USRE42228E1 - Method and apparatus for using data protection code for data integrity in on-chip memory - Google Patents
Method and apparatus for using data protection code for data integrity in on-chip memory Download PDFInfo
- Publication number
- USRE42228E1 USRE42228E1 US10/723,963 US72396303A USRE42228E US RE42228 E1 USRE42228 E1 US RE42228E1 US 72396303 A US72396303 A US 72396303A US RE42228 E USRE42228 E US RE42228E
- Authority
- US
- United States
- Prior art keywords
- data
- frame
- buffer
- memory
- chip
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Lifetime
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
- H04L1/0056—Systems characterized by the type of code used
- H04L1/0061—Error detection codes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F13/14—Handling requests for interconnection or transfer
- G06F13/16—Handling requests for interconnection or transfer for access to memory bus
- G06F13/1668—Details of memory controller
- G06F13/1673—Details of memory controller using buffers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/08—Error detection or correction by redundancy in data representation, e.g. by using checking codes
- G06F11/10—Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's
- G06F11/1004—Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's to protect a block of data words, e.g. CRC or checksum
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F13/14—Handling requests for interconnection or transfer
- G06F13/36—Handling requests for interconnection or transfer for access to common bus or bus system
- G06F13/368—Handling requests for interconnection or transfer for access to common bus or bus system with decentralised access control
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/0614—Improving the reliability of storage systems
- G06F3/0619—Improving the reliability of storage systems in relation to data integrity, e.g. data losses, bit errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0655—Vertical data movement, i.e. input-output transfer; data movement between one or more hosts and one or more storage devices
- G06F3/0656—Data buffering arrangements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/0671—In-line storage system
- G06F3/0673—Single storage device
- G06F3/0674—Disk device
- G06F3/0676—Magnetic disk device
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/0671—In-line storage system
- G06F3/0683—Plurality of storage devices
- G06F3/0689—Disk arrays, e.g. RAID, JBOD
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/24—Testing correct operation
- H04L1/245—Testing correct operation by using the properties of transmission codes
- H04L1/246—Testing correct operation by using the properties of transmission codes two-level transmission codes, e.g. binary
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
- H04L12/42—Loop networks
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/02—Topology update or discovery
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/12—Shortest path evaluation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/48—Routing tree calculation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/90—Buffering arrangements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/90—Buffering arrangements
- H04L49/901—Buffering arrangements using storage descriptor, e.g. read or write pointers
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/90—Buffering arrangements
- H04L49/9031—Wraparound memory, e.g. overrun or underrun detection
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/90—Buffering arrangements
- H04L49/9063—Intermediate storage in different physical parts of a node or terminal
- H04L49/9068—Intermediate storage in different physical parts of a node or terminal in the network interface card
- H04L49/9073—Early interruption upon arrival of a fraction of a packet
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/16—Error detection or correction of the data by redundancy in hardware
- G06F11/20—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
- G06F11/2002—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where interconnections or communication control functionality are redundant
- G06F11/2007—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where interconnections or communication control functionality are redundant using redundant communication media
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2213/00—Indexing scheme relating to interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F2213/0038—System on Chip
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11B—INFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
- G11B5/00—Recording by magnetisation or demagnetisation of a record carrier; Reproducing by magnetic means; Record carriers therefor
- G11B5/012—Recording on, or reproducing or erasing from, magnetic disks
Definitions
- the present invention relates to the field of mass-storage devices. More particularly, this invention relates to an improved fibre-channel arbitrated-loop (“FC-AL”) apparatus and method using fibre-channel cyclic-redundancy-check information for maintaining and/or checking data integrity in on-chip memory.
- FC-AL fibre-channel arbitrated-loop
- One key component of any computer system is a device to store data.
- Computer systems have many different devices where data can be stored.
- One common place for storing massive amounts of data in a computer system is on a disc drive.
- the most basic parts of a disc drive are a disc that is rotated, an actuator that moves a transducer to various locations over the disc, and circuitry that is used to write and read data to and from the disc.
- the disc drive also includes circuitry for encoding data so that it can be successfully retrieved from and written to the disc surface.
- a microprocessor controls most of the operations of the disc drive, in addition to passing the data back to the requesting computer and taking data from a requesting computer for storing to the disc.
- the interface for transferring data between the disc drive and the rest of the computer system is typically a bus or channel, such as the Small Computer Systems Interface (“SCSI”), or the Fibre Channel.
- SCSI Small Computer Systems Interface
- Fibre Channel Fibre Channel
- the fibre channel One standardized interface for exchanging data between various storage devices and various computers is the fibre channel.
- the fibre-channel standard includes arbitrated loops (described further below).
- the fibre-channel standard supports a SCSI-like protocol for controlling data transfers.
- Fibre channels represent significant advantages over Small Computer Standard Interface (“SCSI”) designs. Fibre channels provide significantly higher bandwidths, currently up to about 106 megabytes per second, compared to between two and twenty megabytes per second for traditional SCSI designs. Fibre channels provide greater connectivity in that up to one-hundred twenty-six devices (including the host) may be connected, as compared to a maximum of seven or fifteen devices in typical SCSI environments.
- the fibre channel can be attached with a single connector and does not require a switch.
- a fibre channel using coaxial electrical conductors operates at distances of up to thirty meters between devices, and up to ten kilometers using fibre optics for an entire channel, as compared to a maximum total length of up to twenty-five meters for SCSI environments.
- CRC check cyclic-redundancy-code check
- FC-AL The fibre-channel arbitrated loop
- FC-AL operates at a clock frequency of 106.25 MHZ.
- 8B/10B encoder/decoder is described in U.S. Pat. No. 4,486,739 granted Dec. 4, 1984 for “Byte Oriented DC Balanced (0,4) 8B/10B Partitioned Block Transmission Code” by Franaszek et al., which is incorporated by reference.
- a fibre-channel arbitrated loop (“FC-AL”) allows for multiple devices, each called “a node,” to be connected together.
- a node may be any device (a computer, workstation, printer, disc drive, scanner, etc.) of the computer system having an interface allowing it to be connected to a fibre-channel “topology” (defined just below).
- Each node has at least one port, called an NL port (“node-loop port”) to provide access to other nodes.
- the components that connect two or more ports together are collectively called a “topology” or a “loop.”
- Each node communicates with all other nodes within the provided topology or loop.
- Ports are the connections in a fibre-channel node, though which data may pass over the fibre channel to ports of other nodes (the outside world).
- a typical fibre-channel drive has two ports packaged within the drive's node. Each port includes a pair of “fibers”—one to carry information into the port and one to carry information out of the port.
- Each “fiber” is a serial data connection, and, in one embodiment, each fiber is actually a coaxial wire (e.g., coaxial copper conductors, used when the nodes are in close proximity to one another); in other embodiments, a fiber is implemented as an optical fiber for at least some of its path (e.g., when nodes are separated by an appreciable distance, such as nodes in different cabinets or, especially, different buildings).
- Links carry information or signals packaged in “frames” between nodes.
- Each link can handle multiple types of frames (e.g., initialization, data, and control frames).
- Fibre-channel arbitrated loops attach multiple nodes in the loop without hubs or switches.
- the node ports use arbitration operations to establish a point-to-point data-transfer circuit.
- FC-AL is a distributed topology where each port includes at least the minimum necessary function to establish the circuit.
- the arbitrated-loop topology is used to connect any number of nodes between two and one-hundred twenty-six (126) node ports.
- each node includes dual ports (each connected to a separate loop) which provide redundancy, so that if one loop fails, the other one can fulfill the loop duties. Dual ports also allow two hosts (e.g., two host computers) to share a single drive.
- first- and second-generation FC-AL drives the two ports shared the frame-validation and frame-generation logic. This meant that if one port was receiving or transmitting a frame, the alternate port was effectively busy (since it could not simultaneously use the frame-validation and frame-generation logic), and the alternate port was thus forced to deny its host-bus adapter permission to send frames.
- Some host-bus adapters would continuously have to arbitrate and attempt to send a frame over and over until the primary port closed.
- the drive was only able to transmit on one port at a time. In some cases, an outbound data transfer on a given port would have to be paused in order to send a response or perform loop initialization on the other (alternate) port.
- CRC error checking by which an error code, based on the header and payload data of the transmission, is checked to verify the integrity of the received header and payload data.
- One such error-checking scheme employs cyclic-redundancy-code (“CRC”) information.
- CRC error checking will include a CRC checker to verify the integrity of received data words and a CRC generator to generate CRC information for digital words being transmitted.
- a CRC checker and a CRC generator must be available for each port to handle verification of each received digital word and to generate CRC information for each digital word being transmitted.
- the circuit or loop-interface module transmits on only one port at a time. For example, a disc-drive subsystem communicating through a multi-port interface module to a computer network would prepare and transmit data through only a single port at any given time. However, the loop-interface module might attempt to receive data through plural ports at a given time.
- One approach to reception of data through plural ports is to simply inhibit reception of data through other ports when one port is already receiving data.
- This approach allows common resources, such as the CRC checker or the frame-validation logic, to be shared among the several ports.
- the first port to receive data seizes use of the common resources to the exclusion of the other ports, and the other ports are inhibited from receiving data.
- incoming data cannot be received on the other ports and the other ports are limited to the function of data transmission.
- This approach resulted in the other ports receiving a “busy” condition in response to requests to transmit data, and necessitated repeating the sequence to request transmission of data again and again, until the first port completed the operation it was performing and freed-up the common resource.
- Fiber links have received considerable attention in connection with transmission of data between various devices of a computer network. More particularly, fibre channels offer significant advantages over Small Computer System Interface (“SCSI”) buses in terms of higher bandwidth, greater connectability, greater ease of attachment of modules, greater transmission distance, and other factors. For example, a typical SCSI bus is able to handle up to fifteen (15) modules with a total distance of up to about 25 meters, whereas a fibre channel can handle up to one-hundred twenty-six (126) modules with a distance of about thirty meters between modules using electrical transmission, or up to ten kilometers using optical transmission. Thus, in order to achieve a data-transfer rate of, for example, a terrabyte/second peak, it would require up to seventy SCSI buses but would need only about ten fibre channels.
- SCSI Small Computer System Interface
- Cyclic-redundancy-code information that is received along with a frame from a fibre channel is stored in an on-chip frame buffer, and later checked to ensure the integrity of the data while in the frame buffer.
- data frames along with their CRC information are stored into a data-frame buffer
- non-data frames along with their CRC information are stored into a receive-non-data-frame buffer.
- Dedicated receive buffers are also provided for each port of a two-port node. Cyclic-redundancy code information received from the fibre-channel along with a frame is stored in one of the three frame buffers. These data and CRC are later checked to ensure the integrity of the data while in the frame buffer. Control of a loop is maintained (i.e., the loop connection is held open) as long as a minimum amount of data, which optionally is determined by programming (called a “programmable amount of data”), is available for transmission, in order to reduce the overall amount of time spent arbitrating for control of the loop.
- a minimum amount of data which optionally is determined by programming
- the present invention provides a communications channel system for using fibre-channel cyclic-redundancy code (CRC) for data integrity in an on-chip memory.
- the system includes a first channel node having a first port and a second port, each port supporting a fibre-channel arbitrated-loop communications channel, each communications channel including a cyclic-redundancy code within data transmissions on the communications channel.
- the system includes an on-chip frame memory located on-chip in the first channel node that receives a frame and the received frame's associated CRC from the communications channel, and an integrity apparatus that uses the received associated CRC for data-integrity checking of the received frame that is in the on-chip frame memory.
- One version of this system further includes an off-chip memory operatively coupled to the on-chip frame memory and the integrity apparatus and a verification circuit within the integrity apparatus that verifies the cyclic-redundancy code while moving the received frame from the on-chip memory to the off-chip memory.
- One such embodiment further includes a parity-generation circuit that generates and appends parity to the data as they are moved from the off-chip memory to the on-chip memory, wherein the integrity apparatus further checks and strips away the cyclic-redundancy code while moving the frame of data to the off-chip memory.
- the on-chip frame memory further receives a data frame from the off-chip memory which is devoid of a cyclic-redundancy code
- the system further includes a CRC generator that generates the cyclic-redundancy code based on the data frame received from the off-chip memory, and a transmitter that transmits the data frame from the off-chip memory, including the generated cyclic-redundancy code, onto the communications channel.
- a data frame that is to be transmitted is transferred to the on-chip frame memory from the off-chip memory and is stored in the on-chip frame memory with parity but devoid of CRC information.
- a received data frame transferred to the on-chip frame memory from the communications channel is stored in the on-chip frame memory with CRC but devoid of parity information.
- a system is built according to this invention wherein the system further includes a magnetic-disc-storage drive operatively coupled to the first channel node, and a computer system operatively coupled to a second channel node (or, equivalently, a computer system having a second channel node).
- the second channel node is operatively coupled to the first channel node to transfer data between the first and second channel nodes through the communications channel.
- the disk drive includes a rotatable disc, a transducer in transducing relationship to the rotating disc, and a first channel node having a first port and a second port, each port supporting a fibre-channel arbitrated-loop communications channel.
- Each communications channel includes a cyclic-redundancy code within data transmissions on the communications channel.
- the first channel node is operatively coupled to the transducer to communicate data.
- the a disc drive includes an on-chip frame memory located on-chip in the first channel node that receives a frame and the received frame's associated CRC from the communications channel, and an integrity apparatus that uses the received associated CRC for data-integrity checking of the received frame that is in the on-chip frame memory.
- the disc drive further includes an off-chip memory operatively coupled to the on-chip frame memory and the integrity apparatus, and a verification circuit within the integrity apparatus that verifies the cyclic-redundancy code while moving the received frame from the on-chip memory to the off-chip memory.
- Yet another aspect of the present invention provides a communications method.
- the method includes: (a) supporting a fibre-channel arbitrated-loop communications channel on each of a first port and a second port of a first channel node; (b) receiving a frame from the communications channel, the received frame including a cyclic-redundancy code that is based on other data in the received frame; (c) storing the received frame, including the cyclic-redundancy code, into a frame buffer; (d) moving the received frame to a memory that is separate from the frame buffer; and (e) checking the received frame for accuracy by verifying the cyclic-redundancy code (CRC) while moving the received frame to the separate memory.
- CRC cyclic-redundancy code
- the method includes: (f) placing a frame that is to be transmitted into an on-chip frame buffer; (g) generating the cyclic-redundancy code based on data in the frame to be transmitted; and (h) transmitting the frame to be transmitted, including the cyclic-redundancy code, onto the communications channel.
- the placing step (f) further includes: (f)(i) generating parity for data of the frame to be transmitted; (f)(ii) adding parity to the data of the frame to be transmitted; and the moving step (d) further includes a step of (d)(i) stripping away the cyclic-redundancy code while moving the received frame to the separate memory.
- the receiving step (b) further includes a step of (b)(i) checking the received frame for accuracy by verifying the cyclic-redundancy code while receiving the received frame from the communications channel.
- the method further includes a step of (i) transferring data through the fibre-channel arbitrated-loop communications channel between a magnetic-disc-storage drive that is operatively coupled to the first channel node and a computer system having a second channel node, wherein the second channel node is operatively coupled to the first channel node by the communications channel.
- Still another aspect of the present invention provides a communications channel system that includes a channel node having a first port and a second port, each port supporting a fibre-channel arbitrated-loop communications channel, each communications channel including a cyclic-redundancy code within data transmissions on the communications channel.
- the system also includes a buffer or 55) that receives, from the channel node, a frame that includes a cyclic-redundancy code, and an off-chip memory separate from the buffer.
- the system also includes means (as described throughout this entire specification) for moving the received frame from the buffer to the off-chip memory and checking the received frame for accuracy by verifying the cyclic-redundancy code (CRC) while moving the received frame to the off-chip memory.
- the means for moving further includes means for stripping away the CRC as the frame is checked and moved to the off-chip memory.
- the present invention provides a significant enhancement in data-checking ability by keeping the received CRC of a frame with the frame as the frame is stored in one or more dedicated non-data buffers and/or a dedicated data buffer.
- two non-data buffers are provided, and are operable to simultaneously receive on both inbound fibers of a dual-port fibre-channel interface node.
- one data buffer is provided, and is operable to receive on either inbound fiber of a dual-port fibre-channel interface node.
- a single CRC checker is provided (to save cost) that verifies the cyclic-redundancy code of a frame while moving the frame from the on-chip buffer to the off-chip memory.
- a single CRC generator is provided on each transmit path (to save cost) that generates the appropriate cyclic-redundancy code of a frame while moving the frame from the on-chip buffer to the fibre channel.
- FIG. 1 is a block diagram of a disc drive 100 with a fibre-channel node interface.
- FIG. 2 is a block diagram of an information-handling system 1200 incorporating the present invention.
- FIG. 3 is a block diagram of a fibre-channel node-interface chip 110 .
- FIG. 4 is a block diagram of a fibre-channel loop port circuit 20 .
- FIG. 5 is a block diagram of comparator logic 30 used, in one embodiment, to hold a loop open.
- FIG. 6 is a block diagram of a fibre-channel loop-control circuit 40 .
- FIG. 7 is a block diagram of a fibre-channel receive path circuit 50 .
- FIG. 8 is a block diagram of a fibre-channel pre-buffer-receive path circuit 51 .
- FIG. 9 is a block diagram of a fibre-channel receive-frame non-data buffer circuit 53 .
- FIG. 10 is a block diagram of a fibre-channel data-frame buffer circuit 55 .
- FIG. 11 is a block diagram of a fibre-channel common receive path circuit 59 .
- FIG. 12 is a block diagram of a fibre-channel transfer-control circuit 60 .
- FIG. 13 is a block diagram of a fibre-channel transmit path circuit 70 .
- FIG. 14 is a block diagram of a fibre-channel transmit-frame buffer circuit 73 .
- FIG. 15 is a block diagram of a fibre-channel data transmit path circuit 80 .
- the invention described in this application is useful for all types of disc drives, including hard-disc drives, ZIP drives, floppy-disc drives, optical-disc drives, CDROM (“compact-disc read-only memory”) drives, and any other type of drives, systems of drives (such as a “redundant array of inexpensive/independent disc drives,” or RAID, configuration) or other devices, where data are communicated between drives and other devices or information-handling systems.
- the present invention is useful in node interfaces for non-disc devices, such as hubs and switches (such as are used to connect plural fibre-channel loops to one another), workstations, printers, and other devices or information-handling systems that are connected on a fibre-channel arbitrated loop.
- Section IV is the section primarily related to the details of the present invention; however, the other sections provide relevant information as to the overall environment for the invention.
- FIG. 1 shows a block diagram of a disc-drive device 100 with a fibre-channel node interface.
- a fibre-channel loop-interface circuit 1220 includes a dedicated transmit-frame buffer 73 for loop initialization and responses.
