Abstract: The present invention relates to a method and system for transferring portions of a memory block. A first data mover is configured with a first start address corresponding to a first portion of a source memory block. A second data mover is configured with a second start address corresponding to a second portion of the source memory block sized differently from the first portion. The first portion of the source memory block is transferred by the first data mover and the second portion of the source memory block is transferred by the second data mover.

Abstract: In a data processing device, a processor processes data based on a stored program and a buffer manager accesses the data. The data processing device includes a program memory which stores program codes, the program codes being loaded into the program memory and executed by the processor when processing the data. A shared memory stores one of the program codes and the data. A control unit selectively connects one of the processor and the buffer manager to the shared memory based on a select pattern, wherein the shared memory functions to store the program codes when the select pattern is set in a first condition, and the shared memory functions to store the data when the select pattern is set in a second condition.

Abstract: Various embodiments of a system and method for configuring a set of DMA resources as multiple virtual DMA channels are disclosed. In one embodiment, a system may include a context memory configured to store context parameters for each of the virtual DMA channels, a set of DMA resources, a DMA controller coupled to the context memory, and several I/O resources. The DMA controller is configured to configure the set of DMA resources as different virtual DMA channels using context parameters associated with different respective ones of the virtual DMA channels. Each virtual DMA channel corresponds to one of the I/O resources.

Abstract: A data processing apparatus improves speed and efficiency of transfer of bit data, especially, multivalue data bit plane. For this purpose, a memory 50 holds four 8-bit multivalue data per 1 word, and bit plane coding processing is made by 4×4 (=16) multivalue data (processing block). In a memory area 51, the most significant bit (bit 7) of respective multivalue data (data 0 to 15 in FIG. 5) is collected in the order of multivalue data, and stored in one position (hatched portions in FIG. 5). Similarly, bit 6 is collected from the respective multivalue data and stored in one position.

Abstract: An apparatus comprising a first circuit and a second circuit. The first circuit may be configured to transfer data between a plurality of first ports and a second port via a single port memory in response to one or more control signals. The second circuit may be configured to generate the one or more control signals, wherein the memory is time shared among the second port and the plurality of first ports.

Abstract: A system and method are presented for the encapsulation of a protocol stack in a voice telephony processor. Utilizing the system and method disclosed herein, digital voice telephony signals received in TDM frame-based format are converted to packet-based or cell-based format for transmission on a network, and vice-versa. The system and method may be embodied as a functional block within a specialized high-density integrated circuit voice processor. The voice processor employs on-chip digital signal processors (DSPs) to perform echo cancellation, dynamic range compression/expansion, and other processing on voice data. Advantageously, the encapsulation process of the disclosed herein does not impact the throughput of the DSPs. Instead, voice data is reformatted and prefixed with a header for the appropriate protocol layers using a dedicated on-chip packet control processor and linked list data structures managed by indexed direct memory access (DMA) controllers.

Abstract: A data transaction controller for transferring data responsive to a request from a client. The data transaction controller includes channel circuitry for providing a channel for data transfers. The channel circuitry includes a first storage device for storing channel configuration data. The data transaction controller further includes control circuitry for controlling access by the client to the channel circuitry.

Abstract: A multi-processor arrangement having an interprocessor communication path between each of every possible pair of processors, in addition to I/O paths to and from the arrangement, having signal processing functions configurably embedded in series with the communication paths and/or the I/O paths. Each processor is provided with a local memory which can be accessed by the local processor as well as by the other processors via the communications paths. This allows for efficient data movement from one processor's local memory to another processor's local memory, such as commonly done during signal processing corner turning operations. Configurable signal processing logic may be configured to host one or more signal processing functions which allow data to be autonomously accessed from the processor local memories, processed, and re-deposited in a local memory.

