Data communication system

The method uses two transmission paths with different data rates to notify unlock status without additional lines, addressing the inefficiency in conventional CDR methods by maintaining data transfer and reducing signal lines.

WO2026141385A1PCT designated stage Publication Date: 2026-07-02CEREBRA SYSTEM SOLUTIONS INC +1

Patent Information

Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
CEREBRA SYSTEM SOLUTIONS INC
Filing Date
2025-12-23
Publication Date
2026-07-02

AI Technical Summary

Technical Problem

Conventional data communication methods using the CDR method require additional feedback lines to notify the transmission unit of a lock release, increasing the number of signal lines and complicating system design, and reducing data processing efficiency when one transmission path becomes unlocked.

Method used

A data communication method utilizing two transmission paths with different data rates, where one path maintains its transmission mode and transmits notification data on the other path when the first path becomes unlocked, eliminating the need for a dedicated return line.

Benefits of technology

Efficiently notifies unlock status while reducing the number of signal lines required, maintaining data transfer function and preventing a decrease in data processing efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure JP2025045052_02072026_PF_FP_ABST
    Figure JP2025045052_02072026_PF_FP_ABST
Patent Text Reader

Abstract

This data communication system uses a first transfer path for transmitting data from a first data processing unit to a second data processing unit by a CDR scheme, and a second transfer path for transmitting data from the second data processing unit to the first data processing unit by the CDR scheme, to transmit notification data indicating the reception state of the second data processing unit from the second data processing unit to the first data processing unit via the second transfer path when data communication is performed by the first transfer path, wherein, in cases in which an unlocked state arises when the transfer mode of the data communication in the first transfer path is a high-speed transfer mode, the data communication system transmits notification data indicating that the unlocked state is in effect while maintaining the transfer mode of the second transfer path.
Need to check novelty before this filing date? Find Prior Art

Description

Data communication method

[0006] ,

[0001] This disclosure relates to a data communication method.

[0002] For example, as a method for performing high-speed data communication between LSIs and the like, a clock and data recovery (CDR) method is used. The CDR method is a method of receiving a signal on a transmission line in which a clock is superimposed on data and separating the clock and data, and is used in USB (Universal Serial Bus) and HDMI (registered trademark) (High-Definition Multimedia Interface), etc.

[0003] In the CDR method, in a preliminary initial operation (training operation), a training pattern is sent from a transmission unit, and a reception unit receives the training pattern and adjusts the timing for receiving data. Thereafter, the transmission unit transmits data at the same timing as the training operation, and the reception unit receives data on the premise of the adjusted timing. As long as the transmission unit continues to transmit data at the same timing as the training operation, the reception unit can correctly receive the data of the transmission unit. This is called a locked state.

[0004] For some reason, for example, due to noise or the like, the timing transmitted from the transmission unit may deviate from the training operation. When communicating data at high speed, this timing deviation means that the reception unit cannot immediately receive the data. That is, the lock is released. When the lock is released, it is necessary for the transmission unit to perform some recovery process, such as restarting the training.

[0005] However, the transmission unit has no means of knowing that the lock is released unless it is notified that the data has not been received from the reception unit. For this reason, in a data communication circuit that performs data communication by the CDR method, it is common to provide a feedback line for notifying the transmission unit from the reception unit that the lock is released (see Non-Patent Document 1 and Non-Patent Document 2).

[0006] "V-by-One HS Standard", [online], THine Electronics Inc., [Retrieved November 28, 2024], Internet <URL: https: / / www.thine.co.jp / files / user / img / corporate / VBOSTD-V1P52-0000_Abridged%2BEdition.pdf> "Intel Aria 10 Transceiver PHY User Guide", [online], Intel Corporation, [Retrieved November 28, 2024], Internet <URL: https: / / www.intel.co.jp / jp / content / www / jp / ja / docs / programmable / 683617 / current / lock-to-data-mode. html>

[0007] The aforementioned return lines are required for each data communication circuit, thus increasing the number of signal lines needed for data communication. This increase in signal lines complicates system design and increases the overall system size.

[0008] In response to this, when data communication is bidirectional, a method is being considered in which the other transmission path notifies the other side when one transmission path becomes unlocked. In this method, the data rate of the other transmission path is reduced to a level where data communication is still possible even when the lock is released, and then the notification is sent. As a result, the notification procedure becomes complex, and if data communication is already underway on the other transmission path, it must be interrupted even though it would not normally need to be interrupted, which tends to reduce data processing efficiency.