- Loop initialization is accomplished by transmitting a sequence of one or more specialized non-data frames (and monitoring responses to those frames) to initialize a fibre-channel loop.
- Responses are non-data frames sent in response to commands or inquiries from other nodes.
- the fibre-channel arbitrated-loop communications channel 1250 also called a loop 1250 , or fibre-channel loop 1250 ) can be used to communicate data between disc-storage devices 100 and computers 1202 or other information-handling devices.
- fibre-channel loop 1250 is a serial communications channel; in other embodiments two or more parallel lines (or “fibers”) are used to implement fibre-channel loop 1250 .
- the ports 116 are serial lines, one line 117 for inbound data, and another line 118 for outbound data, both lines 117 and 118 connecting to, and forming part of, communications channel loop 1250 .
- Dedicated receive buffers ( 53 , 53 ′ and 55 ) are also provided for each port 116 of a two-port node.
- each block having a reference numeral with a prime provides the same function as the corresponding block without the prime (e.g., 53 ), but is used for a separate loop port or communications channel.
- Cyclic-redundancy-code information received from the fibre channel 1250 along with a frame is stored in one of one or more frame buffers ( 53 , 53 ′ or 55 ), and later checked to ensure the integrity of the data while in the frame buffer ( 53 , 53 ′ or 55 ).
- Control of a loop 1250 is maintained (i.e., the loop connection is held open) as long as a programmable amount of data is available for transmission, in order to reduce the overall amount of time spent arbitrating for control of the loop 1250 .
- disc drive 100 includes a magnetic-storage head-disc assembly (“HDA”) 114 having one or more disc platters 134 , one or more magnetic read/write transducers 150 per disc platter, and an arm actuator assembly 126 . Signals between the transducers (or “heads”) and HDA interface 113 transfer data to and from the disc platters 134 .
- the “disc drive” of some embodiments e.g., disc drive 1256 of FIG. 1
- the “disc drive” of some embodiments includes HDA 114 and HDA interface 113 (e.g., a conventional SCSI drive), and one or more such conventional disc drives 1256 is connected to an external node interface 1220 in order to connect to a loop or fibre-channel topology, as shown in FIG. 1 .
- a “disc drive” is typified by the disc drive 100 of FIG. 2 , and includes a node interface 1220 integrated with the disc drive 1256 in overall disc drive 100 .
- data are in turn transferred from and to off-chip buffer 111 .
- the invention provides a dedicated on-chip buffer 119 , which, in the embodiment shown, includes a receive-non-data-frame buffer 53 (alternately called an “inbound non-data buffer 53 ”) for each port (i.e., buffers 53 and 53 ′), a transmit-frame buffer 73 which, in one embodiment, can be used simultaneously by both ports (in other embodiments, a single buffer is used by only one port at a time), and a shared data-frame buffer 55 , along with a CRC checker 596 (see FIG. 11 ).
- forty words of transmit-frame buffer 73 are reserved for Port A and forty words are reserved for Port B, so both ports can be initialized at the same time.
- Such an embodiment is equivalent to having two separate forty-word transmit-frame buffers, one for each port, which can be used simultaneously.
- each of these “words” is thirty-six bits wide (thirty-two data bits and four parity bits).
- the CRC validity-checking information that is received with data frames from the fibre-channel loops 1250 is stored with the data in data-frame buffer 55 , and then checked when the data are read out of data-frame buffer 55 , thus providing checking for data errors that may arise as the data frames reside in data-frame buffer 55 or anywhere earlier in the travel of the data frames.
- CRC validity-checking information that is received with non-data frames from the fibre-channel loops 1250 is stored with the data in non-data-frame buffer 53 (or 53 ′), and then checked when the data are read out of non-data-frame buffer 53 (or 53 ′), thus providing checking for data errors that may arise as the non-data frames reside in non-data-frame buffer 53 (or 53 ′) or anywhere earlier in the travel of the data frames.
- Microprocessor 112 is any suitable high-speed processor, and is used to help control the overall data transfer, routing, signalling, error recovery, etc., within disc drive 100 .
- fibre-channel interface chip 110 provides improved frame buffers, error checking, and loop arbitration, as described below.
- loop-port transceiver blocks 115 include port transceivers which serialize and deserialize data transfers through Port A and Port B to the fibre-channel loops 1250 (see FIG. 2 ) connected thereto.
- transceivers 115 are implemented as external transceivers; in other embodiments, these transceivers are located on-chip in block 110 .
- the right-side interfaces i.e., right side relative to the transceiver 115 or 115 ′ of FIG. 1
- are parallel input-output signals that are ten bits wide; in other embodiments, they are twenty bits wide.
- port transceivers 115 and 115 ′ form fibre-channel node interface 1220 .
- port transceivers 115 and 115 ′ are integrated within a single chip 110 .
- the transceivers 115 and 115 ′, including their serializer/deserializer functions are implemented on circuits separate from chip 110 .
- transceivers 115 are merely interfaces between the serial loop 1250 and chip 110 , wherein the serialization/deserialization to ten-bit wide or twenty-bit-wide data occurs on-chip.
- FIG. 2 is a schematic view of a computer system 1200 .
- Computer system 1200 may also be called an electronic system or an information-handling system and includes a central processing unit (“CPU”), a memory and a system bus.
- Computer system 1200 includes a CPU information-handling system 1202 having a central processing unit 1204 , a random-access memory (“RAM”) 1232 , and a system bus 1230 for communicatively coupling the central processing unit 1204 and the random-access memory 1232 .
- the CPU information-handling system 1202 includes the fibre-channel node interface 1220 .
- Each one of the one or more disc-storage information-handling systems 100 through 100 ′ includes one or more disc-drive device 1256 and a fibre-channel node interface 1220 .
- multiple disc drives 1256 are connected to a single node interface 1220 , for example in a RAID (redundant array of inexpensive/independent disc drives) configuration, such that device 100 ′ is a RAID array of disc drives.
- the CPU information-handling system 1202 may also include an input/output interface circuit 1209 that drives an internal input/output bus 1210 and several peripheral devices, such as 1212 , 1214 , and 1216 , that may be attached to the input/output bus 1210 .
- Peripheral devices may include hard-disc drives, magneto-optical drives, floppy-disc drives, monitors, keyboards and other such peripherals.
- any type of disc drive or other peripheral device may use the fibre-channel methods and apparatus (especially, e.g., the improvements in fibre-channel node interface 1220 ) described herein.
- the A port or the B port can be used to connect to any given loop 1250 .
- system 1200 optionally includes a second CPU information-handling system 1202 ′ (which is identical or similar to system 1202 ) having central processing unit 1204 ′ (which is identical to central processing unit 1204 ), a random-access memory (“RAM”) 1232 ′ (which is identical to RAM 1232 ), and a system bus 1230 ′ (which is identical to system bus 1230 ) for communicatively coupling central processing unit 1204 ′ and random-access memory 1232 ′.
- central processing unit 1204 ′ which is identical to central processing unit 1204
- RAM random-access memory
- system bus 1230 ′ which is identical to system bus 1230
- CPU information-handling system 1202 ′ includes its own fibre-channel node interface 1220 ′ (which is identical to node interface 1220 ), but is connected to one or more disc systems 100 (in this illustrated example, it is just connected to disc system 100 ′, but in other examples is connected to all devices or disc systems 100 through 100 ′) through a second fibre-channel loop 1250 ′ (separate and independent from loop 1250 ).
- This configuration allows the two CPU systems 1202 and 1202 ′ to share one or more of the disc systems 100 using separate fibre-channel loops for each CPU system 1202 .
- all devices 100 through 100 ′, and all CPU systems 1202 through 1202 ′ are connected to both loops 1250 and 1250 ′.
- the present invention does not support out-of-order delivery of data frames.
- the fibre-channel controller also implements a protocol for organizing data frames of the code words for transmitting and receiving purposes, which protocol is disclosed in U.S. Pat. No. 5,260,933, entitled ACKNOWLEDGMENT PROTOCOL FOR SERIAL DATA NETWORK WITH OUT-OF-ORDER DELIVERY, by G. L. Rouse.
- the Fibre-Channel Specifications used in building one embodiment of the present invention include the following ANSI Standards:
- frame buffers have been added to the third-generation application-specific integrated circuit (“ASIC”) chip (the fibre-channel interface chip 110 ) to allow both ports to be active simultaneously.
- ASIC application-specific integrated circuit
- Two buffers that receive non-data frames also called “receive-non-data-frame buffers” 53 and 53 ′ of FIG. 1 ) are provided to allow commands and FCP frames (fibre-channel-protocol frames) to be received simultaneously at both ports of the node (and also to permit full-duplex operations, i.e., receiving on one fiber of a port while transmitting on the other fiber of the same port).
- FCP frames short-channel-protocol frames
- the present invention allows the commands to be sorted and optimized while data transfers are progressing, thus improving the performance of the system 1200 .
- the transmit-frame buffer 73 (see description of FIG. 13 ) allows response frames to be transmitted on one port while a data transfer is active on the alternate port. This transmit-frame buffer 73 also allows loop initialization to be performed on one port without pausing or waiting for the transfer to complete on the other port.
- on-chip frame buffers may be used to manage inbound and outbound frames.
- the on-chip RAM may be configured in various ways to strike a balance between performance and silicon real estate.
- the present specification details the use of a dedicated frame buffer 119 and a single-frame transmit path 70 that are used to store and transmit fibre-channel loop-initialization frames and single-frame fibre-channel responses.
- one port may be receiving or transmitting data and using most of the ASIC resources to handle the transfer. Many counters and state machines may be required to handle this type of multi-frame sequence.
- a limited amount of logic is required to provide functions for the alternate port to allow it to receive and transmit frames while the primary port is transferring data.
- a dedicated transmit-frame buffer 73 is provided (shown in FIG. 13 ), along with single-frame-transmit-path circuit 70 , to provide the ability to transmit frames on one port while the other port is transferring data.
- the logic may be configured dynamically, so that either port may transfer data or use the transmit-frame buffer 73 .
- Loop-initialization frames (which are non-data frames) that are received are validated before the frame is written to a frame buffer (e.g., transmit-frame buffer 73 ; see FIG. 13 ).
- a microprocessor 112 has write/read access to the transmit-frame buffer 73 to allow it to examine and modify a received frame before allowing it to be transmitted out.
- the “header” and “payload” of the frame are stored in the frame buffer. (The “header” of a frame includes such information as the Source Identifier, Sequence Count, and Originator Identifier of the frame.
- the “payload” of a frame is the main body of data to be transmitted.)
- the single-frame-transmit path 70 (see FIG. 3 ), under control of microprocessor 112 , assembles the frame and includes the start-of-frame and end-of-frame delimiters and generates the frame cyclic-redundancy-code (“CRC”) information.
- CRC frame cyclic-redundancy-code
- microprocessor 112 places the header and payload of the response frame into the transmit-frame buffer 73 .
- the single-frame-transmit circuit 70 assembles the frame and includes the start-of-frame and end-of-frame delimiters and generates the frame CRC information. Some additional loop control logic is also provided to open the loop 1250 for frame transmission.
- transmit-frame buffer 73 requires write and read pointers ( 733 and 734 , respectively). (Although sometimes denoted as a “single-frame transmit-frame buffer” or a “transmit-frame buffer,” buffer 73 is genetically a “transmit-frame buffer,” and in other embodiments buffer 73 includes one or more transmit frames for each one of one or more ports; the term “transmit-frame buffer” is used herein to include all such embodiments.)
- the single-frame-transmit circuit 70 (a detailed description of which is shown below in FIG. 13 ) requires a frame-length counter 71 , transmit-framing state machine 72 , CRC generator 76 , and transmit multiplexer (“mux”) 74 .
- FIG. 3 is a block diagram of a fibre-channel node-interface chip 110 .
- the fibre-channel node-interface logic 110 in the present invention is responsible for the fibre-channel protocol including the arbitrated loop logic and framing logic.
- One embodiment is optimized for a class-3 SCSI implementation (see the FC AL specifications noted above) using only the SCSI upper-level protocol defined by the fibre-channel protocol (“FCP”) standard.
- the fibre-channel node-interface logic 110 includes four on-chip frame buffers ( 53 , 53 ′, 55 , and 73 ) to assist in dual-port and full-duplex operations, as well as to support a variety of buffer bandwidths.
- the fibre-channel node-interface logic 110 also interfaces to a microprocessor 112 , which allows microprocessor 112 to configure the fibre-channel node-interface logic 110 and to read status information about the present condition of the fibre-channel node-interface logic 110 .
- the fibre-channel node-interface logic 110 includes two loop port circuits 20 (one for Port A and another for Port B, each port having a data-in interface and a data-out interface to support loop communications), loop-control circuit 40 (also called frame-transmit circuit 40 ), receive-path logic 50 , transfer-control logic 60 , single-frame-transmit circuit 70 , transmit-path multiplexer (“mux”) 79 , data-frame-transmit-path logic 80 , and microprocessor interface 90 . These blocks support such functions as receive-frame processing, transmit-data-frame generation, single-frame-transmit generation, transfer control, and processor interfacing.
- Microprocessor interface circuit 90 provides microprocessor 112 access to the registers and counters in the fibre-channel node-interface logic 110 . (When a “microprocessor” is described, it is to be understood that such term includes any suitable programmable logic device.)
- the interface registers are initialized by an external microprocessor 112 prior to responding of the fibre-channel interface. Output transfers are initialized through this interface and status of received transfers is available through this interface.
- the input signals for FIG. 3 include A_IN 3021 which conveys data input from the fibre channel 16 into loop port circuit 20 for Port A, and B_IN 3022 which conveys data input from the fibre channel 16 into loop port circuit 20 for Port B.
- DATA FROM OFF-CHIP BUFFER 3051 conveys data from off-chip buffer 111 to receive path 50 .
- TO OFF-CHIP BUFFER 3052 conveys data to off-chip buffer 111 from receive path 50 .
- BUFFER STATUS 3061 provides status to transfer control 60 .
- MPU ADDRESS 3091 and MPU DATA 3095 into MPU interface 90 provide address and data, respectively, from microprocessor 112 .
- READ_ENABLE 3092 and WRITE_ENABLE 3093 into MPU interface 90 provide enable signals from microprocessor 112 .
- Signals MPU 3076 allow microprocessor 112 to access transmit-frame buffer 73 .
- A_OUT 3023 conveys data to the fibre channel 16 from loop port circuit 20 for Port A
- B_OUT 3024 conveys data to the fibre channel 16 from loop port circuit 20 for Port B.
- FIG. 4 is a block diagram of a fibre-channel loop port circuit 20 .
- the fibre-channel design of one embodiment of the present invention includes two identical loop port circuits 20 to support a dual-ported fibre-channel interface for direct attachment of peripherals.
- fibre-channel loop port circuit 20 includes receive register 21 , 8B/10B decoder logic 22 , word-sync state machine 23 , loss-of-receive-clock detector 24 , loss-of-sync timer 25 , arbitrated-loop logic 26 , and 8B/10B encoder 27 .
- each of loop port circuits 20 interface to external transceivers 115 (see FIG. 1 ) using a ten-bit data interface.
- the transceivers 115 serialize and deserialize serial data to and from a parallel interface (e.g., a ten-bit-wide or a twenty-bit-wide interface).
- these transceivers 115 are integrated into chip 110 .
- the parallel data (input from the fibre channel) are captured using receive clocks from the receiver portion of each transceiver 115 , and are converted to a twenty-bit-wide format before decoding using a parallel 8B/10B decoder.
- the sixteen-bit data plus two k-characters are then checked for word validity before being placed in the arbitrated-loop logic 26 .
- the output of the arbitrated-loop logic 26 is re-synchronized to the transmitter clock and may be passed to the receive-framing logic or re-transmitted on the loop 1250 through encoder 27 .
- encoder 27 converts one eight-bit character to one ten-bit character during each operation; in other embodiments, two or more eight-bit characters are converted into the corresponding number of ten-bit characters in each operation. (See U.S. Pat. No.
- the arbitrated-loop logic 26 includes a loop state machine, an ordered-set decoder, and elasticity insert and delete functions.
- Loop-port circuits 20 implement the Arbitrated-Loop Protocol as defined in the Fibre-Channel Arbitrated-Loop ANSI standard (i.e., FC-AL and/or FC-AL2, described above).
- fibre-channel data are transmitted serially and converted to ten-bit parallel data by the transceiver 115 .
- Receive register 21 captures the ten-bit data (A_IN 3021 or B_IN 3022 ) from the transceiver 115 using clocks generated by the receiver portion of transceiver 115 .
- the data are immediately converted to twenty bits wide (i.e., two ten-bit characters wide) before being passed through the 8B/10B decoder 22 .
- decoder 22 converts one ten-bit character to one eight-bit character during each operation; in other embodiments, two or more ten-bit characters are converted into the corresponding number of eight-bit characters in each operation.
- the 8B/10B decoder logic 22 inputs the encoded data captured by the receive register 21 .
- Two ten-bit characters are decoded in parallel to output two eight-bit characters.
- Running disparity of the input characters is checked and error status is passed to word-sync state machine 23 , as well as to the arbitrated-loop logic 26 .
- Negative running disparity is forced on the next ordered set following a running disparity error.
- Violations to the coding rules are also checked and code-violation status is passed to word-sync state machine 23 .
- the loss-of-receive-clock detector 24 detects when the receive clocks from the transceiver 115 have stopped. When a “loss-of-receive-clock” condition is detected, word-sync state machine 23 is reset and data are prevented from going into the FIFO in arbitrated-loop logic 26 (a FIFO is a first-in first-out memory, typically used to interface between busses or processes having different speeds).
- a FIFO is a first-in first-out memory, typically used to interface between busses or processes having different speeds.
- the current-fill-word (“CFW”, described more below) is transmitted until word sync is re-gained.
- Word-sync state machine 23 logic monitors the input stream for word sync. Word-sync is achieved when three valid ordered sets are detected with proper byte/control character alignment, and no intervening invalid characters are detected. “Loss-of-word-sync” is defined per the FC-PH (i.e., FC-PH Physical & Signalling Interface X3T11/Project 755D/Rev. 4.3) standard. When word-sync is achieved, data are input into the FIFO in arbitrated-loop logic 26 .
- the loss-of-sync timer 25 is used to determine when a loss-of-word-sync condition has been present for more than one maximum frame time (since it may take up to a frame time to detect three valid ordered sets). When this timer expires, microprocessor 112 is interrupted with the LOSS-OF-SYNC interrupt signal 4025 , so that it may take action.
- the arbitrated-loop logic 26 includes a loop-elasticity FIFO, loop FIFO control logic, ordered set-decode logic, loop-state-machine logic, current-fill-word-selection logic, loop-output multiplexer logic, and miscellaneous functions.
