Abstract: An address provided in a request issued by an adapter is converted to an address directly usable in accessing system memory. The address includes a plurality of bits, in which the plurality of bits includes a first portion of bits and a second portion of bits. The second portion of bits is used to index into one or more levels of address translation tables to perform the conversion, while the first portion of bits are ignored for the conversion. The first portion of bits are used to validate the address.

Abstract: A system and a method for failover control comprising: maintaining a primary device table entry (DTE) in a first table activated for a first adapter in communication with a first processor node having a first root complex via a first switch assembly and maintaining a secondary DTE in standby for a second adapter in communication with a second processor node having a second root complex via a second switch assembly; maintaining a primary DTE in a second table activated for the second adapter and maintaining a secondary DTE in standby for the first adapter; and upon a failover, updating the secondary DTE in the first table as an active entry for the second adapter and forming a path to enable traffic to route from the second adapter through the second switch assembly over to the first switch assembly and up to the first root complex of the first processor node.

Abstract: A tiered discovery capability is employed to obtain attributes regarding adapters of an I/O configuration. The first tier obtains a list of the adapter functions accessible to an operating system; the second tier obtains attributes regarding a selected adapter function of the list of adapter functions; and a third tier obtains common attributes of a group of adapter functions, the group including the selected adapter function.

Abstract: An adapter is enabled for use. The enabling includes assigning one or more address spaces to the adapter, based on a request. For each address space assigned to the adapter, a corresponding device table entry is assigned. When the adapter is no longer needed, it is disabled and the assigned device table entries become available.

Abstract: A system and computer implemented method for isolating errors in a computer system is provided. The method includes receiving a direct memory access (DMA) command to access a computer memory, a read response, or an interrupt; associating the DMA command to access the computer memory, the read response, or the interrupt with a stream identified by a stream identification (ID); detecting a memory error caused by the DMA command in the stream, the memory error resulting in stale data in the computer memory; and isolating the memory error in the stream associated with the stream ID from other streams associated with other stream IDs upon detecting the memory error.

Abstract: One or more message signaled interruption requests from one or more input/output (I/O) adapters are converted to I/O adapter event notifications while retaining the message vector indication. An I/O adapter event notification may be routed and presented to a host or to a guest that the host is executing. To present the notification to the correct host or to the correct guest, various data structures in host and/or guest memory are used.

Abstract: Various address translation formats are available for use in obtaining system memory addresses for use by requestors, such as adapter functions, in accessing system memory. The particular address translation format to be used by a given requestor is pre-registered in a device table entry associated with that requestor.

Abstract: Address spaces are resized concurrent to accessing those address spaces. The size of an address space can be increased or decreased concurrent to performing read or write operations on the address space. Further, cache entries associated with an address space being decreased in size are purged.

Abstract: Access to an input/output adapter by a configuration is controlled. For each requested access to an adapter, checks are made to determine whether the configuration is authorized to access the adapter. If it is not authorized, then access is denied. If it is authorized, but access should be temporarily blocked, then instruction execution is altered to indicate such. If access is permitted, but should be blocked for another reason (other than temporarily), then access is denied.

Abstract: The conditions under which adapter interruptions are made pending are controlled. Responsive to an interruption being presented to an operating system, subsequent interruptions are suppressed on all central processing units in the configuration. The operating system processes the interruption, including examining and processing indicators of reported events until the operating system discontinues the suppression. This enables the operating system to control the number of pending interruptions and the number of processors processing those interruptions.

Abstract: A measurement facility is provided for capturing and presenting fine-grained usage information for adapter functions in an input/output subsystem. Adapter specific input/output traffic is tracked on a per function basis and the results are dynamically presented to the user. This information is useful for performance tuning, load balancing and usage based charging, as examples.

Abstract: One or more message signaled interruption requests from one or more input/output (I/O) adapters are converted to I/O adapter event notifications. Each I/O adapter event notification includes the setting of one or more specific indicators in system memory and an interruption request, the first of which results in a pending I/O adapter interruption request. While a request for an I/O adapter interruption is pending, subsequent message signaled interruption requests are converted to I/O adapter event notifications, but do not result in additional requests for I/O adapter interruptions.

Abstract: A system for implementing non-standard input/output (I/O) adapters in a standardized I/O architecture, comprising an I/O hub communicatively coupled to an I/O bus and a plurality of I/O adapters at I/O adapter addresses, the I/O hub including logic for implementing a method comprising receiving requests from the plurality of I/O adapters, storing the I/O adapter addresses of a requester along with their corresponding target recipient addresses and operation codes, receiving a response from a responder, the response indicating that a request has been completed, determining that the response is in a format other than a format supported by the I/O bus, transforming the response into the format supported by the I/O bus, locating a stored I/O adapter address having a corresponding target recipient address that matches the responder address and a corresponding operation code that matches the responder operation code, and transmitting the response to the stored I/O adapter address.

Abstract: A low-latency queue pair (QP) is provided for I/O Adapters that eliminates the overhead associated with work queue elements (WQEs) and defines the mechanisms necessary to allow the placement of the message directly on the queue pair.

Abstract: Inconsistent data in communications networks is detected. The detection is performed by a component of the network that is separate from where the recovery takes place. The detection employs a packet sequence number, which is checked when the packet is received. The packet sequence number is initialized to define a valid window of packet sequence numbers.

Abstract: A transparent PCI-based multi-host switch. A switch is configured with multiple north facing ports to couple the switch to multiple hosts. The multi-host switch can be included in a variety of switch configurations, including configurations having one multi-host switch, configurations having multiple multi-host switches, and configurations including one or more multi-host switches and one or more single host switches. The switch is designed to include controls to accurately route a packet through the switch.

Abstract: Performance counters are provided for virtualized network interfaces of communications networks, while minimizing the use of hardware resources. A virtualized network interface includes physical resources, as well as logical resources. Dedicated performance counters are provided for the physical resources of the virtualized network interface, as well as for logical partitions coupled to that interface, while non-dedicated performance counters are provided for the logical resources. This enables the provision of performance counters for virtualized network interfaces, while minimizing hardware resources consumed by those interfaces.

Abstract: A method, system, and computer program product for target computer processor unit (CPU) determination during cache injection using input/output (I/O) adapter resources are provided. The method includes storing locations of cache lines for pinned or affinity scheduled processes in a table on an input/output (I/O) adapter. The method also includes setting a cache injection hint in an input/output (I/O) transaction when an address in the I/O transaction is found in the table. The cache injection hint is set for performing direct cache injection. The method further includes entering a central processing unit (CPU) identifier and cache type in the I/O transaction, and updating a cache by injecting data values of the I/O transaction into the cache as determined by the CPU identifier and the cache type associated with the address in the table.

Abstract: A method, system, and computer program product for target computer processor unit (CPU) determination during cache injection using I/O hub/chipset resources are provided. The method includes creating a cache injection indirection table on the input/output (I/O) hub or chipset. The cache injection indirection table includes fields for address or address range, CPU identifier, and cache type. In response to receiving an input/output (I/O) transaction, the hub/chipset reads the address in an address field of the I/O transaction, looks up the address in the cache injection indirection table, and injects the address and data of the I/O transaction to a target cache associated with a CPU as identified in the CPU identifier field when, in response to the look up, the address is present in the address field of the cache injection indirection table.

Abstract: Execution of tasks on accelerator units is managed. The managing includes multi-level grouping of tasks into groups based on defined criteria, including start time of tasks and/or deadline of tasks. The task groups and possibly individual tasks are mapped to accelerator units to be executed. During execution, redistribution of a task group and/or an individual task may occur to optimize a defined energy profile.