US20060153238A1

US20060153238A1 - Transfer of control data between network components

Info

Publication number: US20060153238A1
Application number: US11/303,561
Authority: US
Inventors: Gershon Bar-On; Benzi Ende; Simcha Pearl; Sorana Lazarovichi; Luke Chang; Noam Avni
Original assignee: Intel Corp
Current assignee: Intel Corp
Priority date: 2003-12-19
Filing date: 2005-12-15
Publication date: 2006-07-13
Also published as: WO2005067221A1; US20050135421A1; TW200524446A; CN1894906A; US7751442B2; DE112004002503B4; TWI264232B; DE112004002503T5

Abstract

A method and apparatus for transfer of power state data between network components. An embodiment of a method includes determining a command for a computer system, the computer system including a first network component and a second network component. The first network component and the second network component are linked by an interface. The method further includes inserting a message regarding the power state change in a data frame and transferring the data frame from the first network component to the second component via the interface, the data frame being transferred in a period between data packets.

Description

RELATED APPLICATION

This application is a continuation-in-part of U.S. patent application Ser. No. 10/741,314, filed Dec. 19, 2003.

FIELD

An embodiment of the invention relates to computer networks in general, and more specifically to transfer of control data between network components.

BACKGROUND

In a computer network, there are generally multiple layers in operation. For example, Ethernet includes the PHY (physical) and MAC (media access) layers. In many cases these represent separate network devices or components, such as separate Ethernet PHY and MAC devices, which communicate with each other via an interface between the devices.
In the operation of network devices, there is an increasing need to consider power considerations and the power drain of the devices. In order to save power, some conventional systems have certain lower power states to allow a system to reduce power consumption when activity is at reduced levels. Because of the layered nature of a network, there is may be a need to transfer power information between components in order to control power usage.
However, the transfer of control and status information, such as power information, may cause complications for certain network component interfaces. If a limited interface between network components is used to simplify the structure for the network components, then the transfer of control information between the components may be difficult. Adding a control interface, such as a power state interface, may require sideband communications and an increased pin count for the interface, which thus results in more complex design of network devices. For example, conventional stand-alone PHY devices often do not include a power state interface because of the pin-count limitation of the interface for the device. As a result, the devices may consume more energy than would be necessary for a device for which power save features have been implemented.
Semiconductor devices in a printed circuit board (PCB) typically communicate through a device-to-device interconnection (DDI). Such a DDI typically includes copper traces formed in the PCB to transmit signals between devices. A device may be coupled to a DDI by solder bonding or a device socket secured to the PCB.
Cisco Systems has promoted a Serial Gigabit Media Independent Interface (SGMII) format for transmitting Ethernet data frames between devices over a DDI according to a differential pair signal format. In particular, SGMII specifies the transmission of Ethernet data frames as 8 B/10 B code groups. Control information may be transmitted in an out-of-band control channel coupled between the devices.
IEEE Std. 802.3ae-2002, Clause 47 defines a 10 Gigabit Attachment Unit Interface (XAUI) for transmitting data between devices in data lanes. Each data lane typically transmits a serial data signal between the devices using a differential signaling pair. A XAUI is typically coupled to a 10 Gigabit Media Independent Interface (XGMII) which is capable of transmitting or receiving data at a data rate of ten gigabits per second. In addition, the XAUI format may be used in transmitting data over an Infiniband 4x cable as described in the proposed 10GBASE-CX4 standard presently being explored by the IEEE P802.3ak working group. A “device-to-device interconnection” (DDI) as referred to herein relates to a data link to transmit data between devices. For example, a DDI may be formed by conductive traces formed on a circuit board between device sockets to receive devices. A DDI may traverse multiple devices coupled between two devices over a backplane and comprise conductive traces coupling the devices to one another. In another example, a DDI may comprise a cable coupled between two connectors at opposite ends of the cable. Each connector may then transmit data between the cable and a device coupled to the connector by conductive traces. However, these are merely examples of a DDI and embodiments of the present invention are not limited in these respects.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention may be best understood by referring to the following description and accompanying drawings that are used to illustrate embodiments of the invention. In the drawings:
FIG. 1 is an illustration of network devices in an embodiment of the invention;
FIG. 2 is an illustration of a data frame used to transport control data in an embodiment of the invention;
FIG. 3 is a diagram of a possible in-band frame for control data in an embodiment of the invention;
FIG. 4 is a flowchart to illustrate an embodiment of the transfer of power state data;
FIG. 5 is a flowchart to illustrate an embodiment of modification of power state in response to power command data;
FIG. 6 is an illustration of a computer system that may be utilized in conjunction with an embodiment of the invention; and
FIG. 7 shows a schematic diagram of devices coupled by a device-to-device interconnection (DDI) according to an embodiment of the invention.

