24
TS68882
2119A
–
12/01
The memory format on the TS 68882 consists of 96 bits (three long words). Only 80 bits are
actually used, the other 16 bits are for future expandability and for long-word alignment of
floating-point data structures. Extended format has a 15-bit exponent, a 64-bit mantissa, and a
1-bit mantissa sign.
Extended precision numbers are intended for use as temporary variables, intermediate val-
ues, or in places where extra precision is needed. For example, a compiler might select
extended precision arithmetic for evaluation of the right side of an equation with mixed sized
data and then convert the answer to the data type on the left side of the equation. It is antici-
pated that extended precision data will not be stored in large arrays, due to the amount of
memory required by each number.
Packed Decimal
String Real Data
Format
The packed decimal string data format allows packed BCD strings to be input to and output
from the TS 68882. The strings consist of a 3-digit base 10 exponent and a 17-digit base 10
mantissa. Both the exponent and mantissa have a separate sign bit. All digits are packed
BCD, such that an entire string fits in 96 bits (three long words). As is the case with all data for-
mats, when packed BCD strings are input to the TS 68882, the strings are automatically
converted to extended precision real values. This allows packed BCD numbers to be used as
inputs to any operation. For example:
FADD.P # - 6.023E + 24, FP5
BCD numbers can be output from the TS 68882 in a format readily used for printing by a pro-
gram generated by a high-level language compiler. For example:
FMOVE.P FP3.BUFFER (# -5)
instructs the TS 68882 to convert the floating-point data register 3 (FP3) contents into a
packed BCD string with five digits to the right of the decimal point (FORTRAN F format).
Data Format
Summary
All data formats described above are supported orthogonally by all arithmetic and transcen-
dental operations, and by all appropriate TS 68000 Family addressing modes. For example,
all of the following are legal instructions:
On-chip calculations are performed to extended precision format, and the eight floating-point
data registers always contain extended precision values. All data used in an operation is con-
verted to extended precision by the TS 68882 before the specific operation is performed, and
all results are in extended precision. This ensures accuracy without sacrificing performance.
Refer to Figure 17 for a summary of the memory formats for the seven data formats supported
by the TS 68882.
FADD.B
# 3.FP0
FADD.W
D2.FP3
FADD.L
BIGINT.FP7
FADD.S
# 3.14159.FP5
FADD.D
(SP) + .FP6
FADD.X
[(TEMP -PTR.A7)].FP3
FADD.P
# 1.23E25.FP0