Abstract: A direct memory access (DMA) controller is connected with the CPU bus of a computer system through a bus interface and also to an I/O bus, which is connectable to one or more I/O controllers. The DMA controller contains multiple channels, each corresponding to a particular I/O controller, which are coupled to both the bus interface and the I/O bus. Each of the channels contains at least one register set storing information for the transfer and a data buffer holding the data during a transfer between the I/O bus and the CPU bus.

Abstract: A method of avoiding undesirable interferences between services in a telecommunications system that includes basic software for a basic service and supplemental software for services supplemental to the basic service. The supplemental software is divided into action software which acts solely on the basic service, and supplemental software which acts on the remaining supplemental software. A supplemental service is represented by action elements. Combinations of action elements form nodes in a mathematical binomial tree. Only those combinations which correspond to interference, i.e., an undesirable behaviour, between supplemental services will form a number of structures, called interference event trees. Before a supplemental service can be executed, its action elements are compared with nodes in the interference event trees (55), with the intention of ascertaining whether or not the former coincide with action elements belonging to the nodes in the latter.

Abstract: According to the invention an intelligent network backplane interface is provided for an intelligent local area network hub and a method for implementing the common interface is provided. The hub includes a concentrator backplane operating one or more local area access method. Modules are provided for connection to said backplane for providing a local area network function. A common interface, in the form of a carrier unit having an interface management processor is provided for establishing a connection between any one of various modules and the backplane. The processor provides a control for exchanging information between a module and the common interface with a first mailbox for reading information signals from the module and writing information signals to the interface and a second mailbox for reading information from the interface and writing information to the module. Information acquired by the control is used to form a parameter table for listing features of the module.

Abstract: A rate-limited user status broadcast method consists of a broadcast registration procedure and a broadcast procedure and makes use of a database including a user information data table for storing the current status of each user, a broadcast registration table for holding the current registration status of each user for which service is provided, and a destination list containing all concerned destinations for each user. In the broadcast registration procedure, a user is registered to be the subject of a broadcast procedure, and if there has been a change in the user's status, the user's status information is updated. In the periodically activated broadcast procedure, the status changes of registered users are sent to destinations associated with each user.

Abstract: Binary data is transmitted to a network physical layer from a media access controller as a series of multibit nibbles and is encoded into a multi-level data stream and split among a number of transmission channels, thereby reducing the signal frequency necessary to carry the data on each one of the transmission channels. The multi-level signal is then translated at a receiver back into a binary data stream. In a specific embodiment, the symbol transmission frequency on each of the transmission channels is at the same frequency as the nibble transfer rate between the media access controller and the physical layer.

Abstract: A distributed I/O device monitoring logic for power management control. The distributed I/O device monitoring logic reduces the gate count of convention device monitoring logic since the decode logic does not exist at two locations in the system. The distributed I/O device monitoring logic also has the benefits of self configuring monitor circuits, improved functionality, and decreased system power management overhead. The distributed I/O device monitoring logic comprises peripheral control for monitoring an I/O address range of at least one I/O device and for detecting access to the I/O device; system controller means coupled to the peripheral control for providing a ready (RDY #) signal and a system management interrupt (SMI #) signal; and central processing unit (CPU) coupled to the peripheral control and the system controller means for receiving the RDY # signal and the SMI # signal from the system controller and for sending information to both the system controller and the peripheral control.

Abstract: A data processing apparatus having a bidirectional interface which usually remains in a receiving mode during operation ready to receive data signals from an external device over a receiving data bus. When the apparatus is to transmit a data signal to the external device, the data signal placed on a transmitting data bus automatically causes the bidirectional interface to enter transmitting mode in preparation for the transmission.

Abstract: A data processing system having a magnetic disk apparatus that includes a patrol seek system that provides for the altering of a patrol seek method. The patrol seek system provides for the selection of a patrol seek method for executing an optimal patrol seek based on the performance characteristics of a high density magnetic disk apparatus.

Abstract: A superscalar superpipelined microprocessor having a write buffer located between the core and cache is disclosed. The write buffer is controlled to store the results of write operations to memory until such time as the cache becomes available, such as when no high-priority reads are to be performed. The write buffer includes multiple entries that are split into two circular buffer sections for facilitating the interaction with the two pipelines of the core; cross-dependency tables are provided for each write buffer entry to ensure that the data is written from the write buffer to memory in program order, considering the possibility of prior data present in the opposite section. Non-cacheable reads from memory are also ordered in program order with the writing of data from the write buffer. Features for handling speculative execution, detecting and handling data dependencies and exceptions, and performing special write functions (misaligned writes and gathered writes) are also disclosed.

Abstract: All power supplies and LAN adapters required for multiple computer terminals in a networked environment packaged within one portable enclosure that functions as a shipping case. The enclosure is used as an overhead projector table and a printer stand.