DETAILED DESCRIPTION

A method and apparatus are described for transfer of control data between network components.
In one embodiment of the invention, control data is passed from a first network component to a second network component via a limited interface. In one embodiment, the control data includes power state data. In an embodiment, passing of control data is accomplished without additional side-band signals and without requiring software intervention.
In an embodiment of the invention, a network component passes control data to another component by inserting command data in an in-band frame between data packets. In an embodiment, a frame is transferred that includes one or more control fields, which may include, but is not limited to, a power control field. In embodiment, the receiving device receives the in-band frame, interprets the power control field, and adjusts the current power state based at least in part on the data contained in the field.
For the purposes herein, control data includes any data other than content data, and may include messages to notify a recipient of events, status, requests, or configuration commands. Control data includes, but is not limited to, power state data or other power information.
Ethernet is generally a physical and data link layer technology for local area networks (LANs). In the OSI (Open systems interconnection) model of layers, Ethernet technology operates at the physical (PHY) layer and the media access (MAC) portion or sublayer of the data link layer. While structures may vary in different systems, often the MAC and PHY functions are separate devices.
In the operation of computer systems, the conservation of power is often of prime importance. In network operations, a PHY device connected to and driving signals on the physical cabling of a network may consume significant power. However, the structure of the connections between elements complicates efforts to conserve power.
Computer components in general often have different power states, with the power states representing different levels of operation and power consumption. For example, power states are defined in general in relevant specifications, including the PCI (peripheral component interconnect) Bus Power Management specification (for example, PCI Power Management 1.2) and the Advanced Configuration and Power Interface Specification (for example, ACPI Revision 3.0, Sep. 2, 2004). The PCI Bus Power Management specification is intended to enhance the PCI architecture by including standardized power-management capabilities. This specification is architecturally aligned with the ACPI specification, and enables PCI devices to participate in platform-wide and operating system directed power management.
In this power management structure, four power states are defined for devices, with the states being defined for each device in terms of power consumption, device context (how much of the context of the device is retained and does not need to be restored by the operation system), the requirements for the device driver to restore the device to full operation, and the length of time required to restore the device to full operation. These states may be designated as D0 (fully on), intermediate states D1 and D2 (with reduced power consumption and less context retained), and D3 (device off). The power saving level thus is derived from the current power state.
Because a PHY device (or other related device) is often one of the largest power consumers in a system, power savings can be achieved by controlling the power consumption of the PHY through the passing of power control commands. However, if the PHY is not integrated with the MAC, then passing power signals from MAC to the PHY may be difficult. In particular, if the MAC and PHY utilize a reduced MAC/PHY interface, there are fewer options for signal transmission. For example, it may be necessary to utilize extra sideband signals to pass power states because there is commonly no power state interface. As an alternative, power states could be programmed through, for example, a register that could hold the current power state. However, this operation requires extra software intervention in the system operation. A PHY device could in some instances have a pin to shut down or disable the PHY, but this does not provide smart power management. Each port may have its own power state, but there is no per port power savings available.
Under an embodiment of the invention, a control command is passed to the PHY without software intervention. In an embodiment of the invention, a control command is passed to the PHY without additional side-band signals. In an embodiment of the invention, the control command includes a power state. Embodiments of the invention are not limited to power states, and may include any control data that is transferred between components or elements. In an embodiment of the invention, a smart power saving algorithm may be implemented in the PHY through use of transferred control data. Further, a PHY may include smart power saving on a per port basis.
In one particular example, a reduced interface may exist between Ethernet MAC and PHY, with the reduced interface being used to minimize the routing between the MAC and the PHY. In an embodiment of the invention, the interface may use in-band frames to provide control information. For the purposes of this description, an in-band frame is a frame that is transferred with data packets or content information, in contrast with an out-of-band message. In an embodiment of the invention, the in-band data is transferred in the gap between data packets. The interface will define the relevant frame fields for the device.
Ethernet devices are required to allow a minimum idle period between transmission of frames known as the interframe gap (IFG) or interpacket gap (IPG). The IPG provides a recovery time period between frames, which allows a device time to prepare for reception of the subsequent frame. Generally the minimum IPG for Ethernet is 96 bit times, which is 9.6 microseconds for 10 Mb/s Ethernet, 960 nanoseconds for 100 Mb/s Ethernet, and 96 nanoseconds for 1 Gb/s Ethernet. Under an embodiment of the invention, the control data is transported in a frame between the data frames, an in-band frame transferred during the IFG. In one embodiment of the invention, power states are transported in such in-band frames.
In an embodiment of the invention, sending control states between network devices may allow for simplified board design for network devices, while enabling smart power saving algorithms for a network device. For example, a PHY device may be simplified in design because of the limited interface. However, the power consumption of the PHY device may be controlled through the transport of power states between devices.
FIG. 1 is an illustration of network devices in an embodiment of the invention. In this illustration, a first network device is a PHY device 105 and a second network device is a MAC device 110. The PHY device 105 and the MAC device 110 are coupled by an interface 115, which may be linked through a connector. The PHY device 105 is coupled to certain ports 120, which may used for transmitting and receiving data. For example, a series of inbound frames of data 130 may be received at a port and transferred by the PHY device 105 to the MAC device 110. After the MAC device 110 processes and validates a frame of data, the frame is sent on to network devices 125. The data flow may occur in both directions, and an outbound flow of data 135 is also illustrated.
