Sumanadak's Tech Blog

Learning Grails

2011-02-20T16:38:00.002+05:30

While studying IR technique, I was searching a fast,quick web development framework based on Java. After long research and trying with couple with web framework , I settled down with Grails.

I got impression and started building small application based on this framework. Grails is based on groovy dynamic languages. Groovy, you can say a superset of JAVA language. It runs under same JVM what you use for JAVA. So JAVA and groovy programs can run together. Here i am giving couple of link for tutorial. Grails framework is supported by springSource, creator of Spring framework and it supports well. It also supports all modern Object relational Mapping tool like hibernate. A based way to learn to watch screen-cast videos available in grails website and try along with that.

1. grails website

2. groovy website

3. git installation

4. screen-cast tutorial

Information Retrieval study

2011-02-03T20:42:00.004+05:30

Well, It's time to learn . Currently I am reading Information Retrieval Book from this Link.

I feel better while reading this book ,well written.This time my blog post will be quite less as I am more on reading book.Update soon.Meanwhile you can check out some great link about IR.

Resource:

IR Resource

Web Site:

1) http://infolab.stanford.edu

2) http://www.isical.ac.in/~mandar/

3)http://infolab.stanford.edu/~wangz/project/imsearch/review/JOUR/datta.pdf

4) http://eil.stanford.edu/publications/charles_heenan/AcademicInfoRetrievalResearch.pdf

4)http://research.yahoo.com/Search_and_Web_Mining

Started Again..

2011-01-27T13:03:00.002+05:30

After three years!! Oops so many things have been changed. Technology,Me,You.....

After long gap, I realized more to have a blogging. I changed my interests too. I am still interest in Open Source but the fields got change. Currently my interests includes Information retrieval,social data mining and search technology.

I love open source and would always like to work on open source software(Use+Contribute).Hope I would continue from now onwards.

I am back....

2008-08-09T21:19:00.002+05:30

Since November,2006 i didn't write any single post in this blog. I am really sorry to me.
Several things has been changed during this two year.I moved chennai to kolkata,got married,changed my employer.Currently i am in Tata Consultancy Services,Convergence and Multimedia Innovation Lab,Kolkata.I have finished two projects - Rule Engine for RFID based Vehicle Management System and Remote upgrade for Intaractive Set Top Box.Now i am working with real time video streaming in J2ME. I hope i will be posting more and more tech stuff on this blog.

[off-topic] Something about MINIX

2006-11-14T14:42:00.000+05:30

Nearly 4 months back, i came to know to MINIX 3 has been released as a free software. I downloaded this ran also with help of QEMU simulator.The latest relase has been made on 29th May 2006. I think you all people know about this teaching OS. It has been made by Vrije Universiteit, Amsterdam,headed by professor Andrew S. Tanenbaum. Most of the function is written by him. But that time i didn't look much inside of this OS or kernel part. But two days back, one girl from Haldia Institute of Technology , told me that she wants to build a OS like MINIX..then i though let's again visit MINIX. After reading some document , i got realy some good things in my mind realy i am realy getting love with MINIX achitecture. As all of you know MINIX kernel is based on Microkernel architecture. "A microkernel is a minimal computer operating system kernel providing only basic operating system services (system calls), while other services (commonly provided by kernels) are provided by user-space programs called servers. Commonly, microkernels provide services such as address space management, thread management, and inter-process communication, but not networking or display for example."(taken from wikipedia).
Now i am quite familar with Linux kernel and it's architecture is based on monolithic(See wikipedia for more information) kernel architecture. Though it has some good facility for modular programming.
The question which i want to figure out is, most of the device driver has been implemented inside the linux kernel though the biggest device driver (graphics driver ) is in user space. I couls see size of the driver code in linux kernel is 112 MB whereas the Whole kernel itself has 269MB. So nearly more than 50% code in kernel is driver code!!!!.And most of the developers want to make driver in user space and effort is user space device driver.The majority of bugs linux kernel are in device drivers. if you can move out these driver form kernel space to user space i think we can minimise the bugs. I think it would be applicable for Networking also. But their would be some issue like performance and latency. Still i love to see device driver in user space. Why should i take care about kernel when i am making one driver for device? . Let kernel to do that. Thecommunication between driver and kernel should be designed very carefully so that time required for communication would be minimized over a period of time.That's all..That's why i fall love in MINIX. It has fare architecture to do that.I would like to invest some time on that architure and also would like if someone can come with idea like a "conceptual Architecture of MINIX kernel"...Well this post is going to be big. So i decided to stop here, journey with MINIX kernel and hope to post another article in "User space device driver". Those who are realy interested in MINIX kernel please contact me.. And more ever there are very easy chances to be a main developer!!!!or you may say kernel haker!!! in MINIX world.
visit www.minix3.org

