TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Dual XAUI Transceiver • • • • • • • • (1) XGMII Extender Sublayer APPLICATIONS 10 Gigabit Ethernet Servers and Routers TLK3138 TCLK(0) TD(31..0) TDP/N[3:0]0 4 TC(3..0) RCLK(0) 4 RDP/N[3:0]0 RD(31..0) XAUI A/O • RC(3..0) TCLK(0) 4 TDP/N[3:0]1 TD(31..0) TC(3..0) RCLK(0) B/1 • • 4 RDP/N[3:0]1 XAUI • • A/O • • • • ±200 ppm Clock Tolerance in XAUI Transmit and Receive Datapaths Diagnostic LED Output Interface 130-nm CMOS Technology Package: PBGA 484, 23×23mm, 484 ball, 1mm Pitch 2.42 W Typical Power Dissipation (Dual XAUI Mode, Input HSTL Termination Disabled) Commercial Ambient Operating Temp (0°C to 70°C) XGMII • • B/1 • Eight Channel 3.125 Gbps Transceiver Contains Two Complete IEEE P802.3ae-2002 10 Gbps Ethernet XGXS(1) (XAUI) Compliant Cores XAUI (Serial side) Channel Redundancy Mode Support: Fast Switching from Primary to Secondary XAUI/Channel with Provisionable Error Character or Local Fault Code Insertion at Switch Time XGMII (Parallel side) Channel Redundancy Mode Support: Fast Switching from Primary to Secondary XGMII/Channel with Provisionable Error Character or Local Fault Code Insertion at Switch Time Serial Side Transmit Pre-emphasis and Receive Adaptive Equalization to Allow Extended Backplane Reach Selectable Full Duplex Serial Side Retimer Mode Full Datapath Loopback Capability (Serial/Parallel Side) Support PRBS 27-1 and 223-1 Generate/Verify. Support Standard Defined CJPAT, CRPAT, High Frequency, Low Frequency, and Mixed Frequency Testing XGMII: HSTL Class 1 I/O with On-chip 50-Ω Termination on Inputs/Outputs (1.5-V Power Supply) XGMII: Support Receive (Output) Aligned/Centered Timing Modes, and Transmit (Input) Source Centered Mode Supports Jumbo Packet (9600 byte maximum) Operation Align Character Skew Support of 30 bit Times at Chip Pins Dual MDIO: IEEE 802.3ae Clause 45 Compliant Management Data Input / Output Interface (1.2-V and 2.5-V I/O) 1.2-V Core, 1.5-V HSTL I/O Supply, and 2.5-V LVCMOS and Bias Supply JTAG: IEEE 1149.1 Test Interface XGMII FEATURES • • RD(31..0) RC(3..0) Please be aware that an important notice concerning availability, standard warranty, and use in critical applications of Texas Instruments semiconductor products and disclaimers thereto appears at the end of this data sheet. PRODUCTION DATA information is current as of publication date. Products conform to specifications per the terms of the Texas Instruments standard warranty. Production processing does not necessarily include testing of all parameters. Copyright © 2007, Texas Instruments Incorporated TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 These devices have limited built-in ESD protection. The leads should be shorted together or the device placed in conductive foam during storage or handling to prevent electrostatic damage to the MOS gates. DESCRIPTION The TLK3138 is an eight channel serial transceiver. It is compliant with the 10Gbps Ethernet XAUI specification. The TLK3138 provides 10 Gbps high-speed bi-directional point-to-point data transmission. The primary application of this device is in backplanes and front panel connections requiring dual/redundant 10Gbps connections over controlled impedance media of approximately 50Ω. The transmission media can be printed circuit board (PCB) traces, copper cables or fiber-optical media. The ultimate rate and distance of data transfer is dependent upon the attenuation characteristics of the media and the noise coupling into the lines. The TLK3138 performs the parallel-to-serial, serial-to-parallel conversion, and clock extraction functions for a physical layer interface. The TLK3138 provides two complete XGXS/PCS functions defined in Clause 47/48 of the IEEE P802.3ae 10Gbps Ethernet standard. The serial transmitter is implemented using differential Current Mode Logic (CML) with integrated termination resistors. The TLK3138 can be configured as a dual XAUI transceiver. TLK3138 supports two 32-bit data path, 4-bit control, 10 Gigabit Media Independent Interfaces (XGMII) to the protocol device(s). Figure 1 shows an example system block diagram for TLK3138 used to provide the 10Gbps Ethernet Physical Coding Sublayer to Coarse Wave-length Division Multiplexed optical transceiver or parallel optics. Line Card TX TCLK 4 4 XGMII CH A 4 TD(31:0) A CH A 4 XAUI TC(3:0) MAC/ Packet Processor TLK3138 XAUI CWDM or Parallel Optics TLK3138 4 RC(3:0) XGMII XAUI Backplane RX RCLK A RD(31:0) XGMII TX 4 4 XAUI TCLK TD(31:0) B CH B 4 CH B TC(3:0) MAC/ Packet Processor XAUI XGMII RX B RC(3:0) RCLK RD(31:0) Figure 1. System Block Diagram – PCS 2 Submit Documentation Feedback CWDM or Parallel Optics TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Figure 2 shows an example system block diagram for TLK3138 used to provide the system backplane interconnect. Line Card 4 4 RCLK RDP/N[3:0]0 RD(31:0) RC(0:3) RC(3:0) RDP/N[3:0]1 TLK3138 XAUI CH A 4 System Backplane 4 4 4 MAC/ PACKET PROCESSOR PHY/ OPTICS FRAMER/ PCS TDP/N[3:0]0 TCLK TDP/N[3:0]1 TD(31:0) TC(3:0) RCLK RDP/N[3:0]0 RD(31:0) SWITCH FABRIC RDP/N[3:0]1 RC(3:0) TLK3138 XAUI CH A 4 4 TDP/N[3:0]0 TCLK TDP/N[3:0]1 TD(31:0) TC(3:0) Figure 2. System Block Diagram – XAUI Backplane Figure 3 shows the TLK3138 in a Dual XAUI Application: 36 36 36 36 X G M I I A X G M I I B TLK3138 XGXS CORE A XGXS CORE B X A U I A X A U I B 4 4 4 4 Figure 3. DUAL XAUI Application Submit Documentation Feedback 3 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Figure 4 shows the TLK3138 in a XAUI Retimer Application TLK3138 X G M I I A XGXS CORE A X G M I I B XGXS CORE B 4 X A U I A 4 4 X A U I B 4 Figure 4. XAUI Retimer Application Figure 5 shows the TLK3138 in a Redundant XAUI Application. 36 36 X G M I I A TLK3138 XGXS CORE A XGXS CORE B Figure 5. Redundant XAUI Application 4 Submit Documentation Feedback X A U I A X A U I B 4 4 4 4 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Figure 6 shows the TLK3138 in a Redundant XAUI Application. 36 36 36 36 TLK3138 X G M I I A XGXS CORE A X G M I I B XGXS CORE B X A U I A 4 4 Figure 6. Redundant XGMII Application Figure 7 shows the TLK3138 in a XAUI Loopback Application. 36 36 X G M I I A X G M I I B TLK3138 XGXS CORE A XGXS CORE B X A U I A X A U I B 4 4 4 4 Figure 7. XAUI Loopback Application Submit Documentation Feedback 5 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Figure 8 shows the TLK3138 in a XGMII Loopback application. 36 36 X G M I I A X G M I I B TLK3138 XGXS CORE A X A U I A XGXS CORE B X A U I B 4 4 Figure 8. XGMII Loopback Application It is possible to select non-conflicting modes of operation based on the following bits: XAUI_RETIME (See Register Bit 4/5. 32907.2 for definition) REDUNDANT_XAUI (See Register Bit 4/5.32907.6 for definition) REDUNDANT_XGMII (See Register Bit 4/5.32907.7 for definition) XGMII_LOOPBACK (See Register Bit 4/5.32792. 0 for definition (1 bit for each XGXS Core)) XAUI_LOOPBACK (See Register Bit 4/5.32792. 1 for definition (1 bit for each XGXS Core)) The TLK3138 supports the IEEE 802.3 defined Management Data Input/Output (MDIO) Interface to allow ease in configuration and status monitoring of the link. The bi-directional data pin (MDIO) should be externally pulled up to 1.2V or 2.5V per the standard for both MDIO0 and MDIO1. The TLK3138 supports the IEEE 1149.1 defined JTAG test port for ease in board manufacturing test. It also supports a comprehensive series of built-in tests for self-test purposes including PRBS generation and verification, CRPAT, CJPAT, Mixed/High/Low Frequency testing. The TLK3138 operates with a 1.2V core voltage supply, a 1.5V HSTL I/O voltage supply and a 2.5V bias supply. The TLK3138 is packaged in a 23×23mm, 484-ball, 1mm ball pitch Plastic Ball Grid Array (PBGA) package and is characterized for operation from 0°C to 70°C Ambient, 115°C Junction, and 5% power supply variation unless noted otherwise. Note that the junction temperature must be kept below 115°C to meet TI C035 process reliability. 6 Submit Documentation Feedback TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 BLOCK DIAGRAMS Figure 9 provides a high level description of the TLK3138 block diagram. TCLK(3:0) XGMII TXA TXD(31:0) XAUI A Core TXC(3:0) 4 RDP0[3:0] RDN0[3:0] RCLK(3:0) RXD(31:0) 4 TDP0[3:0] TDN0[3:0] XGMII RXA Datapath Connection Block RXC(3:0) XAUI B Core TCLK(3:0) TXD(31:0) XGMII TXB 4 TDP1[3:0] TDN1[3:0] 4 RDP1[3:0] RDN1[3:0] TXC(3:0) TCK RCLK(3:0) RXD(31:0) TDI XGMII RXB JTAG TMS TRSTN RXC(3:0) TDO PRTAD[4:0]/ DVAD[4:0] MDIO0/1 MDIO MDC0/1 Figure 9. TLK3138 Block Diagram Submit Documentation Feedback 7 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Figure 10 is a more detailed block diagram description of XAUI core. Test mode Test mode Test mode Test mode Channel 3 Channel 2 Channel 1 Channel 0 Self Test TCLK TXDP TXCP TXDN TX FIFO/ 8 Bit CTC TXD[7:0] 8b/ 10b enc TXCN TXBP 10-bits TDP0 TXBN TDN0 RXDP RCLK RCLK 8-bits RXD(7:0) 8b/10b Decoding and Self Test Verification/ Reporting RXCP RXDN RXCN RXBP RX FIFO/ CTC 10-bits SERDES Core RCLK REFCLKP REFCLKN Figure 10. Detailed XAUI Core Block Diagram 8 Submit Documentation Feedback RDP0 RXBN RDN0 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Parallel to Serial Parallel Data In REFCLKP /N Multiplying Clock Synthesizer D Q TDP TDN Baud Clock Interpolator and Clock Recovery Recovered Clock RCLK Parallel Data Out Serial to Parallel and Comma Detect RDP RDN Figure 11. Block Diagram of SERDES Core DETAILED DESCRIPTION The TLK3138 has several operational interface modes controlled by register bits A/B and XAUI_RETIME, REDUNDANT_XAUI, and REDUNDANT_XGMII. The major modes of application are discussed below. REDUNDANT XAUI OPERATION The TLK3138 can operate as a redundant XAUI transceiver. In Redundant XAUI MODE (4/5.32907.6), only the A side XGMII interface is active, and both serial interfaces XAUI A and XAUI B are active. It is possible for B side XGMII to receive B side serial data, but the transmit XGMII B interface is ignored. The transmit datapath either duplicates the transmit XGMII data to both serial side data streams, or if the IDLE mode is selected, send only valid A/K/R (idle) sequences on the deselected side. The receive datapath selects one of the two aggregated input serial streams controlled by the A_B (4/5.32907.3) register bit. While communication is occurring on the primary selected channel, the secondary channel is fully functional capable of transmitting and receiving data. All registers are accessible and valid. The only difference between the primary and secondary channels is the primary channel is routed to the A side XGMII receive bus. A completely active secondary channel allows transition from primary to secondary channels within a few clock cycles. During the transition, the data on each byte of the bus is 0xFE (code violation), which is the ERROR indication, or local fault indication (based on a provisioned register value). Submit Documentation Feedback 9 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 DETAILED DESCRIPTION (continued) REDUNDANT XGMII OPERATION The TLK3138 can operate as a redundant XGMII transceiver. In Redundant XGMII MODE (4/5.32907.7), only the A side Serial interface is active, and both Parallel interfaces XGMII A and XGMII B are active. It is possible for B side XAUI to transmit B side parallel data, but the receive XAUI B interface is processed but not used. The receive datapath either duplicates the receive XAUI data to both parallel side data streams, or if the IDLE mode is selected, send only valid (idle) sequences on the deselected side. The transmit datapath selects one of the two aggregated input parallel streams controlled by the A_B (4/5.32907.3) register bit. While communication is occurring on the primary selected channel, the secondary channel is fully functional capable of transmitting data (if desired). All B side transmit related registers are accessible and valid. During the transition, the data on each byte of the bus is 0xFE (code violation), which is the ERROR indication, or local fault indication (based on a provisioned register value). RETIMER OPERATION The TLK3138 can operate as a full duplex serial side re-timer. All the functions of transceiver operations are performed except for the XGMII input interfaces, and optionally the XGMII output interfaces. The recovered data on each XAUI channel is de-serialized, de-skewed, aligned to the reference clock, and re-serialized. In the re-timer mode inputs from the XGMII inputs are ignored. XGMII outputs can provisionally be left on to snoop received data, or turned off to save power. Note that when RETIM is high, the serial side A receive data is routed out to the serial side B transmit serial lines. Similarly, the serial side B receive data is routed out to the serial side A transmit serial lines. If A/B is toggled when in re-timer monitor mode, the data on each byte of the XGMII receive output bus (if not in 3-state) is 0xFE (code violation) for several XGMII clock cycles, or local fault (based on the provisioned register value). PARALLEL CLOCKING MODES The TLK3138 supports source-centered timing and source-aligned timing on the XGMII receive output bus. The source-centered timing supported is the timing defined in P802.3ae Clause 46, with the RX_CLK centered within the receive output data bit timing, as shown in Figure 12. Also shown is source-aligned timing. RX_CLK tSETUP tSETUP Source Centered (DDR) tHOLD tHOLD RXD(31:0) Data Data RXC(3:0) Source Aligned (DDR) RXD(31:0) RXC(3:0) Data Data Data Figure 12. Receive Interface Timing – Source Centered/Aligned On the transmit data path, the data is sampled on the rising edge and falling edge of TCLK as shown in Figure 13. 10 Submit Documentation Feedback TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 DETAILED DESCRIPTION (continued) In the receive datapath A FIFO, placed on the output of the serial to parallel conversion logic for each serial link, compensates for channel skew, clock phase and frequency tolerance differences between the recovered clocks for each serial links and the receive output clock, RX_CLK. This FIFO has a total depth of sixteen ten bit entries, giving 30 bit time deskew (channel to channel skew) alignment capability in standard mode. See HSTL Output Switching Characteristics (DDR) table in the Electrical Characteristics section for more details on XGMII timing. TCLK tSETUP tSETUP tHOLD Source Centered (DDR) TXD(31:0) Data tHOLD Data TXC(3:0) Figure 13. Transmit Interface Timing PARALLEL INTERFACE DATA Data placed on the XGMII transmit input bus is latched and then phase aligned to the internal version of the transmit reference clock, 8b/10b encoded, serialized, then transmitted sequentially beginning with the LSB of the encoded data byte over the differential high speed serial transmit pins. The XGMII receive data bus outputs four bytes on RXD(31:0). Control character (K-characters) reporting for each byte is done by asserting the corresponding control pin, RXC(3:0). When RXC is asserted, the 8 bits of data corresponding to the control pin is to be interpreted as a K-character. If an error is uncovered in decoding the data, the control pin is asserted and 0xFE is output for the corresponding byte. TRANSMISSION LATENCY For each channel, the data transmission latency of the TLK3138 is defined as the delay from the rising or falling edge of the selected transmit clock when valid data is on the transmit data pins to the serial transmission of bit 0, as shown in Figure 14. The maximum transmit latency (TLATENCY) is 650 bit times; the standard allows a combined latency (TX + RX) of 2048 bit times. 10 bit Code Transmitted TXxP TXxN TLATENCY TXD[0-31] Bytes to be Transmitted TCLK Figure 14. Transmission Latency Submit Documentation Feedback 11 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 DETAILED DESCRIPTION (continued) CHANNEL CLOCK TO SERIAL TRANSMIT CLOCK SYNCHRONIZATION In XAUI mode, the TLK3138 allows ±200 ppm difference between the serdes transmit reference on the XAUI side, versus the input TCLK on the XGMII side. There exists a FIFO capable of CTC operations, and has a depth of 32 locations (32 bits wide per location). The reference clock and the transmit data clock(s) may be from a common source, but the design allows for up to ±200 ppm of frequency difference should the application require it. DATA RECEPTION LATENCY For each serial link, the serial-to-parallel data latency is the time from when the first bit arrives at the serial receiver input until it is output in the aligned parallel word on the XGMII, as shown in Figure 15. The maximum receive latency (RLATENCY) is 950 bit times using standard alignment mode; the standard allows a combined latency (TX + RX) of 2048 bit times. 