International Business Machines Corporation

Holger Wetter

Jens Noack

Joachim Keinert

Wilhelm Ernst Haller

Monika Strohmer

Rajiv V Joshi

Sudipto Chakraborty

Maarten Jakob Boersma

Alexander Fritsch

Thomas Ludwig

Ulrich Krauch

Hans-Werner Anderson

Silvia Melitta Mueller

Dieter F Wendel

Jens Leenstra

Uwe Brandt

Jochen Preiss

Tobias Gemmeke

Guenter Gerwig

Thomas Makowski

Christoph W Wandel

Thomas Kalla

Juergen Haas

Armin Windschiegl

Wolfgang Bultmann

Alexander W&ouml;rner

A singled ended current sense amplifier circuit including an input stage having a bitline node, a sense node and a feedback circuit comprising a feedback inverter configured to provide an amplified voltage from the bitline node. The feedback inverter may include first and second NMOS transistors serially connected to a feedback node and first and second PMOS transistors serially connected to the feedback node. The feedback circuit may include a third NMOS transistor having a gate terminal connected to the feedback node and a drain terminal connected to the sense node. The input stage may include a third PMOS transistor operating as a current source to generate a sense current which flows in a current sensing path between the sense node and the bitline node. The input stage may act as a regulator to keep the voltage at the bitline node constant.

Single ended current mode sense amplifier with feedback inverter

A sense amplifier circuit includes a bitline node, a sense node, and a feedback circuit which is connected to the bitline node and to the sense node. The feedback circuit includes a cascode-connected pair of transistors configured to isolate the bitline node from an occurrence of a voltage variation on the sense node.

Single ended bitline current sense amplifiers

The invention relates to an integrated circuit comprising: a row of sink cells, a first driver cell, a second driver cell, an interconnect line connecting the first driver cell to the sink cells of the row; and a shunt line connecting the second driver cell to a point between ends of the interconnect line, wherein a segment of the interconnect line between the point and the first driver cell is bigger than 60% of a length the interconnect line and less than 80% of the length of the interconnect line.

Layouting of interconnect lines in integrated circuits

A computer system may obtain a first schematic design netlist for a first IC design and a second schematic design netlist for a second IC design. The computer system may normalize the first netlist and the second netlist. The computer system may determine that the normalized first netlist is the same as the normalized second netlist. The computer system may obtain a first layout design data for the first IC design and a second layout design data for the second IC design. The computer system may determine that the first layout data is the same as the second layout data. The computer system may copy a sign-off data of the first IC design to the second IC design.

Enabling an incremental sign-off process using design data

Embodiments of the present invention provide systems and methods for reducing power consumption during the operation of a SRAM cell. Embodiments of the present invention reduce power consumption by determining switching activity, and based off a determination of low switching activity, gates off a core which is not written; and limits switching activity on the unaddressed core by applying the highest order bit.

NAND-based write driver for SRAM

An integrated circuit includes a static random access memory array. The static random access memory array includes at least two cores, wherein only one of the cores is written at a time. The integrated circuit further includes a tristate driver. The tristate driver is configured to apply a high impedance state to one of the cores that is not being written. A corresponding electronic dataset product includes a description for the integrated circuit expressed in a hardware description language. A corresponding computer-implemented method generates an electronic description for the integrated circuit expressed in a hardware description language.

Write-bitline control in multicore SRAM arrays

Mechanisms are provided for generating a physical layout of an integrated circuit design. A logic description of the integrated circuit design is received that comprises a first logic description of an irregular logic block of the integrated circuit design and a second logic description of a regular logic block of the integrated circuit design. A manual design of the regular logic block of the integrated circuit design is performed based on user input and an automated design of the irregular logic block of the integrated circuit design is performed without user input. The manual design of the regular logic block and the automated design of the irregular logic block are then integrated into the integrated circuit design to generate a hybrid integrated circuit design.

Incorporating synthesized netlists as subcomponents in a hierarchical custom design

A shifter that includes a plurality of shift stages positioned within the shifter, and receiving and shifting input data to generate a shifted result, and a detection circuit coupled at an input of a final shift stage of the plurality of shifters, in a final stage within the shifter. The detection circuit receives a partially shifted vector at the input of the final shift stage along with a predetermined shift amount, and performing an all-one or all-zero detection operation using a portion of the partially shifted vector and the predetermined shift amount, in parallel, to a shifting operation performed by the final shift stage to generate the shifted result.

