Intel® C++ Compiler 16.0 User and Reference Guide
Inserts instrumentation calls at a function's entry and exit points, and before and after instrumentable loops.
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
Linux and OS X: | -profile-loops=keyword |
Windows: | /Qprofile-loops:keyword |
keyword |
Specifies which type of loops should have instrumentation. Possible values are:
|
OFF |
No instrumentation calls are inserted at a function's entry and exit points, or before and after instrumentable loop. |
This option inserts instrumentation calls at a function's entry and exit points within a single-threaded application. For unthreaded applications, it also inserts instrumentation before and after instrumentable loops of the type listed in keyword.
When the instrumented application is run, this option causes the generation of a loop_prof_funcs_<name>.dump file and a loop_prof_funcs_<name>.dump file, where <name> is a timestamp for the run.
The same timestamp is used for the loop file and function file. This identifies that the loop data and function data were from the same program run.
The same data values are also dumped into the file loop_prof_<name>.xml for use with the data viewer application, unless you turn off the output format by setting the environment variable INTEL_LOOP_PROF_XML_DUMP to 0.
None