Abstract: Systems and methods are described for a load/store micropacket handling system. A method includes interconnecting a compute node with a shared memory node; translating a processor instruction into an interconnect command; transforming the interconnect command into a direct memory access interconnect command; transmitting the direct memory access interconnect command via a link medium; and performing an operation defined by the direct memory access interconnect command. An apparatus includes a computer network, including: a compute node, having: a compute node interconnect interface unit; and a compute node interconnect adapter; a link medium, coupled to the compute node; and a shared memory node, coupled to the link medium, having: a shared memory node interconnect interface unit; and a shared memory node interconnect adapter.

Abstract: The ManArray processor is a scalable indirect VLIW array processor that defines two preferred architectures for indirect VLIW memories. One approach treats the VIM as one composite block of memory using one common address interface to access any VLIW stored in the VIM. The second approach treats the VIM as made up of multiple smaller VIMs each individually associated with the functional units and each individually addressable for loading and reading during XV execution. The VIM memories, contained in each processing element (PE), are accessible by the same type of LV and XV Short Instruction Words (SIWs) as in a single processor instantiation of the indirect VLIW architecture. In the ManArray architecture, the control processor, also called a sequence processor (SP), fetches the instructions from the SIW memory and dispatches them to itself and the PEs. By using the LV instruction, VLIWs can be loaded into VIMs in the SP and the PEs.

Abstract: Aspects of the invention include a method and apparatus to transfer specific data files from a disc drive storage system to an output device such as a printer. In one aspect, the disc drive uses a transfer protocol that determines the files stored from a peripheral device and sent to the output device. In another aspect, the transfer program compares the files sent to the output device to a file structure stored on the disc drive and presents the unsent files to the output device for processing.

Abstract: A distributed direct memory access (DMA) method, apparatus, and system is provided within a system on chip (SOC). DMA controller units are distributed to various functional modules desiring direct memory access. The functional modules interface to a systems bus over which the direct memory access occurs. A global buffer memory, to which the direct memory access is desired, is coupled to the system bus. Bus arbitrators are utilized to arbitrate which functional modules have access to the system bus to perform the direct memory access. Once a functional module is selected by the bus arbitrator to have access to the system bus, it can establish a DMA routine with the global buffer memory.

Abstract: The communication system stores packet data received via a plurality of communication channels in a memory or transmits packet data stored in the memory through the communication channels includes buffer descriptors in which information on packet data is stored. The system comprises a central processing unit (CPU) which stores the information on packet data in the buffer descriptors and indicates whether each of the buffer descriptors is being organized, whether an error occurred in packet data received, or whether the organization of each of the buffer descriptors is completed by allotting a flag bit to each of the buffer descriptors. The system comprises a direct memory access (DMA) controller which stops processing a buffer descriptor currently being accessed and accesses a next buffer descriptor, or processes packet data information stored in the buffer descriptor currently being accessed, after identifying the flag bit of the buffer descriptor currently being accessed.

Abstract: A buffer storage system is provided for storing groupings of data of varying size. The buffer storage system comprises a buffer storage section and a buffer management section. The buffer storage section has a first buffer subsection and a second buffer subsection. The first buffer subsection includes a plurality of buffer units of a first buffer unit size. The second buffer subsection includes a plurality of buffer units of a second buffer unit size wherein the second buffer unit size is larger than the first buffer unit size. The buffer management section is operable to determine the size of an incoming data grouping and to direct the incoming data grouping to one of the buffer subsections based on the size of the data grouping.

Abstract: A buffer control device for controlling a buffer memory includes a comparing unit which compares input data with one or more data patterns, a control unit which stores a code which indicates a data pattern among data patterns into said buffer memory if the input data matches with the data pattern, and a recovering unit which recovers the input data from the code.

Abstract: The present invention manages memory buffers in network device drivers in a flexible operating system (e.g., the Solaris operating system) that increase performance of the operating system at high throughputs with no detriment to the flexible nature of the operating system. Embodiments of the present invention reuse the (same) allocated and dma_binded memory buffers again and again, eliminating the repeated memory management of each data packet. In one embodiment, an rx-descriptor ring is treated as a true circular ring. A new data structure named rxbuffer_id is also defined along with a device-freemsg( ) function. In another embodiment, a device driver allocates and links a memory block (e.g., a rxbuffer_id data structure) with a message block at the time of allocating the message block for relocating incoming data packets. The memory block contains all the needed information for reuse of the message block.

