Abstract: A computer-implemented method, apparatus, and computer program product are disclosed for migrating data from a source physical page to a destination physical page. A migration process is begun to migrate data from the source physical page to the destination physical page which causes a host bridge to enter a first state. The host bridge then suspends processing of direct memory access operations when the host bridge is in the first state. The data is migrated from the source physical page to the destination physical page while the host bridge is in the first state.

Abstract: A computer implemented method, apparatus and mechanism for recovery of an I/O fabric that has become terminally congested or deadlocked due to a failure which causes buffers/queues to fill and thereby causes the root complexes to lose access to their I/O subsystems. Upon detection of a terminally congested or deadlocked transmit queue, access to such queue by other root complexes is suspended while each item in the queue is examined and processed accordingly. Store requests and DMA read reply packets in the queue are discarded, and load requests in the queue are processed by returning a special completion package. Access to the queue by the root complexes is then resumed.

Abstract: Method, apparatus and system for controlling input/output adapter data flow operations in a data processing system that includes at least one of a traffic class mechanism in conjunction with virtual channel resources so as to be able to associate Load/Store and DMA flows to/from an input/output adapter, and a relaxed ordering mechanism for associating a relaxed ordering bit to Load/Store operations to an input/output adapter. Functionality for controlling the input/output adapter data flow is provided in a host bridge that connects the input/output adapter to a system bus of the data processing system.

Abstract: A method and system are provided for dynamically informing an operating system (OS) of a distributed computer system, when a (new) device is added on the network. An OS registers with the subnet administrator (SA) to be notified when a new component is added to the subnet and placed in a partition to which the OS has access. The subnet manager (SM) initializes and configures the new component that is added and, through the partition manager, associates the component to one or more partitions of the subnet. The SA then notifies the OSs with access to the partition(s) to which the component is associated and that have a right to access the new component that the new component is available to the OS. Following, the OS updates a database of available components to include the new component and establishes communication with the new component.

Abstract: An apparatus, program product and method propagate errors detected in an IO fabric element from an IO fabric that is used to couple a plurality of endpoint IO resources to processing elements in a computer. In particular, such errors are propagated to the endpoint IO resources affected by the IO fabric element in connection with recovering from the errors in the IO fabric element. By doing so, a device driver or other program code used to access each affected IO resources may be permitted to asynchronously recover from the propagated error in its associated IO resource, and often without requiring the recovery from the error in the IO fabric element to wait for recovery to be completed for each of the affected IO resources. In addition, an IO fabric may be dynamically configured to support both recoverable and non-recoverable endpoint IO resources.

Abstract: A method in a node for managing authorized attempts to access the node. A packet is received from a source, wherein the packet includes a first key. A determination is made as to whether the first key matches a second key for the node. The packet is dropped without a response to the source if the first key does not match the second key. Information from the packet is stored in response to this absence of a match. The information is sent to a selected recipient in response to a selected event, which may be, for example, either immediately or in response to polling to see if the information is present.

Abstract: A method, system, and product are disclosed for indicating a power status of multiple devices using hierarchically encoded indicators. Multiple nodes are included within a data processing system. Each node includes a different implementation of the devices. Each one of a first level of power indicators are associated with a different one of the nodes. Each one of a second level of power indicators is associated with a different one of the devices. A power status of each node is indicated utilizing one of the first level of power indicators. A power status of each device is indicated utilizing the second level of power indicators.

Abstract: A method for routing System Area Network (SAN) packets to multiple partitions within a single end node is provided. A range of Local Identification addresses (LIDs) are assigned to a channel adapter port within the SAN. Lower order bits within the LID are then assigned to select the particular partition in the end node. The Local ID Mask Control (LMC) field is used to assign multiple LIDs to a single port, using those low order bits to then route the message to the appropriate partition in the end node.

Abstract: A low-level function which enforces logical partitioning establishes a set of virtual indicator lights for certain physical components, the virtual indicator lights being only data in memory, a separate set of virtual indicator lights corresponding to each respective partition. Processes running in a partition can switch and sense the virtual indicator lights corresponding to the partition, but have no direct capability to either switch or to sense the virtual lights of any other partition. The low-level enforcement function alone can switch the state of the physical indicator light, which is generally the logical OR of the virtual indicator lights of the different partitions.

Abstract: A method for associating reliable datagram queue pairs with an underlying end-to-end context of a channel adapter is provided. The method comprises storing a reliable datagram domain (RDD) within the context of a reliable datagram queue pair (RD QP). The same RDD is stored within an end-to-end context (EEC). A partitioning key (P—key) is also stored within the EEC. The RDD cannot be accessed by consumer processes. In the case of incoming messages, the P—keys of the incoming data packet and EEC are compared. If P—keys match, then the RDD's of the RD QP and EEC are compared. If the RDD's match, the packet is processed normally. In the case of outgoing messages, the RDD's of the RD QP and EEC are first compared, and if they match, the P—Key of the EEC is inserted into the transport header of the data packet.

Abstract: A method in a network computing system for managing a plurality of subnet managers in the network computing system. An identification is received of a set of subnet managers within the plurality of subnet managers. The set of subnet managers is allowed to participate in a master election to select a master subnet manager. Subnet managers other than the set of subnet managers are placed in a dormant state. The master subnet manager is elected from the set of subnet managers through the master election, wherein other subnet managers within the number of subnet managers poll the master subnet manager to allow the other subnet managers to elect a new master subnet manager if the master subnet manager fails.

