libstdc++-v3/doc/html/manual/debug.html - gcc - Git at Google

 <?xml version="1.0" encoding="UTF-8" standalone="no"?>
 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><title>Debugging Support</title><meta name="generator" content="DocBook XSL Stylesheets Vsnapshot" /><meta name="keywords" content="C++, debug" /><meta name="keywords" content="ISO C++, library" /><meta name="keywords" content="ISO C++, runtime, library" /><link rel="home" href="../index.html" title="The GNU C++ Library" /><link rel="up" href="using.html" title="Chapter 3. Using" /><link rel="prev" href="using_exceptions.html" title="Exceptions" /><link rel="next" href="std_contents.html" title="Part II.  Standard Contents" /></head><body><div class="navheader"><table width="100%" summary="Navigation header"><tr><th colspan="3" align="center">Debugging Support</th></tr><tr><td width="20%" align="left"><a accesskey="p" href="using_exceptions.html">Prev</a> </td><th width="60%" align="center">Chapter 3. Using</th><td width="20%" align="right"> <a accesskey="n" href="std_contents.html">Next</a></td></tr></table><hr /></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="manual.intro.using.debug"></a>Debugging Support</h2></div></div></div><p>
   There are numerous things that can be done to improve the ease with
   which C++ binaries are debugged when using the GNU tool chain. Here
   are some of them.
 </p><div class="section"><div class="titlepage"><div><div><h3 class="title"><a id="debug.compiler"></a>Using <span class="command"><strong>g++</strong></span></h3></div></div></div><p>
     Compiler flags determine how debug information is transmitted
     between compilation and debug or analysis tools.
   </p><p>
     The default optimizations and debug flags for a libstdc++ build
     are <code class="code">-g -O2</code>. However, both debug and optimization
     flags can be varied to change debugging characteristics. For
     instance, turning off all optimization via the <code class="code">-g -O0
     -fno-inline</code> flags will disable inlining and optimizations,
     and add debugging information, so that stepping through all functions,
     (including inlined constructors and destructors) is possible. In
     addition, <code class="code">-fno-eliminate-unused-debug-types</code> can be
     used when additional debug information, such as nested class info,
     is desired.
 </p><p>
   Or, the debug format that the compiler and debugger use to
   communicate information about source constructs can be changed via
   <code class="code">-gdwarf-2</code> or <code class="code">-gstabs</code> flags: some debugging
   formats permit more expressive type and scope information to be
   shown in GDB. Expressiveness can be enhanced by flags like
   <code class="code">-g3</code>. The default debug information for a particular
   platform can be identified via the value set by the
   PREFERRED_DEBUGGING_TYPE macro in the GCC sources.
 </p><p>
   Many other options are available: please see <a class="link" href="http://gcc.gnu.org/onlinedocs/gcc/Debugging-Options.html#Debugging%20Options" target="_top">"Options
   for Debugging Your Program"</a> in Using the GNU Compiler
   Collection (GCC) for a complete list.
 </p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a id="debug.req"></a>Debug Versions of Library Binary Files</h3></div></div></div><p>
   If you would like debug symbols in libstdc++, there are two ways to
   build libstdc++ with debug flags. The first is to create a separate
   debug build by running make from the top-level of a tree
   freshly-configured with
 </p><pre class="programlisting">
      --enable-libstdcxx-debug
 </pre><p>and perhaps</p><pre class="programlisting">
      --enable-libstdcxx-debug-flags='...'
 </pre><p>
   Both the normal build and the debug build will persist, without
   having to specify <code class="code">CXXFLAGS</code>, and the debug library will
   be installed in a separate directory tree, in <code class="code">(prefix)/lib/debug</code>.
   For more information, look at the
   <a class="link" href="configure.html" title="Configure">configuration</a> section.
 </p><p>
   A second approach is to use the configuration flags
 </p><pre class="programlisting">
      make CXXFLAGS='-g3 -fno-inline -O0' all
 </pre><p>
   This quick and dirty approach is often sufficient for quick
   debugging tasks, when you cannot or don't want to recompile your
   application to use the <a class="link" href="debug_mode.html" title="Chapter 17. Debug Mode">debug mode</a>.</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a id="debug.memory"></a>Memory Leak Hunting</h3></div></div></div><p>
   On many targets GCC supports AddressSanitizer, a fast memory error detector,
   which is enabled by the <code class="option">-fsanitize=address</code> option.
