Transmit driver architecture with JTAG configuration mode, extended equalization range, and multiple power domains

The transmit driver architecture addresses signal integrity and reliability issues in high-speed serial communication by using a current mode driver for mission mode and voltage mode driver for JTAG, with extended equalization range and multiple power domains, ensuring efficient operation and reduced failure rates.

JP2024544150AActive Publication Date: 2024-11-28XILINX INC
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
JP2024529227
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Priority Date
2021-11-16
Filing Date
2022-07-21
Publication Date
2024-11-28
Estimated Expiration
2042-07-21

AI Technical Summary

Technical Problem

High-speed serial communication links face challenges with signal integrity due to electrical degradation and impedance mismatches, and existing transmit driver architectures experience performance degradation during JTAG configuration and reliability issues under large voltage swings.

Method used

A transmit driver architecture with a current mode driver for mission mode and voltage mode driver for JTAG configuration, featuring extended equalization range, multiple power domains, and additional circuitry for reliability, including power switches and programmable configurability to maintain signal integrity and reliability.

Benefits of technology

Enhances signal integrity at high data rates (up to 112 Gbps) by providing efficient JTAG configuration without performance degradation and ensures reliability through multiple power domains and power switches, supporting complex designs with improved yield and reduced failure rates.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024544150000001_ABST
    Figure 2024544150000001_ABST
Patent Text Reader

Abstract

A transmit driver architecture having a test mode (e.g., JTAG configuration mode), an extended equalization range, and / or multiple power domains. An example of a transmit driver circuit generally includes one or more driver unit cells having a differential input node pair configured to receive an input data signal and having a differential output node pair configured to output an output data signal, a plurality of power switches coupled between the differential output node pair and one or more power rails, a first set of one or more drivers coupled between a first test node of the differential test data path and a first output node of the differential output node pair, and a second set of one or more drivers coupled between a second test node of the differential test data path and a second output node of the differential output node pair.
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] TECHNICAL FIELD Examples of the present disclosure relate generally to electronic circuits, and more particularly to transmit driver circuits. [Background technology]

[0002] The use of high-speed serial communication links in electronic systems continues to grow. In some cases, serial links may be used for inter-integrated circuit (IC) communication within an electrical device (e.g., a serial bus between a memory and a processing system) and / or between electrical devices (e.g., a serial bus between a computer and a wearable device). In other cases, serial links may be used for inter-circuit communication within a system-on-chip (SoC). High-speed serial communication links may operate according to various standards, such as Universal Serial Bus (USB), High Definition Multimedia Interface (HDMI), Serial Advanced Technology Attachment (SATA), and Peripheral Component Interconnect Express (PCIe) interfaces. Serializers / Deserializers (SerDes) may be used to transmit and receive data over serial communication links. A SerDes transmitter serializes a multi-bit word into a serial data stream of corresponding bits. A SerDes receiver deserializes the received serial data stream back into the multi-bit word. In some SerDes systems, a clock signal is transmitted along with the serial data stream, while in other SerDes systems the clock signal is instead embedded in the serial data stream. Summary of the Invention

[0003] Certain aspects of the present disclosure relate generally to a transmit (TX) driver architecture for a serializer / deserializer (SerDes) transmitter or the like.

[0004] Some aspects of the present disclosure provide a transmit driver circuit that generally includes one or more driver unit cells having a differential input node pair configured to receive an input data signal and a differential output node pair configured to output an output data signal, a plurality of power switches coupled between the differential output node pair and one or more power rails, a first set of one or more drivers coupled between a first test node of a differential test data path and a first output node of the differential output node pair, and a second set of one or more drivers coupled between a second test node of the differential test data path and a second output node of the differential output node pair.

[0005] Some aspects of the disclosure provide a method of driving data. The method generally includes receiving an input data signal at a differential input node pair of a driver unit cell of a transmit driver circuit, generating an output data signal using the driver unit cell, and outputting the output data signal at a differential output node pair of the transmit driver circuit. The transmit driver circuit generally includes a plurality of power switches coupled between the differential output node pair and one or more power rails, a first set of one or more drivers coupled between a first test node of the differential test data path and a first output node of the differential output node pair, and a second set of one or more drivers coupled between a second test node of the differential test data path and a second output node of the differential output node pair.

[0006] These and other aspects can be understood with reference to the following detailed description.

[0007] So that the above features briefly summarized above may be understood in detail, a more particular description can be made by reference to exemplary implementations, some of which are illustrated in the accompanying drawings. It should be noted, however, that the accompanying drawings illustrate only typical example implementations, and therefore should not be considered as limiting the scope of the claims. [Brief description of the drawings]

[0008] [Figure 1] 1 is a block diagram of an example serializer / deserializer (SerDes) circuit having a SerDes transmitter and a SerDes receiver in which aspects of the present disclosure may be implemented. [Diagram 2] FIG. 2 is a block diagram of an example SerDes transmitter having a current-mode driver in accordance with certain aspects of the present disclosure. [Figure 3A] 1 is a circuit diagram illustrating an example implementation of a transmit driver circuit having a JTAG input in accordance with certain aspects of the present disclosure. [Figure 3B] FIG. 3B illustrates an example operation of the transmit driver circuit of FIG. 3A during JTAG mode in accordance with certain aspects of the disclosure. [Figure 3C] 1 is a circuit diagram illustrating an example implementation of a transmit driver circuit having multiple power domains and power switches in accordance with certain aspects of the disclosure. [Figure 3D] FIG. 3D is a circuit diagram illustrating the power switches of FIG. 3C and an additional switch for reliability during power down mode for the transmit driver circuit, according to some aspects of the disclosure. [Figure 4] 1 is a block diagram and table illustrating a transmit driver architecture with programmable configurability for extended equalization range in accordance with some aspects of the disclosure. [Diagram 5] 4 is a flow diagram of example operations for driving data in accordance with some aspects of the present disclosure. [Figure 6] FIG. 1 is a block diagram illustrating an example architecture for a programmable device in which aspects of the present disclosure may be implemented. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] For ease of understanding, wherever possible, identical reference numbers have been used to designate identical elements common to the figures. It is contemplated that elements of one example may be beneficially incorporated in other examples without specific recitation.