How to enable 3D Desktop

2006-11-10T14:14:00.000+05:30

To enable 3D accelaration in your Destop, You should have some supported graphices card. If it is semi supported then it's also O.K... Now you should have some necessary packages , These are compiz, xgl, xgl-hardware-list, gnome-session and libwnck packages. Just install these packages. Next run gnome-xgl-setting from the command line and enable 3D Desktop option. After than just logout and login again. You will get 3D enabled Desktop. Remind one thing it's for GNOME Desktop not for KDE. To enable 3D Desktop in KDE you have to do configuration. Please follow this link
Link

Increasing HZ value in Linux kernel

2006-10-17T11:26:00.000+05:30

i just came accross to know what are the effect if i increase HZ value ( CPU ) in linux kernel? . I found some good documentation. In this post, i am giving only the link which i found during my research. I will throw more light on that topic in next post. Have a look and think about it..
http://www.kernel.org/pub/linux/utils/kernel/cpufreq/cpufreq.html
http://ubuntu.wordpress.com/2005/11/04/enabling-cpu-frequency-scaling/
http://kerneltrap.org/node/464
http://lxr.linux.no/source/Documentation/cpu-freq/

Happy Hacking..

Know rss field in ps command also mm_struct in linux kernel

2006-09-21T15:54:00.000+05:30

This post is for intended to be answer of one mail from kernelnewbies..Let me tell the question which was asked by someone..
"Task_Struct->mm->RSS different from ps -aux RSS kernelNewbiesHi,
I'm developing kernel modules for educational purposes, and I was trying to find out the amount of memory used by a process. When you go ps -aux it shows for example that a certain process is using 1.7% of memory and has an RSS of 16000, however when I go to the concerned task_struct and printk the rss from mm, I get a value which is in hundreds, compared to thousands? For finding out the memory three items in the mm_struct look interesting to me: RSS, Total_VM and vm_lock, will adding them
together give me the memory used by a process?
regards Irfan "
Rik Van Riel ( i think u knows this guy) if not so please visit his home page http://www.surriel.com/ If you want to see him visit this site http://www.redhat.com/magazine/014dec05/features/xen/Xen_Rik.jpg
Frankly speaking i am little bit fan of this guy.Anyway he has given the answer "The RSS reported by ps is in kilobytes, right ?The kernel counts in pages, which are larger".
Before that i also didn't look at the rss filed value and also franklt speaking i also didn't know what is the exact meaning of this value. I started hunting(should i say hacking!! Ok little bit) the answer..ofcource Rik's clue has given me the thought and i found the answer but still i am not confirmed which i have given explation regarding this question.i am waiting reply from Rik...i will update this blog if I and You find any wrong explanation or information. Please feel free to write comments. I am directly pasting mail which i have send to kernelnewbies mailing list ...
Hi Irfan,

On 9/21/06, Irfan Habib wrote: Hi,
" man ps "says me that
"The SIZE and RSS fields don't count some parts of a process including the page tables, kernel stack, struct thread_info, and struct task_struct. This is usually at least 20 KiB of memory that is always resident. SIZE is the virtual size of the process (code+data+stack). "
and also rss is for "resresident set size, the non-swapped physical memory that a task has used (in kiloBytes). " it means that the total size of the pages currently in memory.Which kernel r u using? I couldn't get rss fileld in mm_struct. I am using 2.6.17. I could get
mm_counter_t _file_rss;
mm_counter_t _anon_rss;
unsigned long hiwater_rss; /* High-watermark of RSS usage */
i think _file_rss and _anon_rss are for holding the value of the total number of resident pages in memory.So size of (_file_rss+_anon_file) should be equal to rss fileld value in ps command for particular process.Now let me prove:
my system i can get output for ps -aux .
[root@ajit pages_program]# ps -aux
Warning: bad syntax, perhaps a bogus '-'? See /usr/share/doc/procps-3.2.6/FAQ
USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
root 1 0.0 0.1 1984 660 ? Ss 12:11 0:00 init [5]
root 2 0.0 0.0 0 0 ? SN 12:11 0:00 [ksoftirqd/0]
root 3 0.0 0.0 0 0 ? S 12:11 0:00 [watchdog/0]
[.....]
Now i can see that RSS is 660kb. In my system i am using 4kb page size. So total number of memory resident page is =660/4=165.Now i have written a small kernel module .....
#include
#include
#include
#include
static int pid_mem ;
MODULE_LICENSE("GPL");
MODULE_AUTHOR("Suman Adak");
module_param(pid_mem, int, 0);
static void print_no_pages(struct task_struct *task)
{
struct mm_struct *mm;
mm = task->mm;
printk("\nThis mm_struct has %d vmas.\n", mm->map_count);
printk("No of pages = %ld",mm->_file_rss + mm->_anon_rss);
}
static int pages_load(void){
struct task_struct *task;
printk("\n the process id to look up as %d.\n", pid_mem);
for_each_process(task) {
if ( task->pid == pid_mem) {
printk("%s[%d]\n", task->comm, task->pid);
print_no_pages(task);
}
}
return 0;
}
static void pages_unload(void)
{
printk("\nbye bye\n");
}
module_init(pages_load);
module_exit(pages_unload);