Shifter with all-one and all-zero detection using a portion of partially shifted vector and shift amount in parallel to generated shifted result

A system for creating layout and wiring diagrams for an integrated circuit (IC) includes a placement engine configured to receive a hierarchical schematic and to create a placed layout. The system also includes a flat layout engine configured to receive the hierarchical schematic and to create a flat layout and a back annotation engine coupled to the placement engine and the flat layout engine, the back annotation engine configured to receive the hierarchical placed layout and the flat unplaced layout and to create a flat placed layout there from.

Fast routing of custom macros

A digital system and method for scanning sequential logic elements are disclosed. The digital system may comprise a plurality of sequential logic elements subdivided into power domains, wherein at least one of the power domains is power gated; a scan chain configured for processing a scan data sequence; a scan enable switch configured for controlling a scan mode; and at least one shadow engine, wherein the at least one shadow engine comprises a control circuit. At least some of the power domains may be interconnected to the scan chain with the scan enable switch, and the scan enable switch may control the scan mode by asserting a scan enable signal. The at least one power gated power domain with one or more sequential logic elements to be power gated may be bypassed via the at least one shadow engine.

System and method for scanning sequential logic elements

The invention proposes a Floating Point Unit with fused multiply-add, with one addend Method for calculating a result of a division with an A-register and a B-register for two multiplicand operands and a C-register for an addend operand, wherein a divide processor using a subtractive method for calculation with a divisor register and a partial remainder register and a multiplier associated to an subtractor uses the C-register as input, wherein while loading the fraction of the dividend through the divisor register into the partial remainder register of the divide processor a calculated shifting is applied for alignment by using the multiplier associated to the subtractor.

Method for calculating a result of a division with a floating point unit with fused multiply-add

A binary and decimal adder unit uses a pre-sum logic for generating pre-sums of the operands A, B under the presumption of one and zero carry inputs into the decimal digit position, and also uses a digits carry network for generating binary carries within the decimal digit positions and a high order carry out signal of said plurality of decimal digits. Each decimal digit position of said adder unit provides a six correction and a pre-sum selection. The pre-sum logic comprises a carry prediction logic for generating decimal digit position carry out signals on the presumption of a zero carry input and of a one carry input into the decimal digit. In response to gating signals derived from the carry out signals of the carry prediction logic and operation control signals a pre-selection is performed for selecting a qualified pre-sum generated on the presumption of a zero carry input into the decimal digit, and for selecting a qualified pre-sum generated on the presumption of a one carry input into the decimal digit. The pre-selection of the qualified pre-sums is performed for all decimal digit positions in parallel to the generation and distribution of the carries in the digits carry network over the total of decimal digit positions. The pre-sum selection logic further comprises a two way selector which is responsive to a digit carry-in signal from the digits carry network for selecting one the qualified pre-sums as the correct sum of the digit position. According to one embodiment of the invention, the pre-sum logic generates six corrected pre-sums which are included in the pre-selection of qualified pre-sums.

Binary and decimal adder unit

A combined adder and logic unit having a reduced operation delay of arithmetic and logic operations, and providing an improved fan in and reduced wiring delays and capacity if implemented in the arithmetic and logic section of a microprocessor chip. The unit comprises a carry network (30) connected to operand inputs for generating carry-out signals of the byte positions (By0-By7) and further comprises a pre-sum logic (32) having a bit function generator (42) and a sum generator (45, 46, 48). Said bit function generator derives from the operands Ai and Bi bit functions Gi, Pi which are provided as logic function output and as input to said sum generator for producing preliminary arithmetic functions (SUM0, SUM1) to anticipate carry-in signals of one or zero. A result selector (70) is controlled by a byte position carry-out signal (Cy55) from the carry network means and by operation control signals to select from the output of said pre-sum logic one of the arithmetic functions (SUM0, SUM1) or one of the logic functions as result of the unit operation.

Growing community of inventors

Weil im Schoenbuch, Germany

Holger Wetter