公开(公告)号：US07657730B2

公开(公告)日：2010-02-02

申请号：US11456102

申请日：2006-07-07

申请人： Salim Ahmed Agha ,  Gary Dean Anderson ,  Wayne Allan Britson ,  Brent William Jacobs ,  William Thomas Truskowski

发明人： Salim Ahmed Agha ,  Gary Dean Anderson ,  Wayne Allan Britson ,  Brent William Jacobs ,  William Thomas Truskowski

IPC分类号： G06F9/00 ,  G06F9/24 ,  G06F15/177

CPC分类号： G06F11/0754 ,  G06F11/0751

摘要： In response to the start of an initialization sequence at a service processor, if power to a main processor was interrupted at a most-recent time that an operating system executed on the main processor, power to the main processor is turned on, the operating system is started executing on the main processor, data from the non-volatile memory of the service processor is provided to the operating system, and the service processor is reset, which restarts the initialization sequence. If the power to the main processor was not interrupted at the most-recent time that the operating system executed on the main processor, and if the operating system is currently executing on the main processor, a monitoring function is started in the service processor, which monitors for errors at a computer system.

摘要翻译： 响应于在服务处理器处的初始化序列的开始，如果在主处理器上执行的操作系统的主要处理器的电源被接通，则主处理器的电力在最近的时间被中断，则操作系统 在主处理器上开始执行，来自服务处理器的非易失性存储器的数据被提供给操作系统，并且服务处理器被重置，其重新启动初始化序列。 如果主处理器的电源在最近的操作系统在主处理器上执行的时间没有中断，如果操作系统当前正在主处理器上执行，则在服务处​​理器中启动监视功能， 监视计算机系统的错误。

公开(公告)号：US20080010445A1

公开(公告)日：2008-01-10

申请号：US11456102

申请日：2006-07-07

申请人： Salim Ahmed Agha ,  Gary Dean Anderson ,  Wayne Allan Britson ,  Brent William Jacobs ,  William Thomas Truskowski

发明人： Salim Ahmed Agha ,  Gary Dean Anderson ,  Wayne Allan Britson ,  Brent William Jacobs ,  William Thomas Truskowski

IPC分类号： G06F15/177

CPC分类号： G06F11/0754 ,  G06F11/0751

摘要： In response to the start of an initialization sequence at a service processor, if power to a main processor was interrupted at a most-recent time that an operating system executed on the main processor, power to the main processor is turned on, the operating system is started executing on the main processor, data from the non-volatile memory of the service processor is provided to the operating system, and the service processor is reset, which restarts the initialization sequence. If the power to the main processor was not interrupted at the most-recent time that the operating system executed on the main processor, and if the operating system is currently executing on the main processor, a monitoring function is started in the service processor, which monitors for errors at a computer system.

摘要翻译： 响应于在服务处理器处的初始化序列的开始，如果在主处理器上执行的操作系统的主要处理器的电源被接通，则主处理器的电力在最近的时间被中断，则操作系统 在主处理器上开始执行，来自服务处理器的非易失性存储器的数据被提供给操作系统，并且服务处理器被重置，其重新启动初始化序列。 如果主处理器的电源在最近的操作系统在主处理器上执行的时间没有中断，如果操作系统当前正在主处理器上执行，则在服务处​​理器中启动监视功能， 监视计算机系统的错误。

公开(公告)号：US07536493B2

公开(公告)日：2009-05-19

申请号：US11279436

申请日：2006-04-12

申请人： Gary Dean Anderson ,  Joseph Donald Armstrong ,  Brent William Jacobs ,  Ajay Kumar Mahajan

发明人： Gary Dean Anderson ,  Joseph Donald Armstrong ,  Brent William Jacobs ,  Ajay Kumar Mahajan

IPC分类号： G06F13/14

CPC分类号： G06F21/31 ,  G06F21/71 ,  G06F21/73 ,  G06F2221/2129

摘要： The aspects of the present invention provide a computer implemented method, an apparatus, and a computer usable program code for identifying a service processor with current setting information. First stored information for a service processor is retrieved from a first memory in the first service processor. Second stored information for the service processor is retrieved from a second memory in a data processing system to which the service processor is connected. The first stored information is compared with the second stored information to form a comparison as to whether the service processor was previously connected to the data processing system and to determine whether the service processor was a last service processor to fully communicate with system firmware in the data processing system. The service processor has current setting information if the service processor was previously connected to the data processing system and was the last service processor to fully communicate with the system firmware.

