JPH1093598A - Method to operate autonomously pci serial bus interface device circuit of packet processing data communication interface device and autonomous boot circuit - Google Patents

Abstract

PROBLEM TO BE SOLVED: To attain autonomous DMA operation for a data packet device by transferring data to at least one program control list to operate a data packet transfer device in the autonomous mode. SOLUTION: While an automatic boot input 370 is in operation, an automatic boot mode is selected, a DMA channel 0 is in use to operate autonomously a PCI interface ASIC 20 such as fetch of an address of a first packet control list and access of a DMA master to an internal link register. The selected automatic boot mode and an external ROM are used to act the PCI interface ASIC 20 like a local processing unit so as to set up a register of all internal PCI interface ASIC 20 and to initialize other function on the PCI bus 24. Then a plurality of other PCLs are built up and linked. Various DMA channels execute the PCL to allow data transfer on an IEEE 1394 bus.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

RELATED PATENT SPECIFICATIONS This specification is hereby incorporated by reference.
No., registered on April 26, 1996 as a provisional patent application, "Method for allocating channel numbers to the received data packet, device and system", which is a continuation of the current U.S. Patent No..

[0002]

FIELD OF THE INVENTION The present invention relates generally to the field of electronic equipment, and more particularly to an improved platform-specific communication interface device, and more particularly to a packetized data communication interface device.
The present invention relates to a method, an apparatus, and a system for autonomously operating a CI bus interface device circuit.

[0003]

BACKGROUND OF THE INVENTION Not only is processing speed important for many multimedia applications, but input / output bandwidth is also a significant factor limiting applications on personal computers. To overcome this limitation, the IEEE 1394 standard uses high-performance multimedia connections for camcorders, televisions,
Stereo, CD exchange, set-top box (set-to
p boxes), mixing consoles and music keyboards, as well as traditional personal computer equipment. "Fire Wire (Fi
IEEE 1394, known as the "reWire)" standard, provides a bus interface standard for portable and desktop computer environments.
The 94 standard serves as an important link technology between the consumer and the computer market, and describes a serial bus driven by advanced communication protocols. IEEE139
The four serial bus is designed to provide the required data rates for a high performance peripheral bus at a low system cost.

[0004] The IEEE developed by the Institute of Electrical and Electronics Engineers (IEEE) and its member computer companies
The 1394 standard is a serial bus interface,
Enables low-cost, high-speed digital data transfer and communication. Transfer rates between devices can reach up to 400 megabits per second (mbps) and can be achieved via both asynchronous and isochronous data transfer modes. IEEE13
The 94 standard interface is itself suitable for video applications because it operates on an equal time interval time division system. For example, if the system is 1
Assuming that a frame is output every 1/15 second, at least one frame must be present in all output packets. The result is that the video looks smooth. For this reason, IEEE1
The 394 standard interface uses an asynchronous transfer mode (ATM), which also operates in isochronous mode.
It has high compatibility with technologies such as ustransfer mode).

[0005] An advantage of the IEEE 1394 standard is the real-time transfer of data, which provides ideal advantages for interconnecting multimedia applications. By using small, durable and flexible cables and cable splicers, the IEEE 1394 standard offers cost savings and eliminates the need for certain cable specifications. The IEEE 1394 standard provides a universal I / O interconnect, which consolidates printed circuit board space while consolidating I / O ports. Furthermore, I
The EEE1394 standard provides a peer-to-peer communication structure, which allows peripherals to communicate directly with each other without imposing a load on the host device.

[0006] The IEEE 1394 standard provides a high-speed serial bus, which uses packetized data including a header. The header contains routing information. Furthermore,
The packetized data includes payload data. The physical properties of the medium need not be designed for long distance transmission.
The IEEE 1394 standard is designed for short distances, such as a local area network running on a desk bus.
Although the distance is often longer than just a desktop, the IEEE 1394 standard is not intended to be used in local area network operation. Thus, basically, the IEEE 1394 standard is for workstations,
Devices such as home computers, televisions, VCRs and camcorders are provided with high-speed interconnects to replace serial and parallel buses for access by various media types such as audio, video and text.

The existing PCI interface has a problem that it lacks flexibility for operating the PCI interface device in a flexible mode. For example, in order to handle the functions of the PCI interface device, it is desirable that the operation is in a stand-alone or autonomous mode in order to support various applications. However, this is not possible with existing PCI interface devices.

[0008]

There is a need for a method and system that enables autonomous DMA operation within a PCI interface device that conforms to the IEEE 1394 standard.

A PC conforming to the IEEE 1394 standard
There is a need for a method and system for controlling access to a DMA machine in an I-interface device and providing an instruction set for autonomous DMA operation functions.

[0010] Furthermore, a PCI interface device conforming to the IEEE 1394 standard is used as a PCI master as a PC.
There is a need for a method and system for enabling the execution of I instructions.

Still further, there is a need for a method and system capable of autonomously turning on power and reconstructing an environment for executing instructions without external intervention.

In accordance with the present invention, there is provided a method and system for autonomously operating a direct memory access circuit, comprising a conventionally developed IEEE 1394.
It essentially eliminates or reduces the drawbacks and problems associated with methods and systems for operating a DMA circuit within a standard or similar communication circuit.

One feature of the present invention is that, for example, the IEEE
A method and system are provided for operating a PCI interface compliant with the E1394 standard in an autonomous mode. The method and system include introducing an autonomous boot mode select signal to a PCI interface device. The method and system form a register for autonomously operating a PCI interface device. The present invention forms at least one packet control list for controlling the operation of the PCI interface device in the autonomous mode. This allows the DMA device associated with the PCI interface device to transfer data associated with the PCI interface bus according to the packet control list. The present invention also provides DMA channel 0
To capture the address of the first program control list that uses. Further, the present invention allows a DMA master to access both external RPL read-only memory (ROM) and internal memory registers associated with a PCI interface device.

The technical features of the present invention are as follows:
It is an object of the present invention to provide a method for a data packet transfer device such as a standard-compliant PCI interface device to operate in a stand-alone mode without the need for an external processor. The present invention provides access so that the DMA engine can be reconfigured to operate independently.

Another technical advantage of the present invention is to provide a method for selecting operation in "autonomous" mode via an external option pin associated with a DMA engine.

Still another technical feature of the present invention is that the RPL
It is an object of the present invention to provide a method for initializing a ROM memory and an SRAM basic address register so that a direct memory access engine can perform read and write access in a stand-alone mode after a power-on operation.

The present invention and its features will be more fully understood from the following description given with reference to the accompanying drawings. The same reference numbers in the drawings refer to the same events.

[0018]

BRIEF DESCRIPTION OF THE DRAWINGS The submitted embodiments of the present invention are illustrated in the accompanying drawings, wherein like numerals refer to like corresponding parts throughout the various views.

FIG. 1 conceptually illustrates a personal computer environment 10, which includes a personal computer 12 indicated by a dashed box 12 and associated peripherals indicated by a dashed box 14. Personal computer 12
There are various buses and nodes inside to control the operation of the personal computer. For example, the interface bus 16 supports communication between a three-port physical layer interface 18 and an interface device 20 according to the present invention. The serial EPROM 22 supports the operation of the PCI interface device 20. PCI interface device 2
0 is a PCI bus 24 and an auxiliary port local bus 26
Interface with Also, the local bus 28 and the PCI handling equipment 3 via the PCI bus 24
0 and 32 are also communicating. PCI host bridge 3
4 executes a host bridge function between the local bus 28 and the PCI bus 24. Auxiliary port local bus 26
Is a flash PROM (or RPL ROM) 36,
Direct Memory Access (DMA) Channel Control Static RA
M (SRAM) 38, user-defined function (AUX) 40,
And video input / output zoom video (ZV) port 42
Communicates with. The host local bus 28 is a host CPU
It communicates with the local memory 44 and the local memory 46.

The peripheral device 14 that communicates with the physical layer interface 18 includes a CD ROM device 48, a laser printer 50, a desktop camera 52, and a digital video cassette recorder (VCR) 54 that interfaces with a video cable set-top box 56. Can be considered.

Referring to FIG. 1, a PCI interface AS
The IC 20 executes a basic function of data packet transmission control between a device operating in an environment supporting the PCI bus 24 and a device operating in a high-speed input / output peripheral device environment established by, for example, an IEEE 1394-1995 standard environment.
The PCI interface ASIC 20 of the present embodiment is an IEEE
E standard 1394-1995 and PCI specification version 2.0
Complies with Further, the PCI interface ASIC 20 has a function of a cycle master, particularly, an IEEE 1394-19.
It runs in a 95 environment and has the ability to detect lost cycle start messages. PCI interface ASI
C20 outputs a 32-bit cyclic redundancy check (CRC) signal to I
Generated for EEE1394 standard packet transmission, and
A 2-bit CRC check is performed when an IEEE 1394 packet is received. PCI interface ASIC20 is PC
It supports an isochronous barrier between the I-interface ASIC 20 and the physical layer interface 18. Furthermore,
The PCI interface ASIC 20 is IEEE1394
Supports standard transmission rates of 100, 200 and 400 Mbps and also has 3 programmable FIFOs
(Eg, asynchronous transmission, isochronous transmission, and general reception).

As described below, PCI interface ASIC 20 implements programmable channel address comparator logic on received input data packets and assigns them a DMA channel. The present invention provides, in at least one embodiment, five scatter-gather DMA channels, wherein the data packet manipulation of each channel includes (1) asynchronous packet transmission, (2), etc. It is programmed to support time interval packet transmission, (3) asynchronous packet reception, and (4) isochronous packet reception functions.

The PCI interface ASIC 20 also provides a PCI master bus function to support DMA operations, as well as a PCI slave function for read and write access to internal registers. To implement a 32-bit PCI address data path, the PCI interface ASIC 20
It provides software control over interrupt events as well as address data parity checking. PCI interface ASIC 20 provides a programmable external local bus to implement a dedicated path to external logic.
In addition, PCI interface ASIC 20 provides an 8-bit or 16-bit interface to a zoom video (ZV) port to transfer video data directly to an external motion video memory carrier.

FIG. 2 presents a block diagram 58, which illustrates the functional partitioning of the PCI interface ASIC 20 of the present invention. The logic in PCI interface ASIC 20 includes PCI bus logic 60, which is a serial EP
A serial EPROM interface 62 for communicating with the ROM 22 is included. PCI master logic 64 and PCI slave logic 66, as well as PCI configuration control and status registers 68, provide the logic necessary for communication with PCI bus 24. Local bus interface logic 70 provides the necessary control logic for interfacing with auxiliary port local bus 26.

D in the PCI interface ASIC 20
MA logic 72 includes a DMA engine 74 and a DMA engine.
A control and status register 76 is included to control operation and communication between PCI bus logic 60 and FIFO logic 78. FIFO logic 78 is a general-purpose receive FIFO
80, asynchronous transmission FIFO 82, equal time interval transmission FIFO
O84, pointer address mapping logic 86, and F
Includes IFO control and status register 88.

The link layer control logic 90 includes a control and status register 92, which controls all functions within the link layer control logic 90 and provides status reporting. Packet transmission control logic 98 and packet reception control logic 102 are logic, cycle timer 94
And works with the cycle monitor 96. The physical link interface logic 104 includes parallel / serial conversion and serial / parallel conversion functions.

Referring to FIG. 2, a PCI bus logic 60 executes interface logic between the PCI interface ASIC 20 and the PCI bus 24. The PCI slave logic 66 provides external PCI handling equipment with read and write functions to the slave interface control logic so that the PCI interface 20 control and status registers 68, 76, 88, 92 can be evaluated. These are required when the application software controls the PCI interface ASIC 20 and monitors the operation status. PC
The I master logic 64 provides the DMA logic 72 with a function of starting data transmission on the PCI bus 24 as a master device. The PCI configuration control and status register 68 is available in the personal computer system 12 application software for configuring and programming the PCI interface ASIC 20. It contains the control and basic registers required by the PCI, as well as the PCI interface ASIC2.
Also includes interrupt control and status signals for 0 and other control and status registers. Local bus interface logic 70 includes RAM, ROM, auxiliary functions, a zoom video port, and auxiliary ports for interfacing and controlling the four GPIO interfaces. Further, the serial EPROM interface 62
Provides certain configuration data required by the PCI and certain system control register information after power-up.

