libstdc++-v3/doc/html/manual/profile_mode.html

   1 <html><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8"><title>Chapter 19. Profile Mode</title><meta name="generator" content="DocBook XSL-NS Stylesheets V1.76.1"><meta name="keywords" content="
   2       C++
   3     ,
   4       library
   5     ,
   6       profile
   7     "><meta name="keywords" content="
   8       ISO C++
   9     ,
  10       library
  11     "><meta name="keywords" content="
  12       ISO C++
  13     ,
  14       runtime
  15     ,
  16       library
  17     "><link rel="home" href="../index.html" title="The GNU C++ Library"><link rel="up" href="extensions.html" title="Part III.  Extensions"><link rel="prev" href="bk01pt03ch18s05.html" title="Testing"><link rel="next" href="bk01pt03ch19s02.html" title="Design"></head><body bgcolor="white" text="black" link="#0000FF" vlink="#840084" alink="#0000FF"><div class="navheader"><table width="100%" summary="Navigation header"><tr><th colspan="3" align="center">Chapter 19. Profile Mode</th></tr><tr><td width="20%" align="left"><a accesskey="p" href="bk01pt03ch18s05.html">Prev</a> </td><th width="60%" align="center">Part III.
  18   Extensions
  19
  20 </th><td width="20%" align="right"> <a accesskey="n" href="bk01pt03ch19s02.html">Next</a></td></tr></table><hr></div><div class="chapter" title="Chapter 19. Profile Mode"><div class="titlepage"><div><div><h2 class="title"><a name="manual.ext.profile_mode"></a>Chapter 19. Profile Mode</h2></div></div></div><div class="toc"><p><b>Table of Contents</b></p><dl><dt><span class="section"><a href="profile_mode.html#manual.ext.profile_mode.intro">Intro</a></span></dt><dd><dl><dt><span class="section"><a href="profile_mode.html#manual.ext.profile_mode.using">Using the Profile Mode</a></span></dt><dt><span class="section"><a href="profile_mode.html#manual.ext.profile_mode.tuning">Tuning the Profile Mode</a></span></dt></dl></dd><dt><span class="section"><a href="bk01pt03ch19s02.html">Design</a></span></dt><dd><dl><dt><span class="section"><a href="bk01pt03ch19s02.html#manual.ext.profile_mode.design.wrapper">Wrapper Model</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s02.html#manual.ext.profile_mode.design.instrumentation">Instrumentation</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s02.html#manual.ext.profile_mode.design.rtlib">Run Time Behavior</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s02.html#manual.ext.profile_mode.design.analysis">Analysis and Diagnostics</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s02.html#manual.ext.profile_mode.design.cost-model">Cost Model</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s02.html#manual.ext.profile_mode.design.reports">Reports</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s02.html#manual.ext.profile_mode.design.testing">Testing</a></span></dt></dl></dd><dt><span class="section"><a href="bk01pt03ch19s03.html">Extensions for Custom Containers</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s04.html">Empirical Cost Model</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s05.html">Implementation Issues</a></span></dt><dd><dl><dt><span class="section"><a href="bk01pt03ch19s05.html#manual.ext.profile_mode.implementation.stack">Stack Traces</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s05.html#manual.ext.profile_mode.implementation.symbols">Symbolization of Instruction Addresses</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s05.html#manual.ext.profile_mode.implementation.concurrency">Concurrency</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s05.html#manual.ext.profile_mode.implementation.stdlib-in-proflib">Using the Standard Library in the Instrumentation Implementation</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s05.html#manual.ext.profile_mode.implementation.malloc-hooks">Malloc Hooks</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s05.html#manual.ext.profile_mode.implementation.construction-destruction">Construction and Destruction of Global Objects</a></span></dt></dl></dd><dt><span class="section"><a href="bk01pt03ch19s06.html">Developer Information</a></span></dt><dd><dl><dt><span class="section"><a href="bk01pt03ch19s06.html#manual.ext.profile_mode.developer.bigpic">Big Picture</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s06.html#manual.ext.profile_mode.developer.howto">How To Add A Diagnostic</a></span></dt></dl></dd><dt><span class="section"><a