i just cut my dmesg here
root@ajit pages_program]# /sbin/insmod pages.ko pid_mem=1
[root@ajit pages_program]# dmesg
kobject pages: registering. parent: , set: module
kobject_uevent
fill_kobj_path: path = '/module/pages'
the process id to look up as 1.
init[1]
This mm_struct has 23 vmas.
No of pages = 165
I could see that process with pid 1 , It has 23 vmas and total no memory resident pages is 165, which is equal to rss field in ps command.i think you understand..So rss and size field doesn't give the exact size of the process...
Rik ,correct me if i am wrong., please clarify me ...How do we set value for hiwater_rss?
Thank

How to find PAGE SIZE in linux kernel

2006-09-21T13:49:00.000+05:30

Sometimes i was finding what is the PAGE SIZE of my linux kernel? . I knew that it could be 4KB,8KB,16KB or 64KB. But the the real question was , how do i find the value? After lot of serching, i came to know the following command from Rik Van Riel (kernelnewbies mailing list)..
[suman@ajit ~]$ getconf PAGE_SIZE
4096
it says that i am using 4KB PAGE SIZE. Now i went to find out where the PAGE SIZE has been wriiten into the source code. I found it /include/asm-i386/page.h in my kernel-source(2.6.17) directory.My system architecture is i386 machine. I could see
#ifndef _I386_PAGE_H
#define _I386_PAGE_H
/* PAGE_SHIFT determines the page size */
#define PAGE_SHIFT 12
#define PAGE_SIZE (1UL << PAGE_SHIFT)
i see that PAGE_SIZE depends on the value of PAGE_SHIFT. i yet to see if i change the PAGE_SHIFT value,what would be the impact of my kernel? Let see..i will be informing....

Also i came to know that PAGE SIZE varries on architecture to architecure. Here is the code snapshot for ia64 archi.

#ifndef _ASM_IA64_PAGE_H
#define _ASM_IA64_PAGE_H
* PAGE_SHIFT determines the actual kernel page size. */
#if defined(CONFIG_IA64_PAGE_SIZE_4KB)
# define PAGE_SHIFT 12
#elif defined(CONFIG_IA64_PAGE_SIZE_8KB)
# define PAGE_SHIFT 13
#elif defined(CONFIG_IA64_PAGE_SIZE_16KB)
# define PAGE_SHIFT 14
#elif defined(CONFIG_IA64_PAGE_SIZE_64KB)
# define PAGE_SHIFT 16
#else
So ia64 machine can use 4KB,8KB,16KB or 64KB depends upon the configuration during compilation.

how to configure multiple internet coneection for single PC or LAN

2006-09-06T16:47:00.000+05:30

Check out this link..
Routing for Multiple uplinks/providers

Linux kernel linked List Implementation

2006-09-01T14:18:00.000+05:30

Linked list is a very important data structure which allows large number of storage with efficiant manipulation on data.Most of the kernel code has been written with help of this data structure.So most of the kernel code (data structure) complexity depends on the implementation of that data structure. You could write your own code but the lack of bad design could give you bad performance. A good operating system should give not only stability but also good performance. So i was eager to know how linux implements linked list. Is it circular singly linked list or doubly linked list or circular doubly linked list ?
does It has some libray file where most of the functions would have been declared? Yes my search ends at
"/usr/src/linux/include/linux/list.h".
I could see all types of fuctions are declared here. Before discuss all these function i should tell somthing about how linux kernel adopts strategy to build efficiant implementation.
It has a unique style to implement the traversing,adding,deleting node into a circular link list.As it is circular , you don't need think about head node and last node.You just simple pick up a node and follow the pointers untill you get back the same original.This approch removes the extra head node implementation.So each routine simple needs a pointer to a single element in the list and it could be any element. You may be wondered to see the implementation.Normally we used to declared linked list as like
struct my_list{
int data,
struct my_list *prev;
struct my_list *next;
};
But if want to adopt linux implementation style then you could write like that
struct my_list{
struct list_head list; //linux kernel list implementation//
int data;
};
struct list_head { //Declared in list.h file
struct list_head *next;
struct list_head *prev;
}