The serial EPROM interface 62 is connected to the PCI interface ASIC 20 and the serial EPRO.
Provides communication with M22 (FIG. 1). At power on,
Serial EPROM interface 62 initializes a small portion of PCI configuration register 68 from serial EPROM 22. Serial EPROM state machine is serial E
While accessing PROM 22, all PCI slave access inputs are terminated in a retry state.

The serial EPROM 22 also contains configuration data for PCI configuration control and status registers 68. This information is read and written by the host CPU 44, which emulates a two-wire serial bus protocol through a control register for the serial EPROM 22. This two-wire serial bus is operated by the host CPU 44 to set the serial EPROM 22 output enable bit to a value of “1”.
Access data and clock bits to emulate a linear serial bus protocol. The PCI configuration control and status register 68 includes a timer bit, which provides a time reference for timing two-wire serial bus protocol events.

The PCI master logic 64 performs necessary control for the PCI interface ASIC 20 to operate the PCI bus 24 as a master device. This logic enables memory read, memory write, memory read line, memory write line, and write disable command operations. Regarding the memory read function, as a result of the DMA read operation of the PCI interface ASIC 20,
A memory read is performed on the PCI bus by a memory read line command. Regarding memory write operation, PCI
As a result of the interface ASIC 20 DMA write operation, a PCI memory write is performed or a memory write line command or a memory write invalid command is issued on the PCI bus.

The PCI slave logic 66 executes control logic necessary for the PCI interface ASIC 20 to operate the PCI bus as a slave device. When enabled, the PCI slave function 66 responds to memory read or write commands in the PCI memory address range specified in the base address register included in 68. The PCI slave logic 66 performs a slave burst transfer when enabled by a slave burst bit in other control registers. The PCI slave logic 66 performs a directed write operation when enabled by control bits in the other control registers.

PCI Configuration Control and Status Register 68
Provides a function for executing PCI operation configuration of the PCI interface ASIC 20 to system and application software.

Local bus interface logic 70
Provides special input / output port groups that share common logic. These ports are either PCI bus 24 or D
It can be accessed from any of the MA engines 74.
External devices on these ports cannot function as master devices. These ports connect the PCI interface ASIC 20 to external devices or interfaces and allow for automatic data transfer with such devices. All local bus interfaces except the zoom video bus interface are synchronized to the local clock, which is provided from the PCI clock. The ZV port clock is programmed based on a PCI clock, an IEEE 1394 clock, or an external clock.

Local bus interface logic 70
Provides a shared local address bus, which in the present invention is a 16-bit address bus, a shared 8-bit or 16-bit read or write data bus, with programmable wait states and ready conditions. Local bus address and data bus are RO
Shared among M, RAM, AUX, and zoom port functions. The ZV output port has horizontal synchronization, vertical synchronization data valid, and ZV-PIXEL. CLR. I will provide a. Other inputs and outputs for local bus interface logic 70 include four general purpose I / O (GPIO) pins, which implement programmable direction and polarity functions. Other signals in this embodiment of the local bus interface logic 70 include a local bus clock output, a reset output, an interrupt input, and an external ready input. PCI configuration control and status register 68 provides control instructions and information necessary to configure the operation of local bus interface logic 70.

The PCI interface ASIC 20 includes a remote program load (RPL) read only memory (ROM), which
Provides a function of reading a boot code from an attached RPL ROM. This allows the system to boot from an IEEE 1394 device, even if the system has lost a particular IEEE 1394 boot code at power reset. Further, the ROM interface can be generalized to provide other functions in addition to accessing the RPL ROM. The local bus interface logic 70 includes the PCI slave and the internal DMA, the flash PROM 36, the SR
It supports read / write access to the AM 38 and other devices such as the RAM equivalent device shown in FIG. Access to the ROM is controlled by the PCI configuration control and status register 68 and is enabled by writing a 1 to the least significant bit of the ROM base address register. ROM interface is 8 bits or 1
Any of the 6-bit wide data can be configured as a designated number of wait states or similar functions. The ROM option is configured via the serial EPROM 22 when the power is reset, and is input via the PCI slave access.

The RAM interface is accessed through a second PCI memory based register in the PCI configuration control and status register 68. This memory is DMA
It can be used as a control structure or data buffer or shared memory interface to other functions such as a digital signal processor. The RAM interface can be configured with either a specified number of wait states or an external ready start with either 8-bit or 16-bit wide data.

The auxiliary interface is a general purpose input / output port, which is used for PCI configuration control and status register 6.
8 is accessed through a third PCI memory base address register. This port is used for compression / decompression (decompre
ssion) Used to implement high speed data paths to external dedicated resources such as logic or video processors / frame buffers. If a ZV port is enabled, one port in the AUX address space
Mapped to port. Otherwise, this space is available as part of the auxiliary address base of the submitted embodiment. In this embodiment, the auxiliary interface is 8
Either bits or 16-bit wide data can be configured as a designated number of wait states or external ready start-stops.

The FIFO logic is designed around a single 256x33 clocked dual port RAM in the preferred embodiment and is divided into three logical FIFOs. Each FIFO is programmed from 0 to 256 words in size. FIFO given
For a combination of sizes, the sum of the three FIFO sizes must be less than 256 words.

The general receive FIFO (GRF) 80 contains read and write pointer pairs and is used in accessing the FIFO dual port RAM. Each pointer counts from 0 to 1 minus its fifo_size_value. The FIFO RAM addressing range for each pointer is set by the logic that generates the offset value. The offset is added to the value of the pointer so that it maps to a unique range of RAM addresses. The active DMA channel uses the read pointer to read asynchronous or isochronous packets from the PCI side of the RAM and write them into host memory. The write pointer is used by the IEEE 1394 receiver to write asynchronous or isochronous packets received on the IEEE 1394 bus to the link side of the FIFO RAM. Two pointers are FIFO RAM
Is communicated through the synchronization logic because both sides of the FIFO dual port RAM are in different clock domains.

The asynchronous transmit FIFO 82 includes a read and write pointer pair for accessing the FIFO dual port RAM. Each pointer is from 0 to its fifo
_Size_value Count in the range of minus one. The FIFO RAM addressing range for each pointer is set by the logic that generates the offset value. The offset is added to the value of the pointer so that it maps to a unique range of addresses. The active DMA channel uses the write pointer to read an asynchronous packet from host memory and transfer it to the PCI
Write in the side. The read pointer is IEEE139
4 used by a transmitter, which uses a read pointer to read an asynchronous packet from the link side of the FIFO RAM and sends it onto the IEEE 1394 bus. The two pointers are communicated through synchronization logic on both sides of the FIFO RAM, which is a FIFO dual port RA.
This is because both sides of M exist in different clock domains.

The isochronous transmit FIFO 84 includes read and write pointer pairs for accessing the FIFO dual port RAM. Each pointer is from 0 to its
fifo_size_value Count in the range of minus one. The FIFO RAM addressing range for each pointer is:
Set by the logic that generates the offset value.
The offset is added to the value of the pointer so that it maps to a unique range of addresses. DM in operation
Channel A uses the write pointer to read an equal time interval packet from host memory and
Write into the PCI side of O RAM. IEEE139
4 The transmitter uses the read pointer to read an equal time interval packet from the link side of the FIFO RAM and transmits it on the IEEE 1394 bus. Read and write pointers are communicated through the synchronization logic on both sides of the FIFO RAM because both sides of the FIFO dual port RAM are in different clock domains.

The pointer dual port address mapping logic 86 uses each of the three size values from the FIFO size register to map each FIFO read / write pointer pair to a unique address range in the FIFO dual port RAM. The pointer address mapping function is generated based on the formula shown in the following table:

[Table 1]

FIFO control and status register 88 implements the control and status register set of FIFO logic 78. FIFO control and status register 88
Is a FIFO for setting the size of each logical FIFO.
Includes O size register. This register has three size parameters, the transmit FIFO 84 at equal time intervals,
The sizes of the asynchronous transmission FIFO 82 and the general reception FIFO 80 are programmed. This register is accessed via a PCI slave 66 read or write operation. The PCI FIFO pointer write / read port has a PCI slave 66 write / read port for software to capture or write the current value of the PCI pointer. The link-side FIFO pointer write / read port has a PCI slave read port for software to capture or write the current value of the link-side pointer. The generic receive FIFO pop-push port receives the 32-bit slave write and is a data quadlet pushed to the top of the GRF 80. A 32-bit slave read from this port is GRF8
Pop off the data quadlet from the beginning of 0. The asynchronous transmit FIFO pop / push port receives a 32-bit slave write to this port and makes it a data quadlet pushed to the top of the asynchronous transmit FIFO. A 32-bit slave read from this port pops off the data quadlet from the beginning of the asynchronous transmit FIFO 82. Equal time interval transmission FI
The FO pop / push port receives a 32-bit slave write to this port and sends the FIFO at equal time intervals.
The data quadlet is pushed to the top of the data quadrant 84. A 32-bit slave read from this port
The data quadlet is popped off from the head of the transmission FIFO 84 at equal time intervals. The FIFO control token status read port comprises a slave read and returns the value of bit 33 of the last data quadlet popped from one of the three FIFOs previously accessed. FI
The FO diagnostic test and control registers include a PCI slave read / write port for software to build FIFO logic to control the diagnostic test and its operation. The transmit FIFO threshold register also includes a software PCI slave read / write port to set the transmit threshold for asynchronous and isochronous transmit FIFOs.

The CRC logic 100 performs the logic to perform various functions, including generating a 32-bit automatic DIN CRC error code on the header portion of the packet data stream generated by the transmitter logic. The transmitter inserts this code into the data stream at the end of the header. For data packets having a data payload, CRC logic 100 generates a 32-bit automatic DIN CRC error code above the data payload portion of the packet stream generated by the transmitter logic. The transmitter inserts this code at the end of the packet stream. The CRC logic 100 generates a 32-bit automatic DIN CRC error code above the header portion of the incoming packet data stream. If the calculated code is equal to the header CRC code sent with the packet, the receiver determines that the header is correct. Further, the CRC logic 100 has 32
A bit automatic DIN CRC error code is generated above the payload portion of the incoming packet data stream. If the calculated code is equal to the data CRC code sent with the packet, the receiver determines that the data payload is correct.

The 1394 link layer logic 90 is based on IEEE
IEEE 1394 link layer control logic (LLC)
Perform as specified in 394-1995 standard. This function controls transmission and reception of IEEE 1394 packet data between the FIFO logic 78 and other devices on the IEEE 1394 bus.

The 1394 link layer control and status register 92 is required by the application software.
Executes control and status register logic to control and monitor the operation of the LLC. The 1394 bus number / node number register provides an interface for application software to program bus and node numbers. The 1394 link layer control register provides an interface for application software to control the operation mode of the LLC. 1
The 394 link layer interrupt status register provides an interface for the application software to interpret the cause of the interrupt generated by the LLC and a mechanism for eliminating this interrupt status. The 1394 link layer interrupt enable register provides an interface for the application software to selectively enable a status bit in the interrupt status register to generate an LLC interrupt or to disable an LLC interrupt. The 1394 cycle timer register provides an interface for application software to program the cycle timer with an initial value or read its current value. When the LLC is operating as a cycle master, this timer is used to time the period at which packet transmission begins every 125 microseconds. The 1394 physical layer access register provides an interface for application software to write data to and read data from the physical layer interface 18 control and status registers.

The 1394 diagnostic test control register provides an interface for application software to perform diagnostic tests on 1394 LLC logic.
The 1394 link layer control and status register 92 also includes a DMA channel 4-0 word 0 receive packet compare value register. Each register is assigned a DMA channel comparator logic function. The DMA channel comparator matches the selected bit position in the comparison value register with the corresponding bit position in the first data quadlet (word 0) of the input packet. The bit position at which the match occurs is specified by the mask value contained in the word 0 received packet comparison mask register. The DMA channel 4-0 word 0 received packet comparison mask register is assigned to the corresponding DMA channel comparator. The DMA channel compare logic uses the mask value in this register to select a bit position in word 0, which is matched against the corresponding bit position in word 0 receive compare value register. DMA channel 4
The -0 word 1 received packet comparison value register is assigned to the DMA channel comparator logic function. The DMA channel comparator matches the selected bit position in the compare value register with the corresponding bit position in the second data quadlet (word 1) of the input packet. The bit position at which the match occurs is specified by the mask value contained in the word 1 received packet comparison mask register. Also DM
The A channel 4-0 word 1 received packet comparison mask register is assigned to the corresponding DMA channel comparator. The DMA channel compare logic uses the mask value in this register to select a bit position in word 1 which is matched with the corresponding bit position in word 1 receive compare value register.