 </p><p>
   There are also various third party memory tracing and debug utilities
   that can be used to provide detailed memory allocation information
   about C++ code. An exhaustive list of tools is not going to be
   attempted, but includes <code class="code">mtrace</code>, <code class="code">valgrind</code>,
   <code class="code">mudflap</code> (no longer supported since GCC 4.9.0), ElectricFence,
   and the non-free commercial product <code class="code">purify</code>.
   In addition, <code class="code">libcwd</code>, jemalloc and TCMalloc have replacements
   for the global <code class="code">new</code> and <code class="code">delete</code> operators
   that can track memory allocation and deallocation and provide useful
   memory statistics.
 </p><p>
   For valgrind, there are some specific items to keep in mind. First
   of all, use a version of valgrind that will work with current GNU
   C++ tools: the first that can do this is valgrind 1.0.4, but later
   versions should work better. Second, using an unoptimized build
   might avoid confusing valgrind.
 </p><p>
   Third, it may be necessary to force deallocation in other libraries
   as well, namely the "C" library. On GNU/Linux, this can be accomplished
   with the appropriate use of the <code class="code">__cxa_atexit</code> or
   <code class="code">atexit</code> functions.
 </p><pre class="programlisting">
    #include &lt;cstdlib&gt;

    extern "C" void __libc_freeres(void);

    void do_something() { }

    int main()
    {
      atexit(__libc_freeres);
      do_something();
      return 0;
    }
 </pre><p>or, using <code class="code">__cxa_atexit</code>:</p><pre class="programlisting">
    extern "C" void __libc_freeres(void);
    extern "C" int __cxa_atexit(void (*func) (void *), void *arg, void *d);

    void do_something() { }

    int main()
    {
       extern void* __dso_handle __attribute__ ((__weak__));
       __cxa_atexit((void (*) (void *)) __libc_freeres, NULL,
 		   &amp;__dso_handle ? __dso_handle : NULL);
       do_test();
       return 0;
    }
 </pre><p>
   Suggested valgrind flags, given the suggestions above about setting
   up the runtime environment, library, and test file, might be:
 </p><pre class="programlisting">
    valgrind -v --num-callers=20 --leak-check=yes --leak-resolution=high --show-reachable=yes a.out
 </pre><div class="section"><div class="titlepage"><div><div><h4 class="title"><a id="debug.memory.mtalloc"></a>Non-memory leaks in Pool and MT allocators</h4></div></div></div><p>
   There are different kinds of allocation schemes that can be used by
   <code class="code">std::allocator</code>. Prior to GCC 3.4.0 the default was to use
   a pooling allocator, <code class="classname">pool_allocator</code>,
   which is still available as the optional
   <code class="classname">__pool_alloc</code> extension.
   Another optional extension, <code class="classname">__mt_alloc</code>,
   is a high-performance pool allocator.
 </p><p>
   In a suspect executable these pooling allocators can give
   the mistaken impression that memory is being leaked,
   when in reality the memory "leak" is a pool being used
   by the library's allocator and is reclaimed after program
   termination.
 </p><p>
   If you're using memory debugging tools on a program that uses
   one of these pooling allocators, you can set the environment variable
   <code class="literal">GLIBCXX_FORCE_NEW</code> to keep extraneous pool allocation
   noise from cluttering debug information.
   For more details, see the
   <a class="link" href="mt_allocator.html" title="Chapter 19. The mt_allocator">mt allocator</a>
   documentation and look specifically for <code class="code">GLIBCXX_FORCE_NEW</code>.
 </p></div></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a id="debug.races"></a>Data Race Hunting</h3></div></div></div><p>
   All synchronization primitives used in the library internals need to be
   understood by race detectors so that they do not produce false reports.