摘要翻译： 本发明的方面提供了一种计算机实现的方法，装置和用于使用当前设置信息识别服务处理器的计算机可用程序代码。 从第一服务处理器中的第一存储器检索用于服务处理器的第一存储信息。 从与服务处理器连接的数据处理系统中的第二存储器检索用于服务处理器的第二存储信息。 将第一存储信息与第二存储信息进行比较以形成关于服务处理器是否先前连接到数据处理系统并且确定服务处理器是否是最后服务处理器以与数据中的系统固件完全通信的比较 处理系统。 如果服务处理器先前连接到数据处理系统，并且是与系统固件完全通信的最后一个服务处理器，则服务处理器具有当前设置信息。

公开(公告)号：US08245233B2

公开(公告)日：2012-08-14

申请号：US12335690

申请日：2008-12-16

申请人： Anis M. Abdul ,  Brent William Jacobs ,  Ajay Kumar Mahajan ,  Atit D. Patel

发明人： Anis M. Abdul ,  Brent William Jacobs ,  Ajay Kumar Mahajan ,  Atit D. Patel

IPC分类号： G06F9/46

CPC分类号： G06F9/5011

摘要： A method, a system and a computer program product for selecting a primary controller for a server system based on the services offered by each controller. A primary controller designator (PCD) utility determines the relative importance of a controller based upon the services provided by the controller and the weighted importance assigned to these services. The PCD utility classifies the services provided by a system-controller according to the following: (1) the number of OS partitions a system-controller is able to communicate with; and (2) the number of hardware devices that a controller has access to. The importance of the services is determined by the host OS partition information and the degree of importance of a partition that utilizes/requires the particular service(s). The PCD utility designates a controller as a “Primary” if the designated “Primary” is capable of providing services that are required for the most important OS partitions, according to the classification of controller services.

摘要翻译： 一种用于基于由每个控制器提供的服务来选择用于服务器系统的主控制器的方法，系统和计算机程序产品。 主控制器指示符（PCD）实用程序根据控制器提供的服务和分配给这些服务的加权重要性来确定控制器的相对重要性。 PCD实用程序根据以下内容对系统控制器提供的服务进行分类：（1）系统控制器能够通信的OS分区数; 和（2）控制器访问的硬件设备的数量。 服务的重要性由主机OS分区信息和利用/要求特定服务的分区的重要程度决定。 如果指定的“主”能够提供最重要的OS分区所需的服务，则根据控制器服务的分类，PCD实用程序将控制器指定为“主”。

公开(公告)号：US07941652B2

公开(公告)日：2011-05-10

申请号：US12110451

申请日：2008-04-28

申请人： Kraig Allan Bottemiller ,  Brent William Jacobs ,  James Albert Pieterick

发明人： Kraig Allan Bottemiller ,  Brent William Jacobs ,  James Albert Pieterick

IPC分类号： G06F9/00

CPC分类号： G06F11/3636

摘要： A method, apparatus and computer program product are provided for implementing atomic data tracing in a processor system including an auxiliary processor unit (APU) coupled to a central processor unit (CPU). The auxiliary processor unit (APU) processes a trace instruction. When a trace instruction is identified by the APU, the APU signals the CPU with a pipeline stall signal for stalling the CPU and checks for an enabled trace engine as specified by the trace instruction. When the trace engine for the trace instruction is enabled, then the trace data is written into a trace buffer. The APU signals the CPU with an op done signal for allowing the CPU to continue with instruction processing.

