International Business Machines Corporation

Sun Microsystems, Inc.

International Business Corporation

Parsotam Trikam Patel

Sharad Mehrotra

Paul Gerard Villarrubia

Robert Paul Masleid

Stephen Thomas Quay

Carol Ivash Gabele

David L Bergeron

Dieter F Wendel

Howard L Kalter

Christopher McCall Durham

Juergen Koehl

Erich Klink

Barbara A Chappell

Asmus Hetzel

David J Widiger

Harish N Kotecha

Alexander K Spencer

Alan C Folta

George Anthony Sai Halasz

Jean-Paul Watson

Phoung Kim Phan

A method of broadcasting N, an even integer, bits of data onto a bus that includes a first plurality of electrical conductors and a second plurality of electrical conductors. The method includes: broadcasting a first portion of data that includes N/2 bits of data onto the first plurality of electrical conductors. Then, after a time period has elapsed that is greater than 0 seconds and less than the time period required to transfer 2 bits of data sequentially on one of the first plurality of electrical conductors, broadcasting a second portion of data that includes N/2 bits of data onto the second plurality of electrical conductors.

Method and apparatus for transmitting data that utilizes delay elements to reduce capacitive coupling

A system that facilitates generating a global routing for a layout of an integrated circuit operates by receiving a netlist to be routed. The system partitions this netlist into global signals, datapath signals, and control signals. Next, the system creates a tiling grid of the integrated circuit and routes connection nets between tiles within this grid. The system then selects an area within the integrated circuit larger than a tile in the first grid. The system creates a second grid of tiles smaller than the tiles of the first grid within this selected area. During this process, connection nets are routed between tiles on the second grid while routings within the first grid are maintained. The system merges connection nets within the first grid with connection nets within the second grid to form the global routing.

Method and apparatus to facilitate global routing for an integrated circuit layout

Provided is a data structure containing wiring information for an integrated circuit and a method for storing the same in a computer readable medium. In one embodiment of the present invention storage of the data structure is achieved by mapping said integrated circuit into memory locations of said medium as a wiring plane having a plurality of data points arranged in a grid. Each of the plurality of data points having state information associated therewith, which includes a plurality of attributes. Subsections of the grid are associated with memory locations in the computer readable medium to form a plurality of tiles. Specifically the tiles and data points are arranged so that a sub-portion of the data points associated with one of the plurality of tiles have common attributes.

Data structure for fine-grid multi-level VLSI routing and method for storing the data structure in a computer readable medium

Disclosed is a method for pre-design estimation of coupling noise and avoidance of coupling noise failures in interconnects. An initial routing of a plurality of nets is estimated utilizing global paths. Then, the worst-case and average-case models for various parameters of each net are evaluated. With these models, a noise analysis is completed by which a determination is made whether coupling noise of any one of the nets is above a threshold level for noise-induced failure (i.e., a noise-failure threshold). When it is determined that the estimated coupling noise of a net falls below the noise-failure threshold, a response mechanism is triggered for later implementation during detailed routing of the nets to prevent the coupling noise from reaching the noise-failure threshold.

Coupled noise estimation and avoidance of noise-failure using global routing information

A method for timing and noise analysis in designing data processing chips is provided. The process begins by wiring all unconnected nets in the design and then using a 2&frac12; D capacitance extraction technique built into a detailed router to extract all of the wired nets. The data from the extracted nets is then process using a timing and analysis tool. Optimization programs are then used to generate fixes for any nets in the design which contribute to timing and noise failures. The present invention gives designers the capability of fast and accurate interconnect extraction within the routing tool. In addition, this technique is incremental. Any wiring changes can be quickly re-extracted, since only local information is required for extraction. This incremental capability allows designers to perform quick iterations of wiring, extraction and timing analysis.

Timing closure and noise avoidance in detailed routing

A method for performing capacitance estimations on an integrated circuit design routed by a global routing tool is disclosed. Routing areas and pin locations of a net within an integrated circuit design are initially obtained from a global routing tool. Common boundaries among the routing areas are then defined. Before the performance of a detailed routing step, congestion information furnished by the global routing tool is utilized to perform probabilistic capacitance calculations for an interconnect that can be routed within the routing areas via the defined common boundaries to connect the pin locations.

