Intel® Inspector User Guide for Linux* OS

User Guide

  • 2021.4
  • 09/27/2021
  • Public Content
Contents

result-dir

Specify the directory where the result is stored.

Syntax

-result-dir
<PATH>
-r
<PATH>
Both short names and long names are case-sensitive. For example,
-R
 is the short name of the
report
 action, and
-r
 is the short name of the
result-dir
 action-option.
Arguments
pathname
A string containing the PATH/name for the
directory where a result is stored. This may be an absolute pathname or a path relative to the current working directory. If the lowest directory in the pathname does not exist, it is created.
By default, the name of the result directory takes this general form: r@@@{at}.
  • The prefix
    r
     indicates that this is a result directory. You may specify a different prefix if desired.
  • The variables
    @@@
     represent an automatic counter that starts from 000 and is incremented by one as each subsequent result is created. There can be only one occurrence of this string in the pathname. When you perform a
    collect
     or
    collect-with
     action, this value is automatically incremented to the next available number. If you use this counter, your first result is numbered 000, the second is 001, and so on. If you perform an action on an existing result without specifying the pathname, the most recent result (which is assumed to be the result with the highest number) is used by default.
  • The suffix
    {at}
     is a code representing the type of
    collect
    or
    collect-with
     action used to generate the result. This code indicates which
    collector,
     analysis type and preset configuration level
    (i.e., knob)
     were selected.
Default
The pathname uses the format r@@@{at} and stores the result in the current working directory.
Description
Use the
result-dir
 action-option to specify the pathname of a result.
  • For the
    collect
     or
    collect-with
     action, use this to specify the directory where you want the result to be created. The result will have the same name as the directory name you specify.
  • For other actions, use this to specify the result you want to use as input.
Using the
@@@
 counter pattern in the directory name argument allows you to collect a result, or perform an action on a the most recent result, without specifying the exact result name.
If you specified a different prefix, such as myRes- (
myRes-@@@{at}
), and then perform a memory error analysis, followed by a threading error analysis, the result directories would be assigned the following names:
myRes-000mi1
 and
myRes-001ti2
.
You cannot put multiple results into the same result directory. If you specify the same result directory for multiple analysis runs, an error is returned. Use the auto-increment counter (@@@) to work around this restriction.
Example
This command:
  • Runs a
    Detect Deadlocks and Data Races (ti2)
     analysis on the application
    myApp
    .
  • Stores the result in the
    myRes000
     result directory in the current working directory the first time
    myRes@@@
     is used. (The next invocation would generate a result directory named
    myRes001
    .)
  • Generates a summary report of detected problems, and writes it to the
    inspxe-cl.txt
     file in the result directory.
  • Generates a suppression file for all detected problems in the
    myRes000
     result (or the highest existing number of similarly named results). 
$ inspxe-cl -collect ti2 -result-dir myRes@@@ -- myApp
This command compares the results of two t1 collections and generates a Summary report to get an overview of regression status. 
$ inspxe-cl -R summary -r myRes000 -r myRes001

Product and Performance Information

1

Performance varies by use, configuration and other factors. Learn more at www.Intel.com/PerformanceIndex.