The invention relates to communication systems and in particular to a method and apparatus for interfacing different transmit rate communication systems.
With the accelerating deployment of Gigabit Ethernet there is a great need for data center equipment supporting a much faster rate in order to handle the aggregation of multiple gigabit links. In response, the IEEE 802.3 working group developed 10 gigabit (10 G) Ethernet. At first, only fiber optic media was specified to support 10 G Ethernet. Soon afterward a very short reach copper media standard was developed, known as 10GBASE-CX4. In June 2006 the 10GBASE-T standard (Clause 55) was approved by IEEE and specifies 10 G Ethernet over unshielded twisted pair (UTP) which is also used with the highly successful 10 megabit (10BASE-T), 100 megabit (100BASE-TX) and gigabit (1000BASE-T) Ethernet copper standards.
The arrival of 10GBASE-T brings a new capability to 10 G Ethernet equipment. That is the capability of operating with link partners of speeds from 10 megabit per second (Mbps) to 10 gigabit per second (Gpbs). However, there is no IEEE defined multi-rate interface between the media access controller (MAC) and the physical layer device (PHY).
802.3 is the IEEE standard for Ethernet networking. 802.3 clauses 44 through 55 define the set of physical coding sublayers (PCS), physical media attachments (PMA) and physical media dependants (PMD) for operation at 10 gigabit per second. Clause 46 defines the 10-gigabit media independent interface (XGMII) that serves as the universal interface between a 10 G media access controller (MAC) and the PCS regardless of the choice of media. XGMII is a 4 byte parallel interface operating at 312.5 MHz. Clause 47 defines the XGMII extension sub-layer (XGXS) and it's interface, the 10-gigabit attachment unit interface (XAUI). XAUI allows the XGMII to be extended across longer distances by serializing the 4 bytes into four serial lanes operating at 3.125 Gbps. Another extension of XGMII known as 10 Gigabit small form factor interface (XFI) is a single lane 10.3 Gbps serial interface using the PCS define in Clause 48 for 10GBASE-R.
802.3 also defines sets of PCS, PMA, and PMD for copper and fiber optic media at rates of 10 Mbps, 100 Mbps, and 1 Gbps each with a corresponding defined interface between the MAC and PCS. For 10 Mbps and 100 Mbps this is the media independent interface (MII) and for a 1 Gbps system it is the gigabit media independent interface (GMII).
However, 802.3 does not define a multi-rate media independent interface. For systems supporting 10 Mbps, 100 Mbps and 1 Gbps, known as 10/100/1000, various solutions were developed within the industry as defacto standards. One of these solutions was SGMII in which GMII is processed by the PCS defined in 802.3 Clause 36 for 1000BASE-X. The GMII is encoded using 8B10B coding and serialized for transmission at 1.25 Gbps.
Several solutions have been proposed, but these proposed solutions do not adequately address the drawbacks of the prior art. For example, the Serial-GMII Specification: ENG-46158 is an industry de-facto standard written and maintained by Cisco Systems. The Serial Gigabit Media Independent Interface (SGMII) is designed to convey network data and port speed between a 10/100/1000 PHY and a MAC. SGMII is specified to operate in both half and full duplex and at all port speeds. However SGMII does not support 10 gigabit operation, and does not support the XGMII interface defined for 10 G Ethernet. Other drawbacks exist with various other prior art systems.
To overcome the drawbacks of the prior art and to provide additional advantages, a universal interface is disclosed. In one embodiment, a rate adaptive interface is provided which is configured to interface a MAC device with a PHY device. In such an embodiment, the interface comprises a rate adaptation module in communication with a MAC device, the rate adaptation module configured to receive data at a rate selected from 10 Mb/s, 100 Mb/s, and 1 Gb/s and process the data to a rate of 1 Gb/s, which is in turn output at a rate of 1 Gb/s. The interface also comprises an encapsulation/recovery module configured to receive the data at a rate of 1 Gb/s from the rate adaptation module and then encapsulate the data at a rate of 1 Gb/s to generated data at a rate of 10 Gb/s. A multiplexer is configured to receive the data at a rate of 10 Gb/s from the encapsulation/recovery module or to received data from a MAC device at a rate of 10 Gb/s and selectively output data at a rate of 10 Gb/s responsive to a mode selection control signal. The mode selection module is configured to control the multiplexer based on control input from a higher layer device. An ordered set generation and detection module is also part of this embodiment and is in communication with the multiplexer and the mode selection module. The ordered set generation and detection module is configured to detect ordered sets which announce a mode change such that the ordered set generation and detection module is in communication with the mode selection module.
In one embodiment, the encapsulation/recovery module is further configured to recover encapsulated data thereby changing the data rate from a 10 G data rate to a 1 G data rate. In one configuration, the encapsulation/recovery module is further configured with an input to receive data at a rate of 10 Gb/s and perform recovery thereon to output the data at a rate of 1 Gb/s. It is also contemplated that the system of Claim 1, wherein the mode selection module adjusts the data rate of operation between 10 Mb/s, 100 Mb/s, or 1 Gb/s.
Also disclosed herein is a rate adaptive interface for use in a network device. In this configuration the interface comprises a MAC device configured to output data from a first port at a variable first rate and from a second port at a second rate. A rate adaptation module is part of this embodiment and configured to receive data from the MAC at the variable first rate and convert the data at the variable first rate to data at the second rate. An encapsulation module configured to receive data at the second rate from the rate adaptation module and convert the data at the second rate to data at a third rate. From there, a switch is configured to interface with the encapsulation module and the MAC device to receive data at a rate of 10 Gb/s from either the encapsulation module or from the second port of the MAC device and then output the data at a rate of 10 Gb/s. The switch may be controlled by a mode selection control signal. In this embodiment, a mode selection module is configured to provide the mode selection control signal to the switch to thereby control operation of the switch.
