Compiler Switches for Performance Analysis on Windows* Targets
Intel® VTune™
can analyze most native binaries on Windows* target systems. However, the settings below are recommended to make the performance analysis more productive and easier:
Profiler
Use This Switch
| To Do This
|
|---|---|
/Zi (highly recommended)
| Enable generating the symbol information required to associate addresses with source lines and to properly walk the call stack in
user-mode sampling and tracing analysis types (Hotspots and Threading).
|
Release build (highly recommended)
| Enable maximum compiler optimization to focus the
VTune
on real performance problems that cannot be optimized with the compiler.
Profiler |
/debug:inline-debug-info | Enable the
VTune
to identify inline functions and, according to the selected
Inline mode, associate the symbols for an inline function with the inline function itself or its caller.
Profiler |
/MD or /MDd | Enable identifying the C runtime calls as system functions and differentiating them from the user code when a proper
Call stack mode is applied to the
VTune
collection result.
Profiler |
/D "TBB_USE_THREADING_TOOLS" | Without
TBB_USE_THREADING_TOOLS set, the
VTune
will not properly identify concurrency issues related to using Intel TBB constructs.
Profiler |
/Qopenmp (highly recommended)
(Intel C++ Compiler)
| Enable the
VTune
to identify parallel regions due to OpenMP* pragmas.
Profiler |
/Qopenmp-link:dynamic (Intel C++ Compiler)
| Enable the Intel Compiler to choose the dynamic version of the OpenMP runtime libraries which has been instrumented for the
VTune
. Usually, this option is enabled for the Intel Compiler by default.
Profiler |
/Qparallel-source-info=2 (Intel C++ Compiler)
| Enable/disable source location emission when OpenMP or auto-parallelism code is generated. 2 is the level of source location emission that tells the compiler to emit path, file, routine name, and line information.
|
-gline-tables-only -fdebug-info-for-profiling
Intel oneAPI DPC++ Compiler
| Enable generating debug information for
GPU analysis of a DPC++ application.
|
-Xsprofile
Intel oneAPI DPC++ Compiler
| Enable source-level mapping of performance data for
FPGA application analysis.
|
The following compiler settings are NOT recommended:
Do Not Use This Switch
| Because Of This
|
|---|---|
debug:parallel | Enables the Intel® Parallel Debugger Extension for the Intel Compiler, which is not used for the
VTune
.
Profiler |
/Qopenmp-link:static | Chooses the static version of the OpenMP runtime libraries for the Intel Compiler. This version of the OpenMP runtime library does not contain the instrumentation data required for the
VTune
analysis.
Profiler |
/Qopenmp_stubs | Prevents OpenMP code from being parallel.
|
Explore the list of libraries recommended or not recommended for the user-mode sampling and tracing analysis types:
Library
| Recommended
| Not Recommended
|
|---|---|---|
OpenMP Runtime (supplied by the Intel Compiler)
| libiomp5md.dll or
libguide40.dll | libiomp5mt.lib ,
libguide.lib ,
vcomp80.dll/vcomp90.dll , or
vcomp80d.dll/vcomp90d.dll |
C Runtime
| msvcr90.dll ,
msvcr80.dll ,
msvcr90d.dll , or
msvcr80d.dll | libcmt.lib |
Product and Performance Information
|
|---|
Performance varies by use, configuration and other factors. Learn more at
www.Intel.com/PerformanceIndex.
Notice revision #20201201
|