10-bit Code Received RDP,RDN RLATENCY Bytes Received RXD[31:0] RCLK Figure 15. Receiver Latency 8B/10B ENCODER All true serial interfaces require a method of encoding to insure sufficient transition density for the receiving PLL to acquire and maintain lock. The encoding scheme also maintains the signal DC balance by keeping the number of ones and zeros are balanced which allows for AC coupled data transmission. The TLK3138 uses the 8B/10B encoding algorithm that is used by 10Gbps and 1Gbps Ethernet and Fibre Channel standards. This provides good transition density for clock recovery and improves error checking. The 8B/10B encoder/decoder function is enabled for all serial links. The TLK3138 internally encodes and decodes the data such that the user reads and writes actual 8-bit data on each channel. The encoder and decoder functions can optionally be enabled or disabled on a per channel basis. The 8B/10B encoder converts 8-bit wide data to a 10-bit wide encoded data character to improve its transition density. This transmission code includes D Characters, used for transmitting data, and K Characters, used for transmitting protocol information. Each K or D character code word can also have both a positive and a negative disparity version. The disparity of a code word is selected by the encoder to balance the running disparity of the serialized data stream. The generation of K-characters to be transmitted on each channel is controlled by transmit control pins, TXC(3:0). When the control pin is asserted along with the 8 bits of data, an 8B/10B K-character is transmitted. Similarly, reception of K-characters is reported by the receive control pins, RXC(3:0). When receive control pin is asserted, the corresponding byte on the receive data bus should be interpreted as a K-character. The TLK3138 transmits and receives all of the twelve valid K-characters as defined in Table 1. 12 Submit Documentation Feedback TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 DETAILED DESCRIPTION (continued) Table 1. Valid K-Codes K-CODE TXC(3:0) or RXC(3:0) DATA BUS BYTES (RXD[x: x-7] or TXD[x: x-7]) 00 through FF 0 K28.0 1 K28.1 ENCODED K-CODE NEGATIVE RUNNING DISPARITY POSITIVE RUNNING DISPARITY DDD DDDDD dddddd dddd dddddd dddd Normal data 000 11100 001111 0100 110000 1011 IdleO/busy 1 001 11100 001111 1001 110000 0110 IdleE/busy K28.2 1 010 11100 001111 0101 110000 1010 K28.3 1 011 11100 001111 0011 110000 1100 K28.4 1 100 11100 001111 0010 110000 1101 K28.5 1 101 11100 001111 1010 110000 0101 K28.6 1 110 11100 001111 0110 110000 1001 K28.7 1 111 11100 001111 1000 110000 0111 Code Violation or Parity Error K23.7 1 111 10111 111010 1000 000101 0111 IdleO/not-busy K27.7 1 111 11011 110110 1000 001001 0111 SOP(S) K29.7 1 111 11101 101110 1000 010001 0111 EOP(T) K30.7 1 111 11110 011110 100 100001 0111 K-code Description Channel Alignment Pre-curser IdleE/not-busy Table 2 provides additional transmit data control coding and descriptions that have been incorporated into 10 Gigabits per second Ethernet. Data patterns put on XGMII transmit data bus other than those defined in Table 2 when the transmit control pin is asserted results in an invalid K-character being transmitted which results in an code error at the receiver. Table 2. Valid XGMII Channel Encodings DATA BUS (TXD[x: x-7] or RXD[x: x-7]) TXC(3:0) or RXC(3:0) DESCRIPTION 00 through FF 0 Normal Data Transmission 00 through 06 1 Reserved 07 1 Idle 08 through 9B 1 Reserved 9C 1 Sequence (only valid in Channel A) 9D through FA 1 Reserved FB 1 Start (only valid in Channel A) FC 1 Reserved FD 1 Terminate FE 1 Transmit error propagation FF 1 Reserved COMMA DETECT AND 8B/10B DECODING When parallel data is clocked into a parallel to serial converter, the byte boundary that was associated with the parallel data is lost in the serialization of the data. When the serial data is received and converted to parallel format again, a method is needed to be able to recognize the byte boundary again. Generally this is accomplished through the use of a synchronization pattern. This is a unique a pattern of 1s and 0s that either cannot occur as part of valid data or is a pattern that repeats at defined intervals. 8B/10B encoding contains a character called the comma (b’0011111’ or b’1100000’) which is used by the comma detect circuit to align the received serial data back to its original byte boundary. The channel synchronization block detects the K28.5 comma, generating a synchronization signal aligning the data to their 10-bit boundaries for decoding. It then converts the data back into 8-bit data. It is important to note that the comma can be either a (b’0011111’) or the inverse (b’1100000’) depending on the running disparity. The TLK3138 decoder detects both patterns. Submit Documentation Feedback 13 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 The reception of K-characters is reported by the assertion of receive control pin, RXC(3:0) for the corresponding byte on the XGMII receive bus. When a code word error or running disparity error is detected in the decoded data received on a serial link, the receive control pin is asserted and an 0xFE is placed on the receive data bus for that channel, as shown in Table 3. Table 3. Receive Data Controls EVENT RECEIVE DATA BUS RXD[x: 7-x] RXC(3:0) Normal Data Normal K-character XX 0 Valid K-code 1 FE 1 Code word error or running disparity error Channel Initialization and Synchronization The TLK3138 has a synchronization state machine which is responsible for handling link initialization and synchronization for each channel. The initialization and synchronization state diagram is provided in Figure 16. The status of any channel can be monitored by reading MDIO register 4/5.24.3:0. Valid Data or Error reset UNSYNC comma Valid Data Error ACQ1 MISS3 4 Consecutive Valid Data Words Error Valid Data comma Error Error ACQ2 MISS2 Error comma Error ACQ3 Valid Data 4 Consecutive Valid Data Words MISS1 Error comma 4 Consecutive Valid Data Words SYNC Valid Data or Comma Figure 16. Channel Synchronization State Machine CHANNEL STATE DESCRIPTIONS UNSYNC – This is the initial state for each channel upon device power up or reset. In this state, the TLK3138 has the comma detect circuit active and makes code word alignment adjustments based on the position of a comma in the incoming data stream. While in this state the TLK3138 sets the Lane Sync bit to '0' for the particular channel in MDIO register bits 4/5.24.3:0, indicating the lane is not synchronized. (1) The channel state transitions to the ACQ1 state upon the detection of a comma. (1) 14 The Lane Sync Bit = '0' bit from any/or all channels causes a local fault to be output on the receive data bus. Submit Documentation Feedback TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 ACQ1 – During this state the comma detect circuit is active but code word re-alignment is disabled. The TLK3138 remains in this state until either a comma is detected in the same code word alignment position as found in state UNSYNC or a decode error is encountered. While in this state, the Lane Sync bit for the particular channel remains de-asserted, indicating the lane is not synchronized. (2) A decode or running disparity error returns the channel state to UNSYNC. A detected comma causes the channel state to transition to ACQ2. ACQ2 – During this state, the comma detect circuit is active but code word re-alignment is disabled. The TLK3138 remains in this state until either a comma is detected in the same code word alignment position as found in state UNSYNC or a decode error is encountered. While in this state, the Lane Sync bit for the particular channel remains de-asserted, indicating the lane is not synchronized.(2) A decode or running disparity error returns the channel state to UNSYNC. A detected comma causes the channel state to transition to ACQ3. ACQ3 –During this state the comma detect circuit is active but code word re-alignment is disabled. The TLK3138 remains in this state until either a comma is detected or a decode error encountered. While in this state, the Lane Sync bit for the particular channel remains de-asserted, indicating the lane is not synchronized.(2) A decode or running disparity error returns the channel state to UNSYNC. A detected comma causes the channel state to transition to SYNC. SYNC –This is the normal state for receiving data. When in this state, the TLK3138 sets the Lane Sync bit to '1' for the particular channel in the MDIO register bits 4/5.24.3:0 indicating the lane has been synchronized. During this state the comma detect circuit is active but code word re-alignment is disabled. A decode or running disparity error causes the channel state to transition to MISS1. MISS1 – When entering this state an internal error counter is cleared. If the next four consecutive codes are decoded without error, the channel state reverts back to SYNC. If a decode or running disparity error is detected, the channel state transitions to MISS2. MISS2 – When entering this state an internal error counter is cleared. If the next four consecutive codes are decoded without error, the channel state reverts back to MISS1. If a decode or running disparity error is detected, the channel state transitions to MISS3. MISS3 –When entering this state an internal error counter is cleared. If the next four consecutive codes are decoded without error, the channel state reverts back to MISS1. If a decode or running disparity error is detected, the channel state transitions to UNSYNC. END OF PACKET ERROR DETECTION Because of their unique data patterns, /A/ (K28.3), /K/ (K28.5), and /T/ (K29.7) catch running disparity errors that may have propagated undetected from previous codes in a packet. Running disparity errors detected by these control codes at the end of packets cause the previous data codes to be reported as errors (0xFE) to allow the protocol device to reject the packet (see Figure 17). XAUI RDP/N0 ... D D D D A RDP/N1 ... D D D T A RDP/N2 ... D D D K A RDP/N3 ... D D D K A XGMII Running Disparity Error Detected by /A/ Yields Running Disparity Error Detected by /T/ Yields Running Disparity Error Detected by /K/ Yields Running Disparity Error Detected by /K/ Yields ... D D D E I ... D D E T I ... D D E I I ... D D E I I D = Data, T = K29.7, A = K28.3, K = K28.5, E = Error (0xFE), I = Idle Figure 17. End of Packet Error Detection (2) The Lane Sync bit = '0' causes a local fault to be output on the receive data bus. Submit Documentation Feedback 15 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 FAULT DETECTION AND REPORTING The TLK3138 detects and reports local faults as well as forward both local and remote faults as defined in the IEEE P802.3ae 10Gbps Ethernet Standard to aid in fault diagnosis. All faults detected by the TLK3138 are reported as local faults to the upper layer protocols. Once a local fault is detected in the TLK3138, MDIO register bit 4/5.1.7 is set. Fault sequences, sequence ordered sets received by the TLK3138, either on the Transmit Data Bus or on the high speed receiver pins, are forwarded without change to the MDIO registers in the TLK3138. Also, note that the TLK3138 is capable of performing CTC operation where only RF and LF or any Q sequences are transported (not generated) in either the transmit or receive direction in XAUI mode. TLK3138 reports a fault by outputting a K28.4 (0x9C) on RXD(7:0), 0x00 on RXD(15:8) and RXD(23:8) and 0x01 for local faults on RXD(31:24). Forwarding of remote faults is handled as a normal transmission. Note that the TLK3138 does not generate a remote fault indication nor any other type of Q. RECEIVE SYNCHRONIZATION AND SKEW COMPENSATION In XAUI mode, the TLK3138 has a FIFO enabled on the receive data path coming from each serial link to compensate for channel skew and clock phase and frequency tolerance differences between the recovered clocks for each channel and the receive output clock RX_CLK. This FIFO has a depth of 32 locations (32 bits wide for each location). The de-skew of the 4 serial links that make up each XAUI channel into a single 32 bit wide column of data is accomplished by alignment of the receive FIFOs on each serial link to a K28.3 control code sent during the inter-packet gap (IPG) between data packets or during initial link synchronization. The K28.3 code (referred to as the A or alignment code) is transmitted on the first column following the end of the data packet as shown in Table 2. The column de-skew state machine is provided in Figure 18. The status of column alignment can be monitored by reading MDIO registers 4/5.24.12 for global alignment. RESET or LOS Valid Data UNALIGN Deskew Error FAIL3 Deskew Error Valid Data FAIL2 ||A|| ||A|| Deskew Deskew Error Error DET1 Valid Data ||A|| FAIL1 Deskew Error ||A|| DET3 ||A|| Valid Data Note: Deskew Error is at least one column containing an A character, but not all four simultaneously. Figure 18. Column De-Skew State Machine 16 Valid Data ||A|| Deskew Error DET2 Valid Data Deskew Error Submit Documentation Feedback ALIGN Valid Data ||A|| TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 COLUMN STATE DESCRIPTIONS UNALIGN – This is the initial state for the column state machine upon device power up or reset. If any of the channel state machines are set to UNSYNC, the column state is set to UNALIGN. In this state, the column state machine searches for alignment character codes (K28.3 or /A/) on each channel and align the FIFO pointers on each channel to the /A/ character code. While in this state, the Column Alignment Sync bit is set to '0' in MDIO registers 4/5.24.12, indicating the column is not aligned. (3) The column state transitions to the DET1 state upon the detection and alignment of /A/ character codes in all four channels. DET1 – During this state, the alignment character code detect circuit is active on each channel but the column re-alignment is disabled. The column state machine remains in this state looking for a column of alignment character codes. If an incomplete alignment column is detected (alignment character codes not found on all channels) or a deskew error is detected, the column state machine transitions to state UNALIGN. While in this state, the Column Alignment Sync bit is set to '0' in MDIO registers 4/5.24.12 indicating the column is not aligned. (4) Detection of a complete alignment column causes the column state machine to transition to state DET2. DET2 – During this state, the alignment character code detect circuit is active on each channel but the column re-alignment is disabled. The column state machine remains in this state looking for a column of alignment character codes. If an incomplete alignment column is detected (alignment character codes not found on all channels) or a deskew error is detected, the column state machine transitions to state UNALIGN. While in this state, the Column Alignment Sync bit is set to '0' in MDIO registers 4/5.24.12 indicating the column is not aligned.(4) Detection of a complete alignment column causes the column state machine to transition to state DET3. DET3 –During this state, the alignment character code detect circuit is active on each channel but the column re-alignment is disabled. The column state machine remains in this state looking for a column of alignment character codes. If an incomplete alignment column is detected (alignment character codes not found on all channels) or a deskew error is detected, the column state machine transitions to state UNALIGN. While in this state, the Column Alignment Sync bit is set to '0' in MDIO registers 4/5.24.12 indicating the column is not aligned.