摘要:
A method of and apparatus for communicating data using a hub. The method includes determining a first estimated unused capacity left in a first frame in which a second transaction is to be performed between a hub and an agent. The method then includes determining an amount of a first data that can fit into the estimated unused capacity and that is to be sent to the hub during a first transaction and then sent by the hub to the agent during the second transaction. The method also includes sending the first data to the hub during the first transaction.
摘要:
A bus controller and its associated device drivers are provided to a digital system to operate and control a peripheral bus, including the selective operation of at least a first portion of the peripheral bus in a store-and-forward manner. The device drivers include a number of programming instructions. Upon programmed with the programming instructions, a digital system is enabled to operate the bus controller to facilitate communication with a first bus agent in this first portion. The programming instructions package a number of request packets destined for the first bus agent into a multi-packet package, schedule the multi-packet package to be transmitted in bulk, at a first communication speed, to a first hub in the first portion, for the first hub to buffer the request packets, and then forward the request packets to the first bus agent, on a packet-by-packet basis and at a second communication speed. In one embodiment, the second communication speed is slower than the first communication speed. In one embodiment, the programming instructions further schedule communications destined for a second bus agent of this first portion, for transmission to the first hub, at the first communication speed, for the first hub to repeat the communications for the second bus agent without buffering, at also the first communication speed.
摘要:
A method includes setting a contention scheme for an asynchronous bus such that the contention delay of isochronous transactions on the bus is bounded, and establishing an isochronous interface between at least two devices, the isochronous interface supporting an X-T contract. A number of isochronous transactions and corresponding return transactions delivered across the bus is measured during a specified time interval.
摘要:
An architecture is provided for isochronous access to memory in a system in which a stream of information may be sent to a memory unit. The stream is divided into a plurality of service periods with a specified maximum amount of information in selected service periods, and selected service periods have a first amount of information associated with asynchronous information and a second amount of information associated with isochronous information. In addition to sending a stream of information, a request for isochronous information from the memory unit may be sent. In this case, a stream of the requested information may be received from a memory unit a predetermined number of service periods after the sending of the request. This stream is also divided into a plurality of service periods with a specified maximum amount of information in selected service periods, and each service period has a first amount of information associated with asynchronous information and a second amount of information associated with isochronous information. In either case, the second amount of information may contain asynchronous information when less than all of the second amount of information is used for isochronous information.
摘要:
An apparatus for handling the removal of a memory card from a computer system. The apparatus includes a socket for receiving the memory card having a first section of card detect pins and a second section of single pins. The second section is aligned with the first section during insertion of the card into the socket. The second section slides out with the card during removal of the card from the socket. The apparatus also includes an interrupt handler for transferring control to the interrupt handler upon a break in contact between the card and card detect pins in the socket. The interrupt handler disables signal lines connected to the signal pins before the second section begins pulling out from the card upon removal of the card from the socket.
摘要:
Device drivers for removable system resources are configured dynamically in a computer system which has a processor, a system memory and an interface for receiving removable system resources (generally denoted feature cards). A feature card includes a card memory area that stores a device driver for controlling the feature card. The feature card device driver is separated into two parts: 1) a full device driver portion, and 2) a stub device driver portion. The full device driver provides all of the device driver functionality necessary to control each and every function of the feature card. The device driver stub is a small compact portion of processing logic for linking the full device driver with operating system software located in the computer system. There is an upper bound for the size of a device driver stub of a given feature card. Any system will have a known number of card slots. Therefore, by allocating enough space per slot in computer system memory to contain the largest permitted card device driver stub, one can ensure that there will always be enough computer system memory to contain every device driver stub required by the system at a given time.