[0010] Some aspects of the present disclosure generally relate to a transmit (TX) driver architecture for a serializer / deserializer (SerDes) transmitter or the like. The TX driver architecture may function as a current-mode driver in mission mode, but support JTAG configuration as a voltage-mode driver at high data rates (e.g., 128 Gbps). In addition, the TX driver architecture may have an extended equalization range for both pre-cursor and post-cursor. In some examples, the TX driver architecture may have multiple power domains and additional circuitry for increased reliability, such as during power-down modes.

[0011] Various features are described below with reference to the drawings. It should be noted that the drawings may or may not be drawn to scale, and elements of similar structure or function are represented by similar reference numerals throughout the drawings. It should be noted that the drawings are intended only to facilitate the description of features. The drawings are not intended as an exhaustive description or as limitations on the scope of the claims. In addition, an illustrated example need not have all aspects or advantages shown. An aspect or advantage described in connection with a particular embodiment is not necessarily limited to that embodiment and may be implemented in any other embodiment, even if not so illustrated or explicitly described.

[0012] Example SerDes Circuit 1 is a block diagram of an example SerDes circuit 100 in which aspects of the present disclosure may be implemented. As shown, the SerDes circuit 100 may include a serializer 102, a SerDes transmitter 104 (labeled “TX”), a phase-locked loop (PLL) 106 for providing a transmit clock, a differential pair of serial channels 108A, 108B, a SerDes receiver 110 (labeled “RX”), a clock data recovery (CDR) circuit 112, a receive-side PLL 114, and a deserializer 116.

[0013] The serializer 102 converts data received on the parallel interface to a serial interface and provides a serial data signal (e.g., a bit stream) to the SerDes transmitter 104. The SerDes transmitter 104 may include a driver that may amplify the serial data signal and output a differential signal on the serial channels 108A, 108B. The driver may be implemented as a voltage mode driver or as a current mode logic (CML) driver. In some aspects, the SerDes transmitter 104 may also perform equalization to improve the signal integrity of the SerDes circuit 100, such as by compensating for (or at least adjusting for) losses in the channels 108A, 108B. The differential signal output by the SerDes transmitter 104 may be, for example, a low voltage differential signal (LVDS). In some aspects, the serial channel 108A may carry a non-inverted signal of a differential pair, and the serial channel 108B may carry an inverted signal (i.e., a complementary signal) of the differential pair. In some aspects, the differential signal output by the SerDes transmitter 104 may be sampled and generated according to a transmit clock output by the PLL 106. The PLL 106 may be used, for example, as a clock synthesizer circuit to generate a high-speed master clock used in serial data transmission.

[0014] The SerDes receiver 110 can receive a differential signal of serial data on the serial channels 108A, 108B and amplify the received differential signal. In some aspects, the SerDes receiver 110 can perform equalization of the received differential signal in addition to the equalization performed by the SerDes transmitter 104. The CDR circuit 112 is coupled to the output of the SerDes receiver 110 and performs timing recovery to recover a master clock used to sample the signal and generate a receive clock at the SerDes receiver 110. In some aspects, the CDR circuit 112 can receive a reference clock signal via the PLL 114. The deserializer 116 converts the serial signal output by the SerDes receiver 110 to a parallel format.

[0015] Signal integrity in transmission lines has been a design concern for many years and remains so as data rates continue to increase for data traversing such transmission lines. As signals travel between interconnects, electrical degradation of the signal can occur as the signal passes through different components. These waveform distortions can be caused by impedance mismatches such as stubs and vias, frequency dependent attenuation, and electromagnetic coupling (e.g., crosstalk) between signal traces. As designs become more sophisticated with additional channels added to increase data rates, overall signal integrity can become even more important. An exemplary modern serial link may use SerDes at each end of the link, with each endpoint having its own transmitter and receiver.

[0016] With transmitters and receivers for a single SerDes circuit residing in different components, the transmitters and receivers may have different types and amounts of equalization. Thus, the SerDes transmitter 104 and / or receiver 110 may incorporate an equalizer, such as a feed-forward equalizer (FFE), a continuous-time linear equalizer (CTLE), or a decision feedback equalizer (DFE), to compensate for or at least adjust for channel anomalies. The equalization method may be linear or nonlinear. In a microelectronic circuit implementation, the equalizer may be either continuous-time (non-sampled) or discrete-time (sampled). The signal amplitude may be discrete (digital) or continuous (analog). The equalizer may also be either adaptive or fixed, and the adaptation algorithm may be, for example, zero-forcing (ZF), least mean square (LMS) (or minimum mean square error (MMSE)), or some nonlinear approach. Furthermore, the target response of the equalizer may be either a full response or a partial response. The equalizer filter can be designed using either a Finite Impulse Response (FIR) filter or an Infinite Impulse Response (IIR) filter. Each equalization scheme has its advantages and disadvantages for a particular application.