(4) Detection of a complete alignment column causes the column state machine to transition to state ALIGN. ALIGN – This is the normal state for receiving data. When in this state, the column state machine sets the Column Alignment Sync bit to '1' in MDIO registers 4/5.24.12 indicating all channels are aligned. During this state the alignment character code detect circuit is active on each channel but the column re-alignment is disabled. If a deskew error is detected in the correct position within the Inter-Packet Gap, the column state machine transitions to state FAIL1. FAIL1 – When in this state, the Column Alignment Sync bit is '1' in MDIO registers 4/5.24.12. During this state the alignment character code detect circuit is active on each channel but the column re-alignment is disabled. If a complete alignment column is not detected in the correct position within the Inter-Packet Gap, the column state machine transitions to state FAIL2. FAIL2 – When in this state, the Column Alignment Sync bit is '1' in MDIO registers 4/5.24.12. During this state the alignment character code detect circuit is active on each channel but the column re-alignment is disabled. If a complete alignment column is not detected in the correct position within the Inter-Packet Gap, the column state machine transitions to state FAIL3. FAIL3 – When in this state, the Column Alignment Sync bit is '1' in MDIO registers 4/5.24.12. During this state the alignment character code detect circuit is active on each channel but the column re-alignment is disabled. If complete alignment column is not detected in the correct position within the Inter-Packet Gap, the column state machine transitions to state UNALIGN. (3) (4) The XGXS Lane Alignment bit = '0' causes a local fault to be output on the receive data bus. The XGXS Lane Alignment bit = '0' causes a local fault to be output on the receive data bus. Submit Documentation Feedback 17 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Any Valid Code K28.3 K28.5 K28.0 RDP/N0 R R X X X X X X X X X X A A A A A A A A A A K K K K K K K K K K R R R R R R R R R R K RDP/N1 R R X X X X X X X X X X A A A A A A A A A A K K K K K K K K K K R R R R R R R R RDP/N2 R R X X X X X X X X X X A A A A A A A A A A K K K K K K K K K K R R R R R R R R R R K K K K RDP/N3 R R X X X X X X X X X X A A A A A A A A A A K K K K K K K K K K R R R R R R R R R R RCLK RXD(7:0) undefined undefined A K RXD(15:8) undefined undefined A K RXD(23:16) undefined undefined A K RXD(31:24) undefined undefined A K Figure 19. Channel De-Skew Using Alignment Code INTER-PACKET GAP MANAGEMENT When in transceiver mode, the TLK3138 replaces the idle codes (see Table 2) during the Inter-Packet Gap (IPG) with the necessary codes to perform all channel alignment, byte alignment, and clock tolerance compensation as defined in IEEE 802.3ae 10Gbps Ethernet Standard. According to the Ethernet Standard, a valid packet must begin on TXD(0:7) of the XGMII. However, due to variable packet sizes, the IPG can begin on any channel. The TLK3138 replaces idle codes latched on the same XGMII clock edge as the end of packet code with /K/ codes (as shown in Figure 20). 18 Submit Documentation Feedback TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Packet IPG TXD(7:0) I I S D D D D ... D D D D I I I I I I TXD(15:8) I I D D D D D ... D D D T I I I I I I TXD(23:16) I I D D D D D ... D D D I I I I I I I TXD(31:24) I I D D D D D ... D D D I I I I I I I TDP/N0 K R S D D D D ... D D D D A R R K K R TDP/N1 K R D D D D D ... D D D T A R R K K R TDP/N2 K R D D D D D ... D D D K A R R K K R TDP/N3 K R D D D D D ... D D D K A R R K K R Input Output S = Start of Packet, D = Data, T = End of Packet, A = K28.3, K = K28.5, R = K28.0, I = Idle Figure 20. Inter-Packet Gap Management The subsequent idles in the IPG are replaced by columns of channel alignment codes (K28.3), byte alignment codes (K28.5), or clock tolerance compensation codes (K28.0). The state machine which governs the IPG replacement procedure is illustrated in Figure 21, with notation defined in Table 2. Note that any IPG management state transitions to send data if the IPG is terminated. The repetition of the /A/ pattern on each serial channel allows the FIFOs to remove or add the required phase and frequency difference to align the data from all four serial links of a XAUI channel and allow output of the aligned 32 bit wide data on a single edge of the receive clock, RX_CLK, as shown in Figure 21. Submit Documentation Feedback 19 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Any State !(||idle|| + ||Q||) ||T|| & nextAK = K + Acnt != 0 ||T|| & nextAK = A * Acnt = 0 Send Data reset Send K Send A ||Q|| !||Q|| Send Q Acnt != 0 * nextKR = R Send R Acnt != 0 * nextKR = K Acnt != 0 * nextKR = K Send K nextKR = R Acnt = 0 Acnt != 0 * nextKR = R !||Q|| * nextKR = R Send A Acnt = 0 ||Q|| !||Q|| * nextKR = K Send Q nextKR = K Figure 21. IPG Management State Machine Table 4. IPG Management State Machine Notation SYMBOL ||idle|| ||Q|| nextAK Acnt nextKR ||T|| DEFINITION XGMII idle. 0x07 on TXD(x:: :x-7), Link status message: K28.4, Dx.y, Dx.y, Dx.y. A Boolean variable. It takes the value K when an A is sent at the beginning of the IPG and the value A when a K is sent at the beginning of the IPG. Its initial value is K. When an A character is sent, variable Acnt is loaded with a random number such that 16 ≤ Acnt ≤ 31. Acnt is decremented each time a column of A characters is generated. A randomly-generated Boolean that can assume the value K or R. Terminate Character Column (Terminate Character in Any Lane) CLOCK TOLERANCE COMPENSATION (CTC) The XAUI interface is defined to allow for separate clock domains on each side of the link. Though the reference clocks for two devices on a XAUI link have the same specified frequencies, there are slight differences that, if not compensated for, leads to over or under run of the FIFOs on the receive/transmit data path. The TLK3138 provides compensation for these differences in clock frequencies via the insertion or the removal of /R/ characters on all channels, as shown in Figure 22 and Figure 23. 20 Submit Documentation Feedback TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Packet IPG RDP/N0 K R S D D D D ... D D D D A R R K S D RDP/N1 K R D D D D D ... D D D T A R R K D D RDP/N2 K R D D D D D ... D D D K A R R K D D RDP/N3 K R D D D D D ... D D D K A R R K D D RXD(7:0) I I S D D D D ... D D D D I I I I I S RXD(15:8) I I D D D D D ... D D D T I I I I I D RXD(23:16) I I D D D D D ... D D D I I I I I I D RXD(31:24) I I D D D D D ... D D D I I I I I I D Input Output S = Start of Packet, D = Data, T = End of Packet, A = K28.3, K = K28.5, R = K28.0, I = Idle Added Column Figure 22. Clock Tolerance Compensation: Add The /R/ code is disparity neutral, allowing its removal or insertion without affecting the current running disparity of each channel’s serial stream. Packet IPG RDP/N0 K R S D D D D ... D D D D A R R K S D RDP/N1 K R D D D D D ... D D D T A R R K D D RDP/N2 K R D D D D D ... D D D K A R R K D D Input RDP/N3 R Dropped Column RXD(7:0) I I S D D D D ... D D D D I I I S D D RXD(15:8) I I D D D D D ... D D D T I I I D D D RXD(23:16) I I D D D D D ... D D D I I I I D D D RXD(31:24) I I D D D D D ... D D D I I I I D D D Output S = Start of Packet, D = Data, T = End of Packet, A = K28.3, K = K28.5, R = K28.0, I = Idle Figure 23. Clock Tolerance Compensation: Drop Submit Documentation Feedback 21 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 PARALLEL TO SERIAL The parallel-to-serial shift register on each channel takes in data and converts it to a serial stream. The shift register is clocked by the internally generated bit clock, which is 10 times the reference clock (REFCLKP/REFCLKN) frequency. The least significant bit (LSB) for each channel is transmitted first. SERIAL TO PARALLEL For each channel, serial data is received on the RDPx/RDNx pins. The interpolator and clock recovery circuit locks to the data stream if the clock to be recovered is within ±200 PPM of the internally generated bit rate clock. The recovered clock is used to retime the input data stream. The serial data is then clocked into the serial-to-parallel shift registers. If enabled, the 10-bit wide parallel data is then fed into 8b/10b decoders. HIGH SPEED CML OUTPUT The high speed data output driver is implemented using Current Mode Logic (CML) with integrated pull up resistors requires no external components. The line can be directly coupled or AC coupled. Under many circumstances, AC couple is desirable. TDP RDP 50 W Transmission Line VDDT 50 W 3 kW 50 W 9 kW GND + - 50 W Transmission Line TDN TRANSMITTER RDN MEDIA RECEIVER Figure 24. Example High Speed I/O AC Coupled Mode Standard Current Mode Logic (CML) drivers usually require external components. The disadvantage of the external edge control is a limited edge rate due to package and line parasitic. The CML driver on TLK3138 has on-chip 50Ω termination resistors terminated to VDDT therefore provides optimum performance for increased speed requirements. The transmitter output driver is highly configurable allowing slew rate, output amplitude and pre-emphasis to be turned to a channel's individual requirements. An internal voltage reference derived from VDDT is also available to provide the target for output amplitude control loop. This reference is enabled by holding register bit 4/5.32900.6 low and results in a nominal output amplitude of ~1400mV differential pk-pk for 100% swing. The receiver input is internally biased to 2×VDDT/3 which is the optimum voltage for input sensitivity. As the input and output references are derived from VDDT, the tolerance of this supply dominates the accuracy of the internal reference. Applications requiring higher tolerance output amplitude are advised to provide a high accuracy external reference. When transmitting data across long lengths of PCB trace or cable, the high frequency content of the signal is attenuated due to the skin effect of the media. This causes a smearing of the data eye when viewed on an oscilloscope. The net result is reduced timing margins for the receiver and clock recovery circuits. In order to provide equalization for the high frequency loss, 2-tap finite impulse response (FIR) transmit pre-emphasis is implemented. In a 1-tap FIR pre-emphasis, differential swing is increased or pre-emphasized for the bit immediately following a transition and subsequently reduced or de-emphasized for run lengths greater than one, as shown in Figure 25. This provides additional high frequency energy to compensate for PCB or cable loss. 22 Submit Documentation Feedback TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 VOD(p) VOD(d) VOD(pd) VCMT VOD(pp) VOD(d) bit time bit time VOD(p) Figure 25. The 2-stage mode operates in a similar manner but considers the logic level of the previous two transmitted bits when determined how much pre-emphasis to apply. The level and mode of the pre-emphasis is programmable via MDIO Register bits 4/5.32900.14:11. Users can control the strength of the pre-emphasis to optimize for a specific system requirement. HIGH SPEED RECEIVER The high speed receiver conforms to the physical layer requirements of IEEE 802.3ae Clause 47(XAUI). The termination impedances of the receiver is configured as 100 Ohms with the center tap weakly tied to 2×VDDT/3 with a capacitor to create an AC ground. AC coupling is always required on receiver inputs. All receive channels incorporate an adaptive equalizer. This circuit compensates for channel insertion loss by amplifying the high frequency components of the signal, reducing inter-symbol interference. Setting 4/5.32900.2 enables adaptive equalization. In this mode, both the gain and bandwidth of the equalizer with be controlled by the receiver equalization logic. Bandwidth selection is based on the setting applied to 4/5.32901.14:13 and 4/5.32900.3. Equalization can be disabled by setting 4/5.32900.2 low. LOOPBACK In XAUI Mode, two internal loopback modes are possible for each XAUI Channel Group A and B. One, called XGMII loopback, allows the data input on the XGMII interface to be returned out the corresponding XGMII interface (A->A, B->B). The other, called XAUI loopback, allows serial data on the XAUI interface to be returned out the corresponding XAUI interface (A->A, B->B). An external loopback (requiring external connection) is also supported, which can be used with the PRBS patterns, as well as the CJPAT, CRPAT, Mixed/High/Low Frequency tests. LINK TEST FUNCTIONS The TLK3138 has an extensive suite of built in test functions to support system diagnostic requirements. Each channel has built-in link test generator and verification logic. Several patterns can be selected via the MDIO that offer extensive test coverage. The patterns are: 27-1 or 223-1 PRBS (Pseudo Random Bit Stream), CJPAT, CRPAT, high and low and mixed frequency patterns. MDIO Management Interface The TLK3138 supports the Management Data Input/Output (MDIO) Interface as defined in Clause 45 of the IEEE 802.3ae Ethernet specification. The MDIO allows register-based management and control of the serial links. Normal operation of the TLK3138 is possible without use of this interface. However, some additional features are accessible only through the MDIO. The following registers in the PROGRAMMERS REFERENCE are implemented as global registers (i.e Same physical register is accessed through XAUI-A register access or XAUI-B register access): 16'h8086, 16'h8088, 16'h8089, 16'h808B, 16'h8093,16'h80A0, 16'h80A1, 16'h80A2 16'h808C, 16'h808E, 16'h8090, 16'h8091, 16'h8092, All other registers are implemented per XAUI channel (unique per channel). Submit Documentation Feedback 23 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 The MDIO Management Interface consists of a bi-directional data path (MDIO) and a clock reference (MDC). The device id and port address are determined by bootstrap pins (see Table 82). In Clause 45, the 5 bootstrap pins determine the port address. Of the port address bits, only the upper four bits are looked at. If the bottom bit of the incoming serial port address is a zero, then XGXS A is addressed. If the bottom bit of the incoming serial port address is a one, then XGXS B is addressed. The device ID is required to be either 4 or 5, so only the bottom device id bit is required. If device id bit 0 is a 0, then a PHY device is selected for both XGXS A and B. If device id bit 0 is a 1, then a DTE device is selected for both XGXS A and B. Note, each register is accessed as either DTE or PHY devices in the TLK3138, although physically there is only one register accessed two different ways. Also note, the XAUI interfaces must either be both DTE devices or both be PHY devices. An even PHY Address (as defined below) indicates an access to XAUI A register space, and an odd PHY Address indicates access to XAUI B register space. Write transactions which address an invalid register or device, or a read only register are ignored. Read transactions which address an invalid register or device return a 0. Timing for a Clause 45 address transaction is shown in Figure 26. The Clause 45 timing required to write to the internal registers is shown in Figure 27. The Clause 45 timing required to read from the internal registers is shown in Figure 28. The Clause 45 timing required to read from the internal registers and then increment the active address for the next transaction is shown in Figure 29. MDC 0 MDIO 0 0 0 A4 A0 DA4 DA0 1 0 A15 A0 1 32 "1's" Preamble Addr Code Start PHY Addr Dev Addr Turn Around Reg Addr Idle Figure 26. CL45 - Management Interface Extended Space Address Timing MDC 0 MDIO 0 0 1 A4 A0 DA4 DA0 1 0 D15 D0 1 32 "1's" Write Code Start Preamble PHY Addr Dev Addr Turn Around Idle Data Figure 27. CL45 - Management Interface Extended Space Write Timing MDC 0 MDIO 0 1 1 A4 A0 DA4 DA0 32 "1's" Read Code Start Preamble PHY Addr Dev Addr 0 Turn Around D15 D0 Data 1 Idle Figure 28. CL45 - Management Interface Extended Space Read Timing MDC 0 MDIO 0 32 "1's" Preamble Start 1 0 Read & Inc. Code A4 A0 PHY Addr DA4 DA0 Dev Addr Z 0 Turn Around D15 D0 Data 1 Idle Figure 29. CL45 - Management Interface Extended Space Read And Increment Timing 24 Submit Documentation Feedback TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 The IEEE 802.3ae Clause 45 specification defines many of the registers, and additional registers have been implemented for expanded functionality. PROGRAMMERS REFERENCE Table 5. XS (1)_CONTROL_1 Address: 0x0000 Bit(s) Default: 0x2040 Name Description 4/5.0.15 Reset 1 = XS reset (including all registers) 0 = Normal operation 4/5.0.14 Loop back 1 = Enable loop back mode. If the device is configured as PHY XS (DVAD(0) = 0), then XAUI_LOOPBACK is performed (RX parallel to TX parallel) If the device is configured as DTE XS (DVAD(0) = 1), then XGMII_LOOPBACK is performed (TX serial to RX serial) 0 = Disable loop back mode 4/5.0.13 Speed Selection This bit always reads 1 indicating operation at 10 Gb/s and above. 4/5.0.11 Low power 1 = Low power mode 0 = Normal operation 4/5.0.6 Speed Selection This bit always reads 1 indicating operation at 10Gbps and above. 