Abstract: Techniques are disclosed relating to managing power consumption and latencies for entry and exit of idle power states. In one embodiment, a processor includes a processing core configured to operate in a plurality of power states (e.g., C-states) that includes an operating power state and at least one idle power state. The processing core is also configured to operate in a plurality of performance states. The processor further includes a power management unit configured to receive a request from the processing core to enter the at least one idle power state. The power management unit is configured to select a first of the plurality of performance states (e.g., P-states) based on the requested idle power state. In one embodiment, the power management unit is further configured to cause the processing core to transition into the selected first performance state prior to entering the requested idle power state.

Abstract: A method of regulating power states in a processing system may begin with a processor component reporting a present processor power state to an input-output hub, where the present processor power state corresponds to one of a plurality of different processor power states ranging from an active state to an inactive state. The input-output hub receives data indicative of the present processor power state and, in response to receiving the present processor power state, establishes a lowest allowable hub power state that corresponds to one of a plurality of different hub power states ranging from an active state to an inactive state. The method continues by determining a present hub power state for the input-output hub, wherein depth of the present hub power state is less than or equal to depth of the lowest allowable hub power state.

Abstract: A method for transitioning power states in a device includes designating a first reduced power state as a target power state. A first expected residency for the target power state is determined based on a counting of activity requests associated with the device. The device is transitioned to the target power state responsive to the expected residency satisfying a first predetermined threshold.

Abstract: An apparatus and method for per-node thermal control of processing nodes is disclosed. The apparatus includes a plurality of processing nodes, and further includes a power management unit configured to set a first frequency limit for at least one of the plurality of processing nodes responsive to receiving an indication of a first detected temperature greater than a first temperature threshold, wherein the first detected temperature is associated with the one of the plurality of processing nodes. The power management unit is further configured to set a second frequency limit for each of the plurality of processing nodes responsive to receiving an indication of a second temperature greater than a second temperature threshold.

Abstract: A system and method for dynamic function based power control is disclosed. In one embodiment, a system includes a bridge unit having a memory controller and a communication hub coupled to the memory controller. The system further includes a power management unit, wherein the power management unit is configured to clock-gate the communication hub responsive to determining that each of a plurality of processor cores are in an idle state and that an I/O interface unit has been idle for an amount of time exceeding a first threshold. The power management unit is further configured to clock-gate the memory controller responsive to clock-gating the communication hub and determining that a memory coupled to the memory controller is in a first low power state. The power management unit may also perform power-gating of functional units subsequent to clock-gating the same.

Abstract: A power management controller (PMC) that interfaces with a processor comprising one or more cores. The PMC may be configured to communicate with each respective core, such that microcode executed by the respective processor core may recognize when a request is made to transition the respective core to a target power-state. For each respective core, the state monitor may monitor active-state residency, non-active-state residency, Direct Memory Access (DMA) transfer activity associated with the respective core, Input/Output (I/O) processes associated with the respective core, and the value of a timer-tick (TT) interval associated with the respective core. The status monitor may derive respective status information for the respective core based on the monitoring and indicate whether the respective core should be allowed to transition to the corresponding target power-state. The PMC may transition the respective processor core to the corresponding target power-state accordingly.

Abstract: An apparatus and method for performance control of processing nodes is disclosed. In one embodiment, a system includes a processing node and a power management unit configured to, for each of a plurality of time intervals, monitor an activity level of the processing node, cause the processing node to operate at a high operating point during one successive time interval if the activity level in the given interval is greater than a high activity threshold, operate at a low operating point at least one successive time interval if the activity level is less than a low activity threshold, or enable operating system software to cause the processing node to operate at one of one or more predefined intermediate operating points of the plurality of operating points if the activity level is less than the high activity threshold and greater than the low activity threshold.

Abstract: A system includes one or more processor cores, and a voltage regulator that provides an operating voltage to the one or more processor cores in response to receiving a voltage identifier signal that is indicative of the operating voltage. The system also includes a power management unit that may provide a first voltage identifier signal corresponding to a first operating voltage in response to determining that the processor cores are operating in a first operating state in which the one or more processor cores may draw up to a maximum load current. The power management unit may also provide a second voltage identifier signal corresponding to a second operating voltage, which is less than the first operating voltage, in response to determining that the processor cores are operating in a second operating state in which the processor cores are incapable of an increase in load current above a predetermined amount.

Abstract: A system may comprise a plurality of processing units, and a control unit and monitoring unit interfacing with the processing units. The control unit may receive requests for transitioning the processing units to respective target power-states, and specify respective target HW power-states corresponding to the respective target power-states. The monitoring unit may monitor operating characteristics of the system, and determine based on operating characteristics whether to allow the processing units to transition to the respective target hardware (HW) power-states. The control unit may be configured to change the respective target HW power-state to a respective updated HW power-state for each processing units for which it is determined that transition to its respective target HW power-state should not be allowed.

