Abstract: Techniques are described for adaptive device power management. The device interface application of a hardware computing unit detects a launch of an application by the operating system (OS) to be executed on the hardware computing unit, in an implementation. The device interface application identifies the launched application and determines whether a hardware profile exists that is associated with the application. The hardware profile includes one or more hardware parameters that yield the optimal performance for power consumption by the hardware computing unit when executing the launched application. Based on determining that the hardware profile exists, the power policy of the OS is updated for the launched application, and thereby, the driver updates the power state(s) of the hardware computing unit based on the new power policy.

Abstract: An electronic device includes a memory and a processor. The processor receives a platform management profile that includes information defining one or more platform management policies, a given platform management policy among the one or more platform management policies including a provided input from a specified hardware or software sensor and/or a provided output action. The processor uses the given platform management policy for controlling operating states of elements in the electronic device.

Abstract: An electronic device includes a memory and a processor. The processor receives a platform management profile that includes information defining one or more platform management policies, a given platform management policy among the one or more platform management policies including a provided input from a specified hardware or software sensor and/or a provided output action. The processor uses the given platform management policy for controlling operating states of elements in the electronic device.

Abstract: Dynamic adjustment of power modes including: detecting an application identified in an application power policy; limiting an application power consumption of a computing component based on the application power policy; monitoring power consumption of a computing component; and selecting a power mode based on the monitored power consumption of the computing component and a power consumption threshold for each of a plurality of power modes.

Abstract: Techniques are described for adaptive device power management. The device interface application of a hardware computing unit detects a launch of an application by the operating system (OS) to be executed on the hardware computing unit, in an implementation. The device interface application identifies the launched application and determines whether a hardware profile exists that is associated with the application. The hardware profile includes one or more hardware parameters that yield the optimal performance for power consumption by the hardware computing unit when executing the launched application. Based on determining that the hardware profile exists, the power policy of the OS is updated for the launched application, and thereby, the driver updates the power state(s) of the hardware computing unit based on the new power policy.

Abstract: An electronic device includes a memory and a processor. The processor acquires a platform management profile, the platform management profile including information defining one or more platform management policies. The processor provides the platform management profile to platform management drivers executing on one or more electronic devices, the platform management profile being configured so that each of the platform management drivers can extract the one or more platform management policies from the platform management profile and use the one or more platform management policies for controlling operating states of elements (e.g., functional blocks, devices, etc.) of the respective electronic device.

Abstract: Power shifting based on bottleneck prediction, including: determining a first plurality of performance metrics for an accelerated processing unit (APU) and a second plurality of performance metrics for a graphics processing unit (GPU); providing the first plurality of performance metrics and the second plurality of performance metrics as an input to a model configured to identify one or more bottlenecks in the APU or the GPU; determining, based on an output of the model, a power distribution between the APU and the GPU; and applying the power distribution.

Abstract: Various methods and apparatus for graphics processing are disclosed. In one aspect, a method of graphics processing using a computing system is provided. The method includes booting the computing system. After booting the computing system operating video memory of the computing system at a non-overclocked frequency, and prior to rebooting having the computing system sequentially increment the frequency of video memory by a selected change in frequency through a series of overclocked frequencies, after each frequency incrementing writing data to the video memory and testing the stability of the video memory data writing, and if the stability testing fails then decrementing the frequency of the video memory to a previous overclocked frequency at which the stability testing did not fail.

Abstract: Various methods and apparatus for graphics processing are disclosed. In one aspect, a method of graphics processing using a computing system is provided. The method includes booting the computing system. After booting the computing system operating video memory of the computing system at a non-overclocked frequency, and prior to rebooting having the computing system sequentially increment the frequency of video memory by a selected change in frequency through a series of overclocked frequencies, after each frequency incrementing writing data to the video memory and testing the stability of the video memory data writing, and if the stability testing fails then decrementing the frequency of the video memory to a previous overclocked frequency at which the stability testing did not fail.

Abstract: A GPU performs dynamic power level management by switching between pre-defined power levels having distinct clock and voltage levels. The dynamic power level management includes identifying a first performance metric associated with processing workloads at the for a consecutive number of measurement cycles. In some embodiments, the consecutive number of measurement cycles includes a current measurement cycle and at least one previous measurement cycle. Based on a determination that the consecutive number of measurement cycles exceeds a minimum hysteresis number, an estimated optimization is determined to be applied to the GPU for a future measurement cycle. A power level setting at the GPU for the future measurement cycle is adjusted based on the estimated optimization. By considering performance metrics including, for example, different processing workloads and hardware configurations, the GPU is able to dynamically adapt its power settings to the particular workload that it is currently processing.

Abstract: A GPU performs dynamic power level management by switching between pre-defined power levels having distinct clock and voltage levels. The dynamic power level management includes identifying a first performance metric associated with processing workloads at the for a consecutive number of measurement cycles. In some embodiments, the consecutive number of measurement cycles includes a current measurement cycle and at least one previous measurement cycle. Based on a determination that the consecutive number of measurement cycles exceeds a minimum hysteresis number, an estimated optimization is determined to be applied to the GPU for a future measurement cycle. A power level setting at the GPU for the future measurement cycle is adjusted based on the estimated optimization. By considering performance metrics including, for example, different processing workloads and hardware configurations, the GPU is able to dynamically adapt its power settings to the particular workload that it is currently processing.