4/5.0.5:2 Speed Selection These bits always read 0 indicating operation at 10Gbps. (1) (2) Access (2) RW SC RW In this section XS refers to either PHY or DTE XS device. RO: Read-Only, RW: Read-Write, SC: Self-Clearing, LL: Latching-Low, LH: Latching-High, COR: Clear-on-Read Table 6. XS_STATUS_1 Address: 0x0001 Bit(s) 4/5.1.7 4/5.1.2 4/5.1.1 Name Fault Default: 0x0082 Description 1 = Fault condition detected (either on TX or RX side. This bit is OR ed version of 4/5.8.10 and 4/5.8.11) 0 = No fault condition detected XS Transmit Link Status Low Power Ability 1 = XS Transmit link is up. 0 = XS Transmit links is down. (This bit is latched low version of 4/5.24.12) This bit always reads 1 indicating support for low power mode Access RO RO/LL RO Table 7. XS_DEVICE_IDENTIFIER_1 Address: 0x0002 Bit(s) 4/5.2.15.0 Name OUI c:r Default: 0x4000 Description Organizationally unique identifier. Access RO Table 8. XS_DEVICE_IDENTIFIER_2 Address: 0x0003 Bit(s) 4/5.3.15:0 Name OUI c:r Default: 0x50D0 Description Device identifier. Manufacturer model and revision number Access RO Table 9. XS_SPEED_ABILITY Address: 0x0004 Bit(s) 4/5.4.0 Name 10G Capable Default: 0x0001 Description This bit always reads 1 indicating operation at 10Gb/s Submit Documentation Feedback Access RO 25 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Table 10. XS_DEVICES_IN_PACKAGE_1 Address: 0x0005 Bit(s) Default: 0x0010 Name Description 4/5.5.5 DTE XS Present 1 = DTE XS present in the package. 0 = DTE XS not present in the package. Read returns 1, when dvad_in is high 4/5.5.4 PHY XS Present 1 = PHY XS present in the package. 0 = PHY XS not present in the package. Read returns 1, when dvad_in is low 4/5.5.3 PCS Present Always reads 0 4/5.5.2 WIS Present Always reads 0 4/5.5.1 PMD/PMA Present Always reads 0 4/5.5.0 Clause 22 registers Present Always reads 0 Access RO Table 11. XS_DEVICES_IN_PACKAGE_2 Address: 0x0006 Bit(s) Default: 0x0000 Name Description 4/5.6.15 Vendor Specific Device 2 Present This bit always reads 0 indicating that vendor specific device 2 not present in package. 4/5.6.14 Vendor Specific Device 1 Present This bit always reads 0 indicating that vendor specific device 1 not present in package . Access RO Table 12. XS_STATUS_2 Address: 0x0008 Bit(s) Default: 0x8C00 Name Description 4/5.8.15:14 Device present Always read 10 indicating that device responds at this address 4/5.8.11 Transmit fault 1 = Fault condition on transmit path 0 = No fault condition on transmit path 4/5.8.10 Receive fault 1 = Fault condition on receive path 0 = No fault condition on receive path Access RO RO/LH Table 13. XS_PACKAGE_IDENTIFIER_1 Address: 0x000E Bit(s) 4/5.14.15:0 Default: 0x4000 Name OUI c:r Description Organizationally unique identifier. Access RO Table 14. XS_PACKAGE_IDENTIFIER_2 Address: 0x000F Bit(s) 4/5.15.15:0 26 Name OUI c:r Default: 0x50D0 Description Organizationally unique identifier Manufacturer model and revision number. Submit Documentation Feedback Access RO TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Table 15. XS_LANE_STATUS Address: 0x0018 Bit(s) Default: 0x0C00 Name Description 4/5.24.12 Align status When 1, indicates all lanes are aligned 4/5.24.11 Pattern testing ability Always reads 1. Able to generate test patterns 4/5.24.10 Loopback ability Always read 1. Has the ability to perform loopback function 4/5.24.3 Lane 3 sync 1 = Lane 3 is synchronized 0 = Lane 3 is not synchronized 4/5.24.2 Lane 2 sync 1 = Lane 2 is synchronized 0 = Lane 2 is not synchronized 4/5.24.1 Lane 1 sync 1 = Lane 1 is synchronized 0 = Lane 1 is not synchronized 4/5.24.0 Lane 0 sync 1 = Lane 0 is synchronized 0 = Lane 0 is not synchronized Access RO Table 16. XS_TEST_CONTROL Address: 0x0019 Bit(s) Default: 0x0000 Name Description 4/5.25.2 Receive test-pattern enable When 1, indicates test pattern function is enabled. 4/5.25.1:0 Test-pattern select 00 01 10 11 = = = = High frequency test pattern Low frequency test pattern Mixed frequency test pattern Reserved Access RW Table 17. TEST_CONFIG Address: 0x8000 Bit(s) Name Default: 0x0000 Description 4/5.32768.1 CRPAT enable When set, enables the CRPAT test pattern on all 4 lanes. 4/5.32768.0 CJPAT enable When set, enables the CJPAT test pattern on all 4 lanes. Access RW Table 18. TEST_VERIFICATION_CONTROL Address: 0x8001 Bit(s) Name Default: 0x0000 Description 4/5.32769.1 CRPAT check enable When set, enables the verification of CRPAT test modes. 4/5.32769.0 CJPAT check enable When set, enables the verification of CJPAT test modes. Access RW Table 19. TX_FIFO_STATUS Address: 0x8002 Bit(s) Name Default: 0x0000 Description 4/5.32770.9 Lane 3 overflow 4/5.32770.8 Lane 2 overflow 4/5.32770.7 Lane 1 overflow 4/5.32770.6 Lane 0 overflow 4/5.32770.5 Lane 3 underflow 4/5.32770.4 Lane 2 underflow 4/5.32770.3 Lane 1 underflow 4/5.32770.2 Lane 0 underflow 4/5.32770.1 Overflow When high, indicates that transmit FIFO overflow condition occurred in any lane. 4/5.32770.0 Underflow When high, indicates that transmit FIFO underflow condition occurred in any lane. Access When high, indicates that transmit FIFO overflow condition occurred for the corresponding lane. RO/LH When high, indicates that transmit FIFO underflow condition occurred for the corresponding lane. RO/LH Submit Documentation Feedback 27 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Table 20. TX_FIFO_DROP_COUNT Address: 0x8003 Bit(s) 4/5.32771.15:0 Default: 0xFFFD Name Drop count Description Access Counter for number of idle drops in the transmit FIFO RO/COR Table 21. TX_FIFO_INSERT_COUNT Address: 0x8004 Bit(s) 4/5.32772.15:0 Default: 0xFFFD Name Description Insert count Access Counter for number of idle inserts in the transmit FIFO RO/COR Table 22. TX_CODEGEN_STATUS Address: 0x8005 Bit(s) Default: 0x0000 Name Description 4/5.32773.6 Invalid XGMII character in lane 3 4/5.32773.5 Invalid XGMII character in lane 2 4/5.32773.4 Invalid XGMII character in lane 1 4/5.32773.3 Invalid XGMII character in lane 0 4/5.32773.2 Invalid XGMII character error When high, indicates invalid XGMII character received in any lane 4/5.32773.1 Invalid T column error When high, indicates invalid Terminate column (column that contains Terminate character not followed by Idle character(s)) received from the XGMII interface. 4/5.32773.0 Invalid S column error When high, indicates invalid Start column (column that contains Start character in a lane other than lane 0) received from the XGMII interface. Access When high, indicates invalid XGMII character received in the corresponding lane. RO/LH RO/LH Table 23. LANE_0_TEST_ERROR_COUNT Address: 0x8006 Bit(s) 4/5.32774.15:0 Default: 0xFFFD Name Description Lane 0 test pattern error counter This counter reflects errors for High, Medium or Low Frequency test patterns for lane 0. This counter increments by 1 for each received character that has error. Access RO/COR Table 24. LANE_1_ TEST_ERROR_COUNT Address: 0x8007 Bit(s) 4/5.32775.15:0 Default: 0xFFFD Name Description Lane 1 test pattern error counter This counter reflects errors for High, Medium or Low Frequency test patterns for lane 1. This counter increments by 1 for each received character that has error. Access RO/COR Table 25. LANE_2_ TEST_ERROR_COUNT Address: 0x8008 Bit(s) 4/5.32776.15:0 Default: 0xFFFD Name Description Lane 2 test pattern error counter This counter reflects errors for High, Medium or Low Frequency test patterns for lane 2. This counter is incremented by 1 for each received character that has error. Access RO/COR Table 26. LANE_3_ TEST_ERROR_COUNT Address: 0x8009 Bit(s) 4/5.32777.15:0 28 Name Lane 3 test pattern error counter Default: 0xFFFD Description This counter reflects errors for High, Medium or Low Frequency test patterns for lane 3. This counter increments by one for each received character that has error. Submit Documentation Feedback Access RO/COR TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Table 27. CRPAT_CJPAT_TEST_ERROR_COUNT_1 (1) Address: 0x800A Bit(s) Name 4/5.32778.15:0 (1) Default: 0xFFFF Description CRPAT/CJPAT test error counter MSB of CRPAT/CJPAT error counter for all 4 lanes Access RO User has to make sure that register 32778 is read first and then register 32779. If user reads register 32779 without reading register 32778 first, then the count value read through 32779 register may not be correct. Table 28. CRPAT_CJPAT_TEST_ERROR_COUNT_2 (1) Address: 0x800B Bit(s) 4/5.32779.15:0 (1) Default: 0xFFFD Name Description CRPAT/CJPAT test error counter LSB of CRPAT/CJPAT error counter for all 4 lanes Access RO User has to make sure that register 32778 is read first and then register 32779. If user reads register 32779 without reading register 32778 first, then the count value read through 32779 register may not be correct. Table 29. LANE_0_EOP_ERROR_COUNT (1) Address: 0x800C Bit(s) Name 4/5.32780.15:0 Lane 0 end of packet error counter Default: 0xFFFD Description End of packet termination error counter for lane 0. End of packet error for lane 0 is detected on the RX side. It is detected when Terminate character is in lane 0 and one or both of the following holds: • Terminate character is not followed by /K/ characters in lanes 1, 2 and 3 Access RO/COR • The column following the terminate column is neither ||K|| or ||A||. (1) Counter is incremented by 1 when EOP error is found on the corresponding lane and when all the lanes are aligned (align_status should be high). Counter holds on to its value when align_status goes low or when the counter reaches its maximum value. It is cleared when it is read. Table 30. LANE_1_EOP_ERROR_COUNT (1) Address: 0x800D Bit(s) Name 4/5.32781.15:0 Lane 1 end of packet error counter Default: 0xFFFD Description End of packet termination error counter for lane 1. End of packet error for lane 1 is detected on the RX side. It is detected when Terminate character is in lane 1 and one or both of the following holds: • Terminate character is not followed by /K/ characters in lanes 2 and 3 Access RO/COR • The column following the terminate column is neither ||K|| or ||A||. (1) Counter is incremented by 1 when EOP error is found on the corresponding lane and when all the lanes are aligned (align_status should be high). Counter holds on to its value when align_status goes low or when the counter reaches its maximum value. It is cleared when it is read. Table 31. LANE_2_EOP_ERROR_COUNT (1) Address: 0x800E Bit(s) 4/5.32782.15:0 Name Lane 2 end of packet error counter Default: 0xFFFD Description End of packet termination error counter for lane 2. End of packet error for lane 2 is detected on the RX side. It is detected when Terminate character is in lane 2 and one or both of the following holds: • Terminate character is not followed by /K/ characters in lane 3 Access RO/COR • The column following the terminate column is neither ||K|| or ||A||. (1) Counter is incremented by 1 when EOP error is found on the corresponding lane and when all the lanes are aligned (align_status should be high). Counter holds on to its value when align_status goes low or when the counter reaches its maximum value. It is cleared when it is read. Submit Documentation Feedback 29 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Table 32. LANE_3_EOP_ERROR_COUNT (1) Address: 0x800F Bit(s) 4/5.32783.15:0 (1) Default: 0xFFFD Name Lane 3 end of packet error counter Description Access End of packet termination error counter for lane 3. End of packet error for lane 3 is detected on the RX side. It is detected when Terminate character is in lane 3 and the column following the terminate column is neither ||K|| or ||A||. RO/COR Counter is incremented by 1 when EOP error is found on the corresponding lane and when all the lanes are aligned (align_status should be high). Counter holds on to its value when align_status goes low or when the counter reaches its maximum value. It is cleared when it is read. Table 33. LANE_0_CODE_ERROR_COUNT (1) Address: 0x8010 Bit(s) 4/5.32784.15:0 (1) Default: 0xFFFD Name Lane 0 code error counter Description Access Output 16-bit counter for invalid code group found in lane 0. Invalid code group is detected when the 8B10B decoder cannot decode the received codeword. RO/COR Counter is incremented by 1 when codeword error is found on the corresponding lane and when all the lanes are aligned (align_status should be high). Counter holds on to its value when align_status goes low or when the counter reaches its maximum value. It is cleared when it is read. Table 34. LANE_1_CODE_ERROR_COUNT (1) Address: 0x8011 Bit(s) 4/5.32785.15:0 (1) Default: 0xFFFD Name Lane 1 code error counter Description Access Output 16-bit counter for invalid code group found in lane 1. Invalid code group is detected when the 8B10B decoder cannot decode the received codeword. RO/COR Counter is incremented by 1 when codeword error is found on the corresponding lane and when all the lanes are aligned (align_status should be high). Counter holds on to its value when align_status goes low or when the counter reaches its maximum value. It is cleared when it is read. Table 35. LANE_2_CODE_ERROR_COUNT (1) Address: 0x8012 Bit(s) 4/5.32786.15:0 (1) Default: 0xFFFD Name Lane 2 code error counter Description Access Output 16-bit counter for invalid code group found in lane 2. Invalid code group is detected when the 8B10B decoder cannot decode the received codeword. RO/COR Counter is incremented by 1 when codeword error is found on the corresponding lane and when all the lanes are aligned (align_status should be high). Counter holds on to its value when align_status goes low or when the counter reaches its maximum value. It is cleared when it is read. Table 36. LANE_3_CODE_ERROR_COUNT (1) Address: 0x8013 Bit(s) 4/5.32787.15:0 (1) 30 Name Lane 3 code error counter Default: 0xFFFD Description Access Output 16-bit counter for invalid code group found in lane 3. Invalid code group is detected when the 8B10B decoder cannot decode the received codeword. RO/COR Counter is incremented by 1 when codeword error is found on the corresponding lane and when all the lanes are aligned (align_status should be high). Counter holds on to its value when align_status goes low or when the counter reaches its maximum value. It is cleared when it is read. Submit Documentation Feedback TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Table 37. RX_CHANNEL_SYNC_STATE Address: 0x8014 Bit(s) Default: 0x0000 Name Description 4/5.32788.11:9 Channel synchronization FSM state for lane 0 Current state of sync state machine in lane 0 4/5.32788.8:6 Channel synchronization FSM state for lane 1 Current state of sync state machine in lane 1 4/5.32788.5:3 Channel synchronization FSM state for lane 2 Current state of sync state machine in lane 2 4/5.32788.2:0 Channel synchronization FSM state for lane 3 Current state of sync state machine in lane 3 Access RO Table 38. RX_LANE_ALIGN_STATUS Address: 0x8015 Bit(s) Default: 0x0000 Name Description 4/5.32789.15:12 Align state Current lane alignment FSM state 4/5.32789.0 Lane Alignment FIFO collision Collision status for lane alignment FIFO. When high, indicates that there is collision error in lane alignment FIFO. Access RO RO/LH Table 39. RX_CHANNEL_SYNC_STATUS Address: 0x8016 Bit(s) Default: 0x0000 Name 4/5.32790.11 Description Channel Synchronization status for all lanes 1 = Channel synchronization is achieved in all lanes. 0 = Channel synchronization is lost in one or more lanes Access RO/LL Table 40. BIT_ORDER Address: 0x8017 Bit(s) Default: 0x0005 Name Description 4/5.32791. 3 XGMII RX bit order When high, reverses the order of bits in the parallel data sent from XAUI RX for each lane. 4/5.32791. 2 XAUI RX bit order When high, reverses the order of bits in the parallel data received from SERDES RX macros for each lane. 4/5.32791. 1 XGMII TX bit order When high, reverses the order of bits in the parallel data received from the XGMII interface each lane. 4/5.32791. 0 XAUI TX bit order When high, reverses the order of bits in the parallel data sent to the SERDES TX macro for each lane. Access RW Table 41. LOOPBACK_CONTROL (1) Address: 0x8018 Bit(s) Default: 0x0000 Name Description 4/5.32792. 1 XAUI side loopback When high, loops back 32 bit data and 4 control bits from the RX path to the TX path. (4/5.0.14 should be 0 else no effect) 4/5.32792. 