公开(公告)号：JP2008102929A

公开(公告)日：2008-05-01

申请号：JP2007267426

申请日：2007-10-15

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： SALZBERG CLAUDIA ANDREA ,  RENATO J RECIO ,  THURBER STEVEN MARK ,  MOHRTUL DANIEL FRANK

IPC: G06F13/10

CPC classification number: G06F13/102

Abstract: PROBLEM TO BE SOLVED: To provide an apparatus and method for communicating with a network adapter using queue data structure and cached address translation. SOLUTION: With the apparatus and method, a device driver calls a device driver service for address translation in a route complex related to a queue data structure and initialization of a protection table (ATPT) item. The device driver service returns an untranslated address to the device driver, and the untranslated address is provided to the network adapter. In response to retrieving a queue element from the queue data structure, the network adapter requests translation of an untranslated address specified by the queue element, and the translated address is stored in the network adapter prior to receiving a data packet targeting a buffer related to the queue element. COPYRIGHT: (C)2008,JPO&INPIT

Abstract translation: 要解决的问题：提供一种使用队列数据结构和缓存地址转换与网络适配器进行通信的装置和方法。 解决方案：使用设备和方法，设备驱动程序在与队列数据结构相关的路由复合和保护表（ATPT）项的初始化中调用设备驱动程序服务进行地址转换。 设备驱动程序服务向设备驱动程序返回未翻译的地址，并将非翻译地址提供给网络适配器。 响应于从队列数据结构检索队列元素，网络适配器请求对由队列元素指定的非翻译地址的转换，并且将转换的地址存储在网络适配器中，然后再接收到针对与 队列元素 版权所有（C）2008，JPO＆INPIT

公开(公告)号：JP2002032327A

公开(公告)日：2002-01-31

申请号：JP2001159963

申请日：2001-05-29

Applicant: IBM

Inventor： BUCKLAND PAT ALLEN ,  NEAL DANIEL MARVIN ,  THURBER STEVEN MARK

IPC: G06F13/36 ,  G06F11/34 ,  G06F13/00

Abstract: PROBLEM TO BE SOLVED: To provide an I/O bus bridge having a performance monitoring function. SOLUTION: In this bus bridge 200 used for a data processing system, the bridge includes a primary bus interface 206 connected to a primary bus, a secondary bus interface 208 connected to a secondary bus 204, a performance- monitoring register 218 and a state machine which is connected to the primary and secondary bus interfaces and records the occurrence of an event designated in the register 218. The register 218 receives posted memory write and a read request in disconnection and division response and monitors various event occurrences, such as a retry according to the mode in which the bridge is an operation target or a master.

公开(公告)号：JP2008152787A

公开(公告)日：2008-07-03

申请号：JP2007324711

申请日：2007-12-17

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： SALZBERG CLAUDIA ANDREA ,  RENATO J RECIO ,  THURBER STEVEN MARK ,  FREIMUTH DOUGLAS M ,  VARGAS JACOBO A

IPC: G06F13/36

CPC classification number: G06F13/4022

Abstract: PROBLEM TO BE SOLVED: To provide a system and method for hot plug/remove of PCIe fabric, etc. SOLUTION: The addition of a new component in the fabric is detected and an event is sent to a multiple root fabric configuration manager. The multiple root fabric configuration manager gathers information about the new component and updates its I/O component tree structure in its configuration data structure to include the new component. The new component may then be utilized via the updated configuration data structure. When a component is to be removed, the multiple root fabric configuration manager receives an event indicating the component to be removed, determines which branches of the tree structure are affected by the removal, and updates its configuration data structure accordingly to remove the component and its associated components from the virtual plane of the removed component. COPYRIGHT: (C)2008,JPO&INPIT

Abstract translation: 要解决的问题：提供用于热插拔/除去PCIe结构等的系统和方法。解决方案：检测到在结构中添加新组件并将事件发送到多根 面料配置经理。 多根结构配置管理器收集有关新组件的信息，并在其配置数据结构中更新其I / O组件树结构以包含新组件。 然后可以通过更新的配置数据结构来利用新的组件。 当要除去组件时，多根结构配置管理器将接收到一个表示要删除的组件的事件，确定树结构的哪些分支受到删除的影响，并相应地更新其配置数据结构以删除该组件及其组件 已删除组件的虚拟平面中的关联组件。 版权所有（C）2008，JPO＆INPIT

