Abstract: In one embodiment, a processor includes a plurality of functional units each to independently execute instructions and a clock distribution circuit having a clock signal generator to generate a clock signal. The clock distribution circuit is coupled to receive a first operating voltage from a first voltage rail and the functional units are coupled to independently receive at least one second operating voltage from one or more second voltage rails. Other embodiments are described and claimed.

Abstract: According to one embodiment of the invention, a processor includes a power control unit, an interface to software during runtime that permits the software to set a plurality of power management constraint parameters for the power control unit during runtime of the processor without a reboot of the processor, and a storage element to store a respective lock bit for each of the plurality of power management constraint parameters to disable the interface from changing a respective constraint parameter when set.

Abstract: In an embodiment, a processor includes a core domain with a plurality of cores and a power controller having a first logic to receive a first request to increase an operating voltage of a first core of the core domain to a second voltage, to instruct a voltage regulator to increase the operating voltage to an interim voltage, and to thereafter instruct the voltage regulator to increase the operating voltage to the second voltage. Other embodiments are described and claimed.

Abstract: Embodiments of systems, apparatuses, and methods for energy efficiency and energy conservation including enabling autonomous hardware-based deep power down of devices are described. In one embodiment, a system includes a device, a static memory, and a power control unit coupled with the device and the static memory. The system further includes a deep power down logic of the power control unit to monitor a status of the device, and to transfer the device to a deep power down state when the device is idle. In the system, the device consumes less power when in the deep power down state than in the idle state.

Abstract: In an embodiment, a processor includes a core domain with a plurality of cores and a power controller having a first logic to receive a first request to increase an operating voltage of a first core of the core domain to a second voltage, to instruct a voltage regulator to increase the operating voltage to an interim voltage, and to thereafter instruct the voltage regulator to increase the operating voltage to the second voltage. Other embodiments are described and claimed.

Abstract: In one embodiment, the present invention includes a processor having multiple domains including at least a core domain and a non-core domain that is transparent to an operating system (OS). The non-core domain can be controlled by a driver. In turn, the processor further includes a memory interconnect to interconnect the core domain and the non-core domain to a memory coupled to the processor. Still further, a power controller, which may be within the processor, can control a frequency of the memory interconnect based on memory boundedness of a workload being executed on the non-core domain. Other embodiments are described and claimed.

Abstract: A heterogeneous processor architecture is described.

Abstract: Remapping technologies for execution context swap between heterogeneous functional hardware units are described. A computing system includes multiple registers configured to store remote contexts of functional units. A mapping table maps the remote context to the functional units. An execution unit is configured to execute a remapping tool that intercepts an operation to access a remote context of a first functional unit of the plurality of functional units that is taken offline. The remapping tool determines that the first functional unit is remapped to a second functional unit using the mapping table. The operation is performed to access the remote context that is remapped to the second functional unit. The first functional unit and the second functional unit may be heterogeneous functional units.

Abstract: A method and apparatus for dynamic power limit sharing among the modules in the platform. In one embodiment of the invention, the platform comprises a processor and memory modules. By expanding the power domain to include the processor and the memory modules, dynamic sharing of the power budget of the platform between the processor and the memory modules is enabled. For low-bandwidth workloads, the dynamic sharing of the power budget offers significant opportunity for the processor to increase its frequency by using the headroom in the memory power and vice versa. This enables higher peak performance for the same total platform power budget in one embodiment of the invention.

Abstract: A method and apparatus for dynamic power limit sharing among the modules in the platform. In one embodiment of the invention, the platform comprises a processor and memory modules. By expanding the power domain to include the processor and the memory modules, dynamic sharing of the power budget of the platform between the processor and the memory modules is enabled. For low-bandwidth workloads, the dynamic sharing of the power budget offers significant opportunity for the processor to increase its frequency by using the headroom in the memory power and vice versa. This enables higher peak performance for the same total platform power budget in one embodiment of the invention.

Abstract: Embodiments of the invention relate to energy efficient and conserving thermal throttling of electronic device processors using a zero voltage processor state. For example, a processor die may include a power control unit (PCU), and an execution unit having power gates and a thermal sensor. The PCU is attached to the thermal sensor to determine if a temperature of the execution unit has increased to greater than an upper threshold, such as while the execution unit is processing data in an active processor power state. The PCU is also attached to the power gates so that upon such detection, it can change the active processor power state to a zero processor power state to reduce the temperature of the execution unit. When the sensor detects that the temperature has decreased to less than a lower threshold, the PCU can change the processor power state back to the active state.

