CL-PS7500FE
System-on-a-Chip for Internet Appliance
19.2.2 Monadic Operations
<ABS|URD|NRM|MVF|MNF|SQT|RND>{cond}<S|D|E>{P|M|Z} Fd, Fn, <Fm|#value>
19.2.3 Library Calls
It is recommended that the following floating-point operators are implemented with calls to an appropriate
library (for example, the C library):
q Power
q Reverse power
q Polar angle
q Logarithm base 10
q Logarithm base 0
q Exponent
q Sine
q Cosine
q Tangent
q Arc sine
q Arc cosine
q Arc tangent
19.2.4 Backwards Compatibility
For backwards compatibility with existing floating-point code, the following floating-point mnemonics are
defined in the ARM floating-point instruction set. These opcodes are treated by the FPA as undefined
instructions, and must be handled by support code, less efficient than using library calls.
<POW|RPW|POL>
<LOG|LGN|EXP|SIN|COS|TAN|ASN|ACS|ATN>
abcdj
00000
00010
00100
00110
01000
01010
01100
01110
10000
10010
Mnemonic
ADF
MUF
SUF
RSF
DVF
RDF
POW
RPW
RMF
FML
Description
Add
Multiply
Subtract
Reverse Subtract
Divide
Reverse Divide
Power
Reverse Power
Remainder
Fast Multiply
Operation
Fd := Fn + Fm
Fd := Fn × Fm
Fd := Fn − Fm
Fd := Fm − Fn
Fd := Fn ÷ Fm
Fd := Fm ÷ Fn
Fd := Fn raised to the power of Fm
Fd := Fm raised to the power of Fn
Fd := IEEE remainder of Fn / Fm
Fd := Fn × Fm
Note
1
1
June 1997
ADVANCE DATA BOOK v2.0
FLOATING-POINT INSTRUCTION SET
175