Now note the name of the structure "list_node". Linux developers has taken this name in mind from the fact that there is no head node in the list. All node could be acted as a head node .The next pointer points the next element and prev pointer will point the previous pointer.Thanks the linux claver list implementation.You could forget about firat node and last node concept.Just just compare list as a big cycle with no first and last.There are many clever implementation of each function. Please go through the list.h.It is well documented.I thnik i couldn't make better documentation than this. I am sure , you can digest it within one hour.
Now some big questions may come:-)

1.) why should i learn linux kernel linked list implementation strategy though i could write my own code or i can take linked list library from somewhere?

2) How can i use linux linked list library in my user space application? What would be the benefit that i could get if i use this library?
Yes , I am trying to give answer of these two question. If you are not satisfied with my answer . please leave a comment...
Answer No 1:
1) If you want to see yourself as a future linux kernel hacker,you must learn this strategy. I can give a prof from a book "linux Kernel Development" by renowned kernel hacker Robert Love.See the Apendix-A
"All new User must use the exiting interfase,we are serious about this, do not reinvent the wheel"

2) Create various type of Data Structure: You can build any data structure as in your mind.
3) Portability: Otherwise it would not have been placed into main linux kerenl source tree.
4) East to learn. readibility: Yes, It is easy to learn, you can learn all function with one hour rigorious study.
5) Complexity: You would be wonderd that all these functions are O(1), means they execute in costant time regardless of the size of the list.

Answer: Q.No.2:
With some minor modification, You could include this list.h header file into your userspace application. It's a single header file. very handy! You need to do
a)Remove #ifdef __KERNE__ and its #endif
b)Remove all #include line
c)Remove rcu related functions(many of them there)

Yes , I am working on this list to publist my own list.h header file. I have already implemeted generic implementation of my own circular doubly linked list and have written same program with help of "list.h" file and compare

You could also ............

P.S: The kernel uses linked list to store the task list, Each process's task_struct is an element in the linked list.
look at this link...You will automatically understand how many times list_add function have been called:-)

Resource: Section
Yes i used to read from good article.Try to understand from the good article.And ofcouse i should share with you.Have a look
Ref 1: "Linux Kernel Development"-Robert Love Appendix -A
Ref 2: Linux kernel linked list for user space
Ref 3: Linux Kernel Linked List Explained
Happy Hacking...

[off-topic]Hope One Day I Would Be..

2006-08-31T18:47:00.000+05:30

I am alwayes very much keen to do my Ph.D in Operating System( It could be Real Time Opearating System also). I don't know when i will get chance to fulfill my dream. But i should do..i must do.. so now i have to prepare myself. But How? While I was thinking , i thouht first i have to find out what should i need? ...i could prepare a list like ..i need extensive knowledge , i have to do hard work, i have to read countless books or I have to ........................................................But first i could guess that , i should know how to do research ...while i was searching , i got a fantastic link which i should share with you...Have a look on this link
Advice on Reserch and Writing
You could visit this page. It's one of my research interest area...
CS Columbia University's Professor Erez Zadok Page

How to install GNU/Linux without any removal media

2006-08-29T15:45:00.000+05:30

If you want to install GNU/Linux withoutany removal media(CD, USB, Floppy) then please cheack out
this given link.You need to just copy the (small) Linux installer on a hard disk and run it from your previous operating system. Finish using the network. That's all. In some cases you can substitute the network by a (big) hard disk if your network is not good..

LINK

Intrusion Detection System

2006-08-28T12:34:00.000+05:30

We were searching for an efficient intrusion detection system. .We wanted to develop some differnt kind of system which can perform as a detection and prevention system.After a long time discussion and searching in the net, we finalized ythis approch..