The interface 115 may vary in different systems. In one example, the interface is a reduced interface that minimizes the number of interconnections between the PHY device 105 and the MAC device 110. However, the use of a reduced interface does not provide for paths for sideband communications that may be used to transfer power states. In an embodiment of the invention, the interface uses an in-band frame 140 sent in a gap between data packets to carry control data. In an embodiment of the invention, an in-band control frame includes a power state field that is used to control power consumption. In an embodiment, the frame includes a power state field that may instruct the PHY to reduce power consumption by moving the PHY into a lower power state. When the power state changes, then in-band status and control frame with the needed power state change is provided. While a single frame is shown, any number of in-band frames may be transferred, with the MAC possibly sending an in-band frame between each two data packets, and may potentially send more than one in-band frames between two data packets. In an embodiment of the invention, the PHY device may also utilize in-band frames to communicate other control messages to the MAC in the inbound data stream 130. Among other types of messages, the PHY may send a confirmation message in response to a command, or an error message if a command appears to contain an error.
FIG. 2 is an illustration of a data frame used to transport control data in an embodiment of the invention. In this illustration, a first packet of data 205 is followed by a second packet of data 210. Between the two frames is the IPG 215, which is the expected gap between two packets of data. In an embodiment of the invention, between the two frames is an in-band frame 220, a frame that begins after the end of the first packet 205 and ends before the beginning of the second packet 210. In an embodiment, the in-band frame 220 includes a power control field, the power control field providing the current power state for the network. In one example, the first packet 205 and the second packet 210 are data packets being transferred from an Ethernet MAC device to an Ethernet PHY device. In this example, the PHY device will read the in-band frame 220 and determine, for e.g., whether a change in power state has occurred or other control change has been made. Based at least in part on the power state information contained in the in-band frame 220, the PHY may reduce operations and transition to a lower power state to conserve power, or may power back up to a higher state to enable more functionality.
FIG. 3 is a diagram of a possible in-band frame for control in an embodiment of the invention. The in-band frame provides one example of a frame that may be used, but embodiments of the invention may utilize any structure or order of fields in the frame. In this example, the frame is a particular length, in this particular case 39 bits long. The frame may provide for multiple control and status states or commands in this example a first bit 305 represents a type, and a second bit a “done” field 310. There are then 14 bits reserved for further use 315 and a bit for reset. Following the reset bit are two bits, with may be used to encode one of four different power states, the possible power states being the D0, D1, D2, and D3 states for the particular device. In this example, there are also four bits for control of LED's for display 330, an additional eight reserved bits 335, and a CRC (cyclic redundancy check) word for error detection for the frame.
FIG. 4 is a flowchart to illustrate an embodiment of the transfer of power state data. In this embodiment, the transfer of data frames from a first network device (such as a MAC device) to a second network device (such as a PHY device) is illustrated for simplicity, but there will likely also be traffic from the second network device to the first network device, in addition to other complicities that are not illustrated here. The flowchart is limited to an illustration of the transfer of power state data, but embodiments are not limited to this example, and other control information may be handled in a similar manner.
In one embodiment, the first device received data packets periodically for transmission 405. The MAC device sends the data packet to the PHY device 410, as normal operations. If there are no control messages to be sent to the PHY device, the MAC may not send an in-band frame and simply waits the needed IPG time between frames 420 before transferring the next frame 410. However if there is a power state change or another control or state signal is needed 415, then an in-band frame, including a power state frame, is inserted in the data stream between data packets 425 before returning to sending the next data packet 410.
FIG. 5 is a flowchart to illustrate an embodiment of modification of power state in response to power command data. In this illustration, a PHY device will receive a data packet from a MAC device 505. The PHY will process and deliver the data packet to the appropriate port for transmission 510. After the data packet, an in-band frame may be received 515. If there is no in-band frame, the PHY device will wait the IPG time period before the possible arrival of another data packet 505. If an in-band frame is received, the PHY will interpret the in-band frame 525. If there is a power control state change command 530, the PHY device will change its power state in response to the command 535. If there are any other commands 540, these commands may also be implemented 545. While this diagram for simplicity illustrates the PHY device complying with a power state change and other commands during the time period between data packets, the timing of the operations may vary in different embodiments. The PHY may implement certain commands after the arrival of the next data packet, or otherwise vary the timing of the implementation of commands as appropriate in the context of the operation.
FIG. 6 is an illustration of a computer system that may be utilized in conjunction with an embodiment of the invention. Under an embodiment of the invention, a computer 600 comprises a bus 605 or other communication means for communicating information, and a processing means such as two or more processors 610 (shown as a first processor 615 and a second processor 620) coupled with the bus 605 for processing information. The processors 610 may comprise one or more physical processors and one or more logical processors. Further, each of the processors 610 may include multiple processor cores.
The computer 600 further comprises a random access memory (RAM) or other dynamic storage device as a main memory 625 for storing information and instructions to be executed by the processors 610. Main memory 625 also may be used for storing temporary variables or other intermediate information during execution of instructions by the processors 610. The computer 600 also may comprise a read only memory (ROM) 630 and/or other static storage device for storing static information and instructions for the processors 610.