Method and system for performing capacitance estimations on an integrated circuit design routed by a global routing tool

A method for performing circuit analysis on an integrated-circuit design having design data available in different forms is disclosed. In accordance with the method and system of the present invention, the integrated-circuit design includes multiple networks, and the different forms of design data may appear within one of the networks. For all of the networks within the integrated-circuit design, different forms of design data are categorized into at least three databases. The first of the at least three databases may contain three-dimensional extraction information, the second of the databases may contain wiring information, and the third of the databases may contain pre-wiring information. For each of the networks, a determination is made as to whether or not three-dimensional extraction information is available. In response to a determination that three-dimensional extraction information is available, performing circuit analysis by utilizing the three-dimensional extraction information.

Method and system for performing circuit analysis on an integrated-circuit design having design data available in different forms

A method and system for performing timing analysis on an integrated circuit design are disclosed. It is always advantageous to be able to conveniently perform a timing analysis on the entire IC design at any stage of the design process in order to gain more accurate timing information about the design. However, at an early stage of the design process, the available physical circuit data are often incomplete, not to mention these preliminary data are usually of a lower quality as far as capability of providing an accurate RC delay and capacitance estimation is concerned. To make the best usage of the preliminary data, the present disclosure describes a method of performing a fleeting timing analysis that can be very useful during an early floor planning stage of the design process when there is no opportunity to buffer or widen any exceptionally long interconnect wires within the IC circuit design. As a result, much faster design turn-around time may be achieved because buffer insertion need not be run for every new pass of the physical circuit design data.

Method and system for performing timing analysis on an integrated circuit design

A structure for reducing cross-talk in VLSI circuits is disclosed. By filling voltage and ground metal lines in free or unused channels of VLSI chips and connecting them efficiently to the regular power image of the chip, the line to line coupling through vertical layers is reduced almost to zero and in-layer line to line coupling is also drastically reduced.

Structure for reducing cross-talk in VLSI circuits and method of making same using filled channels to minimize cross-talk

A method and system for characterizing interconnect data within an integrated circuit in order to facilitate parasitic capacitance estimation is disclosed. An integrated circuit typically includes a substrate layer and several metal layers. In accordance with the method and system of the present invention, an overlapping area of interconnect wires is first identified within the integrated circuit. This overlapping area, which is a polygon, may be formed between the substrate layer and at least one interconnect wire in one of the several metal layers. The overlapping area may also be formed between two interconnect wires, each in a different one of the several metal layers. A netname for the overlapping area is then recorded. Finally, a netname of an interconnect wire in a metal layer that is at the same level of an interconnect wire within the overlapping area and an associated distance from each side of the overlapping area is recorded, for every interconnect wire within the overlapping area. By utilizing these recorded information, the parasitic capacitance of the integrated circuit can be estimated more efficiently.

Method and system for characterizing interconnect data within an

An automated method and system for designing an integrated circuit are disclosed which construct an initial substrate layout of the integrated circuit in response to receipt of a high-level functional description of an integrated circuit. The initial substrate layout, which includes a number of subcircuits electrically connected by a number of interconnects, is constructed based upon estimated timing characteristics of the subcircuits. Next, particular subcircuits are arranged to optimize performance of the substrate layout of the integrated circuit. Performance characteristics of the substrate layout, including timing characteristics of the number of subcircuits and resistive and capacitive characteristics of the number of interconnects, are then determined. In response to a determination of the performance characteristics of the substrate layout, operating power levels of selected subcircuits and resistances of selected interconnects are adjusted to optimize performance of the substrate layout. Thereafter, the step of determining performance characteristics of the substrate layout is repeated. In response to the repeated determination of the performance characteristics of the substrate layout, routing of the number of interconnects electrically connecting the number of subcircuits is finalized. Thus, performance of the substrate layout of the integrated circuit is optimized by iteratively refining the initial substrate layout utilizing performance characteristic data.

