Intel® Advisor Help
See basic workflows for using the CLI version of Advisor
Refer to Workflows
-workflow |
1) Survey the application to determine hotspots. Typically an optimized (non-debug) version of the application is used when surveying an application.
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
1.1) Find trip counts to 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
2.1) Check dependencies for the possible dependencies. Note that annotations mu st be present in the source code for this collection to be successful.
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
2.2) Check memory access patters.
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
3) Update the application to enable the compiler vectorize automatically, or explicitly mark the loops you need to vectorize. Rebuild the application and test.
1) Survey the application to determine hotspots. This step is similar to the fi rst step in the SIMD-parallel workflow (above).
1.1) Find trip counts.
2) Add annotations to the application source code, and rebuild the application. For detailed examples refer to the Getting Started Tutorial.
3) Collect suitability data. Note that annotations must be present in the source code for this collection to be successful. Typically an optimized (non-debug) version of the application is used when collecting suitability data.
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
3) Collect suitability data. Note that annotations must be present in the source code for this collection to be successful. Typically an optimized (non-debug) version of the application is used when collecting suitability data.
4) Check dependencies for the possible dependencies. Note that annotations must be present in the source code for this collection to be successful.
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
5) Display a list of annotations present.
advixe-cl -report=annotations -project-dir ./advi -search-dir src:r=./src -format=text -report-output ./out/annotations.txt
6) Update the application using the chosen parallel coding constructs. Rebuild the application and test.
An option file can be used to reduce typing of options. One option should be entered 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.txtwith an option file that looks like this:
-project-dir ./advi -search-dir src:r=./src -format=text -report-output ./out/annotations.txtFor information on how to analyze an MPI workload, refer to the Intel® Advisor Documentation->Intel Advisor Help->Command Line Interface.