Abstract: A mechanism is provided for optimizing semiconductor packing in a three-dimensional (3D) very-large-scale integration (VLSI) device. The 3D VLSI device comprises a processor layer coupled, via a first set of coupling devices, to at least one signaling and input/output (I/O) layer. The 3D VLSI device further comprises a power delivery layer coupled, via a second set of coupling devices, to the processor layer. In the 3D VLSI device the power delivery layer is dedicated to only delivering power and does not provide data communication signals to the elements of the three-dimensional VLSI device, and the at least one signaling and input/output (I/O) layer is dedicated to only transmitting the data communication signals to and receiving the data communications signals from the processor layer and does not provide power to the elements of the processor layer.

Abstract: A technique for manufacturing a three-dimensional integrated circuit includes stacking a memory unit on a first die that includes a first computational unit. In this case, the memory unit is included in a second die. A second computational unit that is included in a third die is stacked on the second die. Sets of vertical vias that extend through the first, second, and third dies are connected to connect components of the first and second computational units and the memory unit. Multiplexers of the first and second computational units are configured to selectively couple the components to different ones of the sets of vertical vias responsive to respective control words for each of the first and third dies.

Abstract: The present invention relates to a memory controller for an IC with an external DRAM, where the external DRAM has at least one memory bank and communicates with the IC via at least one channel. In line with the invention, the memory controller has a command scheduler which prioritizes the transmission of memory bank commands on the basis of a static priority allocation for commands and a dynamic priority allocation for channels.

Abstract: A method of power gating a microprocessor having an instruction scheduling unit for receiving issued instructions from an instruction decode unit; an execution unit coupled to receive and send signals from and to the instruction scheduling unit; and a state machine located within the execution unit, the method comprises: obtaining a number of instructions per cycle being issued to the instruction scheduling unit; determining, subsequent to obtaining the number of instructions per cycle, if the number of instruction per cycle being issued to the instruction scheduling unit is less than a threshold level, and then determining if at least two of the instructions being issued to the instruction scheduling unit are independent of each other only when the instructions per cycle is less than the threshold level; determining when at least two of the instructions being issued to the instruction scheduling unit are independent of each other; and power gating the microprocessor to gate off power to idle macros with a si

Abstract: A method, system and computer program product for reducing power consumption when processing mathematical operations. Power may be reduced in processor hardware devices that receive one or more operands from an execution unit that executes instructions. A circuit detects when at least one operand of multiple operands is a zero operand, prior to the operand being forwarded to an execution component for completing a mathematical operation. When at least one operand is a zero operand or at least one operand is “unordered”, a flag is set that triggers a gating of a clock signal. The gating of the clock signal disables one or more processing stages and/or devices, which perform the mathematical operation. Disabling the stages and/or devices enables computing the correct result of the mathematical operation on a reduced data path. When a device(s) is disabled, the device may be powered off until the device is again required by subsequent operations.

Abstract: The present invention relates to a method for scheduling and controlling access to a multibank memory having at least two banks, and to an apparatus for reading from and/or writing to recording media using such method. According to the invention, the method comprises the steps of: writing an input stream to the first bank; switching the writing of the input stream to the second bank when a read command for the first bank is received; and switching the writing of the input stream back to the first bank when a read command for the second bank is received.

Abstract: A method of reducing latency in instruction processing in a system, includes calculating a result of a first execution unit, storing the result of the first execution unit in a register file, forwarding the result of the first execution unit, through the bypass unit, to a second execution unit, the second execution unit conducting an instruction dependent on the result, forwarding the result of the first execution unit, from the bypass unit, to a third execution unit, without accessing the register file, the third execution unit conducting an instruction dependent on the result, wherein the execution units can extract the result of the first execution unit through the bypass unit until the new result is calculated, wherein after the new result is calculated, the execution units can access the result of the first execution unit through the register file.

Abstract: A method for implementing sign extension within a multi-precision multiplier is described. The method includes receiving a first input within a multiplier core of the multiplier, receiving a second input within the multiplier core, and creating partial products in the multiplier core using the first and second inputs. The method also includes summing up the partial products in a partial product reduction tree in the multiplier core. The method also includes performing sign extension within the partial product reduction tree of the multiplier core by adding a value to a partial product of the partial product reduction tree. The method further includes computing an output from the partial product reduction tree, the output including a final product of the first and second inputs signed extended to a desired width.

Abstract: A method for reducing leakage current within a register file of a processor is disclosed. The register file within the processor is partitioned into at least two power domains, and each of the two power domains can be powered independently. At least one of the two power domains includes at least as many physical registers as there are architected registers defined in an instruction set architecture of the processor. In response to an occurrence of an idle condition within the processor, all architected register file entries are consolidated into one of power domains that will not be powered off, and the power domains that does not contain any architected register file entries after consolidating are powered off. Afterwards, in response to a detection of an end of the idle condition, all of the power domains are powered back on.

Abstract: A method of dispatching instructions includes dispatching original instructions into an instruction buffer, including at least one operand, renaming the operand, selecting the original instructions from the instruction buffer, sending selected instructions with explicit bits, to an internal operation code exchange table, which includes replacement rules for replacing the selected instructions with a simplified instruction based on the original instructions and the explicit bits, replacing the selected instructions with the simplified instruction in accordance with the explicit bits, and issuing the simplified instructions to an execution unit by sending the simplified instruction and all explicit bits for the operands to a content addressable memory address logic of the internal operation code exchange table, wherein if a bitvector, consisting of the original instruction and the explicit bits, matches a pattern stored in the internal operation code exchange table, the original instruction is replaced by the s

Abstract: The present invention relates to a memory controller for an IC with an external DRAM, where the external DRAM has at least one memory bank and communicates with the IC via at least one channel. In line with the invention, the memory controller has a command scheduler which prioritizes the transmission of memory bank commands on the basis of a static priority allocation for commands and a dynamic priority allocation for channels.

Abstract: Processor instruction pipelines, which split the processing of individual instructions into several sub-stages and thus reduce the complexity of each stage while simultaneously increasing the clock speed, are typical features of RISC architectures. Operands required by the processing are read from a register file. Read-after-write access problems in the pipeline processing can be avoided by using a scoreboard that has an individual entry per address of the register file. Once an instruction enters the pipeline, a flag is set at the address of the destination address of this particular instruction. This flag signals that an instruction inside the pipeline wants to write its result to the respective register address. Hence the result is unavailable as long as the flag is set. It is cleared after the instruction process has successfully written the result into the register file.

Abstract: The present invention relates to a method for scheduling and controlling access to a multibank memory having at least two banks, and to an apparatus for reading from and/or writing to recording media using such method. According to the invention, the method comprises the steps of: writing an input stream to the first bank; switching the writing of the input stream to the second bank when a read command for the first bank is received; and switching the writing of the input stream back to the first bank when a read command for the second bank is received.