Abstract: An error correction module is disclosed whereby two bit cells are used to store a bit of information in a redundant manner so that a redundant error correction module can correct a sporadic data error at one of the two bits.

Abstract: An error correction module is disclosed whereby two bit cells are used to store a bit of information in a redundant manner so that a redundant error correction module can correct a sporadic data error at one of the two bits.

Abstract: A content addressable memory includes a first plurality of search lines, a second plurality of search lines, a first match line, and a storage location. Each search line of the first plurality of search lines receives a corresponding high voltage level or low voltage level during a match detect operation, and each search line of the second plurality of search lines to receive a corresponding high voltage level or low voltage level during the match detect operation. The storage location of the content addressable memory includes a plurality of CAM cells, each CAM cell a first thyristor and second thyristor.

Abstract: A content addressable memory includes a first plurality of search lines, a second plurality of search lines, a first match line, and a storage location. Each search line of the first plurality of search lines receives a corresponding high voltage level or low voltage level during a match detect operation, and each search line of the second plurality of search lines to receive a corresponding high voltage level or low voltage level during the match detect operation. The storage location of the content addressable memory includes a plurality of CAM cells, each CAM cell a first thyristor and second thyristor.

Abstract: A content addressable memory includes a first plurality of search lines, a second plurality of search lines, a first match line, and a storage location. Each search line of the first plurality of search lines receives a corresponding high voltage level or low voltage level during a match detect operation, and each search line of the second plurality of search lines to receive a corresponding high voltage level or low voltage level during the match detect operation. The storage location of the content addressable memory includes a plurality of CAM cells, each CAM cell a first thyristor and second thyristor.

Abstract: A scannable latch is disclosed. The scannable latch includes a dynamic circuit, two cross-coupled NAND gates coupled to the dynamic circuit, and a pair of stacked transistors coupled to the dynamic circuit. One of the stacked transistors is for receiving data signals, and the other stacked transistors is for receiving scan in signals.

Abstract: An apparatus, a method and a computer program product are provided for conserving energy during functional mode of a processor by disabling the scan chain. By inserting logic gating into the scan chain it is possible to disable the scan chain during the processor's functional mode. During functional mode the scan out port of the latch bit in a scan chain toggles, which leads to unnecessary energy consumption. By gating scan control signals and the scan out port of a latch bit, the scan chain segment between latch bits can be disconnected. Therefore, the scan control signals can disable the scan chain during functional mode.

Abstract: A disclosed byte execution unit receives byte instruction information and two operands, and performs an operation specified by the byte instruction information upon one or both of the operands, thereby producing a result. The byte instruction specifies either a count ones in bytes operation, an average bytes operation, an absolute differences of bytes operation, or a sum bytes into halfwords operation. In one embodiment, the byte execution unit includes multiple byte units. Each byte unit includes multiple population counters, two compressor units, adder input multiplexer logic, adder logic, and result multiplexer logic. A data processing system is described including a processor coupled to a memory system. The processor includes the byte execution unit. The memory system includes a byte instruction, wherein the byte instruction specifies either the count ones in bytes operation, the average bytes operation, the absolute differences of bytes operation, or the sum bytes into halfwords operation.

Abstract: An apparatus, a method, and a computer program are provided to efficiently use a microprocessor array. Typically, microprocessor arrays can be divided into multiple subarrays. Also, in the conventional arrays, each of the subarrays were engaged when the microprocessor array is used. To alleviate the power consumed by the microprocessor arrays, row selection logic is employed to engage only specific rows of subarrays. Therefore, power consumed by unused subarrys is saved.

Abstract: Methods and apparatus provide for enabling a digital circuit by biasing at least one switch transistor ON such that a voltage potential of a virtual ground node is substantially equal to a voltage potential of a ground node for a power supply to the digital circuit, wherein the digital circuit is implemented using a plurality of transistors in a silicon-on-insulator (SOI) arrangement and at least some of the transistors are referenced to the virtual ground node; and disabling the digital circuit by biasing a gate terminal of the switch transistor below the voltage potential of the ground node.

Abstract: A preferred embodiment of the present invention provides a method, computer program product, and processor design for supporting high-precision floating-point function estimates are split in two instructions each: a low precision table lookup instruction and a linear interpolation instruction. Estimates of different functions can be implemented using this scheme: A separate table-lookup instruction is provided for each different function, while only a single interpolation instruction is needed, since the single interpolation instruction can perform the interpolation step for any of the functions to be estimated. Thus, a preferred embodiment of the present invention incurs significantly less overhead than would specialized hardware, while still maintaining a uniform FPU latency, which allows for much simpler control logic.

