Abstract: A method and system for determining multi-thread direct memory activity is described. A pipelined circuit for tag availability with multi-threaded direct memory access activity may be employed. The pipelined circuit includes registers for providing a tag to a direct memory access (DMA) thread and receiving the tag upon completion of the DMA thread. The DMA engine is implemented in a multi-threaded environment allowing for out of order completion of the data transfer requests, such as an environment including a peripheral component interconnect extended (PCI-X) bus. The pipelined circuit provides a multi-threaded DMA engine with tags for transactions. In this manner, the number of DMA threads created and executed by the DMA engine may not exceed the number of stages in the pipelined circuit.

Abstract: A multi-threaded DMA engine data transfer system for a data processing system and a method for transferring data in a data processing system. The DMA Engine data transfer system has at least one frame buffer for storing data transmitted or received over an interface. A multi-threaded DMA engine generates a plurality of requests to transfer data over the interface, processes the plurality of requests using the at least one frame buffer, and completes the transfer requests. The multi-threaded DMA engine data transfer system processes a plurality of data transfer requests simultaneously resulting in improved data throughput performance.

Abstract: A silent mirroring protocol is provided, which eliminates the arbitration/selection times associated with all nexuses after the first nexus. During the initial SCSI bus negotiation, the initiator determines the transfer mode capability of all targets. The initiator establishes a group identification. Participants of the group recognize the group and look for an individual identification within the group. The initiator performs arbitration/selection with attention to the leader of the group. The initiator uses a message out phase with a vendor command to select a participant for a data block transfer. Each participant snoops the bus and recognizes when it is the target. If the initiator has more data to mirror, the process is repeated. When the last data block is transferred, the initiator sends a message out to the last participant, which is interpreted by the target leader as a command to release the bus. Each participant reselects the nexus initiator and returns a status.

Abstract: An innovative circuit is disclosed that enhances performance on a SCSI bus by pipelining nexuses in order to associate all nexus attributes on a per nexus basis. For example, a pipeline of nexuses is created so as to associate all of the nexus attributes from different connections involved. A plurality of load stages is provided whereby each load stage can latch all nexus attributes received at that stage. The latched nexus attributes can be loaded and stored at that stage or shifted to the next stage. As a result of the loading and shifting operations, a pipeline of nexuses is created that associates all of the nexus attributes received from the different connections on a per nexus basis. Therefore, all types of data traffic can be processed concurrently on a SCSI bus, which enhances data throughput and bus performance.