Abstract: A method for routing of some embodiments defines global routes for nets in an arbitrary region of a circuit layout in which each net has a set of pins. The method uses a first set of lines of measure the length of the global routes, a second set of lines to measure congestion of the global routes, and a third set of lines to partition the arbitrary region into a first set of sub-regions. For each net, the method identifies a global route that connects a group of first-set sub-regions that contain the net's set of pins.

Abstract: Some embodiments of the invention provide a method of pre-computing routes for nets a region of a design layout. These routes are used by a router that uses a set of partitioning lines to partition the region into a plurality of sub-regions. For each particular set of potential sub-regions, the method initially identifies a set of routes that traverse the particular set of potential sub-regions. For each particular route identified for each particular set of sub-regions, the method then determines whether the particular route is stored in a storage structure. If not, the method stores the particular route in the storage structure.

Abstract: An environment is described in which multiple companies can securely collaborate on a design or other project. The environment includes a set of resources residing on a set of one or more utility servers maintained by a first company, an access control mechanism for controlling access to the set of resources, a secure network connection between the set of utility servers and a second company, and a remote controller for remotely viewing, by an authorized individual from the second company, a user interface of an application while an authorized individual from the first company is executing the application on the set of utility servers. The secure network connection includes a secure association mechanism for establishing a secure association between participating parties, a virtual point-to-point network connection for transmitting data between associated parties, and an encryption/decryption mechanism.

Abstract: The present invention introduces a method for implementing a gridless non Manhattan router by modifying an existing gridless Manhattan router. In the method of the present invention, a tile based router that uses tiles to represent circuit geometry or free space between circuit geometry is first selected. Next, at least one tile routing layer of the tile based router is rotated to implement a diagonal wiring layer. The code of the router is then adjusted to ensure that a via that will connect a Manhattan layer to a non Manhattan layer (a diagonal layer) will fit within a tile on both layers.

Abstract: A method for inferring a requested data input function of a sequential cell from a library of candidate cells, wherein the requested cell and the candidate cell are expressed as polynoms and then divided. The method generates polynomial expressions of the inhibition, transformation and inference steps necessary to convert the candidate cell into the requested cell. The use of polynomial expression and division greatly reduces the number of rules necessary to accommodate the varying combinations of requested cell and candidate cell functions.

Abstract: A machine-implemented, simulations-supporting system creates a hierarchy of data structures for simplifying the task of identifying iso-topological, and iso-geometric, and iso-static instances of subcircuit-definitions. The behaviors of such isomorphic and iso-static instances can be simultaneously predicted by appointing a simulation leader for them and using the simulation leader in combination with a respective simulation model to predict the behavior of the simulation leader. The predicted behavior of the leader is then copied for the followers. In one embodiment, state-describing S-circuit cards each point to a respective, and possibly merged, I-circuit card. The I-circuit cards each point to respective, and possibly merged, element instantiating cards (AG-cards) as well as to respective, and possibly merged, interconnect-topology describing cards (T-circuits).

Abstract: Some embodiments provide a method of pre-computing routes for nets in a region of an integrated circuit (“IC”) layout. The method initially defines a set of partitioning lines for partitioning the region into a plurality of sub-regions during a routing operation. For a particular set of potential sub-regions, the method then identifies a set of routes that traverse the particular set of potential sub-regions, where at least one of the identified routes has at least one diagonal edge. The method then stores the identified routes.

Abstract: In an automated integrated circuit design, if the performances of a layout of circuit devices are not within predetermined tolerances of performance specifications, at least one of the circuit devices is resized or repositioned and an updated value of a device parameter for each resized or repositioned circuit device is determined. A difference between the initial and updated value of each device parameter is then determined and each difference is combined with a ratio formed from changes in the value of one of the device parameters and changes in the value of one of the performances affected by the device parameter. The result of this combination is then combined with the initial value of the performance to determine an updated value therefor.

Abstract: A method for modeling a mixed-language and mixed-signal (MLMS) design is disclosed. The method includes receiving an MLMS design comprising at least a digital driver, a digital receiver, and an analog block connected by an MLMS net in a hierarchical structure and identifying analog-digital boundaries of the MLMS design. For each analog-digital boundary, the method further includes a) selecting a connect module (CM) by using a predetermined discipline resolution procedure; b) determining input driving values of the CM; and c) connecting the digital driver, the digital receiver, and the analog block to the CM. The method repeats steps a), b), and c) on all analog-digital boundaries of the MLMS design.

Abstract: When an IC layout is to include time-constrained signal paths, a placement plan defining positions of cells forming the IC is analyzed to estimate lengths of nets needed to interconnect the cells based on the positions of cells included in those signal paths. A capacitance and resistance of each net is then estimated based on its estimated length. The delay through each time-constrained signal path is then estimated based on the estimated capacitance and resistance of each net to be included in the time-constrained signal path and on the terminal impedances, switching speeds and driving strengths of the cells included in the signal path. The estimated path delay for each signal path is then compared to its timing constraint to determine whether that signal path is a “critical signal path” likely to fail to meet its timing constraint following development of a detailed routing plan.