It is further contemplated that the system further comprise an extension sublayer configured to receive data at a rate of 10 Gb/s from the switch, such that the extension sublayer extends the distance which the data at a rate of 10 Gb/s may be transmitted. In addition, the system may further comprise an ordered set generation and detection module configured interface with the switch to detect ordered sets which determine a mode change, such that the ordered set generation and detection module is in communication with the mode selection module. The encapsulation/recovery module may be further configured to recover encapsulated data thereby changing the data rate from a 10 G data rate to a 1 G data rate. In one embodiment, the data at the variable first rate comprises data at a rate of 10 Mb/s, 100 Mb/s, or 1 Gb/s, and data at the second rate comprises data at a rate of 1 Gb/s and data at the third rate comprises data at a rate of 10 Gb/s. In addition, the data at the third rate may have a format that is different than the data at the second rate. This system may further comprise a PHY Device configured to receive data at a rate of 10 Gb/s from the switch. In one embodiment, the first port and the second port comprise input/output ports and the switch is configured to receive and transmit data to either of the second port of the MAC device or the encapsulation module.
Also disclosed herein is a method for interfacing a multirate MAC device with a PHY device in a network communication device. In this example embodiment, this the method comprises outputting unprocessed data from the multirate MAC device at a variable rate to a rate adaptation module or at a first fixed rate to a multiplexer. The method processes the data at the rate adaptation module to up-convert the data at a variable rate to data at a second fixed rate. Then, the method processes the data at the second fixed rate with an encapsulation module to generate processed data at the first fixed rate. In this embodiment, the method receives, at a multiplexer, the unprocessed data from the MAC device at the first fixed rate or the processed data at the first fixed rate, and also receives, at the multiplexer, a mode selection signal. Responsive to the mode selection signal, the operation outputs from the multiplexer the unprocessed data at the first rate or the processed data at the first rate. In one variation, the variable rate consists of 10 Mb/s, 100 Mb/s and 1 Gb/s. It is contemplated that the first fixed rate comprises 10 Gb/s and the second fixed rate comprises 1 Gb/s. In addition, processing the data at the second fixed rate with an encapsulation module further comprise altering the format of the data. This method may further comprise outputting the data from the multiplexer to an extension sublayer, a PHY device, or a second multiplexer. In this embodiment, up-sampling may comprise padding or repeating data received at the variable rate to create data at a second fixed rate. This method may also comprise establishing a mode of operation and receiving data at the first fixed rate at the multiplexer from any of a PHY device, an extension sublayer, or a second multiplexer. Then, responsive to the mode of operation, the method outputs the data at the first fixed rate to either the MAC device or to a recovery unit. In addition, responsive to the outputting the data at the first fixed rate to a recovery unit, the method converts the data at the first fixed rate to the second fixed rate and converts the data at the second fixed rate to data at a third fixed rate, such that the data at the third fixed rate is a rate selected from a group of variable rates supported by the MAC.
Other systems, methods, features and advantages of the invention will be or will become apparent to one with skill in the art upon examination of the following figures and detailed description. It is intended that all such additional systems, methods, features and advantages be included within this description, be within the scope of the invention, and be protected by the accompanying claims.
The components in the figures are not necessarily to scale, emphasis instead being placed upon illustrating the principles of the invention. In the figures, like reference numerals designate corresponding parts throughout the different views.
An example UGMII (Ultra-serial Gigabit Media Independent Interface) Specification as disclosed herein is a proposed specification for a multi-rate 10 G capable MII. Although one example preferred embodiment is described herein, it is contemplated that numerous other embodiments may be enabled based on the teachings provided herein although such embodiments may differ slightly from the exact structure shown.
In the UGMII system described herein the defined signals for GMII may be fully encapsulated without modification, transported across the (extended) XGMII interface and reconstructed at the MAC RX and PHY TX inputs.
The MAC 130 communicates with the UGMII module 134 via the conductor paths 132 as shown. The conductor paths 132 may comprise 36 parallel lines as shown, or any other type connection. The data exchange format between the MAC 130 and the UGMII module 134 may comprise any of MIIGMII, MII or XGMII as shown. XGMII is defined by one or more IEEE standards and as such, is not described in detail herein, but it does suffer from the drawback of not adapting to rates. One such drawback is that it is fixed at 10 Gb/s.
The UGMII module 134 is configured to convert any rate of incoming or outgoing data between the MAC 130 and a PHY 140 to thereby provide a universal intermediate device capable of interfacing multi-rate MAC and PHY modules. In this embodiment, the UGMII module 134 includes 10 gigabit data rate transfer capability. The UGMII module 134 is described below in detail in connection with
The output of the UGMII module 134connectconnects to a PHY 140. In this embodiment, the PHY (physical layer device) 140 is capable of operation at any data rate, which in this embodiment may be 10 megabits per second (Mb/s), 100 Mb/s, 1 Gigabits per second (Gb/s), or 10 Gb/s. The PHY 140 may comprise any type of device currently in existence, such as defined by one or more IEEE standards, or developed in the future capable of addressing and channel access control. PHY 140 configuration and operation is generally understood by one of ordinary skill in the art and as such, is not described in detail herein.
