摘要:
A system for resetting a cache in a first device connected by a multilinelink to a memory in a second device. A transceiver in the first element connects to one end of each of the link lines and a transceiver in the second device connects to the other end. The transmitter in the first device transceiver is disabled in response to a failure of the transceiver to receive messages from the second device. The transmitter in the first device transceiver also selectively sends a reset sequence to the receiver in the second device. A detector detects when all of the receivers in the second device have either received a reset sequence or have detected that a transmitter in the first device is disabled. The detector sets a latch in response, representing that data in the second device cache is invalid. Optionally, the second device has responders which send responses over the link lines indicating receipt of a reset sequence. The transmitters in the first device switch to a disabled state when the responses are not received within a specified period.
摘要:
A self-timed interface (STI) links two physically separated systems or nodes. A transmit state machine forms each word in a serial bit stream into a plurality of bytes and generates idle and data character sequences. Each byte is separately encoded in a run-length-limited code, along with its idle and data character sequences. Each of the plurality of bytes is transmitted on a separate conducting line along with a transmit clock signal that is also transmitted on a separate line. At the receiver, the data stream on each line is separately phase aligned with the clock, and bit aligned.
摘要:
A method, computer program product, and data processing system for providing system-area network (SAN) multicasting functionality in a logically partitioned (LPAR) data processing system in which a channel adapter is shared among a plurality of logical partitions is disclosed. A preferred embodiment of the present invention allows LPAR “hypervisor” firmware to assume the responsibility for multicast protocol handling and distribution of packets among logical partitions.
摘要:
An apparatus, system and method for controlling access to facilities based on usage class of a requestor are provided. With the apparatus, system and method, a two level protection mechanism is provided for protecting host channel adapter (HCA) facilities from unauthorized access. With the present invention, a first level of access is provided through virtual address translation and a mechanism for determining if the requestor of access may access a system memory address space page associated with a real address to which the virtual address maps. A second level of access is provided through the allocation of usage classes and determining a required usage class for accessing an HCA facility.
摘要:
A method and apparatus for accessing a memory. Access rights for a memory operation are verified using a first data structure in response to receiving a request to perform the operation, wherein the request includes a virtual address for the operation. Responsive to access rights being verified for the memory operation, the virtual address translated into a real address using a second data structure.
摘要:
An apparatus and method for managing work and completion queues using head and tail circular pointers. With the apparatus and method, queue head and tail pointers are maintained in the channel interface and the host channel adapter. The head and tail pointers in the host channel adapter include a queue pointer table index and a queue page index for identifying a position within the queue. For work queues, the tail pointer in the channel interface is used to identify a next position where a work queue entry may be written. The head pointer in the channel interface is used only to determine whether the work queue is full or not. The head pointer in the host channel adapter is used to identify a next work queue entry for processing by the host channel adapter. The tail pointer in the host channel adapter is used by the host channel adapter to determine if the queue is empty. For completion queues, the head pointer in the channel interface is used to identify a next completion queue entry to be processed. The tail pointer in the host channel adapter is used to identify a next position in the completion queue to which the host channel adapter may post a completion queue entry.
摘要:
A mechanism for allowing a single physical IB node to virtualize a plurality of host channel adapters is provided. This includes providing the appearance of both a router and multiple virtual HCA's residing behind that router, to the external REAL subnet components. Each virtual host channel adapter will have unique access control levels. One or more InfiniBand subnets are virtualized in such a way that nodes residing both within the virtual subnets and in separate physical subnets are completely unaware of the virtualization. This virtualization of InfiniBand subnets significantly increases the horizontal scaling capabilities of a single InfiniBand physical component, while at the same time provides “native” network throughput for all the virtual hosts.
摘要:
Defines and handles segments in messages to place pauses and interruptions within the communication of a message between transmitted segments of the message. A common link switch is used in a network to connect links to all nodes, the segment structures in each message is preserved when packets of each message are passed within the switch to a switch transmitter connected to the destination node indicated in each packet of the message for transmitting each of the message segments. Each transmitter stores the source identifier of the first packet it transmits for a segment and then gives priority to transmitting packets which contain source and destination identifiers which match the current transmitter stored source identifier and match the destination node connected to the transmitter. This priority enables each switch transmitter to interleaves segments of concurrent messages while preserving the segmentation of transmitted packets to maintaining a maximum network communication rate for the messages. When an unexpected wait occurs within a transmitting segment, which exceeds a predetermined time-out period, a transmission of any other waiting segment is started, which improves the message transmission efficiency in the network.
摘要:
Defines and handles segments in messages to place pauses and interruptions within the communication of a message between transmitted segments of the message. A port cache of the destination node of each transmitted message obtains a message control block (MCB) which is used to control the reception of inbound segments within each message sent or received by the node. Each MCB stays in the cache only while its message is being communicated to the port and may be castout between segments in its message when there is no empty cache entry to receive a MCB for a current message being communicated but not having its MCB in the cache. Different types of dynamic priorities are written in status fields in each non-empty cache entry to enable a current cache entry to be castout when it is most likely to have the longest wait for being needed next in the cache for a segment communication to its message, which reduces cache castout thrashing to increase the average reception speed for communicating messages in the network. If a common link switch is used in a network to connect links to all nodes, the segment structures in each message is musts be preserved when packets of each message are passed through the switch to a destination node which uses the castout controlled communication cache taught herein.
摘要:
Speeds up a commanded system to read or write data for a large number of data frames transmitted on a link by executing a TRANSFER STRUCTURE instruction that automatically controls the reading or writing of a large number of scattered storage blocks in the storage of the commanded system containing, or to contain, the data transmitted on the link.