0 XGMII side loopback When 1, loops back 40 bit data from TX path to the RX path (4/5.0.14 should be 0 else no effect) (1) Access RW See Loopback section for more information. Submit Documentation Feedback 31 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Table 42. TX_BYPASS_CONTROL Address: 0x8019 Bit(s) Default: 0x0000 Name Description 4/5.32793.15 TX IPG management bypass When high, disables IPG management (replacing Idle XGMII characters with /A/K/R/Q/ code-words) in transmit side. 4/5.32793.11 TX CTC Bypass When high, disables clock tolerance compensation in transmit side 4/5.32793.7 Lane 3 8B10B encoder bypass When high, disables 8B10B encoding on the corresponding lane 4/5.32793.6 Lane 2 8B10B encoder bypass 4/5.32793.5 Lane 1 8B10B encoder bypass 4/5.32793.4 Lane 0 8B10B encoder bypass Access RW RW Table 43. RX_CTC_STATUS Address: 0x801A Bit(s) Name Default: 0x0000 Description 4/5.32794.9 Lane 3 overflow 4/5.32794.8 Lane 2 overflow 4/5.32794.7 Lane 1 overflow 4/5.32794.6 Lane 0 overflow 4/5.32794.5 Lane 3 underflow 4/5.32794.4 Lane 2 underflow 4/5.32794.3 Lane 1 underflow 4/5.32794.2 Lane 0 underflow 4/5.32794.1 Overflow When high, indicates overflow error in any lane. 4/5.32794.0 Underflow When high, indicates underflow error in any lane. Access When high, indicates overflow error in the corresponding lane. RO/LH When high, indicates underflow error in the corresponding lane. RO/LH Table 44. RX_CTC_INSERT_COUNT Address: 0x801B Bit(s) 4/5.32795.15:0 Name Idle insert count Default: 0xFFFD Description Counter for number of idle insertions in RX side Access RO/COR Table 45. RX_CTC_DELETE_COUNT Address: 0x801C Bit(s) 4/5.32796.15:0 Name Idle delete count Default: 0xFFFD Description Counter for number of idle deletions on RX side Access RO/COR Table 46. DATA_DOWN Address: 0x801D Bit(s) Name 4/5.32797.3 Lane 3 data down 4/5.32797.2 Lane 2 data down 4/5.32797.1 Lane 1 data down 4/5.32797.0 Lane 0 data down 32 Default: 0x0000 Description Access When high, indicates that link for the corresponding lane was inactive (data did not toggle) for 4095 cycles of 312.5MHz clock. The 312.5MHz is generated internally by the PLL from the 156MHz Reference clock. Submit Documentation Feedback RO/COR TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Table 47. RX_BYPASS_CONTROL Address: 0x801E Bit(s) Default: 0x0000 Name Description 4/5.32798.15 RX CTC bypass When set, bypasses clock tolerance compensation on the RX side 4/5.32798.14 IPG Checker bypass When set, bypasses the replacement of /A/K/R/ into Idles and also bypasses end-of-packet error checking. 4/5.32798.11 Lane 3 8B/10B decoder bypass When set, disables the 8B/10B decoding for the corresponding lane 4/5.32798.10 Lane 2 8B/10B decoder bypass 4/5.32798.9 Lane 1 8B/10B decoder bypass 4/5.32798.8 Lane 0 8B/10B decoder bypass 4/5.32798.7 Consider sequence column part of IPG When set, sequence columns are counted as part of IPG When low, sequence columns are not counted as IPG 4/5.32798.3 RX Lane align bypass When set, bypasses lane alignment on the RX side Access RW RW Table 48. CLOCK_DOWN_STATUS Address: 0x801F Bit(s) Default: 0x0000 Name 4/5.32799. 7 Lane 3 clock 312 down 4/5.32799. 6 Lane 2 clock 312 down 4/5.32799. 5 Lane 1 clock 312 down 4/5.32799. 4 Lane 0 clock 312 down 4/5.32799. 3 Lane 3 clock 156 down 4/5.32799. 2 Lane 2 clock 156 down 4/5.32799. 1 Lane 1 clock 156 down 4/5.32799. 0 Lane 0 clock 156 down Description When high, indicates that 312MHz clock is down on the corresponding lane for 255 or more cycles. The detection is done on the transmit side. The 312.5MHz is generated internally by the PLL from the 156MHz Reference clock. Access RO/LH When high, indicates that 156MHz XGMII clock is down on the corresponding lane for 255 or more cycles. The detection is done on the transmit side RO/LH Table 49. AUXILIARY_RESET_CONTROL Address: 0x8020 Bit(s) Default: 0x0000 Name Description 4/5.32800. 15 Transmit auxiliary reset When set, resets XAUI transmit data path but does not reset any R/W registers. 4/5.32800.14 Receive auxiliary reset When set, resets XAUI receive data path but does not reset any R/W registers. 4/5.32800.13 TLK3138 auxiliary reset When set, resets the DDR, RETIME muxing, A/B muxing logic but does not reset any R/W registers. Access RW/SC Table 50. TEST_PATTERN_STATUS Address: 0x8021 Bit(s) Default: 0x0000 Name 4/5/32801.15 Test pattern sync status Description Access When high, indicates that preamble for CRPAT/CJPAT has been recovered. RO Table 51. LANE_0_ERROR_CODE Address: 0x8022 Bit(s) 4/5.32802.15:7 Name Lane 0 error code select. Default: 0xCE00 Description Access Error code to be transmitted in case of error condition. This applies to both TX and RX data paths. The msb is the control bit; remaining 8 bits constitute the error code. The default value for lane 0 corresponds to 8′h9C with the control bit being 1′b1. The default values for lanes 0~3 correspond to ||LF|| RW Submit Documentation Feedback 33 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Table 52. LANE_1_ERROR_CODE Address: 0x8023 Bit(s) Default: 0x0000 Name 4/5.32803.15:7 Lane 1 error code select. Description Access Error code to be transmitted in case of error condition. This applies to both TX and RX data paths. The msb is the control bit; remaining 8 bits constitute the error code. The default value for lane 1 corresponds to 8′’h00 with the control bit being 1′’b0. The default values for lanes 0~3 correspond to ||LF|| RW Table 53. LANE_2_ERROR_CODE Address: 0x8024 Bit(s) Default: 0x0000 Name 4/5.32804.15:7 Lane 2 error code select. Description Access Error code to be transmitted in case of error condition. This applies to both TX and RX data paths. The msb is the control bit; remaining 8 bits constitute the error code. The default value for lane 2 corresponds to 8′’h00 with the control bit being 1′’b0. The default values for lanes 0~3 correspond to ||LF|| RW Table 54. LANE_3_ERROR_CODE Address: 0x8025 Bit(s) Default: 0x0080 Name 4/5.32805.15:7 Lane 3 error code select. Description Access Error code to be transmitted in case of error condition. This applies to both TX and RX data paths. The msb is the control bit; remaining 8 bits constitute the error code. The default value for lane 3 corresponds to 8′’h01 with the control bit being 1′’b0. The default values for lanes 0~3 correspond to ||LF|| RW Table 55. RX_PHASE_SHIFT_CONTROL Address: 0x8026 Bit(s) Default: 0x0000 Name 4/5.32806. 15 Lane 3 phase shift 4/5.32806. 14 Lane 2 phase shift 4/5.32806. 13 Lane 1 phase shift 4/5.32806. 12 Lane 0 phase shift 4/5.32806. 11 XGMII Rx Source timing Description Access When set, delays the RX data sent to the XGMII interface by one clock cycle RW When high supports source aligned timing on the XGMII receive output bus When low supports source centered timing on the XGMII receive output bus (Default) Table 56. CHANNEL_SYNC_CONTROL Address: 0x8027 Bit(s) Name Default: 0x0000 Description 4/5.32807. 15 Lane 3 channel sync bypass When set, lane synchronization for the corresponding lane is bypassed 4/5.32807. 14 Lane 2 channel sync bypass 4/5.32807. 13 Lane 1 channel sync bypass 4/5.32807. 12 Lane 0 channel sync bypass 4/5.32807. 11 Lane 3 channel sync freeze 4/5.32807. 10 Lane 2 channel sync freeze 4/5.32807. 9 Lane 1 channel sync freeze 4/5.32807. 8 Lane 0 channel sync freeze 34 Access RW When set, freezes the last acquired word alignment for the corresponding lane RW Submit Documentation Feedback TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Table 57. SERDES_CONFIG_1 (1) Address: 0x8084 Bit(s) 4/5.32900.15 Default: 0x806C Name Description TXBCLKM CFG[23] Access 0 = Individual lane TXBCLK ports are used 1 = TXBCLK[1] is used to time TD for all lanes (default) 4/5.32900.14:1 Pre emphasis 1 (TX_CFG[7:4]) See Table 58: Transmit Pre-emphasis Settings Depends on transmit swing setting controlled by CFG_TX [3:2]. These bits do not have any effect if CFG[3:2] = 2′’b10 (Default 4’b0000) 4/5.32900.10:9 SWING (TX_CFG[3:2]) Output swing setting RW 00 = Maximum transmit amplitude, pre-emphasis available (Default) 01 = 62.5% transmit amplitude, increased pre-emphasis available RW 10 = 37.5% transmit amplitude, pre-emphasis unavailable 11 = 0% transmit amplitude 4/5.32900.8:7 Slew Rate (CFG[16:15]) Slew Rate setting See Table 59: Slew rate control. Tx Rise and Fall times 00 = Fastest edge rate, independent of DATARATE (Default) RW 01 = Intermediate edge rate for given DATARATE 10 = Slower intermediate edge rate for given DATARATE 11 = Slowest edge rate for given DATARATE 4/5.32900.6 EXTREF (CFG[14]) 0 = Internally generated reference is used to set output amplitude 1 = External reference VREF is used to set output amplitude (Default) RW External reference is not supported. 4/5.32900.5 AC Coupled (CFG[13]) 0 = AC coupled operation is disabled 1 = AC coupled operation is enabled (Default) RW Only AC coupled mode is supported. 4/5.32900.4 Enable LOL (CFG[12]) 0 = Loss of link detection is disabled (Default) RW 1 = Loss of link detection is enabled 4/5.32900.3 FASTEQ (CFG[11]) 0 = Adaptive equalization set on low data rate RW 1 = Adaptive equalization set on high data rate (Default) 4/5.32900.2 ENEQ (CFG[10]) 0 = Adaptive equalization is disabled RW 1 = Adaptive equalization is enabled (Default) 4/5.32900.1 FASTUPDT (CFG[9]) 0 = Fast update mode is disabled (Default) RW 1 = Fast update mode is enabled 4/5.32900.0 FASTLOCK (CFG[8]) 0 = Fast-lock mode is disabled (Default) RW 1 = Fast-lock mode is enabled (1) Above control bits are only for vendor testing only. Customer should leave them at their default values. They can be accessed from A side or B side. Table 58. Transmit Pre-Emphasis Settings MODE CFG_TX[7:4] 100% SWING 62.5% SWING 1st BIT 2nd BIT 1st BIT 2nd BIT 0% 0% 0% 0% Disabled 0000 1 – Tap 0001 5% 0% 9% 0% 0010 11% 0% 19% 0% 0011 18% 0% 32% 0% 0100 25% 0% 47% 0% 0101 33% 0% 67% 0% 0110 43% 0% 92% 0% 0111 67% 0% 178% 0% 1000 100% 0% 400% 0% Submit Documentation Feedback 35 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Table 58. Transmit Pre-Emphasis Settings (continued) MODE CFG_TX[7:4] 2 –Tap 100% SWING 62.5% SWING 1st BIT 2nd BIT 1st BIT 2nd BIT 1001 25% 18% 47% 32% 1010 33% 18% 67% 32% 1011 33% 25% 67% 47% 1100 43% 25% 92% 47% 1101 54% 25% 127% 47% 1110 82% 54% 257% 127% 1111 100% 54% 400% 127% The slew rate of the differential driver may be controlled to suit different transmission media and data rates. This is controlled through CFG [16:15] and CFG [6:5], the effects are shown in Table 58. Table 59. Slew Rate Control. Tx Rise and Fall Times CFG[6:5] = 01 or 10 CFG[6:5] = 00 CFG[16:15] MIN MAX MIN MAX 00 90 ps 104 ps 90 ps 104 ps 01 146 ps 202 ps 101 ps 145 ps 10 173 ps 240 ps 126 ps 169 ps 11 281 ps 403 ps 144 ps 196 ps Table 60. SERDES_CONFIG_2 Address: 0x8085 Bit(s) 4/5.32901.15 Default: 0x0AC0 Name JNCSEL (CFG[7]) Description 4/5.32901.14:1 DATARATE (CFG[6:5] / 3 CFG[20:19]) 00 = 01 = 10 = 11 = 4/5.32901.12 EN8 (CFG[4]) 0 = 10 bit operation (Default) 1 = 8 bit operation 4/5.32901.11 PLL_LBW (CFG[3]) 0 = High loop bandwidth 1 = Low loop bandwidth (Default) 4/5.32901.10:8 PLLMUL (CFG[2:0]) Full Rate (Default) Half Rate Quarter Rate Reserved PLL multiply factor. Can be calculated from following equation. LINERATE REFCLK(freq) + PLLMULTIPLY 2 000 = 5x 001 = 25x 010 = 10x (Default) 011 = 15x 100 = 4x 101 = 20x 110 = 8x 111 = Reserved 4/5.32901.7 ENTEST_TX (TX_CFG[1]) 1 = Tx Test Features Enabled (Default) 0 = Tx Test Features Disabled 4/5.32901.6 ENTEST_RX (RX_CFG[1]) 1 = Rx Test Features Enabled (Default) 0 = Rx Test Features Disabled 36 Access 0 = JOGCOM[0..7] controls whether comma alignment is enabled on lane 0..7 (Default) 1 = JOGCOM[0..7] induces an alignment jog on lane 0..7 Submit Documentation Feedback RW TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Table 61. SERDES_DATA_CONTROL (1) Address: 0x8086 Bit(s) Default: 0xFFFF Name Description 4/5.32902.15:8 Enable TX (TX_CFG[0]) 1 = Transmit data pair is enabled for channels 7~0 0 = Transmit data pair is disabled for channels 7~0 (Bit 15 corresponds to channel 7) 4/5.32902.7:0 Enable RX (RX_CFG[0] 1 = Receive data pair is enabled for channels 7~0. 0 = Receive data pair is disabled for channels 7~0 (Bit 7 corresponds to channel 7 ) (1) Access RW When power down mode is enabled using Control register (4/5.0), the SERDES macros go into power down mode where the TX and RX data pairs are disabled for all channels. When A side is powered down, TX and RX data pairs are disabled for channels 3~0. When B side is powered down TX and RX data pairs are disabled for channels 7~4. These low-power modes override the settings in this register. Bits 11:8 corresponds to A side and bits 15:12 corresponds to B side of TX path. Bits 3:0 corresponds to A side and bits 7:4 corresponds to B side of RX path. In normal mode(A side as primary channel, A_B = 1) all the bits needs to be enabled for the normal operation and when B side acts as primary channel(A_B = 0), A side bits can be disabled. Table 62. SERDES_PLL_CONTROL Address: 0x8087 Bit(s) Default: 0x8000 Name 4/5.32903.15 PLL Enable Description 1 = Enabled 0 = Disabled Access RW Table 63. SERDES_SYNC_STATUS Address: 0x8088 Bit(s) Default: 0x0000 Name Description 4/5.32904.15:8 Sync Synchronous detect. When high, indicates a comma character has been detected for lanes 7~0 (Feature not supported. Bits made available for future use) 4/5.32904.7:0 Loss of link Loss of link. When high, indicates that link at the receiver 7~0 lanes is lost (Feature not supported. Bits made available for future use) Access RO Table 64. SERDES_TESTFAIL_CONTROL Address: 0x8089 Bit(s) 4/5.32905.15 Name Test fail select Default: 0x8000 Description Test fail select controls. According to these selection bits, following Test fail signals are given out as testfail_mux_out to the Output Pins 0 = TX TESTFAIL selected 1 = RX TESTFAIL selected Submit Documentation Feedback Access RW 37 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Table 65. SERDES_TEST_CONFIG (1) Address: 0x808A Bit(s) Default: 0x0000 Name Description Access 4/5.32906.15:14 TESTCFG[15:14] Reserved 4/5.32906.13:12 EQTESTMD TESTCFG[13:12] 00 – Equalizer test mode disabled (Default) 01 – 16 Parametric devices observable 10 – 17 Parametric devises are observable 11 – High impedance mode 4/5.32906.11 PADLPBACK TESTCFG[11] 0 – Pad loop back disabled (Default) 1 – Pad loop back enabled TESTCFG[10] should be enabled else no effect 4/5.32906.10 LOOPBACK TESTCFG[10] 0 – Loop back disabled (Default) 1 – Loop back enabled ( For Internal Test Pattern Verification) 4/5.32906.9 BSIN Enable boundary scan inputs. This bit has no affect on transmit SERDES macro. 4/5.32906.8 BSOUT Enable boundary scan outputs. This bit has no affect on receive SERDES macro. 4/5.32906.7:4 AFR TESTCFG[7:4] Asynchronous frequency ramp mode. Refer to Table 66: Asynchronous frequency ramp mode (Default 4’b0000) 4/5.32906.3 TESTCLK TESTCFG[3] 0 – PLL bypass disabled (Default) 1 – PLL bypass enabled 4/5.32906.2:0 TESTPATT TESTCFG[2:0] 000 – 001 – 010 – 011 – 100 – 101 – 110 – 111 – (1) Test pattern gen/verification disabled (Default) Clock pattern gen/verification enabled 27– 1 PRBS gen/verification enabled 223– 1 PRBS gen/verification enabled Low frequency clock pattern Reserved Reserved Reserved Above control bits are only for vendor testing only. Customer should leave them at their default values. Table 66. Asynchronous Frequency Ramp Mode AFR 38 CFG[9] TESTCFG[7:4] RAMP MODE x 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 xxx0 0001 0011 0101 0111 1001 1011 1101 1111 0001 0011 0101 0111 1001 1011 1101 1111 Disabled – 521 ppm +521 ppm –390 ppm +390 ppm –195 ppm +195 ppm –98 ppm +98 ppm –521 ppm +521 ppm – 390 ppm +390 ppm – 195 ppm +195 ppm –695 ppm +695 ppm Submit Documentation Feedback RW TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Table 67. REDUNDANCY_CONTROL Address: 0x808B Bit(s) Default: 0x2700 Name Description Access 4/5.32907.15:13 External voltage divider Setting for External voltage divider for SERDES 1 control 1 [15, 14, 13] 0 0 0 – 0.50 v 0 0 1 – 0.40 v (Default) 0 1 0 – 0.45 v 0 1 1 – 0.35 v 1 0 0 – 0.475 v 1 0 1 – 0.375 v 1 1 0 – 0.425 v 1 1 1 – 0.325 v 4/5.32907.12:10 External voltage divider Setting for External voltage divider for SERDES 0 control 0 [12, 11, 10] 0 0 0 – 0.50 v 0 0 1 – 0.40 v (Default) 0 1 0 – 0.45 v 0 1 1 – 0.35 v 1 0 0 – 0.475 v 1 0 1 – 0.375 v 1 1 0 – 0.425 v 1 1 1 – 0.325 v 4/5.32907.9 HSTL input termination enable 1 Active LOW termination enable. When 0 enables termination resistance on HSTL inputs on B side. Default HIGH (termination disable) 4/5.32907.8 HSTL input termination enable 0 Active LOW termination enable. When 0 enables termination resistance on HSTL inputs on A side. Default HIGH (termination disable) 4/5.32907.7 Redundant XGMII mode When set, puts device in Redundant XGMII Mode 4/5.32907.6 Redundant XAUI mode When set, puts device in Redundant XAUI Mode 4/5.32907.5 XGMII 1 Tristate When set, 3-states the data outputs on the XGMII 1 side. 4/5.32907.4 IDLE When set, during non-retime mode, generates Idle on all lanes of the redundant XAUI channel. 4/5.32907.3 A/B select When set, channel A is selected as primary else channel B acts as primary channel. 