Abstract: An entertainment apparatus comprising a peripheral device and a controller for controlling the peripheral device. The peripheral device and the controller are connected each other by an address bus and a data bus. The peripheral device which receives a DMA acknowledge signal from the controller carries out 32-bit DMA transfer using lower 16 bits of the address bus and the data bus, during assertion of the DMA acknowledge signal.

Abstract: In a data processing system including a plurality of digital signal processor subsystems, selected peripheral components are shared by the digital signal processor subsystems. In particular, the high level data link controller is shared by the subsystems. When a packet is received by a shared high level data link controller, the data signal groups are processed and placed in a temporary storage unit. The address signal group of the received packet is applied to channel block unit where the digital signal processor subsystem, to which the packet is directed, is identified and an INTERRUPT signal corresponding to the identified digital signal processor subsystem is generated. The INTERRUPT signal is applied to a switch. The switch, which receives the signal groups from the temporary storage unit, directs the signal groups to a buffer memory in the channel associated with the identified signal processing subsystem.

Abstract: A method and system for accessing a shared memory in a deterministic schedule. In one embodiment, a system comprises a plurality of processing elements and a system I/O controller where each processing element and system I/O controller comprises a DMA controller. The system further comprises a shared memory coupled to each of the plurality of processing elements where the shared memory comprises a master controller. The master controller may then issue tokens to DMA controllers to grant the right for the associated processing elements and system I/O controller to access the shared memory at deterministic points in time. Each token issued by the master controller grants access to the shared memory for a particular duration of time at a unique deterministic point in time. A processing element or system I/O controller may access the shared memory upon the associated DMA controller relinquishing to the master controller the token that grants the right to access the shared memory at that particular time.

Abstract: A digital signal processor includes a byte direct memory access (DMA) controller and an external memory controller, both of which are coupled to each other. The external memory controller is coupled to a byte memory and other external memories through a common data bus. The byte DMA controller performs a byte DMA operation to the byte memory through the common data bus by controlling the external memory, thereby avoiding an additional data bus. As a result, the digital signal processor according to the present invention has less connecting terminals and achieves a size reduction.

Abstract: A trace and debug support unit (120) that works in conjunction with a bus sniffer (112). The trace and debug support unit (120) maintains in memory one or more configurable filter rules which are used to define parameters of the trace history. A plurality of conditions or rules are provided, satisfaction of one or more of which causes a trace history to be filed. A transfer-specific signal may be issued, whereby all cells of the identified transfer are filed as part of the trace history. Alternatively, a connection-specific flag may be carried with each cell, whereby all cells of the specific connection are filed as part of the trace history.

Abstract: According to one embodiment of the present invention, a system (10) for data transfer is disclosed that comprises a transfer memory (24) having one or more buffers (40, 42, 44, 46, and 48). Accessing units comprising direct memory access units (20 and 22) are coupled to the transfer memory (24) and are operable to access the transfer memory (24). Pointers (50, 52, and 54) stored in the transfer memory (24) direct the accessing units (20 and 22) to selected ones of the buffers (44, 46, and 48) such that no two accessing units (20 and 22) are simultaneously accessing one buffer (44, 46, and 48). More specifically, the pointers (50, 52, and 54) may also direct a memory control unit (30) to a buffer that is not being accessed by an accessing units (20 and 22). According to one embodiment of the present invention, a method for data transfer is disclosed. First, a transfer memory (24) comprising one or more buffers (40, 42, 44, 46, and 48) is provided.