A data storage device 635 may also be coupled to the bus 605 of the computer 600 for storing information and instructions. The data storage device 635 may include a magnetic disk or optical disc and its corresponding drive, flash memory or other nonvolatile memory, or other memory device. Such elements may be combined together or may be separate components, and utilize parts of other elements of the computer 600.
The computer 600 may also be coupled via the bus 605 to a display device 640, such as a cathode ray tube (CRT) display, a liquid crystal display (LCD), a plasma display, or any other display technology, for displaying information to an end user. In some environments, the display device may be a touch-screen that is also utilized as at least a part of an input device. In some environments, display device 640 may be or may include an audio device, such as a speaker for providing audio information. An input device 645 may be coupled to the bus 605 for communicating information and/or command selections to the processors 610. In various implementations, input device 645 may be a keyboard, a keypad, a touch-screen and stylus, a voice-activated system, or other input device, or combinations of such devices. Another type of user input device that may be included is a cursor control device 650, such as a mouse, a trackball, or cursor direction keys for communicating direction information and command selections to the one or more processors 610 and for controlling cursor movement on the display device 640.
A communication device 655 may also be coupled to the bus 605. Depending upon the particular implementation, the communication device 655 may include a transceiver, a wireless modem, a network interface card, or other interface device. In one embodiment, the communication device 655 may include a firewall to protect the computer 600 from improper access. The computer 600 may be linked to a network or to other devices using the communication device 655, which may include links to the Internet, a local area network, or another environment. In an embodiment of the invention, the communication device 655 may comprise an Ethernet or similar network device. In one embodiment, the communication device 655 may comprise multiple components, such as an Ethernet PHY device and an Ethernet MAC device. In an embodiment, the PHY device ad the MAC device are coupled together and transfer control data, including power state information, between them. In an embodiment, the power states and other command data are transferred via in-band frames from the MAC device to the PHY device, which are placed between data frames during the IPG period.
The computer 600 may also comprise a power device or system 460, which may comprise a power supply, a battery, a solar cell, a fuel cell, or other system or device for providing or generating power. The power provided by the power device or system 660 may be distributed as required to elements of the computer 600.
A “serial data signal” as referred to herein relates to a signal comprising information encoded into a series of symbols. For example, a serial data signal may comprise a series of symbols transmitted in a transmission medium where each symbol is transmitted in a symbol period. However, this is merely an example of a serial data signal and embodiments of the present invention are not limited in these respects.
A “differential pair signal” as referred to herein relates to a pair of synchronized signals to transmit encoded data to a destination. For example, differential pair signal may transmit a serial data signal comprising symbols to be decoded for data recovery at a destination. Such a differential pair signal may transmit each symbol as a voltage on each of two transmission media. However, these are merely examples of a differential pair signal and embodiments of the present invention are not limited in these respects.
An “8 B/10 B encoding scheme” as referred to herein relates to a process by which eight-bit data bytes may be encoded into ten-bit “code groups” (e.g., 8 B/10 B code groups), or a process by which ten-bit code groups may be decoded to eight-bit data bytes according to a predetermined “8 B/10 B code group mapping.” An “8 B/10 B encoder” as referred to herein relates to logic to encode an eight-bit data byte to a ten-bit code group, and an “8 B/10 B decoder” as referred to herein relates to logic to decode an eight-bit byte from a ten-bit code group. An “8 B/10 B codec” as referred to herein relates to a combination of an 8 B/10 B encoder and an 8 B/10 B decoder.
“Transmission medium” as referred to herein relates to a medium capable of transmitting data from a source to a destination. For example, a transmission medium may comprise cabling (e.g., coaxial, unshielded twisted wire pair or fiber optic cabling), printed circuit board traces or a wireless transmission medium. However, these are merely examples of a transmission medium and embodiments of the present invention are not limited in these respects.
An “Ethernet data frame” as referred to herein relates to a format for transmitting data in a data link according to a protocol provided in versions of IEEE Std. 802.3 (e.g., to transmit data frames according to 10BASE-X, 100BASE-X, 1000BASE-X or 10GBASE-X protocols). An Ethernet data frame may include, for example, a header portion including a media access control (MAC) address and a payload portion including content data to be processed at a destination. However, this is merely an example of an Ethernet data frame and embodiments of the present invention are not limited in these respects.
An Ethernet data frame may be used to transmit content data between devices or nodes in a data channel. A “control message” as referred to herein relates to messages that may be transmitted between devices or nodes other than content data to notify a node or device receiving the control message of events, status, requests or configuration commands. However, these are merely examples of a control message and embodiments of the present invention are not limited in these respects. A control message may be transmitted in a communication channel which is distinct from a data channel as an “out-of-band” message. Alternatively, a control message may be inserted or interleaved among content data transmitted in a data channel as an “in-band” message.
Briefly, an embodiment of the present invention relates to the transmission of 8 B/10 B code groups including Ethernet data frames in a DDI. Control messages may be inserted among the 8 B/10 B code groups for transmission to a destination device. However, this is merely an example embodiment and other embodiments are not limited in these respects.
FIG. 7 shows a schematic diagram of a system 10 for transmitting data to and receiving data from a node 34 through a transmission medium 32. The transmission medium 32 may comprise any one of several mediums suitable for transmitting data in a data link such as, for example, a cable (e.g., coaxial, unshielded twisted wire pair or fiber optic) or a wireless transmission medium. The transmission medium 32 may transmit data between the node 34 and a data transceiver 12 in Ethernet data frames according to versions of IEEE Std. 802.3 (e.g., 10 BASE-X, 100BASE-X, 1000BASE-X or 10GBASE-X).