[0017] Transmit equalization (e.g., may be used in field programmable gate arrays (FPGAs) and may be referred to as "pre-emphasis") pre-distorts the transmit signal by amplifying the high frequency content of the signal to compensate for, or at least adjust for, the expected amount of loss through the channel. One of the most common techniques applied in TX driver design for equalization is to add pre-cursors and post-cursors to the main signal. In general, better signal integrity can be achieved with a higher range of equalization. A typical range of pre-cursors and post-cursors is 0.28, but as data rates increase (e.g., to 112 Gbps), higher pre-cursor / post-cursor equalization may be associated. Some examples of the present disclosure provide a wider range of equalization (up to 0.40) that provides enhanced signal integrity at higher data rates (e.g., 112 Gbps).

[0018] Example TX Driver Architecture 2 is a block diagram of an example SerDes transmitter 200 implemented with a current-mode driver according to some aspects of the disclosure. The SerDes transmitter 200 may be used as a SerDes transmitter (e.g., SerDes transmitter 104) in a SerDes circuit (e.g., SerDes circuit 100). As shown in FIG. 2, the SerDes transmitter 200 includes one or more data path cells 202, one or more driver unit cells 204, and a load consisting of resistive elements R1, R2.

[0019] In some aspects, the datapath cells 202 may receive input data for transmission and perform signal processing on the received input data. For example, this signal processing may include modulation (e.g., into four-level phase amplitude modulation (PAM4)), equalization (e.g., using one or more multi-tap FIR filters), and / or digital-to-analog conversion (e.g., using one or more current-steering digital-to-analog converters (DACs)). The datapath cells 202 may output differential voltage signals on differential node pairs (labeled “Vinp” and “Vinn”) to the driver unit cells 204. In some aspects, the processed data may be buffered by one or more buffers (also referred to as pre-drivers) before being output to the driver unit cells 204.

[0020] The driver unit cell 204 may be implemented as a current-mode logic (CML) driver, as shown. A CML driver includes a differential transconductance stage configured to receive a differential voltage signal on a differential input node pair Vinp / Vinn and generate a differential output current on a differential output node pair (labeled “txp” and “txn”). As shown in FIG. 2, the transconductance stage may be implemented by n-type transistors Mcw_p and Mcw_n. The driver unit cell 204 also includes a current sink “I bias 2. The transistors Mcw_p and Mcw_n may include a differential input node pair Vinp / Vinn, respectively, and the drains of the transistors Mcw_p and Mcw_n may be connected to load resistive elements R1, R2. The resistive elements R1, R2 may also be coupled to a power supply rail 211 (labeled “avcc”). In some examples, the power supply rail 211 may have a voltage of 0.88V. In some aspects, the resistive elements R1, R2 are adjustable. In this case, either or both of the resistive elements R1, R2 may be implemented by a switched array of resistive elements.

[0021] 2, the driver unit cell 204 may also include an inductive element between the resistive elements R1, R2 and the differential output node pair. In some aspects, the inductive element may be coupled to an electrostatic discharge (ESD) protection circuit for protecting the driver unit cell 204 from ESD.

[0022] The Joint Test Action Group (JTAG) standard has been one of the most efficient and popular ways to enable engineers and others to perform extensive debugging and diagnostics on a system using a small number of dedicated test pins. In JTAG, signals are continuously scanned in and out of the input / output (I / O) cells of a device to control the inputs of the device and test the outputs of the device, sometimes under various conditions. One typical way to implement JTAG on the transmitter side is by adding an additional data path in the transmitter data path that allows the data to be configured between actual data and test signal data. However, as data rates increase and designs become more complex, applying such an additional data path for JTAG in the data path chain may consume excessive power due to the additional load in the data path. Some examples of the present disclosure may solve this problem by implementing a voltage mode topology for the TX driver circuit in test / debug mode (e.g., JTAG mode) and a current mode topology for the TX driver circuit in mission mode, thereby avoiding any performance degradation of the data path or driver during mission mode.

[0023] Furthermore, as TX driver architectures become more complex, reliability may become an increasingly important concern to ensure that designs do not have low yields during mass production or fail within a specified lifetime. Because current-mode driver topologies may experience large voltage swings, driver unit cells may have reliability issues under different swing conditions. Some examples of the present disclosure avoid this issue by providing two different power domains for the TX driver circuit to avoid any large gate-source voltage (Vgs) or drain-source voltage (Vds) issues. In some examples, the TX driver circuit may also include several switches to ensure that reliability is maintained in various power sequence scenarios (e.g., in power-down mode).

[0024] Example TX Driver with JTAG Configuration Support As described above, some examples of the present disclosure provide a transmit (TX) driver circuit capable of operating in a voltage-mode topology in a test / debug mode (e.g., JTAG mode) and in a current-mode topology in a mission mode, thereby avoiding any degradation in the performance of the data path or driver during the mission mode, e.g., supporting JTAG configuration at 128 Gbps data rates.

[0025] 3A is a circuit diagram illustrating an example implementation of a transmit driver circuit 300 having JTAG input circuits 3020 and 3021 (collectively referred to herein as “JTAG input circuits 302”) in accordance with some aspects of the disclosure. In addition to the JTAG input circuit 302, the transmit driver circuit 300 includes a number of driver unit cells 204, inductive elements L1, L2, L3, and L4, ESD protection circuits 3031 and 3032, resistive elements R1 and R2, and a power switch 306.