Abstract: A system may comprise a plurality of processing units and a scheduler configured to maintain a record for each respective processing unit. Each respective record may comprise entries which may indicate 1) how long the respective processing unit has been residing in an idle state, 2) a present power-state in which the respective processing unit resides, and 3) whether the respective processing unit is a designated default (bootstrap) processing unit. The scheduler may select one or more of the plurality of processing units according to their respective records, and assign impending instructions to be executed on the selected one or more processing units. Where additional processing units are required, the scheduler may also insert an instruction to trigger an inter-processor interrupt to transition one or more processing units out of idle-state. The scheduler may then assign some impending instructions to these one or more processing units.

Abstract: A method and apparatus for dynamically controlling a cache size is disclosed. In one embodiment, a method includes changing an operating point of a processor from a first operating point to a second operating point, and selectively removing power from one or more ways of a cache memory responsive to changing the operating point. The method further includes processing one or more instructions in the processor subsequent to removing power from the one or more ways of the cache memory, wherein said processing includes accessing one or more ways of the cache memory from which power was not removed.

Abstract: A method for power management is disclosed. The method may include monitoring requests for access to a memory of a memory subsystem by one or more processor cores; and monitoring requests for access to the memory conveyed by an input/output (I/O) unit. The method may further include determining if at least a first amount of time has elapsed since any one of the processor cores has asserted a memory access request and determining if at least a second amount of time has elapsed since the I/O unit has conveyed a memory access request. A first signal may be asserted if the first and second amounts of time have elapsed. A memory subsystem may be transitioned from operating in a full power state to a first low power state responsive to assertion of the first signal.

Abstract: A processor may comprise one or more cores, where each respective core may comprise one or more state registers, and non-volatile memory configured to store microcode instructions executed by the respective processor core. The processor may further comprise a power management controller (PMC) interfacing with each respective core, and a state monitor (SM) interfacing with the PMC. The PMC may be configured to communicate with each respective core, such that microcode executed by the respective processor core may recognize when a request is made to transition the respective core to a low-power state. The microcode may communicate the request to the PMC, which may in turn determine if the request is for the respective core to transition to a zero-power state. If it is, the PMC may communicate with the SM to determine whether to transition the respective processor core to the zero-power state, and initiate transition to the zero-power state if a determination to transition to the zero-power state is made.

Abstract: A processing node tracks probe activity level associated with its internal caching or memory system. If the probe activity level increases above a threshold probe activity level, the performance state of the processing node is increased above its current performance state to provide enhanced performance capability in responding to the probe requests. After entering the higher performance state in response to the probe activity level being above the threshold probe activity level, the processing nodes returns to a lower performance state in response to a reduction in probe activity. There may be multiple threshold probe activity levels and associated performance states.

Abstract: A processing node tracks probe activity level associated with its cache. The processing node and/or processing system further predicts an idle duration. If the probe activity level increases above a threshold probe activity level, and the idle duration prediction is above a threshold idle duration threshold, the processing node flushes its cache to prevent probes to the cache. If the probe activity level is above the threshold probe activity level but the predicted idle duration is too short, the performance state of the processing node is increased above its current performance state to provide enhanced performance capability in responding to the probe requests.

Abstract: A method for transitioning power states in a device includes designating a first reduced power state as a target power state. A first expected residency for the target power state is determined based on a counting of activity requests associated with the device. The device is transitioned to the target power state responsive to the expected residency satisfying a first predetermined threshold.

Abstract: Performance sensitivities to a change in performance capabilities of computational units of a computer system are determined based on measured utilization metrics for each of the computational units. In order to determine the performance sensitivities, in one approach, the computational units are operated at a first performance level, and respective first utilization metrics are determined. The computational units are then operated at a second performance level and respective second utilization metrics are determined. The sensitivity to performance capability change, e.g., a frequency change, is determined based on the respective first and second utilization metrics. The performance sensitivities of the computational units to a change in performance capability are continually updated in response to, e.g.

Abstract: One or more computational units of a computer system are selectively altered in terms of performance according to which of the one or more computational units has a higher performance sensitivity than others of the computational units.

Abstract: A power allocation strategy limits performance of a subset of a plurality of computational units in a computer system according to performance sensitivity of each of the plurality of computational units to a change performance capability, e.g., frequency change. The performance of the subset of computational units may be limited by setting a power state in which the subset may be operated and/or reducing a current power state of the subset to a lower power state. The subset whose performance is limited includes computational units that are least performance sensitive according to stored sensitivity data. The subset may include one or more processing cores and performance of the one or more processing cores may be limited in response to a CPU-bounded application or graphics processing unit (GPU)-bounded application being executed.

Abstract: A method includes applying a voltage to a first processor core of a plurality of processor cores to deactivate the first processor core, the voltage less than a retention voltage of the first processor core. The application of the voltage can be in response to a software setting. The software setting can be configured via a user input, a software application, an operating system, or a BIOS setting. Alternately, the application of the voltage can be in response to a permanent hardware setting, such as the state of a fuse associated with the first processor core.