Abstract: The present invention provides a method of verification of a RRAM tiling netlist. The method may include steps as follows. Properties “memory_number”, “clock_number” and “netlist_part” of all nets and cells of a RRAM tiling netlist are set to a value 0. A boolean value 0 is assigned to all ground nets of the RRAM tiling netlist, and a boolean value 1 is assigned to all power nets of the RRAM tiling netlist. The RRAM tiling netlist is verified for each customer memory Memk, k=1, 2, . . . , N.

Abstract: A routing multiplexer system provide p outputs based on a selected permutation of p inputs. Each of a plurality of modules has two inputs, two outputs and a control input and is arranged to supply signals at the two inputs to the two outputs in a direct or transposed order based on a value of a bit at the control input. A first p/2 group of the modules are coupled to the n inputs and a second p/2 group of the modules provide the n outputs. A plurality of control bit tables each contains a plurality of bits in an arrangement based on a respective permutation. The memory is responsive to a selected permutation to supply bits to the respective modules based on respective bit values of a respective control bit table, thereby establishing a selected and programmable permutation of the inputs to the outputs.

Abstract: A system for, and method of, allowing conventional memory test circuitry to test parallel memory arrays and an integrated circuit incorporating the system or the method. In one embodiment, the system includes: (1) bit pattern distribution circuitry that causes a probe bit pattern generated by the memory test circuitry to be written to each of the memory arrays, (2) a pseudo-memory, coupled to the bit pattern distribution circuitry, that receives a portion of the probe bit pattern and (3) combinatorial logic, coupled to the pseudo-memory, that employs the portion and data-out bit patterns read from the memory arrays to generate a response bit pattern that matches the probe bit pattern only if all of the data-out bit patterns match the probe bit pattern.

Abstract: The present invention is a method and system for outputting a sequence of commands and data described by a flowchart. The method includes steps as follows. A flowchart describing a sequence of commands and data is received. The flowchart includes a plurality of flowchart symbols. Each of the plurality of flowchart symbols is assigned a ROM (read only memory) record. Assigned ROM records are stored in a ROM. A processor is generated to include the ROM, wherein the processor receives as input a CLOCK signal, a RESET signal, an ENABLE signal and N binary inputs x1, x2, . . . xN, and outputs the sequence of commands and data.

Abstract: A memory timing model is provided, which includes an address input, a multiple-bit data input, a multiple-bit data output, a capacity C and a width N. N one-bit wide memory modules are instantiated in parallel with one another between respective bits of the data input and the data output. Each memory module has a capacity of C bits addressed by the address input.

Abstract: A method and apparatus are provided for creating and using a memory timing database. A plurality of characterization memories are defined, which can be mapped to a memory resource. Each characterization memory has different memory parameters. A plurality of variants of tiling each characterization memory to the memory resource are also defined. Timing characteristics of each tiling variant of each characterization memory are stored in the memory timing database for the memory resource based on sets of input ramptimes and output loads.

Abstract: A search engine architecture substitutes short indices for large data widths, thereby reducing widths required for input to and output from the search engine. The search engine system comprises a search engine responsive to an input address to access an index in the search engine. The index has a width no greater than logarithm on base 2 of the search engine capacity, thereby permitting the search engine to be embodied in an IC chip of reduced area. A driver responds to input commands and to the search engine status to manage indices in the search engine and enable the memory to access its addressable locations based on indices in the search engine.

Abstract: A method and apparatus are provided for receiving a list of design memories, wherein each type of design memory in the list has a name and at least one instance. A pre-placement model is associated with each named memory type in the list. The design memories in the list are mapped to an integrated circuit layout pattern, wherein at least one memory type comprises first and second instances that are mapped differently from one another. After mapping, at least one of the first and second instances is renamed to have a different name than the other. A post-placement model is then associated with each named memory type in the list, including a separate model for each renamed design memory.

Abstract: A process and apparatus are provided for tiling objects, such as design memories, in one or more respective object locations in a layout pattern. For each object, the following steps are performed recursively based on a comparison of at least one of a capacity and a width of the object and that of the respective object location: (1) do nothing; (2) reconfigure the object to have a different capacity and/or width; and (3) split the object into two or more separate objects. The recursion is repeated for each reconfigured object and each separated object.

Abstract: The present invention provides a method and BIST architecture for fast memory testing in a platform-based integrated circuit. The method may include steps as follows. An Mem-BIST controller transmitter is started to generate input signals for a memory in a platform using a deterministic and unconditional test algorithm. The input signals are delayed by a first group of pipelines by n clock cycles. The delayed input signals are received by the memory and an output signal is generated by the memory. The output signal is delayed by a second pipeline by m clock cycles. An Mem-BIST controller receiver is started to receive the delayed output signal for comparison.

Abstract: A method of configuring a random access memory matrix containing partially configured memories in the matrix. The method includes the steps of independently calculating a memory enable signal and a configuration signal for a partially configured memory in each memory tile of the memory matrix. Memory tiles not supported by a memory compiler are determined. A memory wrapper is provided for each tile not supported by the memory compiler. An address controller is inserted in the memory matrix for each tile in a group of tiles. Output signals from each memory location in a memory group having a common group index are combined into a single output signal. A first stripe of memory tiles containing non-configured memory having a first width is selected. A second strip of memory tiles containing configured memory having a second width is also selected.