Abstract: A physical interface card for connection to a data bus associated with a data network node is provided. The physical interface card is adapted to perform without supervision from other data bus connected devices: byte ordering, byte alignment and byte scattering/gathering in conveying data between a data bus connected central memory block and at least one data channel associated with the physical interface card. The functionality is provided via a special function direct memory address device operating in accordance with byte ordering specifications for: data stored in the shared memory block and data conveyed via the at least one data channel. The byte alignment is enabled by direct byte addressing techniques as well as the use of an orphan counter to keep track of processed bytes. An implementation of the orphan counter as a state machine reduces processing overheads.

Abstract: A reconfigurable input/output controller (IOC) allows an adaptive computing engine (ACE) to communicate with external devices. The external devices can comprise a separate system on chip (SOC) or can be other devices or resources such as audio/visual output devices, memory, network or other communications, etc. The IOC allows different modes of transfer and performs necessary translation of input and output commands. In one embodiment, the IOC adheres to standard messaging and communication protocol used by other nodes in the ACE. This approach allows a uniform approach to the ACE design and provides advantages in scalability and adaptability of the ACE system. One feature of the invention provides a physical link adapter for accommodating different external communication types such as, RS231, optical, Firewire, universal synchronous bus (USB), etc.

Abstract: A test pattern generator for generating a test pattern for testing electrical characteristics of an electrical device. The test pattern generator comprises a pattern memory (32), a pattern cache memory (54, 180 and 182), a vector memory (12), a read out controller (14 and 170), and a transfer controller (34 and 178). The pattern memory (32) stores the test pattern. The pattern cache memory (54, 180 and 182) stores the test pattern read out from the pattern memory (32). The vector memory (12) stores a vector instruction indicating an order of the test pattern to be generated. The read out controller (14 and 170) judges whether an address of the test pattern to be read out from the pattern memory (32) is to be jumped or not based on the vector instruction read out from the vector memory (12).

Abstract: Packet-Based Direct Memory Access. The present invention overcomes the oftentimes hardware consumptive and complex implementation of conventional direct memory access (DMA) that employs descriptors. The descriptors that must are employed by conventional DMA must be set up by software, and the handshaking between the hardware and software is typically very cumbersome. The packet-based DMA performed in accordance with the present invention is operable and adaptable to various types of cell-based DMA modes. A flow control regulator, or flow control state machine, is used to control the packet-based DMA performed in accordance with the present invention. Two different multiplexors (MUXs) are employed, one for each of the transmit and the receive packet-based DMA transfers, to select the various cases of packet-based DMA. The present invention is operable within various modes including asynchronous transfer mode (ATM) cell-based asynchronous digital subscriber loop (ADSL) applications.

Abstract: A memory component, on a single integrated circuit, operated as a slave to an external master, includes a RAM, one or more configuration registers, data formatting logic, and associated control logic. The behavior of the memory component, and in particular the selection of a burst transfer format, is controllable via configuration register bits in the one or more configuration registers. Specifically, based on a format selection specified by the configuration bits, the control logic determines the sequence-length of the data transfers between the RAM and the external master. Other than the sequence-length, the data is not otherwise altered during the data transfers.

Abstract: The invention is a method and system for print stream determination. The system's method begins with the initiation of a print stream processing application to which a print stream is directed. A print job is determined from a set of characteristics resident in the print stream. The print processing application will determine the optimal use of the system's peripheral devices for performing the job. The optimal use is determined by comparing each of the job's characteristics with each of the characteristics of the potential device driver. The comparison begins with determination of a value for each of the job characteristics wherein the value is representative of a desired result. A value for each of the device driver characteristics is determined wherein the value is representative of a potential result. Each of the desired results is compared to each corresponding potential result. If no corresponding potential result can be established, then an alternative peripheral device is sought.

Abstract: A PC card control device includes a PC card identifying part configured to identify a type of a card connected with the connector. An identification information acquisition part of the PC card identifying part acquires identification information of the connected card from the connected card with a method conforming to a standard. A first recording part records pieces of first card information and pieces of first identification information of one or more types of PC cards conforming to the standard, and a second recording part records pieces of second card information and pieces of second identification of one or more types of expansion cards not conforming to the standard.