The output of the PHY 140 connects to the channels 144, represented by MDI. MDI channels 144 may comprises a multi-line channel, such as but not limited to CAT 3, CAT 5, CAT 5E, CAT 6, CAT6A, or CAT7. Shown on the left hand side of
In operation, the UGMII system performs numerous different functions. In this example embodiment the first aspect of the invention is mode selection. In this embodiment, mode selection comprises the negotiation between the UGMII extension sublayers located at the MAC and PHY to select between one of several operational modes including: XGMII communication, GMII encapsulation, Clause 22 MDIO register management and Clause 45 MDIO register management. The management aspects of the PHY or MAC set the exact link speed, which may be based on the capability of the channel or hardware. In one embodiment, selection of UGMII and XGMII operating modes, or any other operating mode, are negotiated between the MAC and PHY using ordered sets to announce and acknowledgement a mode change. In one embodiment 802.3 Clause 46 defined ordered sets are utilized. The Clause 22 and 45 define the universal standard set of management registers to give status and control of the physical layer. These clauses defined the address, location, and control data stored therein.
In this example embodiment a device incorporating UGMII system enables the support the following interface modes.
The rate adaptation module 214 may also be configured as hardware, software, or a combination of both and be configured to adapt the rate of lower speed data up to a higher rate, such as GMII 1 Gb/s rate. The MII/GMII I/O port 210 connects the MAC 208 to a rate adaptation module 214 which is configured to adapt the variable rate from the MAC 208 to a constant rate of 1 Gb/s. At a rate of 1 Gb/s, the rate adaptation module 214 connects to a GMII encapsulation/recovery module 216. The GMII encapsulation/recovery module 216 up converts data received from the rate adaptation module 214 at 1 Gb/s to a rate of 10 Gb/s, which is in turn provided to a multiplexer 220. GMII data is used to create small packets or frames to be transmitted across the XGMII and may be padded with zeros or other pad data.
Conversely, data received from the multiplexer 220 at a rate of 10 Gb/s is converted to a 1 Gb/s rate by the GMII encapsulation/recover module 216. The multiplexer 220 also receives data without rate conversion from the MAC 208 via the XGMII I/O port as shown. The input and output from the multiplexer 220 on the MAC 208 side occurs over 36 parallel conductors at a rate of 10 Gb/s, but in other embodiments, other rates may be utilized. The path between the rate adaptation module 214 and the GMII encapsulation/recovery module 216 may comprise any type path, but in this embodiment comprises 12 parallel conductors.
Also connecting to the multiplexer 220 on the MAC 208 side is an ordered set generation and detection unit 224 which is configured to communicate with the multiplexer as shown. The ordered set generation and detection unit 224 monitors and/or generates ordered sets from the MAC 208 or the multiplexer 220, or any higher layer. The higher layer may mean any layer in the OSI model or from an application or user input. In one embodiment, the ordered set generation and detection unit 224 both sends and detects received ordered sets.
Orders sets comprise data utilized to control operation of the MAC, PHY, and/or the UGMII module. The ordered sets may be considered as control data that establish the link characteristics and may set the rate of data exchange. In one embodiment, ordered sets are not exchanged during data mode.
The ordered set generation and detection unit 224 also communicates with the mode selection unit 228. The mode selection unit 228 provides a control input to the multiplexer 220 to selectively control which sets of input/output channels to use for communication during operation. For example, if the PHY is set to operate at 10M, 100M, or 1 G, then the mode selection unit 228 will force the multiplexer 220 to use the I/O ports corresponding to the MAC I/O ports 210. Alternatively, if the rate is set at 10 Gb/s then the multiplexer 220 will be forced to utilized the I/O ports associated with the MAC 208 ports 212. In one embodiment, the mode selection unit 228 is configured to report the detection of ordered sets back to the MAC 208, but it could alternatively be done by the ordered set generation/detection unit 224. It is also possible for control data or ordered sets to be exchanged via a status line from the mode selection unit 228 to the MAC 208. In one implementation the MAC 208 would set the MUX 220 by configuring the mode selection unit 228. However, a more complex design could allow the mode selection unit 228 to set the MUX 220 based on mode setting from the MAC 208 and the detection of ordered sets
The multiplexer I/O port 240 operates a 10 Gb/s and may optionally communicate with an optional extension sublayer 244 or directly with a PHY. The optional extension sublayer 244 may be utilized to extend the distances between the MAC 208 and the PHY.
In this example embodiment another aspect of the invention is data up sampling which is discuss in connection with
Up sampling can be replaced by one of several other means for rate adaptation including idle stuffing between data bytes.
After receipt and at step 412, the operation up samples the octet one time, ten times, or one hundred times for the rates of 1 G, 100M or 10M respectively. For example, to create a data rate of 1 G from a 10M receive rate, up sampling occurs at a rate of 100 times. Up sampling may occur by repeating the received octet or padding with created pad bits. Thereafter, at step 416, the operation passes the up sampled octets to the encapsulation block at a rate of 1 G. As a result, data received at a variable rate is output at a constant rate of 1 G.
Another aspect of the invention is GMII encapsulation.
GMII is defined in Clause 35 of IEEE STD 802.3-2005, Carrier sense multiple access with collision detection (CSMA/CD) access method and physical layer specifications, December 2005, which is incorporated herein in its entirety. In this example embodiment disclosed herein, the MAC transmits to the PHY using data signals TXD<7:0>, transmit enable TX_EN, and transmit error TX_ER. The PHY transmits to the MAC using data signals RXD<7:0>, data valid RX_DV, receive error RX_ER, and carrier sense CRS. In other embodiments different MAC transmit schemes may be utilized. In one embodiment, encapsulation is performed on a block of N GMII transfers (octets) and generates multiple XGMII transfers.
