Browse Prior Art Database

Fast Method for Generating Effective Addresses

IP.com Disclosure Number: IPCOM000106547D
Original Publication Date: 1993-Nov-01
Included in the Prior Art Database: 2005-Mar-21
Document File: 4 page(s) / 84K

Publishing Venue

IBM

Related People

Ishizaki, K: AUTHOR [+2]

Abstract

Disclosed is a fast method for generating effective addresses. On machines with this method, an effective address can be generated faster in the indexed or displacement addressing mode. The method is implemented by replacing specified bits of the base register with a specified offset value in the operand. Unlike the conventional method, The disclosed method does not use an ALU to generate an effective address.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

Fast Method for Generating Effective Addresses

      Disclosed is a fast method for generating effective addresses.
On machines with this method, an effective address can be generated
faster in the indexed or displacement addressing mode.  The method is
implemented by replacing specified bits of the base register with a
specified offset value in the operand.  Unlike the conventional
method, The disclosed method does not use an ALU to generate an
effective address.

      In the conventional method, an operation for generating an
effective address in the indexed or displacement addressing mode
requires an ALU for adding an indexed or displacement offset to the
base address.

      In this new method, a mask bit is newly placed in the operation
for generating an effective address in the indexed or displacement
addressing mode eliminates a mask bit.  The mask bit makes an ALU for
generating an effective address unnecessary in the indexed or
displacement addressing mode.  The mask bit specifies a position and
a number of bits that are replaced in the base register by another
register or value indicated in the operand, when an effective address
is generated.  Assume that M is the address register's bit width and
that L is the bit width specified by the mask bit to be replaced in
the operand.  The generated effective address is the combination of
the upper M - L bits of the base register and the lower L bits of the
offset value in the operand.

      Assume that N is the maximum bit width specified by the mask
bit in the operand.  A compiler or programmer must align the base
address of an object to read and write in units of 2^N words.

      A load operand specifies the destination register number of the
effective address, the base...