公开(公告)号：US5802354A

公开(公告)日：1998-09-01

申请号：US871429

申请日：1997-06-09

申请人： Jeffrey P. Kubala ,  Thomas B. Mathias ,  Ira G. Siegel ,  David E. Whitney

发明人： Jeffrey P. Kubala ,  Thomas B. Mathias ,  Ira G. Siegel ,  David E. Whitney

IPC分类号： G06F1/04 ,  G01R31/3185 ,  G06F1/14 ,  G06F11/267 ,  G06F1/12

CPC分类号： G01R31/318505 ,  G06F11/267

摘要： A method and apparatus for synchronizing selected logical partitions of a partitioned information handling system to a test datesource. A system operator is presented with a display panel in which the operator may specify a set of test partitions and a test clock value. Each partition designated as a test partition is synchronized to the test clock value upon being newly activated, while production partitions are synchronized to a production clock value. The first test partition to be newly activated is synchronized to the test clock value by calculating the difference between the test clock value and a host clock value and storing the difference as an epoch offset for that partition. Each subsequently activated test partition is synchronized to the test clock value by copying the epoch offset of the previously activated test partition.

摘要翻译： 一种用于将分区信息处理系统的选定逻辑分区与测试日期源进行同步的方法和装置。 系统操作员被提供有显示面板，其中操作者可以指定一组测试分区和测试时钟值。 指定为测试分区的每个分区在新激活时与测试时钟值同步，而生产分区与生产时钟值同步。 新激活的第一个测试分区通过计算测试时钟值和主机时钟值之间的差值并将该差值存储为该分区的时期偏移量，与测试时钟值同步。 每个随后激活的测试分区通过复制先前激活的测试分区的历元偏移量与测试时钟值同步。

公开(公告)号：US06209106B1

公开(公告)日：2001-03-27

申请号：US09163539

申请日：1998-09-30

申请人： Jeffrey P. Kubala ,  Ira G. Siegel ,  Kenneth M. Trowell

发明人： Jeffrey P. Kubala ,  Ira G. Siegel ,  Kenneth M. Trowell

IPC分类号： G06F104

CPC分类号： G06F1/04

摘要： A method and apparatus for specifying offsets from an external time reference (ETR) for selected logical partitions of a partitioned information handling system. Each logical partition has a logical time-of-day (TOD) clock which may be set independently of the host system clock by specifying an epoch offset from the host clock. A system operator is presented with a display panel in which the operator may specify the magnitude and sign of an ETR offset for each logical partition selected to have an ETR offset. Each logical partition, upon being activated, issues a Store Clock (STCK) instruction to read its own logical TOD clock as well as a Store ETR (STETR) instruction to read the ETR. If the logical partition has a specified ETR offset, the logical partition manager simulates the STETR instruction by adding or subtracting the ETR offset to the real ETR time and returning the modified value to the partition; otherwise, the logical partition manager returns the real ETR time to the partition. If the read ETR time (as returned by the STETR instruction) differs from the read logical clock time, the logical partition issues a Set Clock (SCK) instruction to reset the logical partition clock to the ETR time. The logical partition manager simulates this latter instruction by computing the difference between the two times and storing the difference in an epoch offset field for the partition. The invention permits sets of logical partitions to be synchronized to different clock values, either for test purposes or operation in different time zones.

摘要翻译： 一种用于指定对于分区信息处理系统的选定逻辑分区的来自外部时间基准（ETR）的偏移的方法和装置。 每个逻辑分区具有逻辑时钟（TOD）时钟，其可以通过从主机时钟指定历元偏移来独立于主机系统时钟设置。 系统操作员被呈现有显示面板，其中操作者可以为选择为具有ETR偏移量的每个逻辑分区指定ETR偏移量的大小和符号。 每个逻辑分区在被激活时发出存储时钟（STCK）指令，以读取其自己的逻辑TOD时钟以及存储ETR（STETR）指令来读取ETR。 如果逻辑分区具有指定的ETR偏移量，则逻辑分区管理器通过将ETR偏移量相加或减去实际ETR时间来模拟STETR指令，并将修改的值返回到分区; 否则，逻辑分区管理器将真正的ETR时间返回到分区。 如果读取的ETR时间（由STETR指令返回）与读取的逻辑时钟时间不同，则逻辑分区发出设置时钟（SCK）指令，将逻辑分区时钟重置为ETR时间。 逻辑分区管理器通过计算两次之间的差异并将该差异存储在分区的时期偏移字段中来模拟后一指令。 本发明允许将逻辑分区集合同步到不同的时钟值，用于测试目的或在不同时区中的操作。