On the transmit side of the MAC N (for N>1) octets of data, TXD(0)<7:0> to TXD(N−1)<7:0>, along with control signals, TX_EN(0:N−1) to TX_ER(0:N−1), are encapsulated within a very short data frame. This frame contains X bytes including the start symbol, /S/, and is at least 8 bytes (the smallest frame supported by the 10GBASE-R PCS). The data frame is followed by an inter-frame gap (IFG) of at least 12 bytes including the terminate symbol, /T/. For N=2, the total encapsulation requires 20 bytes, the equivalent of 2 GMII or 5 XGMII transfers.
Optionally, the remaining bytes in the encapsulation frame may contain management data exchanges that allow for reading and writing of Clause 22 management registers.
At the PHY transmit input, the short frame is received and the N GMII octets are reconstructed (from UGMII rate) along with control signals and presented to the PHY as GMII signals.
In addition, the remaining information within the data frame may be discarded at step 612. During the idle periods, Idle columns, ∥I∥, may be deleted or inserted for clock rate matching purposes as defined in IEEE STD 802.3-2005, Carrier sense multiple access with collision detection (CSMA/CD) access method and physical layer specifications, December 2005.
On the receive side of the PHY the GMII rate adaptation and encapsulation operations are performed similar to those performed on the MAC TX path.
At the MAC receive input, the short encapsulation frame is received and the N GMII octets are reconstructed along with control signals and presented to the MAC as GMII signals. Additional information within the frame may be discarded. During the idle periods (interframe gap) Idle columns, ∥I∥, may be deleted or inserted for clock rate matching purposes as defined in IEEE STD 802.3-2005, Carrier sense multiple access with collision detection (CSMA/CD) access method and physical layer specifications, December 2005. In one embodiment Clause 45 management may occur across the XGMII.
When placed in Clause 45 management mode UGMII may be configured to allow the MAC/host to access management registers across the XGMII or extension. This mode may use ordered sets or short frames to encapsulate the host management interface (MDIO) between the host and PHY.
The UGMII has many advantages and distinctions over prior art system. On such distinction of the invention is that is enables a multi-rate interface between MAC and PHY for 10 G Ethernet capable PHYs.
In various embodiment this invention enables:
Additional advantages of further aspects of the invention are the access to Clause 22 and Clause 45 management registers in the PHY using the same 10 G interface. The separate signals MDC and MDIO, defined by IEEE 802.3, are not required.
One benefit of the invention described herein is the ability to interface a MAC to a 10/100/1000/10 G PHY (or any subset thereof) using a single 2 signal (4-pin) XFI interface, enabling greater port density per MAC IC. The XFI interface is shown in
Table of Contents
Introduction—Overview
Functional Specification
Modes of Operation
Data Upsampling
GMII Encapsulation
Clause 22 Management exchanges
Clause 45 Management
UGMII Mode Selection
Electrical Specification
Applicable Documents
Abbreviations
Disclosed herein is a new interface, the Ultra-serial Gigabit Media Independent Interface (UGMII), for the purpose of connecting a PHY operating in 10/100/1000 mode to a 10 G capable MAC using a 10 G defined interface. This interface may be XGMII (74 signals), XAUI (8 signals), or 10GBASE-R/XFI (2 signals).
UGMII is designed to support one or more of the following:
Encapsulation and transport of the 10/100/1000 MII/GMII interface across a 10 G interconnect.
Operation in full duplex and half duplex modes.
Auto-negotiation of operating mode: either XGMII at 10 Gbps or UGMII operating at 10, 100, or 1000 Mbps
Use of defined 10 G physical coding sublayers (PCS) without modification
An additional capability of UGMII may be access to Clause 22 and Clause 45 management registers in the PHY using the same 10 G interface. Thus, in one embodiment MDC and MDIO are not required.
One benefit of UGMII is the ability to interface MAC to a 10/100/1000/10 G PHY (or any subset thereof) using a single 2 signal (4-pin) XFI interface, which in turn enables greater port density per MAC IC.
Overview
UGMII defines a protocol for the transport of the GMII defined by Clause 35 of [1] over any Ethernet 10-gigabit interface. The purpose of UGMII is to allow a 10 G capable MAC to operate with a multi-rate 10 G PHY using minimal interconnect. For the greatest reduction in interconnect, it is expected that UGMII will be used with a 10GBASE-R PCS across an XFI (10 G serial interface) operating at 10.3125 Gbps, thus requiring only 2 differential pair signals on 4 pins.
An example system utilizing UGMII is shown in the figures provided herewith. UGMII is defined at the XGMII (if implemented) such that any defined extension such as XGXS/XAUI (Clause 48 of [1]), or PCS such as 10GBASE-R over XFI may be used as the interface between MAC and PHY.
In UGMII the defined signals for GMII may be fully encapsulated without modification, transported across the (extended) XGMII interface and reconstructed at the MAC RX and PHY TX inputs.
In the example embodiment described herein, selection of UGMII and XGMII operating modes are negotiated between the MAC and PHY using ordered sets to announce and acknowledgement a mode change.
Modes of Operation
A device incorporating UGMII may be configured to support the following interface modes:
Data encapsulation may be performed on the equivalent implementation of GMII for 10/100/1000 PHYs operating at 125 MHz. For PHYs operating lower than 1000 Mbps data upsampling can be performed prior to encapsulation.
