xref: /linux/Documentation/arch/arm/nwfpe/nwfpe.rst (revision f85f5ae45ad945270a8884261de8249431e8b5a6)
1Introduction
2============
3
4This directory contains the version 0.92 test release of the NetWinder
5Floating Point Emulator.
6
7The majority of the code was written by me, Scott Bambrough It is
8written in C, with a small number of routines in inline assembler
9where required.  It was written quickly, with a goal of implementing a
10working version of all the floating point instructions the compiler
11emits as the first target.  I have attempted to be as optimal as
12possible, but there remains much room for improvement.
13
14I have attempted to make the emulator as portable as possible.  One of
15the problems is with leading underscores on kernel symbols.  Elf
16kernels have no leading underscores, a.out compiled kernels do.  I
17have attempted to use the C_SYMBOL_NAME macro wherever this may be
18important.
19
20Another choice I made was in the file structure.  I have attempted to
21contain all operating system specific code in one module (fpmodule.*).
22All the other files contain emulator specific code.  This should allow
23others to port the emulator to NetBSD for instance relatively easily.
24
25The floating point operations are based on SoftFloat Release 2, by
26John Hauser.  SoftFloat is a software implementation of floating-point
27that conforms to the IEC/IEEE Standard for Binary Floating-point
28Arithmetic.  As many as four formats are supported: single precision,
29double precision, extended double precision, and quadruple precision.
30All operations required by the standard are implemented, except for
31conversions to and from decimal.  We use only the single precision,
32double precision and extended double precision formats.  The port of
33SoftFloat to the ARM was done by Phil Blundell, based on an earlier
34port of SoftFloat version 1 by Neil Carson for NetBSD/arm32.
35
36The file README.FPE contains a description of what has been implemented
37so far in the emulator.  The file TODO contains a information on what
38remains to be done, and other ideas for the emulator.
39
40Bug reports, comments, suggestions should be directed to me at
41<scottb@netwinder.org>.  General reports of "this program doesn't
42work correctly when your emulator is installed" are useful for
43determining that bugs still exist; but are virtually useless when
44attempting to isolate the problem.  Please report them, but don't
45expect quick action.  Bugs still exist.  The problem remains in isolating
46which instruction contains the bug.  Small programs illustrating a specific
47problem are a godsend.
48
49Legal Notices
50-------------
51
52The NetWinder Floating Point Emulator is free software.  Everything Rebel.com
53has written is provided under the GNU GPL.  See the file COPYING for copying
54conditions.  Excluded from the above is the SoftFloat code.  John Hauser's
55legal notice for SoftFloat is included below.
56
57-------------------------------------------------------------------------------
58
59SoftFloat Legal Notice
60
61SoftFloat was written by John R. Hauser.  This work was made possible in
62part by the International Computer Science Institute, located at Suite 600,
631947 Center Street, Berkeley, California 94704.  Funding was partially
64provided by the National Science Foundation under grant MIP-9311980.  The
65original version of this code was written as part of a project to build
66a fixed-point vector processor in collaboration with the University of
67California at Berkeley, overseen by Profs. Nelson Morgan and John Wawrzynek.
68
69THIS SOFTWARE IS DISTRIBUTED AS IS, FOR FREE.  Although reasonable effort
70has been made to avoid it, THIS SOFTWARE MAY CONTAIN FAULTS THAT WILL AT
71TIMES RESULT IN INCORRECT BEHAVIOR.  USE OF THIS SOFTWARE IS RESTRICTED TO
72PERSONS AND ORGANIZATIONS WHO CAN AND WILL TAKE FULL RESPONSIBILITY FOR ANY
73AND ALL LOSSES, COSTS, OR OTHER PROBLEMS ARISING FROM ITS USE.
74-------------------------------------------------------------------------------
75