[0009] This disclosure is made based on the circumstances described above, and aims to provide a data communication method that efficiently notifies the unlock status while reducing the number of signal lines required in CDR data communication.

[0010] A data communication method according to one aspect of this disclosure uses a first transmission path for transmitting data from the first data processing unit to the second data processing unit using the CDR method, and a second transmission path for transmitting data from the second data processing unit to the first data processing unit using the CDR method, when data communication is performed via the first transmission path, the second data processing unit transmits notification data indicating the reception status of the second data processing unit to the first data processing unit via the second transmission path, and when data communication is performed via the second transmission path, the second 1. A data communication method in which notification data indicating the reception status of a data processing unit is transmitted from the first data processing unit to the second data processing unit via the first transmission line, wherein the first and second transmission lines have two transmission modes with different data rates, and when the data communication transmission mode of the first transmission line is a high-speed transmission mode with a large data rate, if the first transmission line enters an unlocked state where the timing of the CDR method is missed, the transmission mode of the second transmission line is maintained, and data indicating that it is in an unlocked state is transmitted as notification data for the second transmission line.

[0011] The data communication method disclosed herein can efficiently notify the unlock status while reducing the number of signal lines required in data communication using the CDR method.

[0012] Figure 1 is a circuit diagram of a data communication circuit used in a data communication system according to one embodiment of the present disclosure. Figure 2 is a schematic diagram showing one-way data communication in the data communication circuit of Figure 1. Figure 3 is a schematic diagram showing two-way data communication in the data communication circuit of Figure 1. Figure 4 is a schematic diagram showing data communication including a refresh operation in the second transmission path of the data communication circuit of Figure 1.

[0013] [Description of Embodiments of This Disclosure] Factors that can lead to an unlocked state include cases where the voltage level or cycle of the data sent by the transmitting side deviates from the specified values, or where the transmitting side's data is correct but the receiving side is unable to keep up with the transmitting side's data. These are likely to occur due to external noise or the accumulation of timing adjustment errors over time. Conventionally, signal lines used for bidirectional communication are often physically located close together, and timing adjustments are likely to be performed simultaneously. Therefore, in bidirectional communication, it has been inferred that if one transmission line is in an unlocked state, the other transmission line is also in an unlocked state or close to it. For this reason, conventional methods reduce the data rate of the transmission line to ensure that notification data can be reliably transmitted.

[0014] However, the Disclosers found through their investigation that it is rare for the other transmission path to be unlocked or nearly unlocked even if one transmission path is unlocked, and that in most cases a lock is established. Based on this, they established the data communication method disclosed herein.

[0015] (1) That is, a data communication system according to one aspect of the present disclosure uses a first transmission path for transmitting data from the first data processing unit to the second data processing unit using the CDR method, and a second transmission path for transmitting data from the second data processing unit to the first data processing unit using the CDR method when data communication is performed via the first transmission path, and when data communication is performed via the second transmission path, the second data processing unit transmits notification data indicating the reception status of the second data processing unit to the first data processing unit via the second transmission path, and when data communication is performed via the second transmission path, A data communication method in which notification data indicating the reception status of the first data processing unit is transmitted from the first data processing unit to the second data processing unit via the first transmission path, wherein the first transmission path and the second transmission path have two transmission modes with different data rates, and when the data communication transmission mode of the first transmission path is a high-speed transmission mode with a large data rate, if the first transmission path enters an unlocked state where the timing of the CDR method is missed, the transmission mode of the second transmission path is maintained, and data indicating that it is in an unlocked state is transmitted as notification data for the second transmission path.

[0016] This data communication method uses a first transmission line and a second transmission line to perform bidirectional data communication using the CDR method. When data communication is performed using the first transmission line, notification data is transmitted using the second transmission line, and when data communication is performed using the second transmission line, notification data is transmitted using the first transmission line. Therefore, this data communication method does not require a dedicated return line, and the number of required signal lines can be reduced. In addition, this data communication method maintains the transmission mode of the second transmission line and transmits data indicating that it is in an unlocked state as notification data for the second transmission line, thus maintaining the data transfer function of the second transmission line and preventing a decrease in data processing efficiency.