For operation in 100 Mbps mode every octet in a frame can be repeated 10 times. For operation in 10 Mbps mode every octet in a frame can be repeated 100 times.
At the far end of the UGMII, after the GMII is reconstructed, the octects will be downsampled by 10 for 100 Mbps operation or by 100 for 10 Mbps operation.
GMII Encapsulation
GMII is defined in Clause 35 of [1]. The MAC transmits to the PHY using data signals TXD<7:0>, transmit enable TX_EN, and transmit error TX_ER. The PHY transmits to the MAC using data signals RXD<7:0>, data valid RX_DV, receive error RX_ER, and carrier sense CRS.
In this example embodiment, encapsulation is performed on a block of 2 GMII transfers (octets) and generates multiple XGMII transfers.
On the transmit side of the MAC two octets of data, TXD0<7:0> and TXD1<7:0>, along with control signals, TX_EN0-1 and TX_ER0-1, are encapsulated within a very short data frame. In this embodiment, this frame contains 8 bytes including the start symbol, /S/, and represents the smallest frame supported by the 10GBASE-R PCS. The data frame is followed by the minimum interframe gap (IFG) of 12 bytes including the terminate symbol, /T/. In this embodiment, the total encapsulation requires 20 bytes, the equivalent of 2 GMII or 5 XGMII transfers.
In one example embodiment, the first byte of the frame following /S/ contains the control signal information and can be constructed as: D1<7:0>=0 0 0 0 TX_ER1 TX_ER0TX_EN1 TX_EN0. The next two bytes may be configured to carry the data: D2<7:0>=TXD0<7:0> and D3<7:0>=TXD1<7:0>. Optionally, the remaining four bytes, D4 to D7, may contain management data exchanges that allow for reading and writing of Clause 22 management registers.
At the PHY transmit input, the short frame is received and the two GMII octets are reconstructed along with control signals and presented to the PHY as GMII signals. Additional information within the frame may be discarded. During the idle periods Idle columns, ∥I∥, may be deleted or inserted for clock rate matching purposes.
On the receive side of the PHY the process may be similar to the MAC TX encapsulation. Thus, two octets of data, RXD0<7:0> and RXD1<7:0>, along with control signals, RX_DV0-1 and RX_ER0-1, may be encapsulated within an 8-byte data frame. A minimum interframe gap follows the data frame.
In this example embodiment, the first byte of the frame following /S/ contains the control signal information and is constructed as: D1<7:0>=0 0 0 0 RX_ER1 RX_ER0 RX_DV1 RX_DV0. The next two bytes may carry the data: D2<7:0>=RXD0<7:0> and D3<7:0>=RXD1<7:0>. Optionally, the remaining four bytes, D4 to D7, may contain management data exchanges that allow for reading and writing of Clause 22 management registers.
At the MAC receive input, the short frame is received and the two GMII octets are reconstructed along with control signals and presented to the MAC as GMII signals. Additional information within the frame may be discarded. During the idle periods, Idle columns, ∥I∥, ma be deleted or inserted for clock rate matching purposes.
Table 1, which follows, illustrates one example encapsulation process.
Clause 22 Management Exchanges
It is further contemplated that the MAC/host may access Clause 22 management register through exchanges with the PHY using the last 4 bytes (or other bytes) of the GMII encapsulation frame. In other embodiments, Clause 22 Management Exchanges may occur in a different manner.
Clause 45 Management
It is further contemplated that when placed in Clause 45 management mode UGMII may allow the MAC/host to access management registers across the XGMII or extension.
In addition, various embodiments may use ordered sets or short frames as above. One potential downside to this arrangement is an inability to access management registers if UGMII is not operating or has a fault. It is contemplated that the UGMII system may be configured to access management registers even during data transmission.
UGMII Mode Selection
In the embodiment disclosed herein, the MAC selects UGMII modes of operation by sending ordered sets across XGMII or its extension.
To place the PHY in any (any particular) mode, the MAC may send an ordered set of /Q/ D1 D2 D3 with D1=96, D2=0 and D1D3 set according to Table 2. When theythe PHY receives one of these ordered sets, it replies with the same ordered set, but sets D2=D3 as an acknowledgement. Upon acknowledgement the MAC stops sending the ordered sets. After detecting the absence of the ordered set from the MAC, the PHY may be configured to stop sending ordered sets and begins operating in the new mode. The MAC then begins operation in the new mode and may begin transmitting data.
The following documents are incorporated by reference in their entirety herein.
IEEE STD 802.3-2005, Carrier sense multiple access with collision detection (CSMA/CD) access method and physical layer specifications, December 2005. This reference is hereby incorporated by referenced in its entirety.
IEEE Standard 802.3an-2006, Amendment 1: Physical Layer and Management Parameters for 10 Gb/s Operation—Type 10GBASET. This reference is hereby incorporated by referenced in its entirety.
The following abbreviations are used herein and defined as set forth below.
While various embodiments of the invention have been described, it will be apparent to those of ordinary skill in the art that many more embodiments and implementations are possible that are within the scope of this invention. In addition, the various features, elements, and embodiments described herein may be claimed or combined in any combination or arrangement.
This application claim priority to U.S. provisional patent application entitled Method And System For A Multi-rate Gigabit Media Independent Interface filed on Aug. 23, 2006 and assigned Ser. No. 60/839,986.