Abstract: A system and method for a processor with memory with combined line and word access are presented. A system performs narrow read/write memory accesses and wide read/write memory accesses to the same memory bank using multiplexers and latches to direct data. The system processes 16 byte load/sore requests using a narrow read/write memory access and also processes 128 byte DMA and instruction fetch requests using a wide read/write memory access. During DMA requests, the system writes/reads sixteen DMA operations to memory on one instruction cycle. By doing this, the memory is available to process load/store or instruction fetch requests during fifteen other instruction cycles.

Abstract: The present invention provides for a method for memory array verification. Initialization commands are received and memory array initialization settings are generated based on received initialization commands. The memory array initialization settings are stored in a memory array. A deterministic read output function for the memory array is identified and a logic built-in self test scan on the memory array is performed based on the identified deterministic read output function.

Abstract: Systems and methods for increasing the amount of current that can flow through the data line pull-down transistors in a sense amplifier by tying the bodies of these transistors to a voltage other than ground. In one embodiment, the bodies of the data line pull-down transistors in a sense amplifier are tied to the intermediate nodes on the opposing side of the sense amplifier to increase the current flow through the data line pull-down transistors, and also to reduce the voltage at the intermediate node that will be pulled low by the action of the bit line transistors. In one embodiment, the sense amplifier also includes pre-charge circuits which pre-charge the intermediate nodes to a predetermined voltage that is not reduced by the threshold voltage of the pull-down transistors.

Abstract: Systems and methods for decreasing the sensitivity of a sense amplifier to variations in the threshold voltages of the data line pull-down transistors by pre-charging the intermediate nodes of the sense amplifier to the voltages on the opposing bit lines when the sense amplifier is not enabled. In one embodiment, the intermediate nodes are coupled to the input bit lines through transistors that are switched on when the sense amplifier is not enabled and switched off when the sense amplifier is enabled. In one embodiment, the intermediate nodes are pre-charged to a predetermined voltage before being pre-charged to the voltages on the bit lines. In one embodiment, the bodies of the data line pull-down transistors may also be body-tied to the opposing intermediate nodes to increase current flow through these transistors, particularly on the side of the sense amplifier that will be pulled low when the sense amplifier is enabled.

Abstract: An apparatus, a method and a computer program are provided to reduce leakage current in a processor. Traditionally, extra logic is employed to reduce leakage currents. However, reducing leakage current without sacrificing fine grain operations and speed can be difficult. Achieving such a goal can be accomplished by incorporating a multiplexer (mux) into the scan-in path of scan registers so that units or sub-units of the processor can be powered down individually. Additionally, the muxes are not incorporated into time paths, so speed can be preserved.

Abstract: A method, an apparatus, and a computer program are provided to more efficiently generate a sticky bit in a Floating Point Design. Traditionally, separate ORing logic or OR trees were employed to compress the stick outputs of a normalization shifter into at least one sticky bit. However, this design has power consumption and area costs associated with it. To overcome these disadvantages, the OR trees of Leading Zero Counters (CLZs) are employed in conjunction with the Edge Vector logic of a Leading Sign Anticipator and an additional OR gate to determine the sticky bit.

Abstract: A scannable latch is disclosed. The scannable latch includes a dynamic circuit, two cross-coupled NAND gates coupled to the dynamic circuit, and a pair of stacked transistors coupled to the dynamic circuit. One of the stacked transistors is for receiving data signals, and the other stacked transistors is for receiving scan in signals.

Abstract: A method, an apparatus, and a computer program are provided for distributing data in a high speed processing unit. Traditionally, true readout data from multiport register files are inverted multiple times when transmitting the readout to data latches, located at multiple physical layers. The inversion of the readout data can be boost the signals and provide the proper true or complement data to the data latches. To reduce the number of inverters, the register files are configured to output true and complement signals. Therefore, power consumption and area are reduced with the elimination of the inverters.

Abstract: A programmable logic array (PLA) latch is disclosed. The PLA latch includes a first logic array, a second logic array and only one output latch. The second logic array is coupled to the first logic array. The output latch is coupled to the second logic array.