1 |
6 |
jidan |
|
2 |
|
|
Package Overview for SoftFloat Release 2b
|
3 |
|
|
|
4 |
|
|
John R. Hauser
|
5 |
|
|
2002 May 27
|
6 |
|
|
|
7 |
|
|
|
8 |
|
|
----------------------------------------------------------------------------
|
9 |
|
|
Overview
|
10 |
|
|
|
11 |
|
|
SoftFloat is a software implementation of floating-point that conforms to
|
12 |
|
|
the IEC/IEEE Standard for Binary Floating-Point Arithmetic. SoftFloat is
|
13 |
|
|
distributed in the form of C source code. Compiling the SoftFloat sources
|
14 |
|
|
generates two things:
|
15 |
|
|
|
16 |
|
|
-- A SoftFloat object file (typically `softfloat.o') containing the complete
|
17 |
|
|
set of IEC/IEEE floating-point routines.
|
18 |
|
|
|
19 |
|
|
-- A `timesoftfloat' program for evaluating the speed of the SoftFloat
|
20 |
|
|
routines. (The SoftFloat module is linked into this program.)
|
21 |
|
|
|
22 |
|
|
The SoftFloat package is documented in four text files:
|
23 |
|
|
|
24 |
|
|
SoftFloat.txt Documentation for using the SoftFloat functions.
|
25 |
|
|
SoftFloat-source.txt Documentation for compiling SoftFloat.
|
26 |
|
|
SoftFloat-history.txt History of major changes to SoftFloat.
|
27 |
|
|
timesoftfloat.txt Documentation for using `timesoftfloat'.
|
28 |
|
|
|
29 |
|
|
Other files in the package comprise the source code for SoftFloat.
|
30 |
|
|
|
31 |
|
|
Please be aware that some work is involved in porting this software to other
|
32 |
|
|
targets. It is not just a matter of getting `make' to complete without
|
33 |
|
|
error messages. I would have written the code that way if I could, but
|
34 |
|
|
there are fundamental differences between systems that can't be hidden.
|
35 |
|
|
You should not attempt to compile SoftFloat without first reading both
|
36 |
|
|
`SoftFloat.txt' and `SoftFloat-source.txt'.
|
37 |
|
|
|
38 |
|
|
|
39 |
|
|
----------------------------------------------------------------------------
|
40 |
|
|
Legal Notice
|
41 |
|
|
|
42 |
|
|
SoftFloat was written by me, John R. Hauser. This work was made possible in
|
43 |
|
|
part by the International Computer Science Institute, located at Suite 600,
|
44 |
|
|
1947 Center Street, Berkeley, California 94704. Funding was partially
|
45 |
|
|
provided by the National Science Foundation under grant MIP-9311980. The
|
46 |
|
|
original version of this code was written as part of a project to build
|
47 |
|
|
a fixed-point vector processor in collaboration with the University of
|
48 |
|
|
California at Berkeley, overseen by Profs. Nelson Morgan and John Wawrzynek.
|
49 |
|
|
|
50 |
|
|
THIS SOFTWARE IS DISTRIBUTED AS IS, FOR FREE. Although reasonable effort
|
51 |
|
|
has been made to avoid it, THIS SOFTWARE MAY CONTAIN FAULTS THAT WILL AT
|
52 |
|
|
TIMES RESULT IN INCORRECT BEHAVIOR. USE OF THIS SOFTWARE IS RESTRICTED TO
|
53 |
|
|
PERSONS AND ORGANIZATIONS WHO CAN AND WILL TAKE FULL RESPONSIBILITY FOR ALL
|
54 |
|
|
LOSSES, COSTS, OR OTHER PROBLEMS THEY INCUR DUE TO THE SOFTWARE, AND WHO
|
55 |
|
|
FURTHERMORE EFFECTIVELY INDEMNIFY JOHN HAUSER AND THE INTERNATIONAL COMPUTER
|
56 |
|
|
SCIENCE INSTITUTE (possibly via similar legal warning) AGAINST ALL LOSSES,
|
57 |
|
|
COSTS, OR OTHER PROBLEMS INCURRED BY THEIR CUSTOMERS AND CLIENTS DUE TO THE
|
58 |
|
|
SOFTWARE.
|
59 |
|
|
|
60 |
|
|
Derivative works are acceptable, even for commercial purposes, provided
|
61 |
|
|
that the minimal documentation requirements stated in the source code are
|
62 |
|
|
satisfied.
|
63 |
|
|
|
64 |
|
|
|
65 |
|
|
----------------------------------------------------------------------------
|
66 |
|
|
Contact Information
|
67 |
|
|
|
68 |
|
|
At the time of this writing, the most up-to-date information about
|
69 |
|
|
SoftFloat and the latest release can be found at the Web page `http://
|
70 |
|
|
www.cs.berkeley.edu/~jhauser/arithmetic/SoftFloat.html'.
|
71 |
|
|
|
72 |
|
|
|