Number | Name | Date | Kind |
---|---|---|---|
4359778 | Lee | Nov 1982 | A |
4583235 | Dömer et al. | Apr 1986 | A |
4878232 | Fisher | Oct 1989 | A |
4956838 | Gilloire et al. | Sep 1990 | A |
5150381 | Forney et al. | Sep 1992 | A |
5222084 | Takahashi | Jun 1993 | A |
5249200 | Chen et al. | Sep 1993 | A |
5293402 | Crespo et al. | Mar 1994 | A |
5297170 | Eyuboglu et al. | Mar 1994 | A |
5301209 | Wei | Apr 1994 | A |
5305307 | Chu | Apr 1994 | A |
5307405 | Sih | Apr 1994 | A |
5388124 | Laroia et al. | Feb 1995 | A |
5633863 | Gysel et al. | May 1997 | A |
5646958 | Tsujimoto | Jul 1997 | A |
5822371 | Goldstein et al. | Oct 1998 | A |
5856970 | Gee et al. | Jan 1999 | A |
5862179 | Goldstein et al. | Jan 1999 | A |
5896452 | Yip et al. | Apr 1999 | A |
5909466 | Labat et al. | Jun 1999 | A |
6081502 | Paneth et al. | Jun 2000 | A |
6088827 | Rao | Jul 2000 | A |
6147979 | Michel et al. | Nov 2000 | A |
6160790 | Bremer | Dec 2000 | A |
6167082 | Ling et al. | Dec 2000 | A |
6201831 | Agazzi et al. | Mar 2001 | B1 |
6226332 | Agazzi et al. | May 2001 | B1 |
6249544 | Agazzi et al. | Jun 2001 | B1 |
6252904 | Agazzi et al. | Jun 2001 | B1 |
6253345 | Agazzi et al. | Jun 2001 | B1 |
6259729 | Seki | Jul 2001 | B1 |
6272173 | Hatamian | Aug 2001 | B1 |
6285653 | Koeman et al. | Sep 2001 | B1 |
6297647 | Kirk et al. | Oct 2001 | B2 |
6304598 | Agazzi et al. | Oct 2001 | B1 |
6351531 | Tahernezhaadi et al. | Feb 2002 | B1 |
6356555 | Rakib et al. | Mar 2002 | B1 |
6359893 | Mills | Mar 2002 | B1 |
6385208 | Findlater | May 2002 | B1 |
6433558 | Sciacero et al. | Aug 2002 | B1 |
6463041 | Agazzi | Oct 2002 | B1 |
6480477 | Treadaway et al. | Nov 2002 | B1 |
6480532 | Vareljian | Nov 2002 | B1 |
6493448 | Mann et al. | Dec 2002 | B1 |
6584160 | Amrany et al. | Jun 2003 | B1 |
6598203 | Tang | Jul 2003 | B1 |
6618480 | Polley et al. | Sep 2003 | B1 |
6665402 | Yue et al. | Dec 2003 | B1 |
6751255 | Reuven et al. | Jun 2004 | B1 |
6813311 | Pal et al. | Nov 2004 | B1 |
6823483 | Creigh | Nov 2004 | B1 |
6826226 | Sahlin et al. | Nov 2004 | B1 |
6862326 | Eran et al. | Mar 2005 | B1 |
6956847 | Heston et al. | Oct 2005 | B2 |
6961373 | Jones | Nov 2005 | B2 |
7031341 | Yu | Apr 2006 | B2 |
7065167 | McClellan | Jun 2006 | B2 |
7230957 | Kang | Jun 2007 | B2 |
7260120 | Kang | Aug 2007 | B2 |
7286622 | Sefidvash | Oct 2007 | B2 |
7305047 | Turner | Dec 2007 | B1 |
7343425 | Lo | Mar 2008 | B1 |
7356047 | Mahalawat | Apr 2008 | B1 |
7376146 | Beverly | May 2008 | B2 |
7593416 | Lo | Sep 2009 | B1 |
7599391 | Lo | Oct 2009 | B1 |
7606945 | Tatapudi | Oct 2009 | B2 |
7664134 | Caldwell | Feb 2010 | B2 |
7720068 | McClellan | May 2010 | B2 |
7751442 | Chang | Jul 2010 | B2 |
7760750 | Lo | Jul 2010 | B1 |
8018924 | Medina | Sep 2011 | B1 |
8259748 | Barrass | Sep 2012 | B2 |
8320400 | Lo | Nov 2012 | B1 |
8565261 | Lo | Oct 2013 | B1 |
8619571 | Sutardja | Dec 2013 | B1 |
8787402 | Barrass | Jul 2014 | B2 |
20010036160 | Curran et al. | Nov 2001 | A1 |
20020067824 | Wang | Jun 2002 | A1 |
20020106016 | Egelmeers et al. | Aug 2002 | A1 |
20020176492 | Zangi et al. | Nov 2002 | A1 |
20020191552 | Watkinson | Dec 2002 | A1 |
20030067888 | Bina et al. | Apr 2003 | A1 |
20030108092 | Gorecki et al. | Jun 2003 | A1 |
20030142659 | Lin et al. | Jul 2003 | A1 |
20030179710 | Huff | Sep 2003 | A1 |
20030214979 | Kang | Nov 2003 | A1 |
20040001540 | Jones | Jan 2004 | A1 |
20040022311 | Zerbe et al. | Feb 2004 | A1 |
20040028156 | Sefidvash | Feb 2004 | A1 |
20040090995 | Kang | May 2004 | A1 |
20040125487 | Sternad et al. | Jul 2004 | A9 |
20050063310 | Caldwell | Mar 2005 | A1 |
20050063413 | Caldwell | Mar 2005 | A1 |
20050196119 | Popovic | Sep 2005 | A1 |
20060153238 | Bar-On | Jul 2006 | A1 |
20070116023 | Tatapudi | May 2007 | A1 |
20070153781 | Tatapudi | Jul 2007 | A1 |
20070174434 | Tatapudi | Jul 2007 | A1 |
20070182489 | Fujimori | Aug 2007 | A1 |
20070258551 | Sefidvash | Nov 2007 | A1 |
20090154473 | Diab | Jun 2009 | A1 |
Entry |
---|
“Producing a Counter EMF”, http://www.tpub.com/neets/book2/5e.htm 3 pages, retrieved from the internet on Sep. 16, 2002. |