Abstract: A printer controller for processing print data includes a data processor, direct memory access controller, first and second memories with corresponding first and second transfer data busses. A bus switch selectively connects the first and second data transfer busses. When uncoupled, the data processor accessed the said first memory via the first data transfer bus and the direct memory access controller may independently accesses the second memory via the second data transfer bus. When connected, either the data processor or the direct memory access controller may access either memory to the exclusion of the other. This permits better allocation of data transfer bandwidth in the memory controller.

Abstract: A memory stores data for transfer over a bus by a computer program that operates according to a bus frame protocol. The memory includes a data structure stored in the memory. The data structure includes a matrix having blocks arranged in N rows and M columns, where N and M are integers that are greater than one. A block of the matrix includes data used with the bus frame protocol and corresponds to a destination port and a time slot for the data.

Abstract: A digital system having a split transaction memory access. The digital system can access data from a system memory through a read buffer (FIFO) located between the processor of the digital system and the system bus. The read buffer is implemented with two FIFOs, a first incoming data FIFO for reading data, and a second outgoing address FIFO for transmitting read requests. The processor of the digital system can access the data FIFO and read data while the data transfer is still in progress. This decreases the processing latency, which allows the processor to be free to perform additional tasks.

Abstract: A system that supports a high performance, scalable, and efficient I/O port protocol to connect to I/O devices is disclosed. A distributed multiprocessing computer system contains a number of processors each coupled to an I/O bridge ASIC implementing the I/O port protocol. One or more I/O devices are coupled to the I/O bridge ASIC, each I/O device capable of accessing machine resources in the computer system by transmitting and receiving message packets. Machine resources in the computer system include data blocks, registers and interrupt queues. Each processor in the computer system is coupled to a memory module capable of storing data blocks shared between the processors. Coherence of the shared data blocks in this shared memory system is maintained using a directory based coherence protocol. Coherence of data blocks transferred during I/O device read and write accesses is maintained using the same coherence protocol as for the memory system.

Abstract: A method of direct memory access (DMA) includes receiving a first notification at a DMA engine that a first list of descriptors has been prepared, each of the descriptors in the list including an instruction for execution by the DMA engine and a link to a succeeding one of the descriptors, except for a final descriptor in the list, which has a null link. The DMA engine reads and executes the descriptors in the first list. When the DMA engine receives a second notification that a second list of the descriptors has been prepared, it rereads at least a part of the final descriptor in the first list to determine a changed value of the link, indicating a first descriptor in the second list. It then reads and executes the descriptors in the second list responsive to the changed value of the link.

Abstract: A method is described for storing and processing/filtering signals, as well as a memory arrangement, a signal processing arrangement and, in particular, a digital filter arrangement having a plurality of filter modules for digital processing/filtering of input values, having a memory area and a signal processing module, which contains in particular at least one multiplier-accumulator which has at least one multiplier and at least one adder. The input values, coefficients, and output values of the arrangement can be stored in the memory area and called up again therefrom as needed. The input values are gated with the coefficients to form output values. In order to alleviate the load on a higher-level microprocessor by digital processing/filtering of the input values, it is proposed that the digital filter arrangement have a Direct Memory Access controller for coordinating data transmission of the filter coefficients, input values and output values between the multiplier-accumulator and the memory area.

Abstract: A single host adapter hardware I/O control block contains information used to specify a transfer of data from a first target device to a host system and in addition information that specifies whether the data is mirrored, and if so, identifies a second target device on which the data is to be read. After transferring the single hardware I/O control block to the host adapter integrated circuit, the host adapter integrated circuit determines whether the hardware I/O control block specifies a mirrored transaction. If a mirrored transaction is specified, the host adapter integrated circuit generates a second hardware I/O control block for the second target device using the information in the first hardware I/O control block. When the execution of both hardware I/O control blocks is complete, the host adapter integrated circuit provides a single completion notification to the host system.