- the loop-elasticity FIFO provides the buffering required to re-synchronize the input data (clocked by the receive clock) with the transmit clock.
- the loop FIFO control logic monitors the status of the arbitrated-loop logic 26 to determine whether an insert or delete operation may be required. Ordered sets are decoded by the ordered-set-recognition logic.
- FC-PH defined ordered sets i.e., FC-PH Physical & Signalling Interface X3T11/Project 755D/Rev. 4.3
- FC-PH Physical & Signalling Interface
- the current-fill-word-selection logic monitors the loop states and decoded ordered sets to determine the current-fill-word (“CFW”).
- CFW current-fill-word
- a hardware state machine uses the ordered-set decodes to perform the loop functions described in the FC-AL standard (i.e., Fibre-Channel FC-AL1 Arbitrated Loop standard X3T11/Project 960D/Rev.
- the inputs LOOP A TRANSMIT CONTROL OUTPUTS 6425 and LOOP B TRANSMIT CONTROL OUTPUTS 6427 provide input to arbitrated-loop logic 26 from the logic in FIG. 6 .
- the outputs LOOP A STATES AND CONTROL 6422 and LOOP B STATES AND CONTROL 6432 control the output of the respective loops and provide status to the loop-control logic, which in turns generates requests to the loop-state machine (see FIG. 6 ).
- the outputs LOOP A DATA 4026 and LOOP B DATA 4027 provide data to the respective local ports (to blocks 51 and 51 ′ respectively of FIG. 7 ).
- 8B/10B encoder logic 27 accepts the sixteen-bit data and 2 k-characters (lower k is always 0) from arbitrated-loop logic 26 .
- the inputs are encoded into two ten-bit characters which are separated and output one at a time to transceiver 115 (see FIG. 1 ), which converts the data to a serial stream.
- both ten-bit characters i.e., twenty bits
- the transmit multiplexer 79 (see FIG.
- the output signals A_OUT 3023 and B_OUT 3024 transmit data to the respective transceivers 115 and 115 ′.
- FIG. 5 shows comparator logic 30 used, in one embodiment, to hold a loop open, once control is achieved by arbitration.
- the amount of OFF-CHIP AVAILABLE DATA 5011 is compared to a predetermined value X-FRAMES 5013 (which, in one embodiment, is a programmable value, and an optimal value is determined empirically; in one embodiment, this value is set to one frame) by comparator 5010 .
- X-FRAMES 5013 which, in one embodiment, is a programmable value, and an optimal value is determined empirically; in one embodiment, this value is set to one frame
- the amount of DATA-FRAME DATA AVAILABLE 5015 is compared to a predetermined value Y-WORDS 5017 (which, in one embodiment, is a programmable value, and an optimal value is determined empirically; in one embodiment, this value is set to the number of words in one-half of a frame; in one embodiment, there are about 2000 words, and Y-WORDS is about 1,000) by comparator 5012 .
- AND gate 5014 determines when both conditions are met, and outputs HOLD LOOP OPEN signal 5019 .
- FIG. 6 is a block diagram of loop-control circuit 40 (also called frame transmit (“XMIT”) circuit 40 ).
- Loop-control circuit 40 includes control logic to generate requests to the appropriate arbitrated-loop state machine (in arbitrated-loop logic 26 of Port A and Port B) as well as to generate requests to the transmit framing state machines 72 (see FIG. 13 ) and 81 (see FIG. 15 ) to begin transmitting frames or R_RDY's.
- Transmit-data-sequencer logic 41 includes logic that is activated when a transfer is requested by microprocessor 112 .
- Transmit-data-sequencer logic 41 monitors the transfer using input signals TRANSMIT STATUS INPUTS 6411 and generates “enables” (i.e., enable signals TRANSMIT CONTROL OUTPUTS 6413 ) for each stage of the transfer. This allows the transfer-ready and FCP responses to be generated without intervention of microprocessor 112 .
- Loop-port A/B open-control state machines 42 (Port A) and 42 ′ (Port B) handle the case where the port is opened by another L_Port or when the loop 1250 is opened to transmit frames.
- This logic generates requests to arbitrate and to close the loop 1250 , and requests to transmit R_RDY's and frames of various sorts, and can be configured for half-duplex or full-duplex operation.
- R_RDY's When the port is configured for half-duplex mode, R_RDY's may be transmitted only when in the Opened state. When the port is configured for full-duplex mode, R_RDY's may be transmitted in either Opened state or Open state. Conditions that cause an R_RDY to be transmitted include “Buffer-to-Buffer Credit (BB_Credit) available and outstanding R_RDY's less than maximum BB_Credit.” (Buffer-to-Buffer Credit control logic 603 , which is described below in FIG. 12 , issues buffer credit to the connected port to allow frames to be sent. This credit is issued by sending R_RDY's.)
- BB_Credit Buffer-to-Buffer Credit
- frames may be transmitted only when in the Open state.
- frames may be transmitted in Open state or in Opened state if the port was opened in full-duplex mode by the frame recipient.
- a request to transmit a frame will be generated when all the following conditions are met:
- loop-port A/B open-init-control state machines 46 (Port A) and 46 ′ (Port B) handle the case of when the loop 1250 is in the Open-Init state.
- This logic 46 and 46 ′ generates requests to transmit frames.
- These state machines will generate a request to transmit a frame when microprocessor 112 requests it, and will monitor for the transmission of the EOF. When the transmission is complete, a transfer-complete is generated to microprocessor 112 .
- the inputs to block 40 include PORT BB_CREDIT AVAILABLE TO TRANSMIT R_RDY 6017 and PORT CREDIT AVAILABLE TO TRANSMIT A FRAME 6020 (see FIG. 12 ), LOOP A STATES AND CONTROL 6422 and LOOP B STATES AND CONTROL 6432 (see FIG. 4 ), and DATA AVAILABLE 6019 (See FIG. 12 ).
- the outputs from block 40 include TRANSMIT CONTROL OUTPUTS 6413 , LOOP A TRANSMIT CONTROL OUTPUTS 6425 , and LOOP B TRANSMIT CONTROL OUTPUTS 6427 .
- the buffers in on-chip frame buffer 119 may be used to manage inbound and outbound frames. Frames received and transmitted are usually stored in a large off-chip area (e.g., off-chip buffer 111 ) at a slower transfer rate. Even when off-chip buffer 111 is capable of the full transfer rate for a single port, for a dual-ported design the bandwidth required will be much greater, adding additional cost.
- On-chip frame buffer 119 in the FC-AL ASIC 110 may be configured in various ways to strike a balance between performance, silicon real estate, and cost.
- the present specification details the use of dedicated frame buffers 53 and 53 ′ (components of the overall on-chip frame buffer 119 ) to receive non-data type frames simultaneously on each port, as well as providing a dedicated large data-frame buffer 55 (also a component of the overall on-chip frame buffer 119 ).
- frames may be received on both ports 116 simultaneously.
- the frames are usually moved to and stored in a larger off-chip memory 111 after they are received.
- Each frame must be validated and the frame cyclic-redundancy-code (“CRC”) information must be checked before transferring the frame off-chip.
- CRC frame cyclic-redundancy-code
- individual receive-non-data-frame buffers 53 and 53 ′ are provided, one at each port 116 , to allow frames to be received simultaneously at full interface rates, and then read one-at-a-time, validated, and transferred off-chip.
- a large common data-frame buffer 55 is also provided to be shared between the ports so that data may be received or transmitted on one port while non-data frames are received on the other port simultaneously. Further, since two individual one-way fibers are provided at each port, a single port can be transmitting and receiving at the same time.
- the Port A receive fiber 117 can receive non-data frames into non-data receive buffer 53 while the Port A transmit fiber 118 is either transmitting data frames from data-frame buffer 55 , or transmitting non-data frames from transmit-frame buffer 73 ; and simultaneously Port B can receive non-data frames to non-data receive buffer 53 ′ while transmitting non-data frames from transmit-frame buffer 73 or data frames from data-frame buffer 55 (in the case where Port A is transmitting non-data frames).
- Either the data-frame buffer 55 , or one of the receive-frame buffers 53 or 53 ′, may be selected to use the receive-frame-validation logic and CRC checker.
- a single data-frame buffer 55 is provided, and can only used for one port 116 at a time, and is used for either transmitting or receiving at any one time. In other embodiments, multiple data-frame buffers 55 are provided to remove such a restriction. Also note that, for one embodiment of the present invention, a single transmit-frame buffer 73 is provided, and can only used for one port 116 at a time. In other embodiments, multiple transmit-frame buffers 73 are provided to remove such a restriction, and to allow truly simultaneous loop initialization operations (or other non-data responses to be sent) on both ports.
- either data or non-data frames may be received on the primary port.
- non-data frames may be received on the alternate port.
- Data frames (which include header, payload, CRC, and frame delimiters) are placed into a large data-frame buffer 55 while non-data frames (which also include header, payload, CRC, and frame delimiters) are placed in smaller receive-frame buffers 53 (or 53 ′).
- receive-frame buffer 53 There is one receive-frame buffer 53 provided for each port 116 .
- one of the three frame buffers ( 53 , 53 ′ or 55 ) has data available, it will be selected to use the receive-validation logic 595 and the CRC-checker logic 596 (see FIG. 11 ).
- non-data frames are transmitted on the primary port.
- non-data frames may be received on either port.
- the data payload is read from off-chip and written to the data-frame buffer 55 and stored until the interface transfer can begin. (The header, CRC, and frame delimiters are added after the frame is read from the frame buffer.)
- non-data frames may be received on either the primary or alternate port.
- the non-data frames are written to the receive-frame buffers 53 or 53 ′ to be held until the frame buffer is given access to the receive-validation logic 595 and CRC-checker logic 596 .
- Priority is given to the data-frame buffer 55 in order to provide the data transfer with the highest performance possible.
- the non-data frames will be handled when the data transfer pauses or is complete. If one of the receive-frame buffers 53 fills, such that loop Buffer-to-Buffer Credit is no longer available, the write/read operation of the data-frame buffer 55 will be suspended to free receive-frame buffer space so that Buffer-to-Buffer Credit will again be available.
- Inbound data frames may accumulate in the data-frame buffer 55 during this time, since new frames may be written to the data-frame RAM 555 while the read was suspended for a short time.
- Outbound data frames available for the interface may be reduced temporarily during this time, since frames may be read from the RAM 555 while the write was suspended for a short time.
- FIG. 7 is a block diagram of fibre-channel receive-path and frame-buffers block 50 (see FIG. 3 ).
- Receive-path and frame-buffers block 50 processes received frame(s) and sends the frame(s) directly off-chip (to off-chip buffer 111 ) or to the single-frame-transmit circuit 70 or stores the frame(s) in one of the three frame buffers that receive frames (receive-non-data-frame buffers 53 or 53 ′, or data-frame buffer 55 ).
- the receive path 50 includes the pre-buffer-receive-frame processing (blocks 51 and 51 ′), data-frame-buffer multiplexer 52 , port-A and port-B receive-non-data-frame buffers 53 and 53 ′ respectively, data-frame buffer 55 , data-frame-buffer transfer-length counters 54 , frame-buffer controller 56 , common receive path 59 , and buffer interface 58 blocks.
- the inputs to block 51 include LOOP A DATA 4026 from FIG. 4 , and LOOP A STATES AND CONTROL 6422 (which also inputs to FIG. 6 ).
- the inputs to block 51 ′ include LOOP B DATA 4027 from FIG. 4 , and LOOP B STATES AND CONTROL 6432 (which also inputs to FIG. 6 ).
- the inputs to data-frame buffer 55 include OFF-CHIP BUFFER DATA 3051 .
- Signal DATA XFER CONTROL 7521 controls data-frame buffer multiplexer 52 .
- Signal BUF_PAUSE 7561 signals to frame-buffer controller 56 that a pause is required (usually due to a buffer not being able to keep up with a transfer-rate bandwidth).
- Signal LD_COUNTERS 7541 signals to data-frame-buffer transfer-length counters 54 to load counter values.
- Output signal BXFR_CNT_ZERO 7542 indicates that all the data of a transfer is in the selected frame buffer.
- Frame buffer controller 56 provides read enable signals RD_ENABLE 7532 to the port A receive-non-data frame buffer 53 , RD_ENABLE 7552 to the data-frame buffer 55 , and RD_ENABLE 7533 to the port B receive-non-data frame buffer 53 ′.
- Buffer-control-interface 58 provides select, strobe, and/or enable signals CONTROLS FOR OFF-CHIP BUFFER 7589 for off-chip buffer 111 .
- Output DATA TO OFF-CHIP BUFFER 3052 provides received data frames and non-data frames to off-chip buffer 111 .
- Data-frame-buffer multiplexer 52 selects the data and the outputs of pre-buffer-receive state machine 512 from the port which has the DATA XFER CTL 7521 bit set.
- the output of this multiplexer 52 provides the data-frame buffer 55 with data and state signals ( 8511 and 8512 , respectively, in FIG. 10 ) so that data can be written into data-frame-buffer RAM 555 (see FIG. 10 ).
- FIG. 8 is a block diagram of a fibre-channel pre-buffer-receive-frame-processing-path circuit 51 , which prepares a frame received from the fibre channel 1250 to be inputted into one of the three frame buffers ( 53 , 53 ′, or 55 ).
- Pre-buffer-receive-path block 51 includes pre-buffer-receive-framing state machine 512 , pre-buffer-receive-frame length counter 515 (and its multiplexer 514 ), and EOF-modifier logic 513 .
- This block 51 is duplicated (i.e., implemented once in each port) for Port A and Port B, since frames may be received on both ports simultaneously.
- Pre-buffer-receive-framing state machine 512 monitors the input stream to determine when frames and R_RDY's are being received. When an SOF is detected, signals are generated for each word of the header, the payload, and frame delimiters. This state machine 512 checks for invalid primitives received during the header or payload and for transfers that violate the maximum frame length (possibly due to a corrupted EOF).
- Pre-buffer-receive-frame length counter 515 is loaded at the beginning of a frame with the maximum frame length (as selected by multiplexer 514 ) of either the command, other or data-buffer areas of the frame buffer which will be the destination of the received frame, based of the R_CTL field of the header of the received frame. If the counter reaches zero before the EOF is detected, a length error is detected. This function helps prevent overrunning the allocated space in the frame buffers.
- EOF-modifier logic 513 checks an inbound frame to see whether it is a data frame, and generates an enable for the frame buffers. EOF-modifier logic 513 captures the Routing-Control field of the inbound frame to be used by the pre-buffer-receive-frame length counter 515 . EOF-modifier logic 513 also modifies the EOF field so that more-detailed status information may be passed through the frame buffers to the common receive path 59 .
- Input signals LOOP A DATA 4026 and LOOP B DATA 4027 from FIG. 4 are coupled to EOF-modifier logic 513 .
- LOOP A STATES AND CONTROL 6422 and LOOP B STATES AND CONTROL 6432 (which also input to FIG. 6 ) provide status information regarding the loop 1250 to state machine 512 .
- MAX FRAME SIZE 8517 provides information regarding the maximum frame size for data frames, control frames and other frames to multiplexer 514 and counter 515 .
- Output signals PRE-BUFFER-RECEIVE DATA 8511 and PRE-BUFFER-RECEIVE STATES 8512 provide data and state information to non-data buffers 53 and 53 ′ (see FIG. 9 ), and data-frame buffer 55 .
- FIG. 9 is a block diagram of a fibre-channel receive-non-data-frame-buffer circuit 53 .
- the receive-non-data-frame buffer 53 includes the receive-frame-buffer write control 533 , receive-frame-buffer read control 534 , receive-frame-buffer RAM 535 , receive-frame-buffer status block 536 , and receive-frame-buffer frame counter 531 .
- This circuit 53 is implemented once each for both Port A and Port B, since frames may be received on both ports simultaneously.
- Receive-frame-buffer write-control block 533 generates address (WPTR 9537 ), data (WDAT 9536 ), and write enables (WE 9539 ) for the random-access memory (“RAM”) in receive-frame-buffer RAM 535 .
- WPTR 9537 address
- WDAT 9536 data
- WE 9539 write enables
- the state enables from pre-buffer-receive state machine 512 are used to develop a write enable (WE 9539 ) to the RAM 535 .
- the address is incremented and a wrap bit (WRAP 9538 ) is provided to be used by receive-frame-buffer status block 536 to determine how much space is available in receive-frame-buffer RAM 535 .
- WRAP 9538 wrap bit
- the data from loop port circuit 20 are translated from sixteen bits wide to thirty-two bits wide and a flag bit is developed to indicate an SOF or EOF delimiter.
- the CRC from the received frame is passed through non-data-frame-buffer RAM 535 to protect the data. That is, the CRC information as received from the fibre channel is stored into the non-data-frame buffer 53 along with the data, and then is checked as the data are read out of the non-data-frame buffer 53 (e.g., as they are transferred to the off-chip buffer 111 ), in order that any errors that arise in the data as they reside in the non-data-frame buffer 53 can be detected (of course, errors that arose in the data in transit on the fibre channel loop 1250 are also detected).
- the input signals to block 53 include PRE-BUFFER-RECEIVE DATA 8511 and PRE-BUFFER-RECEIVE STATES 8512 (see FIG. 8 ), and registered (i.e., versions of signals that are clocked into a register for later use) MPU DATA 9533 and MPU ADDRESS 9534 from microprocessor 112 .
- Receive-frame-buffer read-control block 534 generates a read address (RPTR 9541 ) for the receive-frame-buffer RAM 535 and captures data (RDAT 9540 ) from the RAM 535 .
- RPTR 9541 a read address for the receive-frame-buffer RAM 535
- RDAT 9540 data from the RAM 535 .
- frame-buffer controller 56 selects the receive-non-data-frame buffer 53
- a read to the receive-frame-buffer RAM 535 is enabled.
- the address is incremented and a wrap bit (WRAP 9542 ) is provided to be used by the receive-frame-buffer status block 536 to determine how much space is available in the receive-frame buffer RAM 535 .
- the data from the receive-frame-buffer RAM 535 are captured into a register and monitored for the flag bit to determine the frame's start and end.
- the input signals to block 534 include REGISTERED READ_ENABLE 9535 from microprocessor 112 .
- the output signals from block 534 include RECEIVE NON-DATA BUFFER DATA 9543 , and NON-DATA VALID READ 9546 .
- Receive-frame-buffer RAM 535 includes a synchronous RAM.
- the RAM is thirty-three bits wide (a thirty-two-bit data word plus an SOF/EOF flag bit) and three-hundred and four (304) words long.
- the SOF, header, payload, CRC, and EOF of a received non-data frame are written to the RAM 535 to be held until access to the off-chip buffer 111 and common receive path 59 is available.