4/5.32907.2 RETIME When set, device goes into retime mode else non-retime mode. (Default 0) 4/5.32907.1 XGMII 0 Tristate When set, 3-states the data outputs on the XGMII 0 side. 4/5.32907.0 Transition code Output code to be sent during transition from A to B or B to A. When 0 FE, when 1 LF is transmitted. RW Table 68. TRANSITION_TIME_CONTROL Address: 0x808C Bit(s) Default: 0x0000 Name 4/5.32908.15:0 Transition time Description Transition time interval control in 312.5MHz clock intervals (In Redundant XAUI Mode). Transition time interval control in 156.25MHz clock intervals (In Redundant XGMII Mode). Access RW Table 69. SERDES_JOGCOM_CONTROL (1) Address: 0x808E Bit(s) 4/5.32910.15:8 Name JOGCOM[7:0] RX_CFG[2] Default: 0x0000 Description Jog alignment/ Comma enable bit (one bit per lane). When CFG[7] is set to 0, this bit acts as comma alignment enable control. JOGCOM[i] = 0 : Comma alignment enabled for lane i JOGCOM[i] = 1 : Comma alignment disabled for lane i Access RW When CFG[7] is set to 1, this bit induces alignment jog on lanes 7..0 JOGCOM[i] = 0 : Alignment jog disabled for lane i JOGCOM[i] = 1 : Alignment jog enabled for lane i (1) Above control bits are for vendor testing only. Customer should leave them at their default values. Submit Documentation Feedback 39 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Table 70. DIE_ID_3 Address: 0x8090 Bit(s) Default: 0x0000 Name 4/5.32912.15:0 Die ID [63:48] Description Bits [63:48] of the Die ID. Unique TI DIE identifier. Access RO Table 71. DIE_ID_2 Address: 0x8091 Bit(s) Default: 0x0000 Name 4/5.32913.15:0 Die ID [47:32] Description Bits [47:32] of the Die ID. Unique TI DIE identifier. Access RO Table 72. DIE_ID_1 Address: 0x8092 Bit(s) Default: 0x0000 Name 4/5.32914.15:0 Die ID [31:16] Description Bits [31:16] of the Die ID. Unique TI DIE identifier. Access RO Table 73. : DIE_ID_0 Address: 0x8093 Bit(s) Default: 0x0000 Name 4/5.32915.15:0 Die ID [15:0] Description Bits [15:0] of the Die ID. Unique TI DIE identifier. Access RO Table 74. BIST_CONTROL Address: 0x80A0 Bit(s) Default: 0x0000 Name 4/5.32928.11 bist_enable Description When set enables memory BIST testing Access RW Table 75. BIST_STATUS Address: 0x80A1 Bit(s) Default: 0x0000 Name Description 4/5.32929.15 bist_fail_l When low, indicates memory BIST test has failed. 4/5.32929.14 bist_done When high, indicates memory BIST test has completed Access RO Table 76. RESERVED_CONTROL Address: 0x80A2 Bit(s) 4/5.32930.15 Default: 0x8000 Name Reserved Description Reserved Access RW JITTER TEST PATTERN GENERATION AND VERIFICATION Use one of the following procedures to generate and verify the respective jitter test pattern: • High Frequency Test Pattern: – Issue a hard or soft reset – Read the RX Local Fault bit (4/5/8.10) of the XS_STATUS_2 register to clear. – Read the RX Local Fault bit (4/5/8.10) of the XS_STATUS_2 register and verify it is cleared. This indicates that the RX link is up – Bypass the Lane Alignment by writing 1 to the RX Lane Align Bypass bit of the RX_BYPASS_CONTROL register. – Write 00 to the pattern_select field of the TEST_CONTROL register (4/5.25.1:0). – Start the pattern generation on the XAUI_TX and verification on the XAUI_RX by writing 1 to the test_enable bit of the TEST_CONTROL register (4/5.25.2). 40 Submit Documentation Feedback TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 • • • – Read the test pattern error counters for all channels (CHANNEL_0~3_ TEST_ERR_CNT), to clear the counters. – At this point the pattern verification is in progress and the errors are reported in the error counters. – Reading the counters has no effect on the test except clearing them, i.e. the verification of the pattern continues until the test_enable bit of the TEST_CONTROL register is cleared. Low Frequency Test Pattern: – Issue a hard or soft reset – Read the RX Local Fault bit (4/5/8.10) of the XS_STATUS_2 register to clear – Read the RX Local Fault bit (4/5/8.10) of the XS_STATUS_2 register and verify it is cleared. This indicates that the RX link is up. – Bypass the Lane Alignment by writing 1 to the RX Lane Align Bypass bit of the RX_BYPASS_CONTROL register. – Write 01 to the pattern_select field of the TEST_CONTROL register (4/5.25.1:0). – Start the pattern generation on the XAUI_TX and verification on the XAUI_RX by writing 1 to the test_enable bit of the TEST_CONTROL register (4/5.25.2). – Read the test pattern error counters for all channels (CHANNEL_0~3_ TEST_ERR_CNT), to clear the counters – At this point the pattern verification is in progress and the errors are reported in the error counters. – Reading the counters has no effect on the test except clearing them, i.e. the verification of the pattern continues until the test_enable bit of the TEST_CONTROL register is cleared. Mixed Frequency Test Pattern: – Issue a hard or soft reset – Read the RX Local Fault bit (4/5/8.10) of the XS_STATUS_2 register to clear. – Read the RX Local Fault bit (4/5/8.10) of the XS_STATUS_2 register and verify it is cleared. This indicates that the RX link is up – Bypass the Lane Alignment by writing 1 to the RX Lane Align Bypass bit of the RX_BYPASS_CONTROL register. – Write 10 to the pattern_select field of the TEST_CONTROL register (4/5.25.1:0). – Start the pattern generation on the XAUI_TX and verification on the XAUI_RX by writing 1 to the test_enable bit of the TEST_CONTROL register (4/5.25.2). – Read the test pattern error counters for all channels (CHANNEL_0~3_ TEST_ERR_CNT), to clear the counters. – At this point the pattern verification is in progress and the errors are reported in the error counters. – Reading the counters has no effect on the test except clearing them, i.e. the verification of the pattern continues until the test_enable bit of the TEST_CONTROL register is cleared. Continuous Random Test Pattern (CRPAT): – Issue a hard or soft reset – Read the test pattern error counter cr_cj_err_cnt registers (4/5.32778 – 4/5.32779) to clear – Write 1 to the crpat_enable bit of the Vendor Specific TEST_CONFIG register (4/5.32768.1). – Enable the CRPAT verifier by writing 1 to CRPAT Check Enable bit of the TEST_VERIFICATION_CONTROL register (4/5.32769.1) – In order for the Test Pattern Verifier to start checking the test pattern, it has to receive the Preamble /SFD that is sent at every packet from the test pattern generator. To make sure that the test pattern checking has started, read the 4/5/32801.15 (Test Pattern Status) bit of the Test Pattern Verification Status register. Make sure that the Test Pattern Sync bit is HIGH. If the sync status is not high, this indicates that the verifier never received the Preamble, which may indicate a more severe link problem. – Read Align Status (4/5.1.2) to clear old information. Discard the values read. – Read the CRPAT_CJPAT_TEST_ERROR_COUNT register. Align Status (4/5.1.2) has to read HIGH for the CRPAT_CJPAT_ERR_COUNTER to be valid as this test is packet based. If Align Status reads LOW, the test is not actively comparing data. – The CRPAT_CJPAT_TEST_ERROR_COUNT counter can be read as many times as desired. – If another test is to be performed go to the first step Submit Documentation Feedback 41 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 • Continuous Jitter Test Pattern (CJPAT): – Issue a hard or soft reset – Read the test pattern error counter cr_cj_err_cnt registers (4/5.32778 – 4/5.32779) to clear – Write 1 to the crpat_enable bit of the Vendor Specific TEST_CONFIG register (4/5.32768.0). – Enable the CJPAT verifier by writing 1 to CJPAT Check Enable bit of the TEST_VERIFICATION_CONTROL register (4/5.32769.0) – In order for the Test Pattern Verifier to start checking the test pattern, it has to receive the Preamble /SFD that is sent at every packet from the test pattern generator. To make sure that the test pattern checking has started, read the 4/5/32801.15 (Test Pattern Status) bit of the Test Pattern Verification Status register. Make sure that the Test Pattern Sync bit is HIGH. If the sync status is not high, this indicates that the verifier never received the Preamble, which may indicate a more severe link problem. – Read Align Status (4/5.1.2) to clear old information. Discard the values read. – Read the CRPAT_CJPAT_TEST_ERROR_COUNT register. Align Status (4/5.1.2) has to read HIGH for the CRPAT_CJPAT_ERR_COUNTER to be valid as this test is packet based. If Align Status reads LOW, the test is not actively comparing data. – The CRPAT_CJPAT_TEST_ERROR_COUNT counter can be read as many times as desired. – If another test is to be performed go to the first step If more than one test is specified results are unpredictable. SIGNAL PIN DESCRIPTION Table 77. Global Signals SIGNAL LOCATION VOLTAGE TYPE RST_N D08 VDDP/ VDDO 2.5V LVCMOS I Internal Pull Up Chip Reset (Active Low) When asserted (low logic level), this signal reinitializes the entire device. Must be asserted (low logic level) for at least 10 µs after device power up. REFSEL L18 VDDP/ VDDO 2.5V LVCMOS I Internal Pull Down Reference Select This input selects which reference clock input to use as the default internal clock (0 = REFCLK0, 1 = REFCLK1). REFINEN[1:0] E15, E14 VDDP/ VDDO 2.5V LVCMOS I Internal Pull Down Reference Enable This signal selects which reference clock source is to be used for a given XGXS core (Receive and Transmit Directions). 42 DESCRIPTION REFSEL REFINEN[0] Clock Selected for XGXS 0 0 0 REFCLK0 0 1 REFCLK0 1 0 REFCLK0 1 1 REFCLK1 REFSEL REFINEN[1] Clock Selected for XGXS 1 0 0 REFCLK1 0 1 REFCLK0 1 0 REFCLK1 1 1 REFCLK1 Submit Documentation Feedback TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Table 78. JTAG Signals SIGNAL LOCATION VOLTAGE TYPE DESCRIPTION TDI V19 VDDP/ VDDO 2.5V LVCMOS Input (Internal Pullup) JTAG Input Data. TDI is used to serially shift test data and test instructions into the device during the operation of the test port. TDO W20 VDDP/ VDDO 2.5V LVCMOS Output JTAG Output Data. TDO is used to serially shift test data and test instructions out of the device during operation of the test port. When the JTAG port is not in use, TDO is in a high impedance state. TMS U18 VDDP/ VDDO 2.5V LVCMOS Input (Internal Pullup) JTAG Mode Select. TMS is used to control the state of the internal test-port controller. TCK Y20 VDDP/ VDDO 2.5V LVCMOS Input JTAG Clock. TCK is used to clock state information and test data into and out of the device during the operation of the test port. TRST_N AB20 VDDP/ VDDO 2.5V LVCMOS Input (Internal Pull Down) JTAG Test Reset. TRST_N is used to reset the JTAG logic into system operational mode. Table 79. MDIO Related Signals SIGNAL LOCATION VOLTAGE TYPE DESCRIPTION MDEN0 AB03 VDDM0 1.2/2.5V LVCMOS Input Management Interface Enable Enables MDIO management port 0 when asserted (high). This must be tied to logic zero if management port 0 is unused. It is expected that MDIO Interface 0 and 1 are not both simultaneously active. MDEN1 F14 VDDM1 1.2/2.5V LVCMOS Input Management Interface Enable Enables MDIO management port 1 when asserted (high) . This must be tied to logic zero if management port 1 is unused. It is expected that MDIO Interface 0 and 1 are not both simultaneously active. MDC0 V04 VDDM0 1.2/2.5V LVCMOS Input Management Interface Clock (Port 0) This clock is used to sample the MDIO0 signal. This must be tied to logic zero if management port 0 is unused. MDIO0 AA03 VDDM0 1.2/2.5V LVCMOS Input/ Output Management Interface Data (Port 0) This bidirectional data line for MDIO Port 0 is sampled on the rising edge of MDC0. This must be pulled to logic zero if management port 0 is unused. MDC1 F16 VDDM1 1.2/2.5V LVCMOS Input Management Interface Clock (Port 1) This clock is used to sample the MDIO1 signal. This must be tied to logic zero if management port 1 is unused. MDIO1 F15 VDDM1 1.2/2.5V LVCMOS Input/ Output Management Interface Data (Port 1) This bidirectional data line for MDIO Port 1 is sampled on the rising edge of MDC1. This must be pulled to logic zero if management port 1 is unused. Table 80. Parallel Data Pins SIGNAL LOCATION VOLTAGE TYPE DESCRIPTION TX_CLK0_[3:0] D04, J06, F02, J02 VDDQ0/ VTT0/ VREF0 1.5V HSTL Input Transmit Data Clocks (XGMII 0) These four signals are the XGMII side input clocks per channel. Only TX_CLK0_[1] is used to sample all 36 data/control inputs. TX_CLK0_[3:2,0] are unused, and should be driven (not floating) in the application. TX_CLK1_[3:0] U20, P22, M20, K21 VDDQ1/ VTT1/ VREF1 1.5V HSTL Input Transmit Data Clocks (XGMII 1) These four signals are the XGMII side input clocks per channel. Only TX_CLK1_[1] is used to sample all 36 data/control inputs. TX_CLK1_[3:2,0] are unused, and should be driven (not floating) in the application. Submit Documentation Feedback 43 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Table 80. Parallel Data Pins (continued) SIGNAL LOCATION VOLTAGE TYPE TXD0_[31:0] C04, F05, B06, C05, E04, F04, G05, B04, H05, C03, E03, H04, F03, G03, B02, J05, D02, J04, E02, G02, J03, K05, H02, D01, F01, K03, G01, L06, H01, M05, J01, L04 VDDQ0/ VTT0/ VREF0 1.5V HSTL Input Transmit Data Pins (XGMII 0) Parallel interface data pins. TXD1_[31:0] U22, U21, U19, T18, T20, T19, R18, T21, R21, R19, R20, N22, P21, P20, P19, P18, M21, N19, N20, L22, K22, N18, J22, L21, L20, E22, D22, G22, C22, M19, J21, M18 VDDQ1/ VTT1/ VREF1 1.5V HSTL Input Transmit Data Pins (XGMII 1) Parallel interface data pins. TXC0_[7:0] E05, B03, H03, K06, B05, D03, C02, E01 VDDQ0/ VTT0/ VREF0 1.5V HSTL Input Transmit Data Control (XGMII 0) XGMII Control inputs. Bits 7:4 are unused, and 3:0 are the control bits per lane. TXC1_[7:0] W22, T22, N21, F22, V22, R22, M22, H22 VDDQ1/ VTT1/ VREF1 1.5V HSTL Input Transmit Data Control (XGMII 1) XGMII Control inputs. Bits 7:4 are unused, and 3:0 are the control bits per lane RX_CLK0_[3:0] Y02, V02, P05, M04 VDDQ0 1.5V HSTL Output Receive Data Clocks (XGMII 0) These four signals are the XGMII side output clocks per channel. Only RX_CLK0_[1] is used. RX_CLK0_[3:2,0] are driven static. RX_CLK1_[3:0] C12, E19, J18, K20 VDDQ1 1.5V HSTL Output Receive Data Clocks (XGMII 1) These four signals are the XGMII side output clocks per channel. Only RX_CLK1_[1] is used. RX_CLK1_[3:2,0] are driven static. RXD0_[31:0] Y03, W03, U05, V03, T05, U04, AB01, T06, T04, W02, R06, U02,R05, R04, Y01, W01, T02, P04, R03, V01, N04, U01, N05, R02, R01, P01, P02, N01, N02, M02, M01, M03 VDDQ0 1.5V HSTL Output Receive Data Pins (XGMII 0) Parallel interface data pins. 44 DESCRIPTION Submit Documentation Feedback TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Table 80. Parallel Data Pins (continued) SIGNAL LOCATION VOLTAGE TYPE DESCRIPTION RXD1_[31:0] D12, D13, E17, D15, D14, D16, D17, C10, D18, C17, H18, G18, E18, G19, D19, C18, G20, J19, C19, F20, H19, C20, K18, H20, D21, E21, G21, J20, H21, F21, B21, L19 VDDQ1 1.5V HSTL Output Receive Data Pins (XGMII 1) Parallel interface data pins. RXC0_[7:0] AB02, AA01, T03, N03, AA02, U03, P03, T01 VDDQ0 1.5V HSTL Output Receive Data Control (XGMII 0) XGMII Control inputs. Bits 7:4 are unused, and 3:0 are the control bits per lane. RXC1_[7:0] D11, F19, E20, K19, C11, F18, D20, C21 VDDQ1 1.5V HSTL Output Receive Data Control (XGMII 1) XGMII Control inputs. Bits 7:4 are unused, and 3:0 are the control bits per lane. Table 81. LED Interface Pins SIGNAL LOCATION VOLTAGE TYPE PLL_LOCK1 Y21 VDDP/ VDDO 2.5V LVCMOS Output 12mA PLL Lock Indicates when PLL 1 is locked. (Currently, this pin is for compatibility, and not supported in TLK3138, and is tied high internally). TLK3138 3-states this pin during RESET_N asserted. DESCRIPTION PLL_LOCK0 AB22 VDDP/ VDDO 2.5V LVCMOS Output 12mA PLL Lock Indicates when PLL 0 is locked. (Currently, this pin is for compatibility, and not supported in TLK3138, and ia tied high internally). TLK3138 3-states this pin during RESET_N asserted. LINKLED1 AA21 VDDP/ VDDO 2.5V LVCMOS Output 12mA Link Up Indicates when asserted XGXS 1 link is up. TLK3138 3-states this pin during RESET_N asserted. LINKLED0 AB21 VDDP/ VDDO 2.5V LVCMOS Output 12mA Link Up Indicates when asserted XGXS 0 link is up. TLK3138 3-states this pin during RESET_N asserted. TXLED1 AA22 VDDP/ VDDO 2.5V LVCMOS Output 12mA Transmit Packets Indicates when XGXS 1 is transmitting packets. TLK3138 3-states this pin during RESET_N asserted. This signal toggles at a 6-Hz rate when the chip is transmitting data on the XAUI interface. When not transmitting, this signal is deasserted. TXLED0 V21 VDDP/ VDDO 2.5V LVCMOS Output/ Input* 12mA Internal Pull Down RXLED1 W21 VDDP/ VDDO 2.5V LVCMOS Output 12mA Receive Packets Indicates when XGXS 1 is receiving packets. TLK3138 3-states this pin during RESET_N asserted. This signal toggles at a 6-Hz rate when the chip is receiving data on the XAUI interface. When not receiving, this signal is deasserted. RXLED0 Y22 VDDP/ VDDO 2.5V LVCMOS Output 12mA Receive Packets Indicates when XGXS 0 is receiving packets. TLK3138 3-states this pin during RESET_N asserted. This signal toggles at a 6 Hz rate when the chip is receiving data on the XAUI interface. When not receiving, this signal is deasserted. Transmit Packets Indicates when XGXS 0 is transmitting packets. Should have a pullup or pull down resistor to support bootstrap operation (see Table 82). TLK3138 3-states this pin during RESET_N asserted. This signal toggles at a 6-Hz rate when the chip is transmitting data on the XAUI interface. When not transmitting, this signal is deasserted. Submit Documentation Feedback 45 