[0026] JTAG input circuit 3020 includes a set of one or more drivers (e.g., buffers or inverters 3040, 3041), and JTAG input circuit 3021 includes a set of one or more drivers (e.g., buffers or inverters 3042, 3043). Inverters 3040, 3041, 3042, and 3043 are collectively referred to herein as "inverters 304." Inverter 304 has a power input coupled to power rail 311 (labeled "avccaux"). An input of inverter 3040 is coupled to a test data path node of the differential JTAG input node pair (a first test node labeled "dslow"), and an input of inverter 3043 is coupled to a complementary test data path node of the differential JTAG input node pair (a second test node labeled "dslow_b"). The output of inverter 3040 is coupled to the input of inverter 3041, whose output is coupled to node 312, which may be coupled to the txp node (e.g., via resistive element R1 and inductive elements L1, L2). The output of inverter 3042 is coupled to the input of inverter 3043, whose output is coupled to node 313, which may be coupled to the txn node (e.g., via resistive element R2 and inductive elements L3, L4).

[0027] 3A, resistive element R1 is implemented as a switched array (also called a "switched resistor array" or "switched resistor network") of resistive elements 305. The switched array of resistive elements 305 includes n+1 branches, where n is a positive integer, one branch includes resistive element R1n and the other branches include respective p-type transistors ("MR0" through "MR n-1 ") and resistor elements (labeled "R10" through "R1 n-1"). By selectively turning on and off certain transistors, different resistances can be selected for resistive element R1, thereby implementing a tunable resistive element. Although not shown in FIG. 3A, resistive element R2 may additionally or alternatively be implemented as a switched array of resistive elements having the same number n or a different number of branches.

[0028] The power switch 306 may be coupled between a power supply rail 311 and output nodes 312, 313 of the JTAG input circuit 302. The power switch 306 may be implemented by p-type transistors M7, M8 as shown in FIG. 3A. In this case, the sources of transistors M7, M8 may be coupled to the power supply rail 311 and the gates of transistors M7, M8 may be coupled to a power switch control node for a control signal to control the operation of transistors M7, M8. The drain of transistor M7 may be coupled to an output node 312 and the drain of transistor M8 may be coupled to an output node 313.

[0029] As described above, the transmit driver circuit 300 may support PAM4. Therefore, the transmit driver circuit 300 includes three driver unit cells 204, namely, the LSB <0> Driver unit cell for 2040, MSB <1> Driver unit cell 2041 for MSB <0> The three driver unit cells 204 may be replicated versions of each other. Each driver unit cell may include a pre-driver 320, n-type transistors Mcw_p and Mcw_n (as described above with respect to FIG. 2), pull-up p-type transistors Mpull_p and Mpull_n, and current sink n-type transistors Mcsp and Mcsn biased with a control signal labeled Vb,mn. The reader will understand that the transmit driver circuit may include more or less than three driver unit cells 204.

[0030] 3A , the JTAG input circuit 302 may be disabled or powered down when not in use (e.g., during mission mode) and the transmit driver circuit 300 may function as a current-mode driver, receiving power from the power rails 311 and generating a differential output current at the differential output node pair txp / txn. The load capacitance may be high at the differential output node pair txp / txn, and therefore the additional loading of the JTAG input circuit 302 may not significantly affect the driver performance in mission mode.

[0031] 3B is a diagram illustrating an example operation of the transmit driver circuit 300 of FIG. 3A during JTAG mode, according to some aspects of the disclosure. During JTAG mode, the power switch 306, the driver unit cells 204, and the datapath cells 202 may be turned off or otherwise effectively disabled, as shown in FIG. 3B. In this manner, the test path data in JTAG mode is input via the JTAG input circuit 302 having the test datapath nodes dslow / dslow_b, rather than through the datapath cells 202, and the transmit driver circuit 300 functions as a voltage-mode driver to the differential output node pair txp / txn through the inverter 304, resistive elements R1, R2, and inductive elements L1-L4. This voltage-mode JTAG configuration may allow the test path to operate at maximum efficiency.

[0032] Exemplary TX Driver with Multiple Power Domains and Switches for Reliability As explained above, because the current-mode driver architecture supports a relatively large voltage swing, there may be reliability concerns regarding the driver unit cells 204 due to the large drain-source voltage (Vds) seen by the driver unit cells. Some examples of the present disclosure provide a transmit (TX) driver circuit with two different power domains to avoid issues with large gate-source voltage (Vgs) or Vds.

[0033] 3C is a circuit diagram illustrating an example implementation of a transmit driver circuit 350 having multiple power domains and multiple power switches in each branch according to some aspects of the disclosure. The transmit driver circuit 350 is similar to the transmit driver circuit 300 of FIG. 3A, except that the JTAG input circuit 302 is optional in the transmit driver circuit 350 of FIG. 3C. The transmit driver circuit 350 includes p-type transistors M7 and M8 coupled between a power rail 311 (labeled "avccaux") and nodes 312, 313, respectively. The transmit driver circuit 350 also includes an additional power rail 351 (labeled "avtt") and p-type transistors M9 and M10 coupled between the power rail 351 and nodes 312, 313, respectively.

[0034] The power rail 311 may have a voltage greater than the power rail 351. For example, the power rail 311 may have an (avccaux) voltage of 1.5V, and the power rail 351 may have an (avtt) voltage of 1.2V. During operation of the transmit driver circuit 350, the nodes 312, 313 may be coupled to the power rail 311 through the transistors M7 and M8 in a relatively large swing scenario (e.g., Vds of the driver unit cell 204 is >0.8V). However, in a smaller swing scenario (e.g., Vds <0.8V), the nodes 312, 313 may be coupled to the power rail 351 through the transistors M9 and M10. Such operation not only saves power under low swing conditions, but also improves reliability (in large and small swing scenarios) and prevents the driver unit cell 204 from failing over time.