[0017] (2) In the data communication method described in (1) above, when the transmission mode of the data communication of the second transmission path is in high-speed transmission mode, if the second transmission path becomes unlocked, it is preferable to maintain the transmission mode of the first transmission path and transmit data indicating that it is unlocked as notification data for the first transmission path. By adopting a similar configuration in the case of the second transmission path becoming unlocked, a further decrease in data processing efficiency can be suppressed.

[0018] (3) In the data communication method of (1) or (2) above, when the transmission mode of data communication of the second transmission line is the high-speed transmission mode, if a dormant state in which no signal transition occurs in the second transmission line is maintained for a predetermined period of time, it is preferable to perform a refresh operation that forcibly and temporarily generates a signal transition in the second transmission line. By providing such a refresh operation, it is possible to suppress timing discrepancies in the locked state of the second transmission line and make it easier to maintain the transmission mode of the second transmission line.

[0019] (4) In any of the data communication methods described in (1) to (3) above, if the unlocked state of the first transmission line is not resolved even after a predetermined period has elapsed since sending data indicating that the second transmission line is in an unlocked state as notification data for the second transmission line, it is preferable to perform a recovery process for the second transmission line from the unlocked state. By performing this recovery process, recovery is possible even if the first transmission line and the second transmission line are both in an unlocked state at the same time.

[0020] [Details of Embodiments of the Disclosure] A data communication method according to one embodiment of the Disclosure will be described below.

[0021] [Data Communication Circuit] This data communication method can be implemented using the data communication circuit 1 shown in Figure 1. The data communication circuit 1 is a data communication circuit that performs data communication between the first data processing unit 10 and the second data processing unit 20, and is particularly suitable for use when communicating data at high speed.

[0022] The first data processing unit 10 and the second data processing unit 20 may be mounted on separate semiconductor devices X1 and X2, respectively. Examples of semiconductor devices X1 and X2 include large-scale integrated circuits (LSIs). However, the mounting location is not limited to semiconductor devices, as long as the component performs data communication between components. Furthermore, there is no presumption that it can be used for data communication within a single component, for example, within a semiconductor integrated circuit.

[0023] The data communication circuit 1 comprises a first transmission line 30, a second transmission line 40, and a control unit 50.

[0024] <Transmission Path> The first transmission path 30 transmits data from the first data processing unit 10 to the second data processing unit 20 using the CDR method. The second transmission path 40 transmits data from the second data processing unit 20 to the first data processing unit 10 using the CDR method. In other words, the data communication circuit 1 is capable of bidirectional communication.

[0025] In the data communication circuit 1 shown in Figure 1, the first transmission line 30 and the second transmission line 40 are each composed of two signal lines (first signal lines 31 and 32 and second signal lines 41 and 42), and data is transmitted via differential signals through these two signal lines. For example, in the first transmission line 30, currents with opposite phases flow through the positive first signal line 31 and the negative first signal line 32, and the signal is transmitted by the potential difference between them. In this case, even if the same external noise is applied to the positive and negative first signal lines 31 and 32, the noise is canceled out and malfunctions are less likely because the potential difference between the signal lines is observed. By using differential signals in the first transmission line 30 and the second transmission line 40 in this way, immunity to external noise can be improved.

[0026] The first transmission line 30 and the second transmission line 40 have two transmission modes with different data rates. For example, when performing high-speed data communication using only the first transmission line 30, stable data communication can be achieved by using the second transmission line 40 as a low-speed signal line. Further details will be described later.

[0027] In the two transmission modes described above, the high-speed data rate corresponds to the data rate used when communicating data at high speed. On the other hand, the low-speed data rate should be set to a rate that allows for stable data communication even when disturbances such as clock period jitter occur.

[0028] The lower limit of the high-speed data rate of the data communication circuit 1 is preferably 1 Gbps, more preferably 5 Gbps, and even more preferably 50 Gbps. On the other hand, the upper limit of the high-speed data rate of the data communication circuit 1 is not particularly limited, but is practically 100 Gbps.

[0029] The upper limit of the ratio of the low-speed data rate to the high-speed data rate is preferably 1 / 100, and more preferably 1 / 1000. By keeping the ratio of the low-speed data rate to the high-speed data rate below the above upper limit, more stable data communication can be achieved. On the other hand, there is no particular lower limit to the ratio of the low-speed data rate to the high-speed data rate, but it can be, for example, 1 / 2. If the ratio of the low-speed data rate to the high-speed data rate is above the above lower limit, the effect of improving the stability of data communication by providing a low-speed data rate may be insufficient.

