Intel 80287, 80286 manual Base Architecture

80286 BASE ARCHITECTURE

Separate multiplication and division operations are provided for both signed and unsigned binary numbers. The same addition and subtraction instructions are used with signed or unsigned binary values. Conditional jump instructions, as well as an "interrupt on overflow" instruction, can be used following an unsigned operation on an integer to detect overflow into the sign bit.

Unpacked decimal numbers are stored as unsigned byte quantities. One digit is stored in each byte. The magnitude of the number is determined from the low-order half-byte; hexadecimal values 0-9 are valid and are interpreted as decimal numbers. The high-order half-byte must be zero for multiplication and division; it may contain any value for addition and subtraction.

Arithmetic on unpacked decimal numbers is performed in two steps. The unsigned binary addition, subtraction and multiplication operations are used to produce an intermediate result. An adjustment instruction then changes the value to a final correct unpacked decimal number. Division is performed similarly, except that the adjustment is carried out on the two digit numerator operand in register AX first, followed by an unsigned binary division instruction that produces a correct result.

Unpacked decimal numbers are similar to the ASCII character representations of the digits 0-9. Note, however, that the high-order half-byte of an ASCII numeral is always 3. Unpacked decimal arithmetic may be performed on ASCII numeric characters under the following conditions:

•the high-order half-byte of an ASCII numeral must be set to OR prior to multiplication or division.

unpacked decimal arithmetic leaves the high-order half-byte set to OR; it must be set to 3 to produce a valid ASCII numeral.

Packed decimal numbers are stored as unsigned byte quantities. The byte is treated as having one decimal digit in each half-byte (nibble); the digit in the high-order half-byte is the most significant. Values 0-9 are valid in each half-byte, and the range of a packed decimal number is 0-99. Additions and subtractions are performed in two steps. First, an addition or subtraction instruction is used to produce an intermediate result. Then, an adjustment operation is performed which changes the inter- mediate value to a final correct packed decimal result. Multiplication and division adjustments are only available for unpacked decimal numbers.

Pointers and addresses are described below in section 2.3.3, "Index, Pointer, and Base Registers," and in section 3.8, "Address Manipulation Instructions."

Strings are contiguous bytes or words from 1 to 64K bytes in length. They generaiiy coniain ASCII Of other character data representations. The 80286 provides string manipulation instructions to move, examine, or modify a string (see section 3.7, "Character Translation and String Instructions").

If the 80287 numeric processor extension (NPX) is present in the system - see the 80287 NPX book-the 80286 architecture also supports floating point numbers, 32- and 64-bit integers, and 18-digit BCD data types.

The 80287 Numeric Data Processor supports and stores real numbers in a three-field binary format as required by IEEE standard 754 for floating point numerics (see figure 2-3). The number's significant digits are held in the significand field, the exponent field locates the binary point within the significant digits (and therefore determines the number's magnitude), and the sign field indicates whether the number is positive or negative. (The exponent and significand are analogous to the terms "character- istic" and "mantissa," typically used to describe floating point numbers on some computers.) This format is used by the 80287 with various length significands and exponents to support single precision, double precision and extended (80-bit) precision floating point data types. Negative numbers differ from positive numbers only in their sign bits.

2-6