[0035] Similarly, while the transmit driver circuit is powered down, it may be desirable to ensure that the transmit driver circuit does not have reliability issues during the power down mode. Therefore, the power switches (e.g., transistors M7, M8, M9, and M10) may be designed using thick oxide transistors. Furthermore, as shown in FIG. 3D, several switches may be added to the transmit driver circuit 350 to maintain reliability in the power down mode. For example, one branch of the transmit driver circuit of FIG. 3D (including node 312) includes p-type transistors M11, M12, and M13, and inverters 356 and 358. This same branch of the transmit driver circuit of FIG. 3D also includes n-type transistor M14 and inverter 360. It should be understood that these components may be replicated for the other branch of the transmit driver circuit (including node 313).

[0036] As shown in FIG. 3D, p-type transistor M7 has a source and a body coupled to power supply rail 311 (labeled “avccaux”), a drain coupled to node 312, and a gate coupled to an output (labeled “enb_aux”) of inverter 356. A data input of inverter 356 receives an input labeled “en_aux” and a power input of inverter 356 is coupled to power supply rail 311. In this manner, transistor M7 may be turned on when en_aux is asserted. Transistor M11 has a source coupled to power supply rail 351 (labeled “avtt”), a body coupled to power supply rail 311 via pull-up resistor R11, a gate coupled to power supply rail 311, and a drain coupled to the output of inverter 356. Thus, transistor M11 may be used to prevent any significant leakage current in the transmit driver circuit through the power switch circuit associated with transistor M7 in a power down mode.

[0037] Additionally, transistor M11 may be used to prevent significant leakage during power-up scenarios where the order in which the voltages avccaux and avtt on the power rails 311 and 351, respectively, are pulled up may be random. For example, in a scenario where avtt is on (e.g., 1.2V) but avccaux is still off (e.g., 0V), transistor M11 may be used to pull up the gate of transistor M7 to avtt, preventing significant leakage through transistor M7 that may occur in this scenario without transistor M11.

[0038] P-type transistor M9 has a source coupled to power rail 351, a body coupled to power rail 311 through pull-up resistor R9, a drain coupled to node 312, and a gate coupled to node 371 (labeled "enb_avtt_sw"). Transistor M12 has a source coupled to power rail 351, a body coupled to power rail 311 through pull-up resistor R12, a gate coupled to a complementary power-on register control node (labeled "porb_avtt"), and a drain coupled to node 371. Transistor M13 has a drain coupled to node 371, a gate coupled to the power-on register control node (labeled "por_avtt"), and a source and body coupled to the output of inverter 358 (labeled "enb_avtt"). A data input of inverter 358 receives an input labeled "en_avtt" and a power input of inverter 358 is coupled to power rail 311.

[0039] Transistor M14 has a drain coupled to node 312, a source and a body coupled to a reference potential node (e.g., electrical ground) for the transmit driver circuit, and a gate coupled to the output of inverter 360. Inverter 360 has a data input coupled to power rail 211 (labeled "avcc") and a power input coupled to a voltage approximately half the avccaux voltage (e.g., 0.75V). In this manner, transistor M14 is used to ensure that the output voltage at differential output node pair txp / txn is avccaux / 2 in power down mode to avoid any reliability issues due to large Vds for driver unit cell 204 while powering down the transmit driver circuit. The gate voltages of the transistors in the switched array of resistive elements may be in the avccaux voltage domain to avoid large Vgs issues during power down of the transmit driver circuit.

[0040] Example TX Driver with Extended Pre-Cursor and Post-Cursor Ranges As presented above, some examples of the present disclosure provide a transmit driver architecture with a larger range of equalization (up to 0.40) that provides enhanced signal integrity at higher data rates (e.g., 112 or 224 Gbps). As shown in block diagram 400 and table 420 of FIG. 4, the transmit driver architecture (more specifically, for example, data path cell 202 of FIG. 2) may include a control module 402 for controlling the transmit driver swing and equalization (e.g., using a finite impulse response (FIR) filter). This architecture uses two main cursor slices 408, 406 (labeled "Main0" and "Main1", respectively), one post-cursor slice 404 (labeled "Post1"), and three precursor slices 410, 412, 414 (labeled "Pre1", "Pre2", and "Pre3", respectively), which are controlled by the control module 402. The control module 402 and the (pre- / post-) cursor slices 404, 406, 408, 410, 412, and 414 may be implemented, for example, in a digital signal processor (DSP), a field programmable gate array (FPGA) or other programmable logic device, an application specific integrated circuit (ASIC), or a general purpose processor.

[0041] As the original cursor before any swapping (labeled "ORI cursor" in table 420), the main cursor slices 408, 406 may use twice the current (and have twice the cursor weighting) of the precursor slices 410, 412, 414 and the post-cursor slice 404. For example, the main cursor weighting may be 50% and the precursor and post-cursor weighting may be 25% (i.e., half). Table 420 also indicates in the row labeled "Weight" that for a PAM4 implementation as described above, the most significant bit (MSB) is weighted twice as much as the least significant bit (LSB). In the "Weight" row of table 420 for each cursor slice, the left column represents the relative weighting for the MSB and the right column represents the relative weighting for the LSB.