 </p><p>
   Two annotation macros are used to explain low-level synchronization
   to race detectors:
   <code class="code">_GLIBCXX_SYNCHRONIZATION_HAPPENS_BEFORE()</code> and
   <code class="code"> _GLIBCXX_SYNCHRONIZATION_HAPPENS_AFTER()</code>.
   By default, these macros are defined empty -- anyone who wants
   to use a race detector needs to redefine them to call an
   appropriate API.
   Since these macros are empty by default when the library is built,
   redefining them will only affect inline functions and template
   instantiations which are compiled in user code. This allows annotation
   of templates such as <code class="code">shared_ptr</code>, but not code which is
   only instantiated in the library.  Code which is only instantiated in
   the library needs to be recompiled with the annotation macros defined.
   That can be done by rebuilding the entire
   <code class="filename">libstdc++.so</code> file but a simpler
   alternative exists for ELF platforms such as GNU/Linux, because ELF
   symbol interposition allows symbols defined in the shared library to be
   overridden by symbols with the same name that appear earlier in the
   runtime search path. This means you only need to recompile the functions
   that are affected by the annotation macros, which can be done by
   recompiling individual files.
   Annotating <code class="code">std::string</code> and <code class="code">std::wstring</code>
   reference counting can be done by disabling extern templates (by defining
   <code class="code">_GLIBCXX_EXTERN_TEMPLATE=-1</code>) or by rebuilding the
   <code class="filename">src/string-inst.cc</code> file.
   Annotating the remaining atomic operations (at the time of writing these
   are in <code class="code">ios_base::Init::~Init</code>, <code class="code">locale::_Impl</code>,
   <code class="code">locale::facet</code> and <code class="code">thread::_M_start_thread</code>)
   requires rebuilding the relevant source files.
 </p><p>
   The approach described above is known to work with the following race
   detection tools:
   <a class="link" href="https://valgrind.org/docs/manual/drd-manual.html" target="_top">
   DRD</a>,
   <a class="link" href="https://valgrind.org/docs/manual/hg-manual.html" target="_top">
   Helgrind</a>, and
   <a class="link" href="https://github.com/google/sanitizers" target="_top">
   ThreadSanitizer</a> (this refers to ThreadSanitizer v1, not the
   new "tsan" feature built-in to GCC itself).
 </p><p>
   With DRD, Helgrind and ThreadSanitizer you will need to define
   the macros like this:
 </p><pre class="programlisting">
   #define _GLIBCXX_SYNCHRONIZATION_HAPPENS_BEFORE(A) ANNOTATE_HAPPENS_BEFORE(A)
   #define _GLIBCXX_SYNCHRONIZATION_HAPPENS_AFTER(A)  ANNOTATE_HAPPENS_AFTER(A)
 </pre><p>
   Refer to the documentation of each particular tool for details.
 </p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a id="debug.gdb"></a>Using <span class="command"><strong>gdb</strong></span></h3></div></div></div><p>
   </p><p>
   Many options are available for GDB itself: please see <a class="link" href="http://sourceware.org/gdb/current/onlinedocs/gdb/" target="_top">
   "GDB features for C++" </a> in the GDB documentation. Also
   recommended: the other parts of this manual.
 </p><p>
   These settings can either be switched on in at the GDB command line,
   or put into a <code class="filename">.gdbinit</code> file to establish default
   debugging characteristics, like so:
 </p><pre class="programlisting">
    set print pretty on
    set print object on
    set print static-members on
    set print vtbl on
    set print demangle on
    set demangle-style gnu-v3
 </pre><p>
   Starting with version 7.0, GDB includes support for writing
   pretty-printers in Python.  Pretty printers for containers and other
   classes are distributed with GCC from version 4.5.0 and should be installed
   alongside the libstdc++ shared library files and found automatically by
   GDB.
 </p><p>
   Depending where libstdc++ is installed, GDB might refuse to auto-load
   the python printers and print a warning instead.
   If this happens the python printers can be enabled by following the
   instructions GDB gives for setting your <code class="code">auto-load safe-path</code>
   in your <code class="filename">.gdbinit</code> configuration file.
 </p><p>
   Once loaded, standard library classes that the printers support
   should print in a more human-readable format.  To print the classes
   in the old style, use the <strong class="userinput"><code>/r</code></strong> (raw) switch in the
   print command (i.e., <strong class="userinput"><code>print /r foo</code></strong>).  This will
   print the classes as if the Python pretty-printers were not loaded.