In our system , We have two module. One from userspace called detection system.Another from kernel space called prevention system.System acts like

When packets come from the network to NIC, we capture the packets and send to the userspace program.
Still we are thinking to avoid IP stack for passing each packets to user space.Any way , a user space detection program checks the each and every packets and try to match with predefined rule. If this match happens then detection system calculates the rate of packets whichs falls under this rules. If it overtakes the thresold value then this packet address would be written into a configuration file and immediately send to the kernel module which is ruuning in kernel space. So next time if this type of packets is comming into the NIC, before passing to the network stack, our kernel module will drop theses packets. This procedure will occour recuresively and best way to name this project .....

" Knowledge based intrusion detection and prevention system"
While searching , i got very good metirial from net which i should mention here for further look up:

linux 2.4 Packet Filtering
Writing Network device Driver
Inside the Linux Packet Filter Part 1
Inside the Linux Packet Filter Part II

i will add more......

[off topic]Planning to go pondichery

2006-08-11T14:00:00.000+05:30

Today , i decided to go pondichey as a weekend trip with anubhav and his friends.Last one year i am staying chennai but didn't go anywhere. Though i am not feeling well but i don't want to miss this trip.Next year,i would be a married person.So i would not get a chance to enjoy with friends.

Ok. Hope my trip will give me best enjoyment. Oh i didn't tell my parent that i am going. i thought i will tell when i will reach pondichery.

Wait for next trip blog....

bye bye

Kernel Hacking Free Course 2006

2006-08-07T13:17:00.000+05:30

If you are interested in installing and using a Unix system based on the Linux kernel, you won't have any problem in finding all the documentation you need. In fact, there exist several excellent books that teach how to install Linux and how to run it. Furthermore, many courses are taught in many countries on the same topics.

If your problem is how to modify the Linux kernel and how to evaluate the impact of the changes you made (we call this kind of programmer's behavior kernel hacking) then you won't find as much documentation as in the previous case. In fact, you can find almost anything on the net, except that is scattered around many places and not all the documents you will find are outstanding, while many are obsolete.

So, the ambitious goal of this Kernel hacking course is to describe in a coherent way all you should know to become a reasonable kernel hacker[Written by famous author-Daniel P. Bovet and Marco Cesati, the authors of the popular O'Reilly's book "Understanding the Linux kernel"]

Have a look on this link.

Happy Hacking...

A small tips for C/C++ programming with Vi Editor

2006-07-17T17:39:00.000+05:30

I am using vi editor to write my C program. I found some good feature with Vi editor which will help programmer to write C program clean and readble. Also reduce time to debug the C/C++ source file.

Start C style Indent in Vi Editor
:set cindent

ctags
just make a command " ctags *" on your source directory. Ctags command will create one file called tags which contains defined
macros defined by #define
enumerated values
function definitions, prototypes, and declarations
class, enum, struct, and union names
namespaces
typedefs
variables (definitions and declarations)
class, struct, and union members

Now open vi editor with source file name and put
: tag [i.e if i have a function fun() then it sould be :tag fun or just pressed ,it will list down all the tag name and hit enter . It willl jump where fun() has been defined. Just put ctrl+T to come back.

Syntax highlight:

:syntax on

Auto make

If you prepare makefile for your program then vi will allow you to run your program within vi. Just type
:make
[for more details :h make]

Device Driver Kit for Linux

2006-07-13T12:11:00.000+05:30

Do you want to write Device Driver under linux? What's the easyest way to do that...I have searched couple of time in net to get any device driver kit[ ofcource free] where everything sould be configured and preloaded. You need not to worry about any setup to code for your new device driver. I got one like windriver from Jungo but it is propitary.I don't have money to pay. So i stopped when Gregkh released first LDDK with 2.6.16.18 kernel and full LDD3 book.
You could visit his home page.
If you want to get full iso image then visit this site

Create your own Loopback Device

2006-06-22T16:31:00.000+05:30

A loopback device enables you to simulate a block device using an ordinary disk file.Data can be written or read from a file name disk-image rather than to and from the tracks and sector of an actual physical disk drive or disk partition.(ofcourcd disk-image must reside on an actual disk,which must be larger than the simulated disk). Loopback devices are named like /dev/loop0, /dev/loop1, and so on.Each can be used as a single block device at one time. Be sure when you are stting up loopback device you should be logged as a super user.