[0030] <Data Processing Unit> The first data processing unit 10 includes a first output buffer 11 that drives the first transmission line 30 and a first input buffer 12 that receives signals from the second transmission line 40.

[0031] The first output buffer 11 generates a signal for the data to be transmitted from the first data processing unit 10 to the second data processing unit 20, and its inverted signal, and applies them as voltages to the + side first signal line 31 and the - side first signal line 32, respectively. As a result, the signals are transmitted from the first data processing unit 10 to the second data processing unit 20 through the first transmission line 30.

[0032] Meanwhile, the first input buffer 12 receives signals transmitted from the second data processing unit 20. Since these signals are transmitted via two signal lines, the positive second signal line 41 and the negative second signal line 42, the signals are reproduced by taking the potential difference between these signal lines.

[0033] The second data processing unit 20 includes a second output buffer 21 that drives the second transmission line 40 and a second input buffer 22 that receives signals from the first transmission line 30.

[0034] The configuration of the second output buffer 21 and the second input buffer 22 can be configured in the same way as the first output buffer 11 and the first input buffer 12 of the first data processing unit 10, except that the direction of the transmitted signals is reversed, so a detailed explanation is omitted.

[0035] <Control Unit> The control unit 50 controls data communication. Specifically, when data communication is performed via the first transmission line 30, the control unit 50 controls the transmission of notification data S indicating the reception status of the second data processing unit 20 from the second data processing unit 20 to the first data processing unit 10 via the second transmission line 40, and when data communication is performed via the second transmission line 40, the control unit 50 controls the transmission of notification data S indicating the reception status of the first data processing unit 10 from the first data processing unit 10 to the second data processing unit 20 via the first transmission line 30.

[0036] The control unit 50 is composed of a first control unit 51 located in the first data processing unit 10 and a second control unit 52 located in the second data processing unit 20. The first control unit 51 controls the data to be output to the first output buffer 11 and the data input by the first input buffer 12. The second control unit 52 controls the data to be output to the second output buffer 21 and the data input by the second input buffer 22. The two are controlled independently.

[0037] [Data Communication Method] As described above, the data communication method is performed using the data communication circuit 1 shown in Figure 1. Specifically, the data communication method uses a first transmission line 30 that transmits data from the first data processing unit 10 to the second data processing unit 20 using the CDR method, and a second transmission line 40 that transmits data from the second data processing unit 20 to the first data processing unit 10 using the CDR method. When data communication is performed via the first transmission line 30, notification data S indicating the reception status of the second data processing unit 20 is transmitted from the second data processing unit 20 to the first data processing unit 10 via the second transmission line 40. When data communication is performed via the second transmission line 40, notification data S indicating the reception status of the first data processing unit 10 is transmitted from the first data processing unit 10 to the second data processing unit 20 via the first transmission line 30. The first transmission line 30 and the second transmission line 40 have two transmission modes with different data rates (a high-speed transmission mode with a high data rate and a low-speed transmission mode with a low data rate).

[0038] The operation of the control unit 50 will be explained below using the case where data communication is performed via the first transmission line 30 shown in Figure 2 as an example. The same applies when data communication is performed via the second transmission line 40.

[0039] First, a training operation is performed before data communication takes place. During the training operation, training data is sent from the first output buffer 11 of the first data processing unit 10 to the second data processing unit 20, and the reception timing of the second input buffer 22 of the second data processing unit 20 is determined when transmitting at a high data rate. By determining the reception timing of the second data processing unit 20 in this way, a clock signal is not required to synchronize the first output buffer 11 and the second input buffer 22. When the first control unit 51 transmits normal data D from the first output buffer 11 of the first data processing unit 10 at a high data rate, the second control unit 52 can receive that normal data D in the second input buffer 22 of the second data processing unit 20.

[0040] If, for some unavoidable reason, the transmission timing of the normal data D sent from the first output buffer 11 of the first data processing unit 10 deviates from the timing established during the training operation, the CDR system will enter an unlocked state where the timing is lost, and the second input buffer 22 of the second data processing unit 20 may not be able to receive the normal data D. In such a case, it becomes necessary to perform the training operation again to readjust the timing.

