Intel® C++ Compiler 16.0 User and Reference Guide

profile-loops-report, Qprofile-loops-report

Controls the level of detail for the data collected when instrumentation occurs before and after certain loops.

Architecture Restrictions

Not available on Intel® 64 architecture targeting the Intel® Xeon Phi™ coprocessor x100 product family (formerly code name Knights Corner), on IA-32 architecture targeting Intel® Graphics Technology, or on Intel® 64 architecture targeting Intel® Graphics Technology

Syntax

Linux and OS X:

-profile-loops-report[=n]

Windows:

/Qprofile-loops-report[:n]

Arguments

n

Is a value denoting the level of detail to report. Possible values are:

1

Reports the cycle counts on entry and exits of loops. This is the default if n is not specified.

2

Reports the level 1 default details, but also includes the loop min/max and average loop iteration counts.

To collect the loop iteration counts, additional instrumentation is inserted. This can increase overhead in the instrumented application and slow performance.

Default

1

The report shows the cycle counts on entry and exits of loops.

Description

This option controls the level of detail for the data collected when instrumentation occurs before and after certain loops. To use this option, you must also specify option [Q]profile-loops.

The report appears in file loop_prof_loops_<name>.dump, where <name> is a timestamp value for the run. The columns listed in the report will be based on the level of detail that was selected during instrumentation.

It is recommended that the same report level be used for all files that are instrumented for the application. If different files of the application were instrumented with different levels, the report will contain all the columns of the highest detail level, but with default values for unavailable fields for files that were instrumented at lower levels.

IDE Equivalent

None

Alternate Options

None

See Also