Abstract: In some aspects, the present disclosure provides a method for bandgap voting. In some configurations, the method includes receiving: (i) a first set of votes from a first client of a system, and (ii) a second set of votes from a second client of the system, wherein the first set of votes indicate a first desired set of operational parameters for controlling a plurality of physical resources in the system, wherein the second set of votes indicate a second desired set of operational parameters for controlling the plurality of physical resources, and wherein the plurality of physical resources are shared by the first client and the second client.

Abstract: In some aspects, the present disclosure provides a method for bandgap voting. In some configurations, the method includes receiving: (i) a first set of votes from a first client of a system, and (ii) a second set of votes from a second client of the system, wherein the first set of votes indicate a first desired set of operational parameters for controlling a plurality of physical resources in the system, wherein the second set of votes indicate a second desired set of operational parameters for controlling the plurality of physical resources, and wherein the plurality of physical resources are shared by the first client and the second client.

Abstract: In certain aspects, a method for frequency scaling comprises determining whether only a subset of multiple processors is active, wherein the multiple processors share one or more resources. The method also comprises increasing a frequency of at least one processor in the subset of the multiple processors if a determination is made that only the subset of the multiple processors is active and the frequency of the at least one processor is below a frequency threshold. This may be done, for example, to increase the time duration of an idle mode for the one or more shared resources and achieve an overall power reduction for a system including the multiple processors, the one or more shared resources, and/or other function blocks.

Abstract: A method and system for adjusting bandwidth within a portable computing device based on danger signals monitored from one on more elements of the portable computing device are disclosed. A danger level of an unacceptable deadline miss (“UDM”) element of the portable computing device may be determined with a danger level sensor within the UDM element. Next, a quality of service (“QoS”) controller may adjust a magnitude for one or more danger levels received based on the UDM element type that generated the danger level and based on a potential fault condition type associated with the particular danger level. The danger levels received from one UDM element may be mapped to at least one of another UDM element and a non-UDM element. A quality of service policy for each UDM element and non-UDM element may be mapped in accordance with the danger levels.

Abstract: In certain aspects, a method for frequency scaling comprises determining whether only a subset of multiple processors is active, wherein the multiple processors share one or more resources. The method also comprises increasing a frequency of at least one processor in the subset of the multiple processors if a determination is made that only the subset of the multiple processors is active and the frequency of the at least one processor is below a frequency threshold. This may be done, for example, to increase the time duration of an idle mode for the one or more shared resources and achieve an overall power reduction for a system including the multiple processors, the one or more shared resources, and/or other function blocks.

Abstract: A method and system for managing safe downtime of shared resources within a portable computing device are described. The method may include determining a tolerance for a downtime period for an unacceptable deadline miss element of the portable computing device. Next, the determined tolerance for the downtime period may be transmitted to quality-of-service (“QoS”) controller. The QoS controller may determine if the tolerance for the downtime period needs to be adjusted. The QoS controller may receive a downtime request from one or more shared resources of the portable computing device. The QoS controller may determine if the downtime request needs to be adjusted. Next, the QoS controller may select a downtime request for execution and then identify which one or more unacceptable deadline miss elements of the portable computing device that are impacted by the selected downtime request.

Abstract: A method and system for adjusting bandwidth within a portable computing device based on danger signals monitored from one on more elements of the portable computing device are disclosed. A danger level of an unacceptable deadline miss (“UDM”) element of the portable computing device may be determined with a danger level sensor within the UDM element. Next, a quality of service (“QoS”) controller may adjust a magnitude for one or more danger levels received based on the UDM element type that generated the danger level and based on a potential fault condition type associated with the particular danger level. The danger levels received from one UDM element may be mapped to at least one of another UDM element and a non-UDM element. A quality of service policy for each UDM element and non-UDM element may be mapped in accordance with the danger levels.

