Tutorial: Analyzing Locks and Waits - C++ Sample Code

Última atualização em 6 de setembro de 2016

Intel® VTune™ Amplifier for Linux* OS

Discover how to use the Locks and Waits analysis of Intel® VTune™ Amplifier to identify one of the most common reasons for an inefficient parallel application - threads waiting too long on synchronization objects (locks) while processor cores are underutilized. Focus your tuning efforts on objects with long waits where the system is underutilized.

About This Tutorial	This tutorial uses the sample `tachyon_analyze_locks` application and guides you through basic steps required to analyze the code for locks and waits.
Estimated Duration	10-15 minutes.
Learning Objectives	After you complete this tutorial, you should be able to: Choose an analysis target. Choose the Locks and Waits analysis type. Run the Locks and Waits analysis. Identify the synchronization objects with long waits and poor thread concurrency level. Analyze the source code to locate the most critical code lines. Compare results before and after optimization.
More Resources	Intel VTune Amplifier tutorials (HTML, PDF): https://software.intel.com/en-us/articles/intel-vtune-amplifier-tutorials/ Intel VTune Amplifier support page: https://software.intel.com/en-us/intel-vtune-amplifier-xe-support/ Intel Parallel Studio XE support page: https://software.intel.com/en-us/intel-parallel-studio-xe/ Intel System Studio support page: https://software.intel.com/en-us/intel-system-studio/

Start Here

Para obter informações mais completas sobre otimizações do compilador, consulte nosso aviso de otimização.