METHOD FOR MIGRATING CPU STATE FROM AN INOPERABLE CORE TO A SPARE CORE

    公开(公告)号:US20170293539A1

    公开(公告)日:2017-10-12

    申请号:US15632567

    申请日:2017-06-26

    Abstract: An apparatus is disclosed in which the apparatus may include a plurality of cores, including a first core, a second core and a third core, and circuitry coupled to the first core. The first core may be configured to process a plurality of instructions. The circuitry may be may be configured to detect that the first core stopped committing a subset of the plurality of instructions, and to send an indication to the second core that the first core stopped committing the subset. The second core may be configured to disable the first core from further processing instructions of the subset responsive to receiving the indication, and to copy data from the first core to a third core responsive to disabling the first core. The third core may be configured to resume processing the subset dependent upon the data.

    METHOD AND APPARATUS FOR SYNCHRONIZING THE TIME REFERENCE OF A DYNAMICALLY ACTIVATED PROCESSOR TO THE SYSTEM TIME REFERENCE
    3.
    发明申请
    METHOD AND APPARATUS FOR SYNCHRONIZING THE TIME REFERENCE OF A DYNAMICALLY ACTIVATED PROCESSOR TO THE SYSTEM TIME REFERENCE 有权
    用于将动态加工器的时间参考同步到系统时间参考的方法和装置

    公开(公告)号:US20140143580A1

    公开(公告)日:2014-05-22

    申请号:US13679690

    申请日:2012-11-16

    Inventor: Ali Vahidsafa

    CPC classification number: G06F1/12

    Abstract: Implementations of the present disclosure involve an apparatus and/or method for synchronizing at least one newly activated processor with at least one previously running processor. Each processor is configured to generate a heartbeat and operate according to a STICK. When a previously deactivated processor is added, the heartbeat of each active processor is reset and the current STICK is transmitted to the newly activated processor on the next heartbeat. The newly activated processor may then add the heartbeat period to the acquired STICK and begin incrementing the STICK and normal operation after the next heartbeat.

    Abstract translation: 本公开的实现涉及用于将至少一个新激活的处理器与至少一个先前运行的处理器同步的装置和/或方法。 每个处理器配置为产生心跳并根据STICK进行操作。 当添加先前停用的处理器时,每个活动处理器的心跳被复位,并且当前的STICK在下一个心跳上传输到新激活的处理器。 然后,新激活的处理器可以将心跳周期添加到所获取的STICK,并且在下一个心跳之后开始递增STICK和正常操作。

    COMBO DYNAMIC FLOP WITH SCAN
    4.
    发明申请
    COMBO DYNAMIC FLOP WITH SCAN 有权
    COMBO动态游戏与扫描

    公开(公告)号:US20140136912A1

    公开(公告)日:2014-05-15

    申请号:US13673503

    申请日:2012-11-09

    CPC classification number: G01R31/318541

    Abstract: A combo dynamic flop with scan flop circuit includes a flip-flop circuit, a scan control circuit, and an output buffer circuit. The flip-flop circuit includes a dynamic latch circuit and a static latch circuit. The dynamic latch circuit includes a dynamic latch storage node. The static latch circuit includes a static storage node driven by the dynamic latch. The scan control circuit includes a scan slave feed-forward circuit, a scan latch circuit, and a scan driver circuit driven by the scan feed-back circuit. The scan latch circuit includes a scan feed-back circuit, a scan storage node, and a scan feed-forward circuit driven from the static latch. The output buffer circuit includes a dynamic latch driver driven from the dynamic latch circuit and a static driver driven from the static latch circuit.

    Abstract translation: 具有扫描电路的组合动态触发器包括触发器电路,扫描控制电路和输出缓冲电路。 触发器电路包括动态锁存电路和静态锁存电路。 动态锁存电路包括动态锁存存储节点。 静态锁存电路包括由动态锁存器驱动的静态存储节点。 扫描控制电路包括扫描从机前馈电路,扫描锁存电路和由扫描反馈电路驱动的扫描驱动电路。 扫描锁存电路包括扫描反馈电路,扫描存储节点和从静态锁存器驱动的扫描前馈电路。 输出缓冲电路包括从动态锁存电路驱动的动态锁存驱动器和从静态锁存电路驱动的静态驱动器。