The data transceiver 12 may be coupled to a controller 18 by a DDI. The DDI may transmit a first differential pair signal 14 from the data transceiver 12 to the controller 18 and transmit a second differential pair signal 16 from the controller 18 to the data transceiver 12. According to an embodiment, each of the first and second differential pair signals 14 and 16 may be transmitted in a single pair of conductive traces (e.g., formed in a printed circuit board, not shown) in the DDI coupled between the data transceiver 12 and the controller 18. Accordingly, components containing the data transceiver 12 and the controller 18 may be coupled to one another by four device pins (not shown) on each component (where each component comprises two device pins to transmit or receive differential pair signal 14 and two devices pins to transmit or receive differential pair signal 16). The device pins may be coupled to the DDI by solder bonding or device sockets which are mounted to the DDI and adapted to receive the components containing the data transceiver 12 and controller 18. However, these are merely examples of how device pins may be coupled to a DDI and embodiments of the present invention are not limited in these respects.
The data transceiver 12 may comprise a physical media dependent (PMD) section (not shown) for transmitting data to and receiving data from the transmission medium 32 according to a physical layer data transmission protocol such as Gigabit Ethernet over unshielded twisted wire pair cabling (or 1000BASE-T) or 10 Gigabit Ethernet over unshielded twisted wire pair cabling (or 10GBASE-T). For example, the PMD section may comprise circuitry to detect individual bits in Ethernet data frames received from the transmission medium 32 (e.g., clock and data recovery circuitry) and circuitry to transmit individual bits in Ethernet data frames transmitted to the node 34. The data transceiver 12 may also comprise circuitry (not shown) to encode eight bit bytes making up Ethernet data frames received from the transmission medium 32 (via the PMD section) into ten bit code groups for transmission to the controller 18 on differential pair signal 14 as a serial data signal. The data transceiver 12 may encode the eight bit bytes into ten bit code groups (e.g., 8 B/10 B code groups) as described in IEEE 802.3-2002, Clause 36. Similarly, the data transceiver 12 may comprise circuitry to decode 8 B/10 B code groups received from the differential pair signal 16 into eight bit bytes for transmission in the transmission medium 32 via the PMD section.
The controller 18 may comprise a deserializer 20 to recover 8 B/10 B code groups from the differential pair signal 14 and a serializer 22 to transmit 8 B/10 B code groups to the data transceiver 12 as a serial data signal over the differential pair signal 16. A physical coding sublayer (PCS) section 18 may decode the 8 B/10 B code groups recovered from the deserializer 20 to reconstruct eight-bit bytes of Ethernet data frames received at the data transceiver 12 from node 34. Similarly, the PCS section 18 may encode eight-bit bytes of Ethernet data frames into 8 B/10 B code groups for the serializer 22 to transmit to the data transceiver 12 in differential pair signal 16 (for transmission to node 34).
The PCS section 24 may be coupled to a media access control (MAC) receive block 26 to provide Ethernet data frames reassembled from eight-bit bytes decoded from 8 B/10 B code groups. The PCS section 24 may also be coupled a MAC transmit block 28 to receive Ethernet data frames for transmission through the transmission medium 32. The MAC receive block 26 and MAC transmit block 28 may be coupled at a signaling interface providing a Gigabit Media Independent Interface (GMIT) as defined in IEEE Std. 802.3-2000, Clause 36. However this is merely an example of how portions of a MAC device may be coupled to a PCS section and embodiments of the present invention are not limited in this respect.
The differential pair signals 14 and 16 may transmit Ethernet data frames as 8 B/10 B code groups between the data transceiver 12 and controller 18 as provided in IEEE Std. 802.3-2000, Clause 36.2.4. Such code groups used for the transmission of Ethernet data frames may include, for example, ordered code group sets for establishing bit and code group synchronization, data code groups, idle code group (/I/), start of packet delimiter code group (/S/), end of packet delimiter code group (/T/), carrier extend code group (/R/) and error propagation code group (/V/). In addition to transmitting 8 B/10 B code groups in the differential pair signals 14 and 16 for the transmission of Ethernet data frames, the controller 18 and data transceiver 12 may transmit in-band control messages in the differential pair signals 14 and 16 along with encoded portions of Ethernet data frames. Such in-band control messages may be transmitted as 8 B/10 B code groups inserted among 8 B/10 B code groups transmitting encoded eight-bit bytes of Ethernet data frames. By transmitting the control messages in-band as 8 B/10 B code groups inserted among 8 B/10 B code groups transmitted over the differential pair signals 14 and 16, control messages which would otherwise be transmitted in a management data input/output (MDIO) interface (either at the data transceiver 12 or controller 18) may be transmitted as the inserted 8 B/10 B code groups.
According to an embodiment, control messages may be inserted among the 8 B/10 B code groups in differential pair signals 14 and 16 following an end of packet delimiter /T/ as a six byte (or code group) sequence. When transmitting a control message, for example, the code group sequence:
/T/R/K28.5/Dx.y/(six byte control message)/K28.5/Dx.y/
may be substituted for the typical code group sequence following an Ethernet data frame:
/T/R/K28.5/Dx.y/K28.5/Dx.y/K28.5/Dx.y/K28.5/Dx.y/K28.5/Dx.y/.
In this example, a six byte idle code group sequence “/K28.5/Dx.y/K28.5/Dx.y/K28.5/Dx.y/” in the typical code group sequence may be substituted with six bytes forming the control message. A first byte of the six byte control message may include a special symbol to indicate the presence of a control message (e.g., to access MDIO registers at the destination device) such as “/K28.1/” including a comma. A second byte may specify read or write access to specific MDIO registers. Third and fourth bytes may specify information to be written to an MDIO register and a fifth byte may be reserved. Finally, a sixth byte may include a cyclic redundancy code for error correction (excluding the special symbol /K28.1/). Similar six byte packets may be formatted for read access acknowledge/response control messages or write access acknowledge control messages. However, these are merely an example of how a control message may be inserted among 8 b/10 B code groups for transmitting Ethernet data frames and embodiments of the present invention are not limited in these respects.