Abstract: Devices, systems, methods, and computer-readable mediums for arbitrating bus transactions on a communications bus based on health information are disclosed. Health information of master devices can be used to adjust priorities of bus transactions from master devices to meet quality of service requirements of the master devices. In one embodiment, a bus interconnect is provided and configured to communicate bus transactions from any of a plurality of master devices to slave device(s) coupled the bus interconnect. The bus interconnect is further configured to map health information for each of the plurality of master devices into virtual priority space. The bus interconnect is further configured to translate the virtual priority space into a physical priority level for each of the plurality of master devices. The bus interconnect is further configured to arbitrate bus transactions for the plurality of master devices based on physical priority level for the plurality of master devices.

Abstract: Devices, systems, methods, and computer-readable mediums for arbitrating bus transactions on a communications bus based on health information are disclosed. Health information of master devices can be used to adjust priorities of bus transactions from master devices to meet quality of service requirements of the master devices. In one embodiment, a bus interconnect is provided and configured to communicate bus transactions from any of a plurality of master devices to slave device(s) coupled the bus interconnect. The bus interconnect is further configured to map health information for each of the plurality of master devices into virtual priority space. The bus interconnect is further configured to translate the virtual priority space into a physical priority level for each of the plurality of master devices. The bus interconnect is further configured to arbitrate bus transactions for the plurality of master devices based on physical priority level for the plurality of master devices.

Abstract: A method and system for managing bandwidth demand for a variable bandwidth processing element in a portable computing device (“PCD”) includes monitoring bandwidth requests of a plurality of constant bandwidth processing elements and monitoring bandwidth requests of a variable bandwidth processing element. The variable bandwidth processing element may be either a graphics processing unit (GPU) or a central processing unit (CPU). The method may include determining if the variable bandwidth processing element needs adjustment to its bandwidth for a bus. Next, a message may be communicated to an aggregate bus driver indicating a level of adjustment for the variable bandwidth processing element. Based on the minimum frequency set for the entire PCD and the level of adjustment for the variable bandwidth processing element, the aggregate bus driver may set a new bandwidth value for the predetermined threshold of the bandwidth limiter for the variable bandwidth processing element.

Abstract: A method for computing the alignment sticky bit in floating-point operations is provided. The method includes computing a pre-computed sticky bit. A significand is aligned based on an alignment counter. A shifter sticky OR is computed. The alignment sticky bit is computed based on the pre-computed sticky bit by ORing the pre-computed sticky bit and the shifter sticky OR when the alignment counter comprises a value greater than or equal to a predefined value.

Abstract: Devices, systems, methods, and computer-readable mediums for arbitrating bus transactions on a communications bus based on health information are disclosed. Health information of master devices can be used to adjust priorities of bus transactions from master devices to meet quality of service requirements of the master devices. In one embodiment, a bus interconnect is provided and configured to communicate bus transactions from any of a plurality of master devices to slave device(s) coupled the bus interconnect. The bus interconnect is further configured to map health information for each of the plurality of master devices into virtual priority space. The bus interconnect is further configured to translate the virtual priority space into a physical priority level for each of the plurality of master devices. The bus interconnect is further configured to arbitrate bus transactions for the plurality of master devices based on physical priority level for the plurality of master devices.

Abstract: A floating point unit, a central processing unit, and a method are provided for adjusting the exponent of a floating point number. During an addition or subtraction of two floating point numbers, the significand of the floating point result is rounded, and the exponent of the result may be adjusted due to normalization or renormalization. The exponent adjustment due to renormalization or the exponent adjustment due to normalization and renormalization is combined with the significand rounding operation.

Abstract: A method for computing the alignment sticky bit in floating-point operations is provided. The method includes computing a pre-computed sticky bit. A significand is aligned based on an alignment counter. A shifter sticky OR is computed. The alignment sticky bit is computed based on the pre-computed sticky bit by ORing the pre-computed sticky bit and the shifter sticky OR when the alignment counter comprises a value greater than or equal to a predefined value.

Abstract: A floating point unit, a central processing unit, and a method are provided for adjusting the exponent of a floating point number. During an addition or subtraction of two floating point numbers, the significand of the floating point result is rounded, and the exponent of the result may be adjusted due to normalization or renormalization. The exponent adjustment due to renormalization or the exponent adjustment due to normalization and renormalization is combined with the significand rounding operation.