This loopback device can be used as any other block device(like your hard disk). You can create file system and mount this files system as you would mount the file system on a normal disk or partition.Such type of file system is basically called virtual file system.
Construct the virtual File System and Mount it with loopback device(Follow this step)

1.Create any 10MB file using dd command to hold the virtual file system. You can create this file anywhere as you like but i like to create in /tmp.
[suman@linux tmp]#dd if=/dev/zero of=/tmp/disk-image count=20480
your output should be like

20480+0 records in
20480+0 records out
10485760 bytes (10 MB) copied, 0.218746 seconds, 47.9 MB/s
2.cheack with ls -l command
[suman@linux tmp]# ls -l /tmp/disk-image
-rw-rw-r-- 1 suman suman 10485760 Jun 22 15:43 /tmp/disk-image
3.Now create a file System (i prefered ext2 File System..U can choose as you like) by this command
suman@ajit program]$ mke2fs -q /tmp/disk-image
It will ask for proceed just put y
/tmp/disk-image is not a block special device.
Proceed anyway? (y,n) y
4.Mount the file system using loopback device. Before that you need to create one mount point for the virtual file system.
[suman@linux tmp]# mkdir /tmp/virtual_fs
[suman@linux tmp]# mount -o loop=/dev/loop0 /tmp/disk-image /tmp/virtual_fs

5.If you don't see any error, you have successfully mounted the file System . Now cheack it through this command
[suman@linux tmp]#df -h /tmp/virtual_fs
you would see
Filesystem Size Used Avail Use% Mounted on
/tmp/disk-image 9.7M 92K 9.1M 1% /tmp/virtual_fs
6. Create one test file to test whether it is working correctly or not
[suman@linux tmp]# cd /tmp/virtual_fs
[suman@linux tmp]#echo 'I have done' >test.txt
[suman@linux tmp]#ls -l
lost+found test.txt
[suman@linux tmp]#cat test.txt
I have done
......lost+found is a direcxtory that was automatically added by mke2fs.

If want to unmount the file system just put this command
[suman@linux tmp]# umount virtual_fs

Why should i need loopback device? This question may arriase..let me tell one situation and explain me how we can use loopback device..

You have one CDROM image and you are using frequently. You need to just copy the whole CDROM image into new loopback device...you can do like that
[suman@linux]$ cp /dev/cdrom /tmp/disk-image
or
[suman@linux]$dd if=/dev/cdrom of=/tmp/disk-image

This may take several time depends on the size of your CD ROM. The resulting image file could be as large as the contents of the CDROM.
Now you can mount the CDROM image with out having the original CDROM in your CDROM.For example
[suman@linux]# mount -o loop=/dev/loop0 /tmp/disk-image /mnt/cdrom

You can access much faster than CDROM beacause whole CDROM content is in Hard Disk. Note most CDROM use the file System type iso9660.

You are playing a game which always needs your CDROM. You could not play without CDROM and sometimes you are seeing that your game is going to slow as you expected. Well create a loopback device and copy the CROM contents and mount it......and Enjoy...You are not accessing CDROM phisically which could save your CDROM life.
If you have any comments any question and also if you found any error about this mini tutorial please mail me and please leave your suggestion in comment section that i can improve later. I always apologies my poor english expression. forgive me..

kernel Status

2006-06-21T10:57:00.000+05:30

Please go to LinuxHQ for Linux kernel and development information.

Linux kernel Develoment

2006-06-20T14:38:00.000+05:30

Now a days i am little bit busy. I am learning how to write device driver under linux(maily char and USB). i have spend lots of time to search good meterial (for beginner) from internet.I started with LKMPG 2.6.x from tldp.org. Next I should read LDD3 by rubini(reading..)I listed some online metirial which i found during my study. Hope you could save some time to finding these metirial. I have given all the downloadable link. You just have to click the link and save it. Second tutorial is a link for entire book. You could get chapterwise pdf file from this site.

Tutorial No. 1
The Linux Kernel Module Programming Guide
Tutorial No. 2
Linux Device Driver by Jonathan Corbet, Alessandro Rubini, and Greg Kroah-Hartman(2.6.x kernel) Tutorial No.3
Writing Character Device Driver for Linux
Tutorial No. 4
Programming Guide for Linux USB Device Drivers

Those who are realy interested in kernel development and device driver here is great resource list.

Kernel Resource List

Happy Hacking!

Running FreeBSD inside Fedora Core 5: QEMU A processor Emulator

2006-06-18T20:57:00.000+05:30

Processor emulator is one kind of software which will allow one operating system to run in another operating system.We don't need to reboot the system and write boot sector in MBR.

It is very much usefull for testing cross platform application. Say you want to build one application which should run GNU/Linux,Micro$oft,FreeBSD and MAC OS. Then probabily you would need four different system or one system with four operating System.But every timr you have to reboot your system and test your application. Virtualization(i.e Processor Emulator) will help you to run four operating system in one platform. Just you have to switch one OS to onather by clicking your mouse.
There are many such type of application,both are open source and propitary. Here I have given some list , If you know another please update it.

