Browse Prior Art Database

Computer Aided Packaging and Partitioning System

IP.com Disclosure Number: IPCOM000079618D
Original Publication Date: 1973-Aug-01
Included in the Prior Art Database: 2005-Feb-26
Document File: 4 page(s) / 78K

Publishing Venue

IBM

Related People

Giannuzzi, RJ: AUTHOR [+3]

Abstract

This system of programs provides a logic designer with a tool which enables the efficient partitioning of digital logic into packaging units, such as large-scale integrated circuits, modules, etc. This is accomplished by interacting through a display terminal with on-line data files, containing a block graph representation of his logic design.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 45% of the total text.

Page 1 of 4

Computer Aided Packaging and Partitioning System

This system of programs provides a logic designer with a tool which enables the efficient partitioning of digital logic into packaging units, such as large-scale integrated circuits, modules, etc. This is accomplished by interacting through a display terminal with on-line data files, containing a block graph representation of his logic design.

Fig. 1 displays the major components of the system. A description of the logic is inputted to the system Block File by the LISTB Routine.

Fig. 2 represents the structure of major files of the system. The Block File contains a representation of the graph of the logic design to be partitioned. Each Block in the logic design is identified by its Block Address and contains the information shown in Fig. 3. Note that the parameters such as number of Inputs, Number of Outputs and Number of Sinks, can be used to calculate the location of any object in the record given that the objects have known fixed lengths. Note that the use of Block Address pointers permits ready addressing of all blocks connected to the given block represented in the Block File record.

The Correlation File associates user defined Block Names with Block Address pointers. It is ordered alphabetically on Block Names and can be searched rapidly, to permit quick translation and location of a given Block File record given its Block Name. Reverse translation is effected by the existence of the Block Name in the Block File record.

The Partition File is organized analogously to the Block File. Each record represents the current state of a given partition of the system. The major difference is that a Constituent Block List is incorporated, which identifies as well as provides a pointer to blocks contained in the partition.

WATE Routine permits setting of the weight values in the Block File records. A given set of weights may be set into any given Block File record or the Block File records of all blocks having a user associated block function.

PREP1 is the initiation link of the system and initializes the system.

PREP2 is the central communication link of the system and contains the key routines MERGE, REMOV, and PRTIR. MERGE provides the facility for recursively defining a partition by combining a block and a partition. A more detailed description appears in the "A Procedure for Merging Either a Logic Block or a Partition to a Partition" appearing in the IBM Technical Disclosure Bulletin, Volume 15, No. 12, pages 3626 and 3627, May 1973.

REMOV complements MERGE by providing facility to remove a specified logic block from a partition. More detail is furnished in "An Algorithm for Removal of a Logic Block From a Partition" appearing in the IBM Technical Disclosure Bulletin. Volume 15, No. 12, pages 3624 and 3625, May 1973. The PRTIR routine prints out the contents of any specified Block File or Partition File record with format control. Other subroutines included in PREP2 and their

1

Page 2...