Abstract: Implementations of the present disclosure involve a system and/or method for measuring on-die voltage levels of an integrated circuit through a digital sampling circuit. In particular, the system and/or method utilizes a delay line based analog-to-digital sampling circuit that produces a voltage reading over time, such as at every high frequency clock cycle. The digitized samples are routed to either an on-die memory structure for later analysis or are transmitted to one or more pins of a chip for capture and analysis by an external analyzer.

Abstract: Systems, methods, and other embodiments are disclosed that are configured to provide on-chip current sensing by employing a power distribution network voltage de-convolution technique. A voltage signal on a voltage plane of a system-on-chip device is measured during operation of the system-on-chip device. The voltage signal derives from a power distribution network. The voltage signal is de-convolved, based at least in part on inverse convolution coefficients derived from the power distribution network, to recover a current signal being drawn by the system-on-chip device from the power distribution network.

Abstract: A system includes a power management unit that may monitor the power consumed by a processor including a plurality of processor core. The power management unit may throttle or reduce the operating frequency of the processor cores by applying a number of throttle events in response to determining that the plurality of cores is operating above a predetermined power threshold during a given monitoring cycle. The number of throttle events may be based upon a relative priority of each of the plurality of processor cores to one another and an amount that the processor is operating above the predetermined power threshold. The number of throttle events may correspond to a portion of a total number of throttle events, and which may be dynamically determined during operation based upon a proportionality constant and the difference between the total power consumed by the processor and a predetermined power threshold.

Abstract: Embodiments include systems and methods for optimization of micro-benchmark analysis for microprocessor designs. For example, embodiments seek to generate a suite of micro-benchmarks and associated weighting factors, which can be used to effectively define a weighted aggregate workload condition for a fine-grained (e.g., RTL) simulation in a manner that is a sufficient proxy for predicted commercial workload conditions. The suite of micro-benchmarks can be appreciably more efficient to simulate than the commercial workload, so that using the suite of micro-benchmarks as a proxy for the commercial workload can provide many benefits, including more efficient iterative design.

Abstract: In an integrated circuit that provides a clock signal, an asymmetric frequency-locked loop (AFLL) includes a first digitally controlled oscillator (DCO) that outputs a first signal having a first fundamental frequency, and a second DCO that outputs a second signal having a second fundamental frequency. The integrated circuit includes a voltage regulator that provides a power-supply voltage to the second DCO. Moreover, the AFLL includes control logic that selects one of the first DCO and the second DCO based on an instantaneous value of a power-supply voltage and an average power-supply voltage. Furthermore, the AFLL adjusts a gain of the selected DCO in the first sub-frequency-locked loop based on the instantaneous value of the power-supply voltage and the average power-supply voltage. In this way, an impact of power-supply voltage variations on a time-critical path in the integrated circuit is reduced.

Abstract: Systems and methods for reducing power consumption during data transport across multiple processors when link utilization is low. In a multi-node system, at least one of two nodes may indicate low utilization for a given link between them. In response to further determining no enabled link between the two nodes is over utilized, each of the two nodes may remove the given link from consideration for being scheduled to receive data for transfer and turn off the given link when no more transactions are scheduled for the given link. Disabled links may be re-enabled when high utilization is detected on at least one link between the two nodes.

Abstract: Embodiments of a temperature sensing apparatus are disclosed. The apparatus may include a voltage generator and circuitry. The voltage generator may generate a first voltage level and a second voltage level dependent on an operating temperature. In response to a given change in the operating temperature, the first and second voltage levels may change, with the second voltage level changing by a different amount than the first voltage level. The voltage generator may generate a third voltage level. The circuitry may measure the first voltage level, the second voltage level, and the third voltage level, and may calculate the operating temperature dependent on a ratio of a difference between the first voltage level and the second voltage level and the third voltage level.