Abstract: In accordance with embodiments disclosed herein, there is provided systems and methods for managing shared resources between multiple processing devices. The processor may include a first processing device comprising a first non-coherent hardware block (hb) including a non-coherent data and a second processing device comprising a second non-coherent hb including the non-coherent data. The processor may also include a first hb in communication with the first non-coherent hb and the second non-coherent hb to track and share the non-coherent data between the first and the second processing devices.

Abstract: Processor context save latency is reduced by only restoring context registers with saved state that differs from the reset value of registers. A system agent monitors access to the design blocks and sets a dirty bit to indicate which design block has registers that have changed since the last context save. During a context save operation, the system agent bypasses design blocks that have not had context changes since the latest context save operation. During a context restore operation the system agent does not restore the context registers with saved context values that are equal to the reset value of the context register.

Abstract: In one embodiment, a processor includes a plurality of functional units each to independently execute instructions and a clock distribution circuit having a clock signal generator to generate a clock signal. The clock distribution circuit is coupled to receive a first operating voltage from a first voltage rail and the functional units are coupled to independently receive at least one second operating voltage from one or more second voltage rails. Other embodiments are described and claimed.

Abstract: Technologies for one-level memory (1LM) and two-level memory (2LM) configurations in a common platform are described. A processor includes a first memory interface coupled to a first memory device that is located off-package of the processor and a second memory interface coupled to a second memory device that is located off-package of the processor. The processor also includes a multi-level memory controller (MLMC) coupled to the first memory interface and the second memory interface. The MLMC includes a first configuration and a second configuration. The first memory device is a random access memory (RAM) of a one-level memory (1LM) architecture in the first configuration. The first memory device is a first-level RAM of a two-level memory (2LM) architecture in the second configuration and the second memory device is a second-level non-volatile memory (NVM) of the 2LM architecture in the second configuration.

Abstract: In an embodiment, a processor includes a plurality of cores to independently execute instructions, at least one graphics engine to independently execute graphics instructions, and a power controller including an alignment logic to cause at least one workload to be executed on a first core to be rescheduled to a different time to enable the plurality of cores to be active during an active time window and to be in a low power state during an idle time window. Other embodiments are described and claimed.

Abstract: According to one embodiment, a processor includes an instruction decoder to decode instruction and a execution unit to execute instructions, the execution unit being associated with a capture logic to periodically capture operating heuristics of the execution unit, a detection logic coupled to the execution unit to evaluate the captured operating heuristics to determine whether there is a need to adjust an operating point of the execution unit, and a control logic coupled to the detection logic and the execution unit to adjust the operating point of the execution unit based on the evaluation of the operating heuristics.

Abstract: In an embodiment, a processor includes a core to execute instructions, where the core includes a clock generation circuit to receive and distribute a first clock signal at a first operating frequency provided from a phase lock loop of the processor to a plurality of units of the core. The clock generation circuit may include a dynamic clock logic to receive a dynamic clock frequency command and to cause the clock generation circuit to distribute the first clock signal to at least one of the units at a second operating frequency. Other embodiments are described and claimed.

Abstract: I/O logic can be separated into critical and non-critical portions, with the non-critical portions being powered down during processor idle. The I/O logic is separated into gate logic and ungated logic, where the ungated logic continues to be powered during a processor deep sleep state, and the gated logic is powered off during the deep sleep state. A power control unit can trigger the shutting down of the I/O logic.

Abstract: A processor of an aspect includes a first heterogeneous physical compute element having a first set of supported instructions and architectural features, and a second heterogeneous physical compute element having a second set of supported instructions and architectural features. The second set of supported instructions and architectural features is different than the first set of supported instructions and architectural features. The processor also includes a workload and architectural state migration module coupled with the first and second heterogeneous physical compute elements. The workload and state migration module is operable to migrate a workload and associated architectural state from the first heterogeneous physical compute element to the second heterogeneous physical compute element in response to an attempt by the workload to perform at least one of an unsupported instruction and an unsupported architectural feature on the first heterogeneous physical compute element.