Abstract: A method for modeling a circuit design includes synthesizing the circuit design to create a first gate-level representation of the circuit design. The method also includes analyzing a second gate-level representation of the circuit design to learn architecture information, and resynthesizing the first gate-level representation of the circuit design to incorporate the learned architecture information from the second gate-level representation of the circuit design. A computer-readable storage medium has stored thereon computer instructions that, when executed by a computer, cause the computer to synthesize a circuit design to create a first gate-level representation of the circuit design.

Abstract: A system and method for enabling Intellectual Property (IP) Blocks to be reused at a system level. The present invention represents the IP blocks as blocks that exchange messages without needing to represent the functionality of the IP blocks. The implementations of these IP blocks exchanges messages through complex signaling protocols. In conventional systems, interfacing between IP blocks that use different signaling protocols is a tedious and error prone design task. The present invention uses regular expression based protocol descriptions to show how to map the message onto a signaling protocol. Given two protocols, the present invention builds an interface machine that automatically labels data referenced by all protocols. The present invention is capable of generating the interface even when the data sequencing of the two protocols differs.

Abstract: A method is provided for compiling a model for use in a simulation, the method comprising receiving a description of the model; and automatically converting the description into an implementation of the model that is customized for a selected analysis during simulation.

Abstract: Disclosed is a method, system, and computer program product for performing interblock stitching for electronic designs. According to some approaches, interblock stitching is accomplished by implementing a stitching region between the block and external routing structures. The stitching region is implemented using local preferred direction approaches.

Abstract: A method and system that converges on a global solution to a PCB routing problem using iterations of topology-based routing is described. In some embodiments, the geometric design space is abstracted into a topological graph representing the routing problem. Then, each net is allowed to find its optimal solution path independent of the solution paths for all other nets. The electrical and physical constraints of the system are initially ignored or greatly relaxed. Over each design iteration, the constraints are tightened until a complete, global, topological solution is found. Once a topological solution is found, it is converted into a geometric solution. In the event that no geometric solution exists for that topological solution, then the iteration process is resumed taking into consideration this additional information. The result is the ability to quickly autoroute highly-constrained PCB designs with minimal operator input.

Abstract: To estimate a distribution of voltages or currents in the layers of a multi-layer circuit, an exemplary current flow in each layer is discretized into a number of current vector elements and at least one scalar charge element related to the charge associated with each current vector element. A first distribution of voltages induced in each circuit layer is determined from current vector elements in all of the circuit layers. A second distribution of voltages induced in each circuit layer is determined from the scalar charge elements in all of the circuit layers. For each circuit layer, the first and second distributions of voltages induced therein are combined to determine an actual distribution of voltages in the circuit layer.

Abstract: Integrated proof flow methods and apparatuses are discussed. Integrated proof flow refers to attempting both formal verification and nonformal verification. A coverage metric can be changed by both attempting formal verification and by attempting nonformal verification. Some embodiments of the present invention provide proof flow methods that integrate verification and nonformal verification (e.g., bounded verification, multi-point proof, and/or vector-based simulation) to prove one or more properties in a circuit design.

Abstract: The transmitter circuit architecture is disclosed based on a phase lock loop architecture and which uses a delta-sigma modulator with 2 point modulation. In order to restrict the bandwidth of the PLL, subsidiary analogue modulation is employed, which requires aligning with the delta-sigma modulation. Alignment of the modulation is accomplished by correction of the sensitivity of the PLL voltage controlled oscillator to modulation by correlating residual modulation in the PLL with the modulated signal input. The action of the modulation correlator trims the modulation and the PLL bandwidth without disturbing the normal operation of the transmitter, and allows the use of modulation bandwidths greater than the PLL bandwidth.

Abstract: Some embodiments of the invention provide a method of routing nets in a multi-layer integrated-circuit (“IC”) layout. For each particular net in a set of nets, the method specifies different spacing constraints for routing the particular net in different directions on the same layer. It then defines a particular route for each particular net in the set of nets, where the spacing between at least one particular route and an item adjacent to the route in the layout is different in the different directions on the same layer.

Abstract: Dependencies can be specified between jobs that are constituent to a unit of work, which are automatically determined or identified by processing a work request that defines the work. For example, a second job can be specified as depending on a first job meeting a particular condition. Furthermore, sub-works of the second job are not scheduled for execution until the first job has met the condition, thus allowing the second job to be placed into an active state. First, a work request is received, which specifies a first job that includes a first set of sub-works and a second job that includes a second set of sub-works. The work request is interpreted and processed to determine that the second job has the dependency on the first job. The first job is placed into an active state to enable the first sub-works to be scheduled for execution. The second job is placed in a pending state and it is determined whether the first job has met the condition.