Pages

Monday, September 24, 2012

News: Unusual commands in linux : pv

Today I will speak about pv command.
Pipe viewer - pv is a terminal-based tool for monitoring the progress of data through a pipeline written by Andrew Wood
To install pv on Fedora or CentOS do this:
$ sudo yum install pv
You can read the man file to get more help:
man pv
The result is:
NAME
       pv - monitor the progress of data through a pipe

SYNOPSIS
       pv [OPTION] [FILE]...
       pv [-h|-V]

DESCRIPTION
       pv  allows  a  user  to see the progress of data through a pipeline, by
       giving information such as time  elapsed,  percentage  completed  (with
       progress  bar),  current  throughput  rate, total data transferred, and
       ETA.

       To use it, insert it in a pipeline  between  two  processes,  with  the
       appropriate  options.  Its standard input will be passed through to its
       standard output and progress will be shown on standard error.
Let's see some examples.
You can get precise time how long it will take.
$ pv voronoi.py | python 
 737B 0:00:00 [86.9kB/s] [==================================>] 100%
You can see how fast the computer reads from /dev/zero.
$ pv /dev/zero > /dev/null
1.1GB 0:00:05 [   2GB/s] [    <=>                                            ]
You can use the dialog to show a progress bar.
To do this you need to use -n arg.
$ (pv -n /dev/zero > /dev/null) 2>&1 | dialog --gauge "Please wait" 10 70 0

Wednesday, September 19, 2012

Make one simple kernel module using C programming language.

Today I will try to explain how to make a kernel module.
Basically, a kernel module it's one piece of code that can be loaded and unloaded into the kernel.
You can read about kernel modules from here.
Now, let's try to make a simple kernel module.
Writing your own module can be done if you know some basic rules.
First, about how to use modules.
Suppose you have a new module named kernelmoduletest.
How to load the module.
Just use this command in your module kernel folder:
#insmod kernelmoduletest.ko
Next, test the kernel module messages.
This can be done with:
#dmesg | tail
Also, you need to remove the kernel module.
$ rmmod kernelmoduletest 
ERROR: Removing 'kernelmoduletest': Operation not permitted
This means you need to have superuser rights. So use the sudo or su command.
Then use the rmmod command.
Know how to make the kernel module skeleton.
This is not very simple because you need to know many things about kernel and programming.
But you can try the test with some simple example like this example:
Create the c file named kernelmoduletest and add this source code:
#include <linux module.h="">
#include <linux kernel.h="">

MODULE_LICENSE("GPL");
MODULE_DESCRIPTION("kernelmoduletest");
MODULE_AUTHOR("Catalin George Festila/mythcat/catafest");

int init_module() {

    printk(KERN_INFO "Now I will initialize my  kernel module\n");

    printk(KERN_INFO "Test: Hello World !\n");

    return 0;
}

void cleanup_module() {

    printk(KERN_INFO "Bad!... kernel module unloaded.\n");
}</linux></linux>
Make a new file named Makefile.
Add this to tell how to compile the kernel module.
obj-m += kernelmoduletest.o
all:
 make -C /lib/modules/$(shell uname -r)/build M=$(PWD) modules
clean:
 make -C /lib/modules/$(shell uname -r)/build M=$(PWD) clean
Compile your kernel module
Just use make command.
# make
...
  Building modules, stage 2.
  MODPOST 1 modules
...
make[1]: Leaving directory `/usr/src/linux-headers-2.6.31-14-generic'
If you test your module using the commands from the top of this tutorial, you can see that:

[14694.779227] Now I will initialize my  kernel module
[14694.779233] Test: Hello World !
[15049.825605] Bad!... kernel module unloaded.
If you have already made kernel modules and the subject it's interesting for you, send me an email.
Thank you. Regards.

Sunday, September 2, 2012

News: Tips and tricks: watch and cat.

Sometimes it is necessary to find solutions.
Today's example involves two Linux commands:
watch and cat
Here is a simple method by creating a script that returns an output:
while [ 1 ]; do cat /proc/meminfo; date; echo; sleep 1; done
Make this script executable and name it: meminfo.
The output is then processed by the command:
$ watch -d ./meminfo 
This simple method can be applied to other files ...