union, struct, and bit fields in C and C++

It took me some long time to comprehend the issues involved in using unions, structs, and bit fields in C and C++.

First, for the definition of unions and structs, take a look at this link:
http://publib.boulder.ibm.com/infocenter/comphelp/v8v101/index.jsp?topic=/com.ibm.xlcpp8a.doc/language/ref/strct.htm

Unions could be used together with structs to reinterpret the same location in memory with different interpretations (similar in spirit to the C++ reinterpret_cast). For examples on this see:
http://www.anyexample.com/programming/cplusplus/example_of_using_union_in_cplusplus.xml

C++ extends the unions and structs, for example see:
http://www.yolinux.com/TUTORIALS/LinuxTutorialC++Structures.html

Bit fields are used to define special-sized integers. This is useful for packing data (e.g. flags) together into one memory location. This, however, is a bad habit and should be avoided as the compiler-generated code to access bit fields is usually inefficient! For more examples on this, see:
http://en.wikipedia.org/wiki/Bit_field
http://publib.boulder.ibm.com/infocenter/macxhelp/v6v81/index.jsp?topic=/com.ibm.vacpp6m.doc/language/ref/clrc03defbitf.htm
http://www.cs.cf.ac.uk/Dave/C/node13.html
http://www.daniweb.com/forums/thread55765.html
http://blogs.msdn.com/oldnewthing/archive/2008/11/26/9143050.aspx
http://osr507doc.sco.com/en/tools/ANSI_F.3.9_Structs.unions.html

Finally, unions and structs could be anonymous! You do not have to give a name to a union. In this case its members become directly accessible within the same scope. Wikipedia calls this "untagged union". For an example see:
http://www.velocityreviews.com/forums/t286529-union-with-anonymous-struct.html

Debugging, Profiling, Memory Leaks and more!

I have been searching for open source tools to assist in detecting memory leaks and application profiling. Also, static code analysis was another goal.

For static code analysis I found:
CppCheck (http://en.wikipedia.org/wiki/Cppcheck)

I used it to its full power by invoking (at the top level of the project):

cppcheck --recursive --all --errorsonly --style

Also, SPlint (http://en.wikipedia.org/wiki/SPlint) seems promising, though it is for C code only (no C++).

A comprehensive list of static code analysis could be found in Wikipedia:
http://en.wikipedia.org/wiki/List_of_tools_for_static_code_analysis

About application profiling, the GNU Profiler (gprof) does a good job. It is light-weight and has support in GCC. This is an excellent article about using gprof:

• Programmer's Toolkit: Profiling programs using gprof(http://linuxgazette.net/100/vinayak.html)

Of course, you can best visualize the output of gprof using the (kprof) KDE front-end.

Memory leak detection has many options. Of the best is using Valgrind(http://valgrind.org/). However, I found an interesting and promising feature that has been added to the gcc. This feature is called Mudflap. It is easy to use as a compiler flag and adds memory debugging support directly into your application using libmudflap. More interesting information could be found here:
http://gcc.gnu.org/wiki/Mudflap_Pointer_Debugging
http://gcc.fyxm.net/summit/2003/mudflap.pdf
It is so promising that it will be integrated into the QNX IDE:
http://qnxdevtools.blogspot.com/2008/02/mudflap.html
It can print the memory leaks on program exit via the (-print-leaks) option

Memory debugging tools are so many. There is a list on Wikipedia article:
http://en.wikipedia.org/wiki/Memory_debugger

The task of memory leak detection is achievable using lightwieght tools. These tools basically replaces the default C library malloc() and free() with special debugging implementations. Notable is mtrace(), the memory debugger built in the GNU C library:
http://en.wikipedia.org/wiki/Mtrace

A good article about memory leak detection in C++:
http://www.linuxjournal.com/article/6556

And finally, all what you can dream of are collected in one article, in the GRASS wiki here:
http://josef.fsv.cvut.cz/WIKI/grass-osgeo/index.php/GRASS_Debugging


Hope you find this useful!

Guide to start using Linux in ARABIC

This an excellent piece of work. This book, written in ARABIC, will get you started with using Ubuntu Linux in no time! Get started now:
http://ia310803.us.archive.org/2/items/Practical_Guide_to_use_ubuntu_linux_Arabic_344/Practical_Guide_to_use_ubuntu_linux_arabic.pdf

Egyptian Universities Library!!