In addition, the 1394 link layer control and status register 92 includes a busy retry count register, the contents of which will cause the 1394 transmitter to retry asynchronous packet transmission when a busy acknowledgment is received from the destination node. The number of times should be specified. This register is read and written from the application software by PCI slave access. The busy retry time interval register contains the time interval that the transmitter should delay between subsequent retries when a busy acknowledgment signal is received for each attempt. This register is read and written from the application software via PCI slave access. Also,
There is a state machine vector register, which gives the software the ability to monitor the state vector of each state machine incorporated in the LLC. Furthermore, FIFO
The error counter is an under-run that occurs in the asynchronous and isochronous transmit FIFOs 82, 84 during packet transmission.
Over-ru which occurs in GRF80 during packet reception
n) and count.

The packet receiver control logic 102 executes the necessary logic to receive the input IEEE 1394 packet. The receiver control logic is IEEE
It conforms to the detailed function requirements specified in 1394-1995. These include the following functions for determining whether to receive an incoming asynchronous or isochronous packet using a bus and node ID register and / or a DMA channel receive packet comparator. CRC
The logic function checks the header CRC to make sure that the incoming packet has been received correctly. If the packet has a payload, the data CRC is checked and if the packet passes addressing and CRC checking, the received packet is loaded into GRF 80. Packet receiver control logic 102 generates an acknowledgment on the asynchronous received packet.

The cycle timer logic 96 executes logic for performing the cycle timer function. The periodic timer logic 96 meets the requirements of the periodic timer function specified in the IEEE 1394-1995 standard. The cycle timer includes a cycle counter and a cycle offset timer. The offset timer is either free-timed, reloaded from low to high on the select signal pin, or takes the reload value from the receiver, which depends on the 1394 LLC control and period master and status in the status register 92. Selected based on the state of the periodic source bit. The periodic timer is used to support isochronous data transfer. The cycle time is 32
Bit width. In this embodiment, the lower 12 bits are counted as a modulo 3072 counter, which is a 24.576 MHz clock period,
Or, it is updated every (40.69 nanoseconds). The next 13 high-order bits count 8 kHz (or 125 microseconds), and the highest 7 bits count in seconds.

The cycle monitoring logic 94 executes a logic for performing the cycle monitoring function. Period monitoring logic 9
4 supports isochronous data transfer, monitors LLC operation and handles the schedule of isochronous operations. When the cycle monitor logic 94 receives or sends a cycle start packet, the cycle monitor logic 94 indicates the occurrence of these events by generating a cycle start or cycle receive interrupt. Period monitoring logic 94 also detects the loss of a period start packet and generates a period lost interrupt. When the equal time interval cycle is completed, the cycle monitoring logic 9
No. 4 issues a cycle completion interrupt. Period monitoring logic 94
Signals the transmitter to send a cycle start packet, when the cycle master enable bit is asserted in the 1394 LLC control and status register 92.

The 1394 packet transmission control logic is IE
Logic is executed to control the movement of the IEEE 1394 packet from either the asynchronous transmit FIFO 82 or the isochronous transmit FIFO 84 to cause the physical link interface logic 104 to transmit on the IEEE 1394 bus. The transmission control logic 98 is IEEE1394
Meets functional requirements specified in the 1995 standard.
Packet transmission control logic 98 formats the transmission packet format.

The 1394 packet transmission control logic 98 extracts the quadlets from the asynchronous transmission FIFO 82 and formats them correctly into a 32-bit parallel 1394 asynchronous packet stream. Control logic 98 draws from the isochronous transmit FIFO 84 and formats them correctly into 32-bit parallel 1394 isochronous packet streams. CR for the header and payload portions of the packet using CRC logic
After calculating the C code, the control logic 98 inserts the CRC code into the time slots of the packet stream as required by the type of packet being transmitted. The 1394 packet transmission control logic 98 inputs the parallel packet stream to the physical link interface logic and converts it from parallel to a serial data stream format for transmission to the physical interface.

1394 packet transmission control logic 98
Sends a cycle start packet when the LLC is programmed to operate as a cycle master. 1
The 394 packet transmission control logic 98 sends a 1394 transmission bus request to the physical interface. The physical interface layer arbitrates for the bus and, when a bus grant is received, sends instructions to the transmitter to begin transmission. The 1394 packet transmission control logic 98 performs retry transmissions using the single-phase retry X protocol specified in the IEEE 1394-1995 standard when the busy acknowledgment is returned in response to an asynchronous transmission packet. Further, the 1394 packet transmission control logic 98 sets the packet transmission rate.

Physical link interface logic 104
Executes logic to interface the PCI interface ASIC 20 with the physical layer chip. The physical link interface logic 104 is IEEE1394-1
It meets the requirements of the link physical interface specification in the 995 standard. This function is a PCI interface ASI
Provide C20 with access to physical layer services. The physical link interface logic 104 uses the packet rate code from the transmitter to select the number of serial data streams to generate. Speed code is 100mbp
If set to s, the parallel data stream is converted to two serial data streams, each transmitting at 50 Mbps. Speed code 200mbp
If set to s, the parallel data stream is converted to four serial data streams, each transmitting at 50 Mbps. The physical link interface logic 104 uses the speed indication received by the physical interface to convert the input serial data stream from the physical interface to a parallel data stream input to the reception control logic. For any incoming packet, the physical interface is 100m
PCI interface ASI if received in bps
Generate two serial data streams to C20,
If received at 200 Mbps, four serial data streams are generated. The serial data streams are each clocked at 50 MHz. Physical link interface logic 104 receives serial status responses from the physical interface and converts them to a parallel format. The status response carries a physical interface interrupt indication and / or return data in response to a physical interface register read access request. Physical link interface logic 104 detects and receives serial acknowledgment packets and converts them to a parallel format. In addition, physical link interface logic 104 receives transmitter packet transmission requests or physical interface register read / write access requests and formats them into a serial request stream for transmission to the physical interface. In addition, the physical link interface logic 104 optionally operates as an electrical isochronous barrier between the physical interface and the PCI-LYNX device.

The DMA logic 72 acquires the PCI bus 24 using the PCI master logic 64 and functions as a master device. DMA logic 72 includes a DMA engine 74, which includes a priority state multiplexed common state machine. The DMA engine also includes arbitration logic for operating the channel based on the assigned priority level. In addition, DMA control and status registers 76 provide data path control with the PCI slave 66 for each DMA channel, and access these registers from the PCI interface ASIC 20.

DMA engine 74 executes the state machine logic and fetches control parameters and data buffer pointers from PCL. State machine logic or the packet processing device uses these parameters to control the transfer of data between the data buffers.

FIG. 3 shows the 1394 packet receiving logic 10.
2 illustrates the header comparison logic contained within 2, which is 3
A two-bit packet data stream 106 is received from physical link interface logic 104. The 32-bit packet data stream 106 is an IEEE 1394
Enter the packet receiver logic 102 and the header comparison logic indicated by reference numeral 110. PCI slave 66 also provides 32-bit read / write access to a comparator control register included in number 110.
Line 117 indicates the header compare match output for DMA channel 0. Line 119 indicates a header comparison match for DMA channel 1. Line 121 indicates a header comparison match for DMA channel 2. Line 123 indicates a header comparison match for DMA channel 3. Each of these outputs is connected to the priority encoder 128 and the logical OR circuit 1
Enter 29. The priority encoder 128 generates a DMA channel output 130, which enters the 1394 receiver logic 102. Logical OR circuit 129 provides a comparator match signal to the receiver control logic, where 0 indicates no match, or 1 indicates a match on line 132, which enters 1394 receiver logic 102.

The DMA engine 74 is depicted as having five (5) or more independent DMA channels, all operating simultaneously. In practice, one main control state machine is used, and DMA over time is used.
Multiple division is performed between channels. The priority monitoring isochronous logic continuously examines the current contents of all channels and assigns the highest priority to channels with operations pending execution to the state machine.

FIG. 4 illustrates the general concept described in FIG. 3 in more detail. In FIG. 4, received data 1
06 is header data 1 having words WD0 and WD1.
08. The words WD0 and WD1 are provided in each of the N DMA channel header comparison blocks of the logic 110 (for example, the four channels 0 to 0 in FIG. 3).
3). The DMA header compare register for each DMA channel in logic 110 contains instructions in the circuit, which includes header WD0 at 108 and select register contents 1
20 and the control value logic 122 are compared.
Similarly, WD1 at 109 is compared with the contents of register 124 and the contents of select register 126 using a compare instruction as shown. The outputs of the DMA header comparison register and logic are represented as channel selection [0] through channel selection [N-1]. FIG. 4 further illustrates
The priority encoder 128 receives the channel selection output, generates a channel number on the DMA channel selection line 130, and outputs an address match output 132. As FIG. 3 and the accompanying text explain,
The MA channel select output 130 and the address match output 132 correspond to the IEEE 1394 packet receiver logic 10.
Flowing to 2.

The DMA header comparison register and logic 110 and the priority encoder 128 perform the logic necessary to determine whether an incoming packet is to be received and loaded into the GRF 80.
FIG. 3 shows four DMA channels, and five or more DMA channels are within the scope of the present invention. Each DMA header compare register and logic 110
Are assigned to take charge of one DMA channel. One comparator includes a WD0 field select register 122, a WD1 select register 126, a WD0 compare value register 120, a WD1 compare value register 124, and comparison isochronous logic. The two field selection mask registers specify the bit fields in WD0 and WD1 of the incoming packet, which are matched by the comparator logic with the expected values. The two compare value registers specify the expected bit pattern, which is matched against the selected bit fields in word 0 and word 1 of the incoming packet. The priority encoder 128 collects a DMA channel match indication from each DMA header comparison register and logic 110 and generates a bit code for mapping incoming packets to a particular DMA channel. OR logic 129 combines the DMA header comparison register and the selected indication from logic 110 to produce a single comparator match indication for IEEE 1394 receiver logic 102. The IEEE 1394 receiver logic uses the DMA channel number, and the comparator match indicator to determine whether an incoming packet should be accepted by the GRF 80.

In this embodiment, the DMA packet processing device operates as an independent DMA channel, all of which are executed simultaneously. An actual implementation would use one master control state machine, which may have multiple DMA channels (eg,
Multiplex time division among five DMA channels). Priority monitoring superisochronous logic continuously examines the current contents of all channels and assigns the highest priority of pending functions to that channel so that the state machine executes. To The DMA channel initializes to a static state after reset, waiting for a valid PCL pointer to be written to the packet control list start address register, and the channel enable and link bits to be set in the DMA control register. The valid PCL pointer is determined by the state of bit 0 of the packet control list start address register. 1 indicates an invalid address, and 0 indicates a valid address. The DMA then goes to the address pointed to by the packet control list start address register, fetches a new address and, if valid, starts execution with this as the current PCL address.

If this address is invalid, the link bit is cleared in the DMA control register, a DMA stop interrupt is generated for this channel with the associated state in the interrupt status register, and the channel is disabled. Becomes This mechanism provides a sanity check on the PCL memory structure, as well as a relatively simple way to continue channel PCL execution if the next address link is lost. When a valid next PCL address is detected, the DMA stores BUSY in the DMA control register.
The bit is set and the first control word of the PCL offset section is fetched. Subsequently, a check is performed to determine whether the command is reception, transmission, PCL or local bus direction, or an auxiliary command. From the priority encoder 128, depending on which DMA channel is matched, the highest priority channel with a match identifies the particular DMA channel to be selected.

