摘要:
An I/O control system (19) for a general-purpose computer (13) having multiple bus branches (21) separated by routing circuitry (43, 44) prepares a digital map (48, 53) of I/O device locations on the bus branches, and copies versions of the maps to registers in the routing circuitry (213) on startup and reset (211). The routing maps provide immediate routing information for I/O requests issued by one or more CPUs (40, 41, 42) in a system, allowing therouting circuitry between bus branches to immediately route requests to the proper device with a minimum of wait states. In one aspect I/O devices are polled (205) for location at startup and reset (201), and in another aspect, a universal map protocol (207) is a part of the BIOS or storage accessible by the BIOS, making a system self-configuring and providing the necessary information for the mapping for the routing circuitry.
摘要:
A data processor (21) includes an external bus interface circuit (33) responsive to two internal bus master devices (30, 34) to perform either a fixed or a variable burst access. The data processor (21) activates an external control signal to indicate whether a burst access is a fixed or a variable burst access. The data processor (21) indicates the port size of the accessed memory region by providing a port size signal to the external bus interface circuit (33). The external bus interface circuit (33) is responsive to the port size signal to break up the burst cycle into two or more burst cycles on the external bus (22, 23), if the accessed location corresponds to a memory (24) with a different port size than the internal bus (31).
摘要:
Un tampon bidirectionnel (94) comprend des premier (100) et second (102) tampons unidirectionnels connectés pour retransmettre des signaux dans des sens opposés entre des premier et second bus. Lorsqu'un circuit d'attaque de bus externe fait baisser la tension du premier bus, le premier tampon unidirectionnel (100) fait baisser celle du second bus et génère un signal empêchant le second tampon unidirectionnel d'attaquer activement le premier bus. Lorsque le circuit d'attaque de bus externe permet au premier bus de revenir au niveau logique élevé, le premier tampon unidirectionnel (100) fournit temporairement un courant de charge élevé au second bus afin de faire monter rapidement sa tension. De même, lorsqu'un circuit d'attaque de bus externe fait baisser la tension du second bus, le second tampon unidirectionnel (102) fait baisser celle du premier bus et génère un signal empêchant le premier tampon unidirectionnel d'attaquer activement le second bus. Lorsque le circuit d'attaque de bus externe permet au second bus de revenir au niveau logique élevé, le second tampon (102) fournit temporairement un courant de charge élevé au premier bus afin de faire monter rapidement sa tension.
摘要:
A data processing system includes data processing adapters (2) for controlling the data transfers between end systems (1) and a memory through contention type busses (7 and 11) and a control processor connected to the system bus (7) for through a coupler (8) for controlling the physical resources such as registers or memories of the whole data processing system. A control operation is initiated by the control processor which presents on its processor bus (14) a control operation to be executed in a destination unit which can be the coupler, a selected data processing adapter or a selected end system. Each unit is provided with a control operation interface dedicated to the processing of the control operation or its forwarding toward the destination unit and the preparation of a response to the control operation in the destination unit and the propagation of the response toward the coupler, so as to minimize the bus occupancy times.
摘要:
A Hardware Abstract Data Structure (HADS) includes a General Interface (GI), a Coherence Interface (CI), a Control and Configuration Logic (CCL), an Intelligence Logic (IL) and a Memory Pool (MP), wherein the GI is arranged to implement intercommunion between the HADS and a processor; the CI is arranged to implement coherence storage between multiple processors; the CCL is arranged to, in response to a command received by the GI, configure a hardware data structure for the MP; the IL is arranged to complete a large amount of simple and frequent data processing; and the MP is arranged to store data. Correspondingly, a method and data processing system are also disclosed. Through the disclosure, the HADS which is dynamically configurable, flexible, efficient, universal in interface and good in interconnectivity can be implemented to improve the data processing efficiency.
摘要:
Processor-based system hybrid ring bus interconnects, and related devices, systems, and methods are disclosed. In one embodiment, a processor-based system hybrid ring bus interconnect is provided. The processor-based system hybrid ring bus interconnect includes multiple ring buses, each having a bus width and configured to receive bus transaction messages from a requester device(s). The processor-based system hybrid ring bus interconnect also includes an inter-ring router(s) coupled to the ring buses. The inter-ring router(s) is configured to dynamically direct bus transaction messages among the ring buses based on bandwidth requirements of the requester device(s). Thus, less power is consumed than by a crossbar interconnect due to simpler switching configurations. Further, the inter-ring router(s) allows for provision of multiple ring buses that can be dynamically activated and deactivated based on bandwidth requirements. This provides conservation of power when full bandwidth requirements on the processor-based system hybrid ring bus interconnect are not required.
摘要:
Systems and methods for multiple network access by mobile computing devices are disclosed. In one embodiment, a data bus is used to couple multiple baseband processor endpoints to multiple network access cards, such that each baseband processor endpoint may communicate over the data bus to any of the network access cards. In an exemplary, non-limiting embodiment, the baseband processor endpoint is a modem and the network access cards are subscriber interface module (SIM) cards or universal integrated circuit cards (UICCs). By allowing each of the baseband processor endpoints to use any of the network access cards, different networks may be used for different purposes by the mobile computing device. Further, the use of a single bus in this manner may allow for greater scalability, while also saving pin count, silicon area, board area, and power consumption within the computing device. Such savings ultimately improve the cost of the device.
摘要:
A wireless dockee may access a vendor-specific feature available on a peripheral device in communication with a wireless docking center. Information regarding the vendor-specific feature may be communicated by the wireless docking center to the wireless dockee both during pre-association discovery as well as via a service description that corresponds with the vendor-specific feature. The wireless dockee may use the received information to generate an access request to the peripheral device.
摘要:
In a method and an apparatus provided in embodiments of the present invention, a first switch receives an ARP response from an SDN controller, where the ARP response carries a MAC address of a destination gateway; the first switch acquires, according to the MAC address of the destination gateway, VTEP information corresponding to the MAC address of the destination gateway, where a router corresponding to the VTEP information is located in a first data center; and the first switch sends, according to the VTEP information, an IP packet to the router corresponding to the VTEP information, so that the router corresponding to the VTEP information sends the IP packet to a second virtual machine through a tunnel between the router and a second switch. Therefore, an SDN controller serves as a proxy, which reduces transmission bandwidth occupied by packet broadcasting; in addition, a packet passes only through a router of the first data center, which alleviates roundabout routing of the packet between data centers.