I tried out linux&#39; [perf](https://perf.wiki.kernel.org/index.php/Main_Page) utility today and am having trouble in interpreting its results. I&#39;m used to valgrind&#39;s callgrind which is of course a totally different approach to the sampling based method of perf.

What I did:

    perf record -g -p $(pidof someapp)
    perf report -g -n

Now I see something like this:

&lt;pre&gt;
+     16.92%  kdevelop  libsqlite3.so.0.8.6               [.] 0x3fe57                                                                                                              ↑
+     10.61%  kdevelop  libQtGui.so.4.7.3                 [.] 0x81e344                                                                                                             ▮
+      7.09%  kdevelop  libc-2.14.so                      [.] 0x85804                                                                                                              ▒
+      4.96%  kdevelop  libQtGui.so.4.7.3                 [.] 0x265b69                                                                                                             ▒
+      3.50%  kdevelop  libQtCore.so.4.7.3                [.] 0x18608d                                                                                                             ▒
+      2.68%  kdevelop  libc-2.14.so                      [.] memcpy                                                                                                               ▒
+      1.15%  kdevelop  [kernel.kallsyms]                 [k] copy_user_generic_string                                                                                             ▒
+      0.90%  kdevelop  libQtGui.so.4.7.3                 [.] QTransform::translate(double, double)                                                                                ▒
+      0.88%  kdevelop  libc-2.14.so                      [.] __libc_malloc                                                                                                        ▒
+      0.85%  kdevelop  libc-2.14.so                      [.] memcpy 
...
&lt;/pre&gt;

Ok, these functions might be slow, but how do I find out where they are getting called from? As all these hotspots lie in external libraries I see no way to optimize my code. 

Basically I am looking for some kind of callgraph annotated with accumulated cost, where my functions have a higher inclusive sampling cost than the library functions I call.

Is this possible with perf? If so - how?

Note: I found out that &quot;E&quot; unwraps the callgraph and gives somewhat more information. But the callgraph is often not deep enough and/or terminates randomly without giving information about how much info was spent where. Example:

&lt;pre&gt;
-     10.26%  kate  libkatepartinterfaces.so.4.6.0  [.] Kate::TextLoader::readLine(int&amp;...
     Kate::TextLoader::readLine(int&amp;, int&amp;)                                            
     Kate::TextBuffer::load(QString const&amp;, bool&amp;, bool&amp;)                              
     KateBuffer::openFile(QString const&amp;)                                              
     KateDocument::openFile()                                                          
     0x7fe37a81121c
&lt;/pre&gt;

Could it be an issue that I&#39;m running on 64 bit? See also: http://lists.fedoraproject.org/pipermail/devel/2010-November/144952.html (I&#39;m not using fedora but seems to apply to all 64bit systems).

All Perf Solutions on Gang of Coders

Total of 1 Perf Solutions

linux perf: how to interpret and find hotspots