I couldn't believe it... Check the EUL website NOW
www.eul.edu.eg

The possesive 's

Very informative!

http://www.bbc.co.uk/worldservice/learningenglish/grammar/learnit/learnitv57.shtml

Some stuff about finance and investment

This post may seem a bit strange. I had an opportunity to read some articles about some weird stuff of the financial world! Those guys of finance are crazy, and we are to blame that we leave our destiny in the hands of such a bunch of crazy financial people! Here are some links about possible ways of investment:

• http://ezinearticles.com/?Debentures-Vs.-Stocks-And-Bonds&id=392416
• http://transformyourself.com.jm/faq_fixed_income_investing.html
• http://financial-dictionary.thefreedictionary.com/Debenture+stock

Comics about SW Development

I saw this link posted by a friend, and I followed it. I'm posting it here because it is very realistic and happens to anyone working in the SW industry.

http://contikistrip.kjempekjekt.com/index.php?p=1

Buffer Overflow Exploits

I have recently read the news about Google Chrome Buffer Overflow vulnerability (see this link). Admittedly, I was ignorant enough not to understand how a buffer overflow can cause severe security breaches! Wikipedia came to the rescue, as well as some research into the topic of Assembly-language programming. At last I can understand how SEVERE such an attack could be!

You can dig for it too ;-)
check the Wikipedia article,
http://en.wikipedia.org/wiki/Buffer_overflow

and the excellent book of Programming from the Ground Up, for much better understanding of Assembly:
http://savannah.nongnu.org/projects/pgubook/

Enjoy!

Synthesis Lectures on Biomedical Engineering

At last, lecture notes on Biomedical Engineering =)

http://www.morganclaypool.com/toc/bme/3/1

GNU/Linux from eight miles high

This an excellent article released recently by embedded.com:

http://www.embedded.com/design/opensource/208404045?pgno=1

The article is a bit advanced. But anyone with good understanding of computer architecture should be able to go through it!

A Good article about ECG Signal Processing from LabVIEW

At the time I visited this article it was rated 4 out of 5 which is very good!

http://zone.ni.com/devzone/cda/tut/p/id/6349

GUC Students' Blogs

Here are some blogs by students of the GUC. Currently they are all in Germany for their bachelor projects. I am really proud that I had a chance to work with such amazing and wonderful people! Enjoy:

http://husfahmy.blogspot.com/
http://thoughtsandtips.blogspot.com/
http://lewjee.blogspot.com/

Introduction to PhysioNet

http://www.physionet.org/NIH/NIH-PhysioNet-Intro.ppt

Add Circular Scrolling (and more!) to your laptop touchpad

From the Chinese Linux University!

http://www.chineselinuxuniversity.net/articles/13049.shtml

Add to /etc/X11/xorg.conf (or whatever your xorg.conf is at)

Section "InputDevice"
Identifier "Synaptics Touchpad"
Driver "synaptics"
Option "SendCoreEvents" "true"
Option "Device" "/dev/psaux"
Option "Protocol" "auto-dev"
Option "CircularScrolling" "on"
Option "CircScrollTrigger" "3"
EndSection

Be sure to have Synaptics added to the Server Layout

Section "ServerLayout"
Identifier "Default Layout"
Screen "Default Screen"
InputDevice "Generic Keyboard"
InputDevice "Configured Mouse"
InputDevice "Synaptics Touchpad"
EndSection

Dealing with Debian source packages

I found this interesting post, quoting:

apt-get build-dep glibc
apt-get source -b glibc

should work

or, if you already have the build-dependencies and source files
downloaded (.orig.tar.gz, .diff.gz, and .dsc):

dpkg-source -x glibc*.dsc
cd glibc-*
make -f debian/rules build

randolph
--
Randolph Chung
Debian GNU/Linux Developer, hppa/ia64 ports
http://www.tausq.org/

The dpkg-source utility is found in dpkg-dev package.