Abstract: Embodiments of the invention provide adaptive power ramp control (APRC) in microprocessors. One implementation of the APRC can compute a present core power and a present power ramp condition in the microprocessor, for example, to determine whether the present power is in a particular predefined control zone and whether the present power ramp is greater than a predefined threshold for that control zone. Those determinations can indicate a likelihood of an imminent, undesirable power ramp condition and can inform entry into a control mode. The APRC can generate an appropriate stall control signal in response to its present control mode, and the stall control signal can stall operation of at least one functional unit of the microprocessor according to a predefined stall pattern. This can effectively combat the imminent power ramp condition by reducing the power usage of the microprocessor.

Abstract: Implementations of the present disclosure involve a system and/or method for measuring on-die voltage levels of an integrated circuit through a digital sampling circuit. In particular, the system and/or method utilize a delay line based analog-to-digital sampling circuit that produces a voltage reading over time, such as at every high frequency clock cycle. In one embodiment, the digital sampling circuit may include a clock jitter monitor circuit configured with a constant supply voltage. This clock jitter monitor is configured to measure the clock jitter that is experienced by the digital voltage monitor circuit and, when compared to measured voltage captured by the circuit, may be used to calibrate or otherwise correct the readings provided by the digital voltage monitor circuit.

Abstract: Implementations of the present disclosure involve a system and/or method for measuring on-die voltage levels of an integrated circuit through a digital sampling circuit. In particular, the system and/or method utilize a delay line based analog-to-digital sampling circuit that produces a voltage reading over time, such as at every high frequency clock cycle. In one embodiment, the digital sampling circuit or digital voltage monitor circuit includes a coarse delay component or circuit that further delays the propagation of a clock signal through the delay line. The coarse delay circuit may be programmed to delay the propagation of the signal through the delay line in such a manner as to allow for multiple edges of a clock or test signal to travel simultaneously down the delay line and increase the sensitivity of the circuit.

Abstract: A system includes a power management unit that may monitor the power consumed by a processor including a plurality of processor core. The power management unit may throttle or reduce the operating frequency of the processor cores by applying a number of throttle events in response to determining that the plurality of cores is operating above a predetermined power threshold during a given monitoring cycle. The number of throttle events may be based upon a relative priority of each of the plurality of processor cores to one another and an amount that the processor is operating above the predetermined power threshold. The number of throttle events may correspond to a portion of a total number of throttle events, and which may be dynamically determined during operation based upon a proportionality constant and the difference between the total power consumed by the processor and a predetermined power threshold.

Abstract: In an integrated circuit that provides a clock signal, an asymmetric frequency-locked loop (AFLL) includes a first digitally controlled oscillator (DCO) that outputs a first signal having a first fundamental frequency, and a second DCO that outputs a second signal having a second fundamental frequency. The integrated circuit includes a voltage regulator that provides a power-supply voltage to the second DCO. Moreover, the AFLL includes control logic that selects one of the first DCO and the second DCO based on an instantaneous value of a power-supply voltage and an average power-supply voltage. Furthermore, the AFLL adjusts a gain of the selected DCO in the first sub-frequency-locked loop based on the instantaneous value of the power-supply voltage and the average power-supply voltage. In this way, an impact of power-supply voltage variations on a time-critical path in the integrated circuit is reduced.

Abstract: Embodiments include systems and methods for optimization of micro-benchmark analysis for microprocessor designs. For example, embodiments seek to generate a suite of micro-benchmarks and associated weighting factors, which can be used to effectively define a weighted aggregate workload condition for a fine-grained (e.g., RTL) simulation in a manner that is a sufficient proxy for predicted commercial workload conditions. The suite of micro-benchmarks can be appreciably more efficient to simulate than the commercial workload, so that using the suite of micro-benchmarks as a proxy for the commercial workload can provide many benefits, including more efficient iterative design.

Abstract: Implementations of the present disclosure involve an apparatus and/or method for providing a constant frequency timer signal for a microprocessor that operates with varying core clock signals. The apparatus and/or method utilizes a code generator, such as a gray code generator, operating on a reference clock signal that allows the constant frequency timer signal to be either faster or slower than the core clock frequency. More particularly, the apparatus and/or method may compute a difference between previous gray code samples and add the calculated difference to a software visible reference clock signal such that constant frequency timer signal may be faster or slower than the core clock signal. Through the use of the apparatus and/or method, a core clock signal may be reduced as needed to provide operational power savings to the microprocessor and the computing system employing the techniques described herein, while maintaining synchronization between the executing programs of the computing system.