FIG. 5 shows a designated DM based on this embodiment.
Further details of the operation of the A channel header comparison logic 120 are shown. For example, received data including words WD0, WD1 enters header comparison logic 120. This includes the code segment 134 of the word WD0, which is sent to the destination ID comparison logic 136. Part 13 of WD0
8 is sent to the transfer code comparison logic 140. WD0
Are sent to comparison logic 144 where WD0 bits 15: 6 and 3: 0 are compared. In addition, comparison logic 146 receives bits 31:16, shown as part 148 of WD1. Destination ID comparison logic 1
The comparison logic output from 36, transfer code comparison logic 140, comparison logic 144 acting on WD0 bits 3: 0 and WD0 bits 15: 6, as well as WD1 bit 3
The output of comparison logic 146 for 1:16 is input to AND function 150. The output of AND function 150 is the channel select output [x], which is channel select [0] through channel select [N-1], which priority encoder 128 receives as described above. And the DMA channel selection signal 130 and the address match signal 1
32 is generated.

FIGS. 6, 7, 8 and 9 show details of the received data bits sent to the DMA channel header comparison logic 120. In particular, FIG. 6 shows [7:
4]. Similarly, the transfer code bit 164 is the bit [7:
4] is formed. In the designated DMA channel, the matched data set 168 includes a WD0 matched data header 170 corresponding to header WD0162 and a selected matched data set 172 corresponding to header 166. As shown in FIG. 6, the [7: 4] comparison bits 174 are the comparison selection WD01.
It is changed by the [7: 4] bits 176 of 72, and the matching result for the reception header WD0 bits [7: 4] is determined. The output of this comparison equal time interval is ANDed with the result of the other partial comparison equal time intervals to determine a match value 117 for the particular DMA channel.

FIG. 6 shows a possible target data value array for a particular field. The present invention captures the received data one field at a time and attempts to set a comparison word and a selection word to indicate a possible way to receive a particular result. In particular, FIG. 6 shows a simultaneous comparison for the transfer code 160 field. One of the transfer codes 160 or 164 provides an opcode, which identifies the data type on the receive data bus 16 and 1394 serial data as isochronous or asynchronous data. In the example shown in FIG. 6, the transfer code exists in the same field. This simultaneous comparison works for both asynchronous and isochronous headers as well. There may be different ways to decode the isochronous and asynchronous headers. They may be as shown or integrated with each other. The actual coding will vary, but the operation is similar in that it will ultimately compare or not.

FIG. 7 shows the [3: 0] bits 178 and [15: 8] bits 180 of the asynchronous header 162, as well as the [3: 0] synchronous bits 182 and [15: 8] channel bits of the isochronous header 166. 184 is emphasized. [3: 0] bits of header WD0, primary bit 17
Either 8 or the synchronization bit 182 is compared with the [3: 0] bits 186 of WD0, indicated at 170, modified by the [3: 0] bits 188 of the comparison selection WD0, indicated at 172. Further, [15: of the header WD0 indicated by 170]
7] bits, either the [15: 7] bits 180 of the asynchronous header 162 or the [15: 7] bits 184 of the isochronous header 166, the comparison selection WD0 indicated by 172
Is compared with the [15: 7] bits 190 of the comparison WD0 indicated by 170, modified with the [15: 7] bits of FIG.
The results of these match determinations occurring at WD0 bits [3: 0] and [15:
7].

FIG. 8 shows yet another simultaneous comparison, corresponding to the destination ID comparison logic 136 of the DMA channel comparison logic 110. Asynchronous header 162 indicates destination I
D bits [31:15], which are denoted by reference numeral 200.
It is represented by The [31:16] bits of the header WD0, indicated by these same bit positions, 166, include data length bits [31:16], which are represented by reference numeral 202 for the isochronous data packet. Similarly,
Bits [31: 1] of each link layer control register 92
6], here indicated by reference numeral 204:
Contains node number and bus number data. FIG. 8 shows that the asynchronous reception header WD0 indicated by 162 or the equal time interval reception header WD0 indicated by 166 has the data set 170 bit 2 changed according to the data set 172 bit 208.
FIG. 14 illustrates a simultaneous comparison of whether the bits match the [31:16] bits of 06. Matching data set 210 includes bits [15:11] as destination ID set bits 212.
Based on the simultaneous comparison performed in FIG. 8, a DMA channel number is derived in logic block 146, which is sent to the AND function 150 of the DMA channel comparison logic 110.

The upper bits [31:] of the asynchronous header WD0
16], the IEEE 1394 standard is a method of handling a destination identifier, so that it can have a special meaning. For example, IEEE 1394 encoding specifies messages in broadcast format in different headers. With respect to the receiver header, there is a comparison word 0 and a bit selection word 0, by which a particular reception header word can be selected. In addition, each bit in the comparison selection word 1 [15:11] 212 individually selects a specific destination ID simultaneous comparison formula.

FIG. 9 shows an asynchronous header 162, which includes the source ID bits [31:16] in 220. These bit positions 222 in the isochronous interval header 166 are part of the variable data. Within the WD1 comparison logic 146, the DMA channel 120 indicates bits 16-31 of the matched data set 170 modified based on the comparison selection word 1 bits [31:16] indicated at 226 as header WD1 bits [220 or 222]. 31:16].

FIG. 10 shows a register write circuit 250 for writing an arbitrary number of data register bits in one register write operation. Circuit 25
A 0 illustrates one of any number of similar circuits. Each circuit controls one particular bit of any number of data bits. In the register write circuit 250, a general-purpose I / O (GPIO) write data input 252 is connected to a data flip-flop 254. Data flip-flop 254 also receives clock signal 256 and writable input 258 from AND function 260. AND function 260 is a GPIO address acknowledge signal 2
62, a write standard signal 264 and a GPIO address bit input 266 as inputs. The AND function 260 controls the write enable bit 26 of the data flip-flop 254.
8 is provided with a writable output 258. Each designated address bit 266 is the flip-flop 2 of a1.
The GPIO write data 252 is written into only the 54.

The register writing circuit 250 enables writing only to bits that need to be changed in the register, and saves the previous values of the remaining bits. Register write circuit 250 also allows software to read, specify, or change significant bits without manipulating register bits that are not involved in a particular operation.

If the address of the register exists, the write enable bit 258 and the GPI of 1 to 4
O is written based on the address field. [G
[PIO] Ax in the address number input 266 represents an address bit that is a basis of this writing, and it is determined whether or not to write to this bit. Write strobe 264 signals that this is a write operation. GPIO address acknowledge 262 is a basic address decode signal for all registers. Writable input 258 to flip-flop 250 is the output of the AND function. When write enable is active, input write data 252 is written to flip-flop 254,
Appears at flip-flop output 270.

FIG. 11 shows the address field 272, which includes the GPIO register address bits 274 and the individual bit select field 276 address bits A
0, A1, A2 and A3, which are based on the value of the associated bit in the address field 272, 0 or 1. The last two bits 278 of the address field 272 are usually considered to be a value of zero. The least significant two bits in address 272 are assumed to be zero, which is a requirement of the bus architecture. Next A0, A1,
The four bits designated A2 and A3 allow 4-bit addressing individually or in different combinations. The higher or most significant bits are individual addresses that are specific or assigned to this individual function. Therefore, there is a fixed combination of bit patterns, and a specific address is selected by addressing a given GPIO address value. As a result 0
An entire array of addresses from 000 to 1111 (ie, 16 different possible combinations) can be addressed.

The DMA priority selector is responsible for activating the contents of the channel that has the most urgent need for PCI data movement. DMA is the main state machine (d
ma_fsm. v), which implements the state of the selected channel. The selected channel simultaneously has the stored condition number of any one point. These stored conditions are called the context. Part of the context of the channel is its current state running in the main state machine.

The state machine has several idle states,
In this state, it is necessary to wait until a certain condition that can continue the execution of the channel is satisfied. One obvious case is when the channel is idle or waiting until a valid PCL address is loaded, a channel enable bit is set, and a link bit is set. Otherwise, if the receiving channel is
It is time to wait for the data in O. If one channel is waiting for received data and another channel is
If the FO is ready for a data transfer, we need to have that transfer performed. This type of scenario occurs in many cases.

One of these idle states is the “window of opportunity”.
Where processing of another channel is executed.
The priority selector monitors all these possible idle states and selects which state and channel has the highest priority at that time.

FIG. 12 shows a DMA processed in the environment 300.
Here is an example of an instruction. In FIG. 12, PCI register write data 302 is input to a multiplexer 304. The multiplexer 304 also receives DMA register write data 306. The appropriate data source is selected at 308 based on the operation to be performed. The output of the multiplexer 304 is input to a DMA register 310 and is shown in the figure to include, by way of example, six channels, channel 0 through channel 5. For each channel, the DMA field is the previous address or provisional field 312, the current PCI address field 314, the data buffer address 316, the state 32
0, instruction 322, current state 324, ready bit 326
including. DMA register write arbitration select circuit 328 also provides an input to DMA register 310, which is a PCI slave signal 330, a PCI master signal 33.
2. Respond to DMAPCI master period input 334 and slave end input 336. DMA channel arbitrator 3
40 receives lock channel input 342 and provides an input to multiplexer 346. Slave register read address input 348 controls the operation of multiplexer 344. DMA register read data output 35
0 returns the DMA register data to the PCI interface. Multiplexer 346 provides its output to DMA state machine next state logic 352. The DMA state machine next state value 354 is written back to the state register and other output registers via multiplexer 304. The output from the DMA instruction processing environment 300 is FIFO
Includes read, write, and select signals 356 and a number of signals.

The DMA channel arbiter 340 controls which channel is to be executed. This mediation is FI
Based on signals from FO 358, such as "FIFO active channel requires data" and "FIFO active channel" as well as which FIFO is ready for data transfer and which channel is enabled and active. Done.

The DMA register 310 executes control and setting of the status register, and controls each DMA channel and monitors the status. DMA registers 310 support a number of functions for each DMA channel. For example, when the previous packet control list start address / temporary register 312 processes the queue during the asynchronous transfer, the DMA engine 7
4 provides the register to be updated. It is also used as a temporary save register when auxiliary instructions load and store data. The packet control list start address register 314 is initialized by the application software and specifies the starting point of the first (dummy) PCL in the PCL chain. DMA engine 74 links to the first real PCL using the next address loaded into this PCL. The packet control list start address register 314 stores the active D when the PCL is processed.
Updated by MA channel. The DMA buffer start address register 316 is loaded with the data buffer pointer fetched from the PCL when the operating DMA channel processes the PCL. DMA status register 320 stores an ongoing count of the number of bits transferred during this PCL and includes the completion status of the transfer. PC
After the processing of L is completed, the active DMA channel writes the status information of this register back to the location at offset 0xC in the PCL.

The DMA control register 322 contains control bits that allow application software to enable or disable operation of the DMA channel and recapture the next address of the PCL to establish a link. The DMA control register 322 stores a data buffer transfer control, a transfer bit count value, and an instruction fetched from the PCL. The least significant bit of the DMA ready register 326 indicates that the DMA channel is XMT, RCV, LOA.
Before the execution of the D, STORE, STORE0 or STORE1 instruction is continued, it is possible to wait until the ready state. This ready state is indicated by the control word (1
One or more). The least significant bit of the DMA ready register 326 is set during execution of the BRANCH instruction.
Enable conditional branching of DMA channels. This condition is selected in the control word (s) of the PCL. The current DMA status register 324 stores the status vector of the DMA channel. This register is updated during the active time of the DMA channel and holds the last state vector generated just before the channel stopped.

The DMA register 310 can also include a receive packet count register (not shown),
It contains the currently received packet count. DM
The A engine 74 loads into this register the count of received packets that have passed within the GRF80 token word. This count is decremented each time the data is transferred to the PCI bus 24. It may also include a DMA global register (not shown), which includes a state flag that the state machine uses in tracking the execution of the asynchronous transfer packet. This DMA global register stores the lower limit bits used with the cache line size register and determines the burst size required by the PCI master.

