Qualcomm Incorporated

James Michael Artmeier

Sumit Sur

Jeffrey A Niemann

Jon James Anderson

Qing Li

Bohuslav Rychlik

Philip T Mueller, Jr

Mark D Guzzi

James Lyall Esliger

Robert Dreyer

Michael Douglas Sharp

Systems and methods for dynamic granularity control of parallelized work in a heterogeneous multi-processor portable computing device (PCD) are provided. During operation a first parallelized portion of an application executing on the PCD is identified. The first parallelized portion comprising a plurality of threads for parallel execution on the PCD. Performance information is obtained about a plurality of processors of the PCD, each of the plurality of processors corresponding to one of the plurality of threads. A number M of workload partition granularities for the plurality of threads is determined, and a total execution cost for each of the M workload partition granularities is determined. An optimal granularity comprising a one of the M workload partition granularities with a lowest total execution cost is determined, and the first parallelized portion is partitioned into a plurality of workloads having the optimal granularity.

System and method for dynamic granularity control of parallelized work in a portable computing device (PCD)

Various embodiments of methods and systems for thermally aware scheduling of workloads in a portable computing device that contains a heterogeneous, multi-processor system on a chip ('SoC') are disclosed. Because individual processing components in a heterogeneous, multi-processor SoC may exhibit different processing efficiencies at a given temperature, and because more than one of the processing components may be capable of processing a given block of code, thermally aware workload scheduling techniques that compare performance curves of the individual processing components at their measured operating temperatures can be leveraged to optimize quality of service ('QoS') by allocating workloads in real time, or near real time, to the processing components best positioned to efficiently process the block of code.

Thermally driven workload scheduling in a heterogeneous multi-processor system on a chip

Methods and systems for leveraging temperature sensors in a portable computing device ('PCD') are disclosed. The sensors may be placed within the PCD near known thermal energy producing components such as a central processing unit ('CPU') core, graphical processing unit (&#x201c;GPU&#x201d;) core, power management integrated circuit (&#x201c;PMIC&#x201d;), power amplifier, etc. The signals generated by the sensors may be monitored and used to trigger drivers running on the processing units. The drivers are operable to cause the reallocation of processing loads associated with a given component's generation of thermal energy, as measured by the sensors. In some embodiments, the processing load reallocation is mapped according to parameters associated with pre-identified thermal load scenarios. In other embodiments, the reallocation occurs in real time, or near real time, according to thermal management solutions generated by a thermal management algorithm that may consider CPU and/or GPU performance specifications along with monitored sensor data.

Method and system for thermal load management in a portable computing device

Methods and systems for managing thermal load distribution on a portable computing device ('PCD') include storing on a PCD a plurality of thermal load steering scenarios which identify simulated thermal load conditions for the PCD, corresponding simulated workloads that produced the simulated thermal load conditions, and thermal load steering parameters for steering the simulated thermal load to a predetermined spatial location on the PCD. A scheduled workload for the PCD is monitored to identify a match with one of the thermal load steering scenarios so that the workload may be scheduled according to a thermal load steering parameter. Another method includes initiating a thermal mitigation technique on a PCD and determining a current graphical load being processed by the PCD. A graphics feature associated with the current graphical load is identified. The graphics feature is then disabled while maintaining a frame rate to reduce temperature of the PCD.

Method and system for controlling thermal load distribution in a portable computing device

The aspects enable a computing device or microprocessor to scale the frequency and/or voltage of a processor to an optimal value balancing performance and power savings in view of a current processor workload. Busy and/or idle duration statistics are calculated from the processor during execution. The statistics may include a running average busy and/or idle duration or idle/busy ratio, a variance of the running average and a trend of the running average. Current busy or idle durations or an idle-to-busy ratio may be computed based on collected statistics. The current idle-to-busy ratio may be compared to a target idle-to-busy ratio and the frequency/voltage of the processor may be adjusted based on the results of the comparison to drive the current running average toward the target value. The target value of idle-to-busy ratio may be adjusted based on the calculated variance and/or trend values.

Dynamic voltage and clock scaling control based on running average, variant and trend

Various embodiments of methods and systems for controlling and/or managing thermal energy generation on a portable computing device that contains a heterogeneous multi-core processor are disclosed. Because individual cores in a heterogeneous processor may exhibit different processing efficiencies at a given temperature, thermal mitigation techniques that compare performance curves of the individual cores at their measured operating temperatures can be leveraged to manage thermal energy generation in the PCD by allocating and/or reallocating workloads among the individual cores based on the performance curve comparison.

System and method for managing thermal energy generation in a heterogeneous multi-core processor

Various embodiments of methods and systems for determining the thermal status of processing components within a portable computing device ('PCD') by measuring leakage current on power rails associated with the components are disclosed. One such method involves measuring current on a power rail after a processing component has entered a 'wait for interrupt' mode. Advantageously, because a processing component may &#x201c;power down&#x201d; in such a mode, any current remaining on the power rail associated with the processing component may be attributable to leakage current. Based on the measured leakage current, a thermal status of the processing component may be determined and thermal management policies consistent with the thermal status of the processing component implemented. Notably, it is an advantage of embodiments that the thermal status of a processing component within a PCD may be established without the need to leverage temperature sensors.

Growing community of inventors

Boulder, CO, United States of America

James Michael Artmeier