Browse Prior Art Database

Initialization of Latches from Primary Inputs

IP.com Disclosure Number: IPCOM000120709D
Original Publication Date: 1991-May-01
Included in the Prior Art Database: 2005-Apr-02
Document File: 1 page(s) / 44K

Publishing Venue

IBM

Related People

McNeil, WL: AUTHOR

Abstract

The storage elements (latches) in a VLSI chip often need to be initialized to a known value during a reset sequence. In order to conserve circuitry, the scan chains in an LSSD design are often used during the reset sequence to initialize the latches. The designer must chose one static initialization value for each latch (one or zero), since the scan chains cannot be broken between latches in order to satisfy LSSD design rules.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 91% of the total text.

Initialization of Latches from Primary Inputs

      The storage elements (latches) in a VLSI chip often need to be
initialized to a known value during a reset sequence.  In order to
conserve circuitry, the scan chains in an LSSD design are often used
during the reset sequence to initialize the latches.  The designer
must chose one static initialization value for each latch (one or
zero), since the scan chains cannot be broken between latches in
order to satisfy LSSD design rules.

      It is often desirable to initialize a latch based on primary
input(s) to allow different modes of operation for the chip.  In this
case a dynamic initialization value is necessary.  The solution shown
in Figure 1 below provides a way to use the scan chains for
initialization but also allows a dynamic Initialization Value.

      The Initialization Value may be any combinatorial function of
primary input(s).  During the reset sequence the latches are
"flushed" to known states (A&B clocks are on, C clock is off).  The
Initialization Latch is "flushed" to a zero state, that is the -L2
output=one.  When the reset sequence is over, the A clock turns off
and the C and B clocks are running.  The first C clock pulse will
force the Initialization Value to be placed in the Functional Latch.
The following B clock pulse will clock a zero into the -L2 of the
Initialization Latch, which will allow normal operation of the
Functional latch from that point forward.

      Note that on...