href="bk01pt03ch19s07.html">Diagnostics</a></span></dt><dd><dl><dt><span class="section"><a href="bk01pt03ch19s07.html#manual.ext.profile_mode.analysis.template">Diagnostic Template</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s07.html#manual.ext.profile_mode.analysis.containers">Containers</a></span></dt><dd><dl><dt><span class="section"><a href="bk01pt03ch19s07.html#manual.ext.profile_mode.analysis.hashtable_too_small">Hashtable Too Small</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s07.html#manual.ext.profile_mode.analysis.hashtable_too_large">Hashtable Too Large</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s07.html#manual.ext.profile_mode.analysis.inefficient_hash">Inefficient Hash</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s07.html#manual.ext.profile_mode.analysis.vector_too_small">Vector Too Small</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s07.html#manual.ext.profile_mode.analysis.vector_too_large">Vector Too Large</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s07.html#manual.ext.profile_mode.analysis.vector_to_hashtable">Vector to Hashtable</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s07.html#manual.ext.profile_mode.analysis.hashtable_to_vector">Hashtable to Vector</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s07.html#manual.ext.profile_mode.analysis.vector_to_list">Vector to List</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s07.html#manual.ext.profile_mode.analysis.list_to_vector">List to Vector</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s07.html#manual.ext.profile_mode.analysis.list_to_slist">List to Forward List (Slist)</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s07.html#manual.ext.profile_mode.analysis.assoc_ord_to_unord">Ordered to Unordered Associative Container</a></span></dt></dl></dd><dt><span class="section"><a href="bk01pt03ch19s07.html#manual.ext.profile_mode.analysis.algorithms">Algorithms</a></span></dt><dd><dl><dt><span class="section"><a href="bk01pt03ch19s07.html#manual.ext.profile_mode.analysis.algorithms.sort">Sort Algorithm Performance</a></span></dt></dl></dd><dt><span class="section"><a href="bk01pt03ch19s07.html#manual.ext.profile_mode.analysis.locality">Data Locality</a></span></dt><dd><dl><dt><span class="section"><a href="bk01pt03ch19s07.html#manual.ext.profile_mode.analysis.locality.sw_prefetch">Need Software Prefetch</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s07.html#manual.ext.profile_mode.analysis.locality.linked">Linked Structure Locality</a></span></dt></dl></dd><dt><span class="section"><a href="bk01pt03ch19s07.html#manual.ext.profile_mode.analysis.mthread">Multithreaded Data Access</a></span></dt><dd><dl><dt><span class="section"><a href="bk01pt03ch19s07.html#manual.ext.profile_mode.analysis.mthread.ddtest">Data Dependence Violations at Container Level</a></span></dt><dt><span class="section"><a href="bk01pt03ch19s07.html#manual.ext.profile_mode.analysis.mthread.false_share">False Sharing</a></span></dt></dl></dd><dt><span class="section"><a href="bk01pt03ch19s07.html#manual.ext.profile_mode.analysis.statistics">Statistics</a></span></dt></dl></dd><dt><span class="bibliography"><a href="profile_mode.html#profile_mode.biblio">Bibliography</a></span></dt></dl></div><div class="section" title="Intro"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a name="manual.ext.profile_mode.intro"></a>Intro</h2></div></div></div><p>
  21   <span class="emphasis"><em>Goal: </em></span>Give performance improvement advice based on
  22   recognition of suboptimal usage patterns of the standard library.
  23   </p><p>
  24   <span class="emphasis"><em>Method: </em></span>Wrap the standard library code.  Insert
  25   calls to an instrumentation library to record the internal state of
  26   various components at interesting entry/exit points to/from the standard
  27   library.  Process trace, recognize suboptimal patterns, give advice.
  28   For details, see
  29   <a class="link" href="http://dx.doi.org/10.1109/CGO.2009.36" target="_top">paper presented at
  30    CGO 2009</a>.
  31   </p><p>
  32   <span class="emphasis"><em>Strengths: </em></span>
  33 </p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>
  34   Unintrusive solution.  The application code does not require any
  35   modification.
  36   </p></li><li class="listitem"><p> The advice is call context sensitive, thus capable of
  37   identifying precisely interesting dynamic performance behavior.
