Increasing turbo mode residency of a processor
    2.
    发明授权
    Increasing turbo mode residency of a processor 有权
    增加处理器的turbo模式驻留

    公开(公告)号:US09032125B2

    公开(公告)日:2015-05-12

    申请号:US13780075

    申请日:2013-02-28

    Abstract: In one embodiment, the present invention includes a method for accessing a task stored in an entry of a task queue that identifies the task and a first core of a processor on which the task has been scheduled, reassigning the task to a coldest idle core of the processor, and sending the task to the coldest idle core and maintaining the processor in a turbo mode. Other embodiments are described and claimed.

    Abstract translation: 在一个实施例中,本发明包括一种用于访问存储在任务队列的条目中的任务的方法,所述任务识别任务,并且处理器的第一核心已经被安排在其上,将任务重新分配到最冷空闲核心 处理器,并将任务发送到最冷的空闲核心并将处理器维持在turbo模式。 描述和要求保护其他实施例。

    Apparatus, method, and system for early deep sleep state exit of a processing element

    公开(公告)号:US09720488B2

    公开(公告)日:2017-08-01

    申请号:US15277944

    申请日:2016-09-27

    Abstract: An apparatus and method is described herein for providing an early wake scheme before spawning a new thread. An early wake indication may be provided an amount of time, which may include an amount of time to perform a demotion from a current power state to a lower power state that is closer to an active power state, before a new thread is to be spawned and executed on a processing element (e.g., core or thread). Upon encountering the spawn of the new thread, such as a helper thread, the processing element may further transition from the lower power state to an active power state. The new thread may be executed on the processing element without incurring the latency associated with execution of the new thread waiting for the demotion from the current power state to an active power state after the spawn of the new thread.

    Protect non-memory encryption engine (non-mee) metadata in trusted execution environment

    公开(公告)号:US10031861B2

    公开(公告)日:2018-07-24

    申请号:US14865304

    申请日:2015-09-25

    Abstract: A server, processing device and/or processor includes a processing core and a memory controller, operatively coupled to the processing core, to access data in an off-chip memory. A memory encryption engine (MEE) may be operatively coupled to the memory controller and the off-chip memory. The MEE may store non-MEE metadata bits within a modified version line corresponding to ones of a plurality of data lines stored in a protected region of the off-chip memory, compute an embedded message authentication code (eMAC) using the modified version line, and detect an attempt to modify one of the non-MEE metadata bits by using the eMAC within a MEE tree walk to authenticate access to the plurality of data lines. The non-MEE metadata bits may store coherence bits that track changes to a cache line in a remote socket, poison bits that track error containment within the data lines, and possibly other metadata bits.

    Cache and data organization for memory protection

    公开(公告)号:US10185842B2

    公开(公告)日:2019-01-22

    申请号:US14661044

    申请日:2015-03-18

    Abstract: This disclosure is directed to cache and data organization for memory protection. Memory protection operations in a device may be expedited by organizing cache and/or data structure while providing memory protection for encrypted data. An example device may comprise processing module and a memory module. The processing module may include a memory encryption engine (MEE) to decrypt encrypted data loaded from the memory module, or to encrypt plaintext data prior to storage in the memory module, using security metadata also stored in the memory module. Example security metadata may include version (VER) data, memory authentication code (MAC) data and counter data. Consistent with the present disclosure, a cache associated with the MEE may be partitioned to separate the VER and MAC data from counter data. Data organization may comprise including the VER and MAC data corresponding to particular data in the same data line.

Patent Agency Ranking