- a built-in self-test controller allows the receive-frame-buffer RAM 535 to be tested with data patterns developed specifically for the physical layout of the memory.
- Receive-buffer-status block 536 compares the write and read pointers of the receive-frame-buffer RAM 535 to determine whether the buffer is empty and, if the buffer is not empty, how many frames of space are available in the receive-frame-buffer RAM 535 .
- the outputs of this block 536 (AVAILABLE SPACE 9545 ) are used by the frame-buffer controller 56 to determine whether receive-frame-buffer RAM 535 requires access to the common receive path 59 .
- the outputs are also used by the Buffer-to-Buffer Credit control logic 603 (see FIG. 12 ) to determine whether credit is available.
- Receive-frame-buffer frame-counter block 531 counts the number of frames currently in the receive-frame-buffer RAM 535 .
- the counter 531 is incremented when a frame is written into receive-frame-buffer RAM 535 and decremented when a frame is read from receive-frame-buffer RAM 535 .
- the count (COUNT OF FRAMES IN BUFFER 9544 ) is used by the Buffer-to-Buffer Credit control 603 to determine whether credit is available. In one embodiment, all inputs (except the clock) going into the RAM are delayed, to provide adequate hold time.
- FIG. 10 is a block diagram of a fibre-channel data-frame buffer circuit 55 .
- Data-frame buffer 55 includes data-frame-buffer write control 553 , data-frame-buffer read control 554 , data-frame-buffer RAM 555 , data-frame-buffer status block 556 , data-frame-buffer frame counter 551 , and data-frame-capture block 552 . Only one data transfer is allowed at any given time, so data-frame buffer 55 is shared by Port A and Port B.
- Data-frame-buffer write-control block 553 generates address (WPTR 9555 ), data (WDAT 9554 ), and write enables (WE 9557 ) for the data-frame-buffer RAM 555 .
- WPTR 9555 address
- data data to be written to the disc
- state enables from pre-buffer-receive state machine 512 are used to develop a write enable (WE 9557 ) to the memory 555 .
- enables from frame-buffer controller 56 are used to develop a write enable (WE 9557 ) to the memory 555 .
- wrap bit wrap 9556
- the address is incremented and a wrap bit (wrap 9556 ) is provided to be used by the data-frame-buffer status block 556 to determine how much data/space is available in data-frame-buffer RAM 555 .
- wrap bit wrap 9556
- the data from loop port circuit 20 are translated from sixteen bits wide to thirty-two bits wide and a flag bit is developed to indicate an SOF or EOF delimiter.
- the CRC from the received frame is passed through data-frame-buffer RAM 555 to protect the data.
- the CRC information as received from the fibre channel is stored into the data-frame buffer 55 along with the data, and then is checked as the data are read out of the data-frame buffer 55 (e.g., as they are transferred to the off-chip buffer 111 ), in order that any errors that arise in the data as they reside in the data-frame buffer 55 can be detected (of course, errors that arose in the data in transit on the fibre channel loop 1250 are also detected).
- the data from off-chip buffer 111 are translated (or converted) from sixteen bits wide to thirty-two bits wide and parity is generated to protect the data.
- the input signals to block 553 include PRE-BUFFER-RECEIVE DATA 8511 and PRE-BUFFER-RECEIVE STATES 8512 (see FIG. 8 ), and DATA FROM OFF-CHIP BUFFER 3051 (see FIG. 3 ).
- data in the data-frame-buffer RAM 555 are protected by CRC information if the data are passing from the fibre-channel loop 1250 (see FIG. 2 ) though data-frame-buffer RAM 555 and then to the off-chip buffer 111 , but are protected by parity if they have passed from the off-chip buffer 111 to RAM 555 for transmission to the fibre-channel loop 1250 (in the latter case, CRC information is added to the data going onto the fibre channel after the data leave data-frame-buffer RAM 555 ).
- Data-frame-buffer read-control block 554 generates a read address (RPTR 9559 ) for data-frame-buffer RAM 555 and captures data (RDAT 9558 ) from RAM 555 .
- the frame buffer controller 56 selects the data-frame buffer 55 and a read to the memory is enabled.
- the transmit-frame state machine 81 (see FIG. 15 ) enables the read of the memory.
- the address is incremented and a wrap bit (WRAP 9560 ) is provided to be used by the data-frame buffer status block 556 to determine how much data/space is available in the frame buffer.
- the data from the frame buffer RAM are captured into a register and monitored for the flag bits to determine the frame's start and end, so that an enable can be developed for the common receive path 59 to indicate when data are valid.
- the data from data-frame-buffer RAM 555 are captured into a register and the parity is checked.
- the output signals from block 554 include DATA FRAME BUFFER DATA 9564 , DATA VALID READ 9563 , and DATA PARITY ERROR 9562 .
- Data-frame-buffer RAM 555 includes a synchronous RAM and, in one embodiment, a built-in self-test controller.
- the RAM is thirty-six bits wide (a thirty-two-bit data word plus four SOF/EOF flag bits or parity bits) and 3,232 words long.
- For a write operation (data to be written to the disc), the SOF, header, payload, CRC, and EOF of a received data frame are written to the RAM 555 to be held until access to the off-chip buffer 111 and common receive path 59 is available.
- For a read operation data to be read from the disc, the payload only is written to RAM 555 to be held until the loop 1250 can be opened and the data transmitted. In one embodiment, all inputs (except the clock) going into the RAM are delayed, to provide adequate hold time.
- Data-frame-buffer status block 556 compares the write and read pointers of data-frame-buffer RAM 555 to determine whether the buffer is empty and, if the buffer is not empty, how many frames of data/space are available in the buffer. For a write operation, the outputs of this block 556 (AVAILABLE SPACE 9561 ) are used by the frame-buffer controller 56 to determine that the data-frame buffer 55 requires access to the common receive path 59 . The outputs are also used by the Buffer-to-Buffer Credit control 603 (see FIG. 12 ) to determine whether credit is available. For a read operation (data to be read from the disc), the loop-control block 40 monitors the amount of data in the frame buffer to determine whether a frame can be transmitted.
- a data-frame-buffer threshold must be met to arbitrate to open the loop 1250 .
- a data-frame-buffer-hold threshold is also generated to allow the loop 1250 to be held open in the event that a whole frame is not available but is in process (is being accumulated in data-frame buffer 55 and/or off-chip buffer 111 ).
- Data-frame-buffer frame-counter block 551 counts the number of frames currently in data-frame-buffer RAM 555 using input signals FRAME_OUT 9550 .
- Counter 551 is incremented when a frame is written into RAM 555 and decremented when a frame is read from RAM 555 .
- the count (signal COUNT OF FRAMES IN BUFFER 9566 ) is used by the Buffer-to-Buffer Credit control 603 to determine whether credit is available.
- Data-frame-capture block 552 monitors the received data frame (using input signal ENABLE DATA WRITE DETECT 9551 ) when the enable-capture mode is enabled and captures various fields of the frame header. These values (DATA CAPTURE OUTPUT 9565 ) may then be read by microprocessor 112 .
- data-frame-buffer transfer-length counters 54 control includes two counters that control how read data are prefetched into the data-frame buffer 55 .
- a data-frame-buffer transfer-length counter (in block 54 ) is used to determine how much data should be fetched from the off-chip buffer 111 for a fibre-channel transfer.
- a data-frame-buffer transmit-length counter (in block 54 ) is used to determine how much data should be fetched from the off-chip buffer 111 before pausing for controller 56 to reevaluate which frame buffer should get access to the off-chip buffer 111 (a process described in the following paragraph).
- Frame-buffer controller 56 determines which of the three frame buffers (i.e., data-frame buffer 55 , port-A receive-non-data-frame buffer 53 , or port-B receive-non-data-frame buffer 53 ′) should be granted access to the resources of off-chip buffer 111 . If a port is in a loop-initialization state and that port's receive-non-data-frame buffer 53 (i.e., 53 or 53 ′) is not empty, the loop-initialization frame is given highest priority, so that loop initialization can progress. The data transfer is given the next-highest priority, and will continue as long as the receive-non-data-frame buffers 53 do not fill up. If one of the receive-non-data-frame buffers 53 no longer has room for a frame, that particular receive-non-data-frame buffer 53 will be granted access to the off-chip buffer 111 to drain some frames, and then the data transfer will be resumed.
- the three frame buffers i.e.,
- the first frame buffer ( 55 , 53 or 53 ′) that requires the off-chip buffer resources will be granted access to the off-chip buffer 111 .
- Port A will be given access first.
- the frame-buffer-control block 56 gives a frame buffer access to the off-chip buffer 111 , it continues to service that frame buffer unless the alternate port's receive-non-data-frame buffer 53 fills, or initialization is started, or a data transfer is started. Frames on a given port will be transferred to the off-chip buffer 111 in order-of-delivery to the port.
- the transfer of data between frame-buffer controller 56 and the off-chip buffer 111 may be paused by the buffer-control-interface logic 58 in the event that the data rate from the fibre channel cannot be sustained to off-chip buffer 111 .
- FIG. 11 is a block diagram of common receive path circuit 59 , which takes the frame coming out of one of the frame buffers ( 53 , 53 ′, or 55 ) and prepares the frame for the off-chip buffer 111 .
- This logic 59 recognizes the SOF and captures information from the received header for validation checking. The CRC is checked and the frame is routed to the appropriate buffer area of off-chip buffer 111 .
- Receive-buffer-decode block 591 decodes the start-of-frame and end-of-frame delimiters (derived from input data signal FRM BUFFER DATA 9570 and enable signal VAL_READ 9571 ), and generates signals to be used by the common receive path 59 for frame-validity checks.
- the EOF delimiter may have error status (frame-length error or running-disparity error) embedded which is converted for the receive-path blocks.
- Receive-framing state machine 592 monitors the input stream to determine frame boundaries. When an SOF is detected, signals are generated to allow each word of the header to be captured and to enable the CRC checker 596 , header-validation controls 595 , and buffer controls 598 (along with 5991 , 5992 , and 5993 ). State machine 592 checks for invalid primitives received during the header and for transfers that violate the maximum frame length allowed. Since the common receive path 59 can be paused temporarily, the state outputs of state machine 592 are pulses that go active and then inactive in order that the frame-capture-and-validation blocks are enabled properly.
- Receive-framing length counter (mux 5931 and counter 5932 ) is loaded at the beginning of a frame with the maximum frame length of either the command, other, or data frame (from signals MAX SIZE INPUTS 9572 ), based on the R_CTL field from the header of the received frame. If the counter reaches zero before the EOF is detected, a frame-length error is detected and the frame is marked invalid. This function helps prevent overrunning the allocated space in the off-chip buffer 111 for the frame.
- Receive-frame-header-capture block 594 uses the signals from the receive-framing state machine 592 to capture the various fields of a received header. The captured values are used by the frame-validation logic.
- CRC-checker block 596 checks the CRC at the end of the frame. If a CRC error is detected (indicated by CRC STATUS 9596 ), the frame is marked invalid.
- the CRC checker 596 is enabled by the receive-framing state machine 592 . Contents of the header field, payload field, and CRC word are processed.
- frame buffers that temporarily store fibre-channel frames allow frames to be received at the maximum fibre-channel-interface data-transfer rate.
- the frame may then be transferred to off-chip storage at a slower, more manageable rate.
- Various mechanisms such as parity, CRC, or other redundancy functions, are optionally used to protect the data while they are being stored in the frame buffer.
- data-integrity checking is enhanced by passing the received fibre-channel cyclic-redundancy code (“CRC”) through the frame buffers with the data (i.e., the CRC is stored into the frame buffer with the frame, and then read out with the frame at a later time), extra parity bits that would make the RAM wider may be eliminated.
- the frame buffer is data-frame buffer 55 and/or receive-non-data-frame buffer 53 or 53 ′).
- the CRC is checked after the data are read from the RAM and before they are transferred off-chip (i.e., to off-chip buffer 111 ). Extra parity bits on the inbound data paths can also be eliminated from the interface to the inbound side of the frame buffer and from the outbound side of the frame buffer to the input of the CRC checker 596 .
- the common receive path logic 59 can be shared between the non-data-frame buffers 53 and 53 ′, and the data-frame buffer 55 . Only one CRC checker 596 is necessary, since the CRC is checked just before the frame goes off-chip (i.e., is transferred from the on-chip buffer 53 , 53 ′, or 55 to the off-chip buffer 111 ).
- this mechanism is also used for the opposite direction, when data are transferred from the slower off-chip buffer 111 , temporarily stored in a frame buffer, and transmitted at the maximum data-transfer rate on the fibre-channel interface.
- the R_CTL field is decoded by routing-control-decode block 5933 to determine the type of frame being received.
- the R_CTL field is used to route the frame to the appropriate area of the off-chip buffer 111 , and to determine which validity checks are to be made on the frame.
- Header-validation logic 595 analyzes the contents of the receive-frame header. Based on the R_CTL field of the frame, various fields are verified. If any of the validity checks fail on a non-data frame, the frame is considered invalid. If a validity check fails on a data frame when a data-write transfer is active, microprocessor 112 is notified.
- Receive-frame-status block 597 gathers information about the received frame and blocks invalid data frames from going off-chip, and generates signal FRAME STATUS 9580 . If a frame is not valid, it is essentially ignored (unless a data transfer is active).
- Command counter 5992 is used to track how many command frames are contained in the command area of the off-chip buffer 111 . When a valid command is received, this counter 5992 is incremented. When microprocessor 112 has finished with a command, it must decrement the command counter 5992 (using signal MPU DECREMENTS 9574 from microprocessor 112 ). Block 5992 outputs interrupt request CMD RCVD IRQ 9578 .
- “Other” space counter 5991 is used to track how much space remains in the “other” area of the off-chip buffer 111 (using signal MPU INCREMENTS 9575 from microprocessor 112 ), for giving out Buffer-to-Buffer Credit. This counter 5991 is decremented when a valid frame that is neither a command nor a data frame is received, or when a command frame is received when the off-chip buffer 111 is full. When microprocessor 112 has finished with a frame, it must increment the “other” space counter 5991 to indicate that there is space for another frame. The “other” space counter 5991 indicates (signal OTHER COUNT 9576 , and interrupt request OTHER RCVD IRQ 9577 ) the number of frames with will fit in the “other” area of the off-chip buffer 111 .
- receive-frame-buffer control 598 (which generates signals INCR/DECR 9573 to decrement other counter 5991 and increment command counter 5992 ) and last-4-generation logic 5993 generate signals (CONTROLS FOR OFF-CHIP BUFFER 7589 ) to the logic of off-chip buffer 111 to direct the frame to the proper area.
- FIG. 12 is a block diagram of a fibre-channel transfer-control circuit 60 , which includes control logic for the receive and transmit portions of the logic.
- Transfer-control circuit 60 includes transfer controls and Buffer-to-Buffer Credit control 603 .
- Data-available/space-counter block 604 provides an indication (DATA AVAILABLE 6019 ) of how much off-chip space or data is available, depending on whether the operation is a data-write or a data-read. On a data-write transfer, data-available/space-counter block 604 is used to indicate how much space is available (in terms of frames) in the data portion of the off-chip buffer 111 to allow BB_Credit to be issued. On a data-read transfer, data-available/space-counter block 604 is used to indicate how many data frames are available in the off-chip buffer 111 . The difference in the buffer pointers is compared against microprocessor-loadable (i.e., programmable) values which indicate the number of words per frame(s).
- microprocessor-loadable i.e., programmable
- Buffer-to-Buffer Credit control logic 603 issues buffer credit (signal PORT BB_CREDIT AVAILABLE TO TRANSMIT R_RDY 6017 ) to the connected port to allow frames to be sent. This credit is issued by sending R_RDY's. The amount of credit given to any port is determined by:
- the Buffer-to-Buffer Credit is determined on a per-port basis.
- the Buffer-to-Buffer Credit control block 603 generates signals to the loop-Port A/B open-control blocks 42 and 42 ′ (see FIG. 6 ) to indicate when credit is available.
- the loop-control block 40 will then control the transmission of R_RDY's.
- Received-R_RDY counters 606 determine how many R_RDY's of credit have been received on each port from inputs PORT R_RDY RECEIVED 6010 (one each for Port A and Port B) and PORT FRAME TRANSMITTED 6011 (one each for Port A and Port B), and output signal PORT CREDIT AVAILABLE TO TRANSMIT A FRAME 6020 .
- Transmitted-R_RDY counters 601 determine how many R_RDY's of credit have been transmitted on each port from inputs PORT R_RDY TRANSMITTED 6001 (one each for Port A and Port B) and PORT FRAME RECEIVED 6002 (one each for Port A and Port B).
- Sequence counter 607 is used on data-write transfers to check that received frames arrive in order and is used on data-read transfers to generate the SEQUENCE COUNT 6022 transmitted in the headers of data frames.
- Relative-offset counter 608 is used on data-read transfers to generate the RELATIVE-OFFSET COUNT 6023 transmitted in the headers of data frames.
- Transfer-length counter 609 is used to determine how much data to transfer and provides an indication (TRANSFER LENGTH COUNT 6024 ) when the transfer has completed.
- FIG. 13 is a block diagram of single-frame-transmit-path circuit 70 for a fibre channel.
- One embodiment of the present invention provides a fibre-channel-loop-interface circuit which includes a dedicated transmit-frame buffer 73 for loop initialization and responses. Having such a dedicated transmit-frame buffer 73 allows one port of a dual-port node to be transmitting initialization or response frames while another port is transmitting or receiving (i.e., “communicating”) data frames.
- the loop 1250 must be re-initialized, and the dedicated transmit-frame buffer 73 allows this function to occur without disrupting a data transfer or other function that could be in progress on the other loop connected to the other port of the dual-port node interface. Both ports can be simultaneously initialized as well. Further, this dedicated transmit-frame buffer 73 allows responses, acknowledgments, or other non-data transfers to be transmitted out of one port while the other port is in use.
- single-frame-transmit-path circuit 70 in conjunction with microprocessor 112 operates to provide loop-initialization and response functions for node interface 1220 .
- Transmit-frame buffer 73 accepts inputs MPU LOADABLE 7002 (data from microprocessor 112 ), RECEIVE PATH DATA 7003 and RECEIVE PATH CONTROLS 7004 (frames from the receive path 50 ). Transmit-frame buffer 73 generates outputs MPU READ DATA 3095 , and, through blocks 74 and 76 , XMT_DPTH DATA 7007 (outbound frames including CRC which are to be transmitted).
- the single-frame-transmit-path circuit 70 accepts requests from the loop-control circuit 40 to transmit frames which reside in the single-frame transmit-frame buffer 73 .
- This circuit 70 generates the proper frame delimiter, reads the header and payload from the frame buffer, and generates the proper CRC with CRC generator 76 .