Abstract: A connection apparatus connecting LANs includes LAN control sections, an address searching/registering section and a wiring network. The LAN control sections are respectively provided for LANs which transmit frames, and at least issue a request for address registering and a request for address searching. The address searching/registering section executes address searching/registering processing in accordance with the address searching/registering requests from the LAN control sections. The wiring network includes transmission lines for independently transmitting frames between the LAN control sections.

Abstract: A multiple-task controller includes a plurality of tasks which are controlled with first and second external devices: task memory input/output devices provided respectively for the tasks; a serial input/output processing section and a parallel input/output processing section which process signals transmitted between the first and second external device and the input/output processing section, and perform writing and reading operations with respect to the task memory input/output devices; and a task control section which applies commands written in the task input/output devices respectively to the tasks, and applies the states of process of the tasks respectively to the task memory input/output devices.

Abstract: A system and method for controlling the flow of messages transferred between modules of a computer system is disclosed. The system includes a receiver module comprising a buffer having a capacity to store a predetermined number of messages, and a transmitter module coupled to the receiver module. The transmitter module includes a counter for maintaining a count value representative of free space in the buffer. The transmitter module transmits a message to the buffer when the count value is greater than zero. Then, the transmitter module decrements the count value by a value characteristic of the transmitted message, such that the count value is updated to be representative of free space in the buffer after transmission of the message to the buffer.

Abstract: An input device for a computer includes a mouthpiece and a interface for coupling the mouthpiece to the computer. The mouthpiece includes a plurality of air cells arranged in rows for entering commands to the computer. The are each assigned a command and a chosen command can be entered by inhaling or exhaling through the corresponding cell. The interface identifies which cell a user has selected and produces a distinct digital signal corresponding to that cell. In the preferred embodiment, a MIDI interface is used which discriminates between the different sounds produced by each cell. An on-screen keyboard is selectively displayed on the display screen of the computer to aid a user in selecting and entering commands into the computer. Software allows the user to interact and control different programs loaded on the computer. Potential users include the physically-challenged, astronauts, SCUBA divers, surgeons, construction workers, and musicians.

Abstract: Client computers on a computer network send messages informing their management computer on the network of the status of transmissions such as those for conferences involving the client computers. Managers use the messages to maintain records of active transmissions to (1) discover transmissions that started without explicit authorization from the managers and (2) delete transmission records corresponding to transmissions that terminated without explicitly notifying the managers.

Abstract: The invention relates to a local network operating in the asynchronous transfer mode (ATM), comprising a plurality of stations which are coupled, via network interfaces comprising a coupling device and a control arrangement, to sending and receiving ring lines and which are provided for transmitting user cells containing user-oriented information and at least control cells containing signalling information. The control arrangement assigned to a transmitter station is provided for generating at least a control cell to be sent to a receiver station after the control arrangement has received from the transmitter station a control cell with a realisable request for a unicast or multicast connection from a user of the transmitter station to a user of at least one receiver station, which control cell has in its information field at least information about the users, the address of the transmitter station and a user-related identification for a down channel.

Abstract: A printing system wherein pre-printing commands, print data, and post-printing commands are nested in a data stream, the commands bounding a group of print data pages in the data stream, the group of pages containing one or more page data blocks, and the commands defining one or more operations to be performed relative to the group of pages. Groups of pages are handled in accordance with the commands in the data stream, and the commands are buffered while a print processor processes the group of pages to form printed sheets therefrom.

Abstract: A system for asynchronously duplexing direct access storage device (DASD) data in a plurality of DASD subsystems has the advantage of decoupling the data duplexing operation from the DASD write I/O operation. This ensures the write does not incur unnecessary wait states in the subsystem. By establishing a sequence checkpoint at which time a set of information packets are grouped together and processed as a single sequence unit, this decoupling and independent operation takes place. Through this independence, data copying to a secondary location can take place without affecting the performance of the subsystems and also not affecting the corresponding integrity of the data that is being updated.

Abstract: System timers set by a central processing unit (CPU) provide a variable wait time functionally adapted and related to peripheral device availability. Wait time is iteratively and dynamically adjusted to more closely match peripheral performance to the system, thereby reducing CPU load and enhancing data throughput, with the data output routine executing at a process level rather than an interrupt level.In a preferred embodiment, the prior wait time is cumulated during which the peripheral is unavailable for data in order to form a running average delay utilized as a starting point for a next data input. Large delays are weighted to avoid skewing average wait time too quickly for infrequent longer events. Data throughput is thereby increased, avoiding unnecessary dormant periods with no output from the system, while, at the same time, CPU loading is reduced by avoiding excessive system polling.

Abstract: A method for designing a ring communication network is presented. A candidate ring defined by a candidate subset of nodes and a candidate routing is identified. A direction of communication is determined for each of a set of anticipated demands. The cost of operating the ring is calculated based upon the anticipated demands. Finally, the calculated cost is compared to a selected cost threshold to determine if a candidate ring is feasible, wherein the ring communication network comprises the candidate ring.