摘要翻译： 提供了一种用于在包括耦合到中央处理器单元（CPU）的辅助处理器单元（APU）的处理器系统中实现原子数据跟踪的方法，装置和计算机程序产品。 辅助处理器单元（APU）处理跟踪指令。 当APU识别跟踪指令时，APU用流水线停止信号通知CPU，以阻止CPU并检查跟踪指令指定的启用跟踪引擎。 当跟踪指令的跟踪引擎被使能时，跟踪数据被写入跟踪缓冲区。 APU通过操作完成信号通知CPU，以允许CPU继续进行指令处理。

公开(公告)号：US07996359B2

公开(公告)日：2011-08-09

申请号：US12139537

申请日：2008-06-16

申请人： Nicholas Anthony Pietraniec ,  Ajay Kumar Mahajan ,  Anis M Abdul ,  Brent William Jacobs

发明人： Nicholas Anthony Pietraniec ,  Ajay Kumar Mahajan ,  Anis M Abdul ,  Brent William Jacobs

IPC分类号： G06F17/00

CPC分类号： G06F17/30575

摘要： A method, system, and computer usable program product for managing multi-node multi-version systems are provided in the illustrative embodiments. A process associates a version identifier with a first version of data available at a first node in the multi-node multi-version system. The version identifier corresponds to the first version of the data. The data includes an update. The process sends the data and the associated version identifier to a second node without learning a second version of data available at the second node. In one embodiment, the first version of data may be a subset of the second version of data. In another embodiment the second version of data may be a subset of the first version of data. The process repeats the sending for each update at the first node.

摘要翻译： 在说明性实施例中提供了用于管理多节点多版本系统的方法，系统和计算机可用程序产品。 过程将版本标识符与在多节点多版本系统中的第一节点处可用的第一版本的数据相关联。 版本标识符对应于数据的第一个版本。 数据包括更新。 该过程将数据和相关联的版本标识符发送到第二节点而不学习在第二节点处可用的第二版本的数据。 在一个实施例中，数据的第一版本可以是数据的第二版本的子集。 在另一个实施例中，数据的第二版本可以是数据的第一版本的子集。 该过程重复发送第一个节点上的每个更新。

公开(公告)号：US20080201615A1

公开(公告)日：2008-08-21

申请号：US12110451

申请日：2008-04-28

申请人： Kraig Allan Bottemiller ,  Brent William Jacobs ,  James Albert Pieterick

发明人： Kraig Allan Bottemiller ,  Brent William Jacobs ,  James Albert Pieterick

IPC分类号： G06F11/22

CPC分类号： G06F11/3636

摘要： A method, apparatus and computer program product are provided for implementing atomic data tracing in a processor system including an auxiliary processor unit (APU) coupled to a central processor unit (CPU). The auxiliary processor unit (APU) processes a trace instruction. When a trace instruction is identified by the APU, the APU signals the CPU with a pipeline stall signal for stalling the CPU and checks for an enabled trace engine as specified by the trace instruction. When the trace engine for the trace instruction is enabled, then the trace data is written into a trace buffer. The APU signals the CPU with an op done signal for allowing the CPU to continue with instruction processing.

摘要翻译： 提供了一种用于在包括耦合到中央处理器单元（CPU）的辅助处理器单元（APU）的处理器系统中实现原子数据跟踪的方法，装置和计算机程序产品。 辅助处理器单元（APU）处理跟踪指令。 当APU识别跟踪指令时，APU用流水线停止信号通知CPU，以阻止CPU并检查跟踪指令指定的启用跟踪引擎。 当跟踪指令的跟踪引擎被使能时，跟踪数据被写入跟踪缓冲区。 APU通过操作完成信号通知CPU，以允许CPU继续进行指令处理。

公开(公告)号：US20090313257A1

公开(公告)日：2009-12-17

申请号：US12139537

申请日：2008-06-16

申请人： Nicholas Anthony Pietraniec ,  Ajay Kumar Mahajan ,  Anis M. Abdul ,  Brent William Jacobs

发明人： Nicholas Anthony Pietraniec ,  Ajay Kumar Mahajan ,  Anis M. Abdul ,  Brent William Jacobs

IPC分类号： G06F17/30 ,  G06F7/00

CPC分类号： G06F17/30575

摘要： A method, system, and computer usable program product for managing multi-node multi-version systems are provided in the illustrative embodiments. A process associates a version identifier with a first version of data available at a first node in the multi-node multi-version system. The version identifier corresponds to the first version of the data. The data includes an update. The process sends the data and the associated version identifier to a second node without learning a second version of data available at the second node. In one embodiment, the first version of data may be a subset of the second version of data. In another embodiment the second version of data may be a subset of the first version of data. The process repeats the sending for each update at the first node.

