How to load debug symbols with GDB - Another Approach

I wrote in the post How to load debug symbols with GDB one way do debug a stripped version of an application with full access to debug symbols.

In this post I will explain another way you can do it.

We are going to load debuging information from a file that was generated by our executable just before it was stripped off..

I am going to use the same source and environment settings that I used in my earlier post:

• released.c: source code of the program we wish to debug (listing 1).
• ~/estudo/: Source code of our program will be put here.
  
• ~/local/bin: The stripped off version of binary program will stay here.
• ~/local/symbols: In this place are all files that contain debuging information.
  

<center><tbody></tbody>

#include <stdio.h> #include <stdlib.h> int division(int a, int b); int m; int main(void) { int i; int j; printf("vou setar i\n"); i = 10; printf("vou setar j\n"); j = 1; printf ("i = %d, j = %d\n", i, j); m = division(i, j); printf("m = %d / %d = %d\n", i, j, m); return 0; int division(int a, int b) { return a / b; }

</center>
Let's start:

1 - Create your program with debug information. In our sample:

2 - Now we will generate a file that will contain only debug informations. You generate this file by issuing this command:
In our sample:

The choice of .sym as an extension for the debug info file is totally arbitrary. You can use whatever you wish to.

3 - You remove debugging information:

4 - Move file release to ~/local/bin/:

5 - Move file released.sym to ~/local/symbols/

6 - Go to directory ~/local/bin

7 - Run GDB:

8 - Try list command to see that release executable file doesn't have symbols in it.

Figure 1 shows us what I said.


9 - Let GDB to load symbols from symbols file named release.sym. This file has all symbols that we need to debug. You achieve this by issuing the following command:

From now on you can debug your program as usual.

Figure 2 shows us that debuging symbols where imported successfully and that now the list command shows us the program source code.


As you can see in figure 3, I set a break point at line 17 and I ran the program that stopped there. Then I printed i variable. It is possible just because symbols were loaded.


You can see that this method to load debug symbols is easier than the previous one presented in the post How to load debug symbols with GDB.

My friend Jumpi complained that it is not the unix way to debug applications that just don't have debug information by loading a separated file. We have core files to help us to debug the application in our host.

I agree with him that it is unusual to debug this way, but it may be useful.

The idea behind this method (and behind the previous one) is similar to that used by MS with .pdb files.

It will allow you to save disk space on your host because you can keep only stripped versions of your applications and libraries in it. It is especially important to embedded systems.

At the same time it let you have a colection of symbol files stored at some place (say, a DVD disc or another server). When (and if) you need to do a debug session

I hope this post will make your life easier.

Syndicated 2008-03-17 14:49:00 from Marcio Andrey Oliveira

Precision Resistors

A lot of hobbyists want to make their own circuits but they face a problem: precision resistors.

But what is the difference between a precision resistor and an ordinary resistor? Answer: none.

By the way, the term precision is unappropriated because the tolerance of the component doesn't show us the precision of the resistor but how much its real value can differ from the nominal value.

- What you mean by none?

I will explain. By no means the manufacturer can assure that during a manufacturing process of resistors all of them will have the same value.

So, the manufacturer measures the resistance of each component and then it groups them by ranges of values. Each group has the same tolerance or (in)precision.

But what does tolerance means? It means that manufacturer assures that the real value of the resistor will not be either greater nor lower than nominal value plus or minus a percentage that is indicated by the tolerance band.

Let's see a sample.

Say a manufacturer is producing resistors with resistances of 100 ohms (nominal value) and that he sells them with 1%, 2%, 5%, 10% and 20% of tolerance.

Manufacturer will measure each resistance of 100 ohms and he will group them as follows:

All resistors with resistance varying between 99 and 101 ohms will be put together in the group of components with 1% of tolerance.

All resistors with resistance varying between 98 and 99 ohms or between 101 and 102 ohms will be put together in the group of components with 2% of tolerance (you should remember that all components with resistances varying between 99 and 101 ohms are in the group of resistors with 1% of tolerance).

All resistors with resistance varying between 95 and 98 ohms or between 102 and 105 ohms will be put together in the group of components with 5% of tolerance (you should remember that all components with resistances greater than 98 and less than 102 ohms were put in the groups of resistors with 1% and 2 % of tolerance).

This process continues until all components are classified in their tolerance band.

So I say again: There is no difference between a precision resistor and an ordinary resistor.

There are many projects where the ratio between resistances is more important than the actual values. In these case we can do the same thing as the manufacturer. We can measure our resistors to select the most appropriated.

Let me show an example. Suppose we have an inverter amplifier as shown in figure 1.


Gain is determined by equation

We can see that the circuit's gain is determined by the ratio between Rf and Ri.

Suppose we want a gain of X.

If Rf = 10,000 ohms and Ri = 5,000 ohms the gain is -2.

If Rf = 20,000 ohms and Ri = 10,000 ohms the gain is -2 too.

If we use resistors with tolerance (precision) of 5% the gain could vary between -1.81and -2.21 depending on real values of resistors Rf and Ri.

If we use resistors with tolerance (precision) of 2% the gain could vary between -1.92 e -2.08 what is much better.

What if we don't have precision resistors? What can we do?

We can use our ordinary resistors (tolerance of 5%). Take a bunch of them and measure each one until you find two resistors with a resistance ratio as near as possible of the value we wish (It is 2 in our sample).

Let me illustrate with an example. Say we take a bunch of resistors of 20K and 10K ohms (nominal values) to measure.

Suppose we found some resistors of 20,900 and 10,400 ohms. So we could do Rf = 20,900 ohms and Ri = 10,400. With these values, the gain would be -2.01 that is a better result than estimated with 2% tolerance resistors.

If we found Rf = 21,000 and Ri = 10,500 than the gain would be exactly 2.

