摘要:
A method and apparatus for optimization of multiprocessor synchronization and allocation of system management memory space is herein described. When a system management interrupt (SMI) is received, a first processor checks the state of a second processor, which may be done by checking a storage medium storing values representative of the second processor's state. The first processor handles the SMI or waits for the second processor dependent on the state of the second processor. Furthermore, system management memory is allocated where a first system management memory space assigned to a first processor overlaps a second system management memory space assigned to a second processor, leaving first and second non-overlapping region.
摘要:
A power management technique uses system management interrupt (SMI) to manage performance states of logical processors in a physical processor. Each logical processor is associated with a virtual performance state and an actual performance state. A request to retrieve or to change the virtual performance state causes the SMI to be generated. The virtual performance state is a state known to an operating system (OS). The actual performance state is a state that the logical processor is operating at.
摘要:
A power management technique uses system management interrupt (SMI) to manage states of a processor that includes multiple logical processors. When the SMI is generated, the states of logical processors are verified. When all of the logical processors are idle, the physical processor is placed in a low power state.
摘要:
A method and apparatus for efficient memory allocation and system management interrupt (SMI) handling is herein described. Upon waking a second processor in a multiple processor system, one may use a single SMI to initialize each processor, may use the location of a single default SMI handler as a wake-up vector to the second processor, and may patch an instruction pointer to a non-aligned address during the handling of the SMI with the second processor to forgo the traditional extra aligned memory allocation. In addition, one may use unified handler code to handle software generated SMIs on both the first and second processors and may use exit SMM directly after handling a hardware SMI to save execution time.
摘要:
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.
摘要:
Embodiments of a system for receiving power management guidelines from a first plurality of components of a system, and developing a power management policy to manage one or more of a second plurality of components of the system based at least in part on the received power management guidelines. Other embodiments are described.
摘要:
In some embodiments, an electronic apparatus comprises at least one processor, a plurality of components, and a policy engine comprising logic to receive latency data from one or more components in the electronic device, compute a minimum latency tolerance value from the latency data, and determine a power management policy from the minimum latency tolerance value.
摘要:
Embodiments of a system for receiving power management guidelines from a first plurality of components of a system, and developing a power management policy to manage one or more of a second plurality of components of the system based at least in part on the received power management guidelines. Other embodiments are described.
摘要:
For one disclosed embodiment, a plurality of processor cores may be on a semiconductor die. The processor cores may have at least one corresponding temperature sensor. Circuitry on the semiconductor die may generate thermal event indications based on sensed temperatures from multiple temperature sensors of multiple processor cores. A thermal event indication may indicate that a sensed temperature exceeds a temperature point. Central management logic on the semiconductor die may receive thermal event indications based on sensed temperatures from multiple temperature sensors of multiple processor cores. The central management logic may modify operation of one or more of the processor cores in response to a thermal event indication. Other embodiments are also disclosed.
摘要:
In some embodiments, a system comprises a host system comprising an industry standard interface, a peripheral device coupled to the host device via the industry standard interface, and logic in the host system to confirm that the host device supports an enhanced feature, identify at least one pin on the industry standard interface on which the enhanced feature may be implemented, enable support for the enhanced feature on the at least one pin, and route communication traffic associated with the enhanced feature to the at least one pin. Other embodiments may be described.