User Guide

Contents

Why Not Offloaded: Cannot Be Modeled

Symptoms

A code region of interest has
Cannot Be Modeled
as a reason why it is not offloaded.

Cause and Solution

The
Cannot be modeled
reason is always followed by one of the following messages that clarify what makes offloading impossible.
Message
Cause and Details
Solution
Cannot be modeled: Outside of Marked Region
Offload Advisor cannot model performance for a code region because it is not marked up for analysis.
Make sure a code region satisfies all markup rules or use a different markup strategy. For more information, see Code Region is not Marked Up.
Cannot be modeled: Not Executed
A code region is in the call tree, but the Offload Advisor detected no calls to it for a dataset used during Survey.
This can happen if execution time of the loop is very small and close to the sampling interval of the Intel® Advisor. Such loops can have significant inaccuracies in time measurement. By default, the sampling interval for Survey is 0.01 seconds.
You can try to decrease the sampling interval of the Intel Advisor:
  1. Run the
    collect.py
    script with the
    --dry-run
    option to generate
    advixe-cl
    command lines appropriate for your application.
  2. Copy the first command with
    --collect=survey
    , add the
    --interval=<interval>
    option with a value less than 0.01, and run the command. For example:
    advixe-cl --collect=survey --project-dir=<project-dir> --interval=0.005 --return-app-exitcode --auto-finalize --static-instruction-mix -- <target>
  3. Copy and paste all other commands generated by
    --dry-run
    and run them as they are.
  4. Rerun performance modeling with
    analyze.py
    to generate performance prediction output.
Cannot be modeled: Internal Error
Internal Error
means incorrect data or lack of data because the Offload Advisor encountered issues when collecting or processing data.
For example, the internal error can happen if:
  • The instances of a code region are in the Top-Down view, but some of the instances are not aggregated to the region in the Bottom-Up view.
  • Code region does not have the Instruction Mix data.
You might try to rerun the collection and performance projection to fix the metrics attribution problem. If this does not help, please use the Analyzers community forum for technical support.
Cannot be modeled: System Module
This code region is a system function/loop.
This is not an issue. If this code region is inside an offload region, or a runtime call, its execution time is added to execution time of offloaded regions.
Cannot be modeled: No Execution Count
The Offload Advisor detected no calls to a loop during Trip Count analysis and no information about the Execution Counts is available for this loop.
This can happen if the Offload Advisor incorrectly attributed metrics collected to the application call tree.
Verify the issue:
  1. Locate your loop of interest in the Call Tree section of the HTML report.
  2. Scroll to
    Trip Counts
    column group and check values reported in the
    Call Count
    and
    Average Trip Counts
    (for loops) columns. If there are no execution counts for the loop, these columns do not contain a value or report zero.
  3. If
    Call Count
    and
    Average Trip Counts
    report zero or no value, make sure the call tree is not broken. For more information, see the Broken Call Tree.
If the call tree is not broken and the
Call Count
and
Average Trip Counts
report zero or no value, there is a problem with metrics attribution.
Try to rerun the collection and performance projection to fix the metrics attribution problem.

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 reserverd 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