 </p><p>
   For additional information on STL support and GDB please visit:
   <a class="link" href="http://sourceware.org/gdb/wiki/STLSupport" target="_top"> "GDB Support
   for STL" </a> in the GDB wiki.  Additionally, in-depth
   documentation and discussion of the pretty printing feature can be
   found in "Pretty Printing" node in the GDB manual.  You can find
   on-line versions of the GDB user manual in GDB's homepage, at
   <a class="link" href="http://sourceware.org/gdb/" target="_top"> "GDB: The GNU Project
   Debugger" </a>.
 </p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a id="debug.exceptions"></a>Tracking uncaught exceptions</h3></div></div></div><p>
   The <a class="link" href="termination.html#support.termination.verbose" title="Verbose Terminate Handler">verbose
   termination handler</a> gives information about uncaught
   exceptions which kill the program.
 </p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a id="debug.debug_mode"></a>Debug Mode</h3></div></div></div><p> The <a class="link" href="debug_mode.html" title="Chapter 17. Debug Mode">Debug Mode</a>
   has compile and run-time checks for many containers.
   </p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a id="debug.compile_time_checks"></a>Compile Time Checking</h3></div></div></div><p> The <a class="link" href="ext_compile_checks.html" title="Chapter 16. Compile Time Checks">Compile-Time
   Checks</a> extension has compile-time checks for many algorithms.
   </p></div></div><div class="navfooter"><hr /><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="using_exceptions.html">Prev</a> </td><td width="20%" align="center"><a accesskey="u" href="using.html">Up</a></td><td width="40%" align="right"> <a accesskey="n" href="std_contents.html">Next</a></td></tr><tr><td width="40%" align="left" valign="top">Exceptions </td><td width="20%" align="center"><a accesskey="h" href="../index.html">Home</a></td><td width="40%" align="right" valign="top"> Part II.
     Standard Contents
   </td></tr></table></div></body></html>
	<?xml version="1.0" encoding="UTF-8" standalone="no"?>
	<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"><html xmlns="http://www.w3.org/1999/xhtml"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><title>Debugging Support</title><meta name="generator" content="DocBook XSL Stylesheets Vsnapshot" /><meta name="keywords" content="C++, debug" /><meta name="keywords" content="ISO C++, library" /><meta name="keywords" content="ISO C++, runtime, library" /><link rel="home" href="../index.html" title="The GNU C++ Library" /><link rel="up" href="using.html" title="Chapter 3. Using" /><link rel="prev" href="using_exceptions.html" title="Exceptions" /><link rel="next" href="std_contents.html" title="Part II. Standard Contents" /></head><body><div class="navheader"><table width="100%" summary="Navigation header"><tr><th colspan="3" align="center">Debugging Support</th></tr><tr><td width="20%" align="left"><a accesskey="p" href="using_exceptions.html">Prev</a> </td><th width="60%" align="center">Chapter 3. Using</th><td width="20%" align="right"> <a accesskey="n" href="std_contents.html">Next</a></td></tr></table><hr /></div><div class="section"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="manual.intro.using.debug"></a>Debugging Support</h2></div></div></div><p>
	There are numerous things that can be done to improve the ease with
	which C++ binaries are debugged when using the GNU tool chain. Here
	are some of them.
	</p><div class="section"><div class="titlepage"><div><div><h3 class="title"><a id="debug.compiler"></a>Using <span class="command"><strong>g++</strong></span></h3></div></div></div><p>
	Compiler flags determine how debug information is transmitted
	between compilation and debug or analysis tools.
	</p><p>
	The default optimizations and debug flags for a libstdc++ build
	are <code class="code">-g -O2</code>. However, both debug and optimization
	flags can be varied to change debugging characteristics. For
	instance, turning off all optimization via the <code class="code">-g -O0
	-fno-inline</code> flags will disable inlining and optimizations,
	and add debugging information, so that stepping through all functions,
	(including inlined constructors and destructors) is possible. In
	addition, <code class="code">-fno-eliminate-unused-debug-types</code> can be
	used when additional debug information, such as nested class info,
	is desired.