Compiling over NFS

Ever needed to compile over exported NFS system in Linux. I needed it to compile some programs on my PC using a compiler in a virtual box ;-)

On the NFS server

add the following line to /etc/exports (Debian/Ubuntu systems)

/path/to/exported/directory *(rw,no_subtree_check,insecure,no_root_squash)

On the client machine (the one that will compile)

add the following line to /etc/fstab

10.0.2.2:/path/to/exported/directory /path/to/mnt/point nfs user,rw,exec 0 0

Note: replace 10.0.2.2 with the server IP or hostname

Warning: this setup is TOTALLY INSECURE!!

but who cares ;-)

I need to visit this blog later on

http://blog.akkaya.de/blojsom/blog/jpabel/

Contributors Needed!

Please contribute to this new emerging WikiBook

Biomedical Engineering: A Roadmap

Don't forget to add your name to the authors list ;-)

Israel hospital to deploy largest RFID installation in Middle East

http://healthcareitnews.eu/content/view/1068/43/

What is linux-gate.so.1?

Check this excellent article by Johan Petersson. Be sure to equip yourself with good knowledge about ABIs before reading!!

http://www.trilithium.com/johan/2005/08/linux-gate/

BUG! -- the Biomedical Unix Geeks club

Biomedical Engineering make extensive use of software. Signal processing algorithms and other aspects of the biomedical engineering needs extensive study of computer science and software engineering. Unix platform is of special importance in the medical field due to its stability and ubiquity.
This club is for all biomedical engineers who love Unix, Software Engineering and Computer Science.

IMHO, Unix provides the best environment for the study of algorithms and computer science. The Unix tools and development also implements important point of the software engineering. Hence, came the idea of this club.

Please share your ideas with me on how this should go!

Puppy Linux trials with the eBox series

eBox 2300SX
http://www.murga-linux.com/puppy/viewtopic.php?t=21864

eBox 2300
http://www.murga-linux.com/puppy/viewtopic.php?search_id=21059625&t=13722
http://www.murga-linux.com/puppy/viewtopic.php?t=14719&search_id=21059625

eBox 4300
http://www.murga-linux.com/puppy/viewtopic.php?search_id=21059625&t=26982

Mixing C and C++

Did you ever think of creating C code suitable to be linked from C++ code !!

Just check:
http://developers.sun.com/solaris/articles/mixing.html
http://www.parashift.com/c++-faq-lite/mixing-c-and-cpp.html

e-dsp.com

I just arrived at this amazing website for DSP and Electronics:

http://www.e-dsp.com/

Computer Vision in Hardware

This article is a very nice implementation of computer vision systems in hardware

Automotive vision system recognizes road signs: Part 1 - Basic functions

Name Mangling

Did you ever hear about Name Mangling? I didn't myself! But it is important to take a look at :)

Cross Compiling the hard way...

This tutorial contains many hints about cross compiling. You'll never need to understand all this stuff, but it is worth reading anyway! In practice, cross compilers are already built and you'll just use them!
http://www.landley.net/ols/ols2007/tutorial.txt

Predefined Macros in gcc

From the gcc man page:

-dCHARS
CHARS is a sequence of one or more of the following characters, and must not be preceded by a space. Other characters are interpreted by the compiler proper, or reserved for future versions of GCC, and so are silently ignored. If you specify characters whose behavior conflicts, the result is undefined.

M Instead of the normal output, generate a list of #define directives for all the macros defined during the execution of the preprocessor, including predefined macros. This gives you a way of finding out what is predefined in your version of the preprocessor. Assuming you have no file foo.h, the command

touch foo.h; cpp -dM foo.h

will show all the predefined macros.

Compiling 32-bit Applications on 64-bit Linux Hosts

You need to install the gcc-mutlilib package. Then:
$ gcc -o hello -m32 hello.c

the -m32 flag tells the compiler to target the 32-bit architecture instead of the 64-bit architecture

Also, you may need to take a look at this looooooong discussion:
http://ubuntuforums.org/showthread.php?t=24575

For kernel compiling, you will need to explicitly specify the architecture:

$ make ARCH=i386 menuconfig

$ make ARCH=i386 bzImage

Happy compiling :)

*** Returning to this issue with Makefiles

just run:

$make CFLAGS=-m32 LDFLAGS=-m32

this adds the -m32 option to both the compiler flags and the linker flags

iwl3945 Finally Working :)

This post saved me from throwing away the Linux 2.6.24 kernel :)
http://ubuntuforums.org/showthread.php?p=4612681