Abstract: The present invention provides a method and system for analyzing the quality of an OPC mask. The method includes receiving a target layer from a target design, receiving an OPC mask layer from the OPC mask. The method also includes classifying each cell of at least one of the target layer and the OPC mask layer as either repeating or non-repeating, and for each repeating cell, recognizing geometric points in the target layer to determine quality measuring groups. The method also includes simulating the OPC mask layer based on the quality measuring groups, measuring edge placement errors (EPEs) based on at least one of the geometric points, and providing an EPE layer representing EPEs greater than an EPE threshold.

Abstract: A method of generating a timing model for a customer memory configuration, by generating a plurality of template memory netlists for a given RRAM design. Timing models for the template memory netlists are produced and stored in a first database. The template memory netlists are stored in a second database. A netlist for the customer memory configuration is generated and compared to the template memory netlists to find a match. When a match is found, one of the timing models that is associated with the matching template memory netlist is used as the timing model for the customer memory configuration. When a match is not found, two of the template memory netlists that bound the customer netlist are found, according to at least one parameter, and the timing model for the customer memory configuration is interpolated based on the two bounding template memory netlists.

Abstract: A method of analyzing multimode delay in an integrated circuit design to produce a timing model for the integrated circuit design, by inputting a net list, IO arc delays, interconnection arc delays, and constant nets with assigned Boolean functions for the integrated circuit design, propagating the constant nets and assigning Boolean conditions to the IO arc delays and the interconnection arc delays, evaluating timing path delays and conditions for the integrated circuit design, creating the integrated circuit design timing model parameters, and outputting the integrated circuit design timing model. The method is especially desirable for netlists with very complicated mixing logics that include muxing of clocks. In particular, RRAMs are such netlists.

Abstract: The present invention is directed to a FIFO memory with single port memory modules that may allow simultaneous read and write operations. In an exemplary aspect of the present invention, a method for employing a FIFO memory with single port memory modules of half capacity to perform simultaneous read and write operations includes the following steps: (a) providing a first single port memory module for an even address of a read or write operation; (b) providing a second single port memory module for an odd address of a read or write operation; (c) alternating even address and odd address; and (d) when both a read request and a write request reach either the first single port memory module or the second single port memory module at a clock cycle, fulfilling the read request at the current clock cycle and fulfilling the write request at the next clock cycle.

Abstract: A user-defined memory design is mapped to memories of a base platform for an IC that contains a plurality of memory sets, each containing a plurality of memories of a predetermined type. An optimal memory set is selected from the plurality of memory sets for the design by selecting a preference rate for each memory set from the plurality of sets based on the design and its connections to portions of the IC, and selectively assigning the design to one of the memory sets based on the preference rate. The design is optimally mapped to a plurality of memories of the selected memory set by defining an index of the position of each customer memory in the selected memory set. The customer memories in the selected memory set are arranged in an order, and successive numbers of memories of the selected memory set are assigned to each customer memory in order.

Abstract: A memory generation and placement flow system that receives a customer memory design and places the customer memory design within a customizable standardized integrated circuit design. The memory generation and placement flow system includes a memory librarian tool, a memory estimator tool, and a memory placer tool.

Abstract: A decoder for access data stored in n memories comprises a function matrix containing addresses of the memory locations at unique coordinates. A decomposer sorts addresses from coordinate locations of first and second m×n matrices, such that each row contains no more than one address from the same memory. Positional apparatus stores entries in third and fourth m×n matrices identifying coordinates of addresses in the function matrix such that each entry in the third matrix is at coordinates that matches corresponding coordinates in the first matrix, and each entry in the fourth matrix is at coordinates that matches corresponding coordinates in the second matrix. The decoder is responsive to entries in the matrices for accessing data in parallel from the memories.

Abstract: Objects are assigned to points in a rectangle by dividing the rectangle is divided into a plurality of smaller rectangles and applying an object assignment procedure, such as Kuhn's algorithm, to initially assigned objects in each second rectangle. The initial assignment is performed by calculating a maximal cost of assignment of objects to points, and selecting an assignment of objects having a minimum value of maximal cost, identified by iteratively recalculating the maximal matching assignment based on a midpoint of between the minimum and maximum costs.

Abstract: A decoder for access data stored in n memories comprises a function matrix containing addresses of the memory locations at unique coordinates. A decomposer sorts addresses from coordinate locations of first and second m×n matrices, such that each row contains no more than one address from the same memory. Positional apparatus stores entries in third and fourth m×n matrices identifying coordinates of addresses in the function matrix such that each entry in the third matrix is at coordinates that matches corresponding coordinates in the first matrix, and each entry in the fourth matrix is at coordinates that matches corresponding coordinates in the second matrix. The decoder is responsive to entries in the matrices for accessing data in parallel from the memories.