	</p><p>
	Or, the debug format that the compiler and debugger use to
	communicate information about source constructs can be changed via
	<code class="code">-gdwarf-2</code> or <code class="code">-gstabs</code> flags: some debugging
	formats permit more expressive type and scope information to be
	shown in GDB. Expressiveness can be enhanced by flags like
	<code class="code">-g3</code>. The default debug information for a particular
	platform can be identified via the value set by the
	PREFERRED_DEBUGGING_TYPE macro in the GCC sources.
	</p><p>
	Many other options are available: please see <a class="link" href="http://gcc.gnu.org/onlinedocs/gcc/Debugging-Options.html#Debugging%20Options" target="_top">"Options
	for Debugging Your Program"</a> in Using the GNU Compiler
	Collection (GCC) for a complete list.
	</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a id="debug.req"></a>Debug Versions of Library Binary Files</h3></div></div></div><p>
	If you would like debug symbols in libstdc++, there are two ways to
	build libstdc++ with debug flags. The first is to create a separate
	debug build by running make from the top-level of a tree
	freshly-configured with
	</p><pre class="programlisting">
	--enable-libstdcxx-debug
	</pre><p>and perhaps</p><pre class="programlisting">
	--enable-libstdcxx-debug-flags='...'
	</pre><p>
	Both the normal build and the debug build will persist, without
	having to specify <code class="code">CXXFLAGS</code>, and the debug library will
	be installed in a separate directory tree, in <code class="code">(prefix)/lib/debug</code>.
	For more information, look at the
	<a class="link" href="configure.html" title="Configure">configuration</a> section.
	</p><p>
	A second approach is to use the configuration flags
	</p><pre class="programlisting">
	make CXXFLAGS='-g3 -fno-inline -O0' all
	</pre><p>
	This quick and dirty approach is often sufficient for quick
	debugging tasks, when you cannot or don't want to recompile your
	application to use the <a class="link" href="debug_mode.html" title="Chapter 17. Debug Mode">debug mode</a>.</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a id="debug.memory"></a>Memory Leak Hunting</h3></div></div></div><p>
	On many targets GCC supports AddressSanitizer, a fast memory error detector,
	which is enabled by the <code class="option">-fsanitize=address</code> option.
	</p><p>
	There are also various third party memory tracing and debug utilities
	that can be used to provide detailed memory allocation information
	about C++ code. An exhaustive list of tools is not going to be
	attempted, but includes <code class="code">mtrace</code>, <code class="code">valgrind</code>,
	<code class="code">mudflap</code> (no longer supported since GCC 4.9.0), ElectricFence,
	and the non-free commercial product <code class="code">purify</code>.
	In addition, <code class="code">libcwd</code>, jemalloc and TCMalloc have replacements
	for the global <code class="code">new</code> and <code class="code">delete</code> operators
	that can track memory allocation and deallocation and provide useful
	memory statistics.
	</p><p>
	For valgrind, there are some specific items to keep in mind. First
	of all, use a version of valgrind that will work with current GNU
	C++ tools: the first that can do this is valgrind 1.0.4, but later
	versions should work better. Second, using an unoptimized build
	might avoid confusing valgrind.
	</p><p>
	Third, it may be necessary to force deallocation in other libraries
	as well, namely the "C" library. On GNU/Linux, this can be accomplished
	with the appropriate use of the <code class="code">__cxa_atexit</code> or
	<code class="code">atexit</code> functions.