[0041] When the system enters an unlocked state, for example, some data may be lost or replaced with other data, resulting in the second data processing unit 20 no longer receiving consistent data. From this phenomenon, the second data processing unit 20, which receives the normal data D, can determine that it is no longer receiving the normal data D correctly. However, since the transmission of the normal data D from the first output buffer 11 of the first data processing unit 10 and the reception of the normal data D in the second input buffer 22 of the second data processing unit 20 are performed independently, the first data processing unit 10 cannot recognize that the second data processing unit 20 is no longer receiving the normal data D correctly. Therefore, the second control unit 52 needs to send notification data S from the second data processing unit 20 to the first data processing unit 10 to indicate whether or not the normal data D is being received correctly.

[0042] In the data communication circuit 1, the second transmission line 40 is used to transmit the notification data S. The data length of the notification data S (the number of bits required to represent the data) is shorter than the data length of the normal data D, and the transmission frequency can also be kept low, so that the necessary information can be transmitted to the first data processing unit 10 in a timely manner.

[0043] As shown in Figure 2, regular data D and notification data S can consist of a header indicating which data it is and a body representing the data itself. By predetermining the data lengths of regular data D and notification data S, they can be received and processed separately.

[0044] In the data communication circuit 1 configured as described above, it is possible to simultaneously transmit data to the first transmission line 30 and the second transmission line 40. When simultaneously transmitting data, the normal data D to be transmitted and the notification data S are superimposed and transmitted to the first transmission line 30 and the second transmission line 40. By configuring it in this way, it is possible to simultaneously transmit data to the first transmission line 30 and the second transmission line 40 without providing a dedicated return line.

[0045] When simultaneously transmitting data in both directions, on the first transmission line 30, the normal data D to be transmitted from the first data processing unit 10 to the second data processing unit 20 and the notification data S indicating whether the first data processing unit 10 can normally receive the normal data D from the second data processing unit 20 are superimposed. Also, on the second transmission line 40, the normal data D to be transmitted from the second data processing unit 20 to the first data processing unit 10 and the notification data S indicating whether the second data processing unit 20 can normally receive the normal data D from the first data processing unit 10 are superimposed.

[0046] In the above example, the information of the notification data S is whether the reception is normal, and the second data processing unit 20 can notify spontaneously without a request from the first data processing unit 10. The information of the notification data S is not limited to this. In addition to the case where the second data processing unit 20 notifies spontaneously, for example, the first data processing unit 10 designates the information to be returned by the second data processing unit 20 by the notification data S, and the first data processing unit 10 uses the notification data S from the second data processing unit 20 to return the information required by the first data processing unit 10.

[0047] (Processing in the unlocked state) Here, consider the case where the first transmission line 30 is in an unlocked state when the data communication transmission mode of the first transmission line 30 is the high-speed transmission mode.

[0048] In this data communication method, in this case, while maintaining the transmission mode of the second transmission line 40, data indicating that it is in an unlocked state is transmitted as the notification data S of the second transmission line 40. That is, when the transmission mode of the second transmission line 40 is the low-speed transmission mode, the notification data S is transmitted while remaining in the low-speed transmission mode, and when the transmission mode of the second transmission line 40 is the high-speed transmission mode, the notification data S is transmitted while remaining in the high-speed transmission mode.

[0049] In this data communication method, since the transmission mode of the second transmission line 40 is maintained, for example, as shown in FIG. 3, when the second transmission line 40 is in the high-speed transmission mode and normal data D is being transmitted, the high-speed transmission mode is maintained, so the transmission of normal data D can be continued as it is. Only the transmission of data indicating that it is in an unlocked state as the notification data S in FIG. 3 is added. That is, since there is no need to interrupt the transmission of normal data D, it is difficult for the data processing efficiency to decrease. Even when the second transmission line 40 is in the low-speed transmission mode, since it is not deliberately set to the high-speed transmission mode, the normal data D being transmitted in the low-speed transmission mode can be continuously transmitted, and the same effect is achieved.

[0050] When the transmission mode of data communication of the second transmission line 40 is the high-speed transmission mode, as shown in FIG. 4, when a rest state E in which no signal transition occurs is maintained in the second transmission line 40 for a predetermined period P, a refresh operation R that forcibly and temporarily generates a signal transition in the second transmission line 40 may be performed. By providing the refresh operation R in this way, it is possible to prevent a timing deviation from occurring in the locked state of the second transmission line 40, and it is easy to maintain the transmission mode of the second transmission line 40.

[0051] As the lower limit of the predetermined period P, 1 μs is preferable, and 5 μs is more preferable. On the other hand, as the upper limit of the predetermined period P, 1000 μs is preferable, and 500 μs is more preferable. If the predetermined period P is less than the above lower limit, the power of the data communication circuit 1 may increase unnecessarily. Conversely, if the predetermined period P exceeds the above upper limit, there is a possibility that a timing deviation may easily occur.