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Table 82. Bootstrap Interface Pins SIGNAL LOCATION VOLTAGE TYPE XGMII0_ST D06 VDDP/ VDDO 2.5V Not Used Not used, exists for compatibility. Tie High or Low in LVCMOS Application. Input Internal Pull Down MDIO0_ST E07 VDDP/ VDDO 2.5V Not Used Not used, exists for compatibility. Tie High or Low in LVCMOS Application. Input Internal Pull Down MDMX E06 VDDP/ VDDO 2.5V Not Used Not used, exists for compatibility. Tie High or Low in LVCMOS Application. Input Internal Pull Down PRTAD[4:0] C07, B08, C06, D05, B07 VDDP/ VDDO 2.5V Port Address Used to select the Device Id/Port ID in Clause 45 MDIO LVCMOS modes. Only the upper four bits are used. Input Internal Pull Down DEVID[4:0] AB22, W21, AA22, Y22, V21 VDDP/ VDDO Ishared DESCRIPTION Device ID Selects Device ID for the MDIO interface. Only the lower bit is used (Only DTE or PHY device ID is supported). Shared with PLL_LOCK0, RXLED1, TXLED1, RXLED0, and TXLED0. Table 83. Serial Side Data/Clock Pins SIGNAL LOCATION VOLTAGE REFCLKP0/ REFCLKN0 U10, V10 AVDD LVDS/ Reference Clock 156.25 MHz Differential Input Reference Clock 0. DPECL Input AC Coupled. TYPE DESCRIPTION REFCLKP1/ REFCLKN1 U16, V16 AVDD LVDS/ Reference Clock 156.25 MHz Differential Input Reference Clock 1. DPECL Input AC Coupled. TDP3_0/TDN3_0 TDP2_0/TDN2_0 TDP1_0/TDN1_0 TDP0_0/TDN0_0 AA08, AB08, W09, Y09, AA10, AB10, W11, Y11 AVDD CML Output Transmit Differential Pairs, XAUI Lane A (Lane 0), High speed serial outputs. Minimum bit time 320 pS. AC coupling required. TDP3_1/TDN3_1 TDP2_1/TDN2_1 TDP1_1/TDN1_1 TDP0_1/TDN0_1 AA16, AB16, W17, Y17, AA18, AB18, W19, Y19 AVDD CML Output Transmit Differential Pairs, XAUI Lane B (Lane 1), High speed serial outputs. Minimum bit time 320 pS. AC coupling required. RDP3_0/RDN3_0 RDP2_0/RDN2_0 RDP1_0/RDN1_0 RDP0_0/RDN0_0 W07, Y07, AA06, AB06, W05, Y05, AA04, AB04 AVDD CML Input Receive Differential Pairs, XAUI Lane A (Lane 0), High speed serial inputs with on-chip 100Ω differential termination. Each input pair is terminated differentially across an on-chip 100-Ω resistor. Minimum bit time 320 pS. RDP3_1/RDN3_1 RDP2_1/RDN2_1 RDP1_1/RDN1_1 RDP0_1/RDN0_1 W15, Y15, AA14, AB14, W13, Y13, AA12, AB12 AVDD CML Input Receive Differential Pairs, XAUI Lane B (Lane 1), High speed serial inputs with on-chip 100Ω differential termination. Each input pair is terminated differentially across an onchip 100-Ω resistor. Minimum bit time 320 pS. Table 84. Miscellaneous Pins 46 SIGNAL LOCATION VOLTAGE TYPE GPI4 GPI3 GPI0 R09, R14, R17 VDDP/ VDDO LVCMOS 2.5V Input GPO[5:0] D07, U08, U14, V08, V14, V20 VDDP/ VDDO LVCMOS 2.5V Output DESCRIPTION General Purpose Input– Must Be Grounded in the System Application. General Purpose Output– Must be N/C in the System Application. Submit Documentation Feedback TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Table 85. Voltage Supply and Reference Pins SIGNAL LOCATION TYPE DESCRIPTION DVDD A01, A04, A09, A14, A19, A22, E16, H08, H09, H10, H11, H12, H13, H14, H15, J08, J09, J10, J11, J12, J13, J14, J15, K01, K08, K15, L01, M06, M08, M15, P07 P Digital Core Power (1.2V) VDDP A06, A11, A12, A17, E11, G07, G09, G12, G16, J07, J16, L08, L15, L17, N07, N16, R16, AA20 P Pad Pre-Driver Power (2.5V) VDDO A05, A10, A13, A18, F13, G10, G13, G15, H07, H16, L07, L16 P CMOS Pad Driver Power (2.5V) VDDM[1:0] P16, P06 P Management Pad Power (2.5V, 1.2V) VDDM[1] is used to power MDEN1, MDIO1, MDC1. VDDM[0] is used to power MDEN0, MDIO0, MDC0. These voltages can be at either 2.5V or 1.2V independently. VDDQ1 A15, A20, G11, G17, H17, K17 P XGMII 1 Pad Power (1.5V) VDDQ0 A03, A08, C01, F07, F09, F11, G06, K04, N06 P XGMII 0 Pad Power (1.5V) VREF A02, A16, A21, F12, G14, N17, A07, B01, F10, H06, K07, M07, R07, M17, L05 P XGMII Input Pad Termination Power (.75V) Should track VDDQ1/2 and VDDQ0/2. VREF1 F08 P XGMII Input Pad Termination Power (.75V) Should track VDDQ1/2 and VDDQ0/2. VREF2 G04 P XGMII Input Pad Termination Power (.75V) Should track VDDQ1/2 and VDDQ0/2. VREF3 J17 P XGMII Input Pad Termination Power (.75V) Should track VDDQ1/2 and VDDQ0/2. VREF4 P17 P XGMII Input Pad Termination Power (.75V) Should track VDDQ1/2 and VDDQ0/2. DGND B09, B10, B11, B12, B13, B14, B15, B16, B17, B18, B19, B20, B22, C08, C09, C13, C14, C15, C16, D09, D10, E08, E09, E10, E12, E13, F06, F17, G08, K02, K09, K10, K11, K12, K13, K14, K16, L02, L03, L09, L10, L11, L12, L13, L14, M09, M10, M11, M12, M13, M14, M16, N08, N09, N10, N11, N12, N13, N14, N15, P08, P09, P10, P11, P12, P13, P14, P15, R10, R11, R12, R13 G Digital Ground AVDD U09, U17, V11, Y04, Y06, Y08, Y10, Y12, Y14, Y16, Y18, AB05, AB07, AB09, AB11, AB13, AB15, AB17, AB19 P Analog Power Supply (1.2V) AGND R08, R15, T07, T08, T09, T10, T11, T12, T13, T14, T15, T16, T17, U06, U07, U11, U12, U13, U15, V05, V06, V07, V09, V12, V13, V15, V17, V18, W04, W06, W08, W10, W12, W14, W16, W18, AA05, AA07, AA09, AA11, AA13, AA15, AA17, AA19 G Analog Ground Submit Documentation Feedback 47 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 01 02 03 04 05 06 07 08 09 10 11 A DVDD VREF VDDQ0 DVDD VDDO VDDP VREF VDDQ0 DVDD VDDO VDDP B VREF TXD0_17 TXC0_6 TXD0_24 TXC0_3 TXD0_29 PRTAD0 PRTAD3 DGND DGND DGND C VDDQ0 TXC0_1 TXD0_22 TXD0_31 TXD0_28 PRTAD2 PRTAD4 DGND DGND RXD1_24 RXC1_3 D TXD0_08 TXD0_15 TXC0_2 TX_CLK0 _3 PRTAD1 XGMII0_S T GPO5 RST_N DGND DNGD RXC1_7 DGND DGND DGND VDDP E TXC0_0 TXD0_13 TXD0_21 TXD0_27 TXC0_7 MDMX MDIO0_S T F TXD0_07 TX_CLK0 _1 TXD0_19 TXD0_26 TXD0_30 DGND VDDQ0 VREF1 VDDQ0 VREF VDDQ0 G TXD0_05 TXD0_12 TXD0_18 VREF2 TXD0_25 VDDQ0 VDDP DGND VDDP VDDO VDDQ1 H TXD0_03 TXD0_09 TXC0_5 TXD0_20 TXD0_23 VREF VDDO DVDD DVDD DVDD DVDD J TXD0_01 TX_CLK0 _0 TXD0_11 TXD0_14 TXD0_16 TX_CLK0 _2 VDDP DVDD DVDD DVDD DVDD K DVDD DGND TXD0_06 VDDQ0 TXD0_10 TXC0_4 VREF DVDD DGND DGND DGND L DVDD DGND DGND TXD0_00 VREF TXD0_04 VDDO VDDP DGND DGND DGND M RXD0_01 RXD0_02 RXD0_00 RX_CLK0 _0 TXD0_02 DVDD VREF DVDD DGND DGND DGND N RXD0_04 RXD0_03 RXC0_4 RXD0_11 RXD0_09 VDDQ0 VDDP DGND DGND DGND DGND VDDM0 DVDD DGND DGND DGND DGND P RXD0_06 RXD0_05 RXC0_1 RXD0_14 RX_CLK0 _1 R RXD0_07 RXD0_08 RXD0_13 RXD0_18 RXD0_19 RXD0_21 VREF AGND GPI4 DGND DGND T RXC0_0 RXD0_15 RXC0_5 RXD0_23 RXD0_27 RXD0_24 AGND AGND AGND AGND AGND U RXD0_10 RXD0_20 RXC0_2 RXD0_26 RXD0_29 AGND AGND GPO4 AVDD REFCLKP 0 AGND V RXD0_12 RX_CLK0 _2 RXD0_28 MDC0 AGND AGND AGND GPO2 AGND REFCLK N0 AVDD W RXD0_16 RXD0_22 RXD0_30 AGND RDP1_0 AGND RDP3_0 AGND TDP2_0 AGND TDP0_0 Y RXD0_17 RX_CLK0 _3 RXD0_31 AVDD RDN1_0 AVDD RDN3_0 AVDD TDN2_0 AVDD TDN0_0 AA RXC0_6 RXC0_3 MDIO0 RDP0_0 AGND RDP2_0 AGND TDP3_0 AGND TDP1_0 AGND AB RXD0_25 RXC0_7 MDEN0 RDN0_0 AVDD RDN2_0 AVDD TDN3_0 AVDD TDN1_0 AVDD Figure 30. Device Pinout Diagram (Part 1) (Top View) 48 Submit Documentation Feedback TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 12 13 14 15 16 17 18 19 20 21 22 VDDP VDDO DVDD VDDQ1 VREF VDDP VDDO DVDD VDDQ1 VREF DVDD B DGND DGND DGND DGND DGND DGND DGND DGND DGND RXD1_01 DGND C RX_CLK1 _3 DGND DGND DGND DGND RXD1_22 RXD1_16 RXD1_13 RXD1_10 RXC1_0 TXD1_03 D RXD1_31 RXD1_30 RXD1_27 RXD1_28 RXD1_26 RXD1_25 RXD1_23 RXD1_17 RXC1_1 RXD1_07 TXD1_05 REFINEN 1 DVDD RXD1_29 RXD1_19 RX_CLK1 _2 RXC1_5 RXD1_06 TXD1_06 A E DGND DGND REFINEN 0 F VREF VDD0 MDEN1 MDIO1 MDC1 DGND RXC1_2 RXC1_6 RXD1_12 RXD1_02 TXC1_4 G VDDP VDDO VREF VDDO VDDP VDDQ1 RXD1_20 RXD1_18 RXD1_15 RXD1_05 TXD1_04 H DVDD DVDD DVDD DVDD VDDO VDDQ1 RXD1_21 RXD1_11 RXD1_08 RXD1_03 TXC1_0 RXD1_14 RXD1_04 TXD1_01 TXD1_09 J DVDD DVDD DVDD DVDD VDDP VREF3 RX_CLK1 _1 K DGND DGND DGND DVDD DGND VDDQ1 RXD1_09 RXC1_4 RX_CLK1 _0 TX_CLK1 _0 TXD1_11 L DGND DGND DGND VDDP VDDO VDDP REFSEL RXD1_00 TXD1_07 TXD1_08 TXD1_12 M DGND DGND DGND DVDD DGND VREF TXD1_00 TXD1_02 TX_CLK1 _1 TXD1_15 TXC1_1 N DGND DGND DGND DGND VDDP VREF TXD1_10 TXD1_14 TXD1_13 TXC1_5 TXD1_20 P DGND DGND DGND DGND VDDM1 VREF4 TXD1_16 TXD1_17 TXD1_18 TXD1_19 TX_CLK1 _2 R DGND DGND GPI3 AGND VDDP GPI0 TXD1_25 TXD1_22 TXD1_21 TXD1_23 TXC1_2 T AGND AGND AGND AGND AGND AGND TXD1_28 TXD1_26 TXD1_27 TXD1_24 TXC1_6 U AGND AGND GPO3 AGND REFCLKP 1 AVDD TMS TXD1_29 TX_CLK1 _3 TXD1_30 TXD1_31 AGND AGND TDI GPO0 TXLED0 TXC1_3 TDP2_1 AGND TDP0_1 TDO RXLED1 TXC1_7 RXLED0 V AGND AGND GPO1 AGND REFCLK N1 W AGND RDP1_1 AGND RDP3_1 AGND Y AVDD RDN1_1 AVDD RDN3_1 AVDD TDN2_1 AVDD TDN0_1 TCK PLL_LOC K1 AA RDP0_1 AGND RDP2_1 AGND TDP3_1 AGND TDP1_1 AGND VDDP LINKLED 1 TXLED1 TRST_N LINKLED 0 PLL_LOC K0 AB RDN0_1 AVDD RDN2_1 AVDD TDN3_1 AVDD TDN1_1 AVDD Figure 31. Device Pinout Diagram (Part 2) (Top View) Submit Documentation Feedback 49 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 ABSOLUTE MAXIMUM RATINGS over operating free-air temperature range (unless otherwise noted) (1) VALUE Supply voltage (2) DVDD, AVDD, VTT, VREF -0.3 to 1.5 VDDQ -0.3 to 1.6 VDDP, VDDO, VDDM VI LVCMOS Input Voltage -0.3 to 2 -65 to 150 °C 2 KV CDM 500 V 0 to 70 °C Characterized free-air operating temperature range (2) V HBM Storage temperature (1) -0.3 to 3 -0.3 to Supply 3 HSTL CLASS 1 Electrostatic Discharge UNIT Stresses beyond those listed under absolute maximum ratings may cause permanent damage to the device. These are stress ratings only, and functional operation of the device at these or any other conditions beyond those indicated under recommended operating conditions is not implied. Exposure to absolute-maximum-rated conditions for extended periods may affect device reliability. All voltage values, except differential I/O bus voltages, are with respect to network ground terminal. RECOMMENDED OPERATING CONDITIONS MIN NOM MAX DVDD Core Supply Voltage 1.14 1.2 1.26 AVDD Analog Supply Voltage 1.14 1.2 1.26 VDDQ1/0 Parallel HSTL I/O Supply Votage (1.5V Application) 1.4 1.5 1.6 VDDP VDDO LVCMOS I/O Supply Voltage 2.38 2.5 2.63 VDDM1 Port 1 MDIO I/O Supply Voltage (2.5V Application) 2.38 2.5 2.63 VDDM0 Port 0 MDIO I/O Supply Voltage (2.5V Application) 2.38 2.5 2.63 VDDM1 Port 1 MDIO I/O Supply Voltage (1.2V Application) 1.14 1.2 1.26 VDDM0 Port 0 MDIO I/O Supply Voltage (1.2V Application) 1.14 1.2 1.26 VTT 1/0 VREF1/0 HSTL bias supply voltage (1.5V Application) VDDQ1/0 Divided By Two 0.7 0.75 0.8 IDVDD Digital Core Supply Current Rω = 156.25MHz (DVDD = 1.26V) 815 IVDDM/P/O LVCMOS I/O Supply Current Rω = 156.25MHz (VDDP/VDDO = 2.625V) 195 IVDDQ HSTL I/O Supply Current Rω = 156.25MHz (VDDQ = 1.6V) 450 IAVDD Analog Supply Current Rω = 156.25MHz (AVDD = 1.26V) 740 Total power consumption 1.5V HSTL I/O Mode HSTL Input Terminations Disabled 2.42 3.22 PD HSTL Input Terminations Enabled 2.72 3.63 Isda Analog Shutdown Current Isdd Core Shutdown Current UNIT V mA W 500 µA 11 mA REFERENCE CLOCK TIMING REQUIREMENTS (REFCLKP/N) (1) PARAMETER TEST CONDITIONS Frequency Normal Accuracy Relative to Nominal Accuracy Independent Channel Mode Duty Cycle Jitter (1) 50 Random and deterministic This clock should be crystal referenced to meet the requirements of the above table. Submit Documentation Feedback MIN TYP MAX UNIT 156.25 156.25 156.25 MHz –100 0 100 0 0 0 45% 50% 55% 40 ppm ps TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 REFERENCE CLOCK ELECTRICAL CHARACTERISTICS (REFCLKP/N) PARAMETER Vid Differential input voltage CIN Input capacitance RIN Input differential impedance Trise Rise time TEST CONDITIONS MIN TYP 100 80 20% to 80% 100 50 MAX UNIT 2000 mVPP 3 pF 120 Ω 600 ps MAX UNIT LVCMOS ELECTRICAL CHARACTERISTICS PARAMETER TEST CONDITIONS MIN VDDP/ VDDO–0.2 VOH High-level output voltage IOH = –100 µA, Driver Enabled VOL Low-level output voltage IOL = 100 µA, Driver Enabled VIH High-level input voltage 0.7 × VDDP/ VDDO VIL Low-level input voltage 0 IH, IL Low/High input current IZ Low input current CIN Input capacitance TYP 0.2 V 0.3 × VDDP/ VDDO ±1 ±20 Driver Only, driver disabled 5 µA pF MDIO ELECTRICAL CHARACTERISTICS PARAMETER TEST CONDITIONS MIN TYP MAX VIH High-level input voltage VDDM = 2.5V 2.1 VDDM+ 0.3 VIL Low-level input voltage VDDM = 2.5V 0 0.8 VIH High-level input voltage VDDM = 1.2V 1 VDDM+ 0.3 VIL Low-level input voltage VDDM = 1.2V 0 0.36 VOL Low-level output voltage VDDM = 2.5V 0 0.4 VOL Low=level output voltage VDDM = 1.2V 0 0.2 VOH High-level output voltage VDDM = 1.2/2.5V (Open Drain Driver) Pulled up on the customer board IH, IL Low/high input current IZ Low input current CIN Input capacitance Driver only, driver disabled UNIT V ±1 µA ±20 µA 5 pF HSTL SIGNALS (VDDQ1/0 = 1.5 V) PARAMETER TEST CONDITIONS MIN VOH(dc) High-level output voltage VDDQ–0.4 VOL(dc) Low-level output voltage VOH(ac) High-level output voltage VOL(ac) Low-level output voltage VIH(dc) High-level DC input voltage DC input, logic high VIL(dc) Low-level DC input voltage DC input, logic low VIH(ac) High-level AC input voltage AC input, logic high VIL(ac) Low-level AC input voltage AC input, logic low IOH(dc) High output current -8 IOL(dc) Low output current 8 CIN Input Capacitance TYP MAX UNIT VDDQ 0.4 VDDQ–0.5 VDDQ VREF + 0.1 VDDQ + 0.3 –0.3 VREF –0.1 0.5 VREF + 0.2 VREF – 0.2 mA 4 Submit Documentation Feedback V pF 51 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 SERIAL TRANSMITTER/RECEIVER CHARACTERISTICS PARAMETER TEST CONDITIONS VOD(pp) VOD(pd) TX output differential peak-to-peak voltage swing VOD(pp) MIN TYP MAX Maximum pre-emphasis Enabled Emphasized Bit See Figure 32, 4/5.32900.14:11 = 1000 (Binary) 1100 1280 1450 Maximum Pre-emphasis EnabledDe-emphasized Bit, See Figure 32 4/5.32900.14:11 = 1000 (Binary) 650 760 875 1250 1400 1575 Pre-emphasis disabled 4/5.32900.14:11 = 0000 (Binary) –10 µA 2 pF RX input leakage current CI RX input capacitance tr, tf Differential output signal rise, fall time (20% to 80%) RL = 50 Ω, CL = 5 pF, See Figure 32 JTOL Jitter Tolerance, Total Jitter at Serial Input Zero crossing, See Figure 35 0.6 JDR Serial Input Deterministic Jitter Zero crossing, See Figure 35 0.36 JT Serial Output Total Jitter PRBS at 3.125 GHz, See Figure 35 JD Serial Output Deterministic Jitter PRBS at 3.125 GHz, See Figure 35 0.17 R(LATENC Total delay from RX input to RD output See Figure 15 950 Total delay from TD input to TX output See Figure 14 650 160 0.20 0.35 Y) T(LATENC mVPP 10 ILKG 80 UNIT ps UI (1) Bit Times Y) (1) Unit Interval = 0ne serial bit time (min 320 ps) Table 86. Driver Template Parameters NEAR END VALUE FAR END VALUE X1 (See Figure 33) PARAMETER 0.175 0.275 X2 (See Figure 33) 0.390 0.400 A1 (See Figure 33) 400 100 A2 (See Figure 33) 800 800 UNIT UI mV VOD(p) VOD(d) VOD(pd) VCMT VOD(d) tr , tf bit time VOD(p) Figure 32. Transmit Output Waveform Parameter Definitions 52 Submit Documentation Feedback VOD(pp) TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 A2 Differential Amplitude - mV A1 0 -A1 -A2 X1 X2 1-X2 X2 1.0 Time (UI) Differential Amplitude - mV Figure 33. Transmit Template +50 0 -50 0.325 0.675 Unit Interval 1.0 Figure 34. Receive Template Submit Documentation Feedback 53 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 JDR JR JR JTOL NOTE: JTOL = JR + JDR, where JTOL is the receive jitter tolerance, JDR is the received deterministic jitter, and JR is the Gaussian random edge jitter distribution at a maximum BER = 10-12. Figure 35. Input Jitter HSTL OUTPUT SWITCHING CHARACTERISTICS (DDR) over recommended operating conditions (unless otherwise noted) MIN TYP MAX tsetup RXD0/1_[31:0], RXC0/1_[7:0]setup prior to RX_CLK transition high or low PARAMETER See Figure 36 TEST CONDITIONS 960 960 960 thold RXD0/1_[31:0], RXC0/1_[7:0] hold after RX_CLK transition high or low See Figure 36 960 960 960 Tduty RX_CLK_0/1_[3:0] Duty Cycle Source Centered and Source Aligned tperiod RX_CLK_0/1_[3:0] Period Source Centered and Source Aligned Tpd See Figure 37 RX_CLK_0/1_[3:0] rising or falling to RXD0/1_[31:0], RXC0/1_[7:0] valid 40% UNIT ps 60% 6.25 –640 ns 640 ps tPERIOD RX_CLK_0/1[3:0] VOH(ac) VOL(ac) tSETUP tSETUP tHOLD tHOLD VOH(ac) RXD0/1_[31:0],RXC0/1_[3:0] VOL(ac) Figure 36. HSTL (DDR Mode) Source Centered Output Timing Requirements 54 Submit Documentation Feedback TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 RX_CLK_0/1_[3:0] VOH(ac) VOL(ac) Tpd Tpd VOH(ac) RXD0/1_[31:0],RXC0/1_[3:0] VOL(ac) Figure 37. HSTL (DDR Mode) Source Aligned Output Timing Requirements HSTL (DDR MODE) INPUT TIMING REQUIREMENTS over operating free-air temperature range (unless otherwise noted) PARAMETER TEST CONDITIONS MIN TYP MAX tsetup TXD0/1_[31:0], TXC0/1_[3:0] setup prior to TX_CLK_0/1_[3:0] transition high or low See Figure 38 480 480 480 thold TXD0/1_[31:0], TXC0/1_[3:0] hold after TX_CLK_0/1_[3:0] transition high or low See Figure 38 480 480 480 Tduty TX_CLK_0/1_[3:0] Duty Cycle 40% 156.25 Mhz ±100 ppm tperiod TX_CLK_0/1_[3:0] Period UNIT ps 60% 6.25 ns tPERIOD VIH(ac) TX_CLK_0/1_[3:0] VREF VIL(ac) tSETUP tHOLD tHOLD tSETUP VIH(ac) TXD0/1_[31:0],TXC0/1_[3:0] V REF VIL(ac) Figure 38. HSTL (DDR Mode) Data Input Timing