According to an embodiment, the PCS section 24 may comprise circuitry 30 to detect 8 B/10 B code groups carrying in-band control messages from among 8 B/10 B code groups received from differential pair signal 14, and decode the control messages from the detected 8 B/10 B code groups according to a predetermined mapping of 8 B/10 B code groups to control messages. In response to external signals (not shown), the circuitry 30 may encode control messages for transmission to the data transceiver 12 as 8 B/10 B code groups (e.g., inserted among 8 B/10 B code groups on differential pair signal 16 containing Ethernet data frames) according to the predetermined mapping of 8 B/10 B code groups to control messages.
The data transceiver 12 may also comprise circuitry (not shown) to detect 8 B/10 B code groups carrying in-band control messages from among 8 B/10 B code groups received from differential pair signal 16, and decode the control messages from the detected 8 B/10 B code groups according to the predetermined mapping of 8 B/10 B code groups to control messages. Similarly, the data transceiver 12 may also comprise circuitry to encode control messages for transmission to the data controller 18 as 8 B/10 B code groups (e.g., inserted among 8 B/10 B code groups on differential pair signal 14 containing Ethernet data frames) according to the predetermined mapping of 8 B/10 code groups to control messages.
According to an embodiment, the data transceiver 12 and controller 18 may support multiple Ethernet protocols at different bit rates including 10 BASE-X (at 10 Mbps), 100 BASE-X (at 100 Mbps) and 1000 BASE-X (at 1000 Mbps). In addition, the data transceiver 12 and controller 18 may support an autonegotiation feature to select a data transmission protocol for use between the data transceiver 12 and the node 34 for transmitting Ethernet data frames in the transmission medium 32 as provided in IEEE Std. 802.3-2000, Clause 28. Accordingly, the data transceiver 12 may be capable of negotiating with the node 34 to select the data transmission protocol having the highest data rate from among common data transmission protocols (e.g., 10 BASE-X, 100 BASE-X, 1000BASE-X or 10GBASE-X). Following negotiation between the data transceiver 12 and the node 34 to the common data transmission protocol having the highest data rate, the controller 18 may communicate with the node 34 to identify and negotiate additional capabilities (e.g., abilities to transmit in full or half duplex modes) while communicating according to the selected data transmission protocol as provided in IEEE Std. 802.3-2000, Clause 37.
Among control messages that may be transmitted from the data transceiver 12 to the controller 18 in 8 B/10 B code groups over the differential pair signal 14, the data transceiver 12 may transmit one or more control messages to the controller 18 indicating a data transmission protocol or data rate selected through autonegotiation, or status of the data link between the data transceiver 12 and the node 34 (e.g., active versus inactive, connected versus unconnected, changes in data transmission mode from 10 Gbps to 1 Gbps, etc.). In response to receipt of either of these control messages, the controller 18 my respond by transmitting an acknowledgement in one or more 8 B/10 B code groups over the differential pair signal 16. However, these are merely examples of control messages that may be transmitted from a data transceiver to a controller in 8 B/10 B code groups over a differential pair signal and embodiments of the present invention are not limited in these respects.
Using control messages transmitted as 8 B/10 B code groups in the differential pair signal 14 and in response to data rate selected from autonegotiation with the node 34, the data transceiver 12 and controller 18 may configure the data rate of the differenitial pair signals 14 and 16 according to the selected data rate. For example, if the data rate selected through autonegotiation is 1000 Mbps (e.g., from a selected 1000BASE-X protocol), the data transceiver 12 and controller 18 may configure the differential pair signals 14 and 16 to transmit at a data rate of 1.25 Gbps. (allowing 250 Mbps of overhead for transmitting 8 B/10 B code groups encoded from eight-bit bytes of Ethernet data frames). For a selected data rate of 10 or 100 Mbps, the data transceiver 12 and controller 18 may transmit duplicate Ethernet data frames or code groups in differential pair signals 14 and 16 transmitting at 1.25 Gbps. Alternatively, if the data rate selected through autonegotiation is 10 or 100 Mbps (e.g., from a selected 10 BASE-X or 100BASE-X protocol), the data transceiver 12 and controller 18 may configure the differential pair signals 14 and 16 at a data rate of 125 Mbps. Transmitting differential pair signals 14 and 16 at the lower data rate of 125 Mbps may enable the data transceiver 12 and controller 18 to operate at lower power (over transmitting at the higher 1.25 Gbps. data rate).
According to an embodiment, the controller 18 may be included as part of a computing platform and coupled to a host processing system (e.g., including a host processor, I/O core logic and system memory) hosting an operating system and/or application programs. As such, the computing platform may define certain states and events such as, for example, a software reset event, power states (e.g., full power, standby, snooze, etc.) and events indicating a transition between power states. Among control messages that may be transmitted from the controller 18 to the data transceiver 12 in 8 B/10 B code groups over the differential pair signal 16, the controller 18 may transmit control messages indicating a change in the power state of the computing platform (e.g., change from full power to standby or snooze, or from standby or snooze to resume operation full power) enabling the data transceiver to operate at low voltage when the computing platform is not operating at a full power state. However, these are merely examples of control messages that may be transmitted from a controller to a data transceiver in 8 B/10 B code groups over a differential pair signal and embodiments of the present invention are not limited in these respects.
According to an embodiment, the controller 18 may perform code group and bit synchronization in response to the differential pair signal 14 to ensure the alignment of 8 B/10 B code groups from the data transceiver 12. Similarly, the data transceiver 12 may also perform code group and bit synchronization in response to the differential pair signal 16 to ensure alignment of 8 B/10 B code groups from the controller 18. The controller 18 and data transceiver 12 may perform this code group and bit synchronization as provided in IEEE Std. 802.3-2000, Clauses 36.2.4 and 36.2.5.2.6 to ensure synchronization of multi-code group ordered sets to code group boundaries. However, these are merely examples of how code group and bit synchronization may be established and embodiments of the present invention are not limited in these respects.
While transmitting control messages between the controller 18 and data transceiver 12 as in-band 8 B/10 B code groups and achieving code group and bit synchronization from detection of the received code groups, the controller 18 and data transceiver 12 need only communicate with each other through four device pins (i.e., four device pins on each device to enable transmission of the differential pair signals 14 and 16 between the data transceiver 12 and controller 18). For example, the use of separate pins for an MDIO interface may be avoided by transmitting control messages in-band over the differential pair signals 14 and 16.