[0052] The lower limit of the duration T of the refresh operation R is preferably 1 ns, and more preferably 5 ns. On the other hand, the upper limit of the duration T is preferably 10 ns, and more preferably 5 ns. If the duration T is less than the lower limit, there is a risk that timing discrepancies will occur. Conversely, if the duration T exceeds the upper limit, there is a risk that the power of the data communication circuit 1 will increase unnecessarily.

[0053] The refresh pattern used in the refresh operation S3 is not particularly limited, but it is preferable that the frequency of signal transitions is high. For example, a pattern that repeats 0101 at the minimum time interval (data transmission bitrate) can be adopted.

[0054] Once the refresh operation R is complete, the system returns to the idle state E. Similarly, if the idle state E is maintained for a predetermined period P on the second transmission line 40, the refresh operation R is performed again.

[0055] If the first transmission line 30 becomes unlocked while a refresh operation R is being performed on the second transmission line 40, the high-speed transmission mode of the second transmission line 40 is maintained, and data indicating that it is in an unlocked state is transmitted as notification data S in place of the refresh operation R, or in addition to the refresh operation R.

[0056] When the first data processing unit 10 receives notification data via the second transmission line 40 indicating that it is in an unlocked state, the first data processing unit 10 can know that the first transmission line 30 is in an unlocked state. In such cases, recovery can be achieved by, for example, performing another training operation on the first transmission line 30 to readjust the timing.

[0057] Next, consider the case in the data communication method in question where both the first transmission line 30 and the second transmission line 40 are in high-speed transmission mode, and both the first transmission line 30 and the second transmission line 40 become unlocked. In this case, for example, since the first transmission line 30 is unlocked, notification data S indicating that the first transmission line 30 is unlocked is transmitted via the second transmission line 40. However, since the second transmission line 40 is also unlocked, this notification data S does not reach the first processing unit 10. As a result, the timing of the first transmission line 30 is not readjusted, and the first transmission line 30 may not recover. For similar reasons, the timing of the second transmission line 40 may also not recover.

[0058] In this data communication method, to avoid such a situation, if, after sending data indicating that the second transmission line 40 is in an unlocked state as notification data S, the unlocked state of the first transmission line 30 is not resolved even after a predetermined period has elapsed, a recovery process from the unlocked state is performed on the second transmission line 40. By performing this recovery process, recovery is possible even if the first transmission line 30 and the second transmission line 40 are both in an unlocked state at the same time.

[0059] The lower limit of the above-mentioned predetermined period is preferably 1 μs, and more preferably 5 μs. On the other hand, the upper limit of the above-mentioned predetermined period is preferably 1000 μs, and more preferably 500 μs. If the above-mentioned predetermined period is less than the lower limit, there is a risk that the recovery process will be performed again even if the recovery process is not completed. Conversely, if the above-mentioned predetermined period exceeds the upper limit, the recovery time will be too long, which may make efficient operation difficult.

[0060] If the data communication mode of the first transmission line 30 is the low-speed transmission mode, the processing in the unlocked state may be performed in the same way as in the high-speed transmission mode, provided that data communication is performed using the CDR method. Alternatively, since it is less likely that data will not be received even if there is a delay in transmission timing due to the low speed, the processing may be different from that in the high-speed transmission mode. If the data communication in the low-speed transmission mode is not using the CDR method, the unlocked state as defined in the CDR method does not occur, so the above-mentioned operation is unnecessary.

[0061] Furthermore, when the data communication mode of the second transmission line 40 is high-speed transmission mode, processing can be performed in the same way as when the data communication mode of the first transmission line 30 is high-speed transmission mode. That is, when the second transmission line 40 becomes unlocked while the data communication mode of the second transmission line 40 is high-speed transmission mode, the transmission mode of the first transmission line 30 is maintained, and data indicating that it is unlocked is sent as notification data for the first transmission line 30. By adopting the same configuration when the second transmission line 40 becomes unlocked in this way, a decrease in data processing efficiency can be further suppressed.

