User Guide

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

workflow

Explain typical Intel Advisor user scenarios, with corresponding command lines.

Syntax

--workflow
Usage
Explain typical
Intel Advisor
user scenarios, with corresponding command lines. For example:
Add SIMD Parallelism:
1. Find hotspots.
advixe-cl --collect=survey --project-dir=./advi --search-dir src:r=./src -- ./bin/myApplication advixe-cl --report=survey --project-dir=./advi --search-dir src:r=./src --format=text --report-output=./out/survey.txt
2. Determine the number of loop iterations.
advixe-cl --collect=tripcounts --project-dir=./advi --search-dir src:r=./src --./bin/myApplication advixe-cl --report=tripcounts --project-dir=./advi --search-dir src:r=./src --format=text --report-output=./out/tripcounts.txt
3. Check for possible dependencies.
advixe-cl --collect=dependencies --project-dir=./advi --search-dir src:r=./src -- ./bin/myApplication advixe-cl --report=dependencies --project-dir=./advi --search-dir src:r=./src --format=text --report-output=./out/dependencies.txt
4. Check memory access patterns.
advixe-cl --collect=map --project-dir=./advi --search-dir src:r=./sr c -- ./bin/myApplication advixe-cl --report=map --project-dir=./advi --search-dir src:r=./src --format=text --report-output=./out/map.txt
5. Update the application to enable automatic compiler vectorization, or explicitly mark the loops you need to vectorize. Rebuild the application and test.
Add Threading Parallelism
1. Find hotspots. This step is similar to the first step in the SIMD-parallel workflow (above).
2. Determine the number of loop iterations. This step is similar to the second step in the SIMD parallel workflow (above).
3. Add annotations to the application source code and rebuild the application.
4. Collect suitability data. Note: Annotations must be present in the source code for this collection to be successful.
advixe-cl --collect=suitability --project-dir=./advi --search-dir src:r=./src -- ./bin/myApplication advixe-cl --report=suitability --project-dir=./advi --search-dir src:r=./src --format=text --report-output=./out/suitability.txt
5. Check for the possible dependencies.
advixe-cl --collect=dependencies --project-dir=./advi --search-dir src:r=./src -- ./bin/myApplication advixe-cl --report=dependencies --project-dir=./advi --search-dir src:r=./src --format=text --report-output=./out/dependencies.txt
6. Display a list of currently used annotations.
advixe-cl --report=annotations --project-dir=./advi --search-dir src:r=./src --format=text --report-output=./out/annotations.txt
7. Update the application using the chosen parallel coding constructs. Rebuild the application and test.
Use an option file for efficiency. Enter one option on each line. No spaces are allowed in the option entry; use a new line. The option file must be in UTF-8 format.
advixe-cl --report=annotations --option-file=./advi/option.txt
with an
option.txt
file that looks like this:
--project-dir ./advi --search-dir src:r=./src --format=text --report-output ./out/annotations.txt

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