Abstract: A method, system, and apparatus for preventing input/output (I/O) adapters used by an operating system (OS) image, in a logically partitioned data processing system, from fetching or corrupting data from a memory location allocated to another OS image within the data processing system is provided. A hypervisor prevents transmission of data between an input/output adapter in one of the logical partitions and memory locations assigned to other logical partitions during a direct memory access (DMA) operation by assigning each of the input/output adapters a range of I/O bus DMA addresses. The I/O adapters (IOAs) are connected to PCI host bridges via terminal bridges. A single terminal bridge may support multiple IOAs, in which case every terminal bridge has a plurality of sets of range registers, each associated with a respective one of the IOAs to which it is connected. An arbiter is provided which selects one of the input/output adapters to use the PCI bus.

Abstract: A method in a node within network computing system for selecting a master network manager, wherein the first node is associated with a first priority. Requests are sent to the network computing system to discover other nodes within the network computing system. A second priority from the request is identified in response to receiving a response to one of the requests from another node within the network computing system. The first node shifts to a standby mode if it discovers a master subnet manager or the second priority is higher than the first priority. The first node shifts to a master mode if a response containing a priority higher than the first priority is absent in responses received by the first node and the first node has completed checking all other nodes in the network computing system.

Abstract: A computer system has physical pages of memory subject to access by input/output (“I/O”) devices, and a certain table with entries associating the physical pages with the I/O devices. Responsive to a request for data be moved from a first physical page to a second physical page, an entry is selected for the first physical page in the table. The selected entry indicates an association of the first physical page and one of the I/O devices. Arbitration is temporarily disabled for the selected I/O device so that I/O operations for the I/O device are temporarily disabled. Once arbitration is disabled for the device the data is moved from the first physical page to a second one of the physical pages and the entry is updated in the table to reflect a new association between the I/O device and the second physical page.

Abstract: A method for enabling a Q_key that is tamper proof from applications on a distributed computer system to protect selected network operations is provided. Applications and an operating system (OS) execute on the end nodes and each may access various network resources. In the invention, the network resources are configured for selective access by particular applications or OS. In a preferred embodiment, a control bit of a Q_key, which allows applications to authenticate their use of particular communication resources, i.e., the send and receive queues, is reserved and utilized to signal whether a particular application is allowed access to the resources. Setting the control bit to 0 allows the Q_key to be set by an application directly. When the control bit is set to 1, the Q_key cannot be set by an application and can only be set using a privileged operation performed only by the OS.

Abstract: A method, system, and apparatus for preventing input/output (I/O) adapters used by an operating system (OS) image, in a logically partitioned data processing system, from fetching or corrupting data from a memory location allocated to another OS image within the data processing system is provided. A hypervisor prevents transmission of data between an input/output adapter in one of the logical partitions and memory locations assigned to other logical partitions during a direct memory access (DMA) operation by assigning each of the input/output adapters a range of I/O bus DMA addresses. The I/O adapters (IOAs) are connected to PCI host bridges via terminal bridges. A single terminal bridge may support multiple IOAs, in which case every terminal bridge has a plurality of sets of range registers, each associated with a respective one of the IOAs to which it is connected. An arbiter is provided which selects one of the input/output adapters to use the PCI bus.

Abstract: An apparatus, program product and method propagate errors detected in an IO fabric element from an IO fabric that is used to couple a plurality of endpoint IO resources to processing elements in a computer. In particular, such errors are propagated to the endpoint IO resources affected by the IO fabric element in connection with recovering from the errors in the IO fabric element. By doing so, a device driver or other program code used to access each affected IO resources may be permitted to asynchronously recover from the propagated error in its associated IO resource, and often without requiring the recovery from the error in the IO fabric element to wait for recovery to be completed for each of the affected IO resources. In addition, an IO fabric may be dynamically configured to support both recoverable and non-recoverable endpoint IO resources.

Abstract: A low-level function which enforces logical partitioning establishes a set of virtual indicator lights for certain physical components, the virtual indicator lights being only data in memory, a separate set of virtual indicator lights corresponding to each respective partition. Processes running in a partition can switch and sense the virtual indicator lights corresponding to the partition, but have no direct capability to either switch or to sense the virtual lights of any other partition. The low-level enforcement function alone can switch the state of the physical indicator light, which is generally the logical OR of the virtual indicator lights of the different partitions.

Abstract: A method for managing allocation of network resources within the distributed computer system is provided. Specifically, the network traversal time and the end node response time for requests and/or packets being routed in a switch-connected system area network are utilized to determine the total round trip time for completion of the particular network operation. The sum of the timeout values for all switches that participate in routing the request from a requester (source) to the receptor node (target) is provided to the requester's channel adapter (CA). The time-out values are provided by the switch manufacturer and are sent to a network Subnet Manager (SM) via SM packets (SMP). The timeout values added together represent the SubnetTimeout. The time-out value of the target channel adapter (CA), the ResponseTime, is also provided to the requester. The requester then utilizes one of two timeout equations to calculate the overall response time required for the request to be completed.

Abstract: A bus arbiter for a computer system having a bus for connection to a plurality of bus devices where each bus device requests control of bus by use of a bus request signal. The bus arbiter contains logic which incorporates a fairness scheme for controlling and prioritizing the bus request signals based on a predetermined priority of each bus device and each bus device's prior access within a fairness cycle. Each device's prior access is tracked by bits in a data register and is determined by whether or not the device actually received or sent information over the bus, and not by a simple granting of access which could result in a retry signal.