OpenCores
URL https://opencores.org/ocsvn/openrisc/openrisc/trunk

Subversion Repositories openrisc

[/] [openrisc/] [trunk/] [orpsocv2/] [sw/] [apps/] [testfloat/] [README] - Diff between revs 349 and 393

Go to most recent revision | Show entire file | Details | Blame | View Log

Rev 349 Rev 393
?rev1line?
?rev2line?
 
        TestFloat software tests for floating point operations
 
 
 
From http://www.jhauser.us/arithmetic/TestFloat.html
 
 
 
     TestFloat is a program for testing whether a computer's floating-point
 
     conforms to the IEC/IEEE Standard for Binary Floating-point Arithmetic.
 
     TestFloat works by comparing the behavior of the machine's floating-point
 
     with that of the SoftFloat software implementation of floating-point.
 
 
 
At present only the single precision floating point capability of the software
 
is used to test the OpenRISC's FPU.
 
 
 
The main loop of the applications in TestFloat have been modified so they do
 
not rely on input when executed, rather are hard-coded to run all the tests
 
they possibly can.
 
 
 
All of the output functions (fputs, putc, fprintf) have been replaced with
 
versions available in the OR1K support C library.
 
 
 
The software included here also includes the SoftFloat library, which is
 
is required to test the hardware FPU.
 
 
 
See http://www.jhauser.us/arithmetic/SoftFloat.html for more information on
 
SoftFloat.
 
 
 
The or1ksim simulator actually implements its single precision floating point
 
using functions from SoftFloat (see the softfloat/ path in the or1ksim project)
 
so potentially any non-standard behavior of the floating point unit in or1ksim
 
will not be picked up by this test (as TestFloat uses SoftFloat itself to check
 
against the hardware/system results.) However, TestFloat comes with a different
 
implementation of the IEEE 754 compliant conversion and arithmetic functions to
 
test SoftFloat against, hopefully picking up any inconsistencies that exist.
 
 
 
Note that the bits64 version of SoftFloat's library is included here. It is
 
not a problem for all of the current OR1K GCC implementations to handle 64-bit
 
data types.
 
 
 
The file layout of SoftFloat and TestFloat, in their distributions, has been
 
discarded and all the required files have been included in this path.
 
 
 
testfloat is run during the standard check that can be run after compilation.
 
By default only the testfloat program will be run (not testsoftfloat).
 
testsoftfloat will, actually, be compiled and be under
 
testsuite/test-code-or1k/testfloat and can be run from where if desired.
 
 
 
 
 
Julius Baxter, August 3 2010
 No newline at end of file
 No newline at end of file

powered by: WebSVN 2.1.0

© copyright 1999-2024 OpenCores.org, equivalent to Oliscience, all rights reserved. OpenCores®, registered trademark.