“Electric Machinery: Chap. 2 Transformers—Dot Convention to Denote the Polarity of a Transformer”,http://pemclab.cn.netu.edu.tw/W3elemac/W3slide/ch2.xformers/sld009.htm, 3 pages, Sep. 18, 1995. |
“Transformer Polarity”, Copyright 2002 Kilowatt Classroom, 11C, 3 page. |
“Technical Information—Use of Ferrites in Broadband Transformers”, Fair-Rite Products, Corp., 14th Edition, pp. 170-173, 2001. |
Huiting Chen, et al., “Current Mirror Circuit with Accurate Mirror Gain for Low β Transistors”, supported in part by Dallas Semiconductor Corp., 4 pages, May 2001. |
A. Herkersdorf, et al., “A Scalable SDH/SONET Framer Architecture for DATACOM and TELCO Applications”, IBM Research, Zurich Research Laboratory, Switzerland, 8 pages, Feb. 2000. |
Christopher T. DiMinico of Cable Design Technologies (CDT) Corporation, Massachusetts and Paul Kish of NORDX/CDT, Montreal, Canada, Development of Equal Level Far-End Crosstalk (ELFEXT) and Return Loss Specifications for Gigabit Ethernet Operation on Category 5 Copper Cabling, 10 pages, Oct. 5, 1998. |
Pedro Silva, et al., “Precoder Circuit for Channels with Multipath Dispersion”, Telecommunications Institute, Department of Electronics and Telecommunications Engineering, The University of Aveiro—University Campus, Portugal, 4 pages, 1999. |
“Tutorial: Adaptive Filter, Acoustic Echo Canceller and its Low Power Implementation”, © 2006 of Freehand Communication AB, 6 pages. |
J. Jezek, Institute of Information Theory and Automation, Prague, Czec Republic, et al., “New Algorithm for Spectral Factorization and its Practical Application”, pp. 1-6, May 18, 2001. |
“ADSL Tutorial” http://www.dslforum.org/aboutdsl/ads—1—tutorial.html, pp. 1-3, 2001. |
“Introduction to DSP”, http://www.bores.com/courses/intro/filters/4 fir.htm, pp. 1-2, Apr. 1998. |
“QAM VR-QAM Demodulator with Variable Rate”, DesignObjects™ by sci-worx, pp. 1-2, 2004. |
“Description of Algorithms (Part 1)” http://pwl.netcom.com/˜chip.f/viterbi/algrthms.html. pp. 1-7, retrieved from the internet on Mar. 5, 2002. |
“Fast Fourier Transform”, http://cas.ensmp.fr/˜.chaplais/Wavetour—presentation/transformees/Fourier/FFTUS.html, pp. 1-2, retrieved from the internet on Jul. 5, 2002. |
“Continuous Time Aperiodic Signals: The Fourier Transform”, http://ece.ucsd.edu/˜cruz/ecel01/notes/nodes32.html pp. 1-2, retrieved from the internet on Jul. 5, 2002. |
“Convolution”, http://www.wam.umd.edu/˜toh/spectrum/Convolution.html, pp. 1-2, retrieved from the internet on Jul. 5, 2002. |
“Convolution by DFT”, http://www.gresilog.com/english/excommend/doc/convtd.htm, pp. 1-3, retrieved from the internet on Jul. 5, 2002. |
IEEE Std 802.3-2002, “IEEE Standard for Information technology—Telecommunications and information exchange between systems—Local and metropolitan area networks—Specific requirements, Part 3: Carrier sense multiple access with collision detection (CSMA/CD) access method and physical layer specifications,” The Institute of Electrical and Electronics Engineers, Inc., Mar. 8, 2002. |
IEEE Std 802.3-2005, “IEEE Standard for Information technology—Telecommunications and information exchange between systems—Local and metropolitan area networks—Specific requirements, Part 3: Carrier sense multiple access with collision detection (CSMA/CD) access method and physical layer specifications,” The Institute of Electrical and Electronics Engineers, Inc., , Sections 1-5, Dec. 9, 2005. |
IEEE Draft P802.3ae/D5.0 Supplement to Carrier Sense Multiple Access with Collision Detection (CSMA/CD) Access Method & Physical Layer Specifications—Media Access Control (MAC) Parameters, Physical Layer, and Management Parameters for 10 Gb/s Operation The Institute of Electrical and Electronics Engineers, Inc., May 1, 2002. |
Oscar Agazzi, et al., “10Gb/s PMD Using PAM-5 Trellis Coded Modulation”, Broadcom, IEEE 802.3, Albuquerque, New Mexico, Mar. 6-10, 2000, 38 pages. |
Jamie E. Kardontchik, “4D Encoding in Level-One's Proposal for 1000BASE-T”, Advanced Micro Devices, Aug. 21, 1997—Rev. B, pp. 1-24. |
Erich F. Haratsch, et al., “A 1-Gb/s Joint Equalizer and Trellis Decoder for 1000BASE-T Gigabit Ethernet”, IEEE Journal of Solid-State Circuits, vol. 36, No. 3, Mar. 2001, pp. 374-384. |