    Rotational synchronizer circuit for metastablity resolution
    6.
    发明授权
    Rotational synchronizer circuit for metastablity resolution 有权
    旋转同步电路,用于转移分辨率

    公开(公告)号:US09509317B2

    公开(公告)日:2016-11-29

    申请号:US13755056

    申请日:2013-01-31

    CPC classification number: H03L7/00 H04L7/005 H04L7/02 H04L25/05

    Abstract: A rotational synchronizer for metastability resolution is disclosed. A synchronizer includes a plurality of M+1 latches each coupled to receive data through a common data input. The synchronizer further includes a multiplexer having a N inputs each coupled to receive data from an output of a corresponding one of the M+1 latches, and an output, wherein the multiplexer is configured to select one of its inputs to be coupled to the output. A control circuit is configured to cause the multiplexer to sequentially select outputs of the M+1 latches responsive to N successive clock pulses, and further configured to cause the M+1 latches to sequentially latch data received through the common data input.

    Abstract translation: 公开了一种用于亚稳定性分辨率的旋转同步器。 同步器包括多个M + 1个锁存器,每个锁存器被耦合以通过公共数据输入来接收数据。 所述同步器还包括多路复用器,其具有N个输入端,每个输入端分别被耦合以从所述M + 1锁存器中对应的一个锁存器的输出端接收数据,以及输出端,其中,所述多路复用器被配置为选择其输入之一耦合到所述输出端 。 控制电路被配置为使得多路复用器响应于N个连续时钟脉冲顺序地选择M + 1个锁存器的输出,并且还被配置为使得M + 1锁存器顺序地锁存通过公共数据输入接收到的数据。

    Mixing of low speed and high speed clocks to improve test precision of a digital integrated circuit
    7.
    发明授权
    Mixing of low speed and high speed clocks to improve test precision of a digital integrated circuit 有权
    混合低速和高速时钟,提高数字集成电路的测试精度

    公开(公告)号:US09404967B2

    公开(公告)日:2016-08-02

    申请号:US14535647

    申请日:2014-11-07

    Inventor: Ali Vahidsafa

    Abstract: Implementations of the present disclosure involve an apparatus and/or method for mixing high speed and low speed clock signals during structural testing of a digital integrated circuit to improve the test precision and efficiency. In particular, the apparatus and/or method allow for a testing device to perform stuck-bit testing of the circuit by releasing one or more clock cycles of a low speed clock signal. Further, without having to reset the testing of the circuit, at-speed testing of the circuit may be conducted by the testing device. In one embodiment, at-speed testing occurs by activating a mode signal associated with the circuit design that instructs one or more clock cycles from an internal clock signal to the circuit to be released. The testing device may return to stuck-bit testing at a low speed clock signal, or continue with at-speed testing using the high speed internal clock signal.

    Abstract translation: 本公开的实现涉及用于在数字集成电路的结构测试期间混合高速和低速时钟信号以提高测试精度和效率的装置和/或方法。 具体地,该装置和/或方法允许测试装置通过释放低速时钟信号的一个或多个时钟周期来执行电路的卡位测试。 此外,不必重置电路的测试,电路的高速测试可以由测试装置进行。 在一个实施例中,通过激活与电路设计相关联的模式信号来进行速度测试,其指示从内部时钟信号到要释放的电路的一个或多个时钟周期。 测试设备可以以低速时钟信号返回卡位测试,或使用高速内部时钟信号继续进行高速测试。

    INTER-PROCESSOR BUS LINK AND SWITCH CHIP FAILURE RECOVERY
    8.
    发明申请
    INTER-PROCESSOR BUS LINK AND SWITCH CHIP FAILURE RECOVERY 有权
    处理器总线链路和切换芯片故障恢复

    公开(公告)号:US20160210255A1

    公开(公告)日:2016-07-21

    申请号:US14598640

    申请日:2015-01-16

    CPC classification number: G06F13/4022 G06F11/221 H04L12/00 H04L12/4625

    Abstract: A system is disclosed in which the system may include multiple bus switches, and multiple processors. Each processor may be coupled to each bus switch. Each processor may be configured to initiate a transfer of data to a given bus switch, and detect if a respective link to the given bus switch is inoperable. In response to detecting an inoperable link to a first bus switch, a given processor may be further configured to send a notification message to at least one other processor via at least a second bus switch and to remove routing information corresponding to the inoperable link from a first register. The at least one other processor may be configured to remove additional routing information corresponding to the inoperable link from a second register in response to receiving the notification message from the given processor.