  38   </p></li><li class="listitem"><p>
  39   The overhead model is pay-per-view.  When you turn off a diagnostic class
  40   at compile time, its overhead disappears.
  41   </p></li></ul></div><p>
  42   </p><p>
  43   <span class="emphasis"><em>Drawbacks: </em></span>
  44 </p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>
  45   You must recompile the application code with custom options.
  46   </p></li><li class="listitem"><p>You must run the application on representative input.
  47   The advice is input dependent.
  48   </p></li><li class="listitem"><p>
  49   The execution time will increase, in some cases by factors.
  50   </p></li></ul></div><p>
  51   </p><div class="section" title="Using the Profile Mode"><div class="titlepage"><div><div><h3 class="title"><a name="manual.ext.profile_mode.using"></a>Using the Profile Mode</h3></div></div></div><p>
  52   This is the anticipated common workflow for program <code class="code">foo.cc</code>:
  53 </p><pre class="programlisting">
  54 $ cat foo.cc
  55 #include &lt;vector&gt;
  56 int main() {
  57   vector&lt;int&gt; v;
  58   for (int k = 0; k &lt; 1024; ++k) v.insert(v.begin(), k);
  59 }
  60
  61 $ g++ -D_GLIBCXX_PROFILE foo.cc
  62 $ ./a.out
  63 $ cat libstdcxx-profile.txt
  64 vector-to-list: improvement = 5: call stack = 0x804842c ...
  65     : advice = change std::vector to std::list
  66 vector-size: improvement = 3: call stack = 0x804842c ...
  67     : advice = change initial container size from 0 to 1024
  68 </pre><p>
  69   </p><p>
  70   Anatomy of a warning:
  71   </p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>
  72   Warning id.  This is a short descriptive string for the class
  73   that this warning belongs to.  E.g., "vector-to-list".
  74   </p></li><li class="listitem"><p>
  75   Estimated improvement.  This is an approximation of the benefit expected
  76   from implementing the change suggested by the warning.  It is given on
  77   a log10 scale.  Negative values mean that the alternative would actually
  78   do worse than the current choice.
  79   In the example above, 5 comes from the fact that the overhead of
  80   inserting at the beginning of a vector vs. a list is around 1024 * 1024 / 2,
  81   which is around 10e5.  The improvement from setting the initial size to
  82   1024 is in the range of 10e3, since the overhead of dynamic resizing is
  83   linear in this case.
  84   </p></li><li class="listitem"><p>
  85   Call stack.  Currently, the addresses are printed without
  86   symbol name or code location attribution.
  87   Users are expected to postprocess the output using, for instance, addr2line.
  88   </p></li><li class="listitem"><p>
  89   The warning message.  For some warnings, this is static text, e.g.,
  90   "change vector to list".  For other warnings, such as the one above,
  91   the message contains numeric advice, e.g., the suggested initial size
  92   of the vector.
  93   </p></li></ul></div><p>
  94   </p><p>Three files are generated.  <code class="code">libstdcxx-profile.txt</code>
  95    contains human readable advice.  <code class="code">libstdcxx-profile.raw</code>
  96    contains implementation specific data about each diagnostic.
  97    Their format is not documented.  They are sufficient to generate
  98    all the advice given in <code class="code">libstdcxx-profile.txt</code>.  The advantage
  99    of keeping this raw format is that traces from multiple executions can
 100    be aggregated simply by concatenating the raw traces.  We intend to
 101    offer an external utility program that can issue advice from a trace.
 102    <code class="code">libstdcxx-profile.conf.out</code> lists the actual diagnostic
 103    parameters used.  To alter parameters, edit this file and rename it to
 104    <code class="code">libstdcxx-profile.conf</code>.
 105   </p><p>Advice is given regardless whether the transformation is valid.
 106   For instance, we advise changing a map to an unordered_map even if the
 107   application semantics require that data be ordered.
 108   We believe such warnings can help users understand the performance
 109   behavior of their application better, which can lead to changes
 110   at a higher abstraction level.