- This circuit 70 also generates signals to the loop port circuit 20 for transmitting the current-fill-word and to indicate when the end-of-frame (“EOF”) is being transferred, to allow the encoder to generate the second character of the EOF based on the current running disparity.
- EEF end-of-frame
- Single-frame-transmit state machine 72 accepts requests (via signal SEND_FRAME 7001 ) from the loop-control circuit 40 to transmit a frame.
- this state machine 72 provides selects (i.e., select signals) for each portion of the frame sent to the transmit multiplexer 74 , in order to allow the frame delimiters, header, and payload to be transferred at the proper times.
- This state machine 72 also generates enables to the CRC generator 76 . Control is also provided to the 8B/10B encoder 27 (see FIG. 4 ) to determine when to transmit the EOF signal. When a frame has been transmitted, signals are generated and sent back to loop-control circuit 40 to allow it to continue operation.
- Single-frame-transmit frame-length counter 71 is used to allow the hardware to determine how long the frame is, in order to allow the final CRC to be output at the appropriate time (the CRC is repeatedly calculated one-word-at-a-time and accumulated as the data portion of the frame is output).
- Counter 71 is loaded at the beginning of the frame from a transmit-frame-length register and enabled while the frame is being transmitted.
- Counter 71 is enabled by single-frame-transmit state machine 72 and provides status (i.e., status information) back to state machine 72 to determine when to enable CRC generator 76 .
- the output of single-frame-transmit output multiplexer 74 is an input to the CRC generator 76 to determine the CRC residual word included before the EOF, which feeds through multiplexer 79 to loop-port circuit 20 (see FIG. 3 ).
- the single-frame-transmit state machine 72 generates the selects for this multiplexer 74 to allow the frame delimiters, header, and payload to be transferred at the proper times.
- Single-transmit-fill-character generator block 75 determines when a K-character, current-fill-word, and EOF delimiters are transmitting from the single-frame-transmit circuit 70 .
- the output signals XMIT CONTROLS 7010 go to the transmit-path multiplexer 79 which determines which loop has access to the single-frame-transmit-path circuit 70 .
- FIG. 14 is a block diagram of a fibre-channel transmit-frame buffer 73 .
- a single-frame transmit-frame buffer 73 includes single-frame-buffer write control 733 , single-frame-buffer read control 734 , and single-frame-transmit-frame buffer RAM 735 .
- Single-frame transmit-frame buffer 73 is used to store received loop-initialization frames, outbound loop-initialization frames, and outbound single frames. Forty words of transmit buffer RAM 735 are reserved for Port A and forty are reserved for Port B, in one embodiment.
- Single-frame-buffer write-control block 733 generates address (WPTR 1411 ), data (WDAT 1410 ), and write enables (WE 1412 ) for single-frame-buffer RAM 735 .
- the inputs include DATA FROM RECEIVE PATH 1401 , PORT A FRAME 1403 , PORT B FRAME 1404 , registered MPU DATA 9533 , and registered MPU ADDRESS 9534 .
- the frame is first stored in one of receive-non-data-frame buffers ( 53 or 53 ′) until it can move through the common receive path 59 for validation.
- the frame is written into single-frame transmit-frame buffer 73 .
- This also allows the off-chip buffer 111 to be more dedicated to a data transfer on the alternate port.
- the data from the loop port are translated from sixteen bits wide to thirty-two bits wide and the starting address of the frame is determined on the basis of the port with which the frame is associated.
- Microprocessor 112 also may write to single-frame transmit-frame buffer 73 to modify a loop-initialization frame or to set up an outbound frame. Parity is generated to protect the data in single-frame transmit-frame buffer 73 .
- Single-frame-transmit-buffer read-control block 734 generates a read address (RPTR 1414 ) for the single-frame-transmit buffer RAM 735 and captures data (RDAT 1413 ) from the RAM 735 .
- Inputs include SEND FRAME 1407 and XMIT STATES 1408 .
- a read of the single-frame-transmit buffer RAM 735 is enabled (outputting SINGLE_FRAME TRANSMIT BUFFER DATA 1415 ).
- Microprocessor 112 may also read this frame buffer to access a received loop-initialization frame.
- the data from the transmit-frame buffer RAM 735 are captured into a register and parity is checked (generating output PARITY ERROR 1416 ).
- Single-frame-transmit buffer RAM 735 includes a synchronous RAM and, in one embodiment, a built-in self-test controller.
- the RAM is thirty-six bits wide (a thirty-two-bit data word plus four bits of parity) and eighty locations long.
- the header and payload of a frame are placed into transmit-frame buffer RAM 735 to be held for microprocessor 112 to examine or for transmission on the loop 1250 .
- all inputs (except the clock) going into the RAM are delayed, to provide adequate hold time.
- FIG. 15 is a block diagram of data transmit path circuit 80 , which accepts requests from the loop-control logic 40 to transmit frames. Circuit 80 generates the proper frame delimiter, builds the header from the microprocessor-loadable registers, and generates the CRC. Circuit 80 also generates signals to the loop port circuit 20 for transmitting the current-fill-word and to indicate when the EOF is being transferred, to allow the encoder to generate the second character of the EOF based on the current running disparity.
- Data-transmit-framing state machine 81 accepts requests (signal SEND-FRAME 8001 ) from the loop-control logic 40 to transmit a frame. When a frame is transmitted, this state machine 81 provides selects for each portion of the frame to transmit multiplexer 86 to allow the frame delimiters, header, and payload to be transferred at the proper times. State machine 81 also generates enables to the CRC generator 87 . When a frame has been transmitted, signals are generated back to the loop-control logic 40 to allow it to continue.
- Data-transmit frame-length counter 82 is used to allow the hardware to determine how long the frame is, to allow the CRC to be generated at the appropriate time.
- Data-transmit frame-length counter 82 is loaded at the beginning of the frame from the transmit-frame length-register and enabled while the frame is being transmitted.
- Data-transmit frame-length counter 82 is enabled by the transmit-framing state machine 81 and provides status back to the state machine to determine when to enable the CRC generator 87 .
- the output of data-frame-transmit output multiplexer 86 is an input to the CRC generator 87 which feeds the data transmit path multiplexer 79 .
- the data transmit framing state machine 81 generates the selects for this multiplexer 86 to allow the frame delimiters, header, and payload to be transferred at the proper times.
- the inputs include FRAME BUFFER DATA 8004 , HEADER REGISTERS 8005 , TRANSFER COUNTS 8006 , and TRANSFER-READY PAYLOAD 8007 .
- the outputs include STATES TO FRAME BUFFER READ CONTROL 8009 .
- a delimiter generator in multiplexer 86 determines which start-of-frame (SOF) and end-of-frame (EOF) primitive to use when transmitting frames.
- the output of the data-frame-transmit-output multiplexer 86 is transferred through the CRC generator 87 to determine the CRC residual word included before the EOF. Enables for CRC generator 87 come from the data-transmit-framing state machine 81 . Control is also provided to the 8B/10B encoder 27 (see FIG. 4 ) to determine when to transmit the EOF signal.
- the output of CRC generator 87 is XMT_DPTH DATA 8008 (outbound frames including CRC which are to be transmitted).
- Data Transfer Fill Character block 85 determines when a K-character, current-fill-word, and EOF delimiters are transmitting from the data transmit path circuit 80 .
- the output signals XMIT OUTPUTS 8010 go to transmit path multiplexer 79 which determines which loop has access to the data transmit path circuit 80 .
- transmit-path multiplexer 79 (see FIG. 3 ) is an input to the Port-A and Port-B arbitrated-loop logic 26 within loop-port circuit 20 .
- the data and control signals from the single-frame path 70 and data path 80 are selected by transmit-path multiplexer 79 to the proper port.
- Transmit-path multiplexer 79 also multiplexes the R_RDY primitive to the ports. This allows both ports to transmit simultaneously.
- a node interface 1220 of loop port 116 must arbitrate for access to the loop 1250 .
- a priority system is used to determine which port gains control of the loop 1250 , and a “fairness” scheme is used to assure that ports are not starved.
- the disc drive 100 usually is given a lower priority than a CPU 1202 , with the result that the drive 100 may have to wait to win arbitration until higher-priority devices complete their access.
- the node interface 1220 of loop port 116 gains control of the loop 1250 , it sends as many frames as possible before closing the loop 1250 , in order to avoid unnecessary arbitration cycles.
- the present invention provides a mechanism for enhancing loop performance by changing the rules for the decision of whether or not to close the loop 1250 , based on data availability to the port, which thus reduces overall loop overhead.
- the port determines whether another frame is available. If data are no longer available (for example, if a full frame is not available for transmission), then the loop 1250 is closed. Data may again become available shortly afterwards, so the port must arbitrate again later and win arbitration before continuing the transfer. If this happens as the last frame of the transfer becomes available, completion of the transfer is delayed, which may result in a delay before the next command can be processed.
- the present invention provides a mechanism for controller architecture designs which allows the loop 1250 to be held open by a port if data will shortly be available to the port. This may reduce the number of times a node interface 1220 of loop port 116 must arbitrate during an outgoing data transfer, and may thus allow transfers to complete sooner.
- the loop 1250 is held open in anticipation of sufficient further data becoming available to a port (to justify that port's retaining control of the loop) when both of the following conditions are met:
- the on-chip data-frame buffer 55 is large enough to hold at least six frames of data (i.e., six frames of the maximum frame size of 2112 bytes).
- Data that are being transmitted are first moved into off-chip buffer 111 (e.g., from a disk platter 134 ), and then into on-chip data-frame buffer 55 .
- off-chip buffer 111 e.g., from a disk platter 134
- on-chip data-frame buffer 55 e.g., from a disk platter 134
- data can be transferred out of data-frame buffer 55 at up to about one-hundred-and-six megabytes per second, while, in one embodiment, the transfer from off-chip buffer 111 to on-chip data-frame buffer 55 occurs at a slower rate.
- on-chip data-frame buffer 55 If less than a full frame is already moved into on-chip data-frame buffer 55 , it may still be possible to start a frame transfer and complete transfer of the entire frame out at the full fibre-channel speed of (about) one-hundred-and-six megabytes per second, as long as the last portion of the frame is moved into the on-chip data-frame buffer 55 before those data are needed for transfer out.
- the loop 1250 is held open if at least one-half a frame of data is contained in on-chip data-frame buffer 55 , and at least one frame of data are contained in off-chip buffer 111 .
- this amount of data needed to keep control of the loop 1250 is about one-thousand bytes of data in on-chip buffer 55 (i.e., one-half of a 2,112-byte frame), and about two-thousand bytes of data in off-chip buffer 111 (i.e., one 2,112-byte frame), both amounts being programmable values that are set by microprocessor 112 .
- the loop 1250 is held open if the predetermined amount of data is available, as just described, but the transfer of a frame will start only after an entire frame is available on-chip; in one such embodiment, CFW (current-fill word) signals are transmitted onto the loop 1250 until the entire frame is on-chip and transmission of the frame can start.
- CFW current-fill word
- a count of the number of frames available to transfer out is produced by the controller in off-chip buffer 111 of one embodiment of the present invention.
- a firmware-programmable count X-frames (the number of frames needed in off-chip buffer 111 to keep the loop open), is compared against the count of the number of frames available to transfer, in order to determine whether data are available to transfer into the data-frame buffer.
- the data-frame buffer is an on-chip RAM that is used to temporarily store data from off-chip slower memory, so that the data can be transmitted at the full fibre-channel interface rate.
- a second comparator is used to compare the amount of data available within the data-frame buffer against a firmware programmable count, Y-words (the number of words needed in on-chip buffer 55 to keep the loop open), to determine whether sufficient data are already in the data-frame buffer to hold the loop 1250 open.
- Y-words the number of words needed in on-chip buffer 55 to keep the loop open
- the values of X and Y are firmware programmable to allow this logic to be used with a variety of off-chip random access memory (“RAM”) solutions and transfer rates.
- One goal of the present invention is to hold the loop 1250 open and avoid an extra arbitration cycle when data will shortly be available to a port 116 .
- the loop 1250 must not be held open waiting for data to become available if the wait will be for an extended period of time (for example, for the time required to perform a head switch), since this would prevent other ports on the loop 1250 from performing transfers.
- a fibre-channel loop interface circuit ( 1220 ) has been described that includes a dedicated transmit-frame buffer ( 73 ) for loop initialization and responses. Having such a dedicated transmit-frame buffer ( 73 ) allows one port ( 116 ) of a dual-port node ( 1220 ) to be transmitting initialization or response frames while another port ( 116 ) is transmitting or receiving data frames.
- Dedicated receive buffers ( 53 and 55 ) are also provided for each port ( 116 ) of a two-port node ( 1220 ). Cyclic-redundancy code information received from the fibre-channel ( 1250 ) along with a frame is stored in one of the three frame buffers ( 53 , 53 ′ or 55 ).
- Control of a loop is maintained (i.e., the loop connection is held open) as long as a minimum amount of data, which optionally is determined by programming (called a “programmable amount of data”), is available for transmission, in order to reduce the overall amount of time spent arbitrating for control of the loop ( 1250 ).
- the present invention provides a communications channel system ( 1200 ) for using fibre-channel cyclic-redundancy code (CRC) for data integrity in an on-chip memory.
- the system ( 1200 ) includes a first channel node ( 1220 ) having a first port ( 116 ) and a second port ( 116 ), each port ( 116 ) supporting a fibre-channel arbitrated-loop communications channel ( 1250 ), each communications channel ( 1250 ) including a cyclic-redundancy code within data transmissions on the communications channel ( 1250 ).
- the system ( 1200 ) includes an on-chip frame memory ( 53 or 55 ) located on-chip in the first channel node ( 1220 ) that receives a frame and the received frame's associated CRC from the communications channel ( 1250 ), and an integrity apparatus ( 596 , 553 ) that uses the received associated CRC for data-integrity checking of the received frame that is in the on-chip frame memory ( 53 or 55 ).
- One version of this system further includes an off-chip memory ( 111 ) operatively coupled to the on-chip frame memory ( 53 or 55 ) and the integrity apparatus ( 596 , 553 ) and a verification circuit ( 596 ) within the integrity apparatus that verifies the cyclic-redundancy code while moving the received frame from the on-chip memory ( 53 or 55 ) to the off-chip memory ( 111 ).
- One such embodiment further includes a parity-generation circuit ( 553 ) that generates and appends parity to the data as they are moved from the off-chip memory ( 111 ) to the on-chip memory ( 55 ), wherein the integrity apparatus further checks and strips away the cyclic-redundancy code while moving the frame of data to the off-chip memory ( 111 ).
- a parity-generation circuit 553
- the integrity apparatus further checks and strips away the cyclic-redundancy code while moving the frame of data to the off-chip memory ( 111 ).
- the on-chip frame memory ( 55 ) further receives a data frame from the off-chip memory ( 111 ) which is devoid of a cyclic-redundancy code
- the system further includes a CRC generator ( 87 ) that generates the cyclic-redundancy code based on the data frame received from the off-chip memory ( 111 ), and a transmitter (a portion of transceiver 115 ) that transmits the data frame from the off-chip memory ( 111 ), including the generated cyclic-redundancy code, onto the communications channel ( 1250 ).
- a data frame that is to be transmitted is transferred to the on-chip frame memory ( 55 ) from the off-chip memory ( 111 ) and is stored in the on-chip frame memory ( 55 ) with parity but devoid of CRC information.
- a received data frame transferred to the on-chip frame ( 55 ) memory from the communications channel ( 1250 ) is stored in the on-chip frame memory ( 55 ) with CRC but devoid of parity information.
- a system ( 1200 ) is built according to this invention wherein the system ( 1200 ) further includes a magnetic-disc-storage drive ( 1256 ) operatively coupled to the first channel node ( 1220 ), and a computer system ( 1202 ) operatively coupled to a second channel node ( 1220 ) (or, equivalently, a computer system ( 1202 ) having a second channel node ( 1220 )).
- the second channel node ( 1220 ) is operatively coupled to the first channel node ( 1220 ) to transfer data between the first and second channel nodes ( 1220 ) through the fibre-channel arbitrated-loop communications channel ( 1250 ).
- the disk drive ( 100 ) includes a rotatable disc ( 134 ), a transducer ( 150 ) in transducing relationship to the rotating disc ( 134 ), and a first channel node ( 1220 ) having a first port ( 116 ) and a second port ( 116 ), each port ( 116 ) supporting a fibre-channel arbitrated-loop communications channel ( 1250 ).
- Each communications channel ( 1250 ) includes a cyclic-redundancy code within data transmissions on the communications channel ( 1250 ).
- the first channel node ( 1220 ) is operatively coupled to the transducer ( 150 ) to communicate data.
- the a disc drive ( 100 ) includes an on-chip frame memory ( 53 or 55 ) located on-chip in the first channel node ( 1220 ) that receives a frame and the received frame's associated CRC from the communications channel ( 1250 ), and an integrity apparatus ( 596 , 553 ) that uses the received associated CRC for data-integrity checking of the received frame that is in the on-chip frame memory ( 53 or 55 ).
- the disc drive ( 100 ) further includes an off-chip memory ( 111 ) operatively coupled to the on-chip frame memory ( 53 or 55 ) and the integrity apparatus ( 596 , 553 ), and a verification circuit ( 596 ) within the integrity apparatus ( 596 , 553 ) that verifies the cyclic-redundancy code while moving the received frame from the on-chip memory ( 53 or 55 ) to the off-chip memory ( 111 ).
- Yet another aspect of the present invention provides a communications method.
- the method includes steps of:
- Still another aspect of the present invention provides a communications channel system ( 1200 ) that includes a channel node ( 1220 ) having a first port ( 116 ) and a second port ( 116 ), each port ( 116 ) supporting a fibre-channel arbitrated-loop communications channel ( 1250 ), each communications channel ( 1250 ) including a cyclic-redundancy code within data transmissions on the communications channel ( 1250 ).
- the system also includes a buffer ( 53 or 55 ) that receives, from the channel node ( 1220 ), a frame that includes a cyclic-redundancy code, and an off-chip memory ( 111 ) separate from the buffer ( 53 or 55 ).
- the system ( 1200 ) also includes means (as described throughout this entire specification) for moving the received frame from the buffer to the off-chip memory and checking the received frame for accuracy by verifying the cyclic-redundancy code (CRC) while moving the received frame to the off-chip memory.
- the means for moving further includes means for stripping away the CRC as the frame is checked and moved to the off-chip memory.
- the present invention provides a significant enhancement in data-checking ability by keeping the received CRC of a frame with the frame as the frame is stored in one or more dedicated non-data buffers ( 53 ) and/or a dedicated data buffer ( 55 ).
