few days ago I submitted four patches fixing bugs to the Linux kernel mailinglist. A few minutes (or was it hours?) later, the first two of them were confirmed to be needed and today Jeff Garzik pulled my patches in his BitKeeper kernel tree! Yay! :-)

And today the changelog of 2.6.11 appeared containing my previous patch. A pity they used my e-mail address instead of my name for reference... :-)

woensdag, maart 02, 2005

FOSDEM


FOSDEM
Originally uploaded by t4k1s.
This weekend I attended FOSDEM with a few friends of mine. We ended up taking this group-picture with Linux kernel hacker Alan Cox!

Linux 2.6.11

The 2.6.11 Linux kernel has just been released! 5662 files out of 15140 C files have been modified since 2.6.10... and my Prism54 fix was one of them! :-)

Here's the announcement.

dinsdag, februari 22, 2005

Qemu accelerator kernel module

Last week I tested the new -controversial- Qemu accelerator kernel module.

These are the results of the nbench2 benchmark:

Using the accelerator kernel module:
integer: 4.940
floating point: 9.813
memory: 5.994

Running the test on my machine without any emulator involved:
integer: 5.693
floating point: 11.083
memory: 7.137

Quiet impressive imho... of course -just as with VMware- as these "emulators" actually just execute the assembly code, they should approach native performance. It's a whole different story when IO-bound applications are benchmarked.

maandag, februari 14, 2005

allyesconfig

A nice 2.6.11-rc3-mm3 kernel configuration target in these cold winter days is 'allyesconfig'. As the name implies, it will answer yes to every kernelconfiguration question and therefore it will keep your CPU nice and hot.

For the less well-equipped people, there's the 'allnoconfig' which results in a nice and tidy 412K kernel image.

I've been playing with kexec lately and it seems to work really nice. Unfortunately, it hasn't been merged into Linus' kernel yet. I'm getting the necessary patches out of Andrew Morton's big kernel patch (2.6.10-mm3) and merging them with the ADEOS r9c3 patch for the 2.6.10 kernel. Having these patches combined would allow my LiveCD to switch various kernels without doing a full machine reboot.

maandag, februari 07, 2005

Downset

Today, I accidently noticed that Downset -a band I used to listen to 10 years ago- still existed and actually released CD's after their selftitled album. I was looking through Humo's (a local magazine) concertguide, and noticed that they were giving a concert in the cultural center called "Poorthuis" in a city called "Peer" , which is about a half hour drive from my place. So, I went to see 'em, and they played about an hour and a quarter. As I only know their first record, I noticed how hard it is to understand what their saying if you've never seen the lyrics to their songs before. Luckily, they did play a lot of old songs from their first record. More specifically, they played "Breed the killer", "Downset", "Ritual" and ended the show with "Anger".

woensdag, februari 02, 2005

TOOL lyrics

As I keep relistening to old TOOL albums, and keep losing the links to the lyrics, I'll post 'em here, as a kind of bookmark :-)

http://toolshed.down.net/lyrics/

Elgoibar

Last week the OCEAN project meeting took place in Elgoibar near San Sebastian and Bilbao. I took the opportunity to visit the Guggenheim museum and experience the local habits such as the delicious pintxos and the marvelous ciderhouse-style diner.

Pintxos are similar to Spanish tapas; they are small portions of food, composed of various components, such as fish, meat and vegetables. You eat them with a glass of wine while you're standing in a pub. After finishing your glass in about 10 minutes, you continue to the next pub, where the ritual reoccurs. The friday I arrived at San Sebastian, I had so many pintxos and accompanying glasses of wine, that I can't even recall how many and what I ate :-)

On monday evening, the evening after the first meeting, we went to a ciderhouse which was a apparantly a fairly typical place to have dinner at. The owner of the place would announce at regular intervals that he would open a new cidercase which was the signal for all people to go up and refill their glasses. The food was brought to the tables, where people could take a bit of the dish and eat it with bread. They served various types of fish, prepared in various ways and a truly huge steak.

Jon, one of my collegues explained me how to let the cider drop into the glass from about a meter from the tap but still the local girls insisted on correcting my technique :-)

The local people were all very friendly and there was a very warm and safe atmosphere all over.

This is definitely a place I'd like to visit again.

maandag, januari 10, 2005

Kana

Last Saturday I memorised the Hiragana characters. I'm able to recognise and read them all and their combinations. I started memorising the Katakana characters on Sunday, but since Sunday is family gathering day, I didn't have enough time to spend.

vrijdag, januari 07, 2005

RTAI simplifications

The RTAI developers are preparing to release a new version based on the magma branch. My first impressions are quiet good. They seem to have removed a lot of the unnecessary stuff, which makes the project appear somewhat less complicated.


The number of header and source files in the RTAI Kilauea tree determined with a simple find: 574. In the Vesuvio tree: 569. And finally in the new Magma tree: 325. Which shows that some serious cleanup has occurred.