Automated method and system for designing an optimized integrated circuit

A method for determining an optimal design for wiring interconnect and driver power for a designed target delay begins at the floor planning stages of the chip design and may be repeated during the design process. The designer initially specifies a maximum width that wires are allowed to use and a target delay value. Then the designer gives values to weights used in the calculation of an optimization function G(d,p,w), where d is the delay, p is the power, and w is wire width. An 'ideal' slope ##EQU1## is calculated, assuming zero resistance. The designer chooses a slope decrease value from the 'ideal' slope value. For each set wire width, the delay (at the proper slope) belonging to that particular wire width is obtained. With these inputs, an optimization program according to the invention is run. This program then calculates values of the function G(d,p,w) for increasing wire pitches, starting with the minimum allowed by the technology. The process continues until (1) the target delay is set by the designer is met, (2) the largest pitch value allowed by the designer is reached, or (3) further calculation will not yield a smaller value for the optimization function.

Interconnect and driver optimization for high performance processors

In accordance with the teachings of this invention a method for positioning components of input/output circuits is provided. These components are positioned on a semiconductor substrate. The semiconductor substrate includes other than input/output circuits. Each input/output circuit is provided for processing a single bit of information within a data word containing multiple bits. The method includes the step of (1) dividing each input/output circuit into groups containing similar functional subcomponents; (2) forming a vertical column of subcomponents for each input/output circuit and connecting these subcomponents to perform the function; (3) placing said columns in adjacency to form a plurality of columns with like subcomponents immediately adjacent to each other forming row groups of said like subcomponents; and (4) forming guard rings around the subcomponent row groups if required. Also in accordance with this invention, a group of input/output circuits is provided. The group is positioned on an integrated circuit substrate that contains other than input/output circuits. The input/output circuits group includes a plurality of columns of circuitry component where each column represents all of the input/output circuit components for processing a single bit of information in a group of bits, a plurality of rows positioned across the columns containing like devices among said input/output circuits, and at least one guard ring containing at least one of the plurality of rows.

Bit stack compatible input/output circuits

In accordance with the teachings of this invention a method for positioning components of input/output circuits is provided. These components are positioned on a semiconductor substrate. The semiconductor substrate includes other than input/output circuits. Each input/output circuit is provided for processing a single bit of information within a data word containing multiple bits. The method includes the step of (1) dividing each input/output circuit into groups containing similar functional subcomponents; (2) forming a vertical column of subcomponents for each input/output circuit and connecting these subcomponents to perform the function; (3) placing said columns in adjacency to form a plurality of columns with like subcomponents immediately adjacent to each other other forming row groups of said like subcomponents; and (4) forming at least one guard ring around the subcomponent row groups. Also in accordance with this invention, a group of input/output circuits is provided. The group is positioned on an integrated circuit substrate that contains other than input/output circuits. The input/output circuits group includes a plurality of columns of circuitry component where each column represents all of the input/output circuit components for processing a single bit of information in a group of bits, a plurality of rows positioned across the columns containing like devices among said input/output circuits, and at least one guard ring containing at least one of the plurality of rows.

Method of making bit stack compatible input/output circuits

A saturation-limited bipolar transistor device or circuit and a method of making same are provided which includes a merged NPN transistor and a PNP transistor structure formed so as to produce denser cells or circuits. A simple process is used to form the structure which includes a double diffused technique for making the PNP transistor. The PNP transistor has a double diffused emitter-base arrangement wherein the emitter is asymmetrically positioned with respect to the base so as to also serve as a contact for the base of the NPN transistor. The PNP transistor limits the input current by bypassing excess current to a silicon semiconductor substrate or chip. The structure includes an N type epitaxial layer formed on an N type subcollector with a P type region provided near the surface of the epitaxial layer. The epitaxial layer serves as the NPN collector and as the PNP base contact region. A first N type region is formed through the P type region extending from the surface of the epitaxial layer to the subcollector dividing the P type region into first and second sections which serve as the PNP collector region and the NPN base region, respectively. A second N type region is formed in the second section of the P type region at the surface of the epitaxial layer acting as the NPN emitter and a P+ region is formed in the first N type region at the surface of the epitaxial layer extending into the second section of the P type region which forms the NPN transistor base. This P+ region serves as the PNP emitter and as the NPN base contact.

Growing community of inventors

Austin, TX, United States of America

Parsotam Trikam Patel