公开(公告)号：US06189145B1

公开(公告)日：2001-02-13

申请号：US08864248

申请日：1997-05-28

申请人： Richard A. Bellin, Jr. ,  Jeffrey P. Kubala ,  Ira G. Siegel

发明人： Richard A. Bellin, Jr. ,  Jeffrey P. Kubala ,  Ira G. Siegel

IPC分类号： G06F9445

CPC分类号： G06F9/44521

摘要： A method of replacing a current version of a program module with a replacement version of the module concurrently with the execution of the program on a computer system. While execution of the program is suspended, each address reference in the program to an entry point within the current version of the module is replaced with an address reference to the corresponding entry point within the replacement version of the module. After the address references have been replaced and before execution is resumed, a patch initialization routine is invoked to perform initialization functions such as storage allocation, table initialization and feature detection that are performed before execution of the concurrently applied patch. Execution of the program is resumed when each address reference to the current module has been replaced with one to the replacement module and the patch initialization routine has been performed. A bit vector is used to ensure that each initialization function is performed only once per instantiation of the program being patched. An available storage value indicates the amount of initially allocated storage remaining available after performing an initialization function involving storage allocation.

摘要翻译： 一种在计算机系统上与程序的执行同时替换模块的替代版本的当前版本的程序模块的方法。 当程序的执行被暂停时，程序中的每个地址引用到模块的当前版本中的入口点被替换为模块的替换版本内的相应入口点的地址引用。 在地址引用被替换并且在执行恢复之前，调用补丁初始化例程来执行在执行并发应用的补丁之前执行的诸如存储分配，表初始化和特征检测的初始化功能。 当将当前模块的每个地址引用替换为替换模块并执行补丁初始化程序时，程序的执行恢复。 使用位向量来确保每个初始化功能仅在被修补的程序的实例化时执行一次。 可用存储值指示在执行涉及存储分配的初始化功能之后，剩余可用的初始分配的存储量。

公开(公告)号：US5253344A

公开(公告)日：1993-10-12

申请号：US755246

申请日：1991-09-05

申请人： James E. Bostick ,  Roger E. Hough ,  Suzanne M. John ,  Jeffrey P. Kubala ,  Karen M. Noonan ,  Norman E. Shafa ,  Ira G. Siegel

发明人： James E. Bostick ,  Roger E. Hough ,  Suzanne M. John ,  Jeffrey P. Kubala ,  Karen M. Noonan ,  Norman E. Shafa ,  Ira G. Siegel

IPC分类号： G06F15/177 ,  G06F13/00

CPC分类号： G06F15/177 ,  G06F9/45541 ,  G06F9/45558 ,  G06F2009/45579

摘要： A request is made by a system in a first logical partition, within a logically partitioned data processing system, to dynamically change the I/O configuration of the host system in a way that affects a system in a second logical partition. The hypervisor intercepts the request, ensures the serialization of such dynamic I/O requests, and allows dynamic reconfiguration to proceed. Subsequently, the hypervisor determines the effect of the reconfiguration on the second partition, and notifies the second partition of the change.

摘要翻译： 由逻辑分区数据处理系统中的第一逻辑分区中的系统进行请求，以影响第二逻辑分区中的系统的方式动态地改变主机系统的I / O配置。 管理程序拦截请求，确保这种动态I / O请求的序列化，并允许动态重新配置继续。 随后，管理程序确定重配置对第二分区的影响，并通知第二分区。