公开(公告)号：JP2008102928A

公开(公告)日：2008-05-01

申请号：JP2007267404

申请日：2007-10-15

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： SALZBERG CLAUDIA ANDREA ,  RENATO J RECIO ,  THURBER STEVEN MARK ,  MOHRTUL DANIEL FRANK

IPC: G06F13/14 ,  G06F13/10 ,  G06F13/28

CPC classification number: G06F12/1081

Abstract: PROBLEM TO BE SOLVED: To provide an apparatus and method for communicating with an input/output device or an end point using a queue data structure and pre-translated address related to the queue data structure. SOLUTION: A device driver service is called for address translations in a route complex related to a queue data structure and initialization of protection table item by a device driver, and an untranslated address is returned to the device driver. The untranslated address is next provided for the input/output device or the end point. Subsequently, the input/output device or end point requests translation of these untranslated addresses, and the address is stored in the input/output device or end point before receiving the input/output operation targeting the queue data structure. The cached translation can be used for bypassing an address translation function of the route complex and directly accessing the queue data structure from the input/output device or end point. COPYRIGHT: (C)2008,JPO&INPIT

Abstract translation: 要解决的问题：提供一种使用队列数据结构和与队列数据结构相关的预翻译地址与输入/输出设备或终点进行通信的设备和方法。 解决方案：设备驱动程序服务被称为与队列数据结构相关的路由复合中的地址转换，以及设备驱动程序对保护表项的初始化，并将非翻译地址返回给设备驱动程序。 接下来，为输入/输出设备或终点提供非翻译地址。 随后，输入/输出设备或端点请求翻译这些非翻译地址，并且在接收到针对队列数据结构的输入/输出操作之前，地址被存储在输入/输出设备或终点中。 缓存的转换可用于绕过路由复合体的地址转换功能，并直接从输入/输出设备或终点访问队列数据结构。 版权所有（C）2008，JPO＆INPIT

公开(公告)号：JP2001022686A

公开(公告)日：2001-01-26

申请号：JP2000172663

申请日：2000-06-08

Applicant: IBM

Inventor： KELLEY RICHARD A ,  NEAL DANNY MARVIN ,  THURBER STEVEN MARK ,  ADARUBAATOU GIRAAMO JENS

IPC: G06F13/36 ,  G06F11/00 ,  G06F13/14 ,  G06F13/38 ,  G06F13/40

Abstract: PROBLEM TO BE SOLVED: To provide an improved method and an execution system by which a buffer can be more effectively used in the case of transferring information between devices connected in an information processing system. SOLUTION: In this information processing system, a system bridge circuit can execute or over commits a transaction request from a system device to information transfer exceeding the current capacity of a bridge circuit in order to receive requested information while returning it from an instructed target device such as system memory 1O9 or other system device.

公开(公告)号：JP2008165776A

公开(公告)日：2008-07-17

申请号：JP2007324246

申请日：2007-12-17

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： RENATO J RECIO ,  THURBER STEVEN MARK ,  FREIMUTH DOUGLAS M ,  SALZBERG CLAUDIA ANDREA ,  VARGAS JACOBO A

IPC: G06F15/173

CPC classification number: G06F12/1009 ,  G06F12/145

Abstract: PROBLEM TO BE SOLVED: To provide a system and method for initializing shaped memories for sharing endpoints across a plurality of root complexes. SOLUTION: A multi-root PCIe manager (MR-PCIM) initializes the shared memory between root complexes and endpoints by discovering the PCIe switch fabric by traversing all the links accessible through the interconnected switches of the PCIe switch fabric. As the links are traversed, the MR-PCIM compares information obtained for each of the root complexes and endpoints to determine which endpoints and root complexes reside on the same blade. A virtual PCIe tree data structure is then generated that ties the endpoints available on the PCIe switch fabric to each root complex. COPYRIGHT: (C)2008,JPO&INPIT