Abstract: Implementations of the present disclosure involve an apparatus and/or method for providing one or more clock signals that include a skipped clock cycle to a portion of a computing system. The skipped cycle clock signals may be changed by the computing system during operation of the system by altering masks applied to a global clock signal. However, the flexibility to alter various skipped cycle clock signals may introduce noise or signal disruptions within the system. Thus, the present disclosure may also involve an apparatus and/or method for managing the altering of the clock cycle skipping masks to manage the voltage noise introduced into the system by the adjustment of the operating frequency of the portions of the system. In one embodiment, the method includes prioritizing or otherwise ordering the bits of the masks applied to the global clock signal to attempt to prevent similar bits from being altered simultaneously.

Abstract: Implementations of the present disclosure involve a system and/or method for measuring on-die voltage levels of an integrated circuit through a digital sampling circuit. In particular, the system and/or method utilizes a delay line based analog-to-digital sampling circuit that produces a voltage reading over time, such as at every high frequency clock cycle. The digitized samples are routed to either an on-die memory structure for later analysis or are transmitted to one or more pins of a chip for capture and analysis by an external analyzer.

Abstract: Embodiments include systems and methods for asynchronous, glitch-free clock switching across a wide range of clock frequencies with minimal clock down time. Embodiments effectively provide two stages of synchronization across two independent clock domains. In a first synchronization stage, a received, asynchronous clock select signal is translated into a synchronized clock select signal that is effectively synchronous with respect to a first clock domain and is still effectively asynchronous with respect to a second clock domain. In a second synchronization stage, the synchronized clock select signal is resynchronized so as to be effectively synchronous with respect to the second clock domain. The synchronized select signal can be used to disable the clock of the first clock domain, and the resynchronized clock select signal can be used to enable the clock of the second clock domain.

Abstract: Implementations of the present disclosure involve an apparatus and/or method for providing one or more clock signals that include a skipped clock cycle to a portion of a computing system. The skipped cycle clock signals may be changed by the computing system during operation of the system by altering masks applied to a global clock signal. However, the flexibility to alter various skipped cycle clock signals may introduce noise or signal disruptions within the system. Thus, the present disclosure may also involve an apparatus and/or method for managing the altering of the clock cycle skipping masks to manage the voltage noise introduced into the system by the adjustment of the operating frequency of the portions of the system. In one embodiment, the method includes prioritizing or otherwise ordering the bits of the masks applied to the global clock signal to attempt to prevent similar bits from being altered simultaneously.

Abstract: Embodiments include systems and methods for asynchronous, glitch-free clock switching across a wide range of clock frequencies with minimal clock down time. Embodiments effectively provide two stages of synchronization across two independent clock domains. In a first synchronization stage, a received, asynchronous clock select signal is translated into a synchronized clock select signal that is effectively synchronous with respect to a first clock domain and is still effectively asynchronous with respect to a second clock domain. In a second synchronization stage, the synchronized clock select signal is resynchronized so as to be effectively synchronous with respect to the second clock domain. The synchronized select signal can be used to disable the clock of the first clock domain, and the resynchronized clock select signal can be used to enable the clock of the second clock domain.

Abstract: Systems and methods for efficient data transport across multiple processors when link utilization is congested. In a multi-node system, each of the nodes measures a congestion level for each of the one or more links connected to it. A source node indicates when each of one or more links to a destination node is congested or each non-congested link is unable to send a particular packet type. In response, the source node sets an indication that it is a candidate for seeking a data forwarding path to send a packet of the particular packet type to the destination node. The source node uses measured congestion levels received from other nodes to search for one or more intermediate nodes. An intermediate node in a data forwarding path has non-congested links for data transport. The source node reroutes data to the destination node through the data forwarding path.