Leland Stanford Junior University

Other

Cswitch Corporation

Ivo J Dobbelaere

Abbas A El Gamal

Dana How

Malcolm J Wing

Godfrey P D'Souza

Jay B Patel

Robert K Yu

Colin Neal Murphy

Pohrong Rita Chu

Douglas A Laird

Suresh Subramaniam

Mukunda Krishnappa

Jason Golbus

Dave Trossen

Kevin James

A system including a plurality of programmable logic blocks, a plurality of special-purpose blocks, and a configurable high-speed mesh interconnect fabric operatively connecting the plurality of programmable logic blocks and the plurality of special-purpose blocks, where the configurable high-speed mesh interconnect fabric is configured to implement a plurality of interconnect pipeline buses spanning across the system.

Heterogeneous configurable integrated circuit

A circuit style, which may be employed in fast, area-efficient, flexible programmable interconnect architectures, or in logic circuits, is disclosed. In one embodiment, a plurality of postcharged speed-up circuits, each having a single network node, is connected to the intermediate nodes of a programmable interconnect architecture. Each speed-up circuit monitors the logic level on the network node. When a circuit detects a substantial change in logic level, away from the stand-by level, it temporarily enforces that change by connecting its network node to the signaling logic level. Thus, on each node, a low-impedance enhancement of the signal driving the node temporarily appears. This causes the potential on neighboring nodes, connected through conducting programmable switches, to change towards the signaling level, and their speed-up circuits in turn temporarily enforce the new level. After the temporary enforcement of the signaling level, each speed-up circuit forces its network node back to the stand-by level, for a predetermined period of time. Thus, a forced pulse away from the stand-by logic level towards the signaling level on a node quickly propagates to its connected nodes.

Postcharged interconnection speed-up circuit

A circuit style, which may be employed in fast, area-efficient, flexible programmable interconnect architectures, or in logic circuits, is disclosed. In one embodiment, a plurality of internally delayed logic circuits, each having a single network node, is connected to the intermediate nodes of a programmable interconnect architecture. Each speed-up circuit monitors the logic level on the network node. When a circuit detects a substantial change in logic level, it temporarily enforces that change by connecting its network node to either the high or the low logic level. Thus, on each node, a low-impedance enhancement of the signal driving the node temporarily appears. This causes the potential on neighboring nodes, connected through conducting programmable switches, to change towards the new level, and their speed-up circuits in turn temporarily enforce the new level. Thus, a forced high-to-low or low-to-high level change on a node quickly propagates to its connected nodes.

Self-timed interconnect speed-up circuit

A circuit style, which may be employed in fast, area-efficient, flexible programmable interconnect architectures, is disclosed. In one embodiment, a plurality of clocked dynamic logic circuits, each having a single network node, is connected to the intermediate nodes of a programmable interconnect architecture. During the precharge clock phase, the circuits precharge the intermediate nodes to a high logic level. During the evaluation clock phase, each circuit is initially in the stand-by state, in which it monitors the logic level on its network node. If a substantial deviation from the high level towards the low level is detected, the circuit switches to the discharge state, in which it enforces that level change by connecting its network node to the low level. This causes the potential on neighboring nodes, connected through conducting programmable switches, to change towards a low level, and their circuits in turn switch to the discharge state. Thus, a forced high-to-low level change on a node during the evaluation clock phase quickly propagates to its connected nodes.

Dynamic logic interconnect speed-up circuit

A user-programmable inter-chip interconnect architecture, which may be used for providing programmable interconnections among a plurality of integrated circuits, is disclosed. A plurality of main circuitry in the core region of an integrated circuit is connected through connection nodes to a programmable peripheral switch network in the frame region of the integrated circuit. The peripheral switch network may be programmed by the user to obtain the desired signal-propagating paths between said connection nodes and bonding pads of the peripheral switch network, or among bonding pads of the peripheral switch network. The peripheral switch network has intersecting wiring channels attached to the bonding pads and the connection nodes. Programmable junctions may be present at the intersections of the wiring channels. A substantial number of desired interconnections may be achieved that have only one such programmable junction in the signal-propagating path. When a plurality of integrated circuits including this architecture arc mounted on a substrate that provides fixed conductive leads between the bonding pads of the integrated circuits, user-programmable, highly flexible, high-performance two-point and multi-point connections among the main circuitry of the integrated circuits are obtained.

Electrically programmable inter-chip interconnect architecture

A dynamic logic circuit style, which may be employed in fast logic circuits, is disclosed. The logic functionality is obtained using complementary pass transistor logic networks. Intermediate nodes and the complementary outputs are precharged to a high level during the precharge clock phase. During the evaluation clock phase, the inputs connected to gates of the pass transistors are constrained to either make a low-to-high transition or to maintain their initial low level; such signals are referred to as rising signals. The inputs connected to current electrodes of the pass transistors are constrained to either make a high-to-low transition or maintain their initial high level; such signals are referred to as falling signals. All signals are organized in pairs of true and complementary signals. In a falling signal pair, only one of the signals is allowed to make a high-to-low transition during an evaluation phase, while the other signal remains at the initial high level. In a rising signal pair, only one of the signals is allowed to make a low-to-high transition during an evaluation phase, while the other signal remains at the initial low level. This ensures that one of the true and complementary output nodes of the pass transistor networks makes a high-to-low transition, while the other node remains at a high level. The high-to-low transition is locally buffered by a clocked regenerative feedback circuit present on both output nodes. The signals on the output nodes form a pair of true and complementary signals which may be used as falling input signals for another circuit stage in the same circuit style. In the regenerative feedback circuits, the inverted output signals are produced and these signals form a pair of true and complementary rising signals which may be used as rising input signals for another circuit stage in the same circuit style.

Dynamic complementary pass-transistor logic circuit

Growing community of inventors

Palo Alto, CA, United States of America

Ivo J Dobbelaere