Abstract:
PROBLEM TO BE SOLVED: To provide systems, methods, and apparatuses for virtual to physical address translation with support for page attributes.SOLUTION: A system receives an instruction to translate a virtual memory pointer to a physical memory address for a memory location; in address translation hardware, translates the virtual memory pointer to the physical memory address on the basis of page table information to produce a partial result of the instruction; and, as the result of the instruction, loads the physical memory address and one or more page attributes into one or more processor registers without accessing to a memory hierarchy.
Abstract:
PROBLEM TO BE SOLVED: To improve performance for processing of a task and an event in a multi-core processor.SOLUTION: A device 160 comprises: a plurality of processor elements 131 to 134; a task routing logic 130; and a turbo mode logic 140 which controls the processor elements and operates at least one active processor element with a further high frequency when at least one other processor element is idle. When at least one of the processor elements is in a turbo mode, the task routing logic selects a processor element for executing a task 101 on the basis of at least comparison of performance losses.
Abstract:
PROBLEM TO BE SOLVED: To return a physical memory address about a linear address given as an operand without typically giving additional information in the conventional address translation instruction. SOLUTION: This embodiment relates to a system, method and device for translating from a linear address to a physical address, with support for page attribute. In some embodiments, the system receives an instruction to translate a memory point to a physical memory address about a certain memory position. The system can return the physical memory address to one or a plurality of page attributes. Other embodiments are described and patent claim is made. COPYRIGHT: (C)2009,JPO&INPIT
Abstract:
PROBLEM TO BE SOLVED: To provide systems, methods, and apparatus for virtual to physical address translation with support for page attributes.SOLUTION: The methods include: a step 502 of receiving an instruction to translate a virtual memory pointer to a physical memory address for a memory location; a step 504 of translating the virtual memory pointer into the physical memory address based on page table information; and a step 506 of returning the physical memory address and one or more page attributes.
Abstract:
PROBLEM TO BE SOLVED: To improve the performance for task and event processing in a multicore processor. SOLUTION: The device 160 includes a plurality of processor elements 131-134; a task routing logic 130; and a turbo mode logic 140 which controls the processor elements and operates at least one active processor element with a further high frequency when at least one other processor element is idle. When at least one of the processor elements is in a turbo mode, the task routing logic selects a processor element for executing a task 101 based on at least a comparison of performance losses. COPYRIGHT: (C)2011,JPO&INPIT
Abstract:
Ein System zur dynamischen Leistungsversorgungsschienenumschaltung (DPRS), das eine Mehrfachschienen-Leistungsversorgung umfasst. Die Mehrfachschienen-Leistungsversorgung umfasst eine Hauptschiene und eine Bereitschaftsschiene. Das System zur DPRS umfasst außerdem einen Speicher, der dazu dient, Befehle zu speichern, und kommunikativ mit der Mehrfachschienen-Leistungsversorgung gekoppelt ist. Das System zur DPRS umfasst außerdem einen Prozessor, der kommunikativ mit dem Speicher und der Mehrfachschienen-Leistungsversorgung gekoppelt ist. Wenn der Prozessor Befehle ausführen soll, stellt die Mehrfachschienen-Leistungsversorgung auch Leistung für das System bereit, und als Antwort darauf, dass eine Eintrittsbedingung erfüllt wird, nimmt sie Leistung von der Hauptschiene weg und lässt die Bereitschaftsschiene auf EIN geschaltet. Als Antwort darauf, dass eine Austrittsbedingung erfüllt wird, wird außerdem die Hauptschiene eingeschaltet und beginnt wieder, das System mit Leistung zu versorgen.
Abstract:
In one embodiment, the present invention includes an instruction decoder that can receive an incoming instruction and a path select signal and decode the incoming instruction into a first instruction code or a second instruction code responsive to the path select signal. The two different instruction codes, both representing the same incoming instruction may be used by an execution unit to perform an operation optimized for different data lengths. Other embodiments are described and claimed.
Abstract:
A processor has an instruction to translate a virtual memory pointer to a physical memory address. The instruction returns the address and attributes associated with the page corresponding to the address. The translation may use a translation look aside buffer (TLB) or a page table. It may use the page table if the address is not in the TLB. The instruction may be used by a virtual machine manager to produce a host physical address. It may be used by a guest operating system to produce a guest physical memory address. The instruction may have an operand to define the context of the translation. The returned address may be stored in a dedicated register.
Abstract:
Systeme und Verfahren können die Aggregierung einer ersten Leerlaufzeit von einem ersten, mit der Plattform verbundenen Gerät und einer zweiten Leerlaufzeit von einem zweiten, mit der Plattform verbundenen Gerät bereitstellen. Darüber hinaus kann eine Leerlaufzeit für die Plattform basierend zumindest teilweise auf der ersten Leerlaufzeit und der zweiten Leerlaufzeit ausgewählt werden. In einem Beispiel sind die Leerlaufzeiten als deterministisch, geschätzt oder statistisch eingestuft.