摘要:
An architecture and distributed hierarchical interconnect scheme for field programmable gate arrays (FPGAs). The FPGA is comprised of a number of cells which perform logical functions on input signals. Programmable intraconnections provide connectability between each output of a cell belonging to a logical cluster to at least one input of each of the other cells belonging to that logical cluster. A set of programmable block connectors are used to provide connectability between logical clusters of cells and accessibility to the hierarchical routing network. A uniformly distributed first layer of routing network lines is used to provide connections amongst sets of block connectors. A uniformly distributed second layer of routing network lines is implemented to provide connectability between different first layers of routing network lines. Switching networks are used to provide connectability between the block connectors and routing network lines corresponding to the first layer. Other switching networks provide connectability between the routing network lines corresponding to the first layer to routing network lines corresponding to the second layer. Additional uniformly distributed layers of routing network lines are implemented to provide connectability between different prior layers of routing network lines. An additional routing layer is added when the number of cells is increased as a square function of two of the prior cell count in the array while the length of the routing lines and the number of routing lines increases as a linear function of two. Programmable bi-directional passgates are used as switches to control which of the routing network lines are to be connected.
摘要:
A programmable logic array integrated circuit has a plurality of programmable logic elements grouped into a plurality of mutually exclusive groups. Each group includes signal conductors uniquely associated with that group for conveying signals between the programmable logic elements in that group. Other signal conductors are provided for conveying signals between the groups. Multiplexers can be used in various ways to reduce the number of programmable interconnections required between signal conductors.
摘要:
A method of programming an FPGA to enable encoding of configuration logic block (CLB) outputs enabling an efficient use of FPGA routing resources. The method of the present invention utilizes the one hot approach, wherein only one CLB output is high at a time, to form a state machine using an FPGA. To provide encoding, programmable interconnect points (PIPs) may be programmed to connect CLB outputs to interconnect lines so that the interconnect lines indicate states of the CLB outputs in an encoded form. To provide such encoding, less interconnect lines than CLB outputs provide the encoded form of the CLB outputs. Thus, PIPs can connect a single interconnect line to more than one CLB output. Further, PIPs can connect a single CLB output to interconnect lines provided in separate parallel routing paths. To prevent erroneous results, CLB outputs which are not hot are tri-stated. Output decoding can be provided by an additional decoder in the FPGA connected to the interconnect lines providing the encoded form of the CLB outputs. Output decoding may alternatively be provided using a CLB.
摘要:
The system and method of the present invention provides an innovative bus system of lines (705) which can be programmed and to provide data, control and address information to the logic circuits (701-702) interconnected by the bus system. This flexible structure and process enables a configurable system to be created to programmably connect one or more logic circuits, such as megacells (701-702). The programmability of the bus system enables the cascading of multiple megacells in an arbitrary fashion (i.e., wide, deep or both) and the sharing of common lines for system level communication.
摘要:
The invention relates to an integrated circuit that can be configured to operate as a content addressable memory. The integrated circuit includes a first functional block that stores at least one keyword dataword which is associated with a group of associated data words. The integrated circuit also includes a second functional block that stores the group of associated datawords. The second functional block is connected to the first functional block in such a way that if a request dataword received at the first functional block matches at least one keyword dataword stored therein, then an associated result dataword included in the group of associated data words stored in the second functional block is output by the second functional block. Typically, the integrated circuit chip is a complex programmable logic device architecture (CPLD).
摘要:
An enhanced performance field programmable gate array integrated circuit comprises a field programmable gate array and other functional circuitry such as a mask-programmable gate array in the same integrated circuit. A circuit interface provides communication between the field programmable gate array, the mask-programmable gate array and the integrated circuit I/O.
摘要:
An apparatus and method for reprogramming a reconfigurable-logic array is provided whereby a portion of the array (15) can be reconfigured without disrupting the operation of the entire array. Avoiding total disruption of array operation typically requires that the configuration controls signals, which determine the configuration of the array, remain substantially non-disrupted during a reprogramming operation. In one embodiment, the reprogramming operates by unique decoding in which an electrical path (12) is established only between the particular storage elements being reprogrammed, thereby avoiding disruption of the configuration control signals provided by other storage elements. In other embodiments, buffers and/or read-modify-write techniques are used to minimize disruption of configuration control signals of storage elements not being reprogrammed.