PDF Multiple Page Per Sheet in Linux

To print multiple pages/sheet, you'll need the `pdfjam' package installed. Then,

$pdfnup --nup 1x2 input.pdf

Open Source Alternative to Commercial Software

http://downloadpedia.org/Open_Source_Alternative_to_Commercial_Software

My Ubuntu Sytem borke :(

I spent a scary night yesterday trying to fix my Ubuntu installation. I was a victim for the bug related to libc6. The problem is that after updating the system, the new libc6 package was broken and it rendered the system in a total mess :((

Almost every single application is depending on the standard C library (glibc/libc6/GNU C Library). The fact that the libc6 is not functioning means the system is now totally unstable! Error message of the sort 'null pointer reference' began to show everywhere. I can't start any kind of terminal. Even logging into the Virtual Consoles failed and just thrown some 0x777FFD... addresses on me. I had no access to any package manager and I cannot undo the buggy installation. Even starting a browser to search for the problem failed! I had no other option but to restart and login using a liveCD.

After a lot of searching for the problem, I came across this bug description.
I have learnt a lot :)

To break the normal boot process you need to boot with the kernel option:
break=bottom

This puts you into the initial ram filesystem prompt (initramfs) just before starting the init-bottom scripts

To unpack a .deb archive:
ar -x *.deb

This leaves you with two tarball control.tar.gz and data.tar.gz. I used this to unpack the libc6 deb archive by hand and copy the necessary data.tar.gz files into their correct locations; thus replacing the buggy libc6 installation with a correct installation.

When you 'chroot' into your Ubuntu installation and get the message: 'Unable to execute /bin/bash: Permission Denied'. And even 'chroot /path/to/your/rfs /usr/sbin/apt-get' fails for the same reason. Then most probably the 'ld*' files in /lib/ and /sbin/ldconfig are not give the execute permission.
'chmod +x /lib/ld*' and 'chmod +x /sbin/ldconfig' corrected the problem.

Notice that ld* is the dynamic link loader and is needed by almost every executable file in your system. Also, 'ldconfig' is needed by dpkg, apt, aptitude and all package managers to be able to process your packaging system.

Imperas

Open Virtual Platforms (OVP) portal

http://imperas.demonweb.co.uk/

Best Free Software

Best Free Software ;-)
http://www.pcmag.com/article2/0,2817,2260070,00.asp

Probability ABC

Nice Quote from:
http://blogs.jhu.edu/sukye/2008/probability-abc/

``A priori probability is the probability estimate prior to receiving new information.

Posterior probability is a revised probability that takes into account new available information. For example, let there be two urns, urn A having 5 black balls and 10 red balls and urn B having 10 black balls and 5 red balls. Now if an urn is selected at random, the probability that urn A is chosen is 0.5. This is the a priori probability. If we are given an additional piece of information that a ball was drawn at random from the selected urn, and that ball was black, what is the probability that the chosen urn is urn A? Posterior probability takes into account this additional information and revises the probability downward from 0.5 to 0.333 according to Bayes’ theorem, because a black ball is more probable from urn B than urn A.

Bayes’ Theorem:
Bayes theorem is a formula for revising a priori probabilities after receiving new information. The revised probabilities are called posterior probabilities. For example, consider the probability that you will develop a specific cancer in the next year. An estimate of this probability based on general population data would be a prior estimate; a revised (posterior) estimate would be based on both on the population data and the results of a specific test for cancer.
The best way to understand the terms is to look at an example. Consider a screening test for intestinal tumors. Let Ai = A1 = the event “tumor present”, “B” the event “screening test positive” and “A2″ the event “tumor not present” with no further A’s.
If you have a tumor, the screening test has an 85% chance of catching it — P(B|A1) = .85. However, it also has a 10% chance of falsely indicating “tumor present” when there is no tumor P(B|A2) = .10. The probability of a person having a tumor is .02 P(A1) = .02.
If the screening test is positive, what is the probability that you have a tumor?
.02*.85/(.02*.85+.98*.10)
= .017/(.017+ .098)
= .148''

Take a look here:
http://www.jhu.edu/nthakor/

My Ubuntu :)

I'm currently using this server for my Ubuntu Hardy Heron upgrades:

ftp://ftpserv.tudelft.nl/pub/Linux/archive.ubuntu.com/

Survive for 18 seconds ;-)

http://www.iol.ie/~dluby/escape.htm