- two non-data buffers ( 53 ) are provided, and are operable to simultaneously receive on both inbound fibers ( 117 ) of a dual-port fibre-channel interface node ( 1220 ).
- one data buffer ( 55 ) is provided, and is operable to receive on either inbound fiber ( 117 ) of a dual-port fibre-channel interface node ( 1220 ).
- a single CRC checker ( 596 ) is provided (to save cost) that verifies the cyclic-redundancy code of a frame while moving the frame from the on-chip buffer ( 53 or 55 ) to the off-chip memory ( 111 ).
- a single CRC generator ( 76 for non-data frames and 87 for data frames) is provided on each transmit path (to save cost) that generates the appropriate cyclic-redundancy code of a frame while moving the frame from the on-chip buffer ( 73 or 55 ) to the fibre channel ( 1250 ).
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Computer Security & Cryptography (AREA)
- Quality & Reliability (AREA)
- Small-Scale Networks (AREA)
- Information Transfer Systems (AREA)
- Multi Processors (AREA)
- Detection And Prevention Of Errors In Transmission (AREA)
- Computer And Data Communications (AREA)
- Error Detection And Correction (AREA)
- Detection And Correction Of Errors (AREA)
- Time-Division Multiplex Systems (AREA)
Abstract
Description
Fibre Channel FC-PH | X3T11/Project 755D/Rev. 4.3 | ||
Physical & Signalling Interface | |||
Fibre Channel FC-AL | X3T11/Project 960D/Rev. 4.5 | ||
Arbitrated Loop | |||
Fibre Channel FC-AL2 | X3T11/Project 1133D/Rev. 6.3 | ||
Arbitrated Loop | |||
Fibre Channel FCP | X3T10/ Rev. 012 | ||
Protocol for SCSI | X3.2 | ||
69-199X | |||
-
- a request from
microprocessor 112 to transmit a frame with xmit port enable, - the transmitting port is in the Monitoring state,
- transfer-length count not zero,
- no request from
microprocessor 112 to pause a transfer, and - (non-data transfer, or data-write transfer with transfer-ready not yet transmitted with data threshold met, or data-read transfer with data threshold met and data-frame buffer threshold met).
- a request from
-
- data-
frame buffer 55 has data available - Buffer-to-Buffer Credit is available (R_RDY received)
- non-data transfer, or data-read transfer and transfer-length counter (in
block 609; seeFIG. 12 ) is non-zero
Conditions that cause theloop 1250 to be closed include: - no Buffer-to-Buffer Credit is available when entering the Opened state
- there are no outstanding R_DY's and no more BB_Credit is available when in Opened state
- a processor busy request is active when the port is in Opened state
- the transfer has completed
- data-read transfer operation and data are not available
- a CLS primitive is received and no more BB_Credit is available
- a microprocessor pause request is pending and the logic is between frames
- data-
-
- 1) how much receive space is available in off-
chip buffer 111 fordisc drive 100 - 2) how much receive space is available in receive-non-data-
frame buffer 53 and data-frame buffer 55 fordisc drive 100 - 3) whether, given the credit, the port with which drive 100 is opened could potentially send frames that could occupy the available space.
- 1) how much receive space is available in off-
-
- at least X-frames are available off-chip, and
- at least Y-words of data are available in the data-
frame buffer 55. In one such embodiment, the value of X (where X represents the number of frames which need to be available in the off-chip buffer 111 in order to keep theloop 1250 held open) and the value for Y (where Y represents the number of words which need to be available in the on-chip buffer in order to keep theloop 1250 held open) are each separately programmable (e.g., by firmware code via microprocessor 112). In another embodiment, theloop 1250 is held open when a predetermined amount of data (not necessarily specified as a number of frames) is available in the off-chip buffer 111 (in one such embodiment, the predetermined amount of data required to be available in the off-chip buffer 111 is programmable). In another embodiment, theloop 1250 is held open when a predetermined amount of data (not necessarily specified as a number of words) is available in the on-chip buffer 119 (in one such embodiment, the predetermined amount of data required to be available in the on-chip buffer 119 is programmable). In one embodiment, theloop 1250 is held open if the predetermined amount of data is available (at least one-half frame on-chip and at least one frame available off-chip), but the transfer of a frame will not start until an entire frame is available on-chip.
-
- (a) supporting a fibre-channel arbitrated-loop communications channel (1250) on each of a first port (116) and a second port (116) of a first channel node (1220);
- (b) receiving a frame from the communications channel (1250), the received frame including a cyclic-redundancy code that is based on other data in the received frame;
- (c) storing the received frame, including the cyclic-redundancy code, into a frame buffer (53 or 55);
- (d) moving the received frame to a memory (111) that is separate from the frame buffer (53 OR 55); and
- (e) checking the received frame for accuracy by verifying the cyclic-redundancy code (CRC) while moving the received frame to the separate memory (111).
In one embodiment, the method includes steps of: - (f) placing a frame that is to be transmitted into an on-chip frame buffer (55);
- (g) generating the cyclic-redundancy code based on data in the frame to be transmitted; and
- (h) transmitting the frame to be transmitted, including the cyclic-redundancy code, onto the communications channel (1250).
In one such embodiment of the method, the placing step (f) further includes steps of: - (f)(i) generating parity for data of the frame to be transmitted;
- (f)(ii) adding parity to the data of the frame to be transmitted; and the moving step (d) further includes a step of
- (d)(i) stripping away the cyclic-redundancy code while moving the received frame to the separate memory (111).
In another embodiment of the method, the receiving step (b) further includes a step of
- (d)(i) stripping away the cyclic-redundancy code while moving the received frame to the separate memory (111).
- (b)(i) checking the received frame for accuracy by verifying the cyclic-redundancy code while receiving the received frame from the communications channel (1250).
In one embodiment, the method further includes a step of - (i) transferring data through the fibre-channel arbitrated-loop communications channel (1250) between a magnetic-disc-storage drive (1256) that is operatively coupled to the first channel node (1220) and a computer system (1202) having a second channel node (1220), wherein the second channel node (1220) is operatively coupled to the first channel node (1220) by the communications channel (1250).
Claims (30)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US10/723,963 USRE42228E1 (en) | 1997-11-17 | 2003-11-26 | Method and apparatus for using data protection code for data integrity in on-chip memory |
Applications Claiming Priority (6)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US6592697P | 1997-11-17 | 1997-11-17 | |
US6592097P | 1997-11-17 | 1997-11-17 | |
US6591997P | 1997-11-17 | 1997-11-17 | |
US6721197P | 1997-12-01 | 1997-12-01 | |
US09/193,446 US6324669B1 (en) | 1997-11-17 | 1998-11-17 | Method and apparatus for using CRC for data integrity in on-chip memory |
US10/723,963 USRE42228E1 (en) | 1997-11-17 | 2003-11-26 | Method and apparatus for using data protection code for data integrity in on-chip memory |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US09/193,446 Reissue US6324669B1 (en) | 1997-11-17 | 1998-11-17 | Method and apparatus for using CRC for data integrity in on-chip memory |
Publications (1)
Publication Number | Publication Date |
---|---|
USRE42228E1 true USRE42228E1 (en) | 2011-03-15 |
Family
ID=27490525
Family Applications (6)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US09/193,387 Expired - Lifetime US6502189B1 (en) | 1997-11-17 | 1998-11-17 | Method and dedicated frame buffer for loop initialization and responses |
US09/193,681 Expired - Lifetime US6279057B1 (en) | 1997-11-17 | 1998-11-17 | Communications system having dedicated frame buffers located in a channel node connected to two ports of the channel node for receiving frames |
US09/193,446 Ceased US6324669B1 (en) | 1997-11-17 | 1998-11-17 | Method and apparatus for using CRC for data integrity in on-chip memory |
US09/193,482 Ceased US6317800B1 (en) | 1997-11-17 | 1998-11-17 | System for reducing arbitrated-loop overhead by maintaining control of a communications channel as long as a predetermined amount of data is available within control of channel node |
US10/714,478 Expired - Lifetime USRE40034E1 (en) | 1997-11-17 | 2003-11-13 | Method and apparatus to reduce serial communications path connection overhead |
US10/723,963 Expired - Lifetime USRE42228E1 (en) | 1997-11-17 | 2003-11-26 | Method and apparatus for using data protection code for data integrity in on-chip memory |
Family Applications Before (5)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US09/193,387 Expired - Lifetime US6502189B1 (en) | 1997-11-17 | 1998-11-17 | Method and dedicated frame buffer for loop initialization and responses |
US09/193,681 Expired - Lifetime US6279057B1 (en) | 1997-11-17 | 1998-11-17 | Communications system having dedicated frame buffers located in a channel node connected to two ports of the channel node for receiving frames |
US09/193,446 Ceased US6324669B1 (en) | 1997-11-17 | 1998-11-17 | Method and apparatus for using CRC for data integrity in on-chip memory |
US09/193,482 Ceased US6317800B1 (en) | 1997-11-17 | 1998-11-17 | System for reducing arbitrated-loop overhead by maintaining control of a communications channel as long as a predetermined amount of data is available within control of channel node |
US10/714,478 Expired - Lifetime USRE40034E1 (en) | 1997-11-17 | 2003-11-13 | Method and apparatus to reduce serial communications path connection overhead |
Country Status (7)
Country | Link |
---|---|
US (6) | US6502189B1 (en) |
JP (4) | JP2001523861A (en) |
KR (3) | KR100734649B1 (en) |
CN (3) | CN1158605C (en) |
DE (3) | DE19882830T1 (en) |
GB (3) | GB2346784B (en) |
WO (3) | WO1999026137A1 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110191641A1 (en) * | 2010-02-03 | 2011-08-04 | Fujitsu Limited | Raid device, abnormal device detecting apparatus, and abnormal device detecting method |
Families Citing this family (128)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6356560B1 (en) * | 1997-05-30 | 2002-03-12 | Adtran, Inc. | Arbitration mechanism for statistically multiplexed frame relay switching system |
EP1059589B1 (en) * | 1999-06-09 | 2005-03-30 | Texas Instruments Incorporated | Multi-channel DMA with scheduled ports |
JP2001034427A (en) * | 1999-07-23 | 2001-02-09 | Fujitsu Ltd | Device controller and its method |
US6529945B1 (en) * | 1999-07-26 | 2003-03-04 | International Business Machines Corporation | Data buffer management between two different systems |
US6574696B1 (en) * | 2000-03-07 | 2003-06-03 | Hewlett-Packard Development Company, L.P. | Methods and arrangements for providing bi-directional control between data storage drives and automation controllers |
US6571367B1 (en) * | 2000-03-15 | 2003-05-27 | Emc Corporation | Method and apparatus connecting between a fiber channel and a cache memory |
US6622183B1 (en) * | 2000-03-21 | 2003-09-16 | Lsi Logic Corporation | Data transmission buffer having frame counter feedback for re-transmitting aborted data frames |
FR2812116A1 (en) * | 2000-07-19 | 2002-01-25 | Schlumberger Systems & Service | Method and device for secured writing of data in rewritable memory, for use in microelectronics and chip cards |
US7007097B1 (en) * | 2000-07-20 | 2006-02-28 | Silicon Graphics, Inc. | Method and system for covering multiple resourcces with a single credit in a computer system |
US6483843B1 (en) * | 2000-10-12 | 2002-11-19 | Emulex Corporation | Detecting and counting open ordered sets originating from an attached node port |
US6999460B1 (en) * | 2000-10-16 | 2006-02-14 | Storage Technology Corporation | Arbitrated loop port switching |
US6647528B1 (en) * | 2000-11-02 | 2003-11-11 | Computer Network Technology Corporation | Fiber channel CRC for internal error checking on a switching platform |
US6779150B1 (en) * | 2000-12-21 | 2004-08-17 | Emc Corporation | CRC error detection system and method |
US6868516B1 (en) | 2000-12-21 | 2005-03-15 | Emc Corporation | Method for validating write data to a memory |
US6934904B2 (en) * | 2001-04-30 | 2005-08-23 | Sun Microsystems, Inc. | Data integrity error handling in a redundant storage array |
US7016969B1 (en) * | 2001-05-11 | 2006-03-21 | Cisco Technology, Inc. | System using weighted fairness decisions in spatial reuse protocol forwarding block to determine allowed usage for servicing transmit and transit traffic in a node |
JP2002368768A (en) * | 2001-06-05 | 2002-12-20 | Hitachi Ltd | Electronic device compatible with fiber channel arbitration loop and method for detecting fault in the fiber channel arbitration loop |
US6850410B2 (en) * | 2001-06-29 | 2005-02-01 | Emc Corporation | Advanced technology attachment disk drive module with fibre channel system characteristics |
US7065661B2 (en) * | 2002-12-16 | 2006-06-20 | Emc Corporation | Using request and grant signals to read revision information from an adapter board that interfaces a disk drive |
JP3523616B2 (en) * | 2001-07-24 | 2004-04-26 | 松下電器産業株式会社 | Bus optimization method and communication node |
US7215680B2 (en) * | 2001-07-26 | 2007-05-08 | Nishan Systems, Inc. | Method and apparatus for scheduling packet flow on a fibre channel arbitrated loop |
US20030135577A1 (en) * | 2001-12-19 | 2003-07-17 | Weber Bret S. | Dual porting serial ATA disk drives for fault tolerant applications |
US7433948B2 (en) | 2002-01-23 | 2008-10-07 | Cisco Technology, Inc. | Methods and apparatus for implementing virtualization of storage within a storage area network |
US7391723B2 (en) * | 2002-04-22 | 2008-06-24 | Computer Network Technology Corporation | Frame-level fibre channel CRC on switching platform |
US7808924B2 (en) | 2002-05-24 | 2010-10-05 | Cisco Technology, Inc. | Apparatus and method for preventing disruption of fibre channel fabrics caused by ReConfigure Fabric (RCF) messages |
US7630300B2 (en) * | 2002-07-02 | 2009-12-08 | Emulex Design & Manufacturing Corporation | Methods and apparatus for trunking in fibre channel arbitrated loop systems |
US7660316B2 (en) * | 2002-07-02 | 2010-02-09 | Emulex Design & Manufacturing Corporation | Methods and apparatus for device access fairness in fibre channel arbitrated loop systems |
US7664018B2 (en) | 2002-07-02 | 2010-02-16 | Emulex Design & Manufacturing Corporation | Methods and apparatus for switching fibre channel arbitrated loop devices |
US7397788B2 (en) * | 2002-07-02 | 2008-07-08 | Emulex Design & Manufacturing Corporation | Methods and apparatus for device zoning in fibre channel arbitrated loop systems |
US7171525B1 (en) * | 2002-07-31 | 2007-01-30 | Silicon Image, Inc. | Method and system for arbitrating priority bids sent over serial links to a multi-port storage device |
KR100457046B1 (en) * | 2002-08-07 | 2004-11-10 | 삼성전자주식회사 | Method for forming a contact in semiconductor device process |
DE60321927D1 (en) * | 2002-09-10 | 2008-08-14 | Exagrid Systems Inc | DATA PROTECTION AND DEVICE |
US7397768B1 (en) | 2002-09-11 | 2008-07-08 | Qlogic, Corporation | Zone management in a multi-module fibre channel switch |
US7220908B2 (en) * | 2002-09-12 | 2007-05-22 | Yamaha Corporation | Waveform processing apparatus with versatile data bus |
US6961786B2 (en) * | 2002-09-25 | 2005-11-01 | Lsi Logic Corporation | Fiber loop linking/detecting |
US20040123027A1 (en) * | 2002-10-03 | 2004-06-24 | Workman Michael Lee | Systems and methods of multiple access paths to single ported storage devices |
US20040068591A1 (en) * | 2002-10-03 | 2004-04-08 | Workman Michael Lee | Systems and methods of multiple access paths to single ported storage devices |
US7327680B1 (en) * | 2002-11-05 | 2008-02-05 | Cisco Technology, Inc. | Methods and apparatus for network congestion control |
US6829658B2 (en) | 2002-12-16 | 2004-12-07 | Emc Corporation | Compatible signal-to-pin connector assignments for usage with fibre channel and advanced technology attachment disk drives |
US7194673B2 (en) * | 2002-12-20 | 2007-03-20 | Emc Corporation | Detecting intermittent losses of synchronization in a fibre channel loop |
US7782784B2 (en) * | 2003-01-10 | 2010-08-24 | Cisco Technology, Inc. | Port analyzer adapter |
US7899048B1 (en) | 2003-01-15 | 2011-03-01 | Cisco Technology, Inc. | Method and apparatus for remotely monitoring network traffic through a generic network |
US7516272B2 (en) | 2003-03-28 | 2009-04-07 | Emc Corporation | Midplane-independent implementations of data storage system enclosures |
US6961784B1 (en) * | 2003-04-17 | 2005-11-01 | Qlogic Corporation | Method and system for processing non-data frames in host bus adapters |
US7646767B2 (en) | 2003-07-21 | 2010-01-12 | Qlogic, Corporation | Method and system for programmable data dependant network routing |
JP4432388B2 (en) | 2003-08-12 | 2010-03-17 | 株式会社日立製作所 | I / O controller |
US7234101B1 (en) | 2003-08-27 | 2007-06-19 | Qlogic, Corporation | Method and system for providing data integrity in storage systems |
US8165136B1 (en) | 2003-09-03 | 2012-04-24 | Cisco Technology, Inc. | Virtual port based SPAN |
US7474666B2 (en) | 2003-09-03 | 2009-01-06 | Cisco Technology, Inc. | Switch port analyzers |
