摘要:
A method and structure to optimize computational efficiency in a low-power environment. A design structure is embodied in a machine readable medium used in a design process. The design structure includes a component to determine an optimal point for maximizing computational efficiency in a low-power environment, and a component to selectively control operation of at least one processing unit of a plurality of processing units in accordance with the determined optimal point. The design structure further includes at least one of a component for controlling a frequency of a clock signal transmitted to the at least one processing unit in accordance with the determined optimal point, and a component for determining a present power available.
摘要:
A design structure is embodied in a machine readable medium for designing, manufacturing, or testing a design. The design structure includes a threshold register having a counter, a count register, and a non-volatile storage for storing a state when a value of the count register equals or exceeds a value of the threshold register. Also provided is a method of predicting and/or estimating a power cycle duration in order to save a state in non-volatile memory and a circuit. The method includes setting a threshold value; determining that the threshold value has been equaled or exceeded; and saving the state in the non-volatile memory at a first checkpoint based on the threshold value being equaled or exceeded.
摘要:
A method for increasing the manufacturing yield of field programmable gate arrays (FPGAs) or other programmable logic devices (PLDs). An FPGA or other PLD is formed in several sections, each of the sections having its own power bus and input/output connections. Each section of the FPGA or other PLD is tested to identify defects in the FPGA or other PLD. The FPGA or other PLD is sorted according to whether the section has an acceptable number of defects. An assigned unique number for the FPGA or other PLD chip or part identifies it as partially good. Software for execution and configuring the FPGA or other PLD may use the unique number for programming only the identified functional sections of the FPGA or other PLD. The result is an increase in yield as partially good FPGAs or other PLDs may still be utilized.
摘要:
A method for increasing the manufacturing yield of field programmable gate arrays (FPGAS) or other programmable logic devices (PLDs). An FPGA or other PLD is formed in several sections, each of the sections having its own power bus and input/output connections. Each section of the FPGA or other PLD is tested to identify defects in the FPGA or other PLD. The FPGA or other PLD is sorted according to whether the section has an acceptable number of defects. An assigned unique number for the FPGA or other PLD chip or part identifies it as partially good. Software for execution and configuring the FPGA or other PLD may use the unique number for programming only the identified functional sections of the FPGA or other PLD. The result is an increase in yield as partially good FPGAs or other PLDs may still be utilized.
摘要:
A field programmable gate array (FPGA) device including a non-non-programming-based default power-on electronic configuration. The non-non-programming-based default power-on electronic configuration defines a default state to initial a first logic function. Upon power-up, the FPGA device would be enabled to enter the default state without having first to be configured via a conventional programming mode, thus saving precious processing time during power-up. Several embodiments are disclosed, such as a mask via circuit, an asynchronized set/reset circuit, an unbalanced latch circuit and a flush and scan circuit. A related method is also disclosed to reduce the memory size dedicated to the first logic function to facilitate further programming after power-up. In addition to time saving and further programming, the FPGA device can also allow partial or incremental programming to expand the full functionality to match customer's different needs.
摘要:
A system, method and program product for determining a history state of data in a data retaining device are disclosed. A state of a partially-depleted silicon-on-insulator (PD SOI) device coupled to a data retaining device is measured to indicate a body voltage of the PD SOI device. The body voltage of the PD SOI device may indicate, among others, how long the PD SOI device has been idling, which indirectly indicates how long data in the data retaining device has not been accessed. As such, the current invention may be used efficiently with, e.g., a cache replacement algorithm in a management of the data retaining device.
摘要:
A system, method and program product for determining a relative amount of usage of a data retaining device are disclosed. A charge storing device is coupled to a data retaining device in a manner that a use of the data retaining device triggers a charging of the charge storing device. In a period that the data retaining device idles, charges in the charge storing device decay due to natural means. As such, a potential of the charge storing device may be used to indicate an amount of usage of the data retaining device. A comparison of the potentials of two charge storing devices coupled one-to-one to two data retaining devices may be used as a basis to determine a relative amount of usage of each of the two data retaining devices comparing to the other.
摘要:
Localized logic regions of a circuit include a local comparator electrically connected to a local resistive voltage circuit, to a local resistive ground circuit, and to a local register structure. The local comparator supplies a clock pulse to the local register structures when the local reference voltage is below a local voltage threshold. Activity in the local combinatorial logic structure causes the local reference voltage to drop below the local reference voltage independently of changes in the global reference voltage causing the comparator to output the clock pulse (with sufficient delay to allow the logic results to be stored in the registers only after setup times have been met in the local logic devices). This eliminates the need for a clock distribution tree, thereby saving power when there is no activity in the local combinatorial logic structure.
摘要:
A system, method and program product for retaining a logic state of a processor pipeline architecture are disclosed. A comparator is positioned between two stages of the processor pipeline architecture. A storage capacitor is coupled between a storage node of the comparator and a ground to store an output of the early one of the two stages. A reference logic is provided, which has the same value as the output of the early stage. A logic storing and dividing device is coupled between the reference logic and a reference node of the comparator to generate a logic at the reference node, which is a fraction of the reference logic, and to retain a logic state of the information stored on the storage capacitor. Further mechanisms are provided to determine validity of data stored in the logic storing and dividing device.
摘要:
A reconfigurable logic array (RLA) having a logic capacity and configured to process a function having a total logic requirement that exceeds the logic capacity of the RLA. The RLA includes first and second storage regions and a plurality of programmable logic elements located between the first and second storage regions. When the function is parsed into a plurality of functional blocks, this configuration allows the RLA to process the function by sequentially processing the functional blocks in alternating directions within the RLA, using the plurality of programmable logic elements to sequentially process each of the functional blocks and using the first and second storage regions to temporarily hold the input and output for that one of the functional blocks.