Open Source:Xen,Bochs,QEMU,PeerPC etc
Propitary:Vmware,Micro$oft Virual PC etc

I have successfully installed freeBSD under my Fedora core 5 through QEMU.(Just for showing how it works) Here i am writing very small HOW-TO for Fedora code 5.
(As it is small , I am going to install from RPM not from source)
Step 1: First Get RPM fro FC5(Link)
Step 2: go to root by typing su - and your password
[suman@ajit ~]$ su -
Password:
Step 3: Install RPM
[suman@ajit qemu]# rpm -ivh qemu-0.8.1-1.fc5.i386.rpm
Step 4:put this command
[suman@ajit ~]$ qemu-img create linux.img 2G
after successfull creation of linux.img file it should shows this message
"Formating 'linux.img', fmt=raw, size=2097152 kB"

Step 5: If you download iso image into your home directory then give this command
[suman@ajit ~]$ qemu -boot d -cdrom ~/6.1-RELEASE-i386-bootonly.iso -hda linux.img
Or
If you have CDROM then give this command
[suman@ajit ~]$ qemu -boot d -cdrom /dev/cdrom -hda linux.img

Congratulation! You will see that a freeBSD is booting from a separate window from your present Operating System..

Step 6:Relax and enjoy multiple Operating System from your system...

In this tutorial I didn't include any comment regarding any command. It is your task to read and understand .I am hereby giving some link from internet which will help you to understand..best of Luck!!
link1
Link2
Let me know if you have any comments regarding this mini tutorial..

GDB: A GNU Debugger

2006-06-09T17:38:00.000+05:30

I recently wanted to learn GDB(Briefly). I searched in google with keyword " GDB tutorial"and google gave me this peter page. If you are writing program in linux it would be helpful to you. Peter started this tutorial with memory layout of a program and discussed gdb command with example. Any one who wants to know gdb could start from this tutorial. Anyway peter has also other good stuf . You can visit his site.

More inforation about GDB:
Official GDB User Manual
Norman Matloff tutorial on GDB Princeton University GDB tutorial
RMS's gdb debugger tutorial
uchicago University tutorial

two handful commands

2006-05-31T14:22:00.000+05:30

Do you want to make shortcut in your command ?Then run this command into your command line
command format: alias [shoutcut-command]=[command]
example:
>alias=c=clear
Do you want to revoke this alias? Run this command in to your coomand line
>unalias c

How to open CDROM with command line?
Run this comand
>eject /dev/cdrom or > eject /mnt/cdrom or just run > eject .Bash shell will take CDROM device as a default device.

how to close the CDROM Device with command line?
> eject -t /dev/cdrom or simply put >eject -t

Know about web syndication

2006-05-24T10:00:00.000+05:30

RSS is a family of web feed formats, specified in XML and used for Web syndication. RSS is used by (among other things) news websites, weblogs and podcasting. The abbreviation is variously used to refer to the following standards:
Really Simple Syndication (RSS 2.0)
Rich Site Summary (RSS 0.91, RSS 1.0)
RDF Site Summary (RSS 0.9 and 1.0)

Web feeds provide web content or summaries of web content together with links to the full versions of the content, and other metadata. RSS, in particular, delivers this information as an XML file called an RSS feed, webfeed, RSS stream, or RSS channel. In addition to facilitating syndication, web feeds allow a website's frequent readers to track updates on the site using an aggregator.

(PS: Content has been taken from wikipedia.).

I am using suse 10 with kde 3.5. I used "akregator" to get the feeds from site.

visit Redhat site for more information.

how to write a system tray application?

2006-05-19T15:27:00.000+05:30

The system tray is a cross-desktop method of notifying the user. It will be very usefull for user. Suppose if you have mailclient[Kmail] and if it is in system tray then whenever new mail will come, it will pop up a new windows with notification of new mail. But question is how to write these kind of application? Well i was searching internet for document. Yesterday i got one document from redhat site . It is in perl language.Search the document
Procedure:
1.Your system should have perl and gtk 2.0.
2. We need two library called
a)gnome-python2-libegg-2.10.0-2.1
b)gnome-python2-extras-2.10.0-6
Just Download this two rpm and installl with --nodeps option like
suman@>rpm -ivh gnome-python2-extras-2.10.0-6.i386.rpm gnome-python2-libegg-2.10.0-2.1.i386.rpm --nodeps
Open a vi editor with file name test.py

#! /usr/bin/python
import gtk
import egg.trayicon
t = egg.trayicon.TrayIcon("MyFirstTrayIcon")
t.add(gtk.Label("Hello"))
t.show_all()
gtk.main()

Copy this code to the file. And run as suman> python test.py
Thats's all !!! A "Test" system tray icon will be shown in your desktop system tray.