[0042] To achieve an equalization range up to 0.40, a programmable design technique may be implemented, as shown in table 420 of FIG. 4. In this programmable design, a precursor slice (e.g., Pre1) and / or a post-cursor slice (e.g., Post1) may be configured as one of the main cursor slices, and one of the main cursor slices may instead be configured as a precursor or post-cursor slice. Thus, the precursor slice 410 (Pre1) may function as a main cursor slice with a 25% cursor weighting (rather than the nominal 50% cursor weighting of the main cursor slice), and / or the post-cursor slice 404 (Post1) may function as a main cursor slice with a 25% cursor weighting. Additionally, the main cursor slice 408 may function as a precursor slice with a 50% cursor weighting (rather than the nominal 25% cursor weighting of the precursor slice), and / or the main cursor slice 408 may function as a post-cursor slice with a 50% cursor weighting (rather than the nominal 25% cursor weighting of the post-cursor slice). Because the main cursor slices 406, 408 support a higher amount of current, this programmability feature allows the transmit driver architecture to achieve 0.40 level equalization without using additional cursor slices, thereby saving power and area while still maintaining easy-to-design configurability. This programmable configurability of the cursor slices may be controlled by the control module 402 and may depend on the characteristics of the system in which the transmit driver circuit resides.

[0043] Example TX driver operation 5 is a flow diagram of example operations 500 for driving data in accordance with some aspects of the disclosure. The operations 500 may be performed by a transmit driver circuit (e.g., the transmit driver circuit 300 or 350).

[0044] The operations 500 may begin at block 502 by receiving an input data signal at a differential input node pair (e.g., differential input node pair Vinp / Vinn) of a driver unit cell of a transmit driver circuit (e.g., driver unit cell 204). At block 504, the transmit driver circuit may generate an output data signal using the driver unit cell and output the output data signal at a differential output node pair (e.g., differential output node pair txp / txn) of the transmit driver circuit (e.g., driver unit cell 204). The transmit driver circuit includes a plurality of power switches (e.g., power switch 306) coupled between the differential output node pair and one or more power rails (e.g., power rails 311, 351), a first set (e.g., JTAG input circuit 3020) of one or more drivers (e.g., inverter 304) coupled between a first test node (e.g., node dslow) of the differential test data path and a first output node (e.g., node txp) of the differential output node pair, and a second set (e.g., JTAG input circuit 3021) of one or more drivers (e.g., inverter 304) coupled between a second test node (e.g., node dslow_b) of the differential test data path and a second output node (e.g., node txn) of the differential output node pair.

[0045] According to some aspects, the operations 500 may further involve powering off the driver unit cells and the plurality of power switches at block 508. In some aspects, the transmit driver circuit may receive input test signals at the differential test data paths at block 510, drive the input test signals with one or more drivers of a first set and one or more drivers of a second set at block 512, and output output test signals at differential output node pairs at block 514, while the driver unit cells and the plurality of power switches are powered off (e.g., as in FIG. 3B). In this case, the input test signals may include joint test action group (JTAG) signals.

[0046] According to some aspects, the plurality of power switches include a first switch (e.g., transistor M7) coupled between a first power rail (e.g., power rail 311) of the one or more power rails and a first output node of the differential output node pair, a second switch (e.g., transistor M9) coupled between a second power rail (e.g., power rail 351) of the one or more power rails and a first output node of the differential output node pair, the first power rail having a first voltage (e.g., avccaux=1.5V) different from a second voltage (e.g., avtt=1.2V) of the second power rail, a third switch (e.g., transistor M8) coupled between the first power rail and a second output node of the differential output node pair, and a fourth switch (e.g., transistor M10) coupled between the second power rail and the second output node of the differential output node pair. In this case, generating the output data signal may involve powering the transmit driver circuit with the first and third switches closed and the second and fourth switches open. Based on the swing conditions of the transmit driver circuit (e.g., a low swing scenario), the transmit driver circuit may be changed to be powered with the second and fourth switches closed and the first and third switches open.

[0047] According to some aspects, the operations 500 further include generating an input data signal for equalization based on two main cursors (e.g., main cursor slices 406, 408), one post-cursor (e.g., post-cursor slice 404), and three precursors (e.g., precursor slices 410, 412, 414). In this case, generating the input data signal may include at least one of: programming a first one of the main cursors (e.g., Main0) to function as one of the precursor cursors having a cursor weighting (e.g., 50%) higher than a nominal cursor weighting (e.g., 25%) of one of the precursor cursors (e.g., Pre1); programming a second one of the main cursors (e.g., Main1) to function as a post cursor (e.g., Post1) having a cursor weighting (e.g., 50%) higher than a nominal cursor weighting (e.g., 25%) for the post cursor; programming one of the precursor cursors (e.g., Pre1) to function as a first one of the main cursors with a cursor weighting (e.g., 25%) lower than a nominal cursor weighting (e.g., 50%) of the first one of the main cursors; or programming a post cursor to function as a second one of the main cursors (e.g., having a cursor weighting lower than a nominal cursor weighting for the second one of the main cursors).

[0048] Exemplary Programmable Integrated Circuit The transmit driver architecture described herein may be implemented, for example, in a transmit driver of a serializer / deserializer (SerDes) system. A SerDes transmitter (e.g., SerDes transmitter 104) may be included in any of a variety of suitable devices or systems, such as an integrated circuit (IC) or module.

[0049] One type of IC that may utilize one or more SerDes transmit drivers (e.g., SerDes transmitter 104) is a programmable IC, such as a field programmable gate array (FPGA). An FPGA typically includes an array of programmable tiles. These programmable tiles may include, for example, input / output blocks (IOBs), configurable logic blocks (CLBs), dedicated random access memory blocks (BRAMs), multipliers, digital signal processing blocks (DSPs), processors, clock managers, and delay-locked loops (DLLs). Another type of programmable IC is a complex programmable logic device, or CPLD. A CPLD includes two or more "functional blocks" connected together by an interconnect switch matrix and connected to input / output (I / O) resources. Each functional block in a CPLD includes a two-level AND / OR structure similar to those used in programmable logic array (PLA) and programmable array logic (PAL) devices. Other programmable ICs are programmed by applying processing layers, such as metal layers, that programmably interconnect various elements on the device. These programmable ICs are known as mask-programmable devices. The phrase "programmable IC" can also encompass devices that are only partially programmable, such as application specific integrated circuits (ASICs).