[0062] <Advantages> This data communication method is a data communication method that performs bidirectional data communication using the CDR method with a first transmission line 30 and a second transmission line 40. When data communication is performed using the first transmission line 30, notification data S is transmitted using the second transmission line 40, and when data communication is performed using the second transmission line 40, notification data S is transmitted using the first transmission line 30. For this reason, this data communication method does not require a dedicated return line, and the number of required signal lines can be reduced. In addition, in this data communication method, while maintaining the transmission mode of the second transmission line 40, data indicating that it is in an unlocked state is transmitted as notification data S for the second transmission line 40, so the data transfer function of the second transmission line 40 is maintained and a decrease in data processing efficiency can be suppressed.

[0063] [Other Embodiments] The embodiments described above do not limit the configuration of the present disclosure. Accordingly, the embodiments described above may omit, substitute or add components of each part of the embodiments based on the description herein and common technical knowledge, and all such omissions should be interpreted as falling within the scope of the present disclosure.

[0064] In the above embodiment, we have described a case where, in both the first and second transmission lines, when the data communication transmission mode of one transmission line is a high-speed transmission mode with a large data rate, the one transmission line becomes unlocked, causing the CDR timing to be missed. In this case, the transmission mode of the other transmission line is maintained, and data indicating that it is in an unlocked state is transmitted as notification data for the other transmission line. However, the present disclosure also intends to describe cases where only one transmission line, i.e., only the first transmission line, has the above configuration.

[0065] In the above embodiment, the case in which the first and second transmission lines transmit data using differential signals was described, but it is also possible to adopt a configuration in which either one or both of the first and second transmission lines transmit data using a single signal line. In this case, although noise immunity will decrease, the number of required signal lines can be further reduced.

[0066] In the above embodiment, the control unit is described as being composed of a first control unit located in the first data processing unit and a second control unit located in the second data processing unit. However, it is not essential that the first and second control units are located in the first and second data processing units, respectively, and they may be located elsewhere. Furthermore, the control unit does not need to be divided into two parts, a first control unit and a second control unit; it is also possible to have a configuration in which a single control unit controls everything. However, from the viewpoint of layout efficiency, it is preferable that the control unit is composed of a first control unit located in the first data processing unit and a second control unit located in the second data processing unit.

[0067] The data communication method disclosed herein can efficiently notify the unlock status while reducing the number of signal lines required in data communication using the CDR method.

[0068] 1 Data communication circuit 10 First data processing unit 11 First output buffer 12 First input buffer 20 Second data processing unit 21 Second output buffer 22 Second input buffer 30 First transmission line 31, 32 First signal line 40 Second transmission line 41, 42 Second signal line 50 Control unit 51 First control unit 52 Second control unit X1, X2 Semiconductor device D Normal data S Notification data E Hibernation state R Refresh operation P Predetermined period T Duration

Claims

1. A data communication method comprising: a first transmission path for transmitting data from the first data processing unit to the second data processing unit using the CDR method, and a second transmission path for transmitting data from the second data processing unit to the first data processing unit using the CDR method, wherein when data communication is performed via the first transmission path, notification data indicating the reception status of the second data processing unit is transmitted from the second data processing unit to the first data processing unit via the second transmission path, and when data communication is performed via the second transmission path, notification data indicating the reception status of the first data processing unit is transmitted from the first data processing unit to the second data processing unit via the first transmission path, wherein the first transmission path and the second transmission path have two transmission modes with different data rates. A data communication method in which, when the transmission mode of the data communication of the first transmission line described above is a high-speed transmission mode with a large data rate, the first transmission line enters an unlocked state where the timing of the CDR method is missed, and the transmission mode of the second transmission line described above is maintained, data indicating that the second transmission line is in an unlocked state is transmitted as notification data for the second transmission line described above.

2. When the transmission mode of the data communication of the second transmission line is in high-speed transmission mode, and the second transmission line becomes unlocked, the data communication method according to claim 1, which maintains the transmission mode of the first transmission line and transmits data indicating that it is in an unlocked state as notification data for the first transmission line.

3. The data communication method according to claim 1, wherein when the transmission mode of data communication on the second transmission path is a high-speed transmission mode, and a dormant state in which no signal transition occurs on the second transmission path is maintained for a predetermined period of time, a refresh operation is performed to forcibly and temporarily generate a signal transition on the second transmission path.

4. The data communication method according to claim 1, wherein, after transmitting data indicating that the second transmission line is in an unlocked state as notification data for the second transmission line, if the unlocked state of the first transmission line is not resolved even after a predetermined period of time has elapsed, the second transmission line performs a recovery process from the unlocked state.