Chapter 26 Floating Point Unit Test (fputest)

fputest checks the floating point unit on machines with SPARC architecture.

fputest Subtests

Instruction tests:

• FSR Register test

• Registers test

• NACK test

• Move Registers test

• Positive to Negative test

• Negative to Positive test

• Absolute test

• Single-Precision Integer to Floating Point test

• Double-Precision Integer to Floating Point test

• Single-Precision Floating Point to Integer test

• Double-Precision Floating Point to Integer test

• Single-Precision Round Toward Zero test

• Double-Precision Round Toward Zero test

• Single to Double-Precision Format Conversion test

• Double to Single-Precision Format Conversion test

• Single and Double-Precision Addition, Subtraction, Multiplication, Square-root, Division, and Compare tests

• Single and Double-Precision Compare and Exception if Unordered tests

• Branching and No Branching on Condition Instructions tests

• Single and Double-Precision Chaining tests

• Weitek Status tests

• Lock test

• Single and Double-Precision Datapath tests

• Timing (load) test

Benchmark tests:

• Linpack test

• Cparanoia test

• Kcsqrt test

• Kcdiv test

• Clorenz test

• Cvector test

fputest Options

To reach the dialog box below, 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 may not include the device appropriate to this test. Refer to the SunVTS User's Guide for more details.

Figure 26-1 fputest Test Parameter Options Dialog Box

fputest Test Modes

fputest Command-Line Syntax

/opt/SUNWvts/bin/fputest standard_arguments

64-bit tests are located in the sparcv9 subdirectory: /opt/SUNWvts/bin/sparcv9/testname. If a test is not present in this directory, then it may only be available as a 32-bit test. For more information refer to "32-Bit and 64-Bit Tests".