	</p><pre class="programlisting">
	#include <cstdlib>

	extern "C" void __libc_freeres(void);

	void do_something() { }

	int main()
	{
	atexit(__libc_freeres);
	do_something();
	return 0;
	}
	</pre><p>or, using <code class="code">__cxa_atexit</code>:</p><pre class="programlisting">
	extern "C" void __libc_freeres(void);
	extern "C" int __cxa_atexit(void (func) (void ), void arg, void d);

	void do_something() { }

	int main()
	{
	extern void* __dso_handle __attribute__ ((__weak__));
	__cxa_atexit((void () (void )) __libc_freeres, NULL,
	&__dso_handle ? __dso_handle : NULL);
	do_test();
	return 0;
	}
	</pre><p>
	Suggested valgrind flags, given the suggestions above about setting
	up the runtime environment, library, and test file, might be:
	</p><pre class="programlisting">
	valgrind -v --num-callers=20 --leak-check=yes --leak-resolution=high --show-reachable=yes a.out
	</pre><div class="section"><div class="titlepage"><div><div><h4 class="title"><a id="debug.memory.mtalloc"></a>Non-memory leaks in Pool and MT allocators</h4></div></div></div><p>
	There are different kinds of allocation schemes that can be used by
	<code class="code">std::allocator</code>. Prior to GCC 3.4.0 the default was to use
	a pooling allocator, <code class="classname">pool_allocator</code>,
	which is still available as the optional
	<code class="classname">__pool_alloc</code> extension.
	Another optional extension, <code class="classname">__mt_alloc</code>,
	is a high-performance pool allocator.
	</p><p>
	In a suspect executable these pooling allocators can give
	the mistaken impression that memory is being leaked,
	when in reality the memory "leak" is a pool being used
	by the library's allocator and is reclaimed after program
	termination.
	</p><p>
	If you're using memory debugging tools on a program that uses
	one of these pooling allocators, you can set the environment variable
	<code class="literal">GLIBCXX_FORCE_NEW</code> to keep extraneous pool allocation
	noise from cluttering debug information.
	For more details, see the
	<a class="link" href="mt_allocator.html" title="Chapter 19. The mt_allocator">mt allocator</a>
	documentation and look specifically for <code class="code">GLIBCXX_FORCE_NEW</code>.
	</p></div></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a id="debug.races"></a>Data Race Hunting</h3></div></div></div><p>
	All synchronization primitives used in the library internals need to be
	understood by race detectors so that they do not produce false reports.
	</p><p>
	Two annotation macros are used to explain low-level synchronization
	to race detectors:
	<code class="code">_GLIBCXX_SYNCHRONIZATION_HAPPENS_BEFORE()</code> and
	<code class="code"> _GLIBCXX_SYNCHRONIZATION_HAPPENS_AFTER()</code>.
	By default, these macros are defined empty -- anyone who wants
	to use a race detector needs to redefine them to call an
	appropriate API.
	Since these macros are empty by default when the library is built,
	redefining them will only affect inline functions and template
	instantiations which are compiled in user code. This allows annotation
	of templates such as <code class="code">shared_ptr</code>, but not code which is
	only instantiated in the library. Code which is only instantiated in
	the library needs to be recompiled with the annotation macros defined.
	That can be done by rebuilding the entire
	<code class="filename">libstdc++.so</code> file but a simpler
	alternative exists for ELF platforms such as GNU/Linux, because ELF
	symbol interposition allows symbols defined in the shared library to be
	overridden by symbols with the same name that appear earlier in the
	runtime search path. This means you only need to recompile the functions
	that are affected by the annotation macros, which can be done by
	recompiling individual files.
	Annotating <code class="code">std::string</code> and <code class="code">std::wstring</code>
	reference counting can be done by disabling extern templates (by defining
	<code class="code">_GLIBCXX_EXTERN_TEMPLATE=-1</code>) or by rebuilding the
	<code class="filename">src/string-inst.cc</code> file.
	Annotating the remaining atomic operations (at the time of writing these
	are in <code class="code">ios_base::Init::~Init</code>, <code class="code">locale::_Impl</code>,
	<code class="code">locale::facet</code> and <code class="code">thread::_M_start_thread</code>)
	requires rebuilding the relevant source files.
	</p><p>
	The approach described above is known to work with the following race
	detection tools:
	<a class="link" href="https://valgrind.org/docs/manual/drd-manual.html" target="_top">
	DRD</a>,
	<a class="link" href="https://valgrind.org/docs/manual/hg-manual.html" target="_top">
	Helgrind</a>, and
	<a class="link" href="https://github.com/google/sanitizers" target="_top">
	ThreadSanitizer</a> (this refers to ThreadSanitizer v1, not the
	new "tsan" feature built-in to GCC itself).