Abstract translation: 要解决的问题：提供一种用于初始化整形存储器以用于跨多个根复合体共享端点的系统和方法。 解决方案：多根PCIe管理器（MR-PCIM）通过遍历通过PCIe交换矩阵的互连交换机可访问的所有链接发现PCIe交换结构，初始化根配置和端点之间的共享内存。 随着链接的遍历，MR-PCIM比较了每个根复合体和端点获得的信息，以确定哪个端点和根复合体驻留在同一个刀片上。 然后生成虚拟PCIe树数据结构，将PCIe交换结构上可用的端点与每个根组合相关联。 版权所有（C）2008，JPO＆INPIT

公开(公告)号：JP2008152783A

公开(公告)日：2008-07-03

申请号：JP2007324288

申请日：2007-12-17

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： RENATO J RECIO ,  THURBER STEVEN MARK ,  FREIMUTH DOUGLAS M ,  SALZBERG CLAUDIA ANDREA ,  VARGAS JACOBO A

IPC: G06F13/36 ,  G06F12/00 ,  G06F13/14

CPC classification number: G06F13/4022 ,  G06F13/1652

Abstract: PROBLEM TO BE SOLVED: To provide a system and method for communication between host systems using a transaction protocol and shared memory. SOLUTION: The shared memory is initialized based on a finding process in communication fabric so that at least one end point has an address range in the shared memory of at least two host systems. For communication between the end point and a root complex of the same or different host systems using the shared memory of the host systems, transaction oriented protocol can be established. The transaction oriented protocol specifies a series of transactions to be executed by various elements such as the root complex or the end point in order to push or pull data. Various combinations of push transactions and pull transactions can be used. COPYRIGHT: (C)2008,JPO&INPIT

Abstract translation: 要解决的问题：提供使用事务协议和共享存储器的主机系统之间的通信的系统和方法。 解决方案：基于通信结构中的查找过程来初始化共享存储器，使得至少一个端点在至少两个主机系统的共享存储器中具有地址范围。 对于使用主机系统的共享存储器的相同或不同主机系统的端点和根复合体之间的通信，可以建立面向事务的协议。 面向事务的协议指定要由诸如根复合体或终点的各种元素执行的一系列事务，以便推或拉数据。 可以使用推送事务和拉动事务的各种组合。 版权所有（C）2008，JPO＆INPIT

公开(公告)号：JP2007195166A

公开(公告)日：2007-08-02

申请号：JP2006348980

申请日：2006-12-26

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： RENATO J RECIO ,  STEVEN WADE HUNTER ,  BOYD WILLIAM TODD ,  THURBER STEVEN MARK ,  MADELINE VEGA ,  HOLLAND WILLIAM G ,  FREIMUTH DOUGLAS M

IPC: H04L12/28

CPC classification number: G06F13/4022 ,  H04L45/00 ,  H04L45/54

Abstract: PROBLEM TO BE SOLVED: To share PCI adapters between multiple blades.
SOLUTION: The present invention relates to a method for routing PCI transaction packets through switches between hosts and adapters in a distributed computing system that includes multiple route nodes, one or more PCI switches and PCI adapters, wherein each route node includes multiple hosts and one of the route nodes includes a PCI configuration manager (PCM). A table is created within specified one of switches and when a particular host is connected to the specified switch, a destination identifier including a set of bits designated by operating the PCM is entered to the table. The destination identifier is appended as an address to PCI packet directed through the specified switch from the particular host to one of the adapters. The destination identifier is used to determine that a PCI packet, routed through the specified switch from one of the adapters, is intended for the particular host.
COPYRIGHT: (C)2007,JPO&INPIT