    Abstract translation: 公开了一种系统,其中系统可以包括多个总线开关和多个处理器。 每个处理器可以耦合到每个总线开关。 每个处理器可以被配置为启动数据到给定总线开关的传输,并检测到给定总线开关的相应链路是否不可操作。 响应于检测到不可操作的链接到第一总线交换机,给定处理器可以进一步被配置为经由至少第二总线交换机向至少一个其他处理器发送通知消息,并且从一个第二总线交换机去除对应于不可操作链路的路由信息 首先注册 响应于从给定处理器接收到通知消息,至少一个其他处理器可以被配置为从第二寄存器去除对应于不可操作链路的附加路由信息。

    DRIFT COMPENSATION FOR A REAL TIME CLOCK CIRCUIT
    9.
    发明申请
    DRIFT COMPENSATION FOR A REAL TIME CLOCK CIRCUIT 有权
    实时时钟电路干扰补偿

    公开(公告)号:US20160195894A1

    公开(公告)日:2016-07-07

    申请号:US14590748

    申请日:2015-01-06

    Inventor: Ali Vahidsafa

    Abstract: Implementations of the present disclosure involve an apparatus and/or method for adjusting a counter in a computing system to account for drift of the counter value over time compared to another counter of the system. In particular, a processor of the computing system that includes a local counter component may access a counter component of another processor of the system, referred to as the reference counter. By comparing the value of the reference counter to the local counter, the processor may determine any drift that may have occurred over a period of time in the local counter. The calculated drift, or counter error, may be converted into one or more adjustments to the local counter to synchronize the local counter with the reference counter. In one embodiment, the adjustment to the local counter includes increasing the rate at which the local counter is incremented for a period of time.

    Abstract translation: 本公开的实现涉及用于在计算系统中调整计数器的装置和/或方法,以考虑与系统的另一个计数器相比计算值随时间的漂移。 特别地,包括本地计数器组件的计算系统的处理器可以访问被称为参考计数器的系统的另一个处理器的计数器组件。 通过将参考计数器的值与本地计数器进行比较,处理器可以确定本地计数器中可能在一段时间内发生的漂移。 计算的漂移或计数器错误可以转换为本地计数器的一个或多个调整,以使本地计数器与参考计数器同步。 在一个实施例中,对本地计数器的调整包括增加本地计数器递增一段时间的速率。

    METHOD FOR MIGRATING CPU STATE FROM AN INOPERABLE CORE TO A SPARE CORE
    10.
    发明申请
    METHOD FOR MIGRATING CPU STATE FROM AN INOPERABLE CORE TO A SPARE CORE 有权
    将CPU状态从不可操作的核心迁移到备用核心的方法

    公开(公告)号:US20160147534A1

    公开(公告)日:2016-05-26

    申请号:US14549742

    申请日:2014-11-21

    Abstract: An apparatus is disclosed in which the apparatus may include a plurality of cores, including a first core, a second core and a third core, and circuitry coupled to the first core. The first core may be configured to process a plurality of instructions. The circuitry may be may be configured to detect that the first core stopped committing a subset of the plurality of instructions, and to send an indication to the second core that the first core stopped committing the subset. The second core may be configured to disable the first core from further processing instructions of the subset responsive to receiving the indication, and to copy data from the first core to a third core responsive to disabling the first core. The third core may be configured to resume processing the subset dependent upon the data.

    Abstract translation: 公开了一种装置,其中装置可以包括多个芯,包括第一芯,第二芯和第三芯,以及耦合到第一芯的电路。 第一核可以被配置为处理多个指令。 电路可以被配置为检测第一核心停止提交多个指令的子集,并且向第二核心发送指示第一核心停止提交子集的指示。 第二核心可以被配置为响应于接收到指示而禁止第一核心进一步处理该子集的指令,并且响应于禁用第一核心将数据从第一核心复制到第三核心。 第三核可以被配置为依赖于数据来恢复处理该子集。

Patent Agency Ranking