FIG. 21 shows how the DMA channel is initialized after being reset to the quiescent state in section 135. Each DMA register 310 waits until a valid PCL pointer is written to the packet control list start address register 314, and the channel enable and link bits are set in the DMA control register. Effective PCL
The pointer is determined by bit 0 of the current packet control list (PCL) address register. A value of 1 indicates an invalid address, and a value of 0 indicates a valid address.
Next, the DMA goes to the address specified by the current PCL address register start address register, fetches the next PCL address, and if valid, starts execution with this as the current PCL address. If this address is invalid, the link bit in the DMA control register is cleared, a DMA stop interrupt is generated for this channel, and at the same time the associated status is generated in the interrupt status register. Stopped. This mechanism provides a sanity check of the PCL memory structure, as well as a relatively simple way to continue channel PCL execution if the next address link is lost (ie, invalid). Next PCL in which DMA engine 74 is valid
Upon detecting the address, DMA engine 74 sets the BSY bit in DMA control and status register 76 and fetches the appropriate PCL word. A check is then performed to determine whether the instruction is for receive, transfer, PCI or local bus, or auxiliary instructions.

FIG. 21 is a block diagram showing the DMA in section 137.
FIG. 7 illustrates how the engine 74 performs an operation of receiving the asynchronous data at equal time intervals in the GRF 80. DMA engine 74 performs a check to see if a standby condition exists. If the wait state no longer exists,
The processing unit enters the data transfer phase. Where DMA
The engine 74 enters a loop and checks whether the transfer count value is currently zero. In that case, this is PC
A check is made to see if it is the last data buffer in the L buffer list. If this is the final buffer and the indication of packet boundaries by writing the special control token word into the GRF 80 has not been done by the link layer control logic 90, an error will occur but this will indicate that more packets can be held by the buffer. This is because data will be transferred. In this case, the PKT ERR bit is set in the DMA control and status register 76 and the DMA engine 74 flushes the remaining data to the packet boundary. If the transfer count is currently decremented to zero and another buffer exists in the PCL list, DMA engine 74 obtains the new buffer address and transfer count and performs the transfer.

While moving data from GRF 80 to PCI interface logic 70, DMA engine 74 waits until GRF 80 has enough data to
It issues a transfer execution request to the CI bus master. The transfer threshold is reached when one or two conditions are met. The DMA engine 74 issues a transfer request to the PCI master when the number of bits in the reception FIFO reaches the “high water mark”. This high water mark is equal to or greater than the cache line register or the lower boundary field of the DMA global register.

The DMA obtains information on the data size of the packet from the link when the packet is first
Is written by the link layer control logic 90. The transfer count is used to determine if the data in the GRF 80 is the remaining data in the packet, and if so and its size is smaller than the high water mark, the transfer count is PCI equal to
Request transfer of master logic 64. While the DMA engine 74 transfers data, the data buffer start address register and the data buffer transfer length bit in the DMA control and status register 76 are updated to reflect the current state of the transfer.

When the link layer control logic 90 encounters the end of a packet, as indicated by section 139 of FIGS.
Write to F80 to mark the end of the packet. Included in the control word is a status bit, which indicates the completion status of the packet on the bus. DMA
The engine 74 uses the end mark of this packet to perform GRF8
The data transfer from 0 to the PCI bus 24 ends. If the end-of-packet signal indicates that an IEEE 1394 busy acknowledgment signal is present, the DMA engine 74 reacquires the first buffer address and transfer count of the PCL and begins the entire transfer of the packet. . If the busy acknowledgment state is not indicated from the end of packet indication, the DMA control and status register 76 is loaded with the acknowledgment state passed from the link layer control logic 90 into the end of packet indication and the packet complete bit Is set. The packet completion status is then written to memory in the PCL status word and an interrupt signal is issued and latched in the corresponding bit in the interrupt status register. If the command was a receive and update command, the remaining transfer count and next buffer address are written to the appropriate PCL offset.

FIG. 13 illustrates a table 360, which illustrates the operation of the DMA channel arbitration circuit 340. Table 360 shows that if a channel is currently operating on an IEEE standard 1394 bus, that channel has the highest priority in the DMA channel execution schedule. Otherwise, channel priorities are ordered by channel number, with 0 having the highest priority. "
The "ignore" value is assigned the value "X" in portion 362, which means that the currently active channel has the highest priority, whatever the other values. Channels are scheduled on a first-come, first-served basis.

The operation of the DMA instruction processing is shown in the following example.
Assume that the DMA is currently operating on channel 3.
Therefore, all registers for channel 3 are DMA
Is selected via the multiplexer 346 in the table. DM
A selects an individual slice of the register to appear in the current state and next state logic on the DMA state machine.

FIG. 14 shows a simplified diagram of the personal computer environment 12 of the present invention, which includes an autoboot function 370. Pin 370 of PCI interface ASIC 20 is an auto-boot PCI interface ASIC.
Provide direct input to 20. As a result, the PCI interface ASIC 20 becomes the PCI device # 30.
1. Operates as a host device via the PCI bus 24 for the PCI device # 2 of reference numeral 31. The autonomous operation command for the PCI interface ASIC 20 is RPL
ROM 36 and SRAM 38 would be used.
Further, the PCI interface ASIC 20 communicates with the physical layer interface 18 through the interface bus 16.

When the auto boot pin 370 is operating (ie, when lifted), the auto boot mode of the present invention is selected. Autoboot mode allows for many features, which allow the PCI interface ASIC 20 to function autonomously. The autonomous operation includes the following functions. The first packet control list address is fetched using the DMA channel 0 after a power reset. Also, after the power is reset, the automatic boot mode sets the DMA master to the external RPLR
Provide access to the OM. In addition, after a power reset, the automatic boot mode allows the DMA master to access the internal link registers.

Once enabled as a master on the PCI bus 24, the PCI interface ASIC 20
Issue I configuration, input / output, and memory read and write commands to PCI bus 24 to identify the appropriate address range in controlling the packet control list. During operation in the auto boot mode, the external PCI address space is 30
Limited to bits. The values of the two most significant address bits are always zero. Internally these two bits are used to select a PCI instruction.

The state of the autoboot pin 370 can be read from special bits in other control registers for diagnostic purposes. Selected autoboot mode and external RO
M, the present invention provides a PCI interface ASIC2
0 as a local processing device, and all internal PCs
It is possible to set up the I-interface ASIC 20 registers, initialize other devices on the PCI bus 24, and build and connect multiple other PCLs. Various DMA channels can perform these PCLs to cause data transfer on the IEEE 1394 bus.

By adding an external local bus RAM to the PCI interface ASIC 20, a PCI slave memory is provided in a device on the PCI bus 24, control information can be obtained, and a local memory for data transfer can be provided. . The PCL program can transfer device control / data to other systems via IEEE 1394. Therefore, the environment employing the automatic boot mode of the present invention can be used in peripheral equipment that cannot use a suitable processing device for managing the PCI interface environment.

The circuit of FIG. 14 provides another mode of operation for the chip via external pins. The effect is to enable part of the memory map and pre-specify the DM at power up.
A sufficient resources that can be accessed by the A engine 74,
It is to ensure that all required functions work effectively. The effect is to modify the operation of the DMA engine 74 so that when the DMA machine is in a particular mode, the machine will stop operating instead of requesting new instructions from a dedicated ROM.

The DMA engine 74 is running and acquires an address to get an instruction. This allows PC
The I-interface ASIC 20 can operate in the independent mode. This is also the PCI memory instruction,
And all input / output instructions from the PCI interface ASIC 20 can be generated.

By using the auto-boot pin 370, the present invention enables access and initialization of ROM and RAM based address registers,
The power-on memory map can be reconfigured. This allows the DMA machine read and write access after power up. Further, the present invention changes the behavior of the DMA engine 74 when the automatic boot mode is selected. This makes it possible to obtain a new instruction from the dedicated ROM address when the power is turned on. The initial access address is exactly 0 in this embodiment. The present invention further provides a method for generating bus instructions and protocols that are not required in a normal environment.

The DMA engine 74 can acquire and execute instructions from the RPL ROM 36 when the automatic boot option is selected. This allows the DMA engine 74 to generate the instructions required to operate as a master device on the external PCI bus, and configure, initialize, and manage other PCI devices on this external bus.

The table 380 in FIG.
0 when the internal PCI address bus [31:
30] bit to instruction on the PCI bus. For example, when the value of bit 31 is 0 and the value of bit 30 is any value, a PCI memory instruction is operated. Thereafter, when the value of the address bit 31 becomes 1 and the value of the address bit 30 becomes 0, a PCI input / output instruction occurs. If the value of the address bit 31 is 1 and the value of the address bit 30 is 1, the PCI configuration instruction is executed by the PCI interface ASIC.
Sent to 20.

FIG. 16 shows a diagram of a local bus interface block 390 according to one aspect of the present invention. Local bus interface block 390 includes a local bus configuration register 392, a zoom video (ZV) decode interface block 394 that communicates with the ZV machine 396. Pack / unpack state machine 398 includes an address / data / bit enable holding register 398. Local bus interface block 390 also includes local bus interface state machine 400 and slave acknowledgment interrupt block 402.

The ZV port in FIG. 2 is an output-only port for transferring data from the IEEE 1394 standard bus to an external device on the PCI interface ASIC 20.
When properly programmed, the ZV interface logic provides a way to receive IEEE 1394 digital camera packets and transfer the payload to external ZV compatible equipment with appropriate control signals.

The ZV decoding circuit 394 is accessed via a sub-set of the third PCI memory compliant address register.
When the ZV decoding circuit 394 is enabled, the area between the auxiliary addresses 0xF000 and 0xFFFF is mapped to the ZV port. The ZV port is enabled when one of the six available clock sources 395 is selected as the ZV pixel clock. If none of the six clocks is selected, the ZV port is disabled and the auxiliary interface requires full address space. When the ZV port is disabled, all ZV related outputs are neutral except for the data bus, which is driven during auxiliary, RAM and ROM accesses.

When the ZV port detects that the IEEE 1394 isochronous packet header synchronization field is equal to 0x1, a vertical synchronization signal is generated. In the preferred embodiment, data bits 24 include significant sync field data. When this vertical synchronization bit is detected, a vertical synchronization output is generated. For the remainder of the frame, a horizontal sync output is generated each time a special address is accessed when transferring video data to the zoom port. By properly programming the packet control list, all IEEE 1394 digital camera packets can be transferred through the ZV port.

FIG. 17 shows one embodiment of the synchronization detection circuit 405 of the present invention, which is for generating a vertical synchronization detection signal 407 and a horizontal synchronization detection signal 409.
The value of the PCI slave address bits [15: 0] is 0x
When F000 is assumed and slave data bit 28 is assumed to be a binary value of one, comparison circuits 411 and 413 send a signal to AND gate 415. This produces a vertical sync detect signal on line 407. Similarly, when the slave address bits [15: 0] send 0XF004 to the comparison circuit 417, the horizontal synchronization detection signal appears on the line 409.

FIG. 18 shows one video scan line for a packet zoom port address map 410 according to the present invention. Zoom port address map 410 has address 0X
An equal time interval packet header 412 beginning with F000;
Address 0, shown in detail by reference numerals 414 to 424
Includes an isochronous packet data space that starts with XF004 and ends with 0XFFFF. Each equal time interval packet has the address 0XF000 in the zoom port address space.
The first data payload quadlet 428 is transferred with the header quadlet transferred to
The packet is transferred to the XF004, and the remaining quadlets in the packet are transferred to the addresses in order until the end of the packet payload data reaches the address (0XF000 + (N-1)) in the case of the packet N quadlet length.

FIG. 18 shows that the present invention can autonomously display a video image display in a window managed by hardware without the assistance of software. The present invention allows raw data to be sent out on the IEEE 1394 bus and flow into the zoom port in a format compatible with entering the video controller chip as input data. This allows data to be displayed autonomously while maintaining horizontal and vertical frame synchronization.

In operation, the zoom video port is mapped into an address space. The header contains frame synchronization information or signals. By constructing the correct control structure, the header packet data is forwarded to a specific address based on the address where the logic looks for the field.
For Sony cameras conforming to the IEEE 1394 digital camera specification, the bits in the isochronous header sync field indicate the start of the video frame. By decoding this particular isochronous header synchronization field bit, a vertical synchronization signal is generated. This information can be used to synchronize the reception of zoom port data to the zoom port.