 111   </p></div><div class="section" title="Tuning the Profile Mode"><div class="titlepage"><div><div><h3 class="title"><a name="manual.ext.profile_mode.tuning"></a>Tuning the Profile Mode</h3></div></div></div><p>Compile time switches and environment variables (see also file
 112    profiler.h).  Unless specified otherwise, they can be set at compile time
 113    using -D_&lt;name&gt; or by setting variable &lt;name&gt;
 114    in the environment where the program is run, before starting execution.
 115   </p><div class="itemizedlist"><ul class="itemizedlist" type="disc"><li class="listitem"><p>
 116    <code class="code">_GLIBCXX_PROFILE_NO_&lt;diagnostic&gt;</code>:
 117    disable specific diagnostics.
 118    See section Diagnostics for possible values.
 119    (Environment variables not supported.)
 120    </p></li><li class="listitem"><p>
 121    <code class="code">_GLIBCXX_PROFILE_TRACE_PATH_ROOT</code>: set an alternative root
 122    path for the output files.
 123    </p></li><li class="listitem"><p>_GLIBCXX_PROFILE_MAX_WARN_COUNT: set it to the maximum
 124    number of warnings desired.  The default value is 10.</p></li><li class="listitem"><p>
 125    <code class="code">_GLIBCXX_PROFILE_MAX_STACK_DEPTH</code>: if set to 0,
 126    the advice will
 127    be collected and reported for the program as a whole, and not for each
 128    call context.
 129    This could also be used in continuous regression tests, where you
 130    just need to know whether there is a regression or not.
 131    The default value is 32.
 132    </p></li><li class="listitem"><p>
 133    <code class="code">_GLIBCXX_PROFILE_MEM_PER_DIAGNOSTIC</code>:
 134    set a limit on how much memory to use for the accounting tables for each
 135    diagnostic type.  When this limit is reached, new events are ignored
 136    until the memory usage decreases under the limit.  Generally, this means
 137    that newly created containers will not be instrumented until some
 138    live containers are deleted.  The default is 128 MB.
 139    </p></li><li class="listitem"><p>
 140    <code class="code">_GLIBCXX_PROFILE_NO_THREADS</code>:
 141    Make the library not use threads.  If thread local storage (TLS) is not
 142    available, you will get a preprocessor error asking you to set
 143    -D_GLIBCXX_PROFILE_NO_THREADS if your program is single-threaded.
 144    Multithreaded execution without TLS is not supported.
 145    (Environment variable not supported.)
 146    </p></li><li class="listitem"><p>
 147    <code class="code">_GLIBCXX_HAVE_EXECINFO_H</code>:
 148    This name should be defined automatically at library configuration time.
 149    If your library was configured without <code class="code">execinfo.h</code>, but
 150    you have it in your include path, you can define it explicitly.  Without
 151    it, advice is collected for the program as a whole, and not for each
 152    call context.
 153    (Environment variable not supported.)
 154    </p></li></ul></div><p>
 155   </p></div></div><div class="bibliography" title="Bibliography"><div class="titlepage"><div><div><h2 class="title"><a name="profile_mode.biblio"></a>Bibliography</h2></div></div></div><div class="biblioentry"><a name="id638547"></a><p><span class="citetitle"><em class="citetitle">
 156       Perflint: A Context Sensitive Performance Advisor for C++ Programs
 157     </em>. </span><span class="author"><span class="firstname">Lixia</span> <span class="surname">Liu</span>. </span><span class="author"><span class="firstname">Silvius</span> <span class="surname">Rus</span>. </span><span class="copyright">Copyright © 2009 . </span><span class="publisher"><span class="publishername">
 158         Proceedings of the 2009 International Symposium on Code Generation
 159         and Optimization
 160       . </span></span></p></div></div></div><div class="navfooter"><hr><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="bk01pt03ch18s05.html">Prev</a> </td><td width="20%" align="center"><a accesskey="u" href="extensions.html">Up</a></td><td width="40%" align="right"> <a accesskey="n" href="bk01pt03ch19s02.html">Next</a></td></tr><tr><td width="40%" align="left" valign="top">Testing </td><td width="20%" align="center"><a accesskey="h" href="../index.html">Home</a></td><td width="40%" align="right" valign="top"> Design</td></tr></table></div></body></html>