摘要:
A dedicated bus between a central processing unit and a peripheral unit, such as a graphics controller driving a video display, provides enhanced capability in an environment in which signal processing occurs within the central processing unit. The dedicated bus relieves other data buses, such as the PCI bus, of the need to communicate large amounts of data, such as decompressed video data. The resulting system supports high bandwidth transmissions of decompressed video data, enabling high resolution 24 bit full motion video and multiple data stream video.
摘要:
A dockable computer system is capable of performing symmetrical multi-processing operations. More particularly, the dockable computer system includes a portable computer and a host station (docking station), each including a resident CPU. The dockable computer system is capable of operating in a docked state in which the portable computer is physically joined with the host station and an undocked state in which the portable computer is physically separate from the host station. In the docked state, the dockable computer system is capable of performing demanding computational tasks such as video conferencing as one of the CPUs in either the portable computer or host station is dedicated to the video conferencing operation. The dockable computer system preferably includes a communication channel for transmitting multi-processing support signals between the portable computer and the host station. Multi-processing support signals include synchronization signals, cache coherency signals, and interrupt distribution signals such as the LOCK signal, PLOCK signal, FLUSH signal, EADS signal, INTR signal or INTACK signal. The communication channel may be a dedicated bus or may be provided through a docking bridge between the portable computer and host station. The dockable computer system advantageously optimizes CPU resources when the dockable computer system is in a docked state.
摘要:
A method and apparatus for transferring original data which includes images, between two stations located a distance apart, without actual transmission of the image portion of the data. A library of images are provided at each of the stations. The image to be transferred is processed into a description of the image which allows the reproduction of the image at the receiving end of the transmission using the images contained in the image library in the receiving station.
摘要:
A dedicated bus between a central processing unit and a peripheral unit, such as a graphics controller driving a video display, provides enhanced capability in an environment in which signal processing occurs within the central processing unit. The dedicated bus relieves other data buses, such as the PCI bus, of the need to communicate large amounts of data, such as decompressed video data. The resulting system supports high bandwidth transmissions of decompressed video data, enabling high resolution 24 bit full motion video and multiple data stream video.
摘要:
A data bus for connecting information processing devices is configurable into a plurality of subbusses in order to fully utilize the data bus capacity. The size and data transfer direction of each subbus, as well as the data transfer speed of each subbus, is independent of the other subbusses. Also, the data bus can be reconfigured to meet changing system requirements. A data bus controller is thus provided to accomplish this data bus reconfiguration. The reconfiguration may be accomplished in accordance with one of a plurality of information flow templates which may be stored in a memory. A method of configuring a data bus is also provided wherein information transfer needs of a system are identified and the data bus is configured according to the identified information transfer means. The reconfiguration in accordance with the information transfer needs may be accomplished in accordance with one or more information flow templates which may be stored in a memory. The system may operate in accordance with a self arbitration scheme such that reconfiguration of the system is based on operational experience, such as utilization rates or excess capacity associated with each of the subbusses.
摘要:
A method of allocating bandwidth among a plurality of devices communicatively connected through a data bus provides for determining a data need of at least one of the plurality of devices, allocating portions of the data bus to the devices in response to the data need, and transmitting data between the devices on the allocated portions of the data bus. The portions of the data bus can be subbusses, each comprising at least one bit line. The data need can be based on a measure of fullness of a buffer corresponding to the at least one device. The data need can be provided as feedback from the buffer to a data bus controller which allocates the portions of the data bus. The method can use rules for assigning the subbusses which are stored in a memory. A processor can change the rules to accommodate changing conditions in the data bus. Also provided is a data communication system comprising a dynamically reconfigurable data bus, a data bus controller connected to the dynamically reconfigurable data bus for configuring subbusses of the data bus, a plurality of receiving devices connected to the data bus, and a feedback connection from at least one of the receiving devices to the data bus controller, wherein the data bus controller configures the subbusses in accordance with feedback received over the feedback connection. A memory can be connected to the data bus controller for storing rules for use by the data bus controller in configuring the subbusses. A processor can change the rules to accommodate changing conditions in the data bus.