There are two cases to achieve this, 1
One is horizontal synchronization, and the other is vertical synchronization. For vertical synchronization, a specific address and a specific isochronous header synchronization bit are required for decoding. The horizontal synchronization is generated by decoding another specific address.

The video zoom function of the present invention converts control information, such as horizontal or vertical sync signals, from packetized data, including embedded control information, such as encoded video.
Provide a function to extract autonomously. Simultaneous comparison logic 4
11 and 413 detect the transfer to the destination address specified by the specific data value.

The present invention also provides a software or hardware control structure for specifying appropriate destinations for different segments of packet data. DMA
The architecture allows a particular packet to be broken down into different lengths for different destination addresses and collected or delivered. Therefore, the present invention can write the header field to a specific address and write the data field to another address. Using this feature, the present invention can control which part of the PCI interface device sees which address.

The present invention further provides a programmable counter, which is used to generate basic control signals for multiple access to a particular address location and / or data pattern. This gives the option of using a counter to provide multiple packets per horizontal scan line or during horizontal sync. Therefore, the horizontal synchronization signal is generated immediately after the vertical synchronization. This allows the system to count the number of passes through a particular address. The system can generate another horizontal synchronization signal only after repeating the address four times, for example. This makes it possible to combine four packets into horizontal scanning lines according to a specific camera, resolution, and the like.

FIG. 19 shows a procedure for mapping a single video packet to a plurality of video scan lines using the zoom port address map 410. Zoom port address map 4
10 includes an isochronous packet header 412, which is adjacent to the isochronous packet data 414. The memory spaces 416 to 424 are the same as those described in FIG. Zoom video isochronous packet 440 includes header quadlet 0, which should include the start of frame bit in the isochronous header synchronization field. Quadlet 1 is the first video data payload quadlet, which includes the beginning of video scan A. The horizontal sync signal uses this quadlet at address 0
Generated by writing to XF004. Similarly,
Writing data quadlet 442 to address 0XF004 indicates the beginning of video scan A + 1 by generating a horizontal sync. The writing of data starting from 0XF000 and continuing until [0XF000 + (N-1)] is actual data output through the zoom video port, that is, zoom data. The header is stripped and is not part of the data field.

Accordingly, in this case, the vertical synchronization and the horizontal synchronization are generated for the first packet of the video frame. Subsequent packets represent subsequent scan lines that fill the rest of the video frame header, but these headers do not include a sync field. The data flow is the same for all packets, each point having a header 0X
Starting at F000, if no sync bit is present in the header, no vertical sync bit is set. If written to F004, a horizontal sync is generated. This creates the start of a scan line, which causes the remaining data to flow out as part of the scan line.

Further, there are situations where multiple scanning lines exist for each packet. To handle this, a packet control list is used.

The DMA engine 74 of the present invention is controlled by a data structure called a packet control list or PCL. The PCL contains instruction information, which the DMA retrieves from memory as needed. These instructions tell the DMA the source and destination of the data, and how much bytes to transfer. Some instructions transfer chunks of data to IEEE 1394 transfer FIFO and PCI bus 2.
4 or between the general-purpose receive FIFO 80 and the PCI bus 2
4 and move it. Another instruction moves data between the PCI bus 24 and the auxiliary port local bus 26. Other instructions are secondary functions and are called auxiliary instructions. These auxiliary instructions allow the DMA to pry and manipulate the quadlet of data specified for any PCI address, and allow conditional branching using PCL. The intended use is to allow the DMA to perform special data movement control and to function as a stand-alone processor that can build the PCL in an automatic boot sequence. The full scope of this function is not organized,
Other uses for MA will be developed.

The application software of the present invention programs the operation of the DMA channel using a PCL data structure residing in host memory. The application software is responsible for building the PCL and placing the memory on the storage device. The PCL is organized as a contiguous set of memory arrays, which contains instructions, control parameters,
And a data buffer pointer, which are DMA
Required when a channel transfers one IEEE 1394 data packet, or moves data between PCI bus 24 and auxiliary port local bus 26, or executes one or more auxiliary instructions. is there. The total number of memory arrays required by the PCL is generally limited to 32 quadlets in this implementation.

As a minimum requirement, the PCL start address is aligned at the quadlet boundary. To optimize the performance of the DMA, it is recommended that the PCL start address be a line on a cache line boundary. The data buffer pointer can be arranged on any byte boundary. In order to optimize DMA performance, the present invention arranges data buffer pointers on cache line boundaries. If this is not possible, it is sub-optimal to arrange the data buffer pointers on quadlet boundaries. The total data buffer size specified by PCL is approximately 1 kilobit when the IEEE 1394 bit rate is 100 Mbps, or
When the EEE1394 bit rate is 200 Mbps, it is limited to about 2 kilobits.

In this embodiment, the active DMA channel fetches command and control parameters from the PCL,
Using these, the channel itself is constructed and the transfer command is executed.

An application software program enables a DMA channel to transfer a plurality of IEEE 1394 data packets by chaining a plurality of packet control lists in a PCL list queue. The present invention configures this queue by setting the next address field of each PCL to indicate the starting address of the next PCL in memory. The last PCL in the queue can be programmed to stop DMA processing, return to the head of the queue, or point to a new queue. A PCL containing auxiliary instructions can be incorporated anywhere in the PCL queue, but not both at the same time. The PCL queue can mix receive and transmit or auxiliary instructions; however, an asynchronous transfer instruction must be next to another asynchronous transfer instruction, because it performs pipelined transfers and
Also, there is a possibility of retrying the packet. On the other hand, setting the "wait for status" bit in the asynchronous transfer instruction eliminates this request.

FIG. 20 illustrates an example of a linked list of a plurality of packet control lists according to an embodiment of the present invention. In FIG. 20, PCI queue 450 begins at start address 452. From the start address 452, the processing flow enters the dummy packet control list 454. The dummy packet control list 454 sends the processing flow to the transfer command packet control list 456. In the transfer command packet control list 0, the data buffer count command and the data buffer address information are, for example, data buffer 460,
The data buffer 462 goes to the data buffer 464. In this example, 14 data buffers are used.

The transfer instruction packet control list 0 indicated by reference numeral 456 includes a next list address 466, which is the auxiliary instruction packet control list 1 indicated by reference numeral 470.
Connected to Auxiliary instruction packet control list 1, indicated by reference numeral 470, includes a load instruction 472, a source address at 474, and a store instruction and location 476. Load instruction 472 is connected to data register 478, which provides DMA register 480. Store 1 instruction 4
76 provides input to memory location 482. Reference number 4
Auxiliary command packet control list 1, also indicated at 70, also includes a destination address 484 when the condition is satisfied, which sends the PCI process control to the transfer command packet control list 2, indicated by reference number 486, depending on the condition. The transfer instruction packet control list 2 indicated by reference numeral 486 is the next list address 4
88, which is for connecting to another PCL as indicated by reference numeral 490, while the data count and data buffer address 492 are connected to a data buffer 494, and so on. . If the destination address 484 when the condition is satisfied controls the flow of processing of the PCI queue, the next list address 468
Is connected to the transfer command packet control list 2 indicated by reference numeral 486. This information is then sent from the next list address to another PCI as indicated by reference numeral 490, the data buffer 0 address and the count command information are sent to data buffer 1, and so on.

Typically, when the host CPU 44 operates, a memory is arranged to construct a packet control list and a dummy next PCL address.
Is not present. The dummy next PCL address provides the memory location, which is the start pointer of the first full PCL, PCL0. The host processor then builds the list of required PCLs and causes either type of data transfer or auxiliary instruction to be performed, for example, the operation of an auxiliary instruction that actually performs an instruction equivalent to the processor.

The space in the next PCL address includes an error address, a reserved position, and a state. The next pair is the control state, transfer count. Next to that is the data buffer address. These two long words make up a data transfer to one instruction or one data buffer. An example transfer is a buffer address that points to a data buffer location in memory to which the data is transferred or transferred. If data is continuously transferred, the present embodiment allows a large control loop for continuous data transfer to be formed.

The operation of reception, transmission, or auxiliary operation between the PCI and the port local bus according to the present invention will be described with reference to FIGS. 21 to 30 and the following description. In the receive operation, as shown in FIG. 21, for equal time intervals and asynchronous data in the GRF 80, the process flow begins by checking whether a wait state exists. FIG.
If the channel enable is equal to 1 for an isochronous transmission, as shown at 7, the transmission function is ready and the cycle start bit is acknowledged and there is no wait. The wait state is determined by the wait selector bit of the data buffer control word zero. When the wait state no longer exists, the processing unit enters the data transfer phase. Here, a loop is entered to check whether the current transfer count value is zero. If so, a check is made to see if it is the last data buffer in the PCL buffer list. If the data buffer is the last data buffer and the link layer controller does not indicate the packet boundaries, which is done by writing a special control token word into the GRF FIFO, an error occurs, This is because more packet data than the buffer can hold is about to be transferred. In this case, the packet error bit is set in the DMA status register, and the DMA pushes the remaining data to the packet boundary. If it is determined that the current transfer count has been reduced to zero and there is another buffer in the PCL list, the DMA obtains a new buffer address and transfer count and continues the transfer.

While moving data out of the receive FIFO in the PCI interface, the DMA waits until the FIFO has enough data before making a transfer request to the PCI bus master. The transfer threshold is reached when one or two conditions are met. DMA is a reception FIF
Whenever the number of bytes in O reaches the "high water mark", PC
Request transfer of I master. This high water mark is equal to the upper bound of the cache line size register or the lower bound field of the DMA global register. DMA obtains information on the data size of a packet from the link when the packet is first written into the FIFO by the length layer controller. It uses this transfer count to determine whether the data in the FIFO is the remaining data in the packet, and if so, if its size is less than the high water mark, its transfer count is equal to this remainder. Request transfer to PCI master. While the DMA transfers data, the data buffer start address register in the data buffered in the transfer length bits in the DMA control register is updated to reflect the current status of the transfer.

When the link layer controller reaches the end of the packet, it writes a special control token word in the FIFO to mark the end of the packet. Included in the control word are status bits that indicate the completion status of the packet on the bus. The DMA uses this packet end marker to end the transfer of data from the FIFO to the PCI bus. End of packet marker is IEEE
If it indicates a 1394 busy acknowledgment, the DMA reacquires the PCL first buffer address and transfer count and restarts the packet transfer. If the busy state is not indicated from the end-of-packet marker, the DMA status register is loaded with the acknowledgment state sent in the end-of-packet marker from the link layer controller and the completion of the packet is set. This is then written to memory in the PCL status word along with the number of bits transferred to this PCL. If the INT bit is set for a data buffer control / byte count instruction in the PCL, an interrupt signal is issued and latched into the corresponding DMA PCL bit in the interrupt status register.
If the command is a receive and update command, the remaining transfer count and the next buffer address are written to the PCL. The DMA then determines whether another PCL is linked to the current PCL by taking the next list address field of the current PCL. Linked PC
If L is present, the DMA returns to its linked P
Execution is continued with CL as the current PCL. If another PCL is not linked to the current PCL, a DMA stop interrupt is generated in the interrupt status register with the associated status for this channel and the channel is idle.

With respect to the DMA asynchronous transfer operation, an asynchronous transfer is determined after a valid PCL pointer has been written to the packet control list start address register and the channel enable and link bits have been set, as shown in FIGS. . The ultimate purpose of the asynchronous packet processing device is to allow the link layer controller to convert from FIFO to IEEE.
One packet is to be left behind the current packet transferred to the E1394 bus. From a DMA perspective, this packet on the bus is the previous packet. All states reported by the link layer controller are considered to be for this previous packet; however, by setting the "wait for state" bit in the "data buffer / byte count / instruction" in the PCL, This pipeline operation is prevented. The DMA holds the address of the previous packet control list start address in the previous packet control list start address / temporary register. A flag called "previous PCL valid" is held in the DMA global register by the DMA, and a record is made as to whether it is a valid address at which it was stored. The transfer operation on the asynchronous channel is performed while checking whether a waiting state exists. The wait state is determined by the wait selection bit of "data buffer control / byte count / instruction" in the PCL. "
A flag called "Retry" is maintained by the DMA in the DMA Global Register. The DMA uses this flag to ignore these wait states during retries and to evaluate wait states. Chase.