US7134070B2 (en) * | 2003-09-08 | 2006-11-07 | Intel Corporation | Checksum determination |
US7573870B2 (en) * | 2003-09-25 | 2009-08-11 | Lsi Logic Corporation | Transmit prioritizer context prioritization scheme |
US7353448B1 (en) * | 2003-10-21 | 2008-04-01 | Marvell Semiconductor Israel Ltd. | Methods, architectures, circuits and systems for transmission error determination |
US7219263B1 (en) | 2003-10-29 | 2007-05-15 | Qlogic, Corporation | Method and system for minimizing memory corruption |
US20050114498A1 (en) * | 2003-11-06 | 2005-05-26 | International Business Machines Corporation | Method and apparatus for managing data transfer in a data processing system |
US7219325B1 (en) * | 2003-11-21 | 2007-05-15 | Xilinx, Inc. | Exploiting unused configuration memory cells |
JP4426261B2 (en) * | 2003-11-25 | 2010-03-03 | 株式会社日立製作所 | Channel adapter and disk array device |
US7934023B2 (en) | 2003-12-01 | 2011-04-26 | Cisco Technology, Inc. | Apparatus and method for performing fast fibre channel write operations over relatively high latency networks |
US7145776B2 (en) * | 2003-12-22 | 2006-12-05 | Emc Corporation | Midplane-less data storage enclosure |
US7236361B2 (en) * | 2003-12-22 | 2007-06-26 | Emc Corporation | Fan assembly for installing and removing fans individually and collectively |
US7231581B2 (en) * | 2003-12-31 | 2007-06-12 | Intel Corporation | Communicating using a partial block in a frame |
US20050187979A1 (en) * | 2004-02-09 | 2005-08-25 | Microsoft Corporation | System and method for message-level connection management |
US7986630B1 (en) * | 2004-02-09 | 2011-07-26 | Lsi Corporation | High performance architecture for fiber channel targets and target bridges |
US7360142B1 (en) | 2004-03-03 | 2008-04-15 | Marvell Semiconductor Israel Ltd. | Methods, architectures, circuits, software and systems for CRC determination |
US7434150B1 (en) | 2004-03-03 | 2008-10-07 | Marvell Israel (M.I.S.L.) Ltd. | Methods, circuits, architectures, software and systems for determining a data transmission error and/or checking or confirming such error determinations |
US7930377B2 (en) | 2004-04-23 | 2011-04-19 | Qlogic, Corporation | Method and system for using boot servers in networks |
US7669190B2 (en) | 2004-05-18 | 2010-02-23 | Qlogic, Corporation | Method and system for efficiently recording processor events in host bus adapters |
US20060015659A1 (en) * | 2004-07-19 | 2006-01-19 | Krantz Leon A | System and method for transferring data using storage controllers |
US9201599B2 (en) | 2004-07-19 | 2015-12-01 | Marvell International Ltd. | System and method for transmitting data in storage controllers |
US7719970B1 (en) * | 2004-08-20 | 2010-05-18 | Altera Corporation | Serial communications system with optional data path and control plane features |
US7577772B2 (en) | 2004-09-08 | 2009-08-18 | Qlogic, Corporation | Method and system for optimizing DMA channel selection |
US20060064531A1 (en) * | 2004-09-23 | 2006-03-23 | Alston Jerald K | Method and system for optimizing data transfer in networks |
US7676611B2 (en) | 2004-10-01 | 2010-03-09 | Qlogic, Corporation | Method and system for processing out of orders frames |
US7398335B2 (en) * | 2004-11-22 | 2008-07-08 | Qlogic, Corporation | Method and system for DMA optimization in host bus adapters |
US7164425B2 (en) * | 2004-12-21 | 2007-01-16 | Qlogic Corporation | Method and system for high speed network application |
US7392437B2 (en) | 2005-01-20 | 2008-06-24 | Qlogic, Corporation | Method and system for testing host bus adapters |
US7281077B2 (en) * | 2005-04-06 | 2007-10-09 | Qlogic, Corporation | Elastic buffer module for PCI express devices |
US7231480B2 (en) * | 2005-04-06 | 2007-06-12 | Qlogic, Corporation | Method and system for receiver detection in PCI-Express devices |
US7562284B2 (en) * | 2005-08-26 | 2009-07-14 | International Business Machines Corporation | Apparatus, system, and method for mandatory end to end integrity checking in a storage system |
KR100690274B1 (en) * | 2005-09-12 | 2007-03-09 | 삼성전자주식회사 | Cyclic redundancy check device and communication system comprising the same for multi-channel serial communication |
US7660926B2 (en) * | 2005-11-16 | 2010-02-09 | Sun Microsystems, Inc. | Apparatus and method for a core for implementing a communications port |
US7694204B2 (en) * | 2006-03-09 | 2010-04-06 | Silicon Image, Inc. | Error detection in physical interfaces for point-to-point communications between integrated circuits |
US7461195B1 (en) | 2006-03-17 | 2008-12-02 | Qlogic, Corporation | Method and system for dynamically adjusting data transfer rates in PCI-express devices |
US7933735B2 (en) * | 2007-01-31 | 2011-04-26 | Denso Corporation | Semiconductor integrated circuit |
KR100902051B1 (en) * | 2007-07-12 | 2009-06-15 | 주식회사 하이닉스반도체 | Apparatus and method for generating error detection code |
US7751350B1 (en) * | 2007-08-03 | 2010-07-06 | William George Pabst | Full duplex network radio bridge with low latency and high throughput |
US7975027B2 (en) * | 2007-08-06 | 2011-07-05 | International Business Machines Corporation | Credit depletion notification for transmitting frames between a port pair |
US7787375B2 (en) * | 2007-08-06 | 2010-08-31 | International Business Machines Corporation | Performing a recovery action in response to a credit depletion notification |
US8111610B2 (en) * | 2008-01-25 | 2012-02-07 | Emulex Design & Manufacturing Corporation | Flagging of port conditions in high speed networks |
US8566688B2 (en) * | 2009-09-01 | 2013-10-22 | Ensequence, Inc. | Method of certifying multiple versions of an application |
KR101069725B1 (en) * | 2009-12-24 | 2011-10-04 | 주식회사 하이닉스반도체 | Semiconductor Memory Apparatus, Control Circuit and Method for Fault Address Therefor |
JP5678171B2 (en) | 2010-04-12 | 2015-02-25 | クゥアルコム・インコーポレイテッドQualcomm Incorporated | Channel estimation for low overhead communication in networks |
JP4922442B2 (en) | 2010-07-29 | 2012-04-25 | 株式会社東芝 | Buffer management device, storage device including the same, and buffer management method |
US8504893B1 (en) * | 2010-09-30 | 2013-08-06 | Micron Technology, Inc. | Error detection or correction of a portion of a codeword in a memory device |
GB2484717B (en) * | 2010-10-21 | 2018-06-13 | Advanced Risc Mach Ltd | Security provision for a subject image displayed in a non-secure domain |
CN102468915B (en) * | 2010-11-16 | 2014-07-02 | 北京中电华大电子设计有限责任公司 | Serial port 802.11n wireless network card chip sending path realization method |
CN102158400B (en) * | 2011-03-03 | 2013-12-04 | 北京航空航天大学 | Communication interface of space-based route switching system and space-based route switching system |
CN102999458A (en) * | 2011-09-09 | 2013-03-27 | 中国航天科工集团第三研究院第八三五七研究所 | High-speed intelligent serial port chip |
CN102567276B (en) * | 2011-12-19 | 2014-03-12 | 华为技术有限公司 | Data transmission method based on multiple channels, related nodes and system |
JP5937485B2 (en) * | 2012-10-24 | 2016-06-22 | 富士通フロンテック株式会社 | Transmitter |
US9395924B2 (en) * | 2013-01-22 | 2016-07-19 | Seagate Technology Llc | Management of and region selection for writes to non-volatile memory |
US9569618B2 (en) * | 2013-08-28 | 2017-02-14 | Korea University Research And Business Foundation | Server and method for attesting application in smart device using random executable code |
CN103593261B (en) * | 2013-10-17 | 2015-09-23 | 北京控制工程研究所 | The quick recovery method of communication failure in a kind of interrupt service routine |
US9467389B2 (en) | 2014-04-28 | 2016-10-11 | International Business Machines Corporation | Handling large frames in a virtualized fibre channel over ethernet (FCoE) data forwarder |
US9892065B2 (en) | 2015-05-20 | 2018-02-13 | International Business Machines Corporation | Adjustments of buffer credits for optimizing the number of retry operations and transfer ready operations |
US9864716B2 (en) * | 2015-05-20 | 2018-01-09 | International Business Machines Corporation | Receiving buffer credits by a plurality of channels of one or more host computational devices for transmitting data to a control unit |
US10061734B2 (en) | 2015-05-20 | 2018-08-28 | International Business Machines Corporation | Adjustment of buffer credits and other parameters in a startup phase of communications between a plurality of channels and a control unit |
CN106500739B (en) * | 2016-11-18 | 2019-01-25 | 威科达(东莞)智能控制有限公司 | A method of it is communicated with absolute value encoder |
CN108572883B (en) * | 2017-04-19 | 2021-06-18 | 北京金山云网络技术有限公司 | Data correctness checking method and device |
EP3404546B1 (en) | 2017-05-16 | 2019-09-11 | Melexis Technologies NV | Device for supervising and initializing ports |
CN107229858B (en) * | 2017-05-19 | 2020-04-03 | 郑州云海信息技术有限公司 | PECI bus master-slave machine verification system and method supporting crc8 verification |
US11868804B1 (en) | 2019-11-18 | 2024-01-09 | Groq, Inc. | Processor instruction dispatch configuration |
US11360934B1 (en) | 2017-09-15 | 2022-06-14 | Groq, Inc. | Tensor streaming processor architecture |
US11114138B2 (en) | 2017-09-15 | 2021-09-07 | Groq, Inc. | Data structures with multiple read ports |
US11243880B1 (en) | 2017-09-15 | 2022-02-08 | Groq, Inc. | Processor architecture |
US11170307B1 (en) | 2017-09-21 | 2021-11-09 | Groq, Inc. | Predictive model compiler for generating a statically scheduled binary with known resource constraints |
CN107977284A (en) * | 2017-11-30 | 2018-05-01 | 苏州麦迪斯顿医疗科技股份有限公司 | A kind of data processing method, device, server and medium |
JP7192244B2 (en) | 2018-05-15 | 2022-12-20 | 株式会社デンソー | repeater |
US11301546B2 (en) | 2018-11-19 | 2022-04-12 | Groq, Inc. | Spatial locality transform of matrices |
US11115147B2 (en) * | 2019-01-09 | 2021-09-07 | Groq, Inc. | Multichip fault management |
CN110518978B (en) * | 2019-07-23 | 2022-12-27 | 中国航空无线电电子研究所 | Hybrid data transmission assembly based on optical fiber link |
CN113037428B (en) * | 2019-12-09 | 2022-01-25 | 大唐移动通信设备有限公司 | Data flow processing method and device |
CN111966627B (en) * | 2020-08-13 | 2023-06-30 | 北京中航通用科技有限公司 | 8B/9B encoding and decoding method for serial bus |
TWI778644B (en) * | 2021-05-31 | 2022-09-21 | 新唐科技股份有限公司 | Soc and data burning method thereof |
CN114578299A (en) * | 2021-06-10 | 2022-06-03 | 中国人民解放军63698部队 | Method and system for generating radio frequency signal by wireless remote control beacon device |
US11855655B2 (en) * | 2021-09-28 | 2023-12-26 | Texas Instruments Incorporated | Serial communications module with CRC |
US12045121B2 (en) | 2021-11-12 | 2024-07-23 | Samsung Electronics Co., Ltd. | Security IC and operating method thereof |
CN114117980B (en) * | 2022-01-25 | 2022-06-07 | 飞腾信息技术有限公司 | Chip design method, chip design module, electronic device, and medium |
US20230236756A1 (en) * | 2022-01-25 | 2023-07-27 | Infineon Technologies Ag | Buffer management in an ethernet switch |
Citations (27)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4654654A (en) * | 1983-02-07 | 1987-03-31 | At&T Bell Laboratories | Data network acknowledgement arrangement |
US4819229A (en) * | 1986-03-13 | 1989-04-04 | University Of Strathclyde | Local area network priority control system |
US5168568A (en) * | 1989-02-06 | 1992-12-01 | Compaq Computer Corporation | Delaying arbitration of bus access in digital computers |
US5170466A (en) * | 1989-10-10 | 1992-12-08 | Unisys Corporation | Storage/retrieval system for document |
US5187780A (en) * | 1989-04-07 | 1993-02-16 | Digital Equipment Corporation | Dual-path computer interconnect system with zone manager for packet memory |
JPH07200445A (en) | 1993-11-15 | 1995-08-04 | Internatl Business Mach Corp <Ibm> | Access control method via dynamic communication channel and communication interface system |
JPH0897803A (en) | 1994-09-26 | 1996-04-12 | Nec Eng Ltd | Data transmission system |
US5598541A (en) * | 1994-10-24 | 1997-01-28 | Lsi Logic Corporation | Node loop port communication interface super core for fibre channel |
US5617541A (en) * | 1994-12-21 | 1997-04-01 | International Computer Science Institute | System for packetizing data encoded corresponding to priority levels where reconstructed data corresponds to fractionalized priority level and received fractionalized packets |
US5619497A (en) * | 1994-12-22 | 1997-04-08 | Emc Corporation | Method and apparatus for reordering frames |
US5619647A (en) * | 1994-09-30 | 1997-04-08 | Tandem Computers, Incorporated | System for multiplexing prioritized virtual channels onto physical channels where higher priority virtual will pre-empt a lower priority virtual or a lower priority will wait |
JPH09149067A (en) | 1995-11-21 | 1997-06-06 | Hitachi Cable Ltd | Switching hub |
US5638518A (en) * | 1994-10-24 | 1997-06-10 | Lsi Logic Corporation | Node loop core for implementing transmission protocol in fibre channel |
JPH09238163A (en) | 1995-09-29 | 1997-09-09 | Hewlett Packard Co <Hp> | Buffer memory management device |
US5668809A (en) * | 1993-10-20 | 1997-09-16 | Lsi Logic Corporation | Single chip network hub with dynamic window filter |
US5734918A (en) * | 1994-07-26 | 1998-03-31 | Hitachi, Ltd. | Computer system with an input/output processor which enables direct file transfers between a storage medium and a network |
US5819111A (en) * | 1996-03-15 | 1998-10-06 | Adobe Systems, Inc. | System for managing transfer of data by delaying flow controlling of data through the interface controller until the run length encoded data transfer is complete |
US5818842A (en) * | 1994-01-21 | 1998-10-06 | Newbridge Networks Corporation | Transparent interconnector of LANs by an ATM network |
US5889949A (en) * | 1996-10-11 | 1999-03-30 | C-Cube Microsystems | Processing system with memory arbitrating between memory access requests in a set top box |
US5901280A (en) * | 1996-09-12 | 1999-05-04 | Mitsubishi Denki Kabushiki Kaisha | Transmission monitoring and controlling apparatus and a transmission monitoring and controlling method |
US5917828A (en) * | 1997-01-03 | 1999-06-29 | Ncr Corporation | ATM reassembly controller and method |
US6012128A (en) * | 1994-09-16 | 2000-01-04 | Philips Electronics North America Corporation | Microcontroller having a page address mode |
US6014380A (en) * | 1997-06-30 | 2000-01-11 | Sun Microsystems, Inc. | Mechanism for packet field replacement in a multi-layer distributed network element |
US6061351A (en) * | 1997-02-14 | 2000-05-09 | Advanced Micro Devices, Inc. | Multicopy queue structure with searchable cache area |
US6172990B1 (en) * | 1997-06-19 | 2001-01-09 | Xaqti Corporation | Media access control micro-RISC stream processor and method for implementing the same |
US6185207B1 (en) * | 1997-06-19 | 2001-02-06 | International Business Machines Corporation | Communication system having a local area network adapter for selectively deleting information and method therefor |
US6247060B1 (en) * | 1997-10-14 | 2001-06-12 | Alacritech, Inc. | Passing a communication control block from host to a local device such that a message is processed on the device |
Family Cites Families (57)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4486739A (en) | 1982-06-30 | 1984-12-04 | International Business Machines Corporation | Byte oriented DC balanced (0,4) 8B/10B partitioned block transmission code |
US4933785A (en) | 1988-03-01 | 1990-06-12 | Prairietek Corporation | Disk drive apparatus using dynamic loading/unloading |
US5034837A (en) | 1989-11-09 | 1991-07-23 | Rodime Plc | Magnetic disk drive incorporating a magnetic actuator lock and a very small form factor |
US5235482A (en) | 1989-11-09 | 1993-08-10 | Rodime Plc | Magnetic disk drive incorporating a mechanically damped base |
US5418716A (en) | 1990-07-26 | 1995-05-23 | Nec Corporation | System for recognizing sentence patterns and a system for recognizing sentence patterns and grammatical cases |
US5172538A (en) | 1991-09-05 | 1992-12-22 | Luger Linda S | Stirrup pad |
JP2624064B2 (en) | 1991-11-29 | 1997-06-25 | 日本電気株式会社 | Load / unload mechanism |
JPH05166322A (en) | 1991-12-16 | 1993-07-02 | Nec Ibaraki Ltd | Loading/unloading mechanism for magnetic head |
JPH05244160A (en) * | 1992-02-28 | 1993-09-21 | Toshiba Corp | Bridge device having frame deleting function |
US5260933A (en) | 1992-05-15 | 1993-11-09 | International Business Machines Corporation | Acknowledgement protocol for serial data network with out-of-order delivery |
US5772597A (en) | 1992-09-14 | 1998-06-30 | Sextant Medical Corporation | Surgical tool end effector |
US5341260A (en) | 1992-12-04 | 1994-08-23 | Seagate Technology, Inc. | Reduced torque unloading ramp system for a hard disk drive |
DE69434311D1 (en) | 1993-02-01 | 2005-04-28 | Sun Microsystems Inc | ARCHIVING FILES SYSTEM FOR DATA PROVIDERS IN A DISTRIBUTED NETWORK ENVIRONMENT |
US5617425A (en) | 1993-05-26 | 1997-04-01 | Seagate Technology, Inc. | Disc array having array supporting controllers and interface |
JP3264465B2 (en) | 1993-06-30 | 2002-03-11 | 株式会社日立製作所 | Storage system |
JPH0728746A (en) * | 1993-07-14 | 1995-01-31 | Fuji Xerox Co Ltd | Data transfer device |
JPH0863920A (en) | 1993-08-24 | 1996-03-08 | Sony Corp | Loading/unloading device and its drive controller |
FR2719552B1 (en) * | 1994-05-04 | 1996-07-26 | Eurocopter France | Rectifier blade for anti-torque device with rotor and stator rectifier faired from helicopter. |
US5455723A (en) | 1994-06-02 | 1995-10-03 | International Business Machines Corporation | Method and apparatus for ramp load and unload |
US5828522A (en) | 1994-09-13 | 1998-10-27 | International Business Machines Corporation | Velocity control of head load/unload mechanism in a disk drive using dither |