摘要:
A data communication system includes an express bus, a plurality of local buses, and a plurality of local/express bridges, each local/express bridge connecting a corresponding local bus to the express bus. A plurality of local/local bridges each connect two corresponding local buses. The plurality of local buses and the plurality of local/local bridges comprise a local path. Also provided is a method of communicating information from a sending communication device to a target communication device, comprising the steps of a) determining if the target communication device is on a local bus corresponding to the sending communication device, b) transferring the information from the sending communication device to the target communication device on the local bus corresponding to the sending communication device if the result of step a) is that the target communication device is on the local bus corresponding to the sending communication device, c) transferring the information from the sending communication device to an express bus if the result of step a) is that the target communication device is not on the local bus corresponding to the sending communication device, d) transferring the information from the express bus to a local bus corresponding to the target communication device, and e) transferring the information from the local bus corresponding to the target communication device to the target communication device.
摘要:
A system including a serialized secondary bus architecture. The system may include an LPC bus, an I/O controller, a serialized secondary bus, and at least one slave device. The LPC bus may be connected to the I/O controller, and the at least one slave device may be connected to the I/O controller via the serialized secondary bus. The serialized secondary bus has a reduced pin count relative to the LPC bus. The I/O controller may receive bus transactions from the LPC bus. The I/O controller may translate and forward LPC bus transactions to the at least one device over the secondary bus. The I/O controller may include a processing unit. The processing unit may initiate bus transactions intended for the at least one slave device. The I/O controller may also include a bus arbitration unit. The bus arbitration unit may arbitrate ownership of the secondary bus between the processing unit and the LPC bus.
摘要翻译:一种包括序列化二次总线架构的系统。 该系统可以包括LPC总线,I / O控制器,串行化辅助总线和至少一个从设备。 LPC总线可以连接到I / O控制器,并且至少一个从设备可以经由串行辅助总线连接到I / O控制器。 串行次级总线相对于LPC总线的引脚数量减少。 I / O控制器可以从LPC总线接收总线事务。 I / O控制器可以通过辅助总线将LPC总线事务转换和转发到至少一个设备。 I / O控制器可以包括处理单元。 处理单元可以启动用于至少一个从设备的总线事务。 I / O控制器还可以包括总线仲裁单元。 总线仲裁单元可以仲裁处理单元和LPC总线之间的辅助总线的所有权。
摘要:
A computer system optimized for real-time applications which provides increased performance over current computer architectures. The system includes a standard local system bus or expansion bus, such as the PCI bus, and may also include a dedicated real-time bus or multimedia bus. Various multimedia devices are coupled to one or more of the expansion bus and/or the multimedia bus. The computer system includes byte slicing logic coupled to one or more of the expansion bus and/or the multimedia bus which operates to allow different data streams to use different byte channels simultaneously. Thus the byte sliced multimedia bus allows different peripherals to share the bus simultaneously. The byte slicing logic thus may assign one data stream to a subset of the total byte lanes on the multimedia bus, and fill the unused byte lanes with another data stream. The computer system of the present invention thus provides much greater performance for real-time applications than prior systems.
摘要:
An improved programmable interrupt controller for use in a computer system including one or more interrupt service providers (ISPs), usually central processing units (CPUs). At least one CPU and a main memory system are coupled to a host bus. A bus bridge device couples the host bus to the expansion bus. At least one I/O device is coupled to the expansion bus and generates an interrupt request signal. The bus bridge and other bus devices may also generate interrupt request signals. A programmable interrupt controller receives the interrupt requests and provides processor interrupt signals as well as information regarding resource requirements necessary for servicing the interrupts to the one or more CPUs. The programmable interrupt controller also receives interrupt acknowledge signals from the one or more CPUs.