Mehdi Hatamian, et al., “Design Considerations for Gigabit Ethernet 1000Base-T Twisted Pair Transceivers”, IEEE 1998 Custom Integrated Circuits Conference, pp. 335-342. |
Gottfried Ungerboeck, “Trellis-Coded Modulation with Redundant Signal Sets”, IEEE Communications Magazine, Feb. 1987, vol. 25, No. 2, pp. 5-21. |
“Gigabit Ethernet Over Category 5”, Copyright 2000-2001 Agilent Technologies, 12 pages. |
M. P. Sellers, et al., “Stabilized Precoder for Indoor Radio Communications”, IEEE Communications Letters, vol. 4, No. 10, Oct. 2000, pp. 315-317. |
“Wirescope 350—Understanding ELFEXT”, © 2000 Agilent Technologies, 2 pages. |
David Crawford, “Adaptive Filters”, © David Crawford 1996, pp. 1-5. |
David A. Johns, et al., “Integrated Circuits for Data Transmission Over Twisted-Pair Channels”, IEEE Journal of Solid-State Circuits, vol. 32, No. 3, Mar. 1997, pp. 398-406. |
Prof. David Johns, University of Toronto, “Equalization”, © D.A. Johns 1997, 29 pages. |
David Smalley, “Equalization Concepts: A Tutorial”, Atlanta Regional Technology Center Texas Instruments, Oct. 1994, pp. 1-29. |
Shao-Po Wu, et al., “FIR Filter Design via Spectral Factorization and Convex Optimization”, to appear as Chapter I of Applied Computational Control, Signal and Communications, Biswa Datta Editor, Birkhausesr, 1977, pp. 1-33. |
Richard D. Wesel, et al. “Achievable Rates for Tomlinson-Harashima Precoding”, IEEE Transactions on Information Theory, vol. 44, No. 2, Mar. 1998, pp. 824-831. |
Wolfgang H. Gerstacker, et al., “Blind Equalization Techniques for xDSL Using Channel Coding and Precoding”, submitted to AEU Int. J. Electr. Commun., May 1999, pp. 1-4. |
Chip Fleming, “A Tutorial on Convolutional Coding with Viterbi Decoding”, © 1999-2002, Spectrum Applications, pp. 1-6. |
“ELFEXT—Introduction”, Fluke Networks™, © 2000, pp. 1-2. |
“Iowegian's dspGuru FIR FAQ Part 2: Properties”, © 1999-2000 Iowegian International Corp., pp. 1-4. |
“28.5 Protocol Implementation Conformance Statement (PICS) Proforma for Clause 28, Physical Layer Link Signaling for 10 Mb/s, 100 Mb/s and 1000 Mb/s Auto-Negotiation on Twisted Pair”, IEEE Std. 802.3, 1998 Edition, pp. 6-14 and 18-44. |
M. Tomlinson, “New Automatic Equalizer Employing Modulo Arithmetic”, Electronic Letters, vol. 7, 1971, pp. 138-139. |
H. Harashima and M. Miyakawa, “Matched-Transmission Technique for Channels with Intersymbol Interference”, IEEE Transactions on Communications, vol. COM-20, No. 4, Aug. 1972, pp. 774-780. |
P. Kabal and S. Pasupathy, “Partial-Response Signaling”, IEEE Transactions on Communications, vol. COM-23, No. 9, Sep. 1975, pp. 921-934. |
R.F.H. Fischer, W.H. Gerstacker, and J.B. Huber, “Dynamics Limited Precoding, Shaping, and Blind Equalization for Fast Digital Transmission Over Twisted Pair Lines”, IEEE Journal on Selected Areas in Communications, vol. 13, No. 9, Dec. 1995, pp. 1622-1633. |
R.F.H. Fischer and J.B. Huber, “Comparison of Precoding Schemes for Digital Subscriber Lines”, IEEE Transactions on Communications, vol. 45, No. 3, Mar. 1997, pp. 334-343. |
“Definition of Minimum Phase,” www-ccrma.stanford.edu/˜jos/filters/Definition—Minimum—Phase.html, 3 pages. |
Li, M.; Wang, S., T. Tao, T. Kawasniewski, “FIR Filter Optimization as Pre-emphasis of High Speed Backplane Data Transmission”, Electronics Letters, Jul. 8, 2004, vol. 40, No. 14, 2 pages. |
“Electric Machinery: Chap. 2 Transformers—Dot Convention to Denote the Polarity of a Transformer”, http://pemclab.cn.netu.edu.tw/W3elemac/W3slide/ch2.xformers/sld009.htm, 4 pages (date unknown). |
“Transformer Polarity”, Copyright 2002 Kilowatt Classroom, 11C, 4 page (date unknown). |
“Application Note—Design of H.F. Wideband Power Transformers; Part II—EC07213”, Philips Semiconductors, Mar. 23, 1998, pp. 1-10. |
Shao-Po Wu, et al., “FIR Filter Design via Semidefinite Programming and Spectral Factorization”, Information Systems Laboratory, Stanford University, CA, 6 pages (date unknown). |
Number | Date | Country | |
---|---|---|---|
60839986 | Aug 2006 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 11894990 | Aug 2007 | US |
Child | 13474515 | US |