The differential pair signals 14 and 16 may be transmitted in a DDI extending thirty inches or more over a circuit board coupling the data transceiver 12 and controller 18 to the DDI. According to an embodiment, the system 10 may be provided on a line card in a switch, router or other platform that may be used for forwarding the contents of an Ethernet data frame from the node 34 and another node. The system 10 may provide a single port among multiple ports coupled by switching circuitry (e.g., switch fabric or Ethernet switch, not shown) to forward data frames from a source port (or ingress port) to a destination port (or egress port). For example, the MAC receive block 26 and MAC transmit block 28 may be coupled to the switching circuitry to forward the contents of frames to, and receive frames from, other ports. Also, the MAC receive block 26 or MAC transmit block 28 may be coupled to network processing devices (e.g., network processor, packet processing ASIC or other device for performing packet classification, protocol processing, intrusion detection, etc.). However, these are merely examples of applications of a line card and embodiments of the present invention are not limited in these respects.
In an alternative embodiment, the system 10 may be provided in a system board or motherboard including a host processor (e.g., microprocessor for hosting an operating system and applications) and an I/O core logic chipset (e.g., system memory controller and peripheral I/O controller, not shown). In this embodiment, the controller 18 may be integrated with one or more portions of an 110 core logic chipset while the data transceiver 12 is located near a physical port connection (e.g., cable connection) separated from the I/O core logic chipset. The controller 18 may be coupled to a multiplex data bus as defined in versions of the Peripheral Components Interconnect (PCI) Local Bus Specification 2.3, PCI-X or PCI-Express (e.g., coupled to a “switch” entity). The system board or motherboard of the presently illustrated embodiment may be combined with a system memory for storing machine-readable instructions of an operating system or application programs to be executed by the host processor. For example, the host processor and system memory may host a device driver that defines buffer locations in the system memory that are used to store data packets received from the controller 18 in data frames or store data packets to be transmitted by the controller as Ethernet data frames. Additionally, the controller 18 may comprise a TCP/IP offload engine (not shown) for performing TCP/IP protocol processing on TCP/IP packets received in Ethernet data frames from the node 34.
Particular embodiments described herein relate to the transmission of 8 B/10 B code groups (e.g., including Ethernet data frames and control messages) between the data transceiver 12 and controller 18 in single differential pair signals 14 and 18. In other embodiments, however, the 8 B/10 B code groups may be transmitted between such a data transceiver and controller in multiple differential pair signals. For example, such a data transceiver and controller may be coupled by a DDI comprising a 10 Gigabit Attachment Unit Interface (XAUI) providing four differential pair signals to transmit 8 B/10 B code groups from the data transceiver to the controller and four differential pair signals to transmit 8 B/10 B code groups from the controller to the data transceiver. In this embodiment, the data transceiver and controller may each comprise sixteen device pins for coupling to the DDI, eight pins for transmitting 8 B/10 B code groups and eight pins for receiving 8 B/10 B code groups. Accordingly, 8 B/10 B code groups containing control messages may be inserted among 8 B/10 B code groups (containing Ethernet data frames) transmitted between the controller and data transceiver (in multiple differential pair signals) to obviate the need for an out-of band channel for transmitting the control messages between the data transceiver and the controller.
In the description above, for the purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the present invention. It will be apparent, however, to one skilled in the art that the present invention may be practiced without some of these specific details. In other instances, well-known structures and devices are shown in block diagram form.
The present invention may include various processes. The processes of the present invention may be performed by hardware components or may be embodied in machine-executable instructions, which may be used to cause a general-purpose or special-purpose processor or logic circuits programmed with the instructions to perform the processes. Alternatively, the processes may be performed by a combination of hardware and software.
Portions of the present invention may be provided as a computer program product, which may include a machine-readable medium having stored thereon instructions, which may be used to program a computer (or other electronic devices) to perform a process according to the present invention. The machine-readable medium may include, but is not limited to, floppy diskettes, optical disks, CD-ROMs (compact disk read-only memory), and magneto-optical disks, ROMs (read-only memory), RAMs (random access memory), EPROMs (erasable programmable read-only memory), EEPROMs (electrically-erasable programmable read-only memory), magnet or optical cards, flash memory, or other type of media/machine-readable medium suitable for storing electronic instructions. Moreover, the present invention may also be downloaded as a computer program product, wherein the program may be transferred from a remote computer to a requesting computer by way of data signals embodied in a carrier wave or other propagation medium via a communication link (e.g., a modem or network connection).
Many of the methods are described in their most basic form, but processes can be added to or deleted from any of the methods and information can be added or subtracted from any of the described messages without departing from the basic scope of the present invention. It will be apparent to those skilled in the art that many further modifications and adaptations can be made. The particular embodiments are not provided to limit the invention but to illustrate it. The scope of the present invention is not to be determined by the specific examples provided above but only by the claims below.
It should also be appreciated that reference throughout this specification to “one embodiment” or “an embodiment” means that a particular feature may be included in the practice of the invention. Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. This method of disclosure, however, is not to be interpreted as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims are hereby expressly incorporated into this description, with each claim standing on its own as a separate embodiment of this invention.