Requirements MDIO TIMING REQUIREMENTS over recommended operating conditions (unless otherwise noted) PARAMETER tperiod MDC0/1 period TEST CONDITIONS MIN See Figure 39 100 tsetup MDIO0/1 setup to ↑ MDC0/1 See Figure 39 10 thold MDIO0/1 hold to ↑ MDC0/1 See Figure 39 10 Tvalid MDIO0/1 valid from MDC ↑ 0 Submit Documentation Feedback TYP MAX UNIT nS 300 55 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 MDC0/1 tPERIOD tHOLD tSETUP MDIO0/1 Figure 39. MDIO Read/Write Timing VDDQ 50 W 100 W 50 W Transmission Line + 100 W VDDQ - GND RW RW VREF GND OUTPUT PCB INPUT Figure 40. HSTL I/O Schematic Table 87. Device Thermal Characteristics PACKAGE THERMAL THETA VALUES θJA Airflow (ft/min) Value 0 14.11 100 12.66 250 11.60 400 11.04 600 10.56 θJB 5.42 θJC 2.63 θJL Unit (°C/W) 3.57 Device Maximum Characteristics 56 Maximum Junction Temperature 115 °C Maximum Power Dissipation 3.63 Watts Submit Documentation Feedback TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 DEVICE TEST MODE Table 88 shows how testability pins are muxed with device pins. This is for TI only, and may be ignored by external users of this device. Table 88. Device Test Mode Pin Configuration Functional Functional Mode Device Pin Name Signal Direction (1) (2) (3) (4) Test Mode Signal Direction Functional Mode/JTAG Testing Scan Mode/RedJinn Canned TDL MEMBIST/ Efuse Mode Low Speed SERDES Diagnostics 1 GPI[4] I I 0 O 1 GPI[0] I I 0 1 0 1 PLL_LOCK1 O O — Scan Out 0 Efuse Data Out Test Failure Out 0 PLL_LOCK0 O O — Scan Out 1 Test Failure Out 1 LINKLED1 O O — Scan Out 2 Test Failure Out 2 LINKLED0 O O — Scan Out 3 Test Failure Out 3 TXLED1 O O — Scan Out 4 Test Failure Out 4 Test Failure Out 5 TXLED0 B I/O — Scan Enable RXLED1 O O — Scan Out 5 RXLED0 O O — Burnin Output Burnin Output (MBIST) Test Failure Out 7 REFINEN1 I I — Scan Config 0 Debug HS Clock (1) — REFINEN0 I I — Scan Config 1 RAM Clock Bypass (1) — PRTAD0 I I — Scan Clock MBIST Enable — PRTAD1 I I — Scan In 0 Efuse CTL MODE — PRTAD2 I I — Scan In 1 Efuse Clock — PRTAD3 I I — Scan In 2 Efuse Data In — PRTAD4 I I — Scan In 3 MBIST Resume (2) — XGMII0_ST I I — Scan In 4 MBIST Hold (3) — Test Failure Out 6 MBIST Debug (1) MDIO0_ST I I — Scan In 5 MDEN1 I I — Scan In 6 — — — MDMX I I TESTCLK T — TESTCLK T REFSEL I I — TESTCLK R — TESTCLK R GPI[3] I I IDDQ IDDQ IDDQ IDDQ GPO[5] O O MBIST Start Retention MBIST Start Retention MBIST Start Retention MBIST Start Retention GPO[4] O O MBIST Fail MBIST Fail MBIST Fail MBIST Fail GPO[3] O O MBIST Done MBIST Done MBIST Done MBIST Done GPO[2] O O GPO[1] O O MBIST Restart MBIST Restart MBIST Restart MBIST Restart GPO[0] O O LOW Scan Out 6 LOW LOW TCK (4) I I TCK TCK MBIST Diag Clock TCK MBIST Diag Scan MBIST Diag Scan MBIST Diag Scan MBIST Diag Scan Out Out Out Out This signal is internally connected to LOW (unasserted) when not available at external pins. This signal is internally connected to HIGH (asserted) when not available at external pins. This signal is internally connected to HIGH (unasserted) when not available at external pins. TCK is not a muxed test pin. TCK is used as MBIST diagnostic clock but still retains its functionality. Submit Documentation Feedback 57 TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Table 89. MEMBIST Control DEVICE MODE MDIO (1) JTAG (1) PIN (1) Functional Mode Yes Yes No Scan Mode No No No MEMBIST/Efuse Mode No No Yes Serdes Diagnostic Mode Yes Yes No Procedures for Testing Memory BIST (1) Yes– this control is enabled, No– this control is disabled TLK3138 supports Memory BIST (MBIST) execution through JTAG, PIN and MDIO controls. In all 3 modes MBIST_DONE and MBIST_FAIL_L signals are broadcast on to the pins GPO5 and GPO4 respectively. When MBIST is disabled (MBIST_EN = 0), both MBIST_DONE and MBIST_FAIL_L report LOW value. Before any MBIST testing the user must execute a hardware reset through the RST_N pin. This reset must be held asserted for at least 10µs. MBIST Through JTAG MBIST execution through JTAG allows the user to test the Memory cells inside TLK3138 without accessing many pins on the device. In addition to JTAG related pins, only REFCLK and RST_N pins need to be controlled to test the MBIST. MBIST execution is enabled through the ti_mbist_en instruction, and status is checked via the ti_mbist_status instruction. REFCLK acts as a source clock for the at-speed MBIST execution. The user has to provide a 156.25 MHz clock through REFCLK pins in order to execute MBIST. The following paragraphs explain each MBIST JTAG instruction in detail: ti_mbist_en: This JTAG instruction with op code 4’b1100 allows the user to control the internal MBIST_EN signal. Once the instruction is completed MBIST_EN assumes the TDI value during the update-dr state of instruction. When MBIST_EN is set to HIGH, MBIST execution starts. Once the test is completed and the status signals are checked, the ti_mbist_en instruction can be reissued with TDI set to a logic LOW value and MBIST_EN assumes a logic LOW value. ti_mbist_status: This JTAG instruction with op code 4’b1101 allows the user to check the status of the internal MBIST_DONE and MBIST_FAIL_L signals. Once MBIST_EN is set to HIGH, the status of the MBIST_DONE signal should be checked for a logic HIGH value, which indicates completion of the test and MBIST_FAIL_L signal should be checked for a logic HIGH value, which indicates MBIST execution completed without any errors. The MBIST_FAIL_L signal is LOW true (i.e if MBIST_FAIL_L is LOW, then it indicates that the MBIST execution completed with errors). The MBIST_DONE and MBIST_FAIL_L status signals are captured and scanned out through TDO at the next 2 TCK cycles respectively after entering into capture_dr state. When MBIST is disabled (MBIST_EN = 0), both MBIST_DONE and MBIST_FAIL_L reports logic LOW values. MBIST Through MDIO MBIST can be tested through MDIO by following MDIO instructions. • Write 1’b1 to bit 11 of BIST_CONTROL register (0x80A0) to enable MBIST test. • Read MBIST_DONE status by reading bit 14 of BIST_STATUS register (0x80A1). If the MBIST test is finished then this bit reads HIGH. Once MBIST_DONE is reported HIGH, read MBIST_FAIL_L status by reading bit 15 of BIST_STATUS register (0x80A1). If MBIST test completes without any errors this bit reads HIGH. If any errors are encountered during MBIST tests this bit reads back LOW. Repeat MBIST Testing In order to run repeat MBIST tests the device requires a software initiated or hardware reset to clear the MBIST_DONE signal between executions. For powering down individual blocks and the I/O buffers based on the mode pins (GPI[4], GPI[0]) and the IDDQ mode, the following truth table applies. 58 Submit Documentation Feedback TLK3138 www.ti.com SLLS762A – FEBRUARY 2007 – REVISED APRIL 2007 Table 90. Power Down Scheme IDDQ GPI[4] GPI[0] Software Power Down Operational Mode 0 0 0 0 Normal functional mode. All IO buffers are powered all clocks are enabled. X 0 0 1 Software initiated powerdown. All IO buffers that can be disabled or powered down are disabled/powered down. All clocks are disabled. REDJINN transmit and receive macros and the REDJINN PLL are disabled. 1 0 0 X IDDQ pin initiated powerdown. All I/O buffers that can be disabled or powered down are disabled/powered down. All clocks are disabled. REDJINN transmit and receive macros and the REDJINN PLL are disabled. X 1 1 X Serdes Test Mode. All IO buffers are powered all clocks are enabled. X 1 0 X EFUSE mode. All IO buffers are powered, all clocks are enabled. 0 0 1 X Normal Scan test mode. All IO buffers are powered all clocks are enabled. 1 0 1 X IDDQ Scan test mode. All non scan IO buffers that can be disabled or powered down are disabled/powered down. All clocks are enabled MDIO CHANGES Following are the changes in MDIO programmer’s reference. REGISTER ADRESS BIT(s) CHANGE 1 0x8026 11 Added XGMII Rx Source centered/aligned control bit 2 0x8085 7:6 Added ENTEST bits 3 0x808B 5 Added XGMII 0 3-state control bit. Bit 0 for XGMII 0 Tristate control 4 0x808B 7 Added XGMII Redundancy mode control bit 5 0x808B 9/8 Added HSTL_TERMINATION_EN control bits 6 0x808D All Removed Composite status bits 7 0x809(8-A) All Removed VTP Control registers 8 0x8084 6 Default changed to 1. Ext reference VREF is used to set output amplitude 9 0x808B 15:10 Added SERDES external voltage divider control bits. 10 0x80A0 11 Added BIST control bits 11 0x80A1 15:14 Added BIST status bits 12 0x80A2 15 Added Reserved bit. JTAG LIMITATION RST_N must be asserted (and released if desired) before executing JTAG instructions. Submit Documentation Feedback 59 PACKAGE OPTION ADDENDUM www.ti.com 24-Apr-2007 PACKAGING INFORMATION Orderable Device Status (1) Package Type Package Drawing TLK3138ZDU ACTIVE BGA ZDU Pins Package Eco Plan (2) Qty 484 60 Pb-Free (RoHS) Lead/Ball Finish SNAGCU MSL Peak Temp (3) Level-4-260C-72 HR (1) The marketing status values are defined as follows: ACTIVE: Product device recommended for new designs. LIFEBUY: TI has announced that the device will be discontinued, and a lifetime-buy period is in effect. NRND: Not recommended for new designs. Device is in production to support existing customers, but TI does not recommend using this part in a new design. PREVIEW: Device has been announced but is not in production. Samples may or may not be available. OBSOLETE: TI has discontinued the production of the device. (2) Eco Plan - The planned eco-friendly classification: Pb-Free (RoHS), Pb-Free (RoHS Exempt), or Green (RoHS & no Sb/Br) - please check http://www.ti.com/productcontent for the latest availability information and additional product content details. TBD: The Pb-Free/Green conversion plan has not been defined. Pb-Free (RoHS): TI's terms "Lead-Free" or "Pb-Free" mean semiconductor products that are compatible with the current RoHS requirements for all 6 substances, including the requirement that lead not exceed 0.1% by weight in homogeneous materials. Where designed to be soldered at high temperatures, TI Pb-Free products are suitable for use in specified lead-free processes. Pb-Free (RoHS Exempt): This component has a RoHS exemption for either 1) lead-based flip-chip solder bumps used between the die and package, or 2) lead-based die adhesive used between the die and leadframe. The component is otherwise considered Pb-Free (RoHS compatible) as defined above. Green (RoHS & no Sb/Br): TI defines "Green" to mean Pb-Free (RoHS compatible), and free of Bromine (Br) and Antimony (Sb) based flame retardants (Br or Sb do not exceed 0.1% by weight in homogeneous material) (3) MSL, Peak Temp. -- The Moisture Sensitivity Level rating according to the JEDEC industry standard classifications, and peak solder temperature. Important Information and Disclaimer:The information provided on this page represents TI's knowledge and belief as of the date that it is provided. TI bases its knowledge and belief on information provided by third parties, and makes no representation or warranty as to the accuracy of such information. Efforts are underway to better integrate information from third parties. TI has taken and continues to take reasonable steps to provide representative and accurate information but may not have conducted destructive testing or chemical analysis on incoming materials and chemicals. TI and TI suppliers consider certain information to be proprietary, and thus CAS numbers and other limited information may not be available for release. In no event shall TI's liability arising out of such information exceed the total purchase price of the TI part(s) at issue in this document sold by TI to Customer on an annual basis. Addendum-Page 1 IMPORTANT NOTICE Texas Instruments Incorporated and its subsidiaries (TI) reserve the right to make corrections, modifications, enhancements, improvements, and other changes to its products and services at any time and to discontinue any product or service without notice. Customers should obtain the latest relevant information before placing orders and should verify that such information is current and complete. All products are sold subject to TI’s terms and conditions of sale supplied at the time of order acknowledgment. TI warrants performance of its hardware products to the specifications applicable at the time of sale in accordance with TI’s standard warranty. Testing and other quality control techniques are used to the extent TI deems necessary to support this warranty. Except where mandated by government requirements, testing of all parameters of each product is not necessarily performed. TI assumes no liability for applications assistance or customer product design. Customers are responsible for their products and applications using TI components. To minimize the risks associated with customer products and applications, customers should provide adequate design and operating safeguards. TI does not warrant or represent that any license, either express or implied, is granted under any TI patent right, copyright, mask work right, or other TI intellectual property right relating to any combination, machine, or process in which TI products or services are used. Information published by TI regarding third-party products or services does not constitute a license from TI to use such products or services or a warranty or endorsement thereof. Use of such information may require a license from a third party under the patents or other intellectual property of the third party, or a license from TI under the patents or other intellectual property of TI. Reproduction of information in TI data books or data sheets is permissible only if reproduction is without alteration and is accompanied by all associated warranties, conditions, limitations, and notices. Reproduction of this information with alteration is an unfair and deceptive business practice. TI is not responsible or liable for such altered documentation. Resale of TI products or services with statements different from or beyond the parameters stated by TI for that product or service voids all express and any implied warranties for the associated TI product or service and is an unfair and deceptive business practice. TI is not responsible or liable for any such statements. TI products are not authorized for use in safety-critical applications (such as life support) where a failure of the TI product would reasonably be expected to cause severe personal injury or death, unless officers of the parties have executed an agreement specifically governing such use. Buyers represent that they have all necessary expertise in the safety and regulatory ramifications of their applications, and acknowledge and agree that they are solely responsible for all legal, regulatory and safety-related requirements concerning their products and any use of TI products in such safety-critical applications, notwithstanding any applications-related information or support that may be provided by TI. Further, Buyers must fully indemnify TI and its representatives against any damages arising out of the use of TI products in such safety-critical applications. TI products are neither designed nor intended for use in military/aerospace applications or environments unless the TI products are specifically designated by TI as military-grade or "enhanced plastic." Only products designated by TI as military-grade meet military specifications. Buyers acknowledge and agree that any such use of TI products which TI has not designated as military-grade is solely at the Buyer's risk, and that they are solely responsible for compliance with all legal and regulatory requirements in connection with such use. TI products are neither designed nor intended for use in automotive applications or environments unless the specific TI products are designated by TI as compliant with ISO/TS 16949 requirements. Buyers acknowledge and agree that, if they use any non-designated products in automotive applications, TI will not be responsible for any failure to meet such requirements. Following are URLs where you can obtain information on other Texas Instruments products and application solutions: Products Applications Amplifiers amplifier.ti.com Audio www.ti.com/audio Data Converters dataconverter.ti.com Automotive www.ti.com/automotive DSP dsp.ti.com Broadband www.ti.com/broadband Interface interface.ti.com Digital Control www.ti.com/digitalcontrol Logic logic.ti.com Military www.ti.com/military Power Mgmt power.ti.com Optical Networking www.ti.com/opticalnetwork Microcontrollers microcontroller.ti.com Security www.ti.com/security Low Power Wireless www.ti.com/lpw Telephony www.ti.com/telephony Video & Imaging www.ti.com/video Wireless www.ti.com/wireless Mailing Address: Texas Instruments, Post Office Box 655303, Dallas, Texas 75265 Copyright © 2007, Texas Instruments Incorporated