Abstract: A system and method for increasing the performance of a flash-based storage system, using specialized flash memory controller(s). Several methods of performance improvement are suggested such as adding DMA capability to flash memory controller to reduce the data transfer time; connecting flash chips to a multitude of flash memory controllers, which allow continuation of the data transfer to the system, even after the page programming operation has started; and connecting flash chips to a multitude of DMA-capable flash memory controllers to allow data transfer directly from one flash chip to another. In addition, a multi-controller design is suggested, which efficiently combines these performance-improving methods. In its best mode of operation, the present invention is a Flash-based storage system with several flash controllers or a multi-controller with DMA interface, organized in a way that reduces the page programming, page fetch and page copy time.

Abstract: In a computing system, a computer-readable medium is for storing information. At least one computing device is for receiving at least one installation file and multiple installable files. The installation file includes at least one table for specifying an installation of a subset of the installable files. Also, the computing device is for identifying the subset in response to the table. Moreover, the computing device is for outputting the installation file and the identified subset of the installable files for storage by the computer-readable medium, such that less than all of the installable files are concurrently stored by the computer-readable medium.

Abstract: In a computer system having a plurality of modules connected by a bus, wherein the plurality of modules includes a first module and wherein the system has a word width of two or more bytes, a system and method of byte swapping bytes within a word stored in a location on the first module. An address is constructed, wherein constructing an address includes inserting address bits pointing to the location and activating an attribute bit in the address indicating whether bytes within the word should be swapped. The address is driven on the bus and received at the first module. If the attribute bit is active, byte swapping the word.

Abstract: A plurality of modems or modem types can run on a host processor, a digital signal processor or both, either concurrently or selectively. Modules of more than one modem program can be swapped in and out of DSP memory space. Common modem code can be run on either a host processor or on a DSP using respective command libraries.

Abstract: In a data processing apparatus, an image memory has a descriptor region and an image region, the image region storing a plurality of blocks of image data, the descriptor region storing a corresponding number of descriptor information blocks for the plurality of image data blocks. A DMA controller controls DMA data transfer of the image data blocks from the image region according to each descriptor information block in the descriptor region. The DMA controller comprises a register which stores one of the descriptor information blocks from the descriptor region of the image memory, and a control unit which determines, at a time of occurrence of a CPU interrupt, a start timing of a DMA data output operation of the DMA controller during a DMA data input operation of the DMA controller when an image editing request contained in input image data is received, the CPU interrupt being caused to occur by an interrupt request bit of the descriptor information block read from the register.

Abstract: A architecture for a multifunction peripheral to service a plurality of clients simultaneously. A shared memory receives data from the plurality of clients. A channel multiplexer selects data to be routed to a peripheral, a SCSI emulator is used to logically select the peripheral. The data is then forwarded from the multiplexer via the SCSI emulator to a PCI bus, the PCI bus being physically connected with the peripheral's engine. When data needs to be sent from a peripheral to a client, it is forwarded from the PCI bridge to the SCSI emulator and routed via a demultiplexer to the shared memory wherein it is retrieved by the appropriate client. The multifunction peripheral can be interrupted while performing a first task using a first peripheral, switch to a second task needing a second peripheral, and return to the first task when completed.

Abstract: The ManArray processor is a scalable indirect VLIW array processor that defines two preferred architectures for indirect VLIW memories. One approach treats the VIM as one composite block of memory using one common address interface to access any VLIW stored in the VIM. The second approach treats the VIM as made up of multiple smaller VIMs each individually associated with the functional units and each individually addressable for loading and reading during XV execution. The VIM memories, contained in each processing element (PE), are accessible by the same type of LV and XV Short Instruction Words (SIWs) as in a single processor instantiation of the indirect VLIW architecture. In the ManArray architecture, the control processor, also called a sequence processor (SP), fetches the instructions from the SIW memory and dispatches them to itself and the PEs. By using the LV instruction, VLIWs can be loaded into VIMs in the SP and the PEs.

