What's new? Update 11 - Intel® VTune™ Amplifier XE 2013

Intel® VTune™ Amplifier XE 2013

Intel® VTune™ Amplifier XE is an easy to use performance and thread profiler for C, C++, C#, Fortran, Java and MPI developers. No special recompiles are needed, just start profiling.  Hotspots are highlighted on the source.  A powerful timeline makes it easy to tune your application and scale performance on multicore processors.

New for Update 11!  

Resources

Contents

 

File: vtune_amplifier_xe_2013_update11.tar.gz

Installer for Intel® Vtune™ Amplifier XE 2013 Update 11 for Linux*

File: VTune_Amplifier_XE_2013_update11_setup.exe

Installer for Intel® Vtune™ Amplifier XE 2013 Update 11 for Windows*

* Other names and brands may be claimed as the property of others.

Microsoft, Windows, Visual Studio, Visual C++, and the Windows logo are trademarks, or registered trademarks of Microsoft Corporation in the United States and/or other countries.

NextWhat's New in Update 10?

 

Support for identifying function boundaries using static binary analysis methods for binaries without symbol information

To provide accurate performance data and enable source analysis, the Intel® VTune™ Amplifier requires debug information for the binary files it analyzes. Effective Update 11 if it does not find debug information in the binaries, the VTune Amplifier statically identifies function boundaries and assigns hotspot addresses to generated pseudo names func@address for such functions. For more information please refer to the Using Debug Informationtopic in the product help.

 

NOTE: If debug information is absent, the Call Stack pane may not unwind the call stack correctly for user-mode sampling and tracing analysis types. Additionally in some cases, it can take significantly more time to finalize the results for modules that do not have debug information.

General Exploration metrics summary for hardware event-based sampling analysis results in the command line reports

Command line reports now provide General Exploration metrics summary for hardware event-based sampling analysis results providing a high-level overview of performance problems. The General Exploration Metrics section appears in a Summary report if events were collected during analysis. The set of metrics displayed in the summary depends on the profiled CPU type and list of events. For more information please refer to the Viewing a Summary Report topic in the product help.

Source Function Stack grouping level enabling more accurate result comparison in the Top-down Tree pane

Use Source Function Stack grouping level in the Top-down Tree pane for enabling more accurate result comparison for recompiled binary files when addresses of the same source function or same loop are different, like in these cases:

  • You slightly changed the source and recompiled
  • You changed compilation options and recompiled
  •  You are doing compare between results compiled and collected for different microarchitectures. 

By default, compared functions are grouped by the Function Stack granularity, which is based on function instances. VTune Amplifier treats the same functions with different addresses as separate instances and does not compare them:

When the data is aggregated by Source Function Stack, the VTune Amplifier ignores start addresses and compares functions by source file objects:

For more information please refer to the About Viewing Comparison Datatopic in the product help. 

Change Stack Layout option in the Top-down Tree and Bottom-up panes to switch between chain and tree types of stack layout 

Use the Change Stack Layout option in the Top-down Tree and Bottom-up panes to manage stack data in the grids and switch between chain and tree types of stack layout. Click the Change Stack Layout button to switch between call stack layouts.

Chain layouts  are typically more useful for the bottom-up view:

While tree layouts  are more natural for the top-down view:

 

Support for scientific data representation in the grid

Bottom-up and Top-down Tree panes now support displaying performance values in the scientific notation via Show Data As context menu. Typically this format is recommended for analyzing values < 0.001. For more information please refer to the “Choosing Data Format” topic in the product help.

*Other names and brands may be claimed as the property of others.

Para obtener información más completa sobre las optimizaciones del compilador, consulte nuestro Aviso de optimización.