The method, presented in this post, to use ordinary resistors instead of precision resistors is suitable only in prototype development. In commercial scale it is not feasible and the use of precision resistors is mandatory.

Syndicated 2008-03-17 02:38:00 from Marcio Andrey Oliveira

How to load debug symbols with GDB

My friend Wanderley asked me if it is possible to tell GBD to load debuging symbols from some file and use it to help debuging a program that doesn't have them.

Yes. It is.

There are two solutions to this question.

I going to explain the first solution in this post. The other solution I will explain in the next post.

You can load debuging information from an debug enabled version of executable file.

In order to better explain the first solution, I will setup my sample enviroment as follows:

• released.c: source code of the program we wish to debug (listing 1).
• ~/estudo/: Source code of our program will be put here.
  
• ~/local/bin: The stripped off version of binary program will stay here.
• ~/local/symbols: In this place are all files that contain debuging information.
  

<center><tbody></tbody>

#include <stdio.h> #include <stdlib.h> int division(int a, int b); int m; int main(void) { int i; int j; printf("vou setar i\n"); i = 10; printf("vou setar j\n"); j = 1; printf ("i = %d, j = %d\n", i, j); m = division(i, j); printf("m = %d / %d = %d\n", i, j, m); return 0; int division(int a, int b) { return a / b; }

</center>
I have two versions of the program: with and without debuging information.

1 - You compile your program with debug information. In our sample:

2 - You make a copy of your program. In our sample:

3 - You strip off debuging information:

As you can see on Figure 1, we have two programs. released.full has debuging symbols but release doesn't have them.


4 - Move file release to ~/local/bin/:

5 - Move file released.full to ~/local/symbols/

6 - Go to directory ~/local/bin

7 - Run GDB:

8 - Try list command to see that release executable file doesn't have symbols in it.

Note: if the program was already running you could get its PID then attach GDB to it.

Figure 2 shows us two windows. The first one shows that our executable file has no debug information. In the other window we can see that release is not yet loaded.


9 - Let GDB to load symbols from executable file named release.full. This binary version of our program has all symbols that we need to debug.

Please notice that GDB will not replace the release executable by release.full version of our program. It will just import symbols from release.full into release debugging session.

But GDB needs to know in advance where it must put the symbols it will load. How can you determine the correct memory address?

It is quite simple. You issue command maint inside GDB:

Then you look for .text section. The address that is in the first column is what you want. In our sample, it is 0x08048320. See figure 3.


10 - The next step is to instruct GDB to load debug symbols into .text section. To achieve it you do this:

In our sample it means to type:

From now on you can debug your program as usual.

Figure 4 shows us that debuging symbols where imported successfully and that now the list command (abbreviated as l) shows us the program source code.


As you can see in figure 5, I set a break point at line 17 and I ran the program that stopped there. Then I printed i variable.

In the other terminal I issued ps command. It was done just to show you that the only program running was release executable. There is no instance of release.full program.


I hope this post will make your life easier. In the next time I will teach you another way to import debuging symbols.

Syndicated 2008-03-08 04:07:00 from Marcio Andrey Oliveira

Using Mutt with Gmail

Gmail is the best e-mail service and Firefox is the best browser. But if you keep Gmail's page open for a long time then Firefox will waste a lot of memory of your computer.

If you have an old machine like me, you know that each byte is important. So, I decided to come back to my old mail client: Mutt.

Here I'm showing some steps you need to let mutt access Gmail with IMAP protocol.

I'm using Ubuntu 7.04 with mutt 1.5.13, but it should work with other distros.

You must replace my_mail@gmail.com and my_password occurrences with real values.

1. First of all you must enable IMAP access to your Gmail account. To enable IMAP in your Gmail account:
   1.1. Log in to your Gmail account.
   1.2. Click Settings at the top of any Gmail page.
   1.3. Click Forwarding and POP/IMAP.
   1.4. Select Enable IMAP.
   
   
2. Install mutt, msmtp and ssl. I won't show you how to do it because it is different for each distro.
   
   
3. Create msmtp configuration file (~/.msmtprc) as shown:
   account default
   host smtp.gmail.com
   port 587
   from my_mail@gmail.com
   tls on
   tls_starttls on
   auth on
   user my_mail@gmail.com
   password my_password
   logfile ~/.msmtp.log
   
   
4. Create mutt configuration file (~/.muttrc) as shown:
   set imap_user = "my_mail@gmail.com"
   set imap_pass = "my_password"
   
   set from = "my_mail@gmail.com"
   set realname = "Your Real Name"
   
   set folder = "imaps://imap.gmail.com:993"
   set spoolfile = imaps://imap.gmail.com:993/INBOX
   set record=""
   set postponed="imaps://imap.gmail.com/[Gmail]/Drafts"
   
   set header_cache="~/.mutt/cache/headers"
   set message_cachedir="~/.mutt/cache/bodies"
   set certificate_file=~/.mutt/certificates
   set move = no
   set imap_check_subscribed="yes"
   set imap_list_subscribed="yes"
   set mbox="imaps://imap.gmail.com/[Gmail]/All Mail"
   # this line instructs mutt to refresh my IMAP mailbox every 120 seconds
   set mail_check=60
   #I'm using ^ character to instruct mutt to update e-mail list immediately
   bind index "^" imap-fetch-mail
   

And that's all folks.

Syndicated 2008-02-29 16:46:00 from Marcio Andrey Oliveira

Looking for many words at once

I use Vim daily and my admiration for it always is renewed. It has so may features that I doubt any one can use 100% of its capacity.

One of the Vim's features I like too much is its ability to search different words at once in text.

When we want to find a word we enter in command mode and we type:

Syndicated 2008-01-30 16:37:00 from Marcio Andrey Oliveira