摘要:
The arbitrating method is based on the classification of the users into different categories, and the assignment to all users in a category of an identical privilege level which characterizes the interruption capability of the users in the category. A task performed by a selected user in a category can only be interrupted for granting access to the resource to a user in a category having a higher privilege level. Also a normal preference level is assigned to each user within a category, which determines the selection order of the users in the category. The privilege level of a user category combined with the preference level of each user constitutes the priority level of the user. The access to the resource is granted to a selected user having the highest priority level. If at least one user having a privilege level higher than the privilege level of the selected user, makes a request for the resource, the task of the selected user is interrupted and the interrupted user is assigned an interruption preference level which is higher than the normal preference levels of all other users in the category. A new user having the highest priority level is then selected after the resource is released by the interrupting user.
摘要:
This invention relates to a protocol implemented in a communication system for exchanging data and control messages between adapters to which are attached different users, and a shared memory subsystem comprising a depository storage, a manager of storage and a microprocessor. Such protocol enables the adapters to be the initiators of the transmission and reception of data by using the control lines that connect the manager of storage to all adapters in the same way as the data bus and the address bus. Moreover, the adapters slice the messages into data bursts to which are associated control words specifying the sizes, the owner and the position of the burst in the message. Consequently, those data bursts may be interleaved when transiting on the data bus without the intervention of the microprocessor for the routing, and they will be stored in or read from the depository storage according to the identification of the user in the control word. Therefore, only the shared memory needs to have a high capacity of storage, whereas the manager of storage and the adapters can operate with a reduced capacity of storage, and the time to send or to access a data burst is much more improved.
摘要:
A distribution mechanism includes a scheduling device which partitions a common timing signal with a period T into n slots of t duration each, a configuration table having n addressable locations with each of the n locations storing communication control information and addressable by slot numbers generated by the scheduling device and a distribution buffer device (2) having at least a first and a second part, with each part having n addressable locations addressed by control information provided by the configuration table during each slot period to cause an interface involved in the to be established communications during a selected slot period, to write the information to be transmitted in one part of the distribution buffer and the information to be received by the interface to be read from the other part of the distribution buffer at addresses derived from the communication control information and the slot generated by the schedule means.
摘要:
A device for interconnecting source users and destination users includes a common bus to which a memory with a plurality of independent buffers, a memory interface (22) and a central control apparatus (26) are connected. The memory interface (22) receives messages from source users, stores the messages in selected buffers and chains the buffers together. The central control apparatus generates inbound message queues and outbound message queues in response to commands which it receives from the memory interface.
摘要:
Apparatus for recovering lost buffer contents in a data processing system uses a memory divided into a plurality of buffers provided with buffer control blocks, through which source and destination users exchange information. A buffer management circuit is responsive to requests from users for allocating buffers to source users in order that source users may store the information to be sent to the destination users. This circuit builds buffer queues and dequeues buffers from the queues to send the information contained therein to the destination users and releases the buffers afterwards. A time mark register is settable to n different values in a predetermined order. The value of the time mark register is changed at the expiration of a time period P. Each time a buffer is allocated to one user, the current value of the time mark register is written into a time mark field of the buffer control block and a state field is set to a first value (leased). When the buffer is released, the state field is set to a second value (released). The contents of the buffer control blocks are read at regular time intervals t after period P, and the state field of every buffer control block is tested to determine whether it is set to the second value. If not so set, the time mark field is compared with the value the time mark register had at the time t-xP, where x is a number such as 1
摘要:
The bit streams, transporting the frames, received from lines (6) are placed in register 12 in such a way that n bits are processed in parallel during a time interval T. Parallel processor 10 counts the consecutive logical "1" bits beginning at the low order (left most) bit of the n bits received in interval T and from the bits received in the previous interval T-1, to determine when this number is found equal to 5 which bits have to be deleted, and when this number is found equal to 6 whether a flag is received. As a result, it reassembles N-bit characters, with N
摘要:
A method of operating a packet parser in a computing system includes providing a configurable packet pointer by the packet parser, the packet pointer configured to index a configurable number of atomic parsing elements, the atomic parsing elements having a configurable size, in a data stream received by the computing system for extraction, wherein the indexed atomic parsing elements are non-contiguous in the data stream; and receiving the extracted indexed atomic parsing elements from the data stream by the packet parser.
摘要:
A system and method in accordance with the present invention allows for an adapter to be utilized in a server environment that can accommodate both a 10 G and a 1 G source utilizing the same pins. This is accomplished through the use of a high speed serializer/deserializer (high speed serdes) which can accommodate both data sources. The high speed serdes allows for the use of a relatively low reference clock speed on the NIC to provide the proper clocking of the data sources and also allows for different modes to be set to accommodate the different data sources. Finally the system allows for the adapter to use the same pins for multiple data sources.
摘要:
A system and method for computing a blind checksum includes a host Ethernet adapter (HEA) with a system for receiving a packet. The system determines whether or not the packet is in Internet protocol version four (IPv4). If the packet is not in IPv4, the system computes the checksum of the packet. If the packet is in IPv4, the system determines whether the packet is in transmission control protocol (TCP) or user datagram protocol (UDP). If the packet is not in either of TCP or UDP the system attaches a pseudo-header to the packet and computes the checksum of the packet based on the pseudo-header and the IPv4 standard.
摘要:
Providing communications between operating system partitions and a computer network. In one aspect, an apparatus for distributing network communications among multiple operating system partitions includes a physical port allowing communications between the network and the computer system, and logical ports associated with the physical port, where each logical port is associated with one of the operating system partitions. Each of the logical ports enables communication between a physical port and the associated operating system partition and allows configurability of network resources of the system. Other aspects include a logical switch for logical and physical ports, and packet queues for each connection and for each logical port.