Abstract: A method and apparatus for accommodating the speed requirements of a DMA read request from PCI protocol I/O devices attached via a DMA to a multiprocessor system mesh. A bridge between the device controller and the mesh is described which buffers the data from the memory in cache lines from which the data is delivered finally to the I/O device. The system is adaptive in that the number of cache lines required in past reads are remembered and used to determine if the number of cache lines is reduced or increased.

Abstract: A direct memory access (DMA) circuit which is physically positioned with an input/output device, the DMA circuit storing a first reference value pointing to a data structure which describes a buffer portion of system memory in which data is stored for transfer to the I/O device, a value determining a position within the buffer portion of system memory beginning at which a next sequence of data is to be placed, and a value determining a position within the buffer portion of system memory from which a next sequences of data is to be copied to the I/O device, the DMA circuit including circuitry for reading data from the buffer portion of system memory beginning at the position from which a next sequences of data is to be copied and for writing the data read to the I/O device.

Abstract: Methods of operation and systems for a standardized scatter/gather list processor component within DMACs and intelligent IOPs. The standardized circuit architecture and methods provide a register interface and associated processing capabilities to simplify firmware processing to save and restore context information regarding block transfer operations that are paused and resumed prior to completion. Furthermore, the invention provides for architecture and associated methods for processing of standard scatter/gather list elements by a standardized scatter/gather list processor embedded within DMACs and IOPs. Specifically, as applied in the context of SCSI or Fibre Channel IOPs, the scatter/gather list processor of the present invention simplifies IOP firmware processing to save the current block transfer context on occurrence of a SCSI disconnect and to restore the saved context on occurrence of a SCSI reselect.

Abstract: Apparatus and a method for generating an interrupt when a direct memory access by an I/O device is desired, suspending the operation of the microprocessor in response to the interrupt, placing state of the morph host to a last known correct state in response to the interrupt, determining the memory operation commanded by the I/O device, and utilizing the microprocessor to execute the memory operation commanded by the I/O device.

Abstract: Apparatus and method for providing DMA transfers between an adapter card with or with out DMA capabilities and a system CPU with DMA capabilities. An adapter DMA controller circuit resides between the system CPU and the adapter card. This adapter DMA controller allows the system to run in immediate mode which allows the system CPU to talk to the adapter card as if the adapter DMA controller was not there. The system can also run in DMA mode. In this mode the system CPU sets up the system DMA controller and the adapter DMA controller. The adapter DMA controller takes over sending or receiving data to the adapter card and then requesting a DMA transfer with the system DMA controller. The transfer of data between the adapter DMA controller and the adapter does not use any system CPU resources such as the data and address busses. The system CPU is free to use the system resources to continue operation.

Abstract: A method, data processing system, and I/O subsystem suitable for authorizing DMA accesses requested by a 64-bit I/O adapter are disclosed. The system includes one or more processors that have access to a system memory. A host bridge is connected between the processor(s) and an I/O bus. A first I/O adapter, which generates 32-bit addresses, is coupled to the host bridge. A second I/O adapter coupled to the host bridge is enabled to generate an address with a width greater than 32-bits (such as a 64-bit address). The system may include a Translation Control Entry (TCE) table, that is configured with information needed to translate an address generated by the 32-bit adapter to a wider address (such as a 64-bit address). In addition, the TCE may determine whether DMA access to the translated address by the requesting adapter is authorized. The system further includes an Access Control Table (ACT). The ACT determines whether DMA access to the address generated by the 64-bit I/O adapter is authorized.

Abstract: An external direct memory access unit includes an event recognizer recognizing plural event types, a priority encoder selecting for service one recognized external event, a parameter memory storing service request parameters corresponding to each event type and an external direct memory access controller recalling service request parameters from the parameter memory corresponding to recognized events and submitting them to a centralized transaction processor. The service request parameters include a priority for centralized transaction processor independent of the event recognition priority. The service request parameters may be stored in the form of a linked list. The service requests are preferably direct memory accesses which may include writes to the parameter memory for self modification. The centralized transaction processor may signal an event to event recognizer upon completion of a requested data transfer.