US5535035A (en) * | 1994-09-15 | 1996-07-09 | International Business Machines Corporation | Optical fiber ring communications system and communications method |
US5590006A (en) | 1994-12-23 | 1996-12-31 | International Business Machines Corporation | One-sided, single platter hard disk with center parking features |
US5812564A (en) | 1995-05-08 | 1998-09-22 | Western Digital Corporation | Disk drive with embedded finite field processor for error correction |
US5831795A (en) | 1995-05-10 | 1998-11-03 | Iomega Corporation | Head loading mechanism for a disk drive |
US5694615A (en) * | 1995-06-26 | 1997-12-02 | Hewlett Packard Company | Storage system having storage units interconnected to form multiple loops to provide simultaneous access from multiple hosts |
US5815662A (en) | 1995-08-15 | 1998-09-29 | Ong; Lance | Predictive memory caching for media-on-demand systems |
JPH0955718A (en) * | 1995-08-17 | 1997-02-25 | Nec Corp | Data communication device |
US5768623A (en) | 1995-09-19 | 1998-06-16 | International Business Machines Corporation | System and method for sharing multiple storage arrays by dedicating adapters as primary controller and secondary controller for arrays reside in different host computers |
KR980700639A (en) | 1995-10-10 | 1998-03-30 | 제이. 지. 에이. 롤페스 | Information storage unit with electromagnetic lifting device |
US5778426A (en) | 1995-10-23 | 1998-07-07 | Symbios, Inc. | Methods and structure to maintain a two level cache in a RAID controller and thereby selecting a preferred posting method |
US5864448A (en) | 1995-10-24 | 1999-01-26 | International Business Machines Corporation | Slider and suspension retention and retraction system for high capacity disk drive |
US5822782A (en) | 1995-10-27 | 1998-10-13 | Symbios, Inc. | Methods and structure to maintain raid configuration information on disks of the array |
US5805920A (en) | 1995-11-13 | 1998-09-08 | Tandem Computers Incorporated | Direct bulk data transfers |
JPH09149034A (en) * | 1995-11-20 | 1997-06-06 | Nec Eng Ltd | Atm cell generating method and communication controller |
US5758081A (en) | 1995-12-08 | 1998-05-26 | Aytac; Haluk M. | Computing and communications transmitting, receiving system, with a push button interface, that is continously on, that pairs up with a personal computer and carries out mainly communications related routine tasks |
US5781801A (en) | 1995-12-20 | 1998-07-14 | Emc Corporation | Method and apparatus for receive buffer management in multi-sender communication systems |
US5809328A (en) | 1995-12-21 | 1998-09-15 | Unisys Corp. | Apparatus for fibre channel transmission having interface logic, buffer memory, multiplexor/control device, fibre channel controller, gigabit link module, microprocessor, and bus control device |
US5768530A (en) | 1995-12-28 | 1998-06-16 | Emc Corporation | High speed integrated circuit interface for fibre channel communications |
US5790773A (en) | 1995-12-29 | 1998-08-04 | Symbios, Inc. | Method and apparatus for generating snapshot copies for data backup in a raid subsystem |
US5787242A (en) | 1995-12-29 | 1998-07-28 | Symbios Logic Inc. | Method and apparatus for treatment of deferred write data for a dead raid device |
JP2689973B2 (en) * | 1995-12-29 | 1997-12-10 | 日本電気株式会社 | Switching hub device |
US6115767A (en) * | 1996-03-04 | 2000-09-05 | Matsushita Electric Industrial Co., Ltd. | Apparatus and method of partially transferring data through bus and bus master control device |
US5802080A (en) | 1996-03-28 | 1998-09-01 | Seagate Technology, Inc. | CRC checking using a CRC generator in a multi-port design |
US5663724A (en) | 1996-03-28 | 1997-09-02 | Seagate Technology, Inc. | 16B/20B encoder |
US5761705A (en) | 1996-04-04 | 1998-06-02 | Symbios, Inc. | Methods and structure for maintaining cache consistency in a RAID controller having redundant caches |
US5805788A (en) | 1996-05-20 | 1998-09-08 | Cray Research, Inc. | Raid-5 parity generation and data reconstruction |
US5761534A (en) | 1996-05-20 | 1998-06-02 | Cray Research, Inc. | System for arbitrating packetized data from the network to the peripheral resources and prioritizing the dispatching of packets onto the network |
US5787450A (en) | 1996-05-29 | 1998-07-28 | International Business Machines Corporation | Apparatus and method for constructing a non-linear data object from a common gateway interface |
US5764931A (en) | 1996-05-31 | 1998-06-09 | Sun Microsystems, Inc. | Method and apparatus for passing bus mastership between processors using predefined bus mastership states |
US5822143A (en) | 1996-06-11 | 1998-10-13 | Western Digital Corporation | Decision feedback equalization implementation of partial-response signaling in a magnetic recording channel |
US5790792A (en) | 1996-09-04 | 1998-08-04 | Radiant Systems, Inc. | Method and apparatus for transmitting multimedia data from and application logic server to interactive multimedia workstations |
US5812754A (en) | 1996-09-18 | 1998-09-22 | Silicon Graphics, Inc. | Raid system with fibre channel arbitrated loop |
US5930483A (en) * | 1996-12-09 | 1999-07-27 | International Business Machines Corporation | Method and apparatus for communications control on a small computer system interface |
US5978379A (en) * | 1997-01-23 | 1999-11-02 | Gadzoox Networks, Inc. | Fiber channel learning bridge, learning half bridge, and protocol |
US5875074A (en) | 1997-04-18 | 1999-02-23 | Quinta Corporation | Adjustable head loading apparatus |
JPH10302421A (en) | 1997-04-28 | 1998-11-13 | Internatl Business Mach Corp <Ibm> | Suspension assembly and ramp block and head holding mechanism and disk drive |
US6304910B1 (en) * | 1997-09-24 | 2001-10-16 | Emulex Corporation | Communication processor having buffer list modifier control bits |
-
1998
- 1998-11-17 CN CNB988112086A patent/CN1158605C/en not_active Expired - Lifetime
- 1998-11-17 JP JP2000521448A patent/JP2001523861A/en active Pending
- 1998-11-17 KR KR1020007005386A patent/KR100734649B1/en not_active IP Right Cessation
- 1998-11-17 US US09/193,387 patent/US6502189B1/en not_active Expired - Lifetime
- 1998-11-17 WO PCT/US1998/024571 patent/WO1999026137A1/en active IP Right Grant
- 1998-11-17 WO PCT/US1998/024536 patent/WO1999026151A1/en active IP Right Grant
- 1998-11-17 JP JP2000521449A patent/JP2001523862A/en active Pending
- 1998-11-17 DE DE19882830T patent/DE19882830T1/en not_active Withdrawn
- 1998-11-17 GB GB0009935A patent/GB2346784B/en not_active Expired - Fee Related
- 1998-11-17 DE DE19882831T patent/DE19882831T1/en not_active Withdrawn
- 1998-11-17 US US09/193,681 patent/US6279057B1/en not_active Expired - Lifetime
- 1998-11-17 US US09/193,446 patent/US6324669B1/en not_active Ceased
- 1998-11-17 DE DE19882822T patent/DE19882822T1/en not_active Withdrawn
- 1998-11-17 WO PCT/US1998/024558 patent/WO1999026152A2/en active IP Right Grant
- 1998-11-17 CN CN98811207.8A patent/CN1123832C/en not_active Expired - Lifetime
- 1998-11-17 GB GB9928809A patent/GB2341526B/en not_active Expired - Fee Related
- 1998-11-17 KR KR1020007005385A patent/KR100650818B1/en active IP Right Grant
- 1998-11-17 CN CN98811240.XA patent/CN1304520A/en active Pending
- 1998-11-17 US US09/193,482 patent/US6317800B1/en not_active Ceased
- 1998-11-17 KR KR1020007005387A patent/KR100564665B1/en not_active IP Right Cessation
- 1998-11-17 JP JP2000521437A patent/JP4272813B2/en not_active Expired - Fee Related
- 1998-11-17 GB GB9928812A patent/GB2342021B/en not_active Expired - Fee Related
-
2003
- 2003-11-13 US US10/714,478 patent/USRE40034E1/en not_active Expired - Lifetime
- 2003-11-26 US US10/723,963 patent/USRE42228E1/en not_active Expired - Lifetime
-
2008
- 2008-12-26 JP JP2008333587A patent/JP4547452B2/en not_active Expired - Fee Related
Patent Citations (27)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4654654A (en) * | 1983-02-07 | 1987-03-31 | At&T Bell Laboratories | Data network acknowledgement arrangement |
US4819229A (en) * | 1986-03-13 | 1989-04-04 | University Of Strathclyde | Local area network priority control system |
US5168568A (en) * | 1989-02-06 | 1992-12-01 | Compaq Computer Corporation | Delaying arbitration of bus access in digital computers |
US5187780A (en) * | 1989-04-07 | 1993-02-16 | Digital Equipment Corporation | Dual-path computer interconnect system with zone manager for packet memory |
US5170466A (en) * | 1989-10-10 | 1992-12-08 | Unisys Corporation | Storage/retrieval system for document |
US5668809A (en) * | 1993-10-20 | 1997-09-16 | Lsi Logic Corporation | Single chip network hub with dynamic window filter |
JPH07200445A (en) | 1993-11-15 | 1995-08-04 | Internatl Business Mach Corp <Ibm> | Access control method via dynamic communication channel and communication interface system |
US5818842A (en) * | 1994-01-21 | 1998-10-06 | Newbridge Networks Corporation | Transparent interconnector of LANs by an ATM network |
US5734918A (en) * | 1994-07-26 | 1998-03-31 | Hitachi, Ltd. | Computer system with an input/output processor which enables direct file transfers between a storage medium and a network |
US6012128A (en) * | 1994-09-16 | 2000-01-04 | Philips Electronics North America Corporation | Microcontroller having a page address mode |
JPH0897803A (en) | 1994-09-26 | 1996-04-12 | Nec Eng Ltd | Data transmission system |
US5619647A (en) * | 1994-09-30 | 1997-04-08 | Tandem Computers, Incorporated | System for multiplexing prioritized virtual channels onto physical channels where higher priority virtual will pre-empt a lower priority virtual or a lower priority will wait |
US5638518A (en) * | 1994-10-24 | 1997-06-10 | Lsi Logic Corporation | Node loop core for implementing transmission protocol in fibre channel |
US5598541A (en) * | 1994-10-24 | 1997-01-28 | Lsi Logic Corporation | Node loop port communication interface super core for fibre channel |
US5617541A (en) * | 1994-12-21 | 1997-04-01 | International Computer Science Institute | System for packetizing data encoded corresponding to priority levels where reconstructed data corresponds to fractionalized priority level and received fractionalized packets |
US5619497A (en) * | 1994-12-22 | 1997-04-08 | Emc Corporation | Method and apparatus for reordering frames |
JPH09238163A (en) | 1995-09-29 | 1997-09-09 | Hewlett Packard Co <Hp> | Buffer memory management device |
JPH09149067A (en) | 1995-11-21 | 1997-06-06 | Hitachi Cable Ltd | Switching hub |
US5819111A (en) * | 1996-03-15 | 1998-10-06 | Adobe Systems, Inc. | System for managing transfer of data by delaying flow controlling of data through the interface controller until the run length encoded data transfer is complete |
US5901280A (en) * | 1996-09-12 | 1999-05-04 | Mitsubishi Denki Kabushiki Kaisha | Transmission monitoring and controlling apparatus and a transmission monitoring and controlling method |
US5889949A (en) * | 1996-10-11 | 1999-03-30 | C-Cube Microsystems | Processing system with memory arbitrating between memory access requests in a set top box |
US5917828A (en) * | 1997-01-03 | 1999-06-29 | Ncr Corporation | ATM reassembly controller and method |
US6061351A (en) * | 1997-02-14 | 2000-05-09 | Advanced Micro Devices, Inc. | Multicopy queue structure with searchable cache area |
US6172990B1 (en) * | 1997-06-19 | 2001-01-09 | Xaqti Corporation | Media access control micro-RISC stream processor and method for implementing the same |
US6185207B1 (en) * | 1997-06-19 | 2001-02-06 | International Business Machines Corporation | Communication system having a local area network adapter for selectively deleting information and method therefor |
US6014380A (en) * | 1997-06-30 | 2000-01-11 | Sun Microsystems, Inc. | Mechanism for packet field replacement in a multi-layer distributed network element |
US6247060B1 (en) * | 1997-10-14 | 2001-06-12 | Alacritech, Inc. | Passing a communication control block from host to a local device such that a message is processed on the device |
Non-Patent Citations (6)
Title |
---|
"Fibre Channel, Arbitrated Loop (FC-AL), Rev. 4.5", American National Standard for Information Technology draft proposed, ANSI X3.272-199X, (Jun. 1, 1995). * |
"Fibre Channel, Arbitrated Loop (FC-AL-2), Rev. 6.3", American National Standard for Information Technology draft proposed, ANSI X3.XXX-199X, (May 29, 1998). * |
"Fibre Channel, Physical and Signaling Interface (FC-PH), Rev. 4.3", American National Standard for Information System working draft, ANSi X3/230-199 X, (Jun. 1, 1994). * |
"Information Systems-dpANS Fibre Channel Protocol for SCSI", American National Standard-draft proposed, X3.269-199X revision 12, (Dec. 4, 1995). * |
Georgiou, C.J., et al., "Scalable Protocol Engine for High-Bandwidth Communications"; IEEE, pp. 1121-1126 (1997). * |
Japanese Patent Application No. 200-521437 Decision of Rejection dated Sep. 2, 2008. |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110191641A1 (en) * | 2010-02-03 | 2011-08-04 | Fujitsu Limited | Raid device, abnormal device detecting apparatus, and abnormal device detecting method |
US8607092B2 (en) * | 2010-02-03 | 2013-12-10 | Fujitsu Limited | Raid device, abnormal device detecting apparatus, and abnormal device detecting method |
Also Published As
Similar Documents
Publication | Publication Date | Title |
---|---|---|
USRE42228E1 (en) | Method and apparatus for using data protection code for data integrity in on-chip memory | |
US6553036B1 (en) | Method and apparatus for preserving loop fairness with dynamic half-duplex | |
US6185631B1 (en) | Program for transferring execution of certain channel functions to a control unit and having means for combining certain commands and data packets in one sequence | |
US6697914B1 (en) | Switched node comprising a disk controller with integrated multi-port switching circuitry | |
US7233977B2 (en) | Messaging mechanism employing mailboxes for inter processor communications | |
US6317805B1 (en) | Data transfer interface having protocol conversion device and upper, lower, middle machines: with middle machine arbitrating among lower machine side requesters including selective assembly/disassembly requests | |
US7984252B2 (en) | Storage controllers with dynamic WWN storage modules and methods for managing data and connections between a host and a storage device | |
US6240446B1 (en) | Multiplexing of multiple data packets for multiple input/output operations between multiple input/output devices and a channel subsystem having multiple channels | |
US7359397B2 (en) | Prioritizing transfers across an interface | |
US20160021031A1 (en) | Global shared memory switch | |
US8060696B2 (en) | Positron emission tomography event stream buffering | |
US20070076685A1 (en) | Programmable routing for frame-packet based frame processing | |
US20100202475A1 (en) | Storage device configured to transmit data via fibre channel loop | |
US9201599B2 (en) | System and method for transmitting data in storage controllers | |
US7225274B2 (en) | Method and apparatus for transferring data across a protocol bridge | |
US20060064531A1 (en) | Method and system for optimizing data transfer in networks |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: WELLS FARGO BANK, NATIONAL ASSOCIATION, AS COLLATE Free format text: SECURITY AGREEMENT;ASSIGNORS:MAXTOR CORPORATION;SEAGATE TECHNOLOGY LLC;SEAGATE TECHNOLOGY INTERNATIONAL;REEL/FRAME:022757/0017 Effective date: 20090507 Owner name: JPMORGAN CHASE BANK, N.A., AS ADMINISTRATIVE AGENT Free format text: SECURITY AGREEMENT;ASSIGNORS:MAXTOR CORPORATION;SEAGATE TECHNOLOGY LLC;SEAGATE TECHNOLOGY INTERNATIONAL;REEL/FRAME:022757/0017 Effective date: 20090507 |
|
AS | Assignment |
Owner name: SEAGATE TECHNOLOGY INTERNATIONAL, CALIFORNIA Free format text: RELEASE;ASSIGNOR:JPMORGAN CHASE BANK, N.A., AS ADMINISTRATIVE AGENT;REEL/FRAME:025662/0001 Effective date: 20110114 Owner name: SEAGATE TECHNOLOGY LLC, CALIFORNIA Free format text: RELEASE;ASSIGNOR:JPMORGAN CHASE BANK, N.A., AS ADMINISTRATIVE AGENT;REEL/FRAME:025662/0001 Effective date: 20110114 Owner name: MAXTOR CORPORATION, CALIFORNIA Free format text: RELEASE;ASSIGNOR:JPMORGAN CHASE BANK, N.A., AS ADMINISTRATIVE AGENT;REEL/FRAME:025662/0001 Effective date: 20110114 Owner name: SEAGATE TECHNOLOGY HDD HOLDINGS, CALIFORNIA Free format text: RELEASE;ASSIGNOR:JPMORGAN CHASE BANK, N.A., AS ADMINISTRATIVE AGENT;REEL/FRAME:025662/0001 Effective date: 20110114 |
|
AS | Assignment |
Owner name: THE BANK OF NOVA SCOTIA, AS ADMINISTRATIVE AGENT, Free format text: SECURITY AGREEMENT;ASSIGNOR:SEAGATE TECHNOLOGY LLC;REEL/FRAME:026010/0350 Effective date: 20110118 |
|
FPAY | Fee payment |
Year of fee payment: 12 |
|
AS | Assignment |
Owner name: EVAULT INC. (F/K/A I365 INC.), CALIFORNIA Free format text: TERMINATION AND RELEASE OF SECURITY INTEREST IN PATENT RIGHTS;ASSIGNOR:WELLS FARGO BANK, NATIONAL ASSOCIATION, AS COLLATERAL AGENT AND SECOND PRIORITY REPRESENTATIVE;REEL/FRAME:030833/0001 Effective date: 20130312 Owner name: SEAGATE TECHNOLOGY LLC, CALIFORNIA Free format text: TERMINATION AND RELEASE OF SECURITY INTEREST IN PATENT RIGHTS;ASSIGNOR:WELLS FARGO BANK, NATIONAL ASSOCIATION, AS COLLATERAL AGENT AND SECOND PRIORITY REPRESENTATIVE;REEL/FRAME:030833/0001 Effective date: 20130312 Owner name: SEAGATE TECHNOLOGY INTERNATIONAL, CAYMAN ISLANDS Free format text: TERMINATION AND RELEASE OF SECURITY INTEREST IN PATENT RIGHTS;ASSIGNOR:WELLS FARGO BANK, NATIONAL ASSOCIATION, AS COLLATERAL AGENT AND SECOND PRIORITY REPRESENTATIVE;REEL/FRAME:030833/0001 Effective date: 20130312 Owner name: SEAGATE TECHNOLOGY US HOLDINGS, INC., CALIFORNIA Free format text: TERMINATION AND RELEASE OF SECURITY INTEREST IN PATENT RIGHTS;ASSIGNOR:WELLS FARGO BANK, NATIONAL ASSOCIATION, AS COLLATERAL AGENT AND SECOND PRIORITY REPRESENTATIVE;REEL/FRAME:030833/0001 Effective date: 20130312 |