User Guide

  • 2020
  • 06/18/2020
  • Public Content
Contents

FreeBSD* Targets

Intel® VTune™
Profiler
allows you to collect performance data on a FreeBSD* target system.
Intel VTune
Profiler
is not installed on the FreeBSD target system. Instead, you are able to install
VTune
Profiler
on a Linux*, Windows*, or macOS* host system and use a target package for collecting event-based sampling data on a remote FreeBSD target system in one of the following ways:
  • Using
    VTune
    Profiler
    's automated remote collection capability (command line or user interface)
  • Collecting the results locally on the FreeBSD system and copying them to the host system for viewing with
    VTune
    Profiler
    (command line only)
The following sections explain these options in more detail.

Supported Features

Remote Collection
Local Collection
Collection from Linux, Windows, or macOS host system using the
Intel VTune
Profiler
GUI or command line (
vtune
)
Collection from the FreeBSD system using:
  • Intel VTune
    Profiler
    command line (
    vtune
    )
  • Sampling enabling product (SEP) collectors
Analysis Types:
Analysis types (
VTune
Profiler
command line only):
View results on host system
View results in
VTune
Profiler
on a Linux, Windows, or macOS host system
Use of ITT APIs is not supported on FreeBSD systems.

Remote Collection from Host System

  1. Install
    VTune
    Profiler
    on your Linux*, Windows*, or macOS* host. Refer to the Installation Guide for your host system for detailed instructions.
  2. Install the appropriate sampling drivers on the FreeBSD target system. For more information, see FreeBSD* System Setup.
  3. [Optional] If you want to collect performance data with stacks, build your FreeBSD target application using the
    -fno-omit-frame-pointer
    compiler option, to allow the sampling collector to determine the call chain via frame pointer analysis.
  4. Collect performance data using remote analysis from the host system from the
    VTune
    Profiler
    command line or GUI.
    1. Create or open a project.
    2. Specify your target application and remote system and make sure to specify search directories for symbol/source files required for finalization on the host.
    3. Choose and configure an analysis type.
      Supported
      VTune
      Profiler
      analysis types (event-based sampling analysis only) include:
    4. Run the analysis from the host. Depending on your settings, the application launches and runs automatically. Once collection is finished, the result is finalized and displayed with the
      Summary
      window open.
  5. Review the results on the host system.

Native Collection on FreeBSD System

  1. Install
    VTune
    Profiler
    on your Linux*, Windows*, or macOS* host. Refer to the Installation Guide for your host system for detailed instructions.
  2. Install the appropriate sampling drivers on the FreeBSD target system. For more information, see FreeBSD* System Setup.
  3. [Optional] If you want to collect performance data with stacks, build your FreeBSD target application using the
    -fno-omit-frame-pointer
    compiler option, which allows the sampling collector to determine the call chain via frame pointer analysis.
  4. Collect performance data using one of the following methods. For more information about each of these methods, see Remote Linux Target Setup.
  5. Copy the results to the host system.
  6. Review the results with
    VTune
    Profiler
    .
    • If you used the
      vtune
      command, open the
      *.
      vtune
      file.
    • If you collected SEP data, import the
      *.tb7
      file.

Product and Performance Information

1

Intel's compilers may or may not optimize to the same degree for non-Intel microprocessors for optimizations that are not unique to Intel microprocessors. These optimizations include SSE2, SSE3, and SSSE3 instruction sets and other optimizations. Intel does not guarantee the availability, functionality, or effectiveness of any optimization on microprocessors not manufactured by Intel. Microprocessor-dependent optimizations in this product are intended for use with Intel microprocessors. Certain optimizations not specific to Intel microarchitecture are reserved for Intel microprocessors. Please refer to the applicable product User and Reference Guides for more information regarding the specific instruction sets covered by this notice.

Notice revision #20110804