When the wait state no longer exists, the DMA writes a control token representing the beginning of the packet to the FIFO and enters the data transfer phase. At this point, a loop is entered to check the current transfer count and see if it has reached zero. If so, a check is made to see if this is the last data buffer in the PCL buffer list. If another buffer exists in the PCL, the DMA obtains a new buffer address and transfer count and performs the transfer.
While transferring data from the PCI interface into the asynchronous transfer FIFO, the DMA waits until the FIFO has enough room before requesting the PCI bus master to perform a read transfer. DMA requires a PCI master transfer with a bit count equal to the high water mark defined for the DMA receive operation. While the DMA transfers data, the data buffer start address register and the data buffer transfer link bit in the DMA control register are updated to reflect the current state of the transfer.

When the last bit of data is transferred from the buffer to the Asynchronous Transfer FIFO and the buffer is at the end of the PCL list, as indicated by the last buffer bit in the control / byte count PCL word, the DMA terminates the packet. Know that you have reached. If the previous packet address was valid, the DMA delays checking the status until all packets have been packed into the transfer FIFO. In this case, unless the wait for status bit is set,
The return state is always for the previous packet. If only one packet is transferred, the previous and current packets are the same. If the previous packet address was valid, the DMA checks the packet counter. When the packet has been transferred to the IEEE 1394 bus by the link layer controller and the state of the packet is valid, the link layer controller decrements the packet counter.
The DMA idles and waits until the packet counter goes to zero, indicating that the valid state is applicable to the previous packet. If the status indicates that the packet should be retried the previous time, the DMA makes a FIFO flush request to the link layer controller, which completes the FIFO flush by removing the retry indication. Wait until you indicate that you have done this. The DMA then "returns" to the previous packet and starts the entire transfer. If no retries occur, the DMA updates the DMA status register with the positive acknowledgment status sent from the link layer controller, sets the packet completion, and then sets the completion status to memory in the last PCL status word. , Write along with the number of bits transferred to the currently running PCL, which should not be related to the previous PCL. If an interrupt bit is set in the PCL, an interrupt signal is issued and latched in the corresponding interrupt status register bit.

When the status check is completed, the DMA writes a special control token to the transfer FIFO to indicate the end of the packet. The packet count is reduced to one to indicate to the link layer controller that the end of the packet has been written by the DMA. The current PCL address is saved as the previous PCL address, and the previous packet control list start address register and the "last valid" flag are set in the DMA global register. Then DMA is another PC
Whether L is linked to the current PCL is determined by taking the next list address value. If it is valid, the DMA sets this to the current PCL address and continues execution. If it is not valid or the wait status bit is set, the DMA waits for the current packet to be transferred by the link layer controller. If a valid state is detected, as indicated by the packet counter decrementing to zero, the DMA checks to see if the packet should be retried as indicated by an IEEE 1394 busy state. If so, the FIFO is flushed as described above and the transfer is attempted again.

If a transfer timeout, retry exceeded, or FIFO not reached exists as indicated by the link layer controller, a packet error bit is sent to the DMA status register with an acknowledgment status. This state is updated in the PCL. If the transfer timed out or exceeded retries,
It would be possible for the target node to no longer respond. DMA solves this situation by allowing the PCL that forms the stream of data to this particular node to be skipped. The software sets the entry of the next PCL stream of the PCL to the next stream of transfer data (ie, the next asynchronous transfer to another 1394 node).
It can be set by specifying to CL. If the next PCL stream address is valid, the DMA
Continue execution with. If this address is not valid, the DMA channel will enter an idle state as if it had encountered a case where the next PCL address was marked invalid. If this next stream content is not used, this value should be set to the same value as the next list address. If DMA stops, DMA
Since the stop interrupt state is notified and the introduction of the next PCL stream is invalidated, rewriting of the next PCL stream is required. This is because the DMA is in the next stream acquisition state,
This is because the DMA ignores the next list address. Therefore, it is always necessary to set the "next list address" and "next PCL stream" to the same address if the next stream content is not used, which is all asynchronous transfers that implement the next PCL stream introduction. This is to prevent the channel from dangling due to errors.

FIGS. 28 and 29 show flowcharts similar to those described above for the asynchronous and isochronous transfer operations, but with D
This is applied to the auxiliary operation of the MA engine 74.

As shown in FIG. 30, the transfer operation from the PCI to the local bus and from the local bus to the PCI
Controls data transfer between the PCI bus and the local bus. The PCI address and the number of bits transferred are derived from the PCL data buffer control / bit count / command word in the PCL, as well as other transfer commands such as transfers. The difference is that the destination or source of the transfer is a local bus instead of a FIFO. The local bus address is generated from the AUX_ADR register (see hardware register definition).

The operation between PCI and LOCAL is performed by performing a check to see if a wait state exists. Waiting state is buf0c with PCL offset 0x18
It is determined by the wait selection bit of tl / bit_cnt / cmd. If the wait state no longer exists,
The DMA enters a loop where a check is made to see if the current transfer count has reached zero. If so, a check is made to see if this is the last data buffer in the PCL buffer list. PCL
If another buffer exists in the list, the DMA acquires a new buffer address and transfer count and performs the transfer. While the DMA transfers data, the data buffer start address register and the data buffer transfer length bit in the DMA control register are updated to reflect the current state of the transfer.

If the last bit of data from the buffer is transferred between the local bus and the buffer, and that buffer is at the end of the PCL list as indicated by the last buffer bit of the control / bit count PCL word,
The DMA knows that the end of the transfer has been reached. DMA is DM
Update the A status register with the status of 0x0001, PKT
CMP is set and PCL offset 0x
Written in the PCL status word at C with the number of bytes transferred. INT bit is PCL offset 0x1
When set to 8, buf0ctl / bit_cnt / cmd, an interrupt signal is issued and latched in the corresponding (DMA_PCL [x]) bit in the interrupt status register.

Next, the DMA determines whether or not another PCL is linked to the current PCL by the next list address (PC
This is performed by taking in an L offset (0x00). If it is valid, as indicated by bit 0 = 0, the DMA uses this as the current PCL address.
Execution continues as shown. If another PCL is not linked to the current PCL, indicated by bit 0 = 1, the link and busy bits are cleared in the DMA control register, and a DMA stop interrupt is issued for the state associated with this channel (DMA_HLT). [X])
Is generated in the interrupt status register at the same time, and the channel becomes idle.

FIGS. 31 and 32 show a FIFO high-level functional block diagram 500 including GRF80 and asynchronous transmission FIFO.
82, and FI for an isochronous transmission FIFO 84
The FO operation is described. The FIFO logic of the present invention includes a link side clock region 502 and a PCI side clock region 5.
04. PCI side clock area 504, FIFO
The control and status register 88 is written by the PCI bus interface logic in real time. Also, FI
The FO controller status register 88 provides an input to the pointer address mapping logic 86. Pointer address mapping logic 86 generates a RAM address mapping offset for the FIFO read / write pointer.

The dual port RAM 501 includes a GRF 80, an asynchronous transmission FIFO 82, and an equal time interval transmission FIFO.
The data storage unit 84 is provided at a boundary between the link-side clock area 502 and the PCI-side clock area 504. One PCI-side clock area 504,
Clock domain pointer translation logic 506 is DMAFI
Generate input to FO occupancy logic 508. DM
A FIFO occupancy state logic 508 generates a FIFO state for DMA logic signal 510. One link side clock domain 502 and clock domain pointer translation logic 512
6 generates a FIFO state, 1394 FIFO
An input is provided to the transmitter and receiver occupancy logic 514.

Dual port RAM 501 receives address inputs from multiplexers 518 and 520. Multiplexer 518 receives inputs from general receive write pointer 522, asynchronous transmit read pointer 524, and isochronous transmit read pointer 526. The general receive write pointer 522 is controlled by the 1394 receiver and the address mapping logic.
Asynchronous transmit read pointer 524 and isochronous transmit read pointer 526 are controlled by the 1394 receiver and the address mapping logic. General-purpose reception write pointer 522, asynchronous transmission read pointer 524
And the output from the isochronous transmit read pointer 526 is all the transmitter and receiver occupancy logic 51
4. Clock domain pointer translation logic 506 and multiplexer 518.

The general-purpose receive pointer 528, asynchronous transmit write pointer 530, and isochronous transmit write pointer 532 are all controlled by the DMA logic and the address mapping logic. General-purpose reception pointer 528,
Asynchronous transmit write pointer 530 and isochronous transmit write pointer 532 all provide outputs to DMA FIFO status logic 508, clock domain pointer translation logic 512, and multiplexer 520.

The dual port RAM 501 also has
Data input from the receive logic via the 33-bit data bus 534, link clock 25 MHz signal 536, and control is received from the 1394 transmit / receive logic via line 538. 1394 transmission / reception logic 538
Are sent to the link-side address multiplexer 518. Byte pack logic 540 receives the DMA read data from the host memory via control from the DMA logic and outputs the data to the 33-bit data bus 54.
2 which is the PCI of the dual port RAM 501
Sent to the side data input. Byte unpack logic 5
44 is a dual port RAM 501 PCI side data output 33
It receives the data output from bit bus 546 and provides DMA write data to host memory. Also, D
Control from MA logic is byte unpack logic 5
44, multiplexer 520 and dual port RAM5
01 is sent. From zero to 33 MHz PCI clock signal is sent to dual port RAM 501.

In connection with FIG. 2 and the explanation given in the related explanation, FIFO, GRF80, asynchronous transmission F
A more detailed description of the functions of the FIFO 82 and the isochronous transmission FIFO 84 is provided below.

FIFO Status Logic 514 and 516
Performs the logic necessary to generate the occupancy for each logical FIFO. When calculating the PCI-side FIFO state, the link-to-PCI clock domain translation logic samples the current values of each pointer on the link side of the FIFO and uses these samples to convert the link clock domain to the PCI clock domain. Translate to Each translated link-side pointer has its corresponding PC
The occupation state of each FIFO is generated by comparison with the I-side pointer. This state is used by the DMA logic to pace data transfers between the host memory and the FIFO. Similarly, when calculating the link-side FIFO state,
The translation logic from CI to link clock domain is
The current value of each pointer on the PCI side of the FO is sampled, and these samples are translated from the PCI clock domain to the link clock domain. Each translated PCI
The side pointer is compared with its corresponding link side pointer and the occupancy of each FIFO is calculated. This state is 1
Used in 394 transmit / receive logic to pace data transfer between the 1394 bus and FIFO.

Byte pack logic 540 implements logic that aggregates all quadlet data reads from host memory onto the byte-aligned addresses via the active DMA channel. The logic preferably includes four 8-bit wide registers and four 8-to-1 multiplexers. Each register and multiplexer pair corresponds to one byte sequence. The input of each register is connected to an input byte stream, which is switched to host memory by the active DMA channel. The output of each multiplexer is connected to an output byte stream, which drives a FIFO. For each 8-to-1 multiplexer, four inputs are connected to each register output in a one-to-one correspondence. The remaining four inputs are connected to each register input in a one-to-one correspondence. With this configuration, the bit array DMA read data from the four input byte strings can be switched to the four output byte strings in different orders. Control of the byte multiplexer is performed on the active DMA read channel.

Byte unpacking logic 544 is required to disassemble quadlet data reads from the FIFO to individual selectable bytes for writing via the active DMA channel to byte aligned addresses in host memory. Execute the appropriate logic. This logic consists of four 8-bit wide registers and four 8-to-1 multiplexers. Each register and multiplexer pair corresponds to one byte sequence. The input of each register is connected to an input byte sequence, which is driven from a FIFO. The output of each multiplexer is connected to an output byte stream, which is switched to host memory by a DMA channel. Four inputs for each 8-to-1 multiplexer are connected to each register output in a one-to-one correspondence. The remaining four inputs are connected to each register input in a one-to-one correspondence. With this configuration, quadlet reading from the FIFO can be switched on the output byte sequence in a different order. Control of the byte multiplexer is provided by the DMA write channel during operation.

Although the invention has been described in detail, it should be understood that various changes, substitutions and adaptations can be made without departing from the spirit and scope of the invention as defined in the appended claims.

