C H A P T E R 9 |
Floating Point Unit Test (fputest) |
The fputest checks the functionality of the floating point unit in CPUs. The test verifies the functionality by various arithmetic operations. In addition, the fputest stresses the CPU with the use of benchmarks. Both single and double precision numbers are used for the operations.
When fputest is chosen in Exclusive test mode from the SunVTS graphical user interface, it may run multiple instances in parallel on different CPUs. The number of such instances that may be running in parallel at the same time is dynamically determined depending on system resources.
To reach the following dialog box, right-click on the test name in the System Map and select Test Parameter Options. If you do not see this test in the System Map, you might need to expand the collapsed groups, or your system might not include the device appropriate to this test. Refer to the SunVTS User's Guide for more details.
FIGURE 9-1 fputestTest Parameter Options Dialog Box
Note - It is not advisable to use the Processor Affinity option for this test. Doing so reduces the effectiveness of the test. |
Performs all the instruction tests and all the benchmark tests. |
|
/opt/SUNWvts/bin/fputest standard-arguments -o dev=cpu-unitN
Note - Options -s, -p, -i, and -w are not applicable when tests are invoked from the command line. |
Specifies the CPU unit to be tested. N specifies the numeric ID of online CPU. This is not a required option. |
Note - When using fputest specific arguments, the -o and dev options are required. |
To execute exclusive fputest with verbose messages against CPU 72, use the following syntax:
To execute functional fputest with verbose messages against CPU 3, use the following syntax:
Copyright © 2007, Sun Microsystems, Inc. All Rights Reserved.