Claims

1. A method comprising:

determining a command for a computer system, the computer system including a first network component and a second network component, the first network component and the second network component being linked by an interface;

inserting a control message regarding the command in a data frame; and

transferring the data frame from the first network component to the second component via the interface, the data frame being transferred in a period between data packets.

2. The method of claim 1, wherein the first and second network components are Ethernet components.

3. The method of claim 2, wherein the first network component is a MAC (media access control) Ethernet component.

4. The method of claim 3, wherein the second network component is a PHY (physical) Ethernet component.

5. The method of claim 1, wherein the control message is sent in a gap between data packets.

6. The method of claim 1, wherein the control message includes a power state change.

7. The method of claim 6, further comprising changing the power state of the second network device based at least in part on the message regarding the power state change.

8. A network apparatus comprising:

a first component; and

a second component, the second component to be coupled with the first component via an interface, the second component to transfer a plurality of data packets to the first component, the first component to transfer a data frame in a gap between a first data packet and a second data packet, the data frame including a control message for the first component;

wherein the first component changes from a first state to a second state in response to the data frame.

9. The network apparatus of claim 8, wherein the first component is an Ethernet PHY (physical) component.

10. The network apparatus of claim 9, wherein the second component is an Ethernet MAC (media access) component.

11. The network apparatus of claim 8, wherein the control message is a power control message.

12. The network apparatus of claim 11, wherein the first component changes from a first power state to a second power state in response to the power control message.

13. The network apparatus of claim 11, wherein the interface does not include a power control line.

14. The network apparatus of claim 11, wherein the first component is coupled with a first port and wherein the data packets are to be transferred to the first port.

15. The network apparatus of claim 10, wherein the first component is coupled with a second port, wherein the first component has a first power state for the first port and a second power state for the second port.

16. A system comprising:

a bus;

a processor coupled to the bus;

a dynamic memory coupled to the bus to hold data for transmission;

a communication device coupled to the bus to transmit and receive data, the communication device including:

a physical network device;

and a media access network device, the media access device to transfer a plurality of data packets from the processor to the physical network device, the media access device to transmit a control signal to the physical network device in a period between a first data packet and a second data packet.

17. The system of claim 16, wherein the control signal includes a power control signal.

18. The system of claim 17, wherein the physical network device changes from a first power state to a second power state in response to the power control signal.

19. The system of claim 18, wherein the physical network device consumes less power in the second power state than in the first power state.

20. The system of claim 18, wherein the physical network device is coupled with a plurality of ports.

21. The system of claim 20, wherein the power control signal includes a plurality of power control states, the power control states including a first control power state for a first port and a second power control state for a second port.

22. The system of claim 16, wherein the communication device is an Ethernet device.

23. The system of claim 22, wherein the control signal is transmitted in an interpacket gap between Ethernet data packets.

24. The system of claim 16, wherein the control signal is a part of a control data frame.

25. A machine-readable medium having stored thereon data representing sequences of instructions that, when executed by a machine, cause the machine to perform operations comprising:

sending a first data packet from a first network component to a second network component;

sending a data frame from the first network component to the second network component after the end of the first data frame, the data frame including a control field;

sending a second data packet from the first network component to the second network component, the second data packet being sent after the end of the data frame, the data frame being sent in a time period between the first data packet end the second data packet; and

changing a state of the second network device based at least in part on the control field.

26. The medium of claim 25, wherein the data frame is an in-band frame sent in an inter-packet gap between the first data packet and the second data packet.

27. The medium of claim 25, wherein the control field encodes the current power state for the second network device.

28. The medium of claim 25, wherein the first network component and the second network components are components of an Ethernet communication device.