摘要:
A switching technique allows multiple interconnect bus devices to be connected to a single bus segment, even if the interconnect bus protocol only allows a one of the interconnect devices to be connected at any time. Each of the interconnect devices is connected to the interconnect bus segment with a switch, such that the interconnect device is electrically isolated from the interconnect bus segment when the switch is open. An interconnect sourcing agent connected to the interconnect bus segment controls the switches, closing the switch for one of the interconnect devices when a transaction is destined for that interconnect device, opening all of the other switches so that only one device is connected to the bus at any time.
摘要:
A distributed direct memory access (DMA) architecture where greater than seven DMA channels are provided and utilized. Alternative methods are disclosed for paging or swapping DMA channels so that more than seven may exist in a computer system, but only seven may be available at a time to remain compatible with conventional DMA controller software. In one method, channels may be assigned identical addresses, with one enabled at one time. In another method, channels are assigned unique addresses but the DMA master addresses only a subset of the total number of channels so that up to seven are available to compatible software at any one time.
摘要:
A distributed direct memory access (DMA) architecture where DMA controllers are modified to create isolated DMA channels. Each isolated channel includes its own set of uniquely addressable registers which provide functional compatibility with conventional DMA controllers. A DMA master interacts compatibly with the computer system and transparently communicates special cycles to the isolated DMA channels to cause the distributed DMA architecture to appear as the DMA controllers. The DMA master spawns special cycles to the isolated channels for sharing common write data with multiple channels and merging read data into a single DMA controller compatible register. Channel 4 cascading is also handled via tracking registers and special cycles to maintain disable and masking functionality of channel 4 as it effects channels 0-3.
摘要:
A distributed direct memory access (DMA) architecture where greater than seven DMA channels are provided and utilized. Alternative methods are disclosed for paging or swapping DMA channels so that more than seven may exist in a computer system, but only seven may be available at a time to remain compatible with conventional DMA controller software. In one method, channels may be assigned identical addresses, with one enabled at one time. In another method, channels are assigned unique addresses but the DMA master addresses only a subset of of the total number of channels so that up to seven are available to compatible software at any one time.
摘要:
A hardware device interface supporting transaction authentication is described herein. At least some illustrative embodiments include a device, including an interconnect interface, and processing logic (coupled to the bus interface) that provides access to a plurality of functions of the device through the interconnect interface. A first transaction received by the device, and associated with a function of the plurality of functions, causes a request identifier within the first transaction to be assigned to the function. Access to the function is denied if a request identifier of a second transaction, subsequent to the first transaction, does not match the request identifier assigned to the function.
摘要:
A system and method for remote direct memory access over a network switch fabric. Some illustrative embodiments may include a system comprising a first system node, a direct memory access (DMA) controller, a second system node, and a network switch fabric coupling together the first and second system nodes (the network switch fabric comprises a rooted hierarchical bus). The DMA controller is configured to perform a DMA transfer of data between the first and second system nodes across the network switch fabric. The data is formatted as one or more remote DMA (RDMA) protocol messages that are routed across the network switch fabric based on a bus end-device identifier corresponding to the second system node.
摘要:
In one embodiment, a computer system, comprises at least one host node, at least one input/output node coupled to the host node, at least one multi-function device coupled to the input/output node via a switch, and a middle manager processor comprising logic to block an enumeration process in a host node for the multi-function devices behind the switch hierarchy, initiate an enumeration process for the multi-function devices in a manager processor separate from the host node, store a routing table for the switch hierarchy in a memory module coupled to the manager processor, and allocate, in the manager processor, endpoint device resources to the host node.
摘要:
There is provided a distributed peer-to-peer communication system for interconnect busses of a computer system. More specifically, there is provided a method comprising transmitting a request to establish an isochronous channel between a first device and a second device, establishing the isochronous channel between the first device and the second device, and generating an isochronous transaction across the isochronous channel between the first device and the second device, wherein the isochronous transaction is a message type transaction.
摘要:
Support for indicating and controlling transaction priority on a PCI-X bus. Embodiments of the invention provide indicia that can be set to communicate to PCI-X-to-PCI-X bridges and Completer that a transaction should be handled specially and scheduled ahead of any other transaction not having their corresponding indicia set. A special handling instruction allows the priority transaction to be scheduled first or early. The indicia are implemented by setting a bit(s) in an unused portion of a PCI-X attribute field, or multiplexed with a used portion, to schedule the associated transaction as the priority transaction over other transactions that do not have their corresponding bit set. The present invention can be used for interrupt messaging, audio streams, video streams, isochronous transactions, or for high performance, low bandwidth control structures used for communication in a multiprocessor architecture across PCI-X.
摘要:
A middle manager and methods are provided to enable a plurality of host devices to share one or more input/output devices. The middle manager initializes each shared input/output device and binds one or more functions of each input/output device to a specific host node in the system, such that hosts may only access functions to which they are bound. The middle manager may also utilize a configuration register map to translate values from the actual configuration register into a unique modified value for each of the plurality of host devices such that each host device may access and use the shared input/output device regardless of the firmware or operating system operating thereon.