[0050] 6 is a block diagram illustrating an example architecture 600 for a programmable device. Architecture 600 may be implemented, for example, in a field programmable gate array (FPGA). As shown, architecture 600 includes several different types of programmable circuits (e.g., logic blocks). For example, architecture 600 may include a number of different programmable tiles including multi-gigabit transceivers ("MGTs") 601, configurable logic blocks ("CLBs") 602, blocks of random access memory ("BRAMs") 603, input / output blocks ("IOBs") 604, configuration and clocking logic ("CONFIG / CLOCKS") 605, digital signal processing (DSP) blocks 606, dedicated input / output blocks ("I / O") 607 (e.g., configuration ports and clock ports), and other programmable logic 608, such as digital clock managers, analog-to-digital converters (ADCs), system monitoring logic, etc.

[0051] In some FPGAs, each programmable tile includes a programmable interconnect element (INT) 611 with standardized connections between corresponding interconnect elements 611 in each adjacent tile. The INTs 611 taken together thus implement the programmable interconnect structure of the illustrated FPGA. Each INT 611 also includes connections to and from programmable logic elements within the same tile, as shown by the example included at the right edge of FIG.

[0052] For example, the CLB 602 may include a single INT 611 in addition to a configurable logic element (CLE) 612 that may be programmed to implement a user logic circuit. The BRAM 603 may include a BRAM logic element (BRL) 613 in addition to one or more INTs 611. Typically, the number of INTs 611 included in a tile depends on the width of the tile. In the depicted embodiment, the width of the BRAM tile is equal to five CLBs, although other numbers (e.g., four) may be used. The DSP block 606 may include a DSP logic element ("DSPL") 614 in addition to any appropriate number of INTs 611. The IOB 604 may include, for example, two instances of an I / O logic element (IOL) 615 in addition to one instance of an INT 611. As will be appreciated by those skilled in the art, for example, the actual I / O pads connected to the I / O 615 are typically not limited to the area of ​​the I / O 615.

[0053] In the exemplary architecture 600 depicted in Figure 1, a horizontal region near the center of the die (shown as shaded in Figure 6) is used for configuration, clocks, and other control logic (CONFIG / CLOCKS 105). Another vertical region 609 extending from this central region can be used to distribute clocks and configuration signals across the width of the FPGA.

[0054] Some FPGAs utilizing the architecture 600 shown in Figure 6 include additional logic blocks that interrupt the regular row structure that makes up most of the FPGA. The additional logic blocks may be programmable blocks and / or dedicated logic circuits. For example, a processor block shown as PROC 610 spans several rows of CLB 602 and BRAM 603.

[0055] PROC 610 may be implemented as a hardwired processor that is fabricated as part of a die that implements the programmable circuitry of an FPGA. PROC 610 may represent any of a variety of different processor types and / or systems, ranging in complexity from an individual processor (e.g., a single core capable of executing program code) to an entire processing system having one or more cores, modules, co-processors, interfaces, etc.

[0056] In more complex configurations, for example, PROC 610 may include unidirectional and / or bidirectional interfaces configurable to couple directly to one or more cores (e.g., central processing units), cache memories, memory controllers, I / O pins (e.g., I / O pads) of an IC, and / or to programmable circuitry of an FPGA. The phrase "programmable circuitry" may refer to programmable circuit elements within an IC (e.g., various programmable or configurable circuit blocks or tiles described herein), as well as interconnect circuitry that selectively couples the various circuit blocks, tiles, and / or elements according to configuration data loaded into the FPGA. For example, portions shown in FIG. 6 that are external to PROC 610 may be considered part of the programmable circuitry of an FPGA.

[0057] FIG. 6 is intended to illustrate an example architecture 600 that may be used to implement an FPGA including a programmable circuit (e.g., programmable fabric) and a processing system. For example, the number of logic blocks in a row, the relative width of the row, the number and order of the rows, the types of logic blocks included in a row, the relative sizes of the logic blocks, and the interconnect / logic implementations included on the right side of FIG. 6 are exemplary. In an actual FPGA, for example, two or more adjacent rows of CLBs 602 are included where CLBs typically appear to facilitate efficient implementation of a user circuit design. However, the number of adjacent CLB rows may vary with the overall size of the FPGA. Additionally, the size and / or positioning of PROC 610 within the FPGA is for illustrative purposes only and is not intended as a limitation of one or more examples of the present disclosure.

[0058] The methods disclosed herein comprise one or more steps or actions for achieving the described method. The method steps and / or actions may be interchanged with one another without departing from the scope of the claims. In other words, unless a specific order of steps or actions is specified, the order and / or use of specific steps and / or actions may be modified without departing from the scope of the claims.

[0059] The various processes of the methods described above may be performed by any suitable means capable of performing the corresponding process functions. Such means may include various hardware and / or software components and / or modules, including, but not limited to, circuits, field programmable gate arrays (FPGAs) or other programmable logic, application specific integrated circuits (ASICs), or processors. Generally, where there are operations illustrated in figures, those operations may have corresponding means-plus-function components with similar numbering.