	</p><p>
	With DRD, Helgrind and ThreadSanitizer you will need to define
	the macros like this:
	</p><pre class="programlisting">
	#define _GLIBCXX_SYNCHRONIZATION_HAPPENS_BEFORE(A) ANNOTATE_HAPPENS_BEFORE(A)
	#define _GLIBCXX_SYNCHRONIZATION_HAPPENS_AFTER(A) ANNOTATE_HAPPENS_AFTER(A)
	</pre><p>
	Refer to the documentation of each particular tool for details.
	</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a id="debug.gdb"></a>Using <span class="command"><strong>gdb</strong></span></h3></div></div></div><p>
	</p><p>
	Many options are available for GDB itself: please see <a class="link" href="http://sourceware.org/gdb/current/onlinedocs/gdb/" target="_top">
	"GDB features for C++" </a> in the GDB documentation. Also
	recommended: the other parts of this manual.
	</p><p>
	These settings can either be switched on in at the GDB command line,
	or put into a <code class="filename">.gdbinit</code> file to establish default
	debugging characteristics, like so:
	</p><pre class="programlisting">
	set print pretty on
	set print object on
	set print static-members on
	set print vtbl on
	set print demangle on
	set demangle-style gnu-v3
	</pre><p>
	Starting with version 7.0, GDB includes support for writing
	pretty-printers in Python. Pretty printers for containers and other
	classes are distributed with GCC from version 4.5.0 and should be installed
	alongside the libstdc++ shared library files and found automatically by
	GDB.
	</p><p>
	Depending where libstdc++ is installed, GDB might refuse to auto-load
	the python printers and print a warning instead.
	If this happens the python printers can be enabled by following the
	instructions GDB gives for setting your <code class="code">auto-load safe-path</code>
	in your <code class="filename">.gdbinit</code> configuration file.
	</p><p>
	Once loaded, standard library classes that the printers support
	should print in a more human-readable format. To print the classes
	in the old style, use the <strong class="userinput"><code>/r</code></strong> (raw) switch in the
	print command (i.e., <strong class="userinput"><code>print /r foo</code></strong>). This will
	print the classes as if the Python pretty-printers were not loaded.
	</p><p>
	For additional information on STL support and GDB please visit:
	<a class="link" href="http://sourceware.org/gdb/wiki/STLSupport" target="_top"> "GDB Support
	for STL" </a> in the GDB wiki. Additionally, in-depth
	documentation and discussion of the pretty printing feature can be
	found in "Pretty Printing" node in the GDB manual. You can find
	on-line versions of the GDB user manual in GDB's homepage, at
	<a class="link" href="http://sourceware.org/gdb/" target="_top"> "GDB: The GNU Project
	Debugger" </a>.
	</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a id="debug.exceptions"></a>Tracking uncaught exceptions</h3></div></div></div><p>
	The <a class="link" href="termination.html#support.termination.verbose" title="Verbose Terminate Handler">verbose
	termination handler</a> gives information about uncaught
	exceptions which kill the program.
	</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a id="debug.debug_mode"></a>Debug Mode</h3></div></div></div><p> The <a class="link" href="debug_mode.html" title="Chapter 17. Debug Mode">Debug Mode</a>
	has compile and run-time checks for many containers.
	</p></div><div class="section"><div class="titlepage"><div><div><h3 class="title"><a id="debug.compile_time_checks"></a>Compile Time Checking</h3></div></div></div><p> The <a class="link" href="ext_compile_checks.html" title="Chapter 16. Compile Time Checks">Compile-Time
	Checks</a> extension has compile-time checks for many algorithms.
	</p></div></div><div class="navfooter"><hr /><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="using_exceptions.html">Prev</a> </td><td width="20%" align="center"><a accesskey="u" href="using.html">Up</a></td><td width="40%" align="right"> <a accesskey="n" href="std_contents.html">Next</a></td></tr><tr><td width="40%" align="left" valign="top">Exceptions </td><td width="20%" align="center"><a accesskey="h" href="../index.html">Home</a></td><td width="40%" align="right" valign="top"> Part II.
	Standard Contents
	</td></tr></table></div></body></html>