5.8. Pointer debugging

All the techniques described in section Debugging applies when debugging pointers. What can make debugging pointer errors more difficult is that when things go wrong with a pointer, often memory is corrupted as well, which limits what you can get done with cout.

Tools are your friends.

  1. Learn how to use a debugger.

    The debugging section has a brief overview of the gdb debugger and links to more information.

  2. Consider a dynamic memory checker such as valgrind.

5.8.1. Valgrind

The valgrind program can be used to detect some (but not all) common errors in C and C++ programs that use pointers and dynamic storage allocation. In addition to basic memory checking, valgrind can find many bugs related to uninitialized variables and undefinded behavior. You can run valgrind on your program by putting valgrind at the start of the command line:

valgrind ./my-program

You may pass in arguments to your program or redirect input from the shell:

valgrind ./my-program arg1 arg2 < test-input

This will run your program and produce a report of any memory allocations and de-allocations it did. Valgrind will also warn you about common errors like using uninitialized memory, dereferencing pointers to strange places, writing off the end of memory blocks allocated using malloc or new, or failing to free blocks.

You can suppress all of the output except errors using the -q option, like this:

valgrind -q ./my-program

You can also turn on more tests, e.g.:

valgrind -q --tool=memcheck --leak-check=yes ./my-program

See valgrind --help or man valgrind for more information about the (many) options, or look at the documentation at http://valgrind.org/ for detailed information about what the output means. If you want to run valgrind on your own machine, you may be able to find a version that works at http://valgrind.org. Unfortunately, this is only likely to work if you are running a Unix-like operating system (which includes Linux and MacOS X, but not Windows).

A simple walk-through. Let’s say we have the following code in foo.cpp with a simple error:

#include <iostream>
int main () {
  int x;
  int y = 3;

  std::cout << "sum: " << (x + y) << '\n';
}

Just as with gdb, compile it with debugging symbols.

g++ foo.cpp -g3 -o foo

Valgrind provides a lot of debugging information very quickly, although on some systems, even with debug symbols compiled in, your output looks like this:

==1229== Conditional jump or move depends on uninitialised value(s)
==1229==    at 0x40DDBA3: std::ostreambuf_iterator<char, std::char_traits<char> > std::num_put<char, std::ostreambuf_iterator<char, std::char_traits<char> > >::_M_insert_int<long>(std::ostreambuf_iterator<char, std::char_traits<char> >, std::ios_base&, char, long) const (in /usr/lib/libstdc++.so.6.0.18)
==1229==    by 0x40DE168: std::num_put<char, std::ostreambuf_iterator<char, std::char_traits<char> > >::do_put(std::ostreambuf_iterator<char, std::char_traits<char> >, std::ios_base&, char, long) const (in /usr/lib/libstdc++.so.6.0.18)
==1229==    by 0x40E9F1D: std::ostream& std::ostream::_M_insert<long>(long) (in /usr/lib/libstdc++.so.6.0.18)
==1229==    by 0x40EA112: std::ostream::operator<<(int) (in /usr/lib/libstdc++.so.6.0.18)
==1229==    by 0x41CB9D2: (below main) (in /lib/libc-2.18.so)

In order to get valgrind to report line numbers related to errors in calls to library functions, you need to add the command line argument --track-origins=yes, like this:

$ valgrind --track-origins=yes ./foo

==1302== Conditional jump or move depends on uninitialised value(s)
==1302==    at 0x40DDB47: std::ostreambuf_iterator<char, std::char_traits<char> > std::num_put<char, std::ostreambuf_iterator<char, std::char_traits<char> > >::_M_insert_int<long>(std::ostreambuf_iterator<char, std::char_traits<char> >, std::ios_base&, char, long) const (in /usr/lib/libstdc++.so.6.0.18)
==1302==    by 0x40DE168: std::num_put<char, std::ostreambuf_iterator<char, std::char_traits<char> > >::do_put(std::ostreambuf_iterator<char, std::char_traits<char> >, std::ios_base&, char, long) const (in /usr/lib/libstdc++.so.6.0.18)
==1302==    by 0x40E9F1D: std::ostream& std::ostream::_M_insert<long>(long) (in /usr/lib/libstdc++.so.6.0.18)
==1302==    by 0x40EA112: std::ostream::operator<<(int) (in /usr/lib/libstdc++.so.6.0.18)
==1302==    by 0x41CB9D2: (below main) (in /lib/libc-2.18.so)
==1302==  Uninitialised value was created by a stack allocation
==1302==    at 0x8048717: main (foo.cpp:3)
==1302==
==1302== Use of uninitialised value of size 4

There can be a lot of noise in the output, especially when templates are involved. If you look at the last line in each group, foo.cpp, line 3 is called out each time. And that is exactly where the error lies.

More to Explore

You have attempted of activities on this page