[0060] In the above, reference is made to the embodiments presented in the present disclosure. However, the scope of the present disclosure is not limited to the specific described embodiments. Instead, any combination of the described features and elements, whether related to different embodiments or not, is contemplated for implementing and practicing the contemplated embodiments. Furthermore, the embodiments disclosed herein may achieve other possible solutions or advantages over the prior art, but whether or not a particular advantage is achieved by a given embodiment does not limit the scope of the present disclosure. Thus, the foregoing embodiments, features, and advantages are merely exemplary and are not considered elements or limitations of the appended claims unless expressly recited in the claims. In other words, other and further examples may be devised without departing from the basic scope of the present disclosure, the scope of which is determined by the following "claims".

Claims

1. 1. A transmit driver circuit, comprising: one or more driver unit cells having a differential input node pair configured to receive an input data signal and having a differential output node pair configured to output an output data signal; a plurality of power switches coupled between the differential output node pair and one or more power rails; a first set of one or more drivers coupled between a first test node of the differential test data path and a first output node of the differential output node pair; a second set of one or more drivers coupled between a second test node of the differential test data path and a second output node of the differential output node pair; a transmit driver circuit comprising:

2. 2. The transmit driver circuit of claim 1, wherein the differential test data path comprises a Joint Test Action Group (JTAG) data path.

3. 2. The transmit driver circuit of claim 1, wherein the one or more driver unit cells and the plurality of power switches are configured to be powered off during the test mode for the transmit driver circuit, such that the transmit driver circuit is configured to function as a voltage-mode driver during the test mode.

4. a first set of one or more resistive elements coupled between the outputs of the first set of one or more drivers and the first output node of the differential output node pair; 2. The transmit driver circuit of claim 1, further comprising: a second set of one or more resistive elements coupled between the outputs of the second set of one or more drivers and the second output node of the differential output node pair.

5. 5. The transmit driver circuit of claim 4, wherein at least one of the first set of one or more resistive elements or the second set of one or more resistive elements comprises a switched array of resistive elements.

6. The plurality of power switches a first switch coupled between a first power rail of the one or more power rails and the first output node of the differential output node pair; a second switch coupled between a second power supply rail of the one or more power supply rails and the first output node of the differential output node pair, the first power supply rail having a first voltage that is different from a second voltage of the second power supply rail; a third switch coupled between the first power supply rail and the second output node of the differential output node pair; a fourth switch coupled between the second power supply rail and the second output node of the differential output node pair.

7. 2. The transmit driver circuit of claim 1, wherein the first set of one or more drivers and the second set of one or more drivers each comprise a plurality of inverters coupled in series.

8. 10. The transmit driver circuit of claim 1, further comprising: a control module configured to generate the input data signal based on two main cursors, one post-cursor, and three pre-cursors for equalization.

9. 9. The transmit driver circuit of claim 8, wherein the control module is configured to program the post-cursor to function as one of the main cursors with a cursor weighting that is lower than a nominal cursor weighting for the one of the main cursors.

10. 9. The transmit driver circuit of claim 8, wherein the control module is configured to program a first one of the main cursors to function as the one of the precursors having a cursor weighting higher than a nominal cursor weighting for the one of the precursors.

11. 11. The transmit driver circuit of claim 10, wherein the control module is configured to program a second one of the main cursors to function as the post-cursor with a cursor weighting higher than a nominal cursor weighting for the post-cursor.

12. 10. A programmable integrated circuit (IC) comprising a serializer / deserializer, the serializer comprising the transmit driver circuit of claim 1.

13. 1. A method of driving data, comprising: receiving an input data signal at a differential input node pair of a driver unit cell of a transmit driver circuit; generating an output data signal using the driver unit cell; outputting the output data signal at a differential output node pair of the transmit driver circuit, the transmit driver circuit comprising: a plurality of power switches coupled between the differential output node pair and one or more power rails; a first set of one or more drivers coupled between a first test node of the differential test data path and a first output node of the differential output node pair; a second set of one or more drivers coupled between a second test node of the differential test data path and a second output node of the differential output node pair.

14. powering off the driver unit cell and the plurality of power switches; receiving an input test signal at the differential test data path; driving the input test signals using one or more drivers of the first set and one or more drivers of the second set; 14. The method of claim 13, further comprising: outputting an output test signal at the differential output node pair while the driver unit cell and the plurality of power switches are powered off.

15. The plurality of power switches a first switch coupled between a first power rail of the one or more power rails and the first output node of the differential output node pair; a second switch coupled between a second power supply rail of the one or more power supply rails and the first output node of the differential output node pair, the first power supply rail having a first voltage that is different from a second voltage of the second power supply rail; a third switch coupled between the first power supply rail and the second output node of the differential output node pair; a fourth switch coupled between the second power supply rail and the second output node of the differential output node pair.

16. The method further includes generating the input data signal based on two main cursors, one post-cursor, and three pre-cursors for equalization, wherein generating the input data signal includes: programming a first one of the main cursors to function as one of the precursors having a cursor weighting higher than a nominal cursor weighting for the one of the precursors; programming a second one of the main cursors to function as the post-cursor with a cursor weighting higher than a nominal cursor weighting for the post-cursor; programming the one of the precursors to function as the first one of the main cursors with a cursor weighting that is lower than a nominal cursor weighting for the first one of the main cursors; or and programming the post-cursor to function as the second one of the main cursors with a cursor weighting that is lower than a nominal cursor weighting for the second one of the main cursors.

Citation Information

Patent Citations

  • Differential signal transmitting circuit and differential signal transmitting apparatus

    JP2006340266A

  • Die device having configurable input / output and its control method

    JP2010518630A

  • Transmission driver circuit

    JP2012169905A

  • Transmitter configured to inject test signals to test ac-coupled interconnections

    JP2018523346A

  • Electrostatic discharge (ESD) isolated input / output (I / O) circuitry

    JP2019517127A