The following items are further disclosed with respect to the above description. (1) A method for operating a data packet transfer device in an autonomous mode: guiding an autonomous boot mode selection signal to the data transfer device; constructing a register associated with the data transfer device for autonomous operation of the data transfer device. Including the above steps of enabling direct memory access associated with a data packet transfer device to transfer data to at least one program control list for operating said data packet transfer device in an autonomous mode; Method.

(2) The method according to item 1, further comprising the step of using DMA channel 0 to fetch the address of the first PCL.

(3) The method according to item 1 further comprises an external RP
The above method, comprising the step of enabling DMA master access to the L ROM device.

(4) The method according to item 1 further comprises an external PC
The above method, comprising the steps of allowing DMA master access to the I-interface device.

(5) The method according to item 1, further comprising a step of controlling the execution of the PCI memory instruction by the data packet transfer device according to an autonomous boot mode selection signal.

(6) The method according to (1), further comprising the step of controlling the execution of the PCI I / O instruction by the data packet transfer device according to the autonomous boot mode selection signal.

(7) The method according to (1), further comprising the step of controlling the execution of the PCI construction instruction by the data packet transfer device according to the autonomous boot mode selection signal.

(8) An autonomous boot circuit for operating a data packet transfer device in an autonomous mode: an introduction circuit for leading an autonomous boot mode selection signal to the data transfer device; Instructions for building a register associated with a data transfer device; and wherein the introductory circuit further transfers data to at least one program control list for operating the data packet transfer device in an autonomous mode. The autonomous boot circuit associated with a transfer device.

(9) The autonomous boot circuit according to (8), further comprising a circuit using DMA channel 0 to fetch the autonomous boot circuit address of the first PCL.

(10) The autonomous boot circuit according to claim 8, further comprising a circuit enabling DMA master access to an external RPL ROM device.

(11) The autonomous boot circuit according to item 8, further comprising a circuit enabling DMA master access to an external PCI interface device.

(12) The autonomous boot circuit according to (8), wherein the data packet transfer device further includes a circuit for controlling execution of a PCI memory instruction according to an autonomous boot mode selection signal.

(13) The autonomous boot circuit according to (8), wherein the autonomous boot circuit further includes a circuit for controlling execution of a PCI input / output instruction in accordance with the autonomous boot mode selection signal by the data packet transfer device.

(14) The autonomous boot circuit according to (8), wherein the data packet transfer device further includes a circuit for controlling execution of a PCI construction instruction according to an autonomous boot mode selection signal.

(15) In a personal computer environment, at least one peripheral device employing a data packet transfer bus; and a computer, wherein the computer is a host CPU including a PCI bus;
P that interfaces between the I bus and the peripheral device
A CI interface device, wherein the PCI interface device includes an autonomous boot circuit for operating a data packet transfer device in an autonomous mode, the autonomous boot circuit for: leading an autonomous boot mode selection signal to the data transfer device. An introductory circuit; and instructions for configuring a register associated with the data transfer device for autonomous operation of the data transfer device, the introductory circuit further comprising at least one program for operating the data packet transfer device in an autonomous mode. The personal computer environment associated with a data transfer device for transferring data to a control list.

(16) The personal computer environment as set forth in (15), wherein the autonomous boot circuit according to item 15 further includes a circuit that uses the DMA channel 0 to capture the autonomous boot circuit address of the first PCL.

(17) The personal computer environment as described above, wherein the autonomous boot circuit according to claim 15 further includes a circuit for enabling a DMA master access to an external RPL ROM device.

(18) The personal computer environment according to the item (15), wherein the autonomous boot circuit according to the item (15) further includes a circuit for enabling a DMA master access to an external PCI interface device.

(19) The personal computer environment according to claim 15, wherein the autonomous boot circuit further includes a circuit for controlling execution of a PCI memory instruction by the data packet transfer device according to an autonomous boot mode selection signal.

(20) The personal computer environment as set forth in claim 15, wherein the autonomous boot circuit according to claim 15, further includes a circuit for controlling execution of a PCI input / output instruction in accordance with the autonomous boot mode selection signal by the data packet transfer device.

(21) A method and system for autonomously operating the PCI serial bus interface device 20 in the autonomous mode, including an introduction circuit 370 for guiding the autonomous boot mode selection signal to the data transfer device. The instructions constitute registers 36, 38 associated with the data transfer device for autonomously operating the data transfer device. The introduction circuit 370 includes at least one program control list 456 for operating the data packet transfer device that transfers data in the autonomous mode.
Relate to

[Brief description of the drawings]

FIG. 1 illustrates a schematic diagram of the operation of a personal computer environment incorporating the present invention.

FIG. 2 shows a general operation schematic diagram of the functional logic of the present invention.

FIG. 3 illustrates a conceptual diagram of the comparison logic of the present invention.

FIG. 4 shows a more detailed conceptual representation of the comparison logic of the present invention.

FIG. 5 illustrates further details of the header comparison function of the packet receiver logic of the present invention.

FIG. 6 illustrates possible data values for a matching data set field as applied to the present invention.

FIG. 7 illustrates possible data values for a matching data set field as applied to the present invention.

FIG. 8 illustrates data values considered as matching data set fields as applied to the present invention.

FIG. 9 illustrates possible data values for a matching data set field as applied to the present invention.

FIG. 10 illustrates one embodiment of a programmable selectively writable data flip-flop of the present invention.

FIG. 11 illustrates an example address field for implementing the programmable bit selection function of the present invention.

FIG. 12 shows a diagram of a direct memory access, multi-channel environment being time division multiplexed and illustrating the DMA instruction processing operation of the present invention.

FIG. 13 shows a table illustrating priority assignment based on currently active channels.

FIG. 14 shows a part of the interface device structure of the present invention applicable to the automatic boot function.

FIG. 15 shows a table of PCI master bus instructions corresponding to bits 31 and 30 of the internal PCI address bus when the autoboot pin of FIG. 14 is set to an active state.

FIG. 16 shows the local bus interface of the present invention in block diagram form.

FIG. 17 schematically illustrates the logic of the interface device of the present invention for detecting horizontal and vertical synchronization signals.

FIG. 18 illustrates another feature of the invention, including directing a single data packet to a single scan line.

FIG. 19 illustrates the operation of the present embodiment for mapping a single data packet to multiple video scan lines.

FIG. 20 illustrates a packet control list data structure and a data buffer memory map used in the DMA packet processing apparatus of the present invention.

FIG. 21 shows a partial flow diagram of a DMA machine, illustrating the operation of the packet processor of the present invention.

FIG. 22 shows a partial flow diagram of a DMA machine, illustrating the operation of the packet processor of the present invention.

FIG. 23 illustrates a partial flow diagram of a DMA machine, illustrating the operation of the packet processor of the present invention.

FIG. 24 shows a partial flow diagram of a DMA machine, illustrating the asynchronous transmission operation of the present invention.

FIG. 25 shows a partial flow diagram of a DMA machine, illustrating the asynchronous transmission operation of the present invention.

FIG. 26 shows a partial flow diagram of a DMA machine, illustrating the asynchronous transmission operation of the present invention.

FIG. 27 shows a partial flow diagram of a DMA machine, illustrating the isochronous transmission operation of the present invention.

FIG. 28 shows an example of an auxiliary function flow diagram of the present invention.

FIG. 29 shows an example of an auxiliary function flow diagram of the present invention.

FIG. 30 shows a partial flow diagram of a DMA machine, illustrating the local bus / PCI bus intercommunication operation of the present invention.

FIG. 31 shows functions of the FIFO circuit according to the present invention.
It is a conceptual diagram at a high level.

FIG. 32 shows functions of the FIFO circuit according to the present invention.
It is a conceptual diagram at a high level.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 10 Personal computer environment 12 Personal computer 14 Peripheral device 16 Interface bus 18 3-port physical layer interface 20 PCI interface ASIC 22 Serial EPROM 24 PCI bus 26 Auxiliary port local bus 28 Local bus 30, 31 Address bits 34 PCI host bridge 36 Flash PROM 38 (DMA) Channel control static RAM (SRA)
M) 40 User-defined function (AUX) 42 Zoom video (ZV) port for video input / output 46 Local memory 48 CD ROM device 50 Laser printer 52 Desktop camera 56 Video cable set-top box 60 PCI bus logic 62 Serial EPROM interface 64 PCI Master Logic 66 PCI Slave Logic 68 PCI Configuration Control and Status Register 70 Local Bus Interface Logic 72 DMA Logic 74 DMA Engine 76 DMA Control and Status Register 78 FIFO Logic 80 General-purpose Receive FIFO 82 Asynchronous Transmission FIFO 84 Isochronous Transmission FIFO 86 Pointer Address Mapping logic 88 FIFO control and status register 90 Link layer control logic 92 Control And status register 94 period monitoring logic 96 period timer logic 98 packet transmission control logic 100 CRC logic 102 packet receiver control logic 104 physical link interface logic 106 received data 110 DMA header comparison register and logic 120 DMA channel header comparison logic 128 priority Encoder 129 Logical OR circuit 136 Destination ID comparison logic 140 Transfer code comparison logic 144, 146 Comparison logic 150 AND function 160 Transfer code 162 Asynchronous header 164 Transfer code bit 166 Isochronous interval header 250 Register writing circuit 252 General-purpose I / O ( GPIO) write data input 254 data flip-flop 256 clock signal 258 writable input 260 AND function 262 GPIO address acknowledge signal 264 Write standard signal 266 GPIO address bit input 300 DMA instruction processing environment 302 PCI register write data 304 Multiplexer 306 DMA register write data 310 DMA register 312 Previous address or provisional field 314 Packet control list start address register 316 DMA buffer start address register 320 DMA status register 322 DMA control register 324 Current DMA status register 326 Ready bit 328 DMA register write arbitration selection circuit 330 PCI slave signal 332 PCI master signal 334 DMAPCI master cycle input 336 Slave end input 340 DMA channel Arbitrator 342 Rock Channel Input 344 Multiplexer 346 Multiplexer 348 Slave register read address input 350 DMA register read data output 370 Autoboot input 390 Local bus interface block 392 Local bus configuration register 394 Zoom video (ZV) decode interface block 396 ZV machine 398 Address / data / bit capable Initialization holding register 400 Local bus interface state machine 402 Slave confirmation acknowledge interrupt block 410 Zoom port address map 412 Isochronous packet header 414 Isochronous packet data 416 Memory space 440 Zoom video isochronous packet 450 PCI queue 452 Start address 454 Dummy packet control list 456 Transfer instruction packet Control list 466 Sequential list 472 Logic function 480 DMA register 484 Destination address when condition is satisfied 501 Dual port RAM 502 Link side clock area 504 PCI side clock area 506 Clock area pointer translation logic 508 DMA FIFO occupation state logic 510 DMA logic signal 512 Clock domain pointer translation logic 514 Transmitter and receiver occupancy logic 516 1394 Transmit / receive logic signal 518 Multiplexer 520 Multiplexer 522 General receive write pointer 524 Asynchronous transmit read pointer 526 Current time interval transmit read pointer 528 General receive pointer 530 Asynchronous transmit write Pointer 532 Current time interval transmission write pointer 536 Link clock 25 MHz signal 538 1394 Transmission / reception logic 540 Byte pack logic 542 33-bit data bus 546 PCI-side data output 33-bit bus

Claims (2)

[Claims] 1. A method for operating a data packet transfer device in an autonomous mode, comprising: guiding an autonomous boot mode selection signal to the data transfer device, and setting a register associated with the data transfer device for autonomous operation of the data transfer device. Constructing and enabling direct memory access associated with a data packet forwarding device to transfer data to at least one program control list for operating the data packet forwarding device in an autonomous mode. Method. 2. An autonomous boot circuit for operating a data packet transfer device in an autonomous mode, an introduction circuit for guiding an autonomous boot mode selection signal to the data transfer device, and data for autonomous operation of the data transfer device. Instructions for constructing a register associated with a transfer device, the introductory circuit further comprising: a data transfer device for transferring data to at least one program control list for operating the data packet transfer device in an autonomous mode. Related, said autonomous boot circuit.