Performance-Monitoring Counters Library, for Intel/AMD Processors and Linux
This example introduces
   rabbit command-line option -Clean n

Return to Main Menu


Command-Line Options -Clean n Make n attempts to determine the overhead associated with a call to pmc_read(), and subtract this value from the measured cycles and events when pmc_accumulate() is used. The default is n = 0, so the measurements are not "corrected". Usually n = 10 should be sufficient.
Notes The overhead calculation is done when pmc_select() is first invoked. To avoid overcorrecting, the minimum of n trials is used, and pmc_accumulate() will not subtract more than the measurement.

Performance-Monitoring Counters Library, for Intel/AMD Processors and Linux
Author: Don Heller, dheller@scl.ameslab.gov
Last revised: 2 August 2000