Abstract translation: 要解决的问题：在多个刀片之间共享PCI适配器。 解决方案：本发明涉及一种用于在包括多个路由节点，一个或多个PCI交换机和PCI适配器的分布式计算系统中的主机和适配器之间的交换机路由PCI事务分组的方法，其中每个路由节点包括多个主机 并且其中一个路由节点包括PCI配置管理器（PCM）。 在指定的一个交换机内创建表，当特定主机连接到指定的交换机时，包括通过操作PCM指定的一组位的目的地标识符被输入到表中。 将目的地标识符作为地址提供给通过指定交换机从指定主机指向其中一个适配器的PCI数据包。 目的地标识符用于确定通过指定交换机从其中一个适配器路由的PCI数据包是针对特定主机的。 版权所有（C）2007，JPO＆INPIT

公开(公告)号：JP2009151767A

公开(公告)日：2009-07-09

申请号：JP2008305650

申请日：2008-11-28

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： BROWN AARON CHES ,  RENATO J RECIO ,  THURBER STEVEN MARK ,  DOUGLAS M FREIMUTH

IPC: G06F13/14 ,  G06F9/46 ,  G06F13/10

CPC classification number: G06F13/102 ,  G06F2009/45579

Abstract: PROBLEM TO BE SOLVED: To provide a system and a method for coping with a state for using one input-output (I/O) fabric in common by two or more of logical partitions (LPAR), and for using one input-output adapter (IOA) in common with the other LPAR by each LPAR. SOLUTION: An individual address space of itself is allocated to each LPAR for making access to the virtual function (VF) allocated to its LPAR. In recognition of each LPAR, it has an independent IOA of itself. Each VF is used in common in the whole area of a plurality of LPARs. The function for managing a common resource of the IOA via PF of the IOA, is provided by allocating its physical function (PF) to an input-output virtualization managing partition (IMP). A code performed in the IMP operate as a virtual intermediary to the VF. The IMP also operates as an interface to the PF for making access to the common VF function. COPYRIGHT: (C)2009,JPO&INPIT

Abstract translation: 要解决的问题：提供一种用于处理由两个或多个逻辑分区（LPAR）共同使用一个输入输出（I / O）结构的状态的系统和方法，并且用于使用一个输入 - 每个LPAR与其他LPAR相同的输出适配器（IOA）。 解决方案：将自己的单独地址空间分配给每个LPAR，以访问分配给其LPAR的虚拟功能（VF）。 为了表彰每个LPAR，它具有独立的IOA本身。 每个VF在多个LPAR的整个区域中共同使用。 通过将其物理功能（PF）分配给输入输出虚拟化管理分区（IMP）来提供用于通过IOA的PF管理IOA的公共资源的功能。 在IMP中执行的代码作为VF的虚拟中介。 IMP还作为PF的接口，用于访问公共VF功能。 版权所有（C）2009，JPO＆INPIT

公开(公告)号：JP2004342109A

公开(公告)日：2004-12-02

申请号：JP2004142836

申请日：2004-05-12

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： BAILEY DAVID ALAN ,  NGUYEN TRUNG NGOC ,  NORDSTROM GREGORY MICHAEL ,  PATEL KANISHA ,  THURBER STEVEN MARK

IPC: G06F9/46 ,  G06F9/50 ,  G06F11/00 ,  H02H3/05

CPC classification number: G06F11/0793 ,  G06F11/0712 ,  G06F11/0745

Abstract: PROBLEM TO BE SOLVED: To enable automatic recovery from errors in an I/O fabric. SOLUTION: A device, program products, and a method propagate errors detected in I/O fabric elements within a I/O fabric used for combining one or more end point I/O resource(s) with a processing unit in a computer. More specifically, errors are propagated to an end point I/O resource which is affected by the I/O fabric related to a recovery from the errors in the I/O fabric. Then a device driver or other program codes, used for accessing to each affected I/O resource, can be recovered asynchronously from the errors propagated from the accompanied I/O resources. In this way, recovery from the errors in the I/O fabric elements can often be made, before waiting for all the affected I/O resources to finish recovery. Also, the I/O fabric can be composed dynamically so as to be available for both recoverable end point resources and to unrecoverable endpoint I/O resources. In concrete terms, the I/O fabric element in the I/O fabric can be composed dynamically so that it enables machine check notification in the I/O fabric element, when it detects that the end point I/O resource has essentially become unrecoverable. COPYRIGHT: (C)2005,JPO&NCIPI