After long time back

2006-05-19T15:08:00.000+05:30

Oh! I am writing my blog after 13 days. I went to attend my sister marraige. I couldn't get time to write. Well after comming back to my office, I started coustomizing the Open Source ERP product called Compiere . I successfully made some small changes with modify the source code.

Open Source News:
Okippi is the successor to Blue Security's Blue forge project..Now it is started...

Open Source Shooping cart application

2006-05-05T16:33:00.000+05:30

If you are looking for good open source shopping cart software,visit Interchange. It worked so well that some company began an enterprise-wide migration to open source software that has Linux running everywhere, from the servers to the desktops.

Top 50 Open Source Application

2006-05-05T16:14:00.000+05:30

Now we can do all our work with open source application. Sometimes i have seen that people don't find or don't know which open source applications are available to do certain job. Here is a list of 50 top open source application by devid uhlman which we can use for beginner to enterprise level.This is a pdf file. I am giving here linkwhere you can download. This pdf also contains the regarding application site name where you can get the software.Have a look ,Have a fun!

http://www.uversainc.com/download/top50.pdf

OCFS2 project from Oracle

2006-05-04T14:31:00.000+05:30

OCFS2 is the next generation of the Oracle Cluster File System for Linux. It is an extent based, POSIX compliant file system. Unlike the previous release (OCFS), OCFS2 is a general-purpose file system that can be used for shared Oracle home installations making management of Oracle Real Application Cluster (RAC) installations even easier. Among the new features and benefits are:

Node and architecture local files using Context Dependent Symbolic Links (CDSL)
Network based pluggable DLM
Improved journaling / node recovery using the Linux Kernel "JBD" subsystem
Improved performance of meta-data operations (space allocation, locking, etc).
Improved data caching / locking (for files such as oracle binaries, libraries, etc)

sources: http://oss.oracle.com/projects/ocfs2/

OCFS2 will only work on the 2.6 series of the Linux kernel. As a result, OCFS2 packages are only available for 2.6 based distributions.
SuSE Linux Exterprise Server 9:Bundeled,need configuration
Red Hat Enterprise Linux 4:not bundeled ,need to download and configure
Ubuntu:Bundeled,need configuration

User Guide

Source Control Management Tool

2006-04-27T10:02:00.000+05:30

Now a days we have a good number of source control Management tool.
CVS,SVN,ClearCase,Mercurial are good among of them.I was just evaluating which is best and why? Past two days a good discussion was going on ILUG-Chennai. CVS,SVN,Mercurial are Open Source .
link for CVS,SVN,Mercurial
CVS
SVN
Mercurial

2006-04-26T11:38:00.000+05:30

Yesterday I got a offer letter from Naturesoft,Chennai as a Technical Solution Analyst. My M.Tch will be finished at June,2006. I am very happy that i was employed before i finish my M.Tech.

Now i got too much interested with Linux kernel. I want to make my career with Linux.I found very good metirial to understand the linux kernel
Linux kernel Architecture
Linux Kernel
Now it is time for rock with Linux kernel......Hope i will do my best..

New in Blog

2006-04-21T22:52:00.000+05:30

Today i started my blog. I was inspired while i was browsing gaurav blog.I was priviously thinking that i should have my own website.I started with geocities free webhosting service but after sometimes i feeled that it is very time consuming task to post any comments on that site. So i decided to go for google blog.
Still i should have my own website where i can control each and everything. So what i need for that? One domain name, one good CMS,and little bit html knowledge... CMS should be open source. I decided to go for Drupal.And for hosting still i have to decide. Well my website should have more GNU/Linux related information that whoever will visit my site,they should able to learn something related to open source and GNU/Linux.This blog will not carry my daily activity generally.But sometimes it might be.My goal for creating this blog is to share my knowledge whatever i learned and will learn every day.finally i would like to say

It is very necessary to realize now that "Linux" is only a good monolithic kernel.It is one of the important subsystem of a whole GNU Operating System. We should not call "Linux" as a Operating System.

for more information visit this link:
Linux Kernel
GNU Philosophy