摘要翻译： 在说明性实施例中提供了用于管理多节点多版本系统的方法，系统和计算机可用程序产品。 过程将版本标识符与在多节点多版本系统中的第一节点处可用的第一版本的数据相关联。 版本标识符对应于数据的第一个版本。 数据包括更新。 该过程将数据和相关联的版本标识符发送到第二节点而不学习在第二节点处可用的第二版本的数据。 在一个实施例中，数据的第一版本可以是数据的第二版本的子集。 在另一个实施例中，数据的第二版本可以是数据的第一版本的子集。 该过程重复发送第一个节点上的每个更新。

公开(公告)号：US07383428B2

公开(公告)日：2008-06-03

申请号：US10659975

申请日：2003-09-11

申请人： Kraig Allan Bottemiller ,  Brent William Jacobs ,  James Albert Pieterick

发明人： Kraig Allan Bottemiller ,  Brent William Jacobs ,  James Albert Pieterick

IPC分类号： G06F9/00

CPC分类号： G06F11/3636

摘要： A method, apparatus and computer program product are provided for implementing atomic data tracing in a processor system including an auxiliary processor unit (APU) coupled to a central processor unit (CPU). The auxiliary processor unit (APU) processes a trace instruction. When a trace instruction is identified by the APU, the APU signals the CPU with a pipeline stall signal for stalling the CPU and checks for an enabled trace engine as specified by the trace instruction. When the trace engine for the trace instruction is enabled, then the trace data is written into a trace buffer. The APU signals the CPU with an op done signal for allowing the CPU to continue with instruction processing.

摘要翻译： 提供了一种用于在包括耦合到中央处理器单元（CPU）的辅助处理器单元（APU）的处理器系统中实现原子数据跟踪的方法，装置和计算机程序产品。 辅助处理器单元（APU）处理跟踪指令。 当APU识别跟踪指令时，APU用流水线停止信号通知CPU，以阻止CPU并检查跟踪指令指定的启用跟踪引擎。 当跟踪指令的跟踪引擎被使能时，跟踪数据被写入跟踪缓冲区。 APU通过操作完成信号通知CPU，以允许CPU继续进行指令处理。

公开(公告)号：US07266083B2

公开(公告)日：2007-09-04

申请号：US10374651

申请日：2003-02-26

申请人： Michael Joseph Carnevale ,  Charles Scott Graham ,  Brent William Jacobs ,  Daniel Frank Moertl ,  Timothy Jerry Schimke ,  Lee Anton Sendelbach

发明人： Michael Joseph Carnevale ,  Charles Scott Graham ,  Brent William Jacobs ,  Daniel Frank Moertl ,  Timothy Jerry Schimke ,  Lee Anton Sendelbach

IPC分类号： G01R31/08

CPC分类号： H04L43/0811 ,  H04L41/0663

摘要： A method, apparatus and computer program product are provided for implementing queue pair connection protection over an interconnect network, such as InfiniBand. A message packet is received for a queue pair (QP) and the QP is checked for an imminent connection failure. Responsive to identifying an imminent connection failure, a special message processing mode is established for the QP. After the special message processing mode is established, packets of the message are received without establishing a message queue entry and without storing packet data.

摘要翻译： 提供了一种用于在互连网络（例如InfiniBand）上实现队列对连接保护的方法，装置和计算机程序产品。 接收到队列对（QP）的消息分组，并检查QP是否即将发生连接故障。 响应于识别即将发生的连接故障，为QP建立了特殊的消息处理模式